diff options
Diffstat (limited to 'fs')
247 files changed, 20136 insertions, 13755 deletions
diff --git a/fs/9p/vfs_file.c b/fs/9p/vfs_file.c index d384a8b..aa5ecf4 100644 --- a/fs/9p/vfs_file.c +++ b/fs/9p/vfs_file.c @@ -183,7 +183,7 @@ static int v9fs_file_do_lock(struct file *filp, int cmd, struct file_lock *fl) else flock.length = fl->fl_end - fl->fl_start + 1; flock.proc_id = fl->fl_pid; - flock.client_id = utsname()->nodename; + flock.client_id = fid->clnt->name; if (IS_SETLKW(cmd)) flock.flags = P9_LOCK_FLAGS_BLOCK; @@ -260,7 +260,7 @@ static int v9fs_file_getlock(struct file *filp, struct file_lock *fl) else glock.length = fl->fl_end - fl->fl_start + 1; glock.proc_id = fl->fl_pid; - glock.client_id = utsname()->nodename; + glock.client_id = fid->clnt->name; res = p9_client_getlock_dotl(fid, &glock); if (res < 0) diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c index 25b018e..94de6d1 100644 --- a/fs/9p/vfs_inode.c +++ b/fs/9p/vfs_inode.c @@ -146,7 +146,7 @@ static umode_t p9mode2unixmode(struct v9fs_session_info *v9ses, char type = 0, ext[32]; int major = -1, minor = -1; - strncpy(ext, stat->extension, sizeof(ext)); + strlcpy(ext, stat->extension, sizeof(ext)); sscanf(ext, "%c %u %u", &type, &major, &minor); switch (type) { case 'c': @@ -1186,7 +1186,7 @@ v9fs_stat2inode(struct p9_wstat *stat, struct inode *inode, * this even with .u extension. So check * for non NULL stat->extension */ - strncpy(ext, stat->extension, sizeof(ext)); + strlcpy(ext, stat->extension, sizeof(ext)); /* HARDLINKCOUNT %u */ sscanf(ext, "%13s %u", tag_name, &i_nlink); if (!strncmp(tag_name, "HARDLINKCOUNT", 13)) diff --git a/fs/affs/file.c b/fs/affs/file.c index af3261b..776e393 100644 --- a/fs/affs/file.c +++ b/fs/affs/file.c @@ -836,7 +836,7 @@ affs_truncate(struct inode *inode) struct address_space *mapping = inode->i_mapping; struct page *page; void *fsdata; - u32 size = inode->i_size; + loff_t size = inode->i_size; int res; res = mapping->a_ops->write_begin(NULL, mapping, size, 0, 0, &page, &fsdata); diff --git a/fs/afs/dir.c b/fs/afs/dir.c index 34494fb..646337d 100644 --- a/fs/afs/dir.c +++ b/fs/afs/dir.c @@ -685,16 +685,12 @@ not_found: spin_unlock(&dentry->d_lock); out_bad: - if (dentry->d_inode) { - /* don't unhash if we have submounts */ - if (have_submounts(dentry)) - goto out_skip; - } + /* don't unhash if we have submounts */ + if (check_submounts_and_drop(dentry) != 0) + goto out_skip; _debug("dropping dentry %s/%s", parent->d_name.name, dentry->d_name.name); - shrink_dcache_parent(dentry); - d_drop(dentry); dput(parent); key_put(key); @@ -755,10 +751,6 @@ static int afs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) _enter("{%x:%u},{%s},%ho", dvnode->fid.vid, dvnode->fid.vnode, dentry->d_name.name, mode); - ret = -ENAMETOOLONG; - if (dentry->d_name.len >= AFSNAMEMAX) - goto error; - key = afs_request_key(dvnode->volume->cell); if (IS_ERR(key)) { ret = PTR_ERR(key); @@ -820,10 +812,6 @@ static int afs_rmdir(struct inode *dir, struct dentry *dentry) _enter("{%x:%u},{%s}", dvnode->fid.vid, dvnode->fid.vnode, dentry->d_name.name); - ret = -ENAMETOOLONG; - if (dentry->d_name.len >= AFSNAMEMAX) - goto error; - key = afs_request_key(dvnode->volume->cell); if (IS_ERR(key)) { ret = PTR_ERR(key); @@ -940,10 +928,6 @@ static int afs_create(struct inode *dir, struct dentry *dentry, umode_t mode, _enter("{%x:%u},{%s},%ho,", dvnode->fid.vid, dvnode->fid.vnode, dentry->d_name.name, mode); - ret = -ENAMETOOLONG; - if (dentry->d_name.len >= AFSNAMEMAX) - goto error; - key = afs_request_key(dvnode->volume->cell); if (IS_ERR(key)) { ret = PTR_ERR(key); @@ -1009,10 +993,6 @@ static int afs_link(struct dentry *from, struct inode *dir, dvnode->fid.vid, dvnode->fid.vnode, dentry->d_name.name); - ret = -ENAMETOOLONG; - if (dentry->d_name.len >= AFSNAMEMAX) - goto error; - key = afs_request_key(dvnode->volume->cell); if (IS_ERR(key)) { ret = PTR_ERR(key); @@ -1057,10 +1037,6 @@ static int afs_symlink(struct inode *dir, struct dentry *dentry, dvnode->fid.vid, dvnode->fid.vnode, dentry->d_name.name, content); - ret = -ENAMETOOLONG; - if (dentry->d_name.len >= AFSNAMEMAX) - goto error; - ret = -EINVAL; if (strlen(content) >= AFSPATHMAX) goto error; @@ -1131,10 +1107,6 @@ static int afs_rename(struct inode *old_dir, struct dentry *old_dentry, new_dvnode->fid.vid, new_dvnode->fid.vnode, new_dentry->d_name.name); - ret = -ENAMETOOLONG; - if (new_dentry->d_name.len >= AFSNAMEMAX) - goto error; - key = afs_request_key(orig_dvnode->volume->cell); if (IS_ERR(key)) { ret = PTR_ERR(key); diff --git a/fs/autofs4/dev-ioctl.c b/fs/autofs4/dev-ioctl.c index 743c7c2..0f00da3 100644 --- a/fs/autofs4/dev-ioctl.c +++ b/fs/autofs4/dev-ioctl.c @@ -183,13 +183,14 @@ static int autofs_dev_ioctl_protosubver(struct file *fp, return 0; } +/* Find the topmost mount satisfying test() */ static int find_autofs_mount(const char *pathname, struct path *res, int test(struct path *path, void *data), void *data) { struct path path; - int err = kern_path(pathname, 0, &path); + int err = kern_path_mountpoint(AT_FDCWD, pathname, &path, 0); if (err) return err; err = -ENOENT; @@ -197,10 +198,9 @@ static int find_autofs_mount(const char *pathname, if (path.dentry->d_sb->s_magic == AUTOFS_SUPER_MAGIC) { if (test(&path, data)) { path_get(&path); - if (!err) /* already found some */ - path_put(res); *res = path; err = 0; + break; } } if (!follow_up(&path)) @@ -486,12 +486,11 @@ static int autofs_dev_ioctl_askumount(struct file *fp, * mount if there is one or 0 if it isn't a mountpoint. * * If we aren't supplied with a file descriptor then we - * lookup the nameidata of the path and check if it is the - * root of a mount. If a type is given we are looking for - * a particular autofs mount and if we don't find a match - * we return fail. If the located nameidata path is the - * root of a mount we return 1 along with the super magic - * of the mount or 0 otherwise. + * lookup the path and check if it is the root of a mount. + * If a type is given we are looking for a particular autofs + * mount and if we don't find a match we return fail. If the + * located path is the root of a mount we return 1 along with + * the super magic of the mount or 0 otherwise. * * In both cases the the device number (as returned by * new_encode_dev()) is also returned. @@ -519,9 +518,11 @@ static int autofs_dev_ioctl_ismountpoint(struct file *fp, if (!fp || param->ioctlfd == -1) { if (autofs_type_any(type)) - err = kern_path(name, LOOKUP_FOLLOW, &path); + err = kern_path_mountpoint(AT_FDCWD, + name, &path, LOOKUP_FOLLOW); else - err = find_autofs_mount(name, &path, test_by_type, &type); + err = find_autofs_mount(name, &path, + test_by_type, &type); if (err) goto out; devid = new_encode_dev(path.dentry->d_sb->s_dev); diff --git a/fs/bio-integrity.c b/fs/bio-integrity.c index 8fb42916..6025084 100644 --- a/fs/bio-integrity.c +++ b/fs/bio-integrity.c @@ -716,13 +716,14 @@ int bioset_integrity_create(struct bio_set *bs, int pool_size) return 0; bs->bio_integrity_pool = mempool_create_slab_pool(pool_size, bip_slab); - - bs->bvec_integrity_pool = biovec_create_pool(bs, pool_size); - if (!bs->bvec_integrity_pool) + if (!bs->bio_integrity_pool) return -1; - if (!bs->bio_integrity_pool) + bs->bvec_integrity_pool = biovec_create_pool(bs, pool_size); + if (!bs->bvec_integrity_pool) { + mempool_destroy(bs->bio_integrity_pool); return -1; + } return 0; } diff --git a/fs/cachefiles/interface.c b/fs/cachefiles/interface.c index d4c1206..43eb559 100644 --- a/fs/cachefiles/interface.c +++ b/fs/cachefiles/interface.c @@ -378,6 +378,31 @@ static void cachefiles_sync_cache(struct fscache_cache *_cache) } /* + * check if the backing cache is updated to FS-Cache + * - called by FS-Cache when evaluates if need to invalidate the cache + */ +static bool cachefiles_check_consistency(struct fscache_operation *op) +{ + struct cachefiles_object *object; + struct cachefiles_cache *cache; + const struct cred *saved_cred; + int ret; + + _enter("{OBJ%x}", op->object->debug_id); + + object = container_of(op->object, struct cachefiles_object, fscache); + cache = container_of(object->fscache.cache, + struct cachefiles_cache, cache); + + cachefiles_begin_secure(cache, &saved_cred); + ret = cachefiles_check_auxdata(object); + cachefiles_end_secure(cache, saved_cred); + + _leave(" = %d", ret); + return ret; +} + +/* * notification the attributes on an object have changed * - called with reads/writes excluded by FS-Cache */ @@ -522,4 +547,5 @@ const struct fscache_cache_ops cachefiles_cache_ops = { .write_page = cachefiles_write_page, .uncache_page = cachefiles_uncache_page, .dissociate_pages = cachefiles_dissociate_pages, + .check_consistency = cachefiles_check_consistency, }; diff --git a/fs/cachefiles/internal.h b/fs/cachefiles/internal.h index 4938251..5349473 100644 --- a/fs/cachefiles/internal.h +++ b/fs/cachefiles/internal.h @@ -235,6 +235,7 @@ extern int cachefiles_set_object_xattr(struct cachefiles_object *object, struct cachefiles_xattr *auxdata); extern int cachefiles_update_object_xattr(struct cachefiles_object *object, struct cachefiles_xattr *auxdata); +extern int cachefiles_check_auxdata(struct cachefiles_object *object); extern int cachefiles_check_object_xattr(struct cachefiles_object *object, struct cachefiles_xattr *auxdata); extern int cachefiles_remove_object_xattr(struct cachefiles_cache *cache, diff --git a/fs/cachefiles/xattr.c b/fs/cachefiles/xattr.c index 2476e51..34c88b8 100644 --- a/fs/cachefiles/xattr.c +++ b/fs/cachefiles/xattr.c @@ -157,6 +157,42 @@ int cachefiles_update_object_xattr(struct cachefiles_object *object, } /* + * check the consistency between the backing cache and the FS-Cache cookie + */ +int cachefiles_check_auxdata(struct cachefiles_object *object) +{ + struct cachefiles_xattr *auxbuf; + struct dentry *dentry = object->dentry; + unsigned int dlen; + int ret; + + ASSERT(dentry); + ASSERT(dentry->d_inode); + ASSERT(object->fscache.cookie->def->check_aux); + + auxbuf = kmalloc(sizeof(struct cachefiles_xattr) + 512, GFP_KERNEL); + if (!auxbuf) + return -ENOMEM; + + auxbuf->len = vfs_getxattr(dentry, cachefiles_xattr_cache, + &auxbuf->type, 512 + 1); + if (auxbuf->len < 1) + return -ESTALE; + + if (auxbuf->type != object->fscache.cookie->def->type) + return -ESTALE; + + dlen = auxbuf->len - 1; + ret = fscache_check_aux(&object->fscache, &auxbuf->data, dlen); + + kfree(auxbuf); + if (ret != FSCACHE_CHECKAUX_OKAY) + return -ESTALE; + + return 0; +} + +/* * check the state xattr on a cache file * - return -ESTALE if the object should be deleted */ diff --git a/fs/ceph/Kconfig b/fs/ceph/Kconfig index 49bc782..ac9a2ef 100644 --- a/fs/ceph/Kconfig +++ b/fs/ceph/Kconfig @@ -16,3 +16,12 @@ config CEPH_FS If unsure, say N. +if CEPH_FS +config CEPH_FSCACHE + bool "Enable Ceph client caching support" + depends on CEPH_FS=m && FSCACHE || CEPH_FS=y && FSCACHE=y + help + Choose Y here to enable persistent, read-only local + caching support for Ceph clients using FS-Cache + +endif diff --git a/fs/ceph/Makefile b/fs/ceph/Makefile index bd35212..32e3010 100644 --- a/fs/ceph/Makefile +++ b/fs/ceph/Makefile @@ -9,3 +9,4 @@ ceph-y := super.o inode.o dir.o file.o locks.o addr.o ioctl.o \ mds_client.o mdsmap.o strings.o ceph_frag.o \ debugfs.o +ceph-$(CONFIG_CEPH_FSCACHE) += cache.o diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c index 5318a3b..6df8bd4 100644 --- a/fs/ceph/addr.c +++ b/fs/ceph/addr.c @@ -11,6 +11,7 @@ #include "super.h" #include "mds_client.h" +#include "cache.h" #include <linux/ceph/osd_client.h> /* @@ -70,15 +71,16 @@ static int ceph_set_page_dirty(struct page *page) struct address_space *mapping = page->mapping; struct inode *inode; struct ceph_inode_info *ci; - int undo = 0; struct ceph_snap_context *snapc; + int ret; if (unlikely(!mapping)) return !TestSetPageDirty(page); - if (TestSetPageDirty(page)) { + if (PageDirty(page)) { dout("%p set_page_dirty %p idx %lu -- already dirty\n", mapping->host, page, page->index); + BUG_ON(!PagePrivate(page)); return 0; } @@ -107,35 +109,19 @@ static int ceph_set_page_dirty(struct page *page) snapc, snapc->seq, snapc->num_snaps); spin_unlock(&ci->i_ceph_lock); - /* now adjust page */ - spin_lock_irq(&mapping->tree_lock); - if (page->mapping) { /* Race with truncate? */ - WARN_ON_ONCE(!PageUptodate(page)); - account_page_dirtied(page, page->mapping); - radix_tree_tag_set(&mapping->page_tree, - page_index(page), PAGECACHE_TAG_DIRTY); - - /* - * Reference snap context in page->private. Also set - * PagePrivate so that we get invalidatepage callback. - */ - page->private = (unsigned long)snapc; - SetPagePrivate(page); - } else { - dout("ANON set_page_dirty %p (raced truncate?)\n", page); - undo = 1; - } - - spin_unlock_irq(&mapping->tree_lock); - - if (undo) - /* whoops, we failed to dirty the page */ - ceph_put_wrbuffer_cap_refs(ci, 1, snapc); + /* + * Reference snap context in page->private. Also set + * PagePrivate so that we get invalidatepage callback. + */ + BUG_ON(PagePrivate(page)); + page->private = (unsigned long)snapc; + SetPagePrivate(page); - __mark_inode_dirty(mapping->host, I_DIRTY_PAGES); + ret = __set_page_dirty_nobuffers(page); + WARN_ON(!PageLocked(page)); + WARN_ON(!page->mapping); - BUG_ON(!PageDirty(page)); - return 1; + return ret; } /* @@ -150,11 +136,19 @@ static void ceph_invalidatepage(struct page *page, unsigned int offset, struct ceph_inode_info *ci; struct ceph_snap_context *snapc = page_snap_context(page); - BUG_ON(!PageLocked(page)); - BUG_ON(!PagePrivate(page)); - BUG_ON(!page->mapping); - inode = page->mapping->host; + ci = ceph_inode(inode); + + if (offset != 0 || length != PAGE_CACHE_SIZE) { + dout("%p invalidatepage %p idx %lu partial dirty page %u~%u\n", + inode, page, page->index, offset, length); + return; + } + + ceph_invalidate_fscache_page(inode, page); + + if (!PagePrivate(page)) + return; /* * We can get non-dirty pages here due to races between @@ -164,31 +158,28 @@ static void ceph_invalidatepage(struct page *page, unsigned int offset, if (!PageDirty(page)) pr_err("%p invalidatepage %p page not dirty\n", inode, page); - if (offset == 0 && length == PAGE_CACHE_SIZE) - ClearPageChecked(page); + ClearPageChecked(page); - ci = ceph_inode(inode); - if (offset == 0 && length == PAGE_CACHE_SIZE) { - dout("%p invalidatepage %p idx %lu full dirty page\n", - inode, page, page->index); - ceph_put_wrbuffer_cap_refs(ci, 1, snapc); - ceph_put_snap_context(snapc); - page->private = 0; - ClearPagePrivate(page); - } else { - dout("%p invalidatepage %p idx %lu partial dirty page %u(%u)\n", - inode, page, page->index, offset, length); - } + dout("%p invalidatepage %p idx %lu full dirty page\n", + inode, page, page->index); + + ceph_put_wrbuffer_cap_refs(ci, 1, snapc); + ceph_put_snap_context(snapc); + page->private = 0; + ClearPagePrivate(page); } -/* just a sanity check */ static int ceph_releasepage(struct page *page, gfp_t g) { struct inode *inode = page->mapping ? page->mapping->host : NULL; dout("%p releasepage %p idx %lu\n", inode, page, page->index); WARN_ON(PageDirty(page)); - WARN_ON(PagePrivate(page)); - return 0; + + /* Can we release the page from the cache? */ + if (!ceph_release_fscache_page(page, g)) + return 0; + + return !PagePrivate(page); } /* @@ -198,11 +189,16 @@ static int readpage_nounlock(struct file *filp, struct page *page) { struct inode *inode = file_inode(filp); struct ceph_inode_info *ci = ceph_inode(inode); - struct ceph_osd_client *osdc = + struct ceph_osd_client *osdc = &ceph_inode_to_client(inode)->client->osdc; int err = 0; u64 len = PAGE_CACHE_SIZE; + err = ceph_readpage_from_fscache(inode, page); + + if (err == 0) + goto out; + dout("readpage inode %p file %p page %p index %lu\n", inode, filp, page, page->index); err = ceph_osdc_readpages(osdc, ceph_vino(inode), &ci->i_layout, @@ -220,6 +216,9 @@ static int readpage_nounlock(struct file *filp, struct page *page) } SetPageUptodate(page); + if (err == 0) + ceph_readpage_to_fscache(inode, page); + out: return err < 0 ? err : 0; } @@ -262,6 +261,7 @@ static void finish_read(struct ceph_osd_request *req, struct ceph_msg *msg) page->index); flush_dcache_page(page); SetPageUptodate(page); + ceph_readpage_to_fscache(inode, page); unlock_page(page); page_cache_release(page); bytes -= PAGE_CACHE_SIZE; @@ -331,11 +331,12 @@ static int start_read(struct inode *inode, struct list_head *page_list, int max) page = list_entry(page_list->prev, struct page, lru); BUG_ON(PageLocked(page)); list_del(&page->lru); - + dout("start_read %p adding %p idx %lu\n", inode, page, page->index); if (add_to_page_cache_lru(page, &inode->i_data, page->index, GFP_NOFS)) { + ceph_fscache_uncache_page(inode, page); page_cache_release(page); dout("start_read %p add_to_page_cache failed %p\n", inode, page); @@ -378,6 +379,12 @@ static int ceph_readpages(struct file *file, struct address_space *mapping, int rc = 0; int max = 0; + rc = ceph_readpages_from_fscache(mapping->host, mapping, page_list, + &nr_pages); + + if (rc == 0) + goto out; + if (fsc->mount_options->rsize >= PAGE_CACHE_SIZE) max = (fsc->mount_options->rsize + PAGE_CACHE_SIZE - 1) >> PAGE_SHIFT; @@ -392,6 +399,8 @@ static int ceph_readpages(struct file *file, struct address_space *mapping, BUG_ON(rc == 0); } out: + ceph_fscache_readpages_cancel(inode, page_list); + dout("readpages %p file %p ret %d\n", inode, file, rc); return rc; } @@ -497,6 +506,8 @@ static int writepage_nounlock(struct page *page, struct writeback_control *wbc) CONGESTION_ON_THRESH(fsc->mount_options->congestion_kb)) set_bdi_congested(&fsc->backing_dev_info, BLK_RW_ASYNC); + ceph_readpage_to_fscache(inode, page); + set_page_writeback(page); err = ceph_osdc_writepages(osdc, ceph_vino(inode), &ci->i_layout, snapc, @@ -552,7 +563,6 @@ static void ceph_release_pages(struct page **pages, int num) pagevec_release(&pvec); } - /* * async writeback completion handler. * diff --git a/fs/ceph/cache.c b/fs/ceph/cache.c new file mode 100644 index 0000000..6bfe65e --- /dev/null +++ b/fs/ceph/cache.c @@ -0,0 +1,398 @@ +/* + * Ceph cache definitions. + * + * Copyright (C) 2013 by Adfin Solutions, Inc. All Rights Reserved. + * Written by Milosz Tanski (milosz@adfin.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 + * as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to: + * Free Software Foundation + * 51 Franklin Street, Fifth Floor + * Boston, MA 02111-1301 USA + * + */ + +#include "super.h" +#include "cache.h" + +struct ceph_aux_inode { + struct timespec mtime; + loff_t size; +}; + +struct fscache_netfs ceph_cache_netfs = { + .name = "ceph", + .version = 0, +}; + +static uint16_t ceph_fscache_session_get_key(const void *cookie_netfs_data, + void *buffer, uint16_t maxbuf) +{ + const struct ceph_fs_client* fsc = cookie_netfs_data; + uint16_t klen; + + klen = sizeof(fsc->client->fsid); + if (klen > maxbuf) + return 0; + + memcpy(buffer, &fsc->client->fsid, klen); + return klen; +} + +static const struct fscache_cookie_def ceph_fscache_fsid_object_def = { + .name = "CEPH.fsid", + .type = FSCACHE_COOKIE_TYPE_INDEX, + .get_key = ceph_fscache_session_get_key, +}; + +int ceph_fscache_register(void) +{ + return fscache_register_netfs(&ceph_cache_netfs); +} + +void ceph_fscache_unregister(void) +{ + fscache_unregister_netfs(&ceph_cache_netfs); +} + +int ceph_fscache_register_fs(struct ceph_fs_client* fsc) +{ + fsc->fscache = fscache_acquire_cookie(ceph_cache_netfs.primary_index, + &ceph_fscache_fsid_object_def, + fsc); + + if (fsc->fscache == NULL) { + pr_err("Unable to resgister fsid: %p fscache cookie", fsc); + return 0; + } + + fsc->revalidate_wq = alloc_workqueue("ceph-revalidate", 0, 1); + if (fsc->revalidate_wq == NULL) + return -ENOMEM; + + return 0; +} + +static uint16_t ceph_fscache_inode_get_key(const void *cookie_netfs_data, + void *buffer, uint16_t maxbuf) +{ + const struct ceph_inode_info* ci = cookie_netfs_data; + uint16_t klen; + + /* use ceph virtual inode (id + snaphot) */ + klen = sizeof(ci->i_vino); + if (klen > maxbuf) + return 0; + + memcpy(buffer, &ci->i_vino, klen); + return klen; +} + +static uint16_t ceph_fscache_inode_get_aux(const void *cookie_netfs_data, + void *buffer, uint16_t bufmax) +{ + struct ceph_aux_inode aux; + const struct ceph_inode_info* ci = cookie_netfs_data; + const struct inode* inode = &ci->vfs_inode; + + memset(&aux, 0, sizeof(aux)); + aux.mtime = inode->i_mtime; + aux.size = inode->i_size; + + memcpy(buffer, &aux, sizeof(aux)); + + return sizeof(aux); +} + +static void ceph_fscache_inode_get_attr(const void *cookie_netfs_data, + uint64_t *size) +{ + const struct ceph_inode_info* ci = cookie_netfs_data; + const struct inode* inode = &ci->vfs_inode; + + *size = inode->i_size; +} + +static enum fscache_checkaux ceph_fscache_inode_check_aux( + void *cookie_netfs_data, const void *data, uint16_t dlen) +{ + struct ceph_aux_inode aux; + struct ceph_inode_info* ci = cookie_netfs_data; + struct inode* inode = &ci->vfs_inode; + + if (dlen != sizeof(aux)) + return FSCACHE_CHECKAUX_OBSOLETE; + + memset(&aux, 0, sizeof(aux)); + aux.mtime = inode->i_mtime; + aux.size = inode->i_size; + + if (memcmp(data, &aux, sizeof(aux)) != 0) + return FSCACHE_CHECKAUX_OBSOLETE; + + dout("ceph inode 0x%p cached okay", ci); + return FSCACHE_CHECKAUX_OKAY; +} + +static void ceph_fscache_inode_now_uncached(void* cookie_netfs_data) +{ + struct ceph_inode_info* ci = cookie_netfs_data; + struct pagevec pvec; + pgoff_t first; + int loop, nr_pages; + + pagevec_init(&pvec, 0); + first = 0; + + dout("ceph inode 0x%p now uncached", ci); + + while (1) { + nr_pages = pagevec_lookup(&pvec, ci->vfs_inode.i_mapping, first, + PAGEVEC_SIZE - pagevec_count(&pvec)); + + if (!nr_pages) + break; + + for (loop = 0; loop < nr_pages; loop++) + ClearPageFsCache(pvec.pages[loop]); + + first = pvec.pages[nr_pages - 1]->index + 1; + + pvec.nr = nr_pages; + pagevec_release(&pvec); + cond_resched(); + } +} + +static const struct fscache_cookie_def ceph_fscache_inode_object_def = { + .name = "CEPH.inode", + .type = FSCACHE_COOKIE_TYPE_DATAFILE, + .get_key = ceph_fscache_inode_get_key, + .get_attr = ceph_fscache_inode_get_attr, + .get_aux = ceph_fscache_inode_get_aux, + .check_aux = ceph_fscache_inode_check_aux, + .now_uncached = ceph_fscache_inode_now_uncached, +}; + +void ceph_fscache_register_inode_cookie(struct ceph_fs_client* fsc, + struct ceph_inode_info* ci) +{ + struct inode* inode = &ci->vfs_inode; + + /* No caching for filesystem */ + if (fsc->fscache == NULL) + return; + + /* Only cache for regular files that are read only */ + if ((ci->vfs_inode.i_mode & S_IFREG) == 0) + return; + + /* Avoid multiple racing open requests */ + mutex_lock(&inode->i_mutex); + + if (ci->fscache) + goto done; + + ci->fscache = fscache_acquire_cookie(fsc->fscache, + &ceph_fscache_inode_object_def, + ci); +done: + mutex_unlock(&inode->i_mutex); + +} + +void ceph_fscache_unregister_inode_cookie(struct ceph_inode_info* ci) +{ + struct fscache_cookie* cookie; + + if ((cookie = ci->fscache) == NULL) + return; + + ci->fscache = NULL; + + fscache_uncache_all_inode_pages(cookie, &ci->vfs_inode); + fscache_relinquish_cookie(cookie, 0); +} + +static void ceph_vfs_readpage_complete(struct page *page, void *data, int error) +{ + if (!error) + SetPageUptodate(page); +} + +static void ceph_vfs_readpage_complete_unlock(struct page *page, void *data, int error) +{ + if (!error) + SetPageUptodate(page); + + unlock_page(page); +} + +static inline int cache_valid(struct ceph_inode_info *ci) +{ + return ((ceph_caps_issued(ci) & CEPH_CAP_FILE_CACHE) && + (ci->i_fscache_gen == ci->i_rdcache_gen)); +} + + +/* Atempt to read from the fscache, + * + * This function is called from the readpage_nounlock context. DO NOT attempt to + * unlock the page here (or in the callback). + */ +int ceph_readpage_from_fscache(struct inode *inode, struct page *page) +{ + struct ceph_inode_info *ci = ceph_inode(inode); + int ret; + + if (!cache_valid(ci)) + return -ENOBUFS; + + ret = fscache_read_or_alloc_page(ci->fscache, page, + ceph_vfs_readpage_complete, NULL, + GFP_KERNEL); + + switch (ret) { + case 0: /* Page found */ + dout("page read submitted\n"); + return 0; + case -ENOBUFS: /* Pages were not found, and can't be */ + case -ENODATA: /* Pages were not found */ + dout("page/inode not in cache\n"); + return ret; + default: + dout("%s: unknown error ret = %i\n", __func__, ret); + return ret; + } +} + +int ceph_readpages_from_fscache(struct inode *inode, + struct address_space *mapping, + struct list_head *pages, + unsigned *nr_pages) +{ + struct ceph_inode_info *ci = ceph_inode(inode); + int ret; + + if (!cache_valid(ci)) + return -ENOBUFS; + + ret = fscache_read_or_alloc_pages(ci->fscache, mapping, pages, nr_pages, + ceph_vfs_readpage_complete_unlock, + NULL, mapping_gfp_mask(mapping)); + + switch (ret) { + case 0: /* All pages found */ + dout("all-page read submitted\n"); + return 0; + case -ENOBUFS: /* Some pages were not found, and can't be */ + case -ENODATA: /* some pages were not found */ + dout("page/inode not in cache\n"); + return ret; + default: + dout("%s: unknown error ret = %i\n", __func__, ret); + return ret; + } +} + +void ceph_readpage_to_fscache(struct inode *inode, struct page *page) +{ + struct ceph_inode_info *ci = ceph_inode(inode); + int ret; + + if (!PageFsCache(page)) + return; + + if (!cache_valid(ci)) + return; + + ret = fscache_write_page(ci->fscache, page, GFP_KERNEL); + if (ret) + fscache_uncache_page(ci->fscache, page); +} + +void ceph_invalidate_fscache_page(struct inode* inode, struct page *page) +{ + struct ceph_inode_info *ci = ceph_inode(inode); + + fscache_wait_on_page_write(ci->fscache, page); + fscache_uncache_page(ci->fscache, page); +} + +void ceph_fscache_unregister_fs(struct ceph_fs_client* fsc) +{ + if (fsc->revalidate_wq) + destroy_workqueue(fsc->revalidate_wq); + + fscache_relinquish_cookie(fsc->fscache, 0); + fsc->fscache = NULL; +} + +static void ceph_revalidate_work(struct work_struct *work) +{ + int issued; + u32 orig_gen; + struct ceph_inode_info *ci = container_of(work, struct ceph_inode_info, + i_revalidate_work); + struct inode *inode = &ci->vfs_inode; + + spin_lock(&ci->i_ceph_lock); + issued = __ceph_caps_issued(ci, NULL); + orig_gen = ci->i_rdcache_gen; + spin_unlock(&ci->i_ceph_lock); + + if (!(issued & CEPH_CAP_FILE_CACHE)) { + dout("revalidate_work lost cache before validation %p\n", + inode); + goto out; + } + + if (!fscache_check_consistency(ci->fscache)) + fscache_invalidate(ci->fscache); + + spin_lock(&ci->i_ceph_lock); + /* Update the new valid generation (backwards sanity check too) */ + if (orig_gen > ci->i_fscache_gen) { + ci->i_fscache_gen = orig_gen; + } + spin_unlock(&ci->i_ceph_lock); + +out: + iput(&ci->vfs_inode); +} + +void ceph_queue_revalidate(struct inode *inode) +{ + struct ceph_fs_client *fsc = ceph_sb_to_client(inode->i_sb); + struct ceph_inode_info *ci = ceph_inode(inode); + + if (fsc->revalidate_wq == NULL || ci->fscache == NULL) + return; + + ihold(inode); + + if (queue_work(ceph_sb_to_client(inode->i_sb)->revalidate_wq, + &ci->i_revalidate_work)) { + dout("ceph_queue_revalidate %p\n", inode); + } else { + dout("ceph_queue_revalidate %p failed\n)", inode); + iput(inode); + } +} + +void ceph_fscache_inode_init(struct ceph_inode_info *ci) +{ + ci->fscache = NULL; + /* The first load is verifed cookie open time */ + ci->i_fscache_gen = 1; + INIT_WORK(&ci->i_revalidate_work, ceph_revalidate_work); +} diff --git a/fs/ceph/cache.h b/fs/ceph/cache.h new file mode 100644 index 0000000..ba94940 --- /dev/null +++ b/fs/ceph/cache.h @@ -0,0 +1,159 @@ +/* + * Ceph cache definitions. + * + * Copyright (C) 2013 by Adfin Solutions, Inc. All Rights Reserved. + * Written by Milosz Tanski (milosz@adfin.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 + * as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to: + * Free Software Foundation + * 51 Franklin Street, Fifth Floor + * Boston, MA 02111-1301 USA + * + */ + +#ifndef _CEPH_CACHE_H +#define _CEPH_CACHE_H + +#ifdef CONFIG_CEPH_FSCACHE + +extern struct fscache_netfs ceph_cache_netfs; + +int ceph_fscache_register(void); +void ceph_fscache_unregister(void); + +int ceph_fscache_register_fs(struct ceph_fs_client* fsc); +void ceph_fscache_unregister_fs(struct ceph_fs_client* fsc); + +void ceph_fscache_inode_init(struct ceph_inode_info *ci); +void ceph_fscache_register_inode_cookie(struct ceph_fs_client* fsc, + struct ceph_inode_info* ci); +void ceph_fscache_unregister_inode_cookie(struct ceph_inode_info* ci); + +int ceph_readpage_from_fscache(struct inode *inode, struct page *page); +int ceph_readpages_from_fscache(struct inode *inode, + struct address_space *mapping, + struct list_head *pages, + unsigned *nr_pages); +void ceph_readpage_to_fscache(struct inode *inode, struct page *page); +void ceph_invalidate_fscache_page(struct inode* inode, struct page *page); +void ceph_queue_revalidate(struct inode *inode); + +static inline void ceph_fscache_invalidate(struct inode *inode) +{ + fscache_invalidate(ceph_inode(inode)->fscache); +} + +static inline void ceph_fscache_uncache_page(struct inode *inode, + struct page *page) +{ + struct ceph_inode_info *ci = ceph_inode(inode); + return fscache_uncache_page(ci->fscache, page); +} + +static inline int ceph_release_fscache_page(struct page *page, gfp_t gfp) +{ + struct inode* inode = page->mapping->host; + struct ceph_inode_info *ci = ceph_inode(inode); + return fscache_maybe_release_page(ci->fscache, page, gfp); +} + +static inline void ceph_fscache_readpages_cancel(struct inode *inode, + struct list_head *pages) +{ + struct ceph_inode_info *ci = ceph_inode(inode); + return fscache_readpages_cancel(ci->fscache, pages); +} + +#else + +static inline int ceph_fscache_register(void) +{ + return 0; +} + +static inline void ceph_fscache_unregister(void) +{ +} + +static inline int ceph_fscache_register_fs(struct ceph_fs_client* fsc) +{ + return 0; +} + +static inline void ceph_fscache_unregister_fs(struct ceph_fs_client* fsc) +{ +} + +static inline void ceph_fscache_inode_init(struct ceph_inode_info *ci) +{ +} + +static inline void ceph_fscache_register_inode_cookie(struct ceph_fs_client* parent_fsc, + struct ceph_inode_info* ci) +{ +} + +static inline void ceph_fscache_uncache_page(struct inode *inode, + struct page *pages) +{ +} + +static inline int ceph_readpage_from_fscache(struct inode* inode, + struct page *page) +{ + return -ENOBUFS; +} + +static inline int ceph_readpages_from_fscache(struct inode *inode, + struct address_space *mapping, + struct list_head *pages, + unsigned *nr_pages) +{ + return -ENOBUFS; +} + +static inline void ceph_readpage_to_fscache(struct inode *inode, + struct page *page) +{ +} + +static inline void ceph_fscache_invalidate(struct inode *inode) +{ +} + +static inline void ceph_invalidate_fscache_page(struct inode *inode, + struct page *page) +{ +} + +static inline void ceph_fscache_unregister_inode_cookie(struct ceph_inode_info* ci) +{ +} + +static inline int ceph_release_fscache_page(struct page *page, gfp_t gfp) +{ + return 1; +} + +static inline void ceph_fscache_readpages_cancel(struct inode *inode, + struct list_head *pages) +{ +} + +static inline void ceph_queue_revalidate(struct inode *inode) +{ +} + +#endif + +#endif diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c index 25442b4..13976c3 100644 --- a/fs/ceph/caps.c +++ b/fs/ceph/caps.c @@ -10,6 +10,7 @@ #include "super.h" #include "mds_client.h" +#include "cache.h" #include <linux/ceph/decode.h> #include <linux/ceph/messenger.h> @@ -479,8 +480,9 @@ static void __check_cap_issue(struct ceph_inode_info *ci, struct ceph_cap *cap, * i_rdcache_gen. */ if ((issued & (CEPH_CAP_FILE_CACHE|CEPH_CAP_FILE_LAZYIO)) && - (had & (CEPH_CAP_FILE_CACHE|CEPH_CAP_FILE_LAZYIO)) == 0) + (had & (CEPH_CAP_FILE_CACHE|CEPH_CAP_FILE_LAZYIO)) == 0) { ci->i_rdcache_gen++; + } /* * if we are newly issued FILE_SHARED, mark dir not complete; we @@ -2072,19 +2074,17 @@ static int try_get_cap_refs(struct ceph_inode_info *ci, int need, int want, /* finish pending truncate */ while (ci->i_truncate_pending) { spin_unlock(&ci->i_ceph_lock); - if (!(need & CEPH_CAP_FILE_WR)) - mutex_lock(&inode->i_mutex); __ceph_do_pending_vmtruncate(inode); - if (!(need & CEPH_CAP_FILE_WR)) - mutex_unlock(&inode->i_mutex); spin_lock(&ci->i_ceph_lock); } - if (need & CEPH_CAP_FILE_WR) { + have = __ceph_caps_issued(ci, &implemented); + + if (have & need & CEPH_CAP_FILE_WR) { if (endoff >= 0 && endoff > (loff_t)ci->i_max_size) { dout("get_cap_refs %p endoff %llu > maxsize %llu\n", inode, endoff, ci->i_max_size); - if (endoff > ci->i_wanted_max_size) { + if (endoff > ci->i_requested_max_size) { *check_max = 1; ret = 1; } @@ -2099,7 +2099,6 @@ static int try_get_cap_refs(struct ceph_inode_info *ci, int need, int want, goto out; } } - have = __ceph_caps_issued(ci, &implemented); if ((have & need) == need) { /* @@ -2141,14 +2140,17 @@ static void check_max_size(struct inode *inode, loff_t endoff) /* do we need to explicitly request a larger max_size? */ spin_lock(&ci->i_ceph_lock); - if ((endoff >= ci->i_max_size || - endoff > (inode->i_size << 1)) && - endoff > ci->i_wanted_max_size) { + if (endoff >= ci->i_max_size && endoff > ci->i_wanted_max_size) { dout("write %p at large endoff %llu, req max_size\n", inode, endoff); ci->i_wanted_max_size = endoff; - check = 1; } + /* duplicate ceph_check_caps()'s logic */ + if (ci->i_auth_cap && + (ci->i_auth_cap->issued & CEPH_CAP_FILE_WR) && + ci->i_wanted_max_size > ci->i_max_size && + ci->i_wanted_max_size > ci->i_requested_max_size) + check = 1; spin_unlock(&ci->i_ceph_lock); if (check) ceph_check_caps(ci, CHECK_CAPS_AUTHONLY, NULL); @@ -2334,6 +2336,38 @@ void ceph_put_wrbuffer_cap_refs(struct ceph_inode_info *ci, int nr, } /* + * Invalidate unlinked inode's aliases, so we can drop the inode ASAP. + */ +static void invalidate_aliases(struct inode *inode) +{ + struct dentry *dn, *prev = NULL; + + dout("invalidate_aliases inode %p\n", inode); + d_prune_aliases(inode); + /* + * For non-directory inode, d_find_alias() only returns + * connected dentry. After calling d_invalidate(), the + * dentry become disconnected. + * + * For directory inode, d_find_alias() can return + * disconnected dentry. But directory inode should have + * one alias at most. + */ + while ((dn = d_find_alias(inode))) { + if (dn == prev) { + dput(dn); + break; + } + d_invalidate(dn); + if (prev) + dput(prev); + prev = dn; + } + if (prev) + dput(prev); +} + +/* * Handle a cap GRANT message from the MDS. (Note that a GRANT may * actually be a revocation if it specifies a smaller cap set.) * @@ -2361,8 +2395,9 @@ static void handle_cap_grant(struct inode *inode, struct ceph_mds_caps *grant, int check_caps = 0; int wake = 0; int writeback = 0; - int revoked_rdcache = 0; int queue_invalidate = 0; + int deleted_inode = 0; + int queue_revalidate = 0; dout("handle_cap_grant inode %p cap %p mds%d seq %d %s\n", inode, cap, mds, seq, ceph_cap_string(newcaps)); @@ -2377,9 +2412,7 @@ static void handle_cap_grant(struct inode *inode, struct ceph_mds_caps *grant, if (((cap->issued & ~newcaps) & CEPH_CAP_FILE_CACHE) && (newcaps & CEPH_CAP_FILE_LAZYIO) == 0 && !ci->i_wrbuffer_ref) { - if (try_nonblocking_invalidate(inode) == 0) { - revoked_rdcache = 1; - } else { + if (try_nonblocking_invalidate(inode)) { /* there were locked pages.. invalidate later in a separate thread. */ if (ci->i_rdcache_revoking != ci->i_rdcache_gen) { @@ -2387,6 +2420,8 @@ static void handle_cap_grant(struct inode *inode, struct ceph_mds_caps *grant, ci->i_rdcache_revoking = ci->i_rdcache_gen; } } + + ceph_fscache_invalidate(inode); } /* side effects now are allowed */ @@ -2407,8 +2442,12 @@ static void handle_cap_grant(struct inode *inode, struct ceph_mds_caps *grant, from_kgid(&init_user_ns, inode->i_gid)); } - if ((issued & CEPH_CAP_LINK_EXCL) == 0) + if ((issued & CEPH_CAP_LINK_EXCL) == 0) { set_nlink(inode, le32_to_cpu(grant->nlink)); + if (inode->i_nlink == 0 && + (newcaps & (CEPH_CAP_LINK_SHARED | CEPH_CAP_LINK_EXCL))) + deleted_inode = 1; + } if ((issued & CEPH_CAP_XATTR_EXCL) == 0 && grant->xattr_len) { int len = le32_to_cpu(grant->xattr_len); @@ -2424,6 +2463,11 @@ static void handle_cap_grant(struct inode *inode, struct ceph_mds_caps *grant, } } + /* Do we need to revalidate our fscache cookie. Don't bother on the + * first cache cap as we already validate at cookie creation time. */ + if ((issued & CEPH_CAP_FILE_CACHE) && ci->i_rdcache_gen > 1) + queue_revalidate = 1; + /* size/ctime/mtime/atime? */ ceph_fill_file_size(inode, issued, le32_to_cpu(grant->truncate_seq), @@ -2508,6 +2552,7 @@ static void handle_cap_grant(struct inode *inode, struct ceph_mds_caps *grant, BUG_ON(cap->issued & ~cap->implemented); spin_unlock(&ci->i_ceph_lock); + if (writeback) /* * queue inode for writeback: we can't actually call @@ -2517,6 +2562,10 @@ static void handle_cap_grant(struct inode *inode, struct ceph_mds_caps *grant, ceph_queue_writeback(inode); if (queue_invalidate) ceph_queue_invalidate(inode); + if (deleted_inode) + invalidate_aliases(inode); + if (queue_revalidate) + ceph_queue_revalidate(inode); if (wake) wake_up_all(&ci->i_cap_wq); @@ -2673,8 +2722,10 @@ static void handle_cap_trunc(struct inode *inode, truncate_seq, truncate_size, size); spin_unlock(&ci->i_ceph_lock); - if (queue_trunc) + if (queue_trunc) { ceph_queue_vmtruncate(inode); + ceph_fscache_invalidate(inode); + } } /* diff --git a/fs/ceph/dir.c b/fs/ceph/dir.c index a40ceda..868b61d 100644 --- a/fs/ceph/dir.c +++ b/fs/ceph/dir.c @@ -793,6 +793,8 @@ static int ceph_link(struct dentry *old_dentry, struct inode *dir, req->r_locked_dir = dir; req->r_dentry_drop = CEPH_CAP_FILE_SHARED; req->r_dentry_unless = CEPH_CAP_FILE_EXCL; + /* release LINK_SHARED on source inode (mds will lock it) */ + req->r_old_inode_drop = CEPH_CAP_LINK_SHARED; err = ceph_mdsc_do_request(mdsc, dir, req); if (err) { d_drop(dentry); diff --git a/fs/ceph/file.c b/fs/ceph/file.c index 2ddf061..3de8982 100644 --- a/fs/ceph/file.c +++ b/fs/ceph/file.c @@ -8,9 +8,11 @@ #include <linux/namei.h> #include <linux/writeback.h> #include <linux/aio.h> +#include <linux/falloc.h> #include "super.h" #include "mds_client.h" +#include "cache.h" /* * Ceph file operations @@ -68,9 +70,23 @@ static int ceph_init_file(struct inode *inode, struct file *file, int fmode) { struct ceph_file_info *cf; int ret = 0; + struct ceph_inode_info *ci = ceph_inode(inode); + struct ceph_fs_client *fsc = ceph_sb_to_client(inode->i_sb); + struct ceph_mds_client *mdsc = fsc->mdsc; switch (inode->i_mode & S_IFMT) { case S_IFREG: + /* First file open request creates the cookie, we want to keep + * this cookie around for the filetime of the inode as not to + * have to worry about fscache register / revoke / operation + * races. + * + * Also, if we know the operation is going to invalidate data + * (non readonly) just nuke the cache right away. + */ + ceph_fscache_register_inode_cookie(mdsc->fsc, ci); + if ((fmode & CEPH_FILE_MODE_WR)) + ceph_fscache_invalidate(inode); case S_IFDIR: dout("init_file %p %p 0%o (regular)\n", inode, file, inode->i_mode); @@ -181,6 +197,7 @@ int ceph_open(struct inode *inode, struct file *file) spin_unlock(&ci->i_ceph_lock); return ceph_init_file(inode, file, fmode); } + spin_unlock(&ci->i_ceph_lock); dout("open fmode %d wants %s\n", fmode, ceph_cap_string(wanted)); @@ -191,6 +208,7 @@ int ceph_open(struct inode *inode, struct file *file) } req->r_inode = inode; ihold(inode); + req->r_num_caps = 1; if (flags & (O_CREAT|O_TRUNC)) parent_inode = ceph_get_dentry_parent_inode(file->f_dentry); @@ -313,9 +331,9 @@ static int striped_read(struct inode *inode, { struct ceph_fs_client *fsc = ceph_inode_to_client(inode); struct ceph_inode_info *ci = ceph_inode(inode); - u64 pos, this_len; + u64 pos, this_len, left; int io_align, page_align; - int left, pages_left; + int pages_left; int read; struct page **page_pos; int ret; @@ -346,47 +364,40 @@ more: ret = 0; hit_stripe = this_len < left; was_short = ret >= 0 && ret < this_len; - dout("striped_read %llu~%u (read %u) got %d%s%s\n", pos, left, read, + dout("striped_read %llu~%llu (read %u) got %d%s%s\n", pos, left, read, ret, hit_stripe ? " HITSTRIPE" : "", was_short ? " SHORT" : ""); - if (ret > 0) { - int didpages = (page_align + ret) >> PAGE_CACHE_SHIFT; - - if (read < pos - off) { - dout(" zero gap %llu to %llu\n", off + read, pos); - ceph_zero_page_vector_range(page_align + read, - pos - off - read, pages); + if (ret >= 0) { + int didpages; + if (was_short && (pos + ret < inode->i_size)) { + u64 tmp = min(this_len - ret, + inode->i_size - pos - ret); + dout(" zero gap %llu to %llu\n", + pos + ret, pos + ret + tmp); + ceph_zero_page_vector_range(page_align + read + ret, + tmp, pages); + ret += tmp; } + + didpages = (page_align + ret) >> PAGE_CACHE_SHIFT; pos += ret; read = pos - off; left -= ret; page_pos += didpages; pages_left -= didpages; - /* hit stripe? */ - if (left && hit_stripe) + /* hit stripe and need continue*/ + if (left && hit_stripe && pos < inode->i_size) goto more; } - if (was_short) { + if (read > 0) { + ret = read; /* did we bounce off eof? */ if (pos + left > inode->i_size) *checkeof = 1; - - /* zero trailing bytes (inside i_size) */ - if (left > 0 && pos < inode->i_size) { - if (pos + left > inode->i_size) - left = inode->i_size - pos; - - dout("zero tail %d\n", left); - ceph_zero_page_vector_range(page_align + read, left, - pages); - read += left; - } } - if (ret >= 0) - ret = read; dout("striped_read returns %d\n", ret); return ret; } @@ -618,6 +629,8 @@ out: if (check_caps) ceph_check_caps(ceph_inode(inode), CHECK_CAPS_AUTHONLY, NULL); + } else if (ret != -EOLDSNAPC && written > 0) { + ret = written; } return ret; } @@ -659,7 +672,6 @@ again: if ((got & (CEPH_CAP_FILE_CACHE|CEPH_CAP_FILE_LAZYIO)) == 0 || (iocb->ki_filp->f_flags & O_DIRECT) || - (inode->i_sb->s_flags & MS_SYNCHRONOUS) || (fi->flags & CEPH_F_SYNC)) /* hmm, this isn't really async... */ ret = ceph_sync_read(filp, base, len, ppos, &checkeof); @@ -711,13 +723,11 @@ static ssize_t ceph_aio_write(struct kiocb *iocb, const struct iovec *iov, &ceph_sb_to_client(inode->i_sb)->client->osdc; ssize_t count, written = 0; int err, want, got; - bool hold_mutex; if (ceph_snap(inode) != CEPH_NOSNAP) return -EROFS; mutex_lock(&inode->i_mutex); - hold_mutex = true; err = generic_segment_checks(iov, &nr_segs, &count, VERIFY_READ); if (err) @@ -763,18 +773,31 @@ retry_snap: if ((got & (CEPH_CAP_FILE_BUFFER|CEPH_CAP_FILE_LAZYIO)) == 0 || (iocb->ki_filp->f_flags & O_DIRECT) || - (inode->i_sb->s_flags & MS_SYNCHRONOUS) || (fi->flags & CEPH_F_SYNC)) { mutex_unlock(&inode->i_mutex); written = ceph_sync_write(file, iov->iov_base, count, pos, &iocb->ki_pos); + if (written == -EOLDSNAPC) { + dout("aio_write %p %llx.%llx %llu~%u" + "got EOLDSNAPC, retrying\n", + inode, ceph_vinop(inode), + pos, (unsigned)iov->iov_len); + mutex_lock(&inode->i_mutex); + goto retry_snap; + } } else { + /* + * No need to acquire the i_truncate_mutex. Because + * the MDS revokes Fwb caps before sending truncate + * message to us. We can't get Fwb cap while there + * are pending vmtruncate. So write and vmtruncate + * can not run at the same time + */ written = generic_file_buffered_write(iocb, iov, nr_segs, pos, &iocb->ki_pos, count, 0); mutex_unlock(&inode->i_mutex); } - hold_mutex = false; if (written >= 0) { int dirty; @@ -798,18 +821,12 @@ retry_snap: written = err; } - if (written == -EOLDSNAPC) { - dout("aio_write %p %llx.%llx %llu~%u got EOLDSNAPC, retrying\n", - inode, ceph_vinop(inode), pos, (unsigned)iov->iov_len); - mutex_lock(&inode->i_mutex); - hold_mutex = true; - goto retry_snap; - } + goto out_unlocked; + out: - if (hold_mutex) - mutex_unlock(&inode->i_mutex); + mutex_unlock(&inode->i_mutex); +out_unlocked: current->backing_dev_info = NULL; - return written ? written : err; } @@ -822,7 +839,6 @@ static loff_t ceph_llseek(struct file *file, loff_t offset, int whence) int ret; mutex_lock(&inode->i_mutex); - __ceph_do_pending_vmtruncate(inode); if (whence == SEEK_END || whence == SEEK_DATA || whence == SEEK_HOLE) { ret = ceph_do_getattr(inode, CEPH_STAT_CAP_SIZE); @@ -871,6 +887,204 @@ out: return offset; } +static inline void ceph_zero_partial_page( + struct inode *inode, loff_t offset, unsigned size) +{ + struct page *page; + pgoff_t index = offset >> PAGE_CACHE_SHIFT; + + page = find_lock_page(inode->i_mapping, index); + if (page) { + wait_on_page_writeback(page); + zero_user(page, offset & (PAGE_CACHE_SIZE - 1), size); + unlock_page(page); + page_cache_release(page); + } +} + +static void ceph_zero_pagecache_range(struct inode *inode, loff_t offset, + loff_t length) +{ + loff_t nearly = round_up(offset, PAGE_CACHE_SIZE); + if (offset < nearly) { + loff_t size = nearly - offset; + if (length < size) + size = length; + ceph_zero_partial_page(inode, offset, size); + offset += size; + length -= size; + } + if (length >= PAGE_CACHE_SIZE) { + loff_t size = round_down(length, PAGE_CACHE_SIZE); + truncate_pagecache_range(inode, offset, offset + size - 1); + offset += size; + length -= size; + } + if (length) + ceph_zero_partial_page(inode, offset, length); +} + +static int ceph_zero_partial_object(struct inode *inode, + loff_t offset, loff_t *length) +{ + struct ceph_inode_info *ci = ceph_inode(inode); + struct ceph_fs_client *fsc = ceph_inode_to_client(inode); + struct ceph_osd_request *req; + int ret = 0; + loff_t zero = 0; + int op; + + if (!length) { + op = offset ? CEPH_OSD_OP_DELETE : CEPH_OSD_OP_TRUNCATE; + length = &zero; + } else { + op = CEPH_OSD_OP_ZERO; + } + + req = ceph_osdc_new_request(&fsc->client->osdc, &ci->i_layout, + ceph_vino(inode), + offset, length, + 1, op, + CEPH_OSD_FLAG_WRITE | + CEPH_OSD_FLAG_ONDISK, + NULL, 0, 0, false); + if (IS_ERR(req)) { + ret = PTR_ERR(req); + goto out; + } + + ceph_osdc_build_request(req, offset, NULL, ceph_vino(inode).snap, + &inode->i_mtime); + + ret = ceph_osdc_start_request(&fsc->client->osdc, req, false); + if (!ret) { + ret = ceph_osdc_wait_request(&fsc->client->osdc, req); + if (ret == -ENOENT) + ret = 0; + } + ceph_osdc_put_request(req); + +out: + return ret; +} + +static int ceph_zero_objects(struct inode *inode, loff_t offset, loff_t length) +{ + int ret = 0; + struct ceph_inode_info *ci = ceph_inode(inode); + s32 stripe_unit = ceph_file_layout_su(ci->i_layout); + s32 stripe_count = ceph_file_layout_stripe_count(ci->i_layout); + s32 object_size = ceph_file_layout_object_size(ci->i_layout); + u64 object_set_size = object_size * stripe_count; + u64 nearly, t; + + /* round offset up to next period boundary */ + nearly = offset + object_set_size - 1; + t = nearly; + nearly -= do_div(t, object_set_size); + + while (length && offset < nearly) { + loff_t size = length; + ret = ceph_zero_partial_object(inode, offset, &size); + if (ret < 0) + return ret; + offset += size; + length -= size; + } + while (length >= object_set_size) { + int i; + loff_t pos = offset; + for (i = 0; i < stripe_count; ++i) { + ret = ceph_zero_partial_object(inode, pos, NULL); + if (ret < 0) + return ret; + pos += stripe_unit; + } + offset += object_set_size; + length -= object_set_size; + } + while (length) { + loff_t size = length; + ret = ceph_zero_partial_object(inode, offset, &size); + if (ret < 0) + return ret; + offset += size; + length -= size; + } + return ret; +} + +static long ceph_fallocate(struct file *file, int mode, + loff_t offset, loff_t length) +{ + struct ceph_file_info *fi = file->private_data; + struct inode *inode = file->f_dentry->d_inode; + struct ceph_inode_info *ci = ceph_inode(inode); + struct ceph_osd_client *osdc = + &ceph_inode_to_client(inode)->client->osdc; + int want, got = 0; + int dirty; + int ret = 0; + loff_t endoff = 0; + loff_t size; + + if (!S_ISREG(inode->i_mode)) + return -EOPNOTSUPP; + + if (IS_SWAPFILE(inode)) + return -ETXTBSY; + + mutex_lock(&inode->i_mutex); + + if (ceph_snap(inode) != CEPH_NOSNAP) { + ret = -EROFS; + goto unlock; + } + + if (ceph_osdmap_flag(osdc->osdmap, CEPH_OSDMAP_FULL) && + !(mode & FALLOC_FL_PUNCH_HOLE)) { + ret = -ENOSPC; + goto unlock; + } + + size = i_size_read(inode); + if (!(mode & FALLOC_FL_KEEP_SIZE)) + endoff = offset + length; + + if (fi->fmode & CEPH_FILE_MODE_LAZY) + want = CEPH_CAP_FILE_BUFFER | CEPH_CAP_FILE_LAZYIO; + else + want = CEPH_CAP_FILE_BUFFER; + + ret = ceph_get_caps(ci, CEPH_CAP_FILE_WR, want, &got, endoff); + if (ret < 0) + goto unlock; + + if (mode & FALLOC_FL_PUNCH_HOLE) { + if (offset < size) + ceph_zero_pagecache_range(inode, offset, length); + ret = ceph_zero_objects(inode, offset, length); + } else if (endoff > size) { + truncate_pagecache_range(inode, size, -1); + if (ceph_inode_set_size(inode, endoff)) + ceph_check_caps(ceph_inode(inode), + CHECK_CAPS_AUTHONLY, NULL); + } + + if (!ret) { + spin_lock(&ci->i_ceph_lock); + dirty = __ceph_mark_dirty_caps(ci, CEPH_CAP_FILE_WR); + spin_unlock(&ci->i_ceph_lock); + if (dirty) + __mark_inode_dirty(inode, dirty); + } + + ceph_put_cap_refs(ci, got); +unlock: + mutex_unlock(&inode->i_mutex); + return ret; +} + const struct file_operations ceph_file_fops = { .open = ceph_open, .release = ceph_release, @@ -887,5 +1101,6 @@ const struct file_operations ceph_file_fops = { .splice_write = generic_file_splice_write, .unlocked_ioctl = ceph_ioctl, .compat_ioctl = ceph_ioctl, + .fallocate = ceph_fallocate, }; diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c index f3a2abf..8549a48 100644 --- a/fs/ceph/inode.c +++ b/fs/ceph/inode.c @@ -12,6 +12,7 @@ #include "super.h" #include "mds_client.h" +#include "cache.h" #include <linux/ceph/decode.h> /* @@ -344,6 +345,7 @@ struct inode *ceph_alloc_inode(struct super_block *sb) for (i = 0; i < CEPH_FILE_MODE_NUM; i++) ci->i_nr_by_mode[i] = 0; + mutex_init(&ci->i_truncate_mutex); ci->i_truncate_seq = 0; ci->i_truncate_size = 0; ci->i_truncate_pending = 0; @@ -377,6 +379,8 @@ struct inode *ceph_alloc_inode(struct super_block *sb) INIT_WORK(&ci->i_vmtruncate_work, ceph_vmtruncate_work); + ceph_fscache_inode_init(ci); + return &ci->vfs_inode; } @@ -396,6 +400,8 @@ void ceph_destroy_inode(struct inode *inode) dout("destroy_inode %p ino %llx.%llx\n", inode, ceph_vinop(inode)); + ceph_fscache_unregister_inode_cookie(ci); + ceph_queue_caps_release(inode); /* @@ -430,7 +436,6 @@ void ceph_destroy_inode(struct inode *inode) call_rcu(&inode->i_rcu, ceph_i_callback); } - /* * Helpers to fill in size, ctime, mtime, and atime. We have to be * careful because either the client or MDS may have more up to date @@ -455,16 +460,20 @@ int ceph_fill_file_size(struct inode *inode, int issued, dout("truncate_seq %u -> %u\n", ci->i_truncate_seq, truncate_seq); ci->i_truncate_seq = truncate_seq; + + /* the MDS should have revoked these caps */ + WARN_ON_ONCE(issued & (CEPH_CAP_FILE_EXCL | + CEPH_CAP_FILE_RD | + CEPH_CAP_FILE_WR | + CEPH_CAP_FILE_LAZYIO)); /* * If we hold relevant caps, or in the case where we're * not the only client referencing this file and we * don't hold those caps, then we need to check whether * the file is either opened or mmaped */ - if ((issued & (CEPH_CAP_FILE_CACHE|CEPH_CAP_FILE_RD| - CEPH_CAP_FILE_WR|CEPH_CAP_FILE_BUFFER| - CEPH_CAP_FILE_EXCL| - CEPH_CAP_FILE_LAZYIO)) || + if ((issued & (CEPH_CAP_FILE_CACHE| + CEPH_CAP_FILE_BUFFER)) || mapping_mapped(inode->i_mapping) || __ceph_caps_file_wanted(ci)) { ci->i_truncate_pending++; @@ -478,6 +487,10 @@ int ceph_fill_file_size(struct inode *inode, int issued, truncate_size); ci->i_truncate_size = truncate_size; } + + if (queue_trunc) + ceph_fscache_invalidate(inode); + return queue_trunc; } @@ -1066,7 +1079,7 @@ int ceph_fill_trace(struct super_block *sb, struct ceph_mds_request *req, * complete. */ ceph_set_dentry_offset(req->r_old_dentry); - dout("dn %p gets new offset %lld\n", req->r_old_dentry, + dout("dn %p gets new offset %lld\n", req->r_old_dentry, ceph_dentry(req->r_old_dentry)->offset); dn = req->r_old_dentry; /* use old_dentry */ @@ -1419,18 +1432,20 @@ static void ceph_invalidate_work(struct work_struct *work) u32 orig_gen; int check = 0; + mutex_lock(&ci->i_truncate_mutex); spin_lock(&ci->i_ceph_lock); dout("invalidate_pages %p gen %d revoking %d\n", inode, ci->i_rdcache_gen, ci->i_rdcache_revoking); if (ci->i_rdcache_revoking != ci->i_rdcache_gen) { /* nevermind! */ spin_unlock(&ci->i_ceph_lock); + mutex_unlock(&ci->i_truncate_mutex); goto out; } orig_gen = ci->i_rdcache_gen; spin_unlock(&ci->i_ceph_lock); - truncate_inode_pages(&inode->i_data, 0); + truncate_inode_pages(inode->i_mapping, 0); spin_lock(&ci->i_ceph_lock); if (orig_gen == ci->i_rdcache_gen && @@ -1445,6 +1460,7 @@ static void ceph_invalidate_work(struct work_struct *work) ci->i_rdcache_revoking); } spin_unlock(&ci->i_ceph_lock); + mutex_unlock(&ci->i_truncate_mutex); if (check) ceph_check_caps(ci, 0, NULL); @@ -1465,9 +1481,7 @@ static void ceph_vmtruncate_work(struct work_struct *work) struct inode *inode = &ci->vfs_inode; dout("vmtruncate_work %p\n", inode); - mutex_lock(&inode->i_mutex); __ceph_do_pending_vmtruncate(inode); - mutex_unlock(&inode->i_mutex); iput(inode); } @@ -1480,6 +1494,7 @@ void ceph_queue_vmtruncate(struct inode *inode) struct ceph_inode_info *ci = ceph_inode(inode); ihold(inode); + if (queue_work(ceph_sb_to_client(inode->i_sb)->trunc_wq, &ci->i_vmtruncate_work)) { dout("ceph_queue_vmtruncate %p\n", inode); @@ -1500,11 +1515,13 @@ void __ceph_do_pending_vmtruncate(struct inode *inode) u64 to; int wrbuffer_refs, finish = 0; + mutex_lock(&ci->i_truncate_mutex); retry: spin_lock(&ci->i_ceph_lock); if (ci->i_truncate_pending == 0) { dout("__do_pending_vmtruncate %p none pending\n", inode); spin_unlock(&ci->i_ceph_lock); + mutex_unlock(&ci->i_truncate_mutex); return; } @@ -1521,6 +1538,9 @@ retry: goto retry; } + /* there should be no reader or writer */ + WARN_ON_ONCE(ci->i_rd_ref || ci->i_wr_ref); + to = ci->i_truncate_size; wrbuffer_refs = ci->i_wrbuffer_ref; dout("__do_pending_vmtruncate %p (%d) to %lld\n", inode, @@ -1538,13 +1558,14 @@ retry: if (!finish) goto retry; + mutex_unlock(&ci->i_truncate_mutex); + if (wrbuffer_refs == 0) ceph_check_caps(ci, CHECK_CAPS_AUTHONLY, NULL); wake_up_all(&ci->i_cap_wq); } - /* * symlinks */ @@ -1586,8 +1607,6 @@ int ceph_setattr(struct dentry *dentry, struct iattr *attr) if (ceph_snap(inode) != CEPH_NOSNAP) return -EROFS; - __ceph_do_pending_vmtruncate(inode); - err = inode_change_ok(inode, attr); if (err != 0) return err; @@ -1768,7 +1787,8 @@ int ceph_setattr(struct dentry *dentry, struct iattr *attr) ceph_cap_string(dirtied), mask); ceph_mdsc_put_request(req); - __ceph_do_pending_vmtruncate(inode); + if (mask & CEPH_SETATTR_SIZE) + __ceph_do_pending_vmtruncate(inode); return err; out: spin_unlock(&ci->i_ceph_lock); diff --git a/fs/ceph/ioctl.c b/fs/ceph/ioctl.c index e0b4ef3..669622f 100644 --- a/fs/ceph/ioctl.c +++ b/fs/ceph/ioctl.c @@ -196,8 +196,10 @@ static long ceph_ioctl_get_dataloc(struct file *file, void __user *arg) r = ceph_calc_file_object_mapping(&ci->i_layout, dl.file_offset, len, &dl.object_no, &dl.object_offset, &olen); - if (r < 0) + if (r < 0) { + up_read(&osdc->map_sem); return -EIO; + } dl.file_offset -= dl.object_offset; dl.object_size = ceph_file_layout_object_size(ci->i_layout); dl.block_size = ceph_file_layout_su(ci->i_layout); @@ -209,8 +211,12 @@ static long ceph_ioctl_get_dataloc(struct file *file, void __user *arg) snprintf(dl.object_name, sizeof(dl.object_name), "%llx.%08llx", ceph_ino(inode), dl.object_no); - ceph_calc_ceph_pg(&pgid, dl.object_name, osdc->osdmap, - ceph_file_layout_pg_pool(ci->i_layout)); + r = ceph_calc_ceph_pg(&pgid, dl.object_name, osdc->osdmap, + ceph_file_layout_pg_pool(ci->i_layout)); + if (r < 0) { + up_read(&osdc->map_sem); + return r; + } dl.osd = ceph_calc_pg_primary(osdc->osdmap, pgid); if (dl.osd >= 0) { diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c index 187bf21..b7bda5d 100644 --- a/fs/ceph/mds_client.c +++ b/fs/ceph/mds_client.c @@ -414,6 +414,9 @@ static struct ceph_mds_session *register_session(struct ceph_mds_client *mdsc, { struct ceph_mds_session *s; + if (mds >= mdsc->mdsmap->m_max_mds) + return ERR_PTR(-EINVAL); + s = kzalloc(sizeof(*s), GFP_NOFS); if (!s) return ERR_PTR(-ENOMEM); @@ -1028,6 +1031,37 @@ static void remove_session_caps(struct ceph_mds_session *session) { dout("remove_session_caps on %p\n", session); iterate_session_caps(session, remove_session_caps_cb, NULL); + + spin_lock(&session->s_cap_lock); + if (session->s_nr_caps > 0) { + struct super_block *sb = session->s_mdsc->fsc->sb; + struct inode *inode; + struct ceph_cap *cap, *prev = NULL; + struct ceph_vino vino; + /* + * iterate_session_caps() skips inodes that are being + * deleted, we need to wait until deletions are complete. + * __wait_on_freeing_inode() is designed for the job, + * but it is not exported, so use lookup inode function + * to access it. + */ + while (!list_empty(&session->s_caps)) { + cap = list_entry(session->s_caps.next, + struct ceph_cap, session_caps); + if (cap == prev) + break; + prev = cap; + vino = cap->ci->i_vino; + spin_unlock(&session->s_cap_lock); + + inode = ceph_find_inode(sb, vino); + iput(inode); + + spin_lock(&session->s_cap_lock); + } + } + spin_unlock(&session->s_cap_lock); + BUG_ON(session->s_nr_caps > 0); BUG_ON(!list_empty(&session->s_cap_flushing)); cleanup_cap_releases(session); diff --git a/fs/ceph/super.c b/fs/ceph/super.c index 6627b26..6a0951e 100644 --- a/fs/ceph/super.c +++ b/fs/ceph/super.c @@ -17,6 +17,7 @@ #include "super.h" #include "mds_client.h" +#include "cache.h" #include <linux/ceph/ceph_features.h> #include <linux/ceph/decode.h> @@ -142,6 +143,8 @@ enum { Opt_nodcache, Opt_ino32, Opt_noino32, + Opt_fscache, + Opt_nofscache }; static match_table_t fsopt_tokens = { @@ -167,6 +170,8 @@ static match_table_t fsopt_tokens = { {Opt_nodcache, "nodcache"}, {Opt_ino32, "ino32"}, {Opt_noino32, "noino32"}, + {Opt_fscache, "fsc"}, + {Opt_nofscache, "nofsc"}, {-1, NULL} }; @@ -260,6 +265,12 @@ static int parse_fsopt_token(char *c, void *private) case Opt_noino32: fsopt->flags &= ~CEPH_MOUNT_OPT_INO32; break; + case Opt_fscache: + fsopt->flags |= CEPH_MOUNT_OPT_FSCACHE; + break; + case Opt_nofscache: + fsopt->flags &= ~CEPH_MOUNT_OPT_FSCACHE; + break; default: BUG_ON(token); } @@ -422,6 +433,10 @@ static int ceph_show_options(struct seq_file *m, struct dentry *root) seq_puts(m, ",dcache"); else seq_puts(m, ",nodcache"); + if (fsopt->flags & CEPH_MOUNT_OPT_FSCACHE) + seq_puts(m, ",fsc"); + else + seq_puts(m, ",nofsc"); if (fsopt->wsize) seq_printf(m, ",wsize=%d", fsopt->wsize); @@ -530,11 +545,18 @@ static struct ceph_fs_client *create_fs_client(struct ceph_mount_options *fsopt, if (!fsc->wb_pagevec_pool) goto fail_trunc_wq; + /* setup fscache */ + if ((fsopt->flags & CEPH_MOUNT_OPT_FSCACHE) && + (ceph_fscache_register_fs(fsc) != 0)) + goto fail_fscache; + /* caps */ fsc->min_caps = fsopt->max_readdir; return fsc; +fail_fscache: + ceph_fscache_unregister_fs(fsc); fail_trunc_wq: destroy_workqueue(fsc->trunc_wq); fail_pg_inv_wq: @@ -554,6 +576,8 @@ static void destroy_fs_client(struct ceph_fs_client *fsc) { dout("destroy_fs_client %p\n", fsc); + ceph_fscache_unregister_fs(fsc); + destroy_workqueue(fsc->wb_wq); destroy_workqueue(fsc->pg_inv_wq); destroy_workqueue(fsc->trunc_wq); @@ -588,6 +612,8 @@ static void ceph_inode_init_once(void *foo) static int __init init_caches(void) { + int error = -ENOMEM; + ceph_inode_cachep = kmem_cache_create("ceph_inode_info", sizeof(struct ceph_inode_info), __alignof__(struct ceph_inode_info), @@ -611,15 +637,17 @@ static int __init init_caches(void) if (ceph_file_cachep == NULL) goto bad_file; - return 0; + if ((error = ceph_fscache_register())) + goto bad_file; + return 0; bad_file: kmem_cache_destroy(ceph_dentry_cachep); bad_dentry: kmem_cache_destroy(ceph_cap_cachep); bad_cap: kmem_cache_destroy(ceph_inode_cachep); - return -ENOMEM; + return error; } static void destroy_caches(void) @@ -629,10 +657,13 @@ static void destroy_caches(void) * destroy cache. */ rcu_barrier(); + kmem_cache_destroy(ceph_inode_cachep); kmem_cache_destroy(ceph_cap_cachep); kmem_cache_destroy(ceph_dentry_cachep); kmem_cache_destroy(ceph_file_cachep); + + ceph_fscache_unregister(); } diff --git a/fs/ceph/super.h b/fs/ceph/super.h index cbded57..6014b0a 100644 --- a/fs/ceph/super.h +++ b/fs/ceph/super.h @@ -16,6 +16,10 @@ #include <linux/ceph/libceph.h> +#ifdef CONFIG_CEPH_FSCACHE +#include <linux/fscache.h> +#endif + /* f_type in struct statfs */ #define CEPH_SUPER_MAGIC 0x00c36400 @@ -29,6 +33,7 @@ #define CEPH_MOUNT_OPT_NOASYNCREADDIR (1<<7) /* no dcache readdir */ #define CEPH_MOUNT_OPT_INO32 (1<<8) /* 32 bit inos */ #define CEPH_MOUNT_OPT_DCACHE (1<<9) /* use dcache for readdir etc */ +#define CEPH_MOUNT_OPT_FSCACHE (1<<10) /* use fscache */ #define CEPH_MOUNT_OPT_DEFAULT (CEPH_MOUNT_OPT_RBYTES) @@ -90,6 +95,11 @@ struct ceph_fs_client { struct dentry *debugfs_bdi; struct dentry *debugfs_mdsc, *debugfs_mdsmap; #endif + +#ifdef CONFIG_CEPH_FSCACHE + struct fscache_cookie *fscache; + struct workqueue_struct *revalidate_wq; +#endif }; @@ -288,6 +298,7 @@ struct ceph_inode_info { int i_nr_by_mode[CEPH_FILE_MODE_NUM]; /* open file counts */ + struct mutex i_truncate_mutex; u32 i_truncate_seq; /* last truncate to smaller size */ u64 i_truncate_size; /* and the size we last truncated down to */ int i_truncate_pending; /* still need to call vmtruncate */ @@ -319,6 +330,12 @@ struct ceph_inode_info { struct work_struct i_vmtruncate_work; +#ifdef CONFIG_CEPH_FSCACHE + struct fscache_cookie *fscache; + u32 i_fscache_gen; /* sequence, for delayed fscache validate */ + struct work_struct i_revalidate_work; +#endif + struct inode vfs_inode; /* at end */ }; diff --git a/fs/coredump.c b/fs/coredump.c index 72f816d..9bdeca1 100644 --- a/fs/coredump.c +++ b/fs/coredump.c @@ -190,6 +190,11 @@ static int format_corename(struct core_name *cn, struct coredump_params *cprm) err = cn_printf(cn, "%d", task_tgid_vnr(current)); break; + /* global pid */ + case 'P': + err = cn_printf(cn, "%d", + task_tgid_nr(current)); + break; /* uid */ case 'u': err = cn_printf(cn, "%d", cred->uid); diff --git a/fs/dcache.c b/fs/dcache.c index 5aa53bc..4d9df3c 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -88,6 +88,35 @@ EXPORT_SYMBOL(rename_lock); static struct kmem_cache *dentry_cache __read_mostly; +/** + * read_seqbegin_or_lock - begin a sequence number check or locking block + * lock: sequence lock + * seq : sequence number to be checked + * + * First try it once optimistically without taking the lock. If that fails, + * take the lock. The sequence number is also used as a marker for deciding + * whether to be a reader (even) or writer (odd). + * N.B. seq must be initialized to an even number to begin with. + */ +static inline void read_seqbegin_or_lock(seqlock_t *lock, int *seq) +{ + if (!(*seq & 1)) /* Even */ + *seq = read_seqbegin(lock); + else /* Odd */ + write_seqlock(lock); +} + +static inline int need_seqretry(seqlock_t *lock, int seq) +{ + return !(seq & 1) && read_seqretry(lock, seq); +} + +static inline void done_seqretry(seqlock_t *lock, int seq) +{ + if (seq & 1) + write_sequnlock(lock); +} + /* * This is the single most critical data structure when it comes * to the dcache: the hashtable for lookups. Somebody should try @@ -229,7 +258,7 @@ static void __d_free(struct rcu_head *head) */ static void d_free(struct dentry *dentry) { - BUG_ON(dentry->d_lockref.count); + BUG_ON((int)dentry->d_lockref.count > 0); this_cpu_dec(nr_dentry); if (dentry->d_op && dentry->d_op->d_release) dentry->d_op->d_release(dentry); @@ -308,8 +337,9 @@ static void dentry_unlink_inode(struct dentry * dentry) */ static void dentry_lru_add(struct dentry *dentry) { - if (list_empty(&dentry->d_lru)) { + if (unlikely(!(dentry->d_flags & DCACHE_LRU_LIST))) { spin_lock(&dcache_lru_lock); + dentry->d_flags |= DCACHE_LRU_LIST; list_add(&dentry->d_lru, &dentry->d_sb->s_dentry_lru); dentry->d_sb->s_nr_dentry_unused++; dentry_stat.nr_unused++; @@ -320,7 +350,7 @@ static void dentry_lru_add(struct dentry *dentry) static void __dentry_lru_del(struct dentry *dentry) { list_del_init(&dentry->d_lru); - dentry->d_flags &= ~DCACHE_SHRINK_LIST; + dentry->d_flags &= ~(DCACHE_SHRINK_LIST | DCACHE_LRU_LIST); dentry->d_sb->s_nr_dentry_unused--; dentry_stat.nr_unused--; } @@ -341,6 +371,7 @@ static void dentry_lru_move_list(struct dentry *dentry, struct list_head *list) { spin_lock(&dcache_lru_lock); if (list_empty(&dentry->d_lru)) { + dentry->d_flags |= DCACHE_LRU_LIST; list_add_tail(&dentry->d_lru, list); dentry->d_sb->s_nr_dentry_unused++; dentry_stat.nr_unused++; @@ -443,7 +474,7 @@ EXPORT_SYMBOL(d_drop); * If ref is non-zero, then decrement the refcount too. * Returns dentry requiring refcount drop, or NULL if we're done. */ -static inline struct dentry *dentry_kill(struct dentry *dentry, int ref) +static inline struct dentry *dentry_kill(struct dentry *dentry) __releases(dentry->d_lock) { struct inode *inode; @@ -466,8 +497,11 @@ relock: goto relock; } - if (ref) - dentry->d_lockref.count--; + /* + * The dentry is now unrecoverably dead to the world. + */ + lockref_mark_dead(&dentry->d_lockref); + /* * inform the fs via d_prune that this dentry is about to be * unhashed and destroyed. @@ -509,24 +543,22 @@ relock: */ void dput(struct dentry *dentry) { - if (!dentry) + if (unlikely(!dentry)) return; repeat: - if (dentry->d_lockref.count == 1) - might_sleep(); if (lockref_put_or_lock(&dentry->d_lockref)) return; - if (dentry->d_flags & DCACHE_OP_DELETE) { + /* Unreachable? Get rid of it */ + if (unlikely(d_unhashed(dentry))) + goto kill_it; + + if (unlikely(dentry->d_flags & DCACHE_OP_DELETE)) { if (dentry->d_op->d_delete(dentry)) goto kill_it; } - /* Unreachable? Get rid of it */ - if (d_unhashed(dentry)) - goto kill_it; - dentry->d_flags |= DCACHE_REFERENCED; dentry_lru_add(dentry); @@ -535,7 +567,7 @@ repeat: return; kill_it: - dentry = dentry_kill(dentry, 1); + dentry = dentry_kill(dentry); if (dentry) goto repeat; } @@ -760,7 +792,7 @@ static void try_prune_one_dentry(struct dentry *dentry) { struct dentry *parent; - parent = dentry_kill(dentry, 0); + parent = dentry_kill(dentry); /* * If dentry_kill returns NULL, we have nothing more to do. * if it returns the same dentry, trylocks failed. In either @@ -781,7 +813,7 @@ static void try_prune_one_dentry(struct dentry *dentry) while (dentry) { if (lockref_put_or_lock(&dentry->d_lockref)) return; - dentry = dentry_kill(dentry, 1); + dentry = dentry_kill(dentry); } } @@ -1009,7 +1041,7 @@ void shrink_dcache_for_umount(struct super_block *sb) * the parenthood after dropping the lock and check * that the sequence number still matches. */ -static struct dentry *try_to_ascend(struct dentry *old, int locked, unsigned seq) +static struct dentry *try_to_ascend(struct dentry *old, unsigned seq) { struct dentry *new = old->d_parent; @@ -1023,7 +1055,7 @@ static struct dentry *try_to_ascend(struct dentry *old, int locked, unsigned seq */ if (new != old->d_parent || (old->d_flags & DCACHE_DENTRY_KILLED) || - (!locked && read_seqretry(&rename_lock, seq))) { + need_seqretry(&rename_lock, seq)) { spin_unlock(&new->d_lock); new = NULL; } @@ -1031,34 +1063,55 @@ static struct dentry *try_to_ascend(struct dentry *old, int locked, unsigned seq return new; } +/** + * enum d_walk_ret - action to talke during tree walk + * @D_WALK_CONTINUE: contrinue walk + * @D_WALK_QUIT: quit walk + * @D_WALK_NORETRY: quit when retry is needed + * @D_WALK_SKIP: skip this dentry and its children + */ +enum d_walk_ret { + D_WALK_CONTINUE, + D_WALK_QUIT, + D_WALK_NORETRY, + D_WALK_SKIP, +}; -/* - * Search for at least 1 mount point in the dentry's subdirs. - * We descend to the next level whenever the d_subdirs - * list is non-empty and continue searching. - */ - /** - * have_submounts - check for mounts over a dentry - * @parent: dentry to check. + * d_walk - walk the dentry tree + * @parent: start of walk + * @data: data passed to @enter() and @finish() + * @enter: callback when first entering the dentry + * @finish: callback when successfully finished the walk * - * Return true if the parent or its subdirectories contain - * a mount point + * The @enter() and @finish() callbacks are called with d_lock held. */ -int have_submounts(struct dentry *parent) +static void d_walk(struct dentry *parent, void *data, + enum d_walk_ret (*enter)(void *, struct dentry *), + void (*finish)(void *)) { struct dentry *this_parent; struct list_head *next; - unsigned seq; - int locked = 0; + unsigned seq = 0; + enum d_walk_ret ret; + bool retry = true; - seq = read_seqbegin(&rename_lock); again: + read_seqbegin_or_lock(&rename_lock, &seq); this_parent = parent; - - if (d_mountpoint(parent)) - goto positive; spin_lock(&this_parent->d_lock); + + ret = enter(data, this_parent); + switch (ret) { + case D_WALK_CONTINUE: + break; + case D_WALK_QUIT: + case D_WALK_SKIP: + goto out_unlock; + case D_WALK_NORETRY: + retry = false; + break; + } repeat: next = this_parent->d_subdirs.next; resume: @@ -1068,12 +1121,22 @@ resume: next = tmp->next; spin_lock_nested(&dentry->d_lock, DENTRY_D_LOCK_NESTED); - /* Have we found a mount point ? */ - if (d_mountpoint(dentry)) { + + ret = enter(data, dentry); + switch (ret) { + case D_WALK_CONTINUE: + break; + case D_WALK_QUIT: spin_unlock(&dentry->d_lock); - spin_unlock(&this_parent->d_lock); - goto positive; + goto out_unlock; + case D_WALK_NORETRY: + retry = false; + break; + case D_WALK_SKIP: + spin_unlock(&dentry->d_lock); + continue; } + if (!list_empty(&dentry->d_subdirs)) { spin_unlock(&this_parent->d_lock); spin_release(&dentry->d_lock.dep_map, 1, _RET_IP_); @@ -1088,35 +1151,99 @@ resume: */ if (this_parent != parent) { struct dentry *child = this_parent; - this_parent = try_to_ascend(this_parent, locked, seq); + this_parent = try_to_ascend(this_parent, seq); if (!this_parent) goto rename_retry; next = child->d_u.d_child.next; goto resume; } - spin_unlock(&this_parent->d_lock); - if (!locked && read_seqretry(&rename_lock, seq)) + if (need_seqretry(&rename_lock, seq)) { + spin_unlock(&this_parent->d_lock); goto rename_retry; - if (locked) - write_sequnlock(&rename_lock); - return 0; /* No mount points found in tree */ -positive: - if (!locked && read_seqretry(&rename_lock, seq)) - goto rename_retry; - if (locked) - write_sequnlock(&rename_lock); - return 1; + } + if (finish) + finish(data); + +out_unlock: + spin_unlock(&this_parent->d_lock); + done_seqretry(&rename_lock, seq); + return; rename_retry: - if (locked) - goto again; - locked = 1; - write_seqlock(&rename_lock); + if (!retry) + return; + seq = 1; goto again; } + +/* + * Search for at least 1 mount point in the dentry's subdirs. + * We descend to the next level whenever the d_subdirs + * list is non-empty and continue searching. + */ + +/** + * have_submounts - check for mounts over a dentry + * @parent: dentry to check. + * + * Return true if the parent or its subdirectories contain + * a mount point + */ + +static enum d_walk_ret check_mount(void *data, struct dentry *dentry) +{ + int *ret = data; + if (d_mountpoint(dentry)) { + *ret = 1; + return D_WALK_QUIT; + } + return D_WALK_CONTINUE; +} + +int have_submounts(struct dentry *parent) +{ + int ret = 0; + + d_walk(parent, &ret, check_mount, NULL); + + return ret; +} EXPORT_SYMBOL(have_submounts); /* + * Called by mount code to set a mountpoint and check if the mountpoint is + * reachable (e.g. NFS can unhash a directory dentry and then the complete + * subtree can become unreachable). + * + * Only one of check_submounts_and_drop() and d_set_mounted() must succeed. For + * this reason take rename_lock and d_lock on dentry and ancestors. + */ +int d_set_mounted(struct dentry *dentry) +{ + struct dentry *p; + int ret = -ENOENT; + write_seqlock(&rename_lock); + for (p = dentry->d_parent; !IS_ROOT(p); p = p->d_parent) { + /* Need exclusion wrt. check_submounts_and_drop() */ + spin_lock(&p->d_lock); + if (unlikely(d_unhashed(p))) { + spin_unlock(&p->d_lock); + goto out; + } + spin_unlock(&p->d_lock); + } + spin_lock(&dentry->d_lock); + if (!d_unlinked(dentry)) { + dentry->d_flags |= DCACHE_MOUNTED; + ret = 0; + } + spin_unlock(&dentry->d_lock); +out: + write_sequnlock(&rename_lock); + return ret; +} + +/* * Search the dentry child list of the specified parent, * and move any unused dentries to the end of the unused * list for prune_dcache(). We descend to the next level @@ -1130,93 +1257,46 @@ EXPORT_SYMBOL(have_submounts); * drop the lock and return early due to latency * constraints. */ -static int select_parent(struct dentry *parent, struct list_head *dispose) -{ - struct dentry *this_parent; - struct list_head *next; - unsigned seq; - int found = 0; - int locked = 0; - - seq = read_seqbegin(&rename_lock); -again: - this_parent = parent; - spin_lock(&this_parent->d_lock); -repeat: - next = this_parent->d_subdirs.next; -resume: - while (next != &this_parent->d_subdirs) { - struct list_head *tmp = next; - struct dentry *dentry = list_entry(tmp, struct dentry, d_u.d_child); - next = tmp->next; - spin_lock_nested(&dentry->d_lock, DENTRY_D_LOCK_NESTED); +struct select_data { + struct dentry *start; + struct list_head dispose; + int found; +}; - /* - * move only zero ref count dentries to the dispose list. - * - * Those which are presently on the shrink list, being processed - * by shrink_dentry_list(), shouldn't be moved. Otherwise the - * loop in shrink_dcache_parent() might not make any progress - * and loop forever. - */ - if (dentry->d_lockref.count) { - dentry_lru_del(dentry); - } else if (!(dentry->d_flags & DCACHE_SHRINK_LIST)) { - dentry_lru_move_list(dentry, dispose); - dentry->d_flags |= DCACHE_SHRINK_LIST; - found++; - } - /* - * We can return to the caller if we have found some (this - * ensures forward progress). We'll be coming back to find - * the rest. - */ - if (found && need_resched()) { - spin_unlock(&dentry->d_lock); - goto out; - } +static enum d_walk_ret select_collect(void *_data, struct dentry *dentry) +{ + struct select_data *data = _data; + enum d_walk_ret ret = D_WALK_CONTINUE; - /* - * Descend a level if the d_subdirs list is non-empty. - */ - if (!list_empty(&dentry->d_subdirs)) { - spin_unlock(&this_parent->d_lock); - spin_release(&dentry->d_lock.dep_map, 1, _RET_IP_); - this_parent = dentry; - spin_acquire(&this_parent->d_lock.dep_map, 0, 1, _RET_IP_); - goto repeat; - } + if (data->start == dentry) + goto out; - spin_unlock(&dentry->d_lock); - } /* - * All done at this level ... ascend and resume the search. + * move only zero ref count dentries to the dispose list. + * + * Those which are presently on the shrink list, being processed + * by shrink_dentry_list(), shouldn't be moved. Otherwise the + * loop in shrink_dcache_parent() might not make any progress + * and loop forever. */ - if (this_parent != parent) { - struct dentry *child = this_parent; - this_parent = try_to_ascend(this_parent, locked, seq); - if (!this_parent) - goto rename_retry; - next = child->d_u.d_child.next; - goto resume; + if (dentry->d_lockref.count) { + dentry_lru_del(dentry); + } else if (!(dentry->d_flags & DCACHE_SHRINK_LIST)) { + dentry_lru_move_list(dentry, &data->dispose); + dentry->d_flags |= DCACHE_SHRINK_LIST; + data->found++; + ret = D_WALK_NORETRY; } + /* + * We can return to the caller if we have found some (this + * ensures forward progress). We'll be coming back to find + * the rest. + */ + if (data->found && need_resched()) + ret = D_WALK_QUIT; out: - spin_unlock(&this_parent->d_lock); - if (!locked && read_seqretry(&rename_lock, seq)) - goto rename_retry; - if (locked) - write_sequnlock(&rename_lock); - return found; - -rename_retry: - if (found) - return found; - if (locked) - goto again; - locked = 1; - write_seqlock(&rename_lock); - goto again; + return ret; } /** @@ -1225,18 +1305,90 @@ rename_retry: * * Prune the dcache to remove unused children of the parent dentry. */ -void shrink_dcache_parent(struct dentry * parent) +void shrink_dcache_parent(struct dentry *parent) { - LIST_HEAD(dispose); - int found; + for (;;) { + struct select_data data; + + INIT_LIST_HEAD(&data.dispose); + data.start = parent; + data.found = 0; + + d_walk(parent, &data, select_collect, NULL); + if (!data.found) + break; - while ((found = select_parent(parent, &dispose)) != 0) { - shrink_dentry_list(&dispose); + shrink_dentry_list(&data.dispose); cond_resched(); } } EXPORT_SYMBOL(shrink_dcache_parent); +static enum d_walk_ret check_and_collect(void *_data, struct dentry *dentry) +{ + struct select_data *data = _data; + + if (d_mountpoint(dentry)) { + data->found = -EBUSY; + return D_WALK_QUIT; + } + + return select_collect(_data, dentry); +} + +static void check_and_drop(void *_data) +{ + struct select_data *data = _data; + + if (d_mountpoint(data->start)) + data->found = -EBUSY; + if (!data->found) + __d_drop(data->start); +} + +/** + * check_submounts_and_drop - prune dcache, check for submounts and drop + * + * All done as a single atomic operation relative to has_unlinked_ancestor(). + * Returns 0 if successfully unhashed @parent. If there were submounts then + * return -EBUSY. + * + * @dentry: dentry to prune and drop + */ +int check_submounts_and_drop(struct dentry *dentry) +{ + int ret = 0; + + /* Negative dentries can be dropped without further checks */ + if (!dentry->d_inode) { + d_drop(dentry); + goto out; + } + + for (;;) { + struct select_data data; + + INIT_LIST_HEAD(&data.dispose); + data.start = dentry; + data.found = 0; + + d_walk(dentry, &data, check_and_collect, check_and_drop); + ret = data.found; + + if (!list_empty(&data.dispose)) + shrink_dentry_list(&data.dispose); + + if (ret <= 0) + break; + + cond_resched(); + } + +out: + return ret; +} +EXPORT_SYMBOL(check_submounts_and_drop); + /** * __d_alloc - allocate a dcache entry * @sb: filesystem it will belong to @@ -2519,9 +2671,39 @@ static int prepend(char **buffer, int *buflen, const char *str, int namelen) return 0; } +/** + * prepend_name - prepend a pathname in front of current buffer pointer + * buffer: buffer pointer + * buflen: allocated length of the buffer + * name: name string and length qstr structure + * + * With RCU path tracing, it may race with d_move(). Use ACCESS_ONCE() to + * make sure that either the old or the new name pointer and length are + * fetched. However, there may be mismatch between length and pointer. + * The length cannot be trusted, we need to copy it byte-by-byte until + * the length is reached or a null byte is found. It also prepends "/" at + * the beginning of the name. The sequence number check at the caller will + * retry it again when a d_move() does happen. So any garbage in the buffer + * due to mismatched pointer and length will be discarded. + */ static int prepend_name(char **buffer, int *buflen, struct qstr *name) { - return prepend(buffer, buflen, name->name, name->len); + const char *dname = ACCESS_ONCE(name->name); + u32 dlen = ACCESS_ONCE(name->len); + char *p; + + if (*buflen < dlen + 1) + return -ENAMETOOLONG; + *buflen -= dlen + 1; + p = *buffer -= dlen + 1; + *p++ = '/'; + while (dlen--) { + char c = *dname++; + if (!c) + break; + *p++ = c; + } + return 0; } /** @@ -2531,7 +2713,14 @@ static int prepend_name(char **buffer, int *buflen, struct qstr *name) * @buffer: pointer to the end of the buffer * @buflen: pointer to buffer length * - * Caller holds the rename_lock. + * The function tries to write out the pathname without taking any lock other + * than the RCU read lock to make sure that dentries won't go away. It only + * checks the sequence number of the global rename_lock as any change in the + * dentry's d_seq will be preceded by changes in the rename_lock sequence + * number. If the sequence number had been change, it will restart the whole + * pathname back-tracing sequence again. It performs a total of 3 trials of + * lockless back-tracing sequences before falling back to take the + * rename_lock. */ static int prepend_path(const struct path *path, const struct path *root, @@ -2540,54 +2729,66 @@ static int prepend_path(const struct path *path, struct dentry *dentry = path->dentry; struct vfsmount *vfsmnt = path->mnt; struct mount *mnt = real_mount(vfsmnt); - bool slash = false; int error = 0; + unsigned seq = 0; + char *bptr; + int blen; + rcu_read_lock(); +restart: + bptr = *buffer; + blen = *buflen; + read_seqbegin_or_lock(&rename_lock, &seq); while (dentry != root->dentry || vfsmnt != root->mnt) { struct dentry * parent; if (dentry == vfsmnt->mnt_root || IS_ROOT(dentry)) { /* Global root? */ - if (!mnt_has_parent(mnt)) - goto global_root; - dentry = mnt->mnt_mountpoint; - mnt = mnt->mnt_parent; - vfsmnt = &mnt->mnt; - continue; + if (mnt_has_parent(mnt)) { + dentry = mnt->mnt_mountpoint; + mnt = mnt->mnt_parent; + vfsmnt = &mnt->mnt; + continue; + } + /* + * Filesystems needing to implement special "root names" + * should do so with ->d_dname() + */ + if (IS_ROOT(dentry) && + (dentry->d_name.len != 1 || + dentry->d_name.name[0] != '/')) { + WARN(1, "Root dentry has weird name <%.*s>\n", + (int) dentry->d_name.len, + dentry->d_name.name); + } + if (!error) + error = is_mounted(vfsmnt) ? 1 : 2; + break; } parent = dentry->d_parent; prefetch(parent); - spin_lock(&dentry->d_lock); - error = prepend_name(buffer, buflen, &dentry->d_name); - spin_unlock(&dentry->d_lock); - if (!error) - error = prepend(buffer, buflen, "/", 1); + error = prepend_name(&bptr, &blen, &dentry->d_name); if (error) break; - slash = true; dentry = parent; } + if (!(seq & 1)) + rcu_read_unlock(); + if (need_seqretry(&rename_lock, seq)) { + seq = 1; + goto restart; + } + done_seqretry(&rename_lock, seq); - if (!error && !slash) - error = prepend(buffer, buflen, "/", 1); - - return error; - -global_root: - /* - * Filesystems needing to implement special "root names" - * should do so with ->d_dname() - */ - if (IS_ROOT(dentry) && - (dentry->d_name.len != 1 || dentry->d_name.name[0] != '/')) { - WARN(1, "Root dentry has weird name <%.*s>\n", - (int) dentry->d_name.len, dentry->d_name.name); - } - if (!slash) - error = prepend(buffer, buflen, "/", 1); - if (!error) - error = is_mounted(vfsmnt) ? 1 : 2; + if (error >= 0 && bptr == *buffer) { + if (--blen < 0) + error = -ENAMETOOLONG; + else + *--bptr = '/'; + } + *buffer = bptr; + *buflen = blen; return error; } @@ -2616,9 +2817,7 @@ char *__d_path(const struct path *path, prepend(&res, &buflen, "\0", 1); br_read_lock(&vfsmount_lock); - write_seqlock(&rename_lock); error = prepend_path(path, root, &res, &buflen); - write_sequnlock(&rename_lock); br_read_unlock(&vfsmount_lock); if (error < 0) @@ -2637,9 +2836,7 @@ char *d_absolute_path(const struct path *path, prepend(&res, &buflen, "\0", 1); br_read_lock(&vfsmount_lock); - write_seqlock(&rename_lock); error = prepend_path(path, &root, &res, &buflen); - write_sequnlock(&rename_lock); br_read_unlock(&vfsmount_lock); if (error > 1) @@ -2705,9 +2902,7 @@ char *d_path(const struct path *path, char *buf, int buflen) get_fs_root(current->fs, &root); br_read_lock(&vfsmount_lock); - write_seqlock(&rename_lock); error = path_with_deleted(path, &root, &res, &buflen); - write_sequnlock(&rename_lock); br_read_unlock(&vfsmount_lock); if (error < 0) res = ERR_PTR(error); @@ -2742,10 +2937,10 @@ char *simple_dname(struct dentry *dentry, char *buffer, int buflen) char *end = buffer + buflen; /* these dentries are never renamed, so d_lock is not needed */ if (prepend(&end, &buflen, " (deleted)", 11) || - prepend_name(&end, &buflen, &dentry->d_name) || + prepend(&end, &buflen, dentry->d_name.name, dentry->d_name.len) || prepend(&end, &buflen, "/", 1)) end = ERR_PTR(-ENAMETOOLONG); - return end; + return end; } /* @@ -2753,30 +2948,42 @@ char *simple_dname(struct dentry *dentry, char *buffer, int buflen) */ static char *__dentry_path(struct dentry *dentry, char *buf, int buflen) { - char *end = buf + buflen; - char *retval; + char *end, *retval; + int len, seq = 0; + int error = 0; - prepend(&end, &buflen, "\0", 1); + rcu_read_lock(); +restart: + end = buf + buflen; + len = buflen; + prepend(&end, &len, "\0", 1); if (buflen < 1) goto Elong; /* Get '/' right */ retval = end-1; *retval = '/'; - + read_seqbegin_or_lock(&rename_lock, &seq); while (!IS_ROOT(dentry)) { struct dentry *parent = dentry->d_parent; int error; prefetch(parent); - spin_lock(&dentry->d_lock); - error = prepend_name(&end, &buflen, &dentry->d_name); - spin_unlock(&dentry->d_lock); - if (error != 0 || prepend(&end, &buflen, "/", 1) != 0) - goto Elong; + error = prepend_name(&end, &len, &dentry->d_name); + if (error) + break; retval = end; dentry = parent; } + if (!(seq & 1)) + rcu_read_unlock(); + if (need_seqretry(&rename_lock, seq)) { + seq = 1; + goto restart; + } + done_seqretry(&rename_lock, seq); + if (error) + goto Elong; return retval; Elong: return ERR_PTR(-ENAMETOOLONG); @@ -2784,13 +2991,7 @@ Elong: char *dentry_path_raw(struct dentry *dentry, char *buf, int buflen) { - char *retval; - - write_seqlock(&rename_lock); - retval = __dentry_path(dentry, buf, buflen); - write_sequnlock(&rename_lock); - - return retval; + return __dentry_path(dentry, buf, buflen); } EXPORT_SYMBOL(dentry_path_raw); @@ -2799,7 +3000,6 @@ char *dentry_path(struct dentry *dentry, char *buf, int buflen) char *p = NULL; char *retval; - write_seqlock(&rename_lock); if (d_unlinked(dentry)) { p = buf + buflen; if (prepend(&p, &buflen, "//deleted", 10) != 0) @@ -2807,7 +3007,6 @@ char *dentry_path(struct dentry *dentry, char *buf, int buflen) buflen++; } retval = __dentry_path(dentry, buf, buflen); - write_sequnlock(&rename_lock); if (!IS_ERR(retval) && p) *p = '/'; /* restore '/' overriden with '\0' */ return retval; @@ -2846,7 +3045,6 @@ SYSCALL_DEFINE2(getcwd, char __user *, buf, unsigned long, size) error = -ENOENT; br_read_lock(&vfsmount_lock); - write_seqlock(&rename_lock); if (!d_unlinked(pwd.dentry)) { unsigned long len; char *cwd = page + PAGE_SIZE; @@ -2854,7 +3052,6 @@ SYSCALL_DEFINE2(getcwd, char __user *, buf, unsigned long, size) prepend(&cwd, &buflen, "\0", 1); error = prepend_path(&pwd, &root, &cwd, &buflen); - write_sequnlock(&rename_lock); br_read_unlock(&vfsmount_lock); if (error < 0) @@ -2875,7 +3072,6 @@ SYSCALL_DEFINE2(getcwd, char __user *, buf, unsigned long, size) error = -EFAULT; } } else { - write_sequnlock(&rename_lock); br_read_unlock(&vfsmount_lock); } @@ -2928,68 +3124,24 @@ int is_subdir(struct dentry *new_dentry, struct dentry *old_dentry) return result; } -void d_genocide(struct dentry *root) +static enum d_walk_ret d_genocide_kill(void *data, struct dentry *dentry) { - struct dentry *this_parent; - struct list_head *next; - unsigned seq; - int locked = 0; - - seq = read_seqbegin(&rename_lock); -again: - this_parent = root; - spin_lock(&this_parent->d_lock); -repeat: - next = this_parent->d_subdirs.next; -resume: - while (next != &this_parent->d_subdirs) { - struct list_head *tmp = next; - struct dentry *dentry = list_entry(tmp, struct dentry, d_u.d_child); - next = tmp->next; + struct dentry *root = data; + if (dentry != root) { + if (d_unhashed(dentry) || !dentry->d_inode) + return D_WALK_SKIP; - spin_lock_nested(&dentry->d_lock, DENTRY_D_LOCK_NESTED); - if (d_unhashed(dentry) || !dentry->d_inode) { - spin_unlock(&dentry->d_lock); - continue; - } - if (!list_empty(&dentry->d_subdirs)) { - spin_unlock(&this_parent->d_lock); - spin_release(&dentry->d_lock.dep_map, 1, _RET_IP_); - this_parent = dentry; - spin_acquire(&this_parent->d_lock.dep_map, 0, 1, _RET_IP_); - goto repeat; - } if (!(dentry->d_flags & DCACHE_GENOCIDE)) { dentry->d_flags |= DCACHE_GENOCIDE; dentry->d_lockref.count--; } - spin_unlock(&dentry->d_lock); - } - if (this_parent != root) { - struct dentry *child = this_parent; - if (!(this_parent->d_flags & DCACHE_GENOCIDE)) { - this_parent->d_flags |= DCACHE_GENOCIDE; - this_parent->d_lockref.count--; - } - this_parent = try_to_ascend(this_parent, locked, seq); - if (!this_parent) - goto rename_retry; - next = child->d_u.d_child.next; - goto resume; } - spin_unlock(&this_parent->d_lock); - if (!locked && read_seqretry(&rename_lock, seq)) - goto rename_retry; - if (locked) - write_sequnlock(&rename_lock); - return; + return D_WALK_CONTINUE; +} -rename_retry: - if (locked) - goto again; - locked = 1; - write_seqlock(&rename_lock); - goto again; +void d_genocide(struct dentry *parent) +{ + d_walk(parent, parent, d_genocide_kill, NULL); } void d_tmpfile(struct dentry *dentry, struct inode *inode) diff --git a/fs/direct-io.c b/fs/direct-io.c index 1782023..0e04142 100644 --- a/fs/direct-io.c +++ b/fs/direct-io.c @@ -544,6 +544,7 @@ static inline int dio_bio_reap(struct dio *dio, struct dio_submit *sdio) */ static int sb_init_dio_done_wq(struct super_block *sb) { + struct workqueue_struct *old; struct workqueue_struct *wq = alloc_workqueue("dio/%s", WQ_MEM_RECLAIM, 0, sb->s_id); @@ -552,9 +553,9 @@ static int sb_init_dio_done_wq(struct super_block *sb) /* * This has to be atomic as more DIOs can race to create the workqueue */ - cmpxchg(&sb->s_dio_done_wq, NULL, wq); + old = cmpxchg(&sb->s_dio_done_wq, NULL, wq); /* Someone created workqueue before us? Free ours... */ - if (wq != sb->s_dio_done_wq) + if (old) destroy_workqueue(wq); return 0; } diff --git a/fs/eventpoll.c b/fs/eventpoll.c index 293f867..473e09d 100644 --- a/fs/eventpoll.c +++ b/fs/eventpoll.c @@ -740,6 +740,7 @@ static void ep_free(struct eventpoll *ep) epi = rb_entry(rbp, struct epitem, rbn); ep_unregister_pollwait(ep, epi); + cond_resched(); } /* @@ -754,6 +755,7 @@ static void ep_free(struct eventpoll *ep) while ((rbp = rb_first(&ep->rbr)) != NULL) { epi = rb_entry(rbp, struct epitem, rbn); ep_remove(ep, epi); + cond_resched(); } mutex_unlock(&ep->mtx); @@ -74,6 +74,8 @@ static DEFINE_RWLOCK(binfmt_lock); void __register_binfmt(struct linux_binfmt * fmt, int insert) { BUG_ON(!fmt); + if (WARN_ON(!fmt->load_binary)) + return; write_lock(&binfmt_lock); insert ? list_add(&fmt->lh, &formats) : list_add_tail(&fmt->lh, &formats); @@ -266,7 +268,7 @@ static int __bprm_mm_init(struct linux_binprm *bprm) BUILD_BUG_ON(VM_STACK_FLAGS & VM_STACK_INCOMPLETE_SETUP); vma->vm_end = STACK_TOP_MAX; vma->vm_start = vma->vm_end - PAGE_SIZE; - vma->vm_flags = VM_STACK_FLAGS | VM_STACK_INCOMPLETE_SETUP; + vma->vm_flags = VM_SOFTDIRTY | VM_STACK_FLAGS | VM_STACK_INCOMPLETE_SETUP; vma->vm_page_prot = vm_get_page_prot(vma->vm_flags); INIT_LIST_HEAD(&vma->anon_vma_chain); @@ -1365,18 +1367,18 @@ out: } EXPORT_SYMBOL(remove_arg_zero); +#define printable(c) (((c)=='\t') || ((c)=='\n') || (0x20<=(c) && (c)<=0x7e)) /* * cycle the list of binary formats handler, until one recognizes the image */ int search_binary_handler(struct linux_binprm *bprm) { - unsigned int depth = bprm->recursion_depth; - int try,retval; + bool need_retry = IS_ENABLED(CONFIG_MODULES); struct linux_binfmt *fmt; - pid_t old_pid, old_vpid; + int retval; /* This allows 4 levels of binfmt rewrites before failing hard. */ - if (depth > 5) + if (bprm->recursion_depth > 5) return -ELOOP; retval = security_bprm_check(bprm); @@ -1387,71 +1389,67 @@ int search_binary_handler(struct linux_binprm *bprm) if (retval) return retval; + retval = -ENOENT; + retry: + read_lock(&binfmt_lock); + list_for_each_entry(fmt, &formats, lh) { + if (!try_module_get(fmt->module)) + continue; + read_unlock(&binfmt_lock); + bprm->recursion_depth++; + retval = fmt->load_binary(bprm); + bprm->recursion_depth--; + if (retval >= 0 || retval != -ENOEXEC || + bprm->mm == NULL || bprm->file == NULL) { + put_binfmt(fmt); + return retval; + } + read_lock(&binfmt_lock); + put_binfmt(fmt); + } + read_unlock(&binfmt_lock); + + if (need_retry && retval == -ENOEXEC) { + if (printable(bprm->buf[0]) && printable(bprm->buf[1]) && + printable(bprm->buf[2]) && printable(bprm->buf[3])) + return retval; + if (request_module("binfmt-%04x", *(ushort *)(bprm->buf + 2)) < 0) + return retval; + need_retry = false; + goto retry; + } + + return retval; +} +EXPORT_SYMBOL(search_binary_handler); + +static int exec_binprm(struct linux_binprm *bprm) +{ + pid_t old_pid, old_vpid; + int ret; + /* Need to fetch pid before load_binary changes it */ old_pid = current->pid; rcu_read_lock(); old_vpid = task_pid_nr_ns(current, task_active_pid_ns(current->parent)); rcu_read_unlock(); - retval = -ENOENT; - for (try=0; try<2; try++) { - read_lock(&binfmt_lock); - list_for_each_entry(fmt, &formats, lh) { - int (*fn)(struct linux_binprm *) = fmt->load_binary; - if (!fn) - continue; - if (!try_module_get(fmt->module)) - continue; - read_unlock(&binfmt_lock); - bprm->recursion_depth = depth + 1; - retval = fn(bprm); - bprm->recursion_depth = depth; - if (retval >= 0) { - if (depth == 0) { - trace_sched_process_exec(current, old_pid, bprm); - ptrace_event(PTRACE_EVENT_EXEC, old_vpid); - } - put_binfmt(fmt); - allow_write_access(bprm->file); - if (bprm->file) - fput(bprm->file); - bprm->file = NULL; - current->did_exec = 1; - proc_exec_connector(current); - return retval; - } - read_lock(&binfmt_lock); - put_binfmt(fmt); - if (retval != -ENOEXEC || bprm->mm == NULL) - break; - if (!bprm->file) { - read_unlock(&binfmt_lock); - return retval; - } + ret = search_binary_handler(bprm); + if (ret >= 0) { + trace_sched_process_exec(current, old_pid, bprm); + ptrace_event(PTRACE_EVENT_EXEC, old_vpid); + current->did_exec = 1; + proc_exec_connector(current); + + if (bprm->file) { + allow_write_access(bprm->file); + fput(bprm->file); + bprm->file = NULL; /* to catch use-after-free */ } - read_unlock(&binfmt_lock); -#ifdef CONFIG_MODULES - if (retval != -ENOEXEC || bprm->mm == NULL) { - break; - } else { -#define printable(c) (((c)=='\t') || ((c)=='\n') || (0x20<=(c) && (c)<=0x7e)) - if (printable(bprm->buf[0]) && - printable(bprm->buf[1]) && - printable(bprm->buf[2]) && - printable(bprm->buf[3])) - break; /* -ENOEXEC */ - if (try) - break; /* -ENOEXEC */ - request_module("binfmt-%04x", *(unsigned short *)(&bprm->buf[2])); - } -#else - break; -#endif } - return retval; -} -EXPORT_SYMBOL(search_binary_handler); + return ret; +} /* * sys_execve() executes a new program. @@ -1541,7 +1539,7 @@ static int do_execve_common(const char *filename, if (retval < 0) goto out; - retval = search_binary_handler(bprm); + retval = exec_binprm(bprm); if (retval < 0) goto out; diff --git a/fs/exportfs/expfs.c b/fs/exportfs/expfs.c index 293bc2e..a235f00 100644 --- a/fs/exportfs/expfs.c +++ b/fs/exportfs/expfs.c @@ -231,7 +231,7 @@ static int filldir_one(void * __buf, const char * name, int len, int result = 0; buf->sequence++; - if (buf->ino == ino) { + if (buf->ino == ino && len <= NAME_MAX) { memcpy(buf->name, name, len); buf->name[len] = '\0'; buf->found = 1; diff --git a/fs/file_table.c b/fs/file_table.c index 322cd37..abdd15a 100644 --- a/fs/file_table.c +++ b/fs/file_table.c @@ -311,8 +311,7 @@ void fput(struct file *file) return; /* * After this task has run exit_task_work(), - * task_work_add() will fail. free_ipc_ns()-> - * shm_destroy() can do this. Fall through to delayed + * task_work_add() will fail. Fall through to delayed * fput to avoid leaking *file. */ } diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 68851ff..30f6f27 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -723,7 +723,7 @@ static long __writeback_inodes_wb(struct bdi_writeback *wb, return wrote; } -long writeback_inodes_wb(struct bdi_writeback *wb, long nr_pages, +static long writeback_inodes_wb(struct bdi_writeback *wb, long nr_pages, enum wb_reason reason) { struct wb_writeback_work work = { @@ -1049,10 +1049,8 @@ void wakeup_flusher_threads(long nr_pages, enum wb_reason reason) { struct backing_dev_info *bdi; - if (!nr_pages) { - nr_pages = global_page_state(NR_FILE_DIRTY) + - global_page_state(NR_UNSTABLE_NFS); - } + if (!nr_pages) + nr_pages = get_nr_dirty_pages(); rcu_read_lock(); list_for_each_entry_rcu(bdi, &bdi_list, bdi_list) { @@ -1173,6 +1171,8 @@ void __mark_inode_dirty(struct inode *inode, int flags) bool wakeup_bdi = false; bdi = inode_to_bdi(inode); + spin_unlock(&inode->i_lock); + spin_lock(&bdi->wb.list_lock); if (bdi_cap_writeback_dirty(bdi)) { WARN(!test_bit(BDI_registered, &bdi->state), "bdi-%s not registered\n", bdi->name); @@ -1187,8 +1187,6 @@ void __mark_inode_dirty(struct inode *inode, int flags) wakeup_bdi = true; } - spin_unlock(&inode->i_lock); - spin_lock(&bdi->wb.list_lock); inode->dirtied_when = jiffies; list_move(&inode->i_wb_list, &bdi->wb.b_dirty); spin_unlock(&bdi->wb.list_lock); diff --git a/fs/fscache/cookie.c b/fs/fscache/cookie.c index 0e91a3c..318e843 100644 --- a/fs/fscache/cookie.c +++ b/fs/fscache/cookie.c @@ -558,3 +558,74 @@ void __fscache_cookie_put(struct fscache_cookie *cookie) _leave(""); } + +/* + * check the consistency between the netfs inode and the backing cache + * + * NOTE: it only serves no-index type + */ +int __fscache_check_consistency(struct fscache_cookie *cookie) +{ + struct fscache_operation *op; + struct fscache_object *object; + int ret; + + _enter("%p,", cookie); + + ASSERTCMP(cookie->def->type, ==, FSCACHE_COOKIE_TYPE_DATAFILE); + + if (fscache_wait_for_deferred_lookup(cookie) < 0) + return -ERESTARTSYS; + + if (hlist_empty(&cookie->backing_objects)) + return 0; + + op = kzalloc(sizeof(*op), GFP_NOIO | __GFP_NOMEMALLOC | __GFP_NORETRY); + if (!op) + return -ENOMEM; + + fscache_operation_init(op, NULL, NULL); + op->flags = FSCACHE_OP_MYTHREAD | + (1 << FSCACHE_OP_WAITING); + + spin_lock(&cookie->lock); + + if (hlist_empty(&cookie->backing_objects)) + goto inconsistent; + object = hlist_entry(cookie->backing_objects.first, + struct fscache_object, cookie_link); + if (test_bit(FSCACHE_IOERROR, &object->cache->flags)) + goto inconsistent; + + op->debug_id = atomic_inc_return(&fscache_op_debug_id); + + atomic_inc(&cookie->n_active); + if (fscache_submit_op(object, op) < 0) + goto submit_failed; + + /* the work queue now carries its own ref on the object */ + spin_unlock(&cookie->lock); + + ret = fscache_wait_for_operation_activation(object, op, + NULL, NULL, NULL); + if (ret == 0) { + /* ask the cache to honour the operation */ + ret = object->cache->ops->check_consistency(op); + fscache_op_complete(op, false); + } else if (ret == -ENOBUFS) { + ret = 0; + } + + fscache_put_operation(op); + _leave(" = %d", ret); + return ret; + +submit_failed: + atomic_dec(&cookie->n_active); +inconsistent: + spin_unlock(&cookie->lock); + kfree(op); + _leave(" = -ESTALE"); + return -ESTALE; +} +EXPORT_SYMBOL(__fscache_check_consistency); diff --git a/fs/fscache/internal.h b/fs/fscache/internal.h index 12d505b..4226f66 100644 --- a/fs/fscache/internal.h +++ b/fs/fscache/internal.h @@ -130,6 +130,12 @@ extern void fscache_operation_gc(struct work_struct *); /* * page.c */ +extern int fscache_wait_for_deferred_lookup(struct fscache_cookie *); +extern int fscache_wait_for_operation_activation(struct fscache_object *, + struct fscache_operation *, + atomic_t *, + atomic_t *, + void (*)(struct fscache_operation *)); extern void fscache_invalidate_writes(struct fscache_cookie *); /* diff --git a/fs/fscache/page.c b/fs/fscache/page.c index d479ab3..73899c1 100644 --- a/fs/fscache/page.c +++ b/fs/fscache/page.c @@ -278,7 +278,7 @@ static struct fscache_retrieval *fscache_alloc_retrieval( /* * wait for a deferred lookup to complete */ -static int fscache_wait_for_deferred_lookup(struct fscache_cookie *cookie) +int fscache_wait_for_deferred_lookup(struct fscache_cookie *cookie) { unsigned long jif; @@ -322,42 +322,46 @@ static void fscache_do_cancel_retrieval(struct fscache_operation *_op) /* * wait for an object to become active (or dead) */ -static int fscache_wait_for_retrieval_activation(struct fscache_object *object, - struct fscache_retrieval *op, - atomic_t *stat_op_waits, - atomic_t *stat_object_dead) +int fscache_wait_for_operation_activation(struct fscache_object *object, + struct fscache_operation *op, + atomic_t *stat_op_waits, + atomic_t *stat_object_dead, + void (*do_cancel)(struct fscache_operation *)) { int ret; - if (!test_bit(FSCACHE_OP_WAITING, &op->op.flags)) + if (!test_bit(FSCACHE_OP_WAITING, &op->flags)) goto check_if_dead; _debug(">>> WT"); - fscache_stat(stat_op_waits); - if (wait_on_bit(&op->op.flags, FSCACHE_OP_WAITING, + if (stat_op_waits) + fscache_stat(stat_op_waits); + if (wait_on_bit(&op->flags, FSCACHE_OP_WAITING, fscache_wait_bit_interruptible, TASK_INTERRUPTIBLE) != 0) { - ret = fscache_cancel_op(&op->op, fscache_do_cancel_retrieval); + ret = fscache_cancel_op(op, do_cancel); if (ret == 0) return -ERESTARTSYS; /* it's been removed from the pending queue by another party, * so we should get to run shortly */ - wait_on_bit(&op->op.flags, FSCACHE_OP_WAITING, + wait_on_bit(&op->flags, FSCACHE_OP_WAITING, fscache_wait_bit, TASK_UNINTERRUPTIBLE); } _debug("<<< GO"); check_if_dead: - if (op->op.state == FSCACHE_OP_ST_CANCELLED) { - fscache_stat(stat_object_dead); + if (op->state == FSCACHE_OP_ST_CANCELLED) { + if (stat_object_dead) + fscache_stat(stat_object_dead); _leave(" = -ENOBUFS [cancelled]"); return -ENOBUFS; } if (unlikely(fscache_object_is_dead(object))) { - pr_err("%s() = -ENOBUFS [obj dead %d]\n", __func__, op->op.state); - fscache_cancel_op(&op->op, fscache_do_cancel_retrieval); - fscache_stat(stat_object_dead); + pr_err("%s() = -ENOBUFS [obj dead %d]\n", __func__, op->state); + fscache_cancel_op(op, do_cancel); + if (stat_object_dead) + fscache_stat(stat_object_dead); return -ENOBUFS; } return 0; @@ -432,10 +436,11 @@ int __fscache_read_or_alloc_page(struct fscache_cookie *cookie, /* we wait for the operation to become active, and then process it * *here*, in this thread, and not in the thread pool */ - ret = fscache_wait_for_retrieval_activation( - object, op, + ret = fscache_wait_for_operation_activation( + object, &op->op, __fscache_stat(&fscache_n_retrieval_op_waits), - __fscache_stat(&fscache_n_retrievals_object_dead)); + __fscache_stat(&fscache_n_retrievals_object_dead), + fscache_do_cancel_retrieval); if (ret < 0) goto error; @@ -557,10 +562,11 @@ int __fscache_read_or_alloc_pages(struct fscache_cookie *cookie, /* we wait for the operation to become active, and then process it * *here*, in this thread, and not in the thread pool */ - ret = fscache_wait_for_retrieval_activation( - object, op, + ret = fscache_wait_for_operation_activation( + object, &op->op, __fscache_stat(&fscache_n_retrieval_op_waits), - __fscache_stat(&fscache_n_retrievals_object_dead)); + __fscache_stat(&fscache_n_retrievals_object_dead), + fscache_do_cancel_retrieval); if (ret < 0) goto error; @@ -658,10 +664,11 @@ int __fscache_alloc_page(struct fscache_cookie *cookie, fscache_stat(&fscache_n_alloc_ops); - ret = fscache_wait_for_retrieval_activation( - object, op, + ret = fscache_wait_for_operation_activation( + object, &op->op, __fscache_stat(&fscache_n_alloc_op_waits), - __fscache_stat(&fscache_n_allocs_object_dead)); + __fscache_stat(&fscache_n_allocs_object_dead), + fscache_do_cancel_retrieval); if (ret < 0) goto error; @@ -694,6 +701,22 @@ nobufs: EXPORT_SYMBOL(__fscache_alloc_page); /* + * Unmark pages allocate in the readahead code path (via: + * fscache_readpages_or_alloc) after delegating to the base filesystem + */ +void __fscache_readpages_cancel(struct fscache_cookie *cookie, + struct list_head *pages) +{ + struct page *page; + + list_for_each_entry(page, pages, lru) { + if (PageFsCache(page)) + __fscache_uncache_page(cookie, page); + } +} +EXPORT_SYMBOL(__fscache_readpages_cancel); + +/* * release a write op reference */ static void fscache_release_write_op(struct fscache_operation *_op) @@ -890,7 +913,7 @@ int __fscache_write_page(struct fscache_cookie *cookie, (1 << FSCACHE_OP_WAITING) | (1 << FSCACHE_OP_UNUSE_COOKIE); - ret = radix_tree_preload(gfp & ~__GFP_HIGHMEM); + ret = radix_tree_maybe_preload(gfp & ~__GFP_HIGHMEM); if (ret < 0) goto nomem_free; diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c index 1d55f94..ef74ad5 100644 --- a/fs/fuse/dev.c +++ b/fs/fuse/dev.c @@ -1765,11 +1765,9 @@ static int fuse_notify(struct fuse_conn *fc, enum fuse_notify_code code, /* Look up request on processing list by unique ID */ static struct fuse_req *request_find(struct fuse_conn *fc, u64 unique) { - struct list_head *entry; + struct fuse_req *req; - list_for_each(entry, &fc->processing) { - struct fuse_req *req; - req = list_entry(entry, struct fuse_req, list); + list_for_each_entry(req, &fc->processing, list) { if (req->in.h.unique == unique || req->intr_unique == unique) return req; } diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c index 72a5d5b..3ac9108 100644 --- a/fs/fuse/dir.c +++ b/fs/fuse/dir.c @@ -182,10 +182,11 @@ static int fuse_dentry_revalidate(struct dentry *entry, unsigned int flags) struct inode *inode; struct dentry *parent; struct fuse_conn *fc; + int ret; inode = ACCESS_ONCE(entry->d_inode); if (inode && is_bad_inode(inode)) - return 0; + goto invalid; else if (fuse_dentry_time(entry) < get_jiffies_64()) { int err; struct fuse_entry_out outarg; @@ -195,20 +196,23 @@ static int fuse_dentry_revalidate(struct dentry *entry, unsigned int flags) /* For negative dentries, always do a fresh lookup */ if (!inode) - return 0; + goto invalid; + ret = -ECHILD; if (flags & LOOKUP_RCU) - return -ECHILD; + goto out; fc = get_fuse_conn(inode); req = fuse_get_req_nopages(fc); + ret = PTR_ERR(req); if (IS_ERR(req)) - return 0; + goto out; forget = fuse_alloc_forget(); if (!forget) { fuse_put_request(fc, req); - return 0; + ret = -ENOMEM; + goto out; } attr_version = fuse_get_attr_version(fc); @@ -227,7 +231,7 @@ static int fuse_dentry_revalidate(struct dentry *entry, unsigned int flags) struct fuse_inode *fi = get_fuse_inode(inode); if (outarg.nodeid != get_node_id(inode)) { fuse_queue_forget(fc, forget, outarg.nodeid, 1); - return 0; + goto invalid; } spin_lock(&fc->lock); fi->nlookup++; @@ -235,7 +239,7 @@ static int fuse_dentry_revalidate(struct dentry *entry, unsigned int flags) } kfree(forget); if (err || (outarg.attr.mode ^ inode->i_mode) & S_IFMT) - return 0; + goto invalid; fuse_change_attributes(inode, &outarg.attr, entry_attr_timeout(&outarg), @@ -249,7 +253,15 @@ static int fuse_dentry_revalidate(struct dentry *entry, unsigned int flags) dput(parent); } } - return 1; + ret = 1; +out: + return ret; + +invalid: + ret = 0; + if (check_submounts_and_drop(entry) != 0) + ret = 1; + goto out; } static int invalid_nodeid(u64 nodeid) @@ -267,26 +279,6 @@ int fuse_valid_type(int m) S_ISBLK(m) || S_ISFIFO(m) || S_ISSOCK(m); } -/* - * Add a directory inode to a dentry, ensuring that no other dentry - * refers to this inode. Called with fc->inst_mutex. - */ -static struct dentry *fuse_d_add_directory(struct dentry *entry, - struct inode *inode) -{ - struct dentry *alias = d_find_alias(inode); - if (alias && !(alias->d_flags & DCACHE_DISCONNECTED)) { - /* This tries to shrink the subtree below alias */ - fuse_invalidate_entry(alias); - dput(alias); - if (!hlist_empty(&inode->i_dentry)) - return ERR_PTR(-EBUSY); - } else { - dput(alias); - } - return d_splice_alias(inode, entry); -} - int fuse_lookup_name(struct super_block *sb, u64 nodeid, struct qstr *name, struct fuse_entry_out *outarg, struct inode **inode) { @@ -345,6 +337,24 @@ int fuse_lookup_name(struct super_block *sb, u64 nodeid, struct qstr *name, return err; } +static struct dentry *fuse_materialise_dentry(struct dentry *dentry, + struct inode *inode) +{ + struct dentry *newent; + + if (inode && S_ISDIR(inode->i_mode)) { + struct fuse_conn *fc = get_fuse_conn(inode); + + mutex_lock(&fc->inst_mutex); + newent = d_materialise_unique(dentry, inode); + mutex_unlock(&fc->inst_mutex); + } else { + newent = d_materialise_unique(dentry, inode); + } + + return newent; +} + static struct dentry *fuse_lookup(struct inode *dir, struct dentry *entry, unsigned int flags) { @@ -352,7 +362,6 @@ static struct dentry *fuse_lookup(struct inode *dir, struct dentry *entry, struct fuse_entry_out outarg; struct inode *inode; struct dentry *newent; - struct fuse_conn *fc = get_fuse_conn(dir); bool outarg_valid = true; err = fuse_lookup_name(dir->i_sb, get_node_id(dir), &entry->d_name, @@ -368,16 +377,10 @@ static struct dentry *fuse_lookup(struct inode *dir, struct dentry *entry, if (inode && get_node_id(inode) == FUSE_ROOT_ID) goto out_iput; - if (inode && S_ISDIR(inode->i_mode)) { - mutex_lock(&fc->inst_mutex); - newent = fuse_d_add_directory(entry, inode); - mutex_unlock(&fc->inst_mutex); - err = PTR_ERR(newent); - if (IS_ERR(newent)) - goto out_iput; - } else { - newent = d_splice_alias(inode, entry); - } + newent = fuse_materialise_dentry(entry, inode); + err = PTR_ERR(newent); + if (IS_ERR(newent)) + goto out_err; entry = newent ? newent : entry; if (outarg_valid) @@ -1174,6 +1177,8 @@ static int parse_dirfile(char *buf, size_t nbytes, struct file *file, return -EIO; if (reclen > nbytes) break; + if (memchr(dirent->name, '/', dirent->namelen) != NULL) + return -EIO; if (!dir_emit(ctx, dirent->name, dirent->namelen, dirent->ino, dirent->type)) @@ -1275,18 +1280,10 @@ static int fuse_direntplus_link(struct file *file, if (!inode) goto out; - if (S_ISDIR(inode->i_mode)) { - mutex_lock(&fc->inst_mutex); - alias = fuse_d_add_directory(dentry, inode); - mutex_unlock(&fc->inst_mutex); - err = PTR_ERR(alias); - if (IS_ERR(alias)) { - iput(inode); - goto out; - } - } else { - alias = d_splice_alias(inode, dentry); - } + alias = fuse_materialise_dentry(dentry, inode); + err = PTR_ERR(alias); + if (IS_ERR(alias)) + goto out; if (alias) { dput(dentry); @@ -1320,6 +1317,8 @@ static int parse_dirplusfile(char *buf, size_t nbytes, struct file *file, return -EIO; if (reclen > nbytes) break; + if (memchr(dirent->name, '/', dirent->namelen) != NULL) + return -EIO; if (!over) { /* We fill entries into dstbuf only as much as @@ -1590,6 +1589,7 @@ int fuse_do_setattr(struct inode *inode, struct iattr *attr, struct file *file) { struct fuse_conn *fc = get_fuse_conn(inode); + struct fuse_inode *fi = get_fuse_inode(inode); struct fuse_req *req; struct fuse_setattr_in inarg; struct fuse_attr_out outarg; @@ -1617,8 +1617,10 @@ int fuse_do_setattr(struct inode *inode, struct iattr *attr, if (IS_ERR(req)) return PTR_ERR(req); - if (is_truncate) + if (is_truncate) { fuse_set_nowrite(inode); + set_bit(FUSE_I_SIZE_UNSTABLE, &fi->state); + } memset(&inarg, 0, sizeof(inarg)); memset(&outarg, 0, sizeof(outarg)); @@ -1680,12 +1682,14 @@ int fuse_do_setattr(struct inode *inode, struct iattr *attr, invalidate_inode_pages2(inode->i_mapping); } + clear_bit(FUSE_I_SIZE_UNSTABLE, &fi->state); return 0; error: if (is_truncate) fuse_release_nowrite(inode); + clear_bit(FUSE_I_SIZE_UNSTABLE, &fi->state); return err; } @@ -1749,6 +1753,8 @@ static int fuse_setxattr(struct dentry *entry, const char *name, fc->no_setxattr = 1; err = -EOPNOTSUPP; } + if (!err) + fuse_invalidate_attr(inode); return err; } @@ -1878,6 +1884,8 @@ static int fuse_removexattr(struct dentry *entry, const char *name) fc->no_removexattr = 1; err = -EOPNOTSUPP; } + if (!err) + fuse_invalidate_attr(inode); return err; } diff --git a/fs/fuse/file.c b/fs/fuse/file.c index 5c121fe..d409dea 100644 --- a/fs/fuse/file.c +++ b/fs/fuse/file.c @@ -629,7 +629,8 @@ static void fuse_read_update_size(struct inode *inode, loff_t size, struct fuse_inode *fi = get_fuse_inode(inode); spin_lock(&fc->lock); - if (attr_ver == fi->attr_version && size < inode->i_size) { + if (attr_ver == fi->attr_version && size < inode->i_size && + !test_bit(FUSE_I_SIZE_UNSTABLE, &fi->state)) { fi->attr_version = ++fc->attr_version; i_size_write(inode, size); } @@ -1032,12 +1033,16 @@ static ssize_t fuse_perform_write(struct file *file, { struct inode *inode = mapping->host; struct fuse_conn *fc = get_fuse_conn(inode); + struct fuse_inode *fi = get_fuse_inode(inode); int err = 0; ssize_t res = 0; if (is_bad_inode(inode)) return -EIO; + if (inode->i_size < pos + iov_iter_count(ii)) + set_bit(FUSE_I_SIZE_UNSTABLE, &fi->state); + do { struct fuse_req *req; ssize_t count; @@ -1073,6 +1078,7 @@ static ssize_t fuse_perform_write(struct file *file, if (res > 0) fuse_write_update_size(inode, pos); + clear_bit(FUSE_I_SIZE_UNSTABLE, &fi->state); fuse_invalidate_attr(inode); return res > 0 ? res : err; @@ -1529,7 +1535,6 @@ static int fuse_writepage_locked(struct page *page) inc_bdi_stat(mapping->backing_dev_info, BDI_WRITEBACK); inc_zone_page_state(tmp_page, NR_WRITEBACK_TEMP); - end_page_writeback(page); spin_lock(&fc->lock); list_add(&req->writepages_entry, &fi->writepages); @@ -1537,6 +1542,8 @@ static int fuse_writepage_locked(struct page *page) fuse_flush_writepages(inode); spin_unlock(&fc->lock); + end_page_writeback(page); + return 0; err_free: diff --git a/fs/fuse/fuse_i.h b/fs/fuse/fuse_i.h index fde7249..5ced199 100644 --- a/fs/fuse/fuse_i.h +++ b/fs/fuse/fuse_i.h @@ -115,6 +115,8 @@ struct fuse_inode { enum { /** Advise readdirplus */ FUSE_I_ADVISE_RDPLUS, + /** An operation changing file size is in progress */ + FUSE_I_SIZE_UNSTABLE, }; struct fuse_conn; diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c index 0b57859..8443459 100644 --- a/fs/fuse/inode.c +++ b/fs/fuse/inode.c @@ -201,7 +201,8 @@ void fuse_change_attributes(struct inode *inode, struct fuse_attr *attr, struct timespec old_mtime; spin_lock(&fc->lock); - if (attr_version != 0 && fi->attr_version > attr_version) { + if ((attr_version != 0 && fi->attr_version > attr_version) || + test_bit(FUSE_I_SIZE_UNSTABLE, &fi->state)) { spin_unlock(&fc->lock); return; } @@ -929,7 +930,7 @@ static int fuse_bdi_init(struct fuse_conn *fc, struct super_block *sb) fc->bdi.name = "fuse"; fc->bdi.ra_pages = (VM_MAX_READAHEAD * 1024) / PAGE_CACHE_SIZE; /* fuse does it's own writeback accounting */ - fc->bdi.capabilities = BDI_CAP_NO_ACCT_WB; + fc->bdi.capabilities = BDI_CAP_NO_ACCT_WB | BDI_CAP_STRICTLIMIT; err = bdi_init(&fc->bdi); if (err) diff --git a/fs/gfs2/aops.c b/fs/gfs2/aops.c index ee48ad3..1f7d805 100644 --- a/fs/gfs2/aops.c +++ b/fs/gfs2/aops.c @@ -122,14 +122,13 @@ out: } /** - * gfs2_writeback_writepage - Write page for writeback mappings + * gfs2_writepage - Write page for writeback mappings * @page: The page * @wbc: The writeback control * */ -static int gfs2_writeback_writepage(struct page *page, - struct writeback_control *wbc) +static int gfs2_writepage(struct page *page, struct writeback_control *wbc) { int ret; @@ -141,32 +140,6 @@ static int gfs2_writeback_writepage(struct page *page, } /** - * gfs2_ordered_writepage - Write page for ordered data files - * @page: The page to write - * @wbc: The writeback control - * - */ - -static int gfs2_ordered_writepage(struct page *page, - struct writeback_control *wbc) -{ - struct inode *inode = page->mapping->host; - struct gfs2_inode *ip = GFS2_I(inode); - int ret; - - ret = gfs2_writepage_common(page, wbc); - if (ret <= 0) - return ret; - - if (!page_has_buffers(page)) { - create_empty_buffers(page, inode->i_sb->s_blocksize, - (1 << BH_Dirty)|(1 << BH_Uptodate)); - } - gfs2_page_add_databufs(ip, page, 0, inode->i_sb->s_blocksize-1); - return block_write_full_page(page, gfs2_get_block_noalloc, wbc); -} - -/** * __gfs2_jdata_writepage - The core of jdata writepage * @page: The page to write * @wbc: The writeback control @@ -842,6 +815,8 @@ static int gfs2_write_end(struct file *file, struct address_space *mapping, unsigned int from = pos & (PAGE_CACHE_SIZE - 1); unsigned int to = from + len; int ret; + struct gfs2_trans *tr = current->journal_info; + BUG_ON(!tr); BUG_ON(gfs2_glock_is_locked_by_me(ip->i_gl) == NULL); @@ -852,8 +827,6 @@ static int gfs2_write_end(struct file *file, struct address_space *mapping, goto failed; } - gfs2_trans_add_meta(ip->i_gl, dibh); - if (gfs2_is_stuffed(ip)) return gfs2_stuffed_write_end(inode, dibh, pos, len, copied, page); @@ -861,6 +834,11 @@ static int gfs2_write_end(struct file *file, struct address_space *mapping, gfs2_page_add_databufs(ip, page, from, to); ret = generic_write_end(file, mapping, pos, len, copied, page, fsdata); + if (tr->tr_num_buf_new) + __mark_inode_dirty(inode, I_DIRTY_DATASYNC); + else + gfs2_trans_add_meta(ip->i_gl, dibh); + if (inode == sdp->sd_rindex) { adjust_fs_space(inode); @@ -1107,7 +1085,7 @@ cannot_release: } static const struct address_space_operations gfs2_writeback_aops = { - .writepage = gfs2_writeback_writepage, + .writepage = gfs2_writepage, .writepages = gfs2_writepages, .readpage = gfs2_readpage, .readpages = gfs2_readpages, @@ -1123,7 +1101,7 @@ static const struct address_space_operations gfs2_writeback_aops = { }; static const struct address_space_operations gfs2_ordered_aops = { - .writepage = gfs2_ordered_writepage, + .writepage = gfs2_writepage, .writepages = gfs2_writepages, .readpage = gfs2_readpage, .readpages = gfs2_readpages, diff --git a/fs/gfs2/dentry.c b/fs/gfs2/dentry.c index f2448ab..d3a5d4e 100644 --- a/fs/gfs2/dentry.c +++ b/fs/gfs2/dentry.c @@ -93,12 +93,9 @@ invalid_gunlock: if (!had_lock) gfs2_glock_dq_uninit(&d_gh); invalid: - if (inode && S_ISDIR(inode->i_mode)) { - if (have_submounts(dentry)) - goto valid; - shrink_dcache_parent(dentry); - } - d_drop(dentry); + if (check_submounts_and_drop(dentry) != 0) + goto valid; + dput(parent); return 0; diff --git a/fs/gfs2/file.c b/fs/gfs2/file.c index 72c3866..0621b46 100644 --- a/fs/gfs2/file.c +++ b/fs/gfs2/file.c @@ -650,7 +650,7 @@ static int gfs2_fsync(struct file *file, loff_t start, loff_t end, { struct address_space *mapping = file->f_mapping; struct inode *inode = mapping->host; - int sync_state = inode->i_state & (I_DIRTY_SYNC|I_DIRTY_DATASYNC); + int sync_state = inode->i_state & I_DIRTY; struct gfs2_inode *ip = GFS2_I(inode); int ret = 0, ret1 = 0; @@ -660,6 +660,8 @@ static int gfs2_fsync(struct file *file, loff_t start, loff_t end, return ret1; } + if (!gfs2_is_jdata(ip)) + sync_state &= ~I_DIRTY_PAGES; if (datasync) sync_state &= ~I_DIRTY_SYNC; diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index 544a809..722329c 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -1411,7 +1411,6 @@ __acquires(&lru_lock) if (demote_ok(gl)) handle_callback(gl, LM_ST_UNLOCKED, 0, false); WARN_ON(!test_and_clear_bit(GLF_LOCK, &gl->gl_flags)); - smp_mb__after_clear_bit(); if (queue_delayed_work(glock_workqueue, &gl->gl_work, 0) == 0) gfs2_glock_put_nolock(gl); spin_unlock(&gl->gl_spin); @@ -1488,7 +1487,7 @@ static void examine_bucket(glock_examiner examiner, const struct gfs2_sbd *sdp, rcu_read_lock(); hlist_bl_for_each_entry_rcu(gl, pos, head, gl_list) { - if ((gl->gl_sbd == sdp) && atomic_read(&gl->gl_ref)) + if ((gl->gl_sbd == sdp) && atomic_inc_not_zero(&gl->gl_ref)) examiner(gl); } rcu_read_unlock(); @@ -1508,18 +1507,17 @@ static void glock_hash_walk(glock_examiner examiner, const struct gfs2_sbd *sdp) * thaw_glock - thaw out a glock which has an unprocessed reply waiting * @gl: The glock to thaw * - * N.B. When we freeze a glock, we leave a ref to the glock outstanding, - * so this has to result in the ref count being dropped by one. */ static void thaw_glock(struct gfs2_glock *gl) { if (!test_and_clear_bit(GLF_FROZEN, &gl->gl_flags)) - return; + goto out; set_bit(GLF_REPLY_PENDING, &gl->gl_flags); - gfs2_glock_hold(gl); - if (queue_delayed_work(glock_workqueue, &gl->gl_work, 0) == 0) + if (queue_delayed_work(glock_workqueue, &gl->gl_work, 0) == 0) { +out: gfs2_glock_put(gl); + } } /** @@ -1536,7 +1534,6 @@ static void clear_glock(struct gfs2_glock *gl) if (gl->gl_state != LM_ST_UNLOCKED) handle_callback(gl, LM_ST_UNLOCKED, 0, false); spin_unlock(&gl->gl_spin); - gfs2_glock_hold(gl); if (queue_delayed_work(glock_workqueue, &gl->gl_work, 0) == 0) gfs2_glock_put(gl); } diff --git a/fs/gfs2/lops.c b/fs/gfs2/lops.c index 17c5b5d..010b9fb 100644 --- a/fs/gfs2/lops.c +++ b/fs/gfs2/lops.c @@ -579,6 +579,24 @@ static int buf_lo_scan_elements(struct gfs2_jdesc *jd, unsigned int start, return error; } +/** + * gfs2_meta_sync - Sync all buffers associated with a glock + * @gl: The glock + * + */ + +static void gfs2_meta_sync(struct gfs2_glock *gl) +{ + struct address_space *mapping = gfs2_glock2aspace(gl); + int error; + + filemap_fdatawrite(mapping); + error = filemap_fdatawait(mapping); + + if (error) + gfs2_io_error(gl->gl_sbd); +} + static void buf_lo_after_scan(struct gfs2_jdesc *jd, int error, int pass) { struct gfs2_inode *ip = GFS2_I(jd->jd_inode); diff --git a/fs/gfs2/meta_io.c b/fs/gfs2/meta_io.c index 0da3906..9324150 100644 --- a/fs/gfs2/meta_io.c +++ b/fs/gfs2/meta_io.c @@ -98,24 +98,6 @@ const struct address_space_operations gfs2_meta_aops = { }; /** - * gfs2_meta_sync - Sync all buffers associated with a glock - * @gl: The glock - * - */ - -void gfs2_meta_sync(struct gfs2_glock *gl) -{ - struct address_space *mapping = gfs2_glock2aspace(gl); - int error; - - filemap_fdatawrite(mapping); - error = filemap_fdatawait(mapping); - - if (error) - gfs2_io_error(gl->gl_sbd); -} - -/** * gfs2_getbuf - Get a buffer with a given address space * @gl: the glock * @blkno: the block number (filesystem scope) diff --git a/fs/gfs2/meta_io.h b/fs/gfs2/meta_io.h index 0d4c843..4823b93 100644 --- a/fs/gfs2/meta_io.h +++ b/fs/gfs2/meta_io.h @@ -48,21 +48,17 @@ static inline struct gfs2_sbd *gfs2_mapping2sbd(struct address_space *mapping) return inode->i_sb->s_fs_info; } -void gfs2_meta_sync(struct gfs2_glock *gl); - -struct buffer_head *gfs2_meta_new(struct gfs2_glock *gl, u64 blkno); -int gfs2_meta_read(struct gfs2_glock *gl, u64 blkno, - int flags, struct buffer_head **bhp); -int gfs2_meta_wait(struct gfs2_sbd *sdp, struct buffer_head *bh); -struct buffer_head *gfs2_getbuf(struct gfs2_glock *gl, u64 blkno, int create); - -void gfs2_remove_from_journal(struct buffer_head *bh, struct gfs2_trans *tr, - int meta); - -void gfs2_meta_wipe(struct gfs2_inode *ip, u64 bstart, u32 blen); - -int gfs2_meta_indirect_buffer(struct gfs2_inode *ip, int height, u64 num, - struct buffer_head **bhp); +extern struct buffer_head *gfs2_meta_new(struct gfs2_glock *gl, u64 blkno); +extern int gfs2_meta_read(struct gfs2_glock *gl, u64 blkno, int flags, + struct buffer_head **bhp); +extern int gfs2_meta_wait(struct gfs2_sbd *sdp, struct buffer_head *bh); +extern struct buffer_head *gfs2_getbuf(struct gfs2_glock *gl, u64 blkno, + int create); +extern void gfs2_remove_from_journal(struct buffer_head *bh, + struct gfs2_trans *tr, int meta); +extern void gfs2_meta_wipe(struct gfs2_inode *ip, u64 bstart, u32 blen); +extern int gfs2_meta_indirect_buffer(struct gfs2_inode *ip, int height, u64 num, + struct buffer_head **bhp); static inline int gfs2_meta_inode_buffer(struct gfs2_inode *ip, struct buffer_head **bhp) diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c index 0262c19..19ff5e8 100644 --- a/fs/gfs2/ops_fstype.c +++ b/fs/gfs2/ops_fstype.c @@ -646,6 +646,48 @@ static int gfs2_jindex_hold(struct gfs2_sbd *sdp, struct gfs2_holder *ji_gh) return error; } +/** + * check_journal_clean - Make sure a journal is clean for a spectator mount + * @sdp: The GFS2 superblock + * @jd: The journal descriptor + * + * Returns: 0 if the journal is clean or locked, else an error + */ +static int check_journal_clean(struct gfs2_sbd *sdp, struct gfs2_jdesc *jd) +{ + int error; + struct gfs2_holder j_gh; + struct gfs2_log_header_host head; + struct gfs2_inode *ip; + + ip = GFS2_I(jd->jd_inode); + error = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_NOEXP | + GL_EXACT | GL_NOCACHE, &j_gh); + if (error) { + fs_err(sdp, "Error locking journal for spectator mount.\n"); + return -EPERM; + } + error = gfs2_jdesc_check(jd); + if (error) { + fs_err(sdp, "Error checking journal for spectator mount.\n"); + goto out_unlock; + } + error = gfs2_find_jhead(jd, &head); + if (error) { + fs_err(sdp, "Error parsing journal for spectator mount.\n"); + goto out_unlock; + } + if (!(head.lh_flags & GFS2_LOG_HEAD_UNMOUNT)) { + error = -EPERM; + fs_err(sdp, "jid=%u: Journal is dirty, so the first mounter " + "must not be a spectator.\n", jd->jd_jid); + } + +out_unlock: + gfs2_glock_dq_uninit(&j_gh); + return error; +} + static int init_journal(struct gfs2_sbd *sdp, int undo) { struct inode *master = sdp->sd_master_dir->d_inode; @@ -732,8 +774,15 @@ static int init_journal(struct gfs2_sbd *sdp, int undo) if (sdp->sd_lockstruct.ls_first) { unsigned int x; for (x = 0; x < sdp->sd_journals; x++) { - error = gfs2_recover_journal(gfs2_jdesc_find(sdp, x), - true); + struct gfs2_jdesc *jd = gfs2_jdesc_find(sdp, x); + + if (sdp->sd_args.ar_spectator) { + error = check_journal_clean(sdp, jd); + if (error) + goto fail_jinode_gh; + continue; + } + error = gfs2_recover_journal(jd, true); if (error) { fs_err(sdp, "error recovering journal %u: %d\n", x, error); diff --git a/fs/hfsplus/Kconfig b/fs/hfsplus/Kconfig index a633718..24bc20f 100644 --- a/fs/hfsplus/Kconfig +++ b/fs/hfsplus/Kconfig @@ -11,3 +11,21 @@ config HFSPLUS_FS MacOS 8. It includes all Mac specific filesystem data such as data forks and creator codes, but it also has several UNIX style features such as file ownership and permissions. + +config HFSPLUS_FS_POSIX_ACL + bool "HFS+ POSIX Access Control Lists" + depends on HFSPLUS_FS + select FS_POSIX_ACL + help + POSIX Access Control Lists (ACLs) support permissions for users and + groups beyond the owner/group/world scheme. + + To learn more about Access Control Lists, visit the POSIX ACLs for + Linux website <http://acl.bestbits.at/>. + + It needs to understand that POSIX ACLs are treated only under + Linux. POSIX ACLs doesn't mean something under Mac OS X. + Mac OS X beginning with version 10.4 ("Tiger") support NFSv4 ACLs, + which are part of the NFSv4 standard. + + If you don't know what Access Control Lists are, say N diff --git a/fs/hfsplus/Makefile b/fs/hfsplus/Makefile index 09d278b..683fca2 100644 --- a/fs/hfsplus/Makefile +++ b/fs/hfsplus/Makefile @@ -7,3 +7,5 @@ obj-$(CONFIG_HFSPLUS_FS) += hfsplus.o hfsplus-objs := super.o options.o inode.o ioctl.o extents.o catalog.o dir.o btree.o \ bnode.o brec.o bfind.o tables.o unicode.o wrapper.o bitmap.o part_tbl.o \ attributes.o xattr.o xattr_user.o xattr_security.o xattr_trusted.o + +hfsplus-$(CONFIG_HFSPLUS_FS_POSIX_ACL) += posix_acl.o diff --git a/fs/hfsplus/acl.h b/fs/hfsplus/acl.h new file mode 100644 index 0000000..07c0d49 --- /dev/null +++ b/fs/hfsplus/acl.h @@ -0,0 +1,30 @@ +/* + * linux/fs/hfsplus/acl.h + * + * Vyacheslav Dubeyko <slava@dubeyko.com> + * + * Handler for Posix Access Control Lists (ACLs) support. + */ + +#include <linux/posix_acl_xattr.h> + +#ifdef CONFIG_HFSPLUS_FS_POSIX_ACL + +/* posix_acl.c */ +struct posix_acl *hfsplus_get_posix_acl(struct inode *inode, int type); +extern int hfsplus_posix_acl_chmod(struct inode *); +extern int hfsplus_init_posix_acl(struct inode *, struct inode *); + +#else /* CONFIG_HFSPLUS_FS_POSIX_ACL */ +#define hfsplus_get_posix_acl NULL + +static inline int hfsplus_posix_acl_chmod(struct inode *inode) +{ + return 0; +} + +static inline int hfsplus_init_posix_acl(struct inode *inode, struct inode *dir) +{ + return 0; +} +#endif /* CONFIG_HFSPLUS_FS_POSIX_ACL */ diff --git a/fs/hfsplus/dir.c b/fs/hfsplus/dir.c index d8ce4bd..4a4fea0 100644 --- a/fs/hfsplus/dir.c +++ b/fs/hfsplus/dir.c @@ -16,6 +16,7 @@ #include "hfsplus_fs.h" #include "hfsplus_raw.h" #include "xattr.h" +#include "acl.h" static inline void hfsplus_instantiate(struct dentry *dentry, struct inode *inode, u32 cnid) @@ -529,6 +530,9 @@ const struct inode_operations hfsplus_dir_inode_operations = { .getxattr = generic_getxattr, .listxattr = hfsplus_listxattr, .removexattr = hfsplus_removexattr, +#ifdef CONFIG_HFSPLUS_FS_POSIX_ACL + .get_acl = hfsplus_get_posix_acl, +#endif }; const struct file_operations hfsplus_dir_operations = { diff --git a/fs/hfsplus/hfsplus_fs.h b/fs/hfsplus/hfsplus_fs.h index ede7931..2b9cd01 100644 --- a/fs/hfsplus/hfsplus_fs.h +++ b/fs/hfsplus/hfsplus_fs.h @@ -30,6 +30,7 @@ #define DBG_EXTENT 0x00000020 #define DBG_BITMAP 0x00000040 #define DBG_ATTR_MOD 0x00000080 +#define DBG_ACL_MOD 0x00000100 #if 0 #define DBG_MASK (DBG_EXTENT|DBG_INODE|DBG_BNODE_MOD) diff --git a/fs/hfsplus/inode.c b/fs/hfsplus/inode.c index f833d35..4d2edae 100644 --- a/fs/hfsplus/inode.c +++ b/fs/hfsplus/inode.c @@ -19,6 +19,7 @@ #include "hfsplus_fs.h" #include "hfsplus_raw.h" #include "xattr.h" +#include "acl.h" static int hfsplus_readpage(struct file *file, struct page *page) { @@ -316,6 +317,13 @@ static int hfsplus_setattr(struct dentry *dentry, struct iattr *attr) setattr_copy(inode, attr); mark_inode_dirty(inode); + + if (attr->ia_valid & ATTR_MODE) { + error = hfsplus_posix_acl_chmod(inode); + if (unlikely(error)) + return error; + } + return 0; } @@ -383,6 +391,9 @@ static const struct inode_operations hfsplus_file_inode_operations = { .getxattr = generic_getxattr, .listxattr = hfsplus_listxattr, .removexattr = hfsplus_removexattr, +#ifdef CONFIG_HFSPLUS_FS_POSIX_ACL + .get_acl = hfsplus_get_posix_acl, +#endif }; static const struct file_operations hfsplus_file_operations = { diff --git a/fs/hfsplus/posix_acl.c b/fs/hfsplus/posix_acl.c new file mode 100644 index 0000000..b609cc1 --- /dev/null +++ b/fs/hfsplus/posix_acl.c @@ -0,0 +1,274 @@ +/* + * linux/fs/hfsplus/posix_acl.c + * + * Vyacheslav Dubeyko <slava@dubeyko.com> + * + * Handler for Posix Access Control Lists (ACLs) support. + */ + +#include "hfsplus_fs.h" +#include "xattr.h" +#include "acl.h" + +struct posix_acl *hfsplus_get_posix_acl(struct inode *inode, int type) +{ + struct posix_acl *acl; + char *xattr_name; + char *value = NULL; + ssize_t size; + + acl = get_cached_acl(inode, type); + if (acl != ACL_NOT_CACHED) + return acl; + + switch (type) { + case ACL_TYPE_ACCESS: + xattr_name = POSIX_ACL_XATTR_ACCESS; + break; + case ACL_TYPE_DEFAULT: + xattr_name = POSIX_ACL_XATTR_DEFAULT; + break; + default: + return ERR_PTR(-EINVAL); + } + + size = __hfsplus_getxattr(inode, xattr_name, NULL, 0); + + if (size > 0) { + value = (char *)hfsplus_alloc_attr_entry(); + if (unlikely(!value)) + return ERR_PTR(-ENOMEM); + size = __hfsplus_getxattr(inode, xattr_name, value, size); + } + + if (size > 0) + acl = posix_acl_from_xattr(&init_user_ns, value, size); + else if (size == -ENODATA) + acl = NULL; + else + acl = ERR_PTR(size); + + hfsplus_destroy_attr_entry((hfsplus_attr_entry *)value); + + if (!IS_ERR(acl)) + set_cached_acl(inode, type, acl); + + return acl; +} + +static int hfsplus_set_posix_acl(struct inode *inode, + int type, + struct posix_acl *acl) +{ + int err; + char *xattr_name; + size_t size = 0; + char *value = NULL; + + if (S_ISLNK(inode->i_mode)) + return -EOPNOTSUPP; + + switch (type) { + case ACL_TYPE_ACCESS: + xattr_name = POSIX_ACL_XATTR_ACCESS; + if (acl) { + err = posix_acl_equiv_mode(acl, &inode->i_mode); + if (err < 0) + return err; + } + err = 0; + break; + + case ACL_TYPE_DEFAULT: + xattr_name = POSIX_ACL_XATTR_DEFAULT; + if (!S_ISDIR(inode->i_mode)) + return acl ? -EACCES : 0; + break; + + default: + return -EINVAL; + } + + if (acl) { + size = posix_acl_xattr_size(acl->a_count); + if (unlikely(size > HFSPLUS_MAX_INLINE_DATA_SIZE)) + return -ENOMEM; + value = (char *)hfsplus_alloc_attr_entry(); + if (unlikely(!value)) + return -ENOMEM; + err = posix_acl_to_xattr(&init_user_ns, acl, value, size); + if (unlikely(err < 0)) + goto end_set_acl; + } + + err = __hfsplus_setxattr(inode, xattr_name, value, size, 0); + +end_set_acl: + hfsplus_destroy_attr_entry((hfsplus_attr_entry *)value); + + if (!err) + set_cached_acl(inode, type, acl); + + return err; +} + +int hfsplus_init_posix_acl(struct inode *inode, struct inode *dir) +{ + int err = 0; + struct posix_acl *acl = NULL; + + hfs_dbg(ACL_MOD, + "[%s]: ino %lu, dir->ino %lu\n", + __func__, inode->i_ino, dir->i_ino); + + if (S_ISLNK(inode->i_mode)) + return 0; + + acl = hfsplus_get_posix_acl(dir, ACL_TYPE_DEFAULT); + if (IS_ERR(acl)) + return PTR_ERR(acl); + + if (acl) { + if (S_ISDIR(inode->i_mode)) { + err = hfsplus_set_posix_acl(inode, + ACL_TYPE_DEFAULT, + acl); + if (unlikely(err)) + goto init_acl_cleanup; + } + + err = posix_acl_create(&acl, GFP_NOFS, &inode->i_mode); + if (unlikely(err < 0)) + return err; + + if (err > 0) + err = hfsplus_set_posix_acl(inode, + ACL_TYPE_ACCESS, + acl); + } else + inode->i_mode &= ~current_umask(); + +init_acl_cleanup: + posix_acl_release(acl); + return err; +} + +int hfsplus_posix_acl_chmod(struct inode *inode) +{ + int err; + struct posix_acl *acl; + + hfs_dbg(ACL_MOD, "[%s]: ino %lu\n", __func__, inode->i_ino); + + if (S_ISLNK(inode->i_mode)) + return -EOPNOTSUPP; + + acl = hfsplus_get_posix_acl(inode, ACL_TYPE_ACCESS); + if (IS_ERR(acl) || !acl) + return PTR_ERR(acl); + + err = posix_acl_chmod(&acl, GFP_KERNEL, inode->i_mode); + if (unlikely(err)) + return err; + + err = hfsplus_set_posix_acl(inode, ACL_TYPE_ACCESS, acl); + posix_acl_release(acl); + return err; +} + +static int hfsplus_xattr_get_posix_acl(struct dentry *dentry, + const char *name, + void *buffer, + size_t size, + int type) +{ + int err = 0; + struct posix_acl *acl; + + hfs_dbg(ACL_MOD, + "[%s]: ino %lu, buffer %p, size %zu, type %#x\n", + __func__, dentry->d_inode->i_ino, buffer, size, type); + + if (strcmp(name, "") != 0) + return -EINVAL; + + acl = hfsplus_get_posix_acl(dentry->d_inode, type); + if (IS_ERR(acl)) + return PTR_ERR(acl); + if (acl == NULL) + return -ENODATA; + + err = posix_acl_to_xattr(&init_user_ns, acl, buffer, size); + posix_acl_release(acl); + + return err; +} + +static int hfsplus_xattr_set_posix_acl(struct dentry *dentry, + const char *name, + const void *value, + size_t size, + int flags, + int type) +{ + int err = 0; + struct inode *inode = dentry->d_inode; + struct posix_acl *acl = NULL; + + hfs_dbg(ACL_MOD, + "[%s]: ino %lu, value %p, size %zu, flags %#x, type %#x\n", + __func__, inode->i_ino, value, size, flags, type); + + if (strcmp(name, "") != 0) + return -EINVAL; + + if (!inode_owner_or_capable(inode)) + return -EPERM; + + if (value) { + acl = posix_acl_from_xattr(&init_user_ns, value, size); + if (IS_ERR(acl)) + return PTR_ERR(acl); + else if (acl) { + err = posix_acl_valid(acl); + if (err) + goto end_xattr_set_acl; + } + } + + err = hfsplus_set_posix_acl(inode, type, acl); + +end_xattr_set_acl: + posix_acl_release(acl); + return err; +} + +static size_t hfsplus_xattr_list_posix_acl(struct dentry *dentry, + char *list, + size_t list_size, + const char *name, + size_t name_len, + int type) +{ + /* + * This method is not used. + * It is used hfsplus_listxattr() instead of generic_listxattr(). + */ + return -EOPNOTSUPP; +} + +const struct xattr_handler hfsplus_xattr_acl_access_handler = { + .prefix = POSIX_ACL_XATTR_ACCESS, + .flags = ACL_TYPE_ACCESS, + .list = hfsplus_xattr_list_posix_acl, + .get = hfsplus_xattr_get_posix_acl, + .set = hfsplus_xattr_set_posix_acl, +}; + +const struct xattr_handler hfsplus_xattr_acl_default_handler = { + .prefix = POSIX_ACL_XATTR_DEFAULT, + .flags = ACL_TYPE_DEFAULT, + .list = hfsplus_xattr_list_posix_acl, + .get = hfsplus_xattr_get_posix_acl, + .set = hfsplus_xattr_set_posix_acl, +}; diff --git a/fs/hfsplus/xattr.c b/fs/hfsplus/xattr.c index f663461..bd8471f 100644 --- a/fs/hfsplus/xattr.c +++ b/fs/hfsplus/xattr.c @@ -8,11 +8,16 @@ #include "hfsplus_fs.h" #include "xattr.h" +#include "acl.h" const struct xattr_handler *hfsplus_xattr_handlers[] = { &hfsplus_xattr_osx_handler, &hfsplus_xattr_user_handler, &hfsplus_xattr_trusted_handler, +#ifdef CONFIG_HFSPLUS_FS_POSIX_ACL + &hfsplus_xattr_acl_access_handler, + &hfsplus_xattr_acl_default_handler, +#endif &hfsplus_xattr_security_handler, NULL }; @@ -46,11 +51,58 @@ static inline int is_known_namespace(const char *name) return true; } +static int can_set_system_xattr(struct inode *inode, const char *name, + const void *value, size_t size) +{ +#ifdef CONFIG_HFSPLUS_FS_POSIX_ACL + struct posix_acl *acl; + int err; + + if (!inode_owner_or_capable(inode)) + return -EPERM; + + /* + * POSIX_ACL_XATTR_ACCESS is tied to i_mode + */ + if (strcmp(name, POSIX_ACL_XATTR_ACCESS) == 0) { + acl = posix_acl_from_xattr(&init_user_ns, value, size); + if (IS_ERR(acl)) + return PTR_ERR(acl); + if (acl) { + err = posix_acl_equiv_mode(acl, &inode->i_mode); + posix_acl_release(acl); + if (err < 0) + return err; + mark_inode_dirty(inode); + } + /* + * We're changing the ACL. Get rid of the cached one + */ + forget_cached_acl(inode, ACL_TYPE_ACCESS); + + return 0; + } else if (strcmp(name, POSIX_ACL_XATTR_DEFAULT) == 0) { + acl = posix_acl_from_xattr(&init_user_ns, value, size); + if (IS_ERR(acl)) + return PTR_ERR(acl); + posix_acl_release(acl); + + /* + * We're changing the default ACL. Get rid of the cached one + */ + forget_cached_acl(inode, ACL_TYPE_DEFAULT); + + return 0; + } +#endif /* CONFIG_HFSPLUS_FS_POSIX_ACL */ + return -EOPNOTSUPP; +} + static int can_set_xattr(struct inode *inode, const char *name, const void *value, size_t value_len) { if (!strncmp(name, XATTR_SYSTEM_PREFIX, XATTR_SYSTEM_PREFIX_LEN)) - return -EOPNOTSUPP; /* TODO: implement ACL support */ + return can_set_system_xattr(inode, name, value, value_len); if (!strncmp(name, XATTR_MAC_OSX_PREFIX, XATTR_MAC_OSX_PREFIX_LEN)) { /* @@ -253,11 +305,10 @@ static int copy_name(char *buffer, const char *xattr_name, int name_len) return len; } -static ssize_t hfsplus_getxattr_finder_info(struct dentry *dentry, +static ssize_t hfsplus_getxattr_finder_info(struct inode *inode, void *value, size_t size) { ssize_t res = 0; - struct inode *inode = dentry->d_inode; struct hfs_find_data fd; u16 entry_type; u16 folder_rec_len = sizeof(struct DInfo) + sizeof(struct DXInfo); @@ -304,10 +355,9 @@ end_getxattr_finder_info: return res; } -ssize_t hfsplus_getxattr(struct dentry *dentry, const char *name, +ssize_t __hfsplus_getxattr(struct inode *inode, const char *name, void *value, size_t size) { - struct inode *inode = dentry->d_inode; struct hfs_find_data fd; hfsplus_attr_entry *entry; __be32 xattr_record_type; @@ -333,7 +383,7 @@ ssize_t hfsplus_getxattr(struct dentry *dentry, const char *name, } if (!strcmp_xattr_finder_info(name)) - return hfsplus_getxattr_finder_info(dentry, value, size); + return hfsplus_getxattr_finder_info(inode, value, size); if (!HFSPLUS_SB(inode->i_sb)->attr_tree) return -EOPNOTSUPP; diff --git a/fs/hfsplus/xattr.h b/fs/hfsplus/xattr.h index 847b695..841b569 100644 --- a/fs/hfsplus/xattr.h +++ b/fs/hfsplus/xattr.h @@ -14,8 +14,8 @@ extern const struct xattr_handler hfsplus_xattr_osx_handler; extern const struct xattr_handler hfsplus_xattr_user_handler; extern const struct xattr_handler hfsplus_xattr_trusted_handler; -/*extern const struct xattr_handler hfsplus_xattr_acl_access_handler;*/ -/*extern const struct xattr_handler hfsplus_xattr_acl_default_handler;*/ +extern const struct xattr_handler hfsplus_xattr_acl_access_handler; +extern const struct xattr_handler hfsplus_xattr_acl_default_handler; extern const struct xattr_handler hfsplus_xattr_security_handler; extern const struct xattr_handler *hfsplus_xattr_handlers[]; @@ -29,9 +29,17 @@ static inline int hfsplus_setxattr(struct dentry *dentry, const char *name, return __hfsplus_setxattr(dentry->d_inode, name, value, size, flags); } -ssize_t hfsplus_getxattr(struct dentry *dentry, const char *name, +ssize_t __hfsplus_getxattr(struct inode *inode, const char *name, void *value, size_t size); +static inline ssize_t hfsplus_getxattr(struct dentry *dentry, + const char *name, + void *value, + size_t size) +{ + return __hfsplus_getxattr(dentry->d_inode, name, value, size); +} + ssize_t hfsplus_listxattr(struct dentry *dentry, char *buffer, size_t size); int hfsplus_removexattr(struct dentry *dentry, const char *name); @@ -39,22 +47,7 @@ int hfsplus_removexattr(struct dentry *dentry, const char *name); int hfsplus_init_security(struct inode *inode, struct inode *dir, const struct qstr *qstr); -static inline int hfsplus_init_acl(struct inode *inode, struct inode *dir) -{ - /*TODO: implement*/ - return 0; -} - -static inline int hfsplus_init_inode_security(struct inode *inode, - struct inode *dir, - const struct qstr *qstr) -{ - int err; - - err = hfsplus_init_acl(inode, dir); - if (!err) - err = hfsplus_init_security(inode, dir, qstr); - return err; -} +int hfsplus_init_inode_security(struct inode *inode, struct inode *dir, + const struct qstr *qstr); #endif diff --git a/fs/hfsplus/xattr_security.c b/fs/hfsplus/xattr_security.c index 83b842f..0072276 100644 --- a/fs/hfsplus/xattr_security.c +++ b/fs/hfsplus/xattr_security.c @@ -9,6 +9,7 @@ #include <linux/security.h> #include "hfsplus_fs.h" #include "xattr.h" +#include "acl.h" static int hfsplus_security_getxattr(struct dentry *dentry, const char *name, void *buffer, size_t size, int type) @@ -96,6 +97,18 @@ int hfsplus_init_security(struct inode *inode, struct inode *dir, &hfsplus_initxattrs, NULL); } +int hfsplus_init_inode_security(struct inode *inode, + struct inode *dir, + const struct qstr *qstr) +{ + int err; + + err = hfsplus_init_posix_acl(inode, dir); + if (!err) + err = hfsplus_init_security(inode, dir, qstr); + return err; +} + const struct xattr_handler hfsplus_xattr_security_handler = { .prefix = XATTR_SECURITY_PREFIX, .list = hfsplus_security_listxattr, diff --git a/fs/hostfs/hostfs_kern.c b/fs/hostfs/hostfs_kern.c index cddb052..2543728 100644 --- a/fs/hostfs/hostfs_kern.c +++ b/fs/hostfs/hostfs_kern.c @@ -361,6 +361,13 @@ retry: return 0; } +static int hostfs_file_release(struct inode *inode, struct file *file) +{ + filemap_write_and_wait(inode->i_mapping); + + return 0; +} + int hostfs_fsync(struct file *file, loff_t start, loff_t end, int datasync) { struct inode *inode = file->f_mapping->host; @@ -386,7 +393,7 @@ static const struct file_operations hostfs_file_fops = { .write = do_sync_write, .mmap = generic_file_mmap, .open = hostfs_file_open, - .release = NULL, + .release = hostfs_file_release, .fsync = hostfs_fsync, }; diff --git a/fs/internal.h b/fs/internal.h index 7c5f01c..2be46ea 100644 --- a/fs/internal.h +++ b/fs/internal.h @@ -45,6 +45,9 @@ extern void __init chrdev_init(void); * namei.c */ extern int __inode_permission(struct inode *, int); +extern int user_path_mountpoint_at(int, const char __user *, unsigned int, struct path *); +extern int vfs_path_lookup(struct dentry *, struct vfsmount *, + const char *, unsigned int, struct path *); /* * namespace.c @@ -126,6 +129,7 @@ extern int invalidate_inodes(struct super_block *, bool); * dcache.c */ extern struct dentry *__d_alloc(struct super_block *, const struct qstr *); +extern int d_set_mounted(struct dentry *dentry); /* * read_write.c @@ -494,50 +494,6 @@ static inline void unlock_rcu_walk(void) br_read_unlock(&vfsmount_lock); } -/* - * When we move over from the RCU domain to properly refcounted - * long-lived dentries, we need to check the sequence numbers - * we got before lookup very carefully. - * - * We cannot blindly increment a dentry refcount - even if it - * is not locked - if it is zero, because it may have gone - * through the final d_kill() logic already. - * - * So for a zero refcount, we need to get the spinlock (which is - * safe even for a dead dentry because the de-allocation is - * RCU-delayed), and check the sequence count under the lock. - * - * Once we have checked the sequence count, we know it is live, - * and since we hold the spinlock it cannot die from under us. - * - * In contrast, if the reference count wasn't zero, we can just - * increment the lockref without having to take the spinlock. - * Even if the sequence number ends up being stale, we haven't - * gone through the final dput() and killed the dentry yet. - */ -static inline int d_rcu_to_refcount(struct dentry *dentry, seqcount_t *validate, unsigned seq) -{ - int gotref; - - gotref = lockref_get_or_lock(&dentry->d_lockref); - - /* Does the sequence number still match? */ - if (read_seqcount_retry(validate, seq)) { - if (gotref) - dput(dentry); - else - spin_unlock(&dentry->d_lock); - return -ECHILD; - } - - /* Get the ref now, if we couldn't get it originally */ - if (!gotref) { - dentry->d_lockref.count++; - spin_unlock(&dentry->d_lock); - } - return 0; -} - /** * unlazy_walk - try to switch to ref-walk mode. * @nd: nameidata pathwalk data @@ -552,16 +508,29 @@ static int unlazy_walk(struct nameidata *nd, struct dentry *dentry) { struct fs_struct *fs = current->fs; struct dentry *parent = nd->path.dentry; - int want_root = 0; BUG_ON(!(nd->flags & LOOKUP_RCU)); - if (nd->root.mnt && !(nd->flags & LOOKUP_ROOT)) { - want_root = 1; - spin_lock(&fs->lock); - if (nd->root.mnt != fs->root.mnt || - nd->root.dentry != fs->root.dentry) - goto err_root; - } + + /* + * Get a reference to the parent first: we're + * going to make "path_put(nd->path)" valid in + * non-RCU context for "terminate_walk()". + * + * If this doesn't work, return immediately with + * RCU walking still active (and then we will do + * the RCU walk cleanup in terminate_walk()). + */ + if (!lockref_get_not_dead(&parent->d_lockref)) + return -ECHILD; + + /* + * After the mntget(), we terminate_walk() will do + * the right thing for non-RCU mode, and all our + * subsequent exit cases should unlock_rcu_walk() + * before returning. + */ + mntget(nd->path.mnt); + nd->flags &= ~LOOKUP_RCU; /* * For a negative lookup, the lookup sequence point is the parents @@ -575,30 +544,42 @@ static int unlazy_walk(struct nameidata *nd, struct dentry *dentry) * be valid if the child sequence number is still valid. */ if (!dentry) { - if (d_rcu_to_refcount(parent, &parent->d_seq, nd->seq) < 0) - goto err_root; + if (read_seqcount_retry(&parent->d_seq, nd->seq)) + goto out; BUG_ON(nd->inode != parent->d_inode); } else { - if (d_rcu_to_refcount(dentry, &dentry->d_seq, nd->seq) < 0) - goto err_root; - if (d_rcu_to_refcount(parent, &dentry->d_seq, nd->seq) < 0) - goto err_parent; + if (!lockref_get_not_dead(&dentry->d_lockref)) + goto out; + if (read_seqcount_retry(&dentry->d_seq, nd->seq)) + goto drop_dentry; } - if (want_root) { + + /* + * Sequence counts matched. Now make sure that the root is + * still valid and get it if required. + */ + if (nd->root.mnt && !(nd->flags & LOOKUP_ROOT)) { + spin_lock(&fs->lock); + if (nd->root.mnt != fs->root.mnt || nd->root.dentry != fs->root.dentry) + goto unlock_and_drop_dentry; path_get(&nd->root); spin_unlock(&fs->lock); } - mntget(nd->path.mnt); unlock_rcu_walk(); - nd->flags &= ~LOOKUP_RCU; return 0; -err_parent: +unlock_and_drop_dentry: + spin_unlock(&fs->lock); +drop_dentry: + unlock_rcu_walk(); dput(dentry); -err_root: - if (want_root) - spin_unlock(&fs->lock); + goto drop_root_mnt; +out: + unlock_rcu_walk(); +drop_root_mnt: + if (!(nd->flags & LOOKUP_ROOT)) + nd->root.mnt = NULL; return -ECHILD; } @@ -627,8 +608,13 @@ static int complete_walk(struct nameidata *nd) if (!(nd->flags & LOOKUP_ROOT)) nd->root.mnt = NULL; - if (d_rcu_to_refcount(dentry, &dentry->d_seq, nd->seq) < 0) { + if (unlikely(!lockref_get_not_dead(&dentry->d_lockref))) { + unlock_rcu_walk(); + return -ECHILD; + } + if (read_seqcount_retry(&dentry->d_seq, nd->seq)) { unlock_rcu_walk(); + dput(dentry); return -ECHILD; } mntget(nd->path.mnt); @@ -2223,7 +2209,7 @@ user_path_parent(int dfd, const char __user *path, struct nameidata *nd, } /** - * umount_lookup_last - look up last component for umount + * mountpoint_last - look up last component for umount * @nd: pathwalk nameidata - currently pointing at parent directory of "last" * @path: pointer to container for result * @@ -2250,25 +2236,28 @@ user_path_parent(int dfd, const char __user *path, struct nameidata *nd, * to the link, and nd->path will *not* be put. */ static int -umount_lookup_last(struct nameidata *nd, struct path *path) +mountpoint_last(struct nameidata *nd, struct path *path) { int error = 0; struct dentry *dentry; struct dentry *dir = nd->path.dentry; - if (unlikely(nd->flags & LOOKUP_RCU)) { - WARN_ON_ONCE(1); - error = -ECHILD; - goto error_check; + /* If we're in rcuwalk, drop out of it to handle last component */ + if (nd->flags & LOOKUP_RCU) { + if (unlazy_walk(nd, NULL)) { + error = -ECHILD; + goto out; + } } nd->flags &= ~LOOKUP_PARENT; if (unlikely(nd->last_type != LAST_NORM)) { error = handle_dots(nd, nd->last_type); - if (!error) - dentry = dget(nd->path.dentry); - goto error_check; + if (error) + goto out; + dentry = dget(nd->path.dentry); + goto done; } mutex_lock(&dir->d_inode->i_mutex); @@ -2282,44 +2271,43 @@ umount_lookup_last(struct nameidata *nd, struct path *path) dentry = d_alloc(dir, &nd->last); if (!dentry) { error = -ENOMEM; - } else { - dentry = lookup_real(dir->d_inode, dentry, nd->flags); - if (IS_ERR(dentry)) - error = PTR_ERR(dentry); + goto out; } + dentry = lookup_real(dir->d_inode, dentry, nd->flags); + error = PTR_ERR(dentry); + if (IS_ERR(dentry)) + goto out; } mutex_unlock(&dir->d_inode->i_mutex); -error_check: - if (!error) { - if (!dentry->d_inode) { - error = -ENOENT; - dput(dentry); - } else { - path->dentry = dentry; - path->mnt = mntget(nd->path.mnt); - if (should_follow_link(dentry->d_inode, - nd->flags & LOOKUP_FOLLOW)) - return 1; - follow_mount(path); - } +done: + if (!dentry->d_inode) { + error = -ENOENT; + dput(dentry); + goto out; } + path->dentry = dentry; + path->mnt = mntget(nd->path.mnt); + if (should_follow_link(dentry->d_inode, nd->flags & LOOKUP_FOLLOW)) + return 1; + follow_mount(path); + error = 0; +out: terminate_walk(nd); return error; } /** - * path_umountat - look up a path to be umounted + * path_mountpoint - look up a path to be umounted * @dfd: directory file descriptor to start walk from * @name: full pathname to walk * @flags: lookup flags - * @nd: pathwalk nameidata * * Look up the given name, but don't attempt to revalidate the last component. * Returns 0 and "path" will be valid on success; Retuns error otherwise. */ static int -path_umountat(int dfd, const char *name, struct path *path, unsigned int flags) +path_mountpoint(int dfd, const char *name, struct path *path, unsigned int flags) { struct file *base = NULL; struct nameidata nd; @@ -2334,16 +2322,7 @@ path_umountat(int dfd, const char *name, struct path *path, unsigned int flags) if (err) goto out; - /* If we're in rcuwalk, drop out of it to handle last component */ - if (nd.flags & LOOKUP_RCU) { - err = unlazy_walk(&nd, NULL); - if (err) { - terminate_walk(&nd); - goto out; - } - } - - err = umount_lookup_last(&nd, path); + err = mountpoint_last(&nd, path); while (err > 0) { void *cookie; struct path link = *path; @@ -2354,7 +2333,7 @@ path_umountat(int dfd, const char *name, struct path *path, unsigned int flags) err = follow_link(&link, &nd, &cookie); if (err) break; - err = umount_lookup_last(&nd, path); + err = mountpoint_last(&nd, path); put_link(&nd, &link, cookie); } out: @@ -2367,8 +2346,22 @@ out: return err; } +static int +filename_mountpoint(int dfd, struct filename *s, struct path *path, + unsigned int flags) +{ + int error = path_mountpoint(dfd, s->name, path, flags | LOOKUP_RCU); + if (unlikely(error == -ECHILD)) + error = path_mountpoint(dfd, s->name, path, flags); + if (unlikely(error == -ESTALE)) + error = path_mountpoint(dfd, s->name, path, flags | LOOKUP_REVAL); + if (likely(!error)) + audit_inode(s, path->dentry, 0); + return error; +} + /** - * user_path_umountat - lookup a path from userland in order to umount it + * user_path_mountpoint_at - lookup a path from userland in order to umount it * @dfd: directory file descriptor * @name: pathname from userland * @flags: lookup flags @@ -2382,28 +2375,27 @@ out: * Returns 0 and populates "path" on success. */ int -user_path_umountat(int dfd, const char __user *name, unsigned int flags, +user_path_mountpoint_at(int dfd, const char __user *name, unsigned int flags, struct path *path) { struct filename *s = getname(name); int error; - if (IS_ERR(s)) return PTR_ERR(s); - - error = path_umountat(dfd, s->name, path, flags | LOOKUP_RCU); - if (unlikely(error == -ECHILD)) - error = path_umountat(dfd, s->name, path, flags); - if (unlikely(error == -ESTALE)) - error = path_umountat(dfd, s->name, path, flags | LOOKUP_REVAL); - - if (likely(!error)) - audit_inode(s, path->dentry, 0); - + error = filename_mountpoint(dfd, s, path, flags); putname(s); return error; } +int +kern_path_mountpoint(int dfd, const char *name, struct path *path, + unsigned int flags) +{ + struct filename s = {.name = name}; + return filename_mountpoint(dfd, &s, path, flags); +} +EXPORT_SYMBOL(kern_path_mountpoint); + /* * It's inline, so penalty for filesystems that don't use sticky bit is * minimal. diff --git a/fs/namespace.c b/fs/namespace.c index ad8ea9b..da5c494 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -17,7 +17,7 @@ #include <linux/security.h> #include <linux/idr.h> #include <linux/acct.h> /* acct_auto_close_mnt */ -#include <linux/ramfs.h> /* init_rootfs */ +#include <linux/init.h> /* init_rootfs */ #include <linux/fs_struct.h> /* get_fs_root et.al. */ #include <linux/fsnotify.h> /* fsnotify_vfsmount_delete */ #include <linux/uaccess.h> @@ -611,6 +611,7 @@ static struct mountpoint *new_mountpoint(struct dentry *dentry) { struct list_head *chain = mountpoint_hashtable + hash(NULL, dentry); struct mountpoint *mp; + int ret; list_for_each_entry(mp, chain, m_hash) { if (mp->m_dentry == dentry) { @@ -626,14 +627,12 @@ static struct mountpoint *new_mountpoint(struct dentry *dentry) if (!mp) return ERR_PTR(-ENOMEM); - spin_lock(&dentry->d_lock); - if (d_unlinked(dentry)) { - spin_unlock(&dentry->d_lock); + ret = d_set_mounted(dentry); + if (ret) { kfree(mp); - return ERR_PTR(-ENOENT); + return ERR_PTR(ret); } - dentry->d_flags |= DCACHE_MOUNTED; - spin_unlock(&dentry->d_lock); + mp->m_dentry = dentry; mp->m_count = 1; list_add(&mp->m_hash, chain); @@ -831,6 +830,10 @@ static struct mount *clone_mnt(struct mount *old, struct dentry *root, if ((flag & CL_UNPRIVILEGED) && (mnt->mnt.mnt_flags & MNT_READONLY)) mnt->mnt.mnt_flags |= MNT_LOCK_READONLY; + /* Don't allow unprivileged users to reveal what is under a mount */ + if ((flag & CL_UNPRIVILEGED) && list_empty(&old->mnt_expire)) + mnt->mnt.mnt_flags |= MNT_LOCKED; + atomic_inc(&sb->s_active); mnt->mnt.mnt_sb = sb; mnt->mnt.mnt_root = dget(root); @@ -1318,7 +1321,7 @@ SYSCALL_DEFINE2(umount, char __user *, name, int, flags) if (!(flags & UMOUNT_NOFOLLOW)) lookup_flags |= LOOKUP_FOLLOW; - retval = user_path_umountat(AT_FDCWD, name, lookup_flags, &path); + retval = user_path_mountpoint_at(AT_FDCWD, name, lookup_flags, &path); if (retval) goto out; mnt = real_mount(path.mnt); @@ -1327,6 +1330,8 @@ SYSCALL_DEFINE2(umount, char __user *, name, int, flags) goto dput_and_out; if (!check_mnt(mnt)) goto dput_and_out; + if (mnt->mnt.mnt_flags & MNT_LOCKED) + goto dput_and_out; retval = do_umount(mnt, flags); dput_and_out: @@ -1349,14 +1354,11 @@ SYSCALL_DEFINE1(oldumount, char __user *, name) #endif -static bool mnt_ns_loop(struct path *path) +static bool is_mnt_ns_file(struct dentry *dentry) { - /* Could bind mounting the mount namespace inode cause a - * mount namespace loop? - */ - struct inode *inode = path->dentry->d_inode; + /* Is this a proxy for a mount namespace? */ + struct inode *inode = dentry->d_inode; struct proc_ns *ei; - struct mnt_namespace *mnt_ns; if (!proc_ns_inode(inode)) return false; @@ -1365,7 +1367,19 @@ static bool mnt_ns_loop(struct path *path) if (ei->ns_ops != &mntns_operations) return false; - mnt_ns = ei->ns; + return true; +} + +static bool mnt_ns_loop(struct dentry *dentry) +{ + /* Could bind mounting the mount namespace inode cause a + * mount namespace loop? + */ + struct mnt_namespace *mnt_ns; + if (!is_mnt_ns_file(dentry)) + return false; + + mnt_ns = get_proc_ns(dentry->d_inode)->ns; return current->nsproxy->mnt_ns->seq >= mnt_ns->seq; } @@ -1374,13 +1388,17 @@ struct mount *copy_tree(struct mount *mnt, struct dentry *dentry, { struct mount *res, *p, *q, *r, *parent; - if (!(flag & CL_COPY_ALL) && IS_MNT_UNBINDABLE(mnt)) + if (!(flag & CL_COPY_UNBINDABLE) && IS_MNT_UNBINDABLE(mnt)) + return ERR_PTR(-EINVAL); + + if (!(flag & CL_COPY_MNT_NS_FILE) && is_mnt_ns_file(dentry)) return ERR_PTR(-EINVAL); res = q = clone_mnt(mnt, dentry, flag); if (IS_ERR(q)) return q; + q->mnt.mnt_flags &= ~MNT_LOCKED; q->mnt_mountpoint = mnt->mnt_mountpoint; p = mnt; @@ -1390,7 +1408,13 @@ struct mount *copy_tree(struct mount *mnt, struct dentry *dentry, continue; for (s = r; s; s = next_mnt(s, r)) { - if (!(flag & CL_COPY_ALL) && IS_MNT_UNBINDABLE(s)) { + if (!(flag & CL_COPY_UNBINDABLE) && + IS_MNT_UNBINDABLE(s)) { + s = skip_mnt_tree(s); + continue; + } + if (!(flag & CL_COPY_MNT_NS_FILE) && + is_mnt_ns_file(s->mnt.mnt_root)) { s = skip_mnt_tree(s); continue; } @@ -1696,6 +1720,19 @@ static int do_change_type(struct path *path, int flag) return err; } +static bool has_locked_children(struct mount *mnt, struct dentry *dentry) +{ + struct mount *child; + list_for_each_entry(child, &mnt->mnt_mounts, mnt_child) { + if (!is_subdir(child->mnt_mountpoint, dentry)) + continue; + + if (child->mnt.mnt_flags & MNT_LOCKED) + return true; + } + return false; +} + /* * do loopback mount. */ @@ -1713,7 +1750,7 @@ static int do_loopback(struct path *path, const char *old_name, return err; err = -EINVAL; - if (mnt_ns_loop(&old_path)) + if (mnt_ns_loop(old_path.dentry)) goto out; mp = lock_mount(path); @@ -1731,8 +1768,11 @@ static int do_loopback(struct path *path, const char *old_name, if (!check_mnt(parent) || !check_mnt(old)) goto out2; + if (!recurse && has_locked_children(old, old_path.dentry)) + goto out2; + if (recurse) - mnt = copy_tree(old, old_path.dentry, 0); + mnt = copy_tree(old, old_path.dentry, CL_COPY_MNT_NS_FILE); else mnt = clone_mnt(old, old_path.dentry, 0); @@ -1741,6 +1781,8 @@ static int do_loopback(struct path *path, const char *old_name, goto out2; } + mnt->mnt.mnt_flags &= ~MNT_LOCKED; + err = graft_tree(mnt, parent, mp); if (err) { br_write_lock(&vfsmount_lock); @@ -1853,6 +1895,9 @@ static int do_move_mount(struct path *path, const char *old_name) if (!check_mnt(p) || !check_mnt(old)) goto out1; + if (old->mnt.mnt_flags & MNT_LOCKED) + goto out1; + err = -EINVAL; if (old_path.dentry != old_path.mnt->mnt_root) goto out1; @@ -2389,7 +2434,7 @@ static struct mnt_namespace *dup_mnt_ns(struct mnt_namespace *mnt_ns, namespace_lock(); /* First pass: copy the tree topology */ - copy_flags = CL_COPY_ALL | CL_EXPIRE; + copy_flags = CL_COPY_UNBINDABLE | CL_EXPIRE; if (user_ns != mnt_ns->user_ns) copy_flags |= CL_SHARED_TO_SLAVE | CL_UNPRIVILEGED; new = copy_tree(old, old->mnt.mnt_root, copy_flags); @@ -2424,6 +2469,10 @@ static struct mnt_namespace *dup_mnt_ns(struct mnt_namespace *mnt_ns, } p = next_mnt(p, old); q = next_mnt(q, new); + if (!q) + break; + while (p->mnt.mnt_root != q->mnt.mnt_root) + p = next_mnt(p, old); } namespace_unlock(); @@ -2630,6 +2679,8 @@ SYSCALL_DEFINE2(pivot_root, const char __user *, new_root, goto out4; if (!check_mnt(root_mnt) || !check_mnt(new_mnt)) goto out4; + if (new_mnt->mnt.mnt_flags & MNT_LOCKED) + goto out4; error = -ENOENT; if (d_unlinked(new.dentry)) goto out4; @@ -2653,6 +2704,10 @@ SYSCALL_DEFINE2(pivot_root, const char __user *, new_root, br_write_lock(&vfsmount_lock); detach_mnt(new_mnt, &parent_path); detach_mnt(root_mnt, &root_parent); + if (root_mnt->mnt.mnt_flags & MNT_LOCKED) { + new_mnt->mnt.mnt_flags |= MNT_LOCKED; + root_mnt->mnt.mnt_flags &= ~MNT_LOCKED; + } /* mount old root on put_old */ attach_mnt(root_mnt, old_mnt, old_mp); /* mount new_root on / */ @@ -2811,25 +2866,38 @@ bool current_chrooted(void) return chrooted; } -void update_mnt_policy(struct user_namespace *userns) +bool fs_fully_visible(struct file_system_type *type) { struct mnt_namespace *ns = current->nsproxy->mnt_ns; struct mount *mnt; + bool visible = false; - down_read(&namespace_sem); + if (unlikely(!ns)) + return false; + + namespace_lock(); list_for_each_entry(mnt, &ns->list, mnt_list) { - switch (mnt->mnt.mnt_sb->s_magic) { - case SYSFS_MAGIC: - userns->may_mount_sysfs = true; - break; - case PROC_SUPER_MAGIC: - userns->may_mount_proc = true; - break; + struct mount *child; + if (mnt->mnt.mnt_sb->s_type != type) + continue; + + /* This mount is not fully visible if there are any child mounts + * that cover anything except for empty directories. + */ + list_for_each_entry(child, &mnt->mnt_mounts, mnt_child) { + struct inode *inode = child->mnt_mountpoint->d_inode; + if (!S_ISDIR(inode->i_mode)) + goto next; + if (inode->i_nlink != 2) + goto next; } - if (userns->may_mount_sysfs && userns->may_mount_proc) - break; + visible = true; + goto found; + next: ; } - up_read(&namespace_sem); +found: + namespace_unlock(); + return visible; } static void *mntns_get(struct task_struct *task) @@ -2860,8 +2928,8 @@ static int mntns_install(struct nsproxy *nsproxy, void *ns) struct path root; if (!ns_capable(mnt_ns->user_ns, CAP_SYS_ADMIN) || - !nsown_capable(CAP_SYS_CHROOT) || - !nsown_capable(CAP_SYS_ADMIN)) + !ns_capable(current_user_ns(), CAP_SYS_CHROOT) || + !ns_capable(current_user_ns(), CAP_SYS_ADMIN)) return -EPERM; if (fs->users != 1) diff --git a/fs/nfs/Makefile b/fs/nfs/Makefile index e0bb048..03192a6 100644 --- a/fs/nfs/Makefile +++ b/fs/nfs/Makefile @@ -4,9 +4,10 @@ obj-$(CONFIG_NFS_FS) += nfs.o +CFLAGS_nfstrace.o += -I$(src) nfs-y := client.o dir.o file.o getroot.o inode.o super.o \ direct.o pagelist.o read.o symlink.o unlink.o \ - write.o namespace.o mount_clnt.o + write.o namespace.o mount_clnt.o nfstrace.o nfs-$(CONFIG_ROOT_NFS) += nfsroot.o nfs-$(CONFIG_SYSCTL) += sysctl.o nfs-$(CONFIG_NFS_FSCACHE) += fscache.o fscache-index.o @@ -19,12 +20,14 @@ nfsv3-y := nfs3super.o nfs3client.o nfs3proc.o nfs3xdr.o nfsv3-$(CONFIG_NFS_V3_ACL) += nfs3acl.o obj-$(CONFIG_NFS_V4) += nfsv4.o +CFLAGS_nfs4trace.o += -I$(src) nfsv4-y := nfs4proc.o nfs4xdr.o nfs4state.o nfs4renewd.o nfs4super.o nfs4file.o \ delegation.o idmap.o callback.o callback_xdr.o callback_proc.o \ - nfs4namespace.o nfs4getroot.o nfs4client.o dns_resolve.o + nfs4namespace.o nfs4getroot.o nfs4client.o nfs4session.o \ + dns_resolve.o nfs4trace.o nfsv4-$(CONFIG_NFS_USE_LEGACY_DNS) += cache_lib.o nfsv4-$(CONFIG_SYSCTL) += nfs4sysctl.o -nfsv4-$(CONFIG_NFS_V4_1) += nfs4session.o pnfs.o pnfs_dev.o +nfsv4-$(CONFIG_NFS_V4_1) += pnfs.o pnfs_dev.o obj-$(CONFIG_PNFS_FILE_LAYOUT) += nfs_layout_nfsv41_files.o nfs_layout_nfsv41_files-y := nfs4filelayout.o nfs4filelayoutdev.o diff --git a/fs/nfs/callback_proc.c b/fs/nfs/callback_proc.c index e6ebc4c..ae2e87b 100644 --- a/fs/nfs/callback_proc.c +++ b/fs/nfs/callback_proc.c @@ -15,6 +15,7 @@ #include "internal.h" #include "pnfs.h" #include "nfs4session.h" +#include "nfs4trace.h" #ifdef NFS_DEBUG #define NFSDBG_FACILITY NFSDBG_CALLBACK @@ -93,6 +94,7 @@ __be32 nfs4_callback_recall(struct cb_recallargs *args, void *dummy, default: res = htonl(NFS4ERR_RESOURCE); } + trace_nfs4_recall_delegation(inode, -ntohl(res)); iput(inode); out: dprintk("%s: exit with status = %d\n", __func__, ntohl(res)); @@ -301,14 +303,14 @@ validate_seqid(struct nfs4_slot_table *tbl, struct cb_sequenceargs * args) { struct nfs4_slot *slot; - dprintk("%s enter. slotid %d seqid %d\n", + dprintk("%s enter. slotid %u seqid %u\n", __func__, args->csa_slotid, args->csa_sequenceid); if (args->csa_slotid >= NFS41_BC_MAX_CALLBACKS) return htonl(NFS4ERR_BADSLOT); slot = tbl->slots + args->csa_slotid; - dprintk("%s slot table seqid: %d\n", __func__, slot->seq_nr); + dprintk("%s slot table seqid: %u\n", __func__, slot->seq_nr); /* Normal */ if (likely(args->csa_sequenceid == slot->seq_nr + 1)) { @@ -318,7 +320,7 @@ validate_seqid(struct nfs4_slot_table *tbl, struct cb_sequenceargs * args) /* Replay */ if (args->csa_sequenceid == slot->seq_nr) { - dprintk("%s seqid %d is a replay\n", + dprintk("%s seqid %u is a replay\n", __func__, args->csa_sequenceid); /* Signal process_op to set this error on next op */ if (args->csa_cachethis == 0) @@ -462,6 +464,7 @@ out: } else res->csr_status = status; + trace_nfs4_cb_sequence(args, res, status); dprintk("%s: exit with status = %d res->csr_status %d\n", __func__, ntohl(status), ntohl(res->csr_status)); return status; @@ -518,7 +521,7 @@ __be32 nfs4_callback_recallslot(struct cb_recallslotargs *args, void *dummy, if (!cps->clp) /* set in cb_sequence */ goto out; - dprintk_rcu("NFS: CB_RECALL_SLOT request from %s target highest slotid %d\n", + dprintk_rcu("NFS: CB_RECALL_SLOT request from %s target highest slotid %u\n", rpc_peeraddr2str(cps->clp->cl_rpcclient, RPC_DISPLAY_ADDR), args->crsa_target_highest_slotid); diff --git a/fs/nfs/client.c b/fs/nfs/client.c index 340b1ef..2dceee4 100644 --- a/fs/nfs/client.c +++ b/fs/nfs/client.c @@ -501,8 +501,7 @@ nfs_get_client(const struct nfs_client_initdata *cl_init, &nn->nfs_client_list); spin_unlock(&nn->nfs_client_lock); new->cl_flags = cl_init->init_flags; - return rpc_ops->init_client(new, timeparms, ip_addr, - authflavour); + return rpc_ops->init_client(new, timeparms, ip_addr); } spin_unlock(&nn->nfs_client_lock); @@ -694,13 +693,12 @@ EXPORT_SYMBOL_GPL(nfs_init_server_rpcclient); * @clp: nfs_client to initialise * @timeparms: timeout parameters for underlying RPC transport * @ip_addr: IP presentation address (not used) - * @authflavor: authentication flavor for underlying RPC transport * * Returns pointer to an NFS client, or an ERR_PTR value. */ struct nfs_client *nfs_init_client(struct nfs_client *clp, const struct rpc_timeout *timeparms, - const char *ip_addr, rpc_authflavor_t authflavour) + const char *ip_addr) { int error; diff --git a/fs/nfs/delegation.c b/fs/nfs/delegation.c index 7ec4814..ef792f2 100644 --- a/fs/nfs/delegation.c +++ b/fs/nfs/delegation.c @@ -20,6 +20,7 @@ #include "nfs4_fs.h" #include "delegation.h" #include "internal.h" +#include "nfs4trace.h" static void nfs_free_delegation(struct nfs_delegation *delegation) { @@ -160,6 +161,7 @@ void nfs_inode_reclaim_delegation(struct inode *inode, struct rpc_cred *cred, spin_unlock(&delegation->lock); put_rpccred(oldcred); rcu_read_unlock(); + trace_nfs4_reclaim_delegation(inode, res->delegation_type); } else { /* We appear to have raced with a delegation return. */ spin_unlock(&delegation->lock); @@ -344,6 +346,7 @@ int nfs_inode_set_delegation(struct inode *inode, struct rpc_cred *cred, struct spin_lock(&inode->i_lock); nfsi->cache_validity |= NFS_INO_REVAL_FORCED; spin_unlock(&inode->i_lock); + trace_nfs4_set_delegation(inode, res->delegation_type); out: spin_unlock(&clp->cl_lock); diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index e474ca2b..e79bc6c 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -43,6 +43,8 @@ #include "internal.h" #include "fscache.h" +#include "nfstrace.h" + /* #define NFS_DEBUG_VERBOSE 1 */ static int nfs_opendir(struct inode *, struct file *); @@ -1100,7 +1102,9 @@ static int nfs_lookup_revalidate(struct dentry *dentry, unsigned int flags) if (IS_ERR(label)) goto out_error; + trace_nfs_lookup_revalidate_enter(dir, dentry, flags); error = NFS_PROTO(dir)->lookup(dir, &dentry->d_name, fhandle, fattr, label); + trace_nfs_lookup_revalidate_exit(dir, dentry, flags, error); if (error) goto out_bad; if (nfs_compare_fh(NFS_FH(inode), fhandle)) @@ -1135,14 +1139,13 @@ out_zap_parent: if (inode && S_ISDIR(inode->i_mode)) { /* Purge readdir caches. */ nfs_zap_caches(inode); - /* If we have submounts, don't unhash ! */ - if (have_submounts(dentry)) - goto out_valid; if (dentry->d_flags & DCACHE_DISCONNECTED) goto out_valid; - shrink_dcache_parent(dentry); } - d_drop(dentry); + /* If we have submounts, don't unhash ! */ + if (check_submounts_and_drop(dentry) != 0) + goto out_valid; + dput(parent); dfprintk(LOOKUPCACHE, "NFS: %s(%s/%s) is invalid\n", __func__, dentry->d_parent->d_name.name, @@ -1313,6 +1316,7 @@ struct dentry *nfs_lookup(struct inode *dir, struct dentry * dentry, unsigned in parent = dentry->d_parent; /* Protect against concurrent sillydeletes */ + trace_nfs_lookup_enter(dir, dentry, flags); nfs_block_sillyrename(parent); error = NFS_PROTO(dir)->lookup(dir, &dentry->d_name, fhandle, fattr, label); if (error == -ENOENT) @@ -1339,6 +1343,7 @@ no_entry: nfs_set_verifier(dentry, nfs_save_change_attribute(dir)); out_unblock_sillyrename: nfs_unblock_sillyrename(parent); + trace_nfs_lookup_exit(dir, dentry, flags, error); nfs4_label_free(label); out: nfs_free_fattr(fattr); @@ -1393,7 +1398,6 @@ static int nfs_finish_open(struct nfs_open_context *ctx, nfs_file_set_open_context(file, ctx); out: - put_nfs_open_context(ctx); return err; } @@ -1405,6 +1409,7 @@ int nfs_atomic_open(struct inode *dir, struct dentry *dentry, struct dentry *res; struct iattr attr = { .ia_valid = ATTR_OPEN }; struct inode *inode; + unsigned int lookup_flags = 0; int err; /* Expect a negative dentry */ @@ -1413,6 +1418,10 @@ int nfs_atomic_open(struct inode *dir, struct dentry *dentry, dfprintk(VFS, "NFS: atomic_open(%s/%ld), %s\n", dir->i_sb->s_id, dir->i_ino, dentry->d_name.name); + err = nfs_check_flags(open_flags); + if (err) + return err; + /* NFS only supports OPEN on regular files */ if ((open_flags & O_DIRECTORY)) { if (!d_unhashed(dentry)) { @@ -1423,6 +1432,7 @@ int nfs_atomic_open(struct inode *dir, struct dentry *dentry, */ return -ENOENT; } + lookup_flags = LOOKUP_OPEN|LOOKUP_DIRECTORY; goto no_open; } @@ -1443,12 +1453,14 @@ int nfs_atomic_open(struct inode *dir, struct dentry *dentry, if (IS_ERR(ctx)) goto out; + trace_nfs_atomic_open_enter(dir, ctx, open_flags); nfs_block_sillyrename(dentry->d_parent); inode = NFS_PROTO(dir)->open_context(dir, ctx, open_flags, &attr); nfs_unblock_sillyrename(dentry->d_parent); if (IS_ERR(inode)) { - put_nfs_open_context(ctx); err = PTR_ERR(inode); + trace_nfs_atomic_open_exit(dir, ctx, open_flags, err); + put_nfs_open_context(ctx); switch (err) { case -ENOENT: d_drop(dentry); @@ -1469,11 +1481,13 @@ int nfs_atomic_open(struct inode *dir, struct dentry *dentry, } err = nfs_finish_open(ctx, ctx->dentry, file, open_flags, opened); + trace_nfs_atomic_open_exit(dir, ctx, open_flags, err); + put_nfs_open_context(ctx); out: return err; no_open: - res = nfs_lookup(dir, dentry, 0); + res = nfs_lookup(dir, dentry, lookup_flags); err = PTR_ERR(res); if (IS_ERR(res)) goto out; @@ -1597,7 +1611,9 @@ int nfs_create(struct inode *dir, struct dentry *dentry, attr.ia_mode = mode; attr.ia_valid = ATTR_MODE; + trace_nfs_create_enter(dir, dentry, open_flags); error = NFS_PROTO(dir)->create(dir, dentry, &attr, open_flags); + trace_nfs_create_exit(dir, dentry, open_flags, error); if (error != 0) goto out_err; return 0; @@ -1625,7 +1641,9 @@ nfs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, dev_t rdev) attr.ia_mode = mode; attr.ia_valid = ATTR_MODE; + trace_nfs_mknod_enter(dir, dentry); status = NFS_PROTO(dir)->mknod(dir, dentry, &attr, rdev); + trace_nfs_mknod_exit(dir, dentry, status); if (status != 0) goto out_err; return 0; @@ -1649,7 +1667,9 @@ int nfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) attr.ia_valid = ATTR_MODE; attr.ia_mode = mode | S_IFDIR; + trace_nfs_mkdir_enter(dir, dentry); error = NFS_PROTO(dir)->mkdir(dir, dentry, &attr); + trace_nfs_mkdir_exit(dir, dentry, error); if (error != 0) goto out_err; return 0; @@ -1672,12 +1692,21 @@ int nfs_rmdir(struct inode *dir, struct dentry *dentry) dfprintk(VFS, "NFS: rmdir(%s/%ld), %s\n", dir->i_sb->s_id, dir->i_ino, dentry->d_name.name); - error = NFS_PROTO(dir)->rmdir(dir, &dentry->d_name); - /* Ensure the VFS deletes this inode */ - if (error == 0 && dentry->d_inode != NULL) - clear_nlink(dentry->d_inode); - else if (error == -ENOENT) - nfs_dentry_handle_enoent(dentry); + trace_nfs_rmdir_enter(dir, dentry); + if (dentry->d_inode) { + nfs_wait_on_sillyrename(dentry); + error = NFS_PROTO(dir)->rmdir(dir, &dentry->d_name); + /* Ensure the VFS deletes this inode */ + switch (error) { + case 0: + clear_nlink(dentry->d_inode); + break; + case -ENOENT: + nfs_dentry_handle_enoent(dentry); + } + } else + error = NFS_PROTO(dir)->rmdir(dir, &dentry->d_name); + trace_nfs_rmdir_exit(dir, dentry, error); return error; } @@ -1705,6 +1734,7 @@ static int nfs_safe_remove(struct dentry *dentry) goto out; } + trace_nfs_remove_enter(dir, dentry); if (inode != NULL) { NFS_PROTO(inode)->return_delegation(inode); error = NFS_PROTO(dir)->remove(dir, &dentry->d_name); @@ -1714,6 +1744,7 @@ static int nfs_safe_remove(struct dentry *dentry) error = NFS_PROTO(dir)->remove(dir, &dentry->d_name); if (error == -ENOENT) nfs_dentry_handle_enoent(dentry); + trace_nfs_remove_exit(dir, dentry, error); out: return error; } @@ -1731,13 +1762,14 @@ int nfs_unlink(struct inode *dir, struct dentry *dentry) dfprintk(VFS, "NFS: unlink(%s/%ld, %s)\n", dir->i_sb->s_id, dir->i_ino, dentry->d_name.name); + trace_nfs_unlink_enter(dir, dentry); spin_lock(&dentry->d_lock); if (d_count(dentry) > 1) { spin_unlock(&dentry->d_lock); /* Start asynchronous writeout of the inode */ write_inode_now(dentry->d_inode, 0); error = nfs_sillyrename(dir, dentry); - return error; + goto out; } if (!d_unhashed(dentry)) { __d_drop(dentry); @@ -1749,6 +1781,8 @@ int nfs_unlink(struct inode *dir, struct dentry *dentry) nfs_set_verifier(dentry, nfs_save_change_attribute(dir)); } else if (need_rehash) d_rehash(dentry); +out: + trace_nfs_unlink_exit(dir, dentry, error); return error; } EXPORT_SYMBOL_GPL(nfs_unlink); @@ -1795,7 +1829,9 @@ int nfs_symlink(struct inode *dir, struct dentry *dentry, const char *symname) memset(kaddr + pathlen, 0, PAGE_SIZE - pathlen); kunmap_atomic(kaddr); + trace_nfs_symlink_enter(dir, dentry); error = NFS_PROTO(dir)->symlink(dir, dentry, page, pathlen, &attr); + trace_nfs_symlink_exit(dir, dentry, error); if (error != 0) { dfprintk(VFS, "NFS: symlink(%s/%ld, %s, %s) error %d\n", dir->i_sb->s_id, dir->i_ino, @@ -1830,6 +1866,7 @@ nfs_link(struct dentry *old_dentry, struct inode *dir, struct dentry *dentry) old_dentry->d_parent->d_name.name, old_dentry->d_name.name, dentry->d_parent->d_name.name, dentry->d_name.name); + trace_nfs_link_enter(inode, dir, dentry); NFS_PROTO(inode)->return_delegation(inode); d_drop(dentry); @@ -1838,6 +1875,7 @@ nfs_link(struct dentry *old_dentry, struct inode *dir, struct dentry *dentry) ihold(inode); d_add(dentry, inode); } + trace_nfs_link_exit(inode, dir, dentry, error); return error; } EXPORT_SYMBOL_GPL(nfs_link); @@ -1879,6 +1917,7 @@ int nfs_rename(struct inode *old_dir, struct dentry *old_dentry, new_dentry->d_parent->d_name.name, new_dentry->d_name.name, d_count(new_dentry)); + trace_nfs_rename_enter(old_dir, old_dentry, new_dir, new_dentry); /* * For non-directories, check whether the target is busy and if so, * make a copy of the dentry and then do a silly-rename. If the @@ -1925,6 +1964,8 @@ int nfs_rename(struct inode *old_dir, struct dentry *old_dentry, out: if (rehash) d_rehash(rehash); + trace_nfs_rename_exit(old_dir, old_dentry, + new_dir, new_dentry, error); if (!error) { if (new_inode != NULL) nfs_drop_nlink(new_inode); @@ -2174,9 +2215,11 @@ static int nfs_do_access(struct inode *inode, struct rpc_cred *cred, int mask) struct nfs_access_entry cache; int status; + trace_nfs_access_enter(inode); + status = nfs_access_get_cached(inode, cred, &cache); if (status == 0) - goto out; + goto out_cached; /* Be clever: ask server to check for all possible rights */ cache.mask = MAY_EXEC | MAY_WRITE | MAY_READ; @@ -2189,13 +2232,15 @@ static int nfs_do_access(struct inode *inode, struct rpc_cred *cred, int mask) if (!S_ISDIR(inode->i_mode)) set_bit(NFS_INO_STALE, &NFS_I(inode)->flags); } - return status; + goto out; } nfs_access_add_cache(inode, &cache); +out_cached: + if ((mask & ~cache.mask & (MAY_READ | MAY_WRITE | MAY_EXEC)) != 0) + status = -EACCES; out: - if ((mask & ~cache.mask & (MAY_READ | MAY_WRITE | MAY_EXEC)) == 0) - return 0; - return -EACCES; + trace_nfs_access_exit(inode, status); + return status; } static int nfs_open_permission_mask(int openflags) @@ -2241,11 +2286,6 @@ int nfs_permission(struct inode *inode, int mask) case S_IFLNK: goto out; case S_IFREG: - /* NFSv4 has atomic_open... */ - if (nfs_server_capable(inode, NFS_CAP_ATOMIC_OPEN) - && (mask & MAY_OPEN) - && !(mask & MAY_EXEC)) - goto out; break; case S_IFDIR: /* diff --git a/fs/nfs/file.c b/fs/nfs/file.c index 94e94bd..1e6bfdb 100644 --- a/fs/nfs/file.c +++ b/fs/nfs/file.c @@ -37,6 +37,8 @@ #include "iostat.h" #include "fscache.h" +#include "nfstrace.h" + #define NFSDBG_FACILITY NFSDBG_FILE static const struct vm_operations_struct nfs_file_vm_ops; @@ -294,6 +296,8 @@ nfs_file_fsync(struct file *file, loff_t start, loff_t end, int datasync) int ret; struct inode *inode = file_inode(file); + trace_nfs_fsync_enter(inode); + do { ret = filemap_write_and_wait_range(inode->i_mapping, start, end); if (ret != 0) @@ -310,6 +314,7 @@ nfs_file_fsync(struct file *file, loff_t start, loff_t end, int datasync) end = LLONG_MAX; } while (ret == -EAGAIN); + trace_nfs_fsync_exit(inode, ret); return ret; } @@ -406,6 +411,7 @@ static int nfs_write_end(struct file *file, struct address_space *mapping, struct page *page, void *fsdata) { unsigned offset = pos & (PAGE_CACHE_SIZE - 1); + struct nfs_open_context *ctx = nfs_file_open_context(file); int status; dfprintk(PAGECACHE, "NFS: write_end(%s/%s(%ld), %u@%lld)\n", @@ -441,6 +447,13 @@ static int nfs_write_end(struct file *file, struct address_space *mapping, if (status < 0) return status; NFS_I(mapping->host)->write_io += copied; + + if (nfs_ctx_key_to_expire(ctx)) { + status = nfs_wb_all(mapping->host); + if (status < 0) + return status; + } + return copied; } @@ -637,7 +650,8 @@ static int nfs_need_sync_write(struct file *filp, struct inode *inode) if (IS_SYNC(inode) || (filp->f_flags & O_DSYNC)) return 1; ctx = nfs_file_open_context(filp); - if (test_bit(NFS_CONTEXT_ERROR_WRITE, &ctx->flags)) + if (test_bit(NFS_CONTEXT_ERROR_WRITE, &ctx->flags) || + nfs_ctx_key_to_expire(ctx)) return 1; return 0; } @@ -651,6 +665,10 @@ ssize_t nfs_file_write(struct kiocb *iocb, const struct iovec *iov, ssize_t result; size_t count = iov_length(iov, nr_segs); + result = nfs_key_timeout_notify(iocb->ki_filp, inode); + if (result) + return result; + if (iocb->ki_filp->f_flags & O_DIRECT) return nfs_file_direct_write(iocb, iov, nr_segs, pos, true); diff --git a/fs/nfs/idmap.c b/fs/nfs/idmap.c index c2c4163..567983d 100644 --- a/fs/nfs/idmap.c +++ b/fs/nfs/idmap.c @@ -49,6 +49,7 @@ #include "internal.h" #include "netns.h" +#include "nfs4trace.h" #define NFS_UINT_MAXLEN 11 @@ -63,6 +64,7 @@ struct idmap_legacy_upcalldata { }; struct idmap { + struct rpc_pipe_dir_object idmap_pdo; struct rpc_pipe *idmap_pipe; struct idmap_legacy_upcalldata *idmap_upcall_data; struct mutex idmap_mutex; @@ -310,7 +312,7 @@ static ssize_t nfs_idmap_get_key(const char *name, size_t namelen, if (ret < 0) goto out_up; - payload = rcu_dereference(rkey->payload.data); + payload = rcu_dereference(rkey->payload.rcudata); if (IS_ERR_OR_NULL(payload)) { ret = PTR_ERR(payload); goto out_up; @@ -401,16 +403,23 @@ static struct key_type key_type_id_resolver_legacy = { .request_key = nfs_idmap_legacy_upcall, }; -static void __nfs_idmap_unregister(struct rpc_pipe *pipe) +static void nfs_idmap_pipe_destroy(struct dentry *dir, + struct rpc_pipe_dir_object *pdo) { - if (pipe->dentry) + struct idmap *idmap = pdo->pdo_data; + struct rpc_pipe *pipe = idmap->idmap_pipe; + + if (pipe->dentry) { rpc_unlink(pipe->dentry); + pipe->dentry = NULL; + } } -static int __nfs_idmap_register(struct dentry *dir, - struct idmap *idmap, - struct rpc_pipe *pipe) +static int nfs_idmap_pipe_create(struct dentry *dir, + struct rpc_pipe_dir_object *pdo) { + struct idmap *idmap = pdo->pdo_data; + struct rpc_pipe *pipe = idmap->idmap_pipe; struct dentry *dentry; dentry = rpc_mkpipe_dentry(dir, "idmap", idmap, pipe); @@ -420,36 +429,10 @@ static int __nfs_idmap_register(struct dentry *dir, return 0; } -static void nfs_idmap_unregister(struct nfs_client *clp, - struct rpc_pipe *pipe) -{ - struct net *net = clp->cl_net; - struct super_block *pipefs_sb; - - pipefs_sb = rpc_get_sb_net(net); - if (pipefs_sb) { - __nfs_idmap_unregister(pipe); - rpc_put_sb_net(net); - } -} - -static int nfs_idmap_register(struct nfs_client *clp, - struct idmap *idmap, - struct rpc_pipe *pipe) -{ - struct net *net = clp->cl_net; - struct super_block *pipefs_sb; - int err = 0; - - pipefs_sb = rpc_get_sb_net(net); - if (pipefs_sb) { - if (clp->cl_rpcclient->cl_dentry) - err = __nfs_idmap_register(clp->cl_rpcclient->cl_dentry, - idmap, pipe); - rpc_put_sb_net(net); - } - return err; -} +static const struct rpc_pipe_dir_object_ops nfs_idmap_pipe_dir_object_ops = { + .create = nfs_idmap_pipe_create, + .destroy = nfs_idmap_pipe_destroy, +}; int nfs_idmap_new(struct nfs_client *clp) @@ -462,23 +445,31 @@ nfs_idmap_new(struct nfs_client *clp) if (idmap == NULL) return -ENOMEM; + rpc_init_pipe_dir_object(&idmap->idmap_pdo, + &nfs_idmap_pipe_dir_object_ops, + idmap); + pipe = rpc_mkpipe_data(&idmap_upcall_ops, 0); if (IS_ERR(pipe)) { error = PTR_ERR(pipe); - kfree(idmap); - return error; - } - error = nfs_idmap_register(clp, idmap, pipe); - if (error) { - rpc_destroy_pipe_data(pipe); - kfree(idmap); - return error; + goto err; } idmap->idmap_pipe = pipe; mutex_init(&idmap->idmap_mutex); + error = rpc_add_pipe_dir_object(clp->cl_net, + &clp->cl_rpcclient->cl_pipedir_objects, + &idmap->idmap_pdo); + if (error) + goto err_destroy_pipe; + clp->cl_idmap = idmap; return 0; +err_destroy_pipe: + rpc_destroy_pipe_data(idmap->idmap_pipe); +err: + kfree(idmap); + return error; } void @@ -488,130 +479,26 @@ nfs_idmap_delete(struct nfs_client *clp) if (!idmap) return; - nfs_idmap_unregister(clp, idmap->idmap_pipe); - rpc_destroy_pipe_data(idmap->idmap_pipe); clp->cl_idmap = NULL; + rpc_remove_pipe_dir_object(clp->cl_net, + &clp->cl_rpcclient->cl_pipedir_objects, + &idmap->idmap_pdo); + rpc_destroy_pipe_data(idmap->idmap_pipe); kfree(idmap); } -static int __rpc_pipefs_event(struct nfs_client *clp, unsigned long event, - struct super_block *sb) -{ - int err = 0; - - switch (event) { - case RPC_PIPEFS_MOUNT: - err = __nfs_idmap_register(clp->cl_rpcclient->cl_dentry, - clp->cl_idmap, - clp->cl_idmap->idmap_pipe); - break; - case RPC_PIPEFS_UMOUNT: - if (clp->cl_idmap->idmap_pipe) { - struct dentry *parent; - - parent = clp->cl_idmap->idmap_pipe->dentry->d_parent; - __nfs_idmap_unregister(clp->cl_idmap->idmap_pipe); - /* - * Note: This is a dirty hack. SUNRPC hook has been - * called already but simple_rmdir() call for the - * directory returned with error because of idmap pipe - * inside. Thus now we have to remove this directory - * here. - */ - if (rpc_rmdir(parent)) - printk(KERN_ERR "NFS: %s: failed to remove " - "clnt dir!\n", __func__); - } - break; - default: - printk(KERN_ERR "NFS: %s: unknown event: %ld\n", __func__, - event); - return -ENOTSUPP; - } - return err; -} - -static struct nfs_client *nfs_get_client_for_event(struct net *net, int event) -{ - struct nfs_net *nn = net_generic(net, nfs_net_id); - struct dentry *cl_dentry; - struct nfs_client *clp; - int err; - -restart: - spin_lock(&nn->nfs_client_lock); - list_for_each_entry(clp, &nn->nfs_client_list, cl_share_link) { - /* Wait for initialisation to finish */ - if (clp->cl_cons_state == NFS_CS_INITING) { - atomic_inc(&clp->cl_count); - spin_unlock(&nn->nfs_client_lock); - err = nfs_wait_client_init_complete(clp); - nfs_put_client(clp); - if (err) - return NULL; - goto restart; - } - /* Skip nfs_clients that failed to initialise */ - if (clp->cl_cons_state < 0) - continue; - smp_rmb(); - if (clp->rpc_ops != &nfs_v4_clientops) - continue; - cl_dentry = clp->cl_idmap->idmap_pipe->dentry; - if (((event == RPC_PIPEFS_MOUNT) && cl_dentry) || - ((event == RPC_PIPEFS_UMOUNT) && !cl_dentry)) - continue; - atomic_inc(&clp->cl_count); - spin_unlock(&nn->nfs_client_lock); - return clp; - } - spin_unlock(&nn->nfs_client_lock); - return NULL; -} - -static int rpc_pipefs_event(struct notifier_block *nb, unsigned long event, - void *ptr) -{ - struct super_block *sb = ptr; - struct nfs_client *clp; - int error = 0; - - if (!try_module_get(THIS_MODULE)) - return 0; - - while ((clp = nfs_get_client_for_event(sb->s_fs_info, event))) { - error = __rpc_pipefs_event(clp, event, sb); - nfs_put_client(clp); - if (error) - break; - } - module_put(THIS_MODULE); - return error; -} - -#define PIPEFS_NFS_PRIO 1 - -static struct notifier_block nfs_idmap_block = { - .notifier_call = rpc_pipefs_event, - .priority = SUNRPC_PIPEFS_NFS_PRIO, -}; - int nfs_idmap_init(void) { int ret; ret = nfs_idmap_init_keyring(); if (ret != 0) goto out; - ret = rpc_pipefs_notifier_register(&nfs_idmap_block); - if (ret != 0) - nfs_idmap_quit_keyring(); out: return ret; } void nfs_idmap_quit(void) { - rpc_pipefs_notifier_unregister(&nfs_idmap_block); nfs_idmap_quit_keyring(); } @@ -849,6 +736,7 @@ int nfs_map_name_to_uid(const struct nfs_server *server, const char *name, size_ if (!uid_valid(*uid)) ret = -ERANGE; } + trace_nfs4_map_name_to_uid(name, namelen, id, ret); return ret; } @@ -865,6 +753,7 @@ int nfs_map_group_to_gid(const struct nfs_server *server, const char *name, size if (!gid_valid(*gid)) ret = -ERANGE; } + trace_nfs4_map_group_to_gid(name, namelen, id, ret); return ret; } @@ -879,6 +768,7 @@ int nfs_map_uid_to_name(const struct nfs_server *server, kuid_t uid, char *buf, ret = nfs_idmap_lookup_name(id, "user", buf, buflen, idmap); if (ret < 0) ret = nfs_map_numeric_to_string(id, buf, buflen); + trace_nfs4_map_uid_to_name(buf, ret, id, ret); return ret; } int nfs_map_gid_to_group(const struct nfs_server *server, kgid_t gid, char *buf, size_t buflen) @@ -892,5 +782,6 @@ int nfs_map_gid_to_group(const struct nfs_server *server, kgid_t gid, char *buf, ret = nfs_idmap_lookup_name(id, "group", buf, buflen, idmap); if (ret < 0) ret = nfs_map_numeric_to_string(id, buf, buflen); + trace_nfs4_map_gid_to_group(buf, ret, id, ret); return ret; } diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index 941246f..87e7976 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -38,7 +38,6 @@ #include <linux/slab.h> #include <linux/compat.h> #include <linux/freezer.h> -#include <linux/crc32.h> #include <asm/uaccess.h> @@ -52,6 +51,8 @@ #include "nfs.h" #include "netns.h" +#include "nfstrace.h" + #define NFSDBG_FACILITY NFSDBG_VFS #define NFS_64_BIT_INODE_NUMBERS_ENABLED 1 @@ -503,6 +504,8 @@ nfs_setattr(struct dentry *dentry, struct iattr *attr) if ((attr->ia_valid & ~(ATTR_FILE|ATTR_OPEN)) == 0) return 0; + trace_nfs_setattr_enter(inode); + /* Write all dirty data */ if (S_ISREG(inode->i_mode)) { nfs_inode_dio_wait(inode); @@ -522,6 +525,7 @@ nfs_setattr(struct dentry *dentry, struct iattr *attr) error = nfs_refresh_inode(inode, fattr); nfs_free_fattr(fattr); out: + trace_nfs_setattr_exit(inode, error); return error; } EXPORT_SYMBOL_GPL(nfs_setattr); @@ -591,6 +595,7 @@ int nfs_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat) int need_atime = NFS_I(inode)->cache_validity & NFS_INO_INVALID_ATIME; int err; + trace_nfs_getattr_enter(inode); /* Flush out writes to the server in order to update c/mtime. */ if (S_ISREG(inode->i_mode)) { nfs_inode_dio_wait(inode); @@ -621,6 +626,7 @@ int nfs_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat) stat->ino = nfs_compat_user_ino64(NFS_FILEID(inode)); } out: + trace_nfs_getattr_exit(inode, err); return err; } EXPORT_SYMBOL_GPL(nfs_getattr); @@ -875,6 +881,8 @@ __nfs_revalidate_inode(struct nfs_server *server, struct inode *inode) dfprintk(PAGECACHE, "NFS: revalidating (%s/%Ld)\n", inode->i_sb->s_id, (long long)NFS_FILEID(inode)); + trace_nfs_revalidate_inode_enter(inode); + if (is_bad_inode(inode)) goto out; if (NFS_STALE(inode)) @@ -925,6 +933,7 @@ err_out: nfs4_label_free(label); out: nfs_free_fattr(fattr); + trace_nfs_revalidate_inode_exit(inode, status); return status; } @@ -981,6 +990,7 @@ static int nfs_invalidate_mapping(struct inode *inode, struct address_space *map spin_unlock(&inode->i_lock); nfs_inc_stats(inode, NFSIOS_DATAINVALIDATE); nfs_fscache_wait_on_invalidate(inode); + dfprintk(PAGECACHE, "NFS: (%s/%Ld) data cache invalidated\n", inode->i_sb->s_id, (long long)NFS_FILEID(inode)); return 0; @@ -1014,8 +1024,12 @@ int nfs_revalidate_mapping(struct inode *inode, struct address_space *mapping) if (ret < 0) goto out; } - if (nfsi->cache_validity & NFS_INO_INVALID_DATA) + if (nfsi->cache_validity & NFS_INO_INVALID_DATA) { + trace_nfs_invalidate_mapping_enter(inode); ret = nfs_invalidate_mapping(inode, mapping); + trace_nfs_invalidate_mapping_exit(inode, ret); + } + out: return ret; } @@ -1195,7 +1209,7 @@ u32 _nfs_display_fhandle_hash(const struct nfs_fh *fh) { /* wireshark uses 32-bit AUTODIN crc and does a bitwise * not on the result */ - return ~crc32(0xFFFFFFFF, &fh->data[0], fh->size); + return nfs_fhandle_hash(fh); } /* @@ -1274,9 +1288,17 @@ static int nfs_inode_attrs_need_update(const struct inode *inode, const struct n static int nfs_refresh_inode_locked(struct inode *inode, struct nfs_fattr *fattr) { + int ret; + + trace_nfs_refresh_inode_enter(inode); + if (nfs_inode_attrs_need_update(inode, fattr)) - return nfs_update_inode(inode, fattr); - return nfs_check_inode_attributes(inode, fattr); + ret = nfs_update_inode(inode, fattr); + else + ret = nfs_check_inode_attributes(inode, fattr); + + trace_nfs_refresh_inode_exit(inode, ret); + return ret; } /** diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h index 3c8373f..d388302c 100644 --- a/fs/nfs/internal.h +++ b/fs/nfs/internal.h @@ -5,6 +5,7 @@ #include "nfs4_fs.h" #include <linux/mount.h> #include <linux/security.h> +#include <linux/crc32.h> #define NFS_MS_MASK (MS_RDONLY|MS_NOSUID|MS_NODEV|MS_NOEXEC|MS_SYNCHRONOUS) @@ -185,6 +186,8 @@ extern struct nfs_client *nfs4_set_ds_client(struct nfs_client* mds_clp, int ds_addrlen, int ds_proto, unsigned int ds_timeo, unsigned int ds_retrans); +extern struct rpc_clnt *nfs4_find_or_create_ds_client(struct nfs_client *, + struct inode *); #ifdef CONFIG_PROC_FS extern int __init nfs_fs_proc_init(void); extern void nfs_fs_proc_exit(void); @@ -267,7 +270,7 @@ extern struct rpc_procinfo nfs4_procedures[]; void nfs_close_context(struct nfs_open_context *ctx, int is_sync); extern struct nfs_client *nfs_init_client(struct nfs_client *clp, const struct rpc_timeout *timeparms, - const char *ip_addr, rpc_authflavor_t authflavour); + const char *ip_addr); /* dir.c */ extern int nfs_access_cache_shrinker(struct shrinker *shrink, @@ -355,7 +358,7 @@ extern struct dentry *nfs_get_root(struct super_block *, struct nfs_fh *, extern struct dentry *nfs4_get_root(struct super_block *, struct nfs_fh *, const char *); -extern int nfs4_get_rootfh(struct nfs_server *server, struct nfs_fh *mntfh); +extern int nfs4_get_rootfh(struct nfs_server *server, struct nfs_fh *mntfh, bool); #endif struct nfs_pgio_completion_ops; @@ -430,6 +433,8 @@ void nfs_request_remove_commit_list(struct nfs_page *req, void nfs_init_cinfo(struct nfs_commit_info *cinfo, struct inode *inode, struct nfs_direct_req *dreq); +int nfs_key_timeout_notify(struct file *filp, struct inode *inode); +bool nfs_ctx_key_to_expire(struct nfs_open_context *ctx); #ifdef CONFIG_MIGRATION extern int nfs_migrate_page(struct address_space *, @@ -451,8 +456,7 @@ extern ssize_t nfs_dreq_bytes_left(struct nfs_direct_req *dreq); extern void __nfs4_read_done_cb(struct nfs_read_data *); extern struct nfs_client *nfs4_init_client(struct nfs_client *clp, const struct rpc_timeout *timeparms, - const char *ip_addr, - rpc_authflavor_t authflavour); + const char *ip_addr); extern int nfs40_walk_client_list(struct nfs_client *clp, struct nfs_client **result, struct rpc_cred *cred); @@ -575,3 +579,22 @@ u64 nfs_timespec_to_change_attr(const struct timespec *ts) { return ((u64)ts->tv_sec << 30) + ts->tv_nsec; } + +#ifdef CONFIG_CRC32 +/** + * nfs_fhandle_hash - calculate the crc32 hash for the filehandle + * @fh - pointer to filehandle + * + * returns a crc32 hash for the filehandle that is compatible with + * the one displayed by "wireshark". + */ +static inline u32 nfs_fhandle_hash(const struct nfs_fh *fh) +{ + return ~crc32_le(0xFFFFFFFF, &fh->data[0], fh->size); +} +#else +static inline u32 nfs_fhandle_hash(const struct nfs_fh *fh) +{ + return 0; +} +#endif diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c index f5c84c3..90cb10d 100644 --- a/fs/nfs/nfs3proc.c +++ b/fs/nfs/nfs3proc.c @@ -336,8 +336,8 @@ nfs3_proc_create(struct inode *dir, struct dentry *dentry, struct iattr *sattr, data->arg.create.createmode = NFS3_CREATE_UNCHECKED; if (flags & O_EXCL) { data->arg.create.createmode = NFS3_CREATE_EXCLUSIVE; - data->arg.create.verifier[0] = jiffies; - data->arg.create.verifier[1] = current->pid; + data->arg.create.verifier[0] = cpu_to_be32(jiffies); + data->arg.create.verifier[1] = cpu_to_be32(current->pid); } sattr->ia_mode &= ~current_umask(); @@ -826,9 +826,10 @@ static void nfs3_proc_read_setup(struct nfs_read_data *data, struct rpc_message msg->rpc_proc = &nfs3_procedures[NFS3PROC_READ]; } -static void nfs3_proc_read_rpc_prepare(struct rpc_task *task, struct nfs_read_data *data) +static int nfs3_proc_read_rpc_prepare(struct rpc_task *task, struct nfs_read_data *data) { rpc_call_start(task); + return 0; } static int nfs3_write_done(struct rpc_task *task, struct nfs_write_data *data) @@ -847,9 +848,10 @@ static void nfs3_proc_write_setup(struct nfs_write_data *data, struct rpc_messag msg->rpc_proc = &nfs3_procedures[NFS3PROC_WRITE]; } -static void nfs3_proc_write_rpc_prepare(struct rpc_task *task, struct nfs_write_data *data) +static int nfs3_proc_write_rpc_prepare(struct rpc_task *task, struct nfs_write_data *data) { rpc_call_start(task); + return 0; } static void nfs3_proc_commit_rpc_prepare(struct rpc_task *task, struct nfs_commit_data *data) diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h index ee81e35..f520a11 100644 --- a/fs/nfs/nfs4_fs.h +++ b/fs/nfs/nfs4_fs.h @@ -38,17 +38,15 @@ struct nfs4_minor_version_ops { u32 minor_version; unsigned init_caps; - int (*call_sync)(struct rpc_clnt *clnt, - struct nfs_server *server, - struct rpc_message *msg, - struct nfs4_sequence_args *args, - struct nfs4_sequence_res *res); + int (*init_client)(struct nfs_client *); + void (*shutdown_client)(struct nfs_client *); bool (*match_stateid)(const nfs4_stateid *, const nfs4_stateid *); int (*find_root_sec)(struct nfs_server *, struct nfs_fh *, struct nfs_fsinfo *); int (*free_lock_state)(struct nfs_server *, struct nfs4_lock_state *); + const struct rpc_call_ops *call_sync_ops; const struct nfs4_state_recovery_ops *reboot_recovery_ops; const struct nfs4_state_recovery_ops *nograce_recovery_ops; const struct nfs4_state_maintenance_ops *state_renewal_ops; @@ -135,6 +133,7 @@ struct nfs4_lock_state { struct list_head ls_locks; /* Other lock stateids */ struct nfs4_state * ls_state; /* Pointer to open state */ #define NFS_LOCK_INITIALIZED 0 +#define NFS_LOCK_LOST 1 unsigned long ls_flags; struct nfs_seqid_counter ls_seqid; nfs4_stateid ls_stateid; @@ -193,7 +192,6 @@ struct nfs4_state_recovery_ops { int (*recover_open)(struct nfs4_state_owner *, struct nfs4_state *); int (*recover_lock)(struct nfs4_state *, struct file_lock *); int (*establish_clid)(struct nfs_client *, struct rpc_cred *); - struct rpc_cred * (*get_clid_cred)(struct nfs_client *); int (*reclaim_complete)(struct nfs_client *, struct rpc_cred *); int (*detect_trunking)(struct nfs_client *, struct nfs_client **, struct rpc_cred *); @@ -223,7 +221,7 @@ struct vfsmount *nfs4_submount(struct nfs_server *, struct dentry *, /* nfs4proc.c */ extern int nfs4_proc_setclientid(struct nfs_client *, u32, unsigned short, struct rpc_cred *, struct nfs4_setclientid_res *); extern int nfs4_proc_setclientid_confirm(struct nfs_client *, struct nfs4_setclientid_res *arg, struct rpc_cred *); -extern int nfs4_proc_get_rootfh(struct nfs_server *, struct nfs_fh *, struct nfs_fsinfo *); +extern int nfs4_proc_get_rootfh(struct nfs_server *, struct nfs_fh *, struct nfs_fsinfo *, bool); extern int nfs4_proc_bind_conn_to_session(struct nfs_client *, struct rpc_cred *cred); extern int nfs4_proc_exchange_id(struct nfs_client *clp, struct rpc_cred *cred); extern int nfs4_destroy_clientid(struct nfs_client *clp); @@ -248,9 +246,6 @@ static inline struct nfs4_session *nfs4_get_session(const struct nfs_server *ser return server->nfs_client->cl_session; } -extern int nfs4_setup_sequence(const struct nfs_server *server, - struct nfs4_sequence_args *args, struct nfs4_sequence_res *res, - struct rpc_task *task); extern int nfs41_setup_sequence(struct nfs4_session *session, struct nfs4_sequence_args *args, struct nfs4_sequence_res *res, struct rpc_task *task); @@ -273,18 +268,63 @@ is_ds_client(struct nfs_client *clp) { return clp->cl_exchange_flags & EXCHGID4_FLAG_USE_PNFS_DS; } -#else /* CONFIG_NFS_v4_1 */ -static inline struct nfs4_session *nfs4_get_session(const struct nfs_server *server) + +static inline bool +_nfs4_state_protect(struct nfs_client *clp, unsigned long sp4_mode, + struct rpc_clnt **clntp, struct rpc_message *msg) { - return NULL; + struct rpc_cred *newcred = NULL; + rpc_authflavor_t flavor; + + if (test_bit(sp4_mode, &clp->cl_sp4_flags)) { + spin_lock(&clp->cl_lock); + if (clp->cl_machine_cred != NULL) + newcred = get_rpccred(clp->cl_machine_cred); + spin_unlock(&clp->cl_lock); + if (msg->rpc_cred) + put_rpccred(msg->rpc_cred); + msg->rpc_cred = newcred; + + flavor = clp->cl_rpcclient->cl_auth->au_flavor; + WARN_ON(flavor != RPC_AUTH_GSS_KRB5I && + flavor != RPC_AUTH_GSS_KRB5P); + *clntp = clp->cl_rpcclient; + + return true; + } + return false; } -static inline int nfs4_setup_sequence(const struct nfs_server *server, - struct nfs4_sequence_args *args, struct nfs4_sequence_res *res, - struct rpc_task *task) +/* + * Function responsible for determining if an rpc_message should use the + * machine cred under SP4_MACH_CRED and if so switching the credential and + * authflavor (using the nfs_client's rpc_clnt which will be krb5i/p). + * Should be called before rpc_call_sync/rpc_call_async. + */ +static inline void +nfs4_state_protect(struct nfs_client *clp, unsigned long sp4_mode, + struct rpc_clnt **clntp, struct rpc_message *msg) { - rpc_call_start(task); - return 0; + _nfs4_state_protect(clp, sp4_mode, clntp, msg); +} + +/* + * Special wrapper to nfs4_state_protect for write. + * If WRITE can use machine cred but COMMIT cannot, make sure all writes + * that use machine cred use NFS_FILE_SYNC. + */ +static inline void +nfs4_state_protect_write(struct nfs_client *clp, struct rpc_clnt **clntp, + struct rpc_message *msg, struct nfs_write_data *wdata) +{ + if (_nfs4_state_protect(clp, NFS_SP4_MACH_CRED_WRITE, clntp, msg) && + !test_bit(NFS_SP4_MACH_CRED_COMMIT, &clp->cl_sp4_flags)) + wdata->args.stable = NFS_FILE_SYNC; +} +#else /* CONFIG_NFS_v4_1 */ +static inline struct nfs4_session *nfs4_get_session(const struct nfs_server *server) +{ + return NULL; } static inline bool @@ -298,6 +338,18 @@ is_ds_client(struct nfs_client *clp) { return false; } + +static inline void +nfs4_state_protect(struct nfs_client *clp, unsigned long sp4_flags, + struct rpc_clnt **clntp, struct rpc_message *msg) +{ +} + +static inline void +nfs4_state_protect_write(struct nfs_client *clp, struct rpc_clnt **clntp, + struct rpc_message *msg, struct nfs_write_data *wdata) +{ +} #endif /* CONFIG_NFS_V4_1 */ extern const struct nfs4_minor_version_ops *nfs_v4_minor_ops[]; @@ -308,6 +360,10 @@ extern const u32 nfs4_pathconf_bitmap[3]; extern const u32 nfs4_fsinfo_bitmap[3]; extern const u32 nfs4_fs_locations_bitmap[3]; +void nfs40_shutdown_client(struct nfs_client *); +void nfs41_shutdown_client(struct nfs_client *); +int nfs40_init_client(struct nfs_client *); +int nfs41_init_client(struct nfs_client *); void nfs4_free_client(struct nfs_client *); struct nfs_client *nfs4_alloc_client(const struct nfs_client_initdata *); @@ -319,7 +375,7 @@ extern void nfs4_kill_renewd(struct nfs_client *); extern void nfs4_renew_state(struct work_struct *); /* nfs4state.c */ -struct rpc_cred *nfs4_get_setclientid_cred(struct nfs_client *clp); +struct rpc_cred *nfs4_get_clid_cred(struct nfs_client *clp); struct rpc_cred *nfs4_get_machine_cred_locked(struct nfs_client *clp); struct rpc_cred *nfs4_get_renew_cred_locked(struct nfs_client *clp); int nfs4_discover_server_trunking(struct nfs_client *clp, @@ -327,7 +383,6 @@ int nfs4_discover_server_trunking(struct nfs_client *clp, int nfs40_discover_server_trunking(struct nfs_client *clp, struct nfs_client **, struct rpc_cred *); #if defined(CONFIG_NFS_V4_1) -struct rpc_cred *nfs4_get_exchange_id_cred(struct nfs_client *clp); int nfs41_discover_server_trunking(struct nfs_client *clp, struct nfs_client **, struct rpc_cred *); extern void nfs4_schedule_session_recovery(struct nfs4_session *, int); @@ -382,6 +437,7 @@ struct dentry *nfs4_try_mount(int, const char *, struct nfs_mount_info *, struct extern bool nfs4_disable_idmapping; extern unsigned short max_session_slots; extern unsigned short send_implementation_id; +extern bool recover_lost_locks; #define NFS4_CLIENT_ID_UNIQ_LEN (64) extern char nfs4_client_id_uniquifier[NFS4_CLIENT_ID_UNIQ_LEN]; @@ -429,6 +485,8 @@ static inline bool nfs4_valid_open_stateid(const struct nfs4_state *state) #define nfs4_close_state(a, b) do { } while (0) #define nfs4_close_sync(a, b) do { } while (0) +#define nfs4_state_protect(a, b, c, d) do { } while (0) +#define nfs4_state_protect_write(a, b, c, d) do { } while (0) #endif /* CONFIG_NFS_V4 */ #endif /* __LINUX_FS_NFS_NFS4_FS.H */ diff --git a/fs/nfs/nfs4client.c b/fs/nfs/nfs4client.c index 90dce91..a860ab5 100644 --- a/fs/nfs/nfs4client.c +++ b/fs/nfs/nfs4client.c @@ -41,19 +41,138 @@ static int nfs_get_cb_ident_idr(struct nfs_client *clp, int minorversion) } #ifdef CONFIG_NFS_V4_1 -static void nfs4_shutdown_session(struct nfs_client *clp) +/** + * Per auth flavor data server rpc clients + */ +struct nfs4_ds_server { + struct list_head list; /* ds_clp->cl_ds_clients */ + struct rpc_clnt *rpc_clnt; +}; + +/** + * Common lookup case for DS I/O + */ +static struct nfs4_ds_server * +nfs4_find_ds_client(struct nfs_client *ds_clp, rpc_authflavor_t flavor) +{ + struct nfs4_ds_server *dss; + + rcu_read_lock(); + list_for_each_entry_rcu(dss, &ds_clp->cl_ds_clients, list) { + if (dss->rpc_clnt->cl_auth->au_flavor != flavor) + continue; + goto out; + } + dss = NULL; +out: + rcu_read_unlock(); + return dss; +} + +static struct nfs4_ds_server * +nfs4_add_ds_client(struct nfs_client *ds_clp, rpc_authflavor_t flavor, + struct nfs4_ds_server *new) +{ + struct nfs4_ds_server *dss; + + spin_lock(&ds_clp->cl_lock); + list_for_each_entry(dss, &ds_clp->cl_ds_clients, list) { + if (dss->rpc_clnt->cl_auth->au_flavor != flavor) + continue; + goto out; + } + if (new) + list_add_rcu(&new->list, &ds_clp->cl_ds_clients); + dss = new; +out: + spin_unlock(&ds_clp->cl_lock); /* need some lock to protect list */ + return dss; +} + +static struct nfs4_ds_server * +nfs4_alloc_ds_server(struct nfs_client *ds_clp, rpc_authflavor_t flavor) +{ + struct nfs4_ds_server *dss; + + dss = kmalloc(sizeof(*dss), GFP_NOFS); + if (dss == NULL) + return ERR_PTR(-ENOMEM); + + dss->rpc_clnt = rpc_clone_client_set_auth(ds_clp->cl_rpcclient, flavor); + if (IS_ERR(dss->rpc_clnt)) { + int err = PTR_ERR(dss->rpc_clnt); + kfree (dss); + return ERR_PTR(err); + } + INIT_LIST_HEAD(&dss->list); + + return dss; +} + +static void +nfs4_free_ds_server(struct nfs4_ds_server *dss) +{ + rpc_release_client(dss->rpc_clnt); + kfree(dss); +} + +/** +* Find or create a DS rpc client with th MDS server rpc client auth flavor +* in the nfs_client cl_ds_clients list. +*/ +struct rpc_clnt * +nfs4_find_or_create_ds_client(struct nfs_client *ds_clp, struct inode *inode) +{ + struct nfs4_ds_server *dss, *new; + rpc_authflavor_t flavor = NFS_SERVER(inode)->client->cl_auth->au_flavor; + + dss = nfs4_find_ds_client(ds_clp, flavor); + if (dss != NULL) + goto out; + new = nfs4_alloc_ds_server(ds_clp, flavor); + if (IS_ERR(new)) + return ERR_CAST(new); + dss = nfs4_add_ds_client(ds_clp, flavor, new); + if (dss != new) + nfs4_free_ds_server(new); +out: + return dss->rpc_clnt; +} +EXPORT_SYMBOL_GPL(nfs4_find_or_create_ds_client); + +static void +nfs4_shutdown_ds_clients(struct nfs_client *clp) +{ + struct nfs4_ds_server *dss; + LIST_HEAD(shutdown_list); + + while (!list_empty(&clp->cl_ds_clients)) { + dss = list_entry(clp->cl_ds_clients.next, + struct nfs4_ds_server, list); + list_del(&dss->list); + rpc_shutdown_client(dss->rpc_clnt); + kfree (dss); + } +} + +void nfs41_shutdown_client(struct nfs_client *clp) { if (nfs4_has_session(clp)) { + nfs4_shutdown_ds_clients(clp); nfs4_destroy_session(clp->cl_session); nfs4_destroy_clientid(clp); } } -#else /* CONFIG_NFS_V4_1 */ -static void nfs4_shutdown_session(struct nfs_client *clp) +#endif /* CONFIG_NFS_V4_1 */ + +void nfs40_shutdown_client(struct nfs_client *clp) { + if (clp->cl_slot_tbl) { + nfs4_release_slot_table(clp->cl_slot_tbl); + kfree(clp->cl_slot_tbl); + } } -#endif /* CONFIG_NFS_V4_1 */ struct nfs_client *nfs4_alloc_client(const struct nfs_client_initdata *cl_init) { @@ -73,6 +192,7 @@ struct nfs_client *nfs4_alloc_client(const struct nfs_client_initdata *cl_init) spin_lock_init(&clp->cl_lock); INIT_DELAYED_WORK(&clp->cl_renewd, nfs4_renew_state); + INIT_LIST_HEAD(&clp->cl_ds_clients); rpc_init_wait_queue(&clp->cl_rpcwaitq, "NFS client"); clp->cl_state = 1 << NFS4CLNT_LEASE_EXPIRED; clp->cl_minorversion = cl_init->minorversion; @@ -97,7 +217,7 @@ static void nfs4_shutdown_client(struct nfs_client *clp) { if (__test_and_clear_bit(NFS_CS_RENEWD, &clp->cl_res_state)) nfs4_kill_renewd(clp); - nfs4_shutdown_session(clp); + clp->cl_mvops->shutdown_client(clp); nfs4_destroy_callback(clp); if (__test_and_clear_bit(NFS_CS_IDMAP, &clp->cl_res_state)) nfs_idmap_delete(clp); @@ -144,34 +264,77 @@ static int nfs4_init_callback(struct nfs_client *clp) return 0; } +/** + * nfs40_init_client - nfs_client initialization tasks for NFSv4.0 + * @clp - nfs_client to initialize + * + * Returns zero on success, or a negative errno if some error occurred. + */ +int nfs40_init_client(struct nfs_client *clp) +{ + struct nfs4_slot_table *tbl; + int ret; + + tbl = kzalloc(sizeof(*tbl), GFP_NOFS); + if (tbl == NULL) + return -ENOMEM; + + ret = nfs4_setup_slot_table(tbl, NFS4_MAX_SLOT_TABLE, + "NFSv4.0 transport Slot table"); + if (ret) { + kfree(tbl); + return ret; + } + + clp->cl_slot_tbl = tbl; + return 0; +} + +#if defined(CONFIG_NFS_V4_1) + +/** + * nfs41_init_client - nfs_client initialization tasks for NFSv4.1+ + * @clp - nfs_client to initialize + * + * Returns zero on success, or a negative errno if some error occurred. + */ +int nfs41_init_client(struct nfs_client *clp) +{ + struct nfs4_session *session = NULL; + + /* + * Create the session and mark it expired. + * When a SEQUENCE operation encounters the expired session + * it will do session recovery to initialize it. + */ + session = nfs4_alloc_session(clp); + if (!session) + return -ENOMEM; + + clp->cl_session = session; + + /* + * The create session reply races with the server back + * channel probe. Mark the client NFS_CS_SESSION_INITING + * so that the client back channel can find the + * nfs_client struct + */ + nfs_mark_client_ready(clp, NFS_CS_SESSION_INITING); + return 0; +} + +#endif /* CONFIG_NFS_V4_1 */ + /* * Initialize the minor version specific parts of an NFS4 client record */ static int nfs4_init_client_minor_version(struct nfs_client *clp) { -#if defined(CONFIG_NFS_V4_1) - if (clp->cl_mvops->minor_version) { - struct nfs4_session *session = NULL; - /* - * Create the session and mark it expired. - * When a SEQUENCE operation encounters the expired session - * it will do session recovery to initialize it. - */ - session = nfs4_alloc_session(clp); - if (!session) - return -ENOMEM; - - clp->cl_session = session; - /* - * The create session reply races with the server back - * channel probe. Mark the client NFS_CS_SESSION_INITING - * so that the client back channel can find the - * nfs_client struct - */ - nfs_mark_client_ready(clp, NFS_CS_SESSION_INITING); - } -#endif /* CONFIG_NFS_V4_1 */ + int ret; + ret = clp->cl_mvops->init_client(clp); + if (ret) + return ret; return nfs4_init_callback(clp); } @@ -187,8 +350,7 @@ static int nfs4_init_client_minor_version(struct nfs_client *clp) */ struct nfs_client *nfs4_init_client(struct nfs_client *clp, const struct rpc_timeout *timeparms, - const char *ip_addr, - rpc_authflavor_t authflavour) + const char *ip_addr) { char buf[INET6_ADDRSTRLEN + 1]; struct nfs_client *old; @@ -723,7 +885,7 @@ static void nfs4_session_set_rwsize(struct nfs_server *server) } static int nfs4_server_common_setup(struct nfs_server *server, - struct nfs_fh *mntfh) + struct nfs_fh *mntfh, bool auth_probe) { struct nfs_fattr *fattr; int error; @@ -755,7 +917,7 @@ static int nfs4_server_common_setup(struct nfs_server *server, /* Probe the root fh to retrieve its FSID and filehandle */ - error = nfs4_get_rootfh(server, mntfh); + error = nfs4_get_rootfh(server, mntfh, auth_probe); if (error < 0) goto out; @@ -787,6 +949,7 @@ out: static int nfs4_init_server(struct nfs_server *server, const struct nfs_parsed_mount_data *data) { + rpc_authflavor_t pseudoflavor = RPC_AUTH_UNIX; struct rpc_timeout timeparms; int error; @@ -799,13 +962,16 @@ static int nfs4_init_server(struct nfs_server *server, server->flags = data->flags; server->options = data->options; + if (data->auth_flavor_len >= 1) + pseudoflavor = data->auth_flavors[0]; + /* Get a client record */ error = nfs4_set_client(server, data->nfs_server.hostname, (const struct sockaddr *)&data->nfs_server.address, data->nfs_server.addrlen, data->client_address, - data->auth_flavors[0], + pseudoflavor, data->nfs_server.protocol, &timeparms, data->minorversion, @@ -825,7 +991,7 @@ static int nfs4_init_server(struct nfs_server *server, server->port = data->nfs_server.port; - error = nfs_init_server_rpcclient(server, &timeparms, data->auth_flavors[0]); + error = nfs_init_server_rpcclient(server, &timeparms, pseudoflavor); error: /* Done */ @@ -843,6 +1009,7 @@ struct nfs_server *nfs4_create_server(struct nfs_mount_info *mount_info, struct nfs_subversion *nfs_mod) { struct nfs_server *server; + bool auth_probe; int error; dprintk("--> nfs4_create_server()\n"); @@ -851,12 +1018,14 @@ struct nfs_server *nfs4_create_server(struct nfs_mount_info *mount_info, if (!server) return ERR_PTR(-ENOMEM); + auth_probe = mount_info->parsed->auth_flavor_len < 1; + /* set up the general RPC client */ error = nfs4_init_server(server, mount_info->parsed); if (error < 0) goto error; - error = nfs4_server_common_setup(server, mount_info->mntfh); + error = nfs4_server_common_setup(server, mount_info->mntfh, auth_probe); if (error < 0) goto error; @@ -909,7 +1078,8 @@ struct nfs_server *nfs4_create_referral_server(struct nfs_clone_mount *data, if (error < 0) goto error; - error = nfs4_server_common_setup(server, mntfh); + error = nfs4_server_common_setup(server, mntfh, + !(parent_server->flags & NFS_MOUNT_SECFLAVOUR)); if (error < 0) goto error; diff --git a/fs/nfs/nfs4filelayout.c b/fs/nfs/nfs4filelayout.c index 17ed87e..b86464b 100644 --- a/fs/nfs/nfs4filelayout.c +++ b/fs/nfs/nfs4filelayout.c @@ -39,6 +39,7 @@ #include "internal.h" #include "delegation.h" #include "nfs4filelayout.h" +#include "nfs4trace.h" #define NFSDBG_FACILITY NFSDBG_PNFS_LD @@ -247,6 +248,7 @@ static int filelayout_read_done_cb(struct rpc_task *task, struct nfs_pgio_header *hdr = data->header; int err; + trace_nfs4_pnfs_read(data, task->tk_status); err = filelayout_async_handle_error(task, data->args.context->state, data->ds_clp, hdr->lseg); @@ -363,6 +365,7 @@ static int filelayout_write_done_cb(struct rpc_task *task, struct nfs_pgio_header *hdr = data->header; int err; + trace_nfs4_pnfs_write(data, task->tk_status); err = filelayout_async_handle_error(task, data->args.context->state, data->ds_clp, hdr->lseg); @@ -395,6 +398,7 @@ static int filelayout_commit_done_cb(struct rpc_task *task, { int err; + trace_nfs4_pnfs_commit_ds(data, task->tk_status); err = filelayout_async_handle_error(task, NULL, data->ds_clp, data->lseg); @@ -524,6 +528,7 @@ filelayout_read_pagelist(struct nfs_read_data *data) struct nfs_pgio_header *hdr = data->header; struct pnfs_layout_segment *lseg = hdr->lseg; struct nfs4_pnfs_ds *ds; + struct rpc_clnt *ds_clnt; loff_t offset = data->args.offset; u32 j, idx; struct nfs_fh *fh; @@ -538,6 +543,11 @@ filelayout_read_pagelist(struct nfs_read_data *data) ds = nfs4_fl_prepare_ds(lseg, idx); if (!ds) return PNFS_NOT_ATTEMPTED; + + ds_clnt = nfs4_find_or_create_ds_client(ds->ds_clp, hdr->inode); + if (IS_ERR(ds_clnt)) + return PNFS_NOT_ATTEMPTED; + dprintk("%s USE DS: %s cl_count %d\n", __func__, ds->ds_remotestr, atomic_read(&ds->ds_clp->cl_count)); @@ -552,7 +562,7 @@ filelayout_read_pagelist(struct nfs_read_data *data) data->mds_offset = offset; /* Perform an asynchronous read to ds */ - nfs_initiate_read(ds->ds_clp->cl_rpcclient, data, + nfs_initiate_read(ds_clnt, data, &filelayout_read_call_ops, RPC_TASK_SOFTCONN); return PNFS_ATTEMPTED; } @@ -564,6 +574,7 @@ filelayout_write_pagelist(struct nfs_write_data *data, int sync) struct nfs_pgio_header *hdr = data->header; struct pnfs_layout_segment *lseg = hdr->lseg; struct nfs4_pnfs_ds *ds; + struct rpc_clnt *ds_clnt; loff_t offset = data->args.offset; u32 j, idx; struct nfs_fh *fh; @@ -574,6 +585,11 @@ filelayout_write_pagelist(struct nfs_write_data *data, int sync) ds = nfs4_fl_prepare_ds(lseg, idx); if (!ds) return PNFS_NOT_ATTEMPTED; + + ds_clnt = nfs4_find_or_create_ds_client(ds->ds_clp, hdr->inode); + if (IS_ERR(ds_clnt)) + return PNFS_NOT_ATTEMPTED; + dprintk("%s ino %lu sync %d req %Zu@%llu DS: %s cl_count %d\n", __func__, hdr->inode->i_ino, sync, (size_t) data->args.count, offset, ds->ds_remotestr, atomic_read(&ds->ds_clp->cl_count)); @@ -591,7 +607,7 @@ filelayout_write_pagelist(struct nfs_write_data *data, int sync) data->args.offset = filelayout_get_dserver_offset(lseg, offset); /* Perform an asynchronous write */ - nfs_initiate_write(ds->ds_clp->cl_rpcclient, data, + nfs_initiate_write(ds_clnt, data, &filelayout_write_call_ops, sync, RPC_TASK_SOFTCONN); return PNFS_ATTEMPTED; @@ -1101,16 +1117,19 @@ static int filelayout_initiate_commit(struct nfs_commit_data *data, int how) { struct pnfs_layout_segment *lseg = data->lseg; struct nfs4_pnfs_ds *ds; + struct rpc_clnt *ds_clnt; u32 idx; struct nfs_fh *fh; idx = calc_ds_index_from_commit(lseg, data->ds_commit_index); ds = nfs4_fl_prepare_ds(lseg, idx); - if (!ds) { - prepare_to_resend_writes(data); - filelayout_commit_release(data); - return -EAGAIN; - } + if (!ds) + goto out_err; + + ds_clnt = nfs4_find_or_create_ds_client(ds->ds_clp, data->inode); + if (IS_ERR(ds_clnt)) + goto out_err; + dprintk("%s ino %lu, how %d cl_count %d\n", __func__, data->inode->i_ino, how, atomic_read(&ds->ds_clp->cl_count)); data->commit_done_cb = filelayout_commit_done_cb; @@ -1119,9 +1138,13 @@ static int filelayout_initiate_commit(struct nfs_commit_data *data, int how) fh = select_ds_fh_from_commit(lseg, data->ds_commit_index); if (fh) data->args.fh = fh; - return nfs_initiate_commit(ds->ds_clp->cl_rpcclient, data, + return nfs_initiate_commit(ds_clnt, data, &filelayout_commit_call_ops, how, RPC_TASK_SOFTCONN); +out_err: + prepare_to_resend_writes(data); + filelayout_commit_release(data); + return -EAGAIN; } static int diff --git a/fs/nfs/nfs4getroot.c b/fs/nfs/nfs4getroot.c index 549462e..c0b3a16 100644 --- a/fs/nfs/nfs4getroot.c +++ b/fs/nfs/nfs4getroot.c @@ -9,7 +9,7 @@ #define NFSDBG_FACILITY NFSDBG_CLIENT -int nfs4_get_rootfh(struct nfs_server *server, struct nfs_fh *mntfh) +int nfs4_get_rootfh(struct nfs_server *server, struct nfs_fh *mntfh, bool auth_probe) { struct nfs_fsinfo fsinfo; int ret = -ENOMEM; @@ -21,7 +21,7 @@ int nfs4_get_rootfh(struct nfs_server *server, struct nfs_fh *mntfh) goto out; /* Start by getting the root filehandle from the server */ - ret = nfs4_proc_get_rootfh(server, mntfh, &fsinfo); + ret = nfs4_proc_get_rootfh(server, mntfh, &fsinfo, auth_probe); if (ret < 0) { dprintk("nfs4_get_rootfh: getroot error = %d\n", -ret); goto out; diff --git a/fs/nfs/nfs4namespace.c b/fs/nfs/nfs4namespace.c index cdb0b41..2288cd3 100644 --- a/fs/nfs/nfs4namespace.c +++ b/fs/nfs/nfs4namespace.c @@ -11,6 +11,7 @@ #include <linux/mount.h> #include <linux/namei.h> #include <linux/nfs_fs.h> +#include <linux/nfs_mount.h> #include <linux/slab.h> #include <linux/string.h> #include <linux/sunrpc/clnt.h> @@ -369,21 +370,33 @@ out: struct vfsmount *nfs4_submount(struct nfs_server *server, struct dentry *dentry, struct nfs_fh *fh, struct nfs_fattr *fattr) { + rpc_authflavor_t flavor = server->client->cl_auth->au_flavor; struct dentry *parent = dget_parent(dentry); + struct inode *dir = parent->d_inode; + struct qstr *name = &dentry->d_name; struct rpc_clnt *client; struct vfsmount *mnt; /* Look it up again to get its attributes and sec flavor */ - client = nfs4_proc_lookup_mountpoint(parent->d_inode, &dentry->d_name, fh, fattr); + client = nfs4_proc_lookup_mountpoint(dir, name, fh, fattr); dput(parent); if (IS_ERR(client)) return ERR_CAST(client); - if (fattr->valid & NFS_ATTR_FATTR_V4_REFERRAL) + if (fattr->valid & NFS_ATTR_FATTR_V4_REFERRAL) { mnt = nfs_do_refmount(client, dentry); - else - mnt = nfs_do_submount(dentry, fh, fattr, client->cl_auth->au_flavor); + goto out; + } + if (client->cl_auth->au_flavor != flavor) + flavor = client->cl_auth->au_flavor; + else if (!(server->flags & NFS_MOUNT_SECFLAVOUR)) { + rpc_authflavor_t new = nfs4_negotiate_security(dir, name); + if ((int)new >= 0) + flavor = new; + } + mnt = nfs_do_submount(dentry, fh, fattr, flavor); +out: rpc_shutdown_client(client); return mnt; } diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 108a774..39b6cf2 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -66,6 +66,8 @@ #include "nfs4session.h" #include "fscache.h" +#include "nfs4trace.h" + #define NFSDBG_FACILITY NFSDBG_PROC #define NFS4_POLL_RETRY_MIN (HZ/10) @@ -150,6 +152,7 @@ static int nfs4_map_errors(int err) case -NFS4ERR_RECALLCONFLICT: return -EREMOTEIO; case -NFS4ERR_WRONGSEC: + case -NFS4ERR_WRONG_CRED: return -EPERM; case -NFS4ERR_BADOWNER: case -NFS4ERR_BADNAME: @@ -433,6 +436,20 @@ wait_on_recovery: return ret; } +/* + * Return 'true' if 'clp' is using an rpc_client that is integrity protected + * or 'false' otherwise. + */ +static bool _nfs4_is_integrity_protected(struct nfs_client *clp) +{ + rpc_authflavor_t flavor = clp->cl_rpcclient->cl_auth->au_flavor; + + if (flavor == RPC_AUTH_GSS_KRB5I || + flavor == RPC_AUTH_GSS_KRB5P) + return true; + + return false; +} static void do_renew_lease(struct nfs_client *clp, unsigned long timestamp) { @@ -447,6 +464,88 @@ static void renew_lease(const struct nfs_server *server, unsigned long timestamp do_renew_lease(server->nfs_client, timestamp); } +struct nfs4_call_sync_data { + const struct nfs_server *seq_server; + struct nfs4_sequence_args *seq_args; + struct nfs4_sequence_res *seq_res; +}; + +static void nfs4_init_sequence(struct nfs4_sequence_args *args, + struct nfs4_sequence_res *res, int cache_reply) +{ + args->sa_slot = NULL; + args->sa_cache_this = cache_reply; + args->sa_privileged = 0; + + res->sr_slot = NULL; +} + +static void nfs4_set_sequence_privileged(struct nfs4_sequence_args *args) +{ + args->sa_privileged = 1; +} + +static int nfs40_setup_sequence(const struct nfs_server *server, + struct nfs4_sequence_args *args, + struct nfs4_sequence_res *res, + struct rpc_task *task) +{ + struct nfs4_slot_table *tbl = server->nfs_client->cl_slot_tbl; + struct nfs4_slot *slot; + + /* slot already allocated? */ + if (res->sr_slot != NULL) + goto out_start; + + spin_lock(&tbl->slot_tbl_lock); + if (nfs4_slot_tbl_draining(tbl) && !args->sa_privileged) + goto out_sleep; + + slot = nfs4_alloc_slot(tbl); + if (IS_ERR(slot)) { + if (slot == ERR_PTR(-ENOMEM)) + task->tk_timeout = HZ >> 2; + goto out_sleep; + } + spin_unlock(&tbl->slot_tbl_lock); + + args->sa_slot = slot; + res->sr_slot = slot; + +out_start: + rpc_call_start(task); + return 0; + +out_sleep: + if (args->sa_privileged) + rpc_sleep_on_priority(&tbl->slot_tbl_waitq, task, + NULL, RPC_PRIORITY_PRIVILEGED); + else + rpc_sleep_on(&tbl->slot_tbl_waitq, task, NULL); + spin_unlock(&tbl->slot_tbl_lock); + return -EAGAIN; +} + +static int nfs40_sequence_done(struct rpc_task *task, + struct nfs4_sequence_res *res) +{ + struct nfs4_slot *slot = res->sr_slot; + struct nfs4_slot_table *tbl; + + if (!RPC_WAS_SENT(task)) + goto out; + + tbl = slot->table; + spin_lock(&tbl->slot_tbl_lock); + if (!nfs41_wake_and_assign_slot(tbl, slot)) + nfs4_free_slot(tbl, slot); + spin_unlock(&tbl->slot_tbl_lock); + + res->sr_slot = NULL; +out: + return 1; +} + #if defined(CONFIG_NFS_V4_1) static void nfs41_sequence_free_slot(struct nfs4_sequence_res *res) @@ -506,6 +605,7 @@ static int nfs41_sequence_done(struct rpc_task *task, struct nfs4_sequence_res * interrupted = true; } + trace_nfs4_sequence_done(session, res); /* Check the SEQUENCE operation status */ switch (res->sr_status) { case 0: @@ -591,25 +691,11 @@ static int nfs4_sequence_done(struct rpc_task *task, { if (res->sr_slot == NULL) return 1; + if (!res->sr_slot->table->session) + return nfs40_sequence_done(task, res); return nfs41_sequence_done(task, res); } -static void nfs41_init_sequence(struct nfs4_sequence_args *args, - struct nfs4_sequence_res *res, int cache_reply) -{ - args->sa_slot = NULL; - args->sa_cache_this = 0; - args->sa_privileged = 0; - if (cache_reply) - args->sa_cache_this = 1; - res->sr_slot = NULL; -} - -static void nfs4_set_sequence_privileged(struct nfs4_sequence_args *args) -{ - args->sa_privileged = 1; -} - int nfs41_setup_sequence(struct nfs4_session *session, struct nfs4_sequence_args *args, struct nfs4_sequence_res *res, @@ -647,7 +733,7 @@ int nfs41_setup_sequence(struct nfs4_session *session, args->sa_slot = slot; - dprintk("<-- %s slotid=%d seqid=%d\n", __func__, + dprintk("<-- %s slotid=%u seqid=%u\n", __func__, slot->slot_nr, slot->seq_nr); res->sr_slot = slot; @@ -658,6 +744,7 @@ int nfs41_setup_sequence(struct nfs4_session *session, * set to 1 if an rpc level failure occurs. */ res->sr_status = 1; + trace_nfs4_setup_sequence(session, args); out_success: rpc_call_start(task); return 0; @@ -673,38 +760,30 @@ out_sleep: } EXPORT_SYMBOL_GPL(nfs41_setup_sequence); -int nfs4_setup_sequence(const struct nfs_server *server, - struct nfs4_sequence_args *args, - struct nfs4_sequence_res *res, - struct rpc_task *task) +static int nfs4_setup_sequence(const struct nfs_server *server, + struct nfs4_sequence_args *args, + struct nfs4_sequence_res *res, + struct rpc_task *task) { struct nfs4_session *session = nfs4_get_session(server); int ret = 0; - if (session == NULL) { - rpc_call_start(task); - goto out; - } + if (!session) + return nfs40_setup_sequence(server, args, res, task); - dprintk("--> %s clp %p session %p sr_slot %d\n", + dprintk("--> %s clp %p session %p sr_slot %u\n", __func__, session->clp, session, res->sr_slot ? - res->sr_slot->slot_nr : -1); + res->sr_slot->slot_nr : NFS4_NO_SLOT); ret = nfs41_setup_sequence(session, args, res, task); -out: + dprintk("<-- %s status=%d\n", __func__, ret); return ret; } -struct nfs41_call_sync_data { - const struct nfs_server *seq_server; - struct nfs4_sequence_args *seq_args; - struct nfs4_sequence_res *seq_res; -}; - static void nfs41_call_sync_prepare(struct rpc_task *task, void *calldata) { - struct nfs41_call_sync_data *data = calldata; + struct nfs4_call_sync_data *data = calldata; struct nfs4_session *session = nfs4_get_session(data->seq_server); dprintk("--> %s data->seq_server %p\n", __func__, data->seq_server); @@ -714,7 +793,7 @@ static void nfs41_call_sync_prepare(struct rpc_task *task, void *calldata) static void nfs41_call_sync_done(struct rpc_task *task, void *calldata) { - struct nfs41_call_sync_data *data = calldata; + struct nfs4_call_sync_data *data = calldata; nfs41_sequence_done(task, data->seq_res); } @@ -724,6 +803,42 @@ static const struct rpc_call_ops nfs41_call_sync_ops = { .rpc_call_done = nfs41_call_sync_done, }; +#else /* !CONFIG_NFS_V4_1 */ + +static int nfs4_setup_sequence(const struct nfs_server *server, + struct nfs4_sequence_args *args, + struct nfs4_sequence_res *res, + struct rpc_task *task) +{ + return nfs40_setup_sequence(server, args, res, task); +} + +static int nfs4_sequence_done(struct rpc_task *task, + struct nfs4_sequence_res *res) +{ + return nfs40_sequence_done(task, res); +} + +#endif /* !CONFIG_NFS_V4_1 */ + +static void nfs40_call_sync_prepare(struct rpc_task *task, void *calldata) +{ + struct nfs4_call_sync_data *data = calldata; + nfs4_setup_sequence(data->seq_server, + data->seq_args, data->seq_res, task); +} + +static void nfs40_call_sync_done(struct rpc_task *task, void *calldata) +{ + struct nfs4_call_sync_data *data = calldata; + nfs4_sequence_done(task, data->seq_res); +} + +static const struct rpc_call_ops nfs40_call_sync_ops = { + .rpc_call_prepare = nfs40_call_sync_prepare, + .rpc_call_done = nfs40_call_sync_done, +}; + static int nfs4_call_sync_sequence(struct rpc_clnt *clnt, struct nfs_server *server, struct rpc_message *msg, @@ -732,7 +847,8 @@ static int nfs4_call_sync_sequence(struct rpc_clnt *clnt, { int ret; struct rpc_task *task; - struct nfs41_call_sync_data data = { + struct nfs_client *clp = server->nfs_client; + struct nfs4_call_sync_data data = { .seq_server = server, .seq_args = args, .seq_res = res, @@ -740,7 +856,7 @@ static int nfs4_call_sync_sequence(struct rpc_clnt *clnt, struct rpc_task_setup task_setup = { .rpc_client = clnt, .rpc_message = msg, - .callback_ops = &nfs41_call_sync_ops, + .callback_ops = clp->cl_mvops->call_sync_ops, .callback_data = &data }; @@ -754,35 +870,6 @@ static int nfs4_call_sync_sequence(struct rpc_clnt *clnt, return ret; } -#else -static -void nfs41_init_sequence(struct nfs4_sequence_args *args, - struct nfs4_sequence_res *res, int cache_reply) -{ -} - -static void nfs4_set_sequence_privileged(struct nfs4_sequence_args *args) -{ -} - - -static int nfs4_sequence_done(struct rpc_task *task, - struct nfs4_sequence_res *res) -{ - return 1; -} -#endif /* CONFIG_NFS_V4_1 */ - -static -int _nfs4_call_sync(struct rpc_clnt *clnt, - struct nfs_server *server, - struct rpc_message *msg, - struct nfs4_sequence_args *args, - struct nfs4_sequence_res *res) -{ - return rpc_call_sync(clnt, msg, 0); -} - static int nfs4_call_sync(struct rpc_clnt *clnt, struct nfs_server *server, @@ -791,9 +878,8 @@ int nfs4_call_sync(struct rpc_clnt *clnt, struct nfs4_sequence_res *res, int cache_reply) { - nfs41_init_sequence(args, res, cache_reply); - return server->nfs_client->cl_mvops->call_sync(clnt, server, msg, - args, res); + nfs4_init_sequence(args, res, cache_reply); + return nfs4_call_sync_sequence(clnt, server, msg, args, res); } static void update_changeattr(struct inode *dir, struct nfs4_change_info *cinfo) @@ -933,7 +1019,7 @@ static struct nfs4_opendata *nfs4_opendata_alloc(struct dentry *dentry, p->o_arg.fh = NFS_FH(dentry->d_inode); } if (attrs != NULL && attrs->ia_valid != 0) { - __be32 verf[2]; + __u32 verf[2]; p->o_arg.u.attrs = &p->attrs; memcpy(&p->attrs, attrs, sizeof(p->attrs)); @@ -1103,7 +1189,7 @@ static int update_open_stateid(struct nfs4_state *state, nfs4_stateid *open_stat goto no_delegation; spin_lock(&deleg_cur->lock); - if (nfsi->delegation != deleg_cur || + if (rcu_dereference(nfsi->delegation) != deleg_cur || test_bit(NFS_DELEGATION_RETURNING, &deleg_cur->flags) || (deleg_cur->type & fmode) != fmode) goto no_delegation_unlock; @@ -1440,6 +1526,7 @@ static int nfs4_do_open_reclaim(struct nfs_open_context *ctx, struct nfs4_state int err; do { err = _nfs4_do_open_reclaim(ctx, state); + trace_nfs4_open_reclaim(ctx, 0, err); if (nfs4_clear_cap_atomic_open_v1(server, err, &exception)) continue; if (err != -NFS4ERR_DELAY) @@ -1524,10 +1611,20 @@ int nfs4_open_delegation_recall(struct nfs_open_context *ctx, struct nfs4_state return nfs4_handle_delegation_recall_error(server, state, stateid, err); } +static void nfs4_open_confirm_prepare(struct rpc_task *task, void *calldata) +{ + struct nfs4_opendata *data = calldata; + + nfs40_setup_sequence(data->o_arg.server, &data->o_arg.seq_args, + &data->o_res.seq_res, task); +} + static void nfs4_open_confirm_done(struct rpc_task *task, void *calldata) { struct nfs4_opendata *data = calldata; + nfs40_sequence_done(task, &data->o_res.seq_res); + data->rpc_status = task->tk_status; if (data->rpc_status == 0) { nfs4_stateid_copy(&data->o_res.stateid, &data->c_res.stateid); @@ -1556,6 +1653,7 @@ out_free: } static const struct rpc_call_ops nfs4_open_confirm_ops = { + .rpc_call_prepare = nfs4_open_confirm_prepare, .rpc_call_done = nfs4_open_confirm_done, .rpc_release = nfs4_open_confirm_release, }; @@ -1583,6 +1681,7 @@ static int _nfs4_proc_open_confirm(struct nfs4_opendata *data) }; int status; + nfs4_init_sequence(&data->o_arg.seq_args, &data->o_res.seq_res, 1); kref_get(&data->kref); data->rpc_done = 0; data->rpc_status = 0; @@ -1742,7 +1841,7 @@ static int nfs4_run_open_task(struct nfs4_opendata *data, int isrecover) }; int status; - nfs41_init_sequence(&o_arg->seq_args, &o_res->seq_res, 1); + nfs4_init_sequence(&o_arg->seq_args, &o_res->seq_res, 1); kref_get(&data->kref); data->rpc_done = 0; data->rpc_status = 0; @@ -1895,6 +1994,7 @@ static int nfs4_do_open_expired(struct nfs_open_context *ctx, struct nfs4_state do { err = _nfs4_open_expired(ctx, state); + trace_nfs4_open_expired(ctx, 0, err); if (nfs4_clear_cap_atomic_open_v1(server, err, &exception)) continue; switch (err) { @@ -1944,6 +2044,7 @@ static void nfs41_clear_delegation_stateid(struct nfs4_state *state) cred = get_rpccred(delegation->cred); rcu_read_unlock(); status = nfs41_test_stateid(server, stateid, cred); + trace_nfs4_test_delegation_stateid(state, NULL, status); } else rcu_read_unlock(); @@ -1986,6 +2087,7 @@ static int nfs41_check_open_stateid(struct nfs4_state *state) return -NFS4ERR_BAD_STATEID; status = nfs41_test_stateid(server, stateid, cred); + trace_nfs4_test_open_stateid(state, NULL, status); if (status != NFS_OK) { /* Free the stateid unless the server explicitly * informs us the stateid is unrecognized. */ @@ -2197,6 +2299,7 @@ static struct nfs4_state *nfs4_do_open(struct inode *dir, do { status = _nfs4_do_open(dir, ctx, flags, sattr, label); res = ctx->state; + trace_nfs4_open_file(ctx, flags, status); if (status == 0) break; /* NOTE: BAD_SEQID means the server and client disagree about the @@ -2310,6 +2413,7 @@ static int nfs4_do_setattr(struct inode *inode, struct rpc_cred *cred, int err; do { err = _nfs4_do_setattr(inode, cred, fattr, sattr, state, ilabel, olabel); + trace_nfs4_setattr(inode, err); switch (err) { case -NFS4ERR_OPENMODE: if (!(sattr->ia_valid & ATTR_SIZE)) { @@ -2387,6 +2491,7 @@ static void nfs4_close_done(struct rpc_task *task, void *data) dprintk("%s: begin!\n", __func__); if (!nfs4_sequence_done(task, &calldata->res.seq_res)) return; + trace_nfs4_close(state, &calldata->arg, &calldata->res, task->tk_status); /* hmm. we are done with the inode, and in the process of freeing * the state_owner. we keep this around to process errors */ @@ -2511,10 +2616,13 @@ int nfs4_do_close(struct nfs4_state *state, gfp_t gfp_mask, int wait) }; int status = -ENOMEM; + nfs4_state_protect(server->nfs_client, NFS_SP4_MACH_CRED_CLEANUP, + &task_setup_data.rpc_client, &msg); + calldata = kzalloc(sizeof(*calldata), gfp_mask); if (calldata == NULL) goto out; - nfs41_init_sequence(&calldata->arg.seq_args, &calldata->res.seq_res, 1); + nfs4_init_sequence(&calldata->arg.seq_args, &calldata->res.seq_res, 1); calldata->inode = state->inode; calldata->state = state; calldata->arg.fh = NFS_FH(state->inode); @@ -2690,6 +2798,7 @@ static int nfs4_lookup_root(struct nfs_server *server, struct nfs_fh *fhandle, int err; do { err = _nfs4_lookup_root(server, fhandle, info); + trace_nfs4_lookup_root(server, fhandle, info->fattr, err); switch (err) { case 0: case -NFS4ERR_WRONGSEC: @@ -2705,10 +2814,13 @@ out: static int nfs4_lookup_root_sec(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_fsinfo *info, rpc_authflavor_t flavor) { + struct rpc_auth_create_args auth_args = { + .pseudoflavor = flavor, + }; struct rpc_auth *auth; int ret; - auth = rpcauth_create(flavor, server->client); + auth = rpcauth_create(&auth_args, server->client); if (IS_ERR(auth)) { ret = -EACCES; goto out; @@ -2772,18 +2884,27 @@ static int nfs4_do_find_root_sec(struct nfs_server *server, * @server: initialized nfs_server handle * @fhandle: we fill in the pseudo-fs root file handle * @info: we fill in an FSINFO struct + * @auth_probe: probe the auth flavours * * Returns zero on success, or a negative errno. */ int nfs4_proc_get_rootfh(struct nfs_server *server, struct nfs_fh *fhandle, - struct nfs_fsinfo *info) + struct nfs_fsinfo *info, + bool auth_probe) { int status; - status = nfs4_lookup_root(server, fhandle, info); - if ((status == -NFS4ERR_WRONGSEC) && - !(server->flags & NFS_MOUNT_SECFLAVOUR)) + switch (auth_probe) { + case false: + status = nfs4_lookup_root(server, fhandle, info); + if (status != -NFS4ERR_WRONGSEC) + break; + /* Did user force a 'sec=' mount option? */ + if (server->flags & NFS_MOUNT_SECFLAVOUR) + break; + default: status = nfs4_do_find_root_sec(server, fhandle, info); + } if (status == 0) status = nfs4_server_capabilities(server, fhandle); @@ -2899,8 +3020,9 @@ static int nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs4_exception exception = { }; int err; do { - err = nfs4_handle_exception(server, - _nfs4_proc_getattr(server, fhandle, fattr, label), + err = _nfs4_proc_getattr(server, fhandle, fattr, label); + trace_nfs4_getattr(server, fhandle, fattr, err); + err = nfs4_handle_exception(server, err, &exception); } while (exception.retry); return err; @@ -2940,10 +3062,10 @@ nfs4_proc_setattr(struct dentry *dentry, struct nfs_fattr *fattr, /* Deal with open(O_TRUNC) */ if (sattr->ia_valid & ATTR_OPEN) - sattr->ia_valid &= ~(ATTR_MTIME|ATTR_CTIME|ATTR_OPEN); + sattr->ia_valid &= ~(ATTR_MTIME|ATTR_CTIME); /* Optimization: if the end result is no change, don't RPC */ - if ((sattr->ia_valid & ~(ATTR_FILE)) == 0) + if ((sattr->ia_valid & ~(ATTR_FILE|ATTR_OPEN)) == 0) return 0; /* Search for an existing open(O_WRITE) file */ @@ -3020,6 +3142,7 @@ static int nfs4_proc_lookup_common(struct rpc_clnt **clnt, struct inode *dir, int err; do { err = _nfs4_proc_lookup(client, dir, name, fhandle, fattr, label); + trace_nfs4_lookup(dir, name, err); switch (err) { case -NFS4ERR_BADNAME: err = -ENOENT; @@ -3031,7 +3154,9 @@ static int nfs4_proc_lookup_common(struct rpc_clnt **clnt, struct inode *dir, err = -EPERM; if (client != *clnt) goto out; - + /* No security negotiation if the user specified 'sec=' */ + if (NFS_SERVER(dir)->flags & NFS_MOUNT_SECFLAVOUR) + goto out; client = nfs4_create_sec_client(client, dir, name); if (IS_ERR(client)) return PTR_ERR(client); @@ -3134,8 +3259,9 @@ static int nfs4_proc_access(struct inode *inode, struct nfs_access_entry *entry) struct nfs4_exception exception = { }; int err; do { - err = nfs4_handle_exception(NFS_SERVER(inode), - _nfs4_proc_access(inode, entry), + err = _nfs4_proc_access(inode, entry); + trace_nfs4_access(inode, err); + err = nfs4_handle_exception(NFS_SERVER(inode), err, &exception); } while (exception.retry); return err; @@ -3188,8 +3314,9 @@ static int nfs4_proc_readlink(struct inode *inode, struct page *page, struct nfs4_exception exception = { }; int err; do { - err = nfs4_handle_exception(NFS_SERVER(inode), - _nfs4_proc_readlink(inode, page, pgbase, pglen), + err = _nfs4_proc_readlink(inode, page, pgbase, pglen); + trace_nfs4_readlink(inode, err); + err = nfs4_handle_exception(NFS_SERVER(inode), err, &exception); } while (exception.retry); return err; @@ -3253,8 +3380,9 @@ static int nfs4_proc_remove(struct inode *dir, struct qstr *name) struct nfs4_exception exception = { }; int err; do { - err = nfs4_handle_exception(NFS_SERVER(dir), - _nfs4_proc_remove(dir, name), + err = _nfs4_proc_remove(dir, name); + trace_nfs4_remove(dir, name, err); + err = nfs4_handle_exception(NFS_SERVER(dir), err, &exception); } while (exception.retry); return err; @@ -3268,7 +3396,7 @@ static void nfs4_proc_unlink_setup(struct rpc_message *msg, struct inode *dir) res->server = server; msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_REMOVE]; - nfs41_init_sequence(&args->seq_args, &res->seq_res, 1); + nfs4_init_sequence(&args->seq_args, &res->seq_res, 1); nfs_fattr_init(res->dir_attr); } @@ -3283,7 +3411,8 @@ static void nfs4_proc_unlink_rpc_prepare(struct rpc_task *task, struct nfs_unlin static int nfs4_proc_unlink_done(struct rpc_task *task, struct inode *dir) { - struct nfs_removeres *res = task->tk_msg.rpc_resp; + struct nfs_unlinkdata *data = task->tk_calldata; + struct nfs_removeres *res = &data->res; if (!nfs4_sequence_done(task, &res->seq_res)) return 0; @@ -3301,7 +3430,7 @@ static void nfs4_proc_rename_setup(struct rpc_message *msg, struct inode *dir) msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_RENAME]; res->server = server; - nfs41_init_sequence(&arg->seq_args, &res->seq_res, 1); + nfs4_init_sequence(&arg->seq_args, &res->seq_res, 1); } static void nfs4_proc_rename_rpc_prepare(struct rpc_task *task, struct nfs_renamedata *data) @@ -3315,7 +3444,8 @@ static void nfs4_proc_rename_rpc_prepare(struct rpc_task *task, struct nfs_renam static int nfs4_proc_rename_done(struct rpc_task *task, struct inode *old_dir, struct inode *new_dir) { - struct nfs_renameres *res = task->tk_msg.rpc_resp; + struct nfs_renamedata *data = task->tk_calldata; + struct nfs_renameres *res = &data->res; if (!nfs4_sequence_done(task, &res->seq_res)) return 0; @@ -3361,9 +3491,10 @@ static int nfs4_proc_rename(struct inode *old_dir, struct qstr *old_name, struct nfs4_exception exception = { }; int err; do { - err = nfs4_handle_exception(NFS_SERVER(old_dir), - _nfs4_proc_rename(old_dir, old_name, - new_dir, new_name), + err = _nfs4_proc_rename(old_dir, old_name, + new_dir, new_name); + trace_nfs4_rename(old_dir, old_name, new_dir, new_name, err); + err = nfs4_handle_exception(NFS_SERVER(old_dir), err, &exception); } while (exception.retry); return err; @@ -3525,9 +3656,9 @@ static int nfs4_proc_symlink(struct inode *dir, struct dentry *dentry, label = nfs4_label_init_security(dir, dentry, sattr, &l); do { - err = nfs4_handle_exception(NFS_SERVER(dir), - _nfs4_proc_symlink(dir, dentry, page, - len, sattr, label), + err = _nfs4_proc_symlink(dir, dentry, page, len, sattr, label); + trace_nfs4_symlink(dir, &dentry->d_name, err); + err = nfs4_handle_exception(NFS_SERVER(dir), err, &exception); } while (exception.retry); @@ -3564,8 +3695,9 @@ static int nfs4_proc_mkdir(struct inode *dir, struct dentry *dentry, sattr->ia_mode &= ~current_umask(); do { - err = nfs4_handle_exception(NFS_SERVER(dir), - _nfs4_proc_mkdir(dir, dentry, sattr, label), + err = _nfs4_proc_mkdir(dir, dentry, sattr, label); + trace_nfs4_mkdir(dir, &dentry->d_name, err); + err = nfs4_handle_exception(NFS_SERVER(dir), err, &exception); } while (exception.retry); nfs4_label_release_security(label); @@ -3618,9 +3750,10 @@ static int nfs4_proc_readdir(struct dentry *dentry, struct rpc_cred *cred, struct nfs4_exception exception = { }; int err; do { - err = nfs4_handle_exception(NFS_SERVER(dentry->d_inode), - _nfs4_proc_readdir(dentry, cred, cookie, - pages, count, plus), + err = _nfs4_proc_readdir(dentry, cred, cookie, + pages, count, plus); + trace_nfs4_readdir(dentry->d_inode, err); + err = nfs4_handle_exception(NFS_SERVER(dentry->d_inode), err, &exception); } while (exception.retry); return err; @@ -3672,8 +3805,9 @@ static int nfs4_proc_mknod(struct inode *dir, struct dentry *dentry, sattr->ia_mode &= ~current_umask(); do { - err = nfs4_handle_exception(NFS_SERVER(dir), - _nfs4_proc_mknod(dir, dentry, sattr, label, rdev), + err = _nfs4_proc_mknod(dir, dentry, sattr, label, rdev); + trace_nfs4_mknod(dir, &dentry->d_name, err); + err = nfs4_handle_exception(NFS_SERVER(dir), err, &exception); } while (exception.retry); @@ -3741,6 +3875,7 @@ static int nfs4_do_fsinfo(struct nfs_server *server, struct nfs_fh *fhandle, str do { err = _nfs4_do_fsinfo(server, fhandle, fsinfo); + trace_nfs4_fsinfo(server, fhandle, fsinfo->fattr, err); if (err == 0) { struct nfs_client *clp = server->nfs_client; @@ -3859,6 +3994,7 @@ static int nfs4_read_done_cb(struct rpc_task *task, struct nfs_read_data *data) { struct nfs_server *server = NFS_SERVER(data->header->inode); + trace_nfs4_read(data, task->tk_status); if (nfs4_async_handle_error(task, server, data->args.context->state) == -EAGAIN) { rpc_restart_call_prepare(task); return -EAGAIN; @@ -3902,24 +4038,29 @@ static void nfs4_proc_read_setup(struct nfs_read_data *data, struct rpc_message data->timestamp = jiffies; data->read_done_cb = nfs4_read_done_cb; msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_READ]; - nfs41_init_sequence(&data->args.seq_args, &data->res.seq_res, 0); + nfs4_init_sequence(&data->args.seq_args, &data->res.seq_res, 0); } -static void nfs4_proc_read_rpc_prepare(struct rpc_task *task, struct nfs_read_data *data) +static int nfs4_proc_read_rpc_prepare(struct rpc_task *task, struct nfs_read_data *data) { if (nfs4_setup_sequence(NFS_SERVER(data->header->inode), &data->args.seq_args, &data->res.seq_res, task)) - return; - nfs4_set_rw_stateid(&data->args.stateid, data->args.context, - data->args.lock_context, FMODE_READ); + return 0; + if (nfs4_set_rw_stateid(&data->args.stateid, data->args.context, + data->args.lock_context, FMODE_READ) == -EIO) + return -EIO; + if (unlikely(test_bit(NFS_CONTEXT_BAD, &data->args.context->flags))) + return -EIO; + return 0; } static int nfs4_write_done_cb(struct rpc_task *task, struct nfs_write_data *data) { struct inode *inode = data->header->inode; + trace_nfs4_write(data, task->tk_status); if (nfs4_async_handle_error(task, NFS_SERVER(inode), data->args.context->state) == -EAGAIN) { rpc_restart_call_prepare(task); return -EAGAIN; @@ -3985,18 +4126,22 @@ static void nfs4_proc_write_setup(struct nfs_write_data *data, struct rpc_messag data->timestamp = jiffies; msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_WRITE]; - nfs41_init_sequence(&data->args.seq_args, &data->res.seq_res, 1); + nfs4_init_sequence(&data->args.seq_args, &data->res.seq_res, 1); } -static void nfs4_proc_write_rpc_prepare(struct rpc_task *task, struct nfs_write_data *data) +static int nfs4_proc_write_rpc_prepare(struct rpc_task *task, struct nfs_write_data *data) { if (nfs4_setup_sequence(NFS_SERVER(data->header->inode), &data->args.seq_args, &data->res.seq_res, task)) - return; - nfs4_set_rw_stateid(&data->args.stateid, data->args.context, - data->args.lock_context, FMODE_WRITE); + return 0; + if (nfs4_set_rw_stateid(&data->args.stateid, data->args.context, + data->args.lock_context, FMODE_WRITE) == -EIO) + return -EIO; + if (unlikely(test_bit(NFS_CONTEXT_BAD, &data->args.context->flags))) + return -EIO; + return 0; } static void nfs4_proc_commit_rpc_prepare(struct rpc_task *task, struct nfs_commit_data *data) @@ -4011,6 +4156,7 @@ static int nfs4_commit_done_cb(struct rpc_task *task, struct nfs_commit_data *da { struct inode *inode = data->inode; + trace_nfs4_commit(data, task->tk_status); if (nfs4_async_handle_error(task, NFS_SERVER(inode), NULL) == -EAGAIN) { rpc_restart_call_prepare(task); return -EAGAIN; @@ -4033,7 +4179,7 @@ static void nfs4_proc_commit_setup(struct nfs_commit_data *data, struct rpc_mess data->commit_done_cb = nfs4_commit_done_cb; data->res.server = server; msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_COMMIT]; - nfs41_init_sequence(&data->args.seq_args, &data->res.seq_res, 1); + nfs4_init_sequence(&data->args.seq_args, &data->res.seq_res, 1); } struct nfs4_renewdata { @@ -4062,6 +4208,7 @@ static void nfs4_renew_done(struct rpc_task *task, void *calldata) struct nfs_client *clp = data->client; unsigned long timestamp = data->timestamp; + trace_nfs4_renew_async(clp, task->tk_status); if (task->tk_status < 0) { /* Unless we're shutting down, schedule state recovery! */ if (test_bit(NFS_CS_RENEWD, &clp->cl_res_state) == 0) @@ -4319,6 +4466,7 @@ static ssize_t nfs4_get_acl_uncached(struct inode *inode, void *buf, size_t bufl ssize_t ret; do { ret = __nfs4_get_acl_uncached(inode, buf, buflen); + trace_nfs4_get_acl(inode, ret); if (ret >= 0) break; ret = nfs4_handle_exception(NFS_SERVER(inode), ret, &exception); @@ -4398,8 +4546,9 @@ static int nfs4_proc_set_acl(struct inode *inode, const void *buf, size_t buflen struct nfs4_exception exception = { }; int err; do { - err = nfs4_handle_exception(NFS_SERVER(inode), - __nfs4_proc_set_acl(inode, buf, buflen), + err = __nfs4_proc_set_acl(inode, buf, buflen); + trace_nfs4_set_acl(inode, err); + err = nfs4_handle_exception(NFS_SERVER(inode), err, &exception); } while (exception.retry); return err; @@ -4452,8 +4601,9 @@ static int nfs4_get_security_label(struct inode *inode, void *buf, return -EOPNOTSUPP; do { - err = nfs4_handle_exception(NFS_SERVER(inode), - _nfs4_get_security_label(inode, buf, buflen), + err = _nfs4_get_security_label(inode, buf, buflen); + trace_nfs4_get_security_label(inode, err); + err = nfs4_handle_exception(NFS_SERVER(inode), err, &exception); } while (exception.retry); return err; @@ -4505,9 +4655,10 @@ static int nfs4_do_set_security_label(struct inode *inode, int err; do { - err = nfs4_handle_exception(NFS_SERVER(inode), - _nfs4_do_set_security_label(inode, ilabel, - fattr, olabel), + err = _nfs4_do_set_security_label(inode, ilabel, + fattr, olabel); + trace_nfs4_set_security_label(inode, err); + err = nfs4_handle_exception(NFS_SERVER(inode), err, &exception); } while (exception.retry); return err; @@ -4630,11 +4781,11 @@ static void nfs4_init_boot_verifier(const struct nfs_client *clp, /* An impossible timestamp guarantees this value * will never match a generated boot time. */ verf[0] = 0; - verf[1] = (__be32)(NSEC_PER_SEC + 1); + verf[1] = cpu_to_be32(NSEC_PER_SEC + 1); } else { struct nfs_net *nn = net_generic(clp->cl_net, nfs_net_id); - verf[0] = (__be32)nn->boot_time.tv_sec; - verf[1] = (__be32)nn->boot_time.tv_nsec; + verf[0] = cpu_to_be32(nn->boot_time.tv_sec); + verf[1] = cpu_to_be32(nn->boot_time.tv_nsec); } memcpy(bootverf->data, verf, sizeof(bootverf->data)); } @@ -4660,10 +4811,14 @@ static unsigned int nfs4_init_uniform_client_string(const struct nfs_client *clp, char *buf, size_t len) { - char *nodename = clp->cl_rpcclient->cl_nodename; + const char *nodename = clp->cl_rpcclient->cl_nodename; if (nfs4_client_id_uniquifier[0] != '\0') - nodename = nfs4_client_id_uniquifier; + return scnprintf(buf, len, "Linux NFSv%u.%u %s/%s", + clp->rpc_ops->version, + clp->cl_minorversion, + nfs4_client_id_uniquifier, + nodename); return scnprintf(buf, len, "Linux NFSv%u.%u %s", clp->rpc_ops->version, clp->cl_minorversion, nodename); @@ -4724,6 +4879,7 @@ int nfs4_proc_setclientid(struct nfs_client *clp, u32 program, clp->cl_rpcclient->cl_auth->au_ops->au_name, setclientid.sc_name_len, setclientid.sc_name); status = rpc_call_sync(clp->cl_rpcclient, &msg, RPC_TASK_TIMEOUT); + trace_nfs4_setclientid(clp, status); dprintk("NFS reply setclientid: %d\n", status); return status; } @@ -4751,6 +4907,7 @@ int nfs4_proc_setclientid_confirm(struct nfs_client *clp, clp->cl_rpcclient->cl_auth->au_ops->au_name, clp->cl_clientid); status = rpc_call_sync(clp->cl_rpcclient, &msg, RPC_TASK_TIMEOUT); + trace_nfs4_setclientid_confirm(clp, status); dprintk("NFS reply setclientid_confirm: %d\n", status); return status; } @@ -4772,6 +4929,7 @@ static void nfs4_delegreturn_done(struct rpc_task *task, void *calldata) if (!nfs4_sequence_done(task, &data->res.seq_res)) return; + trace_nfs4_delegreturn_exit(&data->args, &data->res, task->tk_status); switch (task->tk_status) { case -NFS4ERR_STALE_STATEID: case -NFS4ERR_EXPIRED: @@ -4793,7 +4951,6 @@ static void nfs4_delegreturn_release(void *calldata) kfree(calldata); } -#if defined(CONFIG_NFS_V4_1) static void nfs4_delegreturn_prepare(struct rpc_task *task, void *data) { struct nfs4_delegreturndata *d_data; @@ -4805,12 +4962,9 @@ static void nfs4_delegreturn_prepare(struct rpc_task *task, void *data) &d_data->res.seq_res, task); } -#endif /* CONFIG_NFS_V4_1 */ static const struct rpc_call_ops nfs4_delegreturn_ops = { -#if defined(CONFIG_NFS_V4_1) .rpc_call_prepare = nfs4_delegreturn_prepare, -#endif /* CONFIG_NFS_V4_1 */ .rpc_call_done = nfs4_delegreturn_done, .rpc_release = nfs4_delegreturn_release, }; @@ -4835,7 +4989,7 @@ static int _nfs4_proc_delegreturn(struct inode *inode, struct rpc_cred *cred, co data = kzalloc(sizeof(*data), GFP_NOFS); if (data == NULL) return -ENOMEM; - nfs41_init_sequence(&data->args.seq_args, &data->res.seq_res, 1); + nfs4_init_sequence(&data->args.seq_args, &data->res.seq_res, 1); data->args.fhandle = &data->fh; data->args.stateid = &data->stateid; data->args.bitmask = server->cache_consistency_bitmask; @@ -4875,6 +5029,7 @@ int nfs4_proc_delegreturn(struct inode *inode, struct rpc_cred *cred, const nfs4 int err; do { err = _nfs4_proc_delegreturn(inode, cred, stateid, issync); + trace_nfs4_delegreturn(inode, err); switch (err) { case -NFS4ERR_STALE_STATEID: case -NFS4ERR_EXPIRED: @@ -4949,8 +5104,9 @@ static int nfs4_proc_getlk(struct nfs4_state *state, int cmd, struct file_lock * int err; do { - err = nfs4_handle_exception(NFS_SERVER(state->inode), - _nfs4_proc_getlk(state, cmd, request), + err = _nfs4_proc_getlk(state, cmd, request); + trace_nfs4_get_lock(request, state, cmd, err); + err = nfs4_handle_exception(NFS_SERVER(state->inode), err, &exception); } while (exception.retry); return err; @@ -5087,6 +5243,9 @@ static struct rpc_task *nfs4_do_unlck(struct file_lock *fl, .flags = RPC_TASK_ASYNC, }; + nfs4_state_protect(NFS_SERVER(lsp->ls_state->inode)->nfs_client, + NFS_SP4_MACH_CRED_CLEANUP, &task_setup_data.rpc_client, &msg); + /* Ensure this is an unlock - when canceling a lock, the * canceled lock is passed in, and it won't be an unlock. */ @@ -5098,7 +5257,7 @@ static struct rpc_task *nfs4_do_unlck(struct file_lock *fl, return ERR_PTR(-ENOMEM); } - nfs41_init_sequence(&data->arg.seq_args, &data->res.seq_res, 1); + nfs4_init_sequence(&data->arg.seq_args, &data->res.seq_res, 1); msg.rpc_argp = &data->arg; msg.rpc_resp = &data->res; task_setup_data.callback_data = data; @@ -5148,6 +5307,7 @@ static int nfs4_proc_unlck(struct nfs4_state *state, int cmd, struct file_lock * rpc_put_task(task); out: request->fl_flags = fl_flags; + trace_nfs4_unlock(request, state, F_SETLK, status); return status; } @@ -5333,7 +5493,7 @@ static int _nfs4_do_setlk(struct nfs4_state *state, int cmd, struct file_lock *f return -ENOMEM; if (IS_SETLKW(cmd)) data->arg.block = 1; - nfs41_init_sequence(&data->arg.seq_args, &data->res.seq_res, 1); + nfs4_init_sequence(&data->arg.seq_args, &data->res.seq_res, 1); msg.rpc_argp = &data->arg; msg.rpc_resp = &data->res; task_setup_data.callback_data = data; @@ -5371,6 +5531,7 @@ static int nfs4_lock_reclaim(struct nfs4_state *state, struct file_lock *request if (test_bit(NFS_DELEGATED_STATE, &state->flags) != 0) return 0; err = _nfs4_do_setlk(state, F_SETLK, request, NFS_LOCK_RECLAIM); + trace_nfs4_lock_reclaim(request, state, F_SETLK, err); if (err != -NFS4ERR_DELAY) break; nfs4_handle_exception(server, err, &exception); @@ -5389,10 +5550,15 @@ static int nfs4_lock_expired(struct nfs4_state *state, struct file_lock *request err = nfs4_set_lock_state(state, request); if (err != 0) return err; + if (!recover_lost_locks) { + set_bit(NFS_LOCK_LOST, &request->fl_u.nfs4_fl.owner->ls_flags); + return 0; + } do { if (test_bit(NFS_DELEGATED_STATE, &state->flags) != 0) return 0; err = _nfs4_do_setlk(state, F_SETLK, request, NFS_LOCK_EXPIRED); + trace_nfs4_lock_expired(request, state, F_SETLK, err); switch (err) { default: goto out; @@ -5428,6 +5594,7 @@ static int nfs41_check_expired_locks(struct nfs4_state *state) status = nfs41_test_stateid(server, &lsp->ls_stateid, cred); + trace_nfs4_test_lock_stateid(state, lsp, status); if (status != NFS_OK) { /* Free the stateid unless the server * informs us the stateid is unrecognized. */ @@ -5515,6 +5682,7 @@ static int nfs4_proc_setlk(struct nfs4_state *state, int cmd, struct file_lock * do { err = _nfs4_proc_setlk(state, cmd, request); + trace_nfs4_set_lock(request, state, cmd, err); if (err == -NFS4ERR_DENIED) err = -EAGAIN; err = nfs4_handle_exception(NFS_SERVER(state->inode), @@ -5597,8 +5765,23 @@ struct nfs_release_lockowner_data { struct nfs4_lock_state *lsp; struct nfs_server *server; struct nfs_release_lockowner_args args; + struct nfs4_sequence_args seq_args; + struct nfs4_sequence_res seq_res; }; +static void nfs4_release_lockowner_prepare(struct rpc_task *task, void *calldata) +{ + struct nfs_release_lockowner_data *data = calldata; + nfs40_setup_sequence(data->server, + &data->seq_args, &data->seq_res, task); +} + +static void nfs4_release_lockowner_done(struct rpc_task *task, void *calldata) +{ + struct nfs_release_lockowner_data *data = calldata; + nfs40_sequence_done(task, &data->seq_res); +} + static void nfs4_release_lockowner_release(void *calldata) { struct nfs_release_lockowner_data *data = calldata; @@ -5607,6 +5790,8 @@ static void nfs4_release_lockowner_release(void *calldata) } static const struct rpc_call_ops nfs4_release_lockowner_ops = { + .rpc_call_prepare = nfs4_release_lockowner_prepare, + .rpc_call_done = nfs4_release_lockowner_done, .rpc_release = nfs4_release_lockowner_release, }; @@ -5619,14 +5804,17 @@ static int nfs4_release_lockowner(struct nfs_server *server, struct nfs4_lock_st if (server->nfs_client->cl_mvops->minor_version != 0) return -EINVAL; + data = kmalloc(sizeof(*data), GFP_NOFS); if (!data) return -ENOMEM; + nfs4_init_sequence(&data->seq_args, &data->seq_res, 0); data->lsp = lsp; data->server = server; data->args.lock_owner.clientid = server->nfs_client->cl_clientid; data->args.lock_owner.id = lsp->ls_seqid.owner_id; data->args.lock_owner.s_dev = server->s_dev; + msg.rpc_argp = &data->args; rpc_call_async(server->client, &msg, 0, &nfs4_release_lockowner_ops, data); return 0; @@ -5781,14 +5969,23 @@ int nfs4_proc_fs_locations(struct rpc_clnt *client, struct inode *dir, struct nfs4_exception exception = { }; int err; do { - err = nfs4_handle_exception(NFS_SERVER(dir), - _nfs4_proc_fs_locations(client, dir, name, fs_locations, page), + err = _nfs4_proc_fs_locations(client, dir, name, + fs_locations, page); + trace_nfs4_get_fs_locations(dir, name, err); + err = nfs4_handle_exception(NFS_SERVER(dir), err, &exception); } while (exception.retry); return err; } -static int _nfs4_proc_secinfo(struct inode *dir, const struct qstr *name, struct nfs4_secinfo_flavors *flavors) +/** + * If 'use_integrity' is true and the state managment nfs_client + * cl_rpcclient is using krb5i/p, use the integrity protected cl_rpcclient + * and the machine credential as per RFC3530bis and RFC5661 Security + * Considerations sections. Otherwise, just use the user cred with the + * filesystem's rpc_client. + */ +static int _nfs4_proc_secinfo(struct inode *dir, const struct qstr *name, struct nfs4_secinfo_flavors *flavors, bool use_integrity) { int status; struct nfs4_secinfo_arg args = { @@ -5803,10 +6000,25 @@ static int _nfs4_proc_secinfo(struct inode *dir, const struct qstr *name, struct .rpc_argp = &args, .rpc_resp = &res, }; + struct rpc_clnt *clnt = NFS_SERVER(dir)->client; + + if (use_integrity) { + clnt = NFS_SERVER(dir)->nfs_client->cl_rpcclient; + msg.rpc_cred = nfs4_get_clid_cred(NFS_SERVER(dir)->nfs_client); + } dprintk("NFS call secinfo %s\n", name->name); - status = nfs4_call_sync(NFS_SERVER(dir)->client, NFS_SERVER(dir), &msg, &args.seq_args, &res.seq_res, 0); + + nfs4_state_protect(NFS_SERVER(dir)->nfs_client, + NFS_SP4_MACH_CRED_SECINFO, &clnt, &msg); + + status = nfs4_call_sync(clnt, NFS_SERVER(dir), &msg, &args.seq_args, + &res.seq_res, 0); dprintk("NFS reply secinfo: %d\n", status); + + if (msg.rpc_cred) + put_rpccred(msg.rpc_cred); + return status; } @@ -5816,8 +6028,23 @@ int nfs4_proc_secinfo(struct inode *dir, const struct qstr *name, struct nfs4_exception exception = { }; int err; do { - err = nfs4_handle_exception(NFS_SERVER(dir), - _nfs4_proc_secinfo(dir, name, flavors), + err = -NFS4ERR_WRONGSEC; + + /* try to use integrity protection with machine cred */ + if (_nfs4_is_integrity_protected(NFS_SERVER(dir)->nfs_client)) + err = _nfs4_proc_secinfo(dir, name, flavors, true); + + /* + * if unable to use integrity protection, or SECINFO with + * integrity protection returns NFS4ERR_WRONGSEC (which is + * disallowed by spec, but exists in deployed servers) use + * the current filesystem's rpc_client and the user cred. + */ + if (err == -NFS4ERR_WRONGSEC) + err = _nfs4_proc_secinfo(dir, name, flavors, false); + + trace_nfs4_secinfo(dir, name, err); + err = nfs4_handle_exception(NFS_SERVER(dir), err, &exception); } while (exception.retry); return err; @@ -5881,6 +6108,7 @@ int nfs4_proc_bind_conn_to_session(struct nfs_client *clp, struct rpc_cred *cred } status = rpc_call_sync(clp->cl_rpcclient, &msg, RPC_TASK_TIMEOUT); + trace_nfs4_bind_conn_to_session(clp, status); if (status == 0) { if (memcmp(res.session->sess_id.data, clp->cl_session->sess_id.data, NFS4_MAX_SESSIONID_LEN)) { @@ -5909,16 +6137,124 @@ out: } /* - * nfs4_proc_exchange_id() + * Minimum set of SP4_MACH_CRED operations from RFC 5661 in the enforce map + * and operations we'd like to see to enable certain features in the allow map + */ +static const struct nfs41_state_protection nfs4_sp4_mach_cred_request = { + .how = SP4_MACH_CRED, + .enforce.u.words = { + [1] = 1 << (OP_BIND_CONN_TO_SESSION - 32) | + 1 << (OP_EXCHANGE_ID - 32) | + 1 << (OP_CREATE_SESSION - 32) | + 1 << (OP_DESTROY_SESSION - 32) | + 1 << (OP_DESTROY_CLIENTID - 32) + }, + .allow.u.words = { + [0] = 1 << (OP_CLOSE) | + 1 << (OP_LOCKU), + [1] = 1 << (OP_SECINFO - 32) | + 1 << (OP_SECINFO_NO_NAME - 32) | + 1 << (OP_TEST_STATEID - 32) | + 1 << (OP_FREE_STATEID - 32) + } +}; + +/* + * Select the state protection mode for client `clp' given the server results + * from exchange_id in `sp'. * - * Returns zero, a negative errno, or a negative NFS4ERR status code. + * Returns 0 on success, negative errno otherwise. + */ +static int nfs4_sp4_select_mode(struct nfs_client *clp, + struct nfs41_state_protection *sp) +{ + static const u32 supported_enforce[NFS4_OP_MAP_NUM_WORDS] = { + [1] = 1 << (OP_BIND_CONN_TO_SESSION - 32) | + 1 << (OP_EXCHANGE_ID - 32) | + 1 << (OP_CREATE_SESSION - 32) | + 1 << (OP_DESTROY_SESSION - 32) | + 1 << (OP_DESTROY_CLIENTID - 32) + }; + unsigned int i; + + if (sp->how == SP4_MACH_CRED) { + /* Print state protect result */ + dfprintk(MOUNT, "Server SP4_MACH_CRED support:\n"); + for (i = 0; i <= LAST_NFS4_OP; i++) { + if (test_bit(i, sp->enforce.u.longs)) + dfprintk(MOUNT, " enforce op %d\n", i); + if (test_bit(i, sp->allow.u.longs)) + dfprintk(MOUNT, " allow op %d\n", i); + } + + /* make sure nothing is on enforce list that isn't supported */ + for (i = 0; i < NFS4_OP_MAP_NUM_WORDS; i++) { + if (sp->enforce.u.words[i] & ~supported_enforce[i]) { + dfprintk(MOUNT, "sp4_mach_cred: disabled\n"); + return -EINVAL; + } + } + + /* + * Minimal mode - state operations are allowed to use machine + * credential. Note this already happens by default, so the + * client doesn't have to do anything more than the negotiation. + * + * NOTE: we don't care if EXCHANGE_ID is in the list - + * we're already using the machine cred for exchange_id + * and will never use a different cred. + */ + if (test_bit(OP_BIND_CONN_TO_SESSION, sp->enforce.u.longs) && + test_bit(OP_CREATE_SESSION, sp->enforce.u.longs) && + test_bit(OP_DESTROY_SESSION, sp->enforce.u.longs) && + test_bit(OP_DESTROY_CLIENTID, sp->enforce.u.longs)) { + dfprintk(MOUNT, "sp4_mach_cred:\n"); + dfprintk(MOUNT, " minimal mode enabled\n"); + set_bit(NFS_SP4_MACH_CRED_MINIMAL, &clp->cl_sp4_flags); + } else { + dfprintk(MOUNT, "sp4_mach_cred: disabled\n"); + return -EINVAL; + } + + if (test_bit(OP_CLOSE, sp->allow.u.longs) && + test_bit(OP_LOCKU, sp->allow.u.longs)) { + dfprintk(MOUNT, " cleanup mode enabled\n"); + set_bit(NFS_SP4_MACH_CRED_CLEANUP, &clp->cl_sp4_flags); + } + + if (test_bit(OP_SECINFO, sp->allow.u.longs) && + test_bit(OP_SECINFO_NO_NAME, sp->allow.u.longs)) { + dfprintk(MOUNT, " secinfo mode enabled\n"); + set_bit(NFS_SP4_MACH_CRED_SECINFO, &clp->cl_sp4_flags); + } + + if (test_bit(OP_TEST_STATEID, sp->allow.u.longs) && + test_bit(OP_FREE_STATEID, sp->allow.u.longs)) { + dfprintk(MOUNT, " stateid mode enabled\n"); + set_bit(NFS_SP4_MACH_CRED_STATEID, &clp->cl_sp4_flags); + } + + if (test_bit(OP_WRITE, sp->allow.u.longs)) { + dfprintk(MOUNT, " write mode enabled\n"); + set_bit(NFS_SP4_MACH_CRED_WRITE, &clp->cl_sp4_flags); + } + + if (test_bit(OP_COMMIT, sp->allow.u.longs)) { + dfprintk(MOUNT, " commit mode enabled\n"); + set_bit(NFS_SP4_MACH_CRED_COMMIT, &clp->cl_sp4_flags); + } + } + + return 0; +} + +/* + * _nfs4_proc_exchange_id() * - * Since the clientid has expired, all compounds using sessions - * associated with the stale clientid will be returning - * NFS4ERR_BADSESSION in the sequence operation, and will therefore - * be in some phase of session reset. + * Wrapper for EXCHANGE_ID operation. */ -int nfs4_proc_exchange_id(struct nfs_client *clp, struct rpc_cred *cred) +static int _nfs4_proc_exchange_id(struct nfs_client *clp, struct rpc_cred *cred, + u32 sp4_how) { nfs4_verifier verifier; struct nfs41_exchange_id_args args = { @@ -5965,10 +6301,30 @@ int nfs4_proc_exchange_id(struct nfs_client *clp, struct rpc_cred *cred) goto out_server_scope; } + switch (sp4_how) { + case SP4_NONE: + args.state_protect.how = SP4_NONE; + break; + + case SP4_MACH_CRED: + args.state_protect = nfs4_sp4_mach_cred_request; + break; + + default: + /* unsupported! */ + WARN_ON_ONCE(1); + status = -EINVAL; + goto out_server_scope; + } + status = rpc_call_sync(clp->cl_rpcclient, &msg, RPC_TASK_TIMEOUT); + trace_nfs4_exchange_id(clp, status); if (status == 0) status = nfs4_check_cl_exchange_flags(res.flags); + if (status == 0) + status = nfs4_sp4_select_mode(clp, &res.state_protect); + if (status == 0) { clp->cl_clientid = res.clientid; clp->cl_exchange_flags = (res.flags & ~EXCHGID4_FLAG_CONFIRMED_R); @@ -6015,6 +6371,35 @@ out: return status; } +/* + * nfs4_proc_exchange_id() + * + * Returns zero, a negative errno, or a negative NFS4ERR status code. + * + * Since the clientid has expired, all compounds using sessions + * associated with the stale clientid will be returning + * NFS4ERR_BADSESSION in the sequence operation, and will therefore + * be in some phase of session reset. + * + * Will attempt to negotiate SP4_MACH_CRED if krb5i / krb5p auth is used. + */ +int nfs4_proc_exchange_id(struct nfs_client *clp, struct rpc_cred *cred) +{ + rpc_authflavor_t authflavor = clp->cl_rpcclient->cl_auth->au_flavor; + int status; + + /* try SP4_MACH_CRED if krb5i/p */ + if (authflavor == RPC_AUTH_GSS_KRB5I || + authflavor == RPC_AUTH_GSS_KRB5P) { + status = _nfs4_proc_exchange_id(clp, cred, SP4_MACH_CRED); + if (!status) + return 0; + } + + /* try SP4_NONE */ + return _nfs4_proc_exchange_id(clp, cred, SP4_NONE); +} + static int _nfs4_proc_destroy_clientid(struct nfs_client *clp, struct rpc_cred *cred) { @@ -6026,6 +6411,7 @@ static int _nfs4_proc_destroy_clientid(struct nfs_client *clp, int status; status = rpc_call_sync(clp->cl_rpcclient, &msg, RPC_TASK_TIMEOUT); + trace_nfs4_destroy_clientid(clp, status); if (status) dprintk("NFS: Got error %d from the server %s on " "DESTROY_CLIENTID.", status, clp->cl_hostname); @@ -6063,7 +6449,7 @@ int nfs4_destroy_clientid(struct nfs_client *clp) goto out; if (clp->cl_preserve_clid) goto out; - cred = nfs4_get_exchange_id_cred(clp); + cred = nfs4_get_clid_cred(clp); ret = nfs4_proc_destroy_clientid(clp, cred); if (cred) put_rpccred(cred); @@ -6155,7 +6541,7 @@ int nfs4_proc_get_lease_time(struct nfs_client *clp, struct nfs_fsinfo *fsinfo) }; int status; - nfs41_init_sequence(&args.la_seq_args, &res.lr_seq_res, 0); + nfs4_init_sequence(&args.la_seq_args, &res.lr_seq_res, 0); nfs4_set_sequence_privileged(&args.la_seq_args); dprintk("--> %s\n", __func__); task = rpc_run_task(&task_setup); @@ -6289,6 +6675,7 @@ static int _nfs4_proc_create_session(struct nfs_client *clp, args.flags = (SESSION4_PERSIST | SESSION4_BACK_CHAN); status = rpc_call_sync(session->clp->cl_rpcclient, &msg, RPC_TASK_TIMEOUT); + trace_nfs4_create_session(clp, status); if (!status) { /* Verify the session's negotiated channel_attrs values */ @@ -6352,6 +6739,7 @@ int nfs4_proc_destroy_session(struct nfs4_session *session, return status; status = rpc_call_sync(session->clp->cl_rpcclient, &msg, RPC_TASK_TIMEOUT); + trace_nfs4_destroy_session(session->clp, status); if (status) dprintk("NFS: Got error %d from the server on DESTROY_SESSION. " @@ -6401,6 +6789,7 @@ static void nfs41_sequence_call_done(struct rpc_task *task, void *data) if (!nfs41_sequence_done(task, task->tk_msg.rpc_resp)) return; + trace_nfs4_sequence(clp, task->tk_status); if (task->tk_status < 0) { dprintk("%s ERROR %d\n", __func__, task->tk_status); if (atomic_read(&clp->cl_count) == 1) @@ -6458,7 +6847,7 @@ static struct rpc_task *_nfs41_proc_sequence(struct nfs_client *clp, nfs_put_client(clp); return ERR_PTR(-ENOMEM); } - nfs41_init_sequence(&calldata->args, &calldata->res, 0); + nfs4_init_sequence(&calldata->args, &calldata->res, 0); if (is_privileged) nfs4_set_sequence_privileged(&calldata->args); msg.rpc_argp = &calldata->args; @@ -6553,6 +6942,7 @@ static void nfs4_reclaim_complete_done(struct rpc_task *task, void *data) if (!nfs41_sequence_done(task, res)) return; + trace_nfs4_reclaim_complete(clp, task->tk_status); if (nfs41_reclaim_complete_handle_errors(task, clp) == -EAGAIN) { rpc_restart_call_prepare(task); return; @@ -6600,7 +6990,7 @@ static int nfs41_proc_reclaim_complete(struct nfs_client *clp, calldata->clp = clp; calldata->arg.one_fs = 0; - nfs41_init_sequence(&calldata->arg.seq_args, &calldata->res.seq_res, 0); + nfs4_init_sequence(&calldata->arg.seq_args, &calldata->res.seq_res, 0); nfs4_set_sequence_privileged(&calldata->arg.seq_args); msg.rpc_argp = &calldata->arg; msg.rpc_resp = &calldata->res; @@ -6791,7 +7181,7 @@ nfs4_proc_layoutget(struct nfs4_layoutget *lgp, gfp_t gfp_flags) lgp->res.layoutp = &lgp->args.layout; lgp->res.seq_res.sr_slot = NULL; - nfs41_init_sequence(&lgp->args.seq_args, &lgp->res.seq_res, 0); + nfs4_init_sequence(&lgp->args.seq_args, &lgp->res.seq_res, 0); /* nfs4_layoutget_release calls pnfs_put_layout_hdr */ pnfs_get_layout_hdr(NFS_I(inode)->layout); @@ -6802,6 +7192,10 @@ nfs4_proc_layoutget(struct nfs4_layoutget *lgp, gfp_t gfp_flags) status = nfs4_wait_for_completion_rpc_task(task); if (status == 0) status = task->tk_status; + trace_nfs4_layoutget(lgp->args.ctx, + &lgp->args.range, + &lgp->res.range, + status); /* if layoutp->len is 0, nfs4_layoutget_prepare called rpc_exit */ if (status == 0 && lgp->res.layoutp->len) lseg = pnfs_layout_process(lgp); @@ -6874,7 +7268,7 @@ int nfs4_proc_layoutreturn(struct nfs4_layoutreturn *lrp) .rpc_cred = lrp->cred, }; struct rpc_task_setup task_setup_data = { - .rpc_client = lrp->clp->cl_rpcclient, + .rpc_client = NFS_SERVER(lrp->args.inode)->client, .rpc_message = &msg, .callback_ops = &nfs4_layoutreturn_call_ops, .callback_data = lrp, @@ -6882,11 +7276,12 @@ int nfs4_proc_layoutreturn(struct nfs4_layoutreturn *lrp) int status; dprintk("--> %s\n", __func__); - nfs41_init_sequence(&lrp->args.seq_args, &lrp->res.seq_res, 1); + nfs4_init_sequence(&lrp->args.seq_args, &lrp->res.seq_res, 1); task = rpc_run_task(&task_setup_data); if (IS_ERR(task)) return PTR_ERR(task); status = task->tk_status; + trace_nfs4_layoutreturn(lrp->args.inode, status); dprintk("<-- %s status=%d\n", __func__, status); rpc_put_task(task); return status; @@ -7063,7 +7458,7 @@ nfs4_proc_layoutcommit(struct nfs4_layoutcommit_data *data, bool sync) data->args.lastbytewritten, data->args.inode->i_ino); - nfs41_init_sequence(&data->args.seq_args, &data->res.seq_res, 1); + nfs4_init_sequence(&data->args.seq_args, &data->res.seq_res, 1); task = rpc_run_task(&task_setup_data); if (IS_ERR(task)) return PTR_ERR(task); @@ -7073,15 +7468,21 @@ nfs4_proc_layoutcommit(struct nfs4_layoutcommit_data *data, bool sync) if (status != 0) goto out; status = task->tk_status; + trace_nfs4_layoutcommit(data->args.inode, status); out: dprintk("%s: status %d\n", __func__, status); rpc_put_task(task); return status; } +/** + * Use the state managment nfs_client cl_rpcclient, which uses krb5i (if + * possible) as per RFC3530bis and RFC5661 Security Considerations sections + */ static int _nfs41_proc_secinfo_no_name(struct nfs_server *server, struct nfs_fh *fhandle, - struct nfs_fsinfo *info, struct nfs4_secinfo_flavors *flavors) + struct nfs_fsinfo *info, + struct nfs4_secinfo_flavors *flavors, bool use_integrity) { struct nfs41_secinfo_no_name_args args = { .style = SECINFO_STYLE_CURRENT_FH, @@ -7094,7 +7495,23 @@ _nfs41_proc_secinfo_no_name(struct nfs_server *server, struct nfs_fh *fhandle, .rpc_argp = &args, .rpc_resp = &res, }; - return nfs4_call_sync(server->client, server, &msg, &args.seq_args, &res.seq_res, 0); + struct rpc_clnt *clnt = server->client; + int status; + + if (use_integrity) { + clnt = server->nfs_client->cl_rpcclient; + msg.rpc_cred = nfs4_get_clid_cred(server->nfs_client); + } + + dprintk("--> %s\n", __func__); + status = nfs4_call_sync(clnt, server, &msg, &args.seq_args, + &res.seq_res, 0); + dprintk("<-- %s status=%d\n", __func__, status); + + if (msg.rpc_cred) + put_rpccred(msg.rpc_cred); + + return status; } static int @@ -7104,7 +7521,24 @@ nfs41_proc_secinfo_no_name(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs4_exception exception = { }; int err; do { - err = _nfs41_proc_secinfo_no_name(server, fhandle, info, flavors); + /* first try using integrity protection */ + err = -NFS4ERR_WRONGSEC; + + /* try to use integrity protection with machine cred */ + if (_nfs4_is_integrity_protected(server->nfs_client)) + err = _nfs41_proc_secinfo_no_name(server, fhandle, info, + flavors, true); + + /* + * if unable to use integrity protection, or SECINFO with + * integrity protection returns NFS4ERR_WRONGSEC (which is + * disallowed by spec, but exists in deployed servers) use + * the current filesystem's rpc_client and the user cred. + */ + if (err == -NFS4ERR_WRONGSEC) + err = _nfs41_proc_secinfo_no_name(server, fhandle, info, + flavors, false); + switch (err) { case 0: case -NFS4ERR_WRONGSEC: @@ -7174,11 +7608,15 @@ static int _nfs41_test_stateid(struct nfs_server *server, .rpc_resp = &res, .rpc_cred = cred, }; + struct rpc_clnt *rpc_client = server->client; + + nfs4_state_protect(server->nfs_client, NFS_SP4_MACH_CRED_STATEID, + &rpc_client, &msg); dprintk("NFS call test_stateid %p\n", stateid); - nfs41_init_sequence(&args.seq_args, &res.seq_res, 0); + nfs4_init_sequence(&args.seq_args, &res.seq_res, 0); nfs4_set_sequence_privileged(&args.seq_args); - status = nfs4_call_sync_sequence(server->client, server, &msg, + status = nfs4_call_sync_sequence(rpc_client, server, &msg, &args.seq_args, &res.seq_res); if (status != NFS_OK) { dprintk("NFS reply test_stateid: failed, %d\n", status); @@ -7247,7 +7685,7 @@ static void nfs41_free_stateid_release(void *calldata) kfree(calldata); } -const struct rpc_call_ops nfs41_free_stateid_ops = { +static const struct rpc_call_ops nfs41_free_stateid_ops = { .rpc_call_prepare = nfs41_free_stateid_prepare, .rpc_call_done = nfs41_free_stateid_done, .rpc_release = nfs41_free_stateid_release, @@ -7270,6 +7708,9 @@ static struct rpc_task *_nfs41_free_stateid(struct nfs_server *server, }; struct nfs_free_stateid_data *data; + nfs4_state_protect(server->nfs_client, NFS_SP4_MACH_CRED_STATEID, + &task_setup.rpc_client, &msg); + dprintk("NFS call free_stateid %p\n", stateid); data = kmalloc(sizeof(*data), GFP_NOFS); if (!data) @@ -7281,7 +7722,7 @@ static struct rpc_task *_nfs41_free_stateid(struct nfs_server *server, msg.rpc_argp = &data->args; msg.rpc_resp = &data->res; - nfs41_init_sequence(&data->args.seq_args, &data->res.seq_res, 0); + nfs4_init_sequence(&data->args.seq_args, &data->res.seq_res, 0); if (privileged) nfs4_set_sequence_privileged(&data->args.seq_args); @@ -7357,7 +7798,6 @@ static const struct nfs4_state_recovery_ops nfs40_reboot_recovery_ops = { .recover_open = nfs4_open_reclaim, .recover_lock = nfs4_lock_reclaim, .establish_clid = nfs4_init_clientid, - .get_clid_cred = nfs4_get_setclientid_cred, .detect_trunking = nfs40_discover_server_trunking, }; @@ -7368,7 +7808,6 @@ static const struct nfs4_state_recovery_ops nfs41_reboot_recovery_ops = { .recover_open = nfs4_open_reclaim, .recover_lock = nfs4_lock_reclaim, .establish_clid = nfs41_init_clientid, - .get_clid_cred = nfs4_get_exchange_id_cred, .reclaim_complete = nfs41_proc_reclaim_complete, .detect_trunking = nfs41_discover_server_trunking, }; @@ -7380,7 +7819,6 @@ static const struct nfs4_state_recovery_ops nfs40_nograce_recovery_ops = { .recover_open = nfs4_open_expired, .recover_lock = nfs4_lock_expired, .establish_clid = nfs4_init_clientid, - .get_clid_cred = nfs4_get_setclientid_cred, }; #if defined(CONFIG_NFS_V4_1) @@ -7390,7 +7828,6 @@ static const struct nfs4_state_recovery_ops nfs41_nograce_recovery_ops = { .recover_open = nfs41_open_expired, .recover_lock = nfs41_lock_expired, .establish_clid = nfs41_init_clientid, - .get_clid_cred = nfs4_get_exchange_id_cred, }; #endif /* CONFIG_NFS_V4_1 */ @@ -7414,10 +7851,12 @@ static const struct nfs4_minor_version_ops nfs_v4_0_minor_ops = { | NFS_CAP_ATOMIC_OPEN | NFS_CAP_CHANGE_ATTR | NFS_CAP_POSIX_LOCK, - .call_sync = _nfs4_call_sync, + .init_client = nfs40_init_client, + .shutdown_client = nfs40_shutdown_client, .match_stateid = nfs4_match_stateid, .find_root_sec = nfs4_find_root_sec, .free_lock_state = nfs4_release_lockowner, + .call_sync_ops = &nfs40_call_sync_ops, .reboot_recovery_ops = &nfs40_reboot_recovery_ops, .nograce_recovery_ops = &nfs40_nograce_recovery_ops, .state_renewal_ops = &nfs40_state_renewal_ops, @@ -7432,10 +7871,12 @@ static const struct nfs4_minor_version_ops nfs_v4_1_minor_ops = { | NFS_CAP_POSIX_LOCK | NFS_CAP_STATEID_NFSV41 | NFS_CAP_ATOMIC_OPEN_V1, - .call_sync = nfs4_call_sync_sequence, + .init_client = nfs41_init_client, + .shutdown_client = nfs41_shutdown_client, .match_stateid = nfs41_match_stateid, .find_root_sec = nfs41_find_root_sec, .free_lock_state = nfs41_free_lock_state, + .call_sync_ops = &nfs41_call_sync_ops, .reboot_recovery_ops = &nfs41_reboot_recovery_ops, .nograce_recovery_ops = &nfs41_nograce_recovery_ops, .state_renewal_ops = &nfs41_state_renewal_ops, @@ -7451,10 +7892,12 @@ static const struct nfs4_minor_version_ops nfs_v4_2_minor_ops = { | NFS_CAP_POSIX_LOCK | NFS_CAP_STATEID_NFSV41 | NFS_CAP_ATOMIC_OPEN_V1, - .call_sync = nfs4_call_sync_sequence, + .init_client = nfs41_init_client, + .shutdown_client = nfs41_shutdown_client, .match_stateid = nfs41_match_stateid, .find_root_sec = nfs41_find_root_sec, .free_lock_state = nfs41_free_lock_state, + .call_sync_ops = &nfs41_call_sync_ops, .reboot_recovery_ops = &nfs41_reboot_recovery_ops, .nograce_recovery_ops = &nfs41_nograce_recovery_ops, .state_renewal_ops = &nfs41_state_renewal_ops, @@ -7471,7 +7914,7 @@ const struct nfs4_minor_version_ops *nfs_v4_minor_ops[] = { #endif }; -const struct inode_operations nfs4_dir_inode_operations = { +static const struct inode_operations nfs4_dir_inode_operations = { .create = nfs_create, .lookup = nfs_lookup, .atomic_open = nfs_atomic_open, diff --git a/fs/nfs/nfs4session.c b/fs/nfs/nfs4session.c index 36e21cb..cf883c7 100644 --- a/fs/nfs/nfs4session.c +++ b/fs/nfs/nfs4session.c @@ -23,6 +23,14 @@ #define NFSDBG_FACILITY NFSDBG_STATE +static void nfs4_init_slot_table(struct nfs4_slot_table *tbl, const char *queue) +{ + tbl->highest_used_slotid = NFS4_NO_SLOT; + spin_lock_init(&tbl->slot_tbl_lock); + rpc_init_priority_wait_queue(&tbl->slot_tbl_waitq, queue); + init_completion(&tbl->complete); +} + /* * nfs4_shrink_slot_table - free retired slots from the slot table */ @@ -44,6 +52,17 @@ static void nfs4_shrink_slot_table(struct nfs4_slot_table *tbl, u32 newsize) } } +/** + * nfs4_slot_tbl_drain_complete - wake waiters when drain is complete + * @tbl - controlling slot table + * + */ +void nfs4_slot_tbl_drain_complete(struct nfs4_slot_table *tbl) +{ + if (nfs4_slot_tbl_draining(tbl)) + complete(&tbl->complete); +} + /* * nfs4_free_slot - free a slot and efficiently update slot table. * @@ -76,7 +95,7 @@ void nfs4_free_slot(struct nfs4_slot_table *tbl, struct nfs4_slot *slot) nfs4_slot_tbl_drain_complete(tbl); } } - dprintk("%s: slotid %u highest_used_slotid %d\n", __func__, + dprintk("%s: slotid %u highest_used_slotid %u\n", __func__, slotid, tbl->highest_used_slotid); } @@ -146,9 +165,9 @@ struct nfs4_slot *nfs4_alloc_slot(struct nfs4_slot_table *tbl) ret->generation = tbl->generation; out: - dprintk("<-- %s used_slots=%04lx highest_used=%d slotid=%d \n", + dprintk("<-- %s used_slots=%04lx highest_used=%u slotid=%u\n", __func__, tbl->used_slots[0], tbl->highest_used_slotid, - !IS_ERR(ret) ? ret->slot_nr : -1); + !IS_ERR(ret) ? ret->slot_nr : NFS4_NO_SLOT); return ret; } @@ -191,7 +210,7 @@ static int nfs4_realloc_slot_table(struct nfs4_slot_table *tbl, { int ret; - dprintk("--> %s: max_reqs=%u, tbl->max_slots %d\n", __func__, + dprintk("--> %s: max_reqs=%u, tbl->max_slots %u\n", __func__, max_reqs, tbl->max_slots); if (max_reqs > NFS4_MAX_SLOT_TABLE) @@ -205,18 +224,36 @@ static int nfs4_realloc_slot_table(struct nfs4_slot_table *tbl, nfs4_reset_slot_table(tbl, max_reqs - 1, ivalue); spin_unlock(&tbl->slot_tbl_lock); - dprintk("%s: tbl=%p slots=%p max_slots=%d\n", __func__, + dprintk("%s: tbl=%p slots=%p max_slots=%u\n", __func__, tbl, tbl->slots, tbl->max_slots); out: dprintk("<-- %s: return %d\n", __func__, ret); return ret; } -/* Destroy the slot table */ -static void nfs4_destroy_slot_tables(struct nfs4_session *session) +/** + * nfs4_release_slot_table - release resources attached to a slot table + * @tbl: slot table to shut down + * + */ +void nfs4_release_slot_table(struct nfs4_slot_table *tbl) +{ + nfs4_shrink_slot_table(tbl, 0); +} + +/** + * nfs4_setup_slot_table - prepare a stand-alone slot table for use + * @tbl: slot table to set up + * @max_reqs: maximum number of requests allowed + * @queue: name to give RPC wait queue + * + * Returns zero on success, or a negative errno. + */ +int nfs4_setup_slot_table(struct nfs4_slot_table *tbl, unsigned int max_reqs, + const char *queue) { - nfs4_shrink_slot_table(&session->fc_slot_table, 0); - nfs4_shrink_slot_table(&session->bc_slot_table, 0); + nfs4_init_slot_table(tbl, queue); + return nfs4_realloc_slot_table(tbl, max_reqs, 0); } static bool nfs41_assign_slot(struct rpc_task *task, void *pslot) @@ -273,6 +310,8 @@ void nfs41_wake_slot_table(struct nfs4_slot_table *tbl) } } +#if defined(CONFIG_NFS_V4_1) + static void nfs41_set_max_slotid_locked(struct nfs4_slot_table *tbl, u32 target_highest_slotid) { @@ -383,6 +422,12 @@ void nfs41_update_target_slotid(struct nfs4_slot_table *tbl, spin_unlock(&tbl->slot_tbl_lock); } +static void nfs4_destroy_session_slot_tables(struct nfs4_session *session) +{ + nfs4_release_slot_table(&session->fc_slot_table); + nfs4_release_slot_table(&session->bc_slot_table); +} + /* * Initialize or reset the forechannel and backchannel tables */ @@ -405,31 +450,20 @@ int nfs4_setup_session_slot_tables(struct nfs4_session *ses) if (status && tbl->slots == NULL) /* Fore and back channel share a connection so get * both slot tables or neither */ - nfs4_destroy_slot_tables(ses); + nfs4_destroy_session_slot_tables(ses); return status; } struct nfs4_session *nfs4_alloc_session(struct nfs_client *clp) { struct nfs4_session *session; - struct nfs4_slot_table *tbl; session = kzalloc(sizeof(struct nfs4_session), GFP_NOFS); if (!session) return NULL; - tbl = &session->fc_slot_table; - tbl->highest_used_slotid = NFS4_NO_SLOT; - spin_lock_init(&tbl->slot_tbl_lock); - rpc_init_priority_wait_queue(&tbl->slot_tbl_waitq, "ForeChannel Slot table"); - init_completion(&tbl->complete); - - tbl = &session->bc_slot_table; - tbl->highest_used_slotid = NFS4_NO_SLOT; - spin_lock_init(&tbl->slot_tbl_lock); - rpc_init_wait_queue(&tbl->slot_tbl_waitq, "BackChannel Slot table"); - init_completion(&tbl->complete); - + nfs4_init_slot_table(&session->fc_slot_table, "ForeChannel Slot table"); + nfs4_init_slot_table(&session->bc_slot_table, "BackChannel Slot table"); session->session_state = 1<<NFS4_SESSION_INITING; session->clp = clp; @@ -441,7 +475,7 @@ void nfs4_destroy_session(struct nfs4_session *session) struct rpc_xprt *xprt; struct rpc_cred *cred; - cred = nfs4_get_exchange_id_cred(session->clp); + cred = nfs4_get_clid_cred(session->clp); nfs4_proc_destroy_session(session, cred); if (cred) put_rpccred(cred); @@ -452,7 +486,7 @@ void nfs4_destroy_session(struct nfs4_session *session) dprintk("%s Destroy backchannel for xprt %p\n", __func__, xprt); xprt_destroy_backchannel(xprt, NFS41_BC_MIN_CALLBACKS); - nfs4_destroy_slot_tables(session); + nfs4_destroy_session_slot_tables(session); kfree(session); } @@ -513,4 +547,4 @@ int nfs4_init_ds_session(struct nfs_client *clp, unsigned long lease_time) } EXPORT_SYMBOL_GPL(nfs4_init_ds_session); - +#endif /* defined(CONFIG_NFS_V4_1) */ diff --git a/fs/nfs/nfs4session.h b/fs/nfs/nfs4session.h index 3a153d8..2323061 100644 --- a/fs/nfs/nfs4session.h +++ b/fs/nfs/nfs4session.h @@ -8,7 +8,7 @@ #define __LINUX_FS_NFS_NFS4SESSION_H /* maximum number of slots to use */ -#define NFS4_DEF_SLOT_TABLE_SIZE (16U) +#define NFS4_DEF_SLOT_TABLE_SIZE (64U) #define NFS4_MAX_SLOT_TABLE (1024U) #define NFS4_NO_SLOT ((u32)-1) @@ -72,10 +72,22 @@ enum nfs4_session_state { NFS4_SESSION_INITING, }; -#if defined(CONFIG_NFS_V4_1) +extern int nfs4_setup_slot_table(struct nfs4_slot_table *tbl, + unsigned int max_reqs, const char *queue); +extern void nfs4_release_slot_table(struct nfs4_slot_table *tbl); extern struct nfs4_slot *nfs4_alloc_slot(struct nfs4_slot_table *tbl); extern void nfs4_free_slot(struct nfs4_slot_table *tbl, struct nfs4_slot *slot); +extern void nfs4_slot_tbl_drain_complete(struct nfs4_slot_table *tbl); +bool nfs41_wake_and_assign_slot(struct nfs4_slot_table *tbl, + struct nfs4_slot *slot); +void nfs41_wake_slot_table(struct nfs4_slot_table *tbl); + +static inline bool nfs4_slot_tbl_draining(struct nfs4_slot_table *tbl) +{ + return !!test_bit(NFS4_SLOT_TBL_DRAINING, &tbl->slot_tbl_state); +} +#if defined(CONFIG_NFS_V4_1) extern void nfs41_set_target_slotid(struct nfs4_slot_table *tbl, u32 target_highest_slotid); extern void nfs41_update_target_slotid(struct nfs4_slot_table *tbl, @@ -89,17 +101,6 @@ extern void nfs4_destroy_session(struct nfs4_session *session); extern int nfs4_init_session(struct nfs_client *clp); extern int nfs4_init_ds_session(struct nfs_client *, unsigned long); -extern void nfs4_slot_tbl_drain_complete(struct nfs4_slot_table *tbl); - -static inline bool nfs4_slot_tbl_draining(struct nfs4_slot_table *tbl) -{ - return !!test_bit(NFS4_SLOT_TBL_DRAINING, &tbl->slot_tbl_state); -} - -bool nfs41_wake_and_assign_slot(struct nfs4_slot_table *tbl, - struct nfs4_slot *slot); -void nfs41_wake_slot_table(struct nfs4_slot_table *tbl); - /* * Determine if sessions are in use. */ @@ -117,6 +118,16 @@ static inline int nfs4_has_persistent_session(const struct nfs_client *clp) return 0; } +#ifdef CONFIG_CRC32 +/* + * nfs_session_id_hash - calculate the crc32 hash for the session id + * @session - pointer to session + */ +#define nfs_session_id_hash(sess_id) \ + (~crc32_le(0xFFFFFFFF, &(sess_id)->data[0], sizeof((sess_id)->data))) +#else +#define nfs_session_id_hash(session) (0) +#endif #else /* defined(CONFIG_NFS_V4_1) */ static inline int nfs4_init_session(struct nfs_client *clp) diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c index e22862f..cc14cbb 100644 --- a/fs/nfs/nfs4state.c +++ b/fs/nfs/nfs4state.c @@ -154,6 +154,19 @@ struct rpc_cred *nfs4_get_machine_cred_locked(struct nfs_client *clp) return cred; } +static void nfs4_root_machine_cred(struct nfs_client *clp) +{ + struct rpc_cred *cred, *new; + + new = rpc_lookup_machine_cred(NULL); + spin_lock(&clp->cl_lock); + cred = clp->cl_machine_cred; + clp->cl_machine_cred = new; + spin_unlock(&clp->cl_lock); + if (cred != NULL) + put_rpccred(cred); +} + static struct rpc_cred * nfs4_get_renew_cred_server_locked(struct nfs_server *server) { @@ -202,32 +215,6 @@ out: return cred; } -#if defined(CONFIG_NFS_V4_1) - -static int nfs41_setup_state_renewal(struct nfs_client *clp) -{ - int status; - struct nfs_fsinfo fsinfo; - - if (!test_bit(NFS_CS_CHECK_LEASE_TIME, &clp->cl_res_state)) { - nfs4_schedule_state_renewal(clp); - return 0; - } - - status = nfs4_proc_get_lease_time(clp, &fsinfo); - if (status == 0) { - /* Update lease time and schedule renewal */ - spin_lock(&clp->cl_lock); - clp->cl_lease_time = fsinfo.lease_time * HZ; - clp->cl_last_renewal = jiffies; - spin_unlock(&clp->cl_lock); - - nfs4_schedule_state_renewal(clp); - } - - return status; -} - static void nfs4_end_drain_slot_table(struct nfs4_slot_table *tbl) { if (test_and_clear_bit(NFS4_SLOT_TBL_DRAINING, &tbl->slot_tbl_state)) { @@ -241,20 +228,18 @@ static void nfs4_end_drain_session(struct nfs_client *clp) { struct nfs4_session *ses = clp->cl_session; + if (clp->cl_slot_tbl) { + nfs4_end_drain_slot_table(clp->cl_slot_tbl); + return; + } + if (ses != NULL) { nfs4_end_drain_slot_table(&ses->bc_slot_table); nfs4_end_drain_slot_table(&ses->fc_slot_table); } } -/* - * Signal state manager thread if session fore channel is drained - */ -void nfs4_slot_tbl_drain_complete(struct nfs4_slot_table *tbl) -{ - if (nfs4_slot_tbl_draining(tbl)) - complete(&tbl->complete); -} +#if defined(CONFIG_NFS_V4_1) static int nfs4_drain_slot_tbl(struct nfs4_slot_table *tbl) { @@ -274,6 +259,9 @@ static int nfs4_begin_drain_session(struct nfs_client *clp) struct nfs4_session *ses = clp->cl_session; int ret = 0; + if (clp->cl_slot_tbl) + return nfs4_drain_slot_tbl(clp->cl_slot_tbl); + /* back channel */ ret = nfs4_drain_slot_tbl(&ses->bc_slot_table); if (ret) @@ -282,6 +270,30 @@ static int nfs4_begin_drain_session(struct nfs_client *clp) return nfs4_drain_slot_tbl(&ses->fc_slot_table); } +static int nfs41_setup_state_renewal(struct nfs_client *clp) +{ + int status; + struct nfs_fsinfo fsinfo; + + if (!test_bit(NFS_CS_CHECK_LEASE_TIME, &clp->cl_res_state)) { + nfs4_schedule_state_renewal(clp); + return 0; + } + + status = nfs4_proc_get_lease_time(clp, &fsinfo); + if (status == 0) { + /* Update lease time and schedule renewal */ + spin_lock(&clp->cl_lock); + clp->cl_lease_time = fsinfo.lease_time * HZ; + clp->cl_last_renewal = jiffies; + spin_unlock(&clp->cl_lock); + + nfs4_schedule_state_renewal(clp); + } + + return status; +} + static void nfs41_finish_session_reset(struct nfs_client *clp) { clear_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state); @@ -339,62 +351,21 @@ int nfs41_discover_server_trunking(struct nfs_client *clp, return nfs41_walk_client_list(clp, result, cred); } -struct rpc_cred *nfs4_get_exchange_id_cred(struct nfs_client *clp) -{ - struct rpc_cred *cred; - - spin_lock(&clp->cl_lock); - cred = nfs4_get_machine_cred_locked(clp); - spin_unlock(&clp->cl_lock); - return cred; -} - #endif /* CONFIG_NFS_V4_1 */ -static struct rpc_cred * -nfs4_get_setclientid_cred_server(struct nfs_server *server) -{ - struct nfs_client *clp = server->nfs_client; - struct rpc_cred *cred = NULL; - struct nfs4_state_owner *sp; - struct rb_node *pos; - - spin_lock(&clp->cl_lock); - pos = rb_first(&server->state_owners); - if (pos != NULL) { - sp = rb_entry(pos, struct nfs4_state_owner, so_server_node); - cred = get_rpccred(sp->so_cred); - } - spin_unlock(&clp->cl_lock); - return cred; -} - /** - * nfs4_get_setclientid_cred - Acquire credential for a setclientid operation + * nfs4_get_clid_cred - Acquire credential for a setclientid operation * @clp: client state handle * * Returns an rpc_cred with reference count bumped, or NULL. */ -struct rpc_cred *nfs4_get_setclientid_cred(struct nfs_client *clp) +struct rpc_cred *nfs4_get_clid_cred(struct nfs_client *clp) { - struct nfs_server *server; struct rpc_cred *cred; spin_lock(&clp->cl_lock); cred = nfs4_get_machine_cred_locked(clp); spin_unlock(&clp->cl_lock); - if (cred != NULL) - goto out; - - rcu_read_lock(); - list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) { - cred = nfs4_get_setclientid_cred_server(server); - if (cred != NULL) - break; - } - rcu_read_unlock(); - -out: return cred; } @@ -998,7 +969,9 @@ static int nfs4_copy_lock_stateid(nfs4_stateid *dst, fl_pid = lockowner->l_pid; spin_lock(&state->state_lock); lsp = __nfs4_find_lock_state(state, fl_owner, fl_pid, NFS4_ANY_LOCK_TYPE); - if (lsp != NULL && test_bit(NFS_LOCK_INITIALIZED, &lsp->ls_flags) != 0) { + if (lsp && test_bit(NFS_LOCK_LOST, &lsp->ls_flags)) + ret = -EIO; + else if (lsp != NULL && test_bit(NFS_LOCK_INITIALIZED, &lsp->ls_flags) != 0) { nfs4_stateid_copy(dst, &lsp->ls_stateid); ret = 0; smp_rmb(); @@ -1038,11 +1011,17 @@ static int nfs4_copy_open_stateid(nfs4_stateid *dst, struct nfs4_state *state) int nfs4_select_rw_stateid(nfs4_stateid *dst, struct nfs4_state *state, fmode_t fmode, const struct nfs_lockowner *lockowner) { - int ret = 0; + int ret = nfs4_copy_lock_stateid(dst, state, lockowner); + if (ret == -EIO) + /* A lost lock - don't even consider delegations */ + goto out; if (nfs4_copy_delegation_stateid(dst, state->inode, fmode)) goto out; - ret = nfs4_copy_lock_stateid(dst, state, lockowner); if (ret != -ENOENT) + /* nfs4_copy_delegation_stateid() didn't over-write + * dst, so it still has the lock stateid which we now + * choose to use. + */ goto out; ret = nfs4_copy_open_stateid(dst, state); out: @@ -1443,14 +1422,16 @@ restart: if (status >= 0) { status = nfs4_reclaim_locks(state, ops); if (status >= 0) { - spin_lock(&state->state_lock); - list_for_each_entry(lock, &state->lock_states, ls_locks) { - if (!test_bit(NFS_LOCK_INITIALIZED, &lock->ls_flags)) - pr_warn_ratelimited("NFS: " - "%s: Lock reclaim " - "failed!\n", __func__); + if (test_bit(NFS_DELEGATED_STATE, &state->flags) != 0) { + spin_lock(&state->state_lock); + list_for_each_entry(lock, &state->lock_states, ls_locks) { + if (!test_bit(NFS_LOCK_INITIALIZED, &lock->ls_flags)) + pr_warn_ratelimited("NFS: " + "%s: Lock reclaim " + "failed!\n", __func__); + } + spin_unlock(&state->state_lock); } - spin_unlock(&state->state_lock); nfs4_put_open_state(state); spin_lock(&sp->so_lock); goto restart; @@ -1618,7 +1599,7 @@ static void nfs4_state_end_reclaim_reboot(struct nfs_client *clp) if (!nfs4_state_clear_reclaim_reboot(clp)) return; ops = clp->cl_mvops->reboot_recovery_ops; - cred = ops->get_clid_cred(clp); + cred = nfs4_get_clid_cred(clp); nfs4_reclaim_complete(clp, ops, cred); put_rpccred(cred); } @@ -1732,7 +1713,7 @@ static int nfs4_check_lease(struct nfs_client *clp) cred = ops->get_state_renewal_cred_locked(clp); spin_unlock(&clp->cl_lock); if (cred == NULL) { - cred = nfs4_get_setclientid_cred(clp); + cred = nfs4_get_clid_cred(clp); status = -ENOKEY; if (cred == NULL) goto out; @@ -1804,7 +1785,7 @@ static int nfs4_establish_lease(struct nfs_client *clp) clp->cl_mvops->reboot_recovery_ops; int status; - cred = ops->get_clid_cred(clp); + cred = nfs4_get_clid_cred(clp); if (cred == NULL) return -ENOENT; status = ops->establish_clid(clp, cred); @@ -1878,7 +1859,7 @@ int nfs4_discover_server_trunking(struct nfs_client *clp, mutex_lock(&nfs_clid_init_mutex); again: status = -ENOENT; - cred = ops->get_clid_cred(clp); + cred = nfs4_get_clid_cred(clp); if (cred == NULL) goto out_unlock; @@ -1896,7 +1877,11 @@ again: __func__, status); goto again; case -EACCES: - if (i++) + if (i++ == 0) { + nfs4_root_machine_cred(clp); + goto again; + } + if (i > 2) break; case -NFS4ERR_CLID_INUSE: case -NFS4ERR_WRONGSEC: @@ -2052,7 +2037,7 @@ static int nfs4_reset_session(struct nfs_client *clp) if (!nfs4_has_session(clp)) return 0; nfs4_begin_drain_session(clp); - cred = nfs4_get_exchange_id_cred(clp); + cred = nfs4_get_clid_cred(clp); status = nfs4_proc_destroy_session(clp->cl_session, cred); switch (status) { case 0: @@ -2095,7 +2080,7 @@ static int nfs4_bind_conn_to_session(struct nfs_client *clp) if (!nfs4_has_session(clp)) return 0; nfs4_begin_drain_session(clp); - cred = nfs4_get_exchange_id_cred(clp); + cred = nfs4_get_clid_cred(clp); ret = nfs4_proc_bind_conn_to_session(clp, cred); if (cred) put_rpccred(cred); @@ -2116,7 +2101,6 @@ static int nfs4_bind_conn_to_session(struct nfs_client *clp) } #else /* CONFIG_NFS_V4_1 */ static int nfs4_reset_session(struct nfs_client *clp) { return 0; } -static int nfs4_end_drain_session(struct nfs_client *clp) { return 0; } static int nfs4_bind_conn_to_session(struct nfs_client *clp) { diff --git a/fs/nfs/nfs4super.c b/fs/nfs/nfs4super.c index 5dbe2d2..e26acdd 100644 --- a/fs/nfs/nfs4super.c +++ b/fs/nfs/nfs4super.c @@ -253,8 +253,6 @@ struct dentry *nfs4_try_mount(int flags, const char *dev_name, dfprintk(MOUNT, "--> nfs4_try_mount()\n"); - if (data->auth_flavors[0] == RPC_AUTH_MAXFLAVOR) - data->auth_flavors[0] = RPC_AUTH_UNIX; export_path = data->nfs_server.export_path; data->nfs_server.export_path = "/"; root_mnt = nfs_do_root_mount(&nfs4_remote_fs_type, flags, mount_info, diff --git a/fs/nfs/nfs4trace.c b/fs/nfs/nfs4trace.c new file mode 100644 index 0000000..d774335 --- /dev/null +++ b/fs/nfs/nfs4trace.c @@ -0,0 +1,17 @@ +/* + * Copyright (c) 2013 Trond Myklebust <Trond.Myklebust@netapp.com> + */ +#include <linux/nfs_fs.h> +#include "nfs4_fs.h" +#include "internal.h" +#include "nfs4session.h" +#include "callback.h" + +#define CREATE_TRACE_POINTS +#include "nfs4trace.h" + +#ifdef CONFIG_NFS_V4_1 +EXPORT_TRACEPOINT_SYMBOL_GPL(nfs4_pnfs_read); +EXPORT_TRACEPOINT_SYMBOL_GPL(nfs4_pnfs_write); +EXPORT_TRACEPOINT_SYMBOL_GPL(nfs4_pnfs_commit_ds); +#endif diff --git a/fs/nfs/nfs4trace.h b/fs/nfs/nfs4trace.h new file mode 100644 index 0000000..849cf14 --- /dev/null +++ b/fs/nfs/nfs4trace.h @@ -0,0 +1,1148 @@ +/* + * Copyright (c) 2013 Trond Myklebust <Trond.Myklebust@netapp.com> + */ +#undef TRACE_SYSTEM +#define TRACE_SYSTEM nfs4 + +#if !defined(_TRACE_NFS4_H) || defined(TRACE_HEADER_MULTI_READ) +#define _TRACE_NFS4_H + +#include <linux/tracepoint.h> + +#define show_nfsv4_errors(error) \ + __print_symbolic(error, \ + { NFS4_OK, "OK" }, \ + /* Mapped by nfs4_stat_to_errno() */ \ + { -EPERM, "EPERM" }, \ + { -ENOENT, "ENOENT" }, \ + { -EIO, "EIO" }, \ + { -ENXIO, "ENXIO" }, \ + { -EACCES, "EACCES" }, \ + { -EEXIST, "EEXIST" }, \ + { -EXDEV, "EXDEV" }, \ + { -ENOTDIR, "ENOTDIR" }, \ + { -EISDIR, "EISDIR" }, \ + { -EFBIG, "EFBIG" }, \ + { -ENOSPC, "ENOSPC" }, \ + { -EROFS, "EROFS" }, \ + { -EMLINK, "EMLINK" }, \ + { -ENAMETOOLONG, "ENAMETOOLONG" }, \ + { -ENOTEMPTY, "ENOTEMPTY" }, \ + { -EDQUOT, "EDQUOT" }, \ + { -ESTALE, "ESTALE" }, \ + { -EBADHANDLE, "EBADHANDLE" }, \ + { -EBADCOOKIE, "EBADCOOKIE" }, \ + { -ENOTSUPP, "ENOTSUPP" }, \ + { -ETOOSMALL, "ETOOSMALL" }, \ + { -EREMOTEIO, "EREMOTEIO" }, \ + { -EBADTYPE, "EBADTYPE" }, \ + { -EAGAIN, "EAGAIN" }, \ + { -ELOOP, "ELOOP" }, \ + { -EOPNOTSUPP, "EOPNOTSUPP" }, \ + { -EDEADLK, "EDEADLK" }, \ + /* RPC errors */ \ + { -ENOMEM, "ENOMEM" }, \ + { -EKEYEXPIRED, "EKEYEXPIRED" }, \ + { -ETIMEDOUT, "ETIMEDOUT" }, \ + { -ERESTARTSYS, "ERESTARTSYS" }, \ + { -ECONNREFUSED, "ECONNREFUSED" }, \ + { -ECONNRESET, "ECONNRESET" }, \ + { -ENETUNREACH, "ENETUNREACH" }, \ + { -EHOSTUNREACH, "EHOSTUNREACH" }, \ + { -EHOSTDOWN, "EHOSTDOWN" }, \ + { -EPIPE, "EPIPE" }, \ + { -EPFNOSUPPORT, "EPFNOSUPPORT" }, \ + { -EPROTONOSUPPORT, "EPROTONOSUPPORT" }, \ + /* NFSv4 native errors */ \ + { -NFS4ERR_ACCESS, "ACCESS" }, \ + { -NFS4ERR_ATTRNOTSUPP, "ATTRNOTSUPP" }, \ + { -NFS4ERR_ADMIN_REVOKED, "ADMIN_REVOKED" }, \ + { -NFS4ERR_BACK_CHAN_BUSY, "BACK_CHAN_BUSY" }, \ + { -NFS4ERR_BADCHAR, "BADCHAR" }, \ + { -NFS4ERR_BADHANDLE, "BADHANDLE" }, \ + { -NFS4ERR_BADIOMODE, "BADIOMODE" }, \ + { -NFS4ERR_BADLAYOUT, "BADLAYOUT" }, \ + { -NFS4ERR_BADLABEL, "BADLABEL" }, \ + { -NFS4ERR_BADNAME, "BADNAME" }, \ + { -NFS4ERR_BADOWNER, "BADOWNER" }, \ + { -NFS4ERR_BADSESSION, "BADSESSION" }, \ + { -NFS4ERR_BADSLOT, "BADSLOT" }, \ + { -NFS4ERR_BADTYPE, "BADTYPE" }, \ + { -NFS4ERR_BADXDR, "BADXDR" }, \ + { -NFS4ERR_BAD_COOKIE, "BAD_COOKIE" }, \ + { -NFS4ERR_BAD_HIGH_SLOT, "BAD_HIGH_SLOT" }, \ + { -NFS4ERR_BAD_RANGE, "BAD_RANGE" }, \ + { -NFS4ERR_BAD_SEQID, "BAD_SEQID" }, \ + { -NFS4ERR_BAD_SESSION_DIGEST, "BAD_SESSION_DIGEST" }, \ + { -NFS4ERR_BAD_STATEID, "BAD_STATEID" }, \ + { -NFS4ERR_CB_PATH_DOWN, "CB_PATH_DOWN" }, \ + { -NFS4ERR_CLID_INUSE, "CLID_INUSE" }, \ + { -NFS4ERR_CLIENTID_BUSY, "CLIENTID_BUSY" }, \ + { -NFS4ERR_COMPLETE_ALREADY, "COMPLETE_ALREADY" }, \ + { -NFS4ERR_CONN_NOT_BOUND_TO_SESSION, \ + "CONN_NOT_BOUND_TO_SESSION" }, \ + { -NFS4ERR_DEADLOCK, "DEADLOCK" }, \ + { -NFS4ERR_DEADSESSION, "DEAD_SESSION" }, \ + { -NFS4ERR_DELAY, "DELAY" }, \ + { -NFS4ERR_DELEG_ALREADY_WANTED, \ + "DELEG_ALREADY_WANTED" }, \ + { -NFS4ERR_DELEG_REVOKED, "DELEG_REVOKED" }, \ + { -NFS4ERR_DENIED, "DENIED" }, \ + { -NFS4ERR_DIRDELEG_UNAVAIL, "DIRDELEG_UNAVAIL" }, \ + { -NFS4ERR_DQUOT, "DQUOT" }, \ + { -NFS4ERR_ENCR_ALG_UNSUPP, "ENCR_ALG_UNSUPP" }, \ + { -NFS4ERR_EXIST, "EXIST" }, \ + { -NFS4ERR_EXPIRED, "EXPIRED" }, \ + { -NFS4ERR_FBIG, "FBIG" }, \ + { -NFS4ERR_FHEXPIRED, "FHEXPIRED" }, \ + { -NFS4ERR_FILE_OPEN, "FILE_OPEN" }, \ + { -NFS4ERR_GRACE, "GRACE" }, \ + { -NFS4ERR_HASH_ALG_UNSUPP, "HASH_ALG_UNSUPP" }, \ + { -NFS4ERR_INVAL, "INVAL" }, \ + { -NFS4ERR_IO, "IO" }, \ + { -NFS4ERR_ISDIR, "ISDIR" }, \ + { -NFS4ERR_LAYOUTTRYLATER, "LAYOUTTRYLATER" }, \ + { -NFS4ERR_LAYOUTUNAVAILABLE, "LAYOUTUNAVAILABLE" }, \ + { -NFS4ERR_LEASE_MOVED, "LEASE_MOVED" }, \ + { -NFS4ERR_LOCKED, "LOCKED" }, \ + { -NFS4ERR_LOCKS_HELD, "LOCKS_HELD" }, \ + { -NFS4ERR_LOCK_RANGE, "LOCK_RANGE" }, \ + { -NFS4ERR_MINOR_VERS_MISMATCH, "MINOR_VERS_MISMATCH" }, \ + { -NFS4ERR_MLINK, "MLINK" }, \ + { -NFS4ERR_MOVED, "MOVED" }, \ + { -NFS4ERR_NAMETOOLONG, "NAMETOOLONG" }, \ + { -NFS4ERR_NOENT, "NOENT" }, \ + { -NFS4ERR_NOFILEHANDLE, "NOFILEHANDLE" }, \ + { -NFS4ERR_NOMATCHING_LAYOUT, "NOMATCHING_LAYOUT" }, \ + { -NFS4ERR_NOSPC, "NOSPC" }, \ + { -NFS4ERR_NOTDIR, "NOTDIR" }, \ + { -NFS4ERR_NOTEMPTY, "NOTEMPTY" }, \ + { -NFS4ERR_NOTSUPP, "NOTSUPP" }, \ + { -NFS4ERR_NOT_ONLY_OP, "NOT_ONLY_OP" }, \ + { -NFS4ERR_NOT_SAME, "NOT_SAME" }, \ + { -NFS4ERR_NO_GRACE, "NO_GRACE" }, \ + { -NFS4ERR_NXIO, "NXIO" }, \ + { -NFS4ERR_OLD_STATEID, "OLD_STATEID" }, \ + { -NFS4ERR_OPENMODE, "OPENMODE" }, \ + { -NFS4ERR_OP_ILLEGAL, "OP_ILLEGAL" }, \ + { -NFS4ERR_OP_NOT_IN_SESSION, "OP_NOT_IN_SESSION" }, \ + { -NFS4ERR_PERM, "PERM" }, \ + { -NFS4ERR_PNFS_IO_HOLE, "PNFS_IO_HOLE" }, \ + { -NFS4ERR_PNFS_NO_LAYOUT, "PNFS_NO_LAYOUT" }, \ + { -NFS4ERR_RECALLCONFLICT, "RECALLCONFLICT" }, \ + { -NFS4ERR_RECLAIM_BAD, "RECLAIM_BAD" }, \ + { -NFS4ERR_RECLAIM_CONFLICT, "RECLAIM_CONFLICT" }, \ + { -NFS4ERR_REJECT_DELEG, "REJECT_DELEG" }, \ + { -NFS4ERR_REP_TOO_BIG, "REP_TOO_BIG" }, \ + { -NFS4ERR_REP_TOO_BIG_TO_CACHE, \ + "REP_TOO_BIG_TO_CACHE" }, \ + { -NFS4ERR_REQ_TOO_BIG, "REQ_TOO_BIG" }, \ + { -NFS4ERR_RESOURCE, "RESOURCE" }, \ + { -NFS4ERR_RESTOREFH, "RESTOREFH" }, \ + { -NFS4ERR_RETRY_UNCACHED_REP, "RETRY_UNCACHED_REP" }, \ + { -NFS4ERR_RETURNCONFLICT, "RETURNCONFLICT" }, \ + { -NFS4ERR_ROFS, "ROFS" }, \ + { -NFS4ERR_SAME, "SAME" }, \ + { -NFS4ERR_SHARE_DENIED, "SHARE_DENIED" }, \ + { -NFS4ERR_SEQUENCE_POS, "SEQUENCE_POS" }, \ + { -NFS4ERR_SEQ_FALSE_RETRY, "SEQ_FALSE_RETRY" }, \ + { -NFS4ERR_SEQ_MISORDERED, "SEQ_MISORDERED" }, \ + { -NFS4ERR_SERVERFAULT, "SERVERFAULT" }, \ + { -NFS4ERR_STALE, "STALE" }, \ + { -NFS4ERR_STALE_CLIENTID, "STALE_CLIENTID" }, \ + { -NFS4ERR_STALE_STATEID, "STALE_STATEID" }, \ + { -NFS4ERR_SYMLINK, "SYMLINK" }, \ + { -NFS4ERR_TOOSMALL, "TOOSMALL" }, \ + { -NFS4ERR_TOO_MANY_OPS, "TOO_MANY_OPS" }, \ + { -NFS4ERR_UNKNOWN_LAYOUTTYPE, "UNKNOWN_LAYOUTTYPE" }, \ + { -NFS4ERR_UNSAFE_COMPOUND, "UNSAFE_COMPOUND" }, \ + { -NFS4ERR_WRONGSEC, "WRONGSEC" }, \ + { -NFS4ERR_WRONG_CRED, "WRONG_CRED" }, \ + { -NFS4ERR_WRONG_TYPE, "WRONG_TYPE" }, \ + { -NFS4ERR_XDEV, "XDEV" }) + +#define show_open_flags(flags) \ + __print_flags(flags, "|", \ + { O_CREAT, "O_CREAT" }, \ + { O_EXCL, "O_EXCL" }, \ + { O_TRUNC, "O_TRUNC" }, \ + { O_DIRECT, "O_DIRECT" }) + +#define show_fmode_flags(mode) \ + __print_flags(mode, "|", \ + { ((__force unsigned long)FMODE_READ), "READ" }, \ + { ((__force unsigned long)FMODE_WRITE), "WRITE" }, \ + { ((__force unsigned long)FMODE_EXEC), "EXEC" }) + +#define show_nfs_fattr_flags(valid) \ + __print_flags((unsigned long)valid, "|", \ + { NFS_ATTR_FATTR_TYPE, "TYPE" }, \ + { NFS_ATTR_FATTR_MODE, "MODE" }, \ + { NFS_ATTR_FATTR_NLINK, "NLINK" }, \ + { NFS_ATTR_FATTR_OWNER, "OWNER" }, \ + { NFS_ATTR_FATTR_GROUP, "GROUP" }, \ + { NFS_ATTR_FATTR_RDEV, "RDEV" }, \ + { NFS_ATTR_FATTR_SIZE, "SIZE" }, \ + { NFS_ATTR_FATTR_FSID, "FSID" }, \ + { NFS_ATTR_FATTR_FILEID, "FILEID" }, \ + { NFS_ATTR_FATTR_ATIME, "ATIME" }, \ + { NFS_ATTR_FATTR_MTIME, "MTIME" }, \ + { NFS_ATTR_FATTR_CTIME, "CTIME" }, \ + { NFS_ATTR_FATTR_CHANGE, "CHANGE" }, \ + { NFS_ATTR_FATTR_OWNER_NAME, "OWNER_NAME" }, \ + { NFS_ATTR_FATTR_GROUP_NAME, "GROUP_NAME" }) + +DECLARE_EVENT_CLASS(nfs4_clientid_event, + TP_PROTO( + const struct nfs_client *clp, + int error + ), + + TP_ARGS(clp, error), + + TP_STRUCT__entry( + __string(dstaddr, + rpc_peeraddr2str(clp->cl_rpcclient, + RPC_DISPLAY_ADDR)) + __field(int, error) + ), + + TP_fast_assign( + __entry->error = error; + __assign_str(dstaddr, + rpc_peeraddr2str(clp->cl_rpcclient, + RPC_DISPLAY_ADDR)); + ), + + TP_printk( + "error=%d (%s) dstaddr=%s", + __entry->error, + show_nfsv4_errors(__entry->error), + __get_str(dstaddr) + ) +); +#define DEFINE_NFS4_CLIENTID_EVENT(name) \ + DEFINE_EVENT(nfs4_clientid_event, name, \ + TP_PROTO( \ + const struct nfs_client *clp, \ + int error \ + ), \ + TP_ARGS(clp, error)) +DEFINE_NFS4_CLIENTID_EVENT(nfs4_setclientid); +DEFINE_NFS4_CLIENTID_EVENT(nfs4_setclientid_confirm); +DEFINE_NFS4_CLIENTID_EVENT(nfs4_renew); +DEFINE_NFS4_CLIENTID_EVENT(nfs4_renew_async); +#ifdef CONFIG_NFS_V4_1 +DEFINE_NFS4_CLIENTID_EVENT(nfs4_exchange_id); +DEFINE_NFS4_CLIENTID_EVENT(nfs4_create_session); +DEFINE_NFS4_CLIENTID_EVENT(nfs4_destroy_session); +DEFINE_NFS4_CLIENTID_EVENT(nfs4_destroy_clientid); +DEFINE_NFS4_CLIENTID_EVENT(nfs4_bind_conn_to_session); +DEFINE_NFS4_CLIENTID_EVENT(nfs4_sequence); +DEFINE_NFS4_CLIENTID_EVENT(nfs4_reclaim_complete); + +TRACE_EVENT(nfs4_setup_sequence, + TP_PROTO( + const struct nfs4_session *session, + const struct nfs4_sequence_args *args + ), + TP_ARGS(session, args), + + TP_STRUCT__entry( + __field(unsigned int, session) + __field(unsigned int, slot_nr) + __field(unsigned int, seq_nr) + __field(unsigned int, highest_used_slotid) + ), + + TP_fast_assign( + const struct nfs4_slot *sa_slot = args->sa_slot; + __entry->session = nfs_session_id_hash(&session->sess_id); + __entry->slot_nr = sa_slot->slot_nr; + __entry->seq_nr = sa_slot->seq_nr; + __entry->highest_used_slotid = + sa_slot->table->highest_used_slotid; + ), + TP_printk( + "session=0x%08x slot_nr=%u seq_nr=%u " + "highest_used_slotid=%u", + __entry->session, + __entry->slot_nr, + __entry->seq_nr, + __entry->highest_used_slotid + ) +); + +#define show_nfs4_sequence_status_flags(status) \ + __print_flags((unsigned long)status, "|", \ + { SEQ4_STATUS_CB_PATH_DOWN, "CB_PATH_DOWN" }, \ + { SEQ4_STATUS_CB_GSS_CONTEXTS_EXPIRING, \ + "CB_GSS_CONTEXTS_EXPIRING" }, \ + { SEQ4_STATUS_CB_GSS_CONTEXTS_EXPIRED, \ + "CB_GSS_CONTEXTS_EXPIRED" }, \ + { SEQ4_STATUS_EXPIRED_ALL_STATE_REVOKED, \ + "EXPIRED_ALL_STATE_REVOKED" }, \ + { SEQ4_STATUS_EXPIRED_SOME_STATE_REVOKED, \ + "EXPIRED_SOME_STATE_REVOKED" }, \ + { SEQ4_STATUS_ADMIN_STATE_REVOKED, \ + "ADMIN_STATE_REVOKED" }, \ + { SEQ4_STATUS_RECALLABLE_STATE_REVOKED, \ + "RECALLABLE_STATE_REVOKED" }, \ + { SEQ4_STATUS_LEASE_MOVED, "LEASE_MOVED" }, \ + { SEQ4_STATUS_RESTART_RECLAIM_NEEDED, \ + "RESTART_RECLAIM_NEEDED" }, \ + { SEQ4_STATUS_CB_PATH_DOWN_SESSION, \ + "CB_PATH_DOWN_SESSION" }, \ + { SEQ4_STATUS_BACKCHANNEL_FAULT, \ + "BACKCHANNEL_FAULT" }) + +TRACE_EVENT(nfs4_sequence_done, + TP_PROTO( + const struct nfs4_session *session, + const struct nfs4_sequence_res *res + ), + TP_ARGS(session, res), + + TP_STRUCT__entry( + __field(unsigned int, session) + __field(unsigned int, slot_nr) + __field(unsigned int, seq_nr) + __field(unsigned int, highest_slotid) + __field(unsigned int, target_highest_slotid) + __field(unsigned int, status_flags) + __field(int, error) + ), + + TP_fast_assign( + const struct nfs4_slot *sr_slot = res->sr_slot; + __entry->session = nfs_session_id_hash(&session->sess_id); + __entry->slot_nr = sr_slot->slot_nr; + __entry->seq_nr = sr_slot->seq_nr; + __entry->highest_slotid = res->sr_highest_slotid; + __entry->target_highest_slotid = + res->sr_target_highest_slotid; + __entry->error = res->sr_status; + ), + TP_printk( + "error=%d (%s) session=0x%08x slot_nr=%u seq_nr=%u " + "highest_slotid=%u target_highest_slotid=%u " + "status_flags=%u (%s)", + __entry->error, + show_nfsv4_errors(__entry->error), + __entry->session, + __entry->slot_nr, + __entry->seq_nr, + __entry->highest_slotid, + __entry->target_highest_slotid, + __entry->status_flags, + show_nfs4_sequence_status_flags(__entry->status_flags) + ) +); + +struct cb_sequenceargs; +struct cb_sequenceres; + +TRACE_EVENT(nfs4_cb_sequence, + TP_PROTO( + const struct cb_sequenceargs *args, + const struct cb_sequenceres *res, + __be32 status + ), + TP_ARGS(args, res, status), + + TP_STRUCT__entry( + __field(unsigned int, session) + __field(unsigned int, slot_nr) + __field(unsigned int, seq_nr) + __field(unsigned int, highest_slotid) + __field(unsigned int, cachethis) + __field(int, error) + ), + + TP_fast_assign( + __entry->session = nfs_session_id_hash(&args->csa_sessionid); + __entry->slot_nr = args->csa_slotid; + __entry->seq_nr = args->csa_sequenceid; + __entry->highest_slotid = args->csa_highestslotid; + __entry->cachethis = args->csa_cachethis; + __entry->error = -be32_to_cpu(status); + ), + + TP_printk( + "error=%d (%s) session=0x%08x slot_nr=%u seq_nr=%u " + "highest_slotid=%u", + __entry->error, + show_nfsv4_errors(__entry->error), + __entry->session, + __entry->slot_nr, + __entry->seq_nr, + __entry->highest_slotid + ) +); +#endif /* CONFIG_NFS_V4_1 */ + +DECLARE_EVENT_CLASS(nfs4_open_event, + TP_PROTO( + const struct nfs_open_context *ctx, + int flags, + int error + ), + + TP_ARGS(ctx, flags, error), + + TP_STRUCT__entry( + __field(int, error) + __field(unsigned int, flags) + __field(unsigned int, fmode) + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(u64, dir) + __string(name, ctx->dentry->d_name.name) + ), + + TP_fast_assign( + const struct nfs4_state *state = ctx->state; + const struct inode *inode = NULL; + + __entry->error = error; + __entry->flags = flags; + __entry->fmode = (__force unsigned int)ctx->mode; + __entry->dev = ctx->dentry->d_sb->s_dev; + if (!IS_ERR(state)) + inode = state->inode; + if (inode != NULL) { + __entry->fileid = NFS_FILEID(inode); + __entry->fhandle = nfs_fhandle_hash(NFS_FH(inode)); + } else { + __entry->fileid = 0; + __entry->fhandle = 0; + } + __entry->dir = NFS_FILEID(ctx->dentry->d_parent->d_inode); + __assign_str(name, ctx->dentry->d_name.name); + ), + + TP_printk( + "error=%d (%s) flags=%d (%s) fmode=%s " + "fileid=%02x:%02x:%llu fhandle=0x%08x " + "name=%02x:%02x:%llu/%s", + __entry->error, + show_nfsv4_errors(__entry->error), + __entry->flags, + show_open_flags(__entry->flags), + show_fmode_flags(__entry->fmode), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->dir, + __get_str(name) + ) +); + +#define DEFINE_NFS4_OPEN_EVENT(name) \ + DEFINE_EVENT(nfs4_open_event, name, \ + TP_PROTO( \ + const struct nfs_open_context *ctx, \ + int flags, \ + int error \ + ), \ + TP_ARGS(ctx, flags, error)) +DEFINE_NFS4_OPEN_EVENT(nfs4_open_reclaim); +DEFINE_NFS4_OPEN_EVENT(nfs4_open_expired); +DEFINE_NFS4_OPEN_EVENT(nfs4_open_file); + +TRACE_EVENT(nfs4_close, + TP_PROTO( + const struct nfs4_state *state, + const struct nfs_closeargs *args, + const struct nfs_closeres *res, + int error + ), + + TP_ARGS(state, args, res, error), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(unsigned int, fmode) + __field(int, error) + ), + + TP_fast_assign( + const struct inode *inode = state->inode; + + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = NFS_FILEID(inode); + __entry->fhandle = nfs_fhandle_hash(NFS_FH(inode)); + __entry->fmode = (__force unsigned int)state->state; + __entry->error = error; + ), + + TP_printk( + "error=%d (%s) fmode=%s fileid=%02x:%02x:%llu " + "fhandle=0x%08x", + __entry->error, + show_nfsv4_errors(__entry->error), + __entry->fmode ? show_fmode_flags(__entry->fmode) : + "closed", + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle + ) +); + +#define show_lock_cmd(type) \ + __print_symbolic((int)type, \ + { F_GETLK, "GETLK" }, \ + { F_SETLK, "SETLK" }, \ + { F_SETLKW, "SETLKW" }) +#define show_lock_type(type) \ + __print_symbolic((int)type, \ + { F_RDLCK, "RDLCK" }, \ + { F_WRLCK, "WRLCK" }, \ + { F_UNLCK, "UNLCK" }) + +DECLARE_EVENT_CLASS(nfs4_lock_event, + TP_PROTO( + const struct file_lock *request, + const struct nfs4_state *state, + int cmd, + int error + ), + + TP_ARGS(request, state, cmd, error), + + TP_STRUCT__entry( + __field(int, error) + __field(int, cmd) + __field(char, type) + __field(loff_t, start) + __field(loff_t, end) + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + ), + + TP_fast_assign( + const struct inode *inode = state->inode; + + __entry->error = error; + __entry->cmd = cmd; + __entry->type = request->fl_type; + __entry->start = request->fl_start; + __entry->end = request->fl_end; + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = NFS_FILEID(inode); + __entry->fhandle = nfs_fhandle_hash(NFS_FH(inode)); + ), + + TP_printk( + "error=%d (%s) cmd=%s:%s range=%lld:%lld " + "fileid=%02x:%02x:%llu fhandle=0x%08x", + __entry->error, + show_nfsv4_errors(__entry->error), + show_lock_cmd(__entry->cmd), + show_lock_type(__entry->type), + (long long)__entry->start, + (long long)__entry->end, + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle + ) +); + +#define DEFINE_NFS4_LOCK_EVENT(name) \ + DEFINE_EVENT(nfs4_lock_event, name, \ + TP_PROTO( \ + const struct file_lock *request, \ + const struct nfs4_state *state, \ + int cmd, \ + int error \ + ), \ + TP_ARGS(request, state, cmd, error)) +DEFINE_NFS4_LOCK_EVENT(nfs4_get_lock); +DEFINE_NFS4_LOCK_EVENT(nfs4_set_lock); +DEFINE_NFS4_LOCK_EVENT(nfs4_lock_reclaim); +DEFINE_NFS4_LOCK_EVENT(nfs4_lock_expired); +DEFINE_NFS4_LOCK_EVENT(nfs4_unlock); + +DECLARE_EVENT_CLASS(nfs4_set_delegation_event, + TP_PROTO( + const struct inode *inode, + fmode_t fmode + ), + + TP_ARGS(inode, fmode), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(unsigned int, fmode) + ), + + TP_fast_assign( + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = NFS_FILEID(inode); + __entry->fhandle = nfs_fhandle_hash(NFS_FH(inode)); + __entry->fmode = (__force unsigned int)fmode; + ), + + TP_printk( + "fmode=%s fileid=%02x:%02x:%llu fhandle=0x%08x", + show_fmode_flags(__entry->fmode), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle + ) +); +#define DEFINE_NFS4_SET_DELEGATION_EVENT(name) \ + DEFINE_EVENT(nfs4_set_delegation_event, name, \ + TP_PROTO( \ + const struct inode *inode, \ + fmode_t fmode \ + ), \ + TP_ARGS(inode, fmode)) +DEFINE_NFS4_SET_DELEGATION_EVENT(nfs4_set_delegation); +DEFINE_NFS4_SET_DELEGATION_EVENT(nfs4_reclaim_delegation); + +TRACE_EVENT(nfs4_delegreturn_exit, + TP_PROTO( + const struct nfs4_delegreturnargs *args, + const struct nfs4_delegreturnres *res, + int error + ), + + TP_ARGS(args, res, error), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(int, error) + ), + + TP_fast_assign( + __entry->dev = res->server->s_dev; + __entry->fhandle = nfs_fhandle_hash(args->fhandle); + __entry->error = error; + ), + + TP_printk( + "error=%d (%s) dev=%02x:%02x fhandle=0x%08x", + __entry->error, + show_nfsv4_errors(__entry->error), + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->fhandle + ) +); + +#ifdef CONFIG_NFS_V4_1 +DECLARE_EVENT_CLASS(nfs4_test_stateid_event, + TP_PROTO( + const struct nfs4_state *state, + const struct nfs4_lock_state *lsp, + int error + ), + + TP_ARGS(state, lsp, error), + + TP_STRUCT__entry( + __field(int, error) + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + ), + + TP_fast_assign( + const struct inode *inode = state->inode; + + __entry->error = error; + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = NFS_FILEID(inode); + __entry->fhandle = nfs_fhandle_hash(NFS_FH(inode)); + ), + + TP_printk( + "error=%d (%s) fileid=%02x:%02x:%llu fhandle=0x%08x", + __entry->error, + show_nfsv4_errors(__entry->error), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle + ) +); + +#define DEFINE_NFS4_TEST_STATEID_EVENT(name) \ + DEFINE_EVENT(nfs4_test_stateid_event, name, \ + TP_PROTO( \ + const struct nfs4_state *state, \ + const struct nfs4_lock_state *lsp, \ + int error \ + ), \ + TP_ARGS(state, lsp, error)) +DEFINE_NFS4_TEST_STATEID_EVENT(nfs4_test_delegation_stateid); +DEFINE_NFS4_TEST_STATEID_EVENT(nfs4_test_open_stateid); +DEFINE_NFS4_TEST_STATEID_EVENT(nfs4_test_lock_stateid); +#endif /* CONFIG_NFS_V4_1 */ + +DECLARE_EVENT_CLASS(nfs4_lookup_event, + TP_PROTO( + const struct inode *dir, + const struct qstr *name, + int error + ), + + TP_ARGS(dir, name, error), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(int, error) + __field(u64, dir) + __string(name, name->name) + ), + + TP_fast_assign( + __entry->dev = dir->i_sb->s_dev; + __entry->dir = NFS_FILEID(dir); + __entry->error = error; + __assign_str(name, name->name); + ), + + TP_printk( + "error=%d (%s) name=%02x:%02x:%llu/%s", + __entry->error, + show_nfsv4_errors(__entry->error), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->dir, + __get_str(name) + ) +); + +#define DEFINE_NFS4_LOOKUP_EVENT(name) \ + DEFINE_EVENT(nfs4_lookup_event, name, \ + TP_PROTO( \ + const struct inode *dir, \ + const struct qstr *name, \ + int error \ + ), \ + TP_ARGS(dir, name, error)) + +DEFINE_NFS4_LOOKUP_EVENT(nfs4_lookup); +DEFINE_NFS4_LOOKUP_EVENT(nfs4_symlink); +DEFINE_NFS4_LOOKUP_EVENT(nfs4_mkdir); +DEFINE_NFS4_LOOKUP_EVENT(nfs4_mknod); +DEFINE_NFS4_LOOKUP_EVENT(nfs4_remove); +DEFINE_NFS4_LOOKUP_EVENT(nfs4_get_fs_locations); +DEFINE_NFS4_LOOKUP_EVENT(nfs4_secinfo); + +TRACE_EVENT(nfs4_rename, + TP_PROTO( + const struct inode *olddir, + const struct qstr *oldname, + const struct inode *newdir, + const struct qstr *newname, + int error + ), + + TP_ARGS(olddir, oldname, newdir, newname, error), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(int, error) + __field(u64, olddir) + __string(oldname, oldname->name) + __field(u64, newdir) + __string(newname, newname->name) + ), + + TP_fast_assign( + __entry->dev = olddir->i_sb->s_dev; + __entry->olddir = NFS_FILEID(olddir); + __entry->newdir = NFS_FILEID(newdir); + __entry->error = error; + __assign_str(oldname, oldname->name); + __assign_str(newname, newname->name); + ), + + TP_printk( + "error=%d (%s) oldname=%02x:%02x:%llu/%s " + "newname=%02x:%02x:%llu/%s", + __entry->error, + show_nfsv4_errors(__entry->error), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->olddir, + __get_str(oldname), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->newdir, + __get_str(newname) + ) +); + +DECLARE_EVENT_CLASS(nfs4_inode_event, + TP_PROTO( + const struct inode *inode, + int error + ), + + TP_ARGS(inode, error), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(int, error) + ), + + TP_fast_assign( + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = NFS_FILEID(inode); + __entry->fhandle = nfs_fhandle_hash(NFS_FH(inode)); + __entry->error = error; + ), + + TP_printk( + "error=%d (%s) fileid=%02x:%02x:%llu fhandle=0x%08x", + __entry->error, + show_nfsv4_errors(__entry->error), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle + ) +); + +#define DEFINE_NFS4_INODE_EVENT(name) \ + DEFINE_EVENT(nfs4_inode_event, name, \ + TP_PROTO( \ + const struct inode *inode, \ + int error \ + ), \ + TP_ARGS(inode, error)) + +DEFINE_NFS4_INODE_EVENT(nfs4_setattr); +DEFINE_NFS4_INODE_EVENT(nfs4_access); +DEFINE_NFS4_INODE_EVENT(nfs4_readlink); +DEFINE_NFS4_INODE_EVENT(nfs4_readdir); +DEFINE_NFS4_INODE_EVENT(nfs4_get_acl); +DEFINE_NFS4_INODE_EVENT(nfs4_set_acl); +#ifdef CONFIG_NFS_V4_SECURITY_LABEL +DEFINE_NFS4_INODE_EVENT(nfs4_get_security_label); +DEFINE_NFS4_INODE_EVENT(nfs4_set_security_label); +#endif /* CONFIG_NFS_V4_SECURITY_LABEL */ +DEFINE_NFS4_INODE_EVENT(nfs4_recall_delegation); +DEFINE_NFS4_INODE_EVENT(nfs4_delegreturn); + +DECLARE_EVENT_CLASS(nfs4_getattr_event, + TP_PROTO( + const struct nfs_server *server, + const struct nfs_fh *fhandle, + const struct nfs_fattr *fattr, + int error + ), + + TP_ARGS(server, fhandle, fattr, error), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(unsigned int, valid) + __field(int, error) + ), + + TP_fast_assign( + __entry->dev = server->s_dev; + __entry->valid = fattr->valid; + __entry->fhandle = nfs_fhandle_hash(fhandle); + __entry->fileid = (fattr->valid & NFS_ATTR_FATTR_FILEID) ? fattr->fileid : 0; + __entry->error = error; + ), + + TP_printk( + "error=%d (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " + "valid=%s", + __entry->error, + show_nfsv4_errors(__entry->error), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + show_nfs_fattr_flags(__entry->valid) + ) +); + +#define DEFINE_NFS4_GETATTR_EVENT(name) \ + DEFINE_EVENT(nfs4_getattr_event, name, \ + TP_PROTO( \ + const struct nfs_server *server, \ + const struct nfs_fh *fhandle, \ + const struct nfs_fattr *fattr, \ + int error \ + ), \ + TP_ARGS(server, fhandle, fattr, error)) +DEFINE_NFS4_GETATTR_EVENT(nfs4_getattr); +DEFINE_NFS4_GETATTR_EVENT(nfs4_lookup_root); +DEFINE_NFS4_GETATTR_EVENT(nfs4_fsinfo); + +DECLARE_EVENT_CLASS(nfs4_idmap_event, + TP_PROTO( + const char *name, + int len, + u32 id, + int error + ), + + TP_ARGS(name, len, id, error), + + TP_STRUCT__entry( + __field(int, error) + __field(u32, id) + __dynamic_array(char, name, len > 0 ? len + 1 : 1) + ), + + TP_fast_assign( + if (len < 0) + len = 0; + __entry->error = error < 0 ? error : 0; + __entry->id = id; + memcpy(__get_dynamic_array(name), name, len); + ((char *)__get_dynamic_array(name))[len] = 0; + ), + + TP_printk( + "error=%d id=%u name=%s", + __entry->error, + __entry->id, + __get_str(name) + ) +); +#define DEFINE_NFS4_IDMAP_EVENT(name) \ + DEFINE_EVENT(nfs4_idmap_event, name, \ + TP_PROTO( \ + const char *name, \ + int len, \ + u32 id, \ + int error \ + ), \ + TP_ARGS(name, len, id, error)) +DEFINE_NFS4_IDMAP_EVENT(nfs4_map_name_to_uid); +DEFINE_NFS4_IDMAP_EVENT(nfs4_map_group_to_gid); +DEFINE_NFS4_IDMAP_EVENT(nfs4_map_uid_to_name); +DEFINE_NFS4_IDMAP_EVENT(nfs4_map_gid_to_group); + +DECLARE_EVENT_CLASS(nfs4_read_event, + TP_PROTO( + const struct nfs_read_data *data, + int error + ), + + TP_ARGS(data, error), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(loff_t, offset) + __field(size_t, count) + __field(int, error) + ), + + TP_fast_assign( + const struct inode *inode = data->header->inode; + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = NFS_FILEID(inode); + __entry->fhandle = nfs_fhandle_hash(NFS_FH(inode)); + __entry->offset = data->args.offset; + __entry->count = data->args.count; + __entry->error = error; + ), + + TP_printk( + "error=%d (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " + "offset=%lld count=%zu", + __entry->error, + show_nfsv4_errors(__entry->error), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + (long long)__entry->offset, + __entry->count + ) +); +#define DEFINE_NFS4_READ_EVENT(name) \ + DEFINE_EVENT(nfs4_read_event, name, \ + TP_PROTO( \ + const struct nfs_read_data *data, \ + int error \ + ), \ + TP_ARGS(data, error)) +DEFINE_NFS4_READ_EVENT(nfs4_read); +#ifdef CONFIG_NFS_V4_1 +DEFINE_NFS4_READ_EVENT(nfs4_pnfs_read); +#endif /* CONFIG_NFS_V4_1 */ + +DECLARE_EVENT_CLASS(nfs4_write_event, + TP_PROTO( + const struct nfs_write_data *data, + int error + ), + + TP_ARGS(data, error), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(loff_t, offset) + __field(size_t, count) + __field(int, error) + ), + + TP_fast_assign( + const struct inode *inode = data->header->inode; + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = NFS_FILEID(inode); + __entry->fhandle = nfs_fhandle_hash(NFS_FH(inode)); + __entry->offset = data->args.offset; + __entry->count = data->args.count; + __entry->error = error; + ), + + TP_printk( + "error=%d (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " + "offset=%lld count=%zu", + __entry->error, + show_nfsv4_errors(__entry->error), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + (long long)__entry->offset, + __entry->count + ) +); + +#define DEFINE_NFS4_WRITE_EVENT(name) \ + DEFINE_EVENT(nfs4_write_event, name, \ + TP_PROTO( \ + const struct nfs_write_data *data, \ + int error \ + ), \ + TP_ARGS(data, error)) +DEFINE_NFS4_WRITE_EVENT(nfs4_write); +#ifdef CONFIG_NFS_V4_1 +DEFINE_NFS4_WRITE_EVENT(nfs4_pnfs_write); +#endif /* CONFIG_NFS_V4_1 */ + +DECLARE_EVENT_CLASS(nfs4_commit_event, + TP_PROTO( + const struct nfs_commit_data *data, + int error + ), + + TP_ARGS(data, error), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(loff_t, offset) + __field(size_t, count) + __field(int, error) + ), + + TP_fast_assign( + const struct inode *inode = data->inode; + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = NFS_FILEID(inode); + __entry->fhandle = nfs_fhandle_hash(NFS_FH(inode)); + __entry->offset = data->args.offset; + __entry->count = data->args.count; + __entry->error = error; + ), + + TP_printk( + "error=%d (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " + "offset=%lld count=%zu", + __entry->error, + show_nfsv4_errors(__entry->error), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + (long long)__entry->offset, + __entry->count + ) +); +#define DEFINE_NFS4_COMMIT_EVENT(name) \ + DEFINE_EVENT(nfs4_commit_event, name, \ + TP_PROTO( \ + const struct nfs_commit_data *data, \ + int error \ + ), \ + TP_ARGS(data, error)) +DEFINE_NFS4_COMMIT_EVENT(nfs4_commit); +#ifdef CONFIG_NFS_V4_1 +DEFINE_NFS4_COMMIT_EVENT(nfs4_pnfs_commit_ds); + +#define show_pnfs_iomode(iomode) \ + __print_symbolic(iomode, \ + { IOMODE_READ, "READ" }, \ + { IOMODE_RW, "RW" }, \ + { IOMODE_ANY, "ANY" }) + +TRACE_EVENT(nfs4_layoutget, + TP_PROTO( + const struct nfs_open_context *ctx, + const struct pnfs_layout_range *args, + const struct pnfs_layout_range *res, + int error + ), + + TP_ARGS(ctx, args, res, error), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(u32, iomode) + __field(u64, offset) + __field(u64, count) + __field(int, error) + ), + + TP_fast_assign( + const struct inode *inode = ctx->dentry->d_inode; + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = NFS_FILEID(inode); + __entry->fhandle = nfs_fhandle_hash(NFS_FH(inode)); + __entry->iomode = args->iomode; + __entry->offset = args->offset; + __entry->count = args->length; + __entry->error = error; + ), + + TP_printk( + "error=%d (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " + "iomode=%s offset=%llu count=%llu", + __entry->error, + show_nfsv4_errors(__entry->error), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + show_pnfs_iomode(__entry->iomode), + (unsigned long long)__entry->offset, + (unsigned long long)__entry->count + ) +); + +DEFINE_NFS4_INODE_EVENT(nfs4_layoutcommit); +DEFINE_NFS4_INODE_EVENT(nfs4_layoutreturn); + +#endif /* CONFIG_NFS_V4_1 */ + +#endif /* _TRACE_NFS4_H */ + +#undef TRACE_INCLUDE_PATH +#define TRACE_INCLUDE_PATH . +#define TRACE_INCLUDE_FILE nfs4trace +/* This part must be outside protection */ +#include <trace/define_trace.h> diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index 3850b01..fbdad9e 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -294,7 +294,9 @@ static int nfs4_stat_to_errno(int); XDR_QUADLEN(NFS4_EXCHANGE_ID_LEN) + \ 1 /* flags */ + \ 1 /* spa_how */ + \ - 0 /* SP4_NONE (for now) */ + \ + /* max is SP4_MACH_CRED (for now) */ + \ + 1 + NFS4_OP_MAP_NUM_WORDS + \ + 1 + NFS4_OP_MAP_NUM_WORDS + \ 1 /* implementation id array of size 1 */ + \ 1 /* nii_domain */ + \ XDR_QUADLEN(NFS4_OPAQUE_LIMIT) + \ @@ -306,7 +308,9 @@ static int nfs4_stat_to_errno(int); 1 /* eir_sequenceid */ + \ 1 /* eir_flags */ + \ 1 /* spr_how */ + \ - 0 /* SP4_NONE (for now) */ + \ + /* max is SP4_MACH_CRED (for now) */ + \ + 1 + NFS4_OP_MAP_NUM_WORDS + \ + 1 + NFS4_OP_MAP_NUM_WORDS + \ 2 /* eir_server_owner.so_minor_id */ + \ /* eir_server_owner.so_major_id<> */ \ XDR_QUADLEN(NFS4_OPAQUE_LIMIT) + 1 + \ @@ -997,12 +1001,10 @@ static void encode_attrs(struct xdr_stream *xdr, const struct iattr *iap, int owner_namelen = 0; int owner_grouplen = 0; __be32 *p; - __be32 *q; - int len; - uint32_t bmval_len = 2; - uint32_t bmval0 = 0; - uint32_t bmval1 = 0; - uint32_t bmval2 = 0; + unsigned i; + uint32_t len = 0; + uint32_t bmval_len; + uint32_t bmval[3] = { 0 }; /* * We reserve enough space to write the entire attribute buffer at once. @@ -1011,13 +1013,14 @@ static void encode_attrs(struct xdr_stream *xdr, const struct iattr *iap, * = 40 bytes, plus any contribution from variable-length fields * such as owner/group. */ - len = 8; - - /* Sigh */ - if (iap->ia_valid & ATTR_SIZE) + if (iap->ia_valid & ATTR_SIZE) { + bmval[0] |= FATTR4_WORD0_SIZE; len += 8; - if (iap->ia_valid & ATTR_MODE) + } + if (iap->ia_valid & ATTR_MODE) { + bmval[1] |= FATTR4_WORD1_MODE; len += 4; + } if (iap->ia_valid & ATTR_UID) { owner_namelen = nfs_map_uid_to_name(server, iap->ia_uid, owner_name, IDMAP_NAMESZ); if (owner_namelen < 0) { @@ -1028,6 +1031,7 @@ static void encode_attrs(struct xdr_stream *xdr, const struct iattr *iap, owner_namelen = sizeof("nobody") - 1; /* goto out; */ } + bmval[1] |= FATTR4_WORD1_OWNER; len += 4 + (XDR_QUADLEN(owner_namelen) << 2); } if (iap->ia_valid & ATTR_GID) { @@ -1039,92 +1043,73 @@ static void encode_attrs(struct xdr_stream *xdr, const struct iattr *iap, owner_grouplen = sizeof("nobody") - 1; /* goto out; */ } + bmval[1] |= FATTR4_WORD1_OWNER_GROUP; len += 4 + (XDR_QUADLEN(owner_grouplen) << 2); } - if (iap->ia_valid & ATTR_ATIME_SET) + if (iap->ia_valid & ATTR_ATIME_SET) { + bmval[1] |= FATTR4_WORD1_TIME_ACCESS_SET; len += 16; - else if (iap->ia_valid & ATTR_ATIME) + } else if (iap->ia_valid & ATTR_ATIME) { + bmval[1] |= FATTR4_WORD1_TIME_ACCESS_SET; len += 4; - if (iap->ia_valid & ATTR_MTIME_SET) + } + if (iap->ia_valid & ATTR_MTIME_SET) { + bmval[1] |= FATTR4_WORD1_TIME_MODIFY_SET; len += 16; - else if (iap->ia_valid & ATTR_MTIME) + } else if (iap->ia_valid & ATTR_MTIME) { + bmval[1] |= FATTR4_WORD1_TIME_MODIFY_SET; len += 4; + } if (label) { len += 4 + 4 + 4 + (XDR_QUADLEN(label->len) << 2); - bmval_len = 3; + bmval[2] |= FATTR4_WORD2_SECURITY_LABEL; } - len += bmval_len << 2; - p = reserve_space(xdr, len); + if (bmval[2] != 0) + bmval_len = 3; + else if (bmval[1] != 0) + bmval_len = 2; + else + bmval_len = 1; + + p = reserve_space(xdr, 4 + (bmval_len << 2) + 4 + len); - /* - * We write the bitmap length now, but leave the bitmap and the attribute - * buffer length to be backfilled at the end of this routine. - */ *p++ = cpu_to_be32(bmval_len); - q = p; - /* Skip bitmap entries + attrlen */ - p += bmval_len + 1; + for (i = 0; i < bmval_len; i++) + *p++ = cpu_to_be32(bmval[i]); + *p++ = cpu_to_be32(len); - if (iap->ia_valid & ATTR_SIZE) { - bmval0 |= FATTR4_WORD0_SIZE; + if (bmval[0] & FATTR4_WORD0_SIZE) p = xdr_encode_hyper(p, iap->ia_size); - } - if (iap->ia_valid & ATTR_MODE) { - bmval1 |= FATTR4_WORD1_MODE; + if (bmval[1] & FATTR4_WORD1_MODE) *p++ = cpu_to_be32(iap->ia_mode & S_IALLUGO); - } - if (iap->ia_valid & ATTR_UID) { - bmval1 |= FATTR4_WORD1_OWNER; + if (bmval[1] & FATTR4_WORD1_OWNER) p = xdr_encode_opaque(p, owner_name, owner_namelen); - } - if (iap->ia_valid & ATTR_GID) { - bmval1 |= FATTR4_WORD1_OWNER_GROUP; + if (bmval[1] & FATTR4_WORD1_OWNER_GROUP) p = xdr_encode_opaque(p, owner_group, owner_grouplen); + if (bmval[1] & FATTR4_WORD1_TIME_ACCESS_SET) { + if (iap->ia_valid & ATTR_ATIME_SET) { + *p++ = cpu_to_be32(NFS4_SET_TO_CLIENT_TIME); + p = xdr_encode_hyper(p, (s64)iap->ia_atime.tv_sec); + *p++ = cpu_to_be32(iap->ia_atime.tv_nsec); + } else + *p++ = cpu_to_be32(NFS4_SET_TO_SERVER_TIME); } - if (iap->ia_valid & ATTR_ATIME_SET) { - bmval1 |= FATTR4_WORD1_TIME_ACCESS_SET; - *p++ = cpu_to_be32(NFS4_SET_TO_CLIENT_TIME); - p = xdr_encode_hyper(p, (s64)iap->ia_atime.tv_sec); - *p++ = cpu_to_be32(iap->ia_atime.tv_nsec); - } - else if (iap->ia_valid & ATTR_ATIME) { - bmval1 |= FATTR4_WORD1_TIME_ACCESS_SET; - *p++ = cpu_to_be32(NFS4_SET_TO_SERVER_TIME); - } - if (iap->ia_valid & ATTR_MTIME_SET) { - bmval1 |= FATTR4_WORD1_TIME_MODIFY_SET; - *p++ = cpu_to_be32(NFS4_SET_TO_CLIENT_TIME); - p = xdr_encode_hyper(p, (s64)iap->ia_mtime.tv_sec); - *p++ = cpu_to_be32(iap->ia_mtime.tv_nsec); - } - else if (iap->ia_valid & ATTR_MTIME) { - bmval1 |= FATTR4_WORD1_TIME_MODIFY_SET; - *p++ = cpu_to_be32(NFS4_SET_TO_SERVER_TIME); + if (bmval[1] & FATTR4_WORD1_TIME_MODIFY_SET) { + if (iap->ia_valid & ATTR_MTIME_SET) { + *p++ = cpu_to_be32(NFS4_SET_TO_CLIENT_TIME); + p = xdr_encode_hyper(p, (s64)iap->ia_mtime.tv_sec); + *p++ = cpu_to_be32(iap->ia_mtime.tv_nsec); + } else + *p++ = cpu_to_be32(NFS4_SET_TO_SERVER_TIME); } - if (label) { - bmval2 |= FATTR4_WORD2_SECURITY_LABEL; + if (bmval[2] & FATTR4_WORD2_SECURITY_LABEL) { *p++ = cpu_to_be32(label->lfs); *p++ = cpu_to_be32(label->pi); *p++ = cpu_to_be32(label->len); p = xdr_encode_opaque_fixed(p, label->label, label->len); } - /* - * Now we backfill the bitmap and the attribute buffer length. - */ - if (len != ((char *)p - (char *)q) + 4) { - printk(KERN_ERR "NFS: Attr length error, %u != %Zu\n", - len, ((char *)p - (char *)q) + 4); - BUG(); - } - *q++ = htonl(bmval0); - *q++ = htonl(bmval1); - if (bmval_len == 3) - *q++ = htonl(bmval2); - len = (char *)p - (char *)(q + 1); - *q = htonl(len); - /* out: */ } @@ -1745,6 +1730,14 @@ static void encode_bind_conn_to_session(struct xdr_stream *xdr, *p = 0; /* use_conn_in_rdma_mode = False */ } +static void encode_op_map(struct xdr_stream *xdr, struct nfs4_op_map *op_map) +{ + unsigned int i; + encode_uint32(xdr, NFS4_OP_MAP_NUM_WORDS); + for (i = 0; i < NFS4_OP_MAP_NUM_WORDS; i++) + encode_uint32(xdr, op_map->u.words[i]); +} + static void encode_exchange_id(struct xdr_stream *xdr, struct nfs41_exchange_id_args *args, struct compound_hdr *hdr) @@ -1758,9 +1751,20 @@ static void encode_exchange_id(struct xdr_stream *xdr, encode_string(xdr, args->id_len, args->id); - p = reserve_space(xdr, 12); - *p++ = cpu_to_be32(args->flags); - *p++ = cpu_to_be32(0); /* zero length state_protect4_a */ + encode_uint32(xdr, args->flags); + encode_uint32(xdr, args->state_protect.how); + + switch (args->state_protect.how) { + case SP4_NONE: + break; + case SP4_MACH_CRED: + encode_op_map(xdr, &args->state_protect.enforce); + encode_op_map(xdr, &args->state_protect.allow); + break; + default: + WARN_ON_ONCE(1); + break; + } if (send_implementation_id && sizeof(CONFIG_NFS_V4_1_IMPLEMENTATION_ID_DOMAIN) > 1 && @@ -1771,7 +1775,7 @@ static void encode_exchange_id(struct xdr_stream *xdr, utsname()->version, utsname()->machine); if (len > 0) { - *p = cpu_to_be32(1); /* implementation id array length=1 */ + encode_uint32(xdr, 1); /* implementation id array length=1 */ encode_string(xdr, sizeof(CONFIG_NFS_V4_1_IMPLEMENTATION_ID_DOMAIN) - 1, @@ -1782,7 +1786,7 @@ static void encode_exchange_id(struct xdr_stream *xdr, p = xdr_encode_hyper(p, 0); *p = cpu_to_be32(0); } else - *p = cpu_to_be32(0); /* implementation id array length=0 */ + encode_uint32(xdr, 0); /* implementation id array length=0 */ } static void encode_create_session(struct xdr_stream *xdr, @@ -1835,7 +1839,7 @@ static void encode_create_session(struct xdr_stream *xdr, *p++ = cpu_to_be32(RPC_AUTH_UNIX); /* auth_sys */ /* authsys_parms rfc1831 */ - *p++ = (__be32)nn->boot_time.tv_nsec; /* stamp */ + *p++ = cpu_to_be32(nn->boot_time.tv_nsec); /* stamp */ p = xdr_encode_opaque(p, machine_name, len); *p++ = cpu_to_be32(0); /* UID */ *p++ = cpu_to_be32(0); /* GID */ @@ -1877,11 +1881,10 @@ static void encode_sequence(struct xdr_stream *xdr, struct nfs4_slot *slot = args->sa_slot; __be32 *p; - if (slot == NULL) - return; - tp = slot->table; session = tp->session; + if (!session) + return; encode_op_hdr(xdr, OP_SEQUENCE, decode_sequence_maxsz, hdr); @@ -2062,9 +2065,9 @@ static void encode_free_stateid(struct xdr_stream *xdr, static u32 nfs4_xdr_minorversion(const struct nfs4_sequence_args *args) { #if defined(CONFIG_NFS_V4_1) - - if (args->sa_slot) - return args->sa_slot->table->session->clp->cl_mvops->minor_version; + struct nfs4_session *session = args->sa_slot->table->session; + if (session) + return session->clp->cl_mvops->minor_version; #endif /* CONFIG_NFS_V4_1 */ return 0; } @@ -4649,7 +4652,7 @@ static int decode_getfattr(struct xdr_stream *xdr, struct nfs_fattr *fattr, static int decode_first_pnfs_layout_type(struct xdr_stream *xdr, uint32_t *layouttype) { - uint32_t *p; + __be32 *p; int num; p = xdr_inline_decode(xdr, 4); @@ -5394,6 +5397,23 @@ static int decode_secinfo_no_name(struct xdr_stream *xdr, struct nfs4_secinfo_re return decode_secinfo_common(xdr, res); } +static int decode_op_map(struct xdr_stream *xdr, struct nfs4_op_map *op_map) +{ + __be32 *p; + uint32_t bitmap_words; + unsigned int i; + + p = xdr_inline_decode(xdr, 4); + bitmap_words = be32_to_cpup(p++); + if (bitmap_words > NFS4_OP_MAP_NUM_WORDS) + return -EIO; + p = xdr_inline_decode(xdr, 4 * bitmap_words); + for (i = 0; i < bitmap_words; i++) + op_map->u.words[i] = be32_to_cpup(p++); + + return 0; +} + static int decode_exchange_id(struct xdr_stream *xdr, struct nfs41_exchange_id_res *res) { @@ -5417,10 +5437,22 @@ static int decode_exchange_id(struct xdr_stream *xdr, res->seqid = be32_to_cpup(p++); res->flags = be32_to_cpup(p++); - /* We ask for SP4_NONE */ - dummy = be32_to_cpup(p); - if (dummy != SP4_NONE) + res->state_protect.how = be32_to_cpup(p); + switch (res->state_protect.how) { + case SP4_NONE: + break; + case SP4_MACH_CRED: + status = decode_op_map(xdr, &res->state_protect.enforce); + if (status) + return status; + status = decode_op_map(xdr, &res->state_protect.allow); + if (status) + return status; + break; + default: + WARN_ON_ONCE(1); return -EIO; + } /* server_owner4.so_minor_id */ p = xdr_inline_decode(xdr, 8); @@ -5614,6 +5646,8 @@ static int decode_sequence(struct xdr_stream *xdr, if (res->sr_slot == NULL) return 0; + if (!res->sr_slot->table->session) + return 0; status = decode_op_hdr(xdr, OP_SEQUENCE); if (!status) diff --git a/fs/nfs/nfstrace.c b/fs/nfs/nfstrace.c new file mode 100644 index 0000000..4eb0aea --- /dev/null +++ b/fs/nfs/nfstrace.c @@ -0,0 +1,9 @@ +/* + * Copyright (c) 2013 Trond Myklebust <Trond.Myklebust@netapp.com> + */ +#include <linux/nfs_fs.h> +#include <linux/namei.h> +#include "internal.h" + +#define CREATE_TRACE_POINTS +#include "nfstrace.h" diff --git a/fs/nfs/nfstrace.h b/fs/nfs/nfstrace.h new file mode 100644 index 0000000..89fe741 --- /dev/null +++ b/fs/nfs/nfstrace.h @@ -0,0 +1,729 @@ +/* + * Copyright (c) 2013 Trond Myklebust <Trond.Myklebust@netapp.com> + */ +#undef TRACE_SYSTEM +#define TRACE_SYSTEM nfs + +#if !defined(_TRACE_NFS_H) || defined(TRACE_HEADER_MULTI_READ) +#define _TRACE_NFS_H + +#include <linux/tracepoint.h> + +#define nfs_show_file_type(ftype) \ + __print_symbolic(ftype, \ + { DT_UNKNOWN, "UNKNOWN" }, \ + { DT_FIFO, "FIFO" }, \ + { DT_CHR, "CHR" }, \ + { DT_DIR, "DIR" }, \ + { DT_BLK, "BLK" }, \ + { DT_REG, "REG" }, \ + { DT_LNK, "LNK" }, \ + { DT_SOCK, "SOCK" }, \ + { DT_WHT, "WHT" }) + +#define nfs_show_cache_validity(v) \ + __print_flags(v, "|", \ + { NFS_INO_INVALID_ATTR, "INVALID_ATTR" }, \ + { NFS_INO_INVALID_DATA, "INVALID_DATA" }, \ + { NFS_INO_INVALID_ATIME, "INVALID_ATIME" }, \ + { NFS_INO_INVALID_ACCESS, "INVALID_ACCESS" }, \ + { NFS_INO_INVALID_ACL, "INVALID_ACL" }, \ + { NFS_INO_REVAL_PAGECACHE, "REVAL_PAGECACHE" }, \ + { NFS_INO_REVAL_FORCED, "REVAL_FORCED" }, \ + { NFS_INO_INVALID_LABEL, "INVALID_LABEL" }) + +#define nfs_show_nfsi_flags(v) \ + __print_flags(v, "|", \ + { 1 << NFS_INO_ADVISE_RDPLUS, "ADVISE_RDPLUS" }, \ + { 1 << NFS_INO_STALE, "STALE" }, \ + { 1 << NFS_INO_FLUSHING, "FLUSHING" }, \ + { 1 << NFS_INO_FSCACHE, "FSCACHE" }, \ + { 1 << NFS_INO_COMMIT, "COMMIT" }, \ + { 1 << NFS_INO_LAYOUTCOMMIT, "NEED_LAYOUTCOMMIT" }, \ + { 1 << NFS_INO_LAYOUTCOMMITTING, "LAYOUTCOMMIT" }) + +DECLARE_EVENT_CLASS(nfs_inode_event, + TP_PROTO( + const struct inode *inode + ), + + TP_ARGS(inode), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(u64, version) + ), + + TP_fast_assign( + const struct nfs_inode *nfsi = NFS_I(inode); + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = nfsi->fileid; + __entry->fhandle = nfs_fhandle_hash(&nfsi->fh); + __entry->version = inode->i_version; + ), + + TP_printk( + "fileid=%02x:%02x:%llu fhandle=0x%08x version=%llu ", + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + (unsigned long long)__entry->version + ) +); + +DECLARE_EVENT_CLASS(nfs_inode_event_done, + TP_PROTO( + const struct inode *inode, + int error + ), + + TP_ARGS(inode, error), + + TP_STRUCT__entry( + __field(int, error) + __field(dev_t, dev) + __field(u32, fhandle) + __field(unsigned char, type) + __field(u64, fileid) + __field(u64, version) + __field(loff_t, size) + __field(unsigned long, nfsi_flags) + __field(unsigned long, cache_validity) + ), + + TP_fast_assign( + const struct nfs_inode *nfsi = NFS_I(inode); + __entry->error = error; + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = nfsi->fileid; + __entry->fhandle = nfs_fhandle_hash(&nfsi->fh); + __entry->type = nfs_umode_to_dtype(inode->i_mode); + __entry->version = inode->i_version; + __entry->size = i_size_read(inode); + __entry->nfsi_flags = nfsi->flags; + __entry->cache_validity = nfsi->cache_validity; + ), + + TP_printk( + "error=%d fileid=%02x:%02x:%llu fhandle=0x%08x " + "type=%u (%s) version=%llu size=%lld " + "cache_validity=%lu (%s) nfs_flags=%ld (%s)", + __entry->error, + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + __entry->type, + nfs_show_file_type(__entry->type), + (unsigned long long)__entry->version, + (long long)__entry->size, + __entry->cache_validity, + nfs_show_cache_validity(__entry->cache_validity), + __entry->nfsi_flags, + nfs_show_nfsi_flags(__entry->nfsi_flags) + ) +); + +#define DEFINE_NFS_INODE_EVENT(name) \ + DEFINE_EVENT(nfs_inode_event, name, \ + TP_PROTO( \ + const struct inode *inode \ + ), \ + TP_ARGS(inode)) +#define DEFINE_NFS_INODE_EVENT_DONE(name) \ + DEFINE_EVENT(nfs_inode_event_done, name, \ + TP_PROTO( \ + const struct inode *inode, \ + int error \ + ), \ + TP_ARGS(inode, error)) +DEFINE_NFS_INODE_EVENT(nfs_refresh_inode_enter); +DEFINE_NFS_INODE_EVENT_DONE(nfs_refresh_inode_exit); +DEFINE_NFS_INODE_EVENT(nfs_revalidate_inode_enter); +DEFINE_NFS_INODE_EVENT_DONE(nfs_revalidate_inode_exit); +DEFINE_NFS_INODE_EVENT(nfs_invalidate_mapping_enter); +DEFINE_NFS_INODE_EVENT_DONE(nfs_invalidate_mapping_exit); +DEFINE_NFS_INODE_EVENT(nfs_getattr_enter); +DEFINE_NFS_INODE_EVENT_DONE(nfs_getattr_exit); +DEFINE_NFS_INODE_EVENT(nfs_setattr_enter); +DEFINE_NFS_INODE_EVENT_DONE(nfs_setattr_exit); +DEFINE_NFS_INODE_EVENT(nfs_writeback_page_enter); +DEFINE_NFS_INODE_EVENT_DONE(nfs_writeback_page_exit); +DEFINE_NFS_INODE_EVENT(nfs_writeback_inode_enter); +DEFINE_NFS_INODE_EVENT_DONE(nfs_writeback_inode_exit); +DEFINE_NFS_INODE_EVENT(nfs_fsync_enter); +DEFINE_NFS_INODE_EVENT_DONE(nfs_fsync_exit); +DEFINE_NFS_INODE_EVENT(nfs_access_enter); +DEFINE_NFS_INODE_EVENT_DONE(nfs_access_exit); + +#define show_lookup_flags(flags) \ + __print_flags((unsigned long)flags, "|", \ + { LOOKUP_AUTOMOUNT, "AUTOMOUNT" }, \ + { LOOKUP_DIRECTORY, "DIRECTORY" }, \ + { LOOKUP_OPEN, "OPEN" }, \ + { LOOKUP_CREATE, "CREATE" }, \ + { LOOKUP_EXCL, "EXCL" }) + +DECLARE_EVENT_CLASS(nfs_lookup_event, + TP_PROTO( + const struct inode *dir, + const struct dentry *dentry, + unsigned int flags + ), + + TP_ARGS(dir, dentry, flags), + + TP_STRUCT__entry( + __field(unsigned int, flags) + __field(dev_t, dev) + __field(u64, dir) + __string(name, dentry->d_name.name) + ), + + TP_fast_assign( + __entry->dev = dir->i_sb->s_dev; + __entry->dir = NFS_FILEID(dir); + __entry->flags = flags; + __assign_str(name, dentry->d_name.name); + ), + + TP_printk( + "flags=%u (%s) name=%02x:%02x:%llu/%s", + __entry->flags, + show_lookup_flags(__entry->flags), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->dir, + __get_str(name) + ) +); + +#define DEFINE_NFS_LOOKUP_EVENT(name) \ + DEFINE_EVENT(nfs_lookup_event, name, \ + TP_PROTO( \ + const struct inode *dir, \ + const struct dentry *dentry, \ + unsigned int flags \ + ), \ + TP_ARGS(dir, dentry, flags)) + +DECLARE_EVENT_CLASS(nfs_lookup_event_done, + TP_PROTO( + const struct inode *dir, + const struct dentry *dentry, + unsigned int flags, + int error + ), + + TP_ARGS(dir, dentry, flags, error), + + TP_STRUCT__entry( + __field(int, error) + __field(unsigned int, flags) + __field(dev_t, dev) + __field(u64, dir) + __string(name, dentry->d_name.name) + ), + + TP_fast_assign( + __entry->dev = dir->i_sb->s_dev; + __entry->dir = NFS_FILEID(dir); + __entry->error = error; + __entry->flags = flags; + __assign_str(name, dentry->d_name.name); + ), + + TP_printk( + "error=%d flags=%u (%s) name=%02x:%02x:%llu/%s", + __entry->error, + __entry->flags, + show_lookup_flags(__entry->flags), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->dir, + __get_str(name) + ) +); + +#define DEFINE_NFS_LOOKUP_EVENT_DONE(name) \ + DEFINE_EVENT(nfs_lookup_event_done, name, \ + TP_PROTO( \ + const struct inode *dir, \ + const struct dentry *dentry, \ + unsigned int flags, \ + int error \ + ), \ + TP_ARGS(dir, dentry, flags, error)) + +DEFINE_NFS_LOOKUP_EVENT(nfs_lookup_enter); +DEFINE_NFS_LOOKUP_EVENT_DONE(nfs_lookup_exit); +DEFINE_NFS_LOOKUP_EVENT(nfs_lookup_revalidate_enter); +DEFINE_NFS_LOOKUP_EVENT_DONE(nfs_lookup_revalidate_exit); + +#define show_open_flags(flags) \ + __print_flags((unsigned long)flags, "|", \ + { O_CREAT, "O_CREAT" }, \ + { O_EXCL, "O_EXCL" }, \ + { O_TRUNC, "O_TRUNC" }, \ + { O_APPEND, "O_APPEND" }, \ + { O_DSYNC, "O_DSYNC" }, \ + { O_DIRECT, "O_DIRECT" }, \ + { O_DIRECTORY, "O_DIRECTORY" }) + +#define show_fmode_flags(mode) \ + __print_flags(mode, "|", \ + { ((__force unsigned long)FMODE_READ), "READ" }, \ + { ((__force unsigned long)FMODE_WRITE), "WRITE" }, \ + { ((__force unsigned long)FMODE_EXEC), "EXEC" }) + +TRACE_EVENT(nfs_atomic_open_enter, + TP_PROTO( + const struct inode *dir, + const struct nfs_open_context *ctx, + unsigned int flags + ), + + TP_ARGS(dir, ctx, flags), + + TP_STRUCT__entry( + __field(unsigned int, flags) + __field(unsigned int, fmode) + __field(dev_t, dev) + __field(u64, dir) + __string(name, ctx->dentry->d_name.name) + ), + + TP_fast_assign( + __entry->dev = dir->i_sb->s_dev; + __entry->dir = NFS_FILEID(dir); + __entry->flags = flags; + __entry->fmode = (__force unsigned int)ctx->mode; + __assign_str(name, ctx->dentry->d_name.name); + ), + + TP_printk( + "flags=%u (%s) fmode=%s name=%02x:%02x:%llu/%s", + __entry->flags, + show_open_flags(__entry->flags), + show_fmode_flags(__entry->fmode), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->dir, + __get_str(name) + ) +); + +TRACE_EVENT(nfs_atomic_open_exit, + TP_PROTO( + const struct inode *dir, + const struct nfs_open_context *ctx, + unsigned int flags, + int error + ), + + TP_ARGS(dir, ctx, flags, error), + + TP_STRUCT__entry( + __field(int, error) + __field(unsigned int, flags) + __field(unsigned int, fmode) + __field(dev_t, dev) + __field(u64, dir) + __string(name, ctx->dentry->d_name.name) + ), + + TP_fast_assign( + __entry->error = error; + __entry->dev = dir->i_sb->s_dev; + __entry->dir = NFS_FILEID(dir); + __entry->flags = flags; + __entry->fmode = (__force unsigned int)ctx->mode; + __assign_str(name, ctx->dentry->d_name.name); + ), + + TP_printk( + "error=%d flags=%u (%s) fmode=%s " + "name=%02x:%02x:%llu/%s", + __entry->error, + __entry->flags, + show_open_flags(__entry->flags), + show_fmode_flags(__entry->fmode), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->dir, + __get_str(name) + ) +); + +TRACE_EVENT(nfs_create_enter, + TP_PROTO( + const struct inode *dir, + const struct dentry *dentry, + unsigned int flags + ), + + TP_ARGS(dir, dentry, flags), + + TP_STRUCT__entry( + __field(unsigned int, flags) + __field(dev_t, dev) + __field(u64, dir) + __string(name, dentry->d_name.name) + ), + + TP_fast_assign( + __entry->dev = dir->i_sb->s_dev; + __entry->dir = NFS_FILEID(dir); + __entry->flags = flags; + __assign_str(name, dentry->d_name.name); + ), + + TP_printk( + "flags=%u (%s) name=%02x:%02x:%llu/%s", + __entry->flags, + show_open_flags(__entry->flags), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->dir, + __get_str(name) + ) +); + +TRACE_EVENT(nfs_create_exit, + TP_PROTO( + const struct inode *dir, + const struct dentry *dentry, + unsigned int flags, + int error + ), + + TP_ARGS(dir, dentry, flags, error), + + TP_STRUCT__entry( + __field(int, error) + __field(unsigned int, flags) + __field(dev_t, dev) + __field(u64, dir) + __string(name, dentry->d_name.name) + ), + + TP_fast_assign( + __entry->error = error; + __entry->dev = dir->i_sb->s_dev; + __entry->dir = NFS_FILEID(dir); + __entry->flags = flags; + __assign_str(name, dentry->d_name.name); + ), + + TP_printk( + "error=%d flags=%u (%s) name=%02x:%02x:%llu/%s", + __entry->error, + __entry->flags, + show_open_flags(__entry->flags), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->dir, + __get_str(name) + ) +); + +DECLARE_EVENT_CLASS(nfs_directory_event, + TP_PROTO( + const struct inode *dir, + const struct dentry *dentry + ), + + TP_ARGS(dir, dentry), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u64, dir) + __string(name, dentry->d_name.name) + ), + + TP_fast_assign( + __entry->dev = dir->i_sb->s_dev; + __entry->dir = NFS_FILEID(dir); + __assign_str(name, dentry->d_name.name); + ), + + TP_printk( + "name=%02x:%02x:%llu/%s", + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->dir, + __get_str(name) + ) +); + +#define DEFINE_NFS_DIRECTORY_EVENT(name) \ + DEFINE_EVENT(nfs_directory_event, name, \ + TP_PROTO( \ + const struct inode *dir, \ + const struct dentry *dentry \ + ), \ + TP_ARGS(dir, dentry)) + +DECLARE_EVENT_CLASS(nfs_directory_event_done, + TP_PROTO( + const struct inode *dir, + const struct dentry *dentry, + int error + ), + + TP_ARGS(dir, dentry, error), + + TP_STRUCT__entry( + __field(int, error) + __field(dev_t, dev) + __field(u64, dir) + __string(name, dentry->d_name.name) + ), + + TP_fast_assign( + __entry->dev = dir->i_sb->s_dev; + __entry->dir = NFS_FILEID(dir); + __entry->error = error; + __assign_str(name, dentry->d_name.name); + ), + + TP_printk( + "error=%d name=%02x:%02x:%llu/%s", + __entry->error, + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->dir, + __get_str(name) + ) +); + +#define DEFINE_NFS_DIRECTORY_EVENT_DONE(name) \ + DEFINE_EVENT(nfs_directory_event_done, name, \ + TP_PROTO( \ + const struct inode *dir, \ + const struct dentry *dentry, \ + int error \ + ), \ + TP_ARGS(dir, dentry, error)) + +DEFINE_NFS_DIRECTORY_EVENT(nfs_mknod_enter); +DEFINE_NFS_DIRECTORY_EVENT_DONE(nfs_mknod_exit); +DEFINE_NFS_DIRECTORY_EVENT(nfs_mkdir_enter); +DEFINE_NFS_DIRECTORY_EVENT_DONE(nfs_mkdir_exit); +DEFINE_NFS_DIRECTORY_EVENT(nfs_rmdir_enter); +DEFINE_NFS_DIRECTORY_EVENT_DONE(nfs_rmdir_exit); +DEFINE_NFS_DIRECTORY_EVENT(nfs_remove_enter); +DEFINE_NFS_DIRECTORY_EVENT_DONE(nfs_remove_exit); +DEFINE_NFS_DIRECTORY_EVENT(nfs_unlink_enter); +DEFINE_NFS_DIRECTORY_EVENT_DONE(nfs_unlink_exit); +DEFINE_NFS_DIRECTORY_EVENT(nfs_symlink_enter); +DEFINE_NFS_DIRECTORY_EVENT_DONE(nfs_symlink_exit); + +TRACE_EVENT(nfs_link_enter, + TP_PROTO( + const struct inode *inode, + const struct inode *dir, + const struct dentry *dentry + ), + + TP_ARGS(inode, dir, dentry), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u64, fileid) + __field(u64, dir) + __string(name, dentry->d_name.name) + ), + + TP_fast_assign( + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = NFS_FILEID(inode); + __entry->dir = NFS_FILEID(dir); + __assign_str(name, dentry->d_name.name); + ), + + TP_printk( + "fileid=%02x:%02x:%llu name=%02x:%02x:%llu/%s", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->fileid, + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->dir, + __get_str(name) + ) +); + +TRACE_EVENT(nfs_link_exit, + TP_PROTO( + const struct inode *inode, + const struct inode *dir, + const struct dentry *dentry, + int error + ), + + TP_ARGS(inode, dir, dentry, error), + + TP_STRUCT__entry( + __field(int, error) + __field(dev_t, dev) + __field(u64, fileid) + __field(u64, dir) + __string(name, dentry->d_name.name) + ), + + TP_fast_assign( + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = NFS_FILEID(inode); + __entry->dir = NFS_FILEID(dir); + __entry->error = error; + __assign_str(name, dentry->d_name.name); + ), + + TP_printk( + "error=%d fileid=%02x:%02x:%llu name=%02x:%02x:%llu/%s", + __entry->error, + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->fileid, + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->dir, + __get_str(name) + ) +); + +DECLARE_EVENT_CLASS(nfs_rename_event, + TP_PROTO( + const struct inode *old_dir, + const struct dentry *old_dentry, + const struct inode *new_dir, + const struct dentry *new_dentry + ), + + TP_ARGS(old_dir, old_dentry, new_dir, new_dentry), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u64, old_dir) + __field(u64, new_dir) + __string(old_name, old_dentry->d_name.name) + __string(new_name, new_dentry->d_name.name) + ), + + TP_fast_assign( + __entry->dev = old_dir->i_sb->s_dev; + __entry->old_dir = NFS_FILEID(old_dir); + __entry->new_dir = NFS_FILEID(new_dir); + __assign_str(old_name, old_dentry->d_name.name); + __assign_str(new_name, new_dentry->d_name.name); + ), + + TP_printk( + "old_name=%02x:%02x:%llu/%s new_name=%02x:%02x:%llu/%s", + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->old_dir, + __get_str(old_name), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->new_dir, + __get_str(new_name) + ) +); +#define DEFINE_NFS_RENAME_EVENT(name) \ + DEFINE_EVENT(nfs_rename_event, name, \ + TP_PROTO( \ + const struct inode *old_dir, \ + const struct dentry *old_dentry, \ + const struct inode *new_dir, \ + const struct dentry *new_dentry \ + ), \ + TP_ARGS(old_dir, old_dentry, new_dir, new_dentry)) + +DECLARE_EVENT_CLASS(nfs_rename_event_done, + TP_PROTO( + const struct inode *old_dir, + const struct dentry *old_dentry, + const struct inode *new_dir, + const struct dentry *new_dentry, + int error + ), + + TP_ARGS(old_dir, old_dentry, new_dir, new_dentry, error), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(int, error) + __field(u64, old_dir) + __string(old_name, old_dentry->d_name.name) + __field(u64, new_dir) + __string(new_name, new_dentry->d_name.name) + ), + + TP_fast_assign( + __entry->dev = old_dir->i_sb->s_dev; + __entry->old_dir = NFS_FILEID(old_dir); + __entry->new_dir = NFS_FILEID(new_dir); + __entry->error = error; + __assign_str(old_name, old_dentry->d_name.name); + __assign_str(new_name, new_dentry->d_name.name); + ), + + TP_printk( + "error=%d old_name=%02x:%02x:%llu/%s " + "new_name=%02x:%02x:%llu/%s", + __entry->error, + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->old_dir, + __get_str(old_name), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->new_dir, + __get_str(new_name) + ) +); +#define DEFINE_NFS_RENAME_EVENT_DONE(name) \ + DEFINE_EVENT(nfs_rename_event_done, name, \ + TP_PROTO( \ + const struct inode *old_dir, \ + const struct dentry *old_dentry, \ + const struct inode *new_dir, \ + const struct dentry *new_dentry, \ + int error \ + ), \ + TP_ARGS(old_dir, old_dentry, new_dir, \ + new_dentry, error)) + +DEFINE_NFS_RENAME_EVENT(nfs_rename_enter); +DEFINE_NFS_RENAME_EVENT_DONE(nfs_rename_exit); + +DEFINE_NFS_RENAME_EVENT_DONE(nfs_sillyrename_rename); + +TRACE_EVENT(nfs_sillyrename_unlink, + TP_PROTO( + const struct nfs_unlinkdata *data, + int error + ), + + TP_ARGS(data, error), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(int, error) + __field(u64, dir) + __dynamic_array(char, name, data->args.name.len + 1) + ), + + TP_fast_assign( + struct inode *dir = data->dir; + size_t len = data->args.name.len; + __entry->dev = dir->i_sb->s_dev; + __entry->dir = NFS_FILEID(dir); + __entry->error = error; + memcpy(__get_dynamic_array(name), + data->args.name.name, len); + ((char *)__get_dynamic_array(name))[len] = 0; + ), + + TP_printk( + "error=%d name=%02x:%02x:%llu/%s", + __entry->error, + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->dir, + __get_str(name) + ) +); +#endif /* _TRACE_NFS_H */ + +#undef TRACE_INCLUDE_PATH +#define TRACE_INCLUDE_PATH . +#define TRACE_INCLUDE_FILE nfstrace +/* This part must be outside protection */ +#include <trace/define_trace.h> diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c index 29cfb7a..2ffebf2 100644 --- a/fs/nfs/pagelist.c +++ b/fs/nfs/pagelist.c @@ -328,6 +328,19 @@ void nfs_pageio_init(struct nfs_pageio_descriptor *desc, } EXPORT_SYMBOL_GPL(nfs_pageio_init); +static bool nfs_match_open_context(const struct nfs_open_context *ctx1, + const struct nfs_open_context *ctx2) +{ + return ctx1->cred == ctx2->cred && ctx1->state == ctx2->state; +} + +static bool nfs_match_lock_context(const struct nfs_lock_context *l1, + const struct nfs_lock_context *l2) +{ + return l1->lockowner.l_owner == l2->lockowner.l_owner + && l1->lockowner.l_pid == l2->lockowner.l_pid; +} + /** * nfs_can_coalesce_requests - test two requests for compatibility * @prev: pointer to nfs_page @@ -343,13 +356,10 @@ static bool nfs_can_coalesce_requests(struct nfs_page *prev, struct nfs_page *req, struct nfs_pageio_descriptor *pgio) { - if (req->wb_context->cred != prev->wb_context->cred) - return false; - if (req->wb_lock_context->lockowner.l_owner != prev->wb_lock_context->lockowner.l_owner) - return false; - if (req->wb_lock_context->lockowner.l_pid != prev->wb_lock_context->lockowner.l_pid) + if (!nfs_match_open_context(req->wb_context, prev->wb_context)) return false; - if (req->wb_context->state != prev->wb_context->state) + if (req->wb_context->dentry->d_inode->i_flock != NULL && + !nfs_match_lock_context(req->wb_lock_context, prev->wb_lock_context)) return false; if (req->wb_pgbase != 0) return false; diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c index 3a3a79d..d75d938 100644 --- a/fs/nfs/pnfs.c +++ b/fs/nfs/pnfs.c @@ -33,6 +33,7 @@ #include "internal.h" #include "pnfs.h" #include "iostat.h" +#include "nfs4trace.h" #define NFSDBG_FACILITY NFSDBG_PNFS #define PNFS_LAYOUTGET_RETRY_TIMEOUT (120*HZ) @@ -1526,6 +1527,7 @@ void pnfs_ld_write_done(struct nfs_write_data *data) { struct nfs_pgio_header *hdr = data->header; + trace_nfs4_pnfs_write(data, hdr->pnfs_error); if (!hdr->pnfs_error) { pnfs_set_layoutcommit(data); hdr->mds_ops->rpc_call_done(&data->task, data); @@ -1680,6 +1682,7 @@ void pnfs_ld_read_done(struct nfs_read_data *data) { struct nfs_pgio_header *hdr = data->header; + trace_nfs4_pnfs_read(data, hdr->pnfs_error); if (likely(!hdr->pnfs_error)) { __nfs4_read_done_cb(data); hdr->mds_ops->rpc_call_done(&data->task, data); diff --git a/fs/nfs/proc.c b/fs/nfs/proc.c index c041c41..a8f57c7 100644 --- a/fs/nfs/proc.c +++ b/fs/nfs/proc.c @@ -623,9 +623,10 @@ static void nfs_proc_read_setup(struct nfs_read_data *data, struct rpc_message * msg->rpc_proc = &nfs_procedures[NFSPROC_READ]; } -static void nfs_proc_read_rpc_prepare(struct rpc_task *task, struct nfs_read_data *data) +static int nfs_proc_read_rpc_prepare(struct rpc_task *task, struct nfs_read_data *data) { rpc_call_start(task); + return 0; } static int nfs_write_done(struct rpc_task *task, struct nfs_write_data *data) @@ -644,9 +645,10 @@ static void nfs_proc_write_setup(struct nfs_write_data *data, struct rpc_message msg->rpc_proc = &nfs_procedures[NFSPROC_WRITE]; } -static void nfs_proc_write_rpc_prepare(struct rpc_task *task, struct nfs_write_data *data) +static int nfs_proc_write_rpc_prepare(struct rpc_task *task, struct nfs_write_data *data) { rpc_call_start(task); + return 0; } static void nfs_proc_commit_rpc_prepare(struct rpc_task *task, struct nfs_commit_data *data) diff --git a/fs/nfs/read.c b/fs/nfs/read.c index 70a26c6..31db5c3 100644 --- a/fs/nfs/read.c +++ b/fs/nfs/read.c @@ -513,9 +513,10 @@ static void nfs_readpage_release_common(void *calldata) void nfs_read_prepare(struct rpc_task *task, void *calldata) { struct nfs_read_data *data = calldata; - NFS_PROTO(data->header->inode)->read_rpc_prepare(task, data); - if (unlikely(test_bit(NFS_CONTEXT_BAD, &data->args.context->flags))) - rpc_exit(task, -EIO); + int err; + err = NFS_PROTO(data->header->inode)->read_rpc_prepare(task, data); + if (err) + rpc_exit(task, err); } static const struct rpc_call_ops nfs_read_common_ops = { diff --git a/fs/nfs/super.c b/fs/nfs/super.c index f6db66d..5793f24 100644 --- a/fs/nfs/super.c +++ b/fs/nfs/super.c @@ -923,7 +923,7 @@ static struct nfs_parsed_mount_data *nfs_alloc_parsed_mount_data(void) data->nfs_server.port = NFS_UNSPEC_PORT; data->nfs_server.protocol = XPRT_TRANSPORT_TCP; data->auth_flavors[0] = RPC_AUTH_MAXFLAVOR; - data->auth_flavor_len = 1; + data->auth_flavor_len = 0; data->minorversion = 0; data->need_mount = true; data->net = current->nsproxy->net_ns; @@ -1018,6 +1018,13 @@ static void nfs_set_mount_transport_protocol(struct nfs_parsed_mount_data *mnt) } } +static void nfs_set_auth_parsed_mount_data(struct nfs_parsed_mount_data *data, + rpc_authflavor_t pseudoflavor) +{ + data->auth_flavors[0] = pseudoflavor; + data->auth_flavor_len = 1; +} + /* * Parse the value of the 'sec=' option. */ @@ -1025,49 +1032,50 @@ static int nfs_parse_security_flavors(char *value, struct nfs_parsed_mount_data *mnt) { substring_t args[MAX_OPT_ARGS]; + rpc_authflavor_t pseudoflavor; dfprintk(MOUNT, "NFS: parsing sec=%s option\n", value); switch (match_token(value, nfs_secflavor_tokens, args)) { case Opt_sec_none: - mnt->auth_flavors[0] = RPC_AUTH_NULL; + pseudoflavor = RPC_AUTH_NULL; break; case Opt_sec_sys: - mnt->auth_flavors[0] = RPC_AUTH_UNIX; + pseudoflavor = RPC_AUTH_UNIX; break; case Opt_sec_krb5: - mnt->auth_flavors[0] = RPC_AUTH_GSS_KRB5; + pseudoflavor = RPC_AUTH_GSS_KRB5; break; case Opt_sec_krb5i: - mnt->auth_flavors[0] = RPC_AUTH_GSS_KRB5I; + pseudoflavor = RPC_AUTH_GSS_KRB5I; break; case Opt_sec_krb5p: - mnt->auth_flavors[0] = RPC_AUTH_GSS_KRB5P; + pseudoflavor = RPC_AUTH_GSS_KRB5P; break; case Opt_sec_lkey: - mnt->auth_flavors[0] = RPC_AUTH_GSS_LKEY; + pseudoflavor = RPC_AUTH_GSS_LKEY; break; case Opt_sec_lkeyi: - mnt->auth_flavors[0] = RPC_AUTH_GSS_LKEYI; + pseudoflavor = RPC_AUTH_GSS_LKEYI; break; case Opt_sec_lkeyp: - mnt->auth_flavors[0] = RPC_AUTH_GSS_LKEYP; + pseudoflavor = RPC_AUTH_GSS_LKEYP; break; case Opt_sec_spkm: - mnt->auth_flavors[0] = RPC_AUTH_GSS_SPKM; + pseudoflavor = RPC_AUTH_GSS_SPKM; break; case Opt_sec_spkmi: - mnt->auth_flavors[0] = RPC_AUTH_GSS_SPKMI; + pseudoflavor = RPC_AUTH_GSS_SPKMI; break; case Opt_sec_spkmp: - mnt->auth_flavors[0] = RPC_AUTH_GSS_SPKMP; + pseudoflavor = RPC_AUTH_GSS_SPKMP; break; default: return 0; } mnt->flags |= NFS_MOUNT_SECFLAVOUR; - mnt->auth_flavor_len = 1; + nfs_set_auth_parsed_mount_data(mnt, pseudoflavor); return 1; } @@ -1729,7 +1737,7 @@ static struct nfs_server *nfs_try_mount_request(struct nfs_mount_info *mount_inf * Was a sec= authflavor specified in the options? First, verify * whether the server supports it, and then just try to use it if so. */ - if (args->auth_flavors[0] != RPC_AUTH_MAXFLAVOR) { + if (args->auth_flavor_len > 0) { status = nfs_verify_authflavor(args, authlist, authlist_len); dfprintk(MOUNT, "NFS: using auth flavor %u\n", args->auth_flavors[0]); if (status) @@ -1760,7 +1768,7 @@ static struct nfs_server *nfs_try_mount_request(struct nfs_mount_info *mount_inf /* Fallthrough */ } dfprintk(MOUNT, "NFS: attempting to use auth flavor %u\n", flavor); - args->auth_flavors[0] = flavor; + nfs_set_auth_parsed_mount_data(args, flavor); server = nfs_mod->rpc_ops->create_server(mount_info, nfs_mod); if (!IS_ERR(server)) return server; @@ -1776,7 +1784,7 @@ static struct nfs_server *nfs_try_mount_request(struct nfs_mount_info *mount_inf /* Last chance! Try AUTH_UNIX */ dfprintk(MOUNT, "NFS: attempting to use auth flavor %u\n", RPC_AUTH_UNIX); - args->auth_flavors[0] = RPC_AUTH_UNIX; + nfs_set_auth_parsed_mount_data(args, RPC_AUTH_UNIX); return nfs_mod->rpc_ops->create_server(mount_info, nfs_mod); } @@ -1893,6 +1901,7 @@ static int nfs23_validate_mount_data(void *options, { struct nfs_mount_data *data = (struct nfs_mount_data *)options; struct sockaddr *sap = (struct sockaddr *)&args->nfs_server.address; + int extra_flags = NFS_MOUNT_LEGACY_INTERFACE; if (data == NULL) goto out_no_data; @@ -1908,6 +1917,8 @@ static int nfs23_validate_mount_data(void *options, goto out_no_v3; data->root.size = NFS2_FHSIZE; memcpy(data->root.data, data->old_root.data, NFS2_FHSIZE); + /* Turn off security negotiation */ + extra_flags |= NFS_MOUNT_SECFLAVOUR; case 4: if (data->flags & NFS_MOUNT_SECFLAVOUR) goto out_no_sec; @@ -1935,7 +1946,7 @@ static int nfs23_validate_mount_data(void *options, * can deal with. */ args->flags = data->flags & NFS_MOUNT_FLAGMASK; - args->flags |= NFS_MOUNT_LEGACY_INTERFACE; + args->flags |= extra_flags; args->rsize = data->rsize; args->wsize = data->wsize; args->timeo = data->timeo; @@ -1959,9 +1970,10 @@ static int nfs23_validate_mount_data(void *options, args->namlen = data->namlen; args->bsize = data->bsize; - args->auth_flavors[0] = RPC_AUTH_UNIX; if (data->flags & NFS_MOUNT_SECFLAVOUR) - args->auth_flavors[0] = data->pseudoflavor; + nfs_set_auth_parsed_mount_data(args, data->pseudoflavor); + else + nfs_set_auth_parsed_mount_data(args, RPC_AUTH_UNIX); if (!args->nfs_server.hostname) goto out_nomem; @@ -2084,6 +2096,8 @@ static int nfs_validate_text_mount_data(void *options, max_namelen = NFS4_MAXNAMLEN; max_pathlen = NFS4_MAXPATHLEN; nfs_validate_transport_protocol(args); + if (args->nfs_server.protocol == XPRT_TRANSPORT_UDP) + goto out_invalid_transport_udp; nfs4_validate_mount_flags(args); #else goto out_v4_not_compiled; @@ -2106,6 +2120,10 @@ static int nfs_validate_text_mount_data(void *options, out_v4_not_compiled: dfprintk(MOUNT, "NFS: NFSv4 is not compiled into kernel\n"); return -EPROTONOSUPPORT; +#else +out_invalid_transport_udp: + dfprintk(MOUNT, "NFSv4: Unsupported transport protocol udp\n"); + return -EINVAL; #endif /* !CONFIG_NFS_V4 */ out_no_address: @@ -2170,7 +2188,7 @@ nfs_remount(struct super_block *sb, int *flags, char *raw_data) data->rsize = nfss->rsize; data->wsize = nfss->wsize; data->retrans = nfss->client->cl_timeout->to_retries; - data->auth_flavors[0] = nfss->client->cl_auth->au_flavor; + nfs_set_auth_parsed_mount_data(data, nfss->client->cl_auth->au_flavor); data->acregmin = nfss->acregmin / HZ; data->acregmax = nfss->acregmax / HZ; data->acdirmin = nfss->acdirmin / HZ; @@ -2277,6 +2295,18 @@ void nfs_clone_super(struct super_block *sb, struct nfs_mount_info *mount_info) nfs_initialise_sb(sb); } +#define NFS_MOUNT_CMP_FLAGMASK ~(NFS_MOUNT_INTR \ + | NFS_MOUNT_SECURE \ + | NFS_MOUNT_TCP \ + | NFS_MOUNT_VER3 \ + | NFS_MOUNT_KERBEROS \ + | NFS_MOUNT_NONLM \ + | NFS_MOUNT_BROKEN_SUID \ + | NFS_MOUNT_STRICTLOCK \ + | NFS_MOUNT_UNSHARED \ + | NFS_MOUNT_NORESVPORT \ + | NFS_MOUNT_LEGACY_INTERFACE) + static int nfs_compare_mount_options(const struct super_block *s, const struct nfs_server *b, int flags) { const struct nfs_server *a = s->s_fs_info; @@ -2287,7 +2317,7 @@ static int nfs_compare_mount_options(const struct super_block *s, const struct n goto Ebusy; if (a->nfs_client != b->nfs_client) goto Ebusy; - if (a->flags != b->flags) + if ((a->flags ^ b->flags) & NFS_MOUNT_CMP_FLAGMASK) goto Ebusy; if (a->wsize != b->wsize) goto Ebusy; @@ -2301,7 +2331,8 @@ static int nfs_compare_mount_options(const struct super_block *s, const struct n goto Ebusy; if (a->acdirmax != b->acdirmax) goto Ebusy; - if (clnt_a->cl_auth->au_flavor != clnt_b->cl_auth->au_flavor) + if (b->flags & NFS_MOUNT_SECFLAVOUR && + clnt_a->cl_auth->au_flavor != clnt_b->cl_auth->au_flavor) goto Ebusy; return 1; Ebusy: @@ -2673,15 +2704,17 @@ static int nfs4_validate_mount_data(void *options, goto out_no_address; args->nfs_server.port = ntohs(((struct sockaddr_in *)sap)->sin_port); - args->auth_flavors[0] = RPC_AUTH_UNIX; if (data->auth_flavourlen) { + rpc_authflavor_t pseudoflavor; if (data->auth_flavourlen > 1) goto out_inval_auth; - if (copy_from_user(&args->auth_flavors[0], + if (copy_from_user(&pseudoflavor, data->auth_flavours, - sizeof(args->auth_flavors[0]))) + sizeof(pseudoflavor))) return -EFAULT; - } + nfs_set_auth_parsed_mount_data(args, pseudoflavor); + } else + nfs_set_auth_parsed_mount_data(args, RPC_AUTH_UNIX); c = strndup_user(data->hostname.data, NFS4_MAXNAMLEN); if (IS_ERR(c)) @@ -2715,6 +2748,8 @@ static int nfs4_validate_mount_data(void *options, args->acdirmax = data->acdirmax; args->nfs_server.protocol = data->proto; nfs_validate_transport_protocol(args); + if (args->nfs_server.protocol == XPRT_TRANSPORT_UDP) + goto out_invalid_transport_udp; break; default: @@ -2735,6 +2770,10 @@ out_inval_auth: out_no_address: dfprintk(MOUNT, "NFS4: mount program didn't pass remote address\n"); return -EINVAL; + +out_invalid_transport_udp: + dfprintk(MOUNT, "NFSv4: Unsupported transport protocol udp\n"); + return -EINVAL; } /* @@ -2750,6 +2789,7 @@ bool nfs4_disable_idmapping = true; unsigned short max_session_slots = NFS4_DEF_SLOT_TABLE_SIZE; unsigned short send_implementation_id = 1; char nfs4_client_id_uniquifier[NFS4_CLIENT_ID_UNIQ_LEN] = ""; +bool recover_lost_locks = false; EXPORT_SYMBOL_GPL(nfs_callback_set_tcpport); EXPORT_SYMBOL_GPL(nfs_callback_tcpport); @@ -2758,6 +2798,7 @@ EXPORT_SYMBOL_GPL(nfs4_disable_idmapping); EXPORT_SYMBOL_GPL(max_session_slots); EXPORT_SYMBOL_GPL(send_implementation_id); EXPORT_SYMBOL_GPL(nfs4_client_id_uniquifier); +EXPORT_SYMBOL_GPL(recover_lost_locks); #define NFS_CALLBACK_MAXPORTNR (65535U) @@ -2795,4 +2836,10 @@ MODULE_PARM_DESC(send_implementation_id, "Send implementation ID with NFSv4.1 exchange_id"); MODULE_PARM_DESC(nfs4_unique_id, "nfs_client_id4 uniquifier string"); +module_param(recover_lost_locks, bool, 0644); +MODULE_PARM_DESC(recover_lost_locks, + "If the server reports that a lock might be lost, " + "try to recover it risking data corruption."); + + #endif /* CONFIG_NFS_V4 */ diff --git a/fs/nfs/unlink.c b/fs/nfs/unlink.c index 60395ad..bb939ed 100644 --- a/fs/nfs/unlink.c +++ b/fs/nfs/unlink.c @@ -20,6 +20,8 @@ #include "iostat.h" #include "delegation.h" +#include "nfstrace.h" + /** * nfs_free_unlinkdata - release data from a sillydelete operation. * @data: pointer to unlink structure. @@ -77,6 +79,7 @@ static void nfs_async_unlink_done(struct rpc_task *task, void *calldata) struct nfs_unlinkdata *data = calldata; struct inode *dir = data->dir; + trace_nfs_sillyrename_unlink(data, task->tk_status); if (!NFS_PROTO(dir)->unlink_done(task, dir)) rpc_restart_call_prepare(task); } @@ -204,6 +207,13 @@ out_free: return ret; } +void nfs_wait_on_sillyrename(struct dentry *dentry) +{ + struct nfs_inode *nfsi = NFS_I(dentry->d_inode); + + wait_event(nfsi->waitqueue, atomic_read(&nfsi->silly_count) <= 1); +} + void nfs_block_sillyrename(struct dentry *dentry) { struct nfs_inode *nfsi = NFS_I(dentry->d_inode); @@ -336,6 +346,8 @@ static void nfs_async_rename_done(struct rpc_task *task, void *calldata) struct inode *new_dir = data->new_dir; struct dentry *old_dentry = data->old_dentry; + trace_nfs_sillyrename_rename(old_dir, old_dentry, + new_dir, data->new_dentry, task->tk_status); if (!NFS_PROTO(old_dir)->rename_done(task, old_dir, new_dir)) { rpc_restart_call_prepare(task); return; @@ -444,6 +456,14 @@ nfs_async_rename(struct inode *old_dir, struct inode *new_dir, return rpc_run_task(&task_setup_data); } +#define SILLYNAME_PREFIX ".nfs" +#define SILLYNAME_PREFIX_LEN ((unsigned)sizeof(SILLYNAME_PREFIX) - 1) +#define SILLYNAME_FILEID_LEN ((unsigned)sizeof(u64) << 1) +#define SILLYNAME_COUNTER_LEN ((unsigned)sizeof(unsigned int) << 1) +#define SILLYNAME_LEN (SILLYNAME_PREFIX_LEN + \ + SILLYNAME_FILEID_LEN + \ + SILLYNAME_COUNTER_LEN) + /** * nfs_sillyrename - Perform a silly-rename of a dentry * @dir: inode of directory that contains dentry @@ -469,10 +489,8 @@ int nfs_sillyrename(struct inode *dir, struct dentry *dentry) { static unsigned int sillycounter; - const int fileidsize = sizeof(NFS_FILEID(dentry->d_inode))*2; - const int countersize = sizeof(sillycounter)*2; - const int slen = sizeof(".nfs")+fileidsize+countersize-1; - char silly[slen+1]; + unsigned char silly[SILLYNAME_LEN + 1]; + unsigned long long fileid; struct dentry *sdentry; struct rpc_task *task; int error = -EIO; @@ -489,20 +507,20 @@ nfs_sillyrename(struct inode *dir, struct dentry *dentry) if (dentry->d_flags & DCACHE_NFSFS_RENAMED) goto out; - sprintf(silly, ".nfs%*.*Lx", - fileidsize, fileidsize, - (unsigned long long)NFS_FILEID(dentry->d_inode)); + fileid = NFS_FILEID(dentry->d_inode); /* Return delegation in anticipation of the rename */ NFS_PROTO(dentry->d_inode)->return_delegation(dentry->d_inode); sdentry = NULL; do { - char *suffix = silly + slen - countersize; - + int slen; dput(sdentry); sillycounter++; - sprintf(suffix, "%*.*x", countersize, countersize, sillycounter); + slen = scnprintf(silly, sizeof(silly), + SILLYNAME_PREFIX "%0*llx%0*x", + SILLYNAME_FILEID_LEN, fileid, + SILLYNAME_COUNTER_LEN, sillycounter); dfprintk(VFS, "NFS: trying to rename %s to %s\n", dentry->d_name.name, silly); diff --git a/fs/nfs/write.c b/fs/nfs/write.c index f1bdb72..ac1dc33 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -31,6 +31,8 @@ #include "fscache.h" #include "pnfs.h" +#include "nfstrace.h" + #define NFSDBG_FACILITY NFSDBG_PAGECACHE #define MIN_POOL_WRITE (32) @@ -861,7 +863,7 @@ int nfs_flush_incompatible(struct file *file, struct page *page) return 0; l_ctx = req->wb_lock_context; do_flush = req->wb_page != page || req->wb_context != ctx; - if (l_ctx) { + if (l_ctx && ctx->dentry->d_inode->i_flock != NULL) { do_flush |= l_ctx->lockowner.l_owner != current->files || l_ctx->lockowner.l_pid != current->tgid; } @@ -874,6 +876,33 @@ int nfs_flush_incompatible(struct file *file, struct page *page) } /* + * Avoid buffered writes when a open context credential's key would + * expire soon. + * + * Returns -EACCES if the key will expire within RPC_KEY_EXPIRE_FAIL. + * + * Return 0 and set a credential flag which triggers the inode to flush + * and performs NFS_FILE_SYNC writes if the key will expired within + * RPC_KEY_EXPIRE_TIMEO. + */ +int +nfs_key_timeout_notify(struct file *filp, struct inode *inode) +{ + struct nfs_open_context *ctx = nfs_file_open_context(filp); + struct rpc_auth *auth = NFS_SERVER(inode)->client->cl_auth; + + return rpcauth_key_timeout_notify(auth, ctx->cred); +} + +/* + * Test if the open context credential key is marked to expire soon. + */ +bool nfs_ctx_key_to_expire(struct nfs_open_context *ctx) +{ + return rpcauth_cred_key_to_expire(ctx->cred); +} + +/* * If the page cache is marked as unsafe or invalid, then we can't rely on * the PageUptodate() flag. In this case, we will need to turn off * write optimisations that depend on the page contents being correct. @@ -993,6 +1022,9 @@ int nfs_initiate_write(struct rpc_clnt *clnt, data->args.count, (unsigned long long)data->args.offset); + nfs4_state_protect_write(NFS_SERVER(inode)->nfs_client, + &task_setup_data.rpc_client, &msg, data); + task = rpc_run_task(&task_setup_data); if (IS_ERR(task)) { ret = PTR_ERR(task); @@ -1265,9 +1297,10 @@ EXPORT_SYMBOL_GPL(nfs_pageio_reset_write_mds); void nfs_write_prepare(struct rpc_task *task, void *calldata) { struct nfs_write_data *data = calldata; - NFS_PROTO(data->header->inode)->write_rpc_prepare(task, data); - if (unlikely(test_bit(NFS_CONTEXT_BAD, &data->args.context->flags))) - rpc_exit(task, -EIO); + int err; + err = NFS_PROTO(data->header->inode)->write_rpc_prepare(task, data); + if (err) + rpc_exit(task, err); } void nfs_commit_prepare(struct rpc_task *task, void *calldata) @@ -1458,6 +1491,9 @@ int nfs_initiate_commit(struct rpc_clnt *clnt, struct nfs_commit_data *data, dprintk("NFS: %5u initiated commit call\n", data->task.tk_pid); + nfs4_state_protect(NFS_SERVER(data->inode)->nfs_client, + NFS_SP4_MACH_CRED_COMMIT, &task_setup_data.rpc_client, &msg); + task = rpc_run_task(&task_setup_data); if (IS_ERR(task)) return PTR_ERR(task); @@ -1732,8 +1768,14 @@ int nfs_wb_all(struct inode *inode) .range_start = 0, .range_end = LLONG_MAX, }; + int ret; - return sync_inode(inode, &wbc); + trace_nfs_writeback_inode_enter(inode); + + ret = sync_inode(inode, &wbc); + + trace_nfs_writeback_inode_exit(inode, ret); + return ret; } EXPORT_SYMBOL_GPL(nfs_wb_all); @@ -1781,6 +1823,8 @@ int nfs_wb_page(struct inode *inode, struct page *page) }; int ret; + trace_nfs_writeback_page_enter(inode); + for (;;) { wait_on_page_writeback(page); if (clear_page_dirty_for_io(page)) { @@ -1789,14 +1833,15 @@ int nfs_wb_page(struct inode *inode, struct page *page) goto out_error; continue; } + ret = 0; if (!PagePrivate(page)) break; ret = nfs_commit_inode(inode, FLUSH_SYNC); if (ret < 0) goto out_error; } - return 0; out_error: + trace_nfs_writeback_page_exit(inode, ret); return ret; } diff --git a/fs/nfsd/nfs4recover.c b/fs/nfsd/nfs4recover.c index 105a3b0..e0a65a9 100644 --- a/fs/nfsd/nfs4recover.c +++ b/fs/nfsd/nfs4recover.c @@ -173,8 +173,6 @@ nfsd4_create_clid_dir(struct nfs4_client *clp) int status; struct nfsd_net *nn = net_generic(clp->net, nfsd_net_id); - dprintk("NFSD: nfsd4_create_clid_dir for \"%s\"\n", dname); - if (test_and_set_bit(NFSD4_CLIENT_STABLE, &clp->cl_flags)) return; if (!nn->rec_file) diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 43f4229..0874998 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -368,11 +368,8 @@ static struct nfs4_delegation * alloc_init_deleg(struct nfs4_client *clp, struct nfs4_ol_stateid *stp, struct svc_fh *current_fh) { struct nfs4_delegation *dp; - struct nfs4_file *fp = stp->st_file; dprintk("NFSD alloc_init_deleg\n"); - if (fp->fi_had_conflict) - return NULL; if (num_delegations > max_delegations) return NULL; dp = delegstateid(nfs4_alloc_stid(clp, deleg_slab)); @@ -389,8 +386,7 @@ alloc_init_deleg(struct nfs4_client *clp, struct nfs4_ol_stateid *stp, struct sv INIT_LIST_HEAD(&dp->dl_perfile); INIT_LIST_HEAD(&dp->dl_perclnt); INIT_LIST_HEAD(&dp->dl_recall_lru); - get_nfs4_file(fp); - dp->dl_file = fp; + dp->dl_file = NULL; dp->dl_type = NFS4_OPEN_DELEGATE_READ; fh_copy_shallow(&dp->dl_fh, ¤t_fh->fh_handle); dp->dl_time = 0; @@ -3035,7 +3031,7 @@ static int nfs4_setlease(struct nfs4_delegation *dp) if (status) { list_del_init(&dp->dl_perclnt); locks_free_lock(fl); - return -ENOMEM; + return status; } fp->fi_lease = fl; fp->fi_deleg_file = get_file(fl->fl_file); @@ -3044,22 +3040,35 @@ static int nfs4_setlease(struct nfs4_delegation *dp) return 0; } -static int nfs4_set_delegation(struct nfs4_delegation *dp) +static int nfs4_set_delegation(struct nfs4_delegation *dp, struct nfs4_file *fp) { - struct nfs4_file *fp = dp->dl_file; + int status; - if (!fp->fi_lease) - return nfs4_setlease(dp); + if (fp->fi_had_conflict) + return -EAGAIN; + get_nfs4_file(fp); + dp->dl_file = fp; + if (!fp->fi_lease) { + status = nfs4_setlease(dp); + if (status) + goto out_free; + return 0; + } spin_lock(&recall_lock); if (fp->fi_had_conflict) { spin_unlock(&recall_lock); - return -EAGAIN; + status = -EAGAIN; + goto out_free; } atomic_inc(&fp->fi_delegees); list_add(&dp->dl_perfile, &fp->fi_delegations); spin_unlock(&recall_lock); list_add(&dp->dl_perclnt, &dp->dl_stid.sc_client->cl_delegations); return 0; +out_free: + put_nfs4_file(fp); + dp->dl_file = fp; + return status; } static void nfsd4_open_deleg_none_ext(struct nfsd4_open *open, int status) @@ -3134,7 +3143,7 @@ nfs4_open_delegation(struct net *net, struct svc_fh *fh, dp = alloc_init_deleg(oo->oo_owner.so_client, stp, fh); if (dp == NULL) goto out_no_deleg; - status = nfs4_set_delegation(dp); + status = nfs4_set_delegation(dp, stp->st_file); if (status) goto out_free; diff --git a/fs/ocfs2/acl.c b/fs/ocfs2/acl.c index 8a40457..b4f788e 100644 --- a/fs/ocfs2/acl.c +++ b/fs/ocfs2/acl.c @@ -51,10 +51,6 @@ static struct posix_acl *ocfs2_acl_from_xattr(const void *value, size_t size) return ERR_PTR(-EINVAL); count = size / sizeof(struct posix_acl_entry); - if (count < 0) - return ERR_PTR(-EINVAL); - if (count == 0) - return NULL; acl = posix_acl_alloc(count, GFP_NOFS); if (!acl) diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c index 94417a8..f37d3c0 100644 --- a/fs/ocfs2/aops.c +++ b/fs/ocfs2/aops.c @@ -2044,7 +2044,7 @@ int ocfs2_write_end_nolock(struct address_space *mapping, out_write_size: pos += copied; - if (pos > inode->i_size) { + if (pos > i_size_read(inode)) { i_size_write(inode, pos); mark_inode_dirty(inode); } diff --git a/fs/ocfs2/cluster/heartbeat.c b/fs/ocfs2/cluster/heartbeat.c index 5c1c864..363f0dc 100644 --- a/fs/ocfs2/cluster/heartbeat.c +++ b/fs/ocfs2/cluster/heartbeat.c @@ -628,11 +628,9 @@ static void o2hb_fire_callbacks(struct o2hb_callback *hbcall, struct o2nm_node *node, int idx) { - struct list_head *iter; struct o2hb_callback_func *f; - list_for_each(iter, &hbcall->list) { - f = list_entry(iter, struct o2hb_callback_func, hc_item); + list_for_each_entry(f, &hbcall->list, hc_item) { mlog(ML_HEARTBEAT, "calling funcs %p\n", f); (f->hc_func)(node, idx, f->hc_data); } @@ -641,16 +639,9 @@ static void o2hb_fire_callbacks(struct o2hb_callback *hbcall, /* Will run the list in order until we process the passed event */ static void o2hb_run_event_list(struct o2hb_node_event *queued_event) { - int empty; struct o2hb_callback *hbcall; struct o2hb_node_event *event; - spin_lock(&o2hb_live_lock); - empty = list_empty(&queued_event->hn_item); - spin_unlock(&o2hb_live_lock); - if (empty) - return; - /* Holding callback sem assures we don't alter the callback * lists when doing this, and serializes ourselves with other * processes wanting callbacks. */ @@ -709,6 +700,7 @@ static void o2hb_shutdown_slot(struct o2hb_disk_slot *slot) struct o2hb_node_event event = { .hn_item = LIST_HEAD_INIT(event.hn_item), }; struct o2nm_node *node; + int queued = 0; node = o2nm_get_node_by_num(slot->ds_node_num); if (!node) @@ -726,11 +718,13 @@ static void o2hb_shutdown_slot(struct o2hb_disk_slot *slot) o2hb_queue_node_event(&event, O2HB_NODE_DOWN_CB, node, slot->ds_node_num); + queued = 1; } } spin_unlock(&o2hb_live_lock); - o2hb_run_event_list(&event); + if (queued) + o2hb_run_event_list(&event); o2nm_node_put(node); } @@ -790,6 +784,7 @@ static int o2hb_check_slot(struct o2hb_region *reg, unsigned int dead_ms = o2hb_dead_threshold * O2HB_REGION_TIMEOUT_MS; unsigned int slot_dead_ms; int tmp; + int queued = 0; memcpy(hb_block, slot->ds_raw_block, reg->hr_block_bytes); @@ -883,6 +878,7 @@ fire_callbacks: slot->ds_node_num); changed = 1; + queued = 1; } list_add_tail(&slot->ds_live_item, @@ -934,6 +930,7 @@ fire_callbacks: node, slot->ds_node_num); changed = 1; + queued = 1; } /* We don't clear this because the node is still @@ -949,7 +946,8 @@ fire_callbacks: out: spin_unlock(&o2hb_live_lock); - o2hb_run_event_list(&event); + if (queued) + o2hb_run_event_list(&event); if (node) o2nm_node_put(node); @@ -2516,8 +2514,7 @@ unlock: int o2hb_register_callback(const char *region_uuid, struct o2hb_callback_func *hc) { - struct o2hb_callback_func *tmp; - struct list_head *iter; + struct o2hb_callback_func *f; struct o2hb_callback *hbcall; int ret; @@ -2540,10 +2537,9 @@ int o2hb_register_callback(const char *region_uuid, down_write(&o2hb_callback_sem); - list_for_each(iter, &hbcall->list) { - tmp = list_entry(iter, struct o2hb_callback_func, hc_item); - if (hc->hc_priority < tmp->hc_priority) { - list_add_tail(&hc->hc_item, iter); + list_for_each_entry(f, &hbcall->list, hc_item) { + if (hc->hc_priority < f->hc_priority) { + list_add_tail(&hc->hc_item, &f->hc_item); break; } } diff --git a/fs/ocfs2/cluster/tcp.c b/fs/ocfs2/cluster/tcp.c index d644dc6..2cd2406 100644 --- a/fs/ocfs2/cluster/tcp.c +++ b/fs/ocfs2/cluster/tcp.c @@ -543,8 +543,9 @@ static void o2net_set_nn_state(struct o2net_node *nn, } if (was_valid && !valid) { - printk(KERN_NOTICE "o2net: No longer connected to " - SC_NODEF_FMT "\n", SC_NODEF_ARGS(old_sc)); + if (old_sc) + printk(KERN_NOTICE "o2net: No longer connected to " + SC_NODEF_FMT "\n", SC_NODEF_ARGS(old_sc)); o2net_complete_nodes_nsw(nn); } @@ -765,32 +766,32 @@ static struct o2net_msg_handler * o2net_handler_tree_lookup(u32 msg_type, u32 key, struct rb_node ***ret_p, struct rb_node **ret_parent) { - struct rb_node **p = &o2net_handler_tree.rb_node; - struct rb_node *parent = NULL; + struct rb_node **p = &o2net_handler_tree.rb_node; + struct rb_node *parent = NULL; struct o2net_msg_handler *nmh, *ret = NULL; int cmp; - while (*p) { - parent = *p; - nmh = rb_entry(parent, struct o2net_msg_handler, nh_node); + while (*p) { + parent = *p; + nmh = rb_entry(parent, struct o2net_msg_handler, nh_node); cmp = o2net_handler_cmp(nmh, msg_type, key); - if (cmp < 0) - p = &(*p)->rb_left; - else if (cmp > 0) - p = &(*p)->rb_right; - else { + if (cmp < 0) + p = &(*p)->rb_left; + else if (cmp > 0) + p = &(*p)->rb_right; + else { ret = nmh; - break; + break; } - } + } - if (ret_p != NULL) - *ret_p = p; - if (ret_parent != NULL) - *ret_parent = parent; + if (ret_p != NULL) + *ret_p = p; + if (ret_parent != NULL) + *ret_parent = parent; - return ret; + return ret; } static void o2net_handler_kref_release(struct kref *kref) @@ -1695,13 +1696,12 @@ static void o2net_start_connect(struct work_struct *work) ret = 0; out: - if (ret) { + if (ret && sc) { printk(KERN_NOTICE "o2net: Connect attempt to " SC_NODEF_FMT " failed with errno %d\n", SC_NODEF_ARGS(sc), ret); /* 0 err so that another will be queued and attempted * from set_nn_state */ - if (sc) - o2net_ensure_shutdown(nn, sc, 0); + o2net_ensure_shutdown(nn, sc, 0); } if (sc) sc_put(sc); @@ -1873,12 +1873,16 @@ static int o2net_accept_one(struct socket *sock) if (o2nm_this_node() >= node->nd_num) { local_node = o2nm_get_node_by_num(o2nm_this_node()); - printk(KERN_NOTICE "o2net: Unexpected connect attempt seen " - "at node '%s' (%u, %pI4:%d) from node '%s' (%u, " - "%pI4:%d)\n", local_node->nd_name, local_node->nd_num, - &(local_node->nd_ipv4_address), - ntohs(local_node->nd_ipv4_port), node->nd_name, - node->nd_num, &sin.sin_addr.s_addr, ntohs(sin.sin_port)); + if (local_node) + printk(KERN_NOTICE "o2net: Unexpected connect attempt " + "seen at node '%s' (%u, %pI4:%d) from " + "node '%s' (%u, %pI4:%d)\n", + local_node->nd_name, local_node->nd_num, + &(local_node->nd_ipv4_address), + ntohs(local_node->nd_ipv4_port), + node->nd_name, + node->nd_num, &sin.sin_addr.s_addr, + ntohs(sin.sin_port)); ret = -EINVAL; goto out; } diff --git a/fs/ocfs2/dlm/dlmast.c b/fs/ocfs2/dlm/dlmast.c index fbec0be..b46278f 100644 --- a/fs/ocfs2/dlm/dlmast.c +++ b/fs/ocfs2/dlm/dlmast.c @@ -292,7 +292,7 @@ int dlm_proxy_ast_handler(struct o2net_msg *msg, u32 len, void *data, struct dlm_lock *lock = NULL; struct dlm_proxy_ast *past = (struct dlm_proxy_ast *) msg->buf; char *name; - struct list_head *iter, *head=NULL; + struct list_head *head = NULL; __be64 cookie; u32 flags; u8 node; @@ -373,8 +373,7 @@ int dlm_proxy_ast_handler(struct o2net_msg *msg, u32 len, void *data, /* try convert queue for both ast/bast */ head = &res->converting; lock = NULL; - list_for_each(iter, head) { - lock = list_entry (iter, struct dlm_lock, list); + list_for_each_entry(lock, head, list) { if (lock->ml.cookie == cookie) goto do_ast; } @@ -385,8 +384,7 @@ int dlm_proxy_ast_handler(struct o2net_msg *msg, u32 len, void *data, else head = &res->granted; - list_for_each(iter, head) { - lock = list_entry (iter, struct dlm_lock, list); + list_for_each_entry(lock, head, list) { if (lock->ml.cookie == cookie) goto do_ast; } diff --git a/fs/ocfs2/dlm/dlmcommon.h b/fs/ocfs2/dlm/dlmcommon.h index de854cc..e051776 100644 --- a/fs/ocfs2/dlm/dlmcommon.h +++ b/fs/ocfs2/dlm/dlmcommon.h @@ -1079,11 +1079,9 @@ static inline int dlm_lock_compatible(int existing, int request) static inline int dlm_lock_on_list(struct list_head *head, struct dlm_lock *lock) { - struct list_head *iter; struct dlm_lock *tmplock; - list_for_each(iter, head) { - tmplock = list_entry(iter, struct dlm_lock, list); + list_for_each_entry(tmplock, head, list) { if (tmplock == lock) return 1; } diff --git a/fs/ocfs2/dlm/dlmconvert.c b/fs/ocfs2/dlm/dlmconvert.c index 29a886d..e36d63f 100644 --- a/fs/ocfs2/dlm/dlmconvert.c +++ b/fs/ocfs2/dlm/dlmconvert.c @@ -123,7 +123,6 @@ static enum dlm_status __dlmconvert_master(struct dlm_ctxt *dlm, int *kick_thread) { enum dlm_status status = DLM_NORMAL; - struct list_head *iter; struct dlm_lock *tmplock=NULL; assert_spin_locked(&res->spinlock); @@ -185,16 +184,14 @@ static enum dlm_status __dlmconvert_master(struct dlm_ctxt *dlm, /* upconvert from here on */ status = DLM_NORMAL; - list_for_each(iter, &res->granted) { - tmplock = list_entry(iter, struct dlm_lock, list); + list_for_each_entry(tmplock, &res->granted, list) { if (tmplock == lock) continue; if (!dlm_lock_compatible(tmplock->ml.type, type)) goto switch_queues; } - list_for_each(iter, &res->converting) { - tmplock = list_entry(iter, struct dlm_lock, list); + list_for_each_entry(tmplock, &res->converting, list) { if (!dlm_lock_compatible(tmplock->ml.type, type)) goto switch_queues; /* existing conversion requests take precedence */ @@ -424,8 +421,8 @@ int dlm_convert_lock_handler(struct o2net_msg *msg, u32 len, void *data, struct dlm_ctxt *dlm = data; struct dlm_convert_lock *cnv = (struct dlm_convert_lock *)msg->buf; struct dlm_lock_resource *res = NULL; - struct list_head *iter; struct dlm_lock *lock = NULL; + struct dlm_lock *tmp_lock; struct dlm_lockstatus *lksb; enum dlm_status status = DLM_NORMAL; u32 flags; @@ -471,14 +468,13 @@ int dlm_convert_lock_handler(struct o2net_msg *msg, u32 len, void *data, dlm_error(status); goto leave; } - list_for_each(iter, &res->granted) { - lock = list_entry(iter, struct dlm_lock, list); - if (lock->ml.cookie == cnv->cookie && - lock->ml.node == cnv->node_idx) { + list_for_each_entry(tmp_lock, &res->granted, list) { + if (tmp_lock->ml.cookie == cnv->cookie && + tmp_lock->ml.node == cnv->node_idx) { + lock = tmp_lock; dlm_lock_get(lock); break; } - lock = NULL; } spin_unlock(&res->spinlock); if (!lock) { diff --git a/fs/ocfs2/dlm/dlmdebug.c b/fs/ocfs2/dlm/dlmdebug.c index 0e28e24..e33cd7a 100644 --- a/fs/ocfs2/dlm/dlmdebug.c +++ b/fs/ocfs2/dlm/dlmdebug.c @@ -96,7 +96,6 @@ static void __dlm_print_lock(struct dlm_lock *lock) void __dlm_print_one_lock_resource(struct dlm_lock_resource *res) { - struct list_head *iter2; struct dlm_lock *lock; char buf[DLM_LOCKID_NAME_MAX]; @@ -118,18 +117,15 @@ void __dlm_print_one_lock_resource(struct dlm_lock_resource *res) res->inflight_locks, atomic_read(&res->asts_reserved)); dlm_print_lockres_refmap(res); printk(" granted queue:\n"); - list_for_each(iter2, &res->granted) { - lock = list_entry(iter2, struct dlm_lock, list); + list_for_each_entry(lock, &res->granted, list) { __dlm_print_lock(lock); } printk(" converting queue:\n"); - list_for_each(iter2, &res->converting) { - lock = list_entry(iter2, struct dlm_lock, list); + list_for_each_entry(lock, &res->converting, list) { __dlm_print_lock(lock); } printk(" blocked queue:\n"); - list_for_each(iter2, &res->blocked) { - lock = list_entry(iter2, struct dlm_lock, list); + list_for_each_entry(lock, &res->blocked, list) { __dlm_print_lock(lock); } } @@ -446,7 +442,6 @@ static int debug_mle_print(struct dlm_ctxt *dlm, char *buf, int len) { struct dlm_master_list_entry *mle; struct hlist_head *bucket; - struct hlist_node *list; int i, out = 0; unsigned long total = 0, longest = 0, bucket_count = 0; @@ -456,9 +451,7 @@ static int debug_mle_print(struct dlm_ctxt *dlm, char *buf, int len) spin_lock(&dlm->master_lock); for (i = 0; i < DLM_HASH_BUCKETS; i++) { bucket = dlm_master_hash(dlm, i); - hlist_for_each(list, bucket) { - mle = hlist_entry(list, struct dlm_master_list_entry, - master_hash_node); + hlist_for_each_entry(mle, bucket, master_hash_node) { ++total; ++bucket_count; if (len - out < 200) diff --git a/fs/ocfs2/dlm/dlmdomain.c b/fs/ocfs2/dlm/dlmdomain.c index dbb17c0..8b3382a 100644 --- a/fs/ocfs2/dlm/dlmdomain.c +++ b/fs/ocfs2/dlm/dlmdomain.c @@ -193,7 +193,7 @@ struct dlm_lock_resource * __dlm_lookup_lockres_full(struct dlm_ctxt *dlm, unsigned int hash) { struct hlist_head *bucket; - struct hlist_node *list; + struct dlm_lock_resource *res; mlog(0, "%.*s\n", len, name); @@ -201,9 +201,7 @@ struct dlm_lock_resource * __dlm_lookup_lockres_full(struct dlm_ctxt *dlm, bucket = dlm_lockres_hash(dlm, hash); - hlist_for_each(list, bucket) { - struct dlm_lock_resource *res = hlist_entry(list, - struct dlm_lock_resource, hash_node); + hlist_for_each_entry(res, bucket, hash_node) { if (res->lockname.name[0] != name[0]) continue; if (unlikely(res->lockname.len != len)) @@ -262,22 +260,19 @@ struct dlm_lock_resource * dlm_lookup_lockres(struct dlm_ctxt *dlm, static struct dlm_ctxt * __dlm_lookup_domain_full(const char *domain, int len) { - struct dlm_ctxt *tmp = NULL; - struct list_head *iter; + struct dlm_ctxt *tmp; assert_spin_locked(&dlm_domain_lock); /* tmp->name here is always NULL terminated, * but domain may not be! */ - list_for_each(iter, &dlm_domains) { - tmp = list_entry (iter, struct dlm_ctxt, list); + list_for_each_entry(tmp, &dlm_domains, list) { if (strlen(tmp->name) == len && memcmp(tmp->name, domain, len)==0) - break; - tmp = NULL; + return tmp; } - return tmp; + return NULL; } /* For null terminated domain strings ONLY */ @@ -366,25 +361,22 @@ static void __dlm_get(struct dlm_ctxt *dlm) * you shouldn't trust your pointer. */ struct dlm_ctxt *dlm_grab(struct dlm_ctxt *dlm) { - struct list_head *iter; - struct dlm_ctxt *target = NULL; + struct dlm_ctxt *target; + struct dlm_ctxt *ret = NULL; spin_lock(&dlm_domain_lock); - list_for_each(iter, &dlm_domains) { - target = list_entry (iter, struct dlm_ctxt, list); - + list_for_each_entry(target, &dlm_domains, list) { if (target == dlm) { __dlm_get(target); + ret = target; break; } - - target = NULL; } spin_unlock(&dlm_domain_lock); - return target; + return ret; } int dlm_domain_fully_joined(struct dlm_ctxt *dlm) @@ -2296,13 +2288,10 @@ static DECLARE_RWSEM(dlm_callback_sem); void dlm_fire_domain_eviction_callbacks(struct dlm_ctxt *dlm, int node_num) { - struct list_head *iter; struct dlm_eviction_cb *cb; down_read(&dlm_callback_sem); - list_for_each(iter, &dlm->dlm_eviction_callbacks) { - cb = list_entry(iter, struct dlm_eviction_cb, ec_item); - + list_for_each_entry(cb, &dlm->dlm_eviction_callbacks, ec_item) { cb->ec_func(node_num, cb->ec_data); } up_read(&dlm_callback_sem); diff --git a/fs/ocfs2/dlm/dlmlock.c b/fs/ocfs2/dlm/dlmlock.c index 47e67c2..5d32f75 100644 --- a/fs/ocfs2/dlm/dlmlock.c +++ b/fs/ocfs2/dlm/dlmlock.c @@ -91,19 +91,14 @@ void dlm_destroy_lock_cache(void) static int dlm_can_grant_new_lock(struct dlm_lock_resource *res, struct dlm_lock *lock) { - struct list_head *iter; struct dlm_lock *tmplock; - list_for_each(iter, &res->granted) { - tmplock = list_entry(iter, struct dlm_lock, list); - + list_for_each_entry(tmplock, &res->granted, list) { if (!dlm_lock_compatible(tmplock->ml.type, lock->ml.type)) return 0; } - list_for_each(iter, &res->converting) { - tmplock = list_entry(iter, struct dlm_lock, list); - + list_for_each_entry(tmplock, &res->converting, list) { if (!dlm_lock_compatible(tmplock->ml.type, lock->ml.type)) return 0; if (!dlm_lock_compatible(tmplock->ml.convert_type, diff --git a/fs/ocfs2/dlm/dlmmaster.c b/fs/ocfs2/dlm/dlmmaster.c index 33ecbe0..cf0f103 100644 --- a/fs/ocfs2/dlm/dlmmaster.c +++ b/fs/ocfs2/dlm/dlmmaster.c @@ -342,16 +342,13 @@ static int dlm_find_mle(struct dlm_ctxt *dlm, { struct dlm_master_list_entry *tmpmle; struct hlist_head *bucket; - struct hlist_node *list; unsigned int hash; assert_spin_locked(&dlm->master_lock); hash = dlm_lockid_hash(name, namelen); bucket = dlm_master_hash(dlm, hash); - hlist_for_each(list, bucket) { - tmpmle = hlist_entry(list, struct dlm_master_list_entry, - master_hash_node); + hlist_for_each_entry(tmpmle, bucket, master_hash_node) { if (!dlm_mle_equal(dlm, tmpmle, name, namelen)) continue; dlm_get_mle(tmpmle); @@ -3183,7 +3180,7 @@ void dlm_clean_master_list(struct dlm_ctxt *dlm, u8 dead_node) struct dlm_master_list_entry *mle; struct dlm_lock_resource *res; struct hlist_head *bucket; - struct hlist_node *list; + struct hlist_node *tmp; unsigned int i; mlog(0, "dlm=%s, dead node=%u\n", dlm->name, dead_node); @@ -3194,10 +3191,7 @@ top: spin_lock(&dlm->master_lock); for (i = 0; i < DLM_HASH_BUCKETS; i++) { bucket = dlm_master_hash(dlm, i); - hlist_for_each(list, bucket) { - mle = hlist_entry(list, struct dlm_master_list_entry, - master_hash_node); - + hlist_for_each_entry_safe(mle, tmp, bucket, master_hash_node) { BUG_ON(mle->type != DLM_MLE_BLOCK && mle->type != DLM_MLE_MASTER && mle->type != DLM_MLE_MIGRATION); @@ -3378,7 +3372,7 @@ void dlm_force_free_mles(struct dlm_ctxt *dlm) int i; struct hlist_head *bucket; struct dlm_master_list_entry *mle; - struct hlist_node *tmp, *list; + struct hlist_node *tmp; /* * We notified all other nodes that we are exiting the domain and @@ -3394,9 +3388,7 @@ void dlm_force_free_mles(struct dlm_ctxt *dlm) for (i = 0; i < DLM_HASH_BUCKETS; i++) { bucket = dlm_master_hash(dlm, i); - hlist_for_each_safe(list, tmp, bucket) { - mle = hlist_entry(list, struct dlm_master_list_entry, - master_hash_node); + hlist_for_each_entry_safe(mle, tmp, bucket, master_hash_node) { if (mle->type != DLM_MLE_BLOCK) { mlog(ML_ERROR, "bad mle: %p\n", mle); dlm_print_one_mle(mle); diff --git a/fs/ocfs2/dlm/dlmrecovery.c b/fs/ocfs2/dlm/dlmrecovery.c index 773bd32..0b5adca 100644 --- a/fs/ocfs2/dlm/dlmrecovery.c +++ b/fs/ocfs2/dlm/dlmrecovery.c @@ -787,6 +787,7 @@ static int dlm_request_all_locks(struct dlm_ctxt *dlm, u8 request_from, { struct dlm_lock_request lr; int ret; + int status; mlog(0, "\n"); @@ -800,13 +801,15 @@ static int dlm_request_all_locks(struct dlm_ctxt *dlm, u8 request_from, // send message ret = o2net_send_message(DLM_LOCK_REQUEST_MSG, dlm->key, - &lr, sizeof(lr), request_from, NULL); + &lr, sizeof(lr), request_from, &status); /* negative status is handled by caller */ if (ret < 0) mlog(ML_ERROR, "%s: Error %d send LOCK_REQUEST to node %u " "to recover dead node %u\n", dlm->name, ret, request_from, dead_node); + else + ret = status; // return from here, then // sleep until all received or error return ret; @@ -2328,6 +2331,14 @@ static void dlm_do_local_recovery_cleanup(struct dlm_ctxt *dlm, u8 dead_node) } else if (res->owner == dlm->node_num) { dlm_free_dead_locks(dlm, res, dead_node); __dlm_lockres_calc_usage(dlm, res); + } else if (res->owner == DLM_LOCK_RES_OWNER_UNKNOWN) { + if (test_bit(dead_node, res->refmap)) { + mlog(0, "%s:%.*s: dead node %u had a ref, but had " + "no locks and had not purged before dying\n", + dlm->name, res->lockname.len, + res->lockname.name, dead_node); + dlm_lockres_clear_refmap_bit(dlm, res, dead_node); + } } spin_unlock(&res->spinlock); } diff --git a/fs/ocfs2/dlm/dlmthread.c b/fs/ocfs2/dlm/dlmthread.c index e73c833..9db869d 100644 --- a/fs/ocfs2/dlm/dlmthread.c +++ b/fs/ocfs2/dlm/dlmthread.c @@ -286,8 +286,6 @@ static void dlm_shuffle_lists(struct dlm_ctxt *dlm, struct dlm_lock_resource *res) { struct dlm_lock *lock, *target; - struct list_head *iter; - struct list_head *head; int can_grant = 1; /* @@ -314,9 +312,7 @@ converting: dlm->name, res->lockname.len, res->lockname.name); BUG(); } - head = &res->granted; - list_for_each(iter, head) { - lock = list_entry(iter, struct dlm_lock, list); + list_for_each_entry(lock, &res->granted, list) { if (lock==target) continue; if (!dlm_lock_compatible(lock->ml.type, @@ -333,9 +329,8 @@ converting: target->ml.convert_type; } } - head = &res->converting; - list_for_each(iter, head) { - lock = list_entry(iter, struct dlm_lock, list); + + list_for_each_entry(lock, &res->converting, list) { if (lock==target) continue; if (!dlm_lock_compatible(lock->ml.type, @@ -384,9 +379,7 @@ blocked: goto leave; target = list_entry(res->blocked.next, struct dlm_lock, list); - head = &res->granted; - list_for_each(iter, head) { - lock = list_entry(iter, struct dlm_lock, list); + list_for_each_entry(lock, &res->granted, list) { if (lock==target) continue; if (!dlm_lock_compatible(lock->ml.type, target->ml.type)) { @@ -400,9 +393,7 @@ blocked: } } - head = &res->converting; - list_for_each(iter, head) { - lock = list_entry(iter, struct dlm_lock, list); + list_for_each_entry(lock, &res->converting, list) { if (lock==target) continue; if (!dlm_lock_compatible(lock->ml.type, target->ml.type)) { diff --git a/fs/ocfs2/dlm/dlmunlock.c b/fs/ocfs2/dlm/dlmunlock.c index 850aa7e..5698b52 100644 --- a/fs/ocfs2/dlm/dlmunlock.c +++ b/fs/ocfs2/dlm/dlmunlock.c @@ -388,7 +388,6 @@ int dlm_unlock_lock_handler(struct o2net_msg *msg, u32 len, void *data, struct dlm_ctxt *dlm = data; struct dlm_unlock_lock *unlock = (struct dlm_unlock_lock *)msg->buf; struct dlm_lock_resource *res = NULL; - struct list_head *iter; struct dlm_lock *lock = NULL; enum dlm_status status = DLM_NORMAL; int found = 0, i; @@ -458,8 +457,7 @@ int dlm_unlock_lock_handler(struct o2net_msg *msg, u32 len, void *data, } for (i=0; i<3; i++) { - list_for_each(iter, queue) { - lock = list_entry(iter, struct dlm_lock, list); + list_for_each_entry(lock, queue, list) { if (lock->ml.cookie == unlock->cookie && lock->ml.node == unlock->node_idx) { dlm_lock_get(lock); diff --git a/fs/ocfs2/dlmfs/dlmfs.c b/fs/ocfs2/dlmfs/dlmfs.c index 12bafb7..efa2b3d 100644 --- a/fs/ocfs2/dlmfs/dlmfs.c +++ b/fs/ocfs2/dlmfs/dlmfs.c @@ -401,11 +401,8 @@ static struct inode *dlmfs_get_root_inode(struct super_block *sb) { struct inode *inode = new_inode(sb); umode_t mode = S_IFDIR | 0755; - struct dlmfs_inode_private *ip; if (inode) { - ip = DLMFS_I(inode); - inode->i_ino = get_next_ino(); inode_init_owner(inode, NULL, mode); inode->i_mapping->backing_dev_info = &dlmfs_backing_dev_info; diff --git a/fs/ocfs2/extent_map.c b/fs/ocfs2/extent_map.c index 2487116..767370b 100644 --- a/fs/ocfs2/extent_map.c +++ b/fs/ocfs2/extent_map.c @@ -781,7 +781,6 @@ int ocfs2_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, cpos = map_start >> osb->s_clustersize_bits; mapping_end = ocfs2_clusters_for_bytes(inode->i_sb, map_start + map_len); - mapping_end -= cpos; is_last = 0; while (cpos < mapping_end && !is_last) { u32 fe_flags; @@ -852,20 +851,20 @@ int ocfs2_seek_data_hole_offset(struct file *file, loff_t *offset, int whence) down_read(&OCFS2_I(inode)->ip_alloc_sem); - if (*offset >= inode->i_size) { + if (*offset >= i_size_read(inode)) { ret = -ENXIO; goto out_unlock; } if (OCFS2_I(inode)->ip_dyn_features & OCFS2_INLINE_DATA_FL) { if (whence == SEEK_HOLE) - *offset = inode->i_size; + *offset = i_size_read(inode); goto out_unlock; } clen = 0; cpos = *offset >> cs_bits; - cend = ocfs2_clusters_for_bytes(inode->i_sb, inode->i_size); + cend = ocfs2_clusters_for_bytes(inode->i_sb, i_size_read(inode)); while (cpos < cend && !is_last) { ret = ocfs2_get_clusters_nocache(inode, di_bh, cpos, &hole_size, @@ -904,8 +903,8 @@ int ocfs2_seek_data_hole_offset(struct file *file, loff_t *offset, int whence) extlen = clen; extlen <<= cs_bits; - if ((extoff + extlen) > inode->i_size) - extlen = inode->i_size - extoff; + if ((extoff + extlen) > i_size_read(inode)) + extlen = i_size_read(inode) - extoff; extoff += extlen; if (extoff > *offset) *offset = extoff; diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index 3261d71..4f8197c 100644 --- a/fs/ocfs2/file.c +++ b/fs/ocfs2/file.c @@ -671,11 +671,7 @@ restarted_transaction: } else { BUG_ON(why != RESTART_TRANS); - /* TODO: This can be more intelligent. */ - credits = ocfs2_calc_extend_credits(osb->sb, - &fe->id2.i_list, - clusters_to_add); - status = ocfs2_extend_trans(handle, credits); + status = ocfs2_allocate_extend_trans(handle, 1); if (status < 0) { /* handle still has to be committed at * this point. */ @@ -1800,6 +1796,7 @@ static int ocfs2_remove_inode_range(struct inode *inode, ocfs2_truncate_cluster_pages(inode, byte_start, byte_len); out: + ocfs2_free_path(path); ocfs2_schedule_truncate_log_flush(osb, 1); ocfs2_run_deallocs(osb, &dealloc); diff --git a/fs/ocfs2/ioctl.c b/fs/ocfs2/ioctl.c index 0c60ef2..fa32ce9 100644 --- a/fs/ocfs2/ioctl.c +++ b/fs/ocfs2/ioctl.c @@ -303,7 +303,7 @@ int ocfs2_info_handle_journal_size(struct inode *inode, if (o2info_from_user(oij, req)) goto bail; - oij.ij_journal_size = osb->journal->j_inode->i_size; + oij.ij_journal_size = i_size_read(osb->journal->j_inode); o2info_set_request_filled(&oij.ij_req); diff --git a/fs/ocfs2/journal.c b/fs/ocfs2/journal.c index 242170d..44fc3e5 100644 --- a/fs/ocfs2/journal.c +++ b/fs/ocfs2/journal.c @@ -455,6 +455,41 @@ bail: return status; } +/* + * If we have fewer than thresh credits, extend by OCFS2_MAX_TRANS_DATA. + * If that fails, restart the transaction & regain write access for the + * buffer head which is used for metadata modifications. + * Taken from Ext4: extend_or_restart_transaction() + */ +int ocfs2_allocate_extend_trans(handle_t *handle, int thresh) +{ + int status, old_nblks; + + BUG_ON(!handle); + + old_nblks = handle->h_buffer_credits; + trace_ocfs2_allocate_extend_trans(old_nblks, thresh); + + if (old_nblks < thresh) + return 0; + + status = jbd2_journal_extend(handle, OCFS2_MAX_TRANS_DATA); + if (status < 0) { + mlog_errno(status); + goto bail; + } + + if (status > 0) { + status = jbd2_journal_restart(handle, OCFS2_MAX_TRANS_DATA); + if (status < 0) + mlog_errno(status); + } + +bail: + return status; +} + + struct ocfs2_triggers { struct jbd2_buffer_trigger_type ot_triggers; int ot_offset; @@ -801,14 +836,14 @@ int ocfs2_journal_init(struct ocfs2_journal *journal, int *dirty) inode_lock = 1; di = (struct ocfs2_dinode *)bh->b_data; - if (inode->i_size < OCFS2_MIN_JOURNAL_SIZE) { + if (i_size_read(inode) < OCFS2_MIN_JOURNAL_SIZE) { mlog(ML_ERROR, "Journal file size (%lld) is too small!\n", - inode->i_size); + i_size_read(inode)); status = -EINVAL; goto done; } - trace_ocfs2_journal_init(inode->i_size, + trace_ocfs2_journal_init(i_size_read(inode), (unsigned long long)inode->i_blocks, OCFS2_I(inode)->ip_clusters); @@ -1096,7 +1131,7 @@ static int ocfs2_force_read_journal(struct inode *inode) memset(bhs, 0, sizeof(struct buffer_head *) * CONCURRENT_JOURNAL_FILL); - num_blocks = ocfs2_blocks_for_bytes(inode->i_sb, inode->i_size); + num_blocks = ocfs2_blocks_for_bytes(inode->i_sb, i_size_read(inode)); v_blkno = 0; while (v_blkno < num_blocks) { status = ocfs2_extent_map_get_blocks(inode, v_blkno, diff --git a/fs/ocfs2/journal.h b/fs/ocfs2/journal.h index 0a99273..0b479ba 100644 --- a/fs/ocfs2/journal.h +++ b/fs/ocfs2/journal.h @@ -258,6 +258,17 @@ handle_t *ocfs2_start_trans(struct ocfs2_super *osb, int ocfs2_commit_trans(struct ocfs2_super *osb, handle_t *handle); int ocfs2_extend_trans(handle_t *handle, int nblocks); +int ocfs2_allocate_extend_trans(handle_t *handle, + int thresh); + +/* + * Define an arbitrary limit for the amount of data we will anticipate + * writing to any given transaction. For unbounded transactions such as + * fallocate(2) we can write more than this, but we always + * start off at the maximum transaction size and grow the transaction + * optimistically as we go. + */ +#define OCFS2_MAX_TRANS_DATA 64U /* * Create access is for when we get a newly created buffer and we're diff --git a/fs/ocfs2/localalloc.c b/fs/ocfs2/localalloc.c index aebeacd..cd5496b 100644 --- a/fs/ocfs2/localalloc.c +++ b/fs/ocfs2/localalloc.c @@ -1082,7 +1082,7 @@ static int ocfs2_local_alloc_reserve_for_window(struct ocfs2_super *osb, } retry_enospc: - (*ac)->ac_bits_wanted = osb->local_alloc_default_bits; + (*ac)->ac_bits_wanted = osb->local_alloc_bits; status = ocfs2_reserve_cluster_bitmap_bits(osb, *ac); if (status == -ENOSPC) { if (ocfs2_recalc_la_window(osb, OCFS2_LA_EVENT_ENOSPC) == @@ -1154,7 +1154,7 @@ retry_enospc: OCFS2_LA_DISABLED) goto bail; - ac->ac_bits_wanted = osb->local_alloc_default_bits; + ac->ac_bits_wanted = osb->local_alloc_bits; status = ocfs2_claim_clusters(handle, ac, osb->local_alloc_bits, &cluster_off, diff --git a/fs/ocfs2/move_extents.c b/fs/ocfs2/move_extents.c index 452068b..3d3f3c8 100644 --- a/fs/ocfs2/move_extents.c +++ b/fs/ocfs2/move_extents.c @@ -152,6 +152,7 @@ static int __ocfs2_move_extent(handle_t *handle, } out: + ocfs2_free_path(path); return ret; } @@ -845,7 +846,7 @@ static int __ocfs2_move_extents_range(struct buffer_head *di_bh, struct ocfs2_move_extents *range = context->range; struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); - if ((inode->i_size == 0) || (range->me_len == 0)) + if ((i_size_read(inode) == 0) || (range->me_len == 0)) return 0; if (OCFS2_I(inode)->ip_dyn_features & OCFS2_INLINE_DATA_FL) diff --git a/fs/ocfs2/ocfs2_trace.h b/fs/ocfs2/ocfs2_trace.h index 3b481f4..1b60c62 100644 --- a/fs/ocfs2/ocfs2_trace.h +++ b/fs/ocfs2/ocfs2_trace.h @@ -2579,6 +2579,8 @@ DEFINE_OCFS2_INT_INT_EVENT(ocfs2_extend_trans); DEFINE_OCFS2_INT_EVENT(ocfs2_extend_trans_restart); +DEFINE_OCFS2_INT_INT_EVENT(ocfs2_allocate_extend_trans); + DEFINE_OCFS2_ULL_ULL_UINT_UINT_EVENT(ocfs2_journal_access); DEFINE_OCFS2_ULL_EVENT(ocfs2_journal_dirty); diff --git a/fs/ocfs2/quota_global.c b/fs/ocfs2/quota_global.c index 332a281..aaa5061 100644 --- a/fs/ocfs2/quota_global.c +++ b/fs/ocfs2/quota_global.c @@ -234,7 +234,7 @@ ssize_t ocfs2_quota_write(struct super_block *sb, int type, len = sb->s_blocksize - OCFS2_QBLK_RESERVED_SPACE - offset; } - if (gqinode->i_size < off + len) { + if (i_size_read(gqinode) < off + len) { loff_t rounded_end = ocfs2_align_bytes_to_blocks(sb, off + len); @@ -778,8 +778,8 @@ static int ocfs2_acquire_dquot(struct dquot *dquot) */ WARN_ON(journal_current_handle()); status = ocfs2_extend_no_holes(gqinode, NULL, - gqinode->i_size + (need_alloc << sb->s_blocksize_bits), - gqinode->i_size); + i_size_read(gqinode) + (need_alloc << sb->s_blocksize_bits), + i_size_read(gqinode)); if (status < 0) goto out_dq; } diff --git a/fs/ocfs2/quota_local.c b/fs/ocfs2/quota_local.c index 27fe7ee..2e4344b 100644 --- a/fs/ocfs2/quota_local.c +++ b/fs/ocfs2/quota_local.c @@ -982,14 +982,14 @@ static struct ocfs2_quota_chunk *ocfs2_local_quota_add_chunk( /* We are protected by dqio_sem so no locking needed */ status = ocfs2_extend_no_holes(lqinode, NULL, - lqinode->i_size + 2 * sb->s_blocksize, - lqinode->i_size); + i_size_read(lqinode) + 2 * sb->s_blocksize, + i_size_read(lqinode)); if (status < 0) { mlog_errno(status); goto out; } status = ocfs2_simple_size_update(lqinode, oinfo->dqi_lqi_bh, - lqinode->i_size + 2 * sb->s_blocksize); + i_size_read(lqinode) + 2 * sb->s_blocksize); if (status < 0) { mlog_errno(status); goto out; @@ -1125,14 +1125,14 @@ static struct ocfs2_quota_chunk *ocfs2_extend_local_quota_file( /* We are protected by dqio_sem so no locking needed */ status = ocfs2_extend_no_holes(lqinode, NULL, - lqinode->i_size + sb->s_blocksize, - lqinode->i_size); + i_size_read(lqinode) + sb->s_blocksize, + i_size_read(lqinode)); if (status < 0) { mlog_errno(status); goto out; } status = ocfs2_simple_size_update(lqinode, oinfo->dqi_lqi_bh, - lqinode->i_size + sb->s_blocksize); + i_size_read(lqinode) + sb->s_blocksize); if (status < 0) { mlog_errno(status); goto out; diff --git a/fs/ocfs2/refcounttree.c b/fs/ocfs2/refcounttree.c index a70d604..bf4dfc1 100644 --- a/fs/ocfs2/refcounttree.c +++ b/fs/ocfs2/refcounttree.c @@ -3854,7 +3854,10 @@ static int ocfs2_attach_refcount_tree(struct inode *inode, while (cpos < clusters) { ret = ocfs2_get_clusters(inode, cpos, &p_cluster, &num_clusters, &ext_flags); - + if (ret) { + mlog_errno(ret); + goto unlock; + } if (p_cluster && !(ext_flags & OCFS2_EXT_REFCOUNTED)) { ret = ocfs2_add_refcount_flag(inode, &di_et, &ref_tree->rf_ci, @@ -4025,7 +4028,10 @@ static int ocfs2_duplicate_extent_list(struct inode *s_inode, while (cpos < clusters) { ret = ocfs2_get_clusters(s_inode, cpos, &p_cluster, &num_clusters, &ext_flags); - + if (ret) { + mlog_errno(ret); + goto out; + } if (p_cluster) { ret = ocfs2_add_refcounted_extent(t_inode, &et, ref_ci, ref_root_bh, diff --git a/fs/ocfs2/xattr.c b/fs/ocfs2/xattr.c index 317ef0a..6ce0686 100644 --- a/fs/ocfs2/xattr.c +++ b/fs/ocfs2/xattr.c @@ -3505,7 +3505,7 @@ int ocfs2_xattr_set(struct inode *inode, int ret, credits, ref_meta = 0, ref_credits = 0; struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); struct inode *tl_inode = osb->osb_tl_inode; - struct ocfs2_xattr_set_ctxt ctxt = { NULL, NULL, }; + struct ocfs2_xattr_set_ctxt ctxt = { NULL, NULL, NULL, }; struct ocfs2_refcount_tree *ref_tree = NULL; struct ocfs2_xattr_info xi = { @@ -3609,13 +3609,14 @@ int ocfs2_xattr_set(struct inode *inode, if (IS_ERR(ctxt.handle)) { ret = PTR_ERR(ctxt.handle); mlog_errno(ret); - goto cleanup; + goto out_free_ac; } ret = __ocfs2_xattr_set_handle(inode, di, &xi, &xis, &xbs, &ctxt); ocfs2_commit_trans(osb, ctxt.handle); +out_free_ac: if (ctxt.data_ac) ocfs2_free_alloc_context(ctxt.data_ac); if (ctxt.meta_ac) @@ -5881,6 +5882,10 @@ static int ocfs2_xattr_value_attach_refcount(struct inode *inode, while (cpos < clusters) { ret = ocfs2_xattr_get_clusters(inode, cpos, &p_cluster, &num_clusters, el, &ext_flags); + if (ret) { + mlog_errno(ret); + break; + } cpos += num_clusters; if ((ext_flags & OCFS2_EXT_REFCOUNTED)) @@ -6797,7 +6802,7 @@ out: if (ret) { if (*meta_ac) { ocfs2_free_alloc_context(*meta_ac); - meta_ac = NULL; + *meta_ac = NULL; } } diff --git a/fs/ocfs2/xattr.h b/fs/ocfs2/xattr.h index e5c7f15..19f134e 100644 --- a/fs/ocfs2/xattr.h +++ b/fs/ocfs2/xattr.h @@ -32,7 +32,7 @@ enum ocfs2_xattr_type { struct ocfs2_security_xattr_info { int enable; - char *name; + const char *name; void *value; size_t value_len; }; @@ -443,7 +443,7 @@ retry: goto dput_and_out; error = -EPERM; - if (!nsown_capable(CAP_SYS_CHROOT)) + if (!ns_capable(current_user_ns(), CAP_SYS_CHROOT)) goto dput_and_out; error = security_path_chroot(&path); if (error) @@ -19,11 +19,14 @@ #define CL_EXPIRE 0x01 #define CL_SLAVE 0x02 -#define CL_COPY_ALL 0x04 +#define CL_COPY_UNBINDABLE 0x04 #define CL_MAKE_SHARED 0x08 #define CL_PRIVATE 0x10 #define CL_SHARED_TO_SLAVE 0x20 #define CL_UNPRIVILEGED 0x40 +#define CL_COPY_MNT_NS_FILE 0x80 + +#define CL_COPY_ALL (CL_COPY_UNBINDABLE | CL_COPY_MNT_NS_FILE) static inline void set_mnt_shared(struct mount *mnt) { diff --git a/fs/proc/fd.c b/fs/proc/fd.c index 0ff80f9..985ea88 100644 --- a/fs/proc/fd.c +++ b/fs/proc/fd.c @@ -286,7 +286,7 @@ int proc_fd_permission(struct inode *inode, int mask) int rv = generic_permission(inode, mask); if (rv == 0) return 0; - if (task_pid(current) == proc_pid(inode)) + if (task_tgid(current) == proc_pid(inode)) rv = 0; return rv; } diff --git a/fs/proc/root.c b/fs/proc/root.c index e0a790d..87dbcbe 100644 --- a/fs/proc/root.c +++ b/fs/proc/root.c @@ -110,7 +110,11 @@ static struct dentry *proc_mount(struct file_system_type *fs_type, ns = task_active_pid_ns(current); options = data; - if (!current_user_ns()->may_mount_proc) + if (!capable(CAP_SYS_ADMIN) && !fs_fully_visible(fs_type)) + return ERR_PTR(-EPERM); + + /* Does the mounter have privilege over the pid namespace? */ + if (!ns_capable(ns->user_ns, CAP_SYS_ADMIN)) return ERR_PTR(-EPERM); } diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index 107d026..7366e9d 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -740,6 +740,9 @@ static inline void clear_soft_dirty(struct vm_area_struct *vma, ptent = pte_file_clear_soft_dirty(ptent); } + if (vma->vm_flags & VM_SOFTDIRTY) + vma->vm_flags &= ~VM_SOFTDIRTY; + set_pte_at(vma->vm_mm, addr, pte, ptent); #endif } @@ -949,13 +952,15 @@ static void pte_to_pagemap_entry(pagemap_entry_t *pme, struct pagemapread *pm, if (is_migration_entry(entry)) page = migration_entry_to_page(entry); } else { - *pme = make_pme(PM_NOT_PRESENT(pm->v2)); + if (vma->vm_flags & VM_SOFTDIRTY) + flags2 |= __PM_SOFT_DIRTY; + *pme = make_pme(PM_NOT_PRESENT(pm->v2) | PM_STATUS2(pm->v2, flags2)); return; } if (page && !PageAnon(page)) flags |= PM_FILE; - if (pte_soft_dirty(pte)) + if ((vma->vm_flags & VM_SOFTDIRTY) || pte_soft_dirty(pte)) flags2 |= __PM_SOFT_DIRTY; *pme = make_pme(PM_PFRAME(frame) | PM_STATUS2(pm->v2, flags2) | flags); @@ -974,7 +979,7 @@ static void thp_pmd_to_pagemap_entry(pagemap_entry_t *pme, struct pagemapread *p *pme = make_pme(PM_PFRAME(pmd_pfn(pmd) + offset) | PM_STATUS2(pm->v2, pmd_flags2) | PM_PRESENT); else - *pme = make_pme(PM_NOT_PRESENT(pm->v2)); + *pme = make_pme(PM_NOT_PRESENT(pm->v2) | PM_STATUS2(pm->v2, pmd_flags2)); } #else static inline void thp_pmd_to_pagemap_entry(pagemap_entry_t *pme, struct pagemapread *pm, @@ -997,7 +1002,11 @@ static int pagemap_pte_range(pmd_t *pmd, unsigned long addr, unsigned long end, if (vma && pmd_trans_huge_lock(pmd, vma) == 1) { int pmd_flags2; - pmd_flags2 = (pmd_soft_dirty(*pmd) ? __PM_SOFT_DIRTY : 0); + if ((vma->vm_flags & VM_SOFTDIRTY) || pmd_soft_dirty(*pmd)) + pmd_flags2 = __PM_SOFT_DIRTY; + else + pmd_flags2 = 0; + for (; addr != end; addr += PAGE_SIZE) { unsigned long offset; @@ -1015,12 +1024,17 @@ static int pagemap_pte_range(pmd_t *pmd, unsigned long addr, unsigned long end, if (pmd_trans_unstable(pmd)) return 0; for (; addr != end; addr += PAGE_SIZE) { + int flags2; /* check to see if we've left 'vma' behind * and need a new, higher one */ if (vma && (addr >= vma->vm_end)) { vma = find_vma(walk->mm, addr); - pme = make_pme(PM_NOT_PRESENT(pm->v2)); + if (vma && (vma->vm_flags & VM_SOFTDIRTY)) + flags2 = __PM_SOFT_DIRTY; + else + flags2 = 0; + pme = make_pme(PM_NOT_PRESENT(pm->v2) | PM_STATUS2(pm->v2, flags2)); } /* check that 'vma' actually covers this address, @@ -1044,13 +1058,15 @@ static int pagemap_pte_range(pmd_t *pmd, unsigned long addr, unsigned long end, #ifdef CONFIG_HUGETLB_PAGE static void huge_pte_to_pagemap_entry(pagemap_entry_t *pme, struct pagemapread *pm, - pte_t pte, int offset) + pte_t pte, int offset, int flags2) { if (pte_present(pte)) - *pme = make_pme(PM_PFRAME(pte_pfn(pte) + offset) - | PM_STATUS2(pm->v2, 0) | PM_PRESENT); + *pme = make_pme(PM_PFRAME(pte_pfn(pte) + offset) | + PM_STATUS2(pm->v2, flags2) | + PM_PRESENT); else - *pme = make_pme(PM_NOT_PRESENT(pm->v2)); + *pme = make_pme(PM_NOT_PRESENT(pm->v2) | + PM_STATUS2(pm->v2, flags2)); } /* This function walks within one hugetlb entry in the single call */ @@ -1059,12 +1075,22 @@ static int pagemap_hugetlb_range(pte_t *pte, unsigned long hmask, struct mm_walk *walk) { struct pagemapread *pm = walk->private; + struct vm_area_struct *vma; int err = 0; + int flags2; pagemap_entry_t pme; + vma = find_vma(walk->mm, addr); + WARN_ON_ONCE(!vma); + + if (vma && (vma->vm_flags & VM_SOFTDIRTY)) + flags2 = __PM_SOFT_DIRTY; + else + flags2 = 0; + for (; addr != end; addr += PAGE_SIZE) { int offset = (addr & ~hmask) >> PAGE_SHIFT; - huge_pte_to_pagemap_entry(&pme, pm, *pte, offset); + huge_pte_to_pagemap_entry(&pme, pm, *pte, offset, flags2); err = add_to_pagemap(addr, &pme, pm); if (err) return err; @@ -1376,8 +1402,10 @@ static int show_numa_map(struct seq_file *m, void *v, int is_pid) walk.mm = mm; pol = get_vma_policy(task, vma, vma->vm_start); - mpol_to_str(buffer, sizeof(buffer), pol); + n = mpol_to_str(buffer, sizeof(buffer), pol); mpol_cond_put(pol); + if (n < 0) + return n; seq_printf(m, "%08lx %s", vma->vm_start, buffer); diff --git a/fs/proc/vmcore.c b/fs/proc/vmcore.c index a1a16eb..9100d69 100644 --- a/fs/proc/vmcore.c +++ b/fs/proc/vmcore.c @@ -21,6 +21,7 @@ #include <linux/crash_dump.h> #include <linux/list.h> #include <linux/vmalloc.h> +#include <linux/pagemap.h> #include <asm/uaccess.h> #include <asm/io.h> #include "internal.h" @@ -123,11 +124,65 @@ static ssize_t read_from_oldmem(char *buf, size_t count, return read; } +/* + * Architectures may override this function to allocate ELF header in 2nd kernel + */ +int __weak elfcorehdr_alloc(unsigned long long *addr, unsigned long long *size) +{ + return 0; +} + +/* + * Architectures may override this function to free header + */ +void __weak elfcorehdr_free(unsigned long long addr) +{} + +/* + * Architectures may override this function to read from ELF header + */ +ssize_t __weak elfcorehdr_read(char *buf, size_t count, u64 *ppos) +{ + return read_from_oldmem(buf, count, ppos, 0); +} + +/* + * Architectures may override this function to read from notes sections + */ +ssize_t __weak elfcorehdr_read_notes(char *buf, size_t count, u64 *ppos) +{ + return read_from_oldmem(buf, count, ppos, 0); +} + +/* + * Architectures may override this function to map oldmem + */ +int __weak remap_oldmem_pfn_range(struct vm_area_struct *vma, + unsigned long from, unsigned long pfn, + unsigned long size, pgprot_t prot) +{ + return remap_pfn_range(vma, from, pfn, size, prot); +} + +/* + * Copy to either kernel or user space + */ +static int copy_to(void *target, void *src, size_t size, int userbuf) +{ + if (userbuf) { + if (copy_to_user((char __user *) target, src, size)) + return -EFAULT; + } else { + memcpy(target, src, size); + } + return 0; +} + /* Read from the ELF header and then the crash dump. On error, negative value is * returned otherwise number of bytes read are returned. */ -static ssize_t read_vmcore(struct file *file, char __user *buffer, - size_t buflen, loff_t *fpos) +static ssize_t __read_vmcore(char *buffer, size_t buflen, loff_t *fpos, + int userbuf) { ssize_t acc = 0, tmp; size_t tsz; @@ -144,7 +199,7 @@ static ssize_t read_vmcore(struct file *file, char __user *buffer, /* Read ELF core header */ if (*fpos < elfcorebuf_sz) { tsz = min(elfcorebuf_sz - (size_t)*fpos, buflen); - if (copy_to_user(buffer, elfcorebuf + *fpos, tsz)) + if (copy_to(buffer, elfcorebuf + *fpos, tsz, userbuf)) return -EFAULT; buflen -= tsz; *fpos += tsz; @@ -162,7 +217,7 @@ static ssize_t read_vmcore(struct file *file, char __user *buffer, tsz = min(elfcorebuf_sz + elfnotes_sz - (size_t)*fpos, buflen); kaddr = elfnotes_buf + *fpos - elfcorebuf_sz; - if (copy_to_user(buffer, kaddr, tsz)) + if (copy_to(buffer, kaddr, tsz, userbuf)) return -EFAULT; buflen -= tsz; *fpos += tsz; @@ -178,7 +233,7 @@ static ssize_t read_vmcore(struct file *file, char __user *buffer, if (*fpos < m->offset + m->size) { tsz = min_t(size_t, m->offset + m->size - *fpos, buflen); start = m->paddr + *fpos - m->offset; - tmp = read_from_oldmem(buffer, tsz, &start, 1); + tmp = read_from_oldmem(buffer, tsz, &start, userbuf); if (tmp < 0) return tmp; buflen -= tsz; @@ -195,6 +250,55 @@ static ssize_t read_vmcore(struct file *file, char __user *buffer, return acc; } +static ssize_t read_vmcore(struct file *file, char __user *buffer, + size_t buflen, loff_t *fpos) +{ + return __read_vmcore((__force char *) buffer, buflen, fpos, 1); +} + +/* + * The vmcore fault handler uses the page cache and fills data using the + * standard __vmcore_read() function. + * + * On s390 the fault handler is used for memory regions that can't be mapped + * directly with remap_pfn_range(). + */ +static int mmap_vmcore_fault(struct vm_area_struct *vma, struct vm_fault *vmf) +{ +#ifdef CONFIG_S390 + struct address_space *mapping = vma->vm_file->f_mapping; + pgoff_t index = vmf->pgoff; + struct page *page; + loff_t offset; + char *buf; + int rc; + + page = find_or_create_page(mapping, index, GFP_KERNEL); + if (!page) + return VM_FAULT_OOM; + if (!PageUptodate(page)) { + offset = (loff_t) index << PAGE_CACHE_SHIFT; + buf = __va((page_to_pfn(page) << PAGE_SHIFT)); + rc = __read_vmcore(buf, PAGE_SIZE, &offset, 0); + if (rc < 0) { + unlock_page(page); + page_cache_release(page); + return (rc == -ENOMEM) ? VM_FAULT_OOM : VM_FAULT_SIGBUS; + } + SetPageUptodate(page); + } + unlock_page(page); + vmf->page = page; + return 0; +#else + return VM_FAULT_SIGBUS; +#endif +} + +static const struct vm_operations_struct vmcore_mmap_ops = { + .fault = mmap_vmcore_fault, +}; + /** * alloc_elfnotes_buf - allocate buffer for ELF note segment in * vmalloc memory @@ -223,7 +327,7 @@ static inline char *alloc_elfnotes_buf(size_t notes_sz) * regions in the 1st kernel pointed to by PT_LOAD entries) into * virtually contiguous user-space in ELF layout. */ -#if defined(CONFIG_MMU) && !defined(CONFIG_S390) +#ifdef CONFIG_MMU static int mmap_vmcore(struct file *file, struct vm_area_struct *vma) { size_t size = vma->vm_end - vma->vm_start; @@ -241,6 +345,7 @@ static int mmap_vmcore(struct file *file, struct vm_area_struct *vma) vma->vm_flags &= ~(VM_MAYWRITE | VM_MAYEXEC); vma->vm_flags |= VM_MIXEDMAP; + vma->vm_ops = &vmcore_mmap_ops; len = 0; @@ -282,9 +387,9 @@ static int mmap_vmcore(struct file *file, struct vm_area_struct *vma) tsz = min_t(size_t, m->offset + m->size - start, size); paddr = m->paddr + start - m->offset; - if (remap_pfn_range(vma, vma->vm_start + len, - paddr >> PAGE_SHIFT, tsz, - vma->vm_page_prot)) + if (remap_oldmem_pfn_range(vma, vma->vm_start + len, + paddr >> PAGE_SHIFT, tsz, + vma->vm_page_prot)) goto fail; size -= tsz; start += tsz; @@ -357,7 +462,7 @@ static int __init update_note_header_size_elf64(const Elf64_Ehdr *ehdr_ptr) notes_section = kmalloc(max_sz, GFP_KERNEL); if (!notes_section) return -ENOMEM; - rc = read_from_oldmem(notes_section, max_sz, &offset, 0); + rc = elfcorehdr_read_notes(notes_section, max_sz, &offset); if (rc < 0) { kfree(notes_section); return rc; @@ -444,7 +549,8 @@ static int __init copy_notes_elf64(const Elf64_Ehdr *ehdr_ptr, char *notes_buf) if (phdr_ptr->p_type != PT_NOTE) continue; offset = phdr_ptr->p_offset; - rc = read_from_oldmem(notes_buf, phdr_ptr->p_memsz, &offset, 0); + rc = elfcorehdr_read_notes(notes_buf, phdr_ptr->p_memsz, + &offset); if (rc < 0) return rc; notes_buf += phdr_ptr->p_memsz; @@ -536,7 +642,7 @@ static int __init update_note_header_size_elf32(const Elf32_Ehdr *ehdr_ptr) notes_section = kmalloc(max_sz, GFP_KERNEL); if (!notes_section) return -ENOMEM; - rc = read_from_oldmem(notes_section, max_sz, &offset, 0); + rc = elfcorehdr_read_notes(notes_section, max_sz, &offset); if (rc < 0) { kfree(notes_section); return rc; @@ -623,7 +729,8 @@ static int __init copy_notes_elf32(const Elf32_Ehdr *ehdr_ptr, char *notes_buf) if (phdr_ptr->p_type != PT_NOTE) continue; offset = phdr_ptr->p_offset; - rc = read_from_oldmem(notes_buf, phdr_ptr->p_memsz, &offset, 0); + rc = elfcorehdr_read_notes(notes_buf, phdr_ptr->p_memsz, + &offset); if (rc < 0) return rc; notes_buf += phdr_ptr->p_memsz; @@ -810,7 +917,7 @@ static int __init parse_crash_elf64_headers(void) addr = elfcorehdr_addr; /* Read Elf header */ - rc = read_from_oldmem((char*)&ehdr, sizeof(Elf64_Ehdr), &addr, 0); + rc = elfcorehdr_read((char *)&ehdr, sizeof(Elf64_Ehdr), &addr); if (rc < 0) return rc; @@ -837,7 +944,7 @@ static int __init parse_crash_elf64_headers(void) if (!elfcorebuf) return -ENOMEM; addr = elfcorehdr_addr; - rc = read_from_oldmem(elfcorebuf, elfcorebuf_sz_orig, &addr, 0); + rc = elfcorehdr_read(elfcorebuf, elfcorebuf_sz_orig, &addr); if (rc < 0) goto fail; @@ -866,7 +973,7 @@ static int __init parse_crash_elf32_headers(void) addr = elfcorehdr_addr; /* Read Elf header */ - rc = read_from_oldmem((char*)&ehdr, sizeof(Elf32_Ehdr), &addr, 0); + rc = elfcorehdr_read((char *)&ehdr, sizeof(Elf32_Ehdr), &addr); if (rc < 0) return rc; @@ -892,7 +999,7 @@ static int __init parse_crash_elf32_headers(void) if (!elfcorebuf) return -ENOMEM; addr = elfcorehdr_addr; - rc = read_from_oldmem(elfcorebuf, elfcorebuf_sz_orig, &addr, 0); + rc = elfcorehdr_read(elfcorebuf, elfcorebuf_sz_orig, &addr); if (rc < 0) goto fail; @@ -919,7 +1026,7 @@ static int __init parse_crash_elf_headers(void) int rc=0; addr = elfcorehdr_addr; - rc = read_from_oldmem(e_ident, EI_NIDENT, &addr, 0); + rc = elfcorehdr_read(e_ident, EI_NIDENT, &addr); if (rc < 0) return rc; if (memcmp(e_ident, ELFMAG, SELFMAG) != 0) { @@ -952,7 +1059,14 @@ static int __init vmcore_init(void) { int rc = 0; - /* If elfcorehdr= has been passed in cmdline, then capture the dump.*/ + /* Allow architectures to allocate ELF header in 2nd kernel */ + rc = elfcorehdr_alloc(&elfcorehdr_addr, &elfcorehdr_size); + if (rc) + return rc; + /* + * If elfcorehdr= has been passed in cmdline or created in 2nd kernel, + * then capture the dump. + */ if (!(is_vmcore_usable())) return rc; rc = parse_crash_elf_headers(); @@ -960,6 +1074,8 @@ static int __init vmcore_init(void) pr_warn("Kdump: vmcore not initialized\n"); return rc; } + elfcorehdr_free(elfcorehdr_addr); + elfcorehdr_addr = ELFCORE_ADDR_ERR; proc_vmcore = proc_create("vmcore", S_IRUSR, NULL, &proc_vmcore_operations); if (proc_vmcore) diff --git a/fs/quota/quota.c b/fs/quota/quota.c index c7314f1..dea86e8 100644 --- a/fs/quota/quota.c +++ b/fs/quota/quota.c @@ -27,6 +27,7 @@ static int check_quotactl_permission(struct super_block *sb, int type, int cmd, case Q_SYNC: case Q_GETINFO: case Q_XGETQSTAT: + case Q_XGETQSTATV: case Q_XQUOTASYNC: break; /* allow to query information for dquots we "own" */ @@ -217,6 +218,31 @@ static int quota_getxstate(struct super_block *sb, void __user *addr) return ret; } +static int quota_getxstatev(struct super_block *sb, void __user *addr) +{ + struct fs_quota_statv fqs; + int ret; + + if (!sb->s_qcop->get_xstatev) + return -ENOSYS; + + memset(&fqs, 0, sizeof(fqs)); + if (copy_from_user(&fqs, addr, 1)) /* Just read qs_version */ + return -EFAULT; + + /* If this kernel doesn't support user specified version, fail */ + switch (fqs.qs_version) { + case FS_QSTATV_VERSION1: + break; + default: + return -EINVAL; + } + ret = sb->s_qcop->get_xstatev(sb, &fqs); + if (!ret && copy_to_user(addr, &fqs, sizeof(fqs))) + return -EFAULT; + return ret; +} + static int quota_setxquota(struct super_block *sb, int type, qid_t id, void __user *addr) { @@ -293,6 +319,8 @@ static int do_quotactl(struct super_block *sb, int type, int cmd, qid_t id, return quota_setxstate(sb, cmd, addr); case Q_XGETQSTAT: return quota_getxstate(sb, addr); + case Q_XGETQSTATV: + return quota_getxstatev(sb, addr); case Q_XSETQLIM: return quota_setxquota(sb, type, id, addr); case Q_XGETQUOTA: @@ -317,6 +345,7 @@ static int quotactl_cmd_write(int cmd) case Q_GETINFO: case Q_SYNC: case Q_XGETQSTAT: + case Q_XGETQSTATV: case Q_XGETQUOTA: case Q_XQUOTASYNC: return 0; diff --git a/fs/ramfs/inode.c b/fs/ramfs/inode.c index c24f1e1..39d1465 100644 --- a/fs/ramfs/inode.c +++ b/fs/ramfs/inode.c @@ -244,12 +244,6 @@ struct dentry *ramfs_mount(struct file_system_type *fs_type, return mount_nodev(fs_type, flags, data, ramfs_fill_super); } -static struct dentry *rootfs_mount(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data) -{ - return mount_nodev(fs_type, flags|MS_NOUSER, data, ramfs_fill_super); -} - static void ramfs_kill_sb(struct super_block *sb) { kfree(sb->s_fs_info); @@ -262,29 +256,23 @@ static struct file_system_type ramfs_fs_type = { .kill_sb = ramfs_kill_sb, .fs_flags = FS_USERNS_MOUNT, }; -static struct file_system_type rootfs_fs_type = { - .name = "rootfs", - .mount = rootfs_mount, - .kill_sb = kill_litter_super, -}; -static int __init init_ramfs_fs(void) -{ - return register_filesystem(&ramfs_fs_type); -} -module_init(init_ramfs_fs) - -int __init init_rootfs(void) +int __init init_ramfs_fs(void) { + static unsigned long once; int err; + if (test_and_set_bit(0, &once)) + return 0; + err = bdi_init(&ramfs_backing_dev_info); if (err) return err; - err = register_filesystem(&rootfs_fs_type); + err = register_filesystem(&ramfs_fs_type); if (err) bdi_destroy(&ramfs_backing_dev_info); return err; } +module_init(init_ramfs_fs) diff --git a/fs/squashfs/block.c b/fs/squashfs/block.c index fb50652..41d108e 100644 --- a/fs/squashfs/block.c +++ b/fs/squashfs/block.c @@ -167,17 +167,14 @@ int squashfs_read_data(struct super_block *sb, void **buffer, u64 index, /* * Block is uncompressed. */ - int i, in, pg_offset = 0; - - for (i = 0; i < b; i++) { - wait_on_buffer(bh[i]); - if (!buffer_uptodate(bh[i])) - goto block_release; - } + int in, pg_offset = 0; for (bytes = length; k < b; k++) { in = min(bytes, msblk->devblksize - offset); bytes -= in; + wait_on_buffer(bh[k]); + if (!buffer_uptodate(bh[k])) + goto block_release; while (in) { if (pg_offset == PAGE_CACHE_SIZE) { page++; diff --git a/fs/squashfs/dir.c b/fs/squashfs/dir.c index f7f527b..d8c2d74 100644 --- a/fs/squashfs/dir.c +++ b/fs/squashfs/dir.c @@ -54,6 +54,7 @@ static int get_dir_index_using_offset(struct super_block *sb, { struct squashfs_sb_info *msblk = sb->s_fs_info; int err, i, index, length = 0; + unsigned int size; struct squashfs_dir_index dir_index; TRACE("Entered get_dir_index_using_offset, i_count %d, f_pos %lld\n", @@ -81,8 +82,14 @@ static int get_dir_index_using_offset(struct super_block *sb, */ break; + size = le32_to_cpu(dir_index.size) + 1; + + /* size should never be larger than SQUASHFS_NAME_LEN */ + if (size > SQUASHFS_NAME_LEN) + break; + err = squashfs_read_metadata(sb, NULL, &index_start, - &index_offset, le32_to_cpu(dir_index.size) + 1); + &index_offset, size); if (err < 0) break; @@ -105,9 +112,8 @@ static int squashfs_readdir(struct file *file, struct dir_context *ctx) struct inode *inode = file_inode(file); struct squashfs_sb_info *msblk = inode->i_sb->s_fs_info; u64 block = squashfs_i(inode)->start + msblk->directory_table; - int offset = squashfs_i(inode)->offset, length, dir_count, size, - type, err; - unsigned int inode_number; + int offset = squashfs_i(inode)->offset, length, err; + unsigned int inode_number, dir_count, size, type; struct squashfs_dir_header dirh; struct squashfs_dir_entry *dire; @@ -200,6 +206,9 @@ static int squashfs_readdir(struct file *file, struct dir_context *ctx) ((short) le16_to_cpu(dire->inode_number)); type = le16_to_cpu(dire->type); + if (type > SQUASHFS_MAX_DIR_TYPE) + goto failed_read; + if (!dir_emit(ctx, dire->name, size, inode_number, squashfs_filetype_table[type])) diff --git a/fs/squashfs/namei.c b/fs/squashfs/namei.c index 7834a51..67cad77 100644 --- a/fs/squashfs/namei.c +++ b/fs/squashfs/namei.c @@ -79,7 +79,8 @@ static int get_dir_index_using_name(struct super_block *sb, int len) { struct squashfs_sb_info *msblk = sb->s_fs_info; - int i, size, length = 0, err; + int i, length = 0, err; + unsigned int size; struct squashfs_dir_index *index; char *str; @@ -103,6 +104,8 @@ static int get_dir_index_using_name(struct super_block *sb, size = le32_to_cpu(index->size) + 1; + if (size > SQUASHFS_NAME_LEN) + break; err = squashfs_read_metadata(sb, index->name, &index_start, &index_offset, size); @@ -144,7 +147,8 @@ static struct dentry *squashfs_lookup(struct inode *dir, struct dentry *dentry, struct squashfs_dir_entry *dire; u64 block = squashfs_i(dir)->start + msblk->directory_table; int offset = squashfs_i(dir)->offset; - int err, length, dir_count, size; + int err, length; + unsigned int dir_count, size; TRACE("Entered squashfs_lookup [%llx:%x]\n", block, offset); diff --git a/fs/squashfs/squashfs_fs.h b/fs/squashfs/squashfs_fs.h index 9e2349d..4b2beda 100644 --- a/fs/squashfs/squashfs_fs.h +++ b/fs/squashfs/squashfs_fs.h @@ -87,7 +87,7 @@ #define SQUASHFS_COMP_OPTS(flags) SQUASHFS_BIT(flags, \ SQUASHFS_COMP_OPT) -/* Max number of types and file types */ +/* Inode types including extended types */ #define SQUASHFS_DIR_TYPE 1 #define SQUASHFS_REG_TYPE 2 #define SQUASHFS_SYMLINK_TYPE 3 @@ -103,6 +103,9 @@ #define SQUASHFS_LFIFO_TYPE 13 #define SQUASHFS_LSOCKET_TYPE 14 +/* Max type value stored in directory entry */ +#define SQUASHFS_MAX_DIR_TYPE 7 + /* Xattr types */ #define SQUASHFS_XATTR_USER 0 #define SQUASHFS_XATTR_TRUSTED 1 @@ -71,7 +71,7 @@ static int prune_super(struct shrinker *shrink, struct shrink_control *sc) if (!grab_super_passive(sb)) return -1; - if (sb->s_op && sb->s_op->nr_cached_objects) + if (sb->s_op->nr_cached_objects) fs_objects = sb->s_op->nr_cached_objects(sb); total_objects = sb->s_nr_dentry_unused + diff --git a/fs/sysfs/dir.c b/fs/sysfs/dir.c index 99ec5b4..4d83ced 100644 --- a/fs/sysfs/dir.c +++ b/fs/sysfs/dir.c @@ -297,7 +297,6 @@ static int sysfs_dentry_delete(const struct dentry *dentry) static int sysfs_dentry_revalidate(struct dentry *dentry, unsigned int flags) { struct sysfs_dirent *sd; - int is_dir; int type; if (flags & LOOKUP_RCU) @@ -341,18 +340,15 @@ out_bad: * is performed at its new name the dentry will be readded * to the dcache hashes. */ - is_dir = (sysfs_type(sd) == SYSFS_DIR); mutex_unlock(&sysfs_mutex); - if (is_dir) { - /* If we have submounts we must allow the vfs caches - * to lie about the state of the filesystem to prevent - * leaks and other nasty things. - */ - if (have_submounts(dentry)) - goto out_valid; - shrink_dcache_parent(dentry); - } - d_drop(dentry); + + /* If we have submounts we must allow the vfs caches + * to lie about the state of the filesystem to prevent + * leaks and other nasty things. + */ + if (check_submounts_and_drop(dentry) != 0) + goto out_valid; + return 0; } diff --git a/fs/sysfs/mount.c b/fs/sysfs/mount.c index fd7ce7a..834ec2c 100644 --- a/fs/sysfs/mount.c +++ b/fs/sysfs/mount.c @@ -112,8 +112,15 @@ static struct dentry *sysfs_mount(struct file_system_type *fs_type, struct super_block *sb; int error; - if (!(flags & MS_KERNMOUNT) && !current_user_ns()->may_mount_sysfs) - return ERR_PTR(-EPERM); + if (!(flags & MS_KERNMOUNT)) { + if (!capable(CAP_SYS_ADMIN) && !fs_fully_visible(fs_type)) + return ERR_PTR(-EPERM); + + for (type = KOBJ_NS_TYPE_NONE; type < KOBJ_NS_TYPES; type++) { + if (!kobj_ns_current_may_mount(type)) + return ERR_PTR(-EPERM); + } + } info = kzalloc(sizeof(*info), GFP_KERNEL); if (!info) diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile index 4a45080..0719e4d 100644 --- a/fs/xfs/Makefile +++ b/fs/xfs/Makefile @@ -27,9 +27,12 @@ xfs-y += xfs_trace.o # highlevel code xfs-y += xfs_aops.o \ + xfs_attr_inactive.o \ + xfs_attr_list.o \ xfs_bit.o \ + xfs_bmap_util.o \ xfs_buf.o \ - xfs_dfrag.o \ + xfs_dir2_readdir.o \ xfs_discard.o \ xfs_error.o \ xfs_export.o \ @@ -44,11 +47,11 @@ xfs-y += xfs_aops.o \ xfs_iops.o \ xfs_itable.o \ xfs_message.o \ + xfs_mount.o \ xfs_mru_cache.o \ - xfs_rename.o \ xfs_super.o \ - xfs_utils.o \ - xfs_vnodeops.o \ + xfs_symlink.o \ + xfs_trans.o \ xfs_xattr.o \ kmem.o \ uuid.o @@ -73,10 +76,13 @@ xfs-y += xfs_alloc.o \ xfs_ialloc_btree.o \ xfs_icreate_item.o \ xfs_inode.o \ + xfs_inode_fork.o \ + xfs_inode_buf.o \ xfs_log_recover.o \ - xfs_mount.o \ - xfs_symlink.o \ - xfs_trans.o + xfs_log_rlimit.o \ + xfs_sb.o \ + xfs_symlink_remote.o \ + xfs_trans_resv.o # low-level transaction/log code xfs-y += xfs_log.o \ diff --git a/fs/xfs/xfs_acl.c b/fs/xfs/xfs_acl.c index 306d883..6951896 100644 --- a/fs/xfs/xfs_acl.c +++ b/fs/xfs/xfs_acl.c @@ -16,11 +16,13 @@ * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #include "xfs.h" +#include "xfs_log_format.h" +#include "xfs_trans_resv.h" #include "xfs_acl.h" #include "xfs_attr.h" #include "xfs_bmap_btree.h" #include "xfs_inode.h" -#include "xfs_vnodeops.h" +#include "xfs_ag.h" #include "xfs_sb.h" #include "xfs_mount.h" #include "xfs_trace.h" @@ -68,14 +70,15 @@ xfs_acl_from_disk( switch (acl_e->e_tag) { case ACL_USER: + acl_e->e_uid = xfs_uid_to_kuid(be32_to_cpu(ace->ae_id)); + break; case ACL_GROUP: - acl_e->e_id = be32_to_cpu(ace->ae_id); + acl_e->e_gid = xfs_gid_to_kgid(be32_to_cpu(ace->ae_id)); break; case ACL_USER_OBJ: case ACL_GROUP_OBJ: case ACL_MASK: case ACL_OTHER: - acl_e->e_id = ACL_UNDEFINED_ID; break; default: goto fail; @@ -101,7 +104,18 @@ xfs_acl_to_disk(struct xfs_acl *aclp, const struct posix_acl *acl) acl_e = &acl->a_entries[i]; ace->ae_tag = cpu_to_be32(acl_e->e_tag); - ace->ae_id = cpu_to_be32(acl_e->e_id); + switch (acl_e->e_tag) { + case ACL_USER: + ace->ae_id = cpu_to_be32(xfs_kuid_to_uid(acl_e->e_uid)); + break; + case ACL_GROUP: + ace->ae_id = cpu_to_be32(xfs_kgid_to_gid(acl_e->e_gid)); + break; + default: + ace->ae_id = cpu_to_be32(ACL_UNDEFINED_ID); + break; + } + ace->ae_perm = cpu_to_be16(acl_e->e_perm); } } @@ -360,7 +374,7 @@ xfs_xattr_acl_set(struct dentry *dentry, const char *name, return -EINVAL; if (type == ACL_TYPE_DEFAULT && !S_ISDIR(inode->i_mode)) return value ? -EACCES : 0; - if ((current_fsuid() != inode->i_uid) && !capable(CAP_FOWNER)) + if (!inode_owner_or_capable(inode)) return -EPERM; if (!value) diff --git a/fs/xfs/xfs_ag.h b/fs/xfs/xfs_ag.h index 317aa86..1cb740a 100644 --- a/fs/xfs/xfs_ag.h +++ b/fs/xfs/xfs_ag.h @@ -227,59 +227,6 @@ typedef struct xfs_agfl { } xfs_agfl_t; /* - * Per-ag incore structure, copies of information in agf and agi, - * to improve the performance of allocation group selection. - */ -#define XFS_PAGB_NUM_SLOTS 128 - -typedef struct xfs_perag { - struct xfs_mount *pag_mount; /* owner filesystem */ - xfs_agnumber_t pag_agno; /* AG this structure belongs to */ - atomic_t pag_ref; /* perag reference count */ - char pagf_init; /* this agf's entry is initialized */ - char pagi_init; /* this agi's entry is initialized */ - char pagf_metadata; /* the agf is preferred to be metadata */ - char pagi_inodeok; /* The agi is ok for inodes */ - __uint8_t pagf_levels[XFS_BTNUM_AGF]; - /* # of levels in bno & cnt btree */ - __uint32_t pagf_flcount; /* count of blocks in freelist */ - xfs_extlen_t pagf_freeblks; /* total free blocks */ - xfs_extlen_t pagf_longest; /* longest free space */ - __uint32_t pagf_btreeblks; /* # of blocks held in AGF btrees */ - xfs_agino_t pagi_freecount; /* number of free inodes */ - xfs_agino_t pagi_count; /* number of allocated inodes */ - - /* - * Inode allocation search lookup optimisation. - * If the pagino matches, the search for new inodes - * doesn't need to search the near ones again straight away - */ - xfs_agino_t pagl_pagino; - xfs_agino_t pagl_leftrec; - xfs_agino_t pagl_rightrec; -#ifdef __KERNEL__ - spinlock_t pagb_lock; /* lock for pagb_tree */ - struct rb_root pagb_tree; /* ordered tree of busy extents */ - - atomic_t pagf_fstrms; /* # of filestreams active in this AG */ - - spinlock_t pag_ici_lock; /* incore inode cache lock */ - struct radix_tree_root pag_ici_root; /* incore inode cache root */ - int pag_ici_reclaimable; /* reclaimable inodes */ - struct mutex pag_ici_reclaim_lock; /* serialisation point */ - unsigned long pag_ici_reclaim_cursor; /* reclaim restart point */ - - /* buffer cache index */ - spinlock_t pag_buf_lock; /* lock for pag_buf_tree */ - struct rb_root pag_buf_tree; /* ordered tree of active buffers */ - - /* for rcu-safe freeing */ - struct rcu_head rcu_head; -#endif - int pagb_count; /* pagb slots in use */ -} xfs_perag_t; - -/* * tags for inode radix tree */ #define XFS_ICI_NO_TAG (-1) /* special flag for an untagged lookup diff --git a/fs/xfs/xfs_alloc.c b/fs/xfs/xfs_alloc.c index 71596e5..5a1393f 100644 --- a/fs/xfs/xfs_alloc.c +++ b/fs/xfs/xfs_alloc.c @@ -878,7 +878,7 @@ xfs_alloc_ag_vextent_near( xfs_agblock_t ltnew; /* useful start bno of left side */ xfs_extlen_t rlen; /* length of returned extent */ int forced = 0; -#if defined(DEBUG) && defined(__KERNEL__) +#ifdef DEBUG /* * Randomly don't execute the first algorithm. */ @@ -938,8 +938,8 @@ restart: xfs_extlen_t blen=0; xfs_agblock_t bnew=0; -#if defined(DEBUG) && defined(__KERNEL__) - if (!dofirst) +#ifdef DEBUG + if (dofirst) break; #endif /* diff --git a/fs/xfs/xfs_aops.c b/fs/xfs/xfs_aops.c index e11d654..977da0e 100644 --- a/fs/xfs/xfs_aops.c +++ b/fs/xfs/xfs_aops.c @@ -28,9 +28,9 @@ #include "xfs_alloc.h" #include "xfs_error.h" #include "xfs_iomap.h" -#include "xfs_vnodeops.h" #include "xfs_trace.h" #include "xfs_bmap.h" +#include "xfs_bmap_util.h" #include <linux/aio.h> #include <linux/gfp.h> #include <linux/mpage.h> @@ -108,7 +108,7 @@ xfs_setfilesize_trans_alloc( tp = xfs_trans_alloc(mp, XFS_TRANS_FSYNC_TS); - error = xfs_trans_reserve(tp, 0, XFS_FSYNC_TS_LOG_RES(mp), 0, 0, 0); + error = xfs_trans_reserve(tp, &M_RES(mp)->tr_fsyncts, 0, 0); if (error) { xfs_trans_cancel(tp, 0); return error; @@ -440,7 +440,7 @@ xfs_start_page_writeback( end_page_writeback(page); } -static inline int bio_add_buffer(struct bio *bio, struct buffer_head *bh) +static inline int xfs_bio_add_buffer(struct bio *bio, struct buffer_head *bh) { return bio_add_page(bio, bh->b_page, bh->b_size, bh_offset(bh)); } @@ -514,7 +514,7 @@ xfs_submit_ioend( goto retry; } - if (bio_add_buffer(bio, bh) != bh->b_size) { + if (xfs_bio_add_buffer(bio, bh) != bh->b_size) { xfs_submit_ioend_bio(wbc, ioend, bio); goto retry; } @@ -1498,13 +1498,26 @@ xfs_vm_write_failed( loff_t pos, unsigned len) { - loff_t block_offset = pos & PAGE_MASK; + loff_t block_offset; loff_t block_start; loff_t block_end; loff_t from = pos & (PAGE_CACHE_SIZE - 1); loff_t to = from + len; struct buffer_head *bh, *head; + /* + * The request pos offset might be 32 or 64 bit, this is all fine + * on 64-bit platform. However, for 64-bit pos request on 32-bit + * platform, the high 32-bit will be masked off if we evaluate the + * block_offset via (pos & PAGE_MASK) because the PAGE_MASK is + * 0xfffff000 as an unsigned long, hence the result is incorrect + * which could cause the following ASSERT failed in most cases. + * In order to avoid this, we can evaluate the block_offset of the + * start of the page by using shifts rather than masks the mismatch + * problem. + */ + block_offset = (pos >> PAGE_CACHE_SHIFT) << PAGE_CACHE_SHIFT; + ASSERT(block_offset + from == pos); head = page_buffers(page); diff --git a/fs/xfs/xfs_attr.c b/fs/xfs/xfs_attr.c index 20fe3fe..ddcf226 100644 --- a/fs/xfs/xfs_attr.c +++ b/fs/xfs/xfs_attr.c @@ -17,10 +17,11 @@ */ #include "xfs.h" #include "xfs_fs.h" -#include "xfs_types.h" +#include "xfs_format.h" #include "xfs_bit.h" #include "xfs_log.h" #include "xfs_trans.h" +#include "xfs_trans_priv.h" #include "xfs_sb.h" #include "xfs_ag.h" #include "xfs_mount.h" @@ -32,13 +33,13 @@ #include "xfs_alloc.h" #include "xfs_inode_item.h" #include "xfs_bmap.h" +#include "xfs_bmap_util.h" #include "xfs_attr.h" #include "xfs_attr_leaf.h" #include "xfs_attr_remote.h" #include "xfs_error.h" #include "xfs_quota.h" #include "xfs_trans_space.h" -#include "xfs_vnodeops.h" #include "xfs_trace.h" /* @@ -62,7 +63,6 @@ STATIC int xfs_attr_shortform_addname(xfs_da_args_t *args); STATIC int xfs_attr_leaf_get(xfs_da_args_t *args); STATIC int xfs_attr_leaf_addname(xfs_da_args_t *args); STATIC int xfs_attr_leaf_removename(xfs_da_args_t *args); -STATIC int xfs_attr_leaf_list(xfs_attr_list_context_t *context); /* * Internal routines when attribute list is more than one block. @@ -70,7 +70,6 @@ STATIC int xfs_attr_leaf_list(xfs_attr_list_context_t *context); STATIC int xfs_attr_node_get(xfs_da_args_t *args); STATIC int xfs_attr_node_addname(xfs_da_args_t *args); STATIC int xfs_attr_node_removename(xfs_da_args_t *args); -STATIC int xfs_attr_node_list(xfs_attr_list_context_t *context); STATIC int xfs_attr_fillstate(xfs_da_state_t *state); STATIC int xfs_attr_refillstate(xfs_da_state_t *state); @@ -90,7 +89,7 @@ xfs_attr_name_to_xname( return 0; } -STATIC int +int xfs_inode_hasattr( struct xfs_inode *ip) { @@ -227,13 +226,14 @@ xfs_attr_set_int( int valuelen, int flags) { - xfs_da_args_t args; - xfs_fsblock_t firstblock; - xfs_bmap_free_t flist; - int error, err2, committed; - xfs_mount_t *mp = dp->i_mount; - int rsvd = (flags & ATTR_ROOT) != 0; - int local; + xfs_da_args_t args; + xfs_fsblock_t firstblock; + xfs_bmap_free_t flist; + int error, err2, committed; + struct xfs_mount *mp = dp->i_mount; + struct xfs_trans_res tres; + int rsvd = (flags & ATTR_ROOT) != 0; + int local; /* * Attach the dquots to the inode. @@ -293,11 +293,11 @@ xfs_attr_set_int( if (rsvd) args.trans->t_flags |= XFS_TRANS_RESERVE; - error = xfs_trans_reserve(args.trans, args.total, - XFS_ATTRSETM_LOG_RES(mp) + - XFS_ATTRSETRT_LOG_RES(mp) * args.total, - 0, XFS_TRANS_PERM_LOG_RES, - XFS_ATTRSET_LOG_COUNT); + tres.tr_logres = M_RES(mp)->tr_attrsetm.tr_logres + + M_RES(mp)->tr_attrsetrt.tr_logres * args.total; + tres.tr_logcount = XFS_ATTRSET_LOG_COUNT; + tres.tr_logflags = XFS_TRANS_PERM_LOG_RES; + error = xfs_trans_reserve(args.trans, &tres, args.total, 0); if (error) { xfs_trans_cancel(args.trans, 0); return(error); @@ -517,11 +517,9 @@ xfs_attr_remove_int(xfs_inode_t *dp, struct xfs_name *name, int flags) if (flags & ATTR_ROOT) args.trans->t_flags |= XFS_TRANS_RESERVE; - if ((error = xfs_trans_reserve(args.trans, - XFS_ATTRRM_SPACE_RES(mp), - XFS_ATTRRM_LOG_RES(mp), - 0, XFS_TRANS_PERM_LOG_RES, - XFS_ATTRRM_LOG_COUNT))) { + error = xfs_trans_reserve(args.trans, &M_RES(mp)->tr_attrrm, + XFS_ATTRRM_SPACE_RES(mp), 0); + if (error) { xfs_trans_cancel(args.trans, 0); return(error); } @@ -611,228 +609,6 @@ xfs_attr_remove( return xfs_attr_remove_int(dp, &xname, flags); } -int -xfs_attr_list_int(xfs_attr_list_context_t *context) -{ - int error; - xfs_inode_t *dp = context->dp; - - XFS_STATS_INC(xs_attr_list); - - if (XFS_FORCED_SHUTDOWN(dp->i_mount)) - return EIO; - - xfs_ilock(dp, XFS_ILOCK_SHARED); - - /* - * Decide on what work routines to call based on the inode size. - */ - if (!xfs_inode_hasattr(dp)) { - error = 0; - } else if (dp->i_d.di_aformat == XFS_DINODE_FMT_LOCAL) { - error = xfs_attr_shortform_list(context); - } else if (xfs_bmap_one_block(dp, XFS_ATTR_FORK)) { - error = xfs_attr_leaf_list(context); - } else { - error = xfs_attr_node_list(context); - } - - xfs_iunlock(dp, XFS_ILOCK_SHARED); - - return error; -} - -#define ATTR_ENTBASESIZE /* minimum bytes used by an attr */ \ - (((struct attrlist_ent *) 0)->a_name - (char *) 0) -#define ATTR_ENTSIZE(namelen) /* actual bytes used by an attr */ \ - ((ATTR_ENTBASESIZE + (namelen) + 1 + sizeof(u_int32_t)-1) \ - & ~(sizeof(u_int32_t)-1)) - -/* - * Format an attribute and copy it out to the user's buffer. - * Take care to check values and protect against them changing later, - * we may be reading them directly out of a user buffer. - */ -/*ARGSUSED*/ -STATIC int -xfs_attr_put_listent( - xfs_attr_list_context_t *context, - int flags, - unsigned char *name, - int namelen, - int valuelen, - unsigned char *value) -{ - struct attrlist *alist = (struct attrlist *)context->alist; - attrlist_ent_t *aep; - int arraytop; - - ASSERT(!(context->flags & ATTR_KERNOVAL)); - ASSERT(context->count >= 0); - ASSERT(context->count < (ATTR_MAX_VALUELEN/8)); - ASSERT(context->firstu >= sizeof(*alist)); - ASSERT(context->firstu <= context->bufsize); - - /* - * Only list entries in the right namespace. - */ - if (((context->flags & ATTR_SECURE) == 0) != - ((flags & XFS_ATTR_SECURE) == 0)) - return 0; - if (((context->flags & ATTR_ROOT) == 0) != - ((flags & XFS_ATTR_ROOT) == 0)) - return 0; - - arraytop = sizeof(*alist) + - context->count * sizeof(alist->al_offset[0]); - context->firstu -= ATTR_ENTSIZE(namelen); - if (context->firstu < arraytop) { - trace_xfs_attr_list_full(context); - alist->al_more = 1; - context->seen_enough = 1; - return 1; - } - - aep = (attrlist_ent_t *)&context->alist[context->firstu]; - aep->a_valuelen = valuelen; - memcpy(aep->a_name, name, namelen); - aep->a_name[namelen] = 0; - alist->al_offset[context->count++] = context->firstu; - alist->al_count = context->count; - trace_xfs_attr_list_add(context); - return 0; -} - -/* - * Generate a list of extended attribute names and optionally - * also value lengths. Positive return value follows the XFS - * convention of being an error, zero or negative return code - * is the length of the buffer returned (negated), indicating - * success. - */ -int -xfs_attr_list( - xfs_inode_t *dp, - char *buffer, - int bufsize, - int flags, - attrlist_cursor_kern_t *cursor) -{ - xfs_attr_list_context_t context; - struct attrlist *alist; - int error; - - /* - * Validate the cursor. - */ - if (cursor->pad1 || cursor->pad2) - return(XFS_ERROR(EINVAL)); - if ((cursor->initted == 0) && - (cursor->hashval || cursor->blkno || cursor->offset)) - return XFS_ERROR(EINVAL); - - /* - * Check for a properly aligned buffer. - */ - if (((long)buffer) & (sizeof(int)-1)) - return XFS_ERROR(EFAULT); - if (flags & ATTR_KERNOVAL) - bufsize = 0; - - /* - * Initialize the output buffer. - */ - memset(&context, 0, sizeof(context)); - context.dp = dp; - context.cursor = cursor; - context.resynch = 1; - context.flags = flags; - context.alist = buffer; - context.bufsize = (bufsize & ~(sizeof(int)-1)); /* align */ - context.firstu = context.bufsize; - context.put_listent = xfs_attr_put_listent; - - alist = (struct attrlist *)context.alist; - alist->al_count = 0; - alist->al_more = 0; - alist->al_offset[0] = context.bufsize; - - error = xfs_attr_list_int(&context); - ASSERT(error >= 0); - return error; -} - -int /* error */ -xfs_attr_inactive(xfs_inode_t *dp) -{ - xfs_trans_t *trans; - xfs_mount_t *mp; - int error; - - mp = dp->i_mount; - ASSERT(! XFS_NOT_DQATTACHED(mp, dp)); - - xfs_ilock(dp, XFS_ILOCK_SHARED); - if (!xfs_inode_hasattr(dp) || - dp->i_d.di_aformat == XFS_DINODE_FMT_LOCAL) { - xfs_iunlock(dp, XFS_ILOCK_SHARED); - return 0; - } - xfs_iunlock(dp, XFS_ILOCK_SHARED); - - /* - * Start our first transaction of the day. - * - * All future transactions during this code must be "chained" off - * this one via the trans_dup() call. All transactions will contain - * the inode, and the inode will always be marked with trans_ihold(). - * Since the inode will be locked in all transactions, we must log - * the inode in every transaction to let it float upward through - * the log. - */ - trans = xfs_trans_alloc(mp, XFS_TRANS_ATTRINVAL); - if ((error = xfs_trans_reserve(trans, 0, XFS_ATTRINVAL_LOG_RES(mp), 0, - XFS_TRANS_PERM_LOG_RES, - XFS_ATTRINVAL_LOG_COUNT))) { - xfs_trans_cancel(trans, 0); - return(error); - } - xfs_ilock(dp, XFS_ILOCK_EXCL); - - /* - * No need to make quota reservations here. We expect to release some - * blocks, not allocate, in the common case. - */ - xfs_trans_ijoin(trans, dp, 0); - - /* - * Decide on what work routines to call based on the inode size. - */ - if (!xfs_inode_hasattr(dp) || - dp->i_d.di_aformat == XFS_DINODE_FMT_LOCAL) { - error = 0; - goto out; - } - error = xfs_attr3_root_inactive(&trans, dp); - if (error) - goto out; - - error = xfs_itruncate_extents(&trans, dp, XFS_ATTR_FORK, 0); - if (error) - goto out; - - error = xfs_trans_commit(trans, XFS_TRANS_RELEASE_LOG_RES); - xfs_iunlock(dp, XFS_ILOCK_EXCL); - - return(error); - -out: - xfs_trans_cancel(trans, XFS_TRANS_RELEASE_LOG_RES|XFS_TRANS_ABORT); - xfs_iunlock(dp, XFS_ILOCK_EXCL); - return(error); -} - - /*======================================================================== * External routines when attribute list is inside the inode @@ -1166,28 +942,6 @@ xfs_attr_leaf_get(xfs_da_args_t *args) return error; } -/* - * Copy out attribute entries for attr_list(), for leaf attribute lists. - */ -STATIC int -xfs_attr_leaf_list(xfs_attr_list_context_t *context) -{ - int error; - struct xfs_buf *bp; - - trace_xfs_attr_leaf_list(context); - - context->cursor->blkno = 0; - error = xfs_attr3_leaf_read(NULL, context->dp, 0, -1, &bp); - if (error) - return XFS_ERROR(error); - - error = xfs_attr3_leaf_list_int(bp, context); - xfs_trans_brelse(NULL, bp); - return XFS_ERROR(error); -} - - /*======================================================================== * External routines when attribute list size > XFS_LBSIZE(mp). *========================================================================*/ @@ -1260,6 +1014,7 @@ restart: * have been a b-tree. */ xfs_da_state_free(state); + state = NULL; xfs_bmap_init(args->flist, args->firstblock); error = xfs_attr3_leaf_to_node(args); if (!error) { @@ -1780,143 +1535,3 @@ xfs_attr_node_get(xfs_da_args_t *args) xfs_da_state_free(state); return(retval); } - -STATIC int /* error */ -xfs_attr_node_list(xfs_attr_list_context_t *context) -{ - attrlist_cursor_kern_t *cursor; - xfs_attr_leafblock_t *leaf; - xfs_da_intnode_t *node; - struct xfs_attr3_icleaf_hdr leafhdr; - struct xfs_da3_icnode_hdr nodehdr; - struct xfs_da_node_entry *btree; - int error, i; - struct xfs_buf *bp; - - trace_xfs_attr_node_list(context); - - cursor = context->cursor; - cursor->initted = 1; - - /* - * Do all sorts of validation on the passed-in cursor structure. - * If anything is amiss, ignore the cursor and look up the hashval - * starting from the btree root. - */ - bp = NULL; - if (cursor->blkno > 0) { - error = xfs_da3_node_read(NULL, context->dp, cursor->blkno, -1, - &bp, XFS_ATTR_FORK); - if ((error != 0) && (error != EFSCORRUPTED)) - return(error); - if (bp) { - struct xfs_attr_leaf_entry *entries; - - node = bp->b_addr; - switch (be16_to_cpu(node->hdr.info.magic)) { - case XFS_DA_NODE_MAGIC: - case XFS_DA3_NODE_MAGIC: - trace_xfs_attr_list_wrong_blk(context); - xfs_trans_brelse(NULL, bp); - bp = NULL; - break; - case XFS_ATTR_LEAF_MAGIC: - case XFS_ATTR3_LEAF_MAGIC: - leaf = bp->b_addr; - xfs_attr3_leaf_hdr_from_disk(&leafhdr, leaf); - entries = xfs_attr3_leaf_entryp(leaf); - if (cursor->hashval > be32_to_cpu( - entries[leafhdr.count - 1].hashval)) { - trace_xfs_attr_list_wrong_blk(context); - xfs_trans_brelse(NULL, bp); - bp = NULL; - } else if (cursor->hashval <= be32_to_cpu( - entries[0].hashval)) { - trace_xfs_attr_list_wrong_blk(context); - xfs_trans_brelse(NULL, bp); - bp = NULL; - } - break; - default: - trace_xfs_attr_list_wrong_blk(context); - xfs_trans_brelse(NULL, bp); - bp = NULL; - } - } - } - - /* - * We did not find what we expected given the cursor's contents, - * so we start from the top and work down based on the hash value. - * Note that start of node block is same as start of leaf block. - */ - if (bp == NULL) { - cursor->blkno = 0; - for (;;) { - __uint16_t magic; - - error = xfs_da3_node_read(NULL, context->dp, - cursor->blkno, -1, &bp, - XFS_ATTR_FORK); - if (error) - return(error); - node = bp->b_addr; - magic = be16_to_cpu(node->hdr.info.magic); - if (magic == XFS_ATTR_LEAF_MAGIC || - magic == XFS_ATTR3_LEAF_MAGIC) - break; - if (magic != XFS_DA_NODE_MAGIC && - magic != XFS_DA3_NODE_MAGIC) { - XFS_CORRUPTION_ERROR("xfs_attr_node_list(3)", - XFS_ERRLEVEL_LOW, - context->dp->i_mount, - node); - xfs_trans_brelse(NULL, bp); - return XFS_ERROR(EFSCORRUPTED); - } - - xfs_da3_node_hdr_from_disk(&nodehdr, node); - btree = xfs_da3_node_tree_p(node); - for (i = 0; i < nodehdr.count; btree++, i++) { - if (cursor->hashval - <= be32_to_cpu(btree->hashval)) { - cursor->blkno = be32_to_cpu(btree->before); - trace_xfs_attr_list_node_descend(context, - btree); - break; - } - } - if (i == nodehdr.count) { - xfs_trans_brelse(NULL, bp); - return 0; - } - xfs_trans_brelse(NULL, bp); - } - } - ASSERT(bp != NULL); - - /* - * Roll upward through the blocks, processing each leaf block in - * order. As long as there is space in the result buffer, keep - * adding the information. - */ - for (;;) { - leaf = bp->b_addr; - error = xfs_attr3_leaf_list_int(bp, context); - if (error) { - xfs_trans_brelse(NULL, bp); - return error; - } - xfs_attr3_leaf_hdr_from_disk(&leafhdr, leaf); - if (context->seen_enough || leafhdr.forw == 0) - break; - cursor->blkno = leafhdr.forw; - xfs_trans_brelse(NULL, bp); - error = xfs_attr3_leaf_read(NULL, context->dp, cursor->blkno, -1, - &bp); - if (error) - return error; - } - xfs_trans_brelse(NULL, bp); - return 0; -} diff --git a/fs/xfs/xfs_attr.h b/fs/xfs/xfs_attr.h index de8dd58..dd48245 100644 --- a/fs/xfs/xfs_attr.h +++ b/fs/xfs/xfs_attr.h @@ -141,5 +141,14 @@ typedef struct xfs_attr_list_context { */ int xfs_attr_inactive(struct xfs_inode *dp); int xfs_attr_list_int(struct xfs_attr_list_context *); +int xfs_inode_hasattr(struct xfs_inode *ip); +int xfs_attr_get(struct xfs_inode *ip, const unsigned char *name, + unsigned char *value, int *valuelenp, int flags); +int xfs_attr_set(struct xfs_inode *dp, const unsigned char *name, + unsigned char *value, int valuelen, int flags); +int xfs_attr_remove(struct xfs_inode *dp, const unsigned char *name, int flags); +int xfs_attr_list(struct xfs_inode *dp, char *buffer, int bufsize, + int flags, struct attrlist_cursor_kern *cursor); + #endif /* __XFS_ATTR_H__ */ diff --git a/fs/xfs/xfs_attr_inactive.c b/fs/xfs/xfs_attr_inactive.c new file mode 100644 index 0000000..bb24b07 --- /dev/null +++ b/fs/xfs/xfs_attr_inactive.c @@ -0,0 +1,453 @@ +/* + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * Copyright (c) 2013 Red Hat, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#include "xfs.h" +#include "xfs_fs.h" +#include "xfs_format.h" +#include "xfs_bit.h" +#include "xfs_log.h" +#include "xfs_trans.h" +#include "xfs_sb.h" +#include "xfs_ag.h" +#include "xfs_mount.h" +#include "xfs_da_btree.h" +#include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" +#include "xfs_ialloc_btree.h" +#include "xfs_alloc.h" +#include "xfs_btree.h" +#include "xfs_attr_remote.h" +#include "xfs_dinode.h" +#include "xfs_inode.h" +#include "xfs_inode_item.h" +#include "xfs_bmap.h" +#include "xfs_attr.h" +#include "xfs_attr_leaf.h" +#include "xfs_error.h" +#include "xfs_quota.h" +#include "xfs_trace.h" +#include "xfs_trans_priv.h" + +/* + * Look at all the extents for this logical region, + * invalidate any buffers that are incore/in transactions. + */ +STATIC int +xfs_attr3_leaf_freextent( + struct xfs_trans **trans, + struct xfs_inode *dp, + xfs_dablk_t blkno, + int blkcnt) +{ + struct xfs_bmbt_irec map; + struct xfs_buf *bp; + xfs_dablk_t tblkno; + xfs_daddr_t dblkno; + int tblkcnt; + int dblkcnt; + int nmap; + int error; + + /* + * Roll through the "value", invalidating the attribute value's + * blocks. + */ + tblkno = blkno; + tblkcnt = blkcnt; + while (tblkcnt > 0) { + /* + * Try to remember where we decided to put the value. + */ + nmap = 1; + error = xfs_bmapi_read(dp, (xfs_fileoff_t)tblkno, tblkcnt, + &map, &nmap, XFS_BMAPI_ATTRFORK); + if (error) { + return(error); + } + ASSERT(nmap == 1); + ASSERT(map.br_startblock != DELAYSTARTBLOCK); + + /* + * If it's a hole, these are already unmapped + * so there's nothing to invalidate. + */ + if (map.br_startblock != HOLESTARTBLOCK) { + + dblkno = XFS_FSB_TO_DADDR(dp->i_mount, + map.br_startblock); + dblkcnt = XFS_FSB_TO_BB(dp->i_mount, + map.br_blockcount); + bp = xfs_trans_get_buf(*trans, + dp->i_mount->m_ddev_targp, + dblkno, dblkcnt, 0); + if (!bp) + return ENOMEM; + xfs_trans_binval(*trans, bp); + /* + * Roll to next transaction. + */ + error = xfs_trans_roll(trans, dp); + if (error) + return (error); + } + + tblkno += map.br_blockcount; + tblkcnt -= map.br_blockcount; + } + + return(0); +} + +/* + * Invalidate all of the "remote" value regions pointed to by a particular + * leaf block. + * Note that we must release the lock on the buffer so that we are not + * caught holding something that the logging code wants to flush to disk. + */ +STATIC int +xfs_attr3_leaf_inactive( + struct xfs_trans **trans, + struct xfs_inode *dp, + struct xfs_buf *bp) +{ + struct xfs_attr_leafblock *leaf; + struct xfs_attr3_icleaf_hdr ichdr; + struct xfs_attr_leaf_entry *entry; + struct xfs_attr_leaf_name_remote *name_rmt; + struct xfs_attr_inactive_list *list; + struct xfs_attr_inactive_list *lp; + int error; + int count; + int size; + int tmp; + int i; + + leaf = bp->b_addr; + xfs_attr3_leaf_hdr_from_disk(&ichdr, leaf); + + /* + * Count the number of "remote" value extents. + */ + count = 0; + entry = xfs_attr3_leaf_entryp(leaf); + for (i = 0; i < ichdr.count; entry++, i++) { + if (be16_to_cpu(entry->nameidx) && + ((entry->flags & XFS_ATTR_LOCAL) == 0)) { + name_rmt = xfs_attr3_leaf_name_remote(leaf, i); + if (name_rmt->valueblk) + count++; + } + } + + /* + * If there are no "remote" values, we're done. + */ + if (count == 0) { + xfs_trans_brelse(*trans, bp); + return 0; + } + + /* + * Allocate storage for a list of all the "remote" value extents. + */ + size = count * sizeof(xfs_attr_inactive_list_t); + list = kmem_alloc(size, KM_SLEEP); + + /* + * Identify each of the "remote" value extents. + */ + lp = list; + entry = xfs_attr3_leaf_entryp(leaf); + for (i = 0; i < ichdr.count; entry++, i++) { + if (be16_to_cpu(entry->nameidx) && + ((entry->flags & XFS_ATTR_LOCAL) == 0)) { + name_rmt = xfs_attr3_leaf_name_remote(leaf, i); + if (name_rmt->valueblk) { + lp->valueblk = be32_to_cpu(name_rmt->valueblk); + lp->valuelen = xfs_attr3_rmt_blocks(dp->i_mount, + be32_to_cpu(name_rmt->valuelen)); + lp++; + } + } + } + xfs_trans_brelse(*trans, bp); /* unlock for trans. in freextent() */ + + /* + * Invalidate each of the "remote" value extents. + */ + error = 0; + for (lp = list, i = 0; i < count; i++, lp++) { + tmp = xfs_attr3_leaf_freextent(trans, dp, + lp->valueblk, lp->valuelen); + + if (error == 0) + error = tmp; /* save only the 1st errno */ + } + + kmem_free(list); + return error; +} + +/* + * Recurse (gasp!) through the attribute nodes until we find leaves. + * We're doing a depth-first traversal in order to invalidate everything. + */ +STATIC int +xfs_attr3_node_inactive( + struct xfs_trans **trans, + struct xfs_inode *dp, + struct xfs_buf *bp, + int level) +{ + xfs_da_blkinfo_t *info; + xfs_da_intnode_t *node; + xfs_dablk_t child_fsb; + xfs_daddr_t parent_blkno, child_blkno; + int error, i; + struct xfs_buf *child_bp; + struct xfs_da_node_entry *btree; + struct xfs_da3_icnode_hdr ichdr; + + /* + * Since this code is recursive (gasp!) we must protect ourselves. + */ + if (level > XFS_DA_NODE_MAXDEPTH) { + xfs_trans_brelse(*trans, bp); /* no locks for later trans */ + return XFS_ERROR(EIO); + } + + node = bp->b_addr; + xfs_da3_node_hdr_from_disk(&ichdr, node); + parent_blkno = bp->b_bn; + if (!ichdr.count) { + xfs_trans_brelse(*trans, bp); + return 0; + } + btree = xfs_da3_node_tree_p(node); + child_fsb = be32_to_cpu(btree[0].before); + xfs_trans_brelse(*trans, bp); /* no locks for later trans */ + + /* + * If this is the node level just above the leaves, simply loop + * over the leaves removing all of them. If this is higher up + * in the tree, recurse downward. + */ + for (i = 0; i < ichdr.count; i++) { + /* + * Read the subsidiary block to see what we have to work with. + * Don't do this in a transaction. This is a depth-first + * traversal of the tree so we may deal with many blocks + * before we come back to this one. + */ + error = xfs_da3_node_read(*trans, dp, child_fsb, -2, &child_bp, + XFS_ATTR_FORK); + if (error) + return(error); + if (child_bp) { + /* save for re-read later */ + child_blkno = XFS_BUF_ADDR(child_bp); + + /* + * Invalidate the subtree, however we have to. + */ + info = child_bp->b_addr; + switch (info->magic) { + case cpu_to_be16(XFS_DA_NODE_MAGIC): + case cpu_to_be16(XFS_DA3_NODE_MAGIC): + error = xfs_attr3_node_inactive(trans, dp, + child_bp, level + 1); + break; + case cpu_to_be16(XFS_ATTR_LEAF_MAGIC): + case cpu_to_be16(XFS_ATTR3_LEAF_MAGIC): + error = xfs_attr3_leaf_inactive(trans, dp, + child_bp); + break; + default: + error = XFS_ERROR(EIO); + xfs_trans_brelse(*trans, child_bp); + break; + } + if (error) + return error; + + /* + * Remove the subsidiary block from the cache + * and from the log. + */ + error = xfs_da_get_buf(*trans, dp, 0, child_blkno, + &child_bp, XFS_ATTR_FORK); + if (error) + return error; + xfs_trans_binval(*trans, child_bp); + } + + /* + * If we're not done, re-read the parent to get the next + * child block number. + */ + if (i + 1 < ichdr.count) { + error = xfs_da3_node_read(*trans, dp, 0, parent_blkno, + &bp, XFS_ATTR_FORK); + if (error) + return error; + child_fsb = be32_to_cpu(btree[i + 1].before); + xfs_trans_brelse(*trans, bp); + } + /* + * Atomically commit the whole invalidate stuff. + */ + error = xfs_trans_roll(trans, dp); + if (error) + return error; + } + + return 0; +} + +/* + * Indiscriminately delete the entire attribute fork + * + * Recurse (gasp!) through the attribute nodes until we find leaves. + * We're doing a depth-first traversal in order to invalidate everything. + */ +int +xfs_attr3_root_inactive( + struct xfs_trans **trans, + struct xfs_inode *dp) +{ + struct xfs_da_blkinfo *info; + struct xfs_buf *bp; + xfs_daddr_t blkno; + int error; + + /* + * Read block 0 to see what we have to work with. + * We only get here if we have extents, since we remove + * the extents in reverse order the extent containing + * block 0 must still be there. + */ + error = xfs_da3_node_read(*trans, dp, 0, -1, &bp, XFS_ATTR_FORK); + if (error) + return error; + blkno = bp->b_bn; + + /* + * Invalidate the tree, even if the "tree" is only a single leaf block. + * This is a depth-first traversal! + */ + info = bp->b_addr; + switch (info->magic) { + case cpu_to_be16(XFS_DA_NODE_MAGIC): + case cpu_to_be16(XFS_DA3_NODE_MAGIC): + error = xfs_attr3_node_inactive(trans, dp, bp, 1); + break; + case cpu_to_be16(XFS_ATTR_LEAF_MAGIC): + case cpu_to_be16(XFS_ATTR3_LEAF_MAGIC): + error = xfs_attr3_leaf_inactive(trans, dp, bp); + break; + default: + error = XFS_ERROR(EIO); + xfs_trans_brelse(*trans, bp); + break; + } + if (error) + return error; + + /* + * Invalidate the incore copy of the root block. + */ + error = xfs_da_get_buf(*trans, dp, 0, blkno, &bp, XFS_ATTR_FORK); + if (error) + return error; + xfs_trans_binval(*trans, bp); /* remove from cache */ + /* + * Commit the invalidate and start the next transaction. + */ + error = xfs_trans_roll(trans, dp); + + return error; +} + +int +xfs_attr_inactive(xfs_inode_t *dp) +{ + xfs_trans_t *trans; + xfs_mount_t *mp; + int error; + + mp = dp->i_mount; + ASSERT(! XFS_NOT_DQATTACHED(mp, dp)); + + xfs_ilock(dp, XFS_ILOCK_SHARED); + if (!xfs_inode_hasattr(dp) || + dp->i_d.di_aformat == XFS_DINODE_FMT_LOCAL) { + xfs_iunlock(dp, XFS_ILOCK_SHARED); + return 0; + } + xfs_iunlock(dp, XFS_ILOCK_SHARED); + + /* + * Start our first transaction of the day. + * + * All future transactions during this code must be "chained" off + * this one via the trans_dup() call. All transactions will contain + * the inode, and the inode will always be marked with trans_ihold(). + * Since the inode will be locked in all transactions, we must log + * the inode in every transaction to let it float upward through + * the log. + */ + trans = xfs_trans_alloc(mp, XFS_TRANS_ATTRINVAL); + error = xfs_trans_reserve(trans, &M_RES(mp)->tr_attrinval, 0, 0); + if (error) { + xfs_trans_cancel(trans, 0); + return(error); + } + xfs_ilock(dp, XFS_ILOCK_EXCL); + + /* + * No need to make quota reservations here. We expect to release some + * blocks, not allocate, in the common case. + */ + xfs_trans_ijoin(trans, dp, 0); + + /* + * Decide on what work routines to call based on the inode size. + */ + if (!xfs_inode_hasattr(dp) || + dp->i_d.di_aformat == XFS_DINODE_FMT_LOCAL) { + error = 0; + goto out; + } + error = xfs_attr3_root_inactive(&trans, dp); + if (error) + goto out; + + error = xfs_itruncate_extents(&trans, dp, XFS_ATTR_FORK, 0); + if (error) + goto out; + + error = xfs_trans_commit(trans, XFS_TRANS_RELEASE_LOG_RES); + xfs_iunlock(dp, XFS_ILOCK_EXCL); + + return(error); + +out: + xfs_trans_cancel(trans, XFS_TRANS_RELEASE_LOG_RES|XFS_TRANS_ABORT); + xfs_iunlock(dp, XFS_ILOCK_EXCL); + return(error); +} diff --git a/fs/xfs/xfs_attr_leaf.c b/fs/xfs/xfs_attr_leaf.c index b800fbc..86db20a 100644 --- a/fs/xfs/xfs_attr_leaf.c +++ b/fs/xfs/xfs_attr_leaf.c @@ -22,6 +22,7 @@ #include "xfs_bit.h" #include "xfs_log.h" #include "xfs_trans.h" +#include "xfs_trans_priv.h" #include "xfs_sb.h" #include "xfs_ag.h" #include "xfs_mount.h" @@ -78,16 +79,6 @@ STATIC int xfs_attr3_leaf_figure_balance(xfs_da_state_t *state, int *number_usedbytes_in_blk1); /* - * Routines used for shrinking the Btree. - */ -STATIC int xfs_attr3_node_inactive(xfs_trans_t **trans, xfs_inode_t *dp, - struct xfs_buf *bp, int level); -STATIC int xfs_attr3_leaf_inactive(xfs_trans_t **trans, xfs_inode_t *dp, - struct xfs_buf *bp); -STATIC int xfs_attr3_leaf_freextent(xfs_trans_t **trans, xfs_inode_t *dp, - xfs_dablk_t blkno, int blkcnt); - -/* * Utility routines. */ STATIC void xfs_attr3_leaf_moveents(struct xfs_attr_leafblock *src_leaf, @@ -635,7 +626,7 @@ xfs_attr_shortform_getvalue(xfs_da_args_t *args) xfs_attr_sf_entry_t *sfe; int i; - ASSERT(args->dp->i_d.di_aformat == XFS_IFINLINE); + ASSERT(args->dp->i_afp->if_flags == XFS_IFINLINE); sf = (xfs_attr_shortform_t *)args->dp->i_afp->if_u1.if_data; sfe = &sf->list[0]; for (i = 0; i < sf->hdr.count; @@ -751,182 +742,6 @@ out: return(error); } -STATIC int -xfs_attr_shortform_compare(const void *a, const void *b) -{ - xfs_attr_sf_sort_t *sa, *sb; - - sa = (xfs_attr_sf_sort_t *)a; - sb = (xfs_attr_sf_sort_t *)b; - if (sa->hash < sb->hash) { - return(-1); - } else if (sa->hash > sb->hash) { - return(1); - } else { - return(sa->entno - sb->entno); - } -} - - -#define XFS_ISRESET_CURSOR(cursor) \ - (!((cursor)->initted) && !((cursor)->hashval) && \ - !((cursor)->blkno) && !((cursor)->offset)) -/* - * Copy out entries of shortform attribute lists for attr_list(). - * Shortform attribute lists are not stored in hashval sorted order. - * If the output buffer is not large enough to hold them all, then we - * we have to calculate each entries' hashvalue and sort them before - * we can begin returning them to the user. - */ -/*ARGSUSED*/ -int -xfs_attr_shortform_list(xfs_attr_list_context_t *context) -{ - attrlist_cursor_kern_t *cursor; - xfs_attr_sf_sort_t *sbuf, *sbp; - xfs_attr_shortform_t *sf; - xfs_attr_sf_entry_t *sfe; - xfs_inode_t *dp; - int sbsize, nsbuf, count, i; - int error; - - ASSERT(context != NULL); - dp = context->dp; - ASSERT(dp != NULL); - ASSERT(dp->i_afp != NULL); - sf = (xfs_attr_shortform_t *)dp->i_afp->if_u1.if_data; - ASSERT(sf != NULL); - if (!sf->hdr.count) - return(0); - cursor = context->cursor; - ASSERT(cursor != NULL); - - trace_xfs_attr_list_sf(context); - - /* - * If the buffer is large enough and the cursor is at the start, - * do not bother with sorting since we will return everything in - * one buffer and another call using the cursor won't need to be - * made. - * Note the generous fudge factor of 16 overhead bytes per entry. - * If bufsize is zero then put_listent must be a search function - * and can just scan through what we have. - */ - if (context->bufsize == 0 || - (XFS_ISRESET_CURSOR(cursor) && - (dp->i_afp->if_bytes + sf->hdr.count * 16) < context->bufsize)) { - for (i = 0, sfe = &sf->list[0]; i < sf->hdr.count; i++) { - error = context->put_listent(context, - sfe->flags, - sfe->nameval, - (int)sfe->namelen, - (int)sfe->valuelen, - &sfe->nameval[sfe->namelen]); - - /* - * Either search callback finished early or - * didn't fit it all in the buffer after all. - */ - if (context->seen_enough) - break; - - if (error) - return error; - sfe = XFS_ATTR_SF_NEXTENTRY(sfe); - } - trace_xfs_attr_list_sf_all(context); - return(0); - } - - /* do no more for a search callback */ - if (context->bufsize == 0) - return 0; - - /* - * It didn't all fit, so we have to sort everything on hashval. - */ - sbsize = sf->hdr.count * sizeof(*sbuf); - sbp = sbuf = kmem_alloc(sbsize, KM_SLEEP | KM_NOFS); - - /* - * Scan the attribute list for the rest of the entries, storing - * the relevant info from only those that match into a buffer. - */ - nsbuf = 0; - for (i = 0, sfe = &sf->list[0]; i < sf->hdr.count; i++) { - if (unlikely( - ((char *)sfe < (char *)sf) || - ((char *)sfe >= ((char *)sf + dp->i_afp->if_bytes)))) { - XFS_CORRUPTION_ERROR("xfs_attr_shortform_list", - XFS_ERRLEVEL_LOW, - context->dp->i_mount, sfe); - kmem_free(sbuf); - return XFS_ERROR(EFSCORRUPTED); - } - - sbp->entno = i; - sbp->hash = xfs_da_hashname(sfe->nameval, sfe->namelen); - sbp->name = sfe->nameval; - sbp->namelen = sfe->namelen; - /* These are bytes, and both on-disk, don't endian-flip */ - sbp->valuelen = sfe->valuelen; - sbp->flags = sfe->flags; - sfe = XFS_ATTR_SF_NEXTENTRY(sfe); - sbp++; - nsbuf++; - } - - /* - * Sort the entries on hash then entno. - */ - xfs_sort(sbuf, nsbuf, sizeof(*sbuf), xfs_attr_shortform_compare); - - /* - * Re-find our place IN THE SORTED LIST. - */ - count = 0; - cursor->initted = 1; - cursor->blkno = 0; - for (sbp = sbuf, i = 0; i < nsbuf; i++, sbp++) { - if (sbp->hash == cursor->hashval) { - if (cursor->offset == count) { - break; - } - count++; - } else if (sbp->hash > cursor->hashval) { - break; - } - } - if (i == nsbuf) { - kmem_free(sbuf); - return(0); - } - - /* - * Loop putting entries into the user buffer. - */ - for ( ; i < nsbuf; i++, sbp++) { - if (cursor->hashval != sbp->hash) { - cursor->hashval = sbp->hash; - cursor->offset = 0; - } - error = context->put_listent(context, - sbp->flags, - sbp->name, - sbp->namelen, - sbp->valuelen, - &sbp->name[sbp->namelen]); - if (error) - return error; - if (context->seen_enough) - break; - cursor->offset++; - } - - kmem_free(sbuf); - return(0); -} - /* * Check a leaf attribute block to see if all the entries would fit into * a shortform attribute list. @@ -1121,7 +936,6 @@ out: return error; } - /*======================================================================== * Routines used for growing the Btree. *========================================================================*/ @@ -1482,7 +1296,6 @@ xfs_attr3_leaf_compact( ichdr_dst->freemap[0].size = ichdr_dst->firstused - ichdr_dst->freemap[0].base; - /* write the header back to initialise the underlying buffer */ xfs_attr3_leaf_hdr_to_disk(leaf_dst, ichdr_dst); @@ -2643,130 +2456,6 @@ xfs_attr_leaf_newentsize(int namelen, int valuelen, int blocksize, int *local) return size; } -/* - * Copy out attribute list entries for attr_list(), for leaf attribute lists. - */ -int -xfs_attr3_leaf_list_int( - struct xfs_buf *bp, - struct xfs_attr_list_context *context) -{ - struct attrlist_cursor_kern *cursor; - struct xfs_attr_leafblock *leaf; - struct xfs_attr3_icleaf_hdr ichdr; - struct xfs_attr_leaf_entry *entries; - struct xfs_attr_leaf_entry *entry; - int retval; - int i; - - trace_xfs_attr_list_leaf(context); - - leaf = bp->b_addr; - xfs_attr3_leaf_hdr_from_disk(&ichdr, leaf); - entries = xfs_attr3_leaf_entryp(leaf); - - cursor = context->cursor; - cursor->initted = 1; - - /* - * Re-find our place in the leaf block if this is a new syscall. - */ - if (context->resynch) { - entry = &entries[0]; - for (i = 0; i < ichdr.count; entry++, i++) { - if (be32_to_cpu(entry->hashval) == cursor->hashval) { - if (cursor->offset == context->dupcnt) { - context->dupcnt = 0; - break; - } - context->dupcnt++; - } else if (be32_to_cpu(entry->hashval) > - cursor->hashval) { - context->dupcnt = 0; - break; - } - } - if (i == ichdr.count) { - trace_xfs_attr_list_notfound(context); - return 0; - } - } else { - entry = &entries[0]; - i = 0; - } - context->resynch = 0; - - /* - * We have found our place, start copying out the new attributes. - */ - retval = 0; - for (; i < ichdr.count; entry++, i++) { - if (be32_to_cpu(entry->hashval) != cursor->hashval) { - cursor->hashval = be32_to_cpu(entry->hashval); - cursor->offset = 0; - } - - if (entry->flags & XFS_ATTR_INCOMPLETE) - continue; /* skip incomplete entries */ - - if (entry->flags & XFS_ATTR_LOCAL) { - xfs_attr_leaf_name_local_t *name_loc = - xfs_attr3_leaf_name_local(leaf, i); - - retval = context->put_listent(context, - entry->flags, - name_loc->nameval, - (int)name_loc->namelen, - be16_to_cpu(name_loc->valuelen), - &name_loc->nameval[name_loc->namelen]); - if (retval) - return retval; - } else { - xfs_attr_leaf_name_remote_t *name_rmt = - xfs_attr3_leaf_name_remote(leaf, i); - - int valuelen = be32_to_cpu(name_rmt->valuelen); - - if (context->put_value) { - xfs_da_args_t args; - - memset((char *)&args, 0, sizeof(args)); - args.dp = context->dp; - args.whichfork = XFS_ATTR_FORK; - args.valuelen = valuelen; - args.value = kmem_alloc(valuelen, KM_SLEEP | KM_NOFS); - args.rmtblkno = be32_to_cpu(name_rmt->valueblk); - args.rmtblkcnt = xfs_attr3_rmt_blocks( - args.dp->i_mount, valuelen); - retval = xfs_attr_rmtval_get(&args); - if (retval) - return retval; - retval = context->put_listent(context, - entry->flags, - name_rmt->name, - (int)name_rmt->namelen, - valuelen, - args.value); - kmem_free(args.value); - } else { - retval = context->put_listent(context, - entry->flags, - name_rmt->name, - (int)name_rmt->namelen, - valuelen, - NULL); - } - if (retval) - return retval; - } - if (context->seen_enough) - break; - cursor->offset++; - } - trace_xfs_attr_list_leaf_end(context); - return retval; -} - /*======================================================================== * Manage the INCOMPLETE flag in a leaf entry @@ -3011,345 +2700,3 @@ xfs_attr3_leaf_flipflags( return error; } - -/*======================================================================== - * Indiscriminately delete the entire attribute fork - *========================================================================*/ - -/* - * Recurse (gasp!) through the attribute nodes until we find leaves. - * We're doing a depth-first traversal in order to invalidate everything. - */ -int -xfs_attr3_root_inactive( - struct xfs_trans **trans, - struct xfs_inode *dp) -{ - struct xfs_da_blkinfo *info; - struct xfs_buf *bp; - xfs_daddr_t blkno; - int error; - - /* - * Read block 0 to see what we have to work with. - * We only get here if we have extents, since we remove - * the extents in reverse order the extent containing - * block 0 must still be there. - */ - error = xfs_da3_node_read(*trans, dp, 0, -1, &bp, XFS_ATTR_FORK); - if (error) - return error; - blkno = bp->b_bn; - - /* - * Invalidate the tree, even if the "tree" is only a single leaf block. - * This is a depth-first traversal! - */ - info = bp->b_addr; - switch (info->magic) { - case cpu_to_be16(XFS_DA_NODE_MAGIC): - case cpu_to_be16(XFS_DA3_NODE_MAGIC): - error = xfs_attr3_node_inactive(trans, dp, bp, 1); - break; - case cpu_to_be16(XFS_ATTR_LEAF_MAGIC): - case cpu_to_be16(XFS_ATTR3_LEAF_MAGIC): - error = xfs_attr3_leaf_inactive(trans, dp, bp); - break; - default: - error = XFS_ERROR(EIO); - xfs_trans_brelse(*trans, bp); - break; - } - if (error) - return error; - - /* - * Invalidate the incore copy of the root block. - */ - error = xfs_da_get_buf(*trans, dp, 0, blkno, &bp, XFS_ATTR_FORK); - if (error) - return error; - xfs_trans_binval(*trans, bp); /* remove from cache */ - /* - * Commit the invalidate and start the next transaction. - */ - error = xfs_trans_roll(trans, dp); - - return error; -} - -/* - * Recurse (gasp!) through the attribute nodes until we find leaves. - * We're doing a depth-first traversal in order to invalidate everything. - */ -STATIC int -xfs_attr3_node_inactive( - struct xfs_trans **trans, - struct xfs_inode *dp, - struct xfs_buf *bp, - int level) -{ - xfs_da_blkinfo_t *info; - xfs_da_intnode_t *node; - xfs_dablk_t child_fsb; - xfs_daddr_t parent_blkno, child_blkno; - int error, i; - struct xfs_buf *child_bp; - struct xfs_da_node_entry *btree; - struct xfs_da3_icnode_hdr ichdr; - - /* - * Since this code is recursive (gasp!) we must protect ourselves. - */ - if (level > XFS_DA_NODE_MAXDEPTH) { - xfs_trans_brelse(*trans, bp); /* no locks for later trans */ - return XFS_ERROR(EIO); - } - - node = bp->b_addr; - xfs_da3_node_hdr_from_disk(&ichdr, node); - parent_blkno = bp->b_bn; - if (!ichdr.count) { - xfs_trans_brelse(*trans, bp); - return 0; - } - btree = xfs_da3_node_tree_p(node); - child_fsb = be32_to_cpu(btree[0].before); - xfs_trans_brelse(*trans, bp); /* no locks for later trans */ - - /* - * If this is the node level just above the leaves, simply loop - * over the leaves removing all of them. If this is higher up - * in the tree, recurse downward. - */ - for (i = 0; i < ichdr.count; i++) { - /* - * Read the subsidiary block to see what we have to work with. - * Don't do this in a transaction. This is a depth-first - * traversal of the tree so we may deal with many blocks - * before we come back to this one. - */ - error = xfs_da3_node_read(*trans, dp, child_fsb, -2, &child_bp, - XFS_ATTR_FORK); - if (error) - return(error); - if (child_bp) { - /* save for re-read later */ - child_blkno = XFS_BUF_ADDR(child_bp); - - /* - * Invalidate the subtree, however we have to. - */ - info = child_bp->b_addr; - switch (info->magic) { - case cpu_to_be16(XFS_DA_NODE_MAGIC): - case cpu_to_be16(XFS_DA3_NODE_MAGIC): - error = xfs_attr3_node_inactive(trans, dp, - child_bp, level + 1); - break; - case cpu_to_be16(XFS_ATTR_LEAF_MAGIC): - case cpu_to_be16(XFS_ATTR3_LEAF_MAGIC): - error = xfs_attr3_leaf_inactive(trans, dp, - child_bp); - break; - default: - error = XFS_ERROR(EIO); - xfs_trans_brelse(*trans, child_bp); - break; - } - if (error) - return error; - - /* - * Remove the subsidiary block from the cache - * and from the log. - */ - error = xfs_da_get_buf(*trans, dp, 0, child_blkno, - &child_bp, XFS_ATTR_FORK); - if (error) - return error; - xfs_trans_binval(*trans, child_bp); - } - - /* - * If we're not done, re-read the parent to get the next - * child block number. - */ - if (i + 1 < ichdr.count) { - error = xfs_da3_node_read(*trans, dp, 0, parent_blkno, - &bp, XFS_ATTR_FORK); - if (error) - return error; - child_fsb = be32_to_cpu(btree[i + 1].before); - xfs_trans_brelse(*trans, bp); - } - /* - * Atomically commit the whole invalidate stuff. - */ - error = xfs_trans_roll(trans, dp); - if (error) - return error; - } - - return 0; -} - -/* - * Invalidate all of the "remote" value regions pointed to by a particular - * leaf block. - * Note that we must release the lock on the buffer so that we are not - * caught holding something that the logging code wants to flush to disk. - */ -STATIC int -xfs_attr3_leaf_inactive( - struct xfs_trans **trans, - struct xfs_inode *dp, - struct xfs_buf *bp) -{ - struct xfs_attr_leafblock *leaf; - struct xfs_attr3_icleaf_hdr ichdr; - struct xfs_attr_leaf_entry *entry; - struct xfs_attr_leaf_name_remote *name_rmt; - struct xfs_attr_inactive_list *list; - struct xfs_attr_inactive_list *lp; - int error; - int count; - int size; - int tmp; - int i; - - leaf = bp->b_addr; - xfs_attr3_leaf_hdr_from_disk(&ichdr, leaf); - - /* - * Count the number of "remote" value extents. - */ - count = 0; - entry = xfs_attr3_leaf_entryp(leaf); - for (i = 0; i < ichdr.count; entry++, i++) { - if (be16_to_cpu(entry->nameidx) && - ((entry->flags & XFS_ATTR_LOCAL) == 0)) { - name_rmt = xfs_attr3_leaf_name_remote(leaf, i); - if (name_rmt->valueblk) - count++; - } - } - - /* - * If there are no "remote" values, we're done. - */ - if (count == 0) { - xfs_trans_brelse(*trans, bp); - return 0; - } - - /* - * Allocate storage for a list of all the "remote" value extents. - */ - size = count * sizeof(xfs_attr_inactive_list_t); - list = kmem_alloc(size, KM_SLEEP); - - /* - * Identify each of the "remote" value extents. - */ - lp = list; - entry = xfs_attr3_leaf_entryp(leaf); - for (i = 0; i < ichdr.count; entry++, i++) { - if (be16_to_cpu(entry->nameidx) && - ((entry->flags & XFS_ATTR_LOCAL) == 0)) { - name_rmt = xfs_attr3_leaf_name_remote(leaf, i); - if (name_rmt->valueblk) { - lp->valueblk = be32_to_cpu(name_rmt->valueblk); - lp->valuelen = xfs_attr3_rmt_blocks(dp->i_mount, - be32_to_cpu(name_rmt->valuelen)); - lp++; - } - } - } - xfs_trans_brelse(*trans, bp); /* unlock for trans. in freextent() */ - - /* - * Invalidate each of the "remote" value extents. - */ - error = 0; - for (lp = list, i = 0; i < count; i++, lp++) { - tmp = xfs_attr3_leaf_freextent(trans, dp, - lp->valueblk, lp->valuelen); - - if (error == 0) - error = tmp; /* save only the 1st errno */ - } - - kmem_free(list); - return error; -} - -/* - * Look at all the extents for this logical region, - * invalidate any buffers that are incore/in transactions. - */ -STATIC int -xfs_attr3_leaf_freextent( - struct xfs_trans **trans, - struct xfs_inode *dp, - xfs_dablk_t blkno, - int blkcnt) -{ - struct xfs_bmbt_irec map; - struct xfs_buf *bp; - xfs_dablk_t tblkno; - xfs_daddr_t dblkno; - int tblkcnt; - int dblkcnt; - int nmap; - int error; - - /* - * Roll through the "value", invalidating the attribute value's - * blocks. - */ - tblkno = blkno; - tblkcnt = blkcnt; - while (tblkcnt > 0) { - /* - * Try to remember where we decided to put the value. - */ - nmap = 1; - error = xfs_bmapi_read(dp, (xfs_fileoff_t)tblkno, tblkcnt, - &map, &nmap, XFS_BMAPI_ATTRFORK); - if (error) { - return(error); - } - ASSERT(nmap == 1); - ASSERT(map.br_startblock != DELAYSTARTBLOCK); - - /* - * If it's a hole, these are already unmapped - * so there's nothing to invalidate. - */ - if (map.br_startblock != HOLESTARTBLOCK) { - - dblkno = XFS_FSB_TO_DADDR(dp->i_mount, - map.br_startblock); - dblkcnt = XFS_FSB_TO_BB(dp->i_mount, - map.br_blockcount); - bp = xfs_trans_get_buf(*trans, - dp->i_mount->m_ddev_targp, - dblkno, dblkcnt, 0); - if (!bp) - return ENOMEM; - xfs_trans_binval(*trans, bp); - /* - * Roll to next transaction. - */ - error = xfs_trans_roll(trans, dp); - if (error) - return (error); - } - - tblkno += map.br_blockcount; - tblkcnt -= map.br_blockcount; - } - - return(0); -} diff --git a/fs/xfs/xfs_attr_leaf.h b/fs/xfs/xfs_attr_leaf.h index 444a770..c102213 100644 --- a/fs/xfs/xfs_attr_leaf.h +++ b/fs/xfs/xfs_attr_leaf.h @@ -333,6 +333,8 @@ int xfs_attr3_leaf_read(struct xfs_trans *tp, struct xfs_inode *dp, struct xfs_buf **bpp); void xfs_attr3_leaf_hdr_from_disk(struct xfs_attr3_icleaf_hdr *to, struct xfs_attr_leafblock *from); +void xfs_attr3_leaf_hdr_to_disk(struct xfs_attr_leafblock *to, + struct xfs_attr3_icleaf_hdr *from); extern const struct xfs_buf_ops xfs_attr3_leaf_buf_ops; diff --git a/fs/xfs/xfs_attr_list.c b/fs/xfs/xfs_attr_list.c new file mode 100644 index 0000000..cbc80d4 --- /dev/null +++ b/fs/xfs/xfs_attr_list.c @@ -0,0 +1,655 @@ +/* + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * Copyright (c) 2013 Red Hat, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#include "xfs.h" +#include "xfs_fs.h" +#include "xfs_types.h" +#include "xfs_bit.h" +#include "xfs_log.h" +#include "xfs_trans.h" +#include "xfs_sb.h" +#include "xfs_ag.h" +#include "xfs_mount.h" +#include "xfs_da_btree.h" +#include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" +#include "xfs_ialloc_btree.h" +#include "xfs_alloc.h" +#include "xfs_btree.h" +#include "xfs_attr_sf.h" +#include "xfs_attr_remote.h" +#include "xfs_dinode.h" +#include "xfs_inode.h" +#include "xfs_inode_item.h" +#include "xfs_bmap.h" +#include "xfs_attr.h" +#include "xfs_attr_leaf.h" +#include "xfs_error.h" +#include "xfs_trace.h" +#include "xfs_buf_item.h" +#include "xfs_cksum.h" + +STATIC int +xfs_attr_shortform_compare(const void *a, const void *b) +{ + xfs_attr_sf_sort_t *sa, *sb; + + sa = (xfs_attr_sf_sort_t *)a; + sb = (xfs_attr_sf_sort_t *)b; + if (sa->hash < sb->hash) { + return(-1); + } else if (sa->hash > sb->hash) { + return(1); + } else { + return(sa->entno - sb->entno); + } +} + +#define XFS_ISRESET_CURSOR(cursor) \ + (!((cursor)->initted) && !((cursor)->hashval) && \ + !((cursor)->blkno) && !((cursor)->offset)) +/* + * Copy out entries of shortform attribute lists for attr_list(). + * Shortform attribute lists are not stored in hashval sorted order. + * If the output buffer is not large enough to hold them all, then we + * we have to calculate each entries' hashvalue and sort them before + * we can begin returning them to the user. + */ +int +xfs_attr_shortform_list(xfs_attr_list_context_t *context) +{ + attrlist_cursor_kern_t *cursor; + xfs_attr_sf_sort_t *sbuf, *sbp; + xfs_attr_shortform_t *sf; + xfs_attr_sf_entry_t *sfe; + xfs_inode_t *dp; + int sbsize, nsbuf, count, i; + int error; + + ASSERT(context != NULL); + dp = context->dp; + ASSERT(dp != NULL); + ASSERT(dp->i_afp != NULL); + sf = (xfs_attr_shortform_t *)dp->i_afp->if_u1.if_data; + ASSERT(sf != NULL); + if (!sf->hdr.count) + return(0); + cursor = context->cursor; + ASSERT(cursor != NULL); + + trace_xfs_attr_list_sf(context); + + /* + * If the buffer is large enough and the cursor is at the start, + * do not bother with sorting since we will return everything in + * one buffer and another call using the cursor won't need to be + * made. + * Note the generous fudge factor of 16 overhead bytes per entry. + * If bufsize is zero then put_listent must be a search function + * and can just scan through what we have. + */ + if (context->bufsize == 0 || + (XFS_ISRESET_CURSOR(cursor) && + (dp->i_afp->if_bytes + sf->hdr.count * 16) < context->bufsize)) { + for (i = 0, sfe = &sf->list[0]; i < sf->hdr.count; i++) { + error = context->put_listent(context, + sfe->flags, + sfe->nameval, + (int)sfe->namelen, + (int)sfe->valuelen, + &sfe->nameval[sfe->namelen]); + + /* + * Either search callback finished early or + * didn't fit it all in the buffer after all. + */ + if (context->seen_enough) + break; + + if (error) + return error; + sfe = XFS_ATTR_SF_NEXTENTRY(sfe); + } + trace_xfs_attr_list_sf_all(context); + return(0); + } + + /* do no more for a search callback */ + if (context->bufsize == 0) + return 0; + + /* + * It didn't all fit, so we have to sort everything on hashval. + */ + sbsize = sf->hdr.count * sizeof(*sbuf); + sbp = sbuf = kmem_alloc(sbsize, KM_SLEEP | KM_NOFS); + + /* + * Scan the attribute list for the rest of the entries, storing + * the relevant info from only those that match into a buffer. + */ + nsbuf = 0; + for (i = 0, sfe = &sf->list[0]; i < sf->hdr.count; i++) { + if (unlikely( + ((char *)sfe < (char *)sf) || + ((char *)sfe >= ((char *)sf + dp->i_afp->if_bytes)))) { + XFS_CORRUPTION_ERROR("xfs_attr_shortform_list", + XFS_ERRLEVEL_LOW, + context->dp->i_mount, sfe); + kmem_free(sbuf); + return XFS_ERROR(EFSCORRUPTED); + } + + sbp->entno = i; + sbp->hash = xfs_da_hashname(sfe->nameval, sfe->namelen); + sbp->name = sfe->nameval; + sbp->namelen = sfe->namelen; + /* These are bytes, and both on-disk, don't endian-flip */ + sbp->valuelen = sfe->valuelen; + sbp->flags = sfe->flags; + sfe = XFS_ATTR_SF_NEXTENTRY(sfe); + sbp++; + nsbuf++; + } + + /* + * Sort the entries on hash then entno. + */ + xfs_sort(sbuf, nsbuf, sizeof(*sbuf), xfs_attr_shortform_compare); + + /* + * Re-find our place IN THE SORTED LIST. + */ + count = 0; + cursor->initted = 1; + cursor->blkno = 0; + for (sbp = sbuf, i = 0; i < nsbuf; i++, sbp++) { + if (sbp->hash == cursor->hashval) { + if (cursor->offset == count) { + break; + } + count++; + } else if (sbp->hash > cursor->hashval) { + break; + } + } + if (i == nsbuf) { + kmem_free(sbuf); + return(0); + } + + /* + * Loop putting entries into the user buffer. + */ + for ( ; i < nsbuf; i++, sbp++) { + if (cursor->hashval != sbp->hash) { + cursor->hashval = sbp->hash; + cursor->offset = 0; + } + error = context->put_listent(context, + sbp->flags, + sbp->name, + sbp->namelen, + sbp->valuelen, + &sbp->name[sbp->namelen]); + if (error) + return error; + if (context->seen_enough) + break; + cursor->offset++; + } + + kmem_free(sbuf); + return(0); +} + +STATIC int +xfs_attr_node_list(xfs_attr_list_context_t *context) +{ + attrlist_cursor_kern_t *cursor; + xfs_attr_leafblock_t *leaf; + xfs_da_intnode_t *node; + struct xfs_attr3_icleaf_hdr leafhdr; + struct xfs_da3_icnode_hdr nodehdr; + struct xfs_da_node_entry *btree; + int error, i; + struct xfs_buf *bp; + + trace_xfs_attr_node_list(context); + + cursor = context->cursor; + cursor->initted = 1; + + /* + * Do all sorts of validation on the passed-in cursor structure. + * If anything is amiss, ignore the cursor and look up the hashval + * starting from the btree root. + */ + bp = NULL; + if (cursor->blkno > 0) { + error = xfs_da3_node_read(NULL, context->dp, cursor->blkno, -1, + &bp, XFS_ATTR_FORK); + if ((error != 0) && (error != EFSCORRUPTED)) + return(error); + if (bp) { + struct xfs_attr_leaf_entry *entries; + + node = bp->b_addr; + switch (be16_to_cpu(node->hdr.info.magic)) { + case XFS_DA_NODE_MAGIC: + case XFS_DA3_NODE_MAGIC: + trace_xfs_attr_list_wrong_blk(context); + xfs_trans_brelse(NULL, bp); + bp = NULL; + break; + case XFS_ATTR_LEAF_MAGIC: + case XFS_ATTR3_LEAF_MAGIC: + leaf = bp->b_addr; + xfs_attr3_leaf_hdr_from_disk(&leafhdr, leaf); + entries = xfs_attr3_leaf_entryp(leaf); + if (cursor->hashval > be32_to_cpu( + entries[leafhdr.count - 1].hashval)) { + trace_xfs_attr_list_wrong_blk(context); + xfs_trans_brelse(NULL, bp); + bp = NULL; + } else if (cursor->hashval <= be32_to_cpu( + entries[0].hashval)) { + trace_xfs_attr_list_wrong_blk(context); + xfs_trans_brelse(NULL, bp); + bp = NULL; + } + break; + default: + trace_xfs_attr_list_wrong_blk(context); + xfs_trans_brelse(NULL, bp); + bp = NULL; + } + } + } + + /* + * We did not find what we expected given the cursor's contents, + * so we start from the top and work down based on the hash value. + * Note that start of node block is same as start of leaf block. + */ + if (bp == NULL) { + cursor->blkno = 0; + for (;;) { + __uint16_t magic; + + error = xfs_da3_node_read(NULL, context->dp, + cursor->blkno, -1, &bp, + XFS_ATTR_FORK); + if (error) + return(error); + node = bp->b_addr; + magic = be16_to_cpu(node->hdr.info.magic); + if (magic == XFS_ATTR_LEAF_MAGIC || + magic == XFS_ATTR3_LEAF_MAGIC) + break; + if (magic != XFS_DA_NODE_MAGIC && + magic != XFS_DA3_NODE_MAGIC) { + XFS_CORRUPTION_ERROR("xfs_attr_node_list(3)", + XFS_ERRLEVEL_LOW, + context->dp->i_mount, + node); + xfs_trans_brelse(NULL, bp); + return XFS_ERROR(EFSCORRUPTED); + } + + xfs_da3_node_hdr_from_disk(&nodehdr, node); + btree = xfs_da3_node_tree_p(node); + for (i = 0; i < nodehdr.count; btree++, i++) { + if (cursor->hashval + <= be32_to_cpu(btree->hashval)) { + cursor->blkno = be32_to_cpu(btree->before); + trace_xfs_attr_list_node_descend(context, + btree); + break; + } + } + if (i == nodehdr.count) { + xfs_trans_brelse(NULL, bp); + return 0; + } + xfs_trans_brelse(NULL, bp); + } + } + ASSERT(bp != NULL); + + /* + * Roll upward through the blocks, processing each leaf block in + * order. As long as there is space in the result buffer, keep + * adding the information. + */ + for (;;) { + leaf = bp->b_addr; + error = xfs_attr3_leaf_list_int(bp, context); + if (error) { + xfs_trans_brelse(NULL, bp); + return error; + } + xfs_attr3_leaf_hdr_from_disk(&leafhdr, leaf); + if (context->seen_enough || leafhdr.forw == 0) + break; + cursor->blkno = leafhdr.forw; + xfs_trans_brelse(NULL, bp); + error = xfs_attr3_leaf_read(NULL, context->dp, cursor->blkno, -1, + &bp); + if (error) + return error; + } + xfs_trans_brelse(NULL, bp); + return 0; +} + +/* + * Copy out attribute list entries for attr_list(), for leaf attribute lists. + */ +int +xfs_attr3_leaf_list_int( + struct xfs_buf *bp, + struct xfs_attr_list_context *context) +{ + struct attrlist_cursor_kern *cursor; + struct xfs_attr_leafblock *leaf; + struct xfs_attr3_icleaf_hdr ichdr; + struct xfs_attr_leaf_entry *entries; + struct xfs_attr_leaf_entry *entry; + int retval; + int i; + + trace_xfs_attr_list_leaf(context); + + leaf = bp->b_addr; + xfs_attr3_leaf_hdr_from_disk(&ichdr, leaf); + entries = xfs_attr3_leaf_entryp(leaf); + + cursor = context->cursor; + cursor->initted = 1; + + /* + * Re-find our place in the leaf block if this is a new syscall. + */ + if (context->resynch) { + entry = &entries[0]; + for (i = 0; i < ichdr.count; entry++, i++) { + if (be32_to_cpu(entry->hashval) == cursor->hashval) { + if (cursor->offset == context->dupcnt) { + context->dupcnt = 0; + break; + } + context->dupcnt++; + } else if (be32_to_cpu(entry->hashval) > + cursor->hashval) { + context->dupcnt = 0; + break; + } + } + if (i == ichdr.count) { + trace_xfs_attr_list_notfound(context); + return 0; + } + } else { + entry = &entries[0]; + i = 0; + } + context->resynch = 0; + + /* + * We have found our place, start copying out the new attributes. + */ + retval = 0; + for (; i < ichdr.count; entry++, i++) { + if (be32_to_cpu(entry->hashval) != cursor->hashval) { + cursor->hashval = be32_to_cpu(entry->hashval); + cursor->offset = 0; + } + + if (entry->flags & XFS_ATTR_INCOMPLETE) + continue; /* skip incomplete entries */ + + if (entry->flags & XFS_ATTR_LOCAL) { + xfs_attr_leaf_name_local_t *name_loc = + xfs_attr3_leaf_name_local(leaf, i); + + retval = context->put_listent(context, + entry->flags, + name_loc->nameval, + (int)name_loc->namelen, + be16_to_cpu(name_loc->valuelen), + &name_loc->nameval[name_loc->namelen]); + if (retval) + return retval; + } else { + xfs_attr_leaf_name_remote_t *name_rmt = + xfs_attr3_leaf_name_remote(leaf, i); + + int valuelen = be32_to_cpu(name_rmt->valuelen); + + if (context->put_value) { + xfs_da_args_t args; + + memset((char *)&args, 0, sizeof(args)); + args.dp = context->dp; + args.whichfork = XFS_ATTR_FORK; + args.valuelen = valuelen; + args.value = kmem_alloc(valuelen, KM_SLEEP | KM_NOFS); + args.rmtblkno = be32_to_cpu(name_rmt->valueblk); + args.rmtblkcnt = xfs_attr3_rmt_blocks( + args.dp->i_mount, valuelen); + retval = xfs_attr_rmtval_get(&args); + if (retval) + return retval; + retval = context->put_listent(context, + entry->flags, + name_rmt->name, + (int)name_rmt->namelen, + valuelen, + args.value); + kmem_free(args.value); + } else { + retval = context->put_listent(context, + entry->flags, + name_rmt->name, + (int)name_rmt->namelen, + valuelen, + NULL); + } + if (retval) + return retval; + } + if (context->seen_enough) + break; + cursor->offset++; + } + trace_xfs_attr_list_leaf_end(context); + return retval; +} + +/* + * Copy out attribute entries for attr_list(), for leaf attribute lists. + */ +STATIC int +xfs_attr_leaf_list(xfs_attr_list_context_t *context) +{ + int error; + struct xfs_buf *bp; + + trace_xfs_attr_leaf_list(context); + + context->cursor->blkno = 0; + error = xfs_attr3_leaf_read(NULL, context->dp, 0, -1, &bp); + if (error) + return XFS_ERROR(error); + + error = xfs_attr3_leaf_list_int(bp, context); + xfs_trans_brelse(NULL, bp); + return XFS_ERROR(error); +} + +int +xfs_attr_list_int( + xfs_attr_list_context_t *context) +{ + int error; + xfs_inode_t *dp = context->dp; + + XFS_STATS_INC(xs_attr_list); + + if (XFS_FORCED_SHUTDOWN(dp->i_mount)) + return EIO; + + xfs_ilock(dp, XFS_ILOCK_SHARED); + + /* + * Decide on what work routines to call based on the inode size. + */ + if (!xfs_inode_hasattr(dp)) { + error = 0; + } else if (dp->i_d.di_aformat == XFS_DINODE_FMT_LOCAL) { + error = xfs_attr_shortform_list(context); + } else if (xfs_bmap_one_block(dp, XFS_ATTR_FORK)) { + error = xfs_attr_leaf_list(context); + } else { + error = xfs_attr_node_list(context); + } + + xfs_iunlock(dp, XFS_ILOCK_SHARED); + + return error; +} + +#define ATTR_ENTBASESIZE /* minimum bytes used by an attr */ \ + (((struct attrlist_ent *) 0)->a_name - (char *) 0) +#define ATTR_ENTSIZE(namelen) /* actual bytes used by an attr */ \ + ((ATTR_ENTBASESIZE + (namelen) + 1 + sizeof(u_int32_t)-1) \ + & ~(sizeof(u_int32_t)-1)) + +/* + * Format an attribute and copy it out to the user's buffer. + * Take care to check values and protect against them changing later, + * we may be reading them directly out of a user buffer. + */ +STATIC int +xfs_attr_put_listent( + xfs_attr_list_context_t *context, + int flags, + unsigned char *name, + int namelen, + int valuelen, + unsigned char *value) +{ + struct attrlist *alist = (struct attrlist *)context->alist; + attrlist_ent_t *aep; + int arraytop; + + ASSERT(!(context->flags & ATTR_KERNOVAL)); + ASSERT(context->count >= 0); + ASSERT(context->count < (ATTR_MAX_VALUELEN/8)); + ASSERT(context->firstu >= sizeof(*alist)); + ASSERT(context->firstu <= context->bufsize); + + /* + * Only list entries in the right namespace. + */ + if (((context->flags & ATTR_SECURE) == 0) != + ((flags & XFS_ATTR_SECURE) == 0)) + return 0; + if (((context->flags & ATTR_ROOT) == 0) != + ((flags & XFS_ATTR_ROOT) == 0)) + return 0; + + arraytop = sizeof(*alist) + + context->count * sizeof(alist->al_offset[0]); + context->firstu -= ATTR_ENTSIZE(namelen); + if (context->firstu < arraytop) { + trace_xfs_attr_list_full(context); + alist->al_more = 1; + context->seen_enough = 1; + return 1; + } + + aep = (attrlist_ent_t *)&context->alist[context->firstu]; + aep->a_valuelen = valuelen; + memcpy(aep->a_name, name, namelen); + aep->a_name[namelen] = 0; + alist->al_offset[context->count++] = context->firstu; + alist->al_count = context->count; + trace_xfs_attr_list_add(context); + return 0; +} + +/* + * Generate a list of extended attribute names and optionally + * also value lengths. Positive return value follows the XFS + * convention of being an error, zero or negative return code + * is the length of the buffer returned (negated), indicating + * success. + */ +int +xfs_attr_list( + xfs_inode_t *dp, + char *buffer, + int bufsize, + int flags, + attrlist_cursor_kern_t *cursor) +{ + xfs_attr_list_context_t context; + struct attrlist *alist; + int error; + + /* + * Validate the cursor. + */ + if (cursor->pad1 || cursor->pad2) + return(XFS_ERROR(EINVAL)); + if ((cursor->initted == 0) && + (cursor->hashval || cursor->blkno || cursor->offset)) + return XFS_ERROR(EINVAL); + + /* + * Check for a properly aligned buffer. + */ + if (((long)buffer) & (sizeof(int)-1)) + return XFS_ERROR(EFAULT); + if (flags & ATTR_KERNOVAL) + bufsize = 0; + + /* + * Initialize the output buffer. + */ + memset(&context, 0, sizeof(context)); + context.dp = dp; + context.cursor = cursor; + context.resynch = 1; + context.flags = flags; + context.alist = buffer; + context.bufsize = (bufsize & ~(sizeof(int)-1)); /* align */ + context.firstu = context.bufsize; + context.put_listent = xfs_attr_put_listent; + + alist = (struct attrlist *)context.alist; + alist->al_count = 0; + alist->al_more = 0; + alist->al_offset[0] = context.bufsize; + + error = xfs_attr_list_int(&context); + ASSERT(error >= 0); + return error; +} diff --git a/fs/xfs/xfs_attr_remote.c b/fs/xfs/xfs_attr_remote.c index ef6b0c1..712a502 100644 --- a/fs/xfs/xfs_attr_remote.c +++ b/fs/xfs/xfs_attr_remote.c @@ -22,6 +22,7 @@ #include "xfs_bit.h" #include "xfs_log.h" #include "xfs_trans.h" +#include "xfs_trans_priv.h" #include "xfs_sb.h" #include "xfs_ag.h" #include "xfs_mount.h" @@ -33,6 +34,7 @@ #include "xfs_alloc.h" #include "xfs_inode_item.h" #include "xfs_bmap.h" +#include "xfs_bmap_util.h" #include "xfs_attr.h" #include "xfs_attr_leaf.h" #include "xfs_attr_remote.h" @@ -237,7 +239,7 @@ xfs_attr_rmtval_copyout( xfs_ino_t ino, int *offset, int *valuelen, - char **dst) + __uint8_t **dst) { char *src = bp->b_addr; xfs_daddr_t bno = bp->b_bn; @@ -249,7 +251,7 @@ xfs_attr_rmtval_copyout( int hdr_size = 0; int byte_cnt = XFS_ATTR3_RMT_BUF_SPACE(mp, XFS_LBSIZE(mp)); - byte_cnt = min_t(int, *valuelen, byte_cnt); + byte_cnt = min(*valuelen, byte_cnt); if (xfs_sb_version_hascrc(&mp->m_sb)) { if (!xfs_attr3_rmt_hdr_ok(mp, src, ino, *offset, @@ -284,7 +286,7 @@ xfs_attr_rmtval_copyin( xfs_ino_t ino, int *offset, int *valuelen, - char **src) + __uint8_t **src) { char *dst = bp->b_addr; xfs_daddr_t bno = bp->b_bn; @@ -337,7 +339,7 @@ xfs_attr_rmtval_get( struct xfs_mount *mp = args->dp->i_mount; struct xfs_buf *bp; xfs_dablk_t lblkno = args->rmtblkno; - char *dst = args->value; + __uint8_t *dst = args->value; int valuelen = args->valuelen; int nmap; int error; @@ -401,7 +403,7 @@ xfs_attr_rmtval_set( struct xfs_bmbt_irec map; xfs_dablk_t lblkno; xfs_fileoff_t lfileoff = 0; - char *src = args->value; + __uint8_t *src = args->value; int blkcnt; int valuelen; int nmap; @@ -543,11 +545,6 @@ xfs_attr_rmtval_remove( /* * Roll through the "value", invalidating the attribute value's blocks. - * Note that args->rmtblkcnt is the minimum number of data blocks we'll - * see for a CRC enabled remote attribute. Each extent will have a - * header, and so we may have more blocks than we realise here. If we - * fail to map the blocks correctly, we'll have problems with the buffer - * lookups. */ lblkno = args->rmtblkno; blkcnt = args->rmtblkcnt; @@ -628,4 +625,3 @@ xfs_attr_rmtval_remove( } return(0); } - diff --git a/fs/xfs/xfs_bmap.c b/fs/xfs/xfs_bmap.c index 05c698c..92b8309 100644 --- a/fs/xfs/xfs_bmap.c +++ b/fs/xfs/xfs_bmap.c @@ -17,16 +17,17 @@ */ #include "xfs.h" #include "xfs_fs.h" -#include "xfs_types.h" +#include "xfs_format.h" #include "xfs_bit.h" #include "xfs_log.h" #include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" #include "xfs_ag.h" -#include "xfs_dir2.h" #include "xfs_mount.h" #include "xfs_da_btree.h" +#include "xfs_dir2_format.h" +#include "xfs_dir2.h" #include "xfs_bmap_btree.h" #include "xfs_alloc_btree.h" #include "xfs_ialloc_btree.h" @@ -39,6 +40,7 @@ #include "xfs_extfree_item.h" #include "xfs_alloc.h" #include "xfs_bmap.h" +#include "xfs_bmap_util.h" #include "xfs_rtalloc.h" #include "xfs_error.h" #include "xfs_attr_leaf.h" @@ -46,7 +48,6 @@ #include "xfs_trans_space.h" #include "xfs_buf_item.h" #include "xfs_filestream.h" -#include "xfs_vnodeops.h" #include "xfs_trace.h" #include "xfs_symlink.h" @@ -108,19 +109,6 @@ xfs_bmap_compute_maxlevels( mp->m_bm_maxlevels[whichfork] = level; } -/* - * Convert the given file system block to a disk block. We have to treat it - * differently based on whether the file is a real time file or not, because the - * bmap code does. - */ -xfs_daddr_t -xfs_fsb_to_db(struct xfs_inode *ip, xfs_fsblock_t fsb) -{ - return (XFS_IS_REALTIME_INODE(ip) ? \ - (xfs_daddr_t)XFS_FSB_TO_BB((ip)->i_mount, (fsb)) : \ - XFS_FSB_TO_DADDR((ip)->i_mount, (fsb))); -} - STATIC int /* error */ xfs_bmbt_lookup_eq( struct xfs_btree_cur *cur, @@ -263,173 +251,6 @@ xfs_bmap_forkoff_reset( } /* - * Extent tree block counting routines. - */ - -/* - * Count leaf blocks given a range of extent records. - */ -STATIC void -xfs_bmap_count_leaves( - xfs_ifork_t *ifp, - xfs_extnum_t idx, - int numrecs, - int *count) -{ - int b; - - for (b = 0; b < numrecs; b++) { - xfs_bmbt_rec_host_t *frp = xfs_iext_get_ext(ifp, idx + b); - *count += xfs_bmbt_get_blockcount(frp); - } -} - -/* - * Count leaf blocks given a range of extent records originally - * in btree format. - */ -STATIC void -xfs_bmap_disk_count_leaves( - struct xfs_mount *mp, - struct xfs_btree_block *block, - int numrecs, - int *count) -{ - int b; - xfs_bmbt_rec_t *frp; - - for (b = 1; b <= numrecs; b++) { - frp = XFS_BMBT_REC_ADDR(mp, block, b); - *count += xfs_bmbt_disk_get_blockcount(frp); - } -} - -/* - * Recursively walks each level of a btree - * to count total fsblocks is use. - */ -STATIC int /* error */ -xfs_bmap_count_tree( - xfs_mount_t *mp, /* file system mount point */ - xfs_trans_t *tp, /* transaction pointer */ - xfs_ifork_t *ifp, /* inode fork pointer */ - xfs_fsblock_t blockno, /* file system block number */ - int levelin, /* level in btree */ - int *count) /* Count of blocks */ -{ - int error; - xfs_buf_t *bp, *nbp; - int level = levelin; - __be64 *pp; - xfs_fsblock_t bno = blockno; - xfs_fsblock_t nextbno; - struct xfs_btree_block *block, *nextblock; - int numrecs; - - error = xfs_btree_read_bufl(mp, tp, bno, 0, &bp, XFS_BMAP_BTREE_REF, - &xfs_bmbt_buf_ops); - if (error) - return error; - *count += 1; - block = XFS_BUF_TO_BLOCK(bp); - - if (--level) { - /* Not at node above leaves, count this level of nodes */ - nextbno = be64_to_cpu(block->bb_u.l.bb_rightsib); - while (nextbno != NULLFSBLOCK) { - error = xfs_btree_read_bufl(mp, tp, nextbno, 0, &nbp, - XFS_BMAP_BTREE_REF, - &xfs_bmbt_buf_ops); - if (error) - return error; - *count += 1; - nextblock = XFS_BUF_TO_BLOCK(nbp); - nextbno = be64_to_cpu(nextblock->bb_u.l.bb_rightsib); - xfs_trans_brelse(tp, nbp); - } - - /* Dive to the next level */ - pp = XFS_BMBT_PTR_ADDR(mp, block, 1, mp->m_bmap_dmxr[1]); - bno = be64_to_cpu(*pp); - if (unlikely((error = - xfs_bmap_count_tree(mp, tp, ifp, bno, level, count)) < 0)) { - xfs_trans_brelse(tp, bp); - XFS_ERROR_REPORT("xfs_bmap_count_tree(1)", - XFS_ERRLEVEL_LOW, mp); - return XFS_ERROR(EFSCORRUPTED); - } - xfs_trans_brelse(tp, bp); - } else { - /* count all level 1 nodes and their leaves */ - for (;;) { - nextbno = be64_to_cpu(block->bb_u.l.bb_rightsib); - numrecs = be16_to_cpu(block->bb_numrecs); - xfs_bmap_disk_count_leaves(mp, block, numrecs, count); - xfs_trans_brelse(tp, bp); - if (nextbno == NULLFSBLOCK) - break; - bno = nextbno; - error = xfs_btree_read_bufl(mp, tp, bno, 0, &bp, - XFS_BMAP_BTREE_REF, - &xfs_bmbt_buf_ops); - if (error) - return error; - *count += 1; - block = XFS_BUF_TO_BLOCK(bp); - } - } - return 0; -} - -/* - * Count fsblocks of the given fork. - */ -int /* error */ -xfs_bmap_count_blocks( - xfs_trans_t *tp, /* transaction pointer */ - xfs_inode_t *ip, /* incore inode */ - int whichfork, /* data or attr fork */ - int *count) /* out: count of blocks */ -{ - struct xfs_btree_block *block; /* current btree block */ - xfs_fsblock_t bno; /* block # of "block" */ - xfs_ifork_t *ifp; /* fork structure */ - int level; /* btree level, for checking */ - xfs_mount_t *mp; /* file system mount structure */ - __be64 *pp; /* pointer to block address */ - - bno = NULLFSBLOCK; - mp = ip->i_mount; - ifp = XFS_IFORK_PTR(ip, whichfork); - if ( XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_EXTENTS ) { - xfs_bmap_count_leaves(ifp, 0, - ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t), - count); - return 0; - } - - /* - * Root level must use BMAP_BROOT_PTR_ADDR macro to get ptr out. - */ - block = ifp->if_broot; - level = be16_to_cpu(block->bb_level); - ASSERT(level > 0); - pp = XFS_BMAP_BROOT_PTR_ADDR(mp, block, 1, ifp->if_broot_bytes); - bno = be64_to_cpu(*pp); - ASSERT(bno != NULLDFSBNO); - ASSERT(XFS_FSB_TO_AGNO(mp, bno) < mp->m_sb.sb_agcount); - ASSERT(XFS_FSB_TO_AGBNO(mp, bno) < mp->m_sb.sb_agblocks); - - if (unlikely(xfs_bmap_count_tree(mp, tp, ifp, bno, level, count) < 0)) { - XFS_ERROR_REPORT("xfs_bmap_count_blocks(2)", XFS_ERRLEVEL_LOW, - mp); - return XFS_ERROR(EFSCORRUPTED); - } - - return 0; -} - -/* * Debug/sanity checking code */ @@ -724,8 +545,8 @@ xfs_bmap_trace_exlist( /* * Validate that the bmbt_irecs being returned from bmapi are valid - * given the callers original parameters. Specifically check the - * ranges of the returned irecs to ensure that they only extent beyond + * given the caller's original parameters. Specifically check the + * ranges of the returned irecs to ensure that they only extend beyond * the given parameters if the XFS_BMAPI_ENTIRE flag was set. */ STATIC void @@ -823,7 +644,7 @@ xfs_bmap_add_free( * Remove the entry "free" from the free item list. Prev points to the * previous entry, unless "free" is the head of the list. */ -STATIC void +void xfs_bmap_del_free( xfs_bmap_free_t *flist, /* free item list header */ xfs_bmap_free_item_t *prev, /* previous item on list, if any */ @@ -837,92 +658,6 @@ xfs_bmap_del_free( kmem_zone_free(xfs_bmap_free_item_zone, free); } - -/* - * Routine to be called at transaction's end by xfs_bmapi, xfs_bunmapi - * caller. Frees all the extents that need freeing, which must be done - * last due to locking considerations. We never free any extents in - * the first transaction. - * - * Return 1 if the given transaction was committed and a new one - * started, and 0 otherwise in the committed parameter. - */ -int /* error */ -xfs_bmap_finish( - xfs_trans_t **tp, /* transaction pointer addr */ - xfs_bmap_free_t *flist, /* i/o: list extents to free */ - int *committed) /* xact committed or not */ -{ - xfs_efd_log_item_t *efd; /* extent free data */ - xfs_efi_log_item_t *efi; /* extent free intention */ - int error; /* error return value */ - xfs_bmap_free_item_t *free; /* free extent item */ - unsigned int logres; /* new log reservation */ - unsigned int logcount; /* new log count */ - xfs_mount_t *mp; /* filesystem mount structure */ - xfs_bmap_free_item_t *next; /* next item on free list */ - xfs_trans_t *ntp; /* new transaction pointer */ - - ASSERT((*tp)->t_flags & XFS_TRANS_PERM_LOG_RES); - if (flist->xbf_count == 0) { - *committed = 0; - return 0; - } - ntp = *tp; - efi = xfs_trans_get_efi(ntp, flist->xbf_count); - for (free = flist->xbf_first; free; free = free->xbfi_next) - xfs_trans_log_efi_extent(ntp, efi, free->xbfi_startblock, - free->xbfi_blockcount); - logres = ntp->t_log_res; - logcount = ntp->t_log_count; - ntp = xfs_trans_dup(*tp); - error = xfs_trans_commit(*tp, 0); - *tp = ntp; - *committed = 1; - /* - * We have a new transaction, so we should return committed=1, - * even though we're returning an error. - */ - if (error) - return error; - - /* - * transaction commit worked ok so we can drop the extra ticket - * reference that we gained in xfs_trans_dup() - */ - xfs_log_ticket_put(ntp->t_ticket); - - if ((error = xfs_trans_reserve(ntp, 0, logres, 0, XFS_TRANS_PERM_LOG_RES, - logcount))) - return error; - efd = xfs_trans_get_efd(ntp, efi, flist->xbf_count); - for (free = flist->xbf_first; free != NULL; free = next) { - next = free->xbfi_next; - if ((error = xfs_free_extent(ntp, free->xbfi_startblock, - free->xbfi_blockcount))) { - /* - * The bmap free list will be cleaned up at a - * higher level. The EFI will be canceled when - * this transaction is aborted. - * Need to force shutdown here to make sure it - * happens, since this transaction may not be - * dirty yet. - */ - mp = ntp->t_mountp; - if (!XFS_FORCED_SHUTDOWN(mp)) - xfs_force_shutdown(mp, - (error == EFSCORRUPTED) ? - SHUTDOWN_CORRUPT_INCORE : - SHUTDOWN_META_IO_ERROR); - return error; - } - xfs_trans_log_efd_extent(ntp, efd, free->xbfi_startblock, - free->xbfi_blockcount); - xfs_bmap_del_free(flist, NULL, free); - } - return 0; -} - /* * Free up any items left in the list. */ @@ -1413,8 +1148,8 @@ xfs_bmap_add_attrfork( blks = XFS_ADDAFORK_SPACE_RES(mp); if (rsvd) tp->t_flags |= XFS_TRANS_RESERVE; - if ((error = xfs_trans_reserve(tp, blks, XFS_ADDAFORK_LOG_RES(mp), 0, - XFS_TRANS_PERM_LOG_RES, XFS_ADDAFORK_LOG_COUNT))) + error = xfs_trans_reserve(tp, &M_RES(mp)->tr_addafork, blks, 0); + if (error) goto error0; xfs_ilock(ip, XFS_ILOCK_EXCL); error = xfs_trans_reserve_quota_nblks(tp, ip, blks, 0, rsvd ? @@ -1815,7 +1550,7 @@ xfs_bmap_first_unused( } /* - * Returns the file-relative block number of the last block + 1 before + * Returns the file-relative block number of the last block - 1 before * last_block (input value) in the file. * This is not based on i_size, it is based on the extent records. * Returns 0 for local files, as they do not have extent records. @@ -1863,7 +1598,7 @@ xfs_bmap_last_before( return 0; } -STATIC int +int xfs_bmap_last_extent( struct xfs_trans *tp, struct xfs_inode *ip, @@ -1927,29 +1662,6 @@ xfs_bmap_isaeof( } /* - * Check if the endoff is outside the last extent. If so the caller will grow - * the allocation to a stripe unit boundary. All offsets are considered outside - * the end of file for an empty fork, so 1 is returned in *eof in that case. - */ -int -xfs_bmap_eof( - struct xfs_inode *ip, - xfs_fileoff_t endoff, - int whichfork, - int *eof) -{ - struct xfs_bmbt_irec rec; - int error; - - error = xfs_bmap_last_extent(NULL, ip, whichfork, &rec, eof); - if (error || *eof) - return error; - - *eof = endoff >= rec.br_startoff + rec.br_blockcount; - return 0; -} - -/* * Returns the file-relative block number of the first block past eof in * the file. This is not based on i_size, it is based on the extent records. * Returns 0 for local files, as they do not have extent records. @@ -3488,7 +3200,7 @@ done: /* * Adjust the size of the new extent based on di_extsize and rt extsize. */ -STATIC int +int xfs_bmap_extsize_align( xfs_mount_t *mp, xfs_bmbt_irec_t *gotp, /* next extent pointer */ @@ -3650,9 +3362,9 @@ xfs_bmap_extsize_align( #define XFS_ALLOC_GAP_UNITS 4 -STATIC void +void xfs_bmap_adjacent( - xfs_bmalloca_t *ap) /* bmap alloc argument struct */ + struct xfs_bmalloca *ap) /* bmap alloc argument struct */ { xfs_fsblock_t adjust; /* adjustment to block numbers */ xfs_agnumber_t fb_agno; /* ag number of ap->firstblock */ @@ -3799,109 +3511,6 @@ xfs_bmap_adjacent( } STATIC int -xfs_bmap_rtalloc( - xfs_bmalloca_t *ap) /* bmap alloc argument struct */ -{ - xfs_alloctype_t atype = 0; /* type for allocation routines */ - int error; /* error return value */ - xfs_mount_t *mp; /* mount point structure */ - xfs_extlen_t prod = 0; /* product factor for allocators */ - xfs_extlen_t ralen = 0; /* realtime allocation length */ - xfs_extlen_t align; /* minimum allocation alignment */ - xfs_rtblock_t rtb; - - mp = ap->ip->i_mount; - align = xfs_get_extsz_hint(ap->ip); - prod = align / mp->m_sb.sb_rextsize; - error = xfs_bmap_extsize_align(mp, &ap->got, &ap->prev, - align, 1, ap->eof, 0, - ap->conv, &ap->offset, &ap->length); - if (error) - return error; - ASSERT(ap->length); - ASSERT(ap->length % mp->m_sb.sb_rextsize == 0); - - /* - * If the offset & length are not perfectly aligned - * then kill prod, it will just get us in trouble. - */ - if (do_mod(ap->offset, align) || ap->length % align) - prod = 1; - /* - * Set ralen to be the actual requested length in rtextents. - */ - ralen = ap->length / mp->m_sb.sb_rextsize; - /* - * If the old value was close enough to MAXEXTLEN that - * we rounded up to it, cut it back so it's valid again. - * Note that if it's a really large request (bigger than - * MAXEXTLEN), we don't hear about that number, and can't - * adjust the starting point to match it. - */ - if (ralen * mp->m_sb.sb_rextsize >= MAXEXTLEN) - ralen = MAXEXTLEN / mp->m_sb.sb_rextsize; - - /* - * Lock out other modifications to the RT bitmap inode. - */ - xfs_ilock(mp->m_rbmip, XFS_ILOCK_EXCL); - xfs_trans_ijoin(ap->tp, mp->m_rbmip, XFS_ILOCK_EXCL); - - /* - * If it's an allocation to an empty file at offset 0, - * pick an extent that will space things out in the rt area. - */ - if (ap->eof && ap->offset == 0) { - xfs_rtblock_t uninitialized_var(rtx); /* realtime extent no */ - - error = xfs_rtpick_extent(mp, ap->tp, ralen, &rtx); - if (error) - return error; - ap->blkno = rtx * mp->m_sb.sb_rextsize; - } else { - ap->blkno = 0; - } - - xfs_bmap_adjacent(ap); - - /* - * Realtime allocation, done through xfs_rtallocate_extent. - */ - atype = ap->blkno == 0 ? XFS_ALLOCTYPE_ANY_AG : XFS_ALLOCTYPE_NEAR_BNO; - do_div(ap->blkno, mp->m_sb.sb_rextsize); - rtb = ap->blkno; - ap->length = ralen; - if ((error = xfs_rtallocate_extent(ap->tp, ap->blkno, 1, ap->length, - &ralen, atype, ap->wasdel, prod, &rtb))) - return error; - if (rtb == NULLFSBLOCK && prod > 1 && - (error = xfs_rtallocate_extent(ap->tp, ap->blkno, 1, - ap->length, &ralen, atype, - ap->wasdel, 1, &rtb))) - return error; - ap->blkno = rtb; - if (ap->blkno != NULLFSBLOCK) { - ap->blkno *= mp->m_sb.sb_rextsize; - ralen *= mp->m_sb.sb_rextsize; - ap->length = ralen; - ap->ip->i_d.di_nblocks += ralen; - xfs_trans_log_inode(ap->tp, ap->ip, XFS_ILOG_CORE); - if (ap->wasdel) - ap->ip->i_delayed_blks -= ralen; - /* - * Adjust the disk quota also. This was reserved - * earlier. - */ - xfs_trans_mod_dquot_byino(ap->tp, ap->ip, - ap->wasdel ? XFS_TRANS_DQ_DELRTBCOUNT : - XFS_TRANS_DQ_RTBCOUNT, (long) ralen); - } else { - ap->length = 0; - } - return 0; -} - -STATIC int xfs_bmap_btalloc_nullfb( struct xfs_bmalloca *ap, struct xfs_alloc_arg *args, @@ -4018,7 +3627,7 @@ xfs_bmap_btalloc_nullfb( STATIC int xfs_bmap_btalloc( - xfs_bmalloca_t *ap) /* bmap alloc argument struct */ + struct xfs_bmalloca *ap) /* bmap alloc argument struct */ { xfs_mount_t *mp; /* mount point structure */ xfs_alloctype_t atype = 0; /* type for allocation routines */ @@ -4250,7 +3859,7 @@ xfs_bmap_btalloc( */ STATIC int xfs_bmap_alloc( - xfs_bmalloca_t *ap) /* bmap alloc argument struct */ + struct xfs_bmalloca *ap) /* bmap alloc argument struct */ { if (XFS_IS_REALTIME_INODE(ap->ip) && ap->userdata) return xfs_bmap_rtalloc(ap); @@ -4638,7 +4247,7 @@ xfs_bmapi_delay( } -STATIC int +int __xfs_bmapi_allocate( struct xfs_bmalloca *bma) { @@ -4648,12 +4257,9 @@ __xfs_bmapi_allocate( struct xfs_ifork *ifp = XFS_IFORK_PTR(bma->ip, whichfork); int tmp_logflags = 0; int error; - int rt; ASSERT(bma->length > 0); - rt = (whichfork == XFS_DATA_FORK) && XFS_IS_REALTIME_INODE(bma->ip); - /* * For the wasdelay case, we could also just allocate the stuff asked * for in this bmap call but that wouldn't be as good. @@ -4756,45 +4362,6 @@ __xfs_bmapi_allocate( return 0; } -static void -xfs_bmapi_allocate_worker( - struct work_struct *work) -{ - struct xfs_bmalloca *args = container_of(work, - struct xfs_bmalloca, work); - unsigned long pflags; - - /* we are in a transaction context here */ - current_set_flags_nested(&pflags, PF_FSTRANS); - - args->result = __xfs_bmapi_allocate(args); - complete(args->done); - - current_restore_flags_nested(&pflags, PF_FSTRANS); -} - -/* - * Some allocation requests often come in with little stack to work on. Push - * them off to a worker thread so there is lots of stack to use. Otherwise just - * call directly to avoid the context switch overhead here. - */ -int -xfs_bmapi_allocate( - struct xfs_bmalloca *args) -{ - DECLARE_COMPLETION_ONSTACK(done); - - if (!args->stack_switch) - return __xfs_bmapi_allocate(args); - - - args->done = &done; - INIT_WORK_ONSTACK(&args->work, xfs_bmapi_allocate_worker); - queue_work(xfs_alloc_wq, &args->work); - wait_for_completion(&done); - return args->result; -} - STATIC int xfs_bmapi_convert_unwritten( struct xfs_bmalloca *bma, @@ -5789,359 +5356,3 @@ error0: } return error; } - -/* - * returns 1 for success, 0 if we failed to map the extent. - */ -STATIC int -xfs_getbmapx_fix_eof_hole( - xfs_inode_t *ip, /* xfs incore inode pointer */ - struct getbmapx *out, /* output structure */ - int prealloced, /* this is a file with - * preallocated data space */ - __int64_t end, /* last block requested */ - xfs_fsblock_t startblock) -{ - __int64_t fixlen; - xfs_mount_t *mp; /* file system mount point */ - xfs_ifork_t *ifp; /* inode fork pointer */ - xfs_extnum_t lastx; /* last extent pointer */ - xfs_fileoff_t fileblock; - - if (startblock == HOLESTARTBLOCK) { - mp = ip->i_mount; - out->bmv_block = -1; - fixlen = XFS_FSB_TO_BB(mp, XFS_B_TO_FSB(mp, XFS_ISIZE(ip))); - fixlen -= out->bmv_offset; - if (prealloced && out->bmv_offset + out->bmv_length == end) { - /* Came to hole at EOF. Trim it. */ - if (fixlen <= 0) - return 0; - out->bmv_length = fixlen; - } - } else { - if (startblock == DELAYSTARTBLOCK) - out->bmv_block = -2; - else - out->bmv_block = xfs_fsb_to_db(ip, startblock); - fileblock = XFS_BB_TO_FSB(ip->i_mount, out->bmv_offset); - ifp = XFS_IFORK_PTR(ip, XFS_DATA_FORK); - if (xfs_iext_bno_to_ext(ifp, fileblock, &lastx) && - (lastx == (ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t))-1)) - out->bmv_oflags |= BMV_OF_LAST; - } - - return 1; -} - -/* - * Get inode's extents as described in bmv, and format for output. - * Calls formatter to fill the user's buffer until all extents - * are mapped, until the passed-in bmv->bmv_count slots have - * been filled, or until the formatter short-circuits the loop, - * if it is tracking filled-in extents on its own. - */ -int /* error code */ -xfs_getbmap( - xfs_inode_t *ip, - struct getbmapx *bmv, /* user bmap structure */ - xfs_bmap_format_t formatter, /* format to user */ - void *arg) /* formatter arg */ -{ - __int64_t bmvend; /* last block requested */ - int error = 0; /* return value */ - __int64_t fixlen; /* length for -1 case */ - int i; /* extent number */ - int lock; /* lock state */ - xfs_bmbt_irec_t *map; /* buffer for user's data */ - xfs_mount_t *mp; /* file system mount point */ - int nex; /* # of user extents can do */ - int nexleft; /* # of user extents left */ - int subnex; /* # of bmapi's can do */ - int nmap; /* number of map entries */ - struct getbmapx *out; /* output structure */ - int whichfork; /* data or attr fork */ - int prealloced; /* this is a file with - * preallocated data space */ - int iflags; /* interface flags */ - int bmapi_flags; /* flags for xfs_bmapi */ - int cur_ext = 0; - - mp = ip->i_mount; - iflags = bmv->bmv_iflags; - whichfork = iflags & BMV_IF_ATTRFORK ? XFS_ATTR_FORK : XFS_DATA_FORK; - - if (whichfork == XFS_ATTR_FORK) { - if (XFS_IFORK_Q(ip)) { - if (ip->i_d.di_aformat != XFS_DINODE_FMT_EXTENTS && - ip->i_d.di_aformat != XFS_DINODE_FMT_BTREE && - ip->i_d.di_aformat != XFS_DINODE_FMT_LOCAL) - return XFS_ERROR(EINVAL); - } else if (unlikely( - ip->i_d.di_aformat != 0 && - ip->i_d.di_aformat != XFS_DINODE_FMT_EXTENTS)) { - XFS_ERROR_REPORT("xfs_getbmap", XFS_ERRLEVEL_LOW, - ip->i_mount); - return XFS_ERROR(EFSCORRUPTED); - } - - prealloced = 0; - fixlen = 1LL << 32; - } else { - if (ip->i_d.di_format != XFS_DINODE_FMT_EXTENTS && - ip->i_d.di_format != XFS_DINODE_FMT_BTREE && - ip->i_d.di_format != XFS_DINODE_FMT_LOCAL) - return XFS_ERROR(EINVAL); - - if (xfs_get_extsz_hint(ip) || - ip->i_d.di_flags & (XFS_DIFLAG_PREALLOC|XFS_DIFLAG_APPEND)){ - prealloced = 1; - fixlen = mp->m_super->s_maxbytes; - } else { - prealloced = 0; - fixlen = XFS_ISIZE(ip); - } - } - - if (bmv->bmv_length == -1) { - fixlen = XFS_FSB_TO_BB(mp, XFS_B_TO_FSB(mp, fixlen)); - bmv->bmv_length = - max_t(__int64_t, fixlen - bmv->bmv_offset, 0); - } else if (bmv->bmv_length == 0) { - bmv->bmv_entries = 0; - return 0; - } else if (bmv->bmv_length < 0) { - return XFS_ERROR(EINVAL); - } - - nex = bmv->bmv_count - 1; - if (nex <= 0) - return XFS_ERROR(EINVAL); - bmvend = bmv->bmv_offset + bmv->bmv_length; - - - if (bmv->bmv_count > ULONG_MAX / sizeof(struct getbmapx)) - return XFS_ERROR(ENOMEM); - out = kmem_zalloc(bmv->bmv_count * sizeof(struct getbmapx), KM_MAYFAIL); - if (!out) { - out = kmem_zalloc_large(bmv->bmv_count * - sizeof(struct getbmapx)); - if (!out) - return XFS_ERROR(ENOMEM); - } - - xfs_ilock(ip, XFS_IOLOCK_SHARED); - if (whichfork == XFS_DATA_FORK && !(iflags & BMV_IF_DELALLOC)) { - if (ip->i_delayed_blks || XFS_ISIZE(ip) > ip->i_d.di_size) { - error = -filemap_write_and_wait(VFS_I(ip)->i_mapping); - if (error) - goto out_unlock_iolock; - } - /* - * even after flushing the inode, there can still be delalloc - * blocks on the inode beyond EOF due to speculative - * preallocation. These are not removed until the release - * function is called or the inode is inactivated. Hence we - * cannot assert here that ip->i_delayed_blks == 0. - */ - } - - lock = xfs_ilock_map_shared(ip); - - /* - * Don't let nex be bigger than the number of extents - * we can have assuming alternating holes and real extents. - */ - if (nex > XFS_IFORK_NEXTENTS(ip, whichfork) * 2 + 1) - nex = XFS_IFORK_NEXTENTS(ip, whichfork) * 2 + 1; - - bmapi_flags = xfs_bmapi_aflag(whichfork); - if (!(iflags & BMV_IF_PREALLOC)) - bmapi_flags |= XFS_BMAPI_IGSTATE; - - /* - * Allocate enough space to handle "subnex" maps at a time. - */ - error = ENOMEM; - subnex = 16; - map = kmem_alloc(subnex * sizeof(*map), KM_MAYFAIL | KM_NOFS); - if (!map) - goto out_unlock_ilock; - - bmv->bmv_entries = 0; - - if (XFS_IFORK_NEXTENTS(ip, whichfork) == 0 && - (whichfork == XFS_ATTR_FORK || !(iflags & BMV_IF_DELALLOC))) { - error = 0; - goto out_free_map; - } - - nexleft = nex; - - do { - nmap = (nexleft > subnex) ? subnex : nexleft; - error = xfs_bmapi_read(ip, XFS_BB_TO_FSBT(mp, bmv->bmv_offset), - XFS_BB_TO_FSB(mp, bmv->bmv_length), - map, &nmap, bmapi_flags); - if (error) - goto out_free_map; - ASSERT(nmap <= subnex); - - for (i = 0; i < nmap && nexleft && bmv->bmv_length; i++) { - out[cur_ext].bmv_oflags = 0; - if (map[i].br_state == XFS_EXT_UNWRITTEN) - out[cur_ext].bmv_oflags |= BMV_OF_PREALLOC; - else if (map[i].br_startblock == DELAYSTARTBLOCK) - out[cur_ext].bmv_oflags |= BMV_OF_DELALLOC; - out[cur_ext].bmv_offset = - XFS_FSB_TO_BB(mp, map[i].br_startoff); - out[cur_ext].bmv_length = - XFS_FSB_TO_BB(mp, map[i].br_blockcount); - out[cur_ext].bmv_unused1 = 0; - out[cur_ext].bmv_unused2 = 0; - - /* - * delayed allocation extents that start beyond EOF can - * occur due to speculative EOF allocation when the - * delalloc extent is larger than the largest freespace - * extent at conversion time. These extents cannot be - * converted by data writeback, so can exist here even - * if we are not supposed to be finding delalloc - * extents. - */ - if (map[i].br_startblock == DELAYSTARTBLOCK && - map[i].br_startoff <= XFS_B_TO_FSB(mp, XFS_ISIZE(ip))) - ASSERT((iflags & BMV_IF_DELALLOC) != 0); - - if (map[i].br_startblock == HOLESTARTBLOCK && - whichfork == XFS_ATTR_FORK) { - /* came to the end of attribute fork */ - out[cur_ext].bmv_oflags |= BMV_OF_LAST; - goto out_free_map; - } - - if (!xfs_getbmapx_fix_eof_hole(ip, &out[cur_ext], - prealloced, bmvend, - map[i].br_startblock)) - goto out_free_map; - - bmv->bmv_offset = - out[cur_ext].bmv_offset + - out[cur_ext].bmv_length; - bmv->bmv_length = - max_t(__int64_t, 0, bmvend - bmv->bmv_offset); - - /* - * In case we don't want to return the hole, - * don't increase cur_ext so that we can reuse - * it in the next loop. - */ - if ((iflags & BMV_IF_NO_HOLES) && - map[i].br_startblock == HOLESTARTBLOCK) { - memset(&out[cur_ext], 0, sizeof(out[cur_ext])); - continue; - } - - nexleft--; - bmv->bmv_entries++; - cur_ext++; - } - } while (nmap && nexleft && bmv->bmv_length); - - out_free_map: - kmem_free(map); - out_unlock_ilock: - xfs_iunlock_map_shared(ip, lock); - out_unlock_iolock: - xfs_iunlock(ip, XFS_IOLOCK_SHARED); - - for (i = 0; i < cur_ext; i++) { - int full = 0; /* user array is full */ - - /* format results & advance arg */ - error = formatter(&arg, &out[i], &full); - if (error || full) - break; - } - - if (is_vmalloc_addr(out)) - kmem_free_large(out); - else - kmem_free(out); - return error; -} - -/* - * dead simple method of punching delalyed allocation blocks from a range in - * the inode. Walks a block at a time so will be slow, but is only executed in - * rare error cases so the overhead is not critical. This will alays punch out - * both the start and end blocks, even if the ranges only partially overlap - * them, so it is up to the caller to ensure that partial blocks are not - * passed in. - */ -int -xfs_bmap_punch_delalloc_range( - struct xfs_inode *ip, - xfs_fileoff_t start_fsb, - xfs_fileoff_t length) -{ - xfs_fileoff_t remaining = length; - int error = 0; - - ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); - - do { - int done; - xfs_bmbt_irec_t imap; - int nimaps = 1; - xfs_fsblock_t firstblock; - xfs_bmap_free_t flist; - - /* - * Map the range first and check that it is a delalloc extent - * before trying to unmap the range. Otherwise we will be - * trying to remove a real extent (which requires a - * transaction) or a hole, which is probably a bad idea... - */ - error = xfs_bmapi_read(ip, start_fsb, 1, &imap, &nimaps, - XFS_BMAPI_ENTIRE); - - if (error) { - /* something screwed, just bail */ - if (!XFS_FORCED_SHUTDOWN(ip->i_mount)) { - xfs_alert(ip->i_mount, - "Failed delalloc mapping lookup ino %lld fsb %lld.", - ip->i_ino, start_fsb); - } - break; - } - if (!nimaps) { - /* nothing there */ - goto next_block; - } - if (imap.br_startblock != DELAYSTARTBLOCK) { - /* been converted, ignore */ - goto next_block; - } - WARN_ON(imap.br_blockcount == 0); - - /* - * Note: while we initialise the firstblock/flist pair, they - * should never be used because blocks should never be - * allocated or freed for a delalloc extent and hence we need - * don't cancel or finish them after the xfs_bunmapi() call. - */ - xfs_bmap_init(&flist, &firstblock); - error = xfs_bunmapi(NULL, ip, start_fsb, 1, 0, 1, &firstblock, - &flist, &done); - if (error) - break; - - ASSERT(!flist.xbf_count && !flist.xbf_first); -next_block: - start_fsb++; - remaining--; - } while(remaining > 0); - - return error; -} diff --git a/fs/xfs/xfs_bmap.h b/fs/xfs/xfs_bmap.h index 1cf1292..33b41f3 100644 --- a/fs/xfs/xfs_bmap.h +++ b/fs/xfs/xfs_bmap.h @@ -108,41 +108,6 @@ static inline void xfs_bmap_init(xfs_bmap_free_t *flp, xfs_fsblock_t *fbp) } /* - * Argument structure for xfs_bmap_alloc. - */ -typedef struct xfs_bmalloca { - xfs_fsblock_t *firstblock; /* i/o first block allocated */ - struct xfs_bmap_free *flist; /* bmap freelist */ - struct xfs_trans *tp; /* transaction pointer */ - struct xfs_inode *ip; /* incore inode pointer */ - struct xfs_bmbt_irec prev; /* extent before the new one */ - struct xfs_bmbt_irec got; /* extent after, or delayed */ - - xfs_fileoff_t offset; /* offset in file filling in */ - xfs_extlen_t length; /* i/o length asked/allocated */ - xfs_fsblock_t blkno; /* starting block of new extent */ - - struct xfs_btree_cur *cur; /* btree cursor */ - xfs_extnum_t idx; /* current extent index */ - int nallocs;/* number of extents alloc'd */ - int logflags;/* flags for transaction logging */ - - xfs_extlen_t total; /* total blocks needed for xaction */ - xfs_extlen_t minlen; /* minimum allocation size (blocks) */ - xfs_extlen_t minleft; /* amount must be left after alloc */ - char eof; /* set if allocating past last extent */ - char wasdel; /* replacing a delayed allocation */ - char userdata;/* set if is user data */ - char aeof; /* allocated space at eof */ - char conv; /* overwriting unwritten extents */ - char stack_switch; - int flags; - struct completion *done; - struct work_struct work; - int result; -} xfs_bmalloca_t; - -/* * Flags for xfs_bmap_add_extent*. */ #define BMAP_LEFT_CONTIG (1 << 0) @@ -162,7 +127,7 @@ typedef struct xfs_bmalloca { { BMAP_RIGHT_FILLING, "RF" }, \ { BMAP_ATTRFORK, "ATTR" } -#if defined(__KERNEL) && defined(DEBUG) +#ifdef DEBUG void xfs_bmap_trace_exlist(struct xfs_inode *ip, xfs_extnum_t cnt, int whichfork, unsigned long caller_ip); #define XFS_BMAP_TRACE_EXLIST(ip,c,w) \ @@ -205,23 +170,4 @@ int xfs_check_nostate_extents(struct xfs_ifork *ifp, xfs_extnum_t idx, xfs_extnum_t num); uint xfs_default_attroffset(struct xfs_inode *ip); -#ifdef __KERNEL__ -/* bmap to userspace formatter - copy to user & advance pointer */ -typedef int (*xfs_bmap_format_t)(void **, struct getbmapx *, int *); - -int xfs_bmap_finish(struct xfs_trans **tp, struct xfs_bmap_free *flist, - int *committed); -int xfs_getbmap(struct xfs_inode *ip, struct getbmapx *bmv, - xfs_bmap_format_t formatter, void *arg); -int xfs_bmap_eof(struct xfs_inode *ip, xfs_fileoff_t endoff, - int whichfork, int *eof); -int xfs_bmap_count_blocks(struct xfs_trans *tp, struct xfs_inode *ip, - int whichfork, int *count); -int xfs_bmap_punch_delalloc_range(struct xfs_inode *ip, - xfs_fileoff_t start_fsb, xfs_fileoff_t length); - -xfs_daddr_t xfs_fsb_to_db(struct xfs_inode *ip, xfs_fsblock_t fsb); - -#endif /* __KERNEL__ */ - #endif /* __XFS_BMAP_H__ */ diff --git a/fs/xfs/xfs_bmap_btree.c b/fs/xfs/xfs_bmap_btree.c index 0c61a22..cf3bc76 100644 --- a/fs/xfs/xfs_bmap_btree.c +++ b/fs/xfs/xfs_bmap_btree.c @@ -17,7 +17,7 @@ */ #include "xfs.h" #include "xfs_fs.h" -#include "xfs_types.h" +#include "xfs_format.h" #include "xfs_bit.h" #include "xfs_log.h" #include "xfs_trans.h" @@ -722,7 +722,7 @@ xfs_bmbt_key_diff( cur->bc_rec.b.br_startoff; } -static int +static bool xfs_bmbt_verify( struct xfs_buf *bp) { @@ -775,7 +775,6 @@ xfs_bmbt_verify( return false; return true; - } static void @@ -789,7 +788,6 @@ xfs_bmbt_read_verify( bp->b_target->bt_mount, bp->b_addr); xfs_buf_ioerror(bp, EFSCORRUPTED); } - } static void diff --git a/fs/xfs/xfs_bmap_util.c b/fs/xfs/xfs_bmap_util.c new file mode 100644 index 0000000..541d59f --- /dev/null +++ b/fs/xfs/xfs_bmap_util.c @@ -0,0 +1,2026 @@ +/* + * Copyright (c) 2000-2006 Silicon Graphics, Inc. + * Copyright (c) 2012 Red Hat, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#include "xfs.h" +#include "xfs_fs.h" +#include "xfs_format.h" +#include "xfs_bit.h" +#include "xfs_log.h" +#include "xfs_inum.h" +#include "xfs_trans.h" +#include "xfs_sb.h" +#include "xfs_ag.h" +#include "xfs_mount.h" +#include "xfs_da_btree.h" +#include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" +#include "xfs_ialloc_btree.h" +#include "xfs_dinode.h" +#include "xfs_inode.h" +#include "xfs_btree.h" +#include "xfs_extfree_item.h" +#include "xfs_alloc.h" +#include "xfs_bmap.h" +#include "xfs_bmap_util.h" +#include "xfs_rtalloc.h" +#include "xfs_error.h" +#include "xfs_quota.h" +#include "xfs_trans_space.h" +#include "xfs_trace.h" +#include "xfs_icache.h" + +/* Kernel only BMAP related definitions and functions */ + +/* + * Convert the given file system block to a disk block. We have to treat it + * differently based on whether the file is a real time file or not, because the + * bmap code does. + */ +xfs_daddr_t +xfs_fsb_to_db(struct xfs_inode *ip, xfs_fsblock_t fsb) +{ + return (XFS_IS_REALTIME_INODE(ip) ? \ + (xfs_daddr_t)XFS_FSB_TO_BB((ip)->i_mount, (fsb)) : \ + XFS_FSB_TO_DADDR((ip)->i_mount, (fsb))); +} + +/* + * Routine to be called at transaction's end by xfs_bmapi, xfs_bunmapi + * caller. Frees all the extents that need freeing, which must be done + * last due to locking considerations. We never free any extents in + * the first transaction. + * + * Return 1 if the given transaction was committed and a new one + * started, and 0 otherwise in the committed parameter. + */ +int /* error */ +xfs_bmap_finish( + xfs_trans_t **tp, /* transaction pointer addr */ + xfs_bmap_free_t *flist, /* i/o: list extents to free */ + int *committed) /* xact committed or not */ +{ + xfs_efd_log_item_t *efd; /* extent free data */ + xfs_efi_log_item_t *efi; /* extent free intention */ + int error; /* error return value */ + xfs_bmap_free_item_t *free; /* free extent item */ + struct xfs_trans_res tres; /* new log reservation */ + xfs_mount_t *mp; /* filesystem mount structure */ + xfs_bmap_free_item_t *next; /* next item on free list */ + xfs_trans_t *ntp; /* new transaction pointer */ + + ASSERT((*tp)->t_flags & XFS_TRANS_PERM_LOG_RES); + if (flist->xbf_count == 0) { + *committed = 0; + return 0; + } + ntp = *tp; + efi = xfs_trans_get_efi(ntp, flist->xbf_count); + for (free = flist->xbf_first; free; free = free->xbfi_next) + xfs_trans_log_efi_extent(ntp, efi, free->xbfi_startblock, + free->xbfi_blockcount); + + tres.tr_logres = ntp->t_log_res; + tres.tr_logcount = ntp->t_log_count; + tres.tr_logflags = XFS_TRANS_PERM_LOG_RES; + ntp = xfs_trans_dup(*tp); + error = xfs_trans_commit(*tp, 0); + *tp = ntp; + *committed = 1; + /* + * We have a new transaction, so we should return committed=1, + * even though we're returning an error. + */ + if (error) + return error; + + /* + * transaction commit worked ok so we can drop the extra ticket + * reference that we gained in xfs_trans_dup() + */ + xfs_log_ticket_put(ntp->t_ticket); + + error = xfs_trans_reserve(ntp, &tres, 0, 0); + if (error) + return error; + efd = xfs_trans_get_efd(ntp, efi, flist->xbf_count); + for (free = flist->xbf_first; free != NULL; free = next) { + next = free->xbfi_next; + if ((error = xfs_free_extent(ntp, free->xbfi_startblock, + free->xbfi_blockcount))) { + /* + * The bmap free list will be cleaned up at a + * higher level. The EFI will be canceled when + * this transaction is aborted. + * Need to force shutdown here to make sure it + * happens, since this transaction may not be + * dirty yet. + */ + mp = ntp->t_mountp; + if (!XFS_FORCED_SHUTDOWN(mp)) + xfs_force_shutdown(mp, + (error == EFSCORRUPTED) ? + SHUTDOWN_CORRUPT_INCORE : + SHUTDOWN_META_IO_ERROR); + return error; + } + xfs_trans_log_efd_extent(ntp, efd, free->xbfi_startblock, + free->xbfi_blockcount); + xfs_bmap_del_free(flist, NULL, free); + } + return 0; +} + +int +xfs_bmap_rtalloc( + struct xfs_bmalloca *ap) /* bmap alloc argument struct */ +{ + xfs_alloctype_t atype = 0; /* type for allocation routines */ + int error; /* error return value */ + xfs_mount_t *mp; /* mount point structure */ + xfs_extlen_t prod = 0; /* product factor for allocators */ + xfs_extlen_t ralen = 0; /* realtime allocation length */ + xfs_extlen_t align; /* minimum allocation alignment */ + xfs_rtblock_t rtb; + + mp = ap->ip->i_mount; + align = xfs_get_extsz_hint(ap->ip); + prod = align / mp->m_sb.sb_rextsize; + error = xfs_bmap_extsize_align(mp, &ap->got, &ap->prev, + align, 1, ap->eof, 0, + ap->conv, &ap->offset, &ap->length); + if (error) + return error; + ASSERT(ap->length); + ASSERT(ap->length % mp->m_sb.sb_rextsize == 0); + + /* + * If the offset & length are not perfectly aligned + * then kill prod, it will just get us in trouble. + */ + if (do_mod(ap->offset, align) || ap->length % align) + prod = 1; + /* + * Set ralen to be the actual requested length in rtextents. + */ + ralen = ap->length / mp->m_sb.sb_rextsize; + /* + * If the old value was close enough to MAXEXTLEN that + * we rounded up to it, cut it back so it's valid again. + * Note that if it's a really large request (bigger than + * MAXEXTLEN), we don't hear about that number, and can't + * adjust the starting point to match it. + */ + if (ralen * mp->m_sb.sb_rextsize >= MAXEXTLEN) + ralen = MAXEXTLEN / mp->m_sb.sb_rextsize; + + /* + * Lock out other modifications to the RT bitmap inode. + */ + xfs_ilock(mp->m_rbmip, XFS_ILOCK_EXCL); + xfs_trans_ijoin(ap->tp, mp->m_rbmip, XFS_ILOCK_EXCL); + + /* + * If it's an allocation to an empty file at offset 0, + * pick an extent that will space things out in the rt area. + */ + if (ap->eof && ap->offset == 0) { + xfs_rtblock_t uninitialized_var(rtx); /* realtime extent no */ + + error = xfs_rtpick_extent(mp, ap->tp, ralen, &rtx); + if (error) + return error; + ap->blkno = rtx * mp->m_sb.sb_rextsize; + } else { + ap->blkno = 0; + } + + xfs_bmap_adjacent(ap); + + /* + * Realtime allocation, done through xfs_rtallocate_extent. + */ + atype = ap->blkno == 0 ? XFS_ALLOCTYPE_ANY_AG : XFS_ALLOCTYPE_NEAR_BNO; + do_div(ap->blkno, mp->m_sb.sb_rextsize); + rtb = ap->blkno; + ap->length = ralen; + if ((error = xfs_rtallocate_extent(ap->tp, ap->blkno, 1, ap->length, + &ralen, atype, ap->wasdel, prod, &rtb))) + return error; + if (rtb == NULLFSBLOCK && prod > 1 && + (error = xfs_rtallocate_extent(ap->tp, ap->blkno, 1, + ap->length, &ralen, atype, + ap->wasdel, 1, &rtb))) + return error; + ap->blkno = rtb; + if (ap->blkno != NULLFSBLOCK) { + ap->blkno *= mp->m_sb.sb_rextsize; + ralen *= mp->m_sb.sb_rextsize; + ap->length = ralen; + ap->ip->i_d.di_nblocks += ralen; + xfs_trans_log_inode(ap->tp, ap->ip, XFS_ILOG_CORE); + if (ap->wasdel) + ap->ip->i_delayed_blks -= ralen; + /* + * Adjust the disk quota also. This was reserved + * earlier. + */ + xfs_trans_mod_dquot_byino(ap->tp, ap->ip, + ap->wasdel ? XFS_TRANS_DQ_DELRTBCOUNT : + XFS_TRANS_DQ_RTBCOUNT, (long) ralen); + } else { + ap->length = 0; + } + return 0; +} + +/* + * Stack switching interfaces for allocation + */ +static void +xfs_bmapi_allocate_worker( + struct work_struct *work) +{ + struct xfs_bmalloca *args = container_of(work, + struct xfs_bmalloca, work); + unsigned long pflags; + + /* we are in a transaction context here */ + current_set_flags_nested(&pflags, PF_FSTRANS); + + args->result = __xfs_bmapi_allocate(args); + complete(args->done); + + current_restore_flags_nested(&pflags, PF_FSTRANS); +} + +/* + * Some allocation requests often come in with little stack to work on. Push + * them off to a worker thread so there is lots of stack to use. Otherwise just + * call directly to avoid the context switch overhead here. + */ +int +xfs_bmapi_allocate( + struct xfs_bmalloca *args) +{ + DECLARE_COMPLETION_ONSTACK(done); + + if (!args->stack_switch) + return __xfs_bmapi_allocate(args); + + + args->done = &done; + INIT_WORK_ONSTACK(&args->work, xfs_bmapi_allocate_worker); + queue_work(xfs_alloc_wq, &args->work); + wait_for_completion(&done); + return args->result; +} + +/* + * Check if the endoff is outside the last extent. If so the caller will grow + * the allocation to a stripe unit boundary. All offsets are considered outside + * the end of file for an empty fork, so 1 is returned in *eof in that case. + */ +int +xfs_bmap_eof( + struct xfs_inode *ip, + xfs_fileoff_t endoff, + int whichfork, + int *eof) +{ + struct xfs_bmbt_irec rec; + int error; + + error = xfs_bmap_last_extent(NULL, ip, whichfork, &rec, eof); + if (error || *eof) + return error; + + *eof = endoff >= rec.br_startoff + rec.br_blockcount; + return 0; +} + +/* + * Extent tree block counting routines. + */ + +/* + * Count leaf blocks given a range of extent records. + */ +STATIC void +xfs_bmap_count_leaves( + xfs_ifork_t *ifp, + xfs_extnum_t idx, + int numrecs, + int *count) +{ + int b; + + for (b = 0; b < numrecs; b++) { + xfs_bmbt_rec_host_t *frp = xfs_iext_get_ext(ifp, idx + b); + *count += xfs_bmbt_get_blockcount(frp); + } +} + +/* + * Count leaf blocks given a range of extent records originally + * in btree format. + */ +STATIC void +xfs_bmap_disk_count_leaves( + struct xfs_mount *mp, + struct xfs_btree_block *block, + int numrecs, + int *count) +{ + int b; + xfs_bmbt_rec_t *frp; + + for (b = 1; b <= numrecs; b++) { + frp = XFS_BMBT_REC_ADDR(mp, block, b); + *count += xfs_bmbt_disk_get_blockcount(frp); + } +} + +/* + * Recursively walks each level of a btree + * to count total fsblocks in use. + */ +STATIC int /* error */ +xfs_bmap_count_tree( + xfs_mount_t *mp, /* file system mount point */ + xfs_trans_t *tp, /* transaction pointer */ + xfs_ifork_t *ifp, /* inode fork pointer */ + xfs_fsblock_t blockno, /* file system block number */ + int levelin, /* level in btree */ + int *count) /* Count of blocks */ +{ + int error; + xfs_buf_t *bp, *nbp; + int level = levelin; + __be64 *pp; + xfs_fsblock_t bno = blockno; + xfs_fsblock_t nextbno; + struct xfs_btree_block *block, *nextblock; + int numrecs; + + error = xfs_btree_read_bufl(mp, tp, bno, 0, &bp, XFS_BMAP_BTREE_REF, + &xfs_bmbt_buf_ops); + if (error) + return error; + *count += 1; + block = XFS_BUF_TO_BLOCK(bp); + + if (--level) { + /* Not at node above leaves, count this level of nodes */ + nextbno = be64_to_cpu(block->bb_u.l.bb_rightsib); + while (nextbno != NULLFSBLOCK) { + error = xfs_btree_read_bufl(mp, tp, nextbno, 0, &nbp, + XFS_BMAP_BTREE_REF, + &xfs_bmbt_buf_ops); + if (error) + return error; + *count += 1; + nextblock = XFS_BUF_TO_BLOCK(nbp); + nextbno = be64_to_cpu(nextblock->bb_u.l.bb_rightsib); + xfs_trans_brelse(tp, nbp); + } + + /* Dive to the next level */ + pp = XFS_BMBT_PTR_ADDR(mp, block, 1, mp->m_bmap_dmxr[1]); + bno = be64_to_cpu(*pp); + if (unlikely((error = + xfs_bmap_count_tree(mp, tp, ifp, bno, level, count)) < 0)) { + xfs_trans_brelse(tp, bp); + XFS_ERROR_REPORT("xfs_bmap_count_tree(1)", + XFS_ERRLEVEL_LOW, mp); + return XFS_ERROR(EFSCORRUPTED); + } + xfs_trans_brelse(tp, bp); + } else { + /* count all level 1 nodes and their leaves */ + for (;;) { + nextbno = be64_to_cpu(block->bb_u.l.bb_rightsib); + numrecs = be16_to_cpu(block->bb_numrecs); + xfs_bmap_disk_count_leaves(mp, block, numrecs, count); + xfs_trans_brelse(tp, bp); + if (nextbno == NULLFSBLOCK) + break; + bno = nextbno; + error = xfs_btree_read_bufl(mp, tp, bno, 0, &bp, + XFS_BMAP_BTREE_REF, + &xfs_bmbt_buf_ops); + if (error) + return error; + *count += 1; + block = XFS_BUF_TO_BLOCK(bp); + } + } + return 0; +} + +/* + * Count fsblocks of the given fork. + */ +int /* error */ +xfs_bmap_count_blocks( + xfs_trans_t *tp, /* transaction pointer */ + xfs_inode_t *ip, /* incore inode */ + int whichfork, /* data or attr fork */ + int *count) /* out: count of blocks */ +{ + struct xfs_btree_block *block; /* current btree block */ + xfs_fsblock_t bno; /* block # of "block" */ + xfs_ifork_t *ifp; /* fork structure */ + int level; /* btree level, for checking */ + xfs_mount_t *mp; /* file system mount structure */ + __be64 *pp; /* pointer to block address */ + + bno = NULLFSBLOCK; + mp = ip->i_mount; + ifp = XFS_IFORK_PTR(ip, whichfork); + if ( XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_EXTENTS ) { + xfs_bmap_count_leaves(ifp, 0, + ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t), + count); + return 0; + } + + /* + * Root level must use BMAP_BROOT_PTR_ADDR macro to get ptr out. + */ + block = ifp->if_broot; + level = be16_to_cpu(block->bb_level); + ASSERT(level > 0); + pp = XFS_BMAP_BROOT_PTR_ADDR(mp, block, 1, ifp->if_broot_bytes); + bno = be64_to_cpu(*pp); + ASSERT(bno != NULLDFSBNO); + ASSERT(XFS_FSB_TO_AGNO(mp, bno) < mp->m_sb.sb_agcount); + ASSERT(XFS_FSB_TO_AGBNO(mp, bno) < mp->m_sb.sb_agblocks); + + if (unlikely(xfs_bmap_count_tree(mp, tp, ifp, bno, level, count) < 0)) { + XFS_ERROR_REPORT("xfs_bmap_count_blocks(2)", XFS_ERRLEVEL_LOW, + mp); + return XFS_ERROR(EFSCORRUPTED); + } + + return 0; +} + +/* + * returns 1 for success, 0 if we failed to map the extent. + */ +STATIC int +xfs_getbmapx_fix_eof_hole( + xfs_inode_t *ip, /* xfs incore inode pointer */ + struct getbmapx *out, /* output structure */ + int prealloced, /* this is a file with + * preallocated data space */ + __int64_t end, /* last block requested */ + xfs_fsblock_t startblock) +{ + __int64_t fixlen; + xfs_mount_t *mp; /* file system mount point */ + xfs_ifork_t *ifp; /* inode fork pointer */ + xfs_extnum_t lastx; /* last extent pointer */ + xfs_fileoff_t fileblock; + + if (startblock == HOLESTARTBLOCK) { + mp = ip->i_mount; + out->bmv_block = -1; + fixlen = XFS_FSB_TO_BB(mp, XFS_B_TO_FSB(mp, XFS_ISIZE(ip))); + fixlen -= out->bmv_offset; + if (prealloced && out->bmv_offset + out->bmv_length == end) { + /* Came to hole at EOF. Trim it. */ + if (fixlen <= 0) + return 0; + out->bmv_length = fixlen; + } + } else { + if (startblock == DELAYSTARTBLOCK) + out->bmv_block = -2; + else + out->bmv_block = xfs_fsb_to_db(ip, startblock); + fileblock = XFS_BB_TO_FSB(ip->i_mount, out->bmv_offset); + ifp = XFS_IFORK_PTR(ip, XFS_DATA_FORK); + if (xfs_iext_bno_to_ext(ifp, fileblock, &lastx) && + (lastx == (ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t))-1)) + out->bmv_oflags |= BMV_OF_LAST; + } + + return 1; +} + +/* + * Get inode's extents as described in bmv, and format for output. + * Calls formatter to fill the user's buffer until all extents + * are mapped, until the passed-in bmv->bmv_count slots have + * been filled, or until the formatter short-circuits the loop, + * if it is tracking filled-in extents on its own. + */ +int /* error code */ +xfs_getbmap( + xfs_inode_t *ip, + struct getbmapx *bmv, /* user bmap structure */ + xfs_bmap_format_t formatter, /* format to user */ + void *arg) /* formatter arg */ +{ + __int64_t bmvend; /* last block requested */ + int error = 0; /* return value */ + __int64_t fixlen; /* length for -1 case */ + int i; /* extent number */ + int lock; /* lock state */ + xfs_bmbt_irec_t *map; /* buffer for user's data */ + xfs_mount_t *mp; /* file system mount point */ + int nex; /* # of user extents can do */ + int nexleft; /* # of user extents left */ + int subnex; /* # of bmapi's can do */ + int nmap; /* number of map entries */ + struct getbmapx *out; /* output structure */ + int whichfork; /* data or attr fork */ + int prealloced; /* this is a file with + * preallocated data space */ + int iflags; /* interface flags */ + int bmapi_flags; /* flags for xfs_bmapi */ + int cur_ext = 0; + + mp = ip->i_mount; + iflags = bmv->bmv_iflags; + whichfork = iflags & BMV_IF_ATTRFORK ? XFS_ATTR_FORK : XFS_DATA_FORK; + + if (whichfork == XFS_ATTR_FORK) { + if (XFS_IFORK_Q(ip)) { + if (ip->i_d.di_aformat != XFS_DINODE_FMT_EXTENTS && + ip->i_d.di_aformat != XFS_DINODE_FMT_BTREE && + ip->i_d.di_aformat != XFS_DINODE_FMT_LOCAL) + return XFS_ERROR(EINVAL); + } else if (unlikely( + ip->i_d.di_aformat != 0 && + ip->i_d.di_aformat != XFS_DINODE_FMT_EXTENTS)) { + XFS_ERROR_REPORT("xfs_getbmap", XFS_ERRLEVEL_LOW, + ip->i_mount); + return XFS_ERROR(EFSCORRUPTED); + } + + prealloced = 0; + fixlen = 1LL << 32; + } else { + if (ip->i_d.di_format != XFS_DINODE_FMT_EXTENTS && + ip->i_d.di_format != XFS_DINODE_FMT_BTREE && + ip->i_d.di_format != XFS_DINODE_FMT_LOCAL) + return XFS_ERROR(EINVAL); + + if (xfs_get_extsz_hint(ip) || + ip->i_d.di_flags & (XFS_DIFLAG_PREALLOC|XFS_DIFLAG_APPEND)){ + prealloced = 1; + fixlen = mp->m_super->s_maxbytes; + } else { + prealloced = 0; + fixlen = XFS_ISIZE(ip); + } + } + + if (bmv->bmv_length == -1) { + fixlen = XFS_FSB_TO_BB(mp, XFS_B_TO_FSB(mp, fixlen)); + bmv->bmv_length = + max_t(__int64_t, fixlen - bmv->bmv_offset, 0); + } else if (bmv->bmv_length == 0) { + bmv->bmv_entries = 0; + return 0; + } else if (bmv->bmv_length < 0) { + return XFS_ERROR(EINVAL); + } + + nex = bmv->bmv_count - 1; + if (nex <= 0) + return XFS_ERROR(EINVAL); + bmvend = bmv->bmv_offset + bmv->bmv_length; + + + if (bmv->bmv_count > ULONG_MAX / sizeof(struct getbmapx)) + return XFS_ERROR(ENOMEM); + out = kmem_zalloc(bmv->bmv_count * sizeof(struct getbmapx), KM_MAYFAIL); + if (!out) { + out = kmem_zalloc_large(bmv->bmv_count * + sizeof(struct getbmapx)); + if (!out) + return XFS_ERROR(ENOMEM); + } + + xfs_ilock(ip, XFS_IOLOCK_SHARED); + if (whichfork == XFS_DATA_FORK && !(iflags & BMV_IF_DELALLOC)) { + if (ip->i_delayed_blks || XFS_ISIZE(ip) > ip->i_d.di_size) { + error = -filemap_write_and_wait(VFS_I(ip)->i_mapping); + if (error) + goto out_unlock_iolock; + } + /* + * even after flushing the inode, there can still be delalloc + * blocks on the inode beyond EOF due to speculative + * preallocation. These are not removed until the release + * function is called or the inode is inactivated. Hence we + * cannot assert here that ip->i_delayed_blks == 0. + */ + } + + lock = xfs_ilock_map_shared(ip); + + /* + * Don't let nex be bigger than the number of extents + * we can have assuming alternating holes and real extents. + */ + if (nex > XFS_IFORK_NEXTENTS(ip, whichfork) * 2 + 1) + nex = XFS_IFORK_NEXTENTS(ip, whichfork) * 2 + 1; + + bmapi_flags = xfs_bmapi_aflag(whichfork); + if (!(iflags & BMV_IF_PREALLOC)) + bmapi_flags |= XFS_BMAPI_IGSTATE; + + /* + * Allocate enough space to handle "subnex" maps at a time. + */ + error = ENOMEM; + subnex = 16; + map = kmem_alloc(subnex * sizeof(*map), KM_MAYFAIL | KM_NOFS); + if (!map) + goto out_unlock_ilock; + + bmv->bmv_entries = 0; + + if (XFS_IFORK_NEXTENTS(ip, whichfork) == 0 && + (whichfork == XFS_ATTR_FORK || !(iflags & BMV_IF_DELALLOC))) { + error = 0; + goto out_free_map; + } + + nexleft = nex; + + do { + nmap = (nexleft > subnex) ? subnex : nexleft; + error = xfs_bmapi_read(ip, XFS_BB_TO_FSBT(mp, bmv->bmv_offset), + XFS_BB_TO_FSB(mp, bmv->bmv_length), + map, &nmap, bmapi_flags); + if (error) + goto out_free_map; + ASSERT(nmap <= subnex); + + for (i = 0; i < nmap && nexleft && bmv->bmv_length; i++) { + out[cur_ext].bmv_oflags = 0; + if (map[i].br_state == XFS_EXT_UNWRITTEN) + out[cur_ext].bmv_oflags |= BMV_OF_PREALLOC; + else if (map[i].br_startblock == DELAYSTARTBLOCK) + out[cur_ext].bmv_oflags |= BMV_OF_DELALLOC; + out[cur_ext].bmv_offset = + XFS_FSB_TO_BB(mp, map[i].br_startoff); + out[cur_ext].bmv_length = + XFS_FSB_TO_BB(mp, map[i].br_blockcount); + out[cur_ext].bmv_unused1 = 0; + out[cur_ext].bmv_unused2 = 0; + + /* + * delayed allocation extents that start beyond EOF can + * occur due to speculative EOF allocation when the + * delalloc extent is larger than the largest freespace + * extent at conversion time. These extents cannot be + * converted by data writeback, so can exist here even + * if we are not supposed to be finding delalloc + * extents. + */ + if (map[i].br_startblock == DELAYSTARTBLOCK && + map[i].br_startoff <= XFS_B_TO_FSB(mp, XFS_ISIZE(ip))) + ASSERT((iflags & BMV_IF_DELALLOC) != 0); + + if (map[i].br_startblock == HOLESTARTBLOCK && + whichfork == XFS_ATTR_FORK) { + /* came to the end of attribute fork */ + out[cur_ext].bmv_oflags |= BMV_OF_LAST; + goto out_free_map; + } + + if (!xfs_getbmapx_fix_eof_hole(ip, &out[cur_ext], + prealloced, bmvend, + map[i].br_startblock)) + goto out_free_map; + + bmv->bmv_offset = + out[cur_ext].bmv_offset + + out[cur_ext].bmv_length; + bmv->bmv_length = + max_t(__int64_t, 0, bmvend - bmv->bmv_offset); + + /* + * In case we don't want to return the hole, + * don't increase cur_ext so that we can reuse + * it in the next loop. + */ + if ((iflags & BMV_IF_NO_HOLES) && + map[i].br_startblock == HOLESTARTBLOCK) { + memset(&out[cur_ext], 0, sizeof(out[cur_ext])); + continue; + } + + nexleft--; + bmv->bmv_entries++; + cur_ext++; + } + } while (nmap && nexleft && bmv->bmv_length); + + out_free_map: + kmem_free(map); + out_unlock_ilock: + xfs_iunlock_map_shared(ip, lock); + out_unlock_iolock: + xfs_iunlock(ip, XFS_IOLOCK_SHARED); + + for (i = 0; i < cur_ext; i++) { + int full = 0; /* user array is full */ + + /* format results & advance arg */ + error = formatter(&arg, &out[i], &full); + if (error || full) + break; + } + + if (is_vmalloc_addr(out)) + kmem_free_large(out); + else + kmem_free(out); + return error; +} + +/* + * dead simple method of punching delalyed allocation blocks from a range in + * the inode. Walks a block at a time so will be slow, but is only executed in + * rare error cases so the overhead is not critical. This will always punch out + * both the start and end blocks, even if the ranges only partially overlap + * them, so it is up to the caller to ensure that partial blocks are not + * passed in. + */ +int +xfs_bmap_punch_delalloc_range( + struct xfs_inode *ip, + xfs_fileoff_t start_fsb, + xfs_fileoff_t length) +{ + xfs_fileoff_t remaining = length; + int error = 0; + + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); + + do { + int done; + xfs_bmbt_irec_t imap; + int nimaps = 1; + xfs_fsblock_t firstblock; + xfs_bmap_free_t flist; + + /* + * Map the range first and check that it is a delalloc extent + * before trying to unmap the range. Otherwise we will be + * trying to remove a real extent (which requires a + * transaction) or a hole, which is probably a bad idea... + */ + error = xfs_bmapi_read(ip, start_fsb, 1, &imap, &nimaps, + XFS_BMAPI_ENTIRE); + + if (error) { + /* something screwed, just bail */ + if (!XFS_FORCED_SHUTDOWN(ip->i_mount)) { + xfs_alert(ip->i_mount, + "Failed delalloc mapping lookup ino %lld fsb %lld.", + ip->i_ino, start_fsb); + } + break; + } + if (!nimaps) { + /* nothing there */ + goto next_block; + } + if (imap.br_startblock != DELAYSTARTBLOCK) { + /* been converted, ignore */ + goto next_block; + } + WARN_ON(imap.br_blockcount == 0); + + /* + * Note: while we initialise the firstblock/flist pair, they + * should never be used because blocks should never be + * allocated or freed for a delalloc extent and hence we need + * don't cancel or finish them after the xfs_bunmapi() call. + */ + xfs_bmap_init(&flist, &firstblock); + error = xfs_bunmapi(NULL, ip, start_fsb, 1, 0, 1, &firstblock, + &flist, &done); + if (error) + break; + + ASSERT(!flist.xbf_count && !flist.xbf_first); +next_block: + start_fsb++; + remaining--; + } while(remaining > 0); + + return error; +} + +/* + * Test whether it is appropriate to check an inode for and free post EOF + * blocks. The 'force' parameter determines whether we should also consider + * regular files that are marked preallocated or append-only. + */ +bool +xfs_can_free_eofblocks(struct xfs_inode *ip, bool force) +{ + /* prealloc/delalloc exists only on regular files */ + if (!S_ISREG(ip->i_d.di_mode)) + return false; + + /* + * Zero sized files with no cached pages and delalloc blocks will not + * have speculative prealloc/delalloc blocks to remove. + */ + if (VFS_I(ip)->i_size == 0 && + VN_CACHED(VFS_I(ip)) == 0 && + ip->i_delayed_blks == 0) + return false; + + /* If we haven't read in the extent list, then don't do it now. */ + if (!(ip->i_df.if_flags & XFS_IFEXTENTS)) + return false; + + /* + * Do not free real preallocated or append-only files unless the file + * has delalloc blocks and we are forced to remove them. + */ + if (ip->i_d.di_flags & (XFS_DIFLAG_PREALLOC | XFS_DIFLAG_APPEND)) + if (!force || ip->i_delayed_blks == 0) + return false; + + return true; +} + +/* + * This is called by xfs_inactive to free any blocks beyond eof + * when the link count isn't zero and by xfs_dm_punch_hole() when + * punching a hole to EOF. + */ +int +xfs_free_eofblocks( + xfs_mount_t *mp, + xfs_inode_t *ip, + bool need_iolock) +{ + xfs_trans_t *tp; + int error; + xfs_fileoff_t end_fsb; + xfs_fileoff_t last_fsb; + xfs_filblks_t map_len; + int nimaps; + xfs_bmbt_irec_t imap; + + /* + * Figure out if there are any blocks beyond the end + * of the file. If not, then there is nothing to do. + */ + end_fsb = XFS_B_TO_FSB(mp, (xfs_ufsize_t)XFS_ISIZE(ip)); + last_fsb = XFS_B_TO_FSB(mp, mp->m_super->s_maxbytes); + if (last_fsb <= end_fsb) + return 0; + map_len = last_fsb - end_fsb; + + nimaps = 1; + xfs_ilock(ip, XFS_ILOCK_SHARED); + error = xfs_bmapi_read(ip, end_fsb, map_len, &imap, &nimaps, 0); + xfs_iunlock(ip, XFS_ILOCK_SHARED); + + if (!error && (nimaps != 0) && + (imap.br_startblock != HOLESTARTBLOCK || + ip->i_delayed_blks)) { + /* + * Attach the dquots to the inode up front. + */ + error = xfs_qm_dqattach(ip, 0); + if (error) + return error; + + /* + * There are blocks after the end of file. + * Free them up now by truncating the file to + * its current size. + */ + tp = xfs_trans_alloc(mp, XFS_TRANS_INACTIVE); + + if (need_iolock) { + if (!xfs_ilock_nowait(ip, XFS_IOLOCK_EXCL)) { + xfs_trans_cancel(tp, 0); + return EAGAIN; + } + } + + error = xfs_trans_reserve(tp, &M_RES(mp)->tr_itruncate, 0, 0); + if (error) { + ASSERT(XFS_FORCED_SHUTDOWN(mp)); + xfs_trans_cancel(tp, 0); + if (need_iolock) + xfs_iunlock(ip, XFS_IOLOCK_EXCL); + return error; + } + + xfs_ilock(ip, XFS_ILOCK_EXCL); + xfs_trans_ijoin(tp, ip, 0); + + /* + * Do not update the on-disk file size. If we update the + * on-disk file size and then the system crashes before the + * contents of the file are flushed to disk then the files + * may be full of holes (ie NULL files bug). + */ + error = xfs_itruncate_extents(&tp, ip, XFS_DATA_FORK, + XFS_ISIZE(ip)); + if (error) { + /* + * If we get an error at this point we simply don't + * bother truncating the file. + */ + xfs_trans_cancel(tp, + (XFS_TRANS_RELEASE_LOG_RES | + XFS_TRANS_ABORT)); + } else { + error = xfs_trans_commit(tp, + XFS_TRANS_RELEASE_LOG_RES); + if (!error) + xfs_inode_clear_eofblocks_tag(ip); + } + + xfs_iunlock(ip, XFS_ILOCK_EXCL); + if (need_iolock) + xfs_iunlock(ip, XFS_IOLOCK_EXCL); + } + return error; +} + +/* + * xfs_alloc_file_space() + * This routine allocates disk space for the given file. + * + * If alloc_type == 0, this request is for an ALLOCSP type + * request which will change the file size. In this case, no + * DMAPI event will be generated by the call. A TRUNCATE event + * will be generated later by xfs_setattr. + * + * If alloc_type != 0, this request is for a RESVSP type + * request, and a DMAPI DM_EVENT_WRITE will be generated if the + * lower block boundary byte address is less than the file's + * length. + * + * RETURNS: + * 0 on success + * errno on error + * + */ +STATIC int +xfs_alloc_file_space( + xfs_inode_t *ip, + xfs_off_t offset, + xfs_off_t len, + int alloc_type, + int attr_flags) +{ + xfs_mount_t *mp = ip->i_mount; + xfs_off_t count; + xfs_filblks_t allocated_fsb; + xfs_filblks_t allocatesize_fsb; + xfs_extlen_t extsz, temp; + xfs_fileoff_t startoffset_fsb; + xfs_fsblock_t firstfsb; + int nimaps; + int quota_flag; + int rt; + xfs_trans_t *tp; + xfs_bmbt_irec_t imaps[1], *imapp; + xfs_bmap_free_t free_list; + uint qblocks, resblks, resrtextents; + int committed; + int error; + + trace_xfs_alloc_file_space(ip); + + if (XFS_FORCED_SHUTDOWN(mp)) + return XFS_ERROR(EIO); + + error = xfs_qm_dqattach(ip, 0); + if (error) + return error; + + if (len <= 0) + return XFS_ERROR(EINVAL); + + rt = XFS_IS_REALTIME_INODE(ip); + extsz = xfs_get_extsz_hint(ip); + + count = len; + imapp = &imaps[0]; + nimaps = 1; + startoffset_fsb = XFS_B_TO_FSBT(mp, offset); + allocatesize_fsb = XFS_B_TO_FSB(mp, count); + + /* + * Allocate file space until done or until there is an error + */ + while (allocatesize_fsb && !error) { + xfs_fileoff_t s, e; + + /* + * Determine space reservations for data/realtime. + */ + if (unlikely(extsz)) { + s = startoffset_fsb; + do_div(s, extsz); + s *= extsz; + e = startoffset_fsb + allocatesize_fsb; + if ((temp = do_mod(startoffset_fsb, extsz))) + e += temp; + if ((temp = do_mod(e, extsz))) + e += extsz - temp; + } else { + s = 0; + e = allocatesize_fsb; + } + + /* + * The transaction reservation is limited to a 32-bit block + * count, hence we need to limit the number of blocks we are + * trying to reserve to avoid an overflow. We can't allocate + * more than @nimaps extents, and an extent is limited on disk + * to MAXEXTLEN (21 bits), so use that to enforce the limit. + */ + resblks = min_t(xfs_fileoff_t, (e - s), (MAXEXTLEN * nimaps)); + if (unlikely(rt)) { + resrtextents = qblocks = resblks; + resrtextents /= mp->m_sb.sb_rextsize; + resblks = XFS_DIOSTRAT_SPACE_RES(mp, 0); + quota_flag = XFS_QMOPT_RES_RTBLKS; + } else { + resrtextents = 0; + resblks = qblocks = XFS_DIOSTRAT_SPACE_RES(mp, resblks); + quota_flag = XFS_QMOPT_RES_REGBLKS; + } + + /* + * Allocate and setup the transaction. + */ + tp = xfs_trans_alloc(mp, XFS_TRANS_DIOSTRAT); + error = xfs_trans_reserve(tp, &M_RES(mp)->tr_write, + resblks, resrtextents); + /* + * Check for running out of space + */ + if (error) { + /* + * Free the transaction structure. + */ + ASSERT(error == ENOSPC || XFS_FORCED_SHUTDOWN(mp)); + xfs_trans_cancel(tp, 0); + break; + } + xfs_ilock(ip, XFS_ILOCK_EXCL); + error = xfs_trans_reserve_quota_nblks(tp, ip, qblocks, + 0, quota_flag); + if (error) + goto error1; + + xfs_trans_ijoin(tp, ip, 0); + + xfs_bmap_init(&free_list, &firstfsb); + error = xfs_bmapi_write(tp, ip, startoffset_fsb, + allocatesize_fsb, alloc_type, &firstfsb, + 0, imapp, &nimaps, &free_list); + if (error) { + goto error0; + } + + /* + * Complete the transaction + */ + error = xfs_bmap_finish(&tp, &free_list, &committed); + if (error) { + goto error0; + } + + error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES); + xfs_iunlock(ip, XFS_ILOCK_EXCL); + if (error) { + break; + } + + allocated_fsb = imapp->br_blockcount; + + if (nimaps == 0) { + error = XFS_ERROR(ENOSPC); + break; + } + + startoffset_fsb += allocated_fsb; + allocatesize_fsb -= allocated_fsb; + } + + return error; + +error0: /* Cancel bmap, unlock inode, unreserve quota blocks, cancel trans */ + xfs_bmap_cancel(&free_list); + xfs_trans_unreserve_quota_nblks(tp, ip, (long)qblocks, 0, quota_flag); + +error1: /* Just cancel transaction */ + xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES | XFS_TRANS_ABORT); + xfs_iunlock(ip, XFS_ILOCK_EXCL); + return error; +} + +/* + * Zero file bytes between startoff and endoff inclusive. + * The iolock is held exclusive and no blocks are buffered. + * + * This function is used by xfs_free_file_space() to zero + * partial blocks when the range to free is not block aligned. + * When unreserving space with boundaries that are not block + * aligned we round up the start and round down the end + * boundaries and then use this function to zero the parts of + * the blocks that got dropped during the rounding. + */ +STATIC int +xfs_zero_remaining_bytes( + xfs_inode_t *ip, + xfs_off_t startoff, + xfs_off_t endoff) +{ + xfs_bmbt_irec_t imap; + xfs_fileoff_t offset_fsb; + xfs_off_t lastoffset; + xfs_off_t offset; + xfs_buf_t *bp; + xfs_mount_t *mp = ip->i_mount; + int nimap; + int error = 0; + + /* + * Avoid doing I/O beyond eof - it's not necessary + * since nothing can read beyond eof. The space will + * be zeroed when the file is extended anyway. + */ + if (startoff >= XFS_ISIZE(ip)) + return 0; + + if (endoff > XFS_ISIZE(ip)) + endoff = XFS_ISIZE(ip); + + bp = xfs_buf_get_uncached(XFS_IS_REALTIME_INODE(ip) ? + mp->m_rtdev_targp : mp->m_ddev_targp, + BTOBB(mp->m_sb.sb_blocksize), 0); + if (!bp) + return XFS_ERROR(ENOMEM); + + xfs_buf_unlock(bp); + + for (offset = startoff; offset <= endoff; offset = lastoffset + 1) { + offset_fsb = XFS_B_TO_FSBT(mp, offset); + nimap = 1; + error = xfs_bmapi_read(ip, offset_fsb, 1, &imap, &nimap, 0); + if (error || nimap < 1) + break; + ASSERT(imap.br_blockcount >= 1); + ASSERT(imap.br_startoff == offset_fsb); + lastoffset = XFS_FSB_TO_B(mp, imap.br_startoff + 1) - 1; + if (lastoffset > endoff) + lastoffset = endoff; + if (imap.br_startblock == HOLESTARTBLOCK) + continue; + ASSERT(imap.br_startblock != DELAYSTARTBLOCK); + if (imap.br_state == XFS_EXT_UNWRITTEN) + continue; + XFS_BUF_UNDONE(bp); + XFS_BUF_UNWRITE(bp); + XFS_BUF_READ(bp); + XFS_BUF_SET_ADDR(bp, xfs_fsb_to_db(ip, imap.br_startblock)); + xfsbdstrat(mp, bp); + error = xfs_buf_iowait(bp); + if (error) { + xfs_buf_ioerror_alert(bp, + "xfs_zero_remaining_bytes(read)"); + break; + } + memset(bp->b_addr + + (offset - XFS_FSB_TO_B(mp, imap.br_startoff)), + 0, lastoffset - offset + 1); + XFS_BUF_UNDONE(bp); + XFS_BUF_UNREAD(bp); + XFS_BUF_WRITE(bp); + xfsbdstrat(mp, bp); + error = xfs_buf_iowait(bp); + if (error) { + xfs_buf_ioerror_alert(bp, + "xfs_zero_remaining_bytes(write)"); + break; + } + } + xfs_buf_free(bp); + return error; +} + +/* + * xfs_free_file_space() + * This routine frees disk space for the given file. + * + * This routine is only called by xfs_change_file_space + * for an UNRESVSP type call. + * + * RETURNS: + * 0 on success + * errno on error + * + */ +STATIC int +xfs_free_file_space( + xfs_inode_t *ip, + xfs_off_t offset, + xfs_off_t len, + int attr_flags) +{ + int committed; + int done; + xfs_fileoff_t endoffset_fsb; + int error; + xfs_fsblock_t firstfsb; + xfs_bmap_free_t free_list; + xfs_bmbt_irec_t imap; + xfs_off_t ioffset; + xfs_extlen_t mod=0; + xfs_mount_t *mp; + int nimap; + uint resblks; + xfs_off_t rounding; + int rt; + xfs_fileoff_t startoffset_fsb; + xfs_trans_t *tp; + int need_iolock = 1; + + mp = ip->i_mount; + + trace_xfs_free_file_space(ip); + + error = xfs_qm_dqattach(ip, 0); + if (error) + return error; + + error = 0; + if (len <= 0) /* if nothing being freed */ + return error; + rt = XFS_IS_REALTIME_INODE(ip); + startoffset_fsb = XFS_B_TO_FSB(mp, offset); + endoffset_fsb = XFS_B_TO_FSBT(mp, offset + len); + + if (attr_flags & XFS_ATTR_NOLOCK) + need_iolock = 0; + if (need_iolock) { + xfs_ilock(ip, XFS_IOLOCK_EXCL); + /* wait for the completion of any pending DIOs */ + inode_dio_wait(VFS_I(ip)); + } + + rounding = max_t(xfs_off_t, 1 << mp->m_sb.sb_blocklog, PAGE_CACHE_SIZE); + ioffset = offset & ~(rounding - 1); + error = -filemap_write_and_wait_range(VFS_I(ip)->i_mapping, + ioffset, -1); + if (error) + goto out_unlock_iolock; + truncate_pagecache_range(VFS_I(ip), ioffset, -1); + + /* + * Need to zero the stuff we're not freeing, on disk. + * If it's a realtime file & can't use unwritten extents then we + * actually need to zero the extent edges. Otherwise xfs_bunmapi + * will take care of it for us. + */ + if (rt && !xfs_sb_version_hasextflgbit(&mp->m_sb)) { + nimap = 1; + error = xfs_bmapi_read(ip, startoffset_fsb, 1, + &imap, &nimap, 0); + if (error) + goto out_unlock_iolock; + ASSERT(nimap == 0 || nimap == 1); + if (nimap && imap.br_startblock != HOLESTARTBLOCK) { + xfs_daddr_t block; + + ASSERT(imap.br_startblock != DELAYSTARTBLOCK); + block = imap.br_startblock; + mod = do_div(block, mp->m_sb.sb_rextsize); + if (mod) + startoffset_fsb += mp->m_sb.sb_rextsize - mod; + } + nimap = 1; + error = xfs_bmapi_read(ip, endoffset_fsb - 1, 1, + &imap, &nimap, 0); + if (error) + goto out_unlock_iolock; + ASSERT(nimap == 0 || nimap == 1); + if (nimap && imap.br_startblock != HOLESTARTBLOCK) { + ASSERT(imap.br_startblock != DELAYSTARTBLOCK); + mod++; + if (mod && (mod != mp->m_sb.sb_rextsize)) + endoffset_fsb -= mod; + } + } + if ((done = (endoffset_fsb <= startoffset_fsb))) + /* + * One contiguous piece to clear + */ + error = xfs_zero_remaining_bytes(ip, offset, offset + len - 1); + else { + /* + * Some full blocks, possibly two pieces to clear + */ + if (offset < XFS_FSB_TO_B(mp, startoffset_fsb)) + error = xfs_zero_remaining_bytes(ip, offset, + XFS_FSB_TO_B(mp, startoffset_fsb) - 1); + if (!error && + XFS_FSB_TO_B(mp, endoffset_fsb) < offset + len) + error = xfs_zero_remaining_bytes(ip, + XFS_FSB_TO_B(mp, endoffset_fsb), + offset + len - 1); + } + + /* + * free file space until done or until there is an error + */ + resblks = XFS_DIOSTRAT_SPACE_RES(mp, 0); + while (!error && !done) { + + /* + * allocate and setup the transaction. Allow this + * transaction to dip into the reserve blocks to ensure + * the freeing of the space succeeds at ENOSPC. + */ + tp = xfs_trans_alloc(mp, XFS_TRANS_DIOSTRAT); + tp->t_flags |= XFS_TRANS_RESERVE; + error = xfs_trans_reserve(tp, &M_RES(mp)->tr_write, resblks, 0); + + /* + * check for running out of space + */ + if (error) { + /* + * Free the transaction structure. + */ + ASSERT(error == ENOSPC || XFS_FORCED_SHUTDOWN(mp)); + xfs_trans_cancel(tp, 0); + break; + } + xfs_ilock(ip, XFS_ILOCK_EXCL); + error = xfs_trans_reserve_quota(tp, mp, + ip->i_udquot, ip->i_gdquot, ip->i_pdquot, + resblks, 0, XFS_QMOPT_RES_REGBLKS); + if (error) + goto error1; + + xfs_trans_ijoin(tp, ip, 0); + + /* + * issue the bunmapi() call to free the blocks + */ + xfs_bmap_init(&free_list, &firstfsb); + error = xfs_bunmapi(tp, ip, startoffset_fsb, + endoffset_fsb - startoffset_fsb, + 0, 2, &firstfsb, &free_list, &done); + if (error) { + goto error0; + } + + /* + * complete the transaction + */ + error = xfs_bmap_finish(&tp, &free_list, &committed); + if (error) { + goto error0; + } + + error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES); + xfs_iunlock(ip, XFS_ILOCK_EXCL); + } + + out_unlock_iolock: + if (need_iolock) + xfs_iunlock(ip, XFS_IOLOCK_EXCL); + return error; + + error0: + xfs_bmap_cancel(&free_list); + error1: + xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES | XFS_TRANS_ABORT); + xfs_iunlock(ip, need_iolock ? (XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL) : + XFS_ILOCK_EXCL); + return error; +} + + +STATIC int +xfs_zero_file_space( + struct xfs_inode *ip, + xfs_off_t offset, + xfs_off_t len, + int attr_flags) +{ + struct xfs_mount *mp = ip->i_mount; + uint granularity; + xfs_off_t start_boundary; + xfs_off_t end_boundary; + int error; + + granularity = max_t(uint, 1 << mp->m_sb.sb_blocklog, PAGE_CACHE_SIZE); + + /* + * Round the range of extents we are going to convert inwards. If the + * offset is aligned, then it doesn't get changed so we zero from the + * start of the block offset points to. + */ + start_boundary = round_up(offset, granularity); + end_boundary = round_down(offset + len, granularity); + + ASSERT(start_boundary >= offset); + ASSERT(end_boundary <= offset + len); + + if (!(attr_flags & XFS_ATTR_NOLOCK)) + xfs_ilock(ip, XFS_IOLOCK_EXCL); + + if (start_boundary < end_boundary - 1) { + /* punch out the page cache over the conversion range */ + truncate_pagecache_range(VFS_I(ip), start_boundary, + end_boundary - 1); + /* convert the blocks */ + error = xfs_alloc_file_space(ip, start_boundary, + end_boundary - start_boundary - 1, + XFS_BMAPI_PREALLOC | XFS_BMAPI_CONVERT, + attr_flags); + if (error) + goto out_unlock; + + /* We've handled the interior of the range, now for the edges */ + if (start_boundary != offset) + error = xfs_iozero(ip, offset, start_boundary - offset); + if (error) + goto out_unlock; + + if (end_boundary != offset + len) + error = xfs_iozero(ip, end_boundary, + offset + len - end_boundary); + + } else { + /* + * It's either a sub-granularity range or the range spanned lies + * partially across two adjacent blocks. + */ + error = xfs_iozero(ip, offset, len); + } + +out_unlock: + if (!(attr_flags & XFS_ATTR_NOLOCK)) + xfs_iunlock(ip, XFS_IOLOCK_EXCL); + return error; + +} + +/* + * xfs_change_file_space() + * This routine allocates or frees disk space for the given file. + * The user specified parameters are checked for alignment and size + * limitations. + * + * RETURNS: + * 0 on success + * errno on error + * + */ +int +xfs_change_file_space( + xfs_inode_t *ip, + int cmd, + xfs_flock64_t *bf, + xfs_off_t offset, + int attr_flags) +{ + xfs_mount_t *mp = ip->i_mount; + int clrprealloc; + int error; + xfs_fsize_t fsize; + int setprealloc; + xfs_off_t startoffset; + xfs_trans_t *tp; + struct iattr iattr; + + if (!S_ISREG(ip->i_d.di_mode)) + return XFS_ERROR(EINVAL); + + switch (bf->l_whence) { + case 0: /*SEEK_SET*/ + break; + case 1: /*SEEK_CUR*/ + bf->l_start += offset; + break; + case 2: /*SEEK_END*/ + bf->l_start += XFS_ISIZE(ip); + break; + default: + return XFS_ERROR(EINVAL); + } + + /* + * length of <= 0 for resv/unresv/zero is invalid. length for + * alloc/free is ignored completely and we have no idea what userspace + * might have set it to, so set it to zero to allow range + * checks to pass. + */ + switch (cmd) { + case XFS_IOC_ZERO_RANGE: + case XFS_IOC_RESVSP: + case XFS_IOC_RESVSP64: + case XFS_IOC_UNRESVSP: + case XFS_IOC_UNRESVSP64: + if (bf->l_len <= 0) + return XFS_ERROR(EINVAL); + break; + default: + bf->l_len = 0; + break; + } + + if (bf->l_start < 0 || + bf->l_start > mp->m_super->s_maxbytes || + bf->l_start + bf->l_len < 0 || + bf->l_start + bf->l_len >= mp->m_super->s_maxbytes) + return XFS_ERROR(EINVAL); + + bf->l_whence = 0; + + startoffset = bf->l_start; + fsize = XFS_ISIZE(ip); + + setprealloc = clrprealloc = 0; + switch (cmd) { + case XFS_IOC_ZERO_RANGE: + error = xfs_zero_file_space(ip, startoffset, bf->l_len, + attr_flags); + if (error) + return error; + setprealloc = 1; + break; + + case XFS_IOC_RESVSP: + case XFS_IOC_RESVSP64: + error = xfs_alloc_file_space(ip, startoffset, bf->l_len, + XFS_BMAPI_PREALLOC, attr_flags); + if (error) + return error; + setprealloc = 1; + break; + + case XFS_IOC_UNRESVSP: + case XFS_IOC_UNRESVSP64: + if ((error = xfs_free_file_space(ip, startoffset, bf->l_len, + attr_flags))) + return error; + break; + + case XFS_IOC_ALLOCSP: + case XFS_IOC_ALLOCSP64: + case XFS_IOC_FREESP: + case XFS_IOC_FREESP64: + /* + * These operations actually do IO when extending the file, but + * the allocation is done seperately to the zeroing that is + * done. This set of operations need to be serialised against + * other IO operations, such as truncate and buffered IO. We + * need to take the IOLOCK here to serialise the allocation and + * zeroing IO to prevent other IOLOCK holders (e.g. getbmap, + * truncate, direct IO) from racing against the transient + * allocated but not written state we can have here. + */ + xfs_ilock(ip, XFS_IOLOCK_EXCL); + if (startoffset > fsize) { + error = xfs_alloc_file_space(ip, fsize, + startoffset - fsize, 0, + attr_flags | XFS_ATTR_NOLOCK); + if (error) { + xfs_iunlock(ip, XFS_IOLOCK_EXCL); + break; + } + } + + iattr.ia_valid = ATTR_SIZE; + iattr.ia_size = startoffset; + + error = xfs_setattr_size(ip, &iattr, + attr_flags | XFS_ATTR_NOLOCK); + xfs_iunlock(ip, XFS_IOLOCK_EXCL); + + if (error) + return error; + + clrprealloc = 1; + break; + + default: + ASSERT(0); + return XFS_ERROR(EINVAL); + } + + /* + * update the inode timestamp, mode, and prealloc flag bits + */ + tp = xfs_trans_alloc(mp, XFS_TRANS_WRITEID); + error = xfs_trans_reserve(tp, &M_RES(mp)->tr_writeid, 0, 0); + if (error) { + xfs_trans_cancel(tp, 0); + return error; + } + + xfs_ilock(ip, XFS_ILOCK_EXCL); + xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL); + + if ((attr_flags & XFS_ATTR_DMI) == 0) { + ip->i_d.di_mode &= ~S_ISUID; + + /* + * Note that we don't have to worry about mandatory + * file locking being disabled here because we only + * clear the S_ISGID bit if the Group execute bit is + * on, but if it was on then mandatory locking wouldn't + * have been enabled. + */ + if (ip->i_d.di_mode & S_IXGRP) + ip->i_d.di_mode &= ~S_ISGID; + + xfs_trans_ichgtime(tp, ip, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG); + } + if (setprealloc) + ip->i_d.di_flags |= XFS_DIFLAG_PREALLOC; + else if (clrprealloc) + ip->i_d.di_flags &= ~XFS_DIFLAG_PREALLOC; + + xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); + if (attr_flags & XFS_ATTR_SYNC) + xfs_trans_set_sync(tp); + return xfs_trans_commit(tp, 0); +} + +/* + * We need to check that the format of the data fork in the temporary inode is + * valid for the target inode before doing the swap. This is not a problem with + * attr1 because of the fixed fork offset, but attr2 has a dynamically sized + * data fork depending on the space the attribute fork is taking so we can get + * invalid formats on the target inode. + * + * E.g. target has space for 7 extents in extent format, temp inode only has + * space for 6. If we defragment down to 7 extents, then the tmp format is a + * btree, but when swapped it needs to be in extent format. Hence we can't just + * blindly swap data forks on attr2 filesystems. + * + * Note that we check the swap in both directions so that we don't end up with + * a corrupt temporary inode, either. + * + * Note that fixing the way xfs_fsr sets up the attribute fork in the source + * inode will prevent this situation from occurring, so all we do here is + * reject and log the attempt. basically we are putting the responsibility on + * userspace to get this right. + */ +static int +xfs_swap_extents_check_format( + xfs_inode_t *ip, /* target inode */ + xfs_inode_t *tip) /* tmp inode */ +{ + + /* Should never get a local format */ + if (ip->i_d.di_format == XFS_DINODE_FMT_LOCAL || + tip->i_d.di_format == XFS_DINODE_FMT_LOCAL) + return EINVAL; + + /* + * if the target inode has less extents that then temporary inode then + * why did userspace call us? + */ + if (ip->i_d.di_nextents < tip->i_d.di_nextents) + return EINVAL; + + /* + * if the target inode is in extent form and the temp inode is in btree + * form then we will end up with the target inode in the wrong format + * as we already know there are less extents in the temp inode. + */ + if (ip->i_d.di_format == XFS_DINODE_FMT_EXTENTS && + tip->i_d.di_format == XFS_DINODE_FMT_BTREE) + return EINVAL; + + /* Check temp in extent form to max in target */ + if (tip->i_d.di_format == XFS_DINODE_FMT_EXTENTS && + XFS_IFORK_NEXTENTS(tip, XFS_DATA_FORK) > + XFS_IFORK_MAXEXT(ip, XFS_DATA_FORK)) + return EINVAL; + + /* Check target in extent form to max in temp */ + if (ip->i_d.di_format == XFS_DINODE_FMT_EXTENTS && + XFS_IFORK_NEXTENTS(ip, XFS_DATA_FORK) > + XFS_IFORK_MAXEXT(tip, XFS_DATA_FORK)) + return EINVAL; + + /* + * If we are in a btree format, check that the temp root block will fit + * in the target and that it has enough extents to be in btree format + * in the target. + * + * Note that we have to be careful to allow btree->extent conversions + * (a common defrag case) which will occur when the temp inode is in + * extent format... + */ + if (tip->i_d.di_format == XFS_DINODE_FMT_BTREE) { + if (XFS_IFORK_BOFF(ip) && + XFS_BMAP_BMDR_SPACE(tip->i_df.if_broot) > XFS_IFORK_BOFF(ip)) + return EINVAL; + if (XFS_IFORK_NEXTENTS(tip, XFS_DATA_FORK) <= + XFS_IFORK_MAXEXT(ip, XFS_DATA_FORK)) + return EINVAL; + } + + /* Reciprocal target->temp btree format checks */ + if (ip->i_d.di_format == XFS_DINODE_FMT_BTREE) { + if (XFS_IFORK_BOFF(tip) && + XFS_BMAP_BMDR_SPACE(ip->i_df.if_broot) > XFS_IFORK_BOFF(tip)) + return EINVAL; + if (XFS_IFORK_NEXTENTS(ip, XFS_DATA_FORK) <= + XFS_IFORK_MAXEXT(tip, XFS_DATA_FORK)) + return EINVAL; + } + + return 0; +} + +int +xfs_swap_extents( + xfs_inode_t *ip, /* target inode */ + xfs_inode_t *tip, /* tmp inode */ + xfs_swapext_t *sxp) +{ + xfs_mount_t *mp = ip->i_mount; + xfs_trans_t *tp; + xfs_bstat_t *sbp = &sxp->sx_stat; + xfs_ifork_t *tempifp, *ifp, *tifp; + int src_log_flags, target_log_flags; + int error = 0; + int aforkblks = 0; + int taforkblks = 0; + __uint64_t tmp; + + /* + * We have no way of updating owner information in the BMBT blocks for + * each inode on CRC enabled filesystems, so to avoid corrupting the + * this metadata we simply don't allow extent swaps to occur. + */ + if (xfs_sb_version_hascrc(&mp->m_sb)) + return XFS_ERROR(EINVAL); + + tempifp = kmem_alloc(sizeof(xfs_ifork_t), KM_MAYFAIL); + if (!tempifp) { + error = XFS_ERROR(ENOMEM); + goto out; + } + + /* + * we have to do two separate lock calls here to keep lockdep + * happy. If we try to get all the locks in one call, lock will + * report false positives when we drop the ILOCK and regain them + * below. + */ + xfs_lock_two_inodes(ip, tip, XFS_IOLOCK_EXCL); + xfs_lock_two_inodes(ip, tip, XFS_ILOCK_EXCL); + + /* Verify that both files have the same format */ + if ((ip->i_d.di_mode & S_IFMT) != (tip->i_d.di_mode & S_IFMT)) { + error = XFS_ERROR(EINVAL); + goto out_unlock; + } + + /* Verify both files are either real-time or non-realtime */ + if (XFS_IS_REALTIME_INODE(ip) != XFS_IS_REALTIME_INODE(tip)) { + error = XFS_ERROR(EINVAL); + goto out_unlock; + } + + error = -filemap_write_and_wait(VFS_I(tip)->i_mapping); + if (error) + goto out_unlock; + truncate_pagecache_range(VFS_I(tip), 0, -1); + + /* Verify O_DIRECT for ftmp */ + if (VN_CACHED(VFS_I(tip)) != 0) { + error = XFS_ERROR(EINVAL); + goto out_unlock; + } + + /* Verify all data are being swapped */ + if (sxp->sx_offset != 0 || + sxp->sx_length != ip->i_d.di_size || + sxp->sx_length != tip->i_d.di_size) { + error = XFS_ERROR(EFAULT); + goto out_unlock; + } + + trace_xfs_swap_extent_before(ip, 0); + trace_xfs_swap_extent_before(tip, 1); + + /* check inode formats now that data is flushed */ + error = xfs_swap_extents_check_format(ip, tip); + if (error) { + xfs_notice(mp, + "%s: inode 0x%llx format is incompatible for exchanging.", + __func__, ip->i_ino); + goto out_unlock; + } + + /* + * Compare the current change & modify times with that + * passed in. If they differ, we abort this swap. + * This is the mechanism used to ensure the calling + * process that the file was not changed out from + * under it. + */ + if ((sbp->bs_ctime.tv_sec != VFS_I(ip)->i_ctime.tv_sec) || + (sbp->bs_ctime.tv_nsec != VFS_I(ip)->i_ctime.tv_nsec) || + (sbp->bs_mtime.tv_sec != VFS_I(ip)->i_mtime.tv_sec) || + (sbp->bs_mtime.tv_nsec != VFS_I(ip)->i_mtime.tv_nsec)) { + error = XFS_ERROR(EBUSY); + goto out_unlock; + } + + /* We need to fail if the file is memory mapped. Once we have tossed + * all existing pages, the page fault will have no option + * but to go to the filesystem for pages. By making the page fault call + * vop_read (or write in the case of autogrow) they block on the iolock + * until we have switched the extents. + */ + if (VN_MAPPED(VFS_I(ip))) { + error = XFS_ERROR(EBUSY); + goto out_unlock; + } + + xfs_iunlock(ip, XFS_ILOCK_EXCL); + xfs_iunlock(tip, XFS_ILOCK_EXCL); + + /* + * There is a race condition here since we gave up the + * ilock. However, the data fork will not change since + * we have the iolock (locked for truncation too) so we + * are safe. We don't really care if non-io related + * fields change. + */ + truncate_pagecache_range(VFS_I(ip), 0, -1); + + tp = xfs_trans_alloc(mp, XFS_TRANS_SWAPEXT); + error = xfs_trans_reserve(tp, &M_RES(mp)->tr_ichange, 0, 0); + if (error) { + xfs_iunlock(ip, XFS_IOLOCK_EXCL); + xfs_iunlock(tip, XFS_IOLOCK_EXCL); + xfs_trans_cancel(tp, 0); + goto out; + } + xfs_lock_two_inodes(ip, tip, XFS_ILOCK_EXCL); + + /* + * Count the number of extended attribute blocks + */ + if ( ((XFS_IFORK_Q(ip) != 0) && (ip->i_d.di_anextents > 0)) && + (ip->i_d.di_aformat != XFS_DINODE_FMT_LOCAL)) { + error = xfs_bmap_count_blocks(tp, ip, XFS_ATTR_FORK, &aforkblks); + if (error) + goto out_trans_cancel; + } + if ( ((XFS_IFORK_Q(tip) != 0) && (tip->i_d.di_anextents > 0)) && + (tip->i_d.di_aformat != XFS_DINODE_FMT_LOCAL)) { + error = xfs_bmap_count_blocks(tp, tip, XFS_ATTR_FORK, + &taforkblks); + if (error) + goto out_trans_cancel; + } + + /* + * Swap the data forks of the inodes + */ + ifp = &ip->i_df; + tifp = &tip->i_df; + *tempifp = *ifp; /* struct copy */ + *ifp = *tifp; /* struct copy */ + *tifp = *tempifp; /* struct copy */ + + /* + * Fix the on-disk inode values + */ + tmp = (__uint64_t)ip->i_d.di_nblocks; + ip->i_d.di_nblocks = tip->i_d.di_nblocks - taforkblks + aforkblks; + tip->i_d.di_nblocks = tmp + taforkblks - aforkblks; + + tmp = (__uint64_t) ip->i_d.di_nextents; + ip->i_d.di_nextents = tip->i_d.di_nextents; + tip->i_d.di_nextents = tmp; + + tmp = (__uint64_t) ip->i_d.di_format; + ip->i_d.di_format = tip->i_d.di_format; + tip->i_d.di_format = tmp; + + /* + * The extents in the source inode could still contain speculative + * preallocation beyond EOF (e.g. the file is open but not modified + * while defrag is in progress). In that case, we need to copy over the + * number of delalloc blocks the data fork in the source inode is + * tracking beyond EOF so that when the fork is truncated away when the + * temporary inode is unlinked we don't underrun the i_delayed_blks + * counter on that inode. + */ + ASSERT(tip->i_delayed_blks == 0); + tip->i_delayed_blks = ip->i_delayed_blks; + ip->i_delayed_blks = 0; + + src_log_flags = XFS_ILOG_CORE; + switch (ip->i_d.di_format) { + case XFS_DINODE_FMT_EXTENTS: + /* If the extents fit in the inode, fix the + * pointer. Otherwise it's already NULL or + * pointing to the extent. + */ + if (ip->i_d.di_nextents <= XFS_INLINE_EXTS) { + ifp->if_u1.if_extents = + ifp->if_u2.if_inline_ext; + } + src_log_flags |= XFS_ILOG_DEXT; + break; + case XFS_DINODE_FMT_BTREE: + src_log_flags |= XFS_ILOG_DBROOT; + break; + } + + target_log_flags = XFS_ILOG_CORE; + switch (tip->i_d.di_format) { + case XFS_DINODE_FMT_EXTENTS: + /* If the extents fit in the inode, fix the + * pointer. Otherwise it's already NULL or + * pointing to the extent. + */ + if (tip->i_d.di_nextents <= XFS_INLINE_EXTS) { + tifp->if_u1.if_extents = + tifp->if_u2.if_inline_ext; + } + target_log_flags |= XFS_ILOG_DEXT; + break; + case XFS_DINODE_FMT_BTREE: + target_log_flags |= XFS_ILOG_DBROOT; + break; + } + + + xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL); + xfs_trans_ijoin(tp, tip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL); + + xfs_trans_log_inode(tp, ip, src_log_flags); + xfs_trans_log_inode(tp, tip, target_log_flags); + + /* + * If this is a synchronous mount, make sure that the + * transaction goes to disk before returning to the user. + */ + if (mp->m_flags & XFS_MOUNT_WSYNC) + xfs_trans_set_sync(tp); + + error = xfs_trans_commit(tp, 0); + + trace_xfs_swap_extent_after(ip, 0); + trace_xfs_swap_extent_after(tip, 1); +out: + kmem_free(tempifp); + return error; + +out_unlock: + xfs_iunlock(ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL); + xfs_iunlock(tip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL); + goto out; + +out_trans_cancel: + xfs_trans_cancel(tp, 0); + goto out_unlock; +} diff --git a/fs/xfs/xfs_bmap_util.h b/fs/xfs/xfs_bmap_util.h new file mode 100644 index 0000000..0612609 --- /dev/null +++ b/fs/xfs/xfs_bmap_util.h @@ -0,0 +1,110 @@ +/* + * Copyright (c) 2000-2006 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#ifndef __XFS_BMAP_UTIL_H__ +#define __XFS_BMAP_UTIL_H__ + +/* Kernel only BMAP related definitions and functions */ + +struct xfs_bmbt_irec; +struct xfs_bmap_free_item; +struct xfs_ifork; +struct xfs_inode; +struct xfs_mount; +struct xfs_trans; + +/* + * Argument structure for xfs_bmap_alloc. + */ +struct xfs_bmalloca { + xfs_fsblock_t *firstblock; /* i/o first block allocated */ + struct xfs_bmap_free *flist; /* bmap freelist */ + struct xfs_trans *tp; /* transaction pointer */ + struct xfs_inode *ip; /* incore inode pointer */ + struct xfs_bmbt_irec prev; /* extent before the new one */ + struct xfs_bmbt_irec got; /* extent after, or delayed */ + + xfs_fileoff_t offset; /* offset in file filling in */ + xfs_extlen_t length; /* i/o length asked/allocated */ + xfs_fsblock_t blkno; /* starting block of new extent */ + + struct xfs_btree_cur *cur; /* btree cursor */ + xfs_extnum_t idx; /* current extent index */ + int nallocs;/* number of extents alloc'd */ + int logflags;/* flags for transaction logging */ + + xfs_extlen_t total; /* total blocks needed for xaction */ + xfs_extlen_t minlen; /* minimum allocation size (blocks) */ + xfs_extlen_t minleft; /* amount must be left after alloc */ + char eof; /* set if allocating past last extent */ + char wasdel; /* replacing a delayed allocation */ + char userdata;/* set if is user data */ + char aeof; /* allocated space at eof */ + char conv; /* overwriting unwritten extents */ + char stack_switch; + int flags; + struct completion *done; + struct work_struct work; + int result; +}; + +int xfs_bmap_finish(struct xfs_trans **tp, struct xfs_bmap_free *flist, + int *committed); +int xfs_bmap_rtalloc(struct xfs_bmalloca *ap); +int xfs_bmapi_allocate(struct xfs_bmalloca *args); +int __xfs_bmapi_allocate(struct xfs_bmalloca *args); +int xfs_bmap_eof(struct xfs_inode *ip, xfs_fileoff_t endoff, + int whichfork, int *eof); +int xfs_bmap_count_blocks(struct xfs_trans *tp, struct xfs_inode *ip, + int whichfork, int *count); +int xfs_bmap_punch_delalloc_range(struct xfs_inode *ip, + xfs_fileoff_t start_fsb, xfs_fileoff_t length); + +/* bmap to userspace formatter - copy to user & advance pointer */ +typedef int (*xfs_bmap_format_t)(void **, struct getbmapx *, int *); +int xfs_getbmap(struct xfs_inode *ip, struct getbmapx *bmv, + xfs_bmap_format_t formatter, void *arg); + +/* functions in xfs_bmap.c that are only needed by xfs_bmap_util.c */ +void xfs_bmap_del_free(struct xfs_bmap_free *flist, + struct xfs_bmap_free_item *prev, + struct xfs_bmap_free_item *free); +int xfs_bmap_extsize_align(struct xfs_mount *mp, struct xfs_bmbt_irec *gotp, + struct xfs_bmbt_irec *prevp, xfs_extlen_t extsz, + int rt, int eof, int delay, int convert, + xfs_fileoff_t *offp, xfs_extlen_t *lenp); +void xfs_bmap_adjacent(struct xfs_bmalloca *ap); +int xfs_bmap_last_extent(struct xfs_trans *tp, struct xfs_inode *ip, + int whichfork, struct xfs_bmbt_irec *rec, + int *is_empty); + +/* preallocation and hole punch interface */ +int xfs_change_file_space(struct xfs_inode *ip, int cmd, + xfs_flock64_t *bf, xfs_off_t offset, + int attr_flags); + +/* EOF block manipulation functions */ +bool xfs_can_free_eofblocks(struct xfs_inode *ip, bool force); +int xfs_free_eofblocks(struct xfs_mount *mp, struct xfs_inode *ip, + bool need_iolock); + +int xfs_swap_extents(struct xfs_inode *ip, struct xfs_inode *tip, + struct xfs_swapext *sx); + +xfs_daddr_t xfs_fsb_to_db(struct xfs_inode *ip, xfs_fsblock_t fsb); + +#endif /* __XFS_BMAP_UTIL_H__ */ diff --git a/fs/xfs/xfs_btree.c b/fs/xfs/xfs_btree.c index 0903960..7a2b4da 100644 --- a/fs/xfs/xfs_btree.c +++ b/fs/xfs/xfs_btree.c @@ -510,7 +510,7 @@ xfs_btree_ptr_addr( } /* - * Get a the root block which is stored in the inode. + * Get the root block which is stored in the inode. * * For now this btree implementation assumes the btree root is always * stored in the if_broot field of an inode fork. @@ -978,6 +978,7 @@ xfs_btree_init_block_int( buf->bb_u.l.bb_owner = cpu_to_be64(owner); uuid_copy(&buf->bb_u.l.bb_uuid, &mp->m_sb.sb_uuid); buf->bb_u.l.bb_pad = 0; + buf->bb_u.l.bb_lsn = 0; } } else { /* owner is a 32 bit value on short blocks */ @@ -989,6 +990,7 @@ xfs_btree_init_block_int( buf->bb_u.s.bb_blkno = cpu_to_be64(blkno); buf->bb_u.s.bb_owner = cpu_to_be32(__owner); uuid_copy(&buf->bb_u.s.bb_uuid, &mp->m_sb.sb_uuid); + buf->bb_u.s.bb_lsn = 0; } } } @@ -1684,7 +1686,7 @@ xfs_lookup_get_search_key( /* * Lookup the record. The cursor is made to point to it, based on dir. - * Return 0 if can't find any such record, 1 for success. + * stat is set to 0 if can't find any such record, 1 for success. */ int /* error */ xfs_btree_lookup( @@ -2756,7 +2758,6 @@ xfs_btree_make_block_unfull( if (numrecs < cur->bc_ops->get_dmaxrecs(cur, level)) { /* A root block that can be made bigger. */ - xfs_iroot_realloc(ip, 1, cur->bc_private.b.whichfork); } else { /* A root block that needs replacing */ diff --git a/fs/xfs/xfs_btree.h b/fs/xfs/xfs_btree.h index 55e3c7c..c8473c7 100644 --- a/fs/xfs/xfs_btree.h +++ b/fs/xfs/xfs_btree.h @@ -88,13 +88,11 @@ struct xfs_btree_block { #define XFS_BTREE_SBLOCK_CRC_LEN (XFS_BTREE_SBLOCK_LEN + 40) #define XFS_BTREE_LBLOCK_CRC_LEN (XFS_BTREE_LBLOCK_LEN + 48) - #define XFS_BTREE_SBLOCK_CRC_OFF \ offsetof(struct xfs_btree_block, bb_u.s.bb_crc) #define XFS_BTREE_LBLOCK_CRC_OFF \ offsetof(struct xfs_btree_block, bb_u.l.bb_crc) - /* * Generic key, ptr and record wrapper structures. * diff --git a/fs/xfs/xfs_buf.c b/fs/xfs/xfs_buf.c index 1b2472a..c06823f 100644 --- a/fs/xfs/xfs_buf.c +++ b/fs/xfs/xfs_buf.c @@ -35,6 +35,7 @@ #include <linux/freezer.h> #include "xfs_sb.h" +#include "xfs_trans_resv.h" #include "xfs_log.h" #include "xfs_ag.h" #include "xfs_mount.h" @@ -303,7 +304,7 @@ _xfs_buf_free_pages( * Releases the specified buffer. * * The modification state of any associated pages is left unchanged. - * The buffer most not be on any hash - use xfs_buf_rele instead for + * The buffer must not be on any hash - use xfs_buf_rele instead for * hashed and refcounted buffers */ void @@ -1621,7 +1622,7 @@ xfs_setsize_buftarg_flags( /* * When allocating the initial buffer target we have not yet * read in the superblock, so don't know what sized sectors - * are being used is at this early stage. Play safe. + * are being used at this early stage. Play safe. */ STATIC int xfs_setsize_buftarg_early( diff --git a/fs/xfs/xfs_buf_item.c b/fs/xfs/xfs_buf_item.c index bfc4e0c..3a944b1 100644 --- a/fs/xfs/xfs_buf_item.c +++ b/fs/xfs/xfs_buf_item.c @@ -39,6 +39,14 @@ static inline struct xfs_buf_log_item *BUF_ITEM(struct xfs_log_item *lip) STATIC void xfs_buf_do_callbacks(struct xfs_buf *bp); +static inline int +xfs_buf_log_format_size( + struct xfs_buf_log_format *blfp) +{ + return offsetof(struct xfs_buf_log_format, blf_data_map) + + (blfp->blf_map_size * sizeof(blfp->blf_data_map[0])); +} + /* * This returns the number of log iovecs needed to log the * given buf log item. @@ -49,25 +57,27 @@ STATIC void xfs_buf_do_callbacks(struct xfs_buf *bp); * * If the XFS_BLI_STALE flag has been set, then log nothing. */ -STATIC uint +STATIC void xfs_buf_item_size_segment( struct xfs_buf_log_item *bip, - struct xfs_buf_log_format *blfp) + struct xfs_buf_log_format *blfp, + int *nvecs, + int *nbytes) { struct xfs_buf *bp = bip->bli_buf; - uint nvecs; int next_bit; int last_bit; last_bit = xfs_next_bit(blfp->blf_data_map, blfp->blf_map_size, 0); if (last_bit == -1) - return 0; + return; /* * initial count for a dirty buffer is 2 vectors - the format structure * and the first dirty region. */ - nvecs = 2; + *nvecs += 2; + *nbytes += xfs_buf_log_format_size(blfp) + XFS_BLF_CHUNK; while (last_bit != -1) { /* @@ -87,18 +97,17 @@ xfs_buf_item_size_segment( break; } else if (next_bit != last_bit + 1) { last_bit = next_bit; - nvecs++; + (*nvecs)++; } else if (xfs_buf_offset(bp, next_bit * XFS_BLF_CHUNK) != (xfs_buf_offset(bp, last_bit * XFS_BLF_CHUNK) + XFS_BLF_CHUNK)) { last_bit = next_bit; - nvecs++; + (*nvecs)++; } else { last_bit++; } + *nbytes += XFS_BLF_CHUNK; } - - return nvecs; } /* @@ -118,12 +127,13 @@ xfs_buf_item_size_segment( * If the XFS_BLI_STALE flag has been set, then log nothing but the buf log * format structures. */ -STATIC uint +STATIC void xfs_buf_item_size( - struct xfs_log_item *lip) + struct xfs_log_item *lip, + int *nvecs, + int *nbytes) { struct xfs_buf_log_item *bip = BUF_ITEM(lip); - uint nvecs; int i; ASSERT(atomic_read(&bip->bli_refcount) > 0); @@ -135,7 +145,11 @@ xfs_buf_item_size( */ trace_xfs_buf_item_size_stale(bip); ASSERT(bip->__bli_format.blf_flags & XFS_BLF_CANCEL); - return bip->bli_format_count; + *nvecs += bip->bli_format_count; + for (i = 0; i < bip->bli_format_count; i++) { + *nbytes += xfs_buf_log_format_size(&bip->bli_formats[i]); + } + return; } ASSERT(bip->bli_flags & XFS_BLI_LOGGED); @@ -147,7 +161,8 @@ xfs_buf_item_size( * commit, so no vectors are used at all. */ trace_xfs_buf_item_size_ordered(bip); - return XFS_LOG_VEC_ORDERED; + *nvecs = XFS_LOG_VEC_ORDERED; + return; } /* @@ -159,13 +174,11 @@ xfs_buf_item_size( * count for the extra buf log format structure that will need to be * written. */ - nvecs = 0; for (i = 0; i < bip->bli_format_count; i++) { - nvecs += xfs_buf_item_size_segment(bip, &bip->bli_formats[i]); + xfs_buf_item_size_segment(bip, &bip->bli_formats[i], + nvecs, nbytes); } - trace_xfs_buf_item_size(bip); - return nvecs; } static struct xfs_log_iovec * @@ -192,8 +205,7 @@ xfs_buf_item_format_segment( * the actual size of the dirty bitmap rather than the size of the in * memory structure. */ - base_size = offsetof(struct xfs_buf_log_format, blf_data_map) + - (blfp->blf_map_size * sizeof(blfp->blf_data_map[0])); + base_size = xfs_buf_log_format_size(blfp); nvecs = 0; first_bit = xfs_next_bit(blfp->blf_data_map, blfp->blf_map_size, 0); @@ -601,11 +613,9 @@ xfs_buf_item_unlock( } } } - if (clean) - xfs_buf_item_relse(bp); - else if (aborted) { + if (clean || aborted) { if (atomic_dec_and_test(&bip->bli_refcount)) { - ASSERT(XFS_FORCED_SHUTDOWN(lip->li_mountp)); + ASSERT(!aborted || XFS_FORCED_SHUTDOWN(lip->li_mountp)); xfs_buf_item_relse(bp); } } else diff --git a/fs/xfs/xfs_buf_item.h b/fs/xfs/xfs_buf_item.h index 0f1c247..db63710 100644 --- a/fs/xfs/xfs_buf_item.h +++ b/fs/xfs/xfs_buf_item.h @@ -18,101 +18,9 @@ #ifndef __XFS_BUF_ITEM_H__ #define __XFS_BUF_ITEM_H__ -extern kmem_zone_t *xfs_buf_item_zone; - -/* - * This flag indicates that the buffer contains on disk inodes - * and requires special recovery handling. - */ -#define XFS_BLF_INODE_BUF (1<<0) -/* - * This flag indicates that the buffer should not be replayed - * during recovery because its blocks are being freed. - */ -#define XFS_BLF_CANCEL (1<<1) - -/* - * This flag indicates that the buffer contains on disk - * user or group dquots and may require special recovery handling. - */ -#define XFS_BLF_UDQUOT_BUF (1<<2) -#define XFS_BLF_PDQUOT_BUF (1<<3) -#define XFS_BLF_GDQUOT_BUF (1<<4) - -#define XFS_BLF_CHUNK 128 -#define XFS_BLF_SHIFT 7 -#define BIT_TO_WORD_SHIFT 5 -#define NBWORD (NBBY * sizeof(unsigned int)) - -/* - * This is the structure used to lay out a buf log item in the - * log. The data map describes which 128 byte chunks of the buffer - * have been logged. - */ -#define XFS_BLF_DATAMAP_SIZE ((XFS_MAX_BLOCKSIZE / XFS_BLF_CHUNK) / NBWORD) +/* kernel only definitions */ -typedef struct xfs_buf_log_format { - unsigned short blf_type; /* buf log item type indicator */ - unsigned short blf_size; /* size of this item */ - ushort blf_flags; /* misc state */ - ushort blf_len; /* number of blocks in this buf */ - __int64_t blf_blkno; /* starting blkno of this buf */ - unsigned int blf_map_size; /* used size of data bitmap in words */ - unsigned int blf_data_map[XFS_BLF_DATAMAP_SIZE]; /* dirty bitmap */ -} xfs_buf_log_format_t; - -/* - * All buffers now need to tell recovery where the magic number - * is so that it can verify and calculate the CRCs on the buffer correctly - * once the changes have been replayed into the buffer. - * - * The type value is held in the upper 5 bits of the blf_flags field, which is - * an unsigned 16 bit field. Hence we need to shift it 11 bits up and down. - */ -#define XFS_BLFT_BITS 5 -#define XFS_BLFT_SHIFT 11 -#define XFS_BLFT_MASK (((1 << XFS_BLFT_BITS) - 1) << XFS_BLFT_SHIFT) - -enum xfs_blft { - XFS_BLFT_UNKNOWN_BUF = 0, - XFS_BLFT_UDQUOT_BUF, - XFS_BLFT_PDQUOT_BUF, - XFS_BLFT_GDQUOT_BUF, - XFS_BLFT_BTREE_BUF, - XFS_BLFT_AGF_BUF, - XFS_BLFT_AGFL_BUF, - XFS_BLFT_AGI_BUF, - XFS_BLFT_DINO_BUF, - XFS_BLFT_SYMLINK_BUF, - XFS_BLFT_DIR_BLOCK_BUF, - XFS_BLFT_DIR_DATA_BUF, - XFS_BLFT_DIR_FREE_BUF, - XFS_BLFT_DIR_LEAF1_BUF, - XFS_BLFT_DIR_LEAFN_BUF, - XFS_BLFT_DA_NODE_BUF, - XFS_BLFT_ATTR_LEAF_BUF, - XFS_BLFT_ATTR_RMT_BUF, - XFS_BLFT_SB_BUF, - XFS_BLFT_MAX_BUF = (1 << XFS_BLFT_BITS), -}; - -static inline void -xfs_blft_to_flags(struct xfs_buf_log_format *blf, enum xfs_blft type) -{ - ASSERT(type > XFS_BLFT_UNKNOWN_BUF && type < XFS_BLFT_MAX_BUF); - blf->blf_flags &= ~XFS_BLFT_MASK; - blf->blf_flags |= ((type << XFS_BLFT_SHIFT) & XFS_BLFT_MASK); -} - -static inline __uint16_t -xfs_blft_from_flags(struct xfs_buf_log_format *blf) -{ - return (blf->blf_flags & XFS_BLFT_MASK) >> XFS_BLFT_SHIFT; -} - -/* - * buf log item flags - */ +/* buf log item flags */ #define XFS_BLI_HOLD 0x01 #define XFS_BLI_DIRTY 0x02 #define XFS_BLI_STALE 0x04 @@ -133,8 +41,6 @@ xfs_blft_from_flags(struct xfs_buf_log_format *blf) { XFS_BLI_ORDERED, "ORDERED" } -#ifdef __KERNEL__ - struct xfs_buf; struct xfs_mount; struct xfs_buf_log_item; @@ -169,6 +75,6 @@ void xfs_trans_buf_set_type(struct xfs_trans *, struct xfs_buf *, enum xfs_blft); void xfs_trans_buf_copy_type(struct xfs_buf *dst_bp, struct xfs_buf *src_bp); -#endif /* __KERNEL__ */ +extern kmem_zone_t *xfs_buf_item_zone; #endif /* __XFS_BUF_ITEM_H__ */ diff --git a/fs/xfs/xfs_da_btree.c b/fs/xfs/xfs_da_btree.c index 0b8b2a1..d4e59a4 100644 --- a/fs/xfs/xfs_da_btree.c +++ b/fs/xfs/xfs_da_btree.c @@ -27,8 +27,8 @@ #include "xfs_mount.h" #include "xfs_da_btree.h" #include "xfs_bmap_btree.h" -#include "xfs_dir2.h" #include "xfs_dir2_format.h" +#include "xfs_dir2.h" #include "xfs_dir2_priv.h" #include "xfs_dinode.h" #include "xfs_inode.h" @@ -399,7 +399,7 @@ xfs_da3_split( struct xfs_da_intnode *node; struct xfs_buf *bp; int max; - int action; + int action = 0; int error; int i; @@ -2454,9 +2454,9 @@ static int xfs_buf_map_from_irec( struct xfs_mount *mp, struct xfs_buf_map **mapp, - unsigned int *nmaps, + int *nmaps, struct xfs_bmbt_irec *irecs, - unsigned int nirecs) + int nirecs) { struct xfs_buf_map *map; int i; diff --git a/fs/xfs/xfs_da_btree.h b/fs/xfs/xfs_da_btree.h index 6fb3371c..b1f2679 100644 --- a/fs/xfs/xfs_da_btree.h +++ b/fs/xfs/xfs_da_btree.h @@ -133,12 +133,19 @@ extern void xfs_da3_node_hdr_to_disk(struct xfs_da_intnode *to, struct xfs_da3_icnode_hdr *from); static inline int -xfs_da3_node_hdr_size(struct xfs_da_intnode *dap) +__xfs_da3_node_hdr_size(bool v3) { - if (dap->hdr.info.magic == cpu_to_be16(XFS_DA3_NODE_MAGIC)) + if (v3) return sizeof(struct xfs_da3_node_hdr); return sizeof(struct xfs_da_node_hdr); } +static inline int +xfs_da3_node_hdr_size(struct xfs_da_intnode *dap) +{ + bool v3 = dap->hdr.info.magic == cpu_to_be16(XFS_DA3_NODE_MAGIC); + + return __xfs_da3_node_hdr_size(v3); +} static inline struct xfs_da_node_entry * xfs_da3_node_tree_p(struct xfs_da_intnode *dap) @@ -176,6 +183,7 @@ enum xfs_dacmp { typedef struct xfs_da_args { const __uint8_t *name; /* string (maybe not NULL terminated) */ int namelen; /* length of string (maybe no NULL) */ + __uint8_t filetype; /* filetype of inode for directories */ __uint8_t *value; /* set of bytes (maybe contain NULLs) */ int valuelen; /* length of value */ int flags; /* argument flags (eg: ATTR_NOCREATE) */ diff --git a/fs/xfs/xfs_dfrag.c b/fs/xfs/xfs_dfrag.c deleted file mode 100644 index e36445c..0000000 --- a/fs/xfs/xfs_dfrag.c +++ /dev/null @@ -1,459 +0,0 @@ -/* - * Copyright (c) 2000-2006 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#include "xfs.h" -#include "xfs_fs.h" -#include "xfs_types.h" -#include "xfs_log.h" -#include "xfs_trans.h" -#include "xfs_sb.h" -#include "xfs_ag.h" -#include "xfs_mount.h" -#include "xfs_bmap_btree.h" -#include "xfs_alloc_btree.h" -#include "xfs_ialloc_btree.h" -#include "xfs_btree.h" -#include "xfs_dinode.h" -#include "xfs_inode.h" -#include "xfs_inode_item.h" -#include "xfs_bmap.h" -#include "xfs_itable.h" -#include "xfs_dfrag.h" -#include "xfs_error.h" -#include "xfs_vnodeops.h" -#include "xfs_trace.h" - - -static int xfs_swap_extents( - xfs_inode_t *ip, /* target inode */ - xfs_inode_t *tip, /* tmp inode */ - xfs_swapext_t *sxp); - -/* - * ioctl interface for swapext - */ -int -xfs_swapext( - xfs_swapext_t *sxp) -{ - xfs_inode_t *ip, *tip; - struct fd f, tmp; - int error = 0; - - /* Pull information for the target fd */ - f = fdget((int)sxp->sx_fdtarget); - if (!f.file) { - error = XFS_ERROR(EINVAL); - goto out; - } - - if (!(f.file->f_mode & FMODE_WRITE) || - !(f.file->f_mode & FMODE_READ) || - (f.file->f_flags & O_APPEND)) { - error = XFS_ERROR(EBADF); - goto out_put_file; - } - - tmp = fdget((int)sxp->sx_fdtmp); - if (!tmp.file) { - error = XFS_ERROR(EINVAL); - goto out_put_file; - } - - if (!(tmp.file->f_mode & FMODE_WRITE) || - !(tmp.file->f_mode & FMODE_READ) || - (tmp.file->f_flags & O_APPEND)) { - error = XFS_ERROR(EBADF); - goto out_put_tmp_file; - } - - if (IS_SWAPFILE(file_inode(f.file)) || - IS_SWAPFILE(file_inode(tmp.file))) { - error = XFS_ERROR(EINVAL); - goto out_put_tmp_file; - } - - ip = XFS_I(file_inode(f.file)); - tip = XFS_I(file_inode(tmp.file)); - - if (ip->i_mount != tip->i_mount) { - error = XFS_ERROR(EINVAL); - goto out_put_tmp_file; - } - - if (ip->i_ino == tip->i_ino) { - error = XFS_ERROR(EINVAL); - goto out_put_tmp_file; - } - - if (XFS_FORCED_SHUTDOWN(ip->i_mount)) { - error = XFS_ERROR(EIO); - goto out_put_tmp_file; - } - - error = xfs_swap_extents(ip, tip, sxp); - - out_put_tmp_file: - fdput(tmp); - out_put_file: - fdput(f); - out: - return error; -} - -/* - * We need to check that the format of the data fork in the temporary inode is - * valid for the target inode before doing the swap. This is not a problem with - * attr1 because of the fixed fork offset, but attr2 has a dynamically sized - * data fork depending on the space the attribute fork is taking so we can get - * invalid formats on the target inode. - * - * E.g. target has space for 7 extents in extent format, temp inode only has - * space for 6. If we defragment down to 7 extents, then the tmp format is a - * btree, but when swapped it needs to be in extent format. Hence we can't just - * blindly swap data forks on attr2 filesystems. - * - * Note that we check the swap in both directions so that we don't end up with - * a corrupt temporary inode, either. - * - * Note that fixing the way xfs_fsr sets up the attribute fork in the source - * inode will prevent this situation from occurring, so all we do here is - * reject and log the attempt. basically we are putting the responsibility on - * userspace to get this right. - */ -static int -xfs_swap_extents_check_format( - xfs_inode_t *ip, /* target inode */ - xfs_inode_t *tip) /* tmp inode */ -{ - - /* Should never get a local format */ - if (ip->i_d.di_format == XFS_DINODE_FMT_LOCAL || - tip->i_d.di_format == XFS_DINODE_FMT_LOCAL) - return EINVAL; - - /* - * if the target inode has less extents that then temporary inode then - * why did userspace call us? - */ - if (ip->i_d.di_nextents < tip->i_d.di_nextents) - return EINVAL; - - /* - * if the target inode is in extent form and the temp inode is in btree - * form then we will end up with the target inode in the wrong format - * as we already know there are less extents in the temp inode. - */ - if (ip->i_d.di_format == XFS_DINODE_FMT_EXTENTS && - tip->i_d.di_format == XFS_DINODE_FMT_BTREE) - return EINVAL; - - /* Check temp in extent form to max in target */ - if (tip->i_d.di_format == XFS_DINODE_FMT_EXTENTS && - XFS_IFORK_NEXTENTS(tip, XFS_DATA_FORK) > - XFS_IFORK_MAXEXT(ip, XFS_DATA_FORK)) - return EINVAL; - - /* Check target in extent form to max in temp */ - if (ip->i_d.di_format == XFS_DINODE_FMT_EXTENTS && - XFS_IFORK_NEXTENTS(ip, XFS_DATA_FORK) > - XFS_IFORK_MAXEXT(tip, XFS_DATA_FORK)) - return EINVAL; - - /* - * If we are in a btree format, check that the temp root block will fit - * in the target and that it has enough extents to be in btree format - * in the target. - * - * Note that we have to be careful to allow btree->extent conversions - * (a common defrag case) which will occur when the temp inode is in - * extent format... - */ - if (tip->i_d.di_format == XFS_DINODE_FMT_BTREE) { - if (XFS_IFORK_BOFF(ip) && - XFS_BMAP_BMDR_SPACE(tip->i_df.if_broot) > XFS_IFORK_BOFF(ip)) - return EINVAL; - if (XFS_IFORK_NEXTENTS(tip, XFS_DATA_FORK) <= - XFS_IFORK_MAXEXT(ip, XFS_DATA_FORK)) - return EINVAL; - } - - /* Reciprocal target->temp btree format checks */ - if (ip->i_d.di_format == XFS_DINODE_FMT_BTREE) { - if (XFS_IFORK_BOFF(tip) && - XFS_BMAP_BMDR_SPACE(ip->i_df.if_broot) > XFS_IFORK_BOFF(tip)) - return EINVAL; - if (XFS_IFORK_NEXTENTS(ip, XFS_DATA_FORK) <= - XFS_IFORK_MAXEXT(tip, XFS_DATA_FORK)) - return EINVAL; - } - - return 0; -} - -static int -xfs_swap_extents( - xfs_inode_t *ip, /* target inode */ - xfs_inode_t *tip, /* tmp inode */ - xfs_swapext_t *sxp) -{ - xfs_mount_t *mp = ip->i_mount; - xfs_trans_t *tp; - xfs_bstat_t *sbp = &sxp->sx_stat; - xfs_ifork_t *tempifp, *ifp, *tifp; - int src_log_flags, target_log_flags; - int error = 0; - int aforkblks = 0; - int taforkblks = 0; - __uint64_t tmp; - - /* - * We have no way of updating owner information in the BMBT blocks for - * each inode on CRC enabled filesystems, so to avoid corrupting the - * this metadata we simply don't allow extent swaps to occur. - */ - if (xfs_sb_version_hascrc(&mp->m_sb)) - return XFS_ERROR(EINVAL); - - tempifp = kmem_alloc(sizeof(xfs_ifork_t), KM_MAYFAIL); - if (!tempifp) { - error = XFS_ERROR(ENOMEM); - goto out; - } - - /* - * we have to do two separate lock calls here to keep lockdep - * happy. If we try to get all the locks in one call, lock will - * report false positives when we drop the ILOCK and regain them - * below. - */ - xfs_lock_two_inodes(ip, tip, XFS_IOLOCK_EXCL); - xfs_lock_two_inodes(ip, tip, XFS_ILOCK_EXCL); - - /* Verify that both files have the same format */ - if ((ip->i_d.di_mode & S_IFMT) != (tip->i_d.di_mode & S_IFMT)) { - error = XFS_ERROR(EINVAL); - goto out_unlock; - } - - /* Verify both files are either real-time or non-realtime */ - if (XFS_IS_REALTIME_INODE(ip) != XFS_IS_REALTIME_INODE(tip)) { - error = XFS_ERROR(EINVAL); - goto out_unlock; - } - - error = -filemap_write_and_wait(VFS_I(tip)->i_mapping); - if (error) - goto out_unlock; - truncate_pagecache_range(VFS_I(tip), 0, -1); - - /* Verify O_DIRECT for ftmp */ - if (VN_CACHED(VFS_I(tip)) != 0) { - error = XFS_ERROR(EINVAL); - goto out_unlock; - } - - /* Verify all data are being swapped */ - if (sxp->sx_offset != 0 || - sxp->sx_length != ip->i_d.di_size || - sxp->sx_length != tip->i_d.di_size) { - error = XFS_ERROR(EFAULT); - goto out_unlock; - } - - trace_xfs_swap_extent_before(ip, 0); - trace_xfs_swap_extent_before(tip, 1); - - /* check inode formats now that data is flushed */ - error = xfs_swap_extents_check_format(ip, tip); - if (error) { - xfs_notice(mp, - "%s: inode 0x%llx format is incompatible for exchanging.", - __func__, ip->i_ino); - goto out_unlock; - } - - /* - * Compare the current change & modify times with that - * passed in. If they differ, we abort this swap. - * This is the mechanism used to ensure the calling - * process that the file was not changed out from - * under it. - */ - if ((sbp->bs_ctime.tv_sec != VFS_I(ip)->i_ctime.tv_sec) || - (sbp->bs_ctime.tv_nsec != VFS_I(ip)->i_ctime.tv_nsec) || - (sbp->bs_mtime.tv_sec != VFS_I(ip)->i_mtime.tv_sec) || - (sbp->bs_mtime.tv_nsec != VFS_I(ip)->i_mtime.tv_nsec)) { - error = XFS_ERROR(EBUSY); - goto out_unlock; - } - - /* We need to fail if the file is memory mapped. Once we have tossed - * all existing pages, the page fault will have no option - * but to go to the filesystem for pages. By making the page fault call - * vop_read (or write in the case of autogrow) they block on the iolock - * until we have switched the extents. - */ - if (VN_MAPPED(VFS_I(ip))) { - error = XFS_ERROR(EBUSY); - goto out_unlock; - } - - xfs_iunlock(ip, XFS_ILOCK_EXCL); - xfs_iunlock(tip, XFS_ILOCK_EXCL); - - /* - * There is a race condition here since we gave up the - * ilock. However, the data fork will not change since - * we have the iolock (locked for truncation too) so we - * are safe. We don't really care if non-io related - * fields change. - */ - truncate_pagecache_range(VFS_I(ip), 0, -1); - - tp = xfs_trans_alloc(mp, XFS_TRANS_SWAPEXT); - if ((error = xfs_trans_reserve(tp, 0, - XFS_ICHANGE_LOG_RES(mp), 0, - 0, 0))) { - xfs_iunlock(ip, XFS_IOLOCK_EXCL); - xfs_iunlock(tip, XFS_IOLOCK_EXCL); - xfs_trans_cancel(tp, 0); - goto out; - } - xfs_lock_two_inodes(ip, tip, XFS_ILOCK_EXCL); - - /* - * Count the number of extended attribute blocks - */ - if ( ((XFS_IFORK_Q(ip) != 0) && (ip->i_d.di_anextents > 0)) && - (ip->i_d.di_aformat != XFS_DINODE_FMT_LOCAL)) { - error = xfs_bmap_count_blocks(tp, ip, XFS_ATTR_FORK, &aforkblks); - if (error) - goto out_trans_cancel; - } - if ( ((XFS_IFORK_Q(tip) != 0) && (tip->i_d.di_anextents > 0)) && - (tip->i_d.di_aformat != XFS_DINODE_FMT_LOCAL)) { - error = xfs_bmap_count_blocks(tp, tip, XFS_ATTR_FORK, - &taforkblks); - if (error) - goto out_trans_cancel; - } - - /* - * Swap the data forks of the inodes - */ - ifp = &ip->i_df; - tifp = &tip->i_df; - *tempifp = *ifp; /* struct copy */ - *ifp = *tifp; /* struct copy */ - *tifp = *tempifp; /* struct copy */ - - /* - * Fix the on-disk inode values - */ - tmp = (__uint64_t)ip->i_d.di_nblocks; - ip->i_d.di_nblocks = tip->i_d.di_nblocks - taforkblks + aforkblks; - tip->i_d.di_nblocks = tmp + taforkblks - aforkblks; - - tmp = (__uint64_t) ip->i_d.di_nextents; - ip->i_d.di_nextents = tip->i_d.di_nextents; - tip->i_d.di_nextents = tmp; - - tmp = (__uint64_t) ip->i_d.di_format; - ip->i_d.di_format = tip->i_d.di_format; - tip->i_d.di_format = tmp; - - /* - * The extents in the source inode could still contain speculative - * preallocation beyond EOF (e.g. the file is open but not modified - * while defrag is in progress). In that case, we need to copy over the - * number of delalloc blocks the data fork in the source inode is - * tracking beyond EOF so that when the fork is truncated away when the - * temporary inode is unlinked we don't underrun the i_delayed_blks - * counter on that inode. - */ - ASSERT(tip->i_delayed_blks == 0); - tip->i_delayed_blks = ip->i_delayed_blks; - ip->i_delayed_blks = 0; - - src_log_flags = XFS_ILOG_CORE; - switch (ip->i_d.di_format) { - case XFS_DINODE_FMT_EXTENTS: - /* If the extents fit in the inode, fix the - * pointer. Otherwise it's already NULL or - * pointing to the extent. - */ - if (ip->i_d.di_nextents <= XFS_INLINE_EXTS) { - ifp->if_u1.if_extents = - ifp->if_u2.if_inline_ext; - } - src_log_flags |= XFS_ILOG_DEXT; - break; - case XFS_DINODE_FMT_BTREE: - src_log_flags |= XFS_ILOG_DBROOT; - break; - } - - target_log_flags = XFS_ILOG_CORE; - switch (tip->i_d.di_format) { - case XFS_DINODE_FMT_EXTENTS: - /* If the extents fit in the inode, fix the - * pointer. Otherwise it's already NULL or - * pointing to the extent. - */ - if (tip->i_d.di_nextents <= XFS_INLINE_EXTS) { - tifp->if_u1.if_extents = - tifp->if_u2.if_inline_ext; - } - target_log_flags |= XFS_ILOG_DEXT; - break; - case XFS_DINODE_FMT_BTREE: - target_log_flags |= XFS_ILOG_DBROOT; - break; - } - - - xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL); - xfs_trans_ijoin(tp, tip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL); - - xfs_trans_log_inode(tp, ip, src_log_flags); - xfs_trans_log_inode(tp, tip, target_log_flags); - - /* - * If this is a synchronous mount, make sure that the - * transaction goes to disk before returning to the user. - */ - if (mp->m_flags & XFS_MOUNT_WSYNC) - xfs_trans_set_sync(tp); - - error = xfs_trans_commit(tp, 0); - - trace_xfs_swap_extent_after(ip, 0); - trace_xfs_swap_extent_after(tip, 1); -out: - kmem_free(tempifp); - return error; - -out_unlock: - xfs_iunlock(ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL); - xfs_iunlock(tip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL); - goto out; - -out_trans_cancel: - xfs_trans_cancel(tp, 0); - goto out_unlock; -} diff --git a/fs/xfs/xfs_dfrag.h b/fs/xfs/xfs_dfrag.h deleted file mode 100644 index 20bdd93..0000000 --- a/fs/xfs/xfs_dfrag.h +++ /dev/null @@ -1,53 +0,0 @@ -/* - * Copyright (c) 2000,2005 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#ifndef __XFS_DFRAG_H__ -#define __XFS_DFRAG_H__ - -/* - * Structure passed to xfs_swapext - */ - -typedef struct xfs_swapext -{ - __int64_t sx_version; /* version */ - __int64_t sx_fdtarget; /* fd of target file */ - __int64_t sx_fdtmp; /* fd of tmp file */ - xfs_off_t sx_offset; /* offset into file */ - xfs_off_t sx_length; /* leng from offset */ - char sx_pad[16]; /* pad space, unused */ - xfs_bstat_t sx_stat; /* stat of target b4 copy */ -} xfs_swapext_t; - -/* - * Version flag - */ -#define XFS_SX_VERSION 0 - -#ifdef __KERNEL__ -/* - * Prototypes for visible xfs_dfrag.c routines. - */ - -/* - * Syscall interface for xfs_swapext - */ -int xfs_swapext(struct xfs_swapext *sx); - -#endif /* __KERNEL__ */ - -#endif /* __XFS_DFRAG_H__ */ diff --git a/fs/xfs/xfs_dir2.c b/fs/xfs/xfs_dir2.c index 8f023de..edf203a 100644 --- a/fs/xfs/xfs_dir2.c +++ b/fs/xfs/xfs_dir2.c @@ -31,14 +31,14 @@ #include "xfs_inode.h" #include "xfs_inode_item.h" #include "xfs_bmap.h" -#include "xfs_dir2.h" #include "xfs_dir2_format.h" +#include "xfs_dir2.h" #include "xfs_dir2_priv.h" #include "xfs_error.h" -#include "xfs_vnodeops.h" #include "xfs_trace.h" -struct xfs_name xfs_name_dotdot = { (unsigned char *)"..", 2}; +struct xfs_name xfs_name_dotdot = { (unsigned char *)"..", 2, XFS_DIR3_FT_DIR }; + /* * ASCII case-insensitive (ie. A-Z) support for directories that was @@ -90,6 +90,9 @@ void xfs_dir_mount( xfs_mount_t *mp) { + int nodehdr_size; + + ASSERT(xfs_sb_version_hasdirv2(&mp->m_sb)); ASSERT((1 << (mp->m_sb.sb_blocklog + mp->m_sb.sb_dirblklog)) <= XFS_MAX_BLOCKSIZE); @@ -98,12 +101,13 @@ xfs_dir_mount( mp->m_dirdatablk = xfs_dir2_db_to_da(mp, XFS_DIR2_DATA_FIRSTDB(mp)); mp->m_dirleafblk = xfs_dir2_db_to_da(mp, XFS_DIR2_LEAF_FIRSTDB(mp)); mp->m_dirfreeblk = xfs_dir2_db_to_da(mp, XFS_DIR2_FREE_FIRSTDB(mp)); - mp->m_attr_node_ents = - (mp->m_sb.sb_blocksize - (uint)sizeof(xfs_da_node_hdr_t)) / - (uint)sizeof(xfs_da_node_entry_t); - mp->m_dir_node_ents = - (mp->m_dirblksize - (uint)sizeof(xfs_da_node_hdr_t)) / - (uint)sizeof(xfs_da_node_entry_t); + + nodehdr_size = __xfs_da3_node_hdr_size(xfs_sb_version_hascrc(&mp->m_sb)); + mp->m_attr_node_ents = (mp->m_sb.sb_blocksize - nodehdr_size) / + (uint)sizeof(xfs_da_node_entry_t); + mp->m_dir_node_ents = (mp->m_dirblksize - nodehdr_size) / + (uint)sizeof(xfs_da_node_entry_t); + mp->m_dir_magicpct = (mp->m_dirblksize * 37) / 100; if (xfs_sb_version_hasasciici(&mp->m_sb)) mp->m_dirnameops = &xfs_ascii_ci_nameops; @@ -209,6 +213,7 @@ xfs_dir_createname( memset(&args, 0, sizeof(xfs_da_args_t)); args.name = name->name; args.namelen = name->len; + args.filetype = name->type; args.hashval = dp->i_mount->m_dirnameops->hashname(name); args.inumber = inum; args.dp = dp; @@ -283,6 +288,7 @@ xfs_dir_lookup( memset(&args, 0, sizeof(xfs_da_args_t)); args.name = name->name; args.namelen = name->len; + args.filetype = name->type; args.hashval = dp->i_mount->m_dirnameops->hashname(name); args.dp = dp; args.whichfork = XFS_DATA_FORK; @@ -338,6 +344,7 @@ xfs_dir_removename( memset(&args, 0, sizeof(xfs_da_args_t)); args.name = name->name; args.namelen = name->len; + args.filetype = name->type; args.hashval = dp->i_mount->m_dirnameops->hashname(name); args.inumber = ino; args.dp = dp; @@ -363,37 +370,6 @@ xfs_dir_removename( } /* - * Read a directory. - */ -int -xfs_readdir( - xfs_inode_t *dp, - struct dir_context *ctx, - size_t bufsize) -{ - int rval; /* return value */ - int v; /* type-checking value */ - - trace_xfs_readdir(dp); - - if (XFS_FORCED_SHUTDOWN(dp->i_mount)) - return XFS_ERROR(EIO); - - ASSERT(S_ISDIR(dp->i_d.di_mode)); - XFS_STATS_INC(xs_dir_getdents); - - if (dp->i_d.di_format == XFS_DINODE_FMT_LOCAL) - rval = xfs_dir2_sf_getdents(dp, ctx); - else if ((rval = xfs_dir2_isblock(NULL, dp, &v))) - ; - else if (v) - rval = xfs_dir2_block_getdents(dp, ctx); - else - rval = xfs_dir2_leaf_getdents(dp, ctx, bufsize); - return rval; -} - -/* * Replace the inode number of a directory entry. */ int @@ -418,6 +394,7 @@ xfs_dir_replace( memset(&args, 0, sizeof(xfs_da_args_t)); args.name = name->name; args.namelen = name->len; + args.filetype = name->type; args.hashval = dp->i_mount->m_dirnameops->hashname(name); args.inumber = inum; args.dp = dp; @@ -465,6 +442,7 @@ xfs_dir_canenter( memset(&args, 0, sizeof(xfs_da_args_t)); args.name = name->name; args.namelen = name->len; + args.filetype = name->type; args.hashval = dp->i_mount->m_dirnameops->hashname(name); args.dp = dp; args.whichfork = XFS_DATA_FORK; diff --git a/fs/xfs/xfs_dir2.h b/fs/xfs/xfs_dir2.h index e937d99..9910401 100644 --- a/fs/xfs/xfs_dir2.h +++ b/fs/xfs/xfs_dir2.h @@ -23,6 +23,11 @@ struct xfs_da_args; struct xfs_inode; struct xfs_mount; struct xfs_trans; +struct xfs_dir2_sf_hdr; +struct xfs_dir2_sf_entry; +struct xfs_dir2_data_hdr; +struct xfs_dir2_data_entry; +struct xfs_dir2_data_unused; extern struct xfs_name xfs_name_dotdot; @@ -57,4 +62,45 @@ extern int xfs_dir_canenter(struct xfs_trans *tp, struct xfs_inode *dp, */ extern int xfs_dir2_sf_to_block(struct xfs_da_args *args); +/* + * Interface routines used by userspace utilities + */ +extern xfs_ino_t xfs_dir2_sf_get_parent_ino(struct xfs_dir2_sf_hdr *sfp); +extern void xfs_dir2_sf_put_parent_ino(struct xfs_dir2_sf_hdr *sfp, + xfs_ino_t ino); +extern xfs_ino_t xfs_dir3_sfe_get_ino(struct xfs_mount *mp, + struct xfs_dir2_sf_hdr *sfp, struct xfs_dir2_sf_entry *sfep); +extern void xfs_dir3_sfe_put_ino(struct xfs_mount *mp, + struct xfs_dir2_sf_hdr *hdr, struct xfs_dir2_sf_entry *sfep, + xfs_ino_t ino); + +extern int xfs_dir2_isblock(struct xfs_trans *tp, struct xfs_inode *dp, int *r); +extern int xfs_dir2_isleaf(struct xfs_trans *tp, struct xfs_inode *dp, int *r); +extern int xfs_dir2_shrink_inode(struct xfs_da_args *args, xfs_dir2_db_t db, + struct xfs_buf *bp); + +extern void xfs_dir2_data_freescan(struct xfs_mount *mp, + struct xfs_dir2_data_hdr *hdr, int *loghead); +extern void xfs_dir2_data_log_entry(struct xfs_trans *tp, struct xfs_buf *bp, + struct xfs_dir2_data_entry *dep); +extern void xfs_dir2_data_log_header(struct xfs_trans *tp, + struct xfs_buf *bp); +extern void xfs_dir2_data_log_unused(struct xfs_trans *tp, struct xfs_buf *bp, + struct xfs_dir2_data_unused *dup); +extern void xfs_dir2_data_make_free(struct xfs_trans *tp, struct xfs_buf *bp, + xfs_dir2_data_aoff_t offset, xfs_dir2_data_aoff_t len, + int *needlogp, int *needscanp); +extern void xfs_dir2_data_use_free(struct xfs_trans *tp, struct xfs_buf *bp, + struct xfs_dir2_data_unused *dup, xfs_dir2_data_aoff_t offset, + xfs_dir2_data_aoff_t len, int *needlogp, int *needscanp); + +extern struct xfs_dir2_data_free *xfs_dir2_data_freefind( + struct xfs_dir2_data_hdr *hdr, struct xfs_dir2_data_unused *dup); + +extern const struct xfs_buf_ops xfs_dir3_block_buf_ops; +extern const struct xfs_buf_ops xfs_dir3_leafn_buf_ops; +extern const struct xfs_buf_ops xfs_dir3_leaf1_buf_ops; +extern const struct xfs_buf_ops xfs_dir3_free_buf_ops; +extern const struct xfs_buf_ops xfs_dir3_data_buf_ops; + #endif /* __XFS_DIR2_H__ */ diff --git a/fs/xfs/xfs_dir2_block.c b/fs/xfs/xfs_dir2_block.c index 5e7fbd7..0957aa9 100644 --- a/fs/xfs/xfs_dir2_block.c +++ b/fs/xfs/xfs_dir2_block.c @@ -31,8 +31,8 @@ #include "xfs_inode_item.h" #include "xfs_bmap.h" #include "xfs_buf_item.h" -#include "xfs_dir2.h" #include "xfs_dir2_format.h" +#include "xfs_dir2.h" #include "xfs_dir2_priv.h" #include "xfs_error.h" #include "xfs_trace.h" @@ -126,7 +126,7 @@ const struct xfs_buf_ops xfs_dir3_block_buf_ops = { .verify_write = xfs_dir3_block_write_verify, }; -static int +int xfs_dir3_block_read( struct xfs_trans *tp, struct xfs_inode *dp, @@ -369,7 +369,7 @@ xfs_dir2_block_addname( if (error) return error; - len = xfs_dir2_data_entsize(args->namelen); + len = xfs_dir3_data_entsize(mp, args->namelen); /* * Set up pointers to parts of the block. @@ -549,7 +549,8 @@ xfs_dir2_block_addname( dep->inumber = cpu_to_be64(args->inumber); dep->namelen = args->namelen; memcpy(dep->name, args->name, args->namelen); - tagp = xfs_dir2_data_entry_tag_p(dep); + xfs_dir3_dirent_put_ftype(mp, dep, args->filetype); + tagp = xfs_dir3_data_entry_tag_p(mp, dep); *tagp = cpu_to_be16((char *)dep - (char *)hdr); /* * Clean up the bestfree array and log the header, tail, and entry. @@ -565,101 +566,6 @@ xfs_dir2_block_addname( } /* - * Readdir for block directories. - */ -int /* error */ -xfs_dir2_block_getdents( - xfs_inode_t *dp, /* incore inode */ - struct dir_context *ctx) -{ - xfs_dir2_data_hdr_t *hdr; /* block header */ - struct xfs_buf *bp; /* buffer for block */ - xfs_dir2_block_tail_t *btp; /* block tail */ - xfs_dir2_data_entry_t *dep; /* block data entry */ - xfs_dir2_data_unused_t *dup; /* block unused entry */ - char *endptr; /* end of the data entries */ - int error; /* error return value */ - xfs_mount_t *mp; /* filesystem mount point */ - char *ptr; /* current data entry */ - int wantoff; /* starting block offset */ - xfs_off_t cook; - - mp = dp->i_mount; - /* - * If the block number in the offset is out of range, we're done. - */ - if (xfs_dir2_dataptr_to_db(mp, ctx->pos) > mp->m_dirdatablk) - return 0; - - error = xfs_dir3_block_read(NULL, dp, &bp); - if (error) - return error; - - /* - * Extract the byte offset we start at from the seek pointer. - * We'll skip entries before this. - */ - wantoff = xfs_dir2_dataptr_to_off(mp, ctx->pos); - hdr = bp->b_addr; - xfs_dir3_data_check(dp, bp); - /* - * Set up values for the loop. - */ - btp = xfs_dir2_block_tail_p(mp, hdr); - ptr = (char *)xfs_dir3_data_entry_p(hdr); - endptr = (char *)xfs_dir2_block_leaf_p(btp); - - /* - * Loop over the data portion of the block. - * Each object is a real entry (dep) or an unused one (dup). - */ - while (ptr < endptr) { - dup = (xfs_dir2_data_unused_t *)ptr; - /* - * Unused, skip it. - */ - if (be16_to_cpu(dup->freetag) == XFS_DIR2_DATA_FREE_TAG) { - ptr += be16_to_cpu(dup->length); - continue; - } - - dep = (xfs_dir2_data_entry_t *)ptr; - - /* - * Bump pointer for the next iteration. - */ - ptr += xfs_dir2_data_entsize(dep->namelen); - /* - * The entry is before the desired starting point, skip it. - */ - if ((char *)dep - (char *)hdr < wantoff) - continue; - - cook = xfs_dir2_db_off_to_dataptr(mp, mp->m_dirdatablk, - (char *)dep - (char *)hdr); - - ctx->pos = cook & 0x7fffffff; - /* - * If it didn't fit, set the final offset to here & return. - */ - if (!dir_emit(ctx, (char *)dep->name, dep->namelen, - be64_to_cpu(dep->inumber), DT_UNKNOWN)) { - xfs_trans_brelse(NULL, bp); - return 0; - } - } - - /* - * Reached the end of the block. - * Set the offset to a non-existent block 1 and return. - */ - ctx->pos = xfs_dir2_db_off_to_dataptr(mp, mp->m_dirdatablk + 1, 0) & - 0x7fffffff; - xfs_trans_brelse(NULL, bp); - return 0; -} - -/* * Log leaf entries from the block. */ static void @@ -736,6 +642,7 @@ xfs_dir2_block_lookup( * Fill in inode number, CI name if appropriate, release the block. */ args->inumber = be64_to_cpu(dep->inumber); + args->filetype = xfs_dir3_dirent_get_ftype(mp, dep); error = xfs_dir_cilookup_result(args, dep->name, dep->namelen); xfs_trans_brelse(args->trans, bp); return XFS_ERROR(error); @@ -894,7 +801,7 @@ xfs_dir2_block_removename( needlog = needscan = 0; xfs_dir2_data_make_free(tp, bp, (xfs_dir2_data_aoff_t)((char *)dep - (char *)hdr), - xfs_dir2_data_entsize(dep->namelen), &needlog, &needscan); + xfs_dir3_data_entsize(mp, dep->namelen), &needlog, &needscan); /* * Fix up the block tail. */ @@ -968,6 +875,7 @@ xfs_dir2_block_replace( * Change the inode number to the new value. */ dep->inumber = cpu_to_be64(args->inumber); + xfs_dir3_dirent_put_ftype(mp, dep, args->filetype); xfs_dir2_data_log_entry(args->trans, bp, dep); xfs_dir3_data_check(dp, bp); return 0; @@ -1254,7 +1162,8 @@ xfs_dir2_sf_to_block( dep->inumber = cpu_to_be64(dp->i_ino); dep->namelen = 1; dep->name[0] = '.'; - tagp = xfs_dir2_data_entry_tag_p(dep); + xfs_dir3_dirent_put_ftype(mp, dep, XFS_DIR3_FT_DIR); + tagp = xfs_dir3_data_entry_tag_p(mp, dep); *tagp = cpu_to_be16((char *)dep - (char *)hdr); xfs_dir2_data_log_entry(tp, bp, dep); blp[0].hashval = cpu_to_be32(xfs_dir_hash_dot); @@ -1267,7 +1176,8 @@ xfs_dir2_sf_to_block( dep->inumber = cpu_to_be64(xfs_dir2_sf_get_parent_ino(sfp)); dep->namelen = 2; dep->name[0] = dep->name[1] = '.'; - tagp = xfs_dir2_data_entry_tag_p(dep); + xfs_dir3_dirent_put_ftype(mp, dep, XFS_DIR3_FT_DIR); + tagp = xfs_dir3_data_entry_tag_p(mp, dep); *tagp = cpu_to_be16((char *)dep - (char *)hdr); xfs_dir2_data_log_entry(tp, bp, dep); blp[1].hashval = cpu_to_be32(xfs_dir_hash_dotdot); @@ -1312,10 +1222,12 @@ xfs_dir2_sf_to_block( * Copy a real entry. */ dep = (xfs_dir2_data_entry_t *)((char *)hdr + newoffset); - dep->inumber = cpu_to_be64(xfs_dir2_sfe_get_ino(sfp, sfep)); + dep->inumber = cpu_to_be64(xfs_dir3_sfe_get_ino(mp, sfp, sfep)); dep->namelen = sfep->namelen; + xfs_dir3_dirent_put_ftype(mp, dep, + xfs_dir3_sfe_get_ftype(mp, sfp, sfep)); memcpy(dep->name, sfep->name, dep->namelen); - tagp = xfs_dir2_data_entry_tag_p(dep); + tagp = xfs_dir3_data_entry_tag_p(mp, dep); *tagp = cpu_to_be16((char *)dep - (char *)hdr); xfs_dir2_data_log_entry(tp, bp, dep); name.name = sfep->name; @@ -1328,7 +1240,7 @@ xfs_dir2_sf_to_block( if (++i == sfp->count) sfep = NULL; else - sfep = xfs_dir2_sf_nextentry(sfp, sfep); + sfep = xfs_dir3_sf_nextentry(mp, sfp, sfep); } /* Done with the temporary buffer */ kmem_free(sfp); diff --git a/fs/xfs/xfs_dir2_data.c b/fs/xfs/xfs_dir2_data.c index c293023..47e1326 100644 --- a/fs/xfs/xfs_dir2_data.c +++ b/fs/xfs/xfs_dir2_data.c @@ -29,14 +29,12 @@ #include "xfs_dinode.h" #include "xfs_inode.h" #include "xfs_dir2_format.h" +#include "xfs_dir2.h" #include "xfs_dir2_priv.h" #include "xfs_error.h" #include "xfs_buf_item.h" #include "xfs_cksum.h" -STATIC xfs_dir2_data_free_t * -xfs_dir2_data_freefind(xfs_dir2_data_hdr_t *hdr, xfs_dir2_data_unused_t *dup); - /* * Check the consistency of the data block. * The input can also be a block-format directory. @@ -149,8 +147,10 @@ __xfs_dir3_data_check( XFS_WANT_CORRUPTED_RETURN( !xfs_dir_ino_validate(mp, be64_to_cpu(dep->inumber))); XFS_WANT_CORRUPTED_RETURN( - be16_to_cpu(*xfs_dir2_data_entry_tag_p(dep)) == + be16_to_cpu(*xfs_dir3_data_entry_tag_p(mp, dep)) == (char *)dep - (char *)hdr); + XFS_WANT_CORRUPTED_RETURN( + xfs_dir3_dirent_get_ftype(mp, dep) < XFS_DIR3_FT_MAX); count++; lastfree = 0; if (hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC) || @@ -168,7 +168,7 @@ __xfs_dir3_data_check( } XFS_WANT_CORRUPTED_RETURN(i < be32_to_cpu(btp->count)); } - p += xfs_dir2_data_entsize(dep->namelen); + p += xfs_dir3_data_entsize(mp, dep->namelen); } /* * Need to have seen all the entries and all the bestfree slots. @@ -325,7 +325,7 @@ xfs_dir3_data_readahead( * Given a data block and an unused entry from that block, * return the bestfree entry if any that corresponds to it. */ -STATIC xfs_dir2_data_free_t * +xfs_dir2_data_free_t * xfs_dir2_data_freefind( xfs_dir2_data_hdr_t *hdr, /* data block */ xfs_dir2_data_unused_t *dup) /* data unused entry */ @@ -333,7 +333,7 @@ xfs_dir2_data_freefind( xfs_dir2_data_free_t *dfp; /* bestfree entry */ xfs_dir2_data_aoff_t off; /* offset value needed */ struct xfs_dir2_data_free *bf; -#if defined(DEBUG) && defined(__KERNEL__) +#ifdef DEBUG int matched; /* matched the value */ int seenzero; /* saw a 0 bestfree entry */ #endif @@ -341,7 +341,7 @@ xfs_dir2_data_freefind( off = (xfs_dir2_data_aoff_t)((char *)dup - (char *)hdr); bf = xfs_dir3_data_bestfree_p(hdr); -#if defined(DEBUG) && defined(__KERNEL__) +#ifdef DEBUG /* * Validate some consistency in the bestfree table. * Check order, non-overlapping entries, and if we find the @@ -538,8 +538,8 @@ xfs_dir2_data_freescan( else { dep = (xfs_dir2_data_entry_t *)p; ASSERT((char *)dep - (char *)hdr == - be16_to_cpu(*xfs_dir2_data_entry_tag_p(dep))); - p += xfs_dir2_data_entsize(dep->namelen); + be16_to_cpu(*xfs_dir3_data_entry_tag_p(mp, dep))); + p += xfs_dir3_data_entsize(mp, dep->namelen); } } } @@ -629,7 +629,8 @@ xfs_dir2_data_log_entry( struct xfs_buf *bp, xfs_dir2_data_entry_t *dep) /* data entry pointer */ { - xfs_dir2_data_hdr_t *hdr = bp->b_addr; + struct xfs_dir2_data_hdr *hdr = bp->b_addr; + struct xfs_mount *mp = tp->t_mountp; ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC) || hdr->magic == cpu_to_be32(XFS_DIR3_DATA_MAGIC) || @@ -637,7 +638,7 @@ xfs_dir2_data_log_entry( hdr->magic == cpu_to_be32(XFS_DIR3_BLOCK_MAGIC)); xfs_trans_log_buf(tp, bp, (uint)((char *)dep - (char *)hdr), - (uint)((char *)(xfs_dir2_data_entry_tag_p(dep) + 1) - + (uint)((char *)(xfs_dir3_data_entry_tag_p(mp, dep) + 1) - (char *)hdr - 1)); } diff --git a/fs/xfs/xfs_dir2_format.h b/fs/xfs/xfs_dir2_format.h index 7826782b..a0961a6 100644 --- a/fs/xfs/xfs_dir2_format.h +++ b/fs/xfs/xfs_dir2_format.h @@ -69,6 +69,23 @@ #define XFS_DIR3_FREE_MAGIC 0x58444633 /* XDF3: free index blocks */ /* + * Dirents in version 3 directories have a file type field. Additions to this + * list are an on-disk format change, requiring feature bits. Valid values + * are as follows: + */ +#define XFS_DIR3_FT_UNKNOWN 0 +#define XFS_DIR3_FT_REG_FILE 1 +#define XFS_DIR3_FT_DIR 2 +#define XFS_DIR3_FT_CHRDEV 3 +#define XFS_DIR3_FT_BLKDEV 4 +#define XFS_DIR3_FT_FIFO 5 +#define XFS_DIR3_FT_SOCK 6 +#define XFS_DIR3_FT_SYMLINK 7 +#define XFS_DIR3_FT_WHT 8 + +#define XFS_DIR3_FT_MAX 9 + +/* * Byte offset in data block and shortform entry. */ typedef __uint16_t xfs_dir2_data_off_t; @@ -138,6 +155,9 @@ typedef struct xfs_dir2_sf_entry { xfs_dir2_sf_off_t offset; /* saved offset */ __u8 name[]; /* name, variable size */ /* + * A single byte containing the file type field follows the inode + * number for version 3 directory entries. + * * A xfs_dir2_ino8_t or xfs_dir2_ino4_t follows here, at a * variable offset after the name. */ @@ -162,16 +182,6 @@ xfs_dir2_sf_put_offset(xfs_dir2_sf_entry_t *sfep, xfs_dir2_data_aoff_t off) put_unaligned_be16(off, &sfep->offset.i); } -static inline int -xfs_dir2_sf_entsize(struct xfs_dir2_sf_hdr *hdr, int len) -{ - return sizeof(struct xfs_dir2_sf_entry) + /* namelen + offset */ - len + /* name */ - (hdr->i8count ? /* ino */ - sizeof(xfs_dir2_ino8_t) : - sizeof(xfs_dir2_ino4_t)); -} - static inline struct xfs_dir2_sf_entry * xfs_dir2_sf_firstentry(struct xfs_dir2_sf_hdr *hdr) { @@ -179,14 +189,78 @@ xfs_dir2_sf_firstentry(struct xfs_dir2_sf_hdr *hdr) ((char *)hdr + xfs_dir2_sf_hdr_size(hdr->i8count)); } +static inline int +xfs_dir3_sf_entsize( + struct xfs_mount *mp, + struct xfs_dir2_sf_hdr *hdr, + int len) +{ + int count = sizeof(struct xfs_dir2_sf_entry); /* namelen + offset */ + + count += len; /* name */ + count += hdr->i8count ? sizeof(xfs_dir2_ino8_t) : + sizeof(xfs_dir2_ino4_t); /* ino # */ + if (xfs_sb_version_hasftype(&mp->m_sb)) + count += sizeof(__uint8_t); /* file type */ + return count; +} + static inline struct xfs_dir2_sf_entry * -xfs_dir2_sf_nextentry(struct xfs_dir2_sf_hdr *hdr, - struct xfs_dir2_sf_entry *sfep) +xfs_dir3_sf_nextentry( + struct xfs_mount *mp, + struct xfs_dir2_sf_hdr *hdr, + struct xfs_dir2_sf_entry *sfep) { return (struct xfs_dir2_sf_entry *) - ((char *)sfep + xfs_dir2_sf_entsize(hdr, sfep->namelen)); + ((char *)sfep + xfs_dir3_sf_entsize(mp, hdr, sfep->namelen)); } +/* + * in dir3 shortform directories, the file type field is stored at a variable + * offset after the inode number. Because it's only a single byte, endian + * conversion is not necessary. + */ +static inline __uint8_t * +xfs_dir3_sfe_ftypep( + struct xfs_dir2_sf_hdr *hdr, + struct xfs_dir2_sf_entry *sfep) +{ + return (__uint8_t *)&sfep->name[sfep->namelen]; +} + +static inline __uint8_t +xfs_dir3_sfe_get_ftype( + struct xfs_mount *mp, + struct xfs_dir2_sf_hdr *hdr, + struct xfs_dir2_sf_entry *sfep) +{ + __uint8_t *ftp; + + if (!xfs_sb_version_hasftype(&mp->m_sb)) + return XFS_DIR3_FT_UNKNOWN; + + ftp = xfs_dir3_sfe_ftypep(hdr, sfep); + if (*ftp >= XFS_DIR3_FT_MAX) + return XFS_DIR3_FT_UNKNOWN; + return *ftp; +} + +static inline void +xfs_dir3_sfe_put_ftype( + struct xfs_mount *mp, + struct xfs_dir2_sf_hdr *hdr, + struct xfs_dir2_sf_entry *sfep, + __uint8_t ftype) +{ + __uint8_t *ftp; + + ASSERT(ftype < XFS_DIR3_FT_MAX); + + if (!xfs_sb_version_hasftype(&mp->m_sb)) + return; + ftp = xfs_dir3_sfe_ftypep(hdr, sfep); + *ftp = ftype; +} /* * Data block structures. @@ -286,12 +360,18 @@ xfs_dir3_data_bestfree_p(struct xfs_dir2_data_hdr *hdr) * Active entry in a data block. * * Aligned to 8 bytes. After the variable length name field there is a - * 2 byte tag field, which can be accessed using xfs_dir2_data_entry_tag_p. + * 2 byte tag field, which can be accessed using xfs_dir3_data_entry_tag_p. + * + * For dir3 structures, there is file type field between the name and the tag. + * This can only be manipulated by helper functions. It is packed hard against + * the end of the name so any padding for rounding is between the file type and + * the tag. */ typedef struct xfs_dir2_data_entry { __be64 inumber; /* inode number */ __u8 namelen; /* name length */ __u8 name[]; /* name bytes, no null */ + /* __u8 filetype; */ /* type of inode we point to */ /* __be16 tag; */ /* starting offset of us */ } xfs_dir2_data_entry_t; @@ -311,20 +391,67 @@ typedef struct xfs_dir2_data_unused { /* * Size of a data entry. */ -static inline int xfs_dir2_data_entsize(int n) +static inline int +__xfs_dir3_data_entsize( + bool ftype, + int n) { - return (int)roundup(offsetof(struct xfs_dir2_data_entry, name[0]) + n + - (uint)sizeof(xfs_dir2_data_off_t), XFS_DIR2_DATA_ALIGN); + int size = offsetof(struct xfs_dir2_data_entry, name[0]); + + size += n; + size += sizeof(xfs_dir2_data_off_t); + if (ftype) + size += sizeof(__uint8_t); + return roundup(size, XFS_DIR2_DATA_ALIGN); +} +static inline int +xfs_dir3_data_entsize( + struct xfs_mount *mp, + int n) +{ + bool ftype = xfs_sb_version_hasftype(&mp->m_sb) ? true : false; + return __xfs_dir3_data_entsize(ftype, n); +} + +static inline __uint8_t +xfs_dir3_dirent_get_ftype( + struct xfs_mount *mp, + struct xfs_dir2_data_entry *dep) +{ + if (xfs_sb_version_hasftype(&mp->m_sb)) { + __uint8_t type = dep->name[dep->namelen]; + + ASSERT(type < XFS_DIR3_FT_MAX); + if (type < XFS_DIR3_FT_MAX) + return type; + + } + return XFS_DIR3_FT_UNKNOWN; +} + +static inline void +xfs_dir3_dirent_put_ftype( + struct xfs_mount *mp, + struct xfs_dir2_data_entry *dep, + __uint8_t type) +{ + ASSERT(type < XFS_DIR3_FT_MAX); + ASSERT(dep->namelen != 0); + + if (xfs_sb_version_hasftype(&mp->m_sb)) + dep->name[dep->namelen] = type; } /* * Pointer to an entry's tag word. */ static inline __be16 * -xfs_dir2_data_entry_tag_p(struct xfs_dir2_data_entry *dep) +xfs_dir3_data_entry_tag_p( + struct xfs_mount *mp, + struct xfs_dir2_data_entry *dep) { return (__be16 *)((char *)dep + - xfs_dir2_data_entsize(dep->namelen) - sizeof(__be16)); + xfs_dir3_data_entsize(mp, dep->namelen) - sizeof(__be16)); } /* @@ -375,13 +502,17 @@ xfs_dir3_data_unused_p(struct xfs_dir2_data_hdr *hdr) * data block header because the sfe embeds the block offset of the entry into * it so that it doesn't change when format conversion occurs. Bad Things Happen * if we don't follow this rule. + * + * XXX: there is scope for significant optimisation of the logic here. Right + * now we are checking for "dir3 format" over and over again. Ideally we should + * only do it once for each operation. */ #define XFS_DIR3_DATA_DOT_OFFSET(mp) \ xfs_dir3_data_hdr_size(xfs_sb_version_hascrc(&(mp)->m_sb)) #define XFS_DIR3_DATA_DOTDOT_OFFSET(mp) \ - (XFS_DIR3_DATA_DOT_OFFSET(mp) + xfs_dir2_data_entsize(1)) + (XFS_DIR3_DATA_DOT_OFFSET(mp) + xfs_dir3_data_entsize(mp, 1)) #define XFS_DIR3_DATA_FIRST_OFFSET(mp) \ - (XFS_DIR3_DATA_DOTDOT_OFFSET(mp) + xfs_dir2_data_entsize(2)) + (XFS_DIR3_DATA_DOTDOT_OFFSET(mp) + xfs_dir3_data_entsize(mp, 2)) static inline xfs_dir2_data_aoff_t xfs_dir3_data_dot_offset(struct xfs_dir2_data_hdr *hdr) @@ -392,13 +523,19 @@ xfs_dir3_data_dot_offset(struct xfs_dir2_data_hdr *hdr) static inline xfs_dir2_data_aoff_t xfs_dir3_data_dotdot_offset(struct xfs_dir2_data_hdr *hdr) { - return xfs_dir3_data_dot_offset(hdr) + xfs_dir2_data_entsize(1); + bool dir3 = hdr->magic == cpu_to_be32(XFS_DIR3_DATA_MAGIC) || + hdr->magic == cpu_to_be32(XFS_DIR3_BLOCK_MAGIC); + return xfs_dir3_data_dot_offset(hdr) + + __xfs_dir3_data_entsize(dir3, 1); } static inline xfs_dir2_data_aoff_t xfs_dir3_data_first_offset(struct xfs_dir2_data_hdr *hdr) { - return xfs_dir3_data_dotdot_offset(hdr) + xfs_dir2_data_entsize(2); + bool dir3 = hdr->magic == cpu_to_be32(XFS_DIR3_DATA_MAGIC) || + hdr->magic == cpu_to_be32(XFS_DIR3_BLOCK_MAGIC); + return xfs_dir3_data_dotdot_offset(hdr) + + __xfs_dir3_data_entsize(dir3, 2); } /* @@ -519,6 +656,9 @@ struct xfs_dir3_leaf { #define XFS_DIR3_LEAF_CRC_OFF offsetof(struct xfs_dir3_leaf_hdr, info.crc) +extern void xfs_dir3_leaf_hdr_from_disk(struct xfs_dir3_icleaf_hdr *to, + struct xfs_dir2_leaf *from); + static inline int xfs_dir3_leaf_hdr_size(struct xfs_dir2_leaf *lp) { diff --git a/fs/xfs/xfs_dir2_leaf.c b/fs/xfs/xfs_dir2_leaf.c index 2aed25c..08984ee 100644 --- a/fs/xfs/xfs_dir2_leaf.c +++ b/fs/xfs/xfs_dir2_leaf.c @@ -31,6 +31,7 @@ #include "xfs_inode.h" #include "xfs_bmap.h" #include "xfs_dir2_format.h" +#include "xfs_dir2.h" #include "xfs_dir2_priv.h" #include "xfs_error.h" #include "xfs_trace.h" @@ -695,7 +696,7 @@ xfs_dir2_leaf_addname( ents = xfs_dir3_leaf_ents_p(leaf); xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf); bestsp = xfs_dir2_leaf_bests_p(ltp); - length = xfs_dir2_data_entsize(args->namelen); + length = xfs_dir3_data_entsize(mp, args->namelen); /* * See if there are any entries with the same hash value @@ -896,7 +897,8 @@ xfs_dir2_leaf_addname( dep->inumber = cpu_to_be64(args->inumber); dep->namelen = args->namelen; memcpy(dep->name, args->name, dep->namelen); - tagp = xfs_dir2_data_entry_tag_p(dep); + xfs_dir3_dirent_put_ftype(mp, dep, args->filetype); + tagp = xfs_dir3_data_entry_tag_p(mp, dep); *tagp = cpu_to_be16((char *)dep - (char *)hdr); /* * Need to scan fix up the bestfree table. @@ -1083,396 +1085,6 @@ xfs_dir3_leaf_compact_x1( *highstalep = highstale; } -struct xfs_dir2_leaf_map_info { - xfs_extlen_t map_blocks; /* number of fsbs in map */ - xfs_dablk_t map_off; /* last mapped file offset */ - int map_size; /* total entries in *map */ - int map_valid; /* valid entries in *map */ - int nmap; /* mappings to ask xfs_bmapi */ - xfs_dir2_db_t curdb; /* db for current block */ - int ra_current; /* number of read-ahead blks */ - int ra_index; /* *map index for read-ahead */ - int ra_offset; /* map entry offset for ra */ - int ra_want; /* readahead count wanted */ - struct xfs_bmbt_irec map[]; /* map vector for blocks */ -}; - -STATIC int -xfs_dir2_leaf_readbuf( - struct xfs_inode *dp, - size_t bufsize, - struct xfs_dir2_leaf_map_info *mip, - xfs_dir2_off_t *curoff, - struct xfs_buf **bpp) -{ - struct xfs_mount *mp = dp->i_mount; - struct xfs_buf *bp = *bpp; - struct xfs_bmbt_irec *map = mip->map; - struct blk_plug plug; - int error = 0; - int length; - int i; - int j; - - /* - * If we have a buffer, we need to release it and - * take it out of the mapping. - */ - - if (bp) { - xfs_trans_brelse(NULL, bp); - bp = NULL; - mip->map_blocks -= mp->m_dirblkfsbs; - /* - * Loop to get rid of the extents for the - * directory block. - */ - for (i = mp->m_dirblkfsbs; i > 0; ) { - j = min_t(int, map->br_blockcount, i); - map->br_blockcount -= j; - map->br_startblock += j; - map->br_startoff += j; - /* - * If mapping is done, pitch it from - * the table. - */ - if (!map->br_blockcount && --mip->map_valid) - memmove(&map[0], &map[1], - sizeof(map[0]) * mip->map_valid); - i -= j; - } - } - - /* - * Recalculate the readahead blocks wanted. - */ - mip->ra_want = howmany(bufsize + mp->m_dirblksize, - mp->m_sb.sb_blocksize) - 1; - ASSERT(mip->ra_want >= 0); - - /* - * If we don't have as many as we want, and we haven't - * run out of data blocks, get some more mappings. - */ - if (1 + mip->ra_want > mip->map_blocks && - mip->map_off < xfs_dir2_byte_to_da(mp, XFS_DIR2_LEAF_OFFSET)) { - /* - * Get more bmaps, fill in after the ones - * we already have in the table. - */ - mip->nmap = mip->map_size - mip->map_valid; - error = xfs_bmapi_read(dp, mip->map_off, - xfs_dir2_byte_to_da(mp, XFS_DIR2_LEAF_OFFSET) - - mip->map_off, - &map[mip->map_valid], &mip->nmap, 0); - - /* - * Don't know if we should ignore this or try to return an - * error. The trouble with returning errors is that readdir - * will just stop without actually passing the error through. - */ - if (error) - goto out; /* XXX */ - - /* - * If we got all the mappings we asked for, set the final map - * offset based on the last bmap value received. Otherwise, - * we've reached the end. - */ - if (mip->nmap == mip->map_size - mip->map_valid) { - i = mip->map_valid + mip->nmap - 1; - mip->map_off = map[i].br_startoff + map[i].br_blockcount; - } else - mip->map_off = xfs_dir2_byte_to_da(mp, - XFS_DIR2_LEAF_OFFSET); - - /* - * Look for holes in the mapping, and eliminate them. Count up - * the valid blocks. - */ - for (i = mip->map_valid; i < mip->map_valid + mip->nmap; ) { - if (map[i].br_startblock == HOLESTARTBLOCK) { - mip->nmap--; - length = mip->map_valid + mip->nmap - i; - if (length) - memmove(&map[i], &map[i + 1], - sizeof(map[i]) * length); - } else { - mip->map_blocks += map[i].br_blockcount; - i++; - } - } - mip->map_valid += mip->nmap; - } - - /* - * No valid mappings, so no more data blocks. - */ - if (!mip->map_valid) { - *curoff = xfs_dir2_da_to_byte(mp, mip->map_off); - goto out; - } - - /* - * Read the directory block starting at the first mapping. - */ - mip->curdb = xfs_dir2_da_to_db(mp, map->br_startoff); - error = xfs_dir3_data_read(NULL, dp, map->br_startoff, - map->br_blockcount >= mp->m_dirblkfsbs ? - XFS_FSB_TO_DADDR(mp, map->br_startblock) : -1, &bp); - - /* - * Should just skip over the data block instead of giving up. - */ - if (error) - goto out; /* XXX */ - - /* - * Adjust the current amount of read-ahead: we just read a block that - * was previously ra. - */ - if (mip->ra_current) - mip->ra_current -= mp->m_dirblkfsbs; - - /* - * Do we need more readahead? - */ - blk_start_plug(&plug); - for (mip->ra_index = mip->ra_offset = i = 0; - mip->ra_want > mip->ra_current && i < mip->map_blocks; - i += mp->m_dirblkfsbs) { - ASSERT(mip->ra_index < mip->map_valid); - /* - * Read-ahead a contiguous directory block. - */ - if (i > mip->ra_current && - map[mip->ra_index].br_blockcount >= mp->m_dirblkfsbs) { - xfs_dir3_data_readahead(NULL, dp, - map[mip->ra_index].br_startoff + mip->ra_offset, - XFS_FSB_TO_DADDR(mp, - map[mip->ra_index].br_startblock + - mip->ra_offset)); - mip->ra_current = i; - } - - /* - * Read-ahead a non-contiguous directory block. This doesn't - * use our mapping, but this is a very rare case. - */ - else if (i > mip->ra_current) { - xfs_dir3_data_readahead(NULL, dp, - map[mip->ra_index].br_startoff + - mip->ra_offset, -1); - mip->ra_current = i; - } - - /* - * Advance offset through the mapping table. - */ - for (j = 0; j < mp->m_dirblkfsbs; j++) { - /* - * The rest of this extent but not more than a dir - * block. - */ - length = min_t(int, mp->m_dirblkfsbs, - map[mip->ra_index].br_blockcount - - mip->ra_offset); - j += length; - mip->ra_offset += length; - - /* - * Advance to the next mapping if this one is used up. - */ - if (mip->ra_offset == map[mip->ra_index].br_blockcount) { - mip->ra_offset = 0; - mip->ra_index++; - } - } - } - blk_finish_plug(&plug); - -out: - *bpp = bp; - return error; -} - -/* - * Getdents (readdir) for leaf and node directories. - * This reads the data blocks only, so is the same for both forms. - */ -int /* error */ -xfs_dir2_leaf_getdents( - xfs_inode_t *dp, /* incore directory inode */ - struct dir_context *ctx, - size_t bufsize) -{ - struct xfs_buf *bp = NULL; /* data block buffer */ - xfs_dir2_data_hdr_t *hdr; /* data block header */ - xfs_dir2_data_entry_t *dep; /* data entry */ - xfs_dir2_data_unused_t *dup; /* unused entry */ - int error = 0; /* error return value */ - int length; /* temporary length value */ - xfs_mount_t *mp; /* filesystem mount point */ - int byteoff; /* offset in current block */ - xfs_dir2_off_t curoff; /* current overall offset */ - xfs_dir2_off_t newoff; /* new curoff after new blk */ - char *ptr = NULL; /* pointer to current data */ - struct xfs_dir2_leaf_map_info *map_info; - - /* - * If the offset is at or past the largest allowed value, - * give up right away. - */ - if (ctx->pos >= XFS_DIR2_MAX_DATAPTR) - return 0; - - mp = dp->i_mount; - - /* - * Set up to bmap a number of blocks based on the caller's - * buffer size, the directory block size, and the filesystem - * block size. - */ - length = howmany(bufsize + mp->m_dirblksize, - mp->m_sb.sb_blocksize); - map_info = kmem_zalloc(offsetof(struct xfs_dir2_leaf_map_info, map) + - (length * sizeof(struct xfs_bmbt_irec)), - KM_SLEEP | KM_NOFS); - map_info->map_size = length; - - /* - * Inside the loop we keep the main offset value as a byte offset - * in the directory file. - */ - curoff = xfs_dir2_dataptr_to_byte(mp, ctx->pos); - - /* - * Force this conversion through db so we truncate the offset - * down to get the start of the data block. - */ - map_info->map_off = xfs_dir2_db_to_da(mp, - xfs_dir2_byte_to_db(mp, curoff)); - - /* - * Loop over directory entries until we reach the end offset. - * Get more blocks and readahead as necessary. - */ - while (curoff < XFS_DIR2_LEAF_OFFSET) { - /* - * If we have no buffer, or we're off the end of the - * current buffer, need to get another one. - */ - if (!bp || ptr >= (char *)bp->b_addr + mp->m_dirblksize) { - - error = xfs_dir2_leaf_readbuf(dp, bufsize, map_info, - &curoff, &bp); - if (error || !map_info->map_valid) - break; - - /* - * Having done a read, we need to set a new offset. - */ - newoff = xfs_dir2_db_off_to_byte(mp, map_info->curdb, 0); - /* - * Start of the current block. - */ - if (curoff < newoff) - curoff = newoff; - /* - * Make sure we're in the right block. - */ - else if (curoff > newoff) - ASSERT(xfs_dir2_byte_to_db(mp, curoff) == - map_info->curdb); - hdr = bp->b_addr; - xfs_dir3_data_check(dp, bp); - /* - * Find our position in the block. - */ - ptr = (char *)xfs_dir3_data_entry_p(hdr); - byteoff = xfs_dir2_byte_to_off(mp, curoff); - /* - * Skip past the header. - */ - if (byteoff == 0) - curoff += xfs_dir3_data_entry_offset(hdr); - /* - * Skip past entries until we reach our offset. - */ - else { - while ((char *)ptr - (char *)hdr < byteoff) { - dup = (xfs_dir2_data_unused_t *)ptr; - - if (be16_to_cpu(dup->freetag) - == XFS_DIR2_DATA_FREE_TAG) { - - length = be16_to_cpu(dup->length); - ptr += length; - continue; - } - dep = (xfs_dir2_data_entry_t *)ptr; - length = - xfs_dir2_data_entsize(dep->namelen); - ptr += length; - } - /* - * Now set our real offset. - */ - curoff = - xfs_dir2_db_off_to_byte(mp, - xfs_dir2_byte_to_db(mp, curoff), - (char *)ptr - (char *)hdr); - if (ptr >= (char *)hdr + mp->m_dirblksize) { - continue; - } - } - } - /* - * We have a pointer to an entry. - * Is it a live one? - */ - dup = (xfs_dir2_data_unused_t *)ptr; - /* - * No, it's unused, skip over it. - */ - if (be16_to_cpu(dup->freetag) == XFS_DIR2_DATA_FREE_TAG) { - length = be16_to_cpu(dup->length); - ptr += length; - curoff += length; - continue; - } - - dep = (xfs_dir2_data_entry_t *)ptr; - length = xfs_dir2_data_entsize(dep->namelen); - - ctx->pos = xfs_dir2_byte_to_dataptr(mp, curoff) & 0x7fffffff; - if (!dir_emit(ctx, (char *)dep->name, dep->namelen, - be64_to_cpu(dep->inumber), DT_UNKNOWN)) - break; - - /* - * Advance to next entry in the block. - */ - ptr += length; - curoff += length; - /* bufsize may have just been a guess; don't go negative */ - bufsize = bufsize > length ? bufsize - length : 0; - } - - /* - * All done. Set output offset value to current offset. - */ - if (curoff > xfs_dir2_dataptr_to_byte(mp, XFS_DIR2_MAX_DATAPTR)) - ctx->pos = XFS_DIR2_MAX_DATAPTR & 0x7fffffff; - else - ctx->pos = xfs_dir2_byte_to_dataptr(mp, curoff) & 0x7fffffff; - kmem_free(map_info); - if (bp) - xfs_trans_brelse(NULL, bp); - return error; -} - - /* * Log the bests entries indicated from a leaf1 block. */ @@ -1614,6 +1226,7 @@ xfs_dir2_leaf_lookup( * Return the found inode number & CI name if appropriate */ args->inumber = be64_to_cpu(dep->inumber); + args->filetype = xfs_dir3_dirent_get_ftype(dp->i_mount, dep); error = xfs_dir_cilookup_result(args, dep->name, dep->namelen); xfs_trans_brelse(tp, dbp); xfs_trans_brelse(tp, lbp); @@ -1816,7 +1429,7 @@ xfs_dir2_leaf_removename( */ xfs_dir2_data_make_free(tp, dbp, (xfs_dir2_data_aoff_t)((char *)dep - (char *)hdr), - xfs_dir2_data_entsize(dep->namelen), &needlog, &needscan); + xfs_dir3_data_entsize(mp, dep->namelen), &needlog, &needscan); /* * We just mark the leaf entry stale by putting a null in it. */ @@ -1944,6 +1557,7 @@ xfs_dir2_leaf_replace( * Put the new inode number in, log it. */ dep->inumber = cpu_to_be64(args->inumber); + xfs_dir3_dirent_put_ftype(dp->i_mount, dep, args->filetype); tp = args->trans; xfs_dir2_data_log_entry(tp, dbp, dep); xfs_dir3_leaf_check(dp->i_mount, lbp); @@ -1975,10 +1589,6 @@ xfs_dir2_leaf_search_hash( ents = xfs_dir3_leaf_ents_p(leaf); xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf); -#ifndef __KERNEL__ - if (!leafhdr.count) - return 0; -#endif /* * Note, the table cannot be empty, so we have to go through the loop. * Binary search the leaf entries looking for our hash value. diff --git a/fs/xfs/xfs_dir2_node.c b/fs/xfs/xfs_dir2_node.c index 2226a00..4c3dba7 100644 --- a/fs/xfs/xfs_dir2_node.c +++ b/fs/xfs/xfs_dir2_node.c @@ -30,6 +30,7 @@ #include "xfs_inode.h" #include "xfs_bmap.h" #include "xfs_dir2_format.h" +#include "xfs_dir2.h" #include "xfs_dir2_priv.h" #include "xfs_error.h" #include "xfs_trace.h" @@ -312,11 +313,13 @@ xfs_dir2_free_log_header( struct xfs_trans *tp, struct xfs_buf *bp) { +#ifdef DEBUG xfs_dir2_free_t *free; /* freespace structure */ free = bp->b_addr; ASSERT(free->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC) || free->hdr.magic == cpu_to_be32(XFS_DIR3_FREE_MAGIC)); +#endif xfs_trans_log_buf(tp, bp, 0, xfs_dir3_free_hdr_size(tp->t_mountp) - 1); } @@ -602,7 +605,7 @@ xfs_dir2_leafn_lookup_for_addname( ASSERT(free->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC) || free->hdr.magic == cpu_to_be32(XFS_DIR3_FREE_MAGIC)); } - length = xfs_dir2_data_entsize(args->namelen); + length = xfs_dir3_data_entsize(mp, args->namelen); /* * Loop over leaf entries with the right hash value. */ @@ -813,6 +816,7 @@ xfs_dir2_leafn_lookup_for_entry( xfs_trans_brelse(tp, state->extrablk.bp); args->cmpresult = cmp; args->inumber = be64_to_cpu(dep->inumber); + args->filetype = xfs_dir3_dirent_get_ftype(mp, dep); *indexp = index; state->extravalid = 1; state->extrablk.bp = curbp; @@ -1256,7 +1260,7 @@ xfs_dir2_leafn_remove( longest = be16_to_cpu(bf[0].length); needlog = needscan = 0; xfs_dir2_data_make_free(tp, dbp, off, - xfs_dir2_data_entsize(dep->namelen), &needlog, &needscan); + xfs_dir3_data_entsize(mp, dep->namelen), &needlog, &needscan); /* * Rescan the data block freespaces for bestfree. * Log the data block header if needed. @@ -1708,7 +1712,7 @@ xfs_dir2_node_addname_int( dp = args->dp; mp = dp->i_mount; tp = args->trans; - length = xfs_dir2_data_entsize(args->namelen); + length = xfs_dir3_data_entsize(mp, args->namelen); /* * If we came in with a freespace block that means that lookup * found an entry with our hash value. This is the freespace @@ -2004,7 +2008,8 @@ xfs_dir2_node_addname_int( dep->inumber = cpu_to_be64(args->inumber); dep->namelen = args->namelen; memcpy(dep->name, args->name, dep->namelen); - tagp = xfs_dir2_data_entry_tag_p(dep); + xfs_dir3_dirent_put_ftype(mp, dep, args->filetype); + tagp = xfs_dir3_data_entry_tag_p(mp, dep); *tagp = cpu_to_be16((char *)dep - (char *)hdr); xfs_dir2_data_log_entry(tp, dbp, dep); /* @@ -2224,6 +2229,7 @@ xfs_dir2_node_replace( * Fill in the new inode number and log the entry. */ dep->inumber = cpu_to_be64(inum); + xfs_dir3_dirent_put_ftype(state->mp, dep, args->filetype); xfs_dir2_data_log_entry(args->trans, state->extrablk.bp, dep); rval = 0; } diff --git a/fs/xfs/xfs_dir2_priv.h b/fs/xfs/xfs_dir2_priv.h index 0511cda..1bad84c 100644 --- a/fs/xfs/xfs_dir2_priv.h +++ b/fs/xfs/xfs_dir2_priv.h @@ -18,23 +18,26 @@ #ifndef __XFS_DIR2_PRIV_H__ #define __XFS_DIR2_PRIV_H__ +struct dir_context; + /* xfs_dir2.c */ extern int xfs_dir_ino_validate(struct xfs_mount *mp, xfs_ino_t ino); -extern int xfs_dir2_isblock(struct xfs_trans *tp, struct xfs_inode *dp, int *r); -extern int xfs_dir2_isleaf(struct xfs_trans *tp, struct xfs_inode *dp, int *r); extern int xfs_dir2_grow_inode(struct xfs_da_args *args, int space, xfs_dir2_db_t *dbp); -extern int xfs_dir2_shrink_inode(struct xfs_da_args *args, xfs_dir2_db_t db, - struct xfs_buf *bp); extern int xfs_dir_cilookup_result(struct xfs_da_args *args, const unsigned char *name, int len); -/* xfs_dir2_block.c */ -extern const struct xfs_buf_ops xfs_dir3_block_buf_ops; +#define S_SHIFT 12 +extern const unsigned char xfs_mode_to_ftype[]; + +extern unsigned char xfs_dir3_get_dtype(struct xfs_mount *mp, + __uint8_t filetype); + +/* xfs_dir2_block.c */ +extern int xfs_dir3_block_read(struct xfs_trans *tp, struct xfs_inode *dp, + struct xfs_buf **bpp); extern int xfs_dir2_block_addname(struct xfs_da_args *args); -extern int xfs_dir2_block_getdents(struct xfs_inode *dp, - struct dir_context *ctx); extern int xfs_dir2_block_lookup(struct xfs_da_args *args); extern int xfs_dir2_block_removename(struct xfs_da_args *args); extern int xfs_dir2_block_replace(struct xfs_da_args *args); @@ -48,9 +51,6 @@ extern int xfs_dir2_leaf_to_block(struct xfs_da_args *args, #define xfs_dir3_data_check(dp,bp) #endif -extern const struct xfs_buf_ops xfs_dir3_data_buf_ops; -extern const struct xfs_buf_ops xfs_dir3_free_buf_ops; - extern int __xfs_dir3_data_check(struct xfs_inode *dp, struct xfs_buf *bp); extern int xfs_dir3_data_read(struct xfs_trans *tp, struct xfs_inode *dp, xfs_dablk_t bno, xfs_daddr_t mapped_bno, struct xfs_buf **bpp); @@ -60,27 +60,10 @@ extern int xfs_dir3_data_readahead(struct xfs_trans *tp, struct xfs_inode *dp, extern struct xfs_dir2_data_free * xfs_dir2_data_freeinsert(struct xfs_dir2_data_hdr *hdr, struct xfs_dir2_data_unused *dup, int *loghead); -extern void xfs_dir2_data_freescan(struct xfs_mount *mp, - struct xfs_dir2_data_hdr *hdr, int *loghead); extern int xfs_dir3_data_init(struct xfs_da_args *args, xfs_dir2_db_t blkno, struct xfs_buf **bpp); -extern void xfs_dir2_data_log_entry(struct xfs_trans *tp, struct xfs_buf *bp, - struct xfs_dir2_data_entry *dep); -extern void xfs_dir2_data_log_header(struct xfs_trans *tp, - struct xfs_buf *bp); -extern void xfs_dir2_data_log_unused(struct xfs_trans *tp, struct xfs_buf *bp, - struct xfs_dir2_data_unused *dup); -extern void xfs_dir2_data_make_free(struct xfs_trans *tp, struct xfs_buf *bp, - xfs_dir2_data_aoff_t offset, xfs_dir2_data_aoff_t len, - int *needlogp, int *needscanp); -extern void xfs_dir2_data_use_free(struct xfs_trans *tp, struct xfs_buf *bp, - struct xfs_dir2_data_unused *dup, xfs_dir2_data_aoff_t offset, - xfs_dir2_data_aoff_t len, int *needlogp, int *needscanp); /* xfs_dir2_leaf.c */ -extern const struct xfs_buf_ops xfs_dir3_leaf1_buf_ops; -extern const struct xfs_buf_ops xfs_dir3_leafn_buf_ops; - extern int xfs_dir3_leafn_read(struct xfs_trans *tp, struct xfs_inode *dp, xfs_dablk_t fbno, xfs_daddr_t mappedbno, struct xfs_buf **bpp); extern int xfs_dir2_block_to_leaf(struct xfs_da_args *args, @@ -91,8 +74,6 @@ extern void xfs_dir3_leaf_compact(struct xfs_da_args *args, extern void xfs_dir3_leaf_compact_x1(struct xfs_dir3_icleaf_hdr *leafhdr, struct xfs_dir2_leaf_entry *ents, int *indexp, int *lowstalep, int *highstalep, int *lowlogp, int *highlogp); -extern int xfs_dir2_leaf_getdents(struct xfs_inode *dp, struct dir_context *ctx, - size_t bufsize); extern int xfs_dir3_leaf_get_buf(struct xfs_da_args *args, xfs_dir2_db_t bno, struct xfs_buf **bpp, __uint16_t magic); extern void xfs_dir3_leaf_log_ents(struct xfs_trans *tp, struct xfs_buf *bp, @@ -144,18 +125,18 @@ extern int xfs_dir2_free_read(struct xfs_trans *tp, struct xfs_inode *dp, xfs_dablk_t fbno, struct xfs_buf **bpp); /* xfs_dir2_sf.c */ -extern xfs_ino_t xfs_dir2_sf_get_parent_ino(struct xfs_dir2_sf_hdr *sfp); -extern xfs_ino_t xfs_dir2_sfe_get_ino(struct xfs_dir2_sf_hdr *sfp, - struct xfs_dir2_sf_entry *sfep); extern int xfs_dir2_block_sfsize(struct xfs_inode *dp, struct xfs_dir2_data_hdr *block, struct xfs_dir2_sf_hdr *sfhp); extern int xfs_dir2_block_to_sf(struct xfs_da_args *args, struct xfs_buf *bp, int size, xfs_dir2_sf_hdr_t *sfhp); extern int xfs_dir2_sf_addname(struct xfs_da_args *args); extern int xfs_dir2_sf_create(struct xfs_da_args *args, xfs_ino_t pino); -extern int xfs_dir2_sf_getdents(struct xfs_inode *dp, struct dir_context *ctx); extern int xfs_dir2_sf_lookup(struct xfs_da_args *args); extern int xfs_dir2_sf_removename(struct xfs_da_args *args); extern int xfs_dir2_sf_replace(struct xfs_da_args *args); +/* xfs_dir2_readdir.c */ +extern int xfs_readdir(struct xfs_inode *dp, struct dir_context *ctx, + size_t bufsize); + #endif /* __XFS_DIR2_PRIV_H__ */ diff --git a/fs/xfs/xfs_dir2_readdir.c b/fs/xfs/xfs_dir2_readdir.c new file mode 100644 index 0000000..8993ec1 --- /dev/null +++ b/fs/xfs/xfs_dir2_readdir.c @@ -0,0 +1,695 @@ +/* + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * Copyright (c) 2013 Red Hat, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#include "xfs.h" +#include "xfs_fs.h" +#include "xfs_types.h" +#include "xfs_bit.h" +#include "xfs_log.h" +#include "xfs_trans.h" +#include "xfs_sb.h" +#include "xfs_ag.h" +#include "xfs_mount.h" +#include "xfs_da_btree.h" +#include "xfs_bmap_btree.h" +#include "xfs_dinode.h" +#include "xfs_inode.h" +#include "xfs_dir2_format.h" +#include "xfs_dir2.h" +#include "xfs_dir2_priv.h" +#include "xfs_error.h" +#include "xfs_trace.h" +#include "xfs_bmap.h" + +/* + * Directory file type support functions + */ +static unsigned char xfs_dir3_filetype_table[] = { + DT_UNKNOWN, DT_REG, DT_DIR, DT_CHR, DT_BLK, + DT_FIFO, DT_SOCK, DT_LNK, DT_WHT, +}; + +unsigned char +xfs_dir3_get_dtype( + struct xfs_mount *mp, + __uint8_t filetype) +{ + if (!xfs_sb_version_hasftype(&mp->m_sb)) + return DT_UNKNOWN; + + if (filetype >= XFS_DIR3_FT_MAX) + return DT_UNKNOWN; + + return xfs_dir3_filetype_table[filetype]; +} +/* + * @mode, if set, indicates that the type field needs to be set up. + * This uses the transformation from file mode to DT_* as defined in linux/fs.h + * for file type specification. This will be propagated into the directory + * structure if appropriate for the given operation and filesystem config. + */ +const unsigned char xfs_mode_to_ftype[S_IFMT >> S_SHIFT] = { + [0] = XFS_DIR3_FT_UNKNOWN, + [S_IFREG >> S_SHIFT] = XFS_DIR3_FT_REG_FILE, + [S_IFDIR >> S_SHIFT] = XFS_DIR3_FT_DIR, + [S_IFCHR >> S_SHIFT] = XFS_DIR3_FT_CHRDEV, + [S_IFBLK >> S_SHIFT] = XFS_DIR3_FT_BLKDEV, + [S_IFIFO >> S_SHIFT] = XFS_DIR3_FT_FIFO, + [S_IFSOCK >> S_SHIFT] = XFS_DIR3_FT_SOCK, + [S_IFLNK >> S_SHIFT] = XFS_DIR3_FT_SYMLINK, +}; + +STATIC int +xfs_dir2_sf_getdents( + xfs_inode_t *dp, /* incore directory inode */ + struct dir_context *ctx) +{ + int i; /* shortform entry number */ + xfs_mount_t *mp; /* filesystem mount point */ + xfs_dir2_dataptr_t off; /* current entry's offset */ + xfs_dir2_sf_entry_t *sfep; /* shortform directory entry */ + xfs_dir2_sf_hdr_t *sfp; /* shortform structure */ + xfs_dir2_dataptr_t dot_offset; + xfs_dir2_dataptr_t dotdot_offset; + xfs_ino_t ino; + + mp = dp->i_mount; + + ASSERT(dp->i_df.if_flags & XFS_IFINLINE); + /* + * Give up if the directory is way too short. + */ + if (dp->i_d.di_size < offsetof(xfs_dir2_sf_hdr_t, parent)) { + ASSERT(XFS_FORCED_SHUTDOWN(mp)); + return XFS_ERROR(EIO); + } + + ASSERT(dp->i_df.if_bytes == dp->i_d.di_size); + ASSERT(dp->i_df.if_u1.if_data != NULL); + + sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data; + + ASSERT(dp->i_d.di_size >= xfs_dir2_sf_hdr_size(sfp->i8count)); + + /* + * If the block number in the offset is out of range, we're done. + */ + if (xfs_dir2_dataptr_to_db(mp, ctx->pos) > mp->m_dirdatablk) + return 0; + + /* + * Precalculate offsets for . and .. as we will always need them. + * + * XXX(hch): the second argument is sometimes 0 and sometimes + * mp->m_dirdatablk. + */ + dot_offset = xfs_dir2_db_off_to_dataptr(mp, mp->m_dirdatablk, + XFS_DIR3_DATA_DOT_OFFSET(mp)); + dotdot_offset = xfs_dir2_db_off_to_dataptr(mp, mp->m_dirdatablk, + XFS_DIR3_DATA_DOTDOT_OFFSET(mp)); + + /* + * Put . entry unless we're starting past it. + */ + if (ctx->pos <= dot_offset) { + ctx->pos = dot_offset & 0x7fffffff; + if (!dir_emit(ctx, ".", 1, dp->i_ino, DT_DIR)) + return 0; + } + + /* + * Put .. entry unless we're starting past it. + */ + if (ctx->pos <= dotdot_offset) { + ino = xfs_dir2_sf_get_parent_ino(sfp); + ctx->pos = dotdot_offset & 0x7fffffff; + if (!dir_emit(ctx, "..", 2, ino, DT_DIR)) + return 0; + } + + /* + * Loop while there are more entries and put'ing works. + */ + sfep = xfs_dir2_sf_firstentry(sfp); + for (i = 0; i < sfp->count; i++) { + __uint8_t filetype; + + off = xfs_dir2_db_off_to_dataptr(mp, mp->m_dirdatablk, + xfs_dir2_sf_get_offset(sfep)); + + if (ctx->pos > off) { + sfep = xfs_dir3_sf_nextentry(mp, sfp, sfep); + continue; + } + + ino = xfs_dir3_sfe_get_ino(mp, sfp, sfep); + filetype = xfs_dir3_sfe_get_ftype(mp, sfp, sfep); + ctx->pos = off & 0x7fffffff; + if (!dir_emit(ctx, (char *)sfep->name, sfep->namelen, ino, + xfs_dir3_get_dtype(mp, filetype))) + return 0; + sfep = xfs_dir3_sf_nextentry(mp, sfp, sfep); + } + + ctx->pos = xfs_dir2_db_off_to_dataptr(mp, mp->m_dirdatablk + 1, 0) & + 0x7fffffff; + return 0; +} + +/* + * Readdir for block directories. + */ +STATIC int +xfs_dir2_block_getdents( + xfs_inode_t *dp, /* incore inode */ + struct dir_context *ctx) +{ + xfs_dir2_data_hdr_t *hdr; /* block header */ + struct xfs_buf *bp; /* buffer for block */ + xfs_dir2_block_tail_t *btp; /* block tail */ + xfs_dir2_data_entry_t *dep; /* block data entry */ + xfs_dir2_data_unused_t *dup; /* block unused entry */ + char *endptr; /* end of the data entries */ + int error; /* error return value */ + xfs_mount_t *mp; /* filesystem mount point */ + char *ptr; /* current data entry */ + int wantoff; /* starting block offset */ + xfs_off_t cook; + + mp = dp->i_mount; + /* + * If the block number in the offset is out of range, we're done. + */ + if (xfs_dir2_dataptr_to_db(mp, ctx->pos) > mp->m_dirdatablk) + return 0; + + error = xfs_dir3_block_read(NULL, dp, &bp); + if (error) + return error; + + /* + * Extract the byte offset we start at from the seek pointer. + * We'll skip entries before this. + */ + wantoff = xfs_dir2_dataptr_to_off(mp, ctx->pos); + hdr = bp->b_addr; + xfs_dir3_data_check(dp, bp); + /* + * Set up values for the loop. + */ + btp = xfs_dir2_block_tail_p(mp, hdr); + ptr = (char *)xfs_dir3_data_entry_p(hdr); + endptr = (char *)xfs_dir2_block_leaf_p(btp); + + /* + * Loop over the data portion of the block. + * Each object is a real entry (dep) or an unused one (dup). + */ + while (ptr < endptr) { + __uint8_t filetype; + + dup = (xfs_dir2_data_unused_t *)ptr; + /* + * Unused, skip it. + */ + if (be16_to_cpu(dup->freetag) == XFS_DIR2_DATA_FREE_TAG) { + ptr += be16_to_cpu(dup->length); + continue; + } + + dep = (xfs_dir2_data_entry_t *)ptr; + + /* + * Bump pointer for the next iteration. + */ + ptr += xfs_dir3_data_entsize(mp, dep->namelen); + /* + * The entry is before the desired starting point, skip it. + */ + if ((char *)dep - (char *)hdr < wantoff) + continue; + + cook = xfs_dir2_db_off_to_dataptr(mp, mp->m_dirdatablk, + (char *)dep - (char *)hdr); + + ctx->pos = cook & 0x7fffffff; + filetype = xfs_dir3_dirent_get_ftype(mp, dep); + /* + * If it didn't fit, set the final offset to here & return. + */ + if (!dir_emit(ctx, (char *)dep->name, dep->namelen, + be64_to_cpu(dep->inumber), + xfs_dir3_get_dtype(mp, filetype))) { + xfs_trans_brelse(NULL, bp); + return 0; + } + } + + /* + * Reached the end of the block. + * Set the offset to a non-existent block 1 and return. + */ + ctx->pos = xfs_dir2_db_off_to_dataptr(mp, mp->m_dirdatablk + 1, 0) & + 0x7fffffff; + xfs_trans_brelse(NULL, bp); + return 0; +} + +struct xfs_dir2_leaf_map_info { + xfs_extlen_t map_blocks; /* number of fsbs in map */ + xfs_dablk_t map_off; /* last mapped file offset */ + int map_size; /* total entries in *map */ + int map_valid; /* valid entries in *map */ + int nmap; /* mappings to ask xfs_bmapi */ + xfs_dir2_db_t curdb; /* db for current block */ + int ra_current; /* number of read-ahead blks */ + int ra_index; /* *map index for read-ahead */ + int ra_offset; /* map entry offset for ra */ + int ra_want; /* readahead count wanted */ + struct xfs_bmbt_irec map[]; /* map vector for blocks */ +}; + +STATIC int +xfs_dir2_leaf_readbuf( + struct xfs_inode *dp, + size_t bufsize, + struct xfs_dir2_leaf_map_info *mip, + xfs_dir2_off_t *curoff, + struct xfs_buf **bpp) +{ + struct xfs_mount *mp = dp->i_mount; + struct xfs_buf *bp = *bpp; + struct xfs_bmbt_irec *map = mip->map; + struct blk_plug plug; + int error = 0; + int length; + int i; + int j; + + /* + * If we have a buffer, we need to release it and + * take it out of the mapping. + */ + + if (bp) { + xfs_trans_brelse(NULL, bp); + bp = NULL; + mip->map_blocks -= mp->m_dirblkfsbs; + /* + * Loop to get rid of the extents for the + * directory block. + */ + for (i = mp->m_dirblkfsbs; i > 0; ) { + j = min_t(int, map->br_blockcount, i); + map->br_blockcount -= j; + map->br_startblock += j; + map->br_startoff += j; + /* + * If mapping is done, pitch it from + * the table. + */ + if (!map->br_blockcount && --mip->map_valid) + memmove(&map[0], &map[1], + sizeof(map[0]) * mip->map_valid); + i -= j; + } + } + + /* + * Recalculate the readahead blocks wanted. + */ + mip->ra_want = howmany(bufsize + mp->m_dirblksize, + mp->m_sb.sb_blocksize) - 1; + ASSERT(mip->ra_want >= 0); + + /* + * If we don't have as many as we want, and we haven't + * run out of data blocks, get some more mappings. + */ + if (1 + mip->ra_want > mip->map_blocks && + mip->map_off < xfs_dir2_byte_to_da(mp, XFS_DIR2_LEAF_OFFSET)) { + /* + * Get more bmaps, fill in after the ones + * we already have in the table. + */ + mip->nmap = mip->map_size - mip->map_valid; + error = xfs_bmapi_read(dp, mip->map_off, + xfs_dir2_byte_to_da(mp, XFS_DIR2_LEAF_OFFSET) - + mip->map_off, + &map[mip->map_valid], &mip->nmap, 0); + + /* + * Don't know if we should ignore this or try to return an + * error. The trouble with returning errors is that readdir + * will just stop without actually passing the error through. + */ + if (error) + goto out; /* XXX */ + + /* + * If we got all the mappings we asked for, set the final map + * offset based on the last bmap value received. Otherwise, + * we've reached the end. + */ + if (mip->nmap == mip->map_size - mip->map_valid) { + i = mip->map_valid + mip->nmap - 1; + mip->map_off = map[i].br_startoff + map[i].br_blockcount; + } else + mip->map_off = xfs_dir2_byte_to_da(mp, + XFS_DIR2_LEAF_OFFSET); + + /* + * Look for holes in the mapping, and eliminate them. Count up + * the valid blocks. + */ + for (i = mip->map_valid; i < mip->map_valid + mip->nmap; ) { + if (map[i].br_startblock == HOLESTARTBLOCK) { + mip->nmap--; + length = mip->map_valid + mip->nmap - i; + if (length) + memmove(&map[i], &map[i + 1], + sizeof(map[i]) * length); + } else { + mip->map_blocks += map[i].br_blockcount; + i++; + } + } + mip->map_valid += mip->nmap; + } + + /* + * No valid mappings, so no more data blocks. + */ + if (!mip->map_valid) { + *curoff = xfs_dir2_da_to_byte(mp, mip->map_off); + goto out; + } + + /* + * Read the directory block starting at the first mapping. + */ + mip->curdb = xfs_dir2_da_to_db(mp, map->br_startoff); + error = xfs_dir3_data_read(NULL, dp, map->br_startoff, + map->br_blockcount >= mp->m_dirblkfsbs ? + XFS_FSB_TO_DADDR(mp, map->br_startblock) : -1, &bp); + + /* + * Should just skip over the data block instead of giving up. + */ + if (error) + goto out; /* XXX */ + + /* + * Adjust the current amount of read-ahead: we just read a block that + * was previously ra. + */ + if (mip->ra_current) + mip->ra_current -= mp->m_dirblkfsbs; + + /* + * Do we need more readahead? + */ + blk_start_plug(&plug); + for (mip->ra_index = mip->ra_offset = i = 0; + mip->ra_want > mip->ra_current && i < mip->map_blocks; + i += mp->m_dirblkfsbs) { + ASSERT(mip->ra_index < mip->map_valid); + /* + * Read-ahead a contiguous directory block. + */ + if (i > mip->ra_current && + map[mip->ra_index].br_blockcount >= mp->m_dirblkfsbs) { + xfs_dir3_data_readahead(NULL, dp, + map[mip->ra_index].br_startoff + mip->ra_offset, + XFS_FSB_TO_DADDR(mp, + map[mip->ra_index].br_startblock + + mip->ra_offset)); + mip->ra_current = i; + } + + /* + * Read-ahead a non-contiguous directory block. This doesn't + * use our mapping, but this is a very rare case. + */ + else if (i > mip->ra_current) { + xfs_dir3_data_readahead(NULL, dp, + map[mip->ra_index].br_startoff + + mip->ra_offset, -1); + mip->ra_current = i; + } + + /* + * Advance offset through the mapping table. + */ + for (j = 0; j < mp->m_dirblkfsbs; j++) { + /* + * The rest of this extent but not more than a dir + * block. + */ + length = min_t(int, mp->m_dirblkfsbs, + map[mip->ra_index].br_blockcount - + mip->ra_offset); + j += length; + mip->ra_offset += length; + + /* + * Advance to the next mapping if this one is used up. + */ + if (mip->ra_offset == map[mip->ra_index].br_blockcount) { + mip->ra_offset = 0; + mip->ra_index++; + } + } + } + blk_finish_plug(&plug); + +out: + *bpp = bp; + return error; +} + +/* + * Getdents (readdir) for leaf and node directories. + * This reads the data blocks only, so is the same for both forms. + */ +STATIC int +xfs_dir2_leaf_getdents( + xfs_inode_t *dp, /* incore directory inode */ + struct dir_context *ctx, + size_t bufsize) +{ + struct xfs_buf *bp = NULL; /* data block buffer */ + xfs_dir2_data_hdr_t *hdr; /* data block header */ + xfs_dir2_data_entry_t *dep; /* data entry */ + xfs_dir2_data_unused_t *dup; /* unused entry */ + int error = 0; /* error return value */ + int length; /* temporary length value */ + xfs_mount_t *mp; /* filesystem mount point */ + int byteoff; /* offset in current block */ + xfs_dir2_off_t curoff; /* current overall offset */ + xfs_dir2_off_t newoff; /* new curoff after new blk */ + char *ptr = NULL; /* pointer to current data */ + struct xfs_dir2_leaf_map_info *map_info; + + /* + * If the offset is at or past the largest allowed value, + * give up right away. + */ + if (ctx->pos >= XFS_DIR2_MAX_DATAPTR) + return 0; + + mp = dp->i_mount; + + /* + * Set up to bmap a number of blocks based on the caller's + * buffer size, the directory block size, and the filesystem + * block size. + */ + length = howmany(bufsize + mp->m_dirblksize, + mp->m_sb.sb_blocksize); + map_info = kmem_zalloc(offsetof(struct xfs_dir2_leaf_map_info, map) + + (length * sizeof(struct xfs_bmbt_irec)), + KM_SLEEP | KM_NOFS); + map_info->map_size = length; + + /* + * Inside the loop we keep the main offset value as a byte offset + * in the directory file. + */ + curoff = xfs_dir2_dataptr_to_byte(mp, ctx->pos); + + /* + * Force this conversion through db so we truncate the offset + * down to get the start of the data block. + */ + map_info->map_off = xfs_dir2_db_to_da(mp, + xfs_dir2_byte_to_db(mp, curoff)); + + /* + * Loop over directory entries until we reach the end offset. + * Get more blocks and readahead as necessary. + */ + while (curoff < XFS_DIR2_LEAF_OFFSET) { + __uint8_t filetype; + + /* + * If we have no buffer, or we're off the end of the + * current buffer, need to get another one. + */ + if (!bp || ptr >= (char *)bp->b_addr + mp->m_dirblksize) { + + error = xfs_dir2_leaf_readbuf(dp, bufsize, map_info, + &curoff, &bp); + if (error || !map_info->map_valid) + break; + + /* + * Having done a read, we need to set a new offset. + */ + newoff = xfs_dir2_db_off_to_byte(mp, map_info->curdb, 0); + /* + * Start of the current block. + */ + if (curoff < newoff) + curoff = newoff; + /* + * Make sure we're in the right block. + */ + else if (curoff > newoff) + ASSERT(xfs_dir2_byte_to_db(mp, curoff) == + map_info->curdb); + hdr = bp->b_addr; + xfs_dir3_data_check(dp, bp); + /* + * Find our position in the block. + */ + ptr = (char *)xfs_dir3_data_entry_p(hdr); + byteoff = xfs_dir2_byte_to_off(mp, curoff); + /* + * Skip past the header. + */ + if (byteoff == 0) + curoff += xfs_dir3_data_entry_offset(hdr); + /* + * Skip past entries until we reach our offset. + */ + else { + while ((char *)ptr - (char *)hdr < byteoff) { + dup = (xfs_dir2_data_unused_t *)ptr; + + if (be16_to_cpu(dup->freetag) + == XFS_DIR2_DATA_FREE_TAG) { + + length = be16_to_cpu(dup->length); + ptr += length; + continue; + } + dep = (xfs_dir2_data_entry_t *)ptr; + length = + xfs_dir3_data_entsize(mp, dep->namelen); + ptr += length; + } + /* + * Now set our real offset. + */ + curoff = + xfs_dir2_db_off_to_byte(mp, + xfs_dir2_byte_to_db(mp, curoff), + (char *)ptr - (char *)hdr); + if (ptr >= (char *)hdr + mp->m_dirblksize) { + continue; + } + } + } + /* + * We have a pointer to an entry. + * Is it a live one? + */ + dup = (xfs_dir2_data_unused_t *)ptr; + /* + * No, it's unused, skip over it. + */ + if (be16_to_cpu(dup->freetag) == XFS_DIR2_DATA_FREE_TAG) { + length = be16_to_cpu(dup->length); + ptr += length; + curoff += length; + continue; + } + + dep = (xfs_dir2_data_entry_t *)ptr; + length = xfs_dir3_data_entsize(mp, dep->namelen); + filetype = xfs_dir3_dirent_get_ftype(mp, dep); + + ctx->pos = xfs_dir2_byte_to_dataptr(mp, curoff) & 0x7fffffff; + if (!dir_emit(ctx, (char *)dep->name, dep->namelen, + be64_to_cpu(dep->inumber), + xfs_dir3_get_dtype(mp, filetype))) + break; + + /* + * Advance to next entry in the block. + */ + ptr += length; + curoff += length; + /* bufsize may have just been a guess; don't go negative */ + bufsize = bufsize > length ? bufsize - length : 0; + } + + /* + * All done. Set output offset value to current offset. + */ + if (curoff > xfs_dir2_dataptr_to_byte(mp, XFS_DIR2_MAX_DATAPTR)) + ctx->pos = XFS_DIR2_MAX_DATAPTR & 0x7fffffff; + else + ctx->pos = xfs_dir2_byte_to_dataptr(mp, curoff) & 0x7fffffff; + kmem_free(map_info); + if (bp) + xfs_trans_brelse(NULL, bp); + return error; +} + +/* + * Read a directory. + */ +int +xfs_readdir( + xfs_inode_t *dp, + struct dir_context *ctx, + size_t bufsize) +{ + int rval; /* return value */ + int v; /* type-checking value */ + + trace_xfs_readdir(dp); + + if (XFS_FORCED_SHUTDOWN(dp->i_mount)) + return XFS_ERROR(EIO); + + ASSERT(S_ISDIR(dp->i_d.di_mode)); + XFS_STATS_INC(xs_dir_getdents); + + if (dp->i_d.di_format == XFS_DINODE_FMT_LOCAL) + rval = xfs_dir2_sf_getdents(dp, ctx); + else if ((rval = xfs_dir2_isblock(NULL, dp, &v))) + ; + else if (v) + rval = xfs_dir2_block_getdents(dp, ctx); + else + rval = xfs_dir2_leaf_getdents(dp, ctx, bufsize); + return rval; +} diff --git a/fs/xfs/xfs_dir2_sf.c b/fs/xfs/xfs_dir2_sf.c index 97676a3..bb6e284 100644 --- a/fs/xfs/xfs_dir2_sf.c +++ b/fs/xfs/xfs_dir2_sf.c @@ -29,8 +29,8 @@ #include "xfs_inode.h" #include "xfs_inode_item.h" #include "xfs_error.h" -#include "xfs_dir2.h" #include "xfs_dir2_format.h" +#include "xfs_dir2.h" #include "xfs_dir2_priv.h" #include "xfs_trace.h" @@ -95,7 +95,7 @@ xfs_dir2_sf_get_parent_ino( return xfs_dir2_sf_get_ino(hdr, &hdr->parent); } -static void +void xfs_dir2_sf_put_parent_ino( struct xfs_dir2_sf_hdr *hdr, xfs_ino_t ino) @@ -105,31 +105,38 @@ xfs_dir2_sf_put_parent_ino( /* * In short-form directory entries the inode numbers are stored at variable - * offset behind the entry name. The inode numbers may only be accessed - * through the helpers below. + * offset behind the entry name. If the entry stores a filetype value, then it + * sits between the name and the inode number. Hence the inode numbers may only + * be accessed through the helpers below. */ static xfs_dir2_inou_t * -xfs_dir2_sfe_inop( +xfs_dir3_sfe_inop( + struct xfs_mount *mp, struct xfs_dir2_sf_entry *sfep) { - return (xfs_dir2_inou_t *)&sfep->name[sfep->namelen]; + __uint8_t *ptr = &sfep->name[sfep->namelen]; + if (xfs_sb_version_hasftype(&mp->m_sb)) + ptr++; + return (xfs_dir2_inou_t *)ptr; } xfs_ino_t -xfs_dir2_sfe_get_ino( +xfs_dir3_sfe_get_ino( + struct xfs_mount *mp, struct xfs_dir2_sf_hdr *hdr, struct xfs_dir2_sf_entry *sfep) { - return xfs_dir2_sf_get_ino(hdr, xfs_dir2_sfe_inop(sfep)); + return xfs_dir2_sf_get_ino(hdr, xfs_dir3_sfe_inop(mp, sfep)); } -static void -xfs_dir2_sfe_put_ino( +void +xfs_dir3_sfe_put_ino( + struct xfs_mount *mp, struct xfs_dir2_sf_hdr *hdr, struct xfs_dir2_sf_entry *sfep, xfs_ino_t ino) { - xfs_dir2_sf_put_ino(hdr, xfs_dir2_sfe_inop(sfep), ino); + xfs_dir2_sf_put_ino(hdr, xfs_dir3_sfe_inop(mp, sfep), ino); } /* @@ -157,9 +164,16 @@ xfs_dir2_block_sfsize( int namelen; /* total name bytes */ xfs_ino_t parent = 0; /* parent inode number */ int size=0; /* total computed size */ + int has_ftype; mp = dp->i_mount; + /* + * if there is a filetype field, add the extra byte to the namelen + * for each entry that we see. + */ + has_ftype = xfs_sb_version_hasftype(&mp->m_sb) ? 1 : 0; + count = i8count = namelen = 0; btp = xfs_dir2_block_tail_p(mp, hdr); blp = xfs_dir2_block_leaf_p(btp); @@ -188,9 +202,10 @@ xfs_dir2_block_sfsize( if (!isdot) i8count += be64_to_cpu(dep->inumber) > XFS_DIR2_MAX_SHORT_INUM; #endif + /* take into account the file type field */ if (!isdot && !isdotdot) { count++; - namelen += dep->namelen; + namelen += dep->namelen + has_ftype; } else if (isdotdot) parent = be64_to_cpu(dep->inumber); /* @@ -316,12 +331,14 @@ xfs_dir2_block_to_sf( (xfs_dir2_data_aoff_t) ((char *)dep - (char *)hdr)); memcpy(sfep->name, dep->name, dep->namelen); - xfs_dir2_sfe_put_ino(sfp, sfep, + xfs_dir3_sfe_put_ino(mp, sfp, sfep, be64_to_cpu(dep->inumber)); + xfs_dir3_sfe_put_ftype(mp, sfp, sfep, + xfs_dir3_dirent_get_ftype(mp, dep)); - sfep = xfs_dir2_sf_nextentry(sfp, sfep); + sfep = xfs_dir3_sf_nextentry(mp, sfp, sfep); } - ptr += xfs_dir2_data_entsize(dep->namelen); + ptr += xfs_dir3_data_entsize(mp, dep->namelen); } ASSERT((char *)sfep - (char *)sfp == size); xfs_dir2_sf_check(args); @@ -372,7 +389,7 @@ xfs_dir2_sf_addname( /* * Compute entry (and change in) size. */ - add_entsize = xfs_dir2_sf_entsize(sfp, args->namelen); + add_entsize = xfs_dir3_sf_entsize(dp->i_mount, sfp, args->namelen); incr_isize = add_entsize; objchange = 0; #if XFS_BIG_INUMS @@ -466,8 +483,9 @@ xfs_dir2_sf_addname_easy( /* * Grow the in-inode space. */ - xfs_idata_realloc(dp, xfs_dir2_sf_entsize(sfp, args->namelen), - XFS_DATA_FORK); + xfs_idata_realloc(dp, + xfs_dir3_sf_entsize(dp->i_mount, sfp, args->namelen), + XFS_DATA_FORK); /* * Need to set up again due to realloc of the inode data. */ @@ -479,7 +497,9 @@ xfs_dir2_sf_addname_easy( sfep->namelen = args->namelen; xfs_dir2_sf_put_offset(sfep, offset); memcpy(sfep->name, args->name, sfep->namelen); - xfs_dir2_sfe_put_ino(sfp, sfep, args->inumber); + xfs_dir3_sfe_put_ino(dp->i_mount, sfp, sfep, args->inumber); + xfs_dir3_sfe_put_ftype(dp->i_mount, sfp, sfep, args->filetype); + /* * Update the header and inode. */ @@ -519,11 +539,13 @@ xfs_dir2_sf_addname_hard( xfs_dir2_sf_hdr_t *oldsfp; /* original shortform dir */ xfs_dir2_sf_entry_t *sfep; /* entry in new dir */ xfs_dir2_sf_hdr_t *sfp; /* new shortform dir */ + struct xfs_mount *mp; /* * Copy the old directory to the stack buffer. */ dp = args->dp; + mp = dp->i_mount; sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data; old_isize = (int)dp->i_d.di_size; @@ -535,13 +557,13 @@ xfs_dir2_sf_addname_hard( * to insert the new entry. * If it's going to end up at the end then oldsfep will point there. */ - for (offset = XFS_DIR3_DATA_FIRST_OFFSET(dp->i_mount), + for (offset = XFS_DIR3_DATA_FIRST_OFFSET(mp), oldsfep = xfs_dir2_sf_firstentry(oldsfp), - add_datasize = xfs_dir2_data_entsize(args->namelen), + add_datasize = xfs_dir3_data_entsize(mp, args->namelen), eof = (char *)oldsfep == &buf[old_isize]; !eof; - offset = new_offset + xfs_dir2_data_entsize(oldsfep->namelen), - oldsfep = xfs_dir2_sf_nextentry(oldsfp, oldsfep), + offset = new_offset + xfs_dir3_data_entsize(mp, oldsfep->namelen), + oldsfep = xfs_dir3_sf_nextentry(mp, oldsfp, oldsfep), eof = (char *)oldsfep == &buf[old_isize]) { new_offset = xfs_dir2_sf_get_offset(oldsfep); if (offset + add_datasize <= new_offset) @@ -570,7 +592,8 @@ xfs_dir2_sf_addname_hard( sfep->namelen = args->namelen; xfs_dir2_sf_put_offset(sfep, offset); memcpy(sfep->name, args->name, sfep->namelen); - xfs_dir2_sfe_put_ino(sfp, sfep, args->inumber); + xfs_dir3_sfe_put_ino(mp, sfp, sfep, args->inumber); + xfs_dir3_sfe_put_ftype(mp, sfp, sfep, args->filetype); sfp->count++; #if XFS_BIG_INUMS if (args->inumber > XFS_DIR2_MAX_SHORT_INUM && !objchange) @@ -580,7 +603,7 @@ xfs_dir2_sf_addname_hard( * If there's more left to copy, do that. */ if (!eof) { - sfep = xfs_dir2_sf_nextentry(sfp, sfep); + sfep = xfs_dir3_sf_nextentry(mp, sfp, sfep); memcpy(sfep, oldsfep, old_isize - nbytes); } kmem_free(buf); @@ -616,7 +639,7 @@ xfs_dir2_sf_addname_pick( mp = dp->i_mount; sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data; - size = xfs_dir2_data_entsize(args->namelen); + size = xfs_dir3_data_entsize(mp, args->namelen); offset = XFS_DIR3_DATA_FIRST_OFFSET(mp); sfep = xfs_dir2_sf_firstentry(sfp); holefit = 0; @@ -629,8 +652,8 @@ xfs_dir2_sf_addname_pick( if (!holefit) holefit = offset + size <= xfs_dir2_sf_get_offset(sfep); offset = xfs_dir2_sf_get_offset(sfep) + - xfs_dir2_data_entsize(sfep->namelen); - sfep = xfs_dir2_sf_nextentry(sfp, sfep); + xfs_dir3_data_entsize(mp, sfep->namelen); + sfep = xfs_dir3_sf_nextentry(mp, sfp, sfep); } /* * Calculate data bytes used excluding the new entry, if this @@ -684,31 +707,34 @@ xfs_dir2_sf_check( int offset; /* data offset */ xfs_dir2_sf_entry_t *sfep; /* shortform dir entry */ xfs_dir2_sf_hdr_t *sfp; /* shortform structure */ + struct xfs_mount *mp; dp = args->dp; + mp = dp->i_mount; sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data; - offset = XFS_DIR3_DATA_FIRST_OFFSET(dp->i_mount); + offset = XFS_DIR3_DATA_FIRST_OFFSET(mp); ino = xfs_dir2_sf_get_parent_ino(sfp); i8count = ino > XFS_DIR2_MAX_SHORT_INUM; for (i = 0, sfep = xfs_dir2_sf_firstentry(sfp); i < sfp->count; - i++, sfep = xfs_dir2_sf_nextentry(sfp, sfep)) { + i++, sfep = xfs_dir3_sf_nextentry(mp, sfp, sfep)) { ASSERT(xfs_dir2_sf_get_offset(sfep) >= offset); - ino = xfs_dir2_sfe_get_ino(sfp, sfep); + ino = xfs_dir3_sfe_get_ino(mp, sfp, sfep); i8count += ino > XFS_DIR2_MAX_SHORT_INUM; offset = xfs_dir2_sf_get_offset(sfep) + - xfs_dir2_data_entsize(sfep->namelen); + xfs_dir3_data_entsize(mp, sfep->namelen); + ASSERT(xfs_dir3_sfe_get_ftype(mp, sfp, sfep) < + XFS_DIR3_FT_MAX); } ASSERT(i8count == sfp->i8count); ASSERT(XFS_BIG_INUMS || i8count == 0); ASSERT((char *)sfep - (char *)sfp == dp->i_d.di_size); ASSERT(offset + (sfp->count + 2) * (uint)sizeof(xfs_dir2_leaf_entry_t) + - (uint)sizeof(xfs_dir2_block_tail_t) <= - dp->i_mount->m_dirblksize); + (uint)sizeof(xfs_dir2_block_tail_t) <= mp->m_dirblksize); } #endif /* DEBUG */ @@ -765,100 +791,6 @@ xfs_dir2_sf_create( return 0; } -int /* error */ -xfs_dir2_sf_getdents( - xfs_inode_t *dp, /* incore directory inode */ - struct dir_context *ctx) -{ - int i; /* shortform entry number */ - xfs_mount_t *mp; /* filesystem mount point */ - xfs_dir2_dataptr_t off; /* current entry's offset */ - xfs_dir2_sf_entry_t *sfep; /* shortform directory entry */ - xfs_dir2_sf_hdr_t *sfp; /* shortform structure */ - xfs_dir2_dataptr_t dot_offset; - xfs_dir2_dataptr_t dotdot_offset; - xfs_ino_t ino; - - mp = dp->i_mount; - - ASSERT(dp->i_df.if_flags & XFS_IFINLINE); - /* - * Give up if the directory is way too short. - */ - if (dp->i_d.di_size < offsetof(xfs_dir2_sf_hdr_t, parent)) { - ASSERT(XFS_FORCED_SHUTDOWN(mp)); - return XFS_ERROR(EIO); - } - - ASSERT(dp->i_df.if_bytes == dp->i_d.di_size); - ASSERT(dp->i_df.if_u1.if_data != NULL); - - sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data; - - ASSERT(dp->i_d.di_size >= xfs_dir2_sf_hdr_size(sfp->i8count)); - - /* - * If the block number in the offset is out of range, we're done. - */ - if (xfs_dir2_dataptr_to_db(mp, ctx->pos) > mp->m_dirdatablk) - return 0; - - /* - * Precalculate offsets for . and .. as we will always need them. - * - * XXX(hch): the second argument is sometimes 0 and sometimes - * mp->m_dirdatablk. - */ - dot_offset = xfs_dir2_db_off_to_dataptr(mp, mp->m_dirdatablk, - XFS_DIR3_DATA_DOT_OFFSET(mp)); - dotdot_offset = xfs_dir2_db_off_to_dataptr(mp, mp->m_dirdatablk, - XFS_DIR3_DATA_DOTDOT_OFFSET(mp)); - - /* - * Put . entry unless we're starting past it. - */ - if (ctx->pos <= dot_offset) { - ctx->pos = dot_offset & 0x7fffffff; - if (!dir_emit(ctx, ".", 1, dp->i_ino, DT_DIR)) - return 0; - } - - /* - * Put .. entry unless we're starting past it. - */ - if (ctx->pos <= dotdot_offset) { - ino = xfs_dir2_sf_get_parent_ino(sfp); - ctx->pos = dotdot_offset & 0x7fffffff; - if (!dir_emit(ctx, "..", 2, ino, DT_DIR)) - return 0; - } - - /* - * Loop while there are more entries and put'ing works. - */ - sfep = xfs_dir2_sf_firstentry(sfp); - for (i = 0; i < sfp->count; i++) { - off = xfs_dir2_db_off_to_dataptr(mp, mp->m_dirdatablk, - xfs_dir2_sf_get_offset(sfep)); - - if (ctx->pos > off) { - sfep = xfs_dir2_sf_nextentry(sfp, sfep); - continue; - } - - ino = xfs_dir2_sfe_get_ino(sfp, sfep); - ctx->pos = off & 0x7fffffff; - if (!dir_emit(ctx, (char *)sfep->name, sfep->namelen, - ino, DT_UNKNOWN)) - return 0; - sfep = xfs_dir2_sf_nextentry(sfp, sfep); - } - - ctx->pos = xfs_dir2_db_off_to_dataptr(mp, mp->m_dirdatablk + 1, 0) & - 0x7fffffff; - return 0; -} - /* * Lookup an entry in a shortform directory. * Returns EEXIST if found, ENOENT if not found. @@ -898,6 +830,7 @@ xfs_dir2_sf_lookup( if (args->namelen == 1 && args->name[0] == '.') { args->inumber = dp->i_ino; args->cmpresult = XFS_CMP_EXACT; + args->filetype = XFS_DIR3_FT_DIR; return XFS_ERROR(EEXIST); } /* @@ -907,6 +840,7 @@ xfs_dir2_sf_lookup( args->name[0] == '.' && args->name[1] == '.') { args->inumber = xfs_dir2_sf_get_parent_ino(sfp); args->cmpresult = XFS_CMP_EXACT; + args->filetype = XFS_DIR3_FT_DIR; return XFS_ERROR(EEXIST); } /* @@ -914,7 +848,7 @@ xfs_dir2_sf_lookup( */ ci_sfep = NULL; for (i = 0, sfep = xfs_dir2_sf_firstentry(sfp); i < sfp->count; - i++, sfep = xfs_dir2_sf_nextentry(sfp, sfep)) { + i++, sfep = xfs_dir3_sf_nextentry(dp->i_mount, sfp, sfep)) { /* * Compare name and if it's an exact match, return the inode * number. If it's the first case-insensitive match, store the @@ -924,7 +858,10 @@ xfs_dir2_sf_lookup( sfep->namelen); if (cmp != XFS_CMP_DIFFERENT && cmp != args->cmpresult) { args->cmpresult = cmp; - args->inumber = xfs_dir2_sfe_get_ino(sfp, sfep); + args->inumber = xfs_dir3_sfe_get_ino(dp->i_mount, + sfp, sfep); + args->filetype = xfs_dir3_sfe_get_ftype(dp->i_mount, + sfp, sfep); if (cmp == XFS_CMP_EXACT) return XFS_ERROR(EEXIST); ci_sfep = sfep; @@ -980,10 +917,10 @@ xfs_dir2_sf_removename( * Find the one we're deleting. */ for (i = 0, sfep = xfs_dir2_sf_firstentry(sfp); i < sfp->count; - i++, sfep = xfs_dir2_sf_nextentry(sfp, sfep)) { + i++, sfep = xfs_dir3_sf_nextentry(dp->i_mount, sfp, sfep)) { if (xfs_da_compname(args, sfep->name, sfep->namelen) == XFS_CMP_EXACT) { - ASSERT(xfs_dir2_sfe_get_ino(sfp, sfep) == + ASSERT(xfs_dir3_sfe_get_ino(dp->i_mount, sfp, sfep) == args->inumber); break; } @@ -997,7 +934,7 @@ xfs_dir2_sf_removename( * Calculate sizes. */ byteoff = (int)((char *)sfep - (char *)sfp); - entsize = xfs_dir2_sf_entsize(sfp, args->namelen); + entsize = xfs_dir3_sf_entsize(dp->i_mount, sfp, args->namelen); newsize = oldsize - entsize; /* * Copy the part if any after the removed entry, sliding it down. @@ -1113,16 +1050,19 @@ xfs_dir2_sf_replace( * Normal entry, look for the name. */ else { - for (i = 0, sfep = xfs_dir2_sf_firstentry(sfp); - i < sfp->count; - i++, sfep = xfs_dir2_sf_nextentry(sfp, sfep)) { + for (i = 0, sfep = xfs_dir2_sf_firstentry(sfp); i < sfp->count; + i++, sfep = xfs_dir3_sf_nextentry(dp->i_mount, sfp, sfep)) { if (xfs_da_compname(args, sfep->name, sfep->namelen) == XFS_CMP_EXACT) { #if XFS_BIG_INUMS || defined(DEBUG) - ino = xfs_dir2_sfe_get_ino(sfp, sfep); + ino = xfs_dir3_sfe_get_ino(dp->i_mount, + sfp, sfep); ASSERT(args->inumber != ino); #endif - xfs_dir2_sfe_put_ino(sfp, sfep, args->inumber); + xfs_dir3_sfe_put_ino(dp->i_mount, sfp, sfep, + args->inumber); + xfs_dir3_sfe_put_ftype(dp->i_mount, sfp, sfep, + args->filetype); break; } } @@ -1189,10 +1129,12 @@ xfs_dir2_sf_toino4( int oldsize; /* old inode size */ xfs_dir2_sf_entry_t *sfep; /* new sf entry */ xfs_dir2_sf_hdr_t *sfp; /* new sf directory */ + struct xfs_mount *mp; trace_xfs_dir2_sf_toino4(args); dp = args->dp; + mp = dp->i_mount; /* * Copy the old directory to the buffer. @@ -1230,13 +1172,15 @@ xfs_dir2_sf_toino4( for (i = 0, sfep = xfs_dir2_sf_firstentry(sfp), oldsfep = xfs_dir2_sf_firstentry(oldsfp); i < sfp->count; - i++, sfep = xfs_dir2_sf_nextentry(sfp, sfep), - oldsfep = xfs_dir2_sf_nextentry(oldsfp, oldsfep)) { + i++, sfep = xfs_dir3_sf_nextentry(mp, sfp, sfep), + oldsfep = xfs_dir3_sf_nextentry(mp, oldsfp, oldsfep)) { sfep->namelen = oldsfep->namelen; sfep->offset = oldsfep->offset; memcpy(sfep->name, oldsfep->name, sfep->namelen); - xfs_dir2_sfe_put_ino(sfp, sfep, - xfs_dir2_sfe_get_ino(oldsfp, oldsfep)); + xfs_dir3_sfe_put_ino(mp, sfp, sfep, + xfs_dir3_sfe_get_ino(mp, oldsfp, oldsfep)); + xfs_dir3_sfe_put_ftype(mp, sfp, sfep, + xfs_dir3_sfe_get_ftype(mp, oldsfp, oldsfep)); } /* * Clean up the inode. @@ -1264,10 +1208,12 @@ xfs_dir2_sf_toino8( int oldsize; /* old inode size */ xfs_dir2_sf_entry_t *sfep; /* new sf entry */ xfs_dir2_sf_hdr_t *sfp; /* new sf directory */ + struct xfs_mount *mp; trace_xfs_dir2_sf_toino8(args); dp = args->dp; + mp = dp->i_mount; /* * Copy the old directory to the buffer. @@ -1305,13 +1251,15 @@ xfs_dir2_sf_toino8( for (i = 0, sfep = xfs_dir2_sf_firstentry(sfp), oldsfep = xfs_dir2_sf_firstentry(oldsfp); i < sfp->count; - i++, sfep = xfs_dir2_sf_nextentry(sfp, sfep), - oldsfep = xfs_dir2_sf_nextentry(oldsfp, oldsfep)) { + i++, sfep = xfs_dir3_sf_nextentry(mp, sfp, sfep), + oldsfep = xfs_dir3_sf_nextentry(mp, oldsfp, oldsfep)) { sfep->namelen = oldsfep->namelen; sfep->offset = oldsfep->offset; memcpy(sfep->name, oldsfep->name, sfep->namelen); - xfs_dir2_sfe_put_ino(sfp, sfep, - xfs_dir2_sfe_get_ino(oldsfp, oldsfep)); + xfs_dir3_sfe_put_ino(mp, sfp, sfep, + xfs_dir3_sfe_get_ino(mp, oldsfp, oldsfep)); + xfs_dir3_sfe_put_ftype(mp, sfp, sfep, + xfs_dir3_sfe_get_ftype(mp, oldsfp, oldsfep)); } /* * Clean up the inode. diff --git a/fs/xfs/xfs_discard.c b/fs/xfs/xfs_discard.c index 69cf4fc..45560ee 100644 --- a/fs/xfs/xfs_discard.c +++ b/fs/xfs/xfs_discard.c @@ -16,12 +16,13 @@ * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #include "xfs.h" -#include "xfs_sb.h" +#include "xfs_format.h" #include "xfs_log.h" +#include "xfs_trans.h" +#include "xfs_sb.h" #include "xfs_ag.h" #include "xfs_mount.h" #include "xfs_quota.h" -#include "xfs_trans.h" #include "xfs_alloc_btree.h" #include "xfs_bmap_btree.h" #include "xfs_ialloc_btree.h" diff --git a/fs/xfs/xfs_dquot.c b/fs/xfs/xfs_dquot.c index 0adf27e..251c666 100644 --- a/fs/xfs/xfs_dquot.c +++ b/fs/xfs/xfs_dquot.c @@ -17,6 +17,7 @@ */ #include "xfs.h" #include "xfs_fs.h" +#include "xfs_format.h" #include "xfs_bit.h" #include "xfs_log.h" #include "xfs_trans.h" @@ -28,6 +29,7 @@ #include "xfs_bmap_btree.h" #include "xfs_inode.h" #include "xfs_bmap.h" +#include "xfs_bmap_util.h" #include "xfs_rtalloc.h" #include "xfs_error.h" #include "xfs_itable.h" @@ -710,10 +712,8 @@ xfs_qm_dqread( if (flags & XFS_QMOPT_DQALLOC) { tp = xfs_trans_alloc(mp, XFS_TRANS_QM_DQALLOC); - error = xfs_trans_reserve(tp, XFS_QM_DQALLOC_SPACE_RES(mp), - XFS_QM_DQALLOC_LOG_RES(mp), 0, - XFS_TRANS_PERM_LOG_RES, - XFS_WRITE_LOG_COUNT); + error = xfs_trans_reserve(tp, &M_RES(mp)->tr_attrsetm, + XFS_QM_DQALLOC_SPACE_RES(mp), 0); if (error) goto error1; cancelflags = XFS_TRANS_RELEASE_LOG_RES; diff --git a/fs/xfs/xfs_dquot_item.c b/fs/xfs/xfs_dquot_item.c index 57aa4b0..60c6e1f 100644 --- a/fs/xfs/xfs_dquot_item.c +++ b/fs/xfs/xfs_dquot_item.c @@ -17,6 +17,7 @@ */ #include "xfs.h" #include "xfs_fs.h" +#include "xfs_format.h" #include "xfs_log.h" #include "xfs_trans.h" #include "xfs_sb.h" @@ -43,14 +44,15 @@ static inline struct xfs_dq_logitem *DQUOT_ITEM(struct xfs_log_item *lip) /* * returns the number of iovecs needed to log the given dquot item. */ -STATIC uint +STATIC void xfs_qm_dquot_logitem_size( - struct xfs_log_item *lip) + struct xfs_log_item *lip, + int *nvecs, + int *nbytes) { - /* - * we need only two iovecs, one for the format, one for the real thing - */ - return 2; + *nvecs += 2; + *nbytes += sizeof(struct xfs_dq_logformat) + + sizeof(struct xfs_disk_dquot); } /* @@ -285,11 +287,14 @@ static inline struct xfs_qoff_logitem *QOFF_ITEM(struct xfs_log_item *lip) * We only need 1 iovec for an quotaoff item. It just logs the * quotaoff_log_format structure. */ -STATIC uint +STATIC void xfs_qm_qoff_logitem_size( - struct xfs_log_item *lip) + struct xfs_log_item *lip, + int *nvecs, + int *nbytes) { - return 1; + *nvecs += 1; + *nbytes += sizeof(struct xfs_qoff_logitem); } /* diff --git a/fs/xfs/xfs_error.c b/fs/xfs/xfs_error.c index 35d3f5b..1123d93f 100644 --- a/fs/xfs/xfs_error.c +++ b/fs/xfs/xfs_error.c @@ -26,7 +26,6 @@ #include "xfs_bmap_btree.h" #include "xfs_dinode.h" #include "xfs_inode.h" -#include "xfs_utils.h" #include "xfs_error.h" #ifdef DEBUG diff --git a/fs/xfs/xfs_export.c b/fs/xfs/xfs_export.c index c585bc6..066df42 100644 --- a/fs/xfs/xfs_export.c +++ b/fs/xfs/xfs_export.c @@ -21,10 +21,11 @@ #include "xfs_trans.h" #include "xfs_sb.h" #include "xfs_ag.h" -#include "xfs_dir2.h" #include "xfs_mount.h" +#include "xfs_da_btree.h" +#include "xfs_dir2_format.h" +#include "xfs_dir2.h" #include "xfs_export.h" -#include "xfs_vnodeops.h" #include "xfs_bmap_btree.h" #include "xfs_inode.h" #include "xfs_inode_item.h" diff --git a/fs/xfs/xfs_extent_busy.c b/fs/xfs/xfs_extent_busy.c index 85e9f87..86f559f 100644 --- a/fs/xfs/xfs_extent_busy.c +++ b/fs/xfs/xfs_extent_busy.c @@ -147,7 +147,7 @@ xfs_extent_busy_search( * extent. If the overlap covers the beginning, the end, or all of the busy * extent, the overlapping portion can be made unbusy and used for the * allocation. We can't split a busy extent because we can't modify a - * transaction/CIL context busy list, but we can update an entries block + * transaction/CIL context busy list, but we can update an entry's block * number or length. * * Returns true if the extent can safely be reused, or false if the search diff --git a/fs/xfs/xfs_extfree_item.c b/fs/xfs/xfs_extfree_item.c index 452920a..dc53e8f 100644 --- a/fs/xfs/xfs_extfree_item.c +++ b/fs/xfs/xfs_extfree_item.c @@ -73,11 +73,22 @@ __xfs_efi_release( * We only need 1 iovec for an efi item. It just logs the efi_log_format * structure. */ -STATIC uint +static inline int +xfs_efi_item_sizeof( + struct xfs_efi_log_item *efip) +{ + return sizeof(struct xfs_efi_log_format) + + (efip->efi_format.efi_nextents - 1) * sizeof(xfs_extent_t); +} + +STATIC void xfs_efi_item_size( - struct xfs_log_item *lip) + struct xfs_log_item *lip, + int *nvecs, + int *nbytes) { - return 1; + *nvecs += 1; + *nbytes += xfs_efi_item_sizeof(EFI_ITEM(lip)); } /* @@ -93,21 +104,17 @@ xfs_efi_item_format( struct xfs_log_iovec *log_vector) { struct xfs_efi_log_item *efip = EFI_ITEM(lip); - uint size; ASSERT(atomic_read(&efip->efi_next_extent) == efip->efi_format.efi_nextents); efip->efi_format.efi_type = XFS_LI_EFI; - - size = sizeof(xfs_efi_log_format_t); - size += (efip->efi_format.efi_nextents - 1) * sizeof(xfs_extent_t); efip->efi_format.efi_size = 1; log_vector->i_addr = &efip->efi_format; - log_vector->i_len = size; + log_vector->i_len = xfs_efi_item_sizeof(efip); log_vector->i_type = XLOG_REG_TYPE_EFI_FORMAT; - ASSERT(size >= sizeof(xfs_efi_log_format_t)); + ASSERT(log_vector->i_len >= sizeof(xfs_efi_log_format_t)); } @@ -333,11 +340,22 @@ xfs_efd_item_free(struct xfs_efd_log_item *efdp) * We only need 1 iovec for an efd item. It just logs the efd_log_format * structure. */ -STATIC uint +static inline int +xfs_efd_item_sizeof( + struct xfs_efd_log_item *efdp) +{ + return sizeof(xfs_efd_log_format_t) + + (efdp->efd_format.efd_nextents - 1) * sizeof(xfs_extent_t); +} + +STATIC void xfs_efd_item_size( - struct xfs_log_item *lip) + struct xfs_log_item *lip, + int *nvecs, + int *nbytes) { - return 1; + *nvecs += 1; + *nbytes += xfs_efd_item_sizeof(EFD_ITEM(lip)); } /* @@ -353,20 +371,16 @@ xfs_efd_item_format( struct xfs_log_iovec *log_vector) { struct xfs_efd_log_item *efdp = EFD_ITEM(lip); - uint size; ASSERT(efdp->efd_next_extent == efdp->efd_format.efd_nextents); efdp->efd_format.efd_type = XFS_LI_EFD; - - size = sizeof(xfs_efd_log_format_t); - size += (efdp->efd_format.efd_nextents - 1) * sizeof(xfs_extent_t); efdp->efd_format.efd_size = 1; log_vector->i_addr = &efdp->efd_format; - log_vector->i_len = size; + log_vector->i_len = xfs_efd_item_sizeof(efdp); log_vector->i_type = XLOG_REG_TYPE_EFD_FORMAT; - ASSERT(size >= sizeof(xfs_efd_log_format_t)); + ASSERT(log_vector->i_len >= sizeof(xfs_efd_log_format_t)); } /* diff --git a/fs/xfs/xfs_extfree_item.h b/fs/xfs/xfs_extfree_item.h index 4322224..0ffbce3 100644 --- a/fs/xfs/xfs_extfree_item.h +++ b/fs/xfs/xfs_extfree_item.h @@ -18,93 +18,11 @@ #ifndef __XFS_EXTFREE_ITEM_H__ #define __XFS_EXTFREE_ITEM_H__ +/* kernel only EFI/EFD definitions */ + struct xfs_mount; struct kmem_zone; -typedef struct xfs_extent { - xfs_dfsbno_t ext_start; - xfs_extlen_t ext_len; -} xfs_extent_t; - -/* - * Since an xfs_extent_t has types (start:64, len: 32) - * there are different alignments on 32 bit and 64 bit kernels. - * So we provide the different variants for use by a - * conversion routine. - */ - -typedef struct xfs_extent_32 { - __uint64_t ext_start; - __uint32_t ext_len; -} __attribute__((packed)) xfs_extent_32_t; - -typedef struct xfs_extent_64 { - __uint64_t ext_start; - __uint32_t ext_len; - __uint32_t ext_pad; -} xfs_extent_64_t; - -/* - * This is the structure used to lay out an efi log item in the - * log. The efi_extents field is a variable size array whose - * size is given by efi_nextents. - */ -typedef struct xfs_efi_log_format { - __uint16_t efi_type; /* efi log item type */ - __uint16_t efi_size; /* size of this item */ - __uint32_t efi_nextents; /* # extents to free */ - __uint64_t efi_id; /* efi identifier */ - xfs_extent_t efi_extents[1]; /* array of extents to free */ -} xfs_efi_log_format_t; - -typedef struct xfs_efi_log_format_32 { - __uint16_t efi_type; /* efi log item type */ - __uint16_t efi_size; /* size of this item */ - __uint32_t efi_nextents; /* # extents to free */ - __uint64_t efi_id; /* efi identifier */ - xfs_extent_32_t efi_extents[1]; /* array of extents to free */ -} __attribute__((packed)) xfs_efi_log_format_32_t; - -typedef struct xfs_efi_log_format_64 { - __uint16_t efi_type; /* efi log item type */ - __uint16_t efi_size; /* size of this item */ - __uint32_t efi_nextents; /* # extents to free */ - __uint64_t efi_id; /* efi identifier */ - xfs_extent_64_t efi_extents[1]; /* array of extents to free */ -} xfs_efi_log_format_64_t; - -/* - * This is the structure used to lay out an efd log item in the - * log. The efd_extents array is a variable size array whose - * size is given by efd_nextents; - */ -typedef struct xfs_efd_log_format { - __uint16_t efd_type; /* efd log item type */ - __uint16_t efd_size; /* size of this item */ - __uint32_t efd_nextents; /* # of extents freed */ - __uint64_t efd_efi_id; /* id of corresponding efi */ - xfs_extent_t efd_extents[1]; /* array of extents freed */ -} xfs_efd_log_format_t; - -typedef struct xfs_efd_log_format_32 { - __uint16_t efd_type; /* efd log item type */ - __uint16_t efd_size; /* size of this item */ - __uint32_t efd_nextents; /* # of extents freed */ - __uint64_t efd_efi_id; /* id of corresponding efi */ - xfs_extent_32_t efd_extents[1]; /* array of extents freed */ -} __attribute__((packed)) xfs_efd_log_format_32_t; - -typedef struct xfs_efd_log_format_64 { - __uint16_t efd_type; /* efd log item type */ - __uint16_t efd_size; /* size of this item */ - __uint32_t efd_nextents; /* # of extents freed */ - __uint64_t efd_efi_id; /* id of corresponding efi */ - xfs_extent_64_t efd_extents[1]; /* array of extents freed */ -} xfs_efd_log_format_64_t; - - -#ifdef __KERNEL__ - /* * Max number of extents in fast allocation path. */ @@ -160,6 +78,4 @@ int xfs_efi_copy_format(xfs_log_iovec_t *buf, xfs_efi_log_format_t *dst_efi_fmt); void xfs_efi_item_free(xfs_efi_log_item_t *); -#endif /* __KERNEL__ */ - #endif /* __XFS_EXTFREE_ITEM_H__ */ diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c index de3dc98..4c749ab 100644 --- a/fs/xfs/xfs_file.c +++ b/fs/xfs/xfs_file.c @@ -28,10 +28,11 @@ #include "xfs_inode.h" #include "xfs_inode_item.h" #include "xfs_bmap.h" +#include "xfs_bmap_util.h" #include "xfs_error.h" -#include "xfs_vnodeops.h" #include "xfs_da_btree.h" #include "xfs_dir2_format.h" +#include "xfs_dir2.h" #include "xfs_dir2_priv.h" #include "xfs_ioctl.h" #include "xfs_trace.h" diff --git a/fs/xfs/xfs_filestream.c b/fs/xfs/xfs_filestream.c index 5170306..ce78e65 100644 --- a/fs/xfs/xfs_filestream.c +++ b/fs/xfs/xfs_filestream.c @@ -16,18 +16,18 @@ * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #include "xfs.h" +#include "xfs_log.h" #include "xfs_bmap_btree.h" #include "xfs_inum.h" #include "xfs_dinode.h" #include "xfs_inode.h" #include "xfs_ag.h" -#include "xfs_log.h" #include "xfs_trans.h" #include "xfs_sb.h" #include "xfs_mount.h" #include "xfs_bmap.h" +#include "xfs_bmap_util.h" #include "xfs_alloc.h" -#include "xfs_utils.h" #include "xfs_mru_cache.h" #include "xfs_filestream.h" #include "xfs_trace.h" @@ -668,8 +668,8 @@ exit: */ int xfs_filestream_new_ag( - xfs_bmalloca_t *ap, - xfs_agnumber_t *agp) + struct xfs_bmalloca *ap, + xfs_agnumber_t *agp) { int flags, err; xfs_inode_t *ip, *pip = NULL; diff --git a/fs/xfs/xfs_filestream.h b/fs/xfs/xfs_filestream.h index 09dd9af..6d61dbe 100644 --- a/fs/xfs/xfs_filestream.h +++ b/fs/xfs/xfs_filestream.h @@ -18,8 +18,6 @@ #ifndef __XFS_FILESTREAM_H__ #define __XFS_FILESTREAM_H__ -#ifdef __KERNEL__ - struct xfs_mount; struct xfs_inode; struct xfs_perag; @@ -69,6 +67,4 @@ xfs_inode_is_filestream( (ip->i_d.di_flags & XFS_DIFLAG_FILESTREAM); } -#endif /* __KERNEL__ */ - #endif /* __XFS_FILESTREAM_H__ */ diff --git a/fs/xfs/xfs_format.h b/fs/xfs/xfs_format.h new file mode 100644 index 0000000..35c08ff --- /dev/null +++ b/fs/xfs/xfs_format.h @@ -0,0 +1,169 @@ +/* + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#ifndef __XFS_FORMAT_H__ +#define __XFS_FORMAT_H__ + +/* + * XFS On Disk Format Definitions + * + * This header file defines all the on-disk format definitions for + * general XFS objects. Directory and attribute related objects are defined in + * xfs_da_format.h, which log and log item formats are defined in + * xfs_log_format.h. Everything else goes here. + */ + +struct xfs_mount; +struct xfs_trans; +struct xfs_inode; +struct xfs_buf; +struct xfs_ifork; + +/* + * RealTime Device format definitions + */ + +/* Min and max rt extent sizes, specified in bytes */ +#define XFS_MAX_RTEXTSIZE (1024 * 1024 * 1024) /* 1GB */ +#define XFS_DFL_RTEXTSIZE (64 * 1024) /* 64kB */ +#define XFS_MIN_RTEXTSIZE (4 * 1024) /* 4kB */ + +#define XFS_BLOCKSIZE(mp) ((mp)->m_sb.sb_blocksize) +#define XFS_BLOCKMASK(mp) ((mp)->m_blockmask) +#define XFS_BLOCKWSIZE(mp) ((mp)->m_blockwsize) +#define XFS_BLOCKWMASK(mp) ((mp)->m_blockwmask) + +/* + * RT Summary and bit manipulation macros. + */ +#define XFS_SUMOFFS(mp,ls,bb) ((int)((ls) * (mp)->m_sb.sb_rbmblocks + (bb))) +#define XFS_SUMOFFSTOBLOCK(mp,s) \ + (((s) * (uint)sizeof(xfs_suminfo_t)) >> (mp)->m_sb.sb_blocklog) +#define XFS_SUMPTR(mp,bp,so) \ + ((xfs_suminfo_t *)((bp)->b_addr + \ + (((so) * (uint)sizeof(xfs_suminfo_t)) & XFS_BLOCKMASK(mp)))) + +#define XFS_BITTOBLOCK(mp,bi) ((bi) >> (mp)->m_blkbit_log) +#define XFS_BLOCKTOBIT(mp,bb) ((bb) << (mp)->m_blkbit_log) +#define XFS_BITTOWORD(mp,bi) \ + ((int)(((bi) >> XFS_NBWORDLOG) & XFS_BLOCKWMASK(mp))) + +#define XFS_RTMIN(a,b) ((a) < (b) ? (a) : (b)) +#define XFS_RTMAX(a,b) ((a) > (b) ? (a) : (b)) + +#define XFS_RTLOBIT(w) xfs_lowbit32(w) +#define XFS_RTHIBIT(w) xfs_highbit32(w) + +#if XFS_BIG_BLKNOS +#define XFS_RTBLOCKLOG(b) xfs_highbit64(b) +#else +#define XFS_RTBLOCKLOG(b) xfs_highbit32(b) +#endif + +/* + * Dquot and dquot block format definitions + */ +#define XFS_DQUOT_MAGIC 0x4451 /* 'DQ' */ +#define XFS_DQUOT_VERSION (u_int8_t)0x01 /* latest version number */ + +/* + * This is the main portion of the on-disk representation of quota + * information for a user. This is the q_core of the xfs_dquot_t that + * is kept in kernel memory. We pad this with some more expansion room + * to construct the on disk structure. + */ +typedef struct xfs_disk_dquot { + __be16 d_magic; /* dquot magic = XFS_DQUOT_MAGIC */ + __u8 d_version; /* dquot version */ + __u8 d_flags; /* XFS_DQ_USER/PROJ/GROUP */ + __be32 d_id; /* user,project,group id */ + __be64 d_blk_hardlimit;/* absolute limit on disk blks */ + __be64 d_blk_softlimit;/* preferred limit on disk blks */ + __be64 d_ino_hardlimit;/* maximum # allocated inodes */ + __be64 d_ino_softlimit;/* preferred inode limit */ + __be64 d_bcount; /* disk blocks owned by the user */ + __be64 d_icount; /* inodes owned by the user */ + __be32 d_itimer; /* zero if within inode limits if not, + this is when we refuse service */ + __be32 d_btimer; /* similar to above; for disk blocks */ + __be16 d_iwarns; /* warnings issued wrt num inodes */ + __be16 d_bwarns; /* warnings issued wrt disk blocks */ + __be32 d_pad0; /* 64 bit align */ + __be64 d_rtb_hardlimit;/* absolute limit on realtime blks */ + __be64 d_rtb_softlimit;/* preferred limit on RT disk blks */ + __be64 d_rtbcount; /* realtime blocks owned */ + __be32 d_rtbtimer; /* similar to above; for RT disk blocks */ + __be16 d_rtbwarns; /* warnings issued wrt RT disk blocks */ + __be16 d_pad; +} xfs_disk_dquot_t; + +/* + * This is what goes on disk. This is separated from the xfs_disk_dquot because + * carrying the unnecessary padding would be a waste of memory. + */ +typedef struct xfs_dqblk { + xfs_disk_dquot_t dd_diskdq; /* portion that lives incore as well */ + char dd_fill[4]; /* filling for posterity */ + + /* + * These two are only present on filesystems with the CRC bits set. + */ + __be32 dd_crc; /* checksum */ + __be64 dd_lsn; /* last modification in log */ + uuid_t dd_uuid; /* location information */ +} xfs_dqblk_t; + +#define XFS_DQUOT_CRC_OFF offsetof(struct xfs_dqblk, dd_crc) + +/* + * Remote symlink format and access functions. + */ +#define XFS_SYMLINK_MAGIC 0x58534c4d /* XSLM */ + +struct xfs_dsymlink_hdr { + __be32 sl_magic; + __be32 sl_offset; + __be32 sl_bytes; + __be32 sl_crc; + uuid_t sl_uuid; + __be64 sl_owner; + __be64 sl_blkno; + __be64 sl_lsn; +}; + +/* + * The maximum pathlen is 1024 bytes. Since the minimum file system + * blocksize is 512 bytes, we can get a max of 3 extents back from + * bmapi when crc headers are taken into account. + */ +#define XFS_SYMLINK_MAPS 3 + +#define XFS_SYMLINK_BUF_SPACE(mp, bufsize) \ + ((bufsize) - (xfs_sb_version_hascrc(&(mp)->m_sb) ? \ + sizeof(struct xfs_dsymlink_hdr) : 0)) + +int xfs_symlink_blocks(struct xfs_mount *mp, int pathlen); +int xfs_symlink_hdr_set(struct xfs_mount *mp, xfs_ino_t ino, uint32_t offset, + uint32_t size, struct xfs_buf *bp); +bool xfs_symlink_hdr_ok(struct xfs_mount *mp, xfs_ino_t ino, uint32_t offset, + uint32_t size, struct xfs_buf *bp); +void xfs_symlink_local_to_remote(struct xfs_trans *tp, struct xfs_buf *bp, + struct xfs_inode *ip, struct xfs_ifork *ifp); + +extern const struct xfs_buf_ops xfs_symlink_buf_ops; + +#endif /* __XFS_FORMAT_H__ */ diff --git a/fs/xfs/xfs_fs.h b/fs/xfs/xfs_fs.h index d046955..1edb5cc 100644 --- a/fs/xfs/xfs_fs.h +++ b/fs/xfs/xfs_fs.h @@ -240,7 +240,9 @@ typedef struct xfs_fsop_resblks { /* - * Minimum and maximum sizes need for growth checks + * Minimum and maximum sizes need for growth checks. + * + * Block counts are in units of filesystem blocks, not basic blocks. */ #define XFS_MIN_AG_BLOCKS 64 #define XFS_MIN_LOG_BLOCKS 512ULL @@ -311,6 +313,17 @@ typedef struct xfs_bstat { } xfs_bstat_t; /* + * Project quota id helpers (previously projid was 16bit only + * and using two 16bit values to hold new 32bit projid was choosen + * to retain compatibility with "old" filesystems). + */ +static inline __uint32_t +bstat_get_projid(struct xfs_bstat *bs) +{ + return (__uint32_t)bs->bs_projid_hi << 16 | bs->bs_projid_lo; +} + +/* * The user-level BulkStat Request interface structure. */ typedef struct xfs_fsop_bulkreq { @@ -344,7 +357,7 @@ typedef struct xfs_error_injection { * Speculative preallocation trimming. */ #define XFS_EOFBLOCKS_VERSION 1 -struct xfs_eofblocks { +struct xfs_fs_eofblocks { __u32 eof_version; __u32 eof_flags; uid_t eof_uid; @@ -450,6 +463,21 @@ typedef struct xfs_handle { + (handle).ha_fid.fid_len) /* + * Structure passed to XFS_IOC_SWAPEXT + */ +typedef struct xfs_swapext +{ + __int64_t sx_version; /* version */ +#define XFS_SX_VERSION 0 + __int64_t sx_fdtarget; /* fd of target file */ + __int64_t sx_fdtmp; /* fd of tmp file */ + xfs_off_t sx_offset; /* offset into file */ + xfs_off_t sx_length; /* leng from offset */ + char sx_pad[16]; /* pad space, unused */ + xfs_bstat_t sx_stat; /* stat of target b4 copy */ +} xfs_swapext_t; + +/* * Flags for going down operation */ #define XFS_FSOP_GOING_FLAGS_DEFAULT 0x0 /* going down */ @@ -511,8 +539,14 @@ typedef struct xfs_handle { #define XFS_IOC_ERROR_INJECTION _IOW ('X', 116, struct xfs_error_injection) #define XFS_IOC_ERROR_CLEARALL _IOW ('X', 117, struct xfs_error_injection) /* XFS_IOC_ATTRCTL_BY_HANDLE -- deprecated 118 */ + /* XFS_IOC_FREEZE -- FIFREEZE 119 */ /* XFS_IOC_THAW -- FITHAW 120 */ +#ifndef FIFREEZE +#define XFS_IOC_FREEZE _IOWR('X', 119, int) +#define XFS_IOC_THAW _IOWR('X', 120, int) +#endif + #define XFS_IOC_FSSETDM_BY_HANDLE _IOW ('X', 121, struct xfs_fsop_setdm_handlereq) #define XFS_IOC_ATTRLIST_BY_HANDLE _IOW ('X', 122, struct xfs_fsop_attrlist_handlereq) #define XFS_IOC_ATTRMULTI_BY_HANDLE _IOW ('X', 123, struct xfs_fsop_attrmulti_handlereq) diff --git a/fs/xfs/xfs_fsops.c b/fs/xfs/xfs_fsops.c index 614eb0c..e64ee52 100644 --- a/fs/xfs/xfs_fsops.c +++ b/fs/xfs/xfs_fsops.c @@ -203,8 +203,9 @@ xfs_growfs_data_private( tp = xfs_trans_alloc(mp, XFS_TRANS_GROWFS); tp->t_flags |= XFS_TRANS_RESERVE; - if ((error = xfs_trans_reserve(tp, XFS_GROWFS_SPACE_RES(mp), - XFS_GROWDATA_LOG_RES(mp), 0, 0, 0))) { + error = xfs_trans_reserve(tp, &M_RES(mp)->tr_growdata, + XFS_GROWFS_SPACE_RES(mp), 0); + if (error) { xfs_trans_cancel(tp, 0); return error; } @@ -739,8 +740,7 @@ xfs_fs_log_dummy( int error; tp = _xfs_trans_alloc(mp, XFS_TRANS_DUMMY1, KM_SLEEP); - error = xfs_trans_reserve(tp, 0, XFS_SB_LOG_RES(mp), 0, 0, - XFS_DEFAULT_LOG_COUNT); + error = xfs_trans_reserve(tp, &M_RES(mp)->tr_sb, 0, 0); if (error) { xfs_trans_cancel(tp, 0); return error; diff --git a/fs/xfs/xfs_ialloc.c b/fs/xfs/xfs_ialloc.c index 7a0c17d..ccf2fb1 100644 --- a/fs/xfs/xfs_ialloc.c +++ b/fs/xfs/xfs_ialloc.c @@ -39,6 +39,7 @@ #include "xfs_cksum.h" #include "xfs_buf_item.h" #include "xfs_icreate_item.h" +#include "xfs_icache.h" /* @@ -506,7 +507,7 @@ xfs_ialloc_next_ag( /* * Select an allocation group to look for a free inode in, based on the parent - * inode and then mode. Return the allocation group buffer. + * inode and the mode. Return the allocation group buffer. */ STATIC xfs_agnumber_t xfs_ialloc_ag_select( @@ -728,7 +729,7 @@ xfs_dialloc_ag( error = xfs_inobt_get_rec(cur, &rec, &j); if (error) goto error0; - XFS_WANT_CORRUPTED_GOTO(i == 1, error0); + XFS_WANT_CORRUPTED_GOTO(j == 1, error0); if (rec.ir_freecount > 0) { /* @@ -1341,7 +1342,7 @@ xfs_imap( xfs_agblock_t cluster_agbno; /* first block in inode cluster */ int error; /* error code */ int offset; /* index of inode in its buffer */ - int offset_agbno; /* blks from chunk start to inode */ + xfs_agblock_t offset_agbno; /* blks from chunk start to inode */ ASSERT(ino != NULLFSINO); diff --git a/fs/xfs/xfs_icache.c b/fs/xfs/xfs_icache.c index 3f90e1c..16219b9 100644 --- a/fs/xfs/xfs_icache.c +++ b/fs/xfs/xfs_icache.c @@ -17,6 +17,7 @@ */ #include "xfs.h" #include "xfs_fs.h" +#include "xfs_format.h" #include "xfs_types.h" #include "xfs_log.h" #include "xfs_log_priv.h" @@ -31,12 +32,12 @@ #include "xfs_dinode.h" #include "xfs_error.h" #include "xfs_filestream.h" -#include "xfs_vnodeops.h" #include "xfs_inode_item.h" #include "xfs_quota.h" #include "xfs_trace.h" #include "xfs_fsops.h" #include "xfs_icache.h" +#include "xfs_bmap_util.h" #include <linux/kthread.h> #include <linux/freezer.h> @@ -619,7 +620,7 @@ restart: /* * Background scanning to trim post-EOF preallocated space. This is queued - * based on the 'background_prealloc_discard_period' tunable (5m by default). + * based on the 'speculative_prealloc_lifetime' tunable (5m by default). */ STATIC void xfs_queue_eofblocks( @@ -1203,15 +1204,15 @@ xfs_inode_match_id( struct xfs_inode *ip, struct xfs_eofblocks *eofb) { - if (eofb->eof_flags & XFS_EOF_FLAGS_UID && - ip->i_d.di_uid != eofb->eof_uid) + if ((eofb->eof_flags & XFS_EOF_FLAGS_UID) && + !uid_eq(VFS_I(ip)->i_uid, eofb->eof_uid)) return 0; - if (eofb->eof_flags & XFS_EOF_FLAGS_GID && - ip->i_d.di_gid != eofb->eof_gid) + if ((eofb->eof_flags & XFS_EOF_FLAGS_GID) && + !gid_eq(VFS_I(ip)->i_gid, eofb->eof_gid)) return 0; - if (eofb->eof_flags & XFS_EOF_FLAGS_PRID && + if ((eofb->eof_flags & XFS_EOF_FLAGS_PRID) && xfs_get_projid(ip) != eofb->eof_prid) return 0; diff --git a/fs/xfs/xfs_icache.h b/fs/xfs/xfs_icache.h index a01afbb..8a89f7d 100644 --- a/fs/xfs/xfs_icache.h +++ b/fs/xfs/xfs_icache.h @@ -21,9 +21,24 @@ struct xfs_mount; struct xfs_perag; +struct xfs_eofblocks { + __u32 eof_flags; + kuid_t eof_uid; + kgid_t eof_gid; + prid_t eof_prid; + __u64 eof_min_file_size; +}; + #define SYNC_WAIT 0x0001 /* wait for i/o to complete */ #define SYNC_TRYLOCK 0x0002 /* only try to lock inodes */ +/* + * Flags for xfs_iget() + */ +#define XFS_IGET_CREATE 0x1 +#define XFS_IGET_UNTRUSTED 0x2 +#define XFS_IGET_DONTCACHE 0x4 + int xfs_iget(struct xfs_mount *mp, struct xfs_trans *tp, xfs_ino_t ino, uint flags, uint lock_flags, xfs_inode_t **ipp); @@ -49,4 +64,39 @@ int xfs_inode_ag_iterator_tag(struct xfs_mount *mp, int flags, void *args), int flags, void *args, int tag); +static inline int +xfs_fs_eofblocks_from_user( + struct xfs_fs_eofblocks *src, + struct xfs_eofblocks *dst) +{ + if (src->eof_version != XFS_EOFBLOCKS_VERSION) + return EINVAL; + + if (src->eof_flags & ~XFS_EOF_FLAGS_VALID) + return EINVAL; + + if (memchr_inv(&src->pad32, 0, sizeof(src->pad32)) || + memchr_inv(src->pad64, 0, sizeof(src->pad64))) + return EINVAL; + + dst->eof_flags = src->eof_flags; + dst->eof_prid = src->eof_prid; + dst->eof_min_file_size = src->eof_min_file_size; + + dst->eof_uid = INVALID_UID; + if (src->eof_flags & XFS_EOF_FLAGS_UID) { + dst->eof_uid = make_kuid(current_user_ns(), src->eof_uid); + if (!uid_valid(dst->eof_uid)) + return EINVAL; + } + + dst->eof_gid = INVALID_GID; + if (src->eof_flags & XFS_EOF_FLAGS_GID) { + dst->eof_gid = make_kgid(current_user_ns(), src->eof_gid); + if (!gid_valid(dst->eof_gid)) + return EINVAL; + } + return 0; +} + #endif diff --git a/fs/xfs/xfs_icreate_item.c b/fs/xfs/xfs_icreate_item.c index 7716a4e..5a5a593 100644 --- a/fs/xfs/xfs_icreate_item.c +++ b/fs/xfs/xfs_icreate_item.c @@ -20,23 +20,11 @@ #include "xfs_types.h" #include "xfs_bit.h" #include "xfs_log.h" -#include "xfs_inum.h" #include "xfs_trans.h" -#include "xfs_buf_item.h" #include "xfs_sb.h" #include "xfs_ag.h" -#include "xfs_dir2.h" #include "xfs_mount.h" #include "xfs_trans_priv.h" -#include "xfs_bmap_btree.h" -#include "xfs_alloc_btree.h" -#include "xfs_ialloc_btree.h" -#include "xfs_attr_sf.h" -#include "xfs_dinode.h" -#include "xfs_inode.h" -#include "xfs_inode_item.h" -#include "xfs_btree.h" -#include "xfs_ialloc.h" #include "xfs_error.h" #include "xfs_icreate_item.h" @@ -52,11 +40,14 @@ static inline struct xfs_icreate_item *ICR_ITEM(struct xfs_log_item *lip) * * We only need one iovec for the icreate log structure. */ -STATIC uint +STATIC void xfs_icreate_item_size( - struct xfs_log_item *lip) + struct xfs_log_item *lip, + int *nvecs, + int *nbytes) { - return 1; + *nvecs += 1; + *nbytes += sizeof(struct xfs_icreate_log); } /* diff --git a/fs/xfs/xfs_icreate_item.h b/fs/xfs/xfs_icreate_item.h index 88ba8aa..59e89f8 100644 --- a/fs/xfs/xfs_icreate_item.h +++ b/fs/xfs/xfs_icreate_item.h @@ -18,24 +18,6 @@ #ifndef XFS_ICREATE_ITEM_H #define XFS_ICREATE_ITEM_H 1 -/* - * on disk log item structure - * - * Log recovery assumes the first two entries are the type and size and they fit - * in 32 bits. Also in host order (ugh) so they have to be 32 bit aligned so - * decoding can be done correctly. - */ -struct xfs_icreate_log { - __uint16_t icl_type; /* type of log format structure */ - __uint16_t icl_size; /* size of log format structure */ - __be32 icl_ag; /* ag being allocated in */ - __be32 icl_agbno; /* start block of inode range */ - __be32 icl_count; /* number of inodes to initialise */ - __be32 icl_isize; /* size of inodes */ - __be32 icl_length; /* length of extent to initialise */ - __be32 icl_gen; /* inode generation number to use */ -}; - /* in memory log item structure */ struct xfs_icreate_item { struct xfs_log_item ic_item; diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index bb262c2..e3d7538 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -19,18 +19,23 @@ #include "xfs.h" #include "xfs_fs.h" -#include "xfs_types.h" +#include "xfs_format.h" #include "xfs_log.h" #include "xfs_inum.h" #include "xfs_trans.h" +#include "xfs_trans_space.h" #include "xfs_trans_priv.h" #include "xfs_sb.h" #include "xfs_ag.h" #include "xfs_mount.h" +#include "xfs_da_btree.h" +#include "xfs_dir2_format.h" +#include "xfs_dir2.h" #include "xfs_bmap_btree.h" #include "xfs_alloc_btree.h" #include "xfs_ialloc_btree.h" #include "xfs_attr_sf.h" +#include "xfs_attr.h" #include "xfs_dinode.h" #include "xfs_inode.h" #include "xfs_buf_item.h" @@ -39,16 +44,15 @@ #include "xfs_alloc.h" #include "xfs_ialloc.h" #include "xfs_bmap.h" +#include "xfs_bmap_util.h" #include "xfs_error.h" -#include "xfs_utils.h" #include "xfs_quota.h" #include "xfs_filestream.h" -#include "xfs_vnodeops.h" #include "xfs_cksum.h" #include "xfs_trace.h" #include "xfs_icache.h" +#include "xfs_symlink.h" -kmem_zone_t *xfs_ifork_zone; kmem_zone_t *xfs_inode_zone; /* @@ -58,9 +62,6 @@ kmem_zone_t *xfs_inode_zone; #define XFS_ITRUNC_MAX_EXTENTS 2 STATIC int xfs_iflush_int(xfs_inode_t *, xfs_buf_t *); -STATIC int xfs_iformat_local(xfs_inode_t *, xfs_dinode_t *, int, int); -STATIC int xfs_iformat_extents(xfs_inode_t *, xfs_dinode_t *, int); -STATIC int xfs_iformat_btree(xfs_inode_t *, xfs_dinode_t *, int); /* * helper function to extract extent size hint from inode @@ -310,623 +311,202 @@ xfs_isilocked( } #endif -void -__xfs_iflock( - struct xfs_inode *ip) -{ - wait_queue_head_t *wq = bit_waitqueue(&ip->i_flags, __XFS_IFLOCK_BIT); - DEFINE_WAIT_BIT(wait, &ip->i_flags, __XFS_IFLOCK_BIT); - - do { - prepare_to_wait_exclusive(wq, &wait.wait, TASK_UNINTERRUPTIBLE); - if (xfs_isiflocked(ip)) - io_schedule(); - } while (!xfs_iflock_nowait(ip)); - - finish_wait(wq, &wait.wait); -} - #ifdef DEBUG +int xfs_locked_n; +int xfs_small_retries; +int xfs_middle_retries; +int xfs_lots_retries; +int xfs_lock_delays; +#endif + /* - * Make sure that the extents in the given memory buffer - * are valid. + * Bump the subclass so xfs_lock_inodes() acquires each lock with + * a different value */ -STATIC void -xfs_validate_extents( - xfs_ifork_t *ifp, - int nrecs, - xfs_exntfmt_t fmt) +static inline int +xfs_lock_inumorder(int lock_mode, int subclass) { - xfs_bmbt_irec_t irec; - xfs_bmbt_rec_host_t rec; - int i; + if (lock_mode & (XFS_IOLOCK_SHARED|XFS_IOLOCK_EXCL)) + lock_mode |= (subclass + XFS_LOCK_INUMORDER) << XFS_IOLOCK_SHIFT; + if (lock_mode & (XFS_ILOCK_SHARED|XFS_ILOCK_EXCL)) + lock_mode |= (subclass + XFS_LOCK_INUMORDER) << XFS_ILOCK_SHIFT; - for (i = 0; i < nrecs; i++) { - xfs_bmbt_rec_host_t *ep = xfs_iext_get_ext(ifp, i); - rec.l0 = get_unaligned(&ep->l0); - rec.l1 = get_unaligned(&ep->l1); - xfs_bmbt_get_all(&rec, &irec); - if (fmt == XFS_EXTFMT_NOSTATE) - ASSERT(irec.br_state == XFS_EXT_NORM); - } + return lock_mode; } -#else /* DEBUG */ -#define xfs_validate_extents(ifp, nrecs, fmt) -#endif /* DEBUG */ /* - * Check that none of the inode's in the buffer have a next - * unlinked field of 0. + * The following routine will lock n inodes in exclusive mode. + * We assume the caller calls us with the inodes in i_ino order. + * + * We need to detect deadlock where an inode that we lock + * is in the AIL and we start waiting for another inode that is locked + * by a thread in a long running transaction (such as truncate). This can + * result in deadlock since the long running trans might need to wait + * for the inode we just locked in order to push the tail and free space + * in the log. */ -#if defined(DEBUG) void -xfs_inobp_check( - xfs_mount_t *mp, - xfs_buf_t *bp) +xfs_lock_inodes( + xfs_inode_t **ips, + int inodes, + uint lock_mode) { - int i; - int j; - xfs_dinode_t *dip; + int attempts = 0, i, j, try_lock; + xfs_log_item_t *lp; - j = mp->m_inode_cluster_size >> mp->m_sb.sb_inodelog; + ASSERT(ips && (inodes >= 2)); /* we need at least two */ - for (i = 0; i < j; i++) { - dip = (xfs_dinode_t *)xfs_buf_offset(bp, - i * mp->m_sb.sb_inodesize); - if (!dip->di_next_unlinked) { - xfs_alert(mp, - "Detected bogus zero next_unlinked field in incore inode buffer 0x%p.", - bp); - ASSERT(dip->di_next_unlinked); - } - } -} -#endif + try_lock = 0; + i = 0; -static void -xfs_inode_buf_verify( - struct xfs_buf *bp) -{ - struct xfs_mount *mp = bp->b_target->bt_mount; - int i; - int ni; - - /* - * Validate the magic number and version of every inode in the buffer - */ - ni = XFS_BB_TO_FSB(mp, bp->b_length) * mp->m_sb.sb_inopblock; - for (i = 0; i < ni; i++) { - int di_ok; - xfs_dinode_t *dip; - - dip = (struct xfs_dinode *)xfs_buf_offset(bp, - (i << mp->m_sb.sb_inodelog)); - di_ok = dip->di_magic == cpu_to_be16(XFS_DINODE_MAGIC) && - XFS_DINODE_GOOD_VERSION(dip->di_version); - if (unlikely(XFS_TEST_ERROR(!di_ok, mp, - XFS_ERRTAG_ITOBP_INOTOBP, - XFS_RANDOM_ITOBP_INOTOBP))) { - xfs_buf_ioerror(bp, EFSCORRUPTED); - XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_HIGH, - mp, dip); -#ifdef DEBUG - xfs_emerg(mp, - "bad inode magic/vsn daddr %lld #%d (magic=%x)", - (unsigned long long)bp->b_bn, i, - be16_to_cpu(dip->di_magic)); - ASSERT(0); -#endif - } - } - xfs_inobp_check(mp, bp); -} - - -static void -xfs_inode_buf_read_verify( - struct xfs_buf *bp) -{ - xfs_inode_buf_verify(bp); -} - -static void -xfs_inode_buf_write_verify( - struct xfs_buf *bp) -{ - xfs_inode_buf_verify(bp); -} - -const struct xfs_buf_ops xfs_inode_buf_ops = { - .verify_read = xfs_inode_buf_read_verify, - .verify_write = xfs_inode_buf_write_verify, -}; +again: + for (; i < inodes; i++) { + ASSERT(ips[i]); + if (i && (ips[i] == ips[i-1])) /* Already locked */ + continue; -/* - * This routine is called to map an inode to the buffer containing the on-disk - * version of the inode. It returns a pointer to the buffer containing the - * on-disk inode in the bpp parameter, and in the dipp parameter it returns a - * pointer to the on-disk inode within that buffer. - * - * If a non-zero error is returned, then the contents of bpp and dipp are - * undefined. - */ -int -xfs_imap_to_bp( - struct xfs_mount *mp, - struct xfs_trans *tp, - struct xfs_imap *imap, - struct xfs_dinode **dipp, - struct xfs_buf **bpp, - uint buf_flags, - uint iget_flags) -{ - struct xfs_buf *bp; - int error; + /* + * If try_lock is not set yet, make sure all locked inodes + * are not in the AIL. + * If any are, set try_lock to be used later. + */ - buf_flags |= XBF_UNMAPPED; - error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp, imap->im_blkno, - (int)imap->im_len, buf_flags, &bp, - &xfs_inode_buf_ops); - if (error) { - if (error == EAGAIN) { - ASSERT(buf_flags & XBF_TRYLOCK); - return error; + if (!try_lock) { + for (j = (i - 1); j >= 0 && !try_lock; j--) { + lp = (xfs_log_item_t *)ips[j]->i_itemp; + if (lp && (lp->li_flags & XFS_LI_IN_AIL)) { + try_lock++; + } + } } - if (error == EFSCORRUPTED && - (iget_flags & XFS_IGET_UNTRUSTED)) - return XFS_ERROR(EINVAL); - - xfs_warn(mp, "%s: xfs_trans_read_buf() returned error %d.", - __func__, error); - return error; - } - - *bpp = bp; - *dipp = (struct xfs_dinode *)xfs_buf_offset(bp, imap->im_boffset); - return 0; -} - -/* - * Move inode type and inode format specific information from the - * on-disk inode to the in-core inode. For fifos, devs, and sockets - * this means set if_rdev to the proper value. For files, directories, - * and symlinks this means to bring in the in-line data or extent - * pointers. For a file in B-tree format, only the root is immediately - * brought in-core. The rest will be in-lined in if_extents when it - * is first referenced (see xfs_iread_extents()). - */ -STATIC int -xfs_iformat( - xfs_inode_t *ip, - xfs_dinode_t *dip) -{ - xfs_attr_shortform_t *atp; - int size; - int error = 0; - xfs_fsize_t di_size; - - if (unlikely(be32_to_cpu(dip->di_nextents) + - be16_to_cpu(dip->di_anextents) > - be64_to_cpu(dip->di_nblocks))) { - xfs_warn(ip->i_mount, - "corrupt dinode %Lu, extent total = %d, nblocks = %Lu.", - (unsigned long long)ip->i_ino, - (int)(be32_to_cpu(dip->di_nextents) + - be16_to_cpu(dip->di_anextents)), - (unsigned long long) - be64_to_cpu(dip->di_nblocks)); - XFS_CORRUPTION_ERROR("xfs_iformat(1)", XFS_ERRLEVEL_LOW, - ip->i_mount, dip); - return XFS_ERROR(EFSCORRUPTED); - } - - if (unlikely(dip->di_forkoff > ip->i_mount->m_sb.sb_inodesize)) { - xfs_warn(ip->i_mount, "corrupt dinode %Lu, forkoff = 0x%x.", - (unsigned long long)ip->i_ino, - dip->di_forkoff); - XFS_CORRUPTION_ERROR("xfs_iformat(2)", XFS_ERRLEVEL_LOW, - ip->i_mount, dip); - return XFS_ERROR(EFSCORRUPTED); - } - - if (unlikely((ip->i_d.di_flags & XFS_DIFLAG_REALTIME) && - !ip->i_mount->m_rtdev_targp)) { - xfs_warn(ip->i_mount, - "corrupt dinode %Lu, has realtime flag set.", - ip->i_ino); - XFS_CORRUPTION_ERROR("xfs_iformat(realtime)", - XFS_ERRLEVEL_LOW, ip->i_mount, dip); - return XFS_ERROR(EFSCORRUPTED); - } - - switch (ip->i_d.di_mode & S_IFMT) { - case S_IFIFO: - case S_IFCHR: - case S_IFBLK: - case S_IFSOCK: - if (unlikely(dip->di_format != XFS_DINODE_FMT_DEV)) { - XFS_CORRUPTION_ERROR("xfs_iformat(3)", XFS_ERRLEVEL_LOW, - ip->i_mount, dip); - return XFS_ERROR(EFSCORRUPTED); - } - ip->i_d.di_size = 0; - ip->i_df.if_u2.if_rdev = xfs_dinode_get_rdev(dip); - break; + /* + * If any of the previous locks we have locked is in the AIL, + * we must TRY to get the second and subsequent locks. If + * we can't get any, we must release all we have + * and try again. + */ - case S_IFREG: - case S_IFLNK: - case S_IFDIR: - switch (dip->di_format) { - case XFS_DINODE_FMT_LOCAL: + if (try_lock) { + /* try_lock must be 0 if i is 0. */ /* - * no local regular files yet + * try_lock means we have an inode locked + * that is in the AIL. */ - if (unlikely(S_ISREG(be16_to_cpu(dip->di_mode)))) { - xfs_warn(ip->i_mount, - "corrupt inode %Lu (local format for regular file).", - (unsigned long long) ip->i_ino); - XFS_CORRUPTION_ERROR("xfs_iformat(4)", - XFS_ERRLEVEL_LOW, - ip->i_mount, dip); - return XFS_ERROR(EFSCORRUPTED); - } + ASSERT(i != 0); + if (!xfs_ilock_nowait(ips[i], xfs_lock_inumorder(lock_mode, i))) { + attempts++; + + /* + * Unlock all previous guys and try again. + * xfs_iunlock will try to push the tail + * if the inode is in the AIL. + */ + + for(j = i - 1; j >= 0; j--) { + + /* + * Check to see if we've already + * unlocked this one. + * Not the first one going back, + * and the inode ptr is the same. + */ + if ((j != (i - 1)) && ips[j] == + ips[j+1]) + continue; + + xfs_iunlock(ips[j], lock_mode); + } - di_size = be64_to_cpu(dip->di_size); - if (unlikely(di_size > XFS_DFORK_DSIZE(dip, ip->i_mount))) { - xfs_warn(ip->i_mount, - "corrupt inode %Lu (bad size %Ld for local inode).", - (unsigned long long) ip->i_ino, - (long long) di_size); - XFS_CORRUPTION_ERROR("xfs_iformat(5)", - XFS_ERRLEVEL_LOW, - ip->i_mount, dip); - return XFS_ERROR(EFSCORRUPTED); + if ((attempts % 5) == 0) { + delay(1); /* Don't just spin the CPU */ +#ifdef DEBUG + xfs_lock_delays++; +#endif + } + i = 0; + try_lock = 0; + goto again; } - - size = (int)di_size; - error = xfs_iformat_local(ip, dip, XFS_DATA_FORK, size); - break; - case XFS_DINODE_FMT_EXTENTS: - error = xfs_iformat_extents(ip, dip, XFS_DATA_FORK); - break; - case XFS_DINODE_FMT_BTREE: - error = xfs_iformat_btree(ip, dip, XFS_DATA_FORK); - break; - default: - XFS_ERROR_REPORT("xfs_iformat(6)", XFS_ERRLEVEL_LOW, - ip->i_mount); - return XFS_ERROR(EFSCORRUPTED); + } else { + xfs_ilock(ips[i], xfs_lock_inumorder(lock_mode, i)); } - break; - - default: - XFS_ERROR_REPORT("xfs_iformat(7)", XFS_ERRLEVEL_LOW, ip->i_mount); - return XFS_ERROR(EFSCORRUPTED); - } - if (error) { - return error; } - if (!XFS_DFORK_Q(dip)) - return 0; - - ASSERT(ip->i_afp == NULL); - ip->i_afp = kmem_zone_zalloc(xfs_ifork_zone, KM_SLEEP | KM_NOFS); - - switch (dip->di_aformat) { - case XFS_DINODE_FMT_LOCAL: - atp = (xfs_attr_shortform_t *)XFS_DFORK_APTR(dip); - size = be16_to_cpu(atp->hdr.totsize); - - if (unlikely(size < sizeof(struct xfs_attr_sf_hdr))) { - xfs_warn(ip->i_mount, - "corrupt inode %Lu (bad attr fork size %Ld).", - (unsigned long long) ip->i_ino, - (long long) size); - XFS_CORRUPTION_ERROR("xfs_iformat(8)", - XFS_ERRLEVEL_LOW, - ip->i_mount, dip); - return XFS_ERROR(EFSCORRUPTED); - } - error = xfs_iformat_local(ip, dip, XFS_ATTR_FORK, size); - break; - case XFS_DINODE_FMT_EXTENTS: - error = xfs_iformat_extents(ip, dip, XFS_ATTR_FORK); - break; - case XFS_DINODE_FMT_BTREE: - error = xfs_iformat_btree(ip, dip, XFS_ATTR_FORK); - break; - default: - error = XFS_ERROR(EFSCORRUPTED); - break; - } - if (error) { - kmem_zone_free(xfs_ifork_zone, ip->i_afp); - ip->i_afp = NULL; - xfs_idestroy_fork(ip, XFS_DATA_FORK); +#ifdef DEBUG + if (attempts) { + if (attempts < 5) xfs_small_retries++; + else if (attempts < 100) xfs_middle_retries++; + else xfs_lots_retries++; + } else { + xfs_locked_n++; } - return error; +#endif } /* - * The file is in-lined in the on-disk inode. - * If it fits into if_inline_data, then copy - * it there, otherwise allocate a buffer for it - * and copy the data there. Either way, set - * if_data to point at the data. - * If we allocate a buffer for the data, make - * sure that its size is a multiple of 4 and - * record the real size in i_real_bytes. + * xfs_lock_two_inodes() can only be used to lock one type of lock + * at a time - the iolock or the ilock, but not both at once. If + * we lock both at once, lockdep will report false positives saying + * we have violated locking orders. */ -STATIC int -xfs_iformat_local( - xfs_inode_t *ip, - xfs_dinode_t *dip, - int whichfork, - int size) +void +xfs_lock_two_inodes( + xfs_inode_t *ip0, + xfs_inode_t *ip1, + uint lock_mode) { - xfs_ifork_t *ifp; - int real_size; - - /* - * If the size is unreasonable, then something - * is wrong and we just bail out rather than crash in - * kmem_alloc() or memcpy() below. - */ - if (unlikely(size > XFS_DFORK_SIZE(dip, ip->i_mount, whichfork))) { - xfs_warn(ip->i_mount, - "corrupt inode %Lu (bad size %d for local fork, size = %d).", - (unsigned long long) ip->i_ino, size, - XFS_DFORK_SIZE(dip, ip->i_mount, whichfork)); - XFS_CORRUPTION_ERROR("xfs_iformat_local", XFS_ERRLEVEL_LOW, - ip->i_mount, dip); - return XFS_ERROR(EFSCORRUPTED); - } - ifp = XFS_IFORK_PTR(ip, whichfork); - real_size = 0; - if (size == 0) - ifp->if_u1.if_data = NULL; - else if (size <= sizeof(ifp->if_u2.if_inline_data)) - ifp->if_u1.if_data = ifp->if_u2.if_inline_data; - else { - real_size = roundup(size, 4); - ifp->if_u1.if_data = kmem_alloc(real_size, KM_SLEEP | KM_NOFS); - } - ifp->if_bytes = size; - ifp->if_real_bytes = real_size; - if (size) - memcpy(ifp->if_u1.if_data, XFS_DFORK_PTR(dip, whichfork), size); - ifp->if_flags &= ~XFS_IFEXTENTS; - ifp->if_flags |= XFS_IFINLINE; - return 0; -} + xfs_inode_t *temp; + int attempts = 0; + xfs_log_item_t *lp; -/* - * The file consists of a set of extents all - * of which fit into the on-disk inode. - * If there are few enough extents to fit into - * the if_inline_ext, then copy them there. - * Otherwise allocate a buffer for them and copy - * them into it. Either way, set if_extents - * to point at the extents. - */ -STATIC int -xfs_iformat_extents( - xfs_inode_t *ip, - xfs_dinode_t *dip, - int whichfork) -{ - xfs_bmbt_rec_t *dp; - xfs_ifork_t *ifp; - int nex; - int size; - int i; - - ifp = XFS_IFORK_PTR(ip, whichfork); - nex = XFS_DFORK_NEXTENTS(dip, whichfork); - size = nex * (uint)sizeof(xfs_bmbt_rec_t); - - /* - * If the number of extents is unreasonable, then something - * is wrong and we just bail out rather than crash in - * kmem_alloc() or memcpy() below. - */ - if (unlikely(size < 0 || size > XFS_DFORK_SIZE(dip, ip->i_mount, whichfork))) { - xfs_warn(ip->i_mount, "corrupt inode %Lu ((a)extents = %d).", - (unsigned long long) ip->i_ino, nex); - XFS_CORRUPTION_ERROR("xfs_iformat_extents(1)", XFS_ERRLEVEL_LOW, - ip->i_mount, dip); - return XFS_ERROR(EFSCORRUPTED); - } - - ifp->if_real_bytes = 0; - if (nex == 0) - ifp->if_u1.if_extents = NULL; - else if (nex <= XFS_INLINE_EXTS) - ifp->if_u1.if_extents = ifp->if_u2.if_inline_ext; - else - xfs_iext_add(ifp, 0, nex); - - ifp->if_bytes = size; - if (size) { - dp = (xfs_bmbt_rec_t *) XFS_DFORK_PTR(dip, whichfork); - xfs_validate_extents(ifp, nex, XFS_EXTFMT_INODE(ip)); - for (i = 0; i < nex; i++, dp++) { - xfs_bmbt_rec_host_t *ep = xfs_iext_get_ext(ifp, i); - ep->l0 = get_unaligned_be64(&dp->l0); - ep->l1 = get_unaligned_be64(&dp->l1); - } - XFS_BMAP_TRACE_EXLIST(ip, nex, whichfork); - if (whichfork != XFS_DATA_FORK || - XFS_EXTFMT_INODE(ip) == XFS_EXTFMT_NOSTATE) - if (unlikely(xfs_check_nostate_extents( - ifp, 0, nex))) { - XFS_ERROR_REPORT("xfs_iformat_extents(2)", - XFS_ERRLEVEL_LOW, - ip->i_mount); - return XFS_ERROR(EFSCORRUPTED); - } - } - ifp->if_flags |= XFS_IFEXTENTS; - return 0; -} + if (lock_mode & (XFS_IOLOCK_SHARED|XFS_IOLOCK_EXCL)) + ASSERT((lock_mode & (XFS_ILOCK_SHARED|XFS_ILOCK_EXCL)) == 0); + ASSERT(ip0->i_ino != ip1->i_ino); -/* - * The file has too many extents to fit into - * the inode, so they are in B-tree format. - * Allocate a buffer for the root of the B-tree - * and copy the root into it. The i_extents - * field will remain NULL until all of the - * extents are read in (when they are needed). - */ -STATIC int -xfs_iformat_btree( - xfs_inode_t *ip, - xfs_dinode_t *dip, - int whichfork) -{ - struct xfs_mount *mp = ip->i_mount; - xfs_bmdr_block_t *dfp; - xfs_ifork_t *ifp; - /* REFERENCED */ - int nrecs; - int size; - - ifp = XFS_IFORK_PTR(ip, whichfork); - dfp = (xfs_bmdr_block_t *)XFS_DFORK_PTR(dip, whichfork); - size = XFS_BMAP_BROOT_SPACE(mp, dfp); - nrecs = be16_to_cpu(dfp->bb_numrecs); - - /* - * blow out if -- fork has less extents than can fit in - * fork (fork shouldn't be a btree format), root btree - * block has more records than can fit into the fork, - * or the number of extents is greater than the number of - * blocks. - */ - if (unlikely(XFS_IFORK_NEXTENTS(ip, whichfork) <= - XFS_IFORK_MAXEXT(ip, whichfork) || - XFS_BMDR_SPACE_CALC(nrecs) > - XFS_DFORK_SIZE(dip, mp, whichfork) || - XFS_IFORK_NEXTENTS(ip, whichfork) > ip->i_d.di_nblocks)) { - xfs_warn(mp, "corrupt inode %Lu (btree).", - (unsigned long long) ip->i_ino); - XFS_CORRUPTION_ERROR("xfs_iformat_btree", XFS_ERRLEVEL_LOW, - mp, dip); - return XFS_ERROR(EFSCORRUPTED); - } - - ifp->if_broot_bytes = size; - ifp->if_broot = kmem_alloc(size, KM_SLEEP | KM_NOFS); - ASSERT(ifp->if_broot != NULL); - /* - * Copy and convert from the on-disk structure - * to the in-memory structure. - */ - xfs_bmdr_to_bmbt(ip, dfp, XFS_DFORK_SIZE(dip, ip->i_mount, whichfork), - ifp->if_broot, size); - ifp->if_flags &= ~XFS_IFEXTENTS; - ifp->if_flags |= XFS_IFBROOT; + if (ip0->i_ino > ip1->i_ino) { + temp = ip0; + ip0 = ip1; + ip1 = temp; + } - return 0; -} + again: + xfs_ilock(ip0, xfs_lock_inumorder(lock_mode, 0)); -STATIC void -xfs_dinode_from_disk( - xfs_icdinode_t *to, - xfs_dinode_t *from) -{ - to->di_magic = be16_to_cpu(from->di_magic); - to->di_mode = be16_to_cpu(from->di_mode); - to->di_version = from ->di_version; - to->di_format = from->di_format; - to->di_onlink = be16_to_cpu(from->di_onlink); - to->di_uid = be32_to_cpu(from->di_uid); - to->di_gid = be32_to_cpu(from->di_gid); - to->di_nlink = be32_to_cpu(from->di_nlink); - to->di_projid_lo = be16_to_cpu(from->di_projid_lo); - to->di_projid_hi = be16_to_cpu(from->di_projid_hi); - memcpy(to->di_pad, from->di_pad, sizeof(to->di_pad)); - to->di_flushiter = be16_to_cpu(from->di_flushiter); - to->di_atime.t_sec = be32_to_cpu(from->di_atime.t_sec); - to->di_atime.t_nsec = be32_to_cpu(from->di_atime.t_nsec); - to->di_mtime.t_sec = be32_to_cpu(from->di_mtime.t_sec); - to->di_mtime.t_nsec = be32_to_cpu(from->di_mtime.t_nsec); - to->di_ctime.t_sec = be32_to_cpu(from->di_ctime.t_sec); - to->di_ctime.t_nsec = be32_to_cpu(from->di_ctime.t_nsec); - to->di_size = be64_to_cpu(from->di_size); - to->di_nblocks = be64_to_cpu(from->di_nblocks); - to->di_extsize = be32_to_cpu(from->di_extsize); - to->di_nextents = be32_to_cpu(from->di_nextents); - to->di_anextents = be16_to_cpu(from->di_anextents); - to->di_forkoff = from->di_forkoff; - to->di_aformat = from->di_aformat; - to->di_dmevmask = be32_to_cpu(from->di_dmevmask); - to->di_dmstate = be16_to_cpu(from->di_dmstate); - to->di_flags = be16_to_cpu(from->di_flags); - to->di_gen = be32_to_cpu(from->di_gen); - - if (to->di_version == 3) { - to->di_changecount = be64_to_cpu(from->di_changecount); - to->di_crtime.t_sec = be32_to_cpu(from->di_crtime.t_sec); - to->di_crtime.t_nsec = be32_to_cpu(from->di_crtime.t_nsec); - to->di_flags2 = be64_to_cpu(from->di_flags2); - to->di_ino = be64_to_cpu(from->di_ino); - to->di_lsn = be64_to_cpu(from->di_lsn); - memcpy(to->di_pad2, from->di_pad2, sizeof(to->di_pad2)); - uuid_copy(&to->di_uuid, &from->di_uuid); + /* + * If the first lock we have locked is in the AIL, we must TRY to get + * the second lock. If we can't get it, we must release the first one + * and try again. + */ + lp = (xfs_log_item_t *)ip0->i_itemp; + if (lp && (lp->li_flags & XFS_LI_IN_AIL)) { + if (!xfs_ilock_nowait(ip1, xfs_lock_inumorder(lock_mode, 1))) { + xfs_iunlock(ip0, lock_mode); + if ((++attempts % 5) == 0) + delay(1); /* Don't just spin the CPU */ + goto again; + } + } else { + xfs_ilock(ip1, xfs_lock_inumorder(lock_mode, 1)); } } + void -xfs_dinode_to_disk( - xfs_dinode_t *to, - xfs_icdinode_t *from) +__xfs_iflock( + struct xfs_inode *ip) { - to->di_magic = cpu_to_be16(from->di_magic); - to->di_mode = cpu_to_be16(from->di_mode); - to->di_version = from ->di_version; - to->di_format = from->di_format; - to->di_onlink = cpu_to_be16(from->di_onlink); - to->di_uid = cpu_to_be32(from->di_uid); - to->di_gid = cpu_to_be32(from->di_gid); - to->di_nlink = cpu_to_be32(from->di_nlink); - to->di_projid_lo = cpu_to_be16(from->di_projid_lo); - to->di_projid_hi = cpu_to_be16(from->di_projid_hi); - memcpy(to->di_pad, from->di_pad, sizeof(to->di_pad)); - to->di_atime.t_sec = cpu_to_be32(from->di_atime.t_sec); - to->di_atime.t_nsec = cpu_to_be32(from->di_atime.t_nsec); - to->di_mtime.t_sec = cpu_to_be32(from->di_mtime.t_sec); - to->di_mtime.t_nsec = cpu_to_be32(from->di_mtime.t_nsec); - to->di_ctime.t_sec = cpu_to_be32(from->di_ctime.t_sec); - to->di_ctime.t_nsec = cpu_to_be32(from->di_ctime.t_nsec); - to->di_size = cpu_to_be64(from->di_size); - to->di_nblocks = cpu_to_be64(from->di_nblocks); - to->di_extsize = cpu_to_be32(from->di_extsize); - to->di_nextents = cpu_to_be32(from->di_nextents); - to->di_anextents = cpu_to_be16(from->di_anextents); - to->di_forkoff = from->di_forkoff; - to->di_aformat = from->di_aformat; - to->di_dmevmask = cpu_to_be32(from->di_dmevmask); - to->di_dmstate = cpu_to_be16(from->di_dmstate); - to->di_flags = cpu_to_be16(from->di_flags); - to->di_gen = cpu_to_be32(from->di_gen); - - if (from->di_version == 3) { - to->di_changecount = cpu_to_be64(from->di_changecount); - to->di_crtime.t_sec = cpu_to_be32(from->di_crtime.t_sec); - to->di_crtime.t_nsec = cpu_to_be32(from->di_crtime.t_nsec); - to->di_flags2 = cpu_to_be64(from->di_flags2); - to->di_ino = cpu_to_be64(from->di_ino); - to->di_lsn = cpu_to_be64(from->di_lsn); - memcpy(to->di_pad2, from->di_pad2, sizeof(to->di_pad2)); - uuid_copy(&to->di_uuid, &from->di_uuid); - to->di_flushiter = 0; - } else { - to->di_flushiter = cpu_to_be16(from->di_flushiter); - } + wait_queue_head_t *wq = bit_waitqueue(&ip->i_flags, __XFS_IFLOCK_BIT); + DEFINE_WAIT_BIT(wait, &ip->i_flags, __XFS_IFLOCK_BIT); + + do { + prepare_to_wait_exclusive(wq, &wait.wait, TASK_UNINTERRUPTIBLE); + if (xfs_isiflocked(ip)) + io_schedule(); + } while (!xfs_iflock_nowait(ip)); + + finish_wait(wq, &wait.wait); } STATIC uint @@ -987,235 +567,50 @@ xfs_dic2xflags( (XFS_DFORK_Q(dip) ? XFS_XFLAG_HASATTR : 0); } -static bool -xfs_dinode_verify( - struct xfs_mount *mp, - struct xfs_inode *ip, - struct xfs_dinode *dip) -{ - if (dip->di_magic != cpu_to_be16(XFS_DINODE_MAGIC)) - return false; - - /* only version 3 or greater inodes are extensively verified here */ - if (dip->di_version < 3) - return true; - - if (!xfs_sb_version_hascrc(&mp->m_sb)) - return false; - if (!xfs_verify_cksum((char *)dip, mp->m_sb.sb_inodesize, - offsetof(struct xfs_dinode, di_crc))) - return false; - if (be64_to_cpu(dip->di_ino) != ip->i_ino) - return false; - if (!uuid_equal(&dip->di_uuid, &mp->m_sb.sb_uuid)) - return false; - return true; -} - -void -xfs_dinode_calc_crc( - struct xfs_mount *mp, - struct xfs_dinode *dip) -{ - __uint32_t crc; - - if (dip->di_version < 3) - return; - - ASSERT(xfs_sb_version_hascrc(&mp->m_sb)); - crc = xfs_start_cksum((char *)dip, mp->m_sb.sb_inodesize, - offsetof(struct xfs_dinode, di_crc)); - dip->di_crc = xfs_end_cksum(crc); -} - /* - * Read the disk inode attributes into the in-core inode structure. - * - * For version 5 superblocks, if we are initialising a new inode and we are not - * utilising the XFS_MOUNT_IKEEP inode cluster mode, we can simple build the new - * inode core with a random generation number. If we are keeping inodes around, - * we need to read the inode cluster to get the existing generation number off - * disk. Further, if we are using version 4 superblocks (i.e. v1/v2 inode - * format) then log recovery is dependent on the di_flushiter field being - * initialised from the current on-disk value and hence we must also read the - * inode off disk. + * Lookups up an inode from "name". If ci_name is not NULL, then a CI match + * is allowed, otherwise it has to be an exact match. If a CI match is found, + * ci_name->name will point to a the actual name (caller must free) or + * will be set to NULL if an exact match is found. */ int -xfs_iread( - xfs_mount_t *mp, - xfs_trans_t *tp, - xfs_inode_t *ip, - uint iget_flags) +xfs_lookup( + xfs_inode_t *dp, + struct xfs_name *name, + xfs_inode_t **ipp, + struct xfs_name *ci_name) { - xfs_buf_t *bp; - xfs_dinode_t *dip; - int error; - - /* - * Fill in the location information in the in-core inode. - */ - error = xfs_imap(mp, tp, ip->i_ino, &ip->i_imap, iget_flags); - if (error) - return error; - - /* shortcut IO on inode allocation if possible */ - if ((iget_flags & XFS_IGET_CREATE) && - xfs_sb_version_hascrc(&mp->m_sb) && - !(mp->m_flags & XFS_MOUNT_IKEEP)) { - /* initialise the on-disk inode core */ - memset(&ip->i_d, 0, sizeof(ip->i_d)); - ip->i_d.di_magic = XFS_DINODE_MAGIC; - ip->i_d.di_gen = prandom_u32(); - if (xfs_sb_version_hascrc(&mp->m_sb)) { - ip->i_d.di_version = 3; - ip->i_d.di_ino = ip->i_ino; - uuid_copy(&ip->i_d.di_uuid, &mp->m_sb.sb_uuid); - } else - ip->i_d.di_version = 2; - return 0; - } - - /* - * Get pointers to the on-disk inode and the buffer containing it. - */ - error = xfs_imap_to_bp(mp, tp, &ip->i_imap, &dip, &bp, 0, iget_flags); - if (error) - return error; + xfs_ino_t inum; + int error; + uint lock_mode; - /* even unallocated inodes are verified */ - if (!xfs_dinode_verify(mp, ip, dip)) { - xfs_alert(mp, "%s: validation failed for inode %lld failed", - __func__, ip->i_ino); + trace_xfs_lookup(dp, name); - XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, dip); - error = XFS_ERROR(EFSCORRUPTED); - goto out_brelse; - } + if (XFS_FORCED_SHUTDOWN(dp->i_mount)) + return XFS_ERROR(EIO); - /* - * If the on-disk inode is already linked to a directory - * entry, copy all of the inode into the in-core inode. - * xfs_iformat() handles copying in the inode format - * specific information. - * Otherwise, just get the truly permanent information. - */ - if (dip->di_mode) { - xfs_dinode_from_disk(&ip->i_d, dip); - error = xfs_iformat(ip, dip); - if (error) { -#ifdef DEBUG - xfs_alert(mp, "%s: xfs_iformat() returned error %d", - __func__, error); -#endif /* DEBUG */ - goto out_brelse; - } - } else { - /* - * Partial initialisation of the in-core inode. Just the bits - * that xfs_ialloc won't overwrite or relies on being correct. - */ - ip->i_d.di_magic = be16_to_cpu(dip->di_magic); - ip->i_d.di_version = dip->di_version; - ip->i_d.di_gen = be32_to_cpu(dip->di_gen); - ip->i_d.di_flushiter = be16_to_cpu(dip->di_flushiter); - - if (dip->di_version == 3) { - ip->i_d.di_ino = be64_to_cpu(dip->di_ino); - uuid_copy(&ip->i_d.di_uuid, &dip->di_uuid); - } + lock_mode = xfs_ilock_map_shared(dp); + error = xfs_dir_lookup(NULL, dp, name, &inum, ci_name); + xfs_iunlock_map_shared(dp, lock_mode); - /* - * Make sure to pull in the mode here as well in - * case the inode is released without being used. - * This ensures that xfs_inactive() will see that - * the inode is already free and not try to mess - * with the uninitialized part of it. - */ - ip->i_d.di_mode = 0; - } - - /* - * The inode format changed when we moved the link count and - * made it 32 bits long. If this is an old format inode, - * convert it in memory to look like a new one. If it gets - * flushed to disk we will convert back before flushing or - * logging it. We zero out the new projid field and the old link - * count field. We'll handle clearing the pad field (the remains - * of the old uuid field) when we actually convert the inode to - * the new format. We don't change the version number so that we - * can distinguish this from a real new format inode. - */ - if (ip->i_d.di_version == 1) { - ip->i_d.di_nlink = ip->i_d.di_onlink; - ip->i_d.di_onlink = 0; - xfs_set_projid(ip, 0); - } + if (error) + goto out; - ip->i_delayed_blks = 0; + error = xfs_iget(dp->i_mount, NULL, inum, 0, 0, ipp); + if (error) + goto out_free_name; - /* - * Mark the buffer containing the inode as something to keep - * around for a while. This helps to keep recently accessed - * meta-data in-core longer. - */ - xfs_buf_set_ref(bp, XFS_INO_REF); + return 0; - /* - * Use xfs_trans_brelse() to release the buffer containing the on-disk - * inode, because it was acquired with xfs_trans_read_buf() in - * xfs_imap_to_bp() above. If tp is NULL, this is just a normal - * brelse(). If we're within a transaction, then xfs_trans_brelse() - * will only release the buffer if it is not dirty within the - * transaction. It will be OK to release the buffer in this case, - * because inodes on disk are never destroyed and we will be locking the - * new in-core inode before putting it in the cache where other - * processes can find it. Thus we don't have to worry about the inode - * being changed just because we released the buffer. - */ - out_brelse: - xfs_trans_brelse(tp, bp); +out_free_name: + if (ci_name) + kmem_free(ci_name->name); +out: + *ipp = NULL; return error; } /* - * Read in extents from a btree-format inode. - * Allocate and fill in if_extents. Real work is done in xfs_bmap.c. - */ -int -xfs_iread_extents( - xfs_trans_t *tp, - xfs_inode_t *ip, - int whichfork) -{ - int error; - xfs_ifork_t *ifp; - xfs_extnum_t nextents; - - if (unlikely(XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE)) { - XFS_ERROR_REPORT("xfs_iread_extents", XFS_ERRLEVEL_LOW, - ip->i_mount); - return XFS_ERROR(EFSCORRUPTED); - } - nextents = XFS_IFORK_NEXTENTS(ip, whichfork); - ifp = XFS_IFORK_PTR(ip, whichfork); - - /* - * We know that the size is valid (it's checked in iformat_btree) - */ - ifp->if_bytes = ifp->if_real_bytes = 0; - ifp->if_flags |= XFS_IFEXTENTS; - xfs_iext_add(ifp, 0, nextents); - error = xfs_bmap_read_extents(tp, ip, whichfork); - if (error) { - xfs_iext_destroy(ifp); - ifp->if_flags &= ~XFS_IFEXTENTS; - return error; - } - xfs_validate_extents(ifp, nextents, XFS_EXTFMT_INODE(ip)); - return 0; -} - -/* * Allocate an inode on disk and return a copy of its in-core version. * The in-core inode is locked exclusively. Set mode, nlink, and rdev * appropriately within the inode. The uid and gid for the inode are @@ -1295,8 +690,8 @@ xfs_ialloc( ip->i_d.di_onlink = 0; ip->i_d.di_nlink = nlink; ASSERT(ip->i_d.di_nlink == nlink); - ip->i_d.di_uid = current_fsuid(); - ip->i_d.di_gid = current_fsgid(); + ip->i_d.di_uid = xfs_kuid_to_uid(current_fsuid()); + ip->i_d.di_gid = xfs_kgid_to_gid(current_fsgid()); xfs_set_projid(ip, prid); memset(&(ip->i_d.di_pad[0]), 0, sizeof(ip->i_d.di_pad)); @@ -1335,7 +730,7 @@ xfs_ialloc( */ if ((irix_sgid_inherit) && (ip->i_d.di_mode & S_ISGID) && - (!in_group_p((gid_t)ip->i_d.di_gid))) { + (!in_group_p(xfs_gid_to_kgid(ip->i_d.di_gid)))) { ip->i_d.di_mode &= ~S_ISGID; } @@ -1467,6 +862,583 @@ xfs_ialloc( } /* + * Allocates a new inode from disk and return a pointer to the + * incore copy. This routine will internally commit the current + * transaction and allocate a new one if the Space Manager needed + * to do an allocation to replenish the inode free-list. + * + * This routine is designed to be called from xfs_create and + * xfs_create_dir. + * + */ +int +xfs_dir_ialloc( + xfs_trans_t **tpp, /* input: current transaction; + output: may be a new transaction. */ + xfs_inode_t *dp, /* directory within whose allocate + the inode. */ + umode_t mode, + xfs_nlink_t nlink, + xfs_dev_t rdev, + prid_t prid, /* project id */ + int okalloc, /* ok to allocate new space */ + xfs_inode_t **ipp, /* pointer to inode; it will be + locked. */ + int *committed) + +{ + xfs_trans_t *tp; + xfs_trans_t *ntp; + xfs_inode_t *ip; + xfs_buf_t *ialloc_context = NULL; + int code; + void *dqinfo; + uint tflags; + + tp = *tpp; + ASSERT(tp->t_flags & XFS_TRANS_PERM_LOG_RES); + + /* + * xfs_ialloc will return a pointer to an incore inode if + * the Space Manager has an available inode on the free + * list. Otherwise, it will do an allocation and replenish + * the freelist. Since we can only do one allocation per + * transaction without deadlocks, we will need to commit the + * current transaction and start a new one. We will then + * need to call xfs_ialloc again to get the inode. + * + * If xfs_ialloc did an allocation to replenish the freelist, + * it returns the bp containing the head of the freelist as + * ialloc_context. We will hold a lock on it across the + * transaction commit so that no other process can steal + * the inode(s) that we've just allocated. + */ + code = xfs_ialloc(tp, dp, mode, nlink, rdev, prid, okalloc, + &ialloc_context, &ip); + + /* + * Return an error if we were unable to allocate a new inode. + * This should only happen if we run out of space on disk or + * encounter a disk error. + */ + if (code) { + *ipp = NULL; + return code; + } + if (!ialloc_context && !ip) { + *ipp = NULL; + return XFS_ERROR(ENOSPC); + } + + /* + * If the AGI buffer is non-NULL, then we were unable to get an + * inode in one operation. We need to commit the current + * transaction and call xfs_ialloc() again. It is guaranteed + * to succeed the second time. + */ + if (ialloc_context) { + struct xfs_trans_res tres; + + /* + * Normally, xfs_trans_commit releases all the locks. + * We call bhold to hang on to the ialloc_context across + * the commit. Holding this buffer prevents any other + * processes from doing any allocations in this + * allocation group. + */ + xfs_trans_bhold(tp, ialloc_context); + /* + * Save the log reservation so we can use + * them in the next transaction. + */ + tres.tr_logres = xfs_trans_get_log_res(tp); + tres.tr_logcount = xfs_trans_get_log_count(tp); + + /* + * We want the quota changes to be associated with the next + * transaction, NOT this one. So, detach the dqinfo from this + * and attach it to the next transaction. + */ + dqinfo = NULL; + tflags = 0; + if (tp->t_dqinfo) { + dqinfo = (void *)tp->t_dqinfo; + tp->t_dqinfo = NULL; + tflags = tp->t_flags & XFS_TRANS_DQ_DIRTY; + tp->t_flags &= ~(XFS_TRANS_DQ_DIRTY); + } + + ntp = xfs_trans_dup(tp); + code = xfs_trans_commit(tp, 0); + tp = ntp; + if (committed != NULL) { + *committed = 1; + } + /* + * If we get an error during the commit processing, + * release the buffer that is still held and return + * to the caller. + */ + if (code) { + xfs_buf_relse(ialloc_context); + if (dqinfo) { + tp->t_dqinfo = dqinfo; + xfs_trans_free_dqinfo(tp); + } + *tpp = ntp; + *ipp = NULL; + return code; + } + + /* + * transaction commit worked ok so we can drop the extra ticket + * reference that we gained in xfs_trans_dup() + */ + xfs_log_ticket_put(tp->t_ticket); + tres.tr_logflags = XFS_TRANS_PERM_LOG_RES; + code = xfs_trans_reserve(tp, &tres, 0, 0); + + /* + * Re-attach the quota info that we detached from prev trx. + */ + if (dqinfo) { + tp->t_dqinfo = dqinfo; + tp->t_flags |= tflags; + } + + if (code) { + xfs_buf_relse(ialloc_context); + *tpp = ntp; + *ipp = NULL; + return code; + } + xfs_trans_bjoin(tp, ialloc_context); + + /* + * Call ialloc again. Since we've locked out all + * other allocations in this allocation group, + * this call should always succeed. + */ + code = xfs_ialloc(tp, dp, mode, nlink, rdev, prid, + okalloc, &ialloc_context, &ip); + + /* + * If we get an error at this point, return to the caller + * so that the current transaction can be aborted. + */ + if (code) { + *tpp = tp; + *ipp = NULL; + return code; + } + ASSERT(!ialloc_context && ip); + + } else { + if (committed != NULL) + *committed = 0; + } + + *ipp = ip; + *tpp = tp; + + return 0; +} + +/* + * Decrement the link count on an inode & log the change. + * If this causes the link count to go to zero, initiate the + * logging activity required to truncate a file. + */ +int /* error */ +xfs_droplink( + xfs_trans_t *tp, + xfs_inode_t *ip) +{ + int error; + + xfs_trans_ichgtime(tp, ip, XFS_ICHGTIME_CHG); + + ASSERT (ip->i_d.di_nlink > 0); + ip->i_d.di_nlink--; + drop_nlink(VFS_I(ip)); + xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); + + error = 0; + if (ip->i_d.di_nlink == 0) { + /* + * We're dropping the last link to this file. + * Move the on-disk inode to the AGI unlinked list. + * From xfs_inactive() we will pull the inode from + * the list and free it. + */ + error = xfs_iunlink(tp, ip); + } + return error; +} + +/* + * This gets called when the inode's version needs to be changed from 1 to 2. + * Currently this happens when the nlink field overflows the old 16-bit value + * or when chproj is called to change the project for the first time. + * As a side effect the superblock version will also get rev'd + * to contain the NLINK bit. + */ +void +xfs_bump_ino_vers2( + xfs_trans_t *tp, + xfs_inode_t *ip) +{ + xfs_mount_t *mp; + + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); + ASSERT(ip->i_d.di_version == 1); + + ip->i_d.di_version = 2; + ip->i_d.di_onlink = 0; + memset(&(ip->i_d.di_pad[0]), 0, sizeof(ip->i_d.di_pad)); + mp = tp->t_mountp; + if (!xfs_sb_version_hasnlink(&mp->m_sb)) { + spin_lock(&mp->m_sb_lock); + if (!xfs_sb_version_hasnlink(&mp->m_sb)) { + xfs_sb_version_addnlink(&mp->m_sb); + spin_unlock(&mp->m_sb_lock); + xfs_mod_sb(tp, XFS_SB_VERSIONNUM); + } else { + spin_unlock(&mp->m_sb_lock); + } + } + /* Caller must log the inode */ +} + +/* + * Increment the link count on an inode & log the change. + */ +int +xfs_bumplink( + xfs_trans_t *tp, + xfs_inode_t *ip) +{ + xfs_trans_ichgtime(tp, ip, XFS_ICHGTIME_CHG); + + ASSERT(ip->i_d.di_nlink > 0); + ip->i_d.di_nlink++; + inc_nlink(VFS_I(ip)); + if ((ip->i_d.di_version == 1) && + (ip->i_d.di_nlink > XFS_MAXLINK_1)) { + /* + * The inode has increased its number of links beyond + * what can fit in an old format inode. It now needs + * to be converted to a version 2 inode with a 32 bit + * link count. If this is the first inode in the file + * system to do this, then we need to bump the superblock + * version number as well. + */ + xfs_bump_ino_vers2(tp, ip); + } + + xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); + return 0; +} + +int +xfs_create( + xfs_inode_t *dp, + struct xfs_name *name, + umode_t mode, + xfs_dev_t rdev, + xfs_inode_t **ipp) +{ + int is_dir = S_ISDIR(mode); + struct xfs_mount *mp = dp->i_mount; + struct xfs_inode *ip = NULL; + struct xfs_trans *tp = NULL; + int error; + xfs_bmap_free_t free_list; + xfs_fsblock_t first_block; + bool unlock_dp_on_error = false; + uint cancel_flags; + int committed; + prid_t prid; + struct xfs_dquot *udqp = NULL; + struct xfs_dquot *gdqp = NULL; + struct xfs_dquot *pdqp = NULL; + struct xfs_trans_res tres; + uint resblks; + + trace_xfs_create(dp, name); + + if (XFS_FORCED_SHUTDOWN(mp)) + return XFS_ERROR(EIO); + + if (dp->i_d.di_flags & XFS_DIFLAG_PROJINHERIT) + prid = xfs_get_projid(dp); + else + prid = XFS_PROJID_DEFAULT; + + /* + * Make sure that we have allocated dquot(s) on disk. + */ + error = xfs_qm_vop_dqalloc(dp, xfs_kuid_to_uid(current_fsuid()), + xfs_kgid_to_gid(current_fsgid()), prid, + XFS_QMOPT_QUOTALL | XFS_QMOPT_INHERIT, + &udqp, &gdqp, &pdqp); + if (error) + return error; + + if (is_dir) { + rdev = 0; + resblks = XFS_MKDIR_SPACE_RES(mp, name->len); + tres.tr_logres = M_RES(mp)->tr_mkdir.tr_logres; + tres.tr_logcount = XFS_MKDIR_LOG_COUNT; + tp = xfs_trans_alloc(mp, XFS_TRANS_MKDIR); + } else { + resblks = XFS_CREATE_SPACE_RES(mp, name->len); + tres.tr_logres = M_RES(mp)->tr_create.tr_logres; + tres.tr_logcount = XFS_CREATE_LOG_COUNT; + tp = xfs_trans_alloc(mp, XFS_TRANS_CREATE); + } + + cancel_flags = XFS_TRANS_RELEASE_LOG_RES; + + /* + * Initially assume that the file does not exist and + * reserve the resources for that case. If that is not + * the case we'll drop the one we have and get a more + * appropriate transaction later. + */ + tres.tr_logflags = XFS_TRANS_PERM_LOG_RES; + error = xfs_trans_reserve(tp, &tres, resblks, 0); + if (error == ENOSPC) { + /* flush outstanding delalloc blocks and retry */ + xfs_flush_inodes(mp); + error = xfs_trans_reserve(tp, &tres, resblks, 0); + } + if (error == ENOSPC) { + /* No space at all so try a "no-allocation" reservation */ + resblks = 0; + error = xfs_trans_reserve(tp, &tres, 0, 0); + } + if (error) { + cancel_flags = 0; + goto out_trans_cancel; + } + + xfs_ilock(dp, XFS_ILOCK_EXCL | XFS_ILOCK_PARENT); + unlock_dp_on_error = true; + + xfs_bmap_init(&free_list, &first_block); + + /* + * Reserve disk quota and the inode. + */ + error = xfs_trans_reserve_quota(tp, mp, udqp, gdqp, + pdqp, resblks, 1, 0); + if (error) + goto out_trans_cancel; + + error = xfs_dir_canenter(tp, dp, name, resblks); + if (error) + goto out_trans_cancel; + + /* + * A newly created regular or special file just has one directory + * entry pointing to them, but a directory also the "." entry + * pointing to itself. + */ + error = xfs_dir_ialloc(&tp, dp, mode, is_dir ? 2 : 1, rdev, + prid, resblks > 0, &ip, &committed); + if (error) { + if (error == ENOSPC) + goto out_trans_cancel; + goto out_trans_abort; + } + + /* + * Now we join the directory inode to the transaction. We do not do it + * earlier because xfs_dir_ialloc might commit the previous transaction + * (and release all the locks). An error from here on will result in + * the transaction cancel unlocking dp so don't do it explicitly in the + * error path. + */ + xfs_trans_ijoin(tp, dp, XFS_ILOCK_EXCL); + unlock_dp_on_error = false; + + error = xfs_dir_createname(tp, dp, name, ip->i_ino, + &first_block, &free_list, resblks ? + resblks - XFS_IALLOC_SPACE_RES(mp) : 0); + if (error) { + ASSERT(error != ENOSPC); + goto out_trans_abort; + } + xfs_trans_ichgtime(tp, dp, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG); + xfs_trans_log_inode(tp, dp, XFS_ILOG_CORE); + + if (is_dir) { + error = xfs_dir_init(tp, ip, dp); + if (error) + goto out_bmap_cancel; + + error = xfs_bumplink(tp, dp); + if (error) + goto out_bmap_cancel; + } + + /* + * If this is a synchronous mount, make sure that the + * create transaction goes to disk before returning to + * the user. + */ + if (mp->m_flags & (XFS_MOUNT_WSYNC|XFS_MOUNT_DIRSYNC)) + xfs_trans_set_sync(tp); + + /* + * Attach the dquot(s) to the inodes and modify them incore. + * These ids of the inode couldn't have changed since the new + * inode has been locked ever since it was created. + */ + xfs_qm_vop_create_dqattach(tp, ip, udqp, gdqp, pdqp); + + error = xfs_bmap_finish(&tp, &free_list, &committed); + if (error) + goto out_bmap_cancel; + + error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES); + if (error) + goto out_release_inode; + + xfs_qm_dqrele(udqp); + xfs_qm_dqrele(gdqp); + xfs_qm_dqrele(pdqp); + + *ipp = ip; + return 0; + + out_bmap_cancel: + xfs_bmap_cancel(&free_list); + out_trans_abort: + cancel_flags |= XFS_TRANS_ABORT; + out_trans_cancel: + xfs_trans_cancel(tp, cancel_flags); + out_release_inode: + /* + * Wait until after the current transaction is aborted to + * release the inode. This prevents recursive transactions + * and deadlocks from xfs_inactive. + */ + if (ip) + IRELE(ip); + + xfs_qm_dqrele(udqp); + xfs_qm_dqrele(gdqp); + xfs_qm_dqrele(pdqp); + + if (unlock_dp_on_error) + xfs_iunlock(dp, XFS_ILOCK_EXCL); + return error; +} + +int +xfs_link( + xfs_inode_t *tdp, + xfs_inode_t *sip, + struct xfs_name *target_name) +{ + xfs_mount_t *mp = tdp->i_mount; + xfs_trans_t *tp; + int error; + xfs_bmap_free_t free_list; + xfs_fsblock_t first_block; + int cancel_flags; + int committed; + int resblks; + + trace_xfs_link(tdp, target_name); + + ASSERT(!S_ISDIR(sip->i_d.di_mode)); + + if (XFS_FORCED_SHUTDOWN(mp)) + return XFS_ERROR(EIO); + + error = xfs_qm_dqattach(sip, 0); + if (error) + goto std_return; + + error = xfs_qm_dqattach(tdp, 0); + if (error) + goto std_return; + + tp = xfs_trans_alloc(mp, XFS_TRANS_LINK); + cancel_flags = XFS_TRANS_RELEASE_LOG_RES; + resblks = XFS_LINK_SPACE_RES(mp, target_name->len); + error = xfs_trans_reserve(tp, &M_RES(mp)->tr_link, resblks, 0); + if (error == ENOSPC) { + resblks = 0; + error = xfs_trans_reserve(tp, &M_RES(mp)->tr_link, 0, 0); + } + if (error) { + cancel_flags = 0; + goto error_return; + } + + xfs_lock_two_inodes(sip, tdp, XFS_ILOCK_EXCL); + + xfs_trans_ijoin(tp, sip, XFS_ILOCK_EXCL); + xfs_trans_ijoin(tp, tdp, XFS_ILOCK_EXCL); + + /* + * If we are using project inheritance, we only allow hard link + * creation in our tree when the project IDs are the same; else + * the tree quota mechanism could be circumvented. + */ + if (unlikely((tdp->i_d.di_flags & XFS_DIFLAG_PROJINHERIT) && + (xfs_get_projid(tdp) != xfs_get_projid(sip)))) { + error = XFS_ERROR(EXDEV); + goto error_return; + } + + error = xfs_dir_canenter(tp, tdp, target_name, resblks); + if (error) + goto error_return; + + xfs_bmap_init(&free_list, &first_block); + + error = xfs_dir_createname(tp, tdp, target_name, sip->i_ino, + &first_block, &free_list, resblks); + if (error) + goto abort_return; + xfs_trans_ichgtime(tp, tdp, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG); + xfs_trans_log_inode(tp, tdp, XFS_ILOG_CORE); + + error = xfs_bumplink(tp, sip); + if (error) + goto abort_return; + + /* + * If this is a synchronous mount, make sure that the + * link transaction goes to disk before returning to + * the user. + */ + if (mp->m_flags & (XFS_MOUNT_WSYNC|XFS_MOUNT_DIRSYNC)) { + xfs_trans_set_sync(tp); + } + + error = xfs_bmap_finish (&tp, &free_list, &committed); + if (error) { + xfs_bmap_cancel(&free_list); + goto abort_return; + } + + return xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES); + + abort_return: + cancel_flags |= XFS_TRANS_ABORT; + error_return: + xfs_trans_cancel(tp, cancel_flags); + std_return: + return error; +} + +/* * Free up the underlying blocks past new_size. The new size must be smaller * than the current size. This routine can be used both for the attribute and * data fork, and does not modify the inode size, which is left to the caller. @@ -1576,10 +1548,7 @@ xfs_itruncate_extents( * reference that we gained in xfs_trans_dup() */ xfs_log_ticket_put(tp->t_ticket); - error = xfs_trans_reserve(tp, 0, - XFS_ITRUNCATE_LOG_RES(mp), 0, - XFS_TRANS_PERM_LOG_RES, - XFS_ITRUNCATE_LOG_COUNT); + error = xfs_trans_reserve(tp, &M_RES(mp)->tr_itruncate, 0, 0); if (error) goto out; } @@ -1605,6 +1574,271 @@ out_bmap_cancel: goto out; } +int +xfs_release( + xfs_inode_t *ip) +{ + xfs_mount_t *mp = ip->i_mount; + int error; + + if (!S_ISREG(ip->i_d.di_mode) || (ip->i_d.di_mode == 0)) + return 0; + + /* If this is a read-only mount, don't do this (would generate I/O) */ + if (mp->m_flags & XFS_MOUNT_RDONLY) + return 0; + + if (!XFS_FORCED_SHUTDOWN(mp)) { + int truncated; + + /* + * If we are using filestreams, and we have an unlinked + * file that we are processing the last close on, then nothing + * will be able to reopen and write to this file. Purge this + * inode from the filestreams cache so that it doesn't delay + * teardown of the inode. + */ + if ((ip->i_d.di_nlink == 0) && xfs_inode_is_filestream(ip)) + xfs_filestream_deassociate(ip); + + /* + * If we previously truncated this file and removed old data + * in the process, we want to initiate "early" writeout on + * the last close. This is an attempt to combat the notorious + * NULL files problem which is particularly noticeable from a + * truncate down, buffered (re-)write (delalloc), followed by + * a crash. What we are effectively doing here is + * significantly reducing the time window where we'd otherwise + * be exposed to that problem. + */ + truncated = xfs_iflags_test_and_clear(ip, XFS_ITRUNCATED); + if (truncated) { + xfs_iflags_clear(ip, XFS_IDIRTY_RELEASE); + if (VN_DIRTY(VFS_I(ip)) && ip->i_delayed_blks > 0) { + error = -filemap_flush(VFS_I(ip)->i_mapping); + if (error) + return error; + } + } + } + + if (ip->i_d.di_nlink == 0) + return 0; + + if (xfs_can_free_eofblocks(ip, false)) { + + /* + * If we can't get the iolock just skip truncating the blocks + * past EOF because we could deadlock with the mmap_sem + * otherwise. We'll get another chance to drop them once the + * last reference to the inode is dropped, so we'll never leak + * blocks permanently. + * + * Further, check if the inode is being opened, written and + * closed frequently and we have delayed allocation blocks + * outstanding (e.g. streaming writes from the NFS server), + * truncating the blocks past EOF will cause fragmentation to + * occur. + * + * In this case don't do the truncation, either, but we have to + * be careful how we detect this case. Blocks beyond EOF show + * up as i_delayed_blks even when the inode is clean, so we + * need to truncate them away first before checking for a dirty + * release. Hence on the first dirty close we will still remove + * the speculative allocation, but after that we will leave it + * in place. + */ + if (xfs_iflags_test(ip, XFS_IDIRTY_RELEASE)) + return 0; + + error = xfs_free_eofblocks(mp, ip, true); + if (error && error != EAGAIN) + return error; + + /* delalloc blocks after truncation means it really is dirty */ + if (ip->i_delayed_blks) + xfs_iflags_set(ip, XFS_IDIRTY_RELEASE); + } + return 0; +} + +/* + * xfs_inactive + * + * This is called when the vnode reference count for the vnode + * goes to zero. If the file has been unlinked, then it must + * now be truncated. Also, we clear all of the read-ahead state + * kept for the inode here since the file is now closed. + */ +int +xfs_inactive( + xfs_inode_t *ip) +{ + xfs_bmap_free_t free_list; + xfs_fsblock_t first_block; + int committed; + struct xfs_trans *tp; + struct xfs_mount *mp; + struct xfs_trans_res *resp; + int error; + int truncate = 0; + + /* + * If the inode is already free, then there can be nothing + * to clean up here. + */ + if (ip->i_d.di_mode == 0 || is_bad_inode(VFS_I(ip))) { + ASSERT(ip->i_df.if_real_bytes == 0); + ASSERT(ip->i_df.if_broot_bytes == 0); + return VN_INACTIVE_CACHE; + } + + mp = ip->i_mount; + + error = 0; + + /* If this is a read-only mount, don't do this (would generate I/O) */ + if (mp->m_flags & XFS_MOUNT_RDONLY) + goto out; + + if (ip->i_d.di_nlink != 0) { + /* + * force is true because we are evicting an inode from the + * cache. Post-eof blocks must be freed, lest we end up with + * broken free space accounting. + */ + if (xfs_can_free_eofblocks(ip, true)) { + error = xfs_free_eofblocks(mp, ip, false); + if (error) + return VN_INACTIVE_CACHE; + } + goto out; + } + + if (S_ISREG(ip->i_d.di_mode) && + (ip->i_d.di_size != 0 || XFS_ISIZE(ip) != 0 || + ip->i_d.di_nextents > 0 || ip->i_delayed_blks > 0)) + truncate = 1; + + error = xfs_qm_dqattach(ip, 0); + if (error) + return VN_INACTIVE_CACHE; + + tp = xfs_trans_alloc(mp, XFS_TRANS_INACTIVE); + resp = (truncate || S_ISLNK(ip->i_d.di_mode)) ? + &M_RES(mp)->tr_itruncate : &M_RES(mp)->tr_ifree; + + error = xfs_trans_reserve(tp, resp, 0, 0); + if (error) { + ASSERT(XFS_FORCED_SHUTDOWN(mp)); + xfs_trans_cancel(tp, 0); + return VN_INACTIVE_CACHE; + } + + xfs_ilock(ip, XFS_ILOCK_EXCL); + xfs_trans_ijoin(tp, ip, 0); + + if (S_ISLNK(ip->i_d.di_mode)) { + error = xfs_inactive_symlink(ip, &tp); + if (error) + goto out_cancel; + } else if (truncate) { + ip->i_d.di_size = 0; + xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); + + error = xfs_itruncate_extents(&tp, ip, XFS_DATA_FORK, 0); + if (error) + goto out_cancel; + + ASSERT(ip->i_d.di_nextents == 0); + } + + /* + * If there are attributes associated with the file then blow them away + * now. The code calls a routine that recursively deconstructs the + * attribute fork. We need to just commit the current transaction + * because we can't use it for xfs_attr_inactive(). + */ + if (ip->i_d.di_anextents > 0) { + ASSERT(ip->i_d.di_forkoff != 0); + + error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES); + if (error) + goto out_unlock; + + xfs_iunlock(ip, XFS_ILOCK_EXCL); + + error = xfs_attr_inactive(ip); + if (error) + goto out; + + tp = xfs_trans_alloc(mp, XFS_TRANS_INACTIVE); + error = xfs_trans_reserve(tp, &M_RES(mp)->tr_ifree, 0, 0); + if (error) { + xfs_trans_cancel(tp, 0); + goto out; + } + + xfs_ilock(ip, XFS_ILOCK_EXCL); + xfs_trans_ijoin(tp, ip, 0); + } + + if (ip->i_afp) + xfs_idestroy_fork(ip, XFS_ATTR_FORK); + + ASSERT(ip->i_d.di_anextents == 0); + + /* + * Free the inode. + */ + xfs_bmap_init(&free_list, &first_block); + error = xfs_ifree(tp, ip, &free_list); + if (error) { + /* + * If we fail to free the inode, shut down. The cancel + * might do that, we need to make sure. Otherwise the + * inode might be lost for a long time or forever. + */ + if (!XFS_FORCED_SHUTDOWN(mp)) { + xfs_notice(mp, "%s: xfs_ifree returned error %d", + __func__, error); + xfs_force_shutdown(mp, SHUTDOWN_META_IO_ERROR); + } + xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES|XFS_TRANS_ABORT); + } else { + /* + * Credit the quota account(s). The inode is gone. + */ + xfs_trans_mod_dquot_byino(tp, ip, XFS_TRANS_DQ_ICOUNT, -1); + + /* + * Just ignore errors at this point. There is nothing we can + * do except to try to keep going. Make sure it's not a silent + * error. + */ + error = xfs_bmap_finish(&tp, &free_list, &committed); + if (error) + xfs_notice(mp, "%s: xfs_bmap_finish returned error %d", + __func__, error); + error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES); + if (error) + xfs_notice(mp, "%s: xfs_trans_commit returned error %d", + __func__, error); + } + + /* + * Release the dquots held by inode, if any. + */ + xfs_qm_dqdetach(ip); +out_unlock: + xfs_iunlock(ip, XFS_ILOCK_EXCL); +out: + return VN_INACTIVE_CACHE; +out_cancel: + xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES | XFS_TRANS_ABORT); + goto out_unlock; +} + /* * This is called when the inode's link count goes to 0. * We place the on-disk inode on a list in the AGI. It @@ -1861,7 +2095,7 @@ xfs_iunlink_remove( } /* - * A big issue when freeing the inode cluster is is that we _cannot_ skip any + * A big issue when freeing the inode cluster is that we _cannot_ skip any * inodes that are in memory - they all must be marked stale and attached to * the cluster buffer. */ @@ -2094,272 +2328,6 @@ xfs_ifree( } /* - * Reallocate the space for if_broot based on the number of records - * being added or deleted as indicated in rec_diff. Move the records - * and pointers in if_broot to fit the new size. When shrinking this - * will eliminate holes between the records and pointers created by - * the caller. When growing this will create holes to be filled in - * by the caller. - * - * The caller must not request to add more records than would fit in - * the on-disk inode root. If the if_broot is currently NULL, then - * if we adding records one will be allocated. The caller must also - * not request that the number of records go below zero, although - * it can go to zero. - * - * ip -- the inode whose if_broot area is changing - * ext_diff -- the change in the number of records, positive or negative, - * requested for the if_broot array. - */ -void -xfs_iroot_realloc( - xfs_inode_t *ip, - int rec_diff, - int whichfork) -{ - struct xfs_mount *mp = ip->i_mount; - int cur_max; - xfs_ifork_t *ifp; - struct xfs_btree_block *new_broot; - int new_max; - size_t new_size; - char *np; - char *op; - - /* - * Handle the degenerate case quietly. - */ - if (rec_diff == 0) { - return; - } - - ifp = XFS_IFORK_PTR(ip, whichfork); - if (rec_diff > 0) { - /* - * If there wasn't any memory allocated before, just - * allocate it now and get out. - */ - if (ifp->if_broot_bytes == 0) { - new_size = XFS_BMAP_BROOT_SPACE_CALC(mp, rec_diff); - ifp->if_broot = kmem_alloc(new_size, KM_SLEEP | KM_NOFS); - ifp->if_broot_bytes = (int)new_size; - return; - } - - /* - * If there is already an existing if_broot, then we need - * to realloc() it and shift the pointers to their new - * location. The records don't change location because - * they are kept butted up against the btree block header. - */ - cur_max = xfs_bmbt_maxrecs(mp, ifp->if_broot_bytes, 0); - new_max = cur_max + rec_diff; - new_size = XFS_BMAP_BROOT_SPACE_CALC(mp, new_max); - ifp->if_broot = kmem_realloc(ifp->if_broot, new_size, - XFS_BMAP_BROOT_SPACE_CALC(mp, cur_max), - KM_SLEEP | KM_NOFS); - op = (char *)XFS_BMAP_BROOT_PTR_ADDR(mp, ifp->if_broot, 1, - ifp->if_broot_bytes); - np = (char *)XFS_BMAP_BROOT_PTR_ADDR(mp, ifp->if_broot, 1, - (int)new_size); - ifp->if_broot_bytes = (int)new_size; - ASSERT(XFS_BMAP_BMDR_SPACE(ifp->if_broot) <= - XFS_IFORK_SIZE(ip, whichfork)); - memmove(np, op, cur_max * (uint)sizeof(xfs_dfsbno_t)); - return; - } - - /* - * rec_diff is less than 0. In this case, we are shrinking the - * if_broot buffer. It must already exist. If we go to zero - * records, just get rid of the root and clear the status bit. - */ - ASSERT((ifp->if_broot != NULL) && (ifp->if_broot_bytes > 0)); - cur_max = xfs_bmbt_maxrecs(mp, ifp->if_broot_bytes, 0); - new_max = cur_max + rec_diff; - ASSERT(new_max >= 0); - if (new_max > 0) - new_size = XFS_BMAP_BROOT_SPACE_CALC(mp, new_max); - else - new_size = 0; - if (new_size > 0) { - new_broot = kmem_alloc(new_size, KM_SLEEP | KM_NOFS); - /* - * First copy over the btree block header. - */ - memcpy(new_broot, ifp->if_broot, - XFS_BMBT_BLOCK_LEN(ip->i_mount)); - } else { - new_broot = NULL; - ifp->if_flags &= ~XFS_IFBROOT; - } - - /* - * Only copy the records and pointers if there are any. - */ - if (new_max > 0) { - /* - * First copy the records. - */ - op = (char *)XFS_BMBT_REC_ADDR(mp, ifp->if_broot, 1); - np = (char *)XFS_BMBT_REC_ADDR(mp, new_broot, 1); - memcpy(np, op, new_max * (uint)sizeof(xfs_bmbt_rec_t)); - - /* - * Then copy the pointers. - */ - op = (char *)XFS_BMAP_BROOT_PTR_ADDR(mp, ifp->if_broot, 1, - ifp->if_broot_bytes); - np = (char *)XFS_BMAP_BROOT_PTR_ADDR(mp, new_broot, 1, - (int)new_size); - memcpy(np, op, new_max * (uint)sizeof(xfs_dfsbno_t)); - } - kmem_free(ifp->if_broot); - ifp->if_broot = new_broot; - ifp->if_broot_bytes = (int)new_size; - if (ifp->if_broot) - ASSERT(XFS_BMAP_BMDR_SPACE(ifp->if_broot) <= - XFS_IFORK_SIZE(ip, whichfork)); - return; -} - - -/* - * This is called when the amount of space needed for if_data - * is increased or decreased. The change in size is indicated by - * the number of bytes that need to be added or deleted in the - * byte_diff parameter. - * - * If the amount of space needed has decreased below the size of the - * inline buffer, then switch to using the inline buffer. Otherwise, - * use kmem_realloc() or kmem_alloc() to adjust the size of the buffer - * to what is needed. - * - * ip -- the inode whose if_data area is changing - * byte_diff -- the change in the number of bytes, positive or negative, - * requested for the if_data array. - */ -void -xfs_idata_realloc( - xfs_inode_t *ip, - int byte_diff, - int whichfork) -{ - xfs_ifork_t *ifp; - int new_size; - int real_size; - - if (byte_diff == 0) { - return; - } - - ifp = XFS_IFORK_PTR(ip, whichfork); - new_size = (int)ifp->if_bytes + byte_diff; - ASSERT(new_size >= 0); - - if (new_size == 0) { - if (ifp->if_u1.if_data != ifp->if_u2.if_inline_data) { - kmem_free(ifp->if_u1.if_data); - } - ifp->if_u1.if_data = NULL; - real_size = 0; - } else if (new_size <= sizeof(ifp->if_u2.if_inline_data)) { - /* - * If the valid extents/data can fit in if_inline_ext/data, - * copy them from the malloc'd vector and free it. - */ - if (ifp->if_u1.if_data == NULL) { - ifp->if_u1.if_data = ifp->if_u2.if_inline_data; - } else if (ifp->if_u1.if_data != ifp->if_u2.if_inline_data) { - ASSERT(ifp->if_real_bytes != 0); - memcpy(ifp->if_u2.if_inline_data, ifp->if_u1.if_data, - new_size); - kmem_free(ifp->if_u1.if_data); - ifp->if_u1.if_data = ifp->if_u2.if_inline_data; - } - real_size = 0; - } else { - /* - * Stuck with malloc/realloc. - * For inline data, the underlying buffer must be - * a multiple of 4 bytes in size so that it can be - * logged and stay on word boundaries. We enforce - * that here. - */ - real_size = roundup(new_size, 4); - if (ifp->if_u1.if_data == NULL) { - ASSERT(ifp->if_real_bytes == 0); - ifp->if_u1.if_data = kmem_alloc(real_size, - KM_SLEEP | KM_NOFS); - } else if (ifp->if_u1.if_data != ifp->if_u2.if_inline_data) { - /* - * Only do the realloc if the underlying size - * is really changing. - */ - if (ifp->if_real_bytes != real_size) { - ifp->if_u1.if_data = - kmem_realloc(ifp->if_u1.if_data, - real_size, - ifp->if_real_bytes, - KM_SLEEP | KM_NOFS); - } - } else { - ASSERT(ifp->if_real_bytes == 0); - ifp->if_u1.if_data = kmem_alloc(real_size, - KM_SLEEP | KM_NOFS); - memcpy(ifp->if_u1.if_data, ifp->if_u2.if_inline_data, - ifp->if_bytes); - } - } - ifp->if_real_bytes = real_size; - ifp->if_bytes = new_size; - ASSERT(ifp->if_bytes <= XFS_IFORK_SIZE(ip, whichfork)); -} - -void -xfs_idestroy_fork( - xfs_inode_t *ip, - int whichfork) -{ - xfs_ifork_t *ifp; - - ifp = XFS_IFORK_PTR(ip, whichfork); - if (ifp->if_broot != NULL) { - kmem_free(ifp->if_broot); - ifp->if_broot = NULL; - } - - /* - * If the format is local, then we can't have an extents - * array so just look for an inline data array. If we're - * not local then we may or may not have an extents list, - * so check and free it up if we do. - */ - if (XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_LOCAL) { - if ((ifp->if_u1.if_data != ifp->if_u2.if_inline_data) && - (ifp->if_u1.if_data != NULL)) { - ASSERT(ifp->if_real_bytes != 0); - kmem_free(ifp->if_u1.if_data); - ifp->if_u1.if_data = NULL; - ifp->if_real_bytes = 0; - } - } else if ((ifp->if_flags & XFS_IFEXTENTS) && - ((ifp->if_flags & XFS_IFEXTIREC) || - ((ifp->if_u1.if_extents != NULL) && - (ifp->if_u1.if_extents != ifp->if_u2.if_inline_ext)))) { - ASSERT(ifp->if_real_bytes != 0); - xfs_iext_destroy(ifp); - } - ASSERT(ifp->if_u1.if_extents == NULL || - ifp->if_u1.if_extents == ifp->if_u2.if_inline_ext); - ASSERT(ifp->if_real_bytes == 0); - if (whichfork == XFS_ATTR_FORK) { - kmem_zone_free(xfs_ifork_zone, ip->i_afp); - ip->i_afp = NULL; - } -} - -/* * This is called to unpin an inode. The caller must have the inode locked * in at least shared mode so that the buffer cannot be subsequently pinned * once someone is waiting for it to be unpinned. @@ -2402,162 +2370,471 @@ xfs_iunpin_wait( __xfs_iunpin_wait(ip); } -/* - * xfs_iextents_copy() - * - * This is called to copy the REAL extents (as opposed to the delayed - * allocation extents) from the inode into the given buffer. It - * returns the number of bytes copied into the buffer. - * - * If there are no delayed allocation extents, then we can just - * memcpy() the extents into the buffer. Otherwise, we need to - * examine each extent in turn and skip those which are delayed. - */ int -xfs_iextents_copy( - xfs_inode_t *ip, - xfs_bmbt_rec_t *dp, - int whichfork) +xfs_remove( + xfs_inode_t *dp, + struct xfs_name *name, + xfs_inode_t *ip) { - int copied; - int i; - xfs_ifork_t *ifp; - int nrecs; - xfs_fsblock_t start_block; + xfs_mount_t *mp = dp->i_mount; + xfs_trans_t *tp = NULL; + int is_dir = S_ISDIR(ip->i_d.di_mode); + int error = 0; + xfs_bmap_free_t free_list; + xfs_fsblock_t first_block; + int cancel_flags; + int committed; + int link_zero; + uint resblks; + uint log_count; - ifp = XFS_IFORK_PTR(ip, whichfork); - ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_ILOCK_SHARED)); - ASSERT(ifp->if_bytes > 0); + trace_xfs_remove(dp, name); + + if (XFS_FORCED_SHUTDOWN(mp)) + return XFS_ERROR(EIO); + + error = xfs_qm_dqattach(dp, 0); + if (error) + goto std_return; + + error = xfs_qm_dqattach(ip, 0); + if (error) + goto std_return; - nrecs = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t); - XFS_BMAP_TRACE_EXLIST(ip, nrecs, whichfork); - ASSERT(nrecs > 0); + if (is_dir) { + tp = xfs_trans_alloc(mp, XFS_TRANS_RMDIR); + log_count = XFS_DEFAULT_LOG_COUNT; + } else { + tp = xfs_trans_alloc(mp, XFS_TRANS_REMOVE); + log_count = XFS_REMOVE_LOG_COUNT; + } + cancel_flags = XFS_TRANS_RELEASE_LOG_RES; /* - * There are some delayed allocation extents in the - * inode, so copy the extents one at a time and skip - * the delayed ones. There must be at least one - * non-delayed extent. + * We try to get the real space reservation first, + * allowing for directory btree deletion(s) implying + * possible bmap insert(s). If we can't get the space + * reservation then we use 0 instead, and avoid the bmap + * btree insert(s) in the directory code by, if the bmap + * insert tries to happen, instead trimming the LAST + * block from the directory. */ - copied = 0; - for (i = 0; i < nrecs; i++) { - xfs_bmbt_rec_host_t *ep = xfs_iext_get_ext(ifp, i); - start_block = xfs_bmbt_get_startblock(ep); - if (isnullstartblock(start_block)) { - /* - * It's a delayed allocation extent, so skip it. - */ - continue; + resblks = XFS_REMOVE_SPACE_RES(mp); + error = xfs_trans_reserve(tp, &M_RES(mp)->tr_remove, resblks, 0); + if (error == ENOSPC) { + resblks = 0; + error = xfs_trans_reserve(tp, &M_RES(mp)->tr_remove, 0, 0); + } + if (error) { + ASSERT(error != ENOSPC); + cancel_flags = 0; + goto out_trans_cancel; + } + + xfs_lock_two_inodes(dp, ip, XFS_ILOCK_EXCL); + + xfs_trans_ijoin(tp, dp, XFS_ILOCK_EXCL); + xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL); + + /* + * If we're removing a directory perform some additional validation. + */ + if (is_dir) { + ASSERT(ip->i_d.di_nlink >= 2); + if (ip->i_d.di_nlink != 2) { + error = XFS_ERROR(ENOTEMPTY); + goto out_trans_cancel; } + if (!xfs_dir_isempty(ip)) { + error = XFS_ERROR(ENOTEMPTY); + goto out_trans_cancel; + } + } + + xfs_bmap_init(&free_list, &first_block); + error = xfs_dir_removename(tp, dp, name, ip->i_ino, + &first_block, &free_list, resblks); + if (error) { + ASSERT(error != ENOENT); + goto out_bmap_cancel; + } + xfs_trans_ichgtime(tp, dp, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG); - /* Translate to on disk format */ - put_unaligned(cpu_to_be64(ep->l0), &dp->l0); - put_unaligned(cpu_to_be64(ep->l1), &dp->l1); - dp++; - copied++; + if (is_dir) { + /* + * Drop the link from ip's "..". + */ + error = xfs_droplink(tp, dp); + if (error) + goto out_bmap_cancel; + + /* + * Drop the "." link from ip to self. + */ + error = xfs_droplink(tp, ip); + if (error) + goto out_bmap_cancel; + } else { + /* + * When removing a non-directory we need to log the parent + * inode here. For a directory this is done implicitly + * by the xfs_droplink call for the ".." entry. + */ + xfs_trans_log_inode(tp, dp, XFS_ILOG_CORE); } - ASSERT(copied != 0); - xfs_validate_extents(ifp, copied, XFS_EXTFMT_INODE(ip)); - return (copied * (uint)sizeof(xfs_bmbt_rec_t)); + /* + * Drop the link from dp to ip. + */ + error = xfs_droplink(tp, ip); + if (error) + goto out_bmap_cancel; + + /* + * Determine if this is the last link while + * we are in the transaction. + */ + link_zero = (ip->i_d.di_nlink == 0); + + /* + * If this is a synchronous mount, make sure that the + * remove transaction goes to disk before returning to + * the user. + */ + if (mp->m_flags & (XFS_MOUNT_WSYNC|XFS_MOUNT_DIRSYNC)) + xfs_trans_set_sync(tp); + + error = xfs_bmap_finish(&tp, &free_list, &committed); + if (error) + goto out_bmap_cancel; + + error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES); + if (error) + goto std_return; + + /* + * If we are using filestreams, kill the stream association. + * If the file is still open it may get a new one but that + * will get killed on last close in xfs_close() so we don't + * have to worry about that. + */ + if (!is_dir && link_zero && xfs_inode_is_filestream(ip)) + xfs_filestream_deassociate(ip); + + return 0; + + out_bmap_cancel: + xfs_bmap_cancel(&free_list); + cancel_flags |= XFS_TRANS_ABORT; + out_trans_cancel: + xfs_trans_cancel(tp, cancel_flags); + std_return: + return error; } /* - * Each of the following cases stores data into the same region - * of the on-disk inode, so only one of them can be valid at - * any given time. While it is possible to have conflicting formats - * and log flags, e.g. having XFS_ILOG_?DATA set when the fork is - * in EXTENTS format, this can only happen when the fork has - * changed formats after being modified but before being flushed. - * In these cases, the format always takes precedence, because the - * format indicates the current state of the fork. + * Enter all inodes for a rename transaction into a sorted array. */ -/*ARGSUSED*/ STATIC void -xfs_iflush_fork( - xfs_inode_t *ip, - xfs_dinode_t *dip, - xfs_inode_log_item_t *iip, - int whichfork, - xfs_buf_t *bp) -{ - char *cp; - xfs_ifork_t *ifp; - xfs_mount_t *mp; - static const short brootflag[2] = - { XFS_ILOG_DBROOT, XFS_ILOG_ABROOT }; - static const short dataflag[2] = - { XFS_ILOG_DDATA, XFS_ILOG_ADATA }; - static const short extflag[2] = - { XFS_ILOG_DEXT, XFS_ILOG_AEXT }; - - if (!iip) - return; - ifp = XFS_IFORK_PTR(ip, whichfork); - /* - * This can happen if we gave up in iformat in an error path, - * for the attribute fork. - */ - if (!ifp) { - ASSERT(whichfork == XFS_ATTR_FORK); - return; - } - cp = XFS_DFORK_PTR(dip, whichfork); - mp = ip->i_mount; - switch (XFS_IFORK_FORMAT(ip, whichfork)) { - case XFS_DINODE_FMT_LOCAL: - if ((iip->ili_fields & dataflag[whichfork]) && - (ifp->if_bytes > 0)) { - ASSERT(ifp->if_u1.if_data != NULL); - ASSERT(ifp->if_bytes <= XFS_IFORK_SIZE(ip, whichfork)); - memcpy(cp, ifp->if_u1.if_data, ifp->if_bytes); - } - break; +xfs_sort_for_rename( + xfs_inode_t *dp1, /* in: old (source) directory inode */ + xfs_inode_t *dp2, /* in: new (target) directory inode */ + xfs_inode_t *ip1, /* in: inode of old entry */ + xfs_inode_t *ip2, /* in: inode of new entry, if it + already exists, NULL otherwise. */ + xfs_inode_t **i_tab,/* out: array of inode returned, sorted */ + int *num_inodes) /* out: number of inodes in array */ +{ + xfs_inode_t *temp; + int i, j; - case XFS_DINODE_FMT_EXTENTS: - ASSERT((ifp->if_flags & XFS_IFEXTENTS) || - !(iip->ili_fields & extflag[whichfork])); - if ((iip->ili_fields & extflag[whichfork]) && - (ifp->if_bytes > 0)) { - ASSERT(xfs_iext_get_ext(ifp, 0)); - ASSERT(XFS_IFORK_NEXTENTS(ip, whichfork) > 0); - (void)xfs_iextents_copy(ip, (xfs_bmbt_rec_t *)cp, - whichfork); - } - break; + /* + * i_tab contains a list of pointers to inodes. We initialize + * the table here & we'll sort it. We will then use it to + * order the acquisition of the inode locks. + * + * Note that the table may contain duplicates. e.g., dp1 == dp2. + */ + i_tab[0] = dp1; + i_tab[1] = dp2; + i_tab[2] = ip1; + if (ip2) { + *num_inodes = 4; + i_tab[3] = ip2; + } else { + *num_inodes = 3; + i_tab[3] = NULL; + } - case XFS_DINODE_FMT_BTREE: - if ((iip->ili_fields & brootflag[whichfork]) && - (ifp->if_broot_bytes > 0)) { - ASSERT(ifp->if_broot != NULL); - ASSERT(XFS_BMAP_BMDR_SPACE(ifp->if_broot) <= - XFS_IFORK_SIZE(ip, whichfork)); - xfs_bmbt_to_bmdr(mp, ifp->if_broot, ifp->if_broot_bytes, - (xfs_bmdr_block_t *)cp, - XFS_DFORK_SIZE(dip, mp, whichfork)); + /* + * Sort the elements via bubble sort. (Remember, there are at + * most 4 elements to sort, so this is adequate.) + */ + for (i = 0; i < *num_inodes; i++) { + for (j = 1; j < *num_inodes; j++) { + if (i_tab[j]->i_ino < i_tab[j-1]->i_ino) { + temp = i_tab[j]; + i_tab[j] = i_tab[j-1]; + i_tab[j-1] = temp; + } } - break; + } +} + +/* + * xfs_rename + */ +int +xfs_rename( + xfs_inode_t *src_dp, + struct xfs_name *src_name, + xfs_inode_t *src_ip, + xfs_inode_t *target_dp, + struct xfs_name *target_name, + xfs_inode_t *target_ip) +{ + xfs_trans_t *tp = NULL; + xfs_mount_t *mp = src_dp->i_mount; + int new_parent; /* moving to a new dir */ + int src_is_directory; /* src_name is a directory */ + int error; + xfs_bmap_free_t free_list; + xfs_fsblock_t first_block; + int cancel_flags; + int committed; + xfs_inode_t *inodes[4]; + int spaceres; + int num_inodes; + + trace_xfs_rename(src_dp, target_dp, src_name, target_name); + + new_parent = (src_dp != target_dp); + src_is_directory = S_ISDIR(src_ip->i_d.di_mode); + + xfs_sort_for_rename(src_dp, target_dp, src_ip, target_ip, + inodes, &num_inodes); + + xfs_bmap_init(&free_list, &first_block); + tp = xfs_trans_alloc(mp, XFS_TRANS_RENAME); + cancel_flags = XFS_TRANS_RELEASE_LOG_RES; + spaceres = XFS_RENAME_SPACE_RES(mp, target_name->len); + error = xfs_trans_reserve(tp, &M_RES(mp)->tr_rename, spaceres, 0); + if (error == ENOSPC) { + spaceres = 0; + error = xfs_trans_reserve(tp, &M_RES(mp)->tr_rename, 0, 0); + } + if (error) { + xfs_trans_cancel(tp, 0); + goto std_return; + } + + /* + * Attach the dquots to the inodes + */ + error = xfs_qm_vop_rename_dqattach(inodes); + if (error) { + xfs_trans_cancel(tp, cancel_flags); + goto std_return; + } + + /* + * Lock all the participating inodes. Depending upon whether + * the target_name exists in the target directory, and + * whether the target directory is the same as the source + * directory, we can lock from 2 to 4 inodes. + */ + xfs_lock_inodes(inodes, num_inodes, XFS_ILOCK_EXCL); + + /* + * Join all the inodes to the transaction. From this point on, + * we can rely on either trans_commit or trans_cancel to unlock + * them. + */ + xfs_trans_ijoin(tp, src_dp, XFS_ILOCK_EXCL); + if (new_parent) + xfs_trans_ijoin(tp, target_dp, XFS_ILOCK_EXCL); + xfs_trans_ijoin(tp, src_ip, XFS_ILOCK_EXCL); + if (target_ip) + xfs_trans_ijoin(tp, target_ip, XFS_ILOCK_EXCL); + + /* + * If we are using project inheritance, we only allow renames + * into our tree when the project IDs are the same; else the + * tree quota mechanism would be circumvented. + */ + if (unlikely((target_dp->i_d.di_flags & XFS_DIFLAG_PROJINHERIT) && + (xfs_get_projid(target_dp) != xfs_get_projid(src_ip)))) { + error = XFS_ERROR(EXDEV); + goto error_return; + } + + /* + * Set up the target. + */ + if (target_ip == NULL) { + /* + * If there's no space reservation, check the entry will + * fit before actually inserting it. + */ + error = xfs_dir_canenter(tp, target_dp, target_name, spaceres); + if (error) + goto error_return; + /* + * If target does not exist and the rename crosses + * directories, adjust the target directory link count + * to account for the ".." reference from the new entry. + */ + error = xfs_dir_createname(tp, target_dp, target_name, + src_ip->i_ino, &first_block, + &free_list, spaceres); + if (error == ENOSPC) + goto error_return; + if (error) + goto abort_return; + + xfs_trans_ichgtime(tp, target_dp, + XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG); - case XFS_DINODE_FMT_DEV: - if (iip->ili_fields & XFS_ILOG_DEV) { - ASSERT(whichfork == XFS_DATA_FORK); - xfs_dinode_put_rdev(dip, ip->i_df.if_u2.if_rdev); + if (new_parent && src_is_directory) { + error = xfs_bumplink(tp, target_dp); + if (error) + goto abort_return; + } + } else { /* target_ip != NULL */ + /* + * If target exists and it's a directory, check that both + * target and source are directories and that target can be + * destroyed, or that neither is a directory. + */ + if (S_ISDIR(target_ip->i_d.di_mode)) { + /* + * Make sure target dir is empty. + */ + if (!(xfs_dir_isempty(target_ip)) || + (target_ip->i_d.di_nlink > 2)) { + error = XFS_ERROR(EEXIST); + goto error_return; + } } - break; - case XFS_DINODE_FMT_UUID: - if (iip->ili_fields & XFS_ILOG_UUID) { - ASSERT(whichfork == XFS_DATA_FORK); - memcpy(XFS_DFORK_DPTR(dip), - &ip->i_df.if_u2.if_uuid, - sizeof(uuid_t)); + /* + * Link the source inode under the target name. + * If the source inode is a directory and we are moving + * it across directories, its ".." entry will be + * inconsistent until we replace that down below. + * + * In case there is already an entry with the same + * name at the destination directory, remove it first. + */ + error = xfs_dir_replace(tp, target_dp, target_name, + src_ip->i_ino, + &first_block, &free_list, spaceres); + if (error) + goto abort_return; + + xfs_trans_ichgtime(tp, target_dp, + XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG); + + /* + * Decrement the link count on the target since the target + * dir no longer points to it. + */ + error = xfs_droplink(tp, target_ip); + if (error) + goto abort_return; + + if (src_is_directory) { + /* + * Drop the link from the old "." entry. + */ + error = xfs_droplink(tp, target_ip); + if (error) + goto abort_return; } - break; + } /* target_ip != NULL */ - default: - ASSERT(0); - break; + /* + * Remove the source. + */ + if (new_parent && src_is_directory) { + /* + * Rewrite the ".." entry to point to the new + * directory. + */ + error = xfs_dir_replace(tp, src_ip, &xfs_name_dotdot, + target_dp->i_ino, + &first_block, &free_list, spaceres); + ASSERT(error != EEXIST); + if (error) + goto abort_return; + } + + /* + * We always want to hit the ctime on the source inode. + * + * This isn't strictly required by the standards since the source + * inode isn't really being changed, but old unix file systems did + * it and some incremental backup programs won't work without it. + */ + xfs_trans_ichgtime(tp, src_ip, XFS_ICHGTIME_CHG); + xfs_trans_log_inode(tp, src_ip, XFS_ILOG_CORE); + + /* + * Adjust the link count on src_dp. This is necessary when + * renaming a directory, either within one parent when + * the target existed, or across two parent directories. + */ + if (src_is_directory && (new_parent || target_ip != NULL)) { + + /* + * Decrement link count on src_directory since the + * entry that's moved no longer points to it. + */ + error = xfs_droplink(tp, src_dp); + if (error) + goto abort_return; + } + + error = xfs_dir_removename(tp, src_dp, src_name, src_ip->i_ino, + &first_block, &free_list, spaceres); + if (error) + goto abort_return; + + xfs_trans_ichgtime(tp, src_dp, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG); + xfs_trans_log_inode(tp, src_dp, XFS_ILOG_CORE); + if (new_parent) + xfs_trans_log_inode(tp, target_dp, XFS_ILOG_CORE); + + /* + * If this is a synchronous mount, make sure that the + * rename transaction goes to disk before returning to + * the user. + */ + if (mp->m_flags & (XFS_MOUNT_WSYNC|XFS_MOUNT_DIRSYNC)) { + xfs_trans_set_sync(tp); } + + error = xfs_bmap_finish(&tp, &free_list, &committed); + if (error) { + xfs_bmap_cancel(&free_list); + xfs_trans_cancel(tp, (XFS_TRANS_RELEASE_LOG_RES | + XFS_TRANS_ABORT)); + goto std_return; + } + + /* + * trans_commit will unlock src_ip, target_ip & decrement + * the vnode references. + */ + return xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES); + + abort_return: + cancel_flags |= XFS_TRANS_ABORT; + error_return: + xfs_bmap_cancel(&free_list); + xfs_trans_cancel(tp, cancel_flags); + std_return: + return error; } STATIC int @@ -2816,7 +3093,6 @@ abort_out: return error; } - STATIC int xfs_iflush_int( struct xfs_inode *ip, @@ -3004,1072 +3280,3 @@ xfs_iflush_int( corrupt_out: return XFS_ERROR(EFSCORRUPTED); } - -/* - * Return a pointer to the extent record at file index idx. - */ -xfs_bmbt_rec_host_t * -xfs_iext_get_ext( - xfs_ifork_t *ifp, /* inode fork pointer */ - xfs_extnum_t idx) /* index of target extent */ -{ - ASSERT(idx >= 0); - ASSERT(idx < ifp->if_bytes / sizeof(xfs_bmbt_rec_t)); - - if ((ifp->if_flags & XFS_IFEXTIREC) && (idx == 0)) { - return ifp->if_u1.if_ext_irec->er_extbuf; - } else if (ifp->if_flags & XFS_IFEXTIREC) { - xfs_ext_irec_t *erp; /* irec pointer */ - int erp_idx = 0; /* irec index */ - xfs_extnum_t page_idx = idx; /* ext index in target list */ - - erp = xfs_iext_idx_to_irec(ifp, &page_idx, &erp_idx, 0); - return &erp->er_extbuf[page_idx]; - } else if (ifp->if_bytes) { - return &ifp->if_u1.if_extents[idx]; - } else { - return NULL; - } -} - -/* - * Insert new item(s) into the extent records for incore inode - * fork 'ifp'. 'count' new items are inserted at index 'idx'. - */ -void -xfs_iext_insert( - xfs_inode_t *ip, /* incore inode pointer */ - xfs_extnum_t idx, /* starting index of new items */ - xfs_extnum_t count, /* number of inserted items */ - xfs_bmbt_irec_t *new, /* items to insert */ - int state) /* type of extent conversion */ -{ - xfs_ifork_t *ifp = (state & BMAP_ATTRFORK) ? ip->i_afp : &ip->i_df; - xfs_extnum_t i; /* extent record index */ - - trace_xfs_iext_insert(ip, idx, new, state, _RET_IP_); - - ASSERT(ifp->if_flags & XFS_IFEXTENTS); - xfs_iext_add(ifp, idx, count); - for (i = idx; i < idx + count; i++, new++) - xfs_bmbt_set_all(xfs_iext_get_ext(ifp, i), new); -} - -/* - * This is called when the amount of space required for incore file - * extents needs to be increased. The ext_diff parameter stores the - * number of new extents being added and the idx parameter contains - * the extent index where the new extents will be added. If the new - * extents are being appended, then we just need to (re)allocate and - * initialize the space. Otherwise, if the new extents are being - * inserted into the middle of the existing entries, a bit more work - * is required to make room for the new extents to be inserted. The - * caller is responsible for filling in the new extent entries upon - * return. - */ -void -xfs_iext_add( - xfs_ifork_t *ifp, /* inode fork pointer */ - xfs_extnum_t idx, /* index to begin adding exts */ - int ext_diff) /* number of extents to add */ -{ - int byte_diff; /* new bytes being added */ - int new_size; /* size of extents after adding */ - xfs_extnum_t nextents; /* number of extents in file */ - - nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t); - ASSERT((idx >= 0) && (idx <= nextents)); - byte_diff = ext_diff * sizeof(xfs_bmbt_rec_t); - new_size = ifp->if_bytes + byte_diff; - /* - * If the new number of extents (nextents + ext_diff) - * fits inside the inode, then continue to use the inline - * extent buffer. - */ - if (nextents + ext_diff <= XFS_INLINE_EXTS) { - if (idx < nextents) { - memmove(&ifp->if_u2.if_inline_ext[idx + ext_diff], - &ifp->if_u2.if_inline_ext[idx], - (nextents - idx) * sizeof(xfs_bmbt_rec_t)); - memset(&ifp->if_u2.if_inline_ext[idx], 0, byte_diff); - } - ifp->if_u1.if_extents = ifp->if_u2.if_inline_ext; - ifp->if_real_bytes = 0; - } - /* - * Otherwise use a linear (direct) extent list. - * If the extents are currently inside the inode, - * xfs_iext_realloc_direct will switch us from - * inline to direct extent allocation mode. - */ - else if (nextents + ext_diff <= XFS_LINEAR_EXTS) { - xfs_iext_realloc_direct(ifp, new_size); - if (idx < nextents) { - memmove(&ifp->if_u1.if_extents[idx + ext_diff], - &ifp->if_u1.if_extents[idx], - (nextents - idx) * sizeof(xfs_bmbt_rec_t)); - memset(&ifp->if_u1.if_extents[idx], 0, byte_diff); - } - } - /* Indirection array */ - else { - xfs_ext_irec_t *erp; - int erp_idx = 0; - int page_idx = idx; - - ASSERT(nextents + ext_diff > XFS_LINEAR_EXTS); - if (ifp->if_flags & XFS_IFEXTIREC) { - erp = xfs_iext_idx_to_irec(ifp, &page_idx, &erp_idx, 1); - } else { - xfs_iext_irec_init(ifp); - ASSERT(ifp->if_flags & XFS_IFEXTIREC); - erp = ifp->if_u1.if_ext_irec; - } - /* Extents fit in target extent page */ - if (erp && erp->er_extcount + ext_diff <= XFS_LINEAR_EXTS) { - if (page_idx < erp->er_extcount) { - memmove(&erp->er_extbuf[page_idx + ext_diff], - &erp->er_extbuf[page_idx], - (erp->er_extcount - page_idx) * - sizeof(xfs_bmbt_rec_t)); - memset(&erp->er_extbuf[page_idx], 0, byte_diff); - } - erp->er_extcount += ext_diff; - xfs_iext_irec_update_extoffs(ifp, erp_idx + 1, ext_diff); - } - /* Insert a new extent page */ - else if (erp) { - xfs_iext_add_indirect_multi(ifp, - erp_idx, page_idx, ext_diff); - } - /* - * If extent(s) are being appended to the last page in - * the indirection array and the new extent(s) don't fit - * in the page, then erp is NULL and erp_idx is set to - * the next index needed in the indirection array. - */ - else { - int count = ext_diff; - - while (count) { - erp = xfs_iext_irec_new(ifp, erp_idx); - erp->er_extcount = count; - count -= MIN(count, (int)XFS_LINEAR_EXTS); - if (count) { - erp_idx++; - } - } - } - } - ifp->if_bytes = new_size; -} - -/* - * This is called when incore extents are being added to the indirection - * array and the new extents do not fit in the target extent list. The - * erp_idx parameter contains the irec index for the target extent list - * in the indirection array, and the idx parameter contains the extent - * index within the list. The number of extents being added is stored - * in the count parameter. - * - * |-------| |-------| - * | | | | idx - number of extents before idx - * | idx | | count | - * | | | | count - number of extents being inserted at idx - * |-------| |-------| - * | count | | nex2 | nex2 - number of extents after idx + count - * |-------| |-------| - */ -void -xfs_iext_add_indirect_multi( - xfs_ifork_t *ifp, /* inode fork pointer */ - int erp_idx, /* target extent irec index */ - xfs_extnum_t idx, /* index within target list */ - int count) /* new extents being added */ -{ - int byte_diff; /* new bytes being added */ - xfs_ext_irec_t *erp; /* pointer to irec entry */ - xfs_extnum_t ext_diff; /* number of extents to add */ - xfs_extnum_t ext_cnt; /* new extents still needed */ - xfs_extnum_t nex2; /* extents after idx + count */ - xfs_bmbt_rec_t *nex2_ep = NULL; /* temp list for nex2 extents */ - int nlists; /* number of irec's (lists) */ - - ASSERT(ifp->if_flags & XFS_IFEXTIREC); - erp = &ifp->if_u1.if_ext_irec[erp_idx]; - nex2 = erp->er_extcount - idx; - nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ; - - /* - * Save second part of target extent list - * (all extents past */ - if (nex2) { - byte_diff = nex2 * sizeof(xfs_bmbt_rec_t); - nex2_ep = (xfs_bmbt_rec_t *) kmem_alloc(byte_diff, KM_NOFS); - memmove(nex2_ep, &erp->er_extbuf[idx], byte_diff); - erp->er_extcount -= nex2; - xfs_iext_irec_update_extoffs(ifp, erp_idx + 1, -nex2); - memset(&erp->er_extbuf[idx], 0, byte_diff); - } - - /* - * Add the new extents to the end of the target - * list, then allocate new irec record(s) and - * extent buffer(s) as needed to store the rest - * of the new extents. - */ - ext_cnt = count; - ext_diff = MIN(ext_cnt, (int)XFS_LINEAR_EXTS - erp->er_extcount); - if (ext_diff) { - erp->er_extcount += ext_diff; - xfs_iext_irec_update_extoffs(ifp, erp_idx + 1, ext_diff); - ext_cnt -= ext_diff; - } - while (ext_cnt) { - erp_idx++; - erp = xfs_iext_irec_new(ifp, erp_idx); - ext_diff = MIN(ext_cnt, (int)XFS_LINEAR_EXTS); - erp->er_extcount = ext_diff; - xfs_iext_irec_update_extoffs(ifp, erp_idx + 1, ext_diff); - ext_cnt -= ext_diff; - } - - /* Add nex2 extents back to indirection array */ - if (nex2) { - xfs_extnum_t ext_avail; - int i; - - byte_diff = nex2 * sizeof(xfs_bmbt_rec_t); - ext_avail = XFS_LINEAR_EXTS - erp->er_extcount; - i = 0; - /* - * If nex2 extents fit in the current page, append - * nex2_ep after the new extents. - */ - if (nex2 <= ext_avail) { - i = erp->er_extcount; - } - /* - * Otherwise, check if space is available in the - * next page. - */ - else if ((erp_idx < nlists - 1) && - (nex2 <= (ext_avail = XFS_LINEAR_EXTS - - ifp->if_u1.if_ext_irec[erp_idx+1].er_extcount))) { - erp_idx++; - erp++; - /* Create a hole for nex2 extents */ - memmove(&erp->er_extbuf[nex2], erp->er_extbuf, - erp->er_extcount * sizeof(xfs_bmbt_rec_t)); - } - /* - * Final choice, create a new extent page for - * nex2 extents. - */ - else { - erp_idx++; - erp = xfs_iext_irec_new(ifp, erp_idx); - } - memmove(&erp->er_extbuf[i], nex2_ep, byte_diff); - kmem_free(nex2_ep); - erp->er_extcount += nex2; - xfs_iext_irec_update_extoffs(ifp, erp_idx + 1, nex2); - } -} - -/* - * This is called when the amount of space required for incore file - * extents needs to be decreased. The ext_diff parameter stores the - * number of extents to be removed and the idx parameter contains - * the extent index where the extents will be removed from. - * - * If the amount of space needed has decreased below the linear - * limit, XFS_IEXT_BUFSZ, then switch to using the contiguous - * extent array. Otherwise, use kmem_realloc() to adjust the - * size to what is needed. - */ -void -xfs_iext_remove( - xfs_inode_t *ip, /* incore inode pointer */ - xfs_extnum_t idx, /* index to begin removing exts */ - int ext_diff, /* number of extents to remove */ - int state) /* type of extent conversion */ -{ - xfs_ifork_t *ifp = (state & BMAP_ATTRFORK) ? ip->i_afp : &ip->i_df; - xfs_extnum_t nextents; /* number of extents in file */ - int new_size; /* size of extents after removal */ - - trace_xfs_iext_remove(ip, idx, state, _RET_IP_); - - ASSERT(ext_diff > 0); - nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t); - new_size = (nextents - ext_diff) * sizeof(xfs_bmbt_rec_t); - - if (new_size == 0) { - xfs_iext_destroy(ifp); - } else if (ifp->if_flags & XFS_IFEXTIREC) { - xfs_iext_remove_indirect(ifp, idx, ext_diff); - } else if (ifp->if_real_bytes) { - xfs_iext_remove_direct(ifp, idx, ext_diff); - } else { - xfs_iext_remove_inline(ifp, idx, ext_diff); - } - ifp->if_bytes = new_size; -} - -/* - * This removes ext_diff extents from the inline buffer, beginning - * at extent index idx. - */ -void -xfs_iext_remove_inline( - xfs_ifork_t *ifp, /* inode fork pointer */ - xfs_extnum_t idx, /* index to begin removing exts */ - int ext_diff) /* number of extents to remove */ -{ - int nextents; /* number of extents in file */ - - ASSERT(!(ifp->if_flags & XFS_IFEXTIREC)); - ASSERT(idx < XFS_INLINE_EXTS); - nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t); - ASSERT(((nextents - ext_diff) > 0) && - (nextents - ext_diff) < XFS_INLINE_EXTS); - - if (idx + ext_diff < nextents) { - memmove(&ifp->if_u2.if_inline_ext[idx], - &ifp->if_u2.if_inline_ext[idx + ext_diff], - (nextents - (idx + ext_diff)) * - sizeof(xfs_bmbt_rec_t)); - memset(&ifp->if_u2.if_inline_ext[nextents - ext_diff], - 0, ext_diff * sizeof(xfs_bmbt_rec_t)); - } else { - memset(&ifp->if_u2.if_inline_ext[idx], 0, - ext_diff * sizeof(xfs_bmbt_rec_t)); - } -} - -/* - * This removes ext_diff extents from a linear (direct) extent list, - * beginning at extent index idx. If the extents are being removed - * from the end of the list (ie. truncate) then we just need to re- - * allocate the list to remove the extra space. Otherwise, if the - * extents are being removed from the middle of the existing extent - * entries, then we first need to move the extent records beginning - * at idx + ext_diff up in the list to overwrite the records being - * removed, then remove the extra space via kmem_realloc. - */ -void -xfs_iext_remove_direct( - xfs_ifork_t *ifp, /* inode fork pointer */ - xfs_extnum_t idx, /* index to begin removing exts */ - int ext_diff) /* number of extents to remove */ -{ - xfs_extnum_t nextents; /* number of extents in file */ - int new_size; /* size of extents after removal */ - - ASSERT(!(ifp->if_flags & XFS_IFEXTIREC)); - new_size = ifp->if_bytes - - (ext_diff * sizeof(xfs_bmbt_rec_t)); - nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t); - - if (new_size == 0) { - xfs_iext_destroy(ifp); - return; - } - /* Move extents up in the list (if needed) */ - if (idx + ext_diff < nextents) { - memmove(&ifp->if_u1.if_extents[idx], - &ifp->if_u1.if_extents[idx + ext_diff], - (nextents - (idx + ext_diff)) * - sizeof(xfs_bmbt_rec_t)); - } - memset(&ifp->if_u1.if_extents[nextents - ext_diff], - 0, ext_diff * sizeof(xfs_bmbt_rec_t)); - /* - * Reallocate the direct extent list. If the extents - * will fit inside the inode then xfs_iext_realloc_direct - * will switch from direct to inline extent allocation - * mode for us. - */ - xfs_iext_realloc_direct(ifp, new_size); - ifp->if_bytes = new_size; -} - -/* - * This is called when incore extents are being removed from the - * indirection array and the extents being removed span multiple extent - * buffers. The idx parameter contains the file extent index where we - * want to begin removing extents, and the count parameter contains - * how many extents need to be removed. - * - * |-------| |-------| - * | nex1 | | | nex1 - number of extents before idx - * |-------| | count | - * | | | | count - number of extents being removed at idx - * | count | |-------| - * | | | nex2 | nex2 - number of extents after idx + count - * |-------| |-------| - */ -void -xfs_iext_remove_indirect( - xfs_ifork_t *ifp, /* inode fork pointer */ - xfs_extnum_t idx, /* index to begin removing extents */ - int count) /* number of extents to remove */ -{ - xfs_ext_irec_t *erp; /* indirection array pointer */ - int erp_idx = 0; /* indirection array index */ - xfs_extnum_t ext_cnt; /* extents left to remove */ - xfs_extnum_t ext_diff; /* extents to remove in current list */ - xfs_extnum_t nex1; /* number of extents before idx */ - xfs_extnum_t nex2; /* extents after idx + count */ - int page_idx = idx; /* index in target extent list */ - - ASSERT(ifp->if_flags & XFS_IFEXTIREC); - erp = xfs_iext_idx_to_irec(ifp, &page_idx, &erp_idx, 0); - ASSERT(erp != NULL); - nex1 = page_idx; - ext_cnt = count; - while (ext_cnt) { - nex2 = MAX((erp->er_extcount - (nex1 + ext_cnt)), 0); - ext_diff = MIN(ext_cnt, (erp->er_extcount - nex1)); - /* - * Check for deletion of entire list; - * xfs_iext_irec_remove() updates extent offsets. - */ - if (ext_diff == erp->er_extcount) { - xfs_iext_irec_remove(ifp, erp_idx); - ext_cnt -= ext_diff; - nex1 = 0; - if (ext_cnt) { - ASSERT(erp_idx < ifp->if_real_bytes / - XFS_IEXT_BUFSZ); - erp = &ifp->if_u1.if_ext_irec[erp_idx]; - nex1 = 0; - continue; - } else { - break; - } - } - /* Move extents up (if needed) */ - if (nex2) { - memmove(&erp->er_extbuf[nex1], - &erp->er_extbuf[nex1 + ext_diff], - nex2 * sizeof(xfs_bmbt_rec_t)); - } - /* Zero out rest of page */ - memset(&erp->er_extbuf[nex1 + nex2], 0, (XFS_IEXT_BUFSZ - - ((nex1 + nex2) * sizeof(xfs_bmbt_rec_t)))); - /* Update remaining counters */ - erp->er_extcount -= ext_diff; - xfs_iext_irec_update_extoffs(ifp, erp_idx + 1, -ext_diff); - ext_cnt -= ext_diff; - nex1 = 0; - erp_idx++; - erp++; - } - ifp->if_bytes -= count * sizeof(xfs_bmbt_rec_t); - xfs_iext_irec_compact(ifp); -} - -/* - * Create, destroy, or resize a linear (direct) block of extents. - */ -void -xfs_iext_realloc_direct( - xfs_ifork_t *ifp, /* inode fork pointer */ - int new_size) /* new size of extents */ -{ - int rnew_size; /* real new size of extents */ - - rnew_size = new_size; - - ASSERT(!(ifp->if_flags & XFS_IFEXTIREC) || - ((new_size >= 0) && (new_size <= XFS_IEXT_BUFSZ) && - (new_size != ifp->if_real_bytes))); - - /* Free extent records */ - if (new_size == 0) { - xfs_iext_destroy(ifp); - } - /* Resize direct extent list and zero any new bytes */ - else if (ifp->if_real_bytes) { - /* Check if extents will fit inside the inode */ - if (new_size <= XFS_INLINE_EXTS * sizeof(xfs_bmbt_rec_t)) { - xfs_iext_direct_to_inline(ifp, new_size / - (uint)sizeof(xfs_bmbt_rec_t)); - ifp->if_bytes = new_size; - return; - } - if (!is_power_of_2(new_size)){ - rnew_size = roundup_pow_of_two(new_size); - } - if (rnew_size != ifp->if_real_bytes) { - ifp->if_u1.if_extents = - kmem_realloc(ifp->if_u1.if_extents, - rnew_size, - ifp->if_real_bytes, KM_NOFS); - } - if (rnew_size > ifp->if_real_bytes) { - memset(&ifp->if_u1.if_extents[ifp->if_bytes / - (uint)sizeof(xfs_bmbt_rec_t)], 0, - rnew_size - ifp->if_real_bytes); - } - } - /* - * Switch from the inline extent buffer to a direct - * extent list. Be sure to include the inline extent - * bytes in new_size. - */ - else { - new_size += ifp->if_bytes; - if (!is_power_of_2(new_size)) { - rnew_size = roundup_pow_of_two(new_size); - } - xfs_iext_inline_to_direct(ifp, rnew_size); - } - ifp->if_real_bytes = rnew_size; - ifp->if_bytes = new_size; -} - -/* - * Switch from linear (direct) extent records to inline buffer. - */ -void -xfs_iext_direct_to_inline( - xfs_ifork_t *ifp, /* inode fork pointer */ - xfs_extnum_t nextents) /* number of extents in file */ -{ - ASSERT(ifp->if_flags & XFS_IFEXTENTS); - ASSERT(nextents <= XFS_INLINE_EXTS); - /* - * The inline buffer was zeroed when we switched - * from inline to direct extent allocation mode, - * so we don't need to clear it here. - */ - memcpy(ifp->if_u2.if_inline_ext, ifp->if_u1.if_extents, - nextents * sizeof(xfs_bmbt_rec_t)); - kmem_free(ifp->if_u1.if_extents); - ifp->if_u1.if_extents = ifp->if_u2.if_inline_ext; - ifp->if_real_bytes = 0; -} - -/* - * Switch from inline buffer to linear (direct) extent records. - * new_size should already be rounded up to the next power of 2 - * by the caller (when appropriate), so use new_size as it is. - * However, since new_size may be rounded up, we can't update - * if_bytes here. It is the caller's responsibility to update - * if_bytes upon return. - */ -void -xfs_iext_inline_to_direct( - xfs_ifork_t *ifp, /* inode fork pointer */ - int new_size) /* number of extents in file */ -{ - ifp->if_u1.if_extents = kmem_alloc(new_size, KM_NOFS); - memset(ifp->if_u1.if_extents, 0, new_size); - if (ifp->if_bytes) { - memcpy(ifp->if_u1.if_extents, ifp->if_u2.if_inline_ext, - ifp->if_bytes); - memset(ifp->if_u2.if_inline_ext, 0, XFS_INLINE_EXTS * - sizeof(xfs_bmbt_rec_t)); - } - ifp->if_real_bytes = new_size; -} - -/* - * Resize an extent indirection array to new_size bytes. - */ -STATIC void -xfs_iext_realloc_indirect( - xfs_ifork_t *ifp, /* inode fork pointer */ - int new_size) /* new indirection array size */ -{ - int nlists; /* number of irec's (ex lists) */ - int size; /* current indirection array size */ - - ASSERT(ifp->if_flags & XFS_IFEXTIREC); - nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ; - size = nlists * sizeof(xfs_ext_irec_t); - ASSERT(ifp->if_real_bytes); - ASSERT((new_size >= 0) && (new_size != size)); - if (new_size == 0) { - xfs_iext_destroy(ifp); - } else { - ifp->if_u1.if_ext_irec = (xfs_ext_irec_t *) - kmem_realloc(ifp->if_u1.if_ext_irec, - new_size, size, KM_NOFS); - } -} - -/* - * Switch from indirection array to linear (direct) extent allocations. - */ -STATIC void -xfs_iext_indirect_to_direct( - xfs_ifork_t *ifp) /* inode fork pointer */ -{ - xfs_bmbt_rec_host_t *ep; /* extent record pointer */ - xfs_extnum_t nextents; /* number of extents in file */ - int size; /* size of file extents */ - - ASSERT(ifp->if_flags & XFS_IFEXTIREC); - nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t); - ASSERT(nextents <= XFS_LINEAR_EXTS); - size = nextents * sizeof(xfs_bmbt_rec_t); - - xfs_iext_irec_compact_pages(ifp); - ASSERT(ifp->if_real_bytes == XFS_IEXT_BUFSZ); - - ep = ifp->if_u1.if_ext_irec->er_extbuf; - kmem_free(ifp->if_u1.if_ext_irec); - ifp->if_flags &= ~XFS_IFEXTIREC; - ifp->if_u1.if_extents = ep; - ifp->if_bytes = size; - if (nextents < XFS_LINEAR_EXTS) { - xfs_iext_realloc_direct(ifp, size); - } -} - -/* - * Free incore file extents. - */ -void -xfs_iext_destroy( - xfs_ifork_t *ifp) /* inode fork pointer */ -{ - if (ifp->if_flags & XFS_IFEXTIREC) { - int erp_idx; - int nlists; - - nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ; - for (erp_idx = nlists - 1; erp_idx >= 0 ; erp_idx--) { - xfs_iext_irec_remove(ifp, erp_idx); - } - ifp->if_flags &= ~XFS_IFEXTIREC; - } else if (ifp->if_real_bytes) { - kmem_free(ifp->if_u1.if_extents); - } else if (ifp->if_bytes) { - memset(ifp->if_u2.if_inline_ext, 0, XFS_INLINE_EXTS * - sizeof(xfs_bmbt_rec_t)); - } - ifp->if_u1.if_extents = NULL; - ifp->if_real_bytes = 0; - ifp->if_bytes = 0; -} - -/* - * Return a pointer to the extent record for file system block bno. - */ -xfs_bmbt_rec_host_t * /* pointer to found extent record */ -xfs_iext_bno_to_ext( - xfs_ifork_t *ifp, /* inode fork pointer */ - xfs_fileoff_t bno, /* block number to search for */ - xfs_extnum_t *idxp) /* index of target extent */ -{ - xfs_bmbt_rec_host_t *base; /* pointer to first extent */ - xfs_filblks_t blockcount = 0; /* number of blocks in extent */ - xfs_bmbt_rec_host_t *ep = NULL; /* pointer to target extent */ - xfs_ext_irec_t *erp = NULL; /* indirection array pointer */ - int high; /* upper boundary in search */ - xfs_extnum_t idx = 0; /* index of target extent */ - int low; /* lower boundary in search */ - xfs_extnum_t nextents; /* number of file extents */ - xfs_fileoff_t startoff = 0; /* start offset of extent */ - - nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t); - if (nextents == 0) { - *idxp = 0; - return NULL; - } - low = 0; - if (ifp->if_flags & XFS_IFEXTIREC) { - /* Find target extent list */ - int erp_idx = 0; - erp = xfs_iext_bno_to_irec(ifp, bno, &erp_idx); - base = erp->er_extbuf; - high = erp->er_extcount - 1; - } else { - base = ifp->if_u1.if_extents; - high = nextents - 1; - } - /* Binary search extent records */ - while (low <= high) { - idx = (low + high) >> 1; - ep = base + idx; - startoff = xfs_bmbt_get_startoff(ep); - blockcount = xfs_bmbt_get_blockcount(ep); - if (bno < startoff) { - high = idx - 1; - } else if (bno >= startoff + blockcount) { - low = idx + 1; - } else { - /* Convert back to file-based extent index */ - if (ifp->if_flags & XFS_IFEXTIREC) { - idx += erp->er_extoff; - } - *idxp = idx; - return ep; - } - } - /* Convert back to file-based extent index */ - if (ifp->if_flags & XFS_IFEXTIREC) { - idx += erp->er_extoff; - } - if (bno >= startoff + blockcount) { - if (++idx == nextents) { - ep = NULL; - } else { - ep = xfs_iext_get_ext(ifp, idx); - } - } - *idxp = idx; - return ep; -} - -/* - * Return a pointer to the indirection array entry containing the - * extent record for filesystem block bno. Store the index of the - * target irec in *erp_idxp. - */ -xfs_ext_irec_t * /* pointer to found extent record */ -xfs_iext_bno_to_irec( - xfs_ifork_t *ifp, /* inode fork pointer */ - xfs_fileoff_t bno, /* block number to search for */ - int *erp_idxp) /* irec index of target ext list */ -{ - xfs_ext_irec_t *erp = NULL; /* indirection array pointer */ - xfs_ext_irec_t *erp_next; /* next indirection array entry */ - int erp_idx; /* indirection array index */ - int nlists; /* number of extent irec's (lists) */ - int high; /* binary search upper limit */ - int low; /* binary search lower limit */ - - ASSERT(ifp->if_flags & XFS_IFEXTIREC); - nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ; - erp_idx = 0; - low = 0; - high = nlists - 1; - while (low <= high) { - erp_idx = (low + high) >> 1; - erp = &ifp->if_u1.if_ext_irec[erp_idx]; - erp_next = erp_idx < nlists - 1 ? erp + 1 : NULL; - if (bno < xfs_bmbt_get_startoff(erp->er_extbuf)) { - high = erp_idx - 1; - } else if (erp_next && bno >= - xfs_bmbt_get_startoff(erp_next->er_extbuf)) { - low = erp_idx + 1; - } else { - break; - } - } - *erp_idxp = erp_idx; - return erp; -} - -/* - * Return a pointer to the indirection array entry containing the - * extent record at file extent index *idxp. Store the index of the - * target irec in *erp_idxp and store the page index of the target - * extent record in *idxp. - */ -xfs_ext_irec_t * -xfs_iext_idx_to_irec( - xfs_ifork_t *ifp, /* inode fork pointer */ - xfs_extnum_t *idxp, /* extent index (file -> page) */ - int *erp_idxp, /* pointer to target irec */ - int realloc) /* new bytes were just added */ -{ - xfs_ext_irec_t *prev; /* pointer to previous irec */ - xfs_ext_irec_t *erp = NULL; /* pointer to current irec */ - int erp_idx; /* indirection array index */ - int nlists; /* number of irec's (ex lists) */ - int high; /* binary search upper limit */ - int low; /* binary search lower limit */ - xfs_extnum_t page_idx = *idxp; /* extent index in target list */ - - ASSERT(ifp->if_flags & XFS_IFEXTIREC); - ASSERT(page_idx >= 0); - ASSERT(page_idx <= ifp->if_bytes / sizeof(xfs_bmbt_rec_t)); - ASSERT(page_idx < ifp->if_bytes / sizeof(xfs_bmbt_rec_t) || realloc); - - nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ; - erp_idx = 0; - low = 0; - high = nlists - 1; - - /* Binary search extent irec's */ - while (low <= high) { - erp_idx = (low + high) >> 1; - erp = &ifp->if_u1.if_ext_irec[erp_idx]; - prev = erp_idx > 0 ? erp - 1 : NULL; - if (page_idx < erp->er_extoff || (page_idx == erp->er_extoff && - realloc && prev && prev->er_extcount < XFS_LINEAR_EXTS)) { - high = erp_idx - 1; - } else if (page_idx > erp->er_extoff + erp->er_extcount || - (page_idx == erp->er_extoff + erp->er_extcount && - !realloc)) { - low = erp_idx + 1; - } else if (page_idx == erp->er_extoff + erp->er_extcount && - erp->er_extcount == XFS_LINEAR_EXTS) { - ASSERT(realloc); - page_idx = 0; - erp_idx++; - erp = erp_idx < nlists ? erp + 1 : NULL; - break; - } else { - page_idx -= erp->er_extoff; - break; - } - } - *idxp = page_idx; - *erp_idxp = erp_idx; - return(erp); -} - -/* - * Allocate and initialize an indirection array once the space needed - * for incore extents increases above XFS_IEXT_BUFSZ. - */ -void -xfs_iext_irec_init( - xfs_ifork_t *ifp) /* inode fork pointer */ -{ - xfs_ext_irec_t *erp; /* indirection array pointer */ - xfs_extnum_t nextents; /* number of extents in file */ - - ASSERT(!(ifp->if_flags & XFS_IFEXTIREC)); - nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t); - ASSERT(nextents <= XFS_LINEAR_EXTS); - - erp = kmem_alloc(sizeof(xfs_ext_irec_t), KM_NOFS); - - if (nextents == 0) { - ifp->if_u1.if_extents = kmem_alloc(XFS_IEXT_BUFSZ, KM_NOFS); - } else if (!ifp->if_real_bytes) { - xfs_iext_inline_to_direct(ifp, XFS_IEXT_BUFSZ); - } else if (ifp->if_real_bytes < XFS_IEXT_BUFSZ) { - xfs_iext_realloc_direct(ifp, XFS_IEXT_BUFSZ); - } - erp->er_extbuf = ifp->if_u1.if_extents; - erp->er_extcount = nextents; - erp->er_extoff = 0; - - ifp->if_flags |= XFS_IFEXTIREC; - ifp->if_real_bytes = XFS_IEXT_BUFSZ; - ifp->if_bytes = nextents * sizeof(xfs_bmbt_rec_t); - ifp->if_u1.if_ext_irec = erp; - - return; -} - -/* - * Allocate and initialize a new entry in the indirection array. - */ -xfs_ext_irec_t * -xfs_iext_irec_new( - xfs_ifork_t *ifp, /* inode fork pointer */ - int erp_idx) /* index for new irec */ -{ - xfs_ext_irec_t *erp; /* indirection array pointer */ - int i; /* loop counter */ - int nlists; /* number of irec's (ex lists) */ - - ASSERT(ifp->if_flags & XFS_IFEXTIREC); - nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ; - - /* Resize indirection array */ - xfs_iext_realloc_indirect(ifp, ++nlists * - sizeof(xfs_ext_irec_t)); - /* - * Move records down in the array so the - * new page can use erp_idx. - */ - erp = ifp->if_u1.if_ext_irec; - for (i = nlists - 1; i > erp_idx; i--) { - memmove(&erp[i], &erp[i-1], sizeof(xfs_ext_irec_t)); - } - ASSERT(i == erp_idx); - - /* Initialize new extent record */ - erp = ifp->if_u1.if_ext_irec; - erp[erp_idx].er_extbuf = kmem_alloc(XFS_IEXT_BUFSZ, KM_NOFS); - ifp->if_real_bytes = nlists * XFS_IEXT_BUFSZ; - memset(erp[erp_idx].er_extbuf, 0, XFS_IEXT_BUFSZ); - erp[erp_idx].er_extcount = 0; - erp[erp_idx].er_extoff = erp_idx > 0 ? - erp[erp_idx-1].er_extoff + erp[erp_idx-1].er_extcount : 0; - return (&erp[erp_idx]); -} - -/* - * Remove a record from the indirection array. - */ -void -xfs_iext_irec_remove( - xfs_ifork_t *ifp, /* inode fork pointer */ - int erp_idx) /* irec index to remove */ -{ - xfs_ext_irec_t *erp; /* indirection array pointer */ - int i; /* loop counter */ - int nlists; /* number of irec's (ex lists) */ - - ASSERT(ifp->if_flags & XFS_IFEXTIREC); - nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ; - erp = &ifp->if_u1.if_ext_irec[erp_idx]; - if (erp->er_extbuf) { - xfs_iext_irec_update_extoffs(ifp, erp_idx + 1, - -erp->er_extcount); - kmem_free(erp->er_extbuf); - } - /* Compact extent records */ - erp = ifp->if_u1.if_ext_irec; - for (i = erp_idx; i < nlists - 1; i++) { - memmove(&erp[i], &erp[i+1], sizeof(xfs_ext_irec_t)); - } - /* - * Manually free the last extent record from the indirection - * array. A call to xfs_iext_realloc_indirect() with a size - * of zero would result in a call to xfs_iext_destroy() which - * would in turn call this function again, creating a nasty - * infinite loop. - */ - if (--nlists) { - xfs_iext_realloc_indirect(ifp, - nlists * sizeof(xfs_ext_irec_t)); - } else { - kmem_free(ifp->if_u1.if_ext_irec); - } - ifp->if_real_bytes = nlists * XFS_IEXT_BUFSZ; -} - -/* - * This is called to clean up large amounts of unused memory allocated - * by the indirection array. Before compacting anything though, verify - * that the indirection array is still needed and switch back to the - * linear extent list (or even the inline buffer) if possible. The - * compaction policy is as follows: - * - * Full Compaction: Extents fit into a single page (or inline buffer) - * Partial Compaction: Extents occupy less than 50% of allocated space - * No Compaction: Extents occupy at least 50% of allocated space - */ -void -xfs_iext_irec_compact( - xfs_ifork_t *ifp) /* inode fork pointer */ -{ - xfs_extnum_t nextents; /* number of extents in file */ - int nlists; /* number of irec's (ex lists) */ - - ASSERT(ifp->if_flags & XFS_IFEXTIREC); - nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ; - nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t); - - if (nextents == 0) { - xfs_iext_destroy(ifp); - } else if (nextents <= XFS_INLINE_EXTS) { - xfs_iext_indirect_to_direct(ifp); - xfs_iext_direct_to_inline(ifp, nextents); - } else if (nextents <= XFS_LINEAR_EXTS) { - xfs_iext_indirect_to_direct(ifp); - } else if (nextents < (nlists * XFS_LINEAR_EXTS) >> 1) { - xfs_iext_irec_compact_pages(ifp); - } -} - -/* - * Combine extents from neighboring extent pages. - */ -void -xfs_iext_irec_compact_pages( - xfs_ifork_t *ifp) /* inode fork pointer */ -{ - xfs_ext_irec_t *erp, *erp_next;/* pointers to irec entries */ - int erp_idx = 0; /* indirection array index */ - int nlists; /* number of irec's (ex lists) */ - - ASSERT(ifp->if_flags & XFS_IFEXTIREC); - nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ; - while (erp_idx < nlists - 1) { - erp = &ifp->if_u1.if_ext_irec[erp_idx]; - erp_next = erp + 1; - if (erp_next->er_extcount <= - (XFS_LINEAR_EXTS - erp->er_extcount)) { - memcpy(&erp->er_extbuf[erp->er_extcount], - erp_next->er_extbuf, erp_next->er_extcount * - sizeof(xfs_bmbt_rec_t)); - erp->er_extcount += erp_next->er_extcount; - /* - * Free page before removing extent record - * so er_extoffs don't get modified in - * xfs_iext_irec_remove. - */ - kmem_free(erp_next->er_extbuf); - erp_next->er_extbuf = NULL; - xfs_iext_irec_remove(ifp, erp_idx + 1); - nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ; - } else { - erp_idx++; - } - } -} - -/* - * This is called to update the er_extoff field in the indirection - * array when extents have been added or removed from one of the - * extent lists. erp_idx contains the irec index to begin updating - * at and ext_diff contains the number of extents that were added - * or removed. - */ -void -xfs_iext_irec_update_extoffs( - xfs_ifork_t *ifp, /* inode fork pointer */ - int erp_idx, /* irec index to update */ - int ext_diff) /* number of new extents */ -{ - int i; /* loop counter */ - int nlists; /* number of irec's (ex lists */ - - ASSERT(ifp->if_flags & XFS_IFEXTIREC); - nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ; - for (i = erp_idx; i < nlists; i++) { - ifp->if_u1.if_ext_irec[i].er_extoff += ext_diff; - } -} - -/* - * Test whether it is appropriate to check an inode for and free post EOF - * blocks. The 'force' parameter determines whether we should also consider - * regular files that are marked preallocated or append-only. - */ -bool -xfs_can_free_eofblocks(struct xfs_inode *ip, bool force) -{ - /* prealloc/delalloc exists only on regular files */ - if (!S_ISREG(ip->i_d.di_mode)) - return false; - - /* - * Zero sized files with no cached pages and delalloc blocks will not - * have speculative prealloc/delalloc blocks to remove. - */ - if (VFS_I(ip)->i_size == 0 && - VN_CACHED(VFS_I(ip)) == 0 && - ip->i_delayed_blks == 0) - return false; - - /* If we haven't read in the extent list, then don't do it now. */ - if (!(ip->i_df.if_flags & XFS_IFEXTENTS)) - return false; - - /* - * Do not free real preallocated or append-only files unless the file - * has delalloc blocks and we are forced to remove them. - */ - if (ip->i_d.di_flags & (XFS_DIFLAG_PREALLOC | XFS_DIFLAG_APPEND)) - if (!force || ip->i_delayed_blks == 0) - return false; - - return true; -} - diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h index b55fd34..4a91358 100644 --- a/fs/xfs/xfs_inode.h +++ b/fs/xfs/xfs_inode.h @@ -18,225 +18,15 @@ #ifndef __XFS_INODE_H__ #define __XFS_INODE_H__ -struct posix_acl; -struct xfs_dinode; -struct xfs_inode; - -/* - * Fork identifiers. - */ -#define XFS_DATA_FORK 0 -#define XFS_ATTR_FORK 1 - -/* - * The following xfs_ext_irec_t struct introduces a second (top) level - * to the in-core extent allocation scheme. These structs are allocated - * in a contiguous block, creating an indirection array where each entry - * (irec) contains a pointer to a buffer of in-core extent records which - * it manages. Each extent buffer is 4k in size, since 4k is the system - * page size on Linux i386 and systems with larger page sizes don't seem - * to gain much, if anything, by using their native page size as the - * extent buffer size. Also, using 4k extent buffers everywhere provides - * a consistent interface for CXFS across different platforms. - * - * There is currently no limit on the number of irec's (extent lists) - * allowed, so heavily fragmented files may require an indirection array - * which spans multiple system pages of memory. The number of extents - * which would require this amount of contiguous memory is very large - * and should not cause problems in the foreseeable future. However, - * if the memory needed for the contiguous array ever becomes a problem, - * it is possible that a third level of indirection may be required. - */ -typedef struct xfs_ext_irec { - xfs_bmbt_rec_host_t *er_extbuf; /* block of extent records */ - xfs_extnum_t er_extoff; /* extent offset in file */ - xfs_extnum_t er_extcount; /* number of extents in page/block */ -} xfs_ext_irec_t; +#include "xfs_inode_buf.h" +#include "xfs_inode_fork.h" /* - * File incore extent information, present for each of data & attr forks. + * Kernel only inode definitions */ -#define XFS_IEXT_BUFSZ 4096 -#define XFS_LINEAR_EXTS (XFS_IEXT_BUFSZ / (uint)sizeof(xfs_bmbt_rec_t)) -#define XFS_INLINE_EXTS 2 -#define XFS_INLINE_DATA 32 -typedef struct xfs_ifork { - int if_bytes; /* bytes in if_u1 */ - int if_real_bytes; /* bytes allocated in if_u1 */ - struct xfs_btree_block *if_broot; /* file's incore btree root */ - short if_broot_bytes; /* bytes allocated for root */ - unsigned char if_flags; /* per-fork flags */ - union { - xfs_bmbt_rec_host_t *if_extents;/* linear map file exts */ - xfs_ext_irec_t *if_ext_irec; /* irec map file exts */ - char *if_data; /* inline file data */ - } if_u1; - union { - xfs_bmbt_rec_host_t if_inline_ext[XFS_INLINE_EXTS]; - /* very small file extents */ - char if_inline_data[XFS_INLINE_DATA]; - /* very small file data */ - xfs_dev_t if_rdev; /* dev number if special */ - uuid_t if_uuid; /* mount point value */ - } if_u2; -} xfs_ifork_t; - -/* - * Inode location information. Stored in the inode and passed to - * xfs_imap_to_bp() to get a buffer and dinode for a given inode. - */ -struct xfs_imap { - xfs_daddr_t im_blkno; /* starting BB of inode chunk */ - ushort im_len; /* length in BBs of inode chunk */ - ushort im_boffset; /* inode offset in block in bytes */ -}; - -/* - * This is the xfs in-core inode structure. - * Most of the on-disk inode is embedded in the i_d field. - * - * The extent pointers/inline file space, however, are managed - * separately. The memory for this information is pointed to by - * the if_u1 unions depending on the type of the data. - * This is used to linearize the array of extents for fast in-core - * access. This is used until the file's number of extents - * surpasses XFS_MAX_INCORE_EXTENTS, at which point all extent pointers - * are accessed through the buffer cache. - * - * Other state kept in the in-core inode is used for identification, - * locking, transactional updating, etc of the inode. - * - * Generally, we do not want to hold the i_rlock while holding the - * i_ilock. Hierarchy is i_iolock followed by i_rlock. - * - * xfs_iptr_t contains all the inode fields up to and including the - * i_mnext and i_mprev fields, it is used as a marker in the inode - * chain off the mount structure by xfs_sync calls. - */ - -typedef struct xfs_ictimestamp { - __int32_t t_sec; /* timestamp seconds */ - __int32_t t_nsec; /* timestamp nanoseconds */ -} xfs_ictimestamp_t; - -/* - * NOTE: This structure must be kept identical to struct xfs_dinode - * in xfs_dinode.h except for the endianness annotations. - */ -typedef struct xfs_icdinode { - __uint16_t di_magic; /* inode magic # = XFS_DINODE_MAGIC */ - __uint16_t di_mode; /* mode and type of file */ - __int8_t di_version; /* inode version */ - __int8_t di_format; /* format of di_c data */ - __uint16_t di_onlink; /* old number of links to file */ - __uint32_t di_uid; /* owner's user id */ - __uint32_t di_gid; /* owner's group id */ - __uint32_t di_nlink; /* number of links to file */ - __uint16_t di_projid_lo; /* lower part of owner's project id */ - __uint16_t di_projid_hi; /* higher part of owner's project id */ - __uint8_t di_pad[6]; /* unused, zeroed space */ - __uint16_t di_flushiter; /* incremented on flush */ - xfs_ictimestamp_t di_atime; /* time last accessed */ - xfs_ictimestamp_t di_mtime; /* time last modified */ - xfs_ictimestamp_t di_ctime; /* time created/inode modified */ - xfs_fsize_t di_size; /* number of bytes in file */ - xfs_drfsbno_t di_nblocks; /* # of direct & btree blocks used */ - xfs_extlen_t di_extsize; /* basic/minimum extent size for file */ - xfs_extnum_t di_nextents; /* number of extents in data fork */ - xfs_aextnum_t di_anextents; /* number of extents in attribute fork*/ - __uint8_t di_forkoff; /* attr fork offs, <<3 for 64b align */ - __int8_t di_aformat; /* format of attr fork's data */ - __uint32_t di_dmevmask; /* DMIG event mask */ - __uint16_t di_dmstate; /* DMIG state info */ - __uint16_t di_flags; /* random flags, XFS_DIFLAG_... */ - __uint32_t di_gen; /* generation number */ - - /* di_next_unlinked is the only non-core field in the old dinode */ - xfs_agino_t di_next_unlinked;/* agi unlinked list ptr */ - - /* start of the extended dinode, writable fields */ - __uint32_t di_crc; /* CRC of the inode */ - __uint64_t di_changecount; /* number of attribute changes */ - xfs_lsn_t di_lsn; /* flush sequence */ - __uint64_t di_flags2; /* more random flags */ - __uint8_t di_pad2[16]; /* more padding for future expansion */ - - /* fields only written to during inode creation */ - xfs_ictimestamp_t di_crtime; /* time created */ - xfs_ino_t di_ino; /* inode number */ - uuid_t di_uuid; /* UUID of the filesystem */ - - /* structure must be padded to 64 bit alignment */ -} xfs_icdinode_t; - -static inline uint xfs_icdinode_size(int version) -{ - if (version == 3) - return sizeof(struct xfs_icdinode); - return offsetof(struct xfs_icdinode, di_next_unlinked); -} - -/* - * Flags for xfs_ichgtime(). - */ -#define XFS_ICHGTIME_MOD 0x1 /* data fork modification timestamp */ -#define XFS_ICHGTIME_CHG 0x2 /* inode field change timestamp */ -#define XFS_ICHGTIME_CREATE 0x4 /* inode create timestamp */ - -/* - * Per-fork incore inode flags. - */ -#define XFS_IFINLINE 0x01 /* Inline data is read in */ -#define XFS_IFEXTENTS 0x02 /* All extent pointers are read in */ -#define XFS_IFBROOT 0x04 /* i_broot points to the bmap b-tree root */ -#define XFS_IFEXTIREC 0x08 /* Indirection array of extent blocks */ - -/* - * Fork handling. - */ - -#define XFS_IFORK_Q(ip) ((ip)->i_d.di_forkoff != 0) -#define XFS_IFORK_BOFF(ip) ((int)((ip)->i_d.di_forkoff << 3)) - -#define XFS_IFORK_PTR(ip,w) \ - ((w) == XFS_DATA_FORK ? \ - &(ip)->i_df : \ - (ip)->i_afp) -#define XFS_IFORK_DSIZE(ip) \ - (XFS_IFORK_Q(ip) ? \ - XFS_IFORK_BOFF(ip) : \ - XFS_LITINO((ip)->i_mount, (ip)->i_d.di_version)) -#define XFS_IFORK_ASIZE(ip) \ - (XFS_IFORK_Q(ip) ? \ - XFS_LITINO((ip)->i_mount, (ip)->i_d.di_version) - \ - XFS_IFORK_BOFF(ip) : \ - 0) -#define XFS_IFORK_SIZE(ip,w) \ - ((w) == XFS_DATA_FORK ? \ - XFS_IFORK_DSIZE(ip) : \ - XFS_IFORK_ASIZE(ip)) -#define XFS_IFORK_FORMAT(ip,w) \ - ((w) == XFS_DATA_FORK ? \ - (ip)->i_d.di_format : \ - (ip)->i_d.di_aformat) -#define XFS_IFORK_FMT_SET(ip,w,n) \ - ((w) == XFS_DATA_FORK ? \ - ((ip)->i_d.di_format = (n)) : \ - ((ip)->i_d.di_aformat = (n))) -#define XFS_IFORK_NEXTENTS(ip,w) \ - ((w) == XFS_DATA_FORK ? \ - (ip)->i_d.di_nextents : \ - (ip)->i_d.di_anextents) -#define XFS_IFORK_NEXT_SET(ip,w,n) \ - ((w) == XFS_DATA_FORK ? \ - ((ip)->i_d.di_nextents = (n)) : \ - ((ip)->i_d.di_anextents = (n))) -#define XFS_IFORK_MAXEXT(ip, w) \ - (XFS_IFORK_SIZE(ip, w) / sizeof(xfs_bmbt_rec_t)) - - -#ifdef __KERNEL__ +struct xfs_dinode; +struct xfs_inode; struct xfs_buf; struct xfs_bmap_free; struct xfs_bmbt_irec; @@ -525,9 +315,21 @@ static inline int xfs_isiflocked(struct xfs_inode *ip) ((pip)->i_d.di_mode & S_ISGID)) -/* - * xfs_inode.c prototypes. - */ +int xfs_release(struct xfs_inode *ip); +int xfs_inactive(struct xfs_inode *ip); +int xfs_lookup(struct xfs_inode *dp, struct xfs_name *name, + struct xfs_inode **ipp, struct xfs_name *ci_name); +int xfs_create(struct xfs_inode *dp, struct xfs_name *name, + umode_t mode, xfs_dev_t rdev, struct xfs_inode **ipp); +int xfs_remove(struct xfs_inode *dp, struct xfs_name *name, + struct xfs_inode *ip); +int xfs_link(struct xfs_inode *tdp, struct xfs_inode *sip, + struct xfs_name *target_name); +int xfs_rename(struct xfs_inode *src_dp, struct xfs_name *src_name, + struct xfs_inode *src_ip, struct xfs_inode *target_dp, + struct xfs_name *target_name, + struct xfs_inode *target_ip); + void xfs_ilock(xfs_inode_t *, uint); int xfs_ilock_nowait(xfs_inode_t *, uint); void xfs_iunlock(xfs_inode_t *, uint); @@ -548,13 +350,28 @@ int xfs_itruncate_extents(struct xfs_trans **, struct xfs_inode *, int xfs_iunlink(struct xfs_trans *, xfs_inode_t *); void xfs_iext_realloc(xfs_inode_t *, int, int); + void xfs_iunpin_wait(xfs_inode_t *); +#define xfs_ipincount(ip) ((unsigned int) atomic_read(&ip->i_pincount)) + int xfs_iflush(struct xfs_inode *, struct xfs_buf **); void xfs_lock_inodes(xfs_inode_t **, int, uint); void xfs_lock_two_inodes(xfs_inode_t *, xfs_inode_t *, uint); xfs_extlen_t xfs_get_extsz_hint(struct xfs_inode *ip); +int xfs_dir_ialloc(struct xfs_trans **, struct xfs_inode *, umode_t, + xfs_nlink_t, xfs_dev_t, prid_t, int, + struct xfs_inode **, int *); +int xfs_droplink(struct xfs_trans *, struct xfs_inode *); +int xfs_bumplink(struct xfs_trans *, struct xfs_inode *); +void xfs_bump_ino_vers2(struct xfs_trans *, struct xfs_inode *); + +/* from xfs_file.c */ +int xfs_zero_eof(struct xfs_inode *, xfs_off_t, xfs_fsize_t); +int xfs_iozero(struct xfs_inode *, loff_t, size_t); + + #define IHOLD(ip) \ do { \ ASSERT(atomic_read(&VFS_I(ip)->i_count) > 0) ; \ @@ -568,65 +385,6 @@ do { \ iput(VFS_I(ip)); \ } while (0) -#endif /* __KERNEL__ */ - -/* - * Flags for xfs_iget() - */ -#define XFS_IGET_CREATE 0x1 -#define XFS_IGET_UNTRUSTED 0x2 -#define XFS_IGET_DONTCACHE 0x4 - -int xfs_imap_to_bp(struct xfs_mount *, struct xfs_trans *, - struct xfs_imap *, struct xfs_dinode **, - struct xfs_buf **, uint, uint); -int xfs_iread(struct xfs_mount *, struct xfs_trans *, - struct xfs_inode *, uint); -void xfs_dinode_calc_crc(struct xfs_mount *, struct xfs_dinode *); -void xfs_dinode_to_disk(struct xfs_dinode *, - struct xfs_icdinode *); -void xfs_idestroy_fork(struct xfs_inode *, int); -void xfs_idata_realloc(struct xfs_inode *, int, int); -void xfs_iroot_realloc(struct xfs_inode *, int, int); -int xfs_iread_extents(struct xfs_trans *, struct xfs_inode *, int); -int xfs_iextents_copy(struct xfs_inode *, xfs_bmbt_rec_t *, int); - -xfs_bmbt_rec_host_t *xfs_iext_get_ext(xfs_ifork_t *, xfs_extnum_t); -void xfs_iext_insert(xfs_inode_t *, xfs_extnum_t, xfs_extnum_t, - xfs_bmbt_irec_t *, int); -void xfs_iext_add(xfs_ifork_t *, xfs_extnum_t, int); -void xfs_iext_add_indirect_multi(xfs_ifork_t *, int, xfs_extnum_t, int); -void xfs_iext_remove(xfs_inode_t *, xfs_extnum_t, int, int); -void xfs_iext_remove_inline(xfs_ifork_t *, xfs_extnum_t, int); -void xfs_iext_remove_direct(xfs_ifork_t *, xfs_extnum_t, int); -void xfs_iext_remove_indirect(xfs_ifork_t *, xfs_extnum_t, int); -void xfs_iext_realloc_direct(xfs_ifork_t *, int); -void xfs_iext_direct_to_inline(xfs_ifork_t *, xfs_extnum_t); -void xfs_iext_inline_to_direct(xfs_ifork_t *, int); -void xfs_iext_destroy(xfs_ifork_t *); -xfs_bmbt_rec_host_t *xfs_iext_bno_to_ext(xfs_ifork_t *, xfs_fileoff_t, int *); -xfs_ext_irec_t *xfs_iext_bno_to_irec(xfs_ifork_t *, xfs_fileoff_t, int *); -xfs_ext_irec_t *xfs_iext_idx_to_irec(xfs_ifork_t *, xfs_extnum_t *, int *, int); -void xfs_iext_irec_init(xfs_ifork_t *); -xfs_ext_irec_t *xfs_iext_irec_new(xfs_ifork_t *, int); -void xfs_iext_irec_remove(xfs_ifork_t *, int); -void xfs_iext_irec_compact(xfs_ifork_t *); -void xfs_iext_irec_compact_pages(xfs_ifork_t *); -void xfs_iext_irec_compact_full(xfs_ifork_t *); -void xfs_iext_irec_update_extoffs(xfs_ifork_t *, int, int); -bool xfs_can_free_eofblocks(struct xfs_inode *, bool); - -#define xfs_ipincount(ip) ((unsigned int) atomic_read(&ip->i_pincount)) - -#if defined(DEBUG) -void xfs_inobp_check(struct xfs_mount *, struct xfs_buf *); -#else -#define xfs_inobp_check(mp, bp) -#endif /* DEBUG */ - -extern struct kmem_zone *xfs_ifork_zone; extern struct kmem_zone *xfs_inode_zone; -extern struct kmem_zone *xfs_ili_zone; -extern const struct xfs_buf_ops xfs_inode_buf_ops; #endif /* __XFS_INODE_H__ */ diff --git a/fs/xfs/xfs_inode_buf.c b/fs/xfs/xfs_inode_buf.c new file mode 100644 index 0000000..e011d59 --- /dev/null +++ b/fs/xfs/xfs_inode_buf.c @@ -0,0 +1,483 @@ +/* + * Copyright (c) 2000-2006 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#include "xfs.h" +#include "xfs_fs.h" +#include "xfs_format.h" +#include "xfs_log.h" +#include "xfs_trans.h" +#include "xfs_sb.h" +#include "xfs_ag.h" +#include "xfs_mount.h" +#include "xfs_bmap_btree.h" +#include "xfs_ialloc_btree.h" +#include "xfs_dinode.h" +#include "xfs_inode.h" +#include "xfs_error.h" +#include "xfs_cksum.h" +#include "xfs_icache.h" +#include "xfs_ialloc.h" + +/* + * Check that none of the inode's in the buffer have a next + * unlinked field of 0. + */ +#if defined(DEBUG) +void +xfs_inobp_check( + xfs_mount_t *mp, + xfs_buf_t *bp) +{ + int i; + int j; + xfs_dinode_t *dip; + + j = mp->m_inode_cluster_size >> mp->m_sb.sb_inodelog; + + for (i = 0; i < j; i++) { + dip = (xfs_dinode_t *)xfs_buf_offset(bp, + i * mp->m_sb.sb_inodesize); + if (!dip->di_next_unlinked) { + xfs_alert(mp, + "Detected bogus zero next_unlinked field in incore inode buffer 0x%p.", + bp); + ASSERT(dip->di_next_unlinked); + } + } +} +#endif + +/* + * If we are doing readahead on an inode buffer, we might be in log recovery + * reading an inode allocation buffer that hasn't yet been replayed, and hence + * has not had the inode cores stamped into it. Hence for readahead, the buffer + * may be potentially invalid. + * + * If the readahead buffer is invalid, we don't want to mark it with an error, + * but we do want to clear the DONE status of the buffer so that a followup read + * will re-read it from disk. This will ensure that we don't get an unnecessary + * warnings during log recovery and we don't get unnecssary panics on debug + * kernels. + */ +static void +xfs_inode_buf_verify( + struct xfs_buf *bp, + bool readahead) +{ + struct xfs_mount *mp = bp->b_target->bt_mount; + int i; + int ni; + + /* + * Validate the magic number and version of every inode in the buffer + */ + ni = XFS_BB_TO_FSB(mp, bp->b_length) * mp->m_sb.sb_inopblock; + for (i = 0; i < ni; i++) { + int di_ok; + xfs_dinode_t *dip; + + dip = (struct xfs_dinode *)xfs_buf_offset(bp, + (i << mp->m_sb.sb_inodelog)); + di_ok = dip->di_magic == cpu_to_be16(XFS_DINODE_MAGIC) && + XFS_DINODE_GOOD_VERSION(dip->di_version); + if (unlikely(XFS_TEST_ERROR(!di_ok, mp, + XFS_ERRTAG_ITOBP_INOTOBP, + XFS_RANDOM_ITOBP_INOTOBP))) { + if (readahead) { + bp->b_flags &= ~XBF_DONE; + return; + } + + xfs_buf_ioerror(bp, EFSCORRUPTED); + XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_HIGH, + mp, dip); +#ifdef DEBUG + xfs_emerg(mp, + "bad inode magic/vsn daddr %lld #%d (magic=%x)", + (unsigned long long)bp->b_bn, i, + be16_to_cpu(dip->di_magic)); + ASSERT(0); +#endif + } + } + xfs_inobp_check(mp, bp); +} + + +static void +xfs_inode_buf_read_verify( + struct xfs_buf *bp) +{ + xfs_inode_buf_verify(bp, false); +} + +static void +xfs_inode_buf_readahead_verify( + struct xfs_buf *bp) +{ + xfs_inode_buf_verify(bp, true); +} + +static void +xfs_inode_buf_write_verify( + struct xfs_buf *bp) +{ + xfs_inode_buf_verify(bp, false); +} + +const struct xfs_buf_ops xfs_inode_buf_ops = { + .verify_read = xfs_inode_buf_read_verify, + .verify_write = xfs_inode_buf_write_verify, +}; + +const struct xfs_buf_ops xfs_inode_buf_ra_ops = { + .verify_read = xfs_inode_buf_readahead_verify, + .verify_write = xfs_inode_buf_write_verify, +}; + + +/* + * This routine is called to map an inode to the buffer containing the on-disk + * version of the inode. It returns a pointer to the buffer containing the + * on-disk inode in the bpp parameter, and in the dipp parameter it returns a + * pointer to the on-disk inode within that buffer. + * + * If a non-zero error is returned, then the contents of bpp and dipp are + * undefined. + */ +int +xfs_imap_to_bp( + struct xfs_mount *mp, + struct xfs_trans *tp, + struct xfs_imap *imap, + struct xfs_dinode **dipp, + struct xfs_buf **bpp, + uint buf_flags, + uint iget_flags) +{ + struct xfs_buf *bp; + int error; + + buf_flags |= XBF_UNMAPPED; + error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp, imap->im_blkno, + (int)imap->im_len, buf_flags, &bp, + &xfs_inode_buf_ops); + if (error) { + if (error == EAGAIN) { + ASSERT(buf_flags & XBF_TRYLOCK); + return error; + } + + if (error == EFSCORRUPTED && + (iget_flags & XFS_IGET_UNTRUSTED)) + return XFS_ERROR(EINVAL); + + xfs_warn(mp, "%s: xfs_trans_read_buf() returned error %d.", + __func__, error); + return error; + } + + *bpp = bp; + *dipp = (struct xfs_dinode *)xfs_buf_offset(bp, imap->im_boffset); + return 0; +} + +STATIC void +xfs_dinode_from_disk( + xfs_icdinode_t *to, + xfs_dinode_t *from) +{ + to->di_magic = be16_to_cpu(from->di_magic); + to->di_mode = be16_to_cpu(from->di_mode); + to->di_version = from ->di_version; + to->di_format = from->di_format; + to->di_onlink = be16_to_cpu(from->di_onlink); + to->di_uid = be32_to_cpu(from->di_uid); + to->di_gid = be32_to_cpu(from->di_gid); + to->di_nlink = be32_to_cpu(from->di_nlink); + to->di_projid_lo = be16_to_cpu(from->di_projid_lo); + to->di_projid_hi = be16_to_cpu(from->di_projid_hi); + memcpy(to->di_pad, from->di_pad, sizeof(to->di_pad)); + to->di_flushiter = be16_to_cpu(from->di_flushiter); + to->di_atime.t_sec = be32_to_cpu(from->di_atime.t_sec); + to->di_atime.t_nsec = be32_to_cpu(from->di_atime.t_nsec); + to->di_mtime.t_sec = be32_to_cpu(from->di_mtime.t_sec); + to->di_mtime.t_nsec = be32_to_cpu(from->di_mtime.t_nsec); + to->di_ctime.t_sec = be32_to_cpu(from->di_ctime.t_sec); + to->di_ctime.t_nsec = be32_to_cpu(from->di_ctime.t_nsec); + to->di_size = be64_to_cpu(from->di_size); + to->di_nblocks = be64_to_cpu(from->di_nblocks); + to->di_extsize = be32_to_cpu(from->di_extsize); + to->di_nextents = be32_to_cpu(from->di_nextents); + to->di_anextents = be16_to_cpu(from->di_anextents); + to->di_forkoff = from->di_forkoff; + to->di_aformat = from->di_aformat; + to->di_dmevmask = be32_to_cpu(from->di_dmevmask); + to->di_dmstate = be16_to_cpu(from->di_dmstate); + to->di_flags = be16_to_cpu(from->di_flags); + to->di_gen = be32_to_cpu(from->di_gen); + + if (to->di_version == 3) { + to->di_changecount = be64_to_cpu(from->di_changecount); + to->di_crtime.t_sec = be32_to_cpu(from->di_crtime.t_sec); + to->di_crtime.t_nsec = be32_to_cpu(from->di_crtime.t_nsec); + to->di_flags2 = be64_to_cpu(from->di_flags2); + to->di_ino = be64_to_cpu(from->di_ino); + to->di_lsn = be64_to_cpu(from->di_lsn); + memcpy(to->di_pad2, from->di_pad2, sizeof(to->di_pad2)); + uuid_copy(&to->di_uuid, &from->di_uuid); + } +} + +void +xfs_dinode_to_disk( + xfs_dinode_t *to, + xfs_icdinode_t *from) +{ + to->di_magic = cpu_to_be16(from->di_magic); + to->di_mode = cpu_to_be16(from->di_mode); + to->di_version = from ->di_version; + to->di_format = from->di_format; + to->di_onlink = cpu_to_be16(from->di_onlink); + to->di_uid = cpu_to_be32(from->di_uid); + to->di_gid = cpu_to_be32(from->di_gid); + to->di_nlink = cpu_to_be32(from->di_nlink); + to->di_projid_lo = cpu_to_be16(from->di_projid_lo); + to->di_projid_hi = cpu_to_be16(from->di_projid_hi); + memcpy(to->di_pad, from->di_pad, sizeof(to->di_pad)); + to->di_atime.t_sec = cpu_to_be32(from->di_atime.t_sec); + to->di_atime.t_nsec = cpu_to_be32(from->di_atime.t_nsec); + to->di_mtime.t_sec = cpu_to_be32(from->di_mtime.t_sec); + to->di_mtime.t_nsec = cpu_to_be32(from->di_mtime.t_nsec); + to->di_ctime.t_sec = cpu_to_be32(from->di_ctime.t_sec); + to->di_ctime.t_nsec = cpu_to_be32(from->di_ctime.t_nsec); + to->di_size = cpu_to_be64(from->di_size); + to->di_nblocks = cpu_to_be64(from->di_nblocks); + to->di_extsize = cpu_to_be32(from->di_extsize); + to->di_nextents = cpu_to_be32(from->di_nextents); + to->di_anextents = cpu_to_be16(from->di_anextents); + to->di_forkoff = from->di_forkoff; + to->di_aformat = from->di_aformat; + to->di_dmevmask = cpu_to_be32(from->di_dmevmask); + to->di_dmstate = cpu_to_be16(from->di_dmstate); + to->di_flags = cpu_to_be16(from->di_flags); + to->di_gen = cpu_to_be32(from->di_gen); + + if (from->di_version == 3) { + to->di_changecount = cpu_to_be64(from->di_changecount); + to->di_crtime.t_sec = cpu_to_be32(from->di_crtime.t_sec); + to->di_crtime.t_nsec = cpu_to_be32(from->di_crtime.t_nsec); + to->di_flags2 = cpu_to_be64(from->di_flags2); + to->di_ino = cpu_to_be64(from->di_ino); + to->di_lsn = cpu_to_be64(from->di_lsn); + memcpy(to->di_pad2, from->di_pad2, sizeof(to->di_pad2)); + uuid_copy(&to->di_uuid, &from->di_uuid); + to->di_flushiter = 0; + } else { + to->di_flushiter = cpu_to_be16(from->di_flushiter); + } +} + +static bool +xfs_dinode_verify( + struct xfs_mount *mp, + struct xfs_inode *ip, + struct xfs_dinode *dip) +{ + if (dip->di_magic != cpu_to_be16(XFS_DINODE_MAGIC)) + return false; + + /* only version 3 or greater inodes are extensively verified here */ + if (dip->di_version < 3) + return true; + + if (!xfs_sb_version_hascrc(&mp->m_sb)) + return false; + if (!xfs_verify_cksum((char *)dip, mp->m_sb.sb_inodesize, + offsetof(struct xfs_dinode, di_crc))) + return false; + if (be64_to_cpu(dip->di_ino) != ip->i_ino) + return false; + if (!uuid_equal(&dip->di_uuid, &mp->m_sb.sb_uuid)) + return false; + return true; +} + +void +xfs_dinode_calc_crc( + struct xfs_mount *mp, + struct xfs_dinode *dip) +{ + __uint32_t crc; + + if (dip->di_version < 3) + return; + + ASSERT(xfs_sb_version_hascrc(&mp->m_sb)); + crc = xfs_start_cksum((char *)dip, mp->m_sb.sb_inodesize, + offsetof(struct xfs_dinode, di_crc)); + dip->di_crc = xfs_end_cksum(crc); +} + +/* + * Read the disk inode attributes into the in-core inode structure. + * + * For version 5 superblocks, if we are initialising a new inode and we are not + * utilising the XFS_MOUNT_IKEEP inode cluster mode, we can simple build the new + * inode core with a random generation number. If we are keeping inodes around, + * we need to read the inode cluster to get the existing generation number off + * disk. Further, if we are using version 4 superblocks (i.e. v1/v2 inode + * format) then log recovery is dependent on the di_flushiter field being + * initialised from the current on-disk value and hence we must also read the + * inode off disk. + */ +int +xfs_iread( + xfs_mount_t *mp, + xfs_trans_t *tp, + xfs_inode_t *ip, + uint iget_flags) +{ + xfs_buf_t *bp; + xfs_dinode_t *dip; + int error; + + /* + * Fill in the location information in the in-core inode. + */ + error = xfs_imap(mp, tp, ip->i_ino, &ip->i_imap, iget_flags); + if (error) + return error; + + /* shortcut IO on inode allocation if possible */ + if ((iget_flags & XFS_IGET_CREATE) && + xfs_sb_version_hascrc(&mp->m_sb) && + !(mp->m_flags & XFS_MOUNT_IKEEP)) { + /* initialise the on-disk inode core */ + memset(&ip->i_d, 0, sizeof(ip->i_d)); + ip->i_d.di_magic = XFS_DINODE_MAGIC; + ip->i_d.di_gen = prandom_u32(); + if (xfs_sb_version_hascrc(&mp->m_sb)) { + ip->i_d.di_version = 3; + ip->i_d.di_ino = ip->i_ino; + uuid_copy(&ip->i_d.di_uuid, &mp->m_sb.sb_uuid); + } else + ip->i_d.di_version = 2; + return 0; + } + + /* + * Get pointers to the on-disk inode and the buffer containing it. + */ + error = xfs_imap_to_bp(mp, tp, &ip->i_imap, &dip, &bp, 0, iget_flags); + if (error) + return error; + + /* even unallocated inodes are verified */ + if (!xfs_dinode_verify(mp, ip, dip)) { + xfs_alert(mp, "%s: validation failed for inode %lld failed", + __func__, ip->i_ino); + + XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, dip); + error = XFS_ERROR(EFSCORRUPTED); + goto out_brelse; + } + + /* + * If the on-disk inode is already linked to a directory + * entry, copy all of the inode into the in-core inode. + * xfs_iformat_fork() handles copying in the inode format + * specific information. + * Otherwise, just get the truly permanent information. + */ + if (dip->di_mode) { + xfs_dinode_from_disk(&ip->i_d, dip); + error = xfs_iformat_fork(ip, dip); + if (error) { +#ifdef DEBUG + xfs_alert(mp, "%s: xfs_iformat() returned error %d", + __func__, error); +#endif /* DEBUG */ + goto out_brelse; + } + } else { + /* + * Partial initialisation of the in-core inode. Just the bits + * that xfs_ialloc won't overwrite or relies on being correct. + */ + ip->i_d.di_magic = be16_to_cpu(dip->di_magic); + ip->i_d.di_version = dip->di_version; + ip->i_d.di_gen = be32_to_cpu(dip->di_gen); + ip->i_d.di_flushiter = be16_to_cpu(dip->di_flushiter); + + if (dip->di_version == 3) { + ip->i_d.di_ino = be64_to_cpu(dip->di_ino); + uuid_copy(&ip->i_d.di_uuid, &dip->di_uuid); + } + + /* + * Make sure to pull in the mode here as well in + * case the inode is released without being used. + * This ensures that xfs_inactive() will see that + * the inode is already free and not try to mess + * with the uninitialized part of it. + */ + ip->i_d.di_mode = 0; + } + + /* + * The inode format changed when we moved the link count and + * made it 32 bits long. If this is an old format inode, + * convert it in memory to look like a new one. If it gets + * flushed to disk we will convert back before flushing or + * logging it. We zero out the new projid field and the old link + * count field. We'll handle clearing the pad field (the remains + * of the old uuid field) when we actually convert the inode to + * the new format. We don't change the version number so that we + * can distinguish this from a real new format inode. + */ + if (ip->i_d.di_version == 1) { + ip->i_d.di_nlink = ip->i_d.di_onlink; + ip->i_d.di_onlink = 0; + xfs_set_projid(ip, 0); + } + + ip->i_delayed_blks = 0; + + /* + * Mark the buffer containing the inode as something to keep + * around for a while. This helps to keep recently accessed + * meta-data in-core longer. + */ + xfs_buf_set_ref(bp, XFS_INO_REF); + + /* + * Use xfs_trans_brelse() to release the buffer containing the on-disk + * inode, because it was acquired with xfs_trans_read_buf() in + * xfs_imap_to_bp() above. If tp is NULL, this is just a normal + * brelse(). If we're within a transaction, then xfs_trans_brelse() + * will only release the buffer if it is not dirty within the + * transaction. It will be OK to release the buffer in this case, + * because inodes on disk are never destroyed and we will be locking the + * new in-core inode before putting it in the cache where other + * processes can find it. Thus we don't have to worry about the inode + * being changed just because we released the buffer. + */ + out_brelse: + xfs_trans_brelse(tp, bp); + return error; +} diff --git a/fs/xfs/xfs_inode_buf.h b/fs/xfs/xfs_inode_buf.h new file mode 100644 index 0000000..599e6c0 --- /dev/null +++ b/fs/xfs/xfs_inode_buf.h @@ -0,0 +1,53 @@ +/* + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#ifndef __XFS_INODE_BUF_H__ +#define __XFS_INODE_BUF_H__ + +struct xfs_inode; +struct xfs_dinode; +struct xfs_icdinode; + +/* + * Inode location information. Stored in the inode and passed to + * xfs_imap_to_bp() to get a buffer and dinode for a given inode. + */ +struct xfs_imap { + xfs_daddr_t im_blkno; /* starting BB of inode chunk */ + ushort im_len; /* length in BBs of inode chunk */ + ushort im_boffset; /* inode offset in block in bytes */ +}; + +int xfs_imap_to_bp(struct xfs_mount *, struct xfs_trans *, + struct xfs_imap *, struct xfs_dinode **, + struct xfs_buf **, uint, uint); +int xfs_iread(struct xfs_mount *, struct xfs_trans *, + struct xfs_inode *, uint); +void xfs_dinode_calc_crc(struct xfs_mount *, struct xfs_dinode *); +void xfs_dinode_to_disk(struct xfs_dinode *, + struct xfs_icdinode *); + +#if defined(DEBUG) +void xfs_inobp_check(struct xfs_mount *, struct xfs_buf *); +#else +#define xfs_inobp_check(mp, bp) +#endif /* DEBUG */ + +extern const struct xfs_buf_ops xfs_inode_buf_ops; +extern const struct xfs_buf_ops xfs_inode_buf_ra_ops; + +#endif /* __XFS_INODE_BUF_H__ */ diff --git a/fs/xfs/xfs_inode_fork.c b/fs/xfs/xfs_inode_fork.c new file mode 100644 index 0000000..02f1083 --- /dev/null +++ b/fs/xfs/xfs_inode_fork.c @@ -0,0 +1,1920 @@ +/* + * Copyright (c) 2000-2006 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#include <linux/log2.h> + +#include "xfs.h" +#include "xfs_fs.h" +#include "xfs_format.h" +#include "xfs_log.h" +#include "xfs_inum.h" +#include "xfs_trans.h" +#include "xfs_trans_priv.h" +#include "xfs_sb.h" +#include "xfs_ag.h" +#include "xfs_mount.h" +#include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" +#include "xfs_ialloc_btree.h" +#include "xfs_attr_sf.h" +#include "xfs_dinode.h" +#include "xfs_inode.h" +#include "xfs_buf_item.h" +#include "xfs_inode_item.h" +#include "xfs_btree.h" +#include "xfs_alloc.h" +#include "xfs_ialloc.h" +#include "xfs_bmap.h" +#include "xfs_error.h" +#include "xfs_quota.h" +#include "xfs_filestream.h" +#include "xfs_cksum.h" +#include "xfs_trace.h" +#include "xfs_icache.h" + +kmem_zone_t *xfs_ifork_zone; + +STATIC int xfs_iformat_local(xfs_inode_t *, xfs_dinode_t *, int, int); +STATIC int xfs_iformat_extents(xfs_inode_t *, xfs_dinode_t *, int); +STATIC int xfs_iformat_btree(xfs_inode_t *, xfs_dinode_t *, int); + +#ifdef DEBUG +/* + * Make sure that the extents in the given memory buffer + * are valid. + */ +void +xfs_validate_extents( + xfs_ifork_t *ifp, + int nrecs, + xfs_exntfmt_t fmt) +{ + xfs_bmbt_irec_t irec; + xfs_bmbt_rec_host_t rec; + int i; + + for (i = 0; i < nrecs; i++) { + xfs_bmbt_rec_host_t *ep = xfs_iext_get_ext(ifp, i); + rec.l0 = get_unaligned(&ep->l0); + rec.l1 = get_unaligned(&ep->l1); + xfs_bmbt_get_all(&rec, &irec); + if (fmt == XFS_EXTFMT_NOSTATE) + ASSERT(irec.br_state == XFS_EXT_NORM); + } +} +#else /* DEBUG */ +#define xfs_validate_extents(ifp, nrecs, fmt) +#endif /* DEBUG */ + + +/* + * Move inode type and inode format specific information from the + * on-disk inode to the in-core inode. For fifos, devs, and sockets + * this means set if_rdev to the proper value. For files, directories, + * and symlinks this means to bring in the in-line data or extent + * pointers. For a file in B-tree format, only the root is immediately + * brought in-core. The rest will be in-lined in if_extents when it + * is first referenced (see xfs_iread_extents()). + */ +int +xfs_iformat_fork( + xfs_inode_t *ip, + xfs_dinode_t *dip) +{ + xfs_attr_shortform_t *atp; + int size; + int error = 0; + xfs_fsize_t di_size; + + if (unlikely(be32_to_cpu(dip->di_nextents) + + be16_to_cpu(dip->di_anextents) > + be64_to_cpu(dip->di_nblocks))) { + xfs_warn(ip->i_mount, + "corrupt dinode %Lu, extent total = %d, nblocks = %Lu.", + (unsigned long long)ip->i_ino, + (int)(be32_to_cpu(dip->di_nextents) + + be16_to_cpu(dip->di_anextents)), + (unsigned long long) + be64_to_cpu(dip->di_nblocks)); + XFS_CORRUPTION_ERROR("xfs_iformat(1)", XFS_ERRLEVEL_LOW, + ip->i_mount, dip); + return XFS_ERROR(EFSCORRUPTED); + } + + if (unlikely(dip->di_forkoff > ip->i_mount->m_sb.sb_inodesize)) { + xfs_warn(ip->i_mount, "corrupt dinode %Lu, forkoff = 0x%x.", + (unsigned long long)ip->i_ino, + dip->di_forkoff); + XFS_CORRUPTION_ERROR("xfs_iformat(2)", XFS_ERRLEVEL_LOW, + ip->i_mount, dip); + return XFS_ERROR(EFSCORRUPTED); + } + + if (unlikely((ip->i_d.di_flags & XFS_DIFLAG_REALTIME) && + !ip->i_mount->m_rtdev_targp)) { + xfs_warn(ip->i_mount, + "corrupt dinode %Lu, has realtime flag set.", + ip->i_ino); + XFS_CORRUPTION_ERROR("xfs_iformat(realtime)", + XFS_ERRLEVEL_LOW, ip->i_mount, dip); + return XFS_ERROR(EFSCORRUPTED); + } + + switch (ip->i_d.di_mode & S_IFMT) { + case S_IFIFO: + case S_IFCHR: + case S_IFBLK: + case S_IFSOCK: + if (unlikely(dip->di_format != XFS_DINODE_FMT_DEV)) { + XFS_CORRUPTION_ERROR("xfs_iformat(3)", XFS_ERRLEVEL_LOW, + ip->i_mount, dip); + return XFS_ERROR(EFSCORRUPTED); + } + ip->i_d.di_size = 0; + ip->i_df.if_u2.if_rdev = xfs_dinode_get_rdev(dip); + break; + + case S_IFREG: + case S_IFLNK: + case S_IFDIR: + switch (dip->di_format) { + case XFS_DINODE_FMT_LOCAL: + /* + * no local regular files yet + */ + if (unlikely(S_ISREG(be16_to_cpu(dip->di_mode)))) { + xfs_warn(ip->i_mount, + "corrupt inode %Lu (local format for regular file).", + (unsigned long long) ip->i_ino); + XFS_CORRUPTION_ERROR("xfs_iformat(4)", + XFS_ERRLEVEL_LOW, + ip->i_mount, dip); + return XFS_ERROR(EFSCORRUPTED); + } + + di_size = be64_to_cpu(dip->di_size); + if (unlikely(di_size < 0 || + di_size > XFS_DFORK_DSIZE(dip, ip->i_mount))) { + xfs_warn(ip->i_mount, + "corrupt inode %Lu (bad size %Ld for local inode).", + (unsigned long long) ip->i_ino, + (long long) di_size); + XFS_CORRUPTION_ERROR("xfs_iformat(5)", + XFS_ERRLEVEL_LOW, + ip->i_mount, dip); + return XFS_ERROR(EFSCORRUPTED); + } + + size = (int)di_size; + error = xfs_iformat_local(ip, dip, XFS_DATA_FORK, size); + break; + case XFS_DINODE_FMT_EXTENTS: + error = xfs_iformat_extents(ip, dip, XFS_DATA_FORK); + break; + case XFS_DINODE_FMT_BTREE: + error = xfs_iformat_btree(ip, dip, XFS_DATA_FORK); + break; + default: + XFS_ERROR_REPORT("xfs_iformat(6)", XFS_ERRLEVEL_LOW, + ip->i_mount); + return XFS_ERROR(EFSCORRUPTED); + } + break; + + default: + XFS_ERROR_REPORT("xfs_iformat(7)", XFS_ERRLEVEL_LOW, ip->i_mount); + return XFS_ERROR(EFSCORRUPTED); + } + if (error) { + return error; + } + if (!XFS_DFORK_Q(dip)) + return 0; + + ASSERT(ip->i_afp == NULL); + ip->i_afp = kmem_zone_zalloc(xfs_ifork_zone, KM_SLEEP | KM_NOFS); + + switch (dip->di_aformat) { + case XFS_DINODE_FMT_LOCAL: + atp = (xfs_attr_shortform_t *)XFS_DFORK_APTR(dip); + size = be16_to_cpu(atp->hdr.totsize); + + if (unlikely(size < sizeof(struct xfs_attr_sf_hdr))) { + xfs_warn(ip->i_mount, + "corrupt inode %Lu (bad attr fork size %Ld).", + (unsigned long long) ip->i_ino, + (long long) size); + XFS_CORRUPTION_ERROR("xfs_iformat(8)", + XFS_ERRLEVEL_LOW, + ip->i_mount, dip); + return XFS_ERROR(EFSCORRUPTED); + } + + error = xfs_iformat_local(ip, dip, XFS_ATTR_FORK, size); + break; + case XFS_DINODE_FMT_EXTENTS: + error = xfs_iformat_extents(ip, dip, XFS_ATTR_FORK); + break; + case XFS_DINODE_FMT_BTREE: + error = xfs_iformat_btree(ip, dip, XFS_ATTR_FORK); + break; + default: + error = XFS_ERROR(EFSCORRUPTED); + break; + } + if (error) { + kmem_zone_free(xfs_ifork_zone, ip->i_afp); + ip->i_afp = NULL; + xfs_idestroy_fork(ip, XFS_DATA_FORK); + } + return error; +} + +/* + * The file is in-lined in the on-disk inode. + * If it fits into if_inline_data, then copy + * it there, otherwise allocate a buffer for it + * and copy the data there. Either way, set + * if_data to point at the data. + * If we allocate a buffer for the data, make + * sure that its size is a multiple of 4 and + * record the real size in i_real_bytes. + */ +STATIC int +xfs_iformat_local( + xfs_inode_t *ip, + xfs_dinode_t *dip, + int whichfork, + int size) +{ + xfs_ifork_t *ifp; + int real_size; + + /* + * If the size is unreasonable, then something + * is wrong and we just bail out rather than crash in + * kmem_alloc() or memcpy() below. + */ + if (unlikely(size > XFS_DFORK_SIZE(dip, ip->i_mount, whichfork))) { + xfs_warn(ip->i_mount, + "corrupt inode %Lu (bad size %d for local fork, size = %d).", + (unsigned long long) ip->i_ino, size, + XFS_DFORK_SIZE(dip, ip->i_mount, whichfork)); + XFS_CORRUPTION_ERROR("xfs_iformat_local", XFS_ERRLEVEL_LOW, + ip->i_mount, dip); + return XFS_ERROR(EFSCORRUPTED); + } + ifp = XFS_IFORK_PTR(ip, whichfork); + real_size = 0; + if (size == 0) + ifp->if_u1.if_data = NULL; + else if (size <= sizeof(ifp->if_u2.if_inline_data)) + ifp->if_u1.if_data = ifp->if_u2.if_inline_data; + else { + real_size = roundup(size, 4); + ifp->if_u1.if_data = kmem_alloc(real_size, KM_SLEEP | KM_NOFS); + } + ifp->if_bytes = size; + ifp->if_real_bytes = real_size; + if (size) + memcpy(ifp->if_u1.if_data, XFS_DFORK_PTR(dip, whichfork), size); + ifp->if_flags &= ~XFS_IFEXTENTS; + ifp->if_flags |= XFS_IFINLINE; + return 0; +} + +/* + * The file consists of a set of extents all + * of which fit into the on-disk inode. + * If there are few enough extents to fit into + * the if_inline_ext, then copy them there. + * Otherwise allocate a buffer for them and copy + * them into it. Either way, set if_extents + * to point at the extents. + */ +STATIC int +xfs_iformat_extents( + xfs_inode_t *ip, + xfs_dinode_t *dip, + int whichfork) +{ + xfs_bmbt_rec_t *dp; + xfs_ifork_t *ifp; + int nex; + int size; + int i; + + ifp = XFS_IFORK_PTR(ip, whichfork); + nex = XFS_DFORK_NEXTENTS(dip, whichfork); + size = nex * (uint)sizeof(xfs_bmbt_rec_t); + + /* + * If the number of extents is unreasonable, then something + * is wrong and we just bail out rather than crash in + * kmem_alloc() or memcpy() below. + */ + if (unlikely(size < 0 || size > XFS_DFORK_SIZE(dip, ip->i_mount, whichfork))) { + xfs_warn(ip->i_mount, "corrupt inode %Lu ((a)extents = %d).", + (unsigned long long) ip->i_ino, nex); + XFS_CORRUPTION_ERROR("xfs_iformat_extents(1)", XFS_ERRLEVEL_LOW, + ip->i_mount, dip); + return XFS_ERROR(EFSCORRUPTED); + } + + ifp->if_real_bytes = 0; + if (nex == 0) + ifp->if_u1.if_extents = NULL; + else if (nex <= XFS_INLINE_EXTS) + ifp->if_u1.if_extents = ifp->if_u2.if_inline_ext; + else + xfs_iext_add(ifp, 0, nex); + + ifp->if_bytes = size; + if (size) { + dp = (xfs_bmbt_rec_t *) XFS_DFORK_PTR(dip, whichfork); + xfs_validate_extents(ifp, nex, XFS_EXTFMT_INODE(ip)); + for (i = 0; i < nex; i++, dp++) { + xfs_bmbt_rec_host_t *ep = xfs_iext_get_ext(ifp, i); + ep->l0 = get_unaligned_be64(&dp->l0); + ep->l1 = get_unaligned_be64(&dp->l1); + } + XFS_BMAP_TRACE_EXLIST(ip, nex, whichfork); + if (whichfork != XFS_DATA_FORK || + XFS_EXTFMT_INODE(ip) == XFS_EXTFMT_NOSTATE) + if (unlikely(xfs_check_nostate_extents( + ifp, 0, nex))) { + XFS_ERROR_REPORT("xfs_iformat_extents(2)", + XFS_ERRLEVEL_LOW, + ip->i_mount); + return XFS_ERROR(EFSCORRUPTED); + } + } + ifp->if_flags |= XFS_IFEXTENTS; + return 0; +} + +/* + * The file has too many extents to fit into + * the inode, so they are in B-tree format. + * Allocate a buffer for the root of the B-tree + * and copy the root into it. The i_extents + * field will remain NULL until all of the + * extents are read in (when they are needed). + */ +STATIC int +xfs_iformat_btree( + xfs_inode_t *ip, + xfs_dinode_t *dip, + int whichfork) +{ + struct xfs_mount *mp = ip->i_mount; + xfs_bmdr_block_t *dfp; + xfs_ifork_t *ifp; + /* REFERENCED */ + int nrecs; + int size; + + ifp = XFS_IFORK_PTR(ip, whichfork); + dfp = (xfs_bmdr_block_t *)XFS_DFORK_PTR(dip, whichfork); + size = XFS_BMAP_BROOT_SPACE(mp, dfp); + nrecs = be16_to_cpu(dfp->bb_numrecs); + + /* + * blow out if -- fork has less extents than can fit in + * fork (fork shouldn't be a btree format), root btree + * block has more records than can fit into the fork, + * or the number of extents is greater than the number of + * blocks. + */ + if (unlikely(XFS_IFORK_NEXTENTS(ip, whichfork) <= + XFS_IFORK_MAXEXT(ip, whichfork) || + XFS_BMDR_SPACE_CALC(nrecs) > + XFS_DFORK_SIZE(dip, mp, whichfork) || + XFS_IFORK_NEXTENTS(ip, whichfork) > ip->i_d.di_nblocks)) { + xfs_warn(mp, "corrupt inode %Lu (btree).", + (unsigned long long) ip->i_ino); + XFS_CORRUPTION_ERROR("xfs_iformat_btree", XFS_ERRLEVEL_LOW, + mp, dip); + return XFS_ERROR(EFSCORRUPTED); + } + + ifp->if_broot_bytes = size; + ifp->if_broot = kmem_alloc(size, KM_SLEEP | KM_NOFS); + ASSERT(ifp->if_broot != NULL); + /* + * Copy and convert from the on-disk structure + * to the in-memory structure. + */ + xfs_bmdr_to_bmbt(ip, dfp, XFS_DFORK_SIZE(dip, ip->i_mount, whichfork), + ifp->if_broot, size); + ifp->if_flags &= ~XFS_IFEXTENTS; + ifp->if_flags |= XFS_IFBROOT; + + return 0; +} + +/* + * Read in extents from a btree-format inode. + * Allocate and fill in if_extents. Real work is done in xfs_bmap.c. + */ +int +xfs_iread_extents( + xfs_trans_t *tp, + xfs_inode_t *ip, + int whichfork) +{ + int error; + xfs_ifork_t *ifp; + xfs_extnum_t nextents; + + if (unlikely(XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE)) { + XFS_ERROR_REPORT("xfs_iread_extents", XFS_ERRLEVEL_LOW, + ip->i_mount); + return XFS_ERROR(EFSCORRUPTED); + } + nextents = XFS_IFORK_NEXTENTS(ip, whichfork); + ifp = XFS_IFORK_PTR(ip, whichfork); + + /* + * We know that the size is valid (it's checked in iformat_btree) + */ + ifp->if_bytes = ifp->if_real_bytes = 0; + ifp->if_flags |= XFS_IFEXTENTS; + xfs_iext_add(ifp, 0, nextents); + error = xfs_bmap_read_extents(tp, ip, whichfork); + if (error) { + xfs_iext_destroy(ifp); + ifp->if_flags &= ~XFS_IFEXTENTS; + return error; + } + xfs_validate_extents(ifp, nextents, XFS_EXTFMT_INODE(ip)); + return 0; +} +/* + * Reallocate the space for if_broot based on the number of records + * being added or deleted as indicated in rec_diff. Move the records + * and pointers in if_broot to fit the new size. When shrinking this + * will eliminate holes between the records and pointers created by + * the caller. When growing this will create holes to be filled in + * by the caller. + * + * The caller must not request to add more records than would fit in + * the on-disk inode root. If the if_broot is currently NULL, then + * if we are adding records, one will be allocated. The caller must also + * not request that the number of records go below zero, although + * it can go to zero. + * + * ip -- the inode whose if_broot area is changing + * ext_diff -- the change in the number of records, positive or negative, + * requested for the if_broot array. + */ +void +xfs_iroot_realloc( + xfs_inode_t *ip, + int rec_diff, + int whichfork) +{ + struct xfs_mount *mp = ip->i_mount; + int cur_max; + xfs_ifork_t *ifp; + struct xfs_btree_block *new_broot; + int new_max; + size_t new_size; + char *np; + char *op; + + /* + * Handle the degenerate case quietly. + */ + if (rec_diff == 0) { + return; + } + + ifp = XFS_IFORK_PTR(ip, whichfork); + if (rec_diff > 0) { + /* + * If there wasn't any memory allocated before, just + * allocate it now and get out. + */ + if (ifp->if_broot_bytes == 0) { + new_size = XFS_BMAP_BROOT_SPACE_CALC(mp, rec_diff); + ifp->if_broot = kmem_alloc(new_size, KM_SLEEP | KM_NOFS); + ifp->if_broot_bytes = (int)new_size; + return; + } + + /* + * If there is already an existing if_broot, then we need + * to realloc() it and shift the pointers to their new + * location. The records don't change location because + * they are kept butted up against the btree block header. + */ + cur_max = xfs_bmbt_maxrecs(mp, ifp->if_broot_bytes, 0); + new_max = cur_max + rec_diff; + new_size = XFS_BMAP_BROOT_SPACE_CALC(mp, new_max); + ifp->if_broot = kmem_realloc(ifp->if_broot, new_size, + XFS_BMAP_BROOT_SPACE_CALC(mp, cur_max), + KM_SLEEP | KM_NOFS); + op = (char *)XFS_BMAP_BROOT_PTR_ADDR(mp, ifp->if_broot, 1, + ifp->if_broot_bytes); + np = (char *)XFS_BMAP_BROOT_PTR_ADDR(mp, ifp->if_broot, 1, + (int)new_size); + ifp->if_broot_bytes = (int)new_size; + ASSERT(XFS_BMAP_BMDR_SPACE(ifp->if_broot) <= + XFS_IFORK_SIZE(ip, whichfork)); + memmove(np, op, cur_max * (uint)sizeof(xfs_dfsbno_t)); + return; + } + + /* + * rec_diff is less than 0. In this case, we are shrinking the + * if_broot buffer. It must already exist. If we go to zero + * records, just get rid of the root and clear the status bit. + */ + ASSERT((ifp->if_broot != NULL) && (ifp->if_broot_bytes > 0)); + cur_max = xfs_bmbt_maxrecs(mp, ifp->if_broot_bytes, 0); + new_max = cur_max + rec_diff; + ASSERT(new_max >= 0); + if (new_max > 0) + new_size = XFS_BMAP_BROOT_SPACE_CALC(mp, new_max); + else + new_size = 0; + if (new_size > 0) { + new_broot = kmem_alloc(new_size, KM_SLEEP | KM_NOFS); + /* + * First copy over the btree block header. + */ + memcpy(new_broot, ifp->if_broot, + XFS_BMBT_BLOCK_LEN(ip->i_mount)); + } else { + new_broot = NULL; + ifp->if_flags &= ~XFS_IFBROOT; + } + + /* + * Only copy the records and pointers if there are any. + */ + if (new_max > 0) { + /* + * First copy the records. + */ + op = (char *)XFS_BMBT_REC_ADDR(mp, ifp->if_broot, 1); + np = (char *)XFS_BMBT_REC_ADDR(mp, new_broot, 1); + memcpy(np, op, new_max * (uint)sizeof(xfs_bmbt_rec_t)); + + /* + * Then copy the pointers. + */ + op = (char *)XFS_BMAP_BROOT_PTR_ADDR(mp, ifp->if_broot, 1, + ifp->if_broot_bytes); + np = (char *)XFS_BMAP_BROOT_PTR_ADDR(mp, new_broot, 1, + (int)new_size); + memcpy(np, op, new_max * (uint)sizeof(xfs_dfsbno_t)); + } + kmem_free(ifp->if_broot); + ifp->if_broot = new_broot; + ifp->if_broot_bytes = (int)new_size; + if (ifp->if_broot) + ASSERT(XFS_BMAP_BMDR_SPACE(ifp->if_broot) <= + XFS_IFORK_SIZE(ip, whichfork)); + return; +} + + +/* + * This is called when the amount of space needed for if_data + * is increased or decreased. The change in size is indicated by + * the number of bytes that need to be added or deleted in the + * byte_diff parameter. + * + * If the amount of space needed has decreased below the size of the + * inline buffer, then switch to using the inline buffer. Otherwise, + * use kmem_realloc() or kmem_alloc() to adjust the size of the buffer + * to what is needed. + * + * ip -- the inode whose if_data area is changing + * byte_diff -- the change in the number of bytes, positive or negative, + * requested for the if_data array. + */ +void +xfs_idata_realloc( + xfs_inode_t *ip, + int byte_diff, + int whichfork) +{ + xfs_ifork_t *ifp; + int new_size; + int real_size; + + if (byte_diff == 0) { + return; + } + + ifp = XFS_IFORK_PTR(ip, whichfork); + new_size = (int)ifp->if_bytes + byte_diff; + ASSERT(new_size >= 0); + + if (new_size == 0) { + if (ifp->if_u1.if_data != ifp->if_u2.if_inline_data) { + kmem_free(ifp->if_u1.if_data); + } + ifp->if_u1.if_data = NULL; + real_size = 0; + } else if (new_size <= sizeof(ifp->if_u2.if_inline_data)) { + /* + * If the valid extents/data can fit in if_inline_ext/data, + * copy them from the malloc'd vector and free it. + */ + if (ifp->if_u1.if_data == NULL) { + ifp->if_u1.if_data = ifp->if_u2.if_inline_data; + } else if (ifp->if_u1.if_data != ifp->if_u2.if_inline_data) { + ASSERT(ifp->if_real_bytes != 0); + memcpy(ifp->if_u2.if_inline_data, ifp->if_u1.if_data, + new_size); + kmem_free(ifp->if_u1.if_data); + ifp->if_u1.if_data = ifp->if_u2.if_inline_data; + } + real_size = 0; + } else { + /* + * Stuck with malloc/realloc. + * For inline data, the underlying buffer must be + * a multiple of 4 bytes in size so that it can be + * logged and stay on word boundaries. We enforce + * that here. + */ + real_size = roundup(new_size, 4); + if (ifp->if_u1.if_data == NULL) { + ASSERT(ifp->if_real_bytes == 0); + ifp->if_u1.if_data = kmem_alloc(real_size, + KM_SLEEP | KM_NOFS); + } else if (ifp->if_u1.if_data != ifp->if_u2.if_inline_data) { + /* + * Only do the realloc if the underlying size + * is really changing. + */ + if (ifp->if_real_bytes != real_size) { + ifp->if_u1.if_data = + kmem_realloc(ifp->if_u1.if_data, + real_size, + ifp->if_real_bytes, + KM_SLEEP | KM_NOFS); + } + } else { + ASSERT(ifp->if_real_bytes == 0); + ifp->if_u1.if_data = kmem_alloc(real_size, + KM_SLEEP | KM_NOFS); + memcpy(ifp->if_u1.if_data, ifp->if_u2.if_inline_data, + ifp->if_bytes); + } + } + ifp->if_real_bytes = real_size; + ifp->if_bytes = new_size; + ASSERT(ifp->if_bytes <= XFS_IFORK_SIZE(ip, whichfork)); +} + +void +xfs_idestroy_fork( + xfs_inode_t *ip, + int whichfork) +{ + xfs_ifork_t *ifp; + + ifp = XFS_IFORK_PTR(ip, whichfork); + if (ifp->if_broot != NULL) { + kmem_free(ifp->if_broot); + ifp->if_broot = NULL; + } + + /* + * If the format is local, then we can't have an extents + * array so just look for an inline data array. If we're + * not local then we may or may not have an extents list, + * so check and free it up if we do. + */ + if (XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_LOCAL) { + if ((ifp->if_u1.if_data != ifp->if_u2.if_inline_data) && + (ifp->if_u1.if_data != NULL)) { + ASSERT(ifp->if_real_bytes != 0); + kmem_free(ifp->if_u1.if_data); + ifp->if_u1.if_data = NULL; + ifp->if_real_bytes = 0; + } + } else if ((ifp->if_flags & XFS_IFEXTENTS) && + ((ifp->if_flags & XFS_IFEXTIREC) || + ((ifp->if_u1.if_extents != NULL) && + (ifp->if_u1.if_extents != ifp->if_u2.if_inline_ext)))) { + ASSERT(ifp->if_real_bytes != 0); + xfs_iext_destroy(ifp); + } + ASSERT(ifp->if_u1.if_extents == NULL || + ifp->if_u1.if_extents == ifp->if_u2.if_inline_ext); + ASSERT(ifp->if_real_bytes == 0); + if (whichfork == XFS_ATTR_FORK) { + kmem_zone_free(xfs_ifork_zone, ip->i_afp); + ip->i_afp = NULL; + } +} + +/* + * xfs_iextents_copy() + * + * This is called to copy the REAL extents (as opposed to the delayed + * allocation extents) from the inode into the given buffer. It + * returns the number of bytes copied into the buffer. + * + * If there are no delayed allocation extents, then we can just + * memcpy() the extents into the buffer. Otherwise, we need to + * examine each extent in turn and skip those which are delayed. + */ +int +xfs_iextents_copy( + xfs_inode_t *ip, + xfs_bmbt_rec_t *dp, + int whichfork) +{ + int copied; + int i; + xfs_ifork_t *ifp; + int nrecs; + xfs_fsblock_t start_block; + + ifp = XFS_IFORK_PTR(ip, whichfork); + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_ILOCK_SHARED)); + ASSERT(ifp->if_bytes > 0); + + nrecs = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t); + XFS_BMAP_TRACE_EXLIST(ip, nrecs, whichfork); + ASSERT(nrecs > 0); + + /* + * There are some delayed allocation extents in the + * inode, so copy the extents one at a time and skip + * the delayed ones. There must be at least one + * non-delayed extent. + */ + copied = 0; + for (i = 0; i < nrecs; i++) { + xfs_bmbt_rec_host_t *ep = xfs_iext_get_ext(ifp, i); + start_block = xfs_bmbt_get_startblock(ep); + if (isnullstartblock(start_block)) { + /* + * It's a delayed allocation extent, so skip it. + */ + continue; + } + + /* Translate to on disk format */ + put_unaligned_be64(ep->l0, &dp->l0); + put_unaligned_be64(ep->l1, &dp->l1); + dp++; + copied++; + } + ASSERT(copied != 0); + xfs_validate_extents(ifp, copied, XFS_EXTFMT_INODE(ip)); + + return (copied * (uint)sizeof(xfs_bmbt_rec_t)); +} + +/* + * Each of the following cases stores data into the same region + * of the on-disk inode, so only one of them can be valid at + * any given time. While it is possible to have conflicting formats + * and log flags, e.g. having XFS_ILOG_?DATA set when the fork is + * in EXTENTS format, this can only happen when the fork has + * changed formats after being modified but before being flushed. + * In these cases, the format always takes precedence, because the + * format indicates the current state of the fork. + */ +void +xfs_iflush_fork( + xfs_inode_t *ip, + xfs_dinode_t *dip, + xfs_inode_log_item_t *iip, + int whichfork, + xfs_buf_t *bp) +{ + char *cp; + xfs_ifork_t *ifp; + xfs_mount_t *mp; + static const short brootflag[2] = + { XFS_ILOG_DBROOT, XFS_ILOG_ABROOT }; + static const short dataflag[2] = + { XFS_ILOG_DDATA, XFS_ILOG_ADATA }; + static const short extflag[2] = + { XFS_ILOG_DEXT, XFS_ILOG_AEXT }; + + if (!iip) + return; + ifp = XFS_IFORK_PTR(ip, whichfork); + /* + * This can happen if we gave up in iformat in an error path, + * for the attribute fork. + */ + if (!ifp) { + ASSERT(whichfork == XFS_ATTR_FORK); + return; + } + cp = XFS_DFORK_PTR(dip, whichfork); + mp = ip->i_mount; + switch (XFS_IFORK_FORMAT(ip, whichfork)) { + case XFS_DINODE_FMT_LOCAL: + if ((iip->ili_fields & dataflag[whichfork]) && + (ifp->if_bytes > 0)) { + ASSERT(ifp->if_u1.if_data != NULL); + ASSERT(ifp->if_bytes <= XFS_IFORK_SIZE(ip, whichfork)); + memcpy(cp, ifp->if_u1.if_data, ifp->if_bytes); + } + break; + + case XFS_DINODE_FMT_EXTENTS: + ASSERT((ifp->if_flags & XFS_IFEXTENTS) || + !(iip->ili_fields & extflag[whichfork])); + if ((iip->ili_fields & extflag[whichfork]) && + (ifp->if_bytes > 0)) { + ASSERT(xfs_iext_get_ext(ifp, 0)); + ASSERT(XFS_IFORK_NEXTENTS(ip, whichfork) > 0); + (void)xfs_iextents_copy(ip, (xfs_bmbt_rec_t *)cp, + whichfork); + } + break; + + case XFS_DINODE_FMT_BTREE: + if ((iip->ili_fields & brootflag[whichfork]) && + (ifp->if_broot_bytes > 0)) { + ASSERT(ifp->if_broot != NULL); + ASSERT(XFS_BMAP_BMDR_SPACE(ifp->if_broot) <= + XFS_IFORK_SIZE(ip, whichfork)); + xfs_bmbt_to_bmdr(mp, ifp->if_broot, ifp->if_broot_bytes, + (xfs_bmdr_block_t *)cp, + XFS_DFORK_SIZE(dip, mp, whichfork)); + } + break; + + case XFS_DINODE_FMT_DEV: + if (iip->ili_fields & XFS_ILOG_DEV) { + ASSERT(whichfork == XFS_DATA_FORK); + xfs_dinode_put_rdev(dip, ip->i_df.if_u2.if_rdev); + } + break; + + case XFS_DINODE_FMT_UUID: + if (iip->ili_fields & XFS_ILOG_UUID) { + ASSERT(whichfork == XFS_DATA_FORK); + memcpy(XFS_DFORK_DPTR(dip), + &ip->i_df.if_u2.if_uuid, + sizeof(uuid_t)); + } + break; + + default: + ASSERT(0); + break; + } +} + +/* + * Return a pointer to the extent record at file index idx. + */ +xfs_bmbt_rec_host_t * +xfs_iext_get_ext( + xfs_ifork_t *ifp, /* inode fork pointer */ + xfs_extnum_t idx) /* index of target extent */ +{ + ASSERT(idx >= 0); + ASSERT(idx < ifp->if_bytes / sizeof(xfs_bmbt_rec_t)); + + if ((ifp->if_flags & XFS_IFEXTIREC) && (idx == 0)) { + return ifp->if_u1.if_ext_irec->er_extbuf; + } else if (ifp->if_flags & XFS_IFEXTIREC) { + xfs_ext_irec_t *erp; /* irec pointer */ + int erp_idx = 0; /* irec index */ + xfs_extnum_t page_idx = idx; /* ext index in target list */ + + erp = xfs_iext_idx_to_irec(ifp, &page_idx, &erp_idx, 0); + return &erp->er_extbuf[page_idx]; + } else if (ifp->if_bytes) { + return &ifp->if_u1.if_extents[idx]; + } else { + return NULL; + } +} + +/* + * Insert new item(s) into the extent records for incore inode + * fork 'ifp'. 'count' new items are inserted at index 'idx'. + */ +void +xfs_iext_insert( + xfs_inode_t *ip, /* incore inode pointer */ + xfs_extnum_t idx, /* starting index of new items */ + xfs_extnum_t count, /* number of inserted items */ + xfs_bmbt_irec_t *new, /* items to insert */ + int state) /* type of extent conversion */ +{ + xfs_ifork_t *ifp = (state & BMAP_ATTRFORK) ? ip->i_afp : &ip->i_df; + xfs_extnum_t i; /* extent record index */ + + trace_xfs_iext_insert(ip, idx, new, state, _RET_IP_); + + ASSERT(ifp->if_flags & XFS_IFEXTENTS); + xfs_iext_add(ifp, idx, count); + for (i = idx; i < idx + count; i++, new++) + xfs_bmbt_set_all(xfs_iext_get_ext(ifp, i), new); +} + +/* + * This is called when the amount of space required for incore file + * extents needs to be increased. The ext_diff parameter stores the + * number of new extents being added and the idx parameter contains + * the extent index where the new extents will be added. If the new + * extents are being appended, then we just need to (re)allocate and + * initialize the space. Otherwise, if the new extents are being + * inserted into the middle of the existing entries, a bit more work + * is required to make room for the new extents to be inserted. The + * caller is responsible for filling in the new extent entries upon + * return. + */ +void +xfs_iext_add( + xfs_ifork_t *ifp, /* inode fork pointer */ + xfs_extnum_t idx, /* index to begin adding exts */ + int ext_diff) /* number of extents to add */ +{ + int byte_diff; /* new bytes being added */ + int new_size; /* size of extents after adding */ + xfs_extnum_t nextents; /* number of extents in file */ + + nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t); + ASSERT((idx >= 0) && (idx <= nextents)); + byte_diff = ext_diff * sizeof(xfs_bmbt_rec_t); + new_size = ifp->if_bytes + byte_diff; + /* + * If the new number of extents (nextents + ext_diff) + * fits inside the inode, then continue to use the inline + * extent buffer. + */ + if (nextents + ext_diff <= XFS_INLINE_EXTS) { + if (idx < nextents) { + memmove(&ifp->if_u2.if_inline_ext[idx + ext_diff], + &ifp->if_u2.if_inline_ext[idx], + (nextents - idx) * sizeof(xfs_bmbt_rec_t)); + memset(&ifp->if_u2.if_inline_ext[idx], 0, byte_diff); + } + ifp->if_u1.if_extents = ifp->if_u2.if_inline_ext; + ifp->if_real_bytes = 0; + } + /* + * Otherwise use a linear (direct) extent list. + * If the extents are currently inside the inode, + * xfs_iext_realloc_direct will switch us from + * inline to direct extent allocation mode. + */ + else if (nextents + ext_diff <= XFS_LINEAR_EXTS) { + xfs_iext_realloc_direct(ifp, new_size); + if (idx < nextents) { + memmove(&ifp->if_u1.if_extents[idx + ext_diff], + &ifp->if_u1.if_extents[idx], + (nextents - idx) * sizeof(xfs_bmbt_rec_t)); + memset(&ifp->if_u1.if_extents[idx], 0, byte_diff); + } + } + /* Indirection array */ + else { + xfs_ext_irec_t *erp; + int erp_idx = 0; + int page_idx = idx; + + ASSERT(nextents + ext_diff > XFS_LINEAR_EXTS); + if (ifp->if_flags & XFS_IFEXTIREC) { + erp = xfs_iext_idx_to_irec(ifp, &page_idx, &erp_idx, 1); + } else { + xfs_iext_irec_init(ifp); + ASSERT(ifp->if_flags & XFS_IFEXTIREC); + erp = ifp->if_u1.if_ext_irec; + } + /* Extents fit in target extent page */ + if (erp && erp->er_extcount + ext_diff <= XFS_LINEAR_EXTS) { + if (page_idx < erp->er_extcount) { + memmove(&erp->er_extbuf[page_idx + ext_diff], + &erp->er_extbuf[page_idx], + (erp->er_extcount - page_idx) * + sizeof(xfs_bmbt_rec_t)); + memset(&erp->er_extbuf[page_idx], 0, byte_diff); + } + erp->er_extcount += ext_diff; + xfs_iext_irec_update_extoffs(ifp, erp_idx + 1, ext_diff); + } + /* Insert a new extent page */ + else if (erp) { + xfs_iext_add_indirect_multi(ifp, + erp_idx, page_idx, ext_diff); + } + /* + * If extent(s) are being appended to the last page in + * the indirection array and the new extent(s) don't fit + * in the page, then erp is NULL and erp_idx is set to + * the next index needed in the indirection array. + */ + else { + int count = ext_diff; + + while (count) { + erp = xfs_iext_irec_new(ifp, erp_idx); + erp->er_extcount = count; + count -= MIN(count, (int)XFS_LINEAR_EXTS); + if (count) { + erp_idx++; + } + } + } + } + ifp->if_bytes = new_size; +} + +/* + * This is called when incore extents are being added to the indirection + * array and the new extents do not fit in the target extent list. The + * erp_idx parameter contains the irec index for the target extent list + * in the indirection array, and the idx parameter contains the extent + * index within the list. The number of extents being added is stored + * in the count parameter. + * + * |-------| |-------| + * | | | | idx - number of extents before idx + * | idx | | count | + * | | | | count - number of extents being inserted at idx + * |-------| |-------| + * | count | | nex2 | nex2 - number of extents after idx + count + * |-------| |-------| + */ +void +xfs_iext_add_indirect_multi( + xfs_ifork_t *ifp, /* inode fork pointer */ + int erp_idx, /* target extent irec index */ + xfs_extnum_t idx, /* index within target list */ + int count) /* new extents being added */ +{ + int byte_diff; /* new bytes being added */ + xfs_ext_irec_t *erp; /* pointer to irec entry */ + xfs_extnum_t ext_diff; /* number of extents to add */ + xfs_extnum_t ext_cnt; /* new extents still needed */ + xfs_extnum_t nex2; /* extents after idx + count */ + xfs_bmbt_rec_t *nex2_ep = NULL; /* temp list for nex2 extents */ + int nlists; /* number of irec's (lists) */ + + ASSERT(ifp->if_flags & XFS_IFEXTIREC); + erp = &ifp->if_u1.if_ext_irec[erp_idx]; + nex2 = erp->er_extcount - idx; + nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ; + + /* + * Save second part of target extent list + * (all extents past */ + if (nex2) { + byte_diff = nex2 * sizeof(xfs_bmbt_rec_t); + nex2_ep = (xfs_bmbt_rec_t *) kmem_alloc(byte_diff, KM_NOFS); + memmove(nex2_ep, &erp->er_extbuf[idx], byte_diff); + erp->er_extcount -= nex2; + xfs_iext_irec_update_extoffs(ifp, erp_idx + 1, -nex2); + memset(&erp->er_extbuf[idx], 0, byte_diff); + } + + /* + * Add the new extents to the end of the target + * list, then allocate new irec record(s) and + * extent buffer(s) as needed to store the rest + * of the new extents. + */ + ext_cnt = count; + ext_diff = MIN(ext_cnt, (int)XFS_LINEAR_EXTS - erp->er_extcount); + if (ext_diff) { + erp->er_extcount += ext_diff; + xfs_iext_irec_update_extoffs(ifp, erp_idx + 1, ext_diff); + ext_cnt -= ext_diff; + } + while (ext_cnt) { + erp_idx++; + erp = xfs_iext_irec_new(ifp, erp_idx); + ext_diff = MIN(ext_cnt, (int)XFS_LINEAR_EXTS); + erp->er_extcount = ext_diff; + xfs_iext_irec_update_extoffs(ifp, erp_idx + 1, ext_diff); + ext_cnt -= ext_diff; + } + + /* Add nex2 extents back to indirection array */ + if (nex2) { + xfs_extnum_t ext_avail; + int i; + + byte_diff = nex2 * sizeof(xfs_bmbt_rec_t); + ext_avail = XFS_LINEAR_EXTS - erp->er_extcount; + i = 0; + /* + * If nex2 extents fit in the current page, append + * nex2_ep after the new extents. + */ + if (nex2 <= ext_avail) { + i = erp->er_extcount; + } + /* + * Otherwise, check if space is available in the + * next page. + */ + else if ((erp_idx < nlists - 1) && + (nex2 <= (ext_avail = XFS_LINEAR_EXTS - + ifp->if_u1.if_ext_irec[erp_idx+1].er_extcount))) { + erp_idx++; + erp++; + /* Create a hole for nex2 extents */ + memmove(&erp->er_extbuf[nex2], erp->er_extbuf, + erp->er_extcount * sizeof(xfs_bmbt_rec_t)); + } + /* + * Final choice, create a new extent page for + * nex2 extents. + */ + else { + erp_idx++; + erp = xfs_iext_irec_new(ifp, erp_idx); + } + memmove(&erp->er_extbuf[i], nex2_ep, byte_diff); + kmem_free(nex2_ep); + erp->er_extcount += nex2; + xfs_iext_irec_update_extoffs(ifp, erp_idx + 1, nex2); + } +} + +/* + * This is called when the amount of space required for incore file + * extents needs to be decreased. The ext_diff parameter stores the + * number of extents to be removed and the idx parameter contains + * the extent index where the extents will be removed from. + * + * If the amount of space needed has decreased below the linear + * limit, XFS_IEXT_BUFSZ, then switch to using the contiguous + * extent array. Otherwise, use kmem_realloc() to adjust the + * size to what is needed. + */ +void +xfs_iext_remove( + xfs_inode_t *ip, /* incore inode pointer */ + xfs_extnum_t idx, /* index to begin removing exts */ + int ext_diff, /* number of extents to remove */ + int state) /* type of extent conversion */ +{ + xfs_ifork_t *ifp = (state & BMAP_ATTRFORK) ? ip->i_afp : &ip->i_df; + xfs_extnum_t nextents; /* number of extents in file */ + int new_size; /* size of extents after removal */ + + trace_xfs_iext_remove(ip, idx, state, _RET_IP_); + + ASSERT(ext_diff > 0); + nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t); + new_size = (nextents - ext_diff) * sizeof(xfs_bmbt_rec_t); + + if (new_size == 0) { + xfs_iext_destroy(ifp); + } else if (ifp->if_flags & XFS_IFEXTIREC) { + xfs_iext_remove_indirect(ifp, idx, ext_diff); + } else if (ifp->if_real_bytes) { + xfs_iext_remove_direct(ifp, idx, ext_diff); + } else { + xfs_iext_remove_inline(ifp, idx, ext_diff); + } + ifp->if_bytes = new_size; +} + +/* + * This removes ext_diff extents from the inline buffer, beginning + * at extent index idx. + */ +void +xfs_iext_remove_inline( + xfs_ifork_t *ifp, /* inode fork pointer */ + xfs_extnum_t idx, /* index to begin removing exts */ + int ext_diff) /* number of extents to remove */ +{ + int nextents; /* number of extents in file */ + + ASSERT(!(ifp->if_flags & XFS_IFEXTIREC)); + ASSERT(idx < XFS_INLINE_EXTS); + nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t); + ASSERT(((nextents - ext_diff) > 0) && + (nextents - ext_diff) < XFS_INLINE_EXTS); + + if (idx + ext_diff < nextents) { + memmove(&ifp->if_u2.if_inline_ext[idx], + &ifp->if_u2.if_inline_ext[idx + ext_diff], + (nextents - (idx + ext_diff)) * + sizeof(xfs_bmbt_rec_t)); + memset(&ifp->if_u2.if_inline_ext[nextents - ext_diff], + 0, ext_diff * sizeof(xfs_bmbt_rec_t)); + } else { + memset(&ifp->if_u2.if_inline_ext[idx], 0, + ext_diff * sizeof(xfs_bmbt_rec_t)); + } +} + +/* + * This removes ext_diff extents from a linear (direct) extent list, + * beginning at extent index idx. If the extents are being removed + * from the end of the list (ie. truncate) then we just need to re- + * allocate the list to remove the extra space. Otherwise, if the + * extents are being removed from the middle of the existing extent + * entries, then we first need to move the extent records beginning + * at idx + ext_diff up in the list to overwrite the records being + * removed, then remove the extra space via kmem_realloc. + */ +void +xfs_iext_remove_direct( + xfs_ifork_t *ifp, /* inode fork pointer */ + xfs_extnum_t idx, /* index to begin removing exts */ + int ext_diff) /* number of extents to remove */ +{ + xfs_extnum_t nextents; /* number of extents in file */ + int new_size; /* size of extents after removal */ + + ASSERT(!(ifp->if_flags & XFS_IFEXTIREC)); + new_size = ifp->if_bytes - + (ext_diff * sizeof(xfs_bmbt_rec_t)); + nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t); + + if (new_size == 0) { + xfs_iext_destroy(ifp); + return; + } + /* Move extents up in the list (if needed) */ + if (idx + ext_diff < nextents) { + memmove(&ifp->if_u1.if_extents[idx], + &ifp->if_u1.if_extents[idx + ext_diff], + (nextents - (idx + ext_diff)) * + sizeof(xfs_bmbt_rec_t)); + } + memset(&ifp->if_u1.if_extents[nextents - ext_diff], + 0, ext_diff * sizeof(xfs_bmbt_rec_t)); + /* + * Reallocate the direct extent list. If the extents + * will fit inside the inode then xfs_iext_realloc_direct + * will switch from direct to inline extent allocation + * mode for us. + */ + xfs_iext_realloc_direct(ifp, new_size); + ifp->if_bytes = new_size; +} + +/* + * This is called when incore extents are being removed from the + * indirection array and the extents being removed span multiple extent + * buffers. The idx parameter contains the file extent index where we + * want to begin removing extents, and the count parameter contains + * how many extents need to be removed. + * + * |-------| |-------| + * | nex1 | | | nex1 - number of extents before idx + * |-------| | count | + * | | | | count - number of extents being removed at idx + * | count | |-------| + * | | | nex2 | nex2 - number of extents after idx + count + * |-------| |-------| + */ +void +xfs_iext_remove_indirect( + xfs_ifork_t *ifp, /* inode fork pointer */ + xfs_extnum_t idx, /* index to begin removing extents */ + int count) /* number of extents to remove */ +{ + xfs_ext_irec_t *erp; /* indirection array pointer */ + int erp_idx = 0; /* indirection array index */ + xfs_extnum_t ext_cnt; /* extents left to remove */ + xfs_extnum_t ext_diff; /* extents to remove in current list */ + xfs_extnum_t nex1; /* number of extents before idx */ + xfs_extnum_t nex2; /* extents after idx + count */ + int page_idx = idx; /* index in target extent list */ + + ASSERT(ifp->if_flags & XFS_IFEXTIREC); + erp = xfs_iext_idx_to_irec(ifp, &page_idx, &erp_idx, 0); + ASSERT(erp != NULL); + nex1 = page_idx; + ext_cnt = count; + while (ext_cnt) { + nex2 = MAX((erp->er_extcount - (nex1 + ext_cnt)), 0); + ext_diff = MIN(ext_cnt, (erp->er_extcount - nex1)); + /* + * Check for deletion of entire list; + * xfs_iext_irec_remove() updates extent offsets. + */ + if (ext_diff == erp->er_extcount) { + xfs_iext_irec_remove(ifp, erp_idx); + ext_cnt -= ext_diff; + nex1 = 0; + if (ext_cnt) { + ASSERT(erp_idx < ifp->if_real_bytes / + XFS_IEXT_BUFSZ); + erp = &ifp->if_u1.if_ext_irec[erp_idx]; + nex1 = 0; + continue; + } else { + break; + } + } + /* Move extents up (if needed) */ + if (nex2) { + memmove(&erp->er_extbuf[nex1], + &erp->er_extbuf[nex1 + ext_diff], + nex2 * sizeof(xfs_bmbt_rec_t)); + } + /* Zero out rest of page */ + memset(&erp->er_extbuf[nex1 + nex2], 0, (XFS_IEXT_BUFSZ - + ((nex1 + nex2) * sizeof(xfs_bmbt_rec_t)))); + /* Update remaining counters */ + erp->er_extcount -= ext_diff; + xfs_iext_irec_update_extoffs(ifp, erp_idx + 1, -ext_diff); + ext_cnt -= ext_diff; + nex1 = 0; + erp_idx++; + erp++; + } + ifp->if_bytes -= count * sizeof(xfs_bmbt_rec_t); + xfs_iext_irec_compact(ifp); +} + +/* + * Create, destroy, or resize a linear (direct) block of extents. + */ +void +xfs_iext_realloc_direct( + xfs_ifork_t *ifp, /* inode fork pointer */ + int new_size) /* new size of extents */ +{ + int rnew_size; /* real new size of extents */ + + rnew_size = new_size; + + ASSERT(!(ifp->if_flags & XFS_IFEXTIREC) || + ((new_size >= 0) && (new_size <= XFS_IEXT_BUFSZ) && + (new_size != ifp->if_real_bytes))); + + /* Free extent records */ + if (new_size == 0) { + xfs_iext_destroy(ifp); + } + /* Resize direct extent list and zero any new bytes */ + else if (ifp->if_real_bytes) { + /* Check if extents will fit inside the inode */ + if (new_size <= XFS_INLINE_EXTS * sizeof(xfs_bmbt_rec_t)) { + xfs_iext_direct_to_inline(ifp, new_size / + (uint)sizeof(xfs_bmbt_rec_t)); + ifp->if_bytes = new_size; + return; + } + if (!is_power_of_2(new_size)){ + rnew_size = roundup_pow_of_two(new_size); + } + if (rnew_size != ifp->if_real_bytes) { + ifp->if_u1.if_extents = + kmem_realloc(ifp->if_u1.if_extents, + rnew_size, + ifp->if_real_bytes, KM_NOFS); + } + if (rnew_size > ifp->if_real_bytes) { + memset(&ifp->if_u1.if_extents[ifp->if_bytes / + (uint)sizeof(xfs_bmbt_rec_t)], 0, + rnew_size - ifp->if_real_bytes); + } + } + /* + * Switch from the inline extent buffer to a direct + * extent list. Be sure to include the inline extent + * bytes in new_size. + */ + else { + new_size += ifp->if_bytes; + if (!is_power_of_2(new_size)) { + rnew_size = roundup_pow_of_two(new_size); + } + xfs_iext_inline_to_direct(ifp, rnew_size); + } + ifp->if_real_bytes = rnew_size; + ifp->if_bytes = new_size; +} + +/* + * Switch from linear (direct) extent records to inline buffer. + */ +void +xfs_iext_direct_to_inline( + xfs_ifork_t *ifp, /* inode fork pointer */ + xfs_extnum_t nextents) /* number of extents in file */ +{ + ASSERT(ifp->if_flags & XFS_IFEXTENTS); + ASSERT(nextents <= XFS_INLINE_EXTS); + /* + * The inline buffer was zeroed when we switched + * from inline to direct extent allocation mode, + * so we don't need to clear it here. + */ + memcpy(ifp->if_u2.if_inline_ext, ifp->if_u1.if_extents, + nextents * sizeof(xfs_bmbt_rec_t)); + kmem_free(ifp->if_u1.if_extents); + ifp->if_u1.if_extents = ifp->if_u2.if_inline_ext; + ifp->if_real_bytes = 0; +} + +/* + * Switch from inline buffer to linear (direct) extent records. + * new_size should already be rounded up to the next power of 2 + * by the caller (when appropriate), so use new_size as it is. + * However, since new_size may be rounded up, we can't update + * if_bytes here. It is the caller's responsibility to update + * if_bytes upon return. + */ +void +xfs_iext_inline_to_direct( + xfs_ifork_t *ifp, /* inode fork pointer */ + int new_size) /* number of extents in file */ +{ + ifp->if_u1.if_extents = kmem_alloc(new_size, KM_NOFS); + memset(ifp->if_u1.if_extents, 0, new_size); + if (ifp->if_bytes) { + memcpy(ifp->if_u1.if_extents, ifp->if_u2.if_inline_ext, + ifp->if_bytes); + memset(ifp->if_u2.if_inline_ext, 0, XFS_INLINE_EXTS * + sizeof(xfs_bmbt_rec_t)); + } + ifp->if_real_bytes = new_size; +} + +/* + * Resize an extent indirection array to new_size bytes. + */ +STATIC void +xfs_iext_realloc_indirect( + xfs_ifork_t *ifp, /* inode fork pointer */ + int new_size) /* new indirection array size */ +{ + int nlists; /* number of irec's (ex lists) */ + int size; /* current indirection array size */ + + ASSERT(ifp->if_flags & XFS_IFEXTIREC); + nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ; + size = nlists * sizeof(xfs_ext_irec_t); + ASSERT(ifp->if_real_bytes); + ASSERT((new_size >= 0) && (new_size != size)); + if (new_size == 0) { + xfs_iext_destroy(ifp); + } else { + ifp->if_u1.if_ext_irec = (xfs_ext_irec_t *) + kmem_realloc(ifp->if_u1.if_ext_irec, + new_size, size, KM_NOFS); + } +} + +/* + * Switch from indirection array to linear (direct) extent allocations. + */ +STATIC void +xfs_iext_indirect_to_direct( + xfs_ifork_t *ifp) /* inode fork pointer */ +{ + xfs_bmbt_rec_host_t *ep; /* extent record pointer */ + xfs_extnum_t nextents; /* number of extents in file */ + int size; /* size of file extents */ + + ASSERT(ifp->if_flags & XFS_IFEXTIREC); + nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t); + ASSERT(nextents <= XFS_LINEAR_EXTS); + size = nextents * sizeof(xfs_bmbt_rec_t); + + xfs_iext_irec_compact_pages(ifp); + ASSERT(ifp->if_real_bytes == XFS_IEXT_BUFSZ); + + ep = ifp->if_u1.if_ext_irec->er_extbuf; + kmem_free(ifp->if_u1.if_ext_irec); + ifp->if_flags &= ~XFS_IFEXTIREC; + ifp->if_u1.if_extents = ep; + ifp->if_bytes = size; + if (nextents < XFS_LINEAR_EXTS) { + xfs_iext_realloc_direct(ifp, size); + } +} + +/* + * Free incore file extents. + */ +void +xfs_iext_destroy( + xfs_ifork_t *ifp) /* inode fork pointer */ +{ + if (ifp->if_flags & XFS_IFEXTIREC) { + int erp_idx; + int nlists; + + nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ; + for (erp_idx = nlists - 1; erp_idx >= 0 ; erp_idx--) { + xfs_iext_irec_remove(ifp, erp_idx); + } + ifp->if_flags &= ~XFS_IFEXTIREC; + } else if (ifp->if_real_bytes) { + kmem_free(ifp->if_u1.if_extents); + } else if (ifp->if_bytes) { + memset(ifp->if_u2.if_inline_ext, 0, XFS_INLINE_EXTS * + sizeof(xfs_bmbt_rec_t)); + } + ifp->if_u1.if_extents = NULL; + ifp->if_real_bytes = 0; + ifp->if_bytes = 0; +} + +/* + * Return a pointer to the extent record for file system block bno. + */ +xfs_bmbt_rec_host_t * /* pointer to found extent record */ +xfs_iext_bno_to_ext( + xfs_ifork_t *ifp, /* inode fork pointer */ + xfs_fileoff_t bno, /* block number to search for */ + xfs_extnum_t *idxp) /* index of target extent */ +{ + xfs_bmbt_rec_host_t *base; /* pointer to first extent */ + xfs_filblks_t blockcount = 0; /* number of blocks in extent */ + xfs_bmbt_rec_host_t *ep = NULL; /* pointer to target extent */ + xfs_ext_irec_t *erp = NULL; /* indirection array pointer */ + int high; /* upper boundary in search */ + xfs_extnum_t idx = 0; /* index of target extent */ + int low; /* lower boundary in search */ + xfs_extnum_t nextents; /* number of file extents */ + xfs_fileoff_t startoff = 0; /* start offset of extent */ + + nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t); + if (nextents == 0) { + *idxp = 0; + return NULL; + } + low = 0; + if (ifp->if_flags & XFS_IFEXTIREC) { + /* Find target extent list */ + int erp_idx = 0; + erp = xfs_iext_bno_to_irec(ifp, bno, &erp_idx); + base = erp->er_extbuf; + high = erp->er_extcount - 1; + } else { + base = ifp->if_u1.if_extents; + high = nextents - 1; + } + /* Binary search extent records */ + while (low <= high) { + idx = (low + high) >> 1; + ep = base + idx; + startoff = xfs_bmbt_get_startoff(ep); + blockcount = xfs_bmbt_get_blockcount(ep); + if (bno < startoff) { + high = idx - 1; + } else if (bno >= startoff + blockcount) { + low = idx + 1; + } else { + /* Convert back to file-based extent index */ + if (ifp->if_flags & XFS_IFEXTIREC) { + idx += erp->er_extoff; + } + *idxp = idx; + return ep; + } + } + /* Convert back to file-based extent index */ + if (ifp->if_flags & XFS_IFEXTIREC) { + idx += erp->er_extoff; + } + if (bno >= startoff + blockcount) { + if (++idx == nextents) { + ep = NULL; + } else { + ep = xfs_iext_get_ext(ifp, idx); + } + } + *idxp = idx; + return ep; +} + +/* + * Return a pointer to the indirection array entry containing the + * extent record for filesystem block bno. Store the index of the + * target irec in *erp_idxp. + */ +xfs_ext_irec_t * /* pointer to found extent record */ +xfs_iext_bno_to_irec( + xfs_ifork_t *ifp, /* inode fork pointer */ + xfs_fileoff_t bno, /* block number to search for */ + int *erp_idxp) /* irec index of target ext list */ +{ + xfs_ext_irec_t *erp = NULL; /* indirection array pointer */ + xfs_ext_irec_t *erp_next; /* next indirection array entry */ + int erp_idx; /* indirection array index */ + int nlists; /* number of extent irec's (lists) */ + int high; /* binary search upper limit */ + int low; /* binary search lower limit */ + + ASSERT(ifp->if_flags & XFS_IFEXTIREC); + nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ; + erp_idx = 0; + low = 0; + high = nlists - 1; + while (low <= high) { + erp_idx = (low + high) >> 1; + erp = &ifp->if_u1.if_ext_irec[erp_idx]; + erp_next = erp_idx < nlists - 1 ? erp + 1 : NULL; + if (bno < xfs_bmbt_get_startoff(erp->er_extbuf)) { + high = erp_idx - 1; + } else if (erp_next && bno >= + xfs_bmbt_get_startoff(erp_next->er_extbuf)) { + low = erp_idx + 1; + } else { + break; + } + } + *erp_idxp = erp_idx; + return erp; +} + +/* + * Return a pointer to the indirection array entry containing the + * extent record at file extent index *idxp. Store the index of the + * target irec in *erp_idxp and store the page index of the target + * extent record in *idxp. + */ +xfs_ext_irec_t * +xfs_iext_idx_to_irec( + xfs_ifork_t *ifp, /* inode fork pointer */ + xfs_extnum_t *idxp, /* extent index (file -> page) */ + int *erp_idxp, /* pointer to target irec */ + int realloc) /* new bytes were just added */ +{ + xfs_ext_irec_t *prev; /* pointer to previous irec */ + xfs_ext_irec_t *erp = NULL; /* pointer to current irec */ + int erp_idx; /* indirection array index */ + int nlists; /* number of irec's (ex lists) */ + int high; /* binary search upper limit */ + int low; /* binary search lower limit */ + xfs_extnum_t page_idx = *idxp; /* extent index in target list */ + + ASSERT(ifp->if_flags & XFS_IFEXTIREC); + ASSERT(page_idx >= 0); + ASSERT(page_idx <= ifp->if_bytes / sizeof(xfs_bmbt_rec_t)); + ASSERT(page_idx < ifp->if_bytes / sizeof(xfs_bmbt_rec_t) || realloc); + + nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ; + erp_idx = 0; + low = 0; + high = nlists - 1; + + /* Binary search extent irec's */ + while (low <= high) { + erp_idx = (low + high) >> 1; + erp = &ifp->if_u1.if_ext_irec[erp_idx]; + prev = erp_idx > 0 ? erp - 1 : NULL; + if (page_idx < erp->er_extoff || (page_idx == erp->er_extoff && + realloc && prev && prev->er_extcount < XFS_LINEAR_EXTS)) { + high = erp_idx - 1; + } else if (page_idx > erp->er_extoff + erp->er_extcount || + (page_idx == erp->er_extoff + erp->er_extcount && + !realloc)) { + low = erp_idx + 1; + } else if (page_idx == erp->er_extoff + erp->er_extcount && + erp->er_extcount == XFS_LINEAR_EXTS) { + ASSERT(realloc); + page_idx = 0; + erp_idx++; + erp = erp_idx < nlists ? erp + 1 : NULL; + break; + } else { + page_idx -= erp->er_extoff; + break; + } + } + *idxp = page_idx; + *erp_idxp = erp_idx; + return(erp); +} + +/* + * Allocate and initialize an indirection array once the space needed + * for incore extents increases above XFS_IEXT_BUFSZ. + */ +void +xfs_iext_irec_init( + xfs_ifork_t *ifp) /* inode fork pointer */ +{ + xfs_ext_irec_t *erp; /* indirection array pointer */ + xfs_extnum_t nextents; /* number of extents in file */ + + ASSERT(!(ifp->if_flags & XFS_IFEXTIREC)); + nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t); + ASSERT(nextents <= XFS_LINEAR_EXTS); + + erp = kmem_alloc(sizeof(xfs_ext_irec_t), KM_NOFS); + + if (nextents == 0) { + ifp->if_u1.if_extents = kmem_alloc(XFS_IEXT_BUFSZ, KM_NOFS); + } else if (!ifp->if_real_bytes) { + xfs_iext_inline_to_direct(ifp, XFS_IEXT_BUFSZ); + } else if (ifp->if_real_bytes < XFS_IEXT_BUFSZ) { + xfs_iext_realloc_direct(ifp, XFS_IEXT_BUFSZ); + } + erp->er_extbuf = ifp->if_u1.if_extents; + erp->er_extcount = nextents; + erp->er_extoff = 0; + + ifp->if_flags |= XFS_IFEXTIREC; + ifp->if_real_bytes = XFS_IEXT_BUFSZ; + ifp->if_bytes = nextents * sizeof(xfs_bmbt_rec_t); + ifp->if_u1.if_ext_irec = erp; + + return; +} + +/* + * Allocate and initialize a new entry in the indirection array. + */ +xfs_ext_irec_t * +xfs_iext_irec_new( + xfs_ifork_t *ifp, /* inode fork pointer */ + int erp_idx) /* index for new irec */ +{ + xfs_ext_irec_t *erp; /* indirection array pointer */ + int i; /* loop counter */ + int nlists; /* number of irec's (ex lists) */ + + ASSERT(ifp->if_flags & XFS_IFEXTIREC); + nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ; + + /* Resize indirection array */ + xfs_iext_realloc_indirect(ifp, ++nlists * + sizeof(xfs_ext_irec_t)); + /* + * Move records down in the array so the + * new page can use erp_idx. + */ + erp = ifp->if_u1.if_ext_irec; + for (i = nlists - 1; i > erp_idx; i--) { + memmove(&erp[i], &erp[i-1], sizeof(xfs_ext_irec_t)); + } + ASSERT(i == erp_idx); + + /* Initialize new extent record */ + erp = ifp->if_u1.if_ext_irec; + erp[erp_idx].er_extbuf = kmem_alloc(XFS_IEXT_BUFSZ, KM_NOFS); + ifp->if_real_bytes = nlists * XFS_IEXT_BUFSZ; + memset(erp[erp_idx].er_extbuf, 0, XFS_IEXT_BUFSZ); + erp[erp_idx].er_extcount = 0; + erp[erp_idx].er_extoff = erp_idx > 0 ? + erp[erp_idx-1].er_extoff + erp[erp_idx-1].er_extcount : 0; + return (&erp[erp_idx]); +} + +/* + * Remove a record from the indirection array. + */ +void +xfs_iext_irec_remove( + xfs_ifork_t *ifp, /* inode fork pointer */ + int erp_idx) /* irec index to remove */ +{ + xfs_ext_irec_t *erp; /* indirection array pointer */ + int i; /* loop counter */ + int nlists; /* number of irec's (ex lists) */ + + ASSERT(ifp->if_flags & XFS_IFEXTIREC); + nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ; + erp = &ifp->if_u1.if_ext_irec[erp_idx]; + if (erp->er_extbuf) { + xfs_iext_irec_update_extoffs(ifp, erp_idx + 1, + -erp->er_extcount); + kmem_free(erp->er_extbuf); + } + /* Compact extent records */ + erp = ifp->if_u1.if_ext_irec; + for (i = erp_idx; i < nlists - 1; i++) { + memmove(&erp[i], &erp[i+1], sizeof(xfs_ext_irec_t)); + } + /* + * Manually free the last extent record from the indirection + * array. A call to xfs_iext_realloc_indirect() with a size + * of zero would result in a call to xfs_iext_destroy() which + * would in turn call this function again, creating a nasty + * infinite loop. + */ + if (--nlists) { + xfs_iext_realloc_indirect(ifp, + nlists * sizeof(xfs_ext_irec_t)); + } else { + kmem_free(ifp->if_u1.if_ext_irec); + } + ifp->if_real_bytes = nlists * XFS_IEXT_BUFSZ; +} + +/* + * This is called to clean up large amounts of unused memory allocated + * by the indirection array. Before compacting anything though, verify + * that the indirection array is still needed and switch back to the + * linear extent list (or even the inline buffer) if possible. The + * compaction policy is as follows: + * + * Full Compaction: Extents fit into a single page (or inline buffer) + * Partial Compaction: Extents occupy less than 50% of allocated space + * No Compaction: Extents occupy at least 50% of allocated space + */ +void +xfs_iext_irec_compact( + xfs_ifork_t *ifp) /* inode fork pointer */ +{ + xfs_extnum_t nextents; /* number of extents in file */ + int nlists; /* number of irec's (ex lists) */ + + ASSERT(ifp->if_flags & XFS_IFEXTIREC); + nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ; + nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t); + + if (nextents == 0) { + xfs_iext_destroy(ifp); + } else if (nextents <= XFS_INLINE_EXTS) { + xfs_iext_indirect_to_direct(ifp); + xfs_iext_direct_to_inline(ifp, nextents); + } else if (nextents <= XFS_LINEAR_EXTS) { + xfs_iext_indirect_to_direct(ifp); + } else if (nextents < (nlists * XFS_LINEAR_EXTS) >> 1) { + xfs_iext_irec_compact_pages(ifp); + } +} + +/* + * Combine extents from neighboring extent pages. + */ +void +xfs_iext_irec_compact_pages( + xfs_ifork_t *ifp) /* inode fork pointer */ +{ + xfs_ext_irec_t *erp, *erp_next;/* pointers to irec entries */ + int erp_idx = 0; /* indirection array index */ + int nlists; /* number of irec's (ex lists) */ + + ASSERT(ifp->if_flags & XFS_IFEXTIREC); + nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ; + while (erp_idx < nlists - 1) { + erp = &ifp->if_u1.if_ext_irec[erp_idx]; + erp_next = erp + 1; + if (erp_next->er_extcount <= + (XFS_LINEAR_EXTS - erp->er_extcount)) { + memcpy(&erp->er_extbuf[erp->er_extcount], + erp_next->er_extbuf, erp_next->er_extcount * + sizeof(xfs_bmbt_rec_t)); + erp->er_extcount += erp_next->er_extcount; + /* + * Free page before removing extent record + * so er_extoffs don't get modified in + * xfs_iext_irec_remove. + */ + kmem_free(erp_next->er_extbuf); + erp_next->er_extbuf = NULL; + xfs_iext_irec_remove(ifp, erp_idx + 1); + nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ; + } else { + erp_idx++; + } + } +} + +/* + * This is called to update the er_extoff field in the indirection + * array when extents have been added or removed from one of the + * extent lists. erp_idx contains the irec index to begin updating + * at and ext_diff contains the number of extents that were added + * or removed. + */ +void +xfs_iext_irec_update_extoffs( + xfs_ifork_t *ifp, /* inode fork pointer */ + int erp_idx, /* irec index to update */ + int ext_diff) /* number of new extents */ +{ + int i; /* loop counter */ + int nlists; /* number of irec's (ex lists */ + + ASSERT(ifp->if_flags & XFS_IFEXTIREC); + nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ; + for (i = erp_idx; i < nlists; i++) { + ifp->if_u1.if_ext_irec[i].er_extoff += ext_diff; + } +} diff --git a/fs/xfs/xfs_inode_fork.h b/fs/xfs/xfs_inode_fork.h new file mode 100644 index 0000000..28661a0 --- /dev/null +++ b/fs/xfs/xfs_inode_fork.h @@ -0,0 +1,171 @@ +/* + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#ifndef __XFS_INODE_FORK_H__ +#define __XFS_INODE_FORK_H__ + +struct xfs_inode_log_item; + +/* + * The following xfs_ext_irec_t struct introduces a second (top) level + * to the in-core extent allocation scheme. These structs are allocated + * in a contiguous block, creating an indirection array where each entry + * (irec) contains a pointer to a buffer of in-core extent records which + * it manages. Each extent buffer is 4k in size, since 4k is the system + * page size on Linux i386 and systems with larger page sizes don't seem + * to gain much, if anything, by using their native page size as the + * extent buffer size. Also, using 4k extent buffers everywhere provides + * a consistent interface for CXFS across different platforms. + * + * There is currently no limit on the number of irec's (extent lists) + * allowed, so heavily fragmented files may require an indirection array + * which spans multiple system pages of memory. The number of extents + * which would require this amount of contiguous memory is very large + * and should not cause problems in the foreseeable future. However, + * if the memory needed for the contiguous array ever becomes a problem, + * it is possible that a third level of indirection may be required. + */ +typedef struct xfs_ext_irec { + xfs_bmbt_rec_host_t *er_extbuf; /* block of extent records */ + xfs_extnum_t er_extoff; /* extent offset in file */ + xfs_extnum_t er_extcount; /* number of extents in page/block */ +} xfs_ext_irec_t; + +/* + * File incore extent information, present for each of data & attr forks. + */ +#define XFS_IEXT_BUFSZ 4096 +#define XFS_LINEAR_EXTS (XFS_IEXT_BUFSZ / (uint)sizeof(xfs_bmbt_rec_t)) +#define XFS_INLINE_EXTS 2 +#define XFS_INLINE_DATA 32 +typedef struct xfs_ifork { + int if_bytes; /* bytes in if_u1 */ + int if_real_bytes; /* bytes allocated in if_u1 */ + struct xfs_btree_block *if_broot; /* file's incore btree root */ + short if_broot_bytes; /* bytes allocated for root */ + unsigned char if_flags; /* per-fork flags */ + union { + xfs_bmbt_rec_host_t *if_extents;/* linear map file exts */ + xfs_ext_irec_t *if_ext_irec; /* irec map file exts */ + char *if_data; /* inline file data */ + } if_u1; + union { + xfs_bmbt_rec_host_t if_inline_ext[XFS_INLINE_EXTS]; + /* very small file extents */ + char if_inline_data[XFS_INLINE_DATA]; + /* very small file data */ + xfs_dev_t if_rdev; /* dev number if special */ + uuid_t if_uuid; /* mount point value */ + } if_u2; +} xfs_ifork_t; + +/* + * Per-fork incore inode flags. + */ +#define XFS_IFINLINE 0x01 /* Inline data is read in */ +#define XFS_IFEXTENTS 0x02 /* All extent pointers are read in */ +#define XFS_IFBROOT 0x04 /* i_broot points to the bmap b-tree root */ +#define XFS_IFEXTIREC 0x08 /* Indirection array of extent blocks */ + +/* + * Fork handling. + */ + +#define XFS_IFORK_Q(ip) ((ip)->i_d.di_forkoff != 0) +#define XFS_IFORK_BOFF(ip) ((int)((ip)->i_d.di_forkoff << 3)) + +#define XFS_IFORK_PTR(ip,w) \ + ((w) == XFS_DATA_FORK ? \ + &(ip)->i_df : \ + (ip)->i_afp) +#define XFS_IFORK_DSIZE(ip) \ + (XFS_IFORK_Q(ip) ? \ + XFS_IFORK_BOFF(ip) : \ + XFS_LITINO((ip)->i_mount, (ip)->i_d.di_version)) +#define XFS_IFORK_ASIZE(ip) \ + (XFS_IFORK_Q(ip) ? \ + XFS_LITINO((ip)->i_mount, (ip)->i_d.di_version) - \ + XFS_IFORK_BOFF(ip) : \ + 0) +#define XFS_IFORK_SIZE(ip,w) \ + ((w) == XFS_DATA_FORK ? \ + XFS_IFORK_DSIZE(ip) : \ + XFS_IFORK_ASIZE(ip)) +#define XFS_IFORK_FORMAT(ip,w) \ + ((w) == XFS_DATA_FORK ? \ + (ip)->i_d.di_format : \ + (ip)->i_d.di_aformat) +#define XFS_IFORK_FMT_SET(ip,w,n) \ + ((w) == XFS_DATA_FORK ? \ + ((ip)->i_d.di_format = (n)) : \ + ((ip)->i_d.di_aformat = (n))) +#define XFS_IFORK_NEXTENTS(ip,w) \ + ((w) == XFS_DATA_FORK ? \ + (ip)->i_d.di_nextents : \ + (ip)->i_d.di_anextents) +#define XFS_IFORK_NEXT_SET(ip,w,n) \ + ((w) == XFS_DATA_FORK ? \ + ((ip)->i_d.di_nextents = (n)) : \ + ((ip)->i_d.di_anextents = (n))) +#define XFS_IFORK_MAXEXT(ip, w) \ + (XFS_IFORK_SIZE(ip, w) / sizeof(xfs_bmbt_rec_t)) + +int xfs_iformat_fork(struct xfs_inode *, struct xfs_dinode *); +void xfs_iflush_fork(struct xfs_inode *, struct xfs_dinode *, + struct xfs_inode_log_item *, int, + struct xfs_buf *); +void xfs_idestroy_fork(struct xfs_inode *, int); +void xfs_idata_realloc(struct xfs_inode *, int, int); +void xfs_iroot_realloc(struct xfs_inode *, int, int); +int xfs_iread_extents(struct xfs_trans *, struct xfs_inode *, int); +int xfs_iextents_copy(struct xfs_inode *, struct xfs_bmbt_rec *, + int); + +struct xfs_bmbt_rec_host * + xfs_iext_get_ext(struct xfs_ifork *, xfs_extnum_t); +void xfs_iext_insert(struct xfs_inode *, xfs_extnum_t, xfs_extnum_t, + struct xfs_bmbt_irec *, int); +void xfs_iext_add(struct xfs_ifork *, xfs_extnum_t, int); +void xfs_iext_add_indirect_multi(struct xfs_ifork *, int, + xfs_extnum_t, int); +void xfs_iext_remove(struct xfs_inode *, xfs_extnum_t, int, int); +void xfs_iext_remove_inline(struct xfs_ifork *, xfs_extnum_t, int); +void xfs_iext_remove_direct(struct xfs_ifork *, xfs_extnum_t, int); +void xfs_iext_remove_indirect(struct xfs_ifork *, xfs_extnum_t, int); +void xfs_iext_realloc_direct(struct xfs_ifork *, int); +void xfs_iext_direct_to_inline(struct xfs_ifork *, xfs_extnum_t); +void xfs_iext_inline_to_direct(struct xfs_ifork *, int); +void xfs_iext_destroy(struct xfs_ifork *); +struct xfs_bmbt_rec_host * + xfs_iext_bno_to_ext(struct xfs_ifork *, xfs_fileoff_t, int *); +struct xfs_ext_irec * + xfs_iext_bno_to_irec(struct xfs_ifork *, xfs_fileoff_t, int *); +struct xfs_ext_irec * + xfs_iext_idx_to_irec(struct xfs_ifork *, xfs_extnum_t *, int *, + int); +void xfs_iext_irec_init(struct xfs_ifork *); +struct xfs_ext_irec * + xfs_iext_irec_new(struct xfs_ifork *, int); +void xfs_iext_irec_remove(struct xfs_ifork *, int); +void xfs_iext_irec_compact(struct xfs_ifork *); +void xfs_iext_irec_compact_pages(struct xfs_ifork *); +void xfs_iext_irec_compact_full(struct xfs_ifork *); +void xfs_iext_irec_update_extoffs(struct xfs_ifork *, int, int); + +extern struct kmem_zone *xfs_ifork_zone; + +#endif /* __XFS_INODE_FORK_H__ */ diff --git a/fs/xfs/xfs_inode_item.c b/fs/xfs/xfs_inode_item.c index f76ff52..3780811 100644 --- a/fs/xfs/xfs_inode_item.c +++ b/fs/xfs/xfs_inode_item.c @@ -47,32 +47,44 @@ static inline struct xfs_inode_log_item *INODE_ITEM(struct xfs_log_item *lip) * inode core, and possibly one for the inode data/extents/b-tree root * and one for the inode attribute data/extents/b-tree root. */ -STATIC uint +STATIC void xfs_inode_item_size( - struct xfs_log_item *lip) + struct xfs_log_item *lip, + int *nvecs, + int *nbytes) { struct xfs_inode_log_item *iip = INODE_ITEM(lip); struct xfs_inode *ip = iip->ili_inode; - uint nvecs = 2; + + *nvecs += 2; + *nbytes += sizeof(struct xfs_inode_log_format) + + xfs_icdinode_size(ip->i_d.di_version); switch (ip->i_d.di_format) { case XFS_DINODE_FMT_EXTENTS: if ((iip->ili_fields & XFS_ILOG_DEXT) && ip->i_d.di_nextents > 0 && - ip->i_df.if_bytes > 0) - nvecs++; + ip->i_df.if_bytes > 0) { + /* worst case, doesn't subtract delalloc extents */ + *nbytes += XFS_IFORK_DSIZE(ip); + *nvecs += 1; + } break; case XFS_DINODE_FMT_BTREE: if ((iip->ili_fields & XFS_ILOG_DBROOT) && - ip->i_df.if_broot_bytes > 0) - nvecs++; + ip->i_df.if_broot_bytes > 0) { + *nbytes += ip->i_df.if_broot_bytes; + *nvecs += 1; + } break; case XFS_DINODE_FMT_LOCAL: if ((iip->ili_fields & XFS_ILOG_DDATA) && - ip->i_df.if_bytes > 0) - nvecs++; + ip->i_df.if_bytes > 0) { + *nbytes += roundup(ip->i_df.if_bytes, 4); + *nvecs += 1; + } break; case XFS_DINODE_FMT_DEV: @@ -85,7 +97,7 @@ xfs_inode_item_size( } if (!XFS_IFORK_Q(ip)) - return nvecs; + return; /* @@ -95,28 +107,33 @@ xfs_inode_item_size( case XFS_DINODE_FMT_EXTENTS: if ((iip->ili_fields & XFS_ILOG_AEXT) && ip->i_d.di_anextents > 0 && - ip->i_afp->if_bytes > 0) - nvecs++; + ip->i_afp->if_bytes > 0) { + /* worst case, doesn't subtract unused space */ + *nbytes += XFS_IFORK_ASIZE(ip); + *nvecs += 1; + } break; case XFS_DINODE_FMT_BTREE: if ((iip->ili_fields & XFS_ILOG_ABROOT) && - ip->i_afp->if_broot_bytes > 0) - nvecs++; + ip->i_afp->if_broot_bytes > 0) { + *nbytes += ip->i_afp->if_broot_bytes; + *nvecs += 1; + } break; case XFS_DINODE_FMT_LOCAL: if ((iip->ili_fields & XFS_ILOG_ADATA) && - ip->i_afp->if_bytes > 0) - nvecs++; + ip->i_afp->if_bytes > 0) { + *nbytes += roundup(ip->i_afp->if_bytes, 4); + *nvecs += 1; + } break; default: ASSERT(0); break; } - - return nvecs; } /* diff --git a/fs/xfs/xfs_inode_item.h b/fs/xfs/xfs_inode_item.h index 779812f..dce4d65 100644 --- a/fs/xfs/xfs_inode_item.h +++ b/fs/xfs/xfs_inode_item.h @@ -18,123 +18,13 @@ #ifndef __XFS_INODE_ITEM_H__ #define __XFS_INODE_ITEM_H__ -/* - * This is the structure used to lay out an inode log item in the - * log. The size of the inline data/extents/b-tree root to be logged - * (if any) is indicated in the ilf_dsize field. Changes to this structure - * must be added on to the end. - */ -typedef struct xfs_inode_log_format { - __uint16_t ilf_type; /* inode log item type */ - __uint16_t ilf_size; /* size of this item */ - __uint32_t ilf_fields; /* flags for fields logged */ - __uint16_t ilf_asize; /* size of attr d/ext/root */ - __uint16_t ilf_dsize; /* size of data/ext/root */ - __uint64_t ilf_ino; /* inode number */ - union { - __uint32_t ilfu_rdev; /* rdev value for dev inode*/ - uuid_t ilfu_uuid; /* mount point value */ - } ilf_u; - __int64_t ilf_blkno; /* blkno of inode buffer */ - __int32_t ilf_len; /* len of inode buffer */ - __int32_t ilf_boffset; /* off of inode in buffer */ -} xfs_inode_log_format_t; - -typedef struct xfs_inode_log_format_32 { - __uint16_t ilf_type; /* inode log item type */ - __uint16_t ilf_size; /* size of this item */ - __uint32_t ilf_fields; /* flags for fields logged */ - __uint16_t ilf_asize; /* size of attr d/ext/root */ - __uint16_t ilf_dsize; /* size of data/ext/root */ - __uint64_t ilf_ino; /* inode number */ - union { - __uint32_t ilfu_rdev; /* rdev value for dev inode*/ - uuid_t ilfu_uuid; /* mount point value */ - } ilf_u; - __int64_t ilf_blkno; /* blkno of inode buffer */ - __int32_t ilf_len; /* len of inode buffer */ - __int32_t ilf_boffset; /* off of inode in buffer */ -} __attribute__((packed)) xfs_inode_log_format_32_t; - -typedef struct xfs_inode_log_format_64 { - __uint16_t ilf_type; /* inode log item type */ - __uint16_t ilf_size; /* size of this item */ - __uint32_t ilf_fields; /* flags for fields logged */ - __uint16_t ilf_asize; /* size of attr d/ext/root */ - __uint16_t ilf_dsize; /* size of data/ext/root */ - __uint32_t ilf_pad; /* pad for 64 bit boundary */ - __uint64_t ilf_ino; /* inode number */ - union { - __uint32_t ilfu_rdev; /* rdev value for dev inode*/ - uuid_t ilfu_uuid; /* mount point value */ - } ilf_u; - __int64_t ilf_blkno; /* blkno of inode buffer */ - __int32_t ilf_len; /* len of inode buffer */ - __int32_t ilf_boffset; /* off of inode in buffer */ -} xfs_inode_log_format_64_t; - -/* - * Flags for xfs_trans_log_inode flags field. - */ -#define XFS_ILOG_CORE 0x001 /* log standard inode fields */ -#define XFS_ILOG_DDATA 0x002 /* log i_df.if_data */ -#define XFS_ILOG_DEXT 0x004 /* log i_df.if_extents */ -#define XFS_ILOG_DBROOT 0x008 /* log i_df.i_broot */ -#define XFS_ILOG_DEV 0x010 /* log the dev field */ -#define XFS_ILOG_UUID 0x020 /* log the uuid field */ -#define XFS_ILOG_ADATA 0x040 /* log i_af.if_data */ -#define XFS_ILOG_AEXT 0x080 /* log i_af.if_extents */ -#define XFS_ILOG_ABROOT 0x100 /* log i_af.i_broot */ - - -/* - * The timestamps are dirty, but not necessarily anything else in the inode - * core. Unlike the other fields above this one must never make it to disk - * in the ilf_fields of the inode_log_format, but is purely store in-memory in - * ili_fields in the inode_log_item. - */ -#define XFS_ILOG_TIMESTAMP 0x4000 - -#define XFS_ILOG_NONCORE (XFS_ILOG_DDATA | XFS_ILOG_DEXT | \ - XFS_ILOG_DBROOT | XFS_ILOG_DEV | \ - XFS_ILOG_UUID | XFS_ILOG_ADATA | \ - XFS_ILOG_AEXT | XFS_ILOG_ABROOT) - -#define XFS_ILOG_DFORK (XFS_ILOG_DDATA | XFS_ILOG_DEXT | \ - XFS_ILOG_DBROOT) - -#define XFS_ILOG_AFORK (XFS_ILOG_ADATA | XFS_ILOG_AEXT | \ - XFS_ILOG_ABROOT) - -#define XFS_ILOG_ALL (XFS_ILOG_CORE | XFS_ILOG_DDATA | \ - XFS_ILOG_DEXT | XFS_ILOG_DBROOT | \ - XFS_ILOG_DEV | XFS_ILOG_UUID | \ - XFS_ILOG_ADATA | XFS_ILOG_AEXT | \ - XFS_ILOG_ABROOT | XFS_ILOG_TIMESTAMP) - -static inline int xfs_ilog_fbroot(int w) -{ - return (w == XFS_DATA_FORK ? XFS_ILOG_DBROOT : XFS_ILOG_ABROOT); -} - -static inline int xfs_ilog_fext(int w) -{ - return (w == XFS_DATA_FORK ? XFS_ILOG_DEXT : XFS_ILOG_AEXT); -} - -static inline int xfs_ilog_fdata(int w) -{ - return (w == XFS_DATA_FORK ? XFS_ILOG_DDATA : XFS_ILOG_ADATA); -} - -#ifdef __KERNEL__ +/* kernel only definitions */ struct xfs_buf; struct xfs_bmbt_rec; struct xfs_inode; struct xfs_mount; - typedef struct xfs_inode_log_item { xfs_log_item_t ili_item; /* common portion */ struct xfs_inode *ili_inode; /* inode ptr */ @@ -151,7 +41,6 @@ typedef struct xfs_inode_log_item { xfs_inode_log_format_t ili_format; /* logged structure */ } xfs_inode_log_item_t; - static inline int xfs_inode_clean(xfs_inode_t *ip) { return !ip->i_itemp || !(ip->i_itemp->ili_fields & XFS_ILOG_ALL); @@ -165,6 +54,6 @@ extern void xfs_iflush_abort(struct xfs_inode *, bool); extern int xfs_inode_item_format_convert(xfs_log_iovec_t *, xfs_inode_log_format_t *); -#endif /* __KERNEL__ */ +extern struct kmem_zone *xfs_ili_zone; #endif /* __XFS_INODE_ITEM_H__ */ diff --git a/fs/xfs/xfs_ioctl.c b/fs/xfs/xfs_ioctl.c index 6e2bca5..bdebc21 100644 --- a/fs/xfs/xfs_ioctl.c +++ b/fs/xfs/xfs_ioctl.c @@ -17,6 +17,7 @@ */ #include "xfs.h" #include "xfs_fs.h" +#include "xfs_format.h" #include "xfs_log.h" #include "xfs_trans.h" #include "xfs_sb.h" @@ -32,17 +33,16 @@ #include "xfs_error.h" #include "xfs_attr.h" #include "xfs_bmap.h" +#include "xfs_bmap_util.h" #include "xfs_buf_item.h" -#include "xfs_utils.h" -#include "xfs_dfrag.h" #include "xfs_fsops.h" -#include "xfs_vnodeops.h" #include "xfs_discard.h" #include "xfs_quota.h" #include "xfs_inode_item.h" #include "xfs_export.h" #include "xfs_trace.h" #include "xfs_icache.h" +#include "xfs_symlink.h" #include <linux/capability.h> #include <linux/dcache.h> @@ -350,6 +350,40 @@ xfs_readlink_by_handle( return error; } +int +xfs_set_dmattrs( + xfs_inode_t *ip, + u_int evmask, + u_int16_t state) +{ + xfs_mount_t *mp = ip->i_mount; + xfs_trans_t *tp; + int error; + + if (!capable(CAP_SYS_ADMIN)) + return XFS_ERROR(EPERM); + + if (XFS_FORCED_SHUTDOWN(mp)) + return XFS_ERROR(EIO); + + tp = xfs_trans_alloc(mp, XFS_TRANS_SET_DMATTRS); + error = xfs_trans_reserve(tp, &M_RES(mp)->tr_ichange, 0, 0); + if (error) { + xfs_trans_cancel(tp, 0); + return error; + } + xfs_ilock(ip, XFS_ILOCK_EXCL); + xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL); + + ip->i_d.di_dmevmask = evmask; + ip->i_d.di_dmstate = state; + + xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); + error = xfs_trans_commit(tp, 0); + + return error; +} + STATIC int xfs_fssetdm_by_handle( struct file *parfilp, @@ -967,7 +1001,7 @@ xfs_ioctl_setattr( * first do an error checking pass. */ tp = xfs_trans_alloc(mp, XFS_TRANS_SETATTR_NOT_SIZE); - code = xfs_trans_reserve(tp, 0, XFS_ICHANGE_LOG_RES(mp), 0, 0, 0); + code = xfs_trans_reserve(tp, &M_RES(mp)->tr_ichange, 0, 0); if (code) goto error_return; @@ -981,15 +1015,22 @@ xfs_ioctl_setattr( * to the file owner ID, except in cases where the * CAP_FSETID capability is applicable. */ - if (current_fsuid() != ip->i_d.di_uid && !capable(CAP_FOWNER)) { + if (!inode_owner_or_capable(VFS_I(ip))) { code = XFS_ERROR(EPERM); goto error_return; } /* * Do a quota reservation only if projid is actually going to change. + * Only allow changing of projid from init_user_ns since it is a + * non user namespace aware identifier. */ if (mask & FSX_PROJID) { + if (current_user_ns() != &init_user_ns) { + code = XFS_ERROR(EINVAL); + goto error_return; + } + if (XFS_IS_QUOTA_RUNNING(mp) && XFS_IS_PQUOTA_ON(mp) && xfs_get_projid(ip) != fa->fsx_projid) { @@ -1103,7 +1144,7 @@ xfs_ioctl_setattr( * cleared upon successful return from chown() */ if ((ip->i_d.di_mode & (S_ISUID|S_ISGID)) && - !capable(CAP_FSETID)) + !inode_capable(VFS_I(ip), CAP_FSETID)) ip->i_d.di_mode &= ~(S_ISUID|S_ISGID); /* @@ -1328,6 +1369,75 @@ xfs_ioc_getbmapx( return 0; } +int +xfs_ioc_swapext( + xfs_swapext_t *sxp) +{ + xfs_inode_t *ip, *tip; + struct fd f, tmp; + int error = 0; + + /* Pull information for the target fd */ + f = fdget((int)sxp->sx_fdtarget); + if (!f.file) { + error = XFS_ERROR(EINVAL); + goto out; + } + + if (!(f.file->f_mode & FMODE_WRITE) || + !(f.file->f_mode & FMODE_READ) || + (f.file->f_flags & O_APPEND)) { + error = XFS_ERROR(EBADF); + goto out_put_file; + } + + tmp = fdget((int)sxp->sx_fdtmp); + if (!tmp.file) { + error = XFS_ERROR(EINVAL); + goto out_put_file; + } + + if (!(tmp.file->f_mode & FMODE_WRITE) || + !(tmp.file->f_mode & FMODE_READ) || + (tmp.file->f_flags & O_APPEND)) { + error = XFS_ERROR(EBADF); + goto out_put_tmp_file; + } + + if (IS_SWAPFILE(file_inode(f.file)) || + IS_SWAPFILE(file_inode(tmp.file))) { + error = XFS_ERROR(EINVAL); + goto out_put_tmp_file; + } + + ip = XFS_I(file_inode(f.file)); + tip = XFS_I(file_inode(tmp.file)); + + if (ip->i_mount != tip->i_mount) { + error = XFS_ERROR(EINVAL); + goto out_put_tmp_file; + } + + if (ip->i_ino == tip->i_ino) { + error = XFS_ERROR(EINVAL); + goto out_put_tmp_file; + } + + if (XFS_FORCED_SHUTDOWN(ip->i_mount)) { + error = XFS_ERROR(EIO); + goto out_put_tmp_file; + } + + error = xfs_swap_extents(ip, tip, sxp); + + out_put_tmp_file: + fdput(tmp); + out_put_file: + fdput(f); + out: + return error; +} + /* * Note: some of the ioctl's return positive numbers as a * byte count indicating success, such as readlink_by_handle. @@ -1472,7 +1582,7 @@ xfs_file_ioctl( error = mnt_want_write_file(filp); if (error) return error; - error = xfs_swapext(&sxp); + error = xfs_ioc_swapext(&sxp); mnt_drop_write_file(filp); return -error; } @@ -1610,23 +1720,23 @@ xfs_file_ioctl( return -error; case XFS_IOC_FREE_EOFBLOCKS: { - struct xfs_eofblocks eofb; + struct xfs_fs_eofblocks eofb; + struct xfs_eofblocks keofb; - if (copy_from_user(&eofb, arg, sizeof(eofb))) - return -XFS_ERROR(EFAULT); + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; - if (eofb.eof_version != XFS_EOFBLOCKS_VERSION) - return -XFS_ERROR(EINVAL); + if (mp->m_flags & XFS_MOUNT_RDONLY) + return -XFS_ERROR(EROFS); - if (eofb.eof_flags & ~XFS_EOF_FLAGS_VALID) - return -XFS_ERROR(EINVAL); + if (copy_from_user(&eofb, arg, sizeof(eofb))) + return -XFS_ERROR(EFAULT); - if (memchr_inv(&eofb.pad32, 0, sizeof(eofb.pad32)) || - memchr_inv(eofb.pad64, 0, sizeof(eofb.pad64))) - return -XFS_ERROR(EINVAL); + error = xfs_fs_eofblocks_from_user(&eofb, &keofb); + if (error) + return -error; - error = xfs_icache_free_eofblocks(mp, &eofb); - return -error; + return -xfs_icache_free_eofblocks(mp, &keofb); } default: diff --git a/fs/xfs/xfs_ioctl.h b/fs/xfs/xfs_ioctl.h index d56173b..77c02c7 100644 --- a/fs/xfs/xfs_ioctl.h +++ b/fs/xfs/xfs_ioctl.h @@ -27,6 +27,10 @@ xfs_ioc_space( unsigned int cmd, xfs_flock64_t *bf); +int +xfs_ioc_swapext( + xfs_swapext_t *sxp); + extern int xfs_find_handle( unsigned int cmd, @@ -82,4 +86,10 @@ xfs_file_compat_ioctl( unsigned int cmd, unsigned long arg); +extern int +xfs_set_dmattrs( + struct xfs_inode *ip, + u_int evmask, + u_int16_t state); + #endif diff --git a/fs/xfs/xfs_ioctl32.c b/fs/xfs/xfs_ioctl32.c index c0c6625..d3ab953 100644 --- a/fs/xfs/xfs_ioctl32.c +++ b/fs/xfs/xfs_ioctl32.c @@ -33,8 +33,6 @@ #include "xfs_inode.h" #include "xfs_itable.h" #include "xfs_error.h" -#include "xfs_dfrag.h" -#include "xfs_vnodeops.h" #include "xfs_fsops.h" #include "xfs_alloc.h" #include "xfs_rtalloc.h" @@ -644,7 +642,7 @@ xfs_file_compat_ioctl( error = mnt_want_write_file(filp); if (error) return error; - error = xfs_swapext(&sxp); + error = xfs_ioc_swapext(&sxp); mnt_drop_write_file(filp); return -error; } diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c index 6a70964..8d4d49b 100644 --- a/fs/xfs/xfs_iomap.c +++ b/fs/xfs/xfs_iomap.c @@ -17,6 +17,7 @@ */ #include "xfs.h" #include "xfs_fs.h" +#include "xfs_format.h" #include "xfs_log.h" #include "xfs_trans.h" #include "xfs_sb.h" @@ -32,13 +33,13 @@ #include "xfs_inode_item.h" #include "xfs_btree.h" #include "xfs_bmap.h" +#include "xfs_bmap_util.h" #include "xfs_rtalloc.h" #include "xfs_error.h" #include "xfs_itable.h" #include "xfs_attr.h" #include "xfs_buf_item.h" #include "xfs_trans_space.h" -#include "xfs_utils.h" #include "xfs_iomap.h" #include "xfs_trace.h" #include "xfs_icache.h" @@ -187,10 +188,8 @@ xfs_iomap_write_direct( * Allocate and setup the transaction */ tp = xfs_trans_alloc(mp, XFS_TRANS_DIOSTRAT); - error = xfs_trans_reserve(tp, resblks, - XFS_WRITE_LOG_RES(mp), resrtextents, - XFS_TRANS_PERM_LOG_RES, - XFS_WRITE_LOG_COUNT); + error = xfs_trans_reserve(tp, &M_RES(mp)->tr_write, + resblks, resrtextents); /* * Check for running out of space, note: need lock to return */ @@ -698,10 +697,8 @@ xfs_iomap_write_allocate( tp = xfs_trans_alloc(mp, XFS_TRANS_STRAT_WRITE); tp->t_flags |= XFS_TRANS_RESERVE; nres = XFS_EXTENTADD_SPACE_RES(mp, XFS_DATA_FORK); - error = xfs_trans_reserve(tp, nres, - XFS_WRITE_LOG_RES(mp), - 0, XFS_TRANS_PERM_LOG_RES, - XFS_WRITE_LOG_COUNT); + error = xfs_trans_reserve(tp, &M_RES(mp)->tr_write, + nres, 0); if (error) { xfs_trans_cancel(tp, 0); return XFS_ERROR(error); @@ -864,10 +861,8 @@ xfs_iomap_write_unwritten( sb_start_intwrite(mp->m_super); tp = _xfs_trans_alloc(mp, XFS_TRANS_STRAT_WRITE, KM_NOFS); tp->t_flags |= XFS_TRANS_RESERVE | XFS_TRANS_FREEZE_PROT; - error = xfs_trans_reserve(tp, resblks, - XFS_WRITE_LOG_RES(mp), 0, - XFS_TRANS_PERM_LOG_RES, - XFS_WRITE_LOG_COUNT); + error = xfs_trans_reserve(tp, &M_RES(mp)->tr_write, + resblks, 0); if (error) { xfs_trans_cancel(tp, 0); return XFS_ERROR(error); diff --git a/fs/xfs/xfs_iops.c b/fs/xfs/xfs_iops.c index 96dda62..2b8952d 100644 --- a/fs/xfs/xfs_iops.c +++ b/fs/xfs/xfs_iops.c @@ -17,6 +17,7 @@ */ #include "xfs.h" #include "xfs_fs.h" +#include "xfs_format.h" #include "xfs_acl.h" #include "xfs_log.h" #include "xfs_trans.h" @@ -29,16 +30,19 @@ #include "xfs_dinode.h" #include "xfs_inode.h" #include "xfs_bmap.h" +#include "xfs_bmap_util.h" #include "xfs_rtalloc.h" #include "xfs_error.h" #include "xfs_itable.h" #include "xfs_attr.h" #include "xfs_buf_item.h" -#include "xfs_utils.h" -#include "xfs_vnodeops.h" #include "xfs_inode_item.h" #include "xfs_trace.h" #include "xfs_icache.h" +#include "xfs_symlink.h" +#include "xfs_da_btree.h" +#include "xfs_dir2_format.h" +#include "xfs_dir2_priv.h" #include <linux/capability.h> #include <linux/xattr.h> @@ -87,10 +91,12 @@ xfs_init_security( static void xfs_dentry_to_name( struct xfs_name *namep, - struct dentry *dentry) + struct dentry *dentry, + int mode) { namep->name = dentry->d_name.name; namep->len = dentry->d_name.len; + namep->type = xfs_mode_to_ftype[(mode & S_IFMT) >> S_SHIFT]; } STATIC void @@ -106,7 +112,7 @@ xfs_cleanup_inode( * xfs_init_security we must back out. * ENOSPC can hit here, among other things. */ - xfs_dentry_to_name(&teardown, dentry); + xfs_dentry_to_name(&teardown, dentry, 0); xfs_remove(XFS_I(dir), &teardown, XFS_I(inode)); iput(inode); @@ -146,7 +152,7 @@ xfs_vn_mknod( mode &= ~current_umask(); } - xfs_dentry_to_name(&name, dentry); + xfs_dentry_to_name(&name, dentry, mode); error = xfs_create(XFS_I(dir), &name, mode, rdev, &ip); if (unlikely(error)) goto out_free_acl; @@ -207,7 +213,7 @@ xfs_vn_lookup( if (dentry->d_name.len >= MAXNAMELEN) return ERR_PTR(-ENAMETOOLONG); - xfs_dentry_to_name(&name, dentry); + xfs_dentry_to_name(&name, dentry, 0); error = xfs_lookup(XFS_I(dir), &name, &cip, NULL); if (unlikely(error)) { if (unlikely(error != ENOENT)) @@ -234,7 +240,7 @@ xfs_vn_ci_lookup( if (dentry->d_name.len >= MAXNAMELEN) return ERR_PTR(-ENAMETOOLONG); - xfs_dentry_to_name(&xname, dentry); + xfs_dentry_to_name(&xname, dentry, 0); error = xfs_lookup(XFS_I(dir), &xname, &ip, &ci_name); if (unlikely(error)) { if (unlikely(error != ENOENT)) @@ -269,7 +275,7 @@ xfs_vn_link( struct xfs_name name; int error; - xfs_dentry_to_name(&name, dentry); + xfs_dentry_to_name(&name, dentry, inode->i_mode); error = xfs_link(XFS_I(dir), XFS_I(inode), &name); if (unlikely(error)) @@ -288,7 +294,7 @@ xfs_vn_unlink( struct xfs_name name; int error; - xfs_dentry_to_name(&name, dentry); + xfs_dentry_to_name(&name, dentry, 0); error = -xfs_remove(XFS_I(dir), &name, XFS_I(dentry->d_inode)); if (error) @@ -318,7 +324,7 @@ xfs_vn_symlink( mode = S_IFLNK | (irix_symlink_mode ? 0777 & ~current_umask() : S_IRWXUGO); - xfs_dentry_to_name(&name, dentry); + xfs_dentry_to_name(&name, dentry, mode); error = xfs_symlink(XFS_I(dir), &name, symname, mode, &cip); if (unlikely(error)) @@ -350,12 +356,12 @@ xfs_vn_rename( struct xfs_name oname; struct xfs_name nname; - xfs_dentry_to_name(&oname, odentry); - xfs_dentry_to_name(&nname, ndentry); + xfs_dentry_to_name(&oname, odentry, 0); + xfs_dentry_to_name(&nname, ndentry, odentry->d_inode->i_mode); return -xfs_rename(XFS_I(odir), &oname, XFS_I(odentry->d_inode), XFS_I(ndir), &nname, new_inode ? - XFS_I(new_inode) : NULL); + XFS_I(new_inode) : NULL); } /* @@ -420,8 +426,8 @@ xfs_vn_getattr( stat->dev = inode->i_sb->s_dev; stat->mode = ip->i_d.di_mode; stat->nlink = ip->i_d.di_nlink; - stat->uid = ip->i_d.di_uid; - stat->gid = ip->i_d.di_gid; + stat->uid = inode->i_uid; + stat->gid = inode->i_gid; stat->ino = ip->i_ino; stat->atime = inode->i_atime; stat->mtime = inode->i_mtime; @@ -485,8 +491,8 @@ xfs_setattr_nonsize( int mask = iattr->ia_valid; xfs_trans_t *tp; int error; - uid_t uid = 0, iuid = 0; - gid_t gid = 0, igid = 0; + kuid_t uid = GLOBAL_ROOT_UID, iuid = GLOBAL_ROOT_UID; + kgid_t gid = GLOBAL_ROOT_GID, igid = GLOBAL_ROOT_GID; struct xfs_dquot *udqp = NULL, *gdqp = NULL; struct xfs_dquot *olddquot1 = NULL, *olddquot2 = NULL; @@ -522,13 +528,13 @@ xfs_setattr_nonsize( uid = iattr->ia_uid; qflags |= XFS_QMOPT_UQUOTA; } else { - uid = ip->i_d.di_uid; + uid = inode->i_uid; } if ((mask & ATTR_GID) && XFS_IS_GQUOTA_ON(mp)) { gid = iattr->ia_gid; qflags |= XFS_QMOPT_GQUOTA; } else { - gid = ip->i_d.di_gid; + gid = inode->i_gid; } /* @@ -538,14 +544,16 @@ xfs_setattr_nonsize( */ ASSERT(udqp == NULL); ASSERT(gdqp == NULL); - error = xfs_qm_vop_dqalloc(ip, uid, gid, xfs_get_projid(ip), - qflags, &udqp, &gdqp, NULL); + error = xfs_qm_vop_dqalloc(ip, xfs_kuid_to_uid(uid), + xfs_kgid_to_gid(gid), + xfs_get_projid(ip), + qflags, &udqp, &gdqp, NULL); if (error) return error; } tp = xfs_trans_alloc(mp, XFS_TRANS_SETATTR_NOT_SIZE); - error = xfs_trans_reserve(tp, 0, XFS_ICHANGE_LOG_RES(mp), 0, 0, 0); + error = xfs_trans_reserve(tp, &M_RES(mp)->tr_ichange, 0, 0); if (error) goto out_dqrele; @@ -561,8 +569,8 @@ xfs_setattr_nonsize( * while we didn't have the inode locked, inode's dquot(s) * would have changed also. */ - iuid = ip->i_d.di_uid; - igid = ip->i_d.di_gid; + iuid = inode->i_uid; + igid = inode->i_gid; gid = (mask & ATTR_GID) ? iattr->ia_gid : igid; uid = (mask & ATTR_UID) ? iattr->ia_uid : iuid; @@ -571,8 +579,8 @@ xfs_setattr_nonsize( * going to change. */ if (XFS_IS_QUOTA_RUNNING(mp) && - ((XFS_IS_UQUOTA_ON(mp) && iuid != uid) || - (XFS_IS_GQUOTA_ON(mp) && igid != gid))) { + ((XFS_IS_UQUOTA_ON(mp) && !uid_eq(iuid, uid)) || + (XFS_IS_GQUOTA_ON(mp) && !gid_eq(igid, gid)))) { ASSERT(tp); error = xfs_qm_vop_chown_reserve(tp, ip, udqp, gdqp, NULL, capable(CAP_FOWNER) ? @@ -602,17 +610,17 @@ xfs_setattr_nonsize( * Change the ownerships and register quota modifications * in the transaction. */ - if (iuid != uid) { + if (!uid_eq(iuid, uid)) { if (XFS_IS_QUOTA_RUNNING(mp) && XFS_IS_UQUOTA_ON(mp)) { ASSERT(mask & ATTR_UID); ASSERT(udqp); olddquot1 = xfs_qm_vop_chown(tp, ip, &ip->i_udquot, udqp); } - ip->i_d.di_uid = uid; + ip->i_d.di_uid = xfs_kuid_to_uid(uid); inode->i_uid = uid; } - if (igid != gid) { + if (!gid_eq(igid, gid)) { if (XFS_IS_QUOTA_RUNNING(mp) && XFS_IS_GQUOTA_ON(mp)) { ASSERT(!XFS_IS_PQUOTA_ON(mp)); ASSERT(mask & ATTR_GID); @@ -620,7 +628,7 @@ xfs_setattr_nonsize( olddquot2 = xfs_qm_vop_chown(tp, ip, &ip->i_gdquot, gdqp); } - ip->i_d.di_gid = gid; + ip->i_d.di_gid = xfs_kgid_to_gid(gid); inode->i_gid = gid; } } @@ -807,9 +815,7 @@ xfs_setattr_size( goto out_unlock; tp = xfs_trans_alloc(mp, XFS_TRANS_SETATTR_SIZE); - error = xfs_trans_reserve(tp, 0, XFS_ITRUNCATE_LOG_RES(mp), 0, - XFS_TRANS_PERM_LOG_RES, - XFS_ITRUNCATE_LOG_COUNT); + error = xfs_trans_reserve(tp, &M_RES(mp)->tr_itruncate, 0, 0); if (error) goto out_trans_cancel; @@ -932,7 +938,7 @@ xfs_vn_update_time( trace_xfs_update_time(ip); tp = xfs_trans_alloc(mp, XFS_TRANS_FSYNC_TS); - error = xfs_trans_reserve(tp, 0, XFS_FSYNC_TS_LOG_RES(mp), 0, 0, 0); + error = xfs_trans_reserve(tp, &M_RES(mp)->tr_fsyncts, 0, 0); if (error) { xfs_trans_cancel(tp, 0); return -error; @@ -1173,8 +1179,8 @@ xfs_setup_inode( inode->i_mode = ip->i_d.di_mode; set_nlink(inode, ip->i_d.di_nlink); - inode->i_uid = ip->i_d.di_uid; - inode->i_gid = ip->i_d.di_gid; + inode->i_uid = xfs_uid_to_kuid(ip->i_d.di_uid); + inode->i_gid = xfs_gid_to_kgid(ip->i_d.di_gid); switch (inode->i_mode & S_IFMT) { case S_IFBLK: diff --git a/fs/xfs/xfs_iops.h b/fs/xfs/xfs_iops.h index ef41c92..d81fb41 100644 --- a/fs/xfs/xfs_iops.h +++ b/fs/xfs/xfs_iops.h @@ -27,4 +27,17 @@ extern ssize_t xfs_vn_listxattr(struct dentry *, char *data, size_t size); extern void xfs_setup_inode(struct xfs_inode *); +/* + * Internal setattr interfaces. + */ +#define XFS_ATTR_DMI 0x01 /* invocation from a DMI function */ +#define XFS_ATTR_NONBLOCK 0x02 /* return EAGAIN if op would block */ +#define XFS_ATTR_NOLOCK 0x04 /* Don't grab any conflicting locks */ +#define XFS_ATTR_NOACL 0x08 /* Don't call xfs_acl_chmod */ +#define XFS_ATTR_SYNC 0x10 /* synchronous operation required */ + +extern int xfs_setattr_nonsize(struct xfs_inode *ip, struct iattr *vap, + int flags); +extern int xfs_setattr_size(struct xfs_inode *ip, struct iattr *vap, int flags); + #endif /* __XFS_IOPS_H__ */ diff --git a/fs/xfs/xfs_linux.h b/fs/xfs/xfs_linux.h index 800f896..f9bb590 100644 --- a/fs/xfs/xfs_linux.h +++ b/fs/xfs/xfs_linux.h @@ -32,6 +32,38 @@ # define XFS_BIG_INUMS 0 #endif +/* + * Kernel specific type declarations for XFS + */ +typedef signed char __int8_t; +typedef unsigned char __uint8_t; +typedef signed short int __int16_t; +typedef unsigned short int __uint16_t; +typedef signed int __int32_t; +typedef unsigned int __uint32_t; +typedef signed long long int __int64_t; +typedef unsigned long long int __uint64_t; + +typedef __uint32_t inst_t; /* an instruction */ + +typedef __s64 xfs_off_t; /* <file offset> type */ +typedef unsigned long long xfs_ino_t; /* <inode> type */ +typedef __s64 xfs_daddr_t; /* <disk address> type */ +typedef char * xfs_caddr_t; /* <core address> type */ +typedef __u32 xfs_dev_t; +typedef __u32 xfs_nlink_t; + +/* __psint_t is the same size as a pointer */ +#if (BITS_PER_LONG == 32) +typedef __int32_t __psint_t; +typedef __uint32_t __psunsigned_t; +#elif (BITS_PER_LONG == 64) +typedef __int64_t __psint_t; +typedef __uint64_t __psunsigned_t; +#else +#error BITS_PER_LONG must be 32 or 64 +#endif + #include "xfs_types.h" #include "kmem.h" @@ -114,8 +146,6 @@ #define xfs_inherit_sync xfs_params.inherit_sync.val #define xfs_inherit_nodump xfs_params.inherit_nodump.val #define xfs_inherit_noatime xfs_params.inherit_noatim.val -#define xfs_buf_timer_centisecs xfs_params.xfs_buf_timer.val -#define xfs_buf_age_centisecs xfs_params.xfs_buf_age.val #define xfs_inherit_nosymlinks xfs_params.inherit_nosym.val #define xfs_rotorstep xfs_params.rotorstep.val #define xfs_inherit_nodefrag xfs_params.inherit_nodfrg.val @@ -159,6 +189,32 @@ #define MAX(a,b) (max(a,b)) #define howmany(x, y) (((x)+((y)-1))/(y)) +/* Kernel uid/gid conversion. These are used to convert to/from the on disk + * uid_t/gid_t types to the kuid_t/kgid_t types that the kernel uses internally. + * The conversion here is type only, the value will remain the same since we + * are converting to the init_user_ns. The uid is later mapped to a particular + * user namespace value when crossing the kernel/user boundary. + */ +static inline __uint32_t xfs_kuid_to_uid(kuid_t uid) +{ + return from_kuid(&init_user_ns, uid); +} + +static inline kuid_t xfs_uid_to_kuid(__uint32_t uid) +{ + return make_kuid(&init_user_ns, uid); +} + +static inline __uint32_t xfs_kgid_to_gid(kgid_t gid) +{ + return from_kgid(&init_user_ns, gid); +} + +static inline kgid_t xfs_gid_to_kgid(__uint32_t gid) +{ + return make_kgid(&init_user_ns, gid); +} + /* * Various platform dependent calls that don't fit anywhere else */ diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c index d852a2b..5372d58 100644 --- a/fs/xfs/xfs_log.c +++ b/fs/xfs/xfs_log.c @@ -614,7 +614,8 @@ xfs_log_mount( xfs_daddr_t blk_offset, int num_bblks) { - int error; + int error = 0; + int min_logfsbs; if (!(mp->m_flags & XFS_MOUNT_NORECOVERY)) xfs_notice(mp, "Mounting Filesystem"); @@ -631,6 +632,50 @@ xfs_log_mount( } /* + * Validate the given log space and drop a critical message via syslog + * if the log size is too small that would lead to some unexpected + * situations in transaction log space reservation stage. + * + * Note: we can't just reject the mount if the validation fails. This + * would mean that people would have to downgrade their kernel just to + * remedy the situation as there is no way to grow the log (short of + * black magic surgery with xfs_db). + * + * We can, however, reject mounts for CRC format filesystems, as the + * mkfs binary being used to make the filesystem should never create a + * filesystem with a log that is too small. + */ + min_logfsbs = xfs_log_calc_minimum_size(mp); + + if (mp->m_sb.sb_logblocks < min_logfsbs) { + xfs_warn(mp, + "Log size %d blocks too small, minimum size is %d blocks", + mp->m_sb.sb_logblocks, min_logfsbs); + error = EINVAL; + } else if (mp->m_sb.sb_logblocks > XFS_MAX_LOG_BLOCKS) { + xfs_warn(mp, + "Log size %d blocks too large, maximum size is %lld blocks", + mp->m_sb.sb_logblocks, XFS_MAX_LOG_BLOCKS); + error = EINVAL; + } else if (XFS_FSB_TO_B(mp, mp->m_sb.sb_logblocks) > XFS_MAX_LOG_BYTES) { + xfs_warn(mp, + "log size %lld bytes too large, maximum size is %lld bytes", + XFS_FSB_TO_B(mp, mp->m_sb.sb_logblocks), + XFS_MAX_LOG_BYTES); + error = EINVAL; + } + if (error) { + if (xfs_sb_version_hascrc(&mp->m_sb)) { + xfs_crit(mp, "AAIEEE! Log failed size checks. Abort!"); + ASSERT(0); + goto out_free_log; + } + xfs_crit(mp, +"Log size out of supported range. Continuing onwards, but if log hangs are\n" +"experienced then please report this message in the bug report."); + } + + /* * Initialize the AIL now we have a log. */ error = xfs_trans_ail_init(mp); @@ -720,7 +765,7 @@ xfs_log_mount_finish(xfs_mount_t *mp) * Unmount record used to have a string "Unmount filesystem--" in the * data section where the "Un" was really a magic number (XLOG_UNMOUNT_TYPE). * We just write the magic number now since that particular field isn't - * currently architecture converted and "nUmount" is a bit foo. + * currently architecture converted and "Unmount" is a bit foo. * As far as I know, there weren't any dependencies on the old behaviour. */ @@ -1941,7 +1986,7 @@ xlog_print_tic_res( xfs_alert_tag(mp, XFS_PTAG_LOGRES, "xlog_write: reservation ran out. Need to up reservation"); - xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE); + xfs_force_shutdown(mp, SHUTDOWN_LOG_IO_ERROR); } /* @@ -2044,7 +2089,7 @@ xlog_write_setup_ophdr( * Set up the parameters of the region copy into the log. This has * to handle region write split across multiple log buffers - this * state is kept external to this function so that this code can - * can be written in an obvious, self documenting manner. + * be written in an obvious, self documenting manner. */ static int xlog_write_setup_copy( @@ -3391,24 +3436,17 @@ xfs_log_ticket_get( } /* - * Allocate and initialise a new log ticket. + * Figure out the total log space unit (in bytes) that would be + * required for a log ticket. */ -struct xlog_ticket * -xlog_ticket_alloc( - struct xlog *log, - int unit_bytes, - int cnt, - char client, - bool permanent, - xfs_km_flags_t alloc_flags) +int +xfs_log_calc_unit_res( + struct xfs_mount *mp, + int unit_bytes) { - struct xlog_ticket *tic; - uint num_headers; - int iclog_space; - - tic = kmem_zone_zalloc(xfs_log_ticket_zone, alloc_flags); - if (!tic) - return NULL; + struct xlog *log = mp->m_log; + int iclog_space; + uint num_headers; /* * Permanent reservations have up to 'cnt'-1 active log operations @@ -3483,20 +3521,43 @@ xlog_ticket_alloc( unit_bytes += log->l_iclog_hsize; /* for roundoff padding for transaction data and one for commit record */ - if (xfs_sb_version_haslogv2(&log->l_mp->m_sb) && - log->l_mp->m_sb.sb_logsunit > 1) { + if (xfs_sb_version_haslogv2(&mp->m_sb) && mp->m_sb.sb_logsunit > 1) { /* log su roundoff */ - unit_bytes += 2*log->l_mp->m_sb.sb_logsunit; + unit_bytes += 2 * mp->m_sb.sb_logsunit; } else { /* BB roundoff */ - unit_bytes += 2*BBSIZE; + unit_bytes += 2 * BBSIZE; } + return unit_bytes; +} + +/* + * Allocate and initialise a new log ticket. + */ +struct xlog_ticket * +xlog_ticket_alloc( + struct xlog *log, + int unit_bytes, + int cnt, + char client, + bool permanent, + xfs_km_flags_t alloc_flags) +{ + struct xlog_ticket *tic; + int unit_res; + + tic = kmem_zone_zalloc(xfs_log_ticket_zone, alloc_flags); + if (!tic) + return NULL; + + unit_res = xfs_log_calc_unit_res(log->l_mp, unit_bytes); + atomic_set(&tic->t_ref, 1); tic->t_task = current; INIT_LIST_HEAD(&tic->t_queue); - tic->t_unit_res = unit_bytes; - tic->t_curr_res = unit_bytes; + tic->t_unit_res = unit_res; + tic->t_curr_res = unit_res; tic->t_cnt = cnt; tic->t_ocnt = cnt; tic->t_tid = prandom_u32(); diff --git a/fs/xfs/xfs_log.h b/fs/xfs/xfs_log.h index fb630e4..1c45848 100644 --- a/fs/xfs/xfs_log.h +++ b/fs/xfs/xfs_log.h @@ -18,14 +18,30 @@ #ifndef __XFS_LOG_H__ #define __XFS_LOG_H__ -/* get lsn fields */ -#define CYCLE_LSN(lsn) ((uint)((lsn)>>32)) -#define BLOCK_LSN(lsn) ((uint)(lsn)) +#include "xfs_log_format.h" -/* this is used in a spot where we might otherwise double-endian-flip */ -#define CYCLE_LSN_DISK(lsn) (((__be32 *)&(lsn))[0]) +struct xfs_log_vec { + struct xfs_log_vec *lv_next; /* next lv in build list */ + int lv_niovecs; /* number of iovecs in lv */ + struct xfs_log_iovec *lv_iovecp; /* iovec array */ + struct xfs_log_item *lv_item; /* owner */ + char *lv_buf; /* formatted buffer */ + int lv_buf_len; /* size of formatted buffer */ + int lv_size; /* size of allocated lv */ +}; + +#define XFS_LOG_VEC_ORDERED (-1) + +/* + * Structure used to pass callback function and the function's argument + * to the log manager. + */ +typedef struct xfs_log_callback { + struct xfs_log_callback *cb_next; + void (*cb_func)(void *, int); + void *cb_arg; +} xfs_log_callback_t; -#ifdef __KERNEL__ /* * By comparing each component, we don't have to worry about extra * endian issues in treating two 32 bit numbers as one 64 bit number @@ -59,67 +75,6 @@ static inline xfs_lsn_t _lsn_cmp(xfs_lsn_t lsn1, xfs_lsn_t lsn2) */ #define XFS_LOG_SYNC 0x1 -#endif /* __KERNEL__ */ - - -/* Log Clients */ -#define XFS_TRANSACTION 0x69 -#define XFS_VOLUME 0x2 -#define XFS_LOG 0xaa - - -/* Region types for iovec's i_type */ -#define XLOG_REG_TYPE_BFORMAT 1 -#define XLOG_REG_TYPE_BCHUNK 2 -#define XLOG_REG_TYPE_EFI_FORMAT 3 -#define XLOG_REG_TYPE_EFD_FORMAT 4 -#define XLOG_REG_TYPE_IFORMAT 5 -#define XLOG_REG_TYPE_ICORE 6 -#define XLOG_REG_TYPE_IEXT 7 -#define XLOG_REG_TYPE_IBROOT 8 -#define XLOG_REG_TYPE_ILOCAL 9 -#define XLOG_REG_TYPE_IATTR_EXT 10 -#define XLOG_REG_TYPE_IATTR_BROOT 11 -#define XLOG_REG_TYPE_IATTR_LOCAL 12 -#define XLOG_REG_TYPE_QFORMAT 13 -#define XLOG_REG_TYPE_DQUOT 14 -#define XLOG_REG_TYPE_QUOTAOFF 15 -#define XLOG_REG_TYPE_LRHEADER 16 -#define XLOG_REG_TYPE_UNMOUNT 17 -#define XLOG_REG_TYPE_COMMIT 18 -#define XLOG_REG_TYPE_TRANSHDR 19 -#define XLOG_REG_TYPE_ICREATE 20 -#define XLOG_REG_TYPE_MAX 20 - -typedef struct xfs_log_iovec { - void *i_addr; /* beginning address of region */ - int i_len; /* length in bytes of region */ - uint i_type; /* type of region */ -} xfs_log_iovec_t; - -struct xfs_log_vec { - struct xfs_log_vec *lv_next; /* next lv in build list */ - int lv_niovecs; /* number of iovecs in lv */ - struct xfs_log_iovec *lv_iovecp; /* iovec array */ - struct xfs_log_item *lv_item; /* owner */ - char *lv_buf; /* formatted buffer */ - int lv_buf_len; /* size of formatted buffer */ -}; - -#define XFS_LOG_VEC_ORDERED (-1) - -/* - * Structure used to pass callback function and the function's argument - * to the log manager. - */ -typedef struct xfs_log_callback { - struct xfs_log_callback *cb_next; - void (*cb_func)(void *, int); - void *cb_arg; -} xfs_log_callback_t; - - -#ifdef __KERNEL__ /* Log manager interfaces */ struct xfs_mount; struct xlog_in_core; @@ -188,5 +143,4 @@ void xfs_log_work_queue(struct xfs_mount *mp); void xfs_log_worker(struct work_struct *work); void xfs_log_quiesce(struct xfs_mount *mp); -#endif #endif /* __XFS_LOG_H__ */ diff --git a/fs/xfs/xfs_log_cil.c b/fs/xfs/xfs_log_cil.c index 02b9cf3..cfe9797 100644 --- a/fs/xfs/xfs_log_cil.c +++ b/fs/xfs/xfs_log_cil.c @@ -80,6 +80,83 @@ xlog_cil_init_post_recovery( log->l_curr_block); } +STATIC int +xlog_cil_lv_item_format( + struct xfs_log_item *lip, + struct xfs_log_vec *lv) +{ + int index; + char *ptr; + + /* format new vectors into array */ + lip->li_ops->iop_format(lip, lv->lv_iovecp); + + /* copy data into existing array */ + ptr = lv->lv_buf; + for (index = 0; index < lv->lv_niovecs; index++) { + struct xfs_log_iovec *vec = &lv->lv_iovecp[index]; + + memcpy(ptr, vec->i_addr, vec->i_len); + vec->i_addr = ptr; + ptr += vec->i_len; + } + + /* + * some size calculations for log vectors over-estimate, so the caller + * doesn't know the amount of space actually used by the item. Return + * the byte count to the caller so they can check and store it + * appropriately. + */ + return ptr - lv->lv_buf; +} + +/* + * Prepare the log item for insertion into the CIL. Calculate the difference in + * log space and vectors it will consume, and if it is a new item pin it as + * well. + */ +STATIC void +xfs_cil_prepare_item( + struct xlog *log, + struct xfs_log_vec *lv, + struct xfs_log_vec *old_lv, + int *diff_len, + int *diff_iovecs) +{ + /* Account for the new LV being passed in */ + if (lv->lv_buf_len != XFS_LOG_VEC_ORDERED) { + *diff_len += lv->lv_buf_len; + *diff_iovecs += lv->lv_niovecs; + } + + /* + * If there is no old LV, this is the first time we've seen the item in + * this CIL context and so we need to pin it. If we are replacing the + * old_lv, then remove the space it accounts for and free it. + */ + if (!old_lv) + lv->lv_item->li_ops->iop_pin(lv->lv_item); + else if (old_lv != lv) { + ASSERT(lv->lv_buf_len != XFS_LOG_VEC_ORDERED); + + *diff_len -= old_lv->lv_buf_len; + *diff_iovecs -= old_lv->lv_niovecs; + kmem_free(old_lv); + } + + /* attach new log vector to log item */ + lv->lv_item->li_lv = lv; + + /* + * If this is the first time the item is being committed to the + * CIL, store the sequence number on the log item so we can + * tell in future commits whether this is the first checkpoint + * the item is being committed into. + */ + if (!lv->lv_item->li_seq) + lv->lv_item->li_seq = log->l_cilp->xc_ctx->sequence; +} + /* * Format log item into a flat buffers * @@ -106,35 +183,39 @@ xlog_cil_init_post_recovery( * format the regions into the iclog as though they are being formatted * directly out of the objects themselves. */ -static struct xfs_log_vec * -xlog_cil_prepare_log_vecs( - struct xfs_trans *tp) +static void +xlog_cil_insert_format_items( + struct xlog *log, + struct xfs_trans *tp, + int *diff_len, + int *diff_iovecs) { struct xfs_log_item_desc *lidp; - struct xfs_log_vec *lv = NULL; - struct xfs_log_vec *ret_lv = NULL; /* Bail out if we didn't find a log item. */ if (list_empty(&tp->t_items)) { ASSERT(0); - return NULL; + return; } list_for_each_entry(lidp, &tp->t_items, lid_trans) { - struct xfs_log_vec *new_lv; - void *ptr; - int index; - int len = 0; - uint niovecs; + struct xfs_log_item *lip = lidp->lid_item; + struct xfs_log_vec *lv; + struct xfs_log_vec *old_lv; + int niovecs = 0; + int nbytes = 0; + int buf_size; bool ordered = false; /* Skip items which aren't dirty in this transaction. */ if (!(lidp->lid_flags & XFS_LID_DIRTY)) continue; + /* get number of vecs and size of data to be stored */ + lip->li_ops->iop_size(lip, &niovecs, &nbytes); + /* Skip items that do not have any vectors for writing */ - niovecs = IOP_SIZE(lidp->lid_item); if (!niovecs) continue; @@ -146,109 +227,63 @@ xlog_cil_prepare_log_vecs( if (niovecs == XFS_LOG_VEC_ORDERED) { ordered = true; niovecs = 0; + nbytes = 0; } - new_lv = kmem_zalloc(sizeof(*new_lv) + - niovecs * sizeof(struct xfs_log_iovec), - KM_SLEEP|KM_NOFS); - - new_lv->lv_item = lidp->lid_item; - new_lv->lv_niovecs = niovecs; - if (ordered) { - /* track as an ordered logvec */ - new_lv->lv_buf_len = XFS_LOG_VEC_ORDERED; - goto next; - } - - /* The allocated iovec region lies beyond the log vector. */ - new_lv->lv_iovecp = (struct xfs_log_iovec *)&new_lv[1]; + /* grab the old item if it exists for reservation accounting */ + old_lv = lip->li_lv; - /* build the vector array and calculate it's length */ - IOP_FORMAT(new_lv->lv_item, new_lv->lv_iovecp); - for (index = 0; index < new_lv->lv_niovecs; index++) - len += new_lv->lv_iovecp[index].i_len; + /* calc buffer size */ + buf_size = sizeof(struct xfs_log_vec) + nbytes + + niovecs * sizeof(struct xfs_log_iovec); - new_lv->lv_buf_len = len; - new_lv->lv_buf = kmem_alloc(new_lv->lv_buf_len, - KM_SLEEP|KM_NOFS); - ptr = new_lv->lv_buf; + /* compare to existing item size */ + if (lip->li_lv && buf_size <= lip->li_lv->lv_size) { + /* same or smaller, optimise common overwrite case */ + lv = lip->li_lv; + lv->lv_next = NULL; - for (index = 0; index < new_lv->lv_niovecs; index++) { - struct xfs_log_iovec *vec = &new_lv->lv_iovecp[index]; + if (ordered) + goto insert; - memcpy(ptr, vec->i_addr, vec->i_len); - vec->i_addr = ptr; - ptr += vec->i_len; - } - ASSERT(ptr == new_lv->lv_buf + new_lv->lv_buf_len); - -next: - if (!ret_lv) - ret_lv = new_lv; - else - lv->lv_next = new_lv; - lv = new_lv; - } - - return ret_lv; -} - -/* - * Prepare the log item for insertion into the CIL. Calculate the difference in - * log space and vectors it will consume, and if it is a new item pin it as - * well. - */ -STATIC void -xfs_cil_prepare_item( - struct xlog *log, - struct xfs_log_vec *lv, - int *len, - int *diff_iovecs) -{ - struct xfs_log_vec *old = lv->lv_item->li_lv; + /* + * set the item up as though it is a new insertion so + * that the space reservation accounting is correct. + */ + *diff_iovecs -= lv->lv_niovecs; + *diff_len -= lv->lv_buf_len; - if (old) { - /* existing lv on log item, space used is a delta */ - ASSERT((old->lv_buf && old->lv_buf_len && old->lv_niovecs) || - old->lv_buf_len == XFS_LOG_VEC_ORDERED); + /* Ensure the lv is set up according to ->iop_size */ + lv->lv_niovecs = niovecs; + lv->lv_buf = (char *)lv + buf_size - nbytes; - /* - * If the new item is ordered, keep the old one that is already - * tracking dirty or ordered regions - */ - if (lv->lv_buf_len == XFS_LOG_VEC_ORDERED) { - ASSERT(!lv->lv_buf); - kmem_free(lv); - return; + lv->lv_buf_len = xlog_cil_lv_item_format(lip, lv); + goto insert; } - *len += lv->lv_buf_len - old->lv_buf_len; - *diff_iovecs += lv->lv_niovecs - old->lv_niovecs; - kmem_free(old->lv_buf); - kmem_free(old); - } else { - /* new lv, must pin the log item */ - ASSERT(!lv->lv_item->li_lv); - - if (lv->lv_buf_len != XFS_LOG_VEC_ORDERED) { - *len += lv->lv_buf_len; - *diff_iovecs += lv->lv_niovecs; + /* allocate new data chunk */ + lv = kmem_zalloc(buf_size, KM_SLEEP|KM_NOFS); + lv->lv_item = lip; + lv->lv_size = buf_size; + lv->lv_niovecs = niovecs; + if (ordered) { + /* track as an ordered logvec */ + ASSERT(lip->li_lv == NULL); + lv->lv_buf_len = XFS_LOG_VEC_ORDERED; + goto insert; } - IOP_PIN(lv->lv_item); - } + /* The allocated iovec region lies beyond the log vector. */ + lv->lv_iovecp = (struct xfs_log_iovec *)&lv[1]; - /* attach new log vector to log item */ - lv->lv_item->li_lv = lv; + /* The allocated data region lies beyond the iovec region */ + lv->lv_buf = (char *)lv + buf_size - nbytes; - /* - * If this is the first time the item is being committed to the - * CIL, store the sequence number on the log item so we can - * tell in future commits whether this is the first checkpoint - * the item is being committed into. - */ - if (!lv->lv_item->li_seq) - lv->lv_item->li_seq = log->l_cilp->xc_ctx->sequence; + lv->lv_buf_len = xlog_cil_lv_item_format(lip, lv); +insert: + ASSERT(lv->lv_buf_len <= nbytes); + xfs_cil_prepare_item(log, lv, old_lv, diff_len, diff_iovecs); + } } /* @@ -261,53 +296,47 @@ xfs_cil_prepare_item( static void xlog_cil_insert_items( struct xlog *log, - struct xfs_log_vec *log_vector, - struct xlog_ticket *ticket) + struct xfs_trans *tp) { struct xfs_cil *cil = log->l_cilp; struct xfs_cil_ctx *ctx = cil->xc_ctx; - struct xfs_log_vec *lv; + struct xfs_log_item_desc *lidp; int len = 0; int diff_iovecs = 0; int iclog_space; - ASSERT(log_vector); + ASSERT(tp); /* - * Do all the accounting aggregation and switching of log vectors - * around in a separate loop to the insertion of items into the CIL. - * Then we can do a separate loop to update the CIL within a single - * lock/unlock pair. This reduces the number of round trips on the CIL - * lock from O(nr_logvectors) to O(1) and greatly reduces the overall - * hold time for the transaction commit. - * - * If this is the first time the item is being placed into the CIL in - * this context, pin it so it can't be written to disk until the CIL is - * flushed to the iclog and the iclog written to disk. - * * We can do this safely because the context can't checkpoint until we * are done so it doesn't matter exactly how we update the CIL. */ + xlog_cil_insert_format_items(log, tp, &len, &diff_iovecs); + + /* + * Now (re-)position everything modified at the tail of the CIL. + * We do this here so we only need to take the CIL lock once during + * the transaction commit. + */ spin_lock(&cil->xc_cil_lock); - for (lv = log_vector; lv; ) { - struct xfs_log_vec *next = lv->lv_next; + list_for_each_entry(lidp, &tp->t_items, lid_trans) { + struct xfs_log_item *lip = lidp->lid_item; - ASSERT(lv->lv_item->li_lv || list_empty(&lv->lv_item->li_cil)); - lv->lv_next = NULL; + /* Skip items which aren't dirty in this transaction. */ + if (!(lidp->lid_flags & XFS_LID_DIRTY)) + continue; - /* - * xfs_cil_prepare_item() may free the lv, so move the item on - * the CIL first. - */ - list_move_tail(&lv->lv_item->li_cil, &cil->xc_cil); - xfs_cil_prepare_item(log, lv, &len, &diff_iovecs); - lv = next; + list_move_tail(&lip->li_cil, &cil->xc_cil); } /* account for space used by new iovec headers */ len += diff_iovecs * sizeof(xlog_op_header_t); ctx->nvecs += diff_iovecs; + /* attach the transaction to the CIL if it has any busy extents */ + if (!list_empty(&tp->t_busy)) + list_splice_init(&tp->t_busy, &ctx->busy_extents); + /* * Now transfer enough transaction reservation to the context ticket * for the checkpoint. The context ticket is special - the unit @@ -316,10 +345,8 @@ xlog_cil_insert_items( * during the transaction commit. */ if (ctx->ticket->t_curr_res == 0) { - /* first commit in checkpoint, steal the header reservation */ - ASSERT(ticket->t_curr_res >= ctx->ticket->t_unit_res + len); ctx->ticket->t_curr_res = ctx->ticket->t_unit_res; - ticket->t_curr_res -= ctx->ticket->t_unit_res; + tp->t_ticket->t_curr_res -= ctx->ticket->t_unit_res; } /* do we need space for more log record headers? */ @@ -333,10 +360,10 @@ xlog_cil_insert_items( hdrs *= log->l_iclog_hsize + sizeof(struct xlog_op_header); ctx->ticket->t_unit_res += hdrs; ctx->ticket->t_curr_res += hdrs; - ticket->t_curr_res -= hdrs; - ASSERT(ticket->t_curr_res >= len); + tp->t_ticket->t_curr_res -= hdrs; + ASSERT(tp->t_ticket->t_curr_res >= len); } - ticket->t_curr_res -= len; + tp->t_ticket->t_curr_res -= len; ctx->space_used += len; spin_unlock(&cil->xc_cil_lock); @@ -350,7 +377,6 @@ xlog_cil_free_logvec( for (lv = log_vector; lv; ) { struct xfs_log_vec *next = lv->lv_next; - kmem_free(lv->lv_buf); kmem_free(lv); lv = next; } @@ -376,9 +402,9 @@ xlog_cil_committed( xfs_extent_busy_clear(mp, &ctx->busy_extents, (mp->m_flags & XFS_MOUNT_DISCARD) && !abort); - spin_lock(&ctx->cil->xc_cil_lock); + spin_lock(&ctx->cil->xc_push_lock); list_del(&ctx->committing); - spin_unlock(&ctx->cil->xc_cil_lock); + spin_unlock(&ctx->cil->xc_push_lock); xlog_cil_free_logvec(ctx->lv_chain); @@ -433,7 +459,7 @@ xlog_cil_push( down_write(&cil->xc_ctx_lock); ctx = cil->xc_ctx; - spin_lock(&cil->xc_cil_lock); + spin_lock(&cil->xc_push_lock); push_seq = cil->xc_push_seq; ASSERT(push_seq <= ctx->sequence); @@ -444,10 +470,10 @@ xlog_cil_push( */ if (list_empty(&cil->xc_cil)) { cil->xc_push_seq = 0; - spin_unlock(&cil->xc_cil_lock); + spin_unlock(&cil->xc_push_lock); goto out_skip; } - spin_unlock(&cil->xc_cil_lock); + spin_unlock(&cil->xc_push_lock); /* check for a previously pushed seqeunce */ @@ -515,9 +541,9 @@ xlog_cil_push( * that higher sequences will wait for us to write out a commit record * before they do. */ - spin_lock(&cil->xc_cil_lock); + spin_lock(&cil->xc_push_lock); list_add(&ctx->committing, &cil->xc_committing); - spin_unlock(&cil->xc_cil_lock); + spin_unlock(&cil->xc_push_lock); up_write(&cil->xc_ctx_lock); /* @@ -552,7 +578,7 @@ xlog_cil_push( * order the commit records so replay will get them in the right order. */ restart: - spin_lock(&cil->xc_cil_lock); + spin_lock(&cil->xc_push_lock); list_for_each_entry(new_ctx, &cil->xc_committing, committing) { /* * Higher sequences will wait for this one so skip them. @@ -565,11 +591,11 @@ restart: * It is still being pushed! Wait for the push to * complete, then start again from the beginning. */ - xlog_wait(&cil->xc_commit_wait, &cil->xc_cil_lock); + xlog_wait(&cil->xc_commit_wait, &cil->xc_push_lock); goto restart; } } - spin_unlock(&cil->xc_cil_lock); + spin_unlock(&cil->xc_push_lock); /* xfs_log_done always frees the ticket on error. */ commit_lsn = xfs_log_done(log->l_mp, tic, &commit_iclog, 0); @@ -588,10 +614,10 @@ restart: * callbacks to the iclog we can assign the commit LSN to the context * and wake up anyone who is waiting for the commit to complete. */ - spin_lock(&cil->xc_cil_lock); + spin_lock(&cil->xc_push_lock); ctx->commit_lsn = commit_lsn; wake_up_all(&cil->xc_commit_wait); - spin_unlock(&cil->xc_cil_lock); + spin_unlock(&cil->xc_push_lock); /* release the hounds! */ return xfs_log_release_iclog(log->l_mp, commit_iclog); @@ -644,12 +670,12 @@ xlog_cil_push_background( if (cil->xc_ctx->space_used < XLOG_CIL_SPACE_LIMIT(log)) return; - spin_lock(&cil->xc_cil_lock); + spin_lock(&cil->xc_push_lock); if (cil->xc_push_seq < cil->xc_current_sequence) { cil->xc_push_seq = cil->xc_current_sequence; queue_work(log->l_mp->m_cil_workqueue, &cil->xc_push_work); } - spin_unlock(&cil->xc_cil_lock); + spin_unlock(&cil->xc_push_lock); } @@ -672,14 +698,14 @@ xlog_cil_push_foreground( * If the CIL is empty or we've already pushed the sequence then * there's no work we need to do. */ - spin_lock(&cil->xc_cil_lock); + spin_lock(&cil->xc_push_lock); if (list_empty(&cil->xc_cil) || push_seq <= cil->xc_push_seq) { - spin_unlock(&cil->xc_cil_lock); + spin_unlock(&cil->xc_push_lock); return; } cil->xc_push_seq = push_seq; - spin_unlock(&cil->xc_cil_lock); + spin_unlock(&cil->xc_push_lock); /* do the push now */ xlog_cil_push(log); @@ -706,43 +732,25 @@ xfs_log_commit_cil( int flags) { struct xlog *log = mp->m_log; + struct xfs_cil *cil = log->l_cilp; int log_flags = 0; - struct xfs_log_vec *log_vector; if (flags & XFS_TRANS_RELEASE_LOG_RES) log_flags = XFS_LOG_REL_PERM_RESERV; - /* - * Do all the hard work of formatting items (including memory - * allocation) outside the CIL context lock. This prevents stalling CIL - * pushes when we are low on memory and a transaction commit spends a - * lot of time in memory reclaim. - */ - log_vector = xlog_cil_prepare_log_vecs(tp); - if (!log_vector) - return ENOMEM; - /* lock out background commit */ - down_read(&log->l_cilp->xc_ctx_lock); - if (commit_lsn) - *commit_lsn = log->l_cilp->xc_ctx->sequence; + down_read(&cil->xc_ctx_lock); - /* xlog_cil_insert_items() destroys log_vector list */ - xlog_cil_insert_items(log, log_vector, tp->t_ticket); + xlog_cil_insert_items(log, tp); /* check we didn't blow the reservation */ if (tp->t_ticket->t_curr_res < 0) - xlog_print_tic_res(log->l_mp, tp->t_ticket); + xlog_print_tic_res(mp, tp->t_ticket); - /* attach the transaction to the CIL if it has any busy extents */ - if (!list_empty(&tp->t_busy)) { - spin_lock(&log->l_cilp->xc_cil_lock); - list_splice_init(&tp->t_busy, - &log->l_cilp->xc_ctx->busy_extents); - spin_unlock(&log->l_cilp->xc_cil_lock); - } + tp->t_commit_lsn = cil->xc_ctx->sequence; + if (commit_lsn) + *commit_lsn = tp->t_commit_lsn; - tp->t_commit_lsn = *commit_lsn; xfs_log_done(mp, tp->t_ticket, NULL, log_flags); xfs_trans_unreserve_and_mod_sb(tp); @@ -757,11 +765,11 @@ xfs_log_commit_cil( * the log items. This affects (at least) processing of stale buffers, * inodes and EFIs. */ - xfs_trans_free_items(tp, *commit_lsn, 0); + xfs_trans_free_items(tp, tp->t_commit_lsn, 0); xlog_cil_push_background(log); - up_read(&log->l_cilp->xc_ctx_lock); + up_read(&cil->xc_ctx_lock); return 0; } @@ -800,7 +808,7 @@ xlog_cil_force_lsn( * on commits for those as well. */ restart: - spin_lock(&cil->xc_cil_lock); + spin_lock(&cil->xc_push_lock); list_for_each_entry(ctx, &cil->xc_committing, committing) { if (ctx->sequence > sequence) continue; @@ -809,7 +817,7 @@ restart: * It is still being pushed! Wait for the push to * complete, then start again from the beginning. */ - xlog_wait(&cil->xc_commit_wait, &cil->xc_cil_lock); + xlog_wait(&cil->xc_commit_wait, &cil->xc_push_lock); goto restart; } if (ctx->sequence != sequence) @@ -817,7 +825,7 @@ restart: /* found it! */ commit_lsn = ctx->commit_lsn; } - spin_unlock(&cil->xc_cil_lock); + spin_unlock(&cil->xc_push_lock); return commit_lsn; } @@ -875,6 +883,7 @@ xlog_cil_init( INIT_LIST_HEAD(&cil->xc_cil); INIT_LIST_HEAD(&cil->xc_committing); spin_lock_init(&cil->xc_cil_lock); + spin_lock_init(&cil->xc_push_lock); init_rwsem(&cil->xc_ctx_lock); init_waitqueue_head(&cil->xc_commit_wait); diff --git a/fs/xfs/xfs_log_format.h b/fs/xfs/xfs_log_format.h new file mode 100644 index 0000000..31e3a06 --- /dev/null +++ b/fs/xfs/xfs_log_format.h @@ -0,0 +1,852 @@ +/* + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#ifndef __XFS_LOG_FORMAT_H__ +#define __XFS_LOG_FORMAT_H__ + +struct xfs_mount; +struct xfs_trans_res; + +/* + * On-disk Log Format definitions. + * + * This file contains all the on-disk format definitions used within the log. It + * includes the physical log structure itself, as well as all the log item + * format structures that are written into the log and intepreted by log + * recovery. We start with the physical log format definitions, and then work + * through all the log items definitions and everything they encode into the + * log. + */ +typedef __uint32_t xlog_tid_t; + +#define XLOG_MIN_ICLOGS 2 +#define XLOG_MAX_ICLOGS 8 +#define XLOG_HEADER_MAGIC_NUM 0xFEEDbabe /* Invalid cycle number */ +#define XLOG_VERSION_1 1 +#define XLOG_VERSION_2 2 /* Large IClogs, Log sunit */ +#define XLOG_VERSION_OKBITS (XLOG_VERSION_1 | XLOG_VERSION_2) +#define XLOG_MIN_RECORD_BSIZE (16*1024) /* eventually 32k */ +#define XLOG_BIG_RECORD_BSIZE (32*1024) /* 32k buffers */ +#define XLOG_MAX_RECORD_BSIZE (256*1024) +#define XLOG_HEADER_CYCLE_SIZE (32*1024) /* cycle data in header */ +#define XLOG_MIN_RECORD_BSHIFT 14 /* 16384 == 1 << 14 */ +#define XLOG_BIG_RECORD_BSHIFT 15 /* 32k == 1 << 15 */ +#define XLOG_MAX_RECORD_BSHIFT 18 /* 256k == 1 << 18 */ +#define XLOG_BTOLSUNIT(log, b) (((b)+(log)->l_mp->m_sb.sb_logsunit-1) / \ + (log)->l_mp->m_sb.sb_logsunit) +#define XLOG_LSUNITTOB(log, su) ((su) * (log)->l_mp->m_sb.sb_logsunit) + +#define XLOG_HEADER_SIZE 512 + +/* Minimum number of transactions that must fit in the log (defined by mkfs) */ +#define XFS_MIN_LOG_FACTOR 3 + +#define XLOG_REC_SHIFT(log) \ + BTOBB(1 << (xfs_sb_version_haslogv2(&log->l_mp->m_sb) ? \ + XLOG_MAX_RECORD_BSHIFT : XLOG_BIG_RECORD_BSHIFT)) +#define XLOG_TOTAL_REC_SHIFT(log) \ + BTOBB(XLOG_MAX_ICLOGS << (xfs_sb_version_haslogv2(&log->l_mp->m_sb) ? \ + XLOG_MAX_RECORD_BSHIFT : XLOG_BIG_RECORD_BSHIFT)) + +/* get lsn fields */ +#define CYCLE_LSN(lsn) ((uint)((lsn)>>32)) +#define BLOCK_LSN(lsn) ((uint)(lsn)) + +/* this is used in a spot where we might otherwise double-endian-flip */ +#define CYCLE_LSN_DISK(lsn) (((__be32 *)&(lsn))[0]) + +static inline xfs_lsn_t xlog_assign_lsn(uint cycle, uint block) +{ + return ((xfs_lsn_t)cycle << 32) | block; +} + +static inline uint xlog_get_cycle(char *ptr) +{ + if (be32_to_cpu(*(__be32 *)ptr) == XLOG_HEADER_MAGIC_NUM) + return be32_to_cpu(*((__be32 *)ptr + 1)); + else + return be32_to_cpu(*(__be32 *)ptr); +} + +/* Log Clients */ +#define XFS_TRANSACTION 0x69 +#define XFS_VOLUME 0x2 +#define XFS_LOG 0xaa + +#define XLOG_UNMOUNT_TYPE 0x556e /* Un for Unmount */ + +/* Region types for iovec's i_type */ +#define XLOG_REG_TYPE_BFORMAT 1 +#define XLOG_REG_TYPE_BCHUNK 2 +#define XLOG_REG_TYPE_EFI_FORMAT 3 +#define XLOG_REG_TYPE_EFD_FORMAT 4 +#define XLOG_REG_TYPE_IFORMAT 5 +#define XLOG_REG_TYPE_ICORE 6 +#define XLOG_REG_TYPE_IEXT 7 +#define XLOG_REG_TYPE_IBROOT 8 +#define XLOG_REG_TYPE_ILOCAL 9 +#define XLOG_REG_TYPE_IATTR_EXT 10 +#define XLOG_REG_TYPE_IATTR_BROOT 11 +#define XLOG_REG_TYPE_IATTR_LOCAL 12 +#define XLOG_REG_TYPE_QFORMAT 13 +#define XLOG_REG_TYPE_DQUOT 14 +#define XLOG_REG_TYPE_QUOTAOFF 15 +#define XLOG_REG_TYPE_LRHEADER 16 +#define XLOG_REG_TYPE_UNMOUNT 17 +#define XLOG_REG_TYPE_COMMIT 18 +#define XLOG_REG_TYPE_TRANSHDR 19 +#define XLOG_REG_TYPE_ICREATE 20 +#define XLOG_REG_TYPE_MAX 20 + +/* + * Flags to log operation header + * + * The first write of a new transaction will be preceded with a start + * record, XLOG_START_TRANS. Once a transaction is committed, a commit + * record is written, XLOG_COMMIT_TRANS. If a single region can not fit into + * the remainder of the current active in-core log, it is split up into + * multiple regions. Each partial region will be marked with a + * XLOG_CONTINUE_TRANS until the last one, which gets marked with XLOG_END_TRANS. + * + */ +#define XLOG_START_TRANS 0x01 /* Start a new transaction */ +#define XLOG_COMMIT_TRANS 0x02 /* Commit this transaction */ +#define XLOG_CONTINUE_TRANS 0x04 /* Cont this trans into new region */ +#define XLOG_WAS_CONT_TRANS 0x08 /* Cont this trans into new region */ +#define XLOG_END_TRANS 0x10 /* End a continued transaction */ +#define XLOG_UNMOUNT_TRANS 0x20 /* Unmount a filesystem transaction */ + + +typedef struct xlog_op_header { + __be32 oh_tid; /* transaction id of operation : 4 b */ + __be32 oh_len; /* bytes in data region : 4 b */ + __u8 oh_clientid; /* who sent me this : 1 b */ + __u8 oh_flags; /* : 1 b */ + __u16 oh_res2; /* 32 bit align : 2 b */ +} xlog_op_header_t; + +/* valid values for h_fmt */ +#define XLOG_FMT_UNKNOWN 0 +#define XLOG_FMT_LINUX_LE 1 +#define XLOG_FMT_LINUX_BE 2 +#define XLOG_FMT_IRIX_BE 3 + +/* our fmt */ +#ifdef XFS_NATIVE_HOST +#define XLOG_FMT XLOG_FMT_LINUX_BE +#else +#define XLOG_FMT XLOG_FMT_LINUX_LE +#endif + +typedef struct xlog_rec_header { + __be32 h_magicno; /* log record (LR) identifier : 4 */ + __be32 h_cycle; /* write cycle of log : 4 */ + __be32 h_version; /* LR version : 4 */ + __be32 h_len; /* len in bytes; should be 64-bit aligned: 4 */ + __be64 h_lsn; /* lsn of this LR : 8 */ + __be64 h_tail_lsn; /* lsn of 1st LR w/ buffers not committed: 8 */ + __le32 h_crc; /* crc of log record : 4 */ + __be32 h_prev_block; /* block number to previous LR : 4 */ + __be32 h_num_logops; /* number of log operations in this LR : 4 */ + __be32 h_cycle_data[XLOG_HEADER_CYCLE_SIZE / BBSIZE]; + /* new fields */ + __be32 h_fmt; /* format of log record : 4 */ + uuid_t h_fs_uuid; /* uuid of FS : 16 */ + __be32 h_size; /* iclog size : 4 */ +} xlog_rec_header_t; + +typedef struct xlog_rec_ext_header { + __be32 xh_cycle; /* write cycle of log : 4 */ + __be32 xh_cycle_data[XLOG_HEADER_CYCLE_SIZE / BBSIZE]; /* : 256 */ +} xlog_rec_ext_header_t; + +/* + * Quite misnamed, because this union lays out the actual on-disk log buffer. + */ +typedef union xlog_in_core2 { + xlog_rec_header_t hic_header; + xlog_rec_ext_header_t hic_xheader; + char hic_sector[XLOG_HEADER_SIZE]; +} xlog_in_core_2_t; + +/* not an on-disk structure, but needed by log recovery in userspace */ +typedef struct xfs_log_iovec { + void *i_addr; /* beginning address of region */ + int i_len; /* length in bytes of region */ + uint i_type; /* type of region */ +} xfs_log_iovec_t; + + +/* + * Transaction Header definitions. + * + * This is the structure written in the log at the head of every transaction. It + * identifies the type and id of the transaction, and contains the number of + * items logged by the transaction so we know how many to expect during + * recovery. + * + * Do not change the below structure without redoing the code in + * xlog_recover_add_to_trans() and xlog_recover_add_to_cont_trans(). + */ +typedef struct xfs_trans_header { + uint th_magic; /* magic number */ + uint th_type; /* transaction type */ + __int32_t th_tid; /* transaction id (unused) */ + uint th_num_items; /* num items logged by trans */ +} xfs_trans_header_t; + +#define XFS_TRANS_HEADER_MAGIC 0x5452414e /* TRAN */ + +/* + * Log item types. + */ +#define XFS_LI_EFI 0x1236 +#define XFS_LI_EFD 0x1237 +#define XFS_LI_IUNLINK 0x1238 +#define XFS_LI_INODE 0x123b /* aligned ino chunks, var-size ibufs */ +#define XFS_LI_BUF 0x123c /* v2 bufs, variable sized inode bufs */ +#define XFS_LI_DQUOT 0x123d +#define XFS_LI_QUOTAOFF 0x123e +#define XFS_LI_ICREATE 0x123f + +#define XFS_LI_TYPE_DESC \ + { XFS_LI_EFI, "XFS_LI_EFI" }, \ + { XFS_LI_EFD, "XFS_LI_EFD" }, \ + { XFS_LI_IUNLINK, "XFS_LI_IUNLINK" }, \ + { XFS_LI_INODE, "XFS_LI_INODE" }, \ + { XFS_LI_BUF, "XFS_LI_BUF" }, \ + { XFS_LI_DQUOT, "XFS_LI_DQUOT" }, \ + { XFS_LI_QUOTAOFF, "XFS_LI_QUOTAOFF" }, \ + { XFS_LI_ICREATE, "XFS_LI_ICREATE" } + +/* + * Transaction types. Used to distinguish types of buffers. + */ +#define XFS_TRANS_SETATTR_NOT_SIZE 1 +#define XFS_TRANS_SETATTR_SIZE 2 +#define XFS_TRANS_INACTIVE 3 +#define XFS_TRANS_CREATE 4 +#define XFS_TRANS_CREATE_TRUNC 5 +#define XFS_TRANS_TRUNCATE_FILE 6 +#define XFS_TRANS_REMOVE 7 +#define XFS_TRANS_LINK 8 +#define XFS_TRANS_RENAME 9 +#define XFS_TRANS_MKDIR 10 +#define XFS_TRANS_RMDIR 11 +#define XFS_TRANS_SYMLINK 12 +#define XFS_TRANS_SET_DMATTRS 13 +#define XFS_TRANS_GROWFS 14 +#define XFS_TRANS_STRAT_WRITE 15 +#define XFS_TRANS_DIOSTRAT 16 +/* 17 was XFS_TRANS_WRITE_SYNC */ +#define XFS_TRANS_WRITEID 18 +#define XFS_TRANS_ADDAFORK 19 +#define XFS_TRANS_ATTRINVAL 20 +#define XFS_TRANS_ATRUNCATE 21 +#define XFS_TRANS_ATTR_SET 22 +#define XFS_TRANS_ATTR_RM 23 +#define XFS_TRANS_ATTR_FLAG 24 +#define XFS_TRANS_CLEAR_AGI_BUCKET 25 +#define XFS_TRANS_QM_SBCHANGE 26 +/* + * Dummy entries since we use the transaction type to index into the + * trans_type[] in xlog_recover_print_trans_head() + */ +#define XFS_TRANS_DUMMY1 27 +#define XFS_TRANS_DUMMY2 28 +#define XFS_TRANS_QM_QUOTAOFF 29 +#define XFS_TRANS_QM_DQALLOC 30 +#define XFS_TRANS_QM_SETQLIM 31 +#define XFS_TRANS_QM_DQCLUSTER 32 +#define XFS_TRANS_QM_QINOCREATE 33 +#define XFS_TRANS_QM_QUOTAOFF_END 34 +#define XFS_TRANS_SB_UNIT 35 +#define XFS_TRANS_FSYNC_TS 36 +#define XFS_TRANS_GROWFSRT_ALLOC 37 +#define XFS_TRANS_GROWFSRT_ZERO 38 +#define XFS_TRANS_GROWFSRT_FREE 39 +#define XFS_TRANS_SWAPEXT 40 +#define XFS_TRANS_SB_COUNT 41 +#define XFS_TRANS_CHECKPOINT 42 +#define XFS_TRANS_ICREATE 43 +#define XFS_TRANS_TYPE_MAX 43 +/* new transaction types need to be reflected in xfs_logprint(8) */ + +#define XFS_TRANS_TYPES \ + { XFS_TRANS_SETATTR_NOT_SIZE, "SETATTR_NOT_SIZE" }, \ + { XFS_TRANS_SETATTR_SIZE, "SETATTR_SIZE" }, \ + { XFS_TRANS_INACTIVE, "INACTIVE" }, \ + { XFS_TRANS_CREATE, "CREATE" }, \ + { XFS_TRANS_CREATE_TRUNC, "CREATE_TRUNC" }, \ + { XFS_TRANS_TRUNCATE_FILE, "TRUNCATE_FILE" }, \ + { XFS_TRANS_REMOVE, "REMOVE" }, \ + { XFS_TRANS_LINK, "LINK" }, \ + { XFS_TRANS_RENAME, "RENAME" }, \ + { XFS_TRANS_MKDIR, "MKDIR" }, \ + { XFS_TRANS_RMDIR, "RMDIR" }, \ + { XFS_TRANS_SYMLINK, "SYMLINK" }, \ + { XFS_TRANS_SET_DMATTRS, "SET_DMATTRS" }, \ + { XFS_TRANS_GROWFS, "GROWFS" }, \ + { XFS_TRANS_STRAT_WRITE, "STRAT_WRITE" }, \ + { XFS_TRANS_DIOSTRAT, "DIOSTRAT" }, \ + { XFS_TRANS_WRITEID, "WRITEID" }, \ + { XFS_TRANS_ADDAFORK, "ADDAFORK" }, \ + { XFS_TRANS_ATTRINVAL, "ATTRINVAL" }, \ + { XFS_TRANS_ATRUNCATE, "ATRUNCATE" }, \ + { XFS_TRANS_ATTR_SET, "ATTR_SET" }, \ + { XFS_TRANS_ATTR_RM, "ATTR_RM" }, \ + { XFS_TRANS_ATTR_FLAG, "ATTR_FLAG" }, \ + { XFS_TRANS_CLEAR_AGI_BUCKET, "CLEAR_AGI_BUCKET" }, \ + { XFS_TRANS_QM_SBCHANGE, "QM_SBCHANGE" }, \ + { XFS_TRANS_QM_QUOTAOFF, "QM_QUOTAOFF" }, \ + { XFS_TRANS_QM_DQALLOC, "QM_DQALLOC" }, \ + { XFS_TRANS_QM_SETQLIM, "QM_SETQLIM" }, \ + { XFS_TRANS_QM_DQCLUSTER, "QM_DQCLUSTER" }, \ + { XFS_TRANS_QM_QINOCREATE, "QM_QINOCREATE" }, \ + { XFS_TRANS_QM_QUOTAOFF_END, "QM_QOFF_END" }, \ + { XFS_TRANS_SB_UNIT, "SB_UNIT" }, \ + { XFS_TRANS_FSYNC_TS, "FSYNC_TS" }, \ + { XFS_TRANS_GROWFSRT_ALLOC, "GROWFSRT_ALLOC" }, \ + { XFS_TRANS_GROWFSRT_ZERO, "GROWFSRT_ZERO" }, \ + { XFS_TRANS_GROWFSRT_FREE, "GROWFSRT_FREE" }, \ + { XFS_TRANS_SWAPEXT, "SWAPEXT" }, \ + { XFS_TRANS_SB_COUNT, "SB_COUNT" }, \ + { XFS_TRANS_CHECKPOINT, "CHECKPOINT" }, \ + { XFS_TRANS_DUMMY1, "DUMMY1" }, \ + { XFS_TRANS_DUMMY2, "DUMMY2" }, \ + { XLOG_UNMOUNT_REC_TYPE, "UNMOUNT" } + +/* + * This structure is used to track log items associated with + * a transaction. It points to the log item and keeps some + * flags to track the state of the log item. It also tracks + * the amount of space needed to log the item it describes + * once we get to commit processing (see xfs_trans_commit()). + */ +struct xfs_log_item_desc { + struct xfs_log_item *lid_item; + struct list_head lid_trans; + unsigned char lid_flags; +}; + +#define XFS_LID_DIRTY 0x1 + +/* + * Values for t_flags. + */ +#define XFS_TRANS_DIRTY 0x01 /* something needs to be logged */ +#define XFS_TRANS_SB_DIRTY 0x02 /* superblock is modified */ +#define XFS_TRANS_PERM_LOG_RES 0x04 /* xact took a permanent log res */ +#define XFS_TRANS_SYNC 0x08 /* make commit synchronous */ +#define XFS_TRANS_DQ_DIRTY 0x10 /* at least one dquot in trx dirty */ +#define XFS_TRANS_RESERVE 0x20 /* OK to use reserved data blocks */ +#define XFS_TRANS_FREEZE_PROT 0x40 /* Transaction has elevated writer + count in superblock */ + +/* + * Values for call flags parameter. + */ +#define XFS_TRANS_RELEASE_LOG_RES 0x4 +#define XFS_TRANS_ABORT 0x8 + +/* + * Field values for xfs_trans_mod_sb. + */ +#define XFS_TRANS_SB_ICOUNT 0x00000001 +#define XFS_TRANS_SB_IFREE 0x00000002 +#define XFS_TRANS_SB_FDBLOCKS 0x00000004 +#define XFS_TRANS_SB_RES_FDBLOCKS 0x00000008 +#define XFS_TRANS_SB_FREXTENTS 0x00000010 +#define XFS_TRANS_SB_RES_FREXTENTS 0x00000020 +#define XFS_TRANS_SB_DBLOCKS 0x00000040 +#define XFS_TRANS_SB_AGCOUNT 0x00000080 +#define XFS_TRANS_SB_IMAXPCT 0x00000100 +#define XFS_TRANS_SB_REXTSIZE 0x00000200 +#define XFS_TRANS_SB_RBMBLOCKS 0x00000400 +#define XFS_TRANS_SB_RBLOCKS 0x00000800 +#define XFS_TRANS_SB_REXTENTS 0x00001000 +#define XFS_TRANS_SB_REXTSLOG 0x00002000 + +/* + * Here we centralize the specification of XFS meta-data buffer + * reference count values. This determine how hard the buffer + * cache tries to hold onto the buffer. + */ +#define XFS_AGF_REF 4 +#define XFS_AGI_REF 4 +#define XFS_AGFL_REF 3 +#define XFS_INO_BTREE_REF 3 +#define XFS_ALLOC_BTREE_REF 2 +#define XFS_BMAP_BTREE_REF 2 +#define XFS_DIR_BTREE_REF 2 +#define XFS_INO_REF 2 +#define XFS_ATTR_BTREE_REF 1 +#define XFS_DQUOT_REF 1 + +/* + * Flags for xfs_trans_ichgtime(). + */ +#define XFS_ICHGTIME_MOD 0x1 /* data fork modification timestamp */ +#define XFS_ICHGTIME_CHG 0x2 /* inode field change timestamp */ +#define XFS_ICHGTIME_CREATE 0x4 /* inode create timestamp */ + + +/* + * Inode Log Item Format definitions. + * + * This is the structure used to lay out an inode log item in the + * log. The size of the inline data/extents/b-tree root to be logged + * (if any) is indicated in the ilf_dsize field. Changes to this structure + * must be added on to the end. + */ +typedef struct xfs_inode_log_format { + __uint16_t ilf_type; /* inode log item type */ + __uint16_t ilf_size; /* size of this item */ + __uint32_t ilf_fields; /* flags for fields logged */ + __uint16_t ilf_asize; /* size of attr d/ext/root */ + __uint16_t ilf_dsize; /* size of data/ext/root */ + __uint64_t ilf_ino; /* inode number */ + union { + __uint32_t ilfu_rdev; /* rdev value for dev inode*/ + uuid_t ilfu_uuid; /* mount point value */ + } ilf_u; + __int64_t ilf_blkno; /* blkno of inode buffer */ + __int32_t ilf_len; /* len of inode buffer */ + __int32_t ilf_boffset; /* off of inode in buffer */ +} xfs_inode_log_format_t; + +typedef struct xfs_inode_log_format_32 { + __uint16_t ilf_type; /* inode log item type */ + __uint16_t ilf_size; /* size of this item */ + __uint32_t ilf_fields; /* flags for fields logged */ + __uint16_t ilf_asize; /* size of attr d/ext/root */ + __uint16_t ilf_dsize; /* size of data/ext/root */ + __uint64_t ilf_ino; /* inode number */ + union { + __uint32_t ilfu_rdev; /* rdev value for dev inode*/ + uuid_t ilfu_uuid; /* mount point value */ + } ilf_u; + __int64_t ilf_blkno; /* blkno of inode buffer */ + __int32_t ilf_len; /* len of inode buffer */ + __int32_t ilf_boffset; /* off of inode in buffer */ +} __attribute__((packed)) xfs_inode_log_format_32_t; + +typedef struct xfs_inode_log_format_64 { + __uint16_t ilf_type; /* inode log item type */ + __uint16_t ilf_size; /* size of this item */ + __uint32_t ilf_fields; /* flags for fields logged */ + __uint16_t ilf_asize; /* size of attr d/ext/root */ + __uint16_t ilf_dsize; /* size of data/ext/root */ + __uint32_t ilf_pad; /* pad for 64 bit boundary */ + __uint64_t ilf_ino; /* inode number */ + union { + __uint32_t ilfu_rdev; /* rdev value for dev inode*/ + uuid_t ilfu_uuid; /* mount point value */ + } ilf_u; + __int64_t ilf_blkno; /* blkno of inode buffer */ + __int32_t ilf_len; /* len of inode buffer */ + __int32_t ilf_boffset; /* off of inode in buffer */ +} xfs_inode_log_format_64_t; + +/* + * Flags for xfs_trans_log_inode flags field. + */ +#define XFS_ILOG_CORE 0x001 /* log standard inode fields */ +#define XFS_ILOG_DDATA 0x002 /* log i_df.if_data */ +#define XFS_ILOG_DEXT 0x004 /* log i_df.if_extents */ +#define XFS_ILOG_DBROOT 0x008 /* log i_df.i_broot */ +#define XFS_ILOG_DEV 0x010 /* log the dev field */ +#define XFS_ILOG_UUID 0x020 /* log the uuid field */ +#define XFS_ILOG_ADATA 0x040 /* log i_af.if_data */ +#define XFS_ILOG_AEXT 0x080 /* log i_af.if_extents */ +#define XFS_ILOG_ABROOT 0x100 /* log i_af.i_broot */ + + +/* + * The timestamps are dirty, but not necessarily anything else in the inode + * core. Unlike the other fields above this one must never make it to disk + * in the ilf_fields of the inode_log_format, but is purely store in-memory in + * ili_fields in the inode_log_item. + */ +#define XFS_ILOG_TIMESTAMP 0x4000 + +#define XFS_ILOG_NONCORE (XFS_ILOG_DDATA | XFS_ILOG_DEXT | \ + XFS_ILOG_DBROOT | XFS_ILOG_DEV | \ + XFS_ILOG_UUID | XFS_ILOG_ADATA | \ + XFS_ILOG_AEXT | XFS_ILOG_ABROOT) + +#define XFS_ILOG_DFORK (XFS_ILOG_DDATA | XFS_ILOG_DEXT | \ + XFS_ILOG_DBROOT) + +#define XFS_ILOG_AFORK (XFS_ILOG_ADATA | XFS_ILOG_AEXT | \ + XFS_ILOG_ABROOT) + +#define XFS_ILOG_ALL (XFS_ILOG_CORE | XFS_ILOG_DDATA | \ + XFS_ILOG_DEXT | XFS_ILOG_DBROOT | \ + XFS_ILOG_DEV | XFS_ILOG_UUID | \ + XFS_ILOG_ADATA | XFS_ILOG_AEXT | \ + XFS_ILOG_ABROOT | XFS_ILOG_TIMESTAMP) + +static inline int xfs_ilog_fbroot(int w) +{ + return (w == XFS_DATA_FORK ? XFS_ILOG_DBROOT : XFS_ILOG_ABROOT); +} + +static inline int xfs_ilog_fext(int w) +{ + return (w == XFS_DATA_FORK ? XFS_ILOG_DEXT : XFS_ILOG_AEXT); +} + +static inline int xfs_ilog_fdata(int w) +{ + return (w == XFS_DATA_FORK ? XFS_ILOG_DDATA : XFS_ILOG_ADATA); +} + +/* + * Incore version of the on-disk inode core structures. We log this directly + * into the journal in host CPU format (for better or worse) and as such + * directly mirrors the xfs_dinode structure as it must contain all the same + * information. + */ +typedef struct xfs_ictimestamp { + __int32_t t_sec; /* timestamp seconds */ + __int32_t t_nsec; /* timestamp nanoseconds */ +} xfs_ictimestamp_t; + +/* + * NOTE: This structure must be kept identical to struct xfs_dinode + * in xfs_dinode.h except for the endianness annotations. + */ +typedef struct xfs_icdinode { + __uint16_t di_magic; /* inode magic # = XFS_DINODE_MAGIC */ + __uint16_t di_mode; /* mode and type of file */ + __int8_t di_version; /* inode version */ + __int8_t di_format; /* format of di_c data */ + __uint16_t di_onlink; /* old number of links to file */ + __uint32_t di_uid; /* owner's user id */ + __uint32_t di_gid; /* owner's group id */ + __uint32_t di_nlink; /* number of links to file */ + __uint16_t di_projid_lo; /* lower part of owner's project id */ + __uint16_t di_projid_hi; /* higher part of owner's project id */ + __uint8_t di_pad[6]; /* unused, zeroed space */ + __uint16_t di_flushiter; /* incremented on flush */ + xfs_ictimestamp_t di_atime; /* time last accessed */ + xfs_ictimestamp_t di_mtime; /* time last modified */ + xfs_ictimestamp_t di_ctime; /* time created/inode modified */ + xfs_fsize_t di_size; /* number of bytes in file */ + xfs_drfsbno_t di_nblocks; /* # of direct & btree blocks used */ + xfs_extlen_t di_extsize; /* basic/minimum extent size for file */ + xfs_extnum_t di_nextents; /* number of extents in data fork */ + xfs_aextnum_t di_anextents; /* number of extents in attribute fork*/ + __uint8_t di_forkoff; /* attr fork offs, <<3 for 64b align */ + __int8_t di_aformat; /* format of attr fork's data */ + __uint32_t di_dmevmask; /* DMIG event mask */ + __uint16_t di_dmstate; /* DMIG state info */ + __uint16_t di_flags; /* random flags, XFS_DIFLAG_... */ + __uint32_t di_gen; /* generation number */ + + /* di_next_unlinked is the only non-core field in the old dinode */ + xfs_agino_t di_next_unlinked;/* agi unlinked list ptr */ + + /* start of the extended dinode, writable fields */ + __uint32_t di_crc; /* CRC of the inode */ + __uint64_t di_changecount; /* number of attribute changes */ + xfs_lsn_t di_lsn; /* flush sequence */ + __uint64_t di_flags2; /* more random flags */ + __uint8_t di_pad2[16]; /* more padding for future expansion */ + + /* fields only written to during inode creation */ + xfs_ictimestamp_t di_crtime; /* time created */ + xfs_ino_t di_ino; /* inode number */ + uuid_t di_uuid; /* UUID of the filesystem */ + + /* structure must be padded to 64 bit alignment */ +} xfs_icdinode_t; + +static inline uint xfs_icdinode_size(int version) +{ + if (version == 3) + return sizeof(struct xfs_icdinode); + return offsetof(struct xfs_icdinode, di_next_unlinked); +} + +/* + * Buffer Log Format defintions + * + * These are the physical dirty bitmap defintions for the log format structure. + */ +#define XFS_BLF_CHUNK 128 +#define XFS_BLF_SHIFT 7 +#define BIT_TO_WORD_SHIFT 5 +#define NBWORD (NBBY * sizeof(unsigned int)) + +/* + * This flag indicates that the buffer contains on disk inodes + * and requires special recovery handling. + */ +#define XFS_BLF_INODE_BUF (1<<0) + +/* + * This flag indicates that the buffer should not be replayed + * during recovery because its blocks are being freed. + */ +#define XFS_BLF_CANCEL (1<<1) + +/* + * This flag indicates that the buffer contains on disk + * user or group dquots and may require special recovery handling. + */ +#define XFS_BLF_UDQUOT_BUF (1<<2) +#define XFS_BLF_PDQUOT_BUF (1<<3) +#define XFS_BLF_GDQUOT_BUF (1<<4) + +/* + * This is the structure used to lay out a buf log item in the + * log. The data map describes which 128 byte chunks of the buffer + * have been logged. + */ +#define XFS_BLF_DATAMAP_SIZE ((XFS_MAX_BLOCKSIZE / XFS_BLF_CHUNK) / NBWORD) + +typedef struct xfs_buf_log_format { + unsigned short blf_type; /* buf log item type indicator */ + unsigned short blf_size; /* size of this item */ + ushort blf_flags; /* misc state */ + ushort blf_len; /* number of blocks in this buf */ + __int64_t blf_blkno; /* starting blkno of this buf */ + unsigned int blf_map_size; /* used size of data bitmap in words */ + unsigned int blf_data_map[XFS_BLF_DATAMAP_SIZE]; /* dirty bitmap */ +} xfs_buf_log_format_t; + +/* + * All buffers now need to tell recovery where the magic number + * is so that it can verify and calculate the CRCs on the buffer correctly + * once the changes have been replayed into the buffer. + * + * The type value is held in the upper 5 bits of the blf_flags field, which is + * an unsigned 16 bit field. Hence we need to shift it 11 bits up and down. + */ +#define XFS_BLFT_BITS 5 +#define XFS_BLFT_SHIFT 11 +#define XFS_BLFT_MASK (((1 << XFS_BLFT_BITS) - 1) << XFS_BLFT_SHIFT) + +enum xfs_blft { + XFS_BLFT_UNKNOWN_BUF = 0, + XFS_BLFT_UDQUOT_BUF, + XFS_BLFT_PDQUOT_BUF, + XFS_BLFT_GDQUOT_BUF, + XFS_BLFT_BTREE_BUF, + XFS_BLFT_AGF_BUF, + XFS_BLFT_AGFL_BUF, + XFS_BLFT_AGI_BUF, + XFS_BLFT_DINO_BUF, + XFS_BLFT_SYMLINK_BUF, + XFS_BLFT_DIR_BLOCK_BUF, + XFS_BLFT_DIR_DATA_BUF, + XFS_BLFT_DIR_FREE_BUF, + XFS_BLFT_DIR_LEAF1_BUF, + XFS_BLFT_DIR_LEAFN_BUF, + XFS_BLFT_DA_NODE_BUF, + XFS_BLFT_ATTR_LEAF_BUF, + XFS_BLFT_ATTR_RMT_BUF, + XFS_BLFT_SB_BUF, + XFS_BLFT_MAX_BUF = (1 << XFS_BLFT_BITS), +}; + +static inline void +xfs_blft_to_flags(struct xfs_buf_log_format *blf, enum xfs_blft type) +{ + ASSERT(type > XFS_BLFT_UNKNOWN_BUF && type < XFS_BLFT_MAX_BUF); + blf->blf_flags &= ~XFS_BLFT_MASK; + blf->blf_flags |= ((type << XFS_BLFT_SHIFT) & XFS_BLFT_MASK); +} + +static inline __uint16_t +xfs_blft_from_flags(struct xfs_buf_log_format *blf) +{ + return (blf->blf_flags & XFS_BLFT_MASK) >> XFS_BLFT_SHIFT; +} + +/* + * EFI/EFD log format definitions + */ +typedef struct xfs_extent { + xfs_dfsbno_t ext_start; + xfs_extlen_t ext_len; +} xfs_extent_t; + +/* + * Since an xfs_extent_t has types (start:64, len: 32) + * there are different alignments on 32 bit and 64 bit kernels. + * So we provide the different variants for use by a + * conversion routine. + */ +typedef struct xfs_extent_32 { + __uint64_t ext_start; + __uint32_t ext_len; +} __attribute__((packed)) xfs_extent_32_t; + +typedef struct xfs_extent_64 { + __uint64_t ext_start; + __uint32_t ext_len; + __uint32_t ext_pad; +} xfs_extent_64_t; + +/* + * This is the structure used to lay out an efi log item in the + * log. The efi_extents field is a variable size array whose + * size is given by efi_nextents. + */ +typedef struct xfs_efi_log_format { + __uint16_t efi_type; /* efi log item type */ + __uint16_t efi_size; /* size of this item */ + __uint32_t efi_nextents; /* # extents to free */ + __uint64_t efi_id; /* efi identifier */ + xfs_extent_t efi_extents[1]; /* array of extents to free */ +} xfs_efi_log_format_t; + +typedef struct xfs_efi_log_format_32 { + __uint16_t efi_type; /* efi log item type */ + __uint16_t efi_size; /* size of this item */ + __uint32_t efi_nextents; /* # extents to free */ + __uint64_t efi_id; /* efi identifier */ + xfs_extent_32_t efi_extents[1]; /* array of extents to free */ +} __attribute__((packed)) xfs_efi_log_format_32_t; + +typedef struct xfs_efi_log_format_64 { + __uint16_t efi_type; /* efi log item type */ + __uint16_t efi_size; /* size of this item */ + __uint32_t efi_nextents; /* # extents to free */ + __uint64_t efi_id; /* efi identifier */ + xfs_extent_64_t efi_extents[1]; /* array of extents to free */ +} xfs_efi_log_format_64_t; + +/* + * This is the structure used to lay out an efd log item in the + * log. The efd_extents array is a variable size array whose + * size is given by efd_nextents; + */ +typedef struct xfs_efd_log_format { + __uint16_t efd_type; /* efd log item type */ + __uint16_t efd_size; /* size of this item */ + __uint32_t efd_nextents; /* # of extents freed */ + __uint64_t efd_efi_id; /* id of corresponding efi */ + xfs_extent_t efd_extents[1]; /* array of extents freed */ +} xfs_efd_log_format_t; + +typedef struct xfs_efd_log_format_32 { + __uint16_t efd_type; /* efd log item type */ + __uint16_t efd_size; /* size of this item */ + __uint32_t efd_nextents; /* # of extents freed */ + __uint64_t efd_efi_id; /* id of corresponding efi */ + xfs_extent_32_t efd_extents[1]; /* array of extents freed */ +} __attribute__((packed)) xfs_efd_log_format_32_t; + +typedef struct xfs_efd_log_format_64 { + __uint16_t efd_type; /* efd log item type */ + __uint16_t efd_size; /* size of this item */ + __uint32_t efd_nextents; /* # of extents freed */ + __uint64_t efd_efi_id; /* id of corresponding efi */ + xfs_extent_64_t efd_extents[1]; /* array of extents freed */ +} xfs_efd_log_format_64_t; + +/* + * Dquot Log format definitions. + * + * The first two fields must be the type and size fitting into + * 32 bits : log_recovery code assumes that. + */ +typedef struct xfs_dq_logformat { + __uint16_t qlf_type; /* dquot log item type */ + __uint16_t qlf_size; /* size of this item */ + xfs_dqid_t qlf_id; /* usr/grp/proj id : 32 bits */ + __int64_t qlf_blkno; /* blkno of dquot buffer */ + __int32_t qlf_len; /* len of dquot buffer */ + __uint32_t qlf_boffset; /* off of dquot in buffer */ +} xfs_dq_logformat_t; + +/* + * log format struct for QUOTAOFF records. + * The first two fields must be the type and size fitting into + * 32 bits : log_recovery code assumes that. + * We write two LI_QUOTAOFF logitems per quotaoff, the last one keeps a pointer + * to the first and ensures that the first logitem is taken out of the AIL + * only when the last one is securely committed. + */ +typedef struct xfs_qoff_logformat { + unsigned short qf_type; /* quotaoff log item type */ + unsigned short qf_size; /* size of this item */ + unsigned int qf_flags; /* USR and/or GRP */ + char qf_pad[12]; /* padding for future */ +} xfs_qoff_logformat_t; + + +/* + * Disk quotas status in m_qflags, and also sb_qflags. 16 bits. + */ +#define XFS_UQUOTA_ACCT 0x0001 /* user quota accounting ON */ +#define XFS_UQUOTA_ENFD 0x0002 /* user quota limits enforced */ +#define XFS_UQUOTA_CHKD 0x0004 /* quotacheck run on usr quotas */ +#define XFS_PQUOTA_ACCT 0x0008 /* project quota accounting ON */ +#define XFS_OQUOTA_ENFD 0x0010 /* other (grp/prj) quota limits enforced */ +#define XFS_OQUOTA_CHKD 0x0020 /* quotacheck run on other (grp/prj) quotas */ +#define XFS_GQUOTA_ACCT 0x0040 /* group quota accounting ON */ + +/* + * Conversion to and from the combined OQUOTA flag (if necessary) + * is done only in xfs_sb_qflags_to_disk() and xfs_sb_qflags_from_disk() + */ +#define XFS_GQUOTA_ENFD 0x0080 /* group quota limits enforced */ +#define XFS_GQUOTA_CHKD 0x0100 /* quotacheck run on group quotas */ +#define XFS_PQUOTA_ENFD 0x0200 /* project quota limits enforced */ +#define XFS_PQUOTA_CHKD 0x0400 /* quotacheck run on project quotas */ + +#define XFS_ALL_QUOTA_ACCT \ + (XFS_UQUOTA_ACCT | XFS_GQUOTA_ACCT | XFS_PQUOTA_ACCT) +#define XFS_ALL_QUOTA_ENFD \ + (XFS_UQUOTA_ENFD | XFS_GQUOTA_ENFD | XFS_PQUOTA_ENFD) +#define XFS_ALL_QUOTA_CHKD \ + (XFS_UQUOTA_CHKD | XFS_GQUOTA_CHKD | XFS_PQUOTA_CHKD) + +#define XFS_MOUNT_QUOTA_ALL (XFS_UQUOTA_ACCT|XFS_UQUOTA_ENFD|\ + XFS_UQUOTA_CHKD|XFS_GQUOTA_ACCT|\ + XFS_GQUOTA_ENFD|XFS_GQUOTA_CHKD|\ + XFS_PQUOTA_ACCT|XFS_PQUOTA_ENFD|\ + XFS_PQUOTA_CHKD) + +/* + * Inode create log item structure + * + * Log recovery assumes the first two entries are the type and size and they fit + * in 32 bits. Also in host order (ugh) so they have to be 32 bit aligned so + * decoding can be done correctly. + */ +struct xfs_icreate_log { + __uint16_t icl_type; /* type of log format structure */ + __uint16_t icl_size; /* size of log format structure */ + __be32 icl_ag; /* ag being allocated in */ + __be32 icl_agbno; /* start block of inode range */ + __be32 icl_count; /* number of inodes to initialise */ + __be32 icl_isize; /* size of inodes */ + __be32 icl_length; /* length of extent to initialise */ + __be32 icl_gen; /* inode generation number to use */ +}; + +int xfs_log_calc_unit_res(struct xfs_mount *mp, int unit_bytes); +int xfs_log_calc_minimum_size(struct xfs_mount *); + + +#endif /* __XFS_LOG_FORMAT_H__ */ diff --git a/fs/xfs/xfs_log_priv.h b/fs/xfs/xfs_log_priv.h index b9ea262..136654b 100644 --- a/fs/xfs/xfs_log_priv.h +++ b/fs/xfs/xfs_log_priv.h @@ -24,51 +24,13 @@ struct xlog_ticket; struct xfs_mount; /* - * Macros, structures, prototypes for internal log manager use. + * Flags for log structure */ - -#define XLOG_MIN_ICLOGS 2 -#define XLOG_MAX_ICLOGS 8 -#define XLOG_HEADER_MAGIC_NUM 0xFEEDbabe /* Invalid cycle number */ -#define XLOG_VERSION_1 1 -#define XLOG_VERSION_2 2 /* Large IClogs, Log sunit */ -#define XLOG_VERSION_OKBITS (XLOG_VERSION_1 | XLOG_VERSION_2) -#define XLOG_MIN_RECORD_BSIZE (16*1024) /* eventually 32k */ -#define XLOG_BIG_RECORD_BSIZE (32*1024) /* 32k buffers */ -#define XLOG_MAX_RECORD_BSIZE (256*1024) -#define XLOG_HEADER_CYCLE_SIZE (32*1024) /* cycle data in header */ -#define XLOG_MIN_RECORD_BSHIFT 14 /* 16384 == 1 << 14 */ -#define XLOG_BIG_RECORD_BSHIFT 15 /* 32k == 1 << 15 */ -#define XLOG_MAX_RECORD_BSHIFT 18 /* 256k == 1 << 18 */ -#define XLOG_BTOLSUNIT(log, b) (((b)+(log)->l_mp->m_sb.sb_logsunit-1) / \ - (log)->l_mp->m_sb.sb_logsunit) -#define XLOG_LSUNITTOB(log, su) ((su) * (log)->l_mp->m_sb.sb_logsunit) - -#define XLOG_HEADER_SIZE 512 - -#define XLOG_REC_SHIFT(log) \ - BTOBB(1 << (xfs_sb_version_haslogv2(&log->l_mp->m_sb) ? \ - XLOG_MAX_RECORD_BSHIFT : XLOG_BIG_RECORD_BSHIFT)) -#define XLOG_TOTAL_REC_SHIFT(log) \ - BTOBB(XLOG_MAX_ICLOGS << (xfs_sb_version_haslogv2(&log->l_mp->m_sb) ? \ - XLOG_MAX_RECORD_BSHIFT : XLOG_BIG_RECORD_BSHIFT)) - -static inline xfs_lsn_t xlog_assign_lsn(uint cycle, uint block) -{ - return ((xfs_lsn_t)cycle << 32) | block; -} - -static inline uint xlog_get_cycle(char *ptr) -{ - if (be32_to_cpu(*(__be32 *)ptr) == XLOG_HEADER_MAGIC_NUM) - return be32_to_cpu(*((__be32 *)ptr + 1)); - else - return be32_to_cpu(*(__be32 *)ptr); -} - -#define BLK_AVG(blk1, blk2) ((blk1+blk2) >> 1) - -#ifdef __KERNEL__ +#define XLOG_ACTIVE_RECOVERY 0x2 /* in the middle of recovery */ +#define XLOG_RECOVERY_NEEDED 0x4 /* log was recovered */ +#define XLOG_IO_ERROR 0x8 /* log hit an I/O error, and being + shutdown */ +#define XLOG_TAIL_WARN 0x10 /* log tail verify warning issued */ /* * get client id from packed copy. @@ -101,28 +63,8 @@ static inline uint xlog_get_client_id(__be32 i) #define XLOG_STATE_IOERROR 0x0080 /* IO error happened in sync'ing log */ #define XLOG_STATE_ALL 0x7FFF /* All possible valid flags */ #define XLOG_STATE_NOTUSED 0x8000 /* This IC log not being used */ -#endif /* __KERNEL__ */ /* - * Flags to log operation header - * - * The first write of a new transaction will be preceded with a start - * record, XLOG_START_TRANS. Once a transaction is committed, a commit - * record is written, XLOG_COMMIT_TRANS. If a single region can not fit into - * the remainder of the current active in-core log, it is split up into - * multiple regions. Each partial region will be marked with a - * XLOG_CONTINUE_TRANS until the last one, which gets marked with XLOG_END_TRANS. - * - */ -#define XLOG_START_TRANS 0x01 /* Start a new transaction */ -#define XLOG_COMMIT_TRANS 0x02 /* Commit this transaction */ -#define XLOG_CONTINUE_TRANS 0x04 /* Cont this trans into new region */ -#define XLOG_WAS_CONT_TRANS 0x08 /* Cont this trans into new region */ -#define XLOG_END_TRANS 0x10 /* End a continued transaction */ -#define XLOG_UNMOUNT_TRANS 0x20 /* Unmount a filesystem transaction */ - -#ifdef __KERNEL__ -/* * Flags to log ticket */ #define XLOG_TIC_INITED 0x1 /* has been initialized */ @@ -132,22 +74,6 @@ static inline uint xlog_get_client_id(__be32 i) { XLOG_TIC_INITED, "XLOG_TIC_INITED" }, \ { XLOG_TIC_PERM_RESERV, "XLOG_TIC_PERM_RESERV" } -#endif /* __KERNEL__ */ - -#define XLOG_UNMOUNT_TYPE 0x556e /* Un for Unmount */ - -/* - * Flags for log structure - */ -#define XLOG_ACTIVE_RECOVERY 0x2 /* in the middle of recovery */ -#define XLOG_RECOVERY_NEEDED 0x4 /* log was recovered */ -#define XLOG_IO_ERROR 0x8 /* log hit an I/O error, and being - shutdown */ -#define XLOG_TAIL_WARN 0x10 /* log tail verify warning issued */ - -typedef __uint32_t xlog_tid_t; - -#ifdef __KERNEL__ /* * Below are states for covering allocation transactions. * By covering, we mean changing the h_tail_lsn in the last on-disk @@ -223,7 +149,6 @@ typedef __uint32_t xlog_tid_t; #define XLOG_COVER_OPS 5 - /* Ticket reservation region accounting */ #define XLOG_TIC_LEN_MAX 15 @@ -258,64 +183,6 @@ typedef struct xlog_ticket { xlog_res_t t_res_arr[XLOG_TIC_LEN_MAX]; /* array of res : 8 * 15 */ } xlog_ticket_t; -#endif - - -typedef struct xlog_op_header { - __be32 oh_tid; /* transaction id of operation : 4 b */ - __be32 oh_len; /* bytes in data region : 4 b */ - __u8 oh_clientid; /* who sent me this : 1 b */ - __u8 oh_flags; /* : 1 b */ - __u16 oh_res2; /* 32 bit align : 2 b */ -} xlog_op_header_t; - - -/* valid values for h_fmt */ -#define XLOG_FMT_UNKNOWN 0 -#define XLOG_FMT_LINUX_LE 1 -#define XLOG_FMT_LINUX_BE 2 -#define XLOG_FMT_IRIX_BE 3 - -/* our fmt */ -#ifdef XFS_NATIVE_HOST -#define XLOG_FMT XLOG_FMT_LINUX_BE -#else -#define XLOG_FMT XLOG_FMT_LINUX_LE -#endif - -typedef struct xlog_rec_header { - __be32 h_magicno; /* log record (LR) identifier : 4 */ - __be32 h_cycle; /* write cycle of log : 4 */ - __be32 h_version; /* LR version : 4 */ - __be32 h_len; /* len in bytes; should be 64-bit aligned: 4 */ - __be64 h_lsn; /* lsn of this LR : 8 */ - __be64 h_tail_lsn; /* lsn of 1st LR w/ buffers not committed: 8 */ - __le32 h_crc; /* crc of log record : 4 */ - __be32 h_prev_block; /* block number to previous LR : 4 */ - __be32 h_num_logops; /* number of log operations in this LR : 4 */ - __be32 h_cycle_data[XLOG_HEADER_CYCLE_SIZE / BBSIZE]; - /* new fields */ - __be32 h_fmt; /* format of log record : 4 */ - uuid_t h_fs_uuid; /* uuid of FS : 16 */ - __be32 h_size; /* iclog size : 4 */ -} xlog_rec_header_t; - -typedef struct xlog_rec_ext_header { - __be32 xh_cycle; /* write cycle of log : 4 */ - __be32 xh_cycle_data[XLOG_HEADER_CYCLE_SIZE / BBSIZE]; /* : 256 */ -} xlog_rec_ext_header_t; - -#ifdef __KERNEL__ - -/* - * Quite misnamed, because this union lays out the actual on-disk log buffer. - */ -typedef union xlog_in_core2 { - xlog_rec_header_t hic_header; - xlog_rec_ext_header_t hic_xheader; - char hic_sector[XLOG_HEADER_SIZE]; -} xlog_in_core_2_t; - /* * - A log record header is 512 bytes. There is plenty of room to grow the * xlog_rec_header_t into the reserved space. @@ -411,14 +278,17 @@ struct xfs_cil { struct xlog *xc_log; struct list_head xc_cil; spinlock_t xc_cil_lock; + + struct rw_semaphore xc_ctx_lock ____cacheline_aligned_in_smp; struct xfs_cil_ctx *xc_ctx; - struct rw_semaphore xc_ctx_lock; + + spinlock_t xc_push_lock ____cacheline_aligned_in_smp; + xfs_lsn_t xc_push_seq; struct list_head xc_committing; wait_queue_head_t xc_commit_wait; xfs_lsn_t xc_current_sequence; struct work_struct xc_push_work; - xfs_lsn_t xc_push_seq; -}; +} ____cacheline_aligned_in_smp; /* * The amount of log space we allow the CIL to aggregate is difficult to size. @@ -686,6 +556,5 @@ static inline void xlog_wait(wait_queue_head_t *wq, spinlock_t *lock) schedule(); remove_wait_queue(wq, &wait); } -#endif /* __KERNEL__ */ #endif /* __XFS_LOG_PRIV_H__ */ diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c index 7681b19..7c0c1fd 100644 --- a/fs/xfs/xfs_log_recover.c +++ b/fs/xfs/xfs_log_recover.c @@ -17,7 +17,7 @@ */ #include "xfs.h" #include "xfs_fs.h" -#include "xfs_types.h" +#include "xfs_format.h" #include "xfs_bit.h" #include "xfs_log.h" #include "xfs_inum.h" @@ -41,7 +41,6 @@ #include "xfs_extfree_item.h" #include "xfs_trans_priv.h" #include "xfs_quota.h" -#include "xfs_utils.h" #include "xfs_cksum.h" #include "xfs_trace.h" #include "xfs_icache.h" @@ -51,10 +50,12 @@ #include "xfs_symlink.h" #include "xfs_da_btree.h" #include "xfs_dir2_format.h" -#include "xfs_dir2_priv.h" +#include "xfs_dir2.h" #include "xfs_attr_leaf.h" #include "xfs_attr_remote.h" +#define BLK_AVG(blk1, blk2) ((blk1+blk2) >> 1) + STATIC int xlog_find_zeroed( struct xlog *, @@ -607,7 +608,7 @@ out: /* * Head is defined to be the point of the log where the next log write - * write could go. This means that incomplete LR writes at the end are + * could go. This means that incomplete LR writes at the end are * eliminated when calculating the head. We aren't guaranteed that previous * LR have complete transactions. We only know that a cycle number of * current cycle number -1 won't be present in the log if we start writing @@ -963,6 +964,7 @@ xlog_find_tail( } if (!found) { xfs_warn(log->l_mp, "%s: couldn't find sync record", __func__); + xlog_put_bp(bp); ASSERT(0); return XFS_ERROR(EIO); } @@ -1144,7 +1146,8 @@ xlog_find_zeroed( */ xfs_warn(log->l_mp, "Log inconsistent or not a log (last==0, first!=1)"); - return XFS_ERROR(EINVAL); + error = XFS_ERROR(EINVAL); + goto bp_err; } /* we have a partially zeroed log */ @@ -1766,19 +1769,11 @@ xlog_recover_buffer_pass1( /* * Check to see whether the buffer being recovered has a corresponding - * entry in the buffer cancel record table. If it does then return 1 - * so that it will be cancelled, otherwise return 0. If the buffer is - * actually a buffer cancel item (XFS_BLF_CANCEL is set), then decrement - * the refcount on the entry in the table and remove it from the table - * if this is the last reference. - * - * We remove the cancel record from the table when we encounter its - * last occurrence in the log so that if the same buffer is re-used - * again after its last cancellation we actually replay the changes - * made at that point. + * entry in the buffer cancel record table. If it is, return the cancel + * buffer structure to the caller. */ -STATIC int -xlog_check_buffer_cancelled( +STATIC struct xfs_buf_cancel * +xlog_peek_buffer_cancelled( struct xlog *log, xfs_daddr_t blkno, uint len, @@ -1787,22 +1782,16 @@ xlog_check_buffer_cancelled( struct list_head *bucket; struct xfs_buf_cancel *bcp; - if (log->l_buf_cancel_table == NULL) { - /* - * There is nothing in the table built in pass one, - * so this buffer must not be cancelled. - */ + if (!log->l_buf_cancel_table) { + /* empty table means no cancelled buffers in the log */ ASSERT(!(flags & XFS_BLF_CANCEL)); - return 0; + return NULL; } - /* - * Search for an entry in the cancel table that matches our buffer. - */ bucket = XLOG_BUF_CANCEL_BUCKET(log, blkno); list_for_each_entry(bcp, bucket, bc_list) { if (bcp->bc_blkno == blkno && bcp->bc_len == len) - goto found; + return bcp; } /* @@ -1810,9 +1799,32 @@ xlog_check_buffer_cancelled( * that the buffer is NOT cancelled. */ ASSERT(!(flags & XFS_BLF_CANCEL)); - return 0; + return NULL; +} + +/* + * If the buffer is being cancelled then return 1 so that it will be cancelled, + * otherwise return 0. If the buffer is actually a buffer cancel item + * (XFS_BLF_CANCEL is set), then decrement the refcount on the entry in the + * table and remove it from the table if this is the last reference. + * + * We remove the cancel record from the table when we encounter its last + * occurrence in the log so that if the same buffer is re-used again after its + * last cancellation we actually replay the changes made at that point. + */ +STATIC int +xlog_check_buffer_cancelled( + struct xlog *log, + xfs_daddr_t blkno, + uint len, + ushort flags) +{ + struct xfs_buf_cancel *bcp; + + bcp = xlog_peek_buffer_cancelled(log, blkno, len, flags); + if (!bcp) + return 0; -found: /* * We've go a match, so return 1 so that the recovery of this buffer * is cancelled. If this buffer is actually a buffer cancel log @@ -1947,6 +1959,104 @@ xlog_recover_do_inode_buffer( } /* + * V5 filesystems know the age of the buffer on disk being recovered. We can + * have newer objects on disk than we are replaying, and so for these cases we + * don't want to replay the current change as that will make the buffer contents + * temporarily invalid on disk. + * + * The magic number might not match the buffer type we are going to recover + * (e.g. reallocated blocks), so we ignore the xfs_buf_log_format flags. Hence + * extract the LSN of the existing object in the buffer based on it's current + * magic number. If we don't recognise the magic number in the buffer, then + * return a LSN of -1 so that the caller knows it was an unrecognised block and + * so can recover the buffer. + */ +static xfs_lsn_t +xlog_recover_get_buf_lsn( + struct xfs_mount *mp, + struct xfs_buf *bp) +{ + __uint32_t magic32; + __uint16_t magic16; + __uint16_t magicda; + void *blk = bp->b_addr; + + /* v4 filesystems always recover immediately */ + if (!xfs_sb_version_hascrc(&mp->m_sb)) + goto recover_immediately; + + magic32 = be32_to_cpu(*(__be32 *)blk); + switch (magic32) { + case XFS_ABTB_CRC_MAGIC: + case XFS_ABTC_CRC_MAGIC: + case XFS_ABTB_MAGIC: + case XFS_ABTC_MAGIC: + case XFS_IBT_CRC_MAGIC: + case XFS_IBT_MAGIC: + return be64_to_cpu( + ((struct xfs_btree_block *)blk)->bb_u.s.bb_lsn); + case XFS_BMAP_CRC_MAGIC: + case XFS_BMAP_MAGIC: + return be64_to_cpu( + ((struct xfs_btree_block *)blk)->bb_u.l.bb_lsn); + case XFS_AGF_MAGIC: + return be64_to_cpu(((struct xfs_agf *)blk)->agf_lsn); + case XFS_AGFL_MAGIC: + return be64_to_cpu(((struct xfs_agfl *)blk)->agfl_lsn); + case XFS_AGI_MAGIC: + return be64_to_cpu(((struct xfs_agi *)blk)->agi_lsn); + case XFS_SYMLINK_MAGIC: + return be64_to_cpu(((struct xfs_dsymlink_hdr *)blk)->sl_lsn); + case XFS_DIR3_BLOCK_MAGIC: + case XFS_DIR3_DATA_MAGIC: + case XFS_DIR3_FREE_MAGIC: + return be64_to_cpu(((struct xfs_dir3_blk_hdr *)blk)->lsn); + case XFS_ATTR3_RMT_MAGIC: + return be64_to_cpu(((struct xfs_attr3_rmt_hdr *)blk)->rm_lsn); + case XFS_SB_MAGIC: + return be64_to_cpu(((struct xfs_sb *)blk)->sb_lsn); + default: + break; + } + + magicda = be16_to_cpu(((struct xfs_da_blkinfo *)blk)->magic); + switch (magicda) { + case XFS_DIR3_LEAF1_MAGIC: + case XFS_DIR3_LEAFN_MAGIC: + case XFS_DA3_NODE_MAGIC: + return be64_to_cpu(((struct xfs_da3_blkinfo *)blk)->lsn); + default: + break; + } + + /* + * We do individual object checks on dquot and inode buffers as they + * have their own individual LSN records. Also, we could have a stale + * buffer here, so we have to at least recognise these buffer types. + * + * A notd complexity here is inode unlinked list processing - it logs + * the inode directly in the buffer, but we don't know which inodes have + * been modified, and there is no global buffer LSN. Hence we need to + * recover all inode buffer types immediately. This problem will be + * fixed by logical logging of the unlinked list modifications. + */ + magic16 = be16_to_cpu(*(__be16 *)blk); + switch (magic16) { + case XFS_DQUOT_MAGIC: + case XFS_DINODE_MAGIC: + goto recover_immediately; + default: + break; + } + + /* unknown buffer contents, recover immediately */ + +recover_immediately: + return (xfs_lsn_t)-1; + +} + +/* * Validate the recovered buffer is of the correct type and attach the * appropriate buffer operations to them for writeback. Magic numbers are in a * few places: @@ -1955,7 +2065,7 @@ xlog_recover_do_inode_buffer( * inside a struct xfs_da_blkinfo at the start of the buffer. */ static void -xlog_recovery_validate_buf_type( +xlog_recover_validate_buf_type( struct xfs_mount *mp, struct xfs_buf *bp, xfs_buf_log_format_t *buf_f) @@ -2234,7 +2344,7 @@ xlog_recover_do_reg_buffer( * just avoid the verification stage for non-crc filesystems */ if (xfs_sb_version_hascrc(&mp->m_sb)) - xlog_recovery_validate_buf_type(mp, bp, buf_f); + xlog_recover_validate_buf_type(mp, bp, buf_f); } /* @@ -2366,7 +2476,7 @@ xfs_qm_dqcheck( /* * Perform a dquot buffer recovery. - * Simple algorithm: if we have found a QUOTAOFF logitem of the same type + * Simple algorithm: if we have found a QUOTAOFF log item of the same type * (ie. USR or GRP), then just toss this buffer away; don't recover it. * Else, treat it as a regular buffer and do recovery. */ @@ -2425,20 +2535,22 @@ xlog_recover_do_dquot_buffer( * over the log during recovery. During the first we build a table of * those buffers which have been cancelled, and during the second we * only replay those buffers which do not have corresponding cancel - * records in the table. See xlog_recover_do_buffer_pass[1,2] above + * records in the table. See xlog_recover_buffer_pass[1,2] above * for more details on the implementation of the table of cancel records. */ STATIC int xlog_recover_buffer_pass2( struct xlog *log, struct list_head *buffer_list, - struct xlog_recover_item *item) + struct xlog_recover_item *item, + xfs_lsn_t current_lsn) { xfs_buf_log_format_t *buf_f = item->ri_buf[0].i_addr; xfs_mount_t *mp = log->l_mp; xfs_buf_t *bp; int error; uint buf_flags; + xfs_lsn_t lsn; /* * In this pass we only want to recover all the buffers which have @@ -2463,10 +2575,17 @@ xlog_recover_buffer_pass2( error = bp->b_error; if (error) { xfs_buf_ioerror_alert(bp, "xlog_recover_do..(read#1)"); - xfs_buf_relse(bp); - return error; + goto out_release; } + /* + * recover the buffer only if we get an LSN from it and it's less than + * the lsn of the transaction we are replaying. + */ + lsn = xlog_recover_get_buf_lsn(mp, bp); + if (lsn && lsn != -1 && XFS_LSN_CMP(lsn, current_lsn) >= 0) + goto out_release; + if (buf_f->blf_flags & XFS_BLF_INODE_BUF) { error = xlog_recover_do_inode_buffer(mp, item, bp, buf_f); } else if (buf_f->blf_flags & @@ -2476,7 +2595,7 @@ xlog_recover_buffer_pass2( xlog_recover_do_reg_buffer(mp, item, bp, buf_f); } if (error) - return XFS_ERROR(error); + goto out_release; /* * Perform delayed write on the buffer. Asynchronous writes will be @@ -2505,6 +2624,7 @@ xlog_recover_buffer_pass2( xfs_buf_delwri_queue(bp, buffer_list); } +out_release: xfs_buf_relse(bp); return error; } @@ -2513,7 +2633,8 @@ STATIC int xlog_recover_inode_pass2( struct xlog *log, struct list_head *buffer_list, - struct xlog_recover_item *item) + struct xlog_recover_item *item, + xfs_lsn_t current_lsn) { xfs_inode_log_format_t *in_f; xfs_mount_t *mp = log->l_mp; @@ -2593,6 +2714,20 @@ xlog_recover_inode_pass2( } /* + * If the inode has an LSN in it, recover the inode only if it's less + * than the lsn of the transaction we are replaying. + */ + if (dip->di_version >= 3) { + xfs_lsn_t lsn = be64_to_cpu(dip->di_lsn); + + if (lsn && lsn != -1 && XFS_LSN_CMP(lsn, current_lsn) >= 0) { + trace_xfs_log_recover_inode_skip(log, in_f); + error = 0; + goto out_release; + } + } + + /* * di_flushiter is only valid for v1/2 inodes. All changes for v3 inodes * are transactional and if ordering is necessary we can determine that * more accurately by the LSN field in the V3 inode core. Don't trust @@ -2781,6 +2916,8 @@ write_inode_buffer: ASSERT(bp->b_target->bt_mount == mp); bp->b_iodone = xlog_recover_iodone; xfs_buf_delwri_queue(bp, buffer_list); + +out_release: xfs_buf_relse(bp); error: if (need_free) @@ -2822,7 +2959,8 @@ STATIC int xlog_recover_dquot_pass2( struct xlog *log, struct list_head *buffer_list, - struct xlog_recover_item *item) + struct xlog_recover_item *item, + xfs_lsn_t current_lsn) { xfs_mount_t *mp = log->l_mp; xfs_buf_t *bp; @@ -2896,6 +3034,19 @@ xlog_recover_dquot_pass2( return XFS_ERROR(EIO); } + /* + * If the dquot has an LSN in it, recover the dquot only if it's less + * than the lsn of the transaction we are replaying. + */ + if (xfs_sb_version_hascrc(&mp->m_sb)) { + struct xfs_dqblk *dqb = (struct xfs_dqblk *)ddq; + xfs_lsn_t lsn = be64_to_cpu(dqb->dd_lsn); + + if (lsn && lsn != -1 && XFS_LSN_CMP(lsn, current_lsn) >= 0) { + goto out_release; + } + } + memcpy(ddq, recddq, item->ri_buf[1].i_len); if (xfs_sb_version_hascrc(&mp->m_sb)) { xfs_update_cksum((char *)ddq, sizeof(struct xfs_dqblk), @@ -2906,9 +3057,10 @@ xlog_recover_dquot_pass2( ASSERT(bp->b_target->bt_mount == mp); bp->b_iodone = xlog_recover_iodone; xfs_buf_delwri_queue(bp, buffer_list); - xfs_buf_relse(bp); - return (0); +out_release: + xfs_buf_relse(bp); + return 0; } /* @@ -3116,6 +3268,106 @@ xlog_recover_free_trans( kmem_free(trans); } +STATIC void +xlog_recover_buffer_ra_pass2( + struct xlog *log, + struct xlog_recover_item *item) +{ + struct xfs_buf_log_format *buf_f = item->ri_buf[0].i_addr; + struct xfs_mount *mp = log->l_mp; + + if (xlog_peek_buffer_cancelled(log, buf_f->blf_blkno, + buf_f->blf_len, buf_f->blf_flags)) { + return; + } + + xfs_buf_readahead(mp->m_ddev_targp, buf_f->blf_blkno, + buf_f->blf_len, NULL); +} + +STATIC void +xlog_recover_inode_ra_pass2( + struct xlog *log, + struct xlog_recover_item *item) +{ + struct xfs_inode_log_format ilf_buf; + struct xfs_inode_log_format *ilfp; + struct xfs_mount *mp = log->l_mp; + int error; + + if (item->ri_buf[0].i_len == sizeof(struct xfs_inode_log_format)) { + ilfp = item->ri_buf[0].i_addr; + } else { + ilfp = &ilf_buf; + memset(ilfp, 0, sizeof(*ilfp)); + error = xfs_inode_item_format_convert(&item->ri_buf[0], ilfp); + if (error) + return; + } + + if (xlog_peek_buffer_cancelled(log, ilfp->ilf_blkno, ilfp->ilf_len, 0)) + return; + + xfs_buf_readahead(mp->m_ddev_targp, ilfp->ilf_blkno, + ilfp->ilf_len, &xfs_inode_buf_ra_ops); +} + +STATIC void +xlog_recover_dquot_ra_pass2( + struct xlog *log, + struct xlog_recover_item *item) +{ + struct xfs_mount *mp = log->l_mp; + struct xfs_disk_dquot *recddq; + struct xfs_dq_logformat *dq_f; + uint type; + + + if (mp->m_qflags == 0) + return; + + recddq = item->ri_buf[1].i_addr; + if (recddq == NULL) + return; + if (item->ri_buf[1].i_len < sizeof(struct xfs_disk_dquot)) + return; + + type = recddq->d_flags & (XFS_DQ_USER | XFS_DQ_PROJ | XFS_DQ_GROUP); + ASSERT(type); + if (log->l_quotaoffs_flag & type) + return; + + dq_f = item->ri_buf[0].i_addr; + ASSERT(dq_f); + ASSERT(dq_f->qlf_len == 1); + + xfs_buf_readahead(mp->m_ddev_targp, dq_f->qlf_blkno, + XFS_FSB_TO_BB(mp, dq_f->qlf_len), NULL); +} + +STATIC void +xlog_recover_ra_pass2( + struct xlog *log, + struct xlog_recover_item *item) +{ + switch (ITEM_TYPE(item)) { + case XFS_LI_BUF: + xlog_recover_buffer_ra_pass2(log, item); + break; + case XFS_LI_INODE: + xlog_recover_inode_ra_pass2(log, item); + break; + case XFS_LI_DQUOT: + xlog_recover_dquot_ra_pass2(log, item); + break; + case XFS_LI_EFI: + case XFS_LI_EFD: + case XFS_LI_QUOTAOFF: + default: + break; + } +} + STATIC int xlog_recover_commit_pass1( struct xlog *log, @@ -3155,15 +3407,18 @@ xlog_recover_commit_pass2( switch (ITEM_TYPE(item)) { case XFS_LI_BUF: - return xlog_recover_buffer_pass2(log, buffer_list, item); + return xlog_recover_buffer_pass2(log, buffer_list, item, + trans->r_lsn); case XFS_LI_INODE: - return xlog_recover_inode_pass2(log, buffer_list, item); + return xlog_recover_inode_pass2(log, buffer_list, item, + trans->r_lsn); case XFS_LI_EFI: return xlog_recover_efi_pass2(log, item, trans->r_lsn); case XFS_LI_EFD: return xlog_recover_efd_pass2(log, item); case XFS_LI_DQUOT: - return xlog_recover_dquot_pass2(log, buffer_list, item); + return xlog_recover_dquot_pass2(log, buffer_list, item, + trans->r_lsn); case XFS_LI_ICREATE: return xlog_recover_do_icreate_pass2(log, buffer_list, item); case XFS_LI_QUOTAOFF: @@ -3177,6 +3432,26 @@ xlog_recover_commit_pass2( } } +STATIC int +xlog_recover_items_pass2( + struct xlog *log, + struct xlog_recover *trans, + struct list_head *buffer_list, + struct list_head *item_list) +{ + struct xlog_recover_item *item; + int error = 0; + + list_for_each_entry(item, item_list, ri_list) { + error = xlog_recover_commit_pass2(log, trans, + buffer_list, item); + if (error) + return error; + } + + return error; +} + /* * Perform the transaction. * @@ -3189,9 +3464,16 @@ xlog_recover_commit_trans( struct xlog_recover *trans, int pass) { - int error = 0, error2; - xlog_recover_item_t *item; - LIST_HEAD (buffer_list); + int error = 0; + int error2; + int items_queued = 0; + struct xlog_recover_item *item; + struct xlog_recover_item *next; + LIST_HEAD (buffer_list); + LIST_HEAD (ra_list); + LIST_HEAD (done_list); + + #define XLOG_RECOVER_COMMIT_QUEUE_MAX 100 hlist_del(&trans->r_list); @@ -3199,14 +3481,22 @@ xlog_recover_commit_trans( if (error) return error; - list_for_each_entry(item, &trans->r_itemq, ri_list) { + list_for_each_entry_safe(item, next, &trans->r_itemq, ri_list) { switch (pass) { case XLOG_RECOVER_PASS1: error = xlog_recover_commit_pass1(log, trans, item); break; case XLOG_RECOVER_PASS2: - error = xlog_recover_commit_pass2(log, trans, - &buffer_list, item); + xlog_recover_ra_pass2(log, item); + list_move_tail(&item->ri_list, &ra_list); + items_queued++; + if (items_queued >= XLOG_RECOVER_COMMIT_QUEUE_MAX) { + error = xlog_recover_items_pass2(log, trans, + &buffer_list, &ra_list); + list_splice_tail_init(&ra_list, &done_list); + items_queued = 0; + } + break; default: ASSERT(0); @@ -3216,9 +3506,19 @@ xlog_recover_commit_trans( goto out; } +out: + if (!list_empty(&ra_list)) { + if (!error) + error = xlog_recover_items_pass2(log, trans, + &buffer_list, &ra_list); + list_splice_tail_init(&ra_list, &done_list); + } + + if (!list_empty(&done_list)) + list_splice_init(&done_list, &trans->r_itemq); + xlog_recover_free_trans(trans); -out: error2 = xfs_buf_delwri_submit(&buffer_list); return error ? error : error2; } @@ -3376,7 +3676,7 @@ xlog_recover_process_efi( } tp = xfs_trans_alloc(mp, 0); - error = xfs_trans_reserve(tp, 0, XFS_ITRUNCATE_LOG_RES(mp), 0, 0, 0); + error = xfs_trans_reserve(tp, &M_RES(mp)->tr_itruncate, 0, 0); if (error) goto abort_error; efdp = xfs_trans_get_efd(tp, efip, efip->efi_format.efi_nextents); @@ -3482,8 +3782,7 @@ xlog_recover_clear_agi_bucket( int error; tp = xfs_trans_alloc(mp, XFS_TRANS_CLEAR_AGI_BUCKET); - error = xfs_trans_reserve(tp, 0, XFS_CLEAR_AGI_BUCKET_LOG_RES(mp), - 0, 0, 0); + error = xfs_trans_reserve(tp, &M_RES(mp)->tr_clearagi, 0, 0); if (error) goto out_abort; diff --git a/fs/xfs/xfs_log_rlimit.c b/fs/xfs/xfs_log_rlimit.c new file mode 100644 index 0000000..bbcec0b --- /dev/null +++ b/fs/xfs/xfs_log_rlimit.c @@ -0,0 +1,147 @@ +/* + * Copyright (c) 2013 Jie Liu. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#include "xfs.h" +#include "xfs_fs.h" +#include "xfs_log.h" +#include "xfs_trans.h" +#include "xfs_ag.h" +#include "xfs_sb.h" +#include "xfs_mount.h" +#include "xfs_trans_space.h" +#include "xfs_bmap_btree.h" +#include "xfs_inode.h" +#include "xfs_da_btree.h" +#include "xfs_attr_leaf.h" + +/* + * Calculate the maximum length in bytes that would be required for a local + * attribute value as large attributes out of line are not logged. + */ +STATIC int +xfs_log_calc_max_attrsetm_res( + struct xfs_mount *mp) +{ + int size; + int nblks; + + size = xfs_attr_leaf_entsize_local_max(mp->m_sb.sb_blocksize) - + MAXNAMELEN - 1; + nblks = XFS_DAENTER_SPACE_RES(mp, XFS_ATTR_FORK); + nblks += XFS_B_TO_FSB(mp, size); + nblks += XFS_NEXTENTADD_SPACE_RES(mp, size, XFS_ATTR_FORK); + + return M_RES(mp)->tr_attrsetm.tr_logres + + M_RES(mp)->tr_attrsetrt.tr_logres * nblks; +} + +/* + * Iterate over the log space reservation table to figure out and return + * the maximum one in terms of the pre-calculated values which were done + * at mount time. + */ +STATIC void +xfs_log_get_max_trans_res( + struct xfs_mount *mp, + struct xfs_trans_res *max_resp) +{ + struct xfs_trans_res *resp; + struct xfs_trans_res *end_resp; + int log_space = 0; + int attr_space; + + attr_space = xfs_log_calc_max_attrsetm_res(mp); + + resp = (struct xfs_trans_res *)M_RES(mp); + end_resp = (struct xfs_trans_res *)(M_RES(mp) + 1); + for (; resp < end_resp; resp++) { + int tmp = resp->tr_logcount > 1 ? + resp->tr_logres * resp->tr_logcount : + resp->tr_logres; + if (log_space < tmp) { + log_space = tmp; + *max_resp = *resp; /* struct copy */ + } + } + + if (attr_space > log_space) { + *max_resp = M_RES(mp)->tr_attrsetm; /* struct copy */ + max_resp->tr_logres = attr_space; + } +} + +/* + * Calculate the minimum valid log size for the given superblock configuration. + * Used to calculate the minimum log size at mkfs time, and to determine if + * the log is large enough or not at mount time. Returns the minimum size in + * filesystem block size units. + */ +int +xfs_log_calc_minimum_size( + struct xfs_mount *mp) +{ + struct xfs_trans_res tres = {0}; + int max_logres; + int min_logblks = 0; + int lsunit = 0; + + xfs_log_get_max_trans_res(mp, &tres); + + max_logres = xfs_log_calc_unit_res(mp, tres.tr_logres); + if (tres.tr_logcount > 1) + max_logres *= tres.tr_logcount; + + if (xfs_sb_version_haslogv2(&mp->m_sb) && mp->m_sb.sb_logsunit > 1) + lsunit = BTOBB(mp->m_sb.sb_logsunit); + + /* + * Two factors should be taken into account for calculating the minimum + * log space. + * 1) The fundamental limitation is that no single transaction can be + * larger than half size of the log. + * + * From mkfs.xfs, this is considered by the XFS_MIN_LOG_FACTOR + * define, which is set to 3. That means we can definitely fit + * maximally sized 2 transactions in the log. We'll use this same + * value here. + * + * 2) If the lsunit option is specified, a transaction requires 2 LSU + * for the reservation because there are two log writes that can + * require padding - the transaction data and the commit record which + * are written separately and both can require padding to the LSU. + * Consider that we can have an active CIL reservation holding 2*LSU, + * but the CIL is not over a push threshold, in this case, if we + * don't have enough log space for at one new transaction, which + * includes another 2*LSU in the reservation, we will run into dead + * loop situation in log space grant procedure. i.e. + * xlog_grant_head_wait(). + * + * Hence the log size needs to be able to contain two maximally sized + * and padded transactions, which is (2 * (2 * LSU + maxlres)). + * + * Also, the log size should be a multiple of the log stripe unit, round + * it up to lsunit boundary if lsunit is specified. + */ + if (lsunit) { + min_logblks = roundup_64(BTOBB(max_logres), lsunit) + + 2 * lsunit; + } else + min_logblks = BTOBB(max_logres) + 2 * BBSIZE; + min_logblks *= XFS_MIN_LOG_FACTOR; + + return XFS_BB_TO_FSB(mp, min_logblks); +} diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c index 2b0ba35..5dcc680 100644 --- a/fs/xfs/xfs_mount.c +++ b/fs/xfs/xfs_mount.c @@ -17,7 +17,7 @@ */ #include "xfs.h" #include "xfs_fs.h" -#include "xfs_types.h" +#include "xfs_format.h" #include "xfs_bit.h" #include "xfs_log.h" #include "xfs_inum.h" @@ -25,8 +25,10 @@ #include "xfs_trans_priv.h" #include "xfs_sb.h" #include "xfs_ag.h" -#include "xfs_dir2.h" #include "xfs_mount.h" +#include "xfs_da_btree.h" +#include "xfs_dir2_format.h" +#include "xfs_dir2.h" #include "xfs_bmap_btree.h" #include "xfs_alloc_btree.h" #include "xfs_ialloc_btree.h" @@ -40,7 +42,6 @@ #include "xfs_error.h" #include "xfs_quota.h" #include "xfs_fsops.h" -#include "xfs_utils.h" #include "xfs_trace.h" #include "xfs_icache.h" #include "xfs_cksum.h" @@ -59,69 +60,6 @@ STATIC void xfs_icsb_disable_counter(xfs_mount_t *, xfs_sb_field_t); #define xfs_icsb_balance_counter_locked(mp, a, b) do { } while (0) #endif -static const struct { - short offset; - short type; /* 0 = integer - * 1 = binary / string (no translation) - */ -} xfs_sb_info[] = { - { offsetof(xfs_sb_t, sb_magicnum), 0 }, - { offsetof(xfs_sb_t, sb_blocksize), 0 }, - { offsetof(xfs_sb_t, sb_dblocks), 0 }, - { offsetof(xfs_sb_t, sb_rblocks), 0 }, - { offsetof(xfs_sb_t, sb_rextents), 0 }, - { offsetof(xfs_sb_t, sb_uuid), 1 }, - { offsetof(xfs_sb_t, sb_logstart), 0 }, - { offsetof(xfs_sb_t, sb_rootino), 0 }, - { offsetof(xfs_sb_t, sb_rbmino), 0 }, - { offsetof(xfs_sb_t, sb_rsumino), 0 }, - { offsetof(xfs_sb_t, sb_rextsize), 0 }, - { offsetof(xfs_sb_t, sb_agblocks), 0 }, - { offsetof(xfs_sb_t, sb_agcount), 0 }, - { offsetof(xfs_sb_t, sb_rbmblocks), 0 }, - { offsetof(xfs_sb_t, sb_logblocks), 0 }, - { offsetof(xfs_sb_t, sb_versionnum), 0 }, - { offsetof(xfs_sb_t, sb_sectsize), 0 }, - { offsetof(xfs_sb_t, sb_inodesize), 0 }, - { offsetof(xfs_sb_t, sb_inopblock), 0 }, - { offsetof(xfs_sb_t, sb_fname[0]), 1 }, - { offsetof(xfs_sb_t, sb_blocklog), 0 }, - { offsetof(xfs_sb_t, sb_sectlog), 0 }, - { offsetof(xfs_sb_t, sb_inodelog), 0 }, - { offsetof(xfs_sb_t, sb_inopblog), 0 }, - { offsetof(xfs_sb_t, sb_agblklog), 0 }, - { offsetof(xfs_sb_t, sb_rextslog), 0 }, - { offsetof(xfs_sb_t, sb_inprogress), 0 }, - { offsetof(xfs_sb_t, sb_imax_pct), 0 }, - { offsetof(xfs_sb_t, sb_icount), 0 }, - { offsetof(xfs_sb_t, sb_ifree), 0 }, - { offsetof(xfs_sb_t, sb_fdblocks), 0 }, - { offsetof(xfs_sb_t, sb_frextents), 0 }, - { offsetof(xfs_sb_t, sb_uquotino), 0 }, - { offsetof(xfs_sb_t, sb_gquotino), 0 }, - { offsetof(xfs_sb_t, sb_qflags), 0 }, - { offsetof(xfs_sb_t, sb_flags), 0 }, - { offsetof(xfs_sb_t, sb_shared_vn), 0 }, - { offsetof(xfs_sb_t, sb_inoalignmt), 0 }, - { offsetof(xfs_sb_t, sb_unit), 0 }, - { offsetof(xfs_sb_t, sb_width), 0 }, - { offsetof(xfs_sb_t, sb_dirblklog), 0 }, - { offsetof(xfs_sb_t, sb_logsectlog), 0 }, - { offsetof(xfs_sb_t, sb_logsectsize),0 }, - { offsetof(xfs_sb_t, sb_logsunit), 0 }, - { offsetof(xfs_sb_t, sb_features2), 0 }, - { offsetof(xfs_sb_t, sb_bad_features2), 0 }, - { offsetof(xfs_sb_t, sb_features_compat), 0 }, - { offsetof(xfs_sb_t, sb_features_ro_compat), 0 }, - { offsetof(xfs_sb_t, sb_features_incompat), 0 }, - { offsetof(xfs_sb_t, sb_features_log_incompat), 0 }, - { offsetof(xfs_sb_t, sb_crc), 0 }, - { offsetof(xfs_sb_t, sb_pad), 0 }, - { offsetof(xfs_sb_t, sb_pquotino), 0 }, - { offsetof(xfs_sb_t, sb_lsn), 0 }, - { sizeof(xfs_sb_t), 0 } -}; - static DEFINE_MUTEX(xfs_uuid_table_mutex); static int xfs_uuid_table_size; static uuid_t *xfs_uuid_table; @@ -197,64 +135,6 @@ xfs_uuid_unmount( } -/* - * Reference counting access wrappers to the perag structures. - * Because we never free per-ag structures, the only thing we - * have to protect against changes is the tree structure itself. - */ -struct xfs_perag * -xfs_perag_get(struct xfs_mount *mp, xfs_agnumber_t agno) -{ - struct xfs_perag *pag; - int ref = 0; - - rcu_read_lock(); - pag = radix_tree_lookup(&mp->m_perag_tree, agno); - if (pag) { - ASSERT(atomic_read(&pag->pag_ref) >= 0); - ref = atomic_inc_return(&pag->pag_ref); - } - rcu_read_unlock(); - trace_xfs_perag_get(mp, agno, ref, _RET_IP_); - return pag; -} - -/* - * search from @first to find the next perag with the given tag set. - */ -struct xfs_perag * -xfs_perag_get_tag( - struct xfs_mount *mp, - xfs_agnumber_t first, - int tag) -{ - struct xfs_perag *pag; - int found; - int ref; - - rcu_read_lock(); - found = radix_tree_gang_lookup_tag(&mp->m_perag_tree, - (void **)&pag, first, 1, tag); - if (found <= 0) { - rcu_read_unlock(); - return NULL; - } - ref = atomic_inc_return(&pag->pag_ref); - rcu_read_unlock(); - trace_xfs_perag_get_tag(mp, pag->pag_agno, ref, _RET_IP_); - return pag; -} - -void -xfs_perag_put(struct xfs_perag *pag) -{ - int ref; - - ASSERT(atomic_read(&pag->pag_ref) > 0); - ref = atomic_dec_return(&pag->pag_ref); - trace_xfs_perag_put(pag->pag_mount, pag->pag_agno, ref, _RET_IP_); -} - STATIC void __xfs_free_perag( struct rcu_head *head) @@ -307,184 +187,6 @@ xfs_sb_validate_fsb_count( return 0; } -/* - * Check the validity of the SB found. - */ -STATIC int -xfs_mount_validate_sb( - xfs_mount_t *mp, - xfs_sb_t *sbp, - bool check_inprogress, - bool check_version) -{ - - /* - * If the log device and data device have the - * same device number, the log is internal. - * Consequently, the sb_logstart should be non-zero. If - * we have a zero sb_logstart in this case, we may be trying to mount - * a volume filesystem in a non-volume manner. - */ - if (sbp->sb_magicnum != XFS_SB_MAGIC) { - xfs_warn(mp, "bad magic number"); - return XFS_ERROR(EWRONGFS); - } - - - if (!xfs_sb_good_version(sbp)) { - xfs_warn(mp, "bad version"); - return XFS_ERROR(EWRONGFS); - } - - if ((sbp->sb_qflags & (XFS_OQUOTA_ENFD | XFS_OQUOTA_CHKD)) && - (sbp->sb_qflags & (XFS_PQUOTA_ENFD | XFS_GQUOTA_ENFD | - XFS_PQUOTA_CHKD | XFS_GQUOTA_CHKD))) { - xfs_notice(mp, -"Super block has XFS_OQUOTA bits along with XFS_PQUOTA and/or XFS_GQUOTA bits.\n"); - return XFS_ERROR(EFSCORRUPTED); - } - - /* - * Version 5 superblock feature mask validation. Reject combinations the - * kernel cannot support up front before checking anything else. For - * write validation, we don't need to check feature masks. - */ - if (check_version && XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5) { - xfs_alert(mp, -"Version 5 superblock detected. This kernel has EXPERIMENTAL support enabled!\n" -"Use of these features in this kernel is at your own risk!"); - - if (xfs_sb_has_compat_feature(sbp, - XFS_SB_FEAT_COMPAT_UNKNOWN)) { - xfs_warn(mp, -"Superblock has unknown compatible features (0x%x) enabled.\n" -"Using a more recent kernel is recommended.", - (sbp->sb_features_compat & - XFS_SB_FEAT_COMPAT_UNKNOWN)); - } - - if (xfs_sb_has_ro_compat_feature(sbp, - XFS_SB_FEAT_RO_COMPAT_UNKNOWN)) { - xfs_alert(mp, -"Superblock has unknown read-only compatible features (0x%x) enabled.", - (sbp->sb_features_ro_compat & - XFS_SB_FEAT_RO_COMPAT_UNKNOWN)); - if (!(mp->m_flags & XFS_MOUNT_RDONLY)) { - xfs_warn(mp, -"Attempted to mount read-only compatible filesystem read-write.\n" -"Filesystem can only be safely mounted read only."); - return XFS_ERROR(EINVAL); - } - } - if (xfs_sb_has_incompat_feature(sbp, - XFS_SB_FEAT_INCOMPAT_UNKNOWN)) { - xfs_warn(mp, -"Superblock has unknown incompatible features (0x%x) enabled.\n" -"Filesystem can not be safely mounted by this kernel.", - (sbp->sb_features_incompat & - XFS_SB_FEAT_INCOMPAT_UNKNOWN)); - return XFS_ERROR(EINVAL); - } - } - - if (unlikely( - sbp->sb_logstart == 0 && mp->m_logdev_targp == mp->m_ddev_targp)) { - xfs_warn(mp, - "filesystem is marked as having an external log; " - "specify logdev on the mount command line."); - return XFS_ERROR(EINVAL); - } - - if (unlikely( - sbp->sb_logstart != 0 && mp->m_logdev_targp != mp->m_ddev_targp)) { - xfs_warn(mp, - "filesystem is marked as having an internal log; " - "do not specify logdev on the mount command line."); - return XFS_ERROR(EINVAL); - } - - /* - * More sanity checking. Most of these were stolen directly from - * xfs_repair. - */ - if (unlikely( - sbp->sb_agcount <= 0 || - sbp->sb_sectsize < XFS_MIN_SECTORSIZE || - sbp->sb_sectsize > XFS_MAX_SECTORSIZE || - sbp->sb_sectlog < XFS_MIN_SECTORSIZE_LOG || - sbp->sb_sectlog > XFS_MAX_SECTORSIZE_LOG || - sbp->sb_sectsize != (1 << sbp->sb_sectlog) || - sbp->sb_blocksize < XFS_MIN_BLOCKSIZE || - sbp->sb_blocksize > XFS_MAX_BLOCKSIZE || - sbp->sb_blocklog < XFS_MIN_BLOCKSIZE_LOG || - sbp->sb_blocklog > XFS_MAX_BLOCKSIZE_LOG || - sbp->sb_blocksize != (1 << sbp->sb_blocklog) || - sbp->sb_inodesize < XFS_DINODE_MIN_SIZE || - sbp->sb_inodesize > XFS_DINODE_MAX_SIZE || - sbp->sb_inodelog < XFS_DINODE_MIN_LOG || - sbp->sb_inodelog > XFS_DINODE_MAX_LOG || - sbp->sb_inodesize != (1 << sbp->sb_inodelog) || - (sbp->sb_blocklog - sbp->sb_inodelog != sbp->sb_inopblog) || - (sbp->sb_rextsize * sbp->sb_blocksize > XFS_MAX_RTEXTSIZE) || - (sbp->sb_rextsize * sbp->sb_blocksize < XFS_MIN_RTEXTSIZE) || - (sbp->sb_imax_pct > 100 /* zero sb_imax_pct is valid */) || - sbp->sb_dblocks == 0 || - sbp->sb_dblocks > XFS_MAX_DBLOCKS(sbp) || - sbp->sb_dblocks < XFS_MIN_DBLOCKS(sbp))) { - XFS_CORRUPTION_ERROR("SB sanity check failed", - XFS_ERRLEVEL_LOW, mp, sbp); - return XFS_ERROR(EFSCORRUPTED); - } - - /* - * Until this is fixed only page-sized or smaller data blocks work. - */ - if (unlikely(sbp->sb_blocksize > PAGE_SIZE)) { - xfs_warn(mp, - "File system with blocksize %d bytes. " - "Only pagesize (%ld) or less will currently work.", - sbp->sb_blocksize, PAGE_SIZE); - return XFS_ERROR(ENOSYS); - } - - /* - * Currently only very few inode sizes are supported. - */ - switch (sbp->sb_inodesize) { - case 256: - case 512: - case 1024: - case 2048: - break; - default: - xfs_warn(mp, "inode size of %d bytes not supported", - sbp->sb_inodesize); - return XFS_ERROR(ENOSYS); - } - - if (xfs_sb_validate_fsb_count(sbp, sbp->sb_dblocks) || - xfs_sb_validate_fsb_count(sbp, sbp->sb_rblocks)) { - xfs_warn(mp, - "file system too large to be mounted on this system."); - return XFS_ERROR(EFBIG); - } - - if (check_inprogress && sbp->sb_inprogress) { - xfs_warn(mp, "Offline file system operation in progress!"); - return XFS_ERROR(EFSCORRUPTED); - } - - /* - * Version 1 directory format has never worked on Linux. - */ - if (unlikely(!xfs_sb_version_hasdirv2(sbp))) { - xfs_warn(mp, "file system using version 1 directory format"); - return XFS_ERROR(ENOSYS); - } - - return 0; -} - int xfs_initialize_perag( xfs_mount_t *mp, @@ -569,283 +271,15 @@ out_unwind: return error; } -static void -xfs_sb_quota_from_disk(struct xfs_sb *sbp) -{ - if (sbp->sb_qflags & XFS_OQUOTA_ENFD) - sbp->sb_qflags |= (sbp->sb_qflags & XFS_PQUOTA_ACCT) ? - XFS_PQUOTA_ENFD : XFS_GQUOTA_ENFD; - if (sbp->sb_qflags & XFS_OQUOTA_CHKD) - sbp->sb_qflags |= (sbp->sb_qflags & XFS_PQUOTA_ACCT) ? - XFS_PQUOTA_CHKD : XFS_GQUOTA_CHKD; - sbp->sb_qflags &= ~(XFS_OQUOTA_ENFD | XFS_OQUOTA_CHKD); -} - -void -xfs_sb_from_disk( - struct xfs_sb *to, - xfs_dsb_t *from) -{ - to->sb_magicnum = be32_to_cpu(from->sb_magicnum); - to->sb_blocksize = be32_to_cpu(from->sb_blocksize); - to->sb_dblocks = be64_to_cpu(from->sb_dblocks); - to->sb_rblocks = be64_to_cpu(from->sb_rblocks); - to->sb_rextents = be64_to_cpu(from->sb_rextents); - memcpy(&to->sb_uuid, &from->sb_uuid, sizeof(to->sb_uuid)); - to->sb_logstart = be64_to_cpu(from->sb_logstart); - to->sb_rootino = be64_to_cpu(from->sb_rootino); - to->sb_rbmino = be64_to_cpu(from->sb_rbmino); - to->sb_rsumino = be64_to_cpu(from->sb_rsumino); - to->sb_rextsize = be32_to_cpu(from->sb_rextsize); - to->sb_agblocks = be32_to_cpu(from->sb_agblocks); - to->sb_agcount = be32_to_cpu(from->sb_agcount); - to->sb_rbmblocks = be32_to_cpu(from->sb_rbmblocks); - to->sb_logblocks = be32_to_cpu(from->sb_logblocks); - to->sb_versionnum = be16_to_cpu(from->sb_versionnum); - to->sb_sectsize = be16_to_cpu(from->sb_sectsize); - to->sb_inodesize = be16_to_cpu(from->sb_inodesize); - to->sb_inopblock = be16_to_cpu(from->sb_inopblock); - memcpy(&to->sb_fname, &from->sb_fname, sizeof(to->sb_fname)); - to->sb_blocklog = from->sb_blocklog; - to->sb_sectlog = from->sb_sectlog; - to->sb_inodelog = from->sb_inodelog; - to->sb_inopblog = from->sb_inopblog; - to->sb_agblklog = from->sb_agblklog; - to->sb_rextslog = from->sb_rextslog; - to->sb_inprogress = from->sb_inprogress; - to->sb_imax_pct = from->sb_imax_pct; - to->sb_icount = be64_to_cpu(from->sb_icount); - to->sb_ifree = be64_to_cpu(from->sb_ifree); - to->sb_fdblocks = be64_to_cpu(from->sb_fdblocks); - to->sb_frextents = be64_to_cpu(from->sb_frextents); - to->sb_uquotino = be64_to_cpu(from->sb_uquotino); - to->sb_gquotino = be64_to_cpu(from->sb_gquotino); - to->sb_qflags = be16_to_cpu(from->sb_qflags); - to->sb_flags = from->sb_flags; - to->sb_shared_vn = from->sb_shared_vn; - to->sb_inoalignmt = be32_to_cpu(from->sb_inoalignmt); - to->sb_unit = be32_to_cpu(from->sb_unit); - to->sb_width = be32_to_cpu(from->sb_width); - to->sb_dirblklog = from->sb_dirblklog; - to->sb_logsectlog = from->sb_logsectlog; - to->sb_logsectsize = be16_to_cpu(from->sb_logsectsize); - to->sb_logsunit = be32_to_cpu(from->sb_logsunit); - to->sb_features2 = be32_to_cpu(from->sb_features2); - to->sb_bad_features2 = be32_to_cpu(from->sb_bad_features2); - to->sb_features_compat = be32_to_cpu(from->sb_features_compat); - to->sb_features_ro_compat = be32_to_cpu(from->sb_features_ro_compat); - to->sb_features_incompat = be32_to_cpu(from->sb_features_incompat); - to->sb_features_log_incompat = - be32_to_cpu(from->sb_features_log_incompat); - to->sb_pad = 0; - to->sb_pquotino = be64_to_cpu(from->sb_pquotino); - to->sb_lsn = be64_to_cpu(from->sb_lsn); -} - -static inline void -xfs_sb_quota_to_disk( - xfs_dsb_t *to, - xfs_sb_t *from, - __int64_t *fields) -{ - __uint16_t qflags = from->sb_qflags; - - if (*fields & XFS_SB_QFLAGS) { - /* - * The in-core version of sb_qflags do not have - * XFS_OQUOTA_* flags, whereas the on-disk version - * does. So, convert incore XFS_{PG}QUOTA_* flags - * to on-disk XFS_OQUOTA_* flags. - */ - qflags &= ~(XFS_PQUOTA_ENFD | XFS_PQUOTA_CHKD | - XFS_GQUOTA_ENFD | XFS_GQUOTA_CHKD); - - if (from->sb_qflags & - (XFS_PQUOTA_ENFD | XFS_GQUOTA_ENFD)) - qflags |= XFS_OQUOTA_ENFD; - if (from->sb_qflags & - (XFS_PQUOTA_CHKD | XFS_GQUOTA_CHKD)) - qflags |= XFS_OQUOTA_CHKD; - to->sb_qflags = cpu_to_be16(qflags); - *fields &= ~XFS_SB_QFLAGS; - } -} - -/* - * Copy in core superblock to ondisk one. - * - * The fields argument is mask of superblock fields to copy. - */ -void -xfs_sb_to_disk( - xfs_dsb_t *to, - xfs_sb_t *from, - __int64_t fields) -{ - xfs_caddr_t to_ptr = (xfs_caddr_t)to; - xfs_caddr_t from_ptr = (xfs_caddr_t)from; - xfs_sb_field_t f; - int first; - int size; - - ASSERT(fields); - if (!fields) - return; - - xfs_sb_quota_to_disk(to, from, &fields); - while (fields) { - f = (xfs_sb_field_t)xfs_lowbit64((__uint64_t)fields); - first = xfs_sb_info[f].offset; - size = xfs_sb_info[f + 1].offset - first; - - ASSERT(xfs_sb_info[f].type == 0 || xfs_sb_info[f].type == 1); - - if (size == 1 || xfs_sb_info[f].type == 1) { - memcpy(to_ptr + first, from_ptr + first, size); - } else { - switch (size) { - case 2: - *(__be16 *)(to_ptr + first) = - cpu_to_be16(*(__u16 *)(from_ptr + first)); - break; - case 4: - *(__be32 *)(to_ptr + first) = - cpu_to_be32(*(__u32 *)(from_ptr + first)); - break; - case 8: - *(__be64 *)(to_ptr + first) = - cpu_to_be64(*(__u64 *)(from_ptr + first)); - break; - default: - ASSERT(0); - } - } - - fields &= ~(1LL << f); - } -} - -static int -xfs_sb_verify( - struct xfs_buf *bp, - bool check_version) -{ - struct xfs_mount *mp = bp->b_target->bt_mount; - struct xfs_sb sb; - - xfs_sb_from_disk(&sb, XFS_BUF_TO_SBP(bp)); - - /* - * Only check the in progress field for the primary superblock as - * mkfs.xfs doesn't clear it from secondary superblocks. - */ - return xfs_mount_validate_sb(mp, &sb, bp->b_bn == XFS_SB_DADDR, - check_version); -} - -/* - * If the superblock has the CRC feature bit set or the CRC field is non-null, - * check that the CRC is valid. We check the CRC field is non-null because a - * single bit error could clear the feature bit and unused parts of the - * superblock are supposed to be zero. Hence a non-null crc field indicates that - * we've potentially lost a feature bit and we should check it anyway. - */ -static void -xfs_sb_read_verify( - struct xfs_buf *bp) -{ - struct xfs_mount *mp = bp->b_target->bt_mount; - struct xfs_dsb *dsb = XFS_BUF_TO_SBP(bp); - int error; - - /* - * open code the version check to avoid needing to convert the entire - * superblock from disk order just to check the version number - */ - if (dsb->sb_magicnum == cpu_to_be32(XFS_SB_MAGIC) && - (((be16_to_cpu(dsb->sb_versionnum) & XFS_SB_VERSION_NUMBITS) == - XFS_SB_VERSION_5) || - dsb->sb_crc != 0)) { - - if (!xfs_verify_cksum(bp->b_addr, be16_to_cpu(dsb->sb_sectsize), - offsetof(struct xfs_sb, sb_crc))) { - error = EFSCORRUPTED; - goto out_error; - } - } - error = xfs_sb_verify(bp, true); - -out_error: - if (error) { - XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr); - xfs_buf_ioerror(bp, error); - } -} - -/* - * We may be probed for a filesystem match, so we may not want to emit - * messages when the superblock buffer is not actually an XFS superblock. - * If we find an XFS superblock, the run a normal, noisy mount because we are - * really going to mount it and want to know about errors. - */ -static void -xfs_sb_quiet_read_verify( - struct xfs_buf *bp) -{ - struct xfs_dsb *dsb = XFS_BUF_TO_SBP(bp); - - - if (dsb->sb_magicnum == cpu_to_be32(XFS_SB_MAGIC)) { - /* XFS filesystem, verify noisily! */ - xfs_sb_read_verify(bp); - return; - } - /* quietly fail */ - xfs_buf_ioerror(bp, EWRONGFS); -} - -static void -xfs_sb_write_verify( - struct xfs_buf *bp) -{ - struct xfs_mount *mp = bp->b_target->bt_mount; - struct xfs_buf_log_item *bip = bp->b_fspriv; - int error; - - error = xfs_sb_verify(bp, false); - if (error) { - XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr); - xfs_buf_ioerror(bp, error); - return; - } - - if (!xfs_sb_version_hascrc(&mp->m_sb)) - return; - - if (bip) - XFS_BUF_TO_SBP(bp)->sb_lsn = cpu_to_be64(bip->bli_item.li_lsn); - - xfs_update_cksum(bp->b_addr, BBTOB(bp->b_length), - offsetof(struct xfs_sb, sb_crc)); -} - -const struct xfs_buf_ops xfs_sb_buf_ops = { - .verify_read = xfs_sb_read_verify, - .verify_write = xfs_sb_write_verify, -}; - -static const struct xfs_buf_ops xfs_sb_quiet_buf_ops = { - .verify_read = xfs_sb_quiet_read_verify, - .verify_write = xfs_sb_write_verify, -}; - /* * xfs_readsb * * Does the initial read of the superblock. */ int -xfs_readsb(xfs_mount_t *mp, int flags) +xfs_readsb( + struct xfs_mount *mp, + int flags) { unsigned int sector_size; struct xfs_buf *bp; @@ -884,8 +318,8 @@ reread: * Initialize the mount structure from the superblock. */ xfs_sb_from_disk(&mp->m_sb, XFS_BUF_TO_SBP(bp)); - xfs_sb_quota_from_disk(&mp->m_sb); + /* * We must be able to do sector-sized and sector-aligned IO. */ @@ -922,107 +356,6 @@ release_buf: return error; } - -/* - * xfs_mount_common - * - * Mount initialization code establishing various mount - * fields from the superblock associated with the given - * mount structure - */ -STATIC void -xfs_mount_common(xfs_mount_t *mp, xfs_sb_t *sbp) -{ - mp->m_agfrotor = mp->m_agirotor = 0; - spin_lock_init(&mp->m_agirotor_lock); - mp->m_maxagi = mp->m_sb.sb_agcount; - mp->m_blkbit_log = sbp->sb_blocklog + XFS_NBBYLOG; - mp->m_blkbb_log = sbp->sb_blocklog - BBSHIFT; - mp->m_sectbb_log = sbp->sb_sectlog - BBSHIFT; - mp->m_agno_log = xfs_highbit32(sbp->sb_agcount - 1) + 1; - mp->m_agino_log = sbp->sb_inopblog + sbp->sb_agblklog; - mp->m_blockmask = sbp->sb_blocksize - 1; - mp->m_blockwsize = sbp->sb_blocksize >> XFS_WORDLOG; - mp->m_blockwmask = mp->m_blockwsize - 1; - - mp->m_alloc_mxr[0] = xfs_allocbt_maxrecs(mp, sbp->sb_blocksize, 1); - mp->m_alloc_mxr[1] = xfs_allocbt_maxrecs(mp, sbp->sb_blocksize, 0); - mp->m_alloc_mnr[0] = mp->m_alloc_mxr[0] / 2; - mp->m_alloc_mnr[1] = mp->m_alloc_mxr[1] / 2; - - mp->m_inobt_mxr[0] = xfs_inobt_maxrecs(mp, sbp->sb_blocksize, 1); - mp->m_inobt_mxr[1] = xfs_inobt_maxrecs(mp, sbp->sb_blocksize, 0); - mp->m_inobt_mnr[0] = mp->m_inobt_mxr[0] / 2; - mp->m_inobt_mnr[1] = mp->m_inobt_mxr[1] / 2; - - mp->m_bmap_dmxr[0] = xfs_bmbt_maxrecs(mp, sbp->sb_blocksize, 1); - mp->m_bmap_dmxr[1] = xfs_bmbt_maxrecs(mp, sbp->sb_blocksize, 0); - mp->m_bmap_dmnr[0] = mp->m_bmap_dmxr[0] / 2; - mp->m_bmap_dmnr[1] = mp->m_bmap_dmxr[1] / 2; - - mp->m_bsize = XFS_FSB_TO_BB(mp, 1); - mp->m_ialloc_inos = (int)MAX((__uint16_t)XFS_INODES_PER_CHUNK, - sbp->sb_inopblock); - mp->m_ialloc_blks = mp->m_ialloc_inos >> sbp->sb_inopblog; -} - -/* - * xfs_initialize_perag_data - * - * Read in each per-ag structure so we can count up the number of - * allocated inodes, free inodes and used filesystem blocks as this - * information is no longer persistent in the superblock. Once we have - * this information, write it into the in-core superblock structure. - */ -STATIC int -xfs_initialize_perag_data(xfs_mount_t *mp, xfs_agnumber_t agcount) -{ - xfs_agnumber_t index; - xfs_perag_t *pag; - xfs_sb_t *sbp = &mp->m_sb; - uint64_t ifree = 0; - uint64_t ialloc = 0; - uint64_t bfree = 0; - uint64_t bfreelst = 0; - uint64_t btree = 0; - int error; - - for (index = 0; index < agcount; index++) { - /* - * read the agf, then the agi. This gets us - * all the information we need and populates the - * per-ag structures for us. - */ - error = xfs_alloc_pagf_init(mp, NULL, index, 0); - if (error) - return error; - - error = xfs_ialloc_pagi_init(mp, NULL, index); - if (error) - return error; - pag = xfs_perag_get(mp, index); - ifree += pag->pagi_freecount; - ialloc += pag->pagi_count; - bfree += pag->pagf_freeblks; - bfreelst += pag->pagf_flcount; - btree += pag->pagf_btreeblks; - xfs_perag_put(pag); - } - /* - * Overwrite incore superblock counters with just-read data - */ - spin_lock(&mp->m_sb_lock); - sbp->sb_ifree = ifree; - sbp->sb_icount = ialloc; - sbp->sb_fdblocks = bfree + bfreelst + btree; - spin_unlock(&mp->m_sb_lock); - - /* Fixup the per-cpu counters as well. */ - xfs_icsb_reinit_counters(mp); - - return 0; -} - /* * Update alignment values based on mount options and sb values */ @@ -1194,7 +527,7 @@ xfs_set_inoalignment(xfs_mount_t *mp) } /* - * Check that the data (and log if separate) are an ok size. + * Check that the data (and log if separate) is an ok size. */ STATIC int xfs_check_sizes(xfs_mount_t *mp) @@ -1264,8 +597,7 @@ xfs_mount_reset_sbqflags( return 0; tp = xfs_trans_alloc(mp, XFS_TRANS_QM_SBCHANGE); - error = xfs_trans_reserve(tp, 0, XFS_QM_SBCHANGE_LOG_RES(mp), - 0, 0, XFS_DEFAULT_LOG_COUNT); + error = xfs_trans_reserve(tp, &M_RES(mp)->tr_qm_sbchange, 0, 0); if (error) { xfs_trans_cancel(tp, 0); xfs_alert(mp, "%s: Superblock update failed!", __func__); @@ -1315,7 +647,7 @@ xfs_mountfs( uint quotaflags = 0; int error = 0; - xfs_mount_common(mp, sbp); + xfs_sb_mount_common(mp, sbp); /* * Check for a mismatched features2 values. Older kernels @@ -1400,7 +732,7 @@ xfs_mountfs( xfs_set_inoalignment(mp); /* - * Check that the data (and log if separate) are an ok size. + * Check that the data (and log if separate) is an ok size. */ error = xfs_check_sizes(mp); if (error) @@ -1738,8 +1070,7 @@ xfs_log_sbcount(xfs_mount_t *mp) return 0; tp = _xfs_trans_alloc(mp, XFS_TRANS_SB_COUNT, KM_SLEEP); - error = xfs_trans_reserve(tp, 0, XFS_SB_LOG_RES(mp), 0, 0, - XFS_DEFAULT_LOG_COUNT); + error = xfs_trans_reserve(tp, &M_RES(mp)->tr_sb, 0, 0); if (error) { xfs_trans_cancel(tp, 0); return error; @@ -1752,49 +1083,7 @@ xfs_log_sbcount(xfs_mount_t *mp) } /* - * xfs_mod_sb() can be used to copy arbitrary changes to the - * in-core superblock into the superblock buffer to be logged. - * It does not provide the higher level of locking that is - * needed to protect the in-core superblock from concurrent - * access. - */ -void -xfs_mod_sb(xfs_trans_t *tp, __int64_t fields) -{ - xfs_buf_t *bp; - int first; - int last; - xfs_mount_t *mp; - xfs_sb_field_t f; - - ASSERT(fields); - if (!fields) - return; - mp = tp->t_mountp; - bp = xfs_trans_getsb(tp, mp, 0); - first = sizeof(xfs_sb_t); - last = 0; - - /* translate/copy */ - - xfs_sb_to_disk(XFS_BUF_TO_SBP(bp), &mp->m_sb, fields); - - /* find modified range */ - f = (xfs_sb_field_t)xfs_highbit64((__uint64_t)fields); - ASSERT((1LL << f) & XFS_SB_MOD_BITS); - last = xfs_sb_info[f + 1].offset - 1; - - f = (xfs_sb_field_t)xfs_lowbit64((__uint64_t)fields); - ASSERT((1LL << f) & XFS_SB_MOD_BITS); - first = xfs_sb_info[f].offset; - - xfs_trans_buf_set_type(tp, bp, XFS_BLFT_SB_BUF); - xfs_trans_log_buf(tp, bp, first, last); -} - - -/* - * xfs_mod_incore_sb_unlocked() is a utility routine common used to apply + * xfs_mod_incore_sb_unlocked() is a utility routine commonly used to apply * a delta to a specified field in the in-core superblock. Simply * switch on the field indicated and apply the delta to that field. * Fields are not allowed to dip below zero, so if the delta would @@ -2101,8 +1390,7 @@ xfs_mount_log_sb( XFS_SB_VERSIONNUM)); tp = xfs_trans_alloc(mp, XFS_TRANS_SB_UNIT); - error = xfs_trans_reserve(tp, 0, XFS_SB_LOG_RES(mp), 0, 0, - XFS_DEFAULT_LOG_COUNT); + error = xfs_trans_reserve(tp, &M_RES(mp)->tr_sb, 0, 0); if (error) { xfs_trans_cancel(tp, 0); return error; @@ -2260,12 +1548,6 @@ xfs_icsb_init_counters( if (mp->m_sb_cnts == NULL) return -ENOMEM; -#ifdef CONFIG_HOTPLUG_CPU - mp->m_icsb_notifier.notifier_call = xfs_icsb_cpu_notify; - mp->m_icsb_notifier.priority = 0; - register_hotcpu_notifier(&mp->m_icsb_notifier); -#endif /* CONFIG_HOTPLUG_CPU */ - for_each_online_cpu(i) { cntp = (xfs_icsb_cnts_t *)per_cpu_ptr(mp->m_sb_cnts, i); memset(cntp, 0, sizeof(xfs_icsb_cnts_t)); @@ -2278,6 +1560,13 @@ xfs_icsb_init_counters( * initial balance kicks us off correctly */ mp->m_icsb_counters = -1; + +#ifdef CONFIG_HOTPLUG_CPU + mp->m_icsb_notifier.notifier_call = xfs_icsb_cpu_notify; + mp->m_icsb_notifier.priority = 0; + register_hotcpu_notifier(&mp->m_icsb_notifier); +#endif /* CONFIG_HOTPLUG_CPU */ + return 0; } diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h index 4e374d4..1fa0584 100644 --- a/fs/xfs/xfs_mount.h +++ b/fs/xfs/xfs_mount.h @@ -18,45 +18,7 @@ #ifndef __XFS_MOUNT_H__ #define __XFS_MOUNT_H__ -typedef struct xfs_trans_reservations { - uint tr_write; /* extent alloc trans */ - uint tr_itruncate; /* truncate trans */ - uint tr_rename; /* rename trans */ - uint tr_link; /* link trans */ - uint tr_remove; /* unlink trans */ - uint tr_symlink; /* symlink trans */ - uint tr_create; /* create trans */ - uint tr_mkdir; /* mkdir trans */ - uint tr_ifree; /* inode free trans */ - uint tr_ichange; /* inode update trans */ - uint tr_growdata; /* fs data section grow trans */ - uint tr_swrite; /* sync write inode trans */ - uint tr_addafork; /* cvt inode to attributed trans */ - uint tr_writeid; /* write setuid/setgid file */ - uint tr_attrinval; /* attr fork buffer invalidation */ - uint tr_attrsetm; /* set/create an attribute at mount time */ - uint tr_attrsetrt; /* set/create an attribute at runtime */ - uint tr_attrrm; /* remove an attribute */ - uint tr_clearagi; /* clear bad agi unlinked ino bucket */ - uint tr_growrtalloc; /* grow realtime allocations */ - uint tr_growrtzero; /* grow realtime zeroing */ - uint tr_growrtfree; /* grow realtime freeing */ - uint tr_qm_sbchange; /* change quota flags */ - uint tr_qm_setqlim; /* adjust quota limits */ - uint tr_qm_dqalloc; /* allocate quota on disk */ - uint tr_qm_quotaoff; /* turn quota off */ - uint tr_qm_equotaoff;/* end of turn quota off */ - uint tr_sb; /* modify superblock */ -} xfs_trans_reservations_t; - -#ifndef __KERNEL__ - -#define xfs_daddr_to_agno(mp,d) \ - ((xfs_agnumber_t)(XFS_BB_TO_FSBT(mp, d) / (mp)->m_sb.sb_agblocks)) -#define xfs_daddr_to_agbno(mp,d) \ - ((xfs_agblock_t)(XFS_BB_TO_FSBT(mp, d) % (mp)->m_sb.sb_agblocks)) - -#else /* __KERNEL__ */ +#ifdef __KERNEL__ struct xlog; struct xfs_inode; @@ -174,7 +136,7 @@ typedef struct xfs_mount { int m_ialloc_blks; /* blocks in inode allocation */ int m_inoalign_mask;/* mask sb_inoalignmt if used */ uint m_qflags; /* quota status flags */ - xfs_trans_reservations_t m_reservations;/* precomputed res values */ + struct xfs_trans_resv m_resv; /* precomputed res values */ __uint64_t m_maxicount; /* maximum inode count */ __uint64_t m_resblks; /* total reserved blocks */ __uint64_t m_resblks_avail;/* available reserved blocks */ @@ -330,14 +292,6 @@ xfs_daddr_to_agbno(struct xfs_mount *mp, xfs_daddr_t d) } /* - * perag get/put wrappers for ref counting - */ -struct xfs_perag *xfs_perag_get(struct xfs_mount *mp, xfs_agnumber_t agno); -struct xfs_perag *xfs_perag_get_tag(struct xfs_mount *mp, xfs_agnumber_t agno, - int tag); -void xfs_perag_put(struct xfs_perag *pag); - -/* * Per-cpu superblock locking functions */ #ifdef HAVE_PERCPU_SB @@ -366,9 +320,63 @@ typedef struct xfs_mod_sb { int64_t msb_delta; /* Change to make to specified field */ } xfs_mod_sb_t; +/* + * Per-ag incore structure, copies of information in agf and agi, to improve the + * performance of allocation group selection. This is defined for the kernel + * only, and hence is defined here instead of in xfs_ag.h. You need the struct + * xfs_mount to be defined to look up a xfs_perag anyway (via mp->m_perag_tree), + * so this doesn't introduce any strange header file dependencies. + */ +typedef struct xfs_perag { + struct xfs_mount *pag_mount; /* owner filesystem */ + xfs_agnumber_t pag_agno; /* AG this structure belongs to */ + atomic_t pag_ref; /* perag reference count */ + char pagf_init; /* this agf's entry is initialized */ + char pagi_init; /* this agi's entry is initialized */ + char pagf_metadata; /* the agf is preferred to be metadata */ + char pagi_inodeok; /* The agi is ok for inodes */ + __uint8_t pagf_levels[XFS_BTNUM_AGF]; + /* # of levels in bno & cnt btree */ + __uint32_t pagf_flcount; /* count of blocks in freelist */ + xfs_extlen_t pagf_freeblks; /* total free blocks */ + xfs_extlen_t pagf_longest; /* longest free space */ + __uint32_t pagf_btreeblks; /* # of blocks held in AGF btrees */ + xfs_agino_t pagi_freecount; /* number of free inodes */ + xfs_agino_t pagi_count; /* number of allocated inodes */ + + /* + * Inode allocation search lookup optimisation. + * If the pagino matches, the search for new inodes + * doesn't need to search the near ones again straight away + */ + xfs_agino_t pagl_pagino; + xfs_agino_t pagl_leftrec; + xfs_agino_t pagl_rightrec; + spinlock_t pagb_lock; /* lock for pagb_tree */ + struct rb_root pagb_tree; /* ordered tree of busy extents */ + + atomic_t pagf_fstrms; /* # of filestreams active in this AG */ + + spinlock_t pag_ici_lock; /* incore inode cache lock */ + struct radix_tree_root pag_ici_root; /* incore inode cache root */ + int pag_ici_reclaimable; /* reclaimable inodes */ + struct mutex pag_ici_reclaim_lock; /* serialisation point */ + unsigned long pag_ici_reclaim_cursor; /* reclaim restart point */ + + /* buffer cache index */ + spinlock_t pag_buf_lock; /* lock for pag_buf_tree */ + struct rb_root pag_buf_tree; /* ordered tree of active buffers */ + + /* for rcu-safe freeing */ + struct rcu_head rcu_head; + int pagb_count; /* pagb slots in use */ +} xfs_perag_t; + extern int xfs_log_sbcount(xfs_mount_t *); extern __uint64_t xfs_default_resblks(xfs_mount_t *mp); extern int xfs_mountfs(xfs_mount_t *mp); +extern int xfs_initialize_perag(xfs_mount_t *mp, xfs_agnumber_t agcount, + xfs_agnumber_t *maxagi); extern void xfs_unmountfs(xfs_mount_t *); extern int xfs_mod_incore_sb(xfs_mount_t *, xfs_sb_field_t, int64_t, int); @@ -387,13 +395,4 @@ extern void xfs_set_low_space_thresholds(struct xfs_mount *); #endif /* __KERNEL__ */ -extern void xfs_sb_calc_crc(struct xfs_buf *); -extern void xfs_mod_sb(struct xfs_trans *, __int64_t); -extern int xfs_initialize_perag(struct xfs_mount *, xfs_agnumber_t, - xfs_agnumber_t *); -extern void xfs_sb_from_disk(struct xfs_sb *, struct xfs_dsb *); -extern void xfs_sb_to_disk(struct xfs_dsb *, struct xfs_sb *, __int64_t); - -extern const struct xfs_buf_ops xfs_sb_buf_ops; - #endif /* __XFS_MOUNT_H__ */ diff --git a/fs/xfs/xfs_qm.c b/fs/xfs/xfs_qm.c index d320794..6218a0a 100644 --- a/fs/xfs/xfs_qm.c +++ b/fs/xfs/xfs_qm.c @@ -17,6 +17,7 @@ */ #include "xfs.h" #include "xfs_fs.h" +#include "xfs_format.h" #include "xfs_bit.h" #include "xfs_log.h" #include "xfs_trans.h" @@ -37,7 +38,6 @@ #include "xfs_attr.h" #include "xfs_buf_item.h" #include "xfs_trans_space.h" -#include "xfs_utils.h" #include "xfs_qm.h" #include "xfs_trace.h" #include "xfs_icache.h" @@ -834,21 +834,52 @@ xfs_qm_qino_alloc( int error; int committed; + *ip = NULL; + /* + * With superblock that doesn't have separate pquotino, we + * share an inode between gquota and pquota. If the on-disk + * superblock has GQUOTA and the filesystem is now mounted + * with PQUOTA, just use sb_gquotino for sb_pquotino and + * vice-versa. + */ + if (!xfs_sb_version_has_pquotino(&mp->m_sb) && + (flags & (XFS_QMOPT_PQUOTA|XFS_QMOPT_GQUOTA))) { + xfs_ino_t ino = NULLFSINO; + + if ((flags & XFS_QMOPT_PQUOTA) && + (mp->m_sb.sb_gquotino != NULLFSINO)) { + ino = mp->m_sb.sb_gquotino; + ASSERT(mp->m_sb.sb_pquotino == NULLFSINO); + } else if ((flags & XFS_QMOPT_GQUOTA) && + (mp->m_sb.sb_pquotino != NULLFSINO)) { + ino = mp->m_sb.sb_pquotino; + ASSERT(mp->m_sb.sb_gquotino == NULLFSINO); + } + if (ino != NULLFSINO) { + error = xfs_iget(mp, NULL, ino, 0, 0, ip); + if (error) + return error; + mp->m_sb.sb_gquotino = NULLFSINO; + mp->m_sb.sb_pquotino = NULLFSINO; + } + } + tp = xfs_trans_alloc(mp, XFS_TRANS_QM_QINOCREATE); - if ((error = xfs_trans_reserve(tp, - XFS_QM_QINOCREATE_SPACE_RES(mp), - XFS_CREATE_LOG_RES(mp), 0, - XFS_TRANS_PERM_LOG_RES, - XFS_CREATE_LOG_COUNT))) { + error = xfs_trans_reserve(tp, &M_RES(mp)->tr_create, + XFS_QM_QINOCREATE_SPACE_RES(mp), 0); + if (error) { xfs_trans_cancel(tp, 0); return error; } - error = xfs_dir_ialloc(&tp, NULL, S_IFREG, 1, 0, 0, 1, ip, &committed); - if (error) { - xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES | - XFS_TRANS_ABORT); - return error; + if (!*ip) { + error = xfs_dir_ialloc(&tp, NULL, S_IFREG, 1, 0, 0, 1, ip, + &committed); + if (error) { + xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES | + XFS_TRANS_ABORT); + return error; + } } /* @@ -860,21 +891,25 @@ xfs_qm_qino_alloc( if (flags & XFS_QMOPT_SBVERSION) { ASSERT(!xfs_sb_version_hasquota(&mp->m_sb)); ASSERT((sbfields & (XFS_SB_VERSIONNUM | XFS_SB_UQUOTINO | - XFS_SB_GQUOTINO | XFS_SB_QFLAGS)) == - (XFS_SB_VERSIONNUM | XFS_SB_UQUOTINO | - XFS_SB_GQUOTINO | XFS_SB_QFLAGS)); + XFS_SB_GQUOTINO | XFS_SB_PQUOTINO | XFS_SB_QFLAGS)) == + (XFS_SB_VERSIONNUM | XFS_SB_UQUOTINO | + XFS_SB_GQUOTINO | XFS_SB_PQUOTINO | + XFS_SB_QFLAGS)); xfs_sb_version_addquota(&mp->m_sb); mp->m_sb.sb_uquotino = NULLFSINO; mp->m_sb.sb_gquotino = NULLFSINO; + mp->m_sb.sb_pquotino = NULLFSINO; - /* qflags will get updated _after_ quotacheck */ - mp->m_sb.sb_qflags = 0; + /* qflags will get updated fully _after_ quotacheck */ + mp->m_sb.sb_qflags = mp->m_qflags & XFS_ALL_QUOTA_ACCT; } if (flags & XFS_QMOPT_UQUOTA) mp->m_sb.sb_uquotino = (*ip)->i_ino; - else + else if (flags & XFS_QMOPT_GQUOTA) mp->m_sb.sb_gquotino = (*ip)->i_ino; + else + mp->m_sb.sb_pquotino = (*ip)->i_ino; spin_unlock(&mp->m_sb_lock); xfs_mod_sb(tp, sbfields); @@ -1484,11 +1519,10 @@ xfs_qm_init_quotainos( if (error) goto error_rele; } - /* XXX: Use gquotino for now */ if (XFS_IS_PQUOTA_ON(mp) && - mp->m_sb.sb_gquotino != NULLFSINO) { - ASSERT(mp->m_sb.sb_gquotino > 0); - error = xfs_iget(mp, NULL, mp->m_sb.sb_gquotino, + mp->m_sb.sb_pquotino != NULLFSINO) { + ASSERT(mp->m_sb.sb_pquotino > 0); + error = xfs_iget(mp, NULL, mp->m_sb.sb_pquotino, 0, 0, &pip); if (error) goto error_rele; @@ -1496,7 +1530,8 @@ xfs_qm_init_quotainos( } else { flags |= XFS_QMOPT_SBVERSION; sbflags |= (XFS_SB_VERSIONNUM | XFS_SB_UQUOTINO | - XFS_SB_GQUOTINO | XFS_SB_QFLAGS); + XFS_SB_GQUOTINO | XFS_SB_PQUOTINO | + XFS_SB_QFLAGS); } /* @@ -1524,9 +1559,8 @@ xfs_qm_init_quotainos( flags &= ~XFS_QMOPT_SBVERSION; } if (XFS_IS_PQUOTA_ON(mp) && pip == NULL) { - /* XXX: Use XFS_SB_GQUOTINO for now */ error = xfs_qm_qino_alloc(mp, &pip, - sbflags | XFS_SB_GQUOTINO, + sbflags | XFS_SB_PQUOTINO, flags | XFS_QMOPT_PQUOTA); if (error) goto error_rele; @@ -1704,8 +1738,7 @@ xfs_qm_write_sb_changes( int error; tp = xfs_trans_alloc(mp, XFS_TRANS_QM_SBCHANGE); - error = xfs_trans_reserve(tp, 0, XFS_QM_SBCHANGE_LOG_RES(mp), - 0, 0, XFS_DEFAULT_LOG_COUNT); + error = xfs_trans_reserve(tp, &M_RES(mp)->tr_qm_sbchange, 0, 0); if (error) { xfs_trans_cancel(tp, 0); return error; @@ -1734,8 +1767,8 @@ xfs_qm_write_sb_changes( int xfs_qm_vop_dqalloc( struct xfs_inode *ip, - uid_t uid, - gid_t gid, + xfs_dqid_t uid, + xfs_dqid_t gid, prid_t prid, uint flags, struct xfs_dquot **O_udqpp, @@ -1782,7 +1815,7 @@ xfs_qm_vop_dqalloc( * holding ilock. */ xfs_iunlock(ip, lockflags); - error = xfs_qm_dqget(mp, NULL, (xfs_dqid_t) uid, + error = xfs_qm_dqget(mp, NULL, uid, XFS_DQ_USER, XFS_QMOPT_DQALLOC | XFS_QMOPT_DOWARN, @@ -1809,7 +1842,7 @@ xfs_qm_vop_dqalloc( if ((flags & XFS_QMOPT_GQUOTA) && XFS_IS_GQUOTA_ON(mp)) { if (ip->i_d.di_gid != gid) { xfs_iunlock(ip, lockflags); - error = xfs_qm_dqget(mp, NULL, (xfs_dqid_t)gid, + error = xfs_qm_dqget(mp, NULL, gid, XFS_DQ_GROUP, XFS_QMOPT_DQALLOC | XFS_QMOPT_DOWARN, @@ -1943,7 +1976,7 @@ xfs_qm_vop_chown_reserve( XFS_QMOPT_RES_RTBLKS : XFS_QMOPT_RES_REGBLKS; if (XFS_IS_UQUOTA_ON(mp) && udqp && - ip->i_d.di_uid != (uid_t)be32_to_cpu(udqp->q_core.d_id)) { + ip->i_d.di_uid != be32_to_cpu(udqp->q_core.d_id)) { udq_delblks = udqp; /* * If there are delayed allocation blocks, then we have to diff --git a/fs/xfs/xfs_qm.h b/fs/xfs/xfs_qm.h index 579d6a0..670cd44 100644 --- a/fs/xfs/xfs_qm.h +++ b/fs/xfs/xfs_qm.h @@ -160,6 +160,8 @@ extern int xfs_qm_scall_setqlim(struct xfs_mount *, xfs_dqid_t, uint, struct fs_disk_quota *); extern int xfs_qm_scall_getqstat(struct xfs_mount *, struct fs_quota_stat *); +extern int xfs_qm_scall_getqstatv(struct xfs_mount *, + struct fs_quota_statv *); extern int xfs_qm_scall_quotaon(struct xfs_mount *, uint); extern int xfs_qm_scall_quotaoff(struct xfs_mount *, uint); diff --git a/fs/xfs/xfs_qm_bhv.c b/fs/xfs/xfs_qm_bhv.c index 437a52d..3af50cc 100644 --- a/fs/xfs/xfs_qm_bhv.c +++ b/fs/xfs/xfs_qm_bhv.c @@ -17,6 +17,7 @@ */ #include "xfs.h" #include "xfs_fs.h" +#include "xfs_format.h" #include "xfs_log.h" #include "xfs_trans.h" #include "xfs_sb.h" diff --git a/fs/xfs/xfs_qm_syscalls.c b/fs/xfs/xfs_qm_syscalls.c index e4f8b2d..8174aad 100644 --- a/fs/xfs/xfs_qm_syscalls.c +++ b/fs/xfs/xfs_qm_syscalls.c @@ -20,6 +20,7 @@ #include "xfs.h" #include "xfs_fs.h" +#include "xfs_format.h" #include "xfs_bit.h" #include "xfs_log.h" #include "xfs_trans.h" @@ -37,7 +38,6 @@ #include "xfs_error.h" #include "xfs_attr.h" #include "xfs_buf_item.h" -#include "xfs_utils.h" #include "xfs_qm.h" #include "xfs_trace.h" #include "xfs_icache.h" @@ -247,9 +247,7 @@ xfs_qm_scall_trunc_qfile( xfs_ilock(ip, XFS_IOLOCK_EXCL); tp = xfs_trans_alloc(mp, XFS_TRANS_TRUNCATE_FILE); - error = xfs_trans_reserve(tp, 0, XFS_ITRUNCATE_LOG_RES(mp), 0, - XFS_TRANS_PERM_LOG_RES, - XFS_ITRUNCATE_LOG_COUNT); + error = xfs_trans_reserve(tp, &M_RES(mp)->tr_itruncate, 0, 0); if (error) { xfs_trans_cancel(tp, 0); xfs_iunlock(ip, XFS_IOLOCK_EXCL); @@ -296,8 +294,10 @@ xfs_qm_scall_trunc_qfiles( if (flags & XFS_DQ_USER) error = xfs_qm_scall_trunc_qfile(mp, mp->m_sb.sb_uquotino); - if (flags & (XFS_DQ_GROUP|XFS_DQ_PROJ)) + if (flags & XFS_DQ_GROUP) error2 = xfs_qm_scall_trunc_qfile(mp, mp->m_sb.sb_gquotino); + if (flags & XFS_DQ_PROJ) + error2 = xfs_qm_scall_trunc_qfile(mp, mp->m_sb.sb_pquotino); return error ? error : error2; } @@ -404,6 +404,7 @@ xfs_qm_scall_quotaon( /* * Return quota status information, such as uquota-off, enforcements, etc. + * for Q_XGETQSTAT command. */ int xfs_qm_scall_getqstat( @@ -413,8 +414,10 @@ xfs_qm_scall_getqstat( struct xfs_quotainfo *q = mp->m_quotainfo; struct xfs_inode *uip = NULL; struct xfs_inode *gip = NULL; + struct xfs_inode *pip = NULL; bool tempuqip = false; bool tempgqip = false; + bool temppqip = false; memset(out, 0, sizeof(fs_quota_stat_t)); @@ -424,16 +427,106 @@ xfs_qm_scall_getqstat( out->qs_gquota.qfs_ino = NULLFSINO; return (0); } + + out->qs_flags = (__uint16_t) xfs_qm_export_flags(mp->m_qflags & + (XFS_ALL_QUOTA_ACCT| + XFS_ALL_QUOTA_ENFD)); + if (q) { + uip = q->qi_uquotaip; + gip = q->qi_gquotaip; + pip = q->qi_pquotaip; + } + if (!uip && mp->m_sb.sb_uquotino != NULLFSINO) { + if (xfs_iget(mp, NULL, mp->m_sb.sb_uquotino, + 0, 0, &uip) == 0) + tempuqip = true; + } + if (!gip && mp->m_sb.sb_gquotino != NULLFSINO) { + if (xfs_iget(mp, NULL, mp->m_sb.sb_gquotino, + 0, 0, &gip) == 0) + tempgqip = true; + } + /* + * Q_XGETQSTAT doesn't have room for both group and project quotas. + * So, allow the project quota values to be copied out only if + * there is no group quota information available. + */ + if (!gip) { + if (!pip && mp->m_sb.sb_pquotino != NULLFSINO) { + if (xfs_iget(mp, NULL, mp->m_sb.sb_pquotino, + 0, 0, &pip) == 0) + temppqip = true; + } + } else + pip = NULL; + if (uip) { + out->qs_uquota.qfs_ino = mp->m_sb.sb_uquotino; + out->qs_uquota.qfs_nblks = uip->i_d.di_nblocks; + out->qs_uquota.qfs_nextents = uip->i_d.di_nextents; + if (tempuqip) + IRELE(uip); + } + + if (gip) { + out->qs_gquota.qfs_ino = mp->m_sb.sb_gquotino; + out->qs_gquota.qfs_nblks = gip->i_d.di_nblocks; + out->qs_gquota.qfs_nextents = gip->i_d.di_nextents; + if (tempgqip) + IRELE(gip); + } + if (pip) { + out->qs_gquota.qfs_ino = mp->m_sb.sb_gquotino; + out->qs_gquota.qfs_nblks = pip->i_d.di_nblocks; + out->qs_gquota.qfs_nextents = pip->i_d.di_nextents; + if (temppqip) + IRELE(pip); + } + if (q) { + out->qs_incoredqs = q->qi_dquots; + out->qs_btimelimit = q->qi_btimelimit; + out->qs_itimelimit = q->qi_itimelimit; + out->qs_rtbtimelimit = q->qi_rtbtimelimit; + out->qs_bwarnlimit = q->qi_bwarnlimit; + out->qs_iwarnlimit = q->qi_iwarnlimit; + } + return 0; +} + +/* + * Return quota status information, such as uquota-off, enforcements, etc. + * for Q_XGETQSTATV command, to support separate project quota field. + */ +int +xfs_qm_scall_getqstatv( + struct xfs_mount *mp, + struct fs_quota_statv *out) +{ + struct xfs_quotainfo *q = mp->m_quotainfo; + struct xfs_inode *uip = NULL; + struct xfs_inode *gip = NULL; + struct xfs_inode *pip = NULL; + bool tempuqip = false; + bool tempgqip = false; + bool temppqip = false; + + if (!xfs_sb_version_hasquota(&mp->m_sb)) { + out->qs_uquota.qfs_ino = NULLFSINO; + out->qs_gquota.qfs_ino = NULLFSINO; + out->qs_pquota.qfs_ino = NULLFSINO; + return (0); + } + out->qs_flags = (__uint16_t) xfs_qm_export_flags(mp->m_qflags & (XFS_ALL_QUOTA_ACCT| XFS_ALL_QUOTA_ENFD)); - out->qs_pad = 0; out->qs_uquota.qfs_ino = mp->m_sb.sb_uquotino; out->qs_gquota.qfs_ino = mp->m_sb.sb_gquotino; + out->qs_pquota.qfs_ino = mp->m_sb.sb_pquotino; if (q) { uip = q->qi_uquotaip; gip = q->qi_gquotaip; + pip = q->qi_pquotaip; } if (!uip && mp->m_sb.sb_uquotino != NULLFSINO) { if (xfs_iget(mp, NULL, mp->m_sb.sb_uquotino, @@ -445,18 +538,30 @@ xfs_qm_scall_getqstat( 0, 0, &gip) == 0) tempgqip = true; } + if (!pip && mp->m_sb.sb_pquotino != NULLFSINO) { + if (xfs_iget(mp, NULL, mp->m_sb.sb_pquotino, + 0, 0, &pip) == 0) + temppqip = true; + } if (uip) { out->qs_uquota.qfs_nblks = uip->i_d.di_nblocks; out->qs_uquota.qfs_nextents = uip->i_d.di_nextents; if (tempuqip) IRELE(uip); } + if (gip) { out->qs_gquota.qfs_nblks = gip->i_d.di_nblocks; out->qs_gquota.qfs_nextents = gip->i_d.di_nextents; if (tempgqip) IRELE(gip); } + if (pip) { + out->qs_pquota.qfs_nblks = pip->i_d.di_nblocks; + out->qs_pquota.qfs_nextents = pip->i_d.di_nextents; + if (temppqip) + IRELE(pip); + } if (q) { out->qs_incoredqs = q->qi_dquots; out->qs_btimelimit = q->qi_btimelimit; @@ -515,8 +620,7 @@ xfs_qm_scall_setqlim( xfs_dqunlock(dqp); tp = xfs_trans_alloc(mp, XFS_TRANS_QM_SETQLIM); - error = xfs_trans_reserve(tp, 0, XFS_QM_SETQLIM_LOG_RES(mp), - 0, 0, XFS_DEFAULT_LOG_COUNT); + error = xfs_trans_reserve(tp, &M_RES(mp)->tr_qm_setqlim, 0, 0); if (error) { xfs_trans_cancel(tp, 0); goto out_rele; @@ -650,8 +754,7 @@ xfs_qm_log_quotaoff_end( tp = xfs_trans_alloc(mp, XFS_TRANS_QM_QUOTAOFF_END); - error = xfs_trans_reserve(tp, 0, XFS_QM_QUOTAOFF_END_LOG_RES(mp), - 0, 0, XFS_DEFAULT_LOG_COUNT); + error = xfs_trans_reserve(tp, &M_RES(mp)->tr_qm_equotaoff, 0, 0); if (error) { xfs_trans_cancel(tp, 0); return (error); @@ -684,8 +787,7 @@ xfs_qm_log_quotaoff( uint oldsbqflag=0; tp = xfs_trans_alloc(mp, XFS_TRANS_QM_QUOTAOFF); - error = xfs_trans_reserve(tp, 0, XFS_QM_QUOTAOFF_LOG_RES(mp), - 0, 0, XFS_DEFAULT_LOG_COUNT); + error = xfs_trans_reserve(tp, &M_RES(mp)->tr_qm_quotaoff, 0, 0); if (error) goto error0; diff --git a/fs/xfs/xfs_quota.h b/fs/xfs/xfs_quota.h index b14f42c..e7d84d2 100644 --- a/fs/xfs/xfs_quota.h +++ b/fs/xfs/xfs_quota.h @@ -18,267 +18,14 @@ #ifndef __XFS_QUOTA_H__ #define __XFS_QUOTA_H__ -struct xfs_trans; - -/* - * The ondisk form of a dquot structure. - */ -#define XFS_DQUOT_MAGIC 0x4451 /* 'DQ' */ -#define XFS_DQUOT_VERSION (u_int8_t)0x01 /* latest version number */ - -/* - * uid_t and gid_t are hard-coded to 32 bits in the inode. - * Hence, an 'id' in a dquot is 32 bits.. - */ -typedef __uint32_t xfs_dqid_t; - -/* - * Even though users may not have quota limits occupying all 64-bits, - * they may need 64-bit accounting. Hence, 64-bit quota-counters, - * and quota-limits. This is a waste in the common case, but hey ... - */ -typedef __uint64_t xfs_qcnt_t; -typedef __uint16_t xfs_qwarncnt_t; - -/* - * This is the main portion of the on-disk representation of quota - * information for a user. This is the q_core of the xfs_dquot_t that - * is kept in kernel memory. We pad this with some more expansion room - * to construct the on disk structure. - */ -typedef struct xfs_disk_dquot { - __be16 d_magic; /* dquot magic = XFS_DQUOT_MAGIC */ - __u8 d_version; /* dquot version */ - __u8 d_flags; /* XFS_DQ_USER/PROJ/GROUP */ - __be32 d_id; /* user,project,group id */ - __be64 d_blk_hardlimit;/* absolute limit on disk blks */ - __be64 d_blk_softlimit;/* preferred limit on disk blks */ - __be64 d_ino_hardlimit;/* maximum # allocated inodes */ - __be64 d_ino_softlimit;/* preferred inode limit */ - __be64 d_bcount; /* disk blocks owned by the user */ - __be64 d_icount; /* inodes owned by the user */ - __be32 d_itimer; /* zero if within inode limits if not, - this is when we refuse service */ - __be32 d_btimer; /* similar to above; for disk blocks */ - __be16 d_iwarns; /* warnings issued wrt num inodes */ - __be16 d_bwarns; /* warnings issued wrt disk blocks */ - __be32 d_pad0; /* 64 bit align */ - __be64 d_rtb_hardlimit;/* absolute limit on realtime blks */ - __be64 d_rtb_softlimit;/* preferred limit on RT disk blks */ - __be64 d_rtbcount; /* realtime blocks owned */ - __be32 d_rtbtimer; /* similar to above; for RT disk blocks */ - __be16 d_rtbwarns; /* warnings issued wrt RT disk blocks */ - __be16 d_pad; -} xfs_disk_dquot_t; - -/* - * This is what goes on disk. This is separated from the xfs_disk_dquot because - * carrying the unnecessary padding would be a waste of memory. - */ -typedef struct xfs_dqblk { - xfs_disk_dquot_t dd_diskdq; /* portion that lives incore as well */ - char dd_fill[4]; /* filling for posterity */ - - /* - * These two are only present on filesystems with the CRC bits set. - */ - __be32 dd_crc; /* checksum */ - __be64 dd_lsn; /* last modification in log */ - uuid_t dd_uuid; /* location information */ -} xfs_dqblk_t; - -#define XFS_DQUOT_CRC_OFF offsetof(struct xfs_dqblk, dd_crc) - -/* - * flags for q_flags field in the dquot. - */ -#define XFS_DQ_USER 0x0001 /* a user quota */ -#define XFS_DQ_PROJ 0x0002 /* project quota */ -#define XFS_DQ_GROUP 0x0004 /* a group quota */ -#define XFS_DQ_DIRTY 0x0008 /* dquot is dirty */ -#define XFS_DQ_FREEING 0x0010 /* dquot is beeing torn down */ - -#define XFS_DQ_ALLTYPES (XFS_DQ_USER|XFS_DQ_PROJ|XFS_DQ_GROUP) - -#define XFS_DQ_FLAGS \ - { XFS_DQ_USER, "USER" }, \ - { XFS_DQ_PROJ, "PROJ" }, \ - { XFS_DQ_GROUP, "GROUP" }, \ - { XFS_DQ_DIRTY, "DIRTY" }, \ - { XFS_DQ_FREEING, "FREEING" } - -/* - * We have the possibility of all three quota types being active at once, and - * hence free space modification requires modification of all three current - * dquots in a single transaction. For this case we need to have a reservation - * of at least 3 dquots. - * - * However, a chmod operation can change both UID and GID in a single - * transaction, resulting in requiring {old, new} x {uid, gid} dquots to be - * modified. Hence for this case we need to reserve space for at least 4 dquots. - * - * And in the worst case, there's a rename operation that can be modifying up to - * 4 inodes with dquots attached to them. In reality, the only inodes that can - * have their dquots modified are the source and destination directory inodes - * due to directory name creation and removal. That can require space allocation - * and/or freeing on both directory inodes, and hence all three dquots on each - * inode can be modified. And if the directories are world writeable, all the - * dquots can be unique and so 6 dquots can be modified.... - * - * And, of course, we also need to take into account the dquot log format item - * used to describe each dquot. - */ -#define XFS_DQUOT_LOGRES(mp) \ - ((sizeof(struct xfs_dq_logformat) + sizeof(struct xfs_disk_dquot)) * 6) - -/* - * These are the structures used to lay out dquots and quotaoff - * records on the log. Quite similar to those of inodes. - */ - -/* - * log format struct for dquots. - * The first two fields must be the type and size fitting into - * 32 bits : log_recovery code assumes that. - */ -typedef struct xfs_dq_logformat { - __uint16_t qlf_type; /* dquot log item type */ - __uint16_t qlf_size; /* size of this item */ - xfs_dqid_t qlf_id; /* usr/grp/proj id : 32 bits */ - __int64_t qlf_blkno; /* blkno of dquot buffer */ - __int32_t qlf_len; /* len of dquot buffer */ - __uint32_t qlf_boffset; /* off of dquot in buffer */ -} xfs_dq_logformat_t; - -/* - * log format struct for QUOTAOFF records. - * The first two fields must be the type and size fitting into - * 32 bits : log_recovery code assumes that. - * We write two LI_QUOTAOFF logitems per quotaoff, the last one keeps a pointer - * to the first and ensures that the first logitem is taken out of the AIL - * only when the last one is securely committed. - */ -typedef struct xfs_qoff_logformat { - unsigned short qf_type; /* quotaoff log item type */ - unsigned short qf_size; /* size of this item */ - unsigned int qf_flags; /* USR and/or GRP */ - char qf_pad[12]; /* padding for future */ -} xfs_qoff_logformat_t; - - -/* - * Disk quotas status in m_qflags, and also sb_qflags. 16 bits. - */ -#define XFS_UQUOTA_ACCT 0x0001 /* user quota accounting ON */ -#define XFS_UQUOTA_ENFD 0x0002 /* user quota limits enforced */ -#define XFS_UQUOTA_CHKD 0x0004 /* quotacheck run on usr quotas */ -#define XFS_PQUOTA_ACCT 0x0008 /* project quota accounting ON */ -#define XFS_OQUOTA_ENFD 0x0010 /* other (grp/prj) quota limits enforced */ -#define XFS_OQUOTA_CHKD 0x0020 /* quotacheck run on other (grp/prj) quotas */ -#define XFS_GQUOTA_ACCT 0x0040 /* group quota accounting ON */ - -/* - * Conversion to and from the combined OQUOTA flag (if necessary) - * is done only in xfs_sb_qflags_to_disk() and xfs_sb_qflags_from_disk() - */ -#define XFS_GQUOTA_ENFD 0x0080 /* group quota limits enforced */ -#define XFS_GQUOTA_CHKD 0x0100 /* quotacheck run on group quotas */ -#define XFS_PQUOTA_ENFD 0x0200 /* project quota limits enforced */ -#define XFS_PQUOTA_CHKD 0x0400 /* quotacheck run on project quotas */ - -/* - * Quota Accounting/Enforcement flags - */ -#define XFS_ALL_QUOTA_ACCT \ - (XFS_UQUOTA_ACCT | XFS_GQUOTA_ACCT | XFS_PQUOTA_ACCT) -#define XFS_ALL_QUOTA_ENFD \ - (XFS_UQUOTA_ENFD | XFS_GQUOTA_ENFD | XFS_PQUOTA_ENFD) -#define XFS_ALL_QUOTA_CHKD \ - (XFS_UQUOTA_CHKD | XFS_GQUOTA_CHKD | XFS_PQUOTA_CHKD) - -#define XFS_IS_QUOTA_RUNNING(mp) ((mp)->m_qflags & XFS_ALL_QUOTA_ACCT) -#define XFS_IS_UQUOTA_RUNNING(mp) ((mp)->m_qflags & XFS_UQUOTA_ACCT) -#define XFS_IS_PQUOTA_RUNNING(mp) ((mp)->m_qflags & XFS_PQUOTA_ACCT) -#define XFS_IS_GQUOTA_RUNNING(mp) ((mp)->m_qflags & XFS_GQUOTA_ACCT) -#define XFS_IS_UQUOTA_ENFORCED(mp) ((mp)->m_qflags & XFS_UQUOTA_ENFD) -#define XFS_IS_GQUOTA_ENFORCED(mp) ((mp)->m_qflags & XFS_GQUOTA_ENFD) -#define XFS_IS_PQUOTA_ENFORCED(mp) ((mp)->m_qflags & XFS_PQUOTA_ENFD) - -/* - * Incore only flags for quotaoff - these bits get cleared when quota(s) - * are in the process of getting turned off. These flags are in m_qflags but - * never in sb_qflags. - */ -#define XFS_UQUOTA_ACTIVE 0x1000 /* uquotas are being turned off */ -#define XFS_GQUOTA_ACTIVE 0x2000 /* gquotas are being turned off */ -#define XFS_PQUOTA_ACTIVE 0x4000 /* pquotas are being turned off */ -#define XFS_ALL_QUOTA_ACTIVE \ - (XFS_UQUOTA_ACTIVE | XFS_GQUOTA_ACTIVE | XFS_PQUOTA_ACTIVE) +#include "xfs_quota_defs.h" /* - * Checking XFS_IS_*QUOTA_ON() while holding any inode lock guarantees - * quota will be not be switched off as long as that inode lock is held. + * Kernel only quota definitions and functions */ -#define XFS_IS_QUOTA_ON(mp) ((mp)->m_qflags & (XFS_UQUOTA_ACTIVE | \ - XFS_GQUOTA_ACTIVE | \ - XFS_PQUOTA_ACTIVE)) -#define XFS_IS_OQUOTA_ON(mp) ((mp)->m_qflags & (XFS_GQUOTA_ACTIVE | \ - XFS_PQUOTA_ACTIVE)) -#define XFS_IS_UQUOTA_ON(mp) ((mp)->m_qflags & XFS_UQUOTA_ACTIVE) -#define XFS_IS_GQUOTA_ON(mp) ((mp)->m_qflags & XFS_GQUOTA_ACTIVE) -#define XFS_IS_PQUOTA_ON(mp) ((mp)->m_qflags & XFS_PQUOTA_ACTIVE) -/* - * Flags to tell various functions what to do. Not all of these are meaningful - * to a single function. None of these XFS_QMOPT_* flags are meant to have - * persistent values (ie. their values can and will change between versions) - */ -#define XFS_QMOPT_DQALLOC 0x0000002 /* alloc dquot ondisk if needed */ -#define XFS_QMOPT_UQUOTA 0x0000004 /* user dquot requested */ -#define XFS_QMOPT_PQUOTA 0x0000008 /* project dquot requested */ -#define XFS_QMOPT_FORCE_RES 0x0000010 /* ignore quota limits */ -#define XFS_QMOPT_SBVERSION 0x0000040 /* change superblock version num */ -#define XFS_QMOPT_DOWARN 0x0000400 /* increase warning cnt if needed */ -#define XFS_QMOPT_DQREPAIR 0x0001000 /* repair dquot if damaged */ -#define XFS_QMOPT_GQUOTA 0x0002000 /* group dquot requested */ -#define XFS_QMOPT_ENOSPC 0x0004000 /* enospc instead of edquot (prj) */ - -/* - * flags to xfs_trans_mod_dquot to indicate which field needs to be - * modified. - */ -#define XFS_QMOPT_RES_REGBLKS 0x0010000 -#define XFS_QMOPT_RES_RTBLKS 0x0020000 -#define XFS_QMOPT_BCOUNT 0x0040000 -#define XFS_QMOPT_ICOUNT 0x0080000 -#define XFS_QMOPT_RTBCOUNT 0x0100000 -#define XFS_QMOPT_DELBCOUNT 0x0200000 -#define XFS_QMOPT_DELRTBCOUNT 0x0400000 -#define XFS_QMOPT_RES_INOS 0x0800000 - -/* - * flags for dqalloc. - */ -#define XFS_QMOPT_INHERIT 0x1000000 - -/* - * flags to xfs_trans_mod_dquot. - */ -#define XFS_TRANS_DQ_RES_BLKS XFS_QMOPT_RES_REGBLKS -#define XFS_TRANS_DQ_RES_RTBLKS XFS_QMOPT_RES_RTBLKS -#define XFS_TRANS_DQ_RES_INOS XFS_QMOPT_RES_INOS -#define XFS_TRANS_DQ_BCOUNT XFS_QMOPT_BCOUNT -#define XFS_TRANS_DQ_DELBCOUNT XFS_QMOPT_DELBCOUNT -#define XFS_TRANS_DQ_ICOUNT XFS_QMOPT_ICOUNT -#define XFS_TRANS_DQ_RTBCOUNT XFS_QMOPT_RTBCOUNT -#define XFS_TRANS_DQ_DELRTBCOUNT XFS_QMOPT_DELRTBCOUNT - - -#define XFS_QMOPT_QUOTALL \ - (XFS_QMOPT_UQUOTA | XFS_QMOPT_PQUOTA | XFS_QMOPT_GQUOTA) -#define XFS_QMOPT_RESBLK_MASK (XFS_QMOPT_RES_REGBLKS | XFS_QMOPT_RES_RTBLKS) +struct xfs_trans; -#ifdef __KERNEL__ /* * This check is done typically without holding the inode lock; * that may seem racy, but it is harmless in the context that it is used. @@ -301,13 +48,6 @@ typedef struct xfs_qoff_logformat { (XFS_IS_PQUOTA_ON(mp) && \ (mp->m_sb.sb_qflags & XFS_PQUOTA_CHKD) == 0)) -#define XFS_MOUNT_QUOTA_ALL (XFS_UQUOTA_ACCT|XFS_UQUOTA_ENFD|\ - XFS_UQUOTA_CHKD|XFS_GQUOTA_ACCT|\ - XFS_GQUOTA_ENFD|XFS_GQUOTA_CHKD|\ - XFS_PQUOTA_ACCT|XFS_PQUOTA_ENFD|\ - XFS_PQUOTA_CHKD) - - /* * The structure kept inside the xfs_trans_t keep track of dquot changes * within a transaction and apply them later. @@ -340,8 +80,9 @@ extern int xfs_trans_reserve_quota_bydquots(struct xfs_trans *, struct xfs_mount *, struct xfs_dquot *, struct xfs_dquot *, struct xfs_dquot *, long, long, uint); -extern int xfs_qm_vop_dqalloc(struct xfs_inode *, uid_t, gid_t, prid_t, uint, - struct xfs_dquot **, struct xfs_dquot **, struct xfs_dquot **); +extern int xfs_qm_vop_dqalloc(struct xfs_inode *, xfs_dqid_t, xfs_dqid_t, + prid_t, uint, struct xfs_dquot **, struct xfs_dquot **, + struct xfs_dquot **); extern void xfs_qm_vop_create_dqattach(struct xfs_trans *, struct xfs_inode *, struct xfs_dquot *, struct xfs_dquot *, struct xfs_dquot *); extern int xfs_qm_vop_rename_dqattach(struct xfs_inode **); @@ -362,9 +103,9 @@ extern void xfs_qm_unmount_quotas(struct xfs_mount *); #else static inline int -xfs_qm_vop_dqalloc(struct xfs_inode *ip, uid_t uid, gid_t gid, prid_t prid, - uint flags, struct xfs_dquot **udqp, struct xfs_dquot **gdqp, - struct xfs_dquot **pdqp) +xfs_qm_vop_dqalloc(struct xfs_inode *ip, xfs_dqid_t uid, xfs_dqid_t gid, + prid_t prid, uint flags, struct xfs_dquot **udqp, + struct xfs_dquot **gdqp, struct xfs_dquot **pdqp) { *udqp = NULL; *gdqp = NULL; @@ -415,5 +156,4 @@ extern int xfs_mount_reset_sbqflags(struct xfs_mount *); extern const struct xfs_buf_ops xfs_dquot_buf_ops; -#endif /* __KERNEL__ */ #endif /* __XFS_QUOTA_H__ */ diff --git a/fs/xfs/xfs_quota_defs.h b/fs/xfs/xfs_quota_defs.h new file mode 100644 index 0000000..e6b0d6e --- /dev/null +++ b/fs/xfs/xfs_quota_defs.h @@ -0,0 +1,157 @@ +/* + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#ifndef __XFS_QUOTA_DEFS_H__ +#define __XFS_QUOTA_DEFS_H__ + +/* + * Quota definitions shared between user and kernel source trees. + */ + +/* + * Even though users may not have quota limits occupying all 64-bits, + * they may need 64-bit accounting. Hence, 64-bit quota-counters, + * and quota-limits. This is a waste in the common case, but hey ... + */ +typedef __uint64_t xfs_qcnt_t; +typedef __uint16_t xfs_qwarncnt_t; + +/* + * flags for q_flags field in the dquot. + */ +#define XFS_DQ_USER 0x0001 /* a user quota */ +#define XFS_DQ_PROJ 0x0002 /* project quota */ +#define XFS_DQ_GROUP 0x0004 /* a group quota */ +#define XFS_DQ_DIRTY 0x0008 /* dquot is dirty */ +#define XFS_DQ_FREEING 0x0010 /* dquot is beeing torn down */ + +#define XFS_DQ_ALLTYPES (XFS_DQ_USER|XFS_DQ_PROJ|XFS_DQ_GROUP) + +#define XFS_DQ_FLAGS \ + { XFS_DQ_USER, "USER" }, \ + { XFS_DQ_PROJ, "PROJ" }, \ + { XFS_DQ_GROUP, "GROUP" }, \ + { XFS_DQ_DIRTY, "DIRTY" }, \ + { XFS_DQ_FREEING, "FREEING" } + +/* + * We have the possibility of all three quota types being active at once, and + * hence free space modification requires modification of all three current + * dquots in a single transaction. For this case we need to have a reservation + * of at least 3 dquots. + * + * However, a chmod operation can change both UID and GID in a single + * transaction, resulting in requiring {old, new} x {uid, gid} dquots to be + * modified. Hence for this case we need to reserve space for at least 4 dquots. + * + * And in the worst case, there's a rename operation that can be modifying up to + * 4 inodes with dquots attached to them. In reality, the only inodes that can + * have their dquots modified are the source and destination directory inodes + * due to directory name creation and removal. That can require space allocation + * and/or freeing on both directory inodes, and hence all three dquots on each + * inode can be modified. And if the directories are world writeable, all the + * dquots can be unique and so 6 dquots can be modified.... + * + * And, of course, we also need to take into account the dquot log format item + * used to describe each dquot. + */ +#define XFS_DQUOT_LOGRES(mp) \ + ((sizeof(struct xfs_dq_logformat) + sizeof(struct xfs_disk_dquot)) * 6) + +#define XFS_IS_QUOTA_RUNNING(mp) ((mp)->m_qflags & XFS_ALL_QUOTA_ACCT) +#define XFS_IS_UQUOTA_RUNNING(mp) ((mp)->m_qflags & XFS_UQUOTA_ACCT) +#define XFS_IS_PQUOTA_RUNNING(mp) ((mp)->m_qflags & XFS_PQUOTA_ACCT) +#define XFS_IS_GQUOTA_RUNNING(mp) ((mp)->m_qflags & XFS_GQUOTA_ACCT) +#define XFS_IS_UQUOTA_ENFORCED(mp) ((mp)->m_qflags & XFS_UQUOTA_ENFD) +#define XFS_IS_GQUOTA_ENFORCED(mp) ((mp)->m_qflags & XFS_GQUOTA_ENFD) +#define XFS_IS_PQUOTA_ENFORCED(mp) ((mp)->m_qflags & XFS_PQUOTA_ENFD) + +/* + * Incore only flags for quotaoff - these bits get cleared when quota(s) + * are in the process of getting turned off. These flags are in m_qflags but + * never in sb_qflags. + */ +#define XFS_UQUOTA_ACTIVE 0x1000 /* uquotas are being turned off */ +#define XFS_GQUOTA_ACTIVE 0x2000 /* gquotas are being turned off */ +#define XFS_PQUOTA_ACTIVE 0x4000 /* pquotas are being turned off */ +#define XFS_ALL_QUOTA_ACTIVE \ + (XFS_UQUOTA_ACTIVE | XFS_GQUOTA_ACTIVE | XFS_PQUOTA_ACTIVE) + +/* + * Checking XFS_IS_*QUOTA_ON() while holding any inode lock guarantees + * quota will be not be switched off as long as that inode lock is held. + */ +#define XFS_IS_QUOTA_ON(mp) ((mp)->m_qflags & (XFS_UQUOTA_ACTIVE | \ + XFS_GQUOTA_ACTIVE | \ + XFS_PQUOTA_ACTIVE)) +#define XFS_IS_OQUOTA_ON(mp) ((mp)->m_qflags & (XFS_GQUOTA_ACTIVE | \ + XFS_PQUOTA_ACTIVE)) +#define XFS_IS_UQUOTA_ON(mp) ((mp)->m_qflags & XFS_UQUOTA_ACTIVE) +#define XFS_IS_GQUOTA_ON(mp) ((mp)->m_qflags & XFS_GQUOTA_ACTIVE) +#define XFS_IS_PQUOTA_ON(mp) ((mp)->m_qflags & XFS_PQUOTA_ACTIVE) + +/* + * Flags to tell various functions what to do. Not all of these are meaningful + * to a single function. None of these XFS_QMOPT_* flags are meant to have + * persistent values (ie. their values can and will change between versions) + */ +#define XFS_QMOPT_DQALLOC 0x0000002 /* alloc dquot ondisk if needed */ +#define XFS_QMOPT_UQUOTA 0x0000004 /* user dquot requested */ +#define XFS_QMOPT_PQUOTA 0x0000008 /* project dquot requested */ +#define XFS_QMOPT_FORCE_RES 0x0000010 /* ignore quota limits */ +#define XFS_QMOPT_SBVERSION 0x0000040 /* change superblock version num */ +#define XFS_QMOPT_DOWARN 0x0000400 /* increase warning cnt if needed */ +#define XFS_QMOPT_DQREPAIR 0x0001000 /* repair dquot if damaged */ +#define XFS_QMOPT_GQUOTA 0x0002000 /* group dquot requested */ +#define XFS_QMOPT_ENOSPC 0x0004000 /* enospc instead of edquot (prj) */ + +/* + * flags to xfs_trans_mod_dquot to indicate which field needs to be + * modified. + */ +#define XFS_QMOPT_RES_REGBLKS 0x0010000 +#define XFS_QMOPT_RES_RTBLKS 0x0020000 +#define XFS_QMOPT_BCOUNT 0x0040000 +#define XFS_QMOPT_ICOUNT 0x0080000 +#define XFS_QMOPT_RTBCOUNT 0x0100000 +#define XFS_QMOPT_DELBCOUNT 0x0200000 +#define XFS_QMOPT_DELRTBCOUNT 0x0400000 +#define XFS_QMOPT_RES_INOS 0x0800000 + +/* + * flags for dqalloc. + */ +#define XFS_QMOPT_INHERIT 0x1000000 + +/* + * flags to xfs_trans_mod_dquot. + */ +#define XFS_TRANS_DQ_RES_BLKS XFS_QMOPT_RES_REGBLKS +#define XFS_TRANS_DQ_RES_RTBLKS XFS_QMOPT_RES_RTBLKS +#define XFS_TRANS_DQ_RES_INOS XFS_QMOPT_RES_INOS +#define XFS_TRANS_DQ_BCOUNT XFS_QMOPT_BCOUNT +#define XFS_TRANS_DQ_DELBCOUNT XFS_QMOPT_DELBCOUNT +#define XFS_TRANS_DQ_ICOUNT XFS_QMOPT_ICOUNT +#define XFS_TRANS_DQ_RTBCOUNT XFS_QMOPT_RTBCOUNT +#define XFS_TRANS_DQ_DELRTBCOUNT XFS_QMOPT_DELRTBCOUNT + + +#define XFS_QMOPT_QUOTALL \ + (XFS_QMOPT_UQUOTA | XFS_QMOPT_PQUOTA | XFS_QMOPT_GQUOTA) +#define XFS_QMOPT_RESBLK_MASK (XFS_QMOPT_RES_REGBLKS | XFS_QMOPT_RES_RTBLKS) + +#endif /* __XFS_QUOTA_H__ */ diff --git a/fs/xfs/xfs_quotaops.c b/fs/xfs/xfs_quotaops.c index 20e30f9..1326d81 100644 --- a/fs/xfs/xfs_quotaops.c +++ b/fs/xfs/xfs_quotaops.c @@ -16,8 +16,10 @@ * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #include "xfs.h" -#include "xfs_sb.h" +#include "xfs_format.h" +#include "xfs_trans_resv.h" #include "xfs_log.h" +#include "xfs_sb.h" #include "xfs_ag.h" #include "xfs_mount.h" #include "xfs_quota.h" @@ -54,6 +56,18 @@ xfs_fs_get_xstate( } STATIC int +xfs_fs_get_xstatev( + struct super_block *sb, + struct fs_quota_statv *fqs) +{ + struct xfs_mount *mp = XFS_M(sb); + + if (!XFS_IS_QUOTA_RUNNING(mp)) + return -ENOSYS; + return -xfs_qm_scall_getqstatv(mp, fqs); +} + +STATIC int xfs_fs_set_xstate( struct super_block *sb, unsigned int uflags, @@ -133,6 +147,7 @@ xfs_fs_set_dqblk( } const struct quotactl_ops xfs_quotactl_operations = { + .get_xstatev = xfs_fs_get_xstatev, .get_xstate = xfs_fs_get_xstate, .set_xstate = xfs_fs_set_xstate, .get_dqblk = xfs_fs_get_dqblk, diff --git a/fs/xfs/xfs_rename.c b/fs/xfs/xfs_rename.c deleted file mode 100644 index 30ff5f4..0000000 --- a/fs/xfs/xfs_rename.c +++ /dev/null @@ -1,346 +0,0 @@ -/* - * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#include "xfs.h" -#include "xfs_fs.h" -#include "xfs_types.h" -#include "xfs_log.h" -#include "xfs_trans.h" -#include "xfs_sb.h" -#include "xfs_ag.h" -#include "xfs_dir2.h" -#include "xfs_mount.h" -#include "xfs_da_btree.h" -#include "xfs_bmap_btree.h" -#include "xfs_dinode.h" -#include "xfs_inode.h" -#include "xfs_inode_item.h" -#include "xfs_bmap.h" -#include "xfs_error.h" -#include "xfs_quota.h" -#include "xfs_utils.h" -#include "xfs_trans_space.h" -#include "xfs_vnodeops.h" -#include "xfs_trace.h" - - -/* - * Enter all inodes for a rename transaction into a sorted array. - */ -STATIC void -xfs_sort_for_rename( - xfs_inode_t *dp1, /* in: old (source) directory inode */ - xfs_inode_t *dp2, /* in: new (target) directory inode */ - xfs_inode_t *ip1, /* in: inode of old entry */ - xfs_inode_t *ip2, /* in: inode of new entry, if it - already exists, NULL otherwise. */ - xfs_inode_t **i_tab,/* out: array of inode returned, sorted */ - int *num_inodes) /* out: number of inodes in array */ -{ - xfs_inode_t *temp; - int i, j; - - /* - * i_tab contains a list of pointers to inodes. We initialize - * the table here & we'll sort it. We will then use it to - * order the acquisition of the inode locks. - * - * Note that the table may contain duplicates. e.g., dp1 == dp2. - */ - i_tab[0] = dp1; - i_tab[1] = dp2; - i_tab[2] = ip1; - if (ip2) { - *num_inodes = 4; - i_tab[3] = ip2; - } else { - *num_inodes = 3; - i_tab[3] = NULL; - } - - /* - * Sort the elements via bubble sort. (Remember, there are at - * most 4 elements to sort, so this is adequate.) - */ - for (i = 0; i < *num_inodes; i++) { - for (j = 1; j < *num_inodes; j++) { - if (i_tab[j]->i_ino < i_tab[j-1]->i_ino) { - temp = i_tab[j]; - i_tab[j] = i_tab[j-1]; - i_tab[j-1] = temp; - } - } - } -} - -/* - * xfs_rename - */ -int -xfs_rename( - xfs_inode_t *src_dp, - struct xfs_name *src_name, - xfs_inode_t *src_ip, - xfs_inode_t *target_dp, - struct xfs_name *target_name, - xfs_inode_t *target_ip) -{ - xfs_trans_t *tp = NULL; - xfs_mount_t *mp = src_dp->i_mount; - int new_parent; /* moving to a new dir */ - int src_is_directory; /* src_name is a directory */ - int error; - xfs_bmap_free_t free_list; - xfs_fsblock_t first_block; - int cancel_flags; - int committed; - xfs_inode_t *inodes[4]; - int spaceres; - int num_inodes; - - trace_xfs_rename(src_dp, target_dp, src_name, target_name); - - new_parent = (src_dp != target_dp); - src_is_directory = S_ISDIR(src_ip->i_d.di_mode); - - xfs_sort_for_rename(src_dp, target_dp, src_ip, target_ip, - inodes, &num_inodes); - - xfs_bmap_init(&free_list, &first_block); - tp = xfs_trans_alloc(mp, XFS_TRANS_RENAME); - cancel_flags = XFS_TRANS_RELEASE_LOG_RES; - spaceres = XFS_RENAME_SPACE_RES(mp, target_name->len); - error = xfs_trans_reserve(tp, spaceres, XFS_RENAME_LOG_RES(mp), 0, - XFS_TRANS_PERM_LOG_RES, XFS_RENAME_LOG_COUNT); - if (error == ENOSPC) { - spaceres = 0; - error = xfs_trans_reserve(tp, 0, XFS_RENAME_LOG_RES(mp), 0, - XFS_TRANS_PERM_LOG_RES, XFS_RENAME_LOG_COUNT); - } - if (error) { - xfs_trans_cancel(tp, 0); - goto std_return; - } - - /* - * Attach the dquots to the inodes - */ - error = xfs_qm_vop_rename_dqattach(inodes); - if (error) { - xfs_trans_cancel(tp, cancel_flags); - goto std_return; - } - - /* - * Lock all the participating inodes. Depending upon whether - * the target_name exists in the target directory, and - * whether the target directory is the same as the source - * directory, we can lock from 2 to 4 inodes. - */ - xfs_lock_inodes(inodes, num_inodes, XFS_ILOCK_EXCL); - - /* - * Join all the inodes to the transaction. From this point on, - * we can rely on either trans_commit or trans_cancel to unlock - * them. - */ - xfs_trans_ijoin(tp, src_dp, XFS_ILOCK_EXCL); - if (new_parent) - xfs_trans_ijoin(tp, target_dp, XFS_ILOCK_EXCL); - xfs_trans_ijoin(tp, src_ip, XFS_ILOCK_EXCL); - if (target_ip) - xfs_trans_ijoin(tp, target_ip, XFS_ILOCK_EXCL); - - /* - * If we are using project inheritance, we only allow renames - * into our tree when the project IDs are the same; else the - * tree quota mechanism would be circumvented. - */ - if (unlikely((target_dp->i_d.di_flags & XFS_DIFLAG_PROJINHERIT) && - (xfs_get_projid(target_dp) != xfs_get_projid(src_ip)))) { - error = XFS_ERROR(EXDEV); - goto error_return; - } - - /* - * Set up the target. - */ - if (target_ip == NULL) { - /* - * If there's no space reservation, check the entry will - * fit before actually inserting it. - */ - error = xfs_dir_canenter(tp, target_dp, target_name, spaceres); - if (error) - goto error_return; - /* - * If target does not exist and the rename crosses - * directories, adjust the target directory link count - * to account for the ".." reference from the new entry. - */ - error = xfs_dir_createname(tp, target_dp, target_name, - src_ip->i_ino, &first_block, - &free_list, spaceres); - if (error == ENOSPC) - goto error_return; - if (error) - goto abort_return; - - xfs_trans_ichgtime(tp, target_dp, - XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG); - - if (new_parent && src_is_directory) { - error = xfs_bumplink(tp, target_dp); - if (error) - goto abort_return; - } - } else { /* target_ip != NULL */ - /* - * If target exists and it's a directory, check that both - * target and source are directories and that target can be - * destroyed, or that neither is a directory. - */ - if (S_ISDIR(target_ip->i_d.di_mode)) { - /* - * Make sure target dir is empty. - */ - if (!(xfs_dir_isempty(target_ip)) || - (target_ip->i_d.di_nlink > 2)) { - error = XFS_ERROR(EEXIST); - goto error_return; - } - } - - /* - * Link the source inode under the target name. - * If the source inode is a directory and we are moving - * it across directories, its ".." entry will be - * inconsistent until we replace that down below. - * - * In case there is already an entry with the same - * name at the destination directory, remove it first. - */ - error = xfs_dir_replace(tp, target_dp, target_name, - src_ip->i_ino, - &first_block, &free_list, spaceres); - if (error) - goto abort_return; - - xfs_trans_ichgtime(tp, target_dp, - XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG); - - /* - * Decrement the link count on the target since the target - * dir no longer points to it. - */ - error = xfs_droplink(tp, target_ip); - if (error) - goto abort_return; - - if (src_is_directory) { - /* - * Drop the link from the old "." entry. - */ - error = xfs_droplink(tp, target_ip); - if (error) - goto abort_return; - } - } /* target_ip != NULL */ - - /* - * Remove the source. - */ - if (new_parent && src_is_directory) { - /* - * Rewrite the ".." entry to point to the new - * directory. - */ - error = xfs_dir_replace(tp, src_ip, &xfs_name_dotdot, - target_dp->i_ino, - &first_block, &free_list, spaceres); - ASSERT(error != EEXIST); - if (error) - goto abort_return; - } - - /* - * We always want to hit the ctime on the source inode. - * - * This isn't strictly required by the standards since the source - * inode isn't really being changed, but old unix file systems did - * it and some incremental backup programs won't work without it. - */ - xfs_trans_ichgtime(tp, src_ip, XFS_ICHGTIME_CHG); - xfs_trans_log_inode(tp, src_ip, XFS_ILOG_CORE); - - /* - * Adjust the link count on src_dp. This is necessary when - * renaming a directory, either within one parent when - * the target existed, or across two parent directories. - */ - if (src_is_directory && (new_parent || target_ip != NULL)) { - - /* - * Decrement link count on src_directory since the - * entry that's moved no longer points to it. - */ - error = xfs_droplink(tp, src_dp); - if (error) - goto abort_return; - } - - error = xfs_dir_removename(tp, src_dp, src_name, src_ip->i_ino, - &first_block, &free_list, spaceres); - if (error) - goto abort_return; - - xfs_trans_ichgtime(tp, src_dp, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG); - xfs_trans_log_inode(tp, src_dp, XFS_ILOG_CORE); - if (new_parent) - xfs_trans_log_inode(tp, target_dp, XFS_ILOG_CORE); - - /* - * If this is a synchronous mount, make sure that the - * rename transaction goes to disk before returning to - * the user. - */ - if (mp->m_flags & (XFS_MOUNT_WSYNC|XFS_MOUNT_DIRSYNC)) { - xfs_trans_set_sync(tp); - } - - error = xfs_bmap_finish(&tp, &free_list, &committed); - if (error) { - xfs_bmap_cancel(&free_list); - xfs_trans_cancel(tp, (XFS_TRANS_RELEASE_LOG_RES | - XFS_TRANS_ABORT)); - goto std_return; - } - - /* - * trans_commit will unlock src_ip, target_ip & decrement - * the vnode references. - */ - return xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES); - - abort_return: - cancel_flags |= XFS_TRANS_ABORT; - error_return: - xfs_bmap_cancel(&free_list); - xfs_trans_cancel(tp, cancel_flags); - std_return: - return error; -} diff --git a/fs/xfs/xfs_rtalloc.c b/fs/xfs/xfs_rtalloc.c index 98dc670..6f9e63c 100644 --- a/fs/xfs/xfs_rtalloc.c +++ b/fs/xfs/xfs_rtalloc.c @@ -17,25 +17,24 @@ */ #include "xfs.h" #include "xfs_fs.h" -#include "xfs_types.h" +#include "xfs_format.h" #include "xfs_bit.h" #include "xfs_log.h" #include "xfs_trans.h" #include "xfs_sb.h" #include "xfs_ag.h" -#include "xfs_dir2.h" #include "xfs_mount.h" #include "xfs_bmap_btree.h" #include "xfs_dinode.h" #include "xfs_inode.h" #include "xfs_alloc.h" #include "xfs_bmap.h" +#include "xfs_bmap_util.h" #include "xfs_rtalloc.h" #include "xfs_fsops.h" #include "xfs_error.h" #include "xfs_inode_item.h" #include "xfs_trans_space.h" -#include "xfs_utils.h" #include "xfs_trace.h" #include "xfs_buf.h" #include "xfs_icache.h" @@ -101,10 +100,9 @@ xfs_growfs_rt_alloc( /* * Reserve space & log for one extent added to the file. */ - if ((error = xfs_trans_reserve(tp, resblks, - XFS_GROWRTALLOC_LOG_RES(mp), 0, - XFS_TRANS_PERM_LOG_RES, - XFS_DEFAULT_PERM_LOG_COUNT))) + error = xfs_trans_reserve(tp, &M_RES(mp)->tr_growdata, + resblks, 0); + if (error) goto error_cancel; cancelflags = XFS_TRANS_RELEASE_LOG_RES; /* @@ -147,8 +145,9 @@ xfs_growfs_rt_alloc( /* * Reserve log for one block zeroing. */ - if ((error = xfs_trans_reserve(tp, 0, - XFS_GROWRTZERO_LOG_RES(mp), 0, 0, 0))) + error = xfs_trans_reserve(tp, &M_RES(mp)->tr_growrtzero, + 0, 0); + if (error) goto error_cancel; /* * Lock the bitmap inode. @@ -736,8 +735,8 @@ xfs_rtallocate_range( { xfs_rtblock_t end; /* end of the allocated extent */ int error; /* error value */ - xfs_rtblock_t postblock; /* first block allocated > end */ - xfs_rtblock_t preblock; /* first block allocated < start */ + xfs_rtblock_t postblock = 0; /* first block allocated > end */ + xfs_rtblock_t preblock = 0; /* first block allocated < start */ end = start + len - 1; /* @@ -1958,8 +1957,9 @@ xfs_growfs_rt( * Start a transaction, get the log reservation. */ tp = xfs_trans_alloc(mp, XFS_TRANS_GROWFSRT_FREE); - if ((error = xfs_trans_reserve(tp, 0, - XFS_GROWRTFREE_LOG_RES(nmp), 0, 0, 0))) + error = xfs_trans_reserve(tp, &M_RES(mp)->tr_growrtfree, + 0, 0); + if (error) goto error_cancel; /* * Lock out other callers by grabbing the bitmap inode lock. @@ -2148,7 +2148,7 @@ xfs_rtfree_extent( ASSERT(mp->m_rbmip->i_itemp != NULL); ASSERT(xfs_isilocked(mp->m_rbmip, XFS_ILOCK_EXCL)); -#if defined(__KERNEL__) && defined(DEBUG) +#ifdef DEBUG /* * Check to see that this whole range is currently allocated. */ diff --git a/fs/xfs/xfs_rtalloc.h b/fs/xfs/xfs_rtalloc.h index f7f3a35..b2a1a24 100644 --- a/fs/xfs/xfs_rtalloc.h +++ b/fs/xfs/xfs_rtalloc.h @@ -18,58 +18,11 @@ #ifndef __XFS_RTALLOC_H__ #define __XFS_RTALLOC_H__ +/* kernel only definitions and functions */ + struct xfs_mount; struct xfs_trans; -/* Min and max rt extent sizes, specified in bytes */ -#define XFS_MAX_RTEXTSIZE (1024 * 1024 * 1024) /* 1GB */ -#define XFS_DFL_RTEXTSIZE (64 * 1024) /* 64kB */ -#define XFS_MIN_RTEXTSIZE (4 * 1024) /* 4kB */ - -/* - * Constants for bit manipulations. - */ -#define XFS_NBBYLOG 3 /* log2(NBBY) */ -#define XFS_WORDLOG 2 /* log2(sizeof(xfs_rtword_t)) */ -#define XFS_NBWORDLOG (XFS_NBBYLOG + XFS_WORDLOG) -#define XFS_NBWORD (1 << XFS_NBWORDLOG) -#define XFS_WORDMASK ((1 << XFS_WORDLOG) - 1) - -#define XFS_BLOCKSIZE(mp) ((mp)->m_sb.sb_blocksize) -#define XFS_BLOCKMASK(mp) ((mp)->m_blockmask) -#define XFS_BLOCKWSIZE(mp) ((mp)->m_blockwsize) -#define XFS_BLOCKWMASK(mp) ((mp)->m_blockwmask) - -/* - * Summary and bit manipulation macros. - */ -#define XFS_SUMOFFS(mp,ls,bb) ((int)((ls) * (mp)->m_sb.sb_rbmblocks + (bb))) -#define XFS_SUMOFFSTOBLOCK(mp,s) \ - (((s) * (uint)sizeof(xfs_suminfo_t)) >> (mp)->m_sb.sb_blocklog) -#define XFS_SUMPTR(mp,bp,so) \ - ((xfs_suminfo_t *)((bp)->b_addr + \ - (((so) * (uint)sizeof(xfs_suminfo_t)) & XFS_BLOCKMASK(mp)))) - -#define XFS_BITTOBLOCK(mp,bi) ((bi) >> (mp)->m_blkbit_log) -#define XFS_BLOCKTOBIT(mp,bb) ((bb) << (mp)->m_blkbit_log) -#define XFS_BITTOWORD(mp,bi) \ - ((int)(((bi) >> XFS_NBWORDLOG) & XFS_BLOCKWMASK(mp))) - -#define XFS_RTMIN(a,b) ((a) < (b) ? (a) : (b)) -#define XFS_RTMAX(a,b) ((a) > (b) ? (a) : (b)) - -#define XFS_RTLOBIT(w) xfs_lowbit32(w) -#define XFS_RTHIBIT(w) xfs_highbit32(w) - -#if XFS_BIG_BLKNOS -#define XFS_RTBLOCKLOG(b) xfs_highbit64(b) -#else -#define XFS_RTBLOCKLOG(b) xfs_highbit32(b) -#endif - - -#ifdef __KERNEL__ - #ifdef CONFIG_XFS_RT /* * Function prototypes for exported functions. @@ -161,6 +114,4 @@ xfs_rtmount_init( # define xfs_rtunmount_inodes(m) #endif /* CONFIG_XFS_RT */ -#endif /* __KERNEL__ */ - #endif /* __XFS_RTALLOC_H__ */ diff --git a/fs/xfs/xfs_sb.c b/fs/xfs/xfs_sb.c new file mode 100644 index 0000000..a5b59d9 --- /dev/null +++ b/fs/xfs/xfs_sb.c @@ -0,0 +1,834 @@ +/* + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#include "xfs.h" +#include "xfs_fs.h" +#include "xfs_format.h" +#include "xfs_bit.h" +#include "xfs_log.h" +#include "xfs_inum.h" +#include "xfs_trans.h" +#include "xfs_trans_priv.h" +#include "xfs_sb.h" +#include "xfs_ag.h" +#include "xfs_mount.h" +#include "xfs_da_btree.h" +#include "xfs_dir2_format.h" +#include "xfs_dir2.h" +#include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" +#include "xfs_ialloc_btree.h" +#include "xfs_dinode.h" +#include "xfs_inode.h" +#include "xfs_btree.h" +#include "xfs_ialloc.h" +#include "xfs_alloc.h" +#include "xfs_rtalloc.h" +#include "xfs_bmap.h" +#include "xfs_error.h" +#include "xfs_quota.h" +#include "xfs_fsops.h" +#include "xfs_trace.h" +#include "xfs_cksum.h" +#include "xfs_buf_item.h" + +/* + * Physical superblock buffer manipulations. Shared with libxfs in userspace. + */ + +static const struct { + short offset; + short type; /* 0 = integer + * 1 = binary / string (no translation) + */ +} xfs_sb_info[] = { + { offsetof(xfs_sb_t, sb_magicnum), 0 }, + { offsetof(xfs_sb_t, sb_blocksize), 0 }, + { offsetof(xfs_sb_t, sb_dblocks), 0 }, + { offsetof(xfs_sb_t, sb_rblocks), 0 }, + { offsetof(xfs_sb_t, sb_rextents), 0 }, + { offsetof(xfs_sb_t, sb_uuid), 1 }, + { offsetof(xfs_sb_t, sb_logstart), 0 }, + { offsetof(xfs_sb_t, sb_rootino), 0 }, + { offsetof(xfs_sb_t, sb_rbmino), 0 }, + { offsetof(xfs_sb_t, sb_rsumino), 0 }, + { offsetof(xfs_sb_t, sb_rextsize), 0 }, + { offsetof(xfs_sb_t, sb_agblocks), 0 }, + { offsetof(xfs_sb_t, sb_agcount), 0 }, + { offsetof(xfs_sb_t, sb_rbmblocks), 0 }, + { offsetof(xfs_sb_t, sb_logblocks), 0 }, + { offsetof(xfs_sb_t, sb_versionnum), 0 }, + { offsetof(xfs_sb_t, sb_sectsize), 0 }, + { offsetof(xfs_sb_t, sb_inodesize), 0 }, + { offsetof(xfs_sb_t, sb_inopblock), 0 }, + { offsetof(xfs_sb_t, sb_fname[0]), 1 }, + { offsetof(xfs_sb_t, sb_blocklog), 0 }, + { offsetof(xfs_sb_t, sb_sectlog), 0 }, + { offsetof(xfs_sb_t, sb_inodelog), 0 }, + { offsetof(xfs_sb_t, sb_inopblog), 0 }, + { offsetof(xfs_sb_t, sb_agblklog), 0 }, + { offsetof(xfs_sb_t, sb_rextslog), 0 }, + { offsetof(xfs_sb_t, sb_inprogress), 0 }, + { offsetof(xfs_sb_t, sb_imax_pct), 0 }, + { offsetof(xfs_sb_t, sb_icount), 0 }, + { offsetof(xfs_sb_t, sb_ifree), 0 }, + { offsetof(xfs_sb_t, sb_fdblocks), 0 }, + { offsetof(xfs_sb_t, sb_frextents), 0 }, + { offsetof(xfs_sb_t, sb_uquotino), 0 }, + { offsetof(xfs_sb_t, sb_gquotino), 0 }, + { offsetof(xfs_sb_t, sb_qflags), 0 }, + { offsetof(xfs_sb_t, sb_flags), 0 }, + { offsetof(xfs_sb_t, sb_shared_vn), 0 }, + { offsetof(xfs_sb_t, sb_inoalignmt), 0 }, + { offsetof(xfs_sb_t, sb_unit), 0 }, + { offsetof(xfs_sb_t, sb_width), 0 }, + { offsetof(xfs_sb_t, sb_dirblklog), 0 }, + { offsetof(xfs_sb_t, sb_logsectlog), 0 }, + { offsetof(xfs_sb_t, sb_logsectsize), 0 }, + { offsetof(xfs_sb_t, sb_logsunit), 0 }, + { offsetof(xfs_sb_t, sb_features2), 0 }, + { offsetof(xfs_sb_t, sb_bad_features2), 0 }, + { offsetof(xfs_sb_t, sb_features_compat), 0 }, + { offsetof(xfs_sb_t, sb_features_ro_compat), 0 }, + { offsetof(xfs_sb_t, sb_features_incompat), 0 }, + { offsetof(xfs_sb_t, sb_features_log_incompat), 0 }, + { offsetof(xfs_sb_t, sb_crc), 0 }, + { offsetof(xfs_sb_t, sb_pad), 0 }, + { offsetof(xfs_sb_t, sb_pquotino), 0 }, + { offsetof(xfs_sb_t, sb_lsn), 0 }, + { sizeof(xfs_sb_t), 0 } +}; + +/* + * Reference counting access wrappers to the perag structures. + * Because we never free per-ag structures, the only thing we + * have to protect against changes is the tree structure itself. + */ +struct xfs_perag * +xfs_perag_get( + struct xfs_mount *mp, + xfs_agnumber_t agno) +{ + struct xfs_perag *pag; + int ref = 0; + + rcu_read_lock(); + pag = radix_tree_lookup(&mp->m_perag_tree, agno); + if (pag) { + ASSERT(atomic_read(&pag->pag_ref) >= 0); + ref = atomic_inc_return(&pag->pag_ref); + } + rcu_read_unlock(); + trace_xfs_perag_get(mp, agno, ref, _RET_IP_); + return pag; +} + +/* + * search from @first to find the next perag with the given tag set. + */ +struct xfs_perag * +xfs_perag_get_tag( + struct xfs_mount *mp, + xfs_agnumber_t first, + int tag) +{ + struct xfs_perag *pag; + int found; + int ref; + + rcu_read_lock(); + found = radix_tree_gang_lookup_tag(&mp->m_perag_tree, + (void **)&pag, first, 1, tag); + if (found <= 0) { + rcu_read_unlock(); + return NULL; + } + ref = atomic_inc_return(&pag->pag_ref); + rcu_read_unlock(); + trace_xfs_perag_get_tag(mp, pag->pag_agno, ref, _RET_IP_); + return pag; +} + +void +xfs_perag_put( + struct xfs_perag *pag) +{ + int ref; + + ASSERT(atomic_read(&pag->pag_ref) > 0); + ref = atomic_dec_return(&pag->pag_ref); + trace_xfs_perag_put(pag->pag_mount, pag->pag_agno, ref, _RET_IP_); +} + +/* + * Check the validity of the SB found. + */ +STATIC int +xfs_mount_validate_sb( + xfs_mount_t *mp, + xfs_sb_t *sbp, + bool check_inprogress, + bool check_version) +{ + + /* + * If the log device and data device have the + * same device number, the log is internal. + * Consequently, the sb_logstart should be non-zero. If + * we have a zero sb_logstart in this case, we may be trying to mount + * a volume filesystem in a non-volume manner. + */ + if (sbp->sb_magicnum != XFS_SB_MAGIC) { + xfs_warn(mp, "bad magic number"); + return XFS_ERROR(EWRONGFS); + } + + + if (!xfs_sb_good_version(sbp)) { + xfs_warn(mp, "bad version"); + return XFS_ERROR(EWRONGFS); + } + + /* + * Version 5 superblock feature mask validation. Reject combinations the + * kernel cannot support up front before checking anything else. For + * write validation, we don't need to check feature masks. + */ + if (check_version && XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5) { + xfs_alert(mp, +"Version 5 superblock detected. This kernel has EXPERIMENTAL support enabled!\n" +"Use of these features in this kernel is at your own risk!"); + + if (xfs_sb_has_compat_feature(sbp, + XFS_SB_FEAT_COMPAT_UNKNOWN)) { + xfs_warn(mp, +"Superblock has unknown compatible features (0x%x) enabled.\n" +"Using a more recent kernel is recommended.", + (sbp->sb_features_compat & + XFS_SB_FEAT_COMPAT_UNKNOWN)); + } + + if (xfs_sb_has_ro_compat_feature(sbp, + XFS_SB_FEAT_RO_COMPAT_UNKNOWN)) { + xfs_alert(mp, +"Superblock has unknown read-only compatible features (0x%x) enabled.", + (sbp->sb_features_ro_compat & + XFS_SB_FEAT_RO_COMPAT_UNKNOWN)); + if (!(mp->m_flags & XFS_MOUNT_RDONLY)) { + xfs_warn(mp, +"Attempted to mount read-only compatible filesystem read-write.\n" +"Filesystem can only be safely mounted read only."); + return XFS_ERROR(EINVAL); + } + } + if (xfs_sb_has_incompat_feature(sbp, + XFS_SB_FEAT_INCOMPAT_UNKNOWN)) { + xfs_warn(mp, +"Superblock has unknown incompatible features (0x%x) enabled.\n" +"Filesystem can not be safely mounted by this kernel.", + (sbp->sb_features_incompat & + XFS_SB_FEAT_INCOMPAT_UNKNOWN)); + return XFS_ERROR(EINVAL); + } + } + + if (xfs_sb_version_has_pquotino(sbp)) { + if (sbp->sb_qflags & (XFS_OQUOTA_ENFD | XFS_OQUOTA_CHKD)) { + xfs_notice(mp, + "Version 5 of Super block has XFS_OQUOTA bits.\n"); + return XFS_ERROR(EFSCORRUPTED); + } + } else if (sbp->sb_qflags & (XFS_PQUOTA_ENFD | XFS_GQUOTA_ENFD | + XFS_PQUOTA_CHKD | XFS_GQUOTA_CHKD)) { + xfs_notice(mp, +"Superblock earlier than Version 5 has XFS_[PQ]UOTA_{ENFD|CHKD} bits.\n"); + return XFS_ERROR(EFSCORRUPTED); + } + + if (unlikely( + sbp->sb_logstart == 0 && mp->m_logdev_targp == mp->m_ddev_targp)) { + xfs_warn(mp, + "filesystem is marked as having an external log; " + "specify logdev on the mount command line."); + return XFS_ERROR(EINVAL); + } + + if (unlikely( + sbp->sb_logstart != 0 && mp->m_logdev_targp != mp->m_ddev_targp)) { + xfs_warn(mp, + "filesystem is marked as having an internal log; " + "do not specify logdev on the mount command line."); + return XFS_ERROR(EINVAL); + } + + /* + * More sanity checking. Most of these were stolen directly from + * xfs_repair. + */ + if (unlikely( + sbp->sb_agcount <= 0 || + sbp->sb_sectsize < XFS_MIN_SECTORSIZE || + sbp->sb_sectsize > XFS_MAX_SECTORSIZE || + sbp->sb_sectlog < XFS_MIN_SECTORSIZE_LOG || + sbp->sb_sectlog > XFS_MAX_SECTORSIZE_LOG || + sbp->sb_sectsize != (1 << sbp->sb_sectlog) || + sbp->sb_blocksize < XFS_MIN_BLOCKSIZE || + sbp->sb_blocksize > XFS_MAX_BLOCKSIZE || + sbp->sb_blocklog < XFS_MIN_BLOCKSIZE_LOG || + sbp->sb_blocklog > XFS_MAX_BLOCKSIZE_LOG || + sbp->sb_blocksize != (1 << sbp->sb_blocklog) || + sbp->sb_inodesize < XFS_DINODE_MIN_SIZE || + sbp->sb_inodesize > XFS_DINODE_MAX_SIZE || + sbp->sb_inodelog < XFS_DINODE_MIN_LOG || + sbp->sb_inodelog > XFS_DINODE_MAX_LOG || + sbp->sb_inodesize != (1 << sbp->sb_inodelog) || + (sbp->sb_blocklog - sbp->sb_inodelog != sbp->sb_inopblog) || + (sbp->sb_rextsize * sbp->sb_blocksize > XFS_MAX_RTEXTSIZE) || + (sbp->sb_rextsize * sbp->sb_blocksize < XFS_MIN_RTEXTSIZE) || + (sbp->sb_imax_pct > 100 /* zero sb_imax_pct is valid */) || + sbp->sb_dblocks == 0 || + sbp->sb_dblocks > XFS_MAX_DBLOCKS(sbp) || + sbp->sb_dblocks < XFS_MIN_DBLOCKS(sbp))) { + XFS_CORRUPTION_ERROR("SB sanity check failed", + XFS_ERRLEVEL_LOW, mp, sbp); + return XFS_ERROR(EFSCORRUPTED); + } + + /* + * Until this is fixed only page-sized or smaller data blocks work. + */ + if (unlikely(sbp->sb_blocksize > PAGE_SIZE)) { + xfs_warn(mp, + "File system with blocksize %d bytes. " + "Only pagesize (%ld) or less will currently work.", + sbp->sb_blocksize, PAGE_SIZE); + return XFS_ERROR(ENOSYS); + } + + /* + * Currently only very few inode sizes are supported. + */ + switch (sbp->sb_inodesize) { + case 256: + case 512: + case 1024: + case 2048: + break; + default: + xfs_warn(mp, "inode size of %d bytes not supported", + sbp->sb_inodesize); + return XFS_ERROR(ENOSYS); + } + + if (xfs_sb_validate_fsb_count(sbp, sbp->sb_dblocks) || + xfs_sb_validate_fsb_count(sbp, sbp->sb_rblocks)) { + xfs_warn(mp, + "file system too large to be mounted on this system."); + return XFS_ERROR(EFBIG); + } + + if (check_inprogress && sbp->sb_inprogress) { + xfs_warn(mp, "Offline file system operation in progress!"); + return XFS_ERROR(EFSCORRUPTED); + } + + /* + * Version 1 directory format has never worked on Linux. + */ + if (unlikely(!xfs_sb_version_hasdirv2(sbp))) { + xfs_warn(mp, "file system using version 1 directory format"); + return XFS_ERROR(ENOSYS); + } + + return 0; +} + +void +xfs_sb_quota_from_disk(struct xfs_sb *sbp) +{ + /* + * older mkfs doesn't initialize quota inodes to NULLFSINO. This + * leads to in-core values having two different values for a quota + * inode to be invalid: 0 and NULLFSINO. Change it to a single value + * NULLFSINO. + * + * Note that this change affect only the in-core values. These + * values are not written back to disk unless any quota information + * is written to the disk. Even in that case, sb_pquotino field is + * not written to disk unless the superblock supports pquotino. + */ + if (sbp->sb_uquotino == 0) + sbp->sb_uquotino = NULLFSINO; + if (sbp->sb_gquotino == 0) + sbp->sb_gquotino = NULLFSINO; + if (sbp->sb_pquotino == 0) + sbp->sb_pquotino = NULLFSINO; + + /* + * We need to do these manipilations only if we are working + * with an older version of on-disk superblock. + */ + if (xfs_sb_version_has_pquotino(sbp)) + return; + + if (sbp->sb_qflags & XFS_OQUOTA_ENFD) + sbp->sb_qflags |= (sbp->sb_qflags & XFS_PQUOTA_ACCT) ? + XFS_PQUOTA_ENFD : XFS_GQUOTA_ENFD; + if (sbp->sb_qflags & XFS_OQUOTA_CHKD) + sbp->sb_qflags |= (sbp->sb_qflags & XFS_PQUOTA_ACCT) ? + XFS_PQUOTA_CHKD : XFS_GQUOTA_CHKD; + sbp->sb_qflags &= ~(XFS_OQUOTA_ENFD | XFS_OQUOTA_CHKD); + + if (sbp->sb_qflags & XFS_PQUOTA_ACCT) { + /* + * In older version of superblock, on-disk superblock only + * has sb_gquotino, and in-core superblock has both sb_gquotino + * and sb_pquotino. But, only one of them is supported at any + * point of time. So, if PQUOTA is set in disk superblock, + * copy over sb_gquotino to sb_pquotino. + */ + sbp->sb_pquotino = sbp->sb_gquotino; + sbp->sb_gquotino = NULLFSINO; + } +} + +void +xfs_sb_from_disk( + struct xfs_sb *to, + xfs_dsb_t *from) +{ + to->sb_magicnum = be32_to_cpu(from->sb_magicnum); + to->sb_blocksize = be32_to_cpu(from->sb_blocksize); + to->sb_dblocks = be64_to_cpu(from->sb_dblocks); + to->sb_rblocks = be64_to_cpu(from->sb_rblocks); + to->sb_rextents = be64_to_cpu(from->sb_rextents); + memcpy(&to->sb_uuid, &from->sb_uuid, sizeof(to->sb_uuid)); + to->sb_logstart = be64_to_cpu(from->sb_logstart); + to->sb_rootino = be64_to_cpu(from->sb_rootino); + to->sb_rbmino = be64_to_cpu(from->sb_rbmino); + to->sb_rsumino = be64_to_cpu(from->sb_rsumino); + to->sb_rextsize = be32_to_cpu(from->sb_rextsize); + to->sb_agblocks = be32_to_cpu(from->sb_agblocks); + to->sb_agcount = be32_to_cpu(from->sb_agcount); + to->sb_rbmblocks = be32_to_cpu(from->sb_rbmblocks); + to->sb_logblocks = be32_to_cpu(from->sb_logblocks); + to->sb_versionnum = be16_to_cpu(from->sb_versionnum); + to->sb_sectsize = be16_to_cpu(from->sb_sectsize); + to->sb_inodesize = be16_to_cpu(from->sb_inodesize); + to->sb_inopblock = be16_to_cpu(from->sb_inopblock); + memcpy(&to->sb_fname, &from->sb_fname, sizeof(to->sb_fname)); + to->sb_blocklog = from->sb_blocklog; + to->sb_sectlog = from->sb_sectlog; + to->sb_inodelog = from->sb_inodelog; + to->sb_inopblog = from->sb_inopblog; + to->sb_agblklog = from->sb_agblklog; + to->sb_rextslog = from->sb_rextslog; + to->sb_inprogress = from->sb_inprogress; + to->sb_imax_pct = from->sb_imax_pct; + to->sb_icount = be64_to_cpu(from->sb_icount); + to->sb_ifree = be64_to_cpu(from->sb_ifree); + to->sb_fdblocks = be64_to_cpu(from->sb_fdblocks); + to->sb_frextents = be64_to_cpu(from->sb_frextents); + to->sb_uquotino = be64_to_cpu(from->sb_uquotino); + to->sb_gquotino = be64_to_cpu(from->sb_gquotino); + to->sb_qflags = be16_to_cpu(from->sb_qflags); + to->sb_flags = from->sb_flags; + to->sb_shared_vn = from->sb_shared_vn; + to->sb_inoalignmt = be32_to_cpu(from->sb_inoalignmt); + to->sb_unit = be32_to_cpu(from->sb_unit); + to->sb_width = be32_to_cpu(from->sb_width); + to->sb_dirblklog = from->sb_dirblklog; + to->sb_logsectlog = from->sb_logsectlog; + to->sb_logsectsize = be16_to_cpu(from->sb_logsectsize); + to->sb_logsunit = be32_to_cpu(from->sb_logsunit); + to->sb_features2 = be32_to_cpu(from->sb_features2); + to->sb_bad_features2 = be32_to_cpu(from->sb_bad_features2); + to->sb_features_compat = be32_to_cpu(from->sb_features_compat); + to->sb_features_ro_compat = be32_to_cpu(from->sb_features_ro_compat); + to->sb_features_incompat = be32_to_cpu(from->sb_features_incompat); + to->sb_features_log_incompat = + be32_to_cpu(from->sb_features_log_incompat); + to->sb_pad = 0; + to->sb_pquotino = be64_to_cpu(from->sb_pquotino); + to->sb_lsn = be64_to_cpu(from->sb_lsn); +} + +static inline void +xfs_sb_quota_to_disk( + xfs_dsb_t *to, + xfs_sb_t *from, + __int64_t *fields) +{ + __uint16_t qflags = from->sb_qflags; + + /* + * We need to do these manipilations only if we are working + * with an older version of on-disk superblock. + */ + if (xfs_sb_version_has_pquotino(from)) + return; + + if (*fields & XFS_SB_QFLAGS) { + /* + * The in-core version of sb_qflags do not have + * XFS_OQUOTA_* flags, whereas the on-disk version + * does. So, convert incore XFS_{PG}QUOTA_* flags + * to on-disk XFS_OQUOTA_* flags. + */ + qflags &= ~(XFS_PQUOTA_ENFD | XFS_PQUOTA_CHKD | + XFS_GQUOTA_ENFD | XFS_GQUOTA_CHKD); + + if (from->sb_qflags & + (XFS_PQUOTA_ENFD | XFS_GQUOTA_ENFD)) + qflags |= XFS_OQUOTA_ENFD; + if (from->sb_qflags & + (XFS_PQUOTA_CHKD | XFS_GQUOTA_CHKD)) + qflags |= XFS_OQUOTA_CHKD; + to->sb_qflags = cpu_to_be16(qflags); + *fields &= ~XFS_SB_QFLAGS; + } + + /* + * GQUOTINO and PQUOTINO cannot be used together in versions + * of superblock that do not have pquotino. from->sb_flags + * tells us which quota is active and should be copied to + * disk. + */ + if ((*fields & XFS_SB_GQUOTINO) && + (from->sb_qflags & XFS_GQUOTA_ACCT)) + to->sb_gquotino = cpu_to_be64(from->sb_gquotino); + else if ((*fields & XFS_SB_PQUOTINO) && + (from->sb_qflags & XFS_PQUOTA_ACCT)) + to->sb_gquotino = cpu_to_be64(from->sb_pquotino); + + *fields &= ~(XFS_SB_PQUOTINO | XFS_SB_GQUOTINO); +} + +/* + * Copy in core superblock to ondisk one. + * + * The fields argument is mask of superblock fields to copy. + */ +void +xfs_sb_to_disk( + xfs_dsb_t *to, + xfs_sb_t *from, + __int64_t fields) +{ + xfs_caddr_t to_ptr = (xfs_caddr_t)to; + xfs_caddr_t from_ptr = (xfs_caddr_t)from; + xfs_sb_field_t f; + int first; + int size; + + ASSERT(fields); + if (!fields) + return; + + xfs_sb_quota_to_disk(to, from, &fields); + while (fields) { + f = (xfs_sb_field_t)xfs_lowbit64((__uint64_t)fields); + first = xfs_sb_info[f].offset; + size = xfs_sb_info[f + 1].offset - first; + + ASSERT(xfs_sb_info[f].type == 0 || xfs_sb_info[f].type == 1); + + if (size == 1 || xfs_sb_info[f].type == 1) { + memcpy(to_ptr + first, from_ptr + first, size); + } else { + switch (size) { + case 2: + *(__be16 *)(to_ptr + first) = + cpu_to_be16(*(__u16 *)(from_ptr + first)); + break; + case 4: + *(__be32 *)(to_ptr + first) = + cpu_to_be32(*(__u32 *)(from_ptr + first)); + break; + case 8: + *(__be64 *)(to_ptr + first) = + cpu_to_be64(*(__u64 *)(from_ptr + first)); + break; + default: + ASSERT(0); + } + } + + fields &= ~(1LL << f); + } +} + +static int +xfs_sb_verify( + struct xfs_buf *bp, + bool check_version) +{ + struct xfs_mount *mp = bp->b_target->bt_mount; + struct xfs_sb sb; + + xfs_sb_from_disk(&sb, XFS_BUF_TO_SBP(bp)); + + /* + * Only check the in progress field for the primary superblock as + * mkfs.xfs doesn't clear it from secondary superblocks. + */ + return xfs_mount_validate_sb(mp, &sb, bp->b_bn == XFS_SB_DADDR, + check_version); +} + +/* + * If the superblock has the CRC feature bit set or the CRC field is non-null, + * check that the CRC is valid. We check the CRC field is non-null because a + * single bit error could clear the feature bit and unused parts of the + * superblock are supposed to be zero. Hence a non-null crc field indicates that + * we've potentially lost a feature bit and we should check it anyway. + */ +static void +xfs_sb_read_verify( + struct xfs_buf *bp) +{ + struct xfs_mount *mp = bp->b_target->bt_mount; + struct xfs_dsb *dsb = XFS_BUF_TO_SBP(bp); + int error; + + /* + * open code the version check to avoid needing to convert the entire + * superblock from disk order just to check the version number + */ + if (dsb->sb_magicnum == cpu_to_be32(XFS_SB_MAGIC) && + (((be16_to_cpu(dsb->sb_versionnum) & XFS_SB_VERSION_NUMBITS) == + XFS_SB_VERSION_5) || + dsb->sb_crc != 0)) { + + if (!xfs_verify_cksum(bp->b_addr, be16_to_cpu(dsb->sb_sectsize), + offsetof(struct xfs_sb, sb_crc))) { + error = EFSCORRUPTED; + goto out_error; + } + } + error = xfs_sb_verify(bp, true); + +out_error: + if (error) { + XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, + mp, bp->b_addr); + xfs_buf_ioerror(bp, error); + } +} + +/* + * We may be probed for a filesystem match, so we may not want to emit + * messages when the superblock buffer is not actually an XFS superblock. + * If we find an XFS superblock, then run a normal, noisy mount because we are + * really going to mount it and want to know about errors. + */ +static void +xfs_sb_quiet_read_verify( + struct xfs_buf *bp) +{ + struct xfs_dsb *dsb = XFS_BUF_TO_SBP(bp); + + + if (dsb->sb_magicnum == cpu_to_be32(XFS_SB_MAGIC)) { + /* XFS filesystem, verify noisily! */ + xfs_sb_read_verify(bp); + return; + } + /* quietly fail */ + xfs_buf_ioerror(bp, EWRONGFS); +} + +static void +xfs_sb_write_verify( + struct xfs_buf *bp) +{ + struct xfs_mount *mp = bp->b_target->bt_mount; + struct xfs_buf_log_item *bip = bp->b_fspriv; + int error; + + error = xfs_sb_verify(bp, false); + if (error) { + XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, + mp, bp->b_addr); + xfs_buf_ioerror(bp, error); + return; + } + + if (!xfs_sb_version_hascrc(&mp->m_sb)) + return; + + if (bip) + XFS_BUF_TO_SBP(bp)->sb_lsn = cpu_to_be64(bip->bli_item.li_lsn); + + xfs_update_cksum(bp->b_addr, BBTOB(bp->b_length), + offsetof(struct xfs_sb, sb_crc)); +} + +const struct xfs_buf_ops xfs_sb_buf_ops = { + .verify_read = xfs_sb_read_verify, + .verify_write = xfs_sb_write_verify, +}; + +const struct xfs_buf_ops xfs_sb_quiet_buf_ops = { + .verify_read = xfs_sb_quiet_read_verify, + .verify_write = xfs_sb_write_verify, +}; + +/* + * xfs_mount_common + * + * Mount initialization code establishing various mount + * fields from the superblock associated with the given + * mount structure + */ +void +xfs_sb_mount_common( + struct xfs_mount *mp, + struct xfs_sb *sbp) +{ + mp->m_agfrotor = mp->m_agirotor = 0; + spin_lock_init(&mp->m_agirotor_lock); + mp->m_maxagi = mp->m_sb.sb_agcount; + mp->m_blkbit_log = sbp->sb_blocklog + XFS_NBBYLOG; + mp->m_blkbb_log = sbp->sb_blocklog - BBSHIFT; + mp->m_sectbb_log = sbp->sb_sectlog - BBSHIFT; + mp->m_agno_log = xfs_highbit32(sbp->sb_agcount - 1) + 1; + mp->m_agino_log = sbp->sb_inopblog + sbp->sb_agblklog; + mp->m_blockmask = sbp->sb_blocksize - 1; + mp->m_blockwsize = sbp->sb_blocksize >> XFS_WORDLOG; + mp->m_blockwmask = mp->m_blockwsize - 1; + + mp->m_alloc_mxr[0] = xfs_allocbt_maxrecs(mp, sbp->sb_blocksize, 1); + mp->m_alloc_mxr[1] = xfs_allocbt_maxrecs(mp, sbp->sb_blocksize, 0); + mp->m_alloc_mnr[0] = mp->m_alloc_mxr[0] / 2; + mp->m_alloc_mnr[1] = mp->m_alloc_mxr[1] / 2; + + mp->m_inobt_mxr[0] = xfs_inobt_maxrecs(mp, sbp->sb_blocksize, 1); + mp->m_inobt_mxr[1] = xfs_inobt_maxrecs(mp, sbp->sb_blocksize, 0); + mp->m_inobt_mnr[0] = mp->m_inobt_mxr[0] / 2; + mp->m_inobt_mnr[1] = mp->m_inobt_mxr[1] / 2; + + mp->m_bmap_dmxr[0] = xfs_bmbt_maxrecs(mp, sbp->sb_blocksize, 1); + mp->m_bmap_dmxr[1] = xfs_bmbt_maxrecs(mp, sbp->sb_blocksize, 0); + mp->m_bmap_dmnr[0] = mp->m_bmap_dmxr[0] / 2; + mp->m_bmap_dmnr[1] = mp->m_bmap_dmxr[1] / 2; + + mp->m_bsize = XFS_FSB_TO_BB(mp, 1); + mp->m_ialloc_inos = (int)MAX((__uint16_t)XFS_INODES_PER_CHUNK, + sbp->sb_inopblock); + mp->m_ialloc_blks = mp->m_ialloc_inos >> sbp->sb_inopblog; +} + +/* + * xfs_initialize_perag_data + * + * Read in each per-ag structure so we can count up the number of + * allocated inodes, free inodes and used filesystem blocks as this + * information is no longer persistent in the superblock. Once we have + * this information, write it into the in-core superblock structure. + */ +int +xfs_initialize_perag_data( + struct xfs_mount *mp, + xfs_agnumber_t agcount) +{ + xfs_agnumber_t index; + xfs_perag_t *pag; + xfs_sb_t *sbp = &mp->m_sb; + uint64_t ifree = 0; + uint64_t ialloc = 0; + uint64_t bfree = 0; + uint64_t bfreelst = 0; + uint64_t btree = 0; + int error; + + for (index = 0; index < agcount; index++) { + /* + * read the agf, then the agi. This gets us + * all the information we need and populates the + * per-ag structures for us. + */ + error = xfs_alloc_pagf_init(mp, NULL, index, 0); + if (error) + return error; + + error = xfs_ialloc_pagi_init(mp, NULL, index); + if (error) + return error; + pag = xfs_perag_get(mp, index); + ifree += pag->pagi_freecount; + ialloc += pag->pagi_count; + bfree += pag->pagf_freeblks; + bfreelst += pag->pagf_flcount; + btree += pag->pagf_btreeblks; + xfs_perag_put(pag); + } + /* + * Overwrite incore superblock counters with just-read data + */ + spin_lock(&mp->m_sb_lock); + sbp->sb_ifree = ifree; + sbp->sb_icount = ialloc; + sbp->sb_fdblocks = bfree + bfreelst + btree; + spin_unlock(&mp->m_sb_lock); + + /* Fixup the per-cpu counters as well. */ + xfs_icsb_reinit_counters(mp); + + return 0; +} + +/* + * xfs_mod_sb() can be used to copy arbitrary changes to the + * in-core superblock into the superblock buffer to be logged. + * It does not provide the higher level of locking that is + * needed to protect the in-core superblock from concurrent + * access. + */ +void +xfs_mod_sb(xfs_trans_t *tp, __int64_t fields) +{ + xfs_buf_t *bp; + int first; + int last; + xfs_mount_t *mp; + xfs_sb_field_t f; + + ASSERT(fields); + if (!fields) + return; + mp = tp->t_mountp; + bp = xfs_trans_getsb(tp, mp, 0); + first = sizeof(xfs_sb_t); + last = 0; + + /* translate/copy */ + + xfs_sb_to_disk(XFS_BUF_TO_SBP(bp), &mp->m_sb, fields); + + /* find modified range */ + f = (xfs_sb_field_t)xfs_highbit64((__uint64_t)fields); + ASSERT((1LL << f) & XFS_SB_MOD_BITS); + last = xfs_sb_info[f + 1].offset - 1; + + f = (xfs_sb_field_t)xfs_lowbit64((__uint64_t)fields); + ASSERT((1LL << f) & XFS_SB_MOD_BITS); + first = xfs_sb_info[f].offset; + + xfs_trans_buf_set_type(tp, bp, XFS_BLFT_SB_BUF); + xfs_trans_log_buf(tp, bp, first, last); +} diff --git a/fs/xfs/xfs_sb.h b/fs/xfs/xfs_sb.h index 78f9e70..6835b44 100644 --- a/fs/xfs/xfs_sb.h +++ b/fs/xfs/xfs_sb.h @@ -26,6 +26,7 @@ struct xfs_buf; struct xfs_mount; +struct xfs_trans; #define XFS_SB_MAGIC 0x58465342 /* 'XFSB' */ #define XFS_SB_VERSION_1 1 /* 5.3, 6.0.1, 6.1 */ @@ -83,11 +84,13 @@ struct xfs_mount; #define XFS_SB_VERSION2_PARENTBIT 0x00000010 /* parent pointers */ #define XFS_SB_VERSION2_PROJID32BIT 0x00000080 /* 32 bit project id */ #define XFS_SB_VERSION2_CRCBIT 0x00000100 /* metadata CRCs */ +#define XFS_SB_VERSION2_FTYPE 0x00000200 /* inode type in dir */ #define XFS_SB_VERSION2_OKREALFBITS \ (XFS_SB_VERSION2_LAZYSBCOUNTBIT | \ XFS_SB_VERSION2_ATTR2BIT | \ - XFS_SB_VERSION2_PROJID32BIT) + XFS_SB_VERSION2_PROJID32BIT | \ + XFS_SB_VERSION2_FTYPE) #define XFS_SB_VERSION2_OKSASHFBITS \ (0) #define XFS_SB_VERSION2_OKREALBITS \ @@ -354,15 +357,8 @@ static inline int xfs_sb_good_version(xfs_sb_t *sbp) (sbp->sb_features2 & ~XFS_SB_VERSION2_OKREALBITS))) return 0; -#ifdef __KERNEL__ if (sbp->sb_shared_vn > XFS_SB_MAX_SHARED_VN) return 0; -#else - if ((sbp->sb_versionnum & XFS_SB_VERSION_SHAREDBIT) && - sbp->sb_shared_vn > XFS_SB_MAX_SHARED_VN) - return 0; -#endif - return 1; } if (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5) @@ -554,12 +550,13 @@ static inline int xfs_sb_version_hasprojid32bit(xfs_sb_t *sbp) (sbp->sb_features2 & XFS_SB_VERSION2_PROJID32BIT)); } -static inline int xfs_sb_version_hascrc(xfs_sb_t *sbp) +static inline void xfs_sb_version_addprojid32bit(xfs_sb_t *sbp) { - return XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5; + sbp->sb_versionnum |= XFS_SB_VERSION_MOREBITSBIT; + sbp->sb_features2 |= XFS_SB_VERSION2_PROJID32BIT; + sbp->sb_bad_features2 |= XFS_SB_VERSION2_PROJID32BIT; } - /* * Extended v5 superblock feature masks. These are to be used for new v5 * superblock features only. @@ -598,7 +595,10 @@ xfs_sb_has_ro_compat_feature( return (sbp->sb_features_ro_compat & feature) != 0; } -#define XFS_SB_FEAT_INCOMPAT_ALL 0 +#define XFS_SB_FEAT_INCOMPAT_FTYPE (1 << 0) /* filetype in dirent */ +#define XFS_SB_FEAT_INCOMPAT_ALL \ + (XFS_SB_FEAT_INCOMPAT_FTYPE) + #define XFS_SB_FEAT_INCOMPAT_UNKNOWN ~XFS_SB_FEAT_INCOMPAT_ALL static inline bool xfs_sb_has_incompat_feature( @@ -618,16 +618,39 @@ xfs_sb_has_incompat_log_feature( return (sbp->sb_features_log_incompat & feature) != 0; } -static inline bool -xfs_is_quota_inode(struct xfs_sb *sbp, xfs_ino_t ino) +/* + * V5 superblock specific feature checks + */ +static inline int xfs_sb_version_hascrc(xfs_sb_t *sbp) { - return (ino == sbp->sb_uquotino || ino == sbp->sb_gquotino); + return XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5; +} + +static inline int xfs_sb_version_has_pquotino(xfs_sb_t *sbp) +{ + return XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5; +} + +static inline int xfs_sb_version_hasftype(struct xfs_sb *sbp) +{ + return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5 && + xfs_sb_has_incompat_feature(sbp, XFS_SB_FEAT_INCOMPAT_FTYPE)) || + (xfs_sb_version_hasmorebits(sbp) && + (sbp->sb_features2 & XFS_SB_VERSION2_FTYPE)); } /* * end of superblock version macros */ +static inline bool +xfs_is_quota_inode(struct xfs_sb *sbp, xfs_ino_t ino) +{ + return (ino == sbp->sb_uquotino || + ino == sbp->sb_gquotino || + ino == sbp->sb_pquotino); +} + #define XFS_SB_DADDR ((xfs_daddr_t)0) /* daddr in filesystem/ag */ #define XFS_SB_BLOCK(mp) XFS_HDR_BLOCK(mp, XFS_SB_DADDR) #define XFS_BUF_TO_SBP(bp) ((xfs_dsb_t *)((bp)->b_addr)) @@ -660,4 +683,23 @@ xfs_is_quota_inode(struct xfs_sb *sbp, xfs_ino_t ino) #define XFS_B_TO_FSBT(mp,b) (((__uint64_t)(b)) >> (mp)->m_sb.sb_blocklog) #define XFS_B_FSB_OFFSET(mp,b) ((b) & (mp)->m_blockmask) +/* + * perag get/put wrappers for ref counting + */ +extern struct xfs_perag *xfs_perag_get(struct xfs_mount *, xfs_agnumber_t); +extern struct xfs_perag *xfs_perag_get_tag(struct xfs_mount *, xfs_agnumber_t, + int tag); +extern void xfs_perag_put(struct xfs_perag *pag); +extern int xfs_initialize_perag_data(struct xfs_mount *, xfs_agnumber_t); + +extern void xfs_sb_calc_crc(struct xfs_buf *); +extern void xfs_mod_sb(struct xfs_trans *, __int64_t); +extern void xfs_sb_mount_common(struct xfs_mount *, struct xfs_sb *); +extern void xfs_sb_from_disk(struct xfs_sb *, struct xfs_dsb *); +extern void xfs_sb_to_disk(struct xfs_dsb *, struct xfs_sb *, __int64_t); +extern void xfs_sb_quota_from_disk(struct xfs_sb *sbp); + +extern const struct xfs_buf_ops xfs_sb_buf_ops; +extern const struct xfs_buf_ops xfs_sb_quiet_buf_ops; + #endif /* __XFS_SB_H__ */ diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c index 1d68ffc..979a77d 100644 --- a/fs/xfs/xfs_super.c +++ b/fs/xfs/xfs_super.c @@ -17,12 +17,12 @@ */ #include "xfs.h" +#include "xfs_format.h" #include "xfs_log.h" #include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" #include "xfs_ag.h" -#include "xfs_dir2.h" #include "xfs_alloc.h" #include "xfs_quota.h" #include "xfs_mount.h" @@ -40,12 +40,12 @@ #include "xfs_fsops.h" #include "xfs_attr.h" #include "xfs_buf_item.h" -#include "xfs_utils.h" -#include "xfs_vnodeops.h" #include "xfs_log_priv.h" #include "xfs_trans_priv.h" #include "xfs_filestream.h" #include "xfs_da_btree.h" +#include "xfs_dir2_format.h" +#include "xfs_dir2.h" #include "xfs_extfree_item.h" #include "xfs_mru_cache.h" #include "xfs_inode_item.h" @@ -421,12 +421,6 @@ xfs_parseargs( } #endif - if ((mp->m_qflags & (XFS_GQUOTA_ACCT | XFS_GQUOTA_ACTIVE)) && - (mp->m_qflags & (XFS_PQUOTA_ACCT | XFS_PQUOTA_ACTIVE))) { - xfs_warn(mp, "cannot mount with both project and group quota"); - return EINVAL; - } - if ((dsunit && !dswidth) || (!dsunit && dswidth)) { xfs_warn(mp, "sunit and swidth must be specified together"); return EINVAL; @@ -556,14 +550,13 @@ xfs_showargs( else if (mp->m_qflags & XFS_UQUOTA_ACCT) seq_puts(m, "," MNTOPT_UQUOTANOENF); - /* Either project or group quotas can be active, not both */ - if (mp->m_qflags & XFS_PQUOTA_ACCT) { if (mp->m_qflags & XFS_PQUOTA_ENFD) seq_puts(m, "," MNTOPT_PRJQUOTA); else seq_puts(m, "," MNTOPT_PQUOTANOENF); - } else if (mp->m_qflags & XFS_GQUOTA_ACCT) { + } + if (mp->m_qflags & XFS_GQUOTA_ACCT) { if (mp->m_qflags & XFS_GQUOTA_ENFD) seq_puts(m, "," MNTOPT_GRPQUOTA); else @@ -870,17 +863,17 @@ xfs_init_mount_workqueues( goto out_destroy_unwritten; mp->m_reclaim_workqueue = alloc_workqueue("xfs-reclaim/%s", - WQ_NON_REENTRANT, 0, mp->m_fsname); + 0, 0, mp->m_fsname); if (!mp->m_reclaim_workqueue) goto out_destroy_cil; mp->m_log_workqueue = alloc_workqueue("xfs-log/%s", - WQ_NON_REENTRANT, 0, mp->m_fsname); + 0, 0, mp->m_fsname); if (!mp->m_log_workqueue) goto out_destroy_reclaim; mp->m_eofblocks_workqueue = alloc_workqueue("xfs-eofblocks/%s", - WQ_NON_REENTRANT, 0, mp->m_fsname); + 0, 0, mp->m_fsname); if (!mp->m_eofblocks_workqueue) goto out_destroy_log; @@ -1396,6 +1389,14 @@ xfs_finish_flags( return XFS_ERROR(EROFS); } + if ((mp->m_qflags & (XFS_GQUOTA_ACCT | XFS_GQUOTA_ACTIVE)) && + (mp->m_qflags & (XFS_PQUOTA_ACCT | XFS_PQUOTA_ACTIVE)) && + !xfs_sb_version_has_pquotino(&mp->m_sb)) { + xfs_warn(mp, + "Super block does not support project and group quota together"); + return XFS_ERROR(EINVAL); + } + return 0; } diff --git a/fs/xfs/xfs_symlink.c b/fs/xfs/xfs_symlink.c index f4895b6..2f2a7c0 100644 --- a/fs/xfs/xfs_symlink.c +++ b/fs/xfs/xfs_symlink.c @@ -18,200 +18,29 @@ */ #include "xfs.h" #include "xfs_fs.h" -#include "xfs_types.h" +#include "xfs_format.h" #include "xfs_bit.h" #include "xfs_log.h" #include "xfs_trans.h" #include "xfs_sb.h" #include "xfs_ag.h" -#include "xfs_dir2.h" #include "xfs_mount.h" #include "xfs_da_btree.h" +#include "xfs_dir2_format.h" +#include "xfs_dir2.h" #include "xfs_bmap_btree.h" #include "xfs_ialloc_btree.h" #include "xfs_dinode.h" #include "xfs_inode.h" -#include "xfs_inode_item.h" -#include "xfs_itable.h" #include "xfs_ialloc.h" #include "xfs_alloc.h" #include "xfs_bmap.h" +#include "xfs_bmap_util.h" #include "xfs_error.h" #include "xfs_quota.h" -#include "xfs_utils.h" #include "xfs_trans_space.h" -#include "xfs_log_priv.h" #include "xfs_trace.h" #include "xfs_symlink.h" -#include "xfs_cksum.h" -#include "xfs_buf_item.h" - - -/* - * Each contiguous block has a header, so it is not just a simple pathlen - * to FSB conversion. - */ -int -xfs_symlink_blocks( - struct xfs_mount *mp, - int pathlen) -{ - int buflen = XFS_SYMLINK_BUF_SPACE(mp, mp->m_sb.sb_blocksize); - - return (pathlen + buflen - 1) / buflen; -} - -static int -xfs_symlink_hdr_set( - struct xfs_mount *mp, - xfs_ino_t ino, - uint32_t offset, - uint32_t size, - struct xfs_buf *bp) -{ - struct xfs_dsymlink_hdr *dsl = bp->b_addr; - - if (!xfs_sb_version_hascrc(&mp->m_sb)) - return 0; - - dsl->sl_magic = cpu_to_be32(XFS_SYMLINK_MAGIC); - dsl->sl_offset = cpu_to_be32(offset); - dsl->sl_bytes = cpu_to_be32(size); - uuid_copy(&dsl->sl_uuid, &mp->m_sb.sb_uuid); - dsl->sl_owner = cpu_to_be64(ino); - dsl->sl_blkno = cpu_to_be64(bp->b_bn); - bp->b_ops = &xfs_symlink_buf_ops; - - return sizeof(struct xfs_dsymlink_hdr); -} - -/* - * Checking of the symlink header is split into two parts. the verifier does - * CRC, location and bounds checking, the unpacking function checks the path - * parameters and owner. - */ -bool -xfs_symlink_hdr_ok( - struct xfs_mount *mp, - xfs_ino_t ino, - uint32_t offset, - uint32_t size, - struct xfs_buf *bp) -{ - struct xfs_dsymlink_hdr *dsl = bp->b_addr; - - if (offset != be32_to_cpu(dsl->sl_offset)) - return false; - if (size != be32_to_cpu(dsl->sl_bytes)) - return false; - if (ino != be64_to_cpu(dsl->sl_owner)) - return false; - - /* ok */ - return true; -} - -static bool -xfs_symlink_verify( - struct xfs_buf *bp) -{ - struct xfs_mount *mp = bp->b_target->bt_mount; - struct xfs_dsymlink_hdr *dsl = bp->b_addr; - - if (!xfs_sb_version_hascrc(&mp->m_sb)) - return false; - if (dsl->sl_magic != cpu_to_be32(XFS_SYMLINK_MAGIC)) - return false; - if (!uuid_equal(&dsl->sl_uuid, &mp->m_sb.sb_uuid)) - return false; - if (bp->b_bn != be64_to_cpu(dsl->sl_blkno)) - return false; - if (be32_to_cpu(dsl->sl_offset) + - be32_to_cpu(dsl->sl_bytes) >= MAXPATHLEN) - return false; - if (dsl->sl_owner == 0) - return false; - - return true; -} - -static void -xfs_symlink_read_verify( - struct xfs_buf *bp) -{ - struct xfs_mount *mp = bp->b_target->bt_mount; - - /* no verification of non-crc buffers */ - if (!xfs_sb_version_hascrc(&mp->m_sb)) - return; - - if (!xfs_verify_cksum(bp->b_addr, BBTOB(bp->b_length), - offsetof(struct xfs_dsymlink_hdr, sl_crc)) || - !xfs_symlink_verify(bp)) { - XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr); - xfs_buf_ioerror(bp, EFSCORRUPTED); - } -} - -static void -xfs_symlink_write_verify( - struct xfs_buf *bp) -{ - struct xfs_mount *mp = bp->b_target->bt_mount; - struct xfs_buf_log_item *bip = bp->b_fspriv; - - /* no verification of non-crc buffers */ - if (!xfs_sb_version_hascrc(&mp->m_sb)) - return; - - if (!xfs_symlink_verify(bp)) { - XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr); - xfs_buf_ioerror(bp, EFSCORRUPTED); - return; - } - - if (bip) { - struct xfs_dsymlink_hdr *dsl = bp->b_addr; - dsl->sl_lsn = cpu_to_be64(bip->bli_item.li_lsn); - } - xfs_update_cksum(bp->b_addr, BBTOB(bp->b_length), - offsetof(struct xfs_dsymlink_hdr, sl_crc)); -} - -const struct xfs_buf_ops xfs_symlink_buf_ops = { - .verify_read = xfs_symlink_read_verify, - .verify_write = xfs_symlink_write_verify, -}; - -void -xfs_symlink_local_to_remote( - struct xfs_trans *tp, - struct xfs_buf *bp, - struct xfs_inode *ip, - struct xfs_ifork *ifp) -{ - struct xfs_mount *mp = ip->i_mount; - char *buf; - - if (!xfs_sb_version_hascrc(&mp->m_sb)) { - bp->b_ops = NULL; - memcpy(bp->b_addr, ifp->if_u1.if_data, ifp->if_bytes); - return; - } - - /* - * As this symlink fits in an inode literal area, it must also fit in - * the smallest buffer the filesystem supports. - */ - ASSERT(BBTOB(bp->b_length) >= - ifp->if_bytes + sizeof(struct xfs_dsymlink_hdr)); - - bp->b_ops = &xfs_symlink_buf_ops; - - buf = bp->b_addr; - buf += xfs_symlink_hdr_set(mp, ip->i_ino, 0, ifp->if_bytes, bp); - memcpy(buf, ifp->if_u1.if_data, ifp->if_bytes); -} /* ----- Kernel only functions below ----- */ STATIC int @@ -386,8 +215,11 @@ xfs_symlink( /* * Make sure that we have allocated dquot(s) on disk. */ - error = xfs_qm_vop_dqalloc(dp, current_fsuid(), current_fsgid(), prid, - XFS_QMOPT_QUOTALL | XFS_QMOPT_INHERIT, &udqp, &gdqp, &pdqp); + error = xfs_qm_vop_dqalloc(dp, + xfs_kuid_to_uid(current_fsuid()), + xfs_kgid_to_gid(current_fsgid()), prid, + XFS_QMOPT_QUOTALL | XFS_QMOPT_INHERIT, + &udqp, &gdqp, &pdqp); if (error) goto std_return; @@ -402,12 +234,10 @@ xfs_symlink( else fs_blocks = xfs_symlink_blocks(mp, pathlen); resblks = XFS_SYMLINK_SPACE_RES(mp, link_name->len, fs_blocks); - error = xfs_trans_reserve(tp, resblks, XFS_SYMLINK_LOG_RES(mp), 0, - XFS_TRANS_PERM_LOG_RES, XFS_SYMLINK_LOG_COUNT); + error = xfs_trans_reserve(tp, &M_RES(mp)->tr_symlink, resblks, 0); if (error == ENOSPC && fs_blocks == 0) { resblks = 0; - error = xfs_trans_reserve(tp, 0, XFS_SYMLINK_LOG_RES(mp), 0, - XFS_TRANS_PERM_LOG_RES, XFS_SYMLINK_LOG_COUNT); + error = xfs_trans_reserve(tp, &M_RES(mp)->tr_symlink, 0, 0); } if (error) { cancel_flags = 0; @@ -710,8 +540,8 @@ xfs_inactive_symlink_rmt( * Put an itruncate log reservation in the new transaction * for our caller. */ - if ((error = xfs_trans_reserve(tp, 0, XFS_ITRUNCATE_LOG_RES(mp), 0, - XFS_TRANS_PERM_LOG_RES, XFS_ITRUNCATE_LOG_COUNT))) { + error = xfs_trans_reserve(tp, &M_RES(mp)->tr_itruncate, 0, 0); + if (error) { ASSERT(XFS_FORCED_SHUTDOWN(mp)); goto error0; } diff --git a/fs/xfs/xfs_symlink.h b/fs/xfs/xfs_symlink.h index 3743948..99338ba 100644 --- a/fs/xfs/xfs_symlink.h +++ b/fs/xfs/xfs_symlink.h @@ -17,50 +17,11 @@ #ifndef __XFS_SYMLINK_H #define __XFS_SYMLINK_H 1 -struct xfs_mount; -struct xfs_trans; -struct xfs_inode; -struct xfs_buf; -struct xfs_ifork; -struct xfs_name; - -#define XFS_SYMLINK_MAGIC 0x58534c4d /* XSLM */ - -struct xfs_dsymlink_hdr { - __be32 sl_magic; - __be32 sl_offset; - __be32 sl_bytes; - __be32 sl_crc; - uuid_t sl_uuid; - __be64 sl_owner; - __be64 sl_blkno; - __be64 sl_lsn; -}; - -/* - * The maximum pathlen is 1024 bytes. Since the minimum file system - * blocksize is 512 bytes, we can get a max of 3 extents back from - * bmapi when crc headers are taken into account. - */ -#define XFS_SYMLINK_MAPS 3 - -#define XFS_SYMLINK_BUF_SPACE(mp, bufsize) \ - ((bufsize) - (xfs_sb_version_hascrc(&(mp)->m_sb) ? \ - sizeof(struct xfs_dsymlink_hdr) : 0)) - -int xfs_symlink_blocks(struct xfs_mount *mp, int pathlen); - -void xfs_symlink_local_to_remote(struct xfs_trans *tp, struct xfs_buf *bp, - struct xfs_inode *ip, struct xfs_ifork *ifp); - -extern const struct xfs_buf_ops xfs_symlink_buf_ops; - -#ifdef __KERNEL__ +/* Kernel only symlink defintions */ int xfs_symlink(struct xfs_inode *dp, struct xfs_name *link_name, const char *target_path, umode_t mode, struct xfs_inode **ipp); int xfs_readlink(struct xfs_inode *ip, char *link); int xfs_inactive_symlink(struct xfs_inode *ip, struct xfs_trans **tpp); -#endif /* __KERNEL__ */ #endif /* __XFS_SYMLINK_H */ diff --git a/fs/xfs/xfs_symlink_remote.c b/fs/xfs/xfs_symlink_remote.c new file mode 100644 index 0000000..01c85e3 --- /dev/null +++ b/fs/xfs/xfs_symlink_remote.c @@ -0,0 +1,200 @@ +/* + * Copyright (c) 2000-2006 Silicon Graphics, Inc. + * Copyright (c) 2012-2013 Red Hat, Inc. + * All rights reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#include "xfs.h" +#include "xfs_fs.h" +#include "xfs_format.h" +#include "xfs_log.h" +#include "xfs_trans.h" +#include "xfs_ag.h" +#include "xfs_sb.h" +#include "xfs_mount.h" +#include "xfs_bmap_btree.h" +#include "xfs_inode.h" +#include "xfs_error.h" +#include "xfs_trace.h" +#include "xfs_symlink.h" +#include "xfs_cksum.h" +#include "xfs_buf_item.h" + + +/* + * Each contiguous block has a header, so it is not just a simple pathlen + * to FSB conversion. + */ +int +xfs_symlink_blocks( + struct xfs_mount *mp, + int pathlen) +{ + int buflen = XFS_SYMLINK_BUF_SPACE(mp, mp->m_sb.sb_blocksize); + + return (pathlen + buflen - 1) / buflen; +} + +int +xfs_symlink_hdr_set( + struct xfs_mount *mp, + xfs_ino_t ino, + uint32_t offset, + uint32_t size, + struct xfs_buf *bp) +{ + struct xfs_dsymlink_hdr *dsl = bp->b_addr; + + if (!xfs_sb_version_hascrc(&mp->m_sb)) + return 0; + + dsl->sl_magic = cpu_to_be32(XFS_SYMLINK_MAGIC); + dsl->sl_offset = cpu_to_be32(offset); + dsl->sl_bytes = cpu_to_be32(size); + uuid_copy(&dsl->sl_uuid, &mp->m_sb.sb_uuid); + dsl->sl_owner = cpu_to_be64(ino); + dsl->sl_blkno = cpu_to_be64(bp->b_bn); + bp->b_ops = &xfs_symlink_buf_ops; + + return sizeof(struct xfs_dsymlink_hdr); +} + +/* + * Checking of the symlink header is split into two parts. the verifier does + * CRC, location and bounds checking, the unpacking function checks the path + * parameters and owner. + */ +bool +xfs_symlink_hdr_ok( + struct xfs_mount *mp, + xfs_ino_t ino, + uint32_t offset, + uint32_t size, + struct xfs_buf *bp) +{ + struct xfs_dsymlink_hdr *dsl = bp->b_addr; + + if (offset != be32_to_cpu(dsl->sl_offset)) + return false; + if (size != be32_to_cpu(dsl->sl_bytes)) + return false; + if (ino != be64_to_cpu(dsl->sl_owner)) + return false; + + /* ok */ + return true; +} + +static bool +xfs_symlink_verify( + struct xfs_buf *bp) +{ + struct xfs_mount *mp = bp->b_target->bt_mount; + struct xfs_dsymlink_hdr *dsl = bp->b_addr; + + if (!xfs_sb_version_hascrc(&mp->m_sb)) + return false; + if (dsl->sl_magic != cpu_to_be32(XFS_SYMLINK_MAGIC)) + return false; + if (!uuid_equal(&dsl->sl_uuid, &mp->m_sb.sb_uuid)) + return false; + if (bp->b_bn != be64_to_cpu(dsl->sl_blkno)) + return false; + if (be32_to_cpu(dsl->sl_offset) + + be32_to_cpu(dsl->sl_bytes) >= MAXPATHLEN) + return false; + if (dsl->sl_owner == 0) + return false; + + return true; +} + +static void +xfs_symlink_read_verify( + struct xfs_buf *bp) +{ + struct xfs_mount *mp = bp->b_target->bt_mount; + + /* no verification of non-crc buffers */ + if (!xfs_sb_version_hascrc(&mp->m_sb)) + return; + + if (!xfs_verify_cksum(bp->b_addr, BBTOB(bp->b_length), + offsetof(struct xfs_dsymlink_hdr, sl_crc)) || + !xfs_symlink_verify(bp)) { + XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr); + xfs_buf_ioerror(bp, EFSCORRUPTED); + } +} + +static void +xfs_symlink_write_verify( + struct xfs_buf *bp) +{ + struct xfs_mount *mp = bp->b_target->bt_mount; + struct xfs_buf_log_item *bip = bp->b_fspriv; + + /* no verification of non-crc buffers */ + if (!xfs_sb_version_hascrc(&mp->m_sb)) + return; + + if (!xfs_symlink_verify(bp)) { + XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr); + xfs_buf_ioerror(bp, EFSCORRUPTED); + return; + } + + if (bip) { + struct xfs_dsymlink_hdr *dsl = bp->b_addr; + dsl->sl_lsn = cpu_to_be64(bip->bli_item.li_lsn); + } + xfs_update_cksum(bp->b_addr, BBTOB(bp->b_length), + offsetof(struct xfs_dsymlink_hdr, sl_crc)); +} + +const struct xfs_buf_ops xfs_symlink_buf_ops = { + .verify_read = xfs_symlink_read_verify, + .verify_write = xfs_symlink_write_verify, +}; + +void +xfs_symlink_local_to_remote( + struct xfs_trans *tp, + struct xfs_buf *bp, + struct xfs_inode *ip, + struct xfs_ifork *ifp) +{ + struct xfs_mount *mp = ip->i_mount; + char *buf; + + if (!xfs_sb_version_hascrc(&mp->m_sb)) { + bp->b_ops = NULL; + memcpy(bp->b_addr, ifp->if_u1.if_data, ifp->if_bytes); + return; + } + + /* + * As this symlink fits in an inode literal area, it must also fit in + * the smallest buffer the filesystem supports. + */ + ASSERT(BBTOB(bp->b_length) >= + ifp->if_bytes + sizeof(struct xfs_dsymlink_hdr)); + + bp->b_ops = &xfs_symlink_buf_ops; + + buf = bp->b_addr; + buf += xfs_symlink_hdr_set(mp, ip->i_ino, 0, ifp->if_bytes, bp); + memcpy(buf, ifp->if_u1.if_data, ifp->if_bytes); +} diff --git a/fs/xfs/xfs_trace.c b/fs/xfs/xfs_trace.c index b6e3897..5d7b3e4 100644 --- a/fs/xfs/xfs_trace.c +++ b/fs/xfs/xfs_trace.c @@ -18,6 +18,7 @@ #include "xfs.h" #include "xfs_fs.h" #include "xfs_types.h" +#include "xfs_format.h" #include "xfs_log.h" #include "xfs_trans.h" #include "xfs_sb.h" diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c index 35a2299..5411e01 100644 --- a/fs/xfs/xfs_trans.c +++ b/fs/xfs/xfs_trans.c @@ -18,7 +18,7 @@ */ #include "xfs.h" #include "xfs_fs.h" -#include "xfs_types.h" +#include "xfs_format.h" #include "xfs_log.h" #include "xfs_trans.h" #include "xfs_sb.h" @@ -49,629 +49,6 @@ kmem_zone_t *xfs_trans_zone; kmem_zone_t *xfs_log_item_desc_zone; /* - * A buffer has a format structure overhead in the log in addition - * to the data, so we need to take this into account when reserving - * space in a transaction for a buffer. Round the space required up - * to a multiple of 128 bytes so that we don't change the historical - * reservation that has been used for this overhead. - */ -STATIC uint -xfs_buf_log_overhead(void) -{ - return round_up(sizeof(struct xlog_op_header) + - sizeof(struct xfs_buf_log_format), 128); -} - -/* - * Calculate out transaction log reservation per item in bytes. - * - * The nbufs argument is used to indicate the number of items that - * will be changed in a transaction. size is used to tell how many - * bytes should be reserved per item. - */ -STATIC uint -xfs_calc_buf_res( - uint nbufs, - uint size) -{ - return nbufs * (size + xfs_buf_log_overhead()); -} - -/* - * Various log reservation values. - * - * These are based on the size of the file system block because that is what - * most transactions manipulate. Each adds in an additional 128 bytes per - * item logged to try to account for the overhead of the transaction mechanism. - * - * Note: Most of the reservations underestimate the number of allocation - * groups into which they could free extents in the xfs_bmap_finish() call. - * This is because the number in the worst case is quite high and quite - * unusual. In order to fix this we need to change xfs_bmap_finish() to free - * extents in only a single AG at a time. This will require changes to the - * EFI code as well, however, so that the EFI for the extents not freed is - * logged again in each transaction. See SGI PV #261917. - * - * Reservation functions here avoid a huge stack in xfs_trans_init due to - * register overflow from temporaries in the calculations. - */ - - -/* - * In a write transaction we can allocate a maximum of 2 - * extents. This gives: - * the inode getting the new extents: inode size - * the inode's bmap btree: max depth * block size - * the agfs of the ags from which the extents are allocated: 2 * sector - * the superblock free block counter: sector size - * the allocation btrees: 2 exts * 2 trees * (2 * max depth - 1) * block size - * And the bmap_finish transaction can free bmap blocks in a join: - * the agfs of the ags containing the blocks: 2 * sector size - * the agfls of the ags containing the blocks: 2 * sector size - * the super block free block counter: sector size - * the allocation btrees: 2 exts * 2 trees * (2 * max depth - 1) * block size - */ -STATIC uint -xfs_calc_write_reservation( - struct xfs_mount *mp) -{ - return XFS_DQUOT_LOGRES(mp) + - MAX((xfs_calc_buf_res(1, mp->m_sb.sb_inodesize) + - xfs_calc_buf_res(XFS_BM_MAXLEVELS(mp, XFS_DATA_FORK), - XFS_FSB_TO_B(mp, 1)) + - xfs_calc_buf_res(3, mp->m_sb.sb_sectsize) + - xfs_calc_buf_res(XFS_ALLOCFREE_LOG_COUNT(mp, 2), - XFS_FSB_TO_B(mp, 1))), - (xfs_calc_buf_res(5, mp->m_sb.sb_sectsize) + - xfs_calc_buf_res(XFS_ALLOCFREE_LOG_COUNT(mp, 2), - XFS_FSB_TO_B(mp, 1)))); -} - -/* - * In truncating a file we free up to two extents at once. We can modify: - * the inode being truncated: inode size - * the inode's bmap btree: (max depth + 1) * block size - * And the bmap_finish transaction can free the blocks and bmap blocks: - * the agf for each of the ags: 4 * sector size - * the agfl for each of the ags: 4 * sector size - * the super block to reflect the freed blocks: sector size - * worst case split in allocation btrees per extent assuming 4 extents: - * 4 exts * 2 trees * (2 * max depth - 1) * block size - * the inode btree: max depth * blocksize - * the allocation btrees: 2 trees * (max depth - 1) * block size - */ -STATIC uint -xfs_calc_itruncate_reservation( - struct xfs_mount *mp) -{ - return XFS_DQUOT_LOGRES(mp) + - MAX((xfs_calc_buf_res(1, mp->m_sb.sb_inodesize) + - xfs_calc_buf_res(XFS_BM_MAXLEVELS(mp, XFS_DATA_FORK) + 1, - XFS_FSB_TO_B(mp, 1))), - (xfs_calc_buf_res(9, mp->m_sb.sb_sectsize) + - xfs_calc_buf_res(XFS_ALLOCFREE_LOG_COUNT(mp, 4), - XFS_FSB_TO_B(mp, 1)) + - xfs_calc_buf_res(5, 0) + - xfs_calc_buf_res(XFS_ALLOCFREE_LOG_COUNT(mp, 1), - XFS_FSB_TO_B(mp, 1)) + - xfs_calc_buf_res(2 + XFS_IALLOC_BLOCKS(mp) + - mp->m_in_maxlevels, 0))); -} - -/* - * In renaming a files we can modify: - * the four inodes involved: 4 * inode size - * the two directory btrees: 2 * (max depth + v2) * dir block size - * the two directory bmap btrees: 2 * max depth * block size - * And the bmap_finish transaction can free dir and bmap blocks (two sets - * of bmap blocks) giving: - * the agf for the ags in which the blocks live: 3 * sector size - * the agfl for the ags in which the blocks live: 3 * sector size - * the superblock for the free block count: sector size - * the allocation btrees: 3 exts * 2 trees * (2 * max depth - 1) * block size - */ -STATIC uint -xfs_calc_rename_reservation( - struct xfs_mount *mp) -{ - return XFS_DQUOT_LOGRES(mp) + - MAX((xfs_calc_buf_res(4, mp->m_sb.sb_inodesize) + - xfs_calc_buf_res(2 * XFS_DIROP_LOG_COUNT(mp), - XFS_FSB_TO_B(mp, 1))), - (xfs_calc_buf_res(7, mp->m_sb.sb_sectsize) + - xfs_calc_buf_res(XFS_ALLOCFREE_LOG_COUNT(mp, 3), - XFS_FSB_TO_B(mp, 1)))); -} - -/* - * For creating a link to an inode: - * the parent directory inode: inode size - * the linked inode: inode size - * the directory btree could split: (max depth + v2) * dir block size - * the directory bmap btree could join or split: (max depth + v2) * blocksize - * And the bmap_finish transaction can free some bmap blocks giving: - * the agf for the ag in which the blocks live: sector size - * the agfl for the ag in which the blocks live: sector size - * the superblock for the free block count: sector size - * the allocation btrees: 2 trees * (2 * max depth - 1) * block size - */ -STATIC uint -xfs_calc_link_reservation( - struct xfs_mount *mp) -{ - return XFS_DQUOT_LOGRES(mp) + - MAX((xfs_calc_buf_res(2, mp->m_sb.sb_inodesize) + - xfs_calc_buf_res(XFS_DIROP_LOG_COUNT(mp), - XFS_FSB_TO_B(mp, 1))), - (xfs_calc_buf_res(3, mp->m_sb.sb_sectsize) + - xfs_calc_buf_res(XFS_ALLOCFREE_LOG_COUNT(mp, 1), - XFS_FSB_TO_B(mp, 1)))); -} - -/* - * For removing a directory entry we can modify: - * the parent directory inode: inode size - * the removed inode: inode size - * the directory btree could join: (max depth + v2) * dir block size - * the directory bmap btree could join or split: (max depth + v2) * blocksize - * And the bmap_finish transaction can free the dir and bmap blocks giving: - * the agf for the ag in which the blocks live: 2 * sector size - * the agfl for the ag in which the blocks live: 2 * sector size - * the superblock for the free block count: sector size - * the allocation btrees: 2 exts * 2 trees * (2 * max depth - 1) * block size - */ -STATIC uint -xfs_calc_remove_reservation( - struct xfs_mount *mp) -{ - return XFS_DQUOT_LOGRES(mp) + - MAX((xfs_calc_buf_res(2, mp->m_sb.sb_inodesize) + - xfs_calc_buf_res(XFS_DIROP_LOG_COUNT(mp), - XFS_FSB_TO_B(mp, 1))), - (xfs_calc_buf_res(5, mp->m_sb.sb_sectsize) + - xfs_calc_buf_res(XFS_ALLOCFREE_LOG_COUNT(mp, 2), - XFS_FSB_TO_B(mp, 1)))); -} - -/* - * For create, break it in to the two cases that the transaction - * covers. We start with the modify case - allocation done by modification - * of the state of existing inodes - and the allocation case. - */ - -/* - * For create we can modify: - * the parent directory inode: inode size - * the new inode: inode size - * the inode btree entry: block size - * the superblock for the nlink flag: sector size - * the directory btree: (max depth + v2) * dir block size - * the directory inode's bmap btree: (max depth + v2) * block size - */ -STATIC uint -xfs_calc_create_resv_modify( - struct xfs_mount *mp) -{ - return xfs_calc_buf_res(2, mp->m_sb.sb_inodesize) + - xfs_calc_buf_res(1, mp->m_sb.sb_sectsize) + - (uint)XFS_FSB_TO_B(mp, 1) + - xfs_calc_buf_res(XFS_DIROP_LOG_COUNT(mp), XFS_FSB_TO_B(mp, 1)); -} - -/* - * For create we can allocate some inodes giving: - * the agi and agf of the ag getting the new inodes: 2 * sectorsize - * the superblock for the nlink flag: sector size - * the inode blocks allocated: XFS_IALLOC_BLOCKS * blocksize - * the inode btree: max depth * blocksize - * the allocation btrees: 2 trees * (max depth - 1) * block size - */ -STATIC uint -xfs_calc_create_resv_alloc( - struct xfs_mount *mp) -{ - return xfs_calc_buf_res(2, mp->m_sb.sb_sectsize) + - mp->m_sb.sb_sectsize + - xfs_calc_buf_res(XFS_IALLOC_BLOCKS(mp), XFS_FSB_TO_B(mp, 1)) + - xfs_calc_buf_res(mp->m_in_maxlevels, XFS_FSB_TO_B(mp, 1)) + - xfs_calc_buf_res(XFS_ALLOCFREE_LOG_COUNT(mp, 1), - XFS_FSB_TO_B(mp, 1)); -} - -STATIC uint -__xfs_calc_create_reservation( - struct xfs_mount *mp) -{ - return XFS_DQUOT_LOGRES(mp) + - MAX(xfs_calc_create_resv_alloc(mp), - xfs_calc_create_resv_modify(mp)); -} - -/* - * For icreate we can allocate some inodes giving: - * the agi and agf of the ag getting the new inodes: 2 * sectorsize - * the superblock for the nlink flag: sector size - * the inode btree: max depth * blocksize - * the allocation btrees: 2 trees * (max depth - 1) * block size - */ -STATIC uint -xfs_calc_icreate_resv_alloc( - struct xfs_mount *mp) -{ - return xfs_calc_buf_res(2, mp->m_sb.sb_sectsize) + - mp->m_sb.sb_sectsize + - xfs_calc_buf_res(mp->m_in_maxlevels, XFS_FSB_TO_B(mp, 1)) + - xfs_calc_buf_res(XFS_ALLOCFREE_LOG_COUNT(mp, 1), - XFS_FSB_TO_B(mp, 1)); -} - -STATIC uint -xfs_calc_icreate_reservation(xfs_mount_t *mp) -{ - return XFS_DQUOT_LOGRES(mp) + - MAX(xfs_calc_icreate_resv_alloc(mp), - xfs_calc_create_resv_modify(mp)); -} - -STATIC uint -xfs_calc_create_reservation( - struct xfs_mount *mp) -{ - if (xfs_sb_version_hascrc(&mp->m_sb)) - return xfs_calc_icreate_reservation(mp); - return __xfs_calc_create_reservation(mp); - -} - -/* - * Making a new directory is the same as creating a new file. - */ -STATIC uint -xfs_calc_mkdir_reservation( - struct xfs_mount *mp) -{ - return xfs_calc_create_reservation(mp); -} - - -/* - * Making a new symplink is the same as creating a new file, but - * with the added blocks for remote symlink data which can be up to 1kB in - * length (MAXPATHLEN). - */ -STATIC uint -xfs_calc_symlink_reservation( - struct xfs_mount *mp) -{ - return xfs_calc_create_reservation(mp) + - xfs_calc_buf_res(1, MAXPATHLEN); -} - -/* - * In freeing an inode we can modify: - * the inode being freed: inode size - * the super block free inode counter: sector size - * the agi hash list and counters: sector size - * the inode btree entry: block size - * the on disk inode before ours in the agi hash list: inode cluster size - * the inode btree: max depth * blocksize - * the allocation btrees: 2 trees * (max depth - 1) * block size - */ -STATIC uint -xfs_calc_ifree_reservation( - struct xfs_mount *mp) -{ - return XFS_DQUOT_LOGRES(mp) + - xfs_calc_buf_res(1, mp->m_sb.sb_inodesize) + - xfs_calc_buf_res(2, mp->m_sb.sb_sectsize) + - xfs_calc_buf_res(1, XFS_FSB_TO_B(mp, 1)) + - MAX((__uint16_t)XFS_FSB_TO_B(mp, 1), - XFS_INODE_CLUSTER_SIZE(mp)) + - xfs_calc_buf_res(1, 0) + - xfs_calc_buf_res(2 + XFS_IALLOC_BLOCKS(mp) + - mp->m_in_maxlevels, 0) + - xfs_calc_buf_res(XFS_ALLOCFREE_LOG_COUNT(mp, 1), - XFS_FSB_TO_B(mp, 1)); -} - -/* - * When only changing the inode we log the inode and possibly the superblock - * We also add a bit of slop for the transaction stuff. - */ -STATIC uint -xfs_calc_ichange_reservation( - struct xfs_mount *mp) -{ - return XFS_DQUOT_LOGRES(mp) + - mp->m_sb.sb_inodesize + - mp->m_sb.sb_sectsize + - 512; - -} - -/* - * Growing the data section of the filesystem. - * superblock - * agi and agf - * allocation btrees - */ -STATIC uint -xfs_calc_growdata_reservation( - struct xfs_mount *mp) -{ - return xfs_calc_buf_res(3, mp->m_sb.sb_sectsize) + - xfs_calc_buf_res(XFS_ALLOCFREE_LOG_COUNT(mp, 1), - XFS_FSB_TO_B(mp, 1)); -} - -/* - * Growing the rt section of the filesystem. - * In the first set of transactions (ALLOC) we allocate space to the - * bitmap or summary files. - * superblock: sector size - * agf of the ag from which the extent is allocated: sector size - * bmap btree for bitmap/summary inode: max depth * blocksize - * bitmap/summary inode: inode size - * allocation btrees for 1 block alloc: 2 * (2 * maxdepth - 1) * blocksize - */ -STATIC uint -xfs_calc_growrtalloc_reservation( - struct xfs_mount *mp) -{ - return xfs_calc_buf_res(2, mp->m_sb.sb_sectsize) + - xfs_calc_buf_res(XFS_BM_MAXLEVELS(mp, XFS_DATA_FORK), - XFS_FSB_TO_B(mp, 1)) + - xfs_calc_buf_res(1, mp->m_sb.sb_inodesize) + - xfs_calc_buf_res(XFS_ALLOCFREE_LOG_COUNT(mp, 1), - XFS_FSB_TO_B(mp, 1)); -} - -/* - * Growing the rt section of the filesystem. - * In the second set of transactions (ZERO) we zero the new metadata blocks. - * one bitmap/summary block: blocksize - */ -STATIC uint -xfs_calc_growrtzero_reservation( - struct xfs_mount *mp) -{ - return xfs_calc_buf_res(1, mp->m_sb.sb_blocksize); -} - -/* - * Growing the rt section of the filesystem. - * In the third set of transactions (FREE) we update metadata without - * allocating any new blocks. - * superblock: sector size - * bitmap inode: inode size - * summary inode: inode size - * one bitmap block: blocksize - * summary blocks: new summary size - */ -STATIC uint -xfs_calc_growrtfree_reservation( - struct xfs_mount *mp) -{ - return xfs_calc_buf_res(1, mp->m_sb.sb_sectsize) + - xfs_calc_buf_res(2, mp->m_sb.sb_inodesize) + - xfs_calc_buf_res(1, mp->m_sb.sb_blocksize) + - xfs_calc_buf_res(1, mp->m_rsumsize); -} - -/* - * Logging the inode modification timestamp on a synchronous write. - * inode - */ -STATIC uint -xfs_calc_swrite_reservation( - struct xfs_mount *mp) -{ - return xfs_calc_buf_res(1, mp->m_sb.sb_inodesize); -} - -/* - * Logging the inode mode bits when writing a setuid/setgid file - * inode - */ -STATIC uint -xfs_calc_writeid_reservation(xfs_mount_t *mp) -{ - return xfs_calc_buf_res(1, mp->m_sb.sb_inodesize); -} - -/* - * Converting the inode from non-attributed to attributed. - * the inode being converted: inode size - * agf block and superblock (for block allocation) - * the new block (directory sized) - * bmap blocks for the new directory block - * allocation btrees - */ -STATIC uint -xfs_calc_addafork_reservation( - struct xfs_mount *mp) -{ - return XFS_DQUOT_LOGRES(mp) + - xfs_calc_buf_res(1, mp->m_sb.sb_inodesize) + - xfs_calc_buf_res(2, mp->m_sb.sb_sectsize) + - xfs_calc_buf_res(1, mp->m_dirblksize) + - xfs_calc_buf_res(XFS_DAENTER_BMAP1B(mp, XFS_DATA_FORK) + 1, - XFS_FSB_TO_B(mp, 1)) + - xfs_calc_buf_res(XFS_ALLOCFREE_LOG_COUNT(mp, 1), - XFS_FSB_TO_B(mp, 1)); -} - -/* - * Removing the attribute fork of a file - * the inode being truncated: inode size - * the inode's bmap btree: max depth * block size - * And the bmap_finish transaction can free the blocks and bmap blocks: - * the agf for each of the ags: 4 * sector size - * the agfl for each of the ags: 4 * sector size - * the super block to reflect the freed blocks: sector size - * worst case split in allocation btrees per extent assuming 4 extents: - * 4 exts * 2 trees * (2 * max depth - 1) * block size - */ -STATIC uint -xfs_calc_attrinval_reservation( - struct xfs_mount *mp) -{ - return MAX((xfs_calc_buf_res(1, mp->m_sb.sb_inodesize) + - xfs_calc_buf_res(XFS_BM_MAXLEVELS(mp, XFS_ATTR_FORK), - XFS_FSB_TO_B(mp, 1))), - (xfs_calc_buf_res(9, mp->m_sb.sb_sectsize) + - xfs_calc_buf_res(XFS_ALLOCFREE_LOG_COUNT(mp, 4), - XFS_FSB_TO_B(mp, 1)))); -} - -/* - * Setting an attribute at mount time. - * the inode getting the attribute - * the superblock for allocations - * the agfs extents are allocated from - * the attribute btree * max depth - * the inode allocation btree - * Since attribute transaction space is dependent on the size of the attribute, - * the calculation is done partially at mount time and partially at runtime(see - * below). - */ -STATIC uint -xfs_calc_attrsetm_reservation( - struct xfs_mount *mp) -{ - return XFS_DQUOT_LOGRES(mp) + - xfs_calc_buf_res(1, mp->m_sb.sb_inodesize) + - xfs_calc_buf_res(1, mp->m_sb.sb_sectsize) + - xfs_calc_buf_res(XFS_DA_NODE_MAXDEPTH, XFS_FSB_TO_B(mp, 1)); -} - -/* - * Setting an attribute at runtime, transaction space unit per block. - * the superblock for allocations: sector size - * the inode bmap btree could join or split: max depth * block size - * Since the runtime attribute transaction space is dependent on the total - * blocks needed for the 1st bmap, here we calculate out the space unit for - * one block so that the caller could figure out the total space according - * to the attibute extent length in blocks by: ext * XFS_ATTRSETRT_LOG_RES(mp). - */ -STATIC uint -xfs_calc_attrsetrt_reservation( - struct xfs_mount *mp) -{ - return xfs_calc_buf_res(1, mp->m_sb.sb_sectsize) + - xfs_calc_buf_res(XFS_BM_MAXLEVELS(mp, XFS_ATTR_FORK), - XFS_FSB_TO_B(mp, 1)); -} - -/* - * Removing an attribute. - * the inode: inode size - * the attribute btree could join: max depth * block size - * the inode bmap btree could join or split: max depth * block size - * And the bmap_finish transaction can free the attr blocks freed giving: - * the agf for the ag in which the blocks live: 2 * sector size - * the agfl for the ag in which the blocks live: 2 * sector size - * the superblock for the free block count: sector size - * the allocation btrees: 2 exts * 2 trees * (2 * max depth - 1) * block size - */ -STATIC uint -xfs_calc_attrrm_reservation( - struct xfs_mount *mp) -{ - return XFS_DQUOT_LOGRES(mp) + - MAX((xfs_calc_buf_res(1, mp->m_sb.sb_inodesize) + - xfs_calc_buf_res(XFS_DA_NODE_MAXDEPTH, - XFS_FSB_TO_B(mp, 1)) + - (uint)XFS_FSB_TO_B(mp, - XFS_BM_MAXLEVELS(mp, XFS_ATTR_FORK)) + - xfs_calc_buf_res(XFS_BM_MAXLEVELS(mp, XFS_DATA_FORK), 0)), - (xfs_calc_buf_res(5, mp->m_sb.sb_sectsize) + - xfs_calc_buf_res(XFS_ALLOCFREE_LOG_COUNT(mp, 2), - XFS_FSB_TO_B(mp, 1)))); -} - -/* - * Clearing a bad agino number in an agi hash bucket. - */ -STATIC uint -xfs_calc_clear_agi_bucket_reservation( - struct xfs_mount *mp) -{ - return xfs_calc_buf_res(1, mp->m_sb.sb_sectsize); -} - -/* - * Clearing the quotaflags in the superblock. - * the super block for changing quota flags: sector size - */ -STATIC uint -xfs_calc_qm_sbchange_reservation( - struct xfs_mount *mp) -{ - return xfs_calc_buf_res(1, mp->m_sb.sb_sectsize); -} - -/* - * Adjusting quota limits. - * the xfs_disk_dquot_t: sizeof(struct xfs_disk_dquot) - */ -STATIC uint -xfs_calc_qm_setqlim_reservation( - struct xfs_mount *mp) -{ - return xfs_calc_buf_res(1, sizeof(struct xfs_disk_dquot)); -} - -/* - * Allocating quota on disk if needed. - * the write transaction log space: XFS_WRITE_LOG_RES(mp) - * the unit of quota allocation: one system block size - */ -STATIC uint -xfs_calc_qm_dqalloc_reservation( - struct xfs_mount *mp) -{ - return XFS_WRITE_LOG_RES(mp) + - xfs_calc_buf_res(1, - XFS_FSB_TO_B(mp, XFS_DQUOT_CLUSTER_SIZE_FSB) - 1); -} - -/* - * Turning off quotas. - * the xfs_qoff_logitem_t: sizeof(struct xfs_qoff_logitem) * 2 - * the superblock for the quota flags: sector size - */ -STATIC uint -xfs_calc_qm_quotaoff_reservation( - struct xfs_mount *mp) -{ - return sizeof(struct xfs_qoff_logitem) * 2 + - xfs_calc_buf_res(1, mp->m_sb.sb_sectsize); -} - -/* - * End of turning off quotas. - * the xfs_qoff_logitem_t: sizeof(struct xfs_qoff_logitem) * 2 - */ -STATIC uint -xfs_calc_qm_quotaoff_end_reservation( - struct xfs_mount *mp) -{ - return sizeof(struct xfs_qoff_logitem) * 2; -} - -/* - * Syncing the incore super block changes to disk. - * the super block to reflect the changes: sector size - */ -STATIC uint -xfs_calc_sb_reservation( - struct xfs_mount *mp) -{ - return xfs_calc_buf_res(1, mp->m_sb.sb_sectsize); -} - -/* * Initialize the precomputed transaction reservation values * in the mount structure. */ @@ -679,36 +56,7 @@ void xfs_trans_init( struct xfs_mount *mp) { - struct xfs_trans_reservations *resp = &mp->m_reservations; - - resp->tr_write = xfs_calc_write_reservation(mp); - resp->tr_itruncate = xfs_calc_itruncate_reservation(mp); - resp->tr_rename = xfs_calc_rename_reservation(mp); - resp->tr_link = xfs_calc_link_reservation(mp); - resp->tr_remove = xfs_calc_remove_reservation(mp); - resp->tr_symlink = xfs_calc_symlink_reservation(mp); - resp->tr_create = xfs_calc_create_reservation(mp); - resp->tr_mkdir = xfs_calc_mkdir_reservation(mp); - resp->tr_ifree = xfs_calc_ifree_reservation(mp); - resp->tr_ichange = xfs_calc_ichange_reservation(mp); - resp->tr_growdata = xfs_calc_growdata_reservation(mp); - resp->tr_swrite = xfs_calc_swrite_reservation(mp); - resp->tr_writeid = xfs_calc_writeid_reservation(mp); - resp->tr_addafork = xfs_calc_addafork_reservation(mp); - resp->tr_attrinval = xfs_calc_attrinval_reservation(mp); - resp->tr_attrsetm = xfs_calc_attrsetm_reservation(mp); - resp->tr_attrsetrt = xfs_calc_attrsetrt_reservation(mp); - resp->tr_attrrm = xfs_calc_attrrm_reservation(mp); - resp->tr_clearagi = xfs_calc_clear_agi_bucket_reservation(mp); - resp->tr_growrtalloc = xfs_calc_growrtalloc_reservation(mp); - resp->tr_growrtzero = xfs_calc_growrtzero_reservation(mp); - resp->tr_growrtfree = xfs_calc_growrtfree_reservation(mp); - resp->tr_qm_sbchange = xfs_calc_qm_sbchange_reservation(mp); - resp->tr_qm_setqlim = xfs_calc_qm_setqlim_reservation(mp); - resp->tr_qm_dqalloc = xfs_calc_qm_dqalloc_reservation(mp); - resp->tr_qm_quotaoff = xfs_calc_qm_quotaoff_reservation(mp); - resp->tr_qm_equotaoff = xfs_calc_qm_quotaoff_end_reservation(mp); - resp->tr_sb = xfs_calc_sb_reservation(mp); + xfs_trans_resv_calc(mp, M_RES(mp)); } /* @@ -744,7 +92,7 @@ _xfs_trans_alloc( atomic_inc(&mp->m_active_trans); tp = kmem_zone_zalloc(xfs_trans_zone, memflags); - tp->t_magic = XFS_TRANS_MAGIC; + tp->t_magic = XFS_TRANS_HEADER_MAGIC; tp->t_type = type; tp->t_mountp = mp; INIT_LIST_HEAD(&tp->t_items); @@ -789,7 +137,7 @@ xfs_trans_dup( /* * Initialize the new transaction structure. */ - ntp->t_magic = XFS_TRANS_MAGIC; + ntp->t_magic = XFS_TRANS_HEADER_MAGIC; ntp->t_type = tp->t_type; ntp->t_mountp = tp->t_mountp; INIT_LIST_HEAD(&ntp->t_items); @@ -832,12 +180,10 @@ xfs_trans_dup( */ int xfs_trans_reserve( - xfs_trans_t *tp, - uint blocks, - uint logspace, - uint rtextents, - uint flags, - uint logcount) + struct xfs_trans *tp, + struct xfs_trans_res *resp, + uint blocks, + uint rtextents) { int error = 0; int rsvd = (tp->t_flags & XFS_TRANS_RESERVE) != 0; @@ -863,13 +209,15 @@ xfs_trans_reserve( /* * Reserve the log space needed for this transaction. */ - if (logspace > 0) { + if (resp->tr_logres > 0) { bool permanent = false; - ASSERT(tp->t_log_res == 0 || tp->t_log_res == logspace); - ASSERT(tp->t_log_count == 0 || tp->t_log_count == logcount); + ASSERT(tp->t_log_res == 0 || + tp->t_log_res == resp->tr_logres); + ASSERT(tp->t_log_count == 0 || + tp->t_log_count == resp->tr_logcount); - if (flags & XFS_TRANS_PERM_LOG_RES) { + if (resp->tr_logflags & XFS_TRANS_PERM_LOG_RES) { tp->t_flags |= XFS_TRANS_PERM_LOG_RES; permanent = true; } else { @@ -878,20 +226,21 @@ xfs_trans_reserve( } if (tp->t_ticket != NULL) { - ASSERT(flags & XFS_TRANS_PERM_LOG_RES); + ASSERT(resp->tr_logflags & XFS_TRANS_PERM_LOG_RES); error = xfs_log_regrant(tp->t_mountp, tp->t_ticket); } else { - error = xfs_log_reserve(tp->t_mountp, logspace, - logcount, &tp->t_ticket, - XFS_TRANSACTION, permanent, - tp->t_type); + error = xfs_log_reserve(tp->t_mountp, + resp->tr_logres, + resp->tr_logcount, + &tp->t_ticket, XFS_TRANSACTION, + permanent, tp->t_type); } if (error) goto undo_blocks; - tp->t_log_res = logspace; - tp->t_log_count = logcount; + tp->t_log_res = resp->tr_logres; + tp->t_log_count = resp->tr_logcount; } /* @@ -916,10 +265,10 @@ xfs_trans_reserve( * reservations which have already been performed. */ undo_log: - if (logspace > 0) { + if (resp->tr_logres > 0) { int log_flags; - if (flags & XFS_TRANS_PERM_LOG_RES) { + if (resp->tr_logflags & XFS_TRANS_PERM_LOG_RES) { log_flags = XFS_LOG_REL_PERM_RESERV; } else { log_flags = 0; @@ -1367,10 +716,10 @@ xfs_trans_free_items( lip->li_desc = NULL; if (commit_lsn != NULLCOMMITLSN) - IOP_COMMITTING(lip, commit_lsn); + lip->li_ops->iop_committing(lip, commit_lsn); if (flags & XFS_TRANS_ABORT) lip->li_flags |= XFS_LI_ABORTED; - IOP_UNLOCK(lip); + lip->li_ops->iop_unlock(lip); xfs_trans_free_item_desc(lidp); } @@ -1390,8 +739,11 @@ xfs_log_item_batch_insert( /* xfs_trans_ail_update_bulk drops ailp->xa_lock */ xfs_trans_ail_update_bulk(ailp, cur, log_items, nr_items, commit_lsn); - for (i = 0; i < nr_items; i++) - IOP_UNPIN(log_items[i], 0); + for (i = 0; i < nr_items; i++) { + struct xfs_log_item *lip = log_items[i]; + + lip->li_ops->iop_unpin(lip, 0); + } } /* @@ -1401,11 +753,11 @@ xfs_log_item_batch_insert( * * If we are called with the aborted flag set, it is because a log write during * a CIL checkpoint commit has failed. In this case, all the items in the - * checkpoint have already gone through IOP_COMMITED and IOP_UNLOCK, which + * checkpoint have already gone through iop_commited and iop_unlock, which * means that checkpoint commit abort handling is treated exactly the same * as an iclog write error even though we haven't started any IO yet. Hence in - * this case all we need to do is IOP_COMMITTED processing, followed by an - * IOP_UNPIN(aborted) call. + * this case all we need to do is iop_committed processing, followed by an + * iop_unpin(aborted) call. * * The AIL cursor is used to optimise the insert process. If commit_lsn is not * at the end of the AIL, the insert cursor avoids the need to walk @@ -1438,7 +790,7 @@ xfs_trans_committed_bulk( if (aborted) lip->li_flags |= XFS_LI_ABORTED; - item_lsn = IOP_COMMITTED(lip, commit_lsn); + item_lsn = lip->li_ops->iop_committed(lip, commit_lsn); /* item_lsn of -1 means the item needs no further processing */ if (XFS_LSN_CMP(item_lsn, (xfs_lsn_t)-1) == 0) @@ -1450,7 +802,7 @@ xfs_trans_committed_bulk( */ if (aborted) { ASSERT(XFS_FORCED_SHUTDOWN(ailp->xa_mount)); - IOP_UNPIN(lip, 1); + lip->li_ops->iop_unpin(lip, 1); continue; } @@ -1468,7 +820,7 @@ xfs_trans_committed_bulk( xfs_trans_ail_update(ailp, lip, item_lsn); else spin_unlock(&ailp->xa_lock); - IOP_UNPIN(lip, 0); + lip->li_ops->iop_unpin(lip, 0); continue; } @@ -1666,7 +1018,7 @@ xfs_trans_roll( struct xfs_inode *dp) { struct xfs_trans *trans; - unsigned int logres, count; + struct xfs_trans_res tres; int error; /* @@ -1678,8 +1030,8 @@ xfs_trans_roll( /* * Copy the critical parameters from one trans to the next. */ - logres = trans->t_log_res; - count = trans->t_log_count; + tres.tr_logres = trans->t_log_res; + tres.tr_logcount = trans->t_log_count; *tpp = xfs_trans_dup(trans); /* @@ -1710,8 +1062,8 @@ xfs_trans_roll( * across this call, or that anything that is locked be logged in * the prior and the next transactions. */ - error = xfs_trans_reserve(trans, 0, logres, 0, - XFS_TRANS_PERM_LOG_RES, count); + tres.tr_logflags = XFS_TRANS_PERM_LOG_RES; + error = xfs_trans_reserve(trans, &tres, 0, 0); /* * Ensure that the inode is in the new transaction and locked. */ diff --git a/fs/xfs/xfs_trans.h b/fs/xfs/xfs_trans.h index 2b49463..09cf40b 100644 --- a/fs/xfs/xfs_trans.h +++ b/fs/xfs/xfs_trans.h @@ -20,285 +20,9 @@ struct xfs_log_item; -/* - * This is the structure written in the log at the head of - * every transaction. It identifies the type and id of the - * transaction, and contains the number of items logged by - * the transaction so we know how many to expect during recovery. - * - * Do not change the below structure without redoing the code in - * xlog_recover_add_to_trans() and xlog_recover_add_to_cont_trans(). - */ -typedef struct xfs_trans_header { - uint th_magic; /* magic number */ - uint th_type; /* transaction type */ - __int32_t th_tid; /* transaction id (unused) */ - uint th_num_items; /* num items logged by trans */ -} xfs_trans_header_t; - -#define XFS_TRANS_HEADER_MAGIC 0x5452414e /* TRAN */ - -/* - * Log item types. - */ -#define XFS_LI_EFI 0x1236 -#define XFS_LI_EFD 0x1237 -#define XFS_LI_IUNLINK 0x1238 -#define XFS_LI_INODE 0x123b /* aligned ino chunks, var-size ibufs */ -#define XFS_LI_BUF 0x123c /* v2 bufs, variable sized inode bufs */ -#define XFS_LI_DQUOT 0x123d -#define XFS_LI_QUOTAOFF 0x123e -#define XFS_LI_ICREATE 0x123f - -#define XFS_LI_TYPE_DESC \ - { XFS_LI_EFI, "XFS_LI_EFI" }, \ - { XFS_LI_EFD, "XFS_LI_EFD" }, \ - { XFS_LI_IUNLINK, "XFS_LI_IUNLINK" }, \ - { XFS_LI_INODE, "XFS_LI_INODE" }, \ - { XFS_LI_BUF, "XFS_LI_BUF" }, \ - { XFS_LI_DQUOT, "XFS_LI_DQUOT" }, \ - { XFS_LI_QUOTAOFF, "XFS_LI_QUOTAOFF" } - -/* - * Transaction types. Used to distinguish types of buffers. - */ -#define XFS_TRANS_SETATTR_NOT_SIZE 1 -#define XFS_TRANS_SETATTR_SIZE 2 -#define XFS_TRANS_INACTIVE 3 -#define XFS_TRANS_CREATE 4 -#define XFS_TRANS_CREATE_TRUNC 5 -#define XFS_TRANS_TRUNCATE_FILE 6 -#define XFS_TRANS_REMOVE 7 -#define XFS_TRANS_LINK 8 -#define XFS_TRANS_RENAME 9 -#define XFS_TRANS_MKDIR 10 -#define XFS_TRANS_RMDIR 11 -#define XFS_TRANS_SYMLINK 12 -#define XFS_TRANS_SET_DMATTRS 13 -#define XFS_TRANS_GROWFS 14 -#define XFS_TRANS_STRAT_WRITE 15 -#define XFS_TRANS_DIOSTRAT 16 -/* 17 was XFS_TRANS_WRITE_SYNC */ -#define XFS_TRANS_WRITEID 18 -#define XFS_TRANS_ADDAFORK 19 -#define XFS_TRANS_ATTRINVAL 20 -#define XFS_TRANS_ATRUNCATE 21 -#define XFS_TRANS_ATTR_SET 22 -#define XFS_TRANS_ATTR_RM 23 -#define XFS_TRANS_ATTR_FLAG 24 -#define XFS_TRANS_CLEAR_AGI_BUCKET 25 -#define XFS_TRANS_QM_SBCHANGE 26 -/* - * Dummy entries since we use the transaction type to index into the - * trans_type[] in xlog_recover_print_trans_head() - */ -#define XFS_TRANS_DUMMY1 27 -#define XFS_TRANS_DUMMY2 28 -#define XFS_TRANS_QM_QUOTAOFF 29 -#define XFS_TRANS_QM_DQALLOC 30 -#define XFS_TRANS_QM_SETQLIM 31 -#define XFS_TRANS_QM_DQCLUSTER 32 -#define XFS_TRANS_QM_QINOCREATE 33 -#define XFS_TRANS_QM_QUOTAOFF_END 34 -#define XFS_TRANS_SB_UNIT 35 -#define XFS_TRANS_FSYNC_TS 36 -#define XFS_TRANS_GROWFSRT_ALLOC 37 -#define XFS_TRANS_GROWFSRT_ZERO 38 -#define XFS_TRANS_GROWFSRT_FREE 39 -#define XFS_TRANS_SWAPEXT 40 -#define XFS_TRANS_SB_COUNT 41 -#define XFS_TRANS_CHECKPOINT 42 -#define XFS_TRANS_ICREATE 43 -#define XFS_TRANS_TYPE_MAX 43 -/* new transaction types need to be reflected in xfs_logprint(8) */ - -#define XFS_TRANS_TYPES \ - { XFS_TRANS_SETATTR_NOT_SIZE, "SETATTR_NOT_SIZE" }, \ - { XFS_TRANS_SETATTR_SIZE, "SETATTR_SIZE" }, \ - { XFS_TRANS_INACTIVE, "INACTIVE" }, \ - { XFS_TRANS_CREATE, "CREATE" }, \ - { XFS_TRANS_CREATE_TRUNC, "CREATE_TRUNC" }, \ - { XFS_TRANS_TRUNCATE_FILE, "TRUNCATE_FILE" }, \ - { XFS_TRANS_REMOVE, "REMOVE" }, \ - { XFS_TRANS_LINK, "LINK" }, \ - { XFS_TRANS_RENAME, "RENAME" }, \ - { XFS_TRANS_MKDIR, "MKDIR" }, \ - { XFS_TRANS_RMDIR, "RMDIR" }, \ - { XFS_TRANS_SYMLINK, "SYMLINK" }, \ - { XFS_TRANS_SET_DMATTRS, "SET_DMATTRS" }, \ - { XFS_TRANS_GROWFS, "GROWFS" }, \ - { XFS_TRANS_STRAT_WRITE, "STRAT_WRITE" }, \ - { XFS_TRANS_DIOSTRAT, "DIOSTRAT" }, \ - { XFS_TRANS_WRITEID, "WRITEID" }, \ - { XFS_TRANS_ADDAFORK, "ADDAFORK" }, \ - { XFS_TRANS_ATTRINVAL, "ATTRINVAL" }, \ - { XFS_TRANS_ATRUNCATE, "ATRUNCATE" }, \ - { XFS_TRANS_ATTR_SET, "ATTR_SET" }, \ - { XFS_TRANS_ATTR_RM, "ATTR_RM" }, \ - { XFS_TRANS_ATTR_FLAG, "ATTR_FLAG" }, \ - { XFS_TRANS_CLEAR_AGI_BUCKET, "CLEAR_AGI_BUCKET" }, \ - { XFS_TRANS_QM_SBCHANGE, "QM_SBCHANGE" }, \ - { XFS_TRANS_QM_QUOTAOFF, "QM_QUOTAOFF" }, \ - { XFS_TRANS_QM_DQALLOC, "QM_DQALLOC" }, \ - { XFS_TRANS_QM_SETQLIM, "QM_SETQLIM" }, \ - { XFS_TRANS_QM_DQCLUSTER, "QM_DQCLUSTER" }, \ - { XFS_TRANS_QM_QINOCREATE, "QM_QINOCREATE" }, \ - { XFS_TRANS_QM_QUOTAOFF_END, "QM_QOFF_END" }, \ - { XFS_TRANS_SB_UNIT, "SB_UNIT" }, \ - { XFS_TRANS_FSYNC_TS, "FSYNC_TS" }, \ - { XFS_TRANS_GROWFSRT_ALLOC, "GROWFSRT_ALLOC" }, \ - { XFS_TRANS_GROWFSRT_ZERO, "GROWFSRT_ZERO" }, \ - { XFS_TRANS_GROWFSRT_FREE, "GROWFSRT_FREE" }, \ - { XFS_TRANS_SWAPEXT, "SWAPEXT" }, \ - { XFS_TRANS_SB_COUNT, "SB_COUNT" }, \ - { XFS_TRANS_CHECKPOINT, "CHECKPOINT" }, \ - { XFS_TRANS_DUMMY1, "DUMMY1" }, \ - { XFS_TRANS_DUMMY2, "DUMMY2" }, \ - { XLOG_UNMOUNT_REC_TYPE, "UNMOUNT" } - -/* - * This structure is used to track log items associated with - * a transaction. It points to the log item and keeps some - * flags to track the state of the log item. It also tracks - * the amount of space needed to log the item it describes - * once we get to commit processing (see xfs_trans_commit()). - */ -struct xfs_log_item_desc { - struct xfs_log_item *lid_item; - struct list_head lid_trans; - unsigned char lid_flags; -}; - -#define XFS_LID_DIRTY 0x1 - -#define XFS_TRANS_MAGIC 0x5452414E /* 'TRAN' */ -/* - * Values for t_flags. - */ -#define XFS_TRANS_DIRTY 0x01 /* something needs to be logged */ -#define XFS_TRANS_SB_DIRTY 0x02 /* superblock is modified */ -#define XFS_TRANS_PERM_LOG_RES 0x04 /* xact took a permanent log res */ -#define XFS_TRANS_SYNC 0x08 /* make commit synchronous */ -#define XFS_TRANS_DQ_DIRTY 0x10 /* at least one dquot in trx dirty */ -#define XFS_TRANS_RESERVE 0x20 /* OK to use reserved data blocks */ -#define XFS_TRANS_FREEZE_PROT 0x40 /* Transaction has elevated writer - count in superblock */ - -/* - * Values for call flags parameter. - */ -#define XFS_TRANS_RELEASE_LOG_RES 0x4 -#define XFS_TRANS_ABORT 0x8 - -/* - * Field values for xfs_trans_mod_sb. - */ -#define XFS_TRANS_SB_ICOUNT 0x00000001 -#define XFS_TRANS_SB_IFREE 0x00000002 -#define XFS_TRANS_SB_FDBLOCKS 0x00000004 -#define XFS_TRANS_SB_RES_FDBLOCKS 0x00000008 -#define XFS_TRANS_SB_FREXTENTS 0x00000010 -#define XFS_TRANS_SB_RES_FREXTENTS 0x00000020 -#define XFS_TRANS_SB_DBLOCKS 0x00000040 -#define XFS_TRANS_SB_AGCOUNT 0x00000080 -#define XFS_TRANS_SB_IMAXPCT 0x00000100 -#define XFS_TRANS_SB_REXTSIZE 0x00000200 -#define XFS_TRANS_SB_RBMBLOCKS 0x00000400 -#define XFS_TRANS_SB_RBLOCKS 0x00000800 -#define XFS_TRANS_SB_REXTENTS 0x00001000 -#define XFS_TRANS_SB_REXTSLOG 0x00002000 - - -/* - * Per-extent log reservation for the allocation btree changes - * involved in freeing or allocating an extent. - * 2 trees * (2 blocks/level * max depth - 1) - */ -#define XFS_ALLOCFREE_LOG_COUNT(mp,nx) \ - ((nx) * (2 * (2 * XFS_AG_MAXLEVELS(mp) - 1))) - -/* - * Per-directory log reservation for any directory change. - * dir blocks: (1 btree block per level + data block + free block) - * bmap btree: (levels + 2) * max depth - * v2 directory blocks can be fragmented below the dirblksize down to the fsb - * size, so account for that in the DAENTER macros. - */ -#define XFS_DIROP_LOG_COUNT(mp) \ - (XFS_DAENTER_BLOCKS(mp, XFS_DATA_FORK) + \ - XFS_DAENTER_BMAPS(mp, XFS_DATA_FORK) + 1) - +#include "xfs_trans_resv.h" -#define XFS_WRITE_LOG_RES(mp) ((mp)->m_reservations.tr_write) -#define XFS_ITRUNCATE_LOG_RES(mp) ((mp)->m_reservations.tr_itruncate) -#define XFS_RENAME_LOG_RES(mp) ((mp)->m_reservations.tr_rename) -#define XFS_LINK_LOG_RES(mp) ((mp)->m_reservations.tr_link) -#define XFS_REMOVE_LOG_RES(mp) ((mp)->m_reservations.tr_remove) -#define XFS_SYMLINK_LOG_RES(mp) ((mp)->m_reservations.tr_symlink) -#define XFS_CREATE_LOG_RES(mp) ((mp)->m_reservations.tr_create) -#define XFS_MKDIR_LOG_RES(mp) ((mp)->m_reservations.tr_mkdir) -#define XFS_IFREE_LOG_RES(mp) ((mp)->m_reservations.tr_ifree) -#define XFS_ICHANGE_LOG_RES(mp) ((mp)->m_reservations.tr_ichange) -#define XFS_GROWDATA_LOG_RES(mp) ((mp)->m_reservations.tr_growdata) -#define XFS_GROWRTALLOC_LOG_RES(mp) ((mp)->m_reservations.tr_growrtalloc) -#define XFS_GROWRTZERO_LOG_RES(mp) ((mp)->m_reservations.tr_growrtzero) -#define XFS_GROWRTFREE_LOG_RES(mp) ((mp)->m_reservations.tr_growrtfree) -#define XFS_SWRITE_LOG_RES(mp) ((mp)->m_reservations.tr_swrite) -/* - * Logging the inode timestamps on an fsync -- same as SWRITE - * as long as SWRITE logs the entire inode core - */ -#define XFS_FSYNC_TS_LOG_RES(mp) ((mp)->m_reservations.tr_swrite) -#define XFS_WRITEID_LOG_RES(mp) ((mp)->m_reservations.tr_swrite) -#define XFS_ADDAFORK_LOG_RES(mp) ((mp)->m_reservations.tr_addafork) -#define XFS_ATTRINVAL_LOG_RES(mp) ((mp)->m_reservations.tr_attrinval) -#define XFS_ATTRSETM_LOG_RES(mp) ((mp)->m_reservations.tr_attrsetm) -#define XFS_ATTRSETRT_LOG_RES(mp) ((mp)->m_reservations.tr_attrsetrt) -#define XFS_ATTRRM_LOG_RES(mp) ((mp)->m_reservations.tr_attrrm) -#define XFS_CLEAR_AGI_BUCKET_LOG_RES(mp) ((mp)->m_reservations.tr_clearagi) -#define XFS_QM_SBCHANGE_LOG_RES(mp) ((mp)->m_reservations.tr_qm_sbchange) -#define XFS_QM_SETQLIM_LOG_RES(mp) ((mp)->m_reservations.tr_qm_setqlim) -#define XFS_QM_DQALLOC_LOG_RES(mp) ((mp)->m_reservations.tr_qm_dqalloc) -#define XFS_QM_QUOTAOFF_LOG_RES(mp) ((mp)->m_reservations.tr_qm_quotaoff) -#define XFS_QM_QUOTAOFF_END_LOG_RES(mp) ((mp)->m_reservations.tr_qm_equotaoff) -#define XFS_SB_LOG_RES(mp) ((mp)->m_reservations.tr_sb) - -/* - * Various log count values. - */ -#define XFS_DEFAULT_LOG_COUNT 1 -#define XFS_DEFAULT_PERM_LOG_COUNT 2 -#define XFS_ITRUNCATE_LOG_COUNT 2 -#define XFS_INACTIVE_LOG_COUNT 2 -#define XFS_CREATE_LOG_COUNT 2 -#define XFS_MKDIR_LOG_COUNT 3 -#define XFS_SYMLINK_LOG_COUNT 3 -#define XFS_REMOVE_LOG_COUNT 2 -#define XFS_LINK_LOG_COUNT 2 -#define XFS_RENAME_LOG_COUNT 2 -#define XFS_WRITE_LOG_COUNT 2 -#define XFS_ADDAFORK_LOG_COUNT 2 -#define XFS_ATTRINVAL_LOG_COUNT 1 -#define XFS_ATTRSET_LOG_COUNT 3 -#define XFS_ATTRRM_LOG_COUNT 3 - -/* - * Here we centralize the specification of XFS meta-data buffer - * reference count values. This determine how hard the buffer - * cache tries to hold onto the buffer. - */ -#define XFS_AGF_REF 4 -#define XFS_AGI_REF 4 -#define XFS_AGFL_REF 3 -#define XFS_INO_BTREE_REF 3 -#define XFS_ALLOC_BTREE_REF 2 -#define XFS_BMAP_BTREE_REF 2 -#define XFS_DIR_BTREE_REF 2 -#define XFS_INO_REF 2 -#define XFS_ATTR_BTREE_REF 1 -#define XFS_DQUOT_REF 1 - -#ifdef __KERNEL__ +/* kernel only transaction subsystem defines */ struct xfs_buf; struct xfs_buftarg; @@ -310,6 +34,7 @@ struct xfs_log_iovec; struct xfs_log_item_desc; struct xfs_mount; struct xfs_trans; +struct xfs_trans_res; struct xfs_dquot_acct; struct xfs_busy_extent; @@ -342,7 +67,7 @@ typedef struct xfs_log_item { { XFS_LI_ABORTED, "ABORTED" } struct xfs_item_ops { - uint (*iop_size)(xfs_log_item_t *); + void (*iop_size)(xfs_log_item_t *, int *, int *); void (*iop_format)(xfs_log_item_t *, struct xfs_log_iovec *); void (*iop_pin)(xfs_log_item_t *); void (*iop_unpin)(xfs_log_item_t *, int remove); @@ -352,17 +77,8 @@ struct xfs_item_ops { void (*iop_committing)(xfs_log_item_t *, xfs_lsn_t); }; -#define IOP_SIZE(ip) (*(ip)->li_ops->iop_size)(ip) -#define IOP_FORMAT(ip,vp) (*(ip)->li_ops->iop_format)(ip, vp) -#define IOP_PIN(ip) (*(ip)->li_ops->iop_pin)(ip) -#define IOP_UNPIN(ip, remove) (*(ip)->li_ops->iop_unpin)(ip, remove) -#define IOP_PUSH(ip, list) (*(ip)->li_ops->iop_push)(ip, list) -#define IOP_UNLOCK(ip) (*(ip)->li_ops->iop_unlock)(ip) -#define IOP_COMMITTED(ip, lsn) (*(ip)->li_ops->iop_committed)(ip, lsn) -#define IOP_COMMITTING(ip, lsn) (*(ip)->li_ops->iop_committing)(ip, lsn) - /* - * Return values for the IOP_PUSH() routines. + * Return values for the iop_push() routines. */ #define XFS_ITEM_SUCCESS 0 #define XFS_ITEM_PINNED 1 @@ -446,7 +162,7 @@ typedef struct xfs_trans { xfs_trans_t *xfs_trans_alloc(struct xfs_mount *, uint); xfs_trans_t *_xfs_trans_alloc(struct xfs_mount *, uint, xfs_km_flags_t); xfs_trans_t *xfs_trans_dup(xfs_trans_t *); -int xfs_trans_reserve(xfs_trans_t *, uint, uint, uint, +int xfs_trans_reserve(struct xfs_trans *, struct xfs_trans_res *, uint, uint); void xfs_trans_mod_sb(xfs_trans_t *, uint, int64_t); @@ -528,9 +244,4 @@ void xfs_trans_ail_destroy(struct xfs_mount *); extern kmem_zone_t *xfs_trans_zone; extern kmem_zone_t *xfs_log_item_desc_zone; -#endif /* __KERNEL__ */ - -void xfs_trans_init(struct xfs_mount *); -int xfs_trans_roll(struct xfs_trans **, struct xfs_inode *); - #endif /* __XFS_TRANS_H__ */ diff --git a/fs/xfs/xfs_trans_ail.c b/fs/xfs/xfs_trans_ail.c index 0eda725..21c6d7d 100644 --- a/fs/xfs/xfs_trans_ail.c +++ b/fs/xfs/xfs_trans_ail.c @@ -61,20 +61,6 @@ xfs_ail_check( #endif /* DEBUG */ /* - * Return a pointer to the first item in the AIL. If the AIL is empty, then - * return NULL. - */ -xfs_log_item_t * -xfs_ail_min( - struct xfs_ail *ailp) -{ - if (list_empty(&ailp->xa_ail)) - return NULL; - - return list_first_entry(&ailp->xa_ail, xfs_log_item_t, li_ail); -} - - /* * Return a pointer to the last item in the AIL. If the AIL is empty, then * return NULL. */ @@ -393,11 +379,11 @@ xfsaild_push( int lock_result; /* - * Note that IOP_PUSH may unlock and reacquire the AIL lock. We + * Note that iop_push may unlock and reacquire the AIL lock. We * rely on the AIL cursor implementation to be able to deal with * the dropped lock. */ - lock_result = IOP_PUSH(lip, &ailp->xa_buf_list); + lock_result = lip->li_ops->iop_push(lip, &ailp->xa_buf_list); switch (lock_result) { case XFS_ITEM_SUCCESS: XFS_STATS_INC(xs_push_ail_success); diff --git a/fs/xfs/xfs_trans_buf.c b/fs/xfs/xfs_trans_buf.c index aa5a04b..8c75b8f 100644 --- a/fs/xfs/xfs_trans_buf.c +++ b/fs/xfs/xfs_trans_buf.c @@ -505,7 +505,7 @@ xfs_trans_brelse(xfs_trans_t *tp, /* * Mark the buffer as not needing to be unlocked when the buf item's - * IOP_UNLOCK() routine is called. The buffer must already be locked + * iop_unlock() routine is called. The buffer must already be locked * and associated with the given transaction. */ /* ARGSUSED */ diff --git a/fs/xfs/xfs_trans_dquot.c b/fs/xfs/xfs_trans_dquot.c index 61407a8..54ee3c5 100644 --- a/fs/xfs/xfs_trans_dquot.c +++ b/fs/xfs/xfs_trans_dquot.c @@ -17,6 +17,7 @@ */ #include "xfs.h" #include "xfs_fs.h" +#include "xfs_format.h" #include "xfs_log.h" #include "xfs_trans.h" #include "xfs_sb.h" diff --git a/fs/xfs/xfs_trans_priv.h b/fs/xfs/xfs_trans_priv.h index 53b7c9b..c52def0 100644 --- a/fs/xfs/xfs_trans_priv.h +++ b/fs/xfs/xfs_trans_priv.h @@ -25,6 +25,9 @@ struct xfs_trans; struct xfs_ail; struct xfs_log_vec; + +void xfs_trans_init(struct xfs_mount *); +int xfs_trans_roll(struct xfs_trans **, struct xfs_inode *); void xfs_trans_add_item(struct xfs_trans *, struct xfs_log_item *); void xfs_trans_del_item(struct xfs_log_item *); void xfs_trans_free_items(struct xfs_trans *tp, xfs_lsn_t commit_lsn, @@ -83,6 +86,18 @@ void xfs_trans_ail_update_bulk(struct xfs_ail *ailp, struct xfs_ail_cursor *cur, struct xfs_log_item **log_items, int nr_items, xfs_lsn_t lsn) __releases(ailp->xa_lock); +/* + * Return a pointer to the first item in the AIL. If the AIL is empty, then + * return NULL. + */ +static inline struct xfs_log_item * +xfs_ail_min( + struct xfs_ail *ailp) +{ + return list_first_entry_or_null(&ailp->xa_ail, struct xfs_log_item, + li_ail); +} + static inline void xfs_trans_ail_update( struct xfs_ail *ailp, diff --git a/fs/xfs/xfs_trans_resv.c b/fs/xfs/xfs_trans_resv.c new file mode 100644 index 0000000..a65a3cc4 --- /dev/null +++ b/fs/xfs/xfs_trans_resv.c @@ -0,0 +1,803 @@ +/* + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * Copyright (C) 2010 Red Hat, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#include "xfs.h" +#include "xfs_fs.h" +#include "xfs_format.h" +#include "xfs_log.h" +#include "xfs_trans_resv.h" +#include "xfs_trans.h" +#include "xfs_sb.h" +#include "xfs_ag.h" +#include "xfs_mount.h" +#include "xfs_error.h" +#include "xfs_da_btree.h" +#include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" +#include "xfs_ialloc_btree.h" +#include "xfs_dinode.h" +#include "xfs_inode.h" +#include "xfs_btree.h" +#include "xfs_ialloc.h" +#include "xfs_alloc.h" +#include "xfs_extent_busy.h" +#include "xfs_bmap.h" +#include "xfs_bmap_util.h" +#include "xfs_quota.h" +#include "xfs_qm.h" +#include "xfs_trans_space.h" +#include "xfs_trace.h" + +/* + * A buffer has a format structure overhead in the log in addition + * to the data, so we need to take this into account when reserving + * space in a transaction for a buffer. Round the space required up + * to a multiple of 128 bytes so that we don't change the historical + * reservation that has been used for this overhead. + */ +STATIC uint +xfs_buf_log_overhead(void) +{ + return round_up(sizeof(struct xlog_op_header) + + sizeof(struct xfs_buf_log_format), 128); +} + +/* + * Calculate out transaction log reservation per item in bytes. + * + * The nbufs argument is used to indicate the number of items that + * will be changed in a transaction. size is used to tell how many + * bytes should be reserved per item. + */ +STATIC uint +xfs_calc_buf_res( + uint nbufs, + uint size) +{ + return nbufs * (size + xfs_buf_log_overhead()); +} + +/* + * Logging inodes is really tricksy. They are logged in memory format, + * which means that what we write into the log doesn't directly translate into + * the amount of space they use on disk. + * + * Case in point - btree format forks in memory format use more space than the + * on-disk format. In memory, the buffer contains a normal btree block header so + * the btree code can treat it as though it is just another generic buffer. + * However, when we write it to the inode fork, we don't write all of this + * header as it isn't needed. e.g. the root is only ever in the inode, so + * there's no need for sibling pointers which would waste 16 bytes of space. + * + * Hence when we have an inode with a maximally sized btree format fork, then + * amount of information we actually log is greater than the size of the inode + * on disk. Hence we need an inode reservation function that calculates all this + * correctly. So, we log: + * + * - log op headers for object + * - inode log format object + * - the entire inode contents (core + 2 forks) + * - two bmap btree block headers + */ +STATIC uint +xfs_calc_inode_res( + struct xfs_mount *mp, + uint ninodes) +{ + return ninodes * (sizeof(struct xlog_op_header) + + sizeof(struct xfs_inode_log_format) + + mp->m_sb.sb_inodesize + + 2 * XFS_BMBT_BLOCK_LEN(mp)); +} + +/* + * Various log reservation values. + * + * These are based on the size of the file system block because that is what + * most transactions manipulate. Each adds in an additional 128 bytes per + * item logged to try to account for the overhead of the transaction mechanism. + * + * Note: Most of the reservations underestimate the number of allocation + * groups into which they could free extents in the xfs_bmap_finish() call. + * This is because the number in the worst case is quite high and quite + * unusual. In order to fix this we need to change xfs_bmap_finish() to free + * extents in only a single AG at a time. This will require changes to the + * EFI code as well, however, so that the EFI for the extents not freed is + * logged again in each transaction. See SGI PV #261917. + * + * Reservation functions here avoid a huge stack in xfs_trans_init due to + * register overflow from temporaries in the calculations. + */ + + +/* + * In a write transaction we can allocate a maximum of 2 + * extents. This gives: + * the inode getting the new extents: inode size + * the inode's bmap btree: max depth * block size + * the agfs of the ags from which the extents are allocated: 2 * sector + * the superblock free block counter: sector size + * the allocation btrees: 2 exts * 2 trees * (2 * max depth - 1) * block size + * And the bmap_finish transaction can free bmap blocks in a join: + * the agfs of the ags containing the blocks: 2 * sector size + * the agfls of the ags containing the blocks: 2 * sector size + * the super block free block counter: sector size + * the allocation btrees: 2 exts * 2 trees * (2 * max depth - 1) * block size + */ +STATIC uint +xfs_calc_write_reservation( + struct xfs_mount *mp) +{ + return XFS_DQUOT_LOGRES(mp) + + MAX((xfs_calc_inode_res(mp, 1) + + xfs_calc_buf_res(XFS_BM_MAXLEVELS(mp, XFS_DATA_FORK), + XFS_FSB_TO_B(mp, 1)) + + xfs_calc_buf_res(3, mp->m_sb.sb_sectsize) + + xfs_calc_buf_res(XFS_ALLOCFREE_LOG_COUNT(mp, 2), + XFS_FSB_TO_B(mp, 1))), + (xfs_calc_buf_res(5, mp->m_sb.sb_sectsize) + + xfs_calc_buf_res(XFS_ALLOCFREE_LOG_COUNT(mp, 2), + XFS_FSB_TO_B(mp, 1)))); +} + +/* + * In truncating a file we free up to two extents at once. We can modify: + * the inode being truncated: inode size + * the inode's bmap btree: (max depth + 1) * block size + * And the bmap_finish transaction can free the blocks and bmap blocks: + * the agf for each of the ags: 4 * sector size + * the agfl for each of the ags: 4 * sector size + * the super block to reflect the freed blocks: sector size + * worst case split in allocation btrees per extent assuming 4 extents: + * 4 exts * 2 trees * (2 * max depth - 1) * block size + * the inode btree: max depth * blocksize + * the allocation btrees: 2 trees * (max depth - 1) * block size + */ +STATIC uint +xfs_calc_itruncate_reservation( + struct xfs_mount *mp) +{ + return XFS_DQUOT_LOGRES(mp) + + MAX((xfs_calc_inode_res(mp, 1) + + xfs_calc_buf_res(XFS_BM_MAXLEVELS(mp, XFS_DATA_FORK) + 1, + XFS_FSB_TO_B(mp, 1))), + (xfs_calc_buf_res(9, mp->m_sb.sb_sectsize) + + xfs_calc_buf_res(XFS_ALLOCFREE_LOG_COUNT(mp, 4), + XFS_FSB_TO_B(mp, 1)) + + xfs_calc_buf_res(5, 0) + + xfs_calc_buf_res(XFS_ALLOCFREE_LOG_COUNT(mp, 1), + XFS_FSB_TO_B(mp, 1)) + + xfs_calc_buf_res(2 + XFS_IALLOC_BLOCKS(mp) + + mp->m_in_maxlevels, 0))); +} + +/* + * In renaming a files we can modify: + * the four inodes involved: 4 * inode size + * the two directory btrees: 2 * (max depth + v2) * dir block size + * the two directory bmap btrees: 2 * max depth * block size + * And the bmap_finish transaction can free dir and bmap blocks (two sets + * of bmap blocks) giving: + * the agf for the ags in which the blocks live: 3 * sector size + * the agfl for the ags in which the blocks live: 3 * sector size + * the superblock for the free block count: sector size + * the allocation btrees: 3 exts * 2 trees * (2 * max depth - 1) * block size + */ +STATIC uint +xfs_calc_rename_reservation( + struct xfs_mount *mp) +{ + return XFS_DQUOT_LOGRES(mp) + + MAX((xfs_calc_inode_res(mp, 4) + + xfs_calc_buf_res(2 * XFS_DIROP_LOG_COUNT(mp), + XFS_FSB_TO_B(mp, 1))), + (xfs_calc_buf_res(7, mp->m_sb.sb_sectsize) + + xfs_calc_buf_res(XFS_ALLOCFREE_LOG_COUNT(mp, 3), + XFS_FSB_TO_B(mp, 1)))); +} + +/* + * For creating a link to an inode: + * the parent directory inode: inode size + * the linked inode: inode size + * the directory btree could split: (max depth + v2) * dir block size + * the directory bmap btree could join or split: (max depth + v2) * blocksize + * And the bmap_finish transaction can free some bmap blocks giving: + * the agf for the ag in which the blocks live: sector size + * the agfl for the ag in which the blocks live: sector size + * the superblock for the free block count: sector size + * the allocation btrees: 2 trees * (2 * max depth - 1) * block size + */ +STATIC uint +xfs_calc_link_reservation( + struct xfs_mount *mp) +{ + return XFS_DQUOT_LOGRES(mp) + + MAX((xfs_calc_inode_res(mp, 2) + + xfs_calc_buf_res(XFS_DIROP_LOG_COUNT(mp), + XFS_FSB_TO_B(mp, 1))), + (xfs_calc_buf_res(3, mp->m_sb.sb_sectsize) + + xfs_calc_buf_res(XFS_ALLOCFREE_LOG_COUNT(mp, 1), + XFS_FSB_TO_B(mp, 1)))); +} + +/* + * For removing a directory entry we can modify: + * the parent directory inode: inode size + * the removed inode: inode size + * the directory btree could join: (max depth + v2) * dir block size + * the directory bmap btree could join or split: (max depth + v2) * blocksize + * And the bmap_finish transaction can free the dir and bmap blocks giving: + * the agf for the ag in which the blocks live: 2 * sector size + * the agfl for the ag in which the blocks live: 2 * sector size + * the superblock for the free block count: sector size + * the allocation btrees: 2 exts * 2 trees * (2 * max depth - 1) * block size + */ +STATIC uint +xfs_calc_remove_reservation( + struct xfs_mount *mp) +{ + return XFS_DQUOT_LOGRES(mp) + + MAX((xfs_calc_inode_res(mp, 2) + + xfs_calc_buf_res(XFS_DIROP_LOG_COUNT(mp), + XFS_FSB_TO_B(mp, 1))), + (xfs_calc_buf_res(5, mp->m_sb.sb_sectsize) + + xfs_calc_buf_res(XFS_ALLOCFREE_LOG_COUNT(mp, 2), + XFS_FSB_TO_B(mp, 1)))); +} + +/* + * For create, break it in to the two cases that the transaction + * covers. We start with the modify case - allocation done by modification + * of the state of existing inodes - and the allocation case. + */ + +/* + * For create we can modify: + * the parent directory inode: inode size + * the new inode: inode size + * the inode btree entry: block size + * the superblock for the nlink flag: sector size + * the directory btree: (max depth + v2) * dir block size + * the directory inode's bmap btree: (max depth + v2) * block size + */ +STATIC uint +xfs_calc_create_resv_modify( + struct xfs_mount *mp) +{ + return xfs_calc_inode_res(mp, 2) + + xfs_calc_buf_res(1, mp->m_sb.sb_sectsize) + + (uint)XFS_FSB_TO_B(mp, 1) + + xfs_calc_buf_res(XFS_DIROP_LOG_COUNT(mp), XFS_FSB_TO_B(mp, 1)); +} + +/* + * For create we can allocate some inodes giving: + * the agi and agf of the ag getting the new inodes: 2 * sectorsize + * the superblock for the nlink flag: sector size + * the inode blocks allocated: XFS_IALLOC_BLOCKS * blocksize + * the inode btree: max depth * blocksize + * the allocation btrees: 2 trees * (max depth - 1) * block size + */ +STATIC uint +xfs_calc_create_resv_alloc( + struct xfs_mount *mp) +{ + return xfs_calc_buf_res(2, mp->m_sb.sb_sectsize) + + mp->m_sb.sb_sectsize + + xfs_calc_buf_res(XFS_IALLOC_BLOCKS(mp), XFS_FSB_TO_B(mp, 1)) + + xfs_calc_buf_res(mp->m_in_maxlevels, XFS_FSB_TO_B(mp, 1)) + + xfs_calc_buf_res(XFS_ALLOCFREE_LOG_COUNT(mp, 1), + XFS_FSB_TO_B(mp, 1)); +} + +STATIC uint +__xfs_calc_create_reservation( + struct xfs_mount *mp) +{ + return XFS_DQUOT_LOGRES(mp) + + MAX(xfs_calc_create_resv_alloc(mp), + xfs_calc_create_resv_modify(mp)); +} + +/* + * For icreate we can allocate some inodes giving: + * the agi and agf of the ag getting the new inodes: 2 * sectorsize + * the superblock for the nlink flag: sector size + * the inode btree: max depth * blocksize + * the allocation btrees: 2 trees * (max depth - 1) * block size + */ +STATIC uint +xfs_calc_icreate_resv_alloc( + struct xfs_mount *mp) +{ + return xfs_calc_buf_res(2, mp->m_sb.sb_sectsize) + + mp->m_sb.sb_sectsize + + xfs_calc_buf_res(mp->m_in_maxlevels, XFS_FSB_TO_B(mp, 1)) + + xfs_calc_buf_res(XFS_ALLOCFREE_LOG_COUNT(mp, 1), + XFS_FSB_TO_B(mp, 1)); +} + +STATIC uint +xfs_calc_icreate_reservation(xfs_mount_t *mp) +{ + return XFS_DQUOT_LOGRES(mp) + + MAX(xfs_calc_icreate_resv_alloc(mp), + xfs_calc_create_resv_modify(mp)); +} + +STATIC uint +xfs_calc_create_reservation( + struct xfs_mount *mp) +{ + if (xfs_sb_version_hascrc(&mp->m_sb)) + return xfs_calc_icreate_reservation(mp); + return __xfs_calc_create_reservation(mp); + +} + +/* + * Making a new directory is the same as creating a new file. + */ +STATIC uint +xfs_calc_mkdir_reservation( + struct xfs_mount *mp) +{ + return xfs_calc_create_reservation(mp); +} + + +/* + * Making a new symplink is the same as creating a new file, but + * with the added blocks for remote symlink data which can be up to 1kB in + * length (MAXPATHLEN). + */ +STATIC uint +xfs_calc_symlink_reservation( + struct xfs_mount *mp) +{ + return xfs_calc_create_reservation(mp) + + xfs_calc_buf_res(1, MAXPATHLEN); +} + +/* + * In freeing an inode we can modify: + * the inode being freed: inode size + * the super block free inode counter: sector size + * the agi hash list and counters: sector size + * the inode btree entry: block size + * the on disk inode before ours in the agi hash list: inode cluster size + * the inode btree: max depth * blocksize + * the allocation btrees: 2 trees * (max depth - 1) * block size + */ +STATIC uint +xfs_calc_ifree_reservation( + struct xfs_mount *mp) +{ + return XFS_DQUOT_LOGRES(mp) + + xfs_calc_inode_res(mp, 1) + + xfs_calc_buf_res(2, mp->m_sb.sb_sectsize) + + xfs_calc_buf_res(1, XFS_FSB_TO_B(mp, 1)) + + MAX((__uint16_t)XFS_FSB_TO_B(mp, 1), + XFS_INODE_CLUSTER_SIZE(mp)) + + xfs_calc_buf_res(1, 0) + + xfs_calc_buf_res(2 + XFS_IALLOC_BLOCKS(mp) + + mp->m_in_maxlevels, 0) + + xfs_calc_buf_res(XFS_ALLOCFREE_LOG_COUNT(mp, 1), + XFS_FSB_TO_B(mp, 1)); +} + +/* + * When only changing the inode we log the inode and possibly the superblock + * We also add a bit of slop for the transaction stuff. + */ +STATIC uint +xfs_calc_ichange_reservation( + struct xfs_mount *mp) +{ + return XFS_DQUOT_LOGRES(mp) + + xfs_calc_inode_res(mp, 1) + + xfs_calc_buf_res(1, mp->m_sb.sb_sectsize); + +} + +/* + * Growing the data section of the filesystem. + * superblock + * agi and agf + * allocation btrees + */ +STATIC uint +xfs_calc_growdata_reservation( + struct xfs_mount *mp) +{ + return xfs_calc_buf_res(3, mp->m_sb.sb_sectsize) + + xfs_calc_buf_res(XFS_ALLOCFREE_LOG_COUNT(mp, 1), + XFS_FSB_TO_B(mp, 1)); +} + +/* + * Growing the rt section of the filesystem. + * In the first set of transactions (ALLOC) we allocate space to the + * bitmap or summary files. + * superblock: sector size + * agf of the ag from which the extent is allocated: sector size + * bmap btree for bitmap/summary inode: max depth * blocksize + * bitmap/summary inode: inode size + * allocation btrees for 1 block alloc: 2 * (2 * maxdepth - 1) * blocksize + */ +STATIC uint +xfs_calc_growrtalloc_reservation( + struct xfs_mount *mp) +{ + return xfs_calc_buf_res(2, mp->m_sb.sb_sectsize) + + xfs_calc_buf_res(XFS_BM_MAXLEVELS(mp, XFS_DATA_FORK), + XFS_FSB_TO_B(mp, 1)) + + xfs_calc_inode_res(mp, 1) + + xfs_calc_buf_res(XFS_ALLOCFREE_LOG_COUNT(mp, 1), + XFS_FSB_TO_B(mp, 1)); +} + +/* + * Growing the rt section of the filesystem. + * In the second set of transactions (ZERO) we zero the new metadata blocks. + * one bitmap/summary block: blocksize + */ +STATIC uint +xfs_calc_growrtzero_reservation( + struct xfs_mount *mp) +{ + return xfs_calc_buf_res(1, mp->m_sb.sb_blocksize); +} + +/* + * Growing the rt section of the filesystem. + * In the third set of transactions (FREE) we update metadata without + * allocating any new blocks. + * superblock: sector size + * bitmap inode: inode size + * summary inode: inode size + * one bitmap block: blocksize + * summary blocks: new summary size + */ +STATIC uint +xfs_calc_growrtfree_reservation( + struct xfs_mount *mp) +{ + return xfs_calc_buf_res(1, mp->m_sb.sb_sectsize) + + xfs_calc_inode_res(mp, 2) + + xfs_calc_buf_res(1, mp->m_sb.sb_blocksize) + + xfs_calc_buf_res(1, mp->m_rsumsize); +} + +/* + * Logging the inode modification timestamp on a synchronous write. + * inode + */ +STATIC uint +xfs_calc_swrite_reservation( + struct xfs_mount *mp) +{ + return xfs_calc_inode_res(mp, 1); +} + +/* + * Logging the inode mode bits when writing a setuid/setgid file + * inode + */ +STATIC uint +xfs_calc_writeid_reservation( + struct xfs_mount *mp) +{ + return xfs_calc_inode_res(mp, 1); +} + +/* + * Converting the inode from non-attributed to attributed. + * the inode being converted: inode size + * agf block and superblock (for block allocation) + * the new block (directory sized) + * bmap blocks for the new directory block + * allocation btrees + */ +STATIC uint +xfs_calc_addafork_reservation( + struct xfs_mount *mp) +{ + return XFS_DQUOT_LOGRES(mp) + + xfs_calc_inode_res(mp, 1) + + xfs_calc_buf_res(2, mp->m_sb.sb_sectsize) + + xfs_calc_buf_res(1, mp->m_dirblksize) + + xfs_calc_buf_res(XFS_DAENTER_BMAP1B(mp, XFS_DATA_FORK) + 1, + XFS_FSB_TO_B(mp, 1)) + + xfs_calc_buf_res(XFS_ALLOCFREE_LOG_COUNT(mp, 1), + XFS_FSB_TO_B(mp, 1)); +} + +/* + * Removing the attribute fork of a file + * the inode being truncated: inode size + * the inode's bmap btree: max depth * block size + * And the bmap_finish transaction can free the blocks and bmap blocks: + * the agf for each of the ags: 4 * sector size + * the agfl for each of the ags: 4 * sector size + * the super block to reflect the freed blocks: sector size + * worst case split in allocation btrees per extent assuming 4 extents: + * 4 exts * 2 trees * (2 * max depth - 1) * block size + */ +STATIC uint +xfs_calc_attrinval_reservation( + struct xfs_mount *mp) +{ + return MAX((xfs_calc_inode_res(mp, 1) + + xfs_calc_buf_res(XFS_BM_MAXLEVELS(mp, XFS_ATTR_FORK), + XFS_FSB_TO_B(mp, 1))), + (xfs_calc_buf_res(9, mp->m_sb.sb_sectsize) + + xfs_calc_buf_res(XFS_ALLOCFREE_LOG_COUNT(mp, 4), + XFS_FSB_TO_B(mp, 1)))); +} + +/* + * Setting an attribute at mount time. + * the inode getting the attribute + * the superblock for allocations + * the agfs extents are allocated from + * the attribute btree * max depth + * the inode allocation btree + * Since attribute transaction space is dependent on the size of the attribute, + * the calculation is done partially at mount time and partially at runtime(see + * below). + */ +STATIC uint +xfs_calc_attrsetm_reservation( + struct xfs_mount *mp) +{ + return XFS_DQUOT_LOGRES(mp) + + xfs_calc_inode_res(mp, 1) + + xfs_calc_buf_res(1, mp->m_sb.sb_sectsize) + + xfs_calc_buf_res(XFS_DA_NODE_MAXDEPTH, XFS_FSB_TO_B(mp, 1)); +} + +/* + * Setting an attribute at runtime, transaction space unit per block. + * the superblock for allocations: sector size + * the inode bmap btree could join or split: max depth * block size + * Since the runtime attribute transaction space is dependent on the total + * blocks needed for the 1st bmap, here we calculate out the space unit for + * one block so that the caller could figure out the total space according + * to the attibute extent length in blocks by: + * ext * M_RES(mp)->tr_attrsetrt.tr_logres + */ +STATIC uint +xfs_calc_attrsetrt_reservation( + struct xfs_mount *mp) +{ + return xfs_calc_buf_res(1, mp->m_sb.sb_sectsize) + + xfs_calc_buf_res(XFS_BM_MAXLEVELS(mp, XFS_ATTR_FORK), + XFS_FSB_TO_B(mp, 1)); +} + +/* + * Removing an attribute. + * the inode: inode size + * the attribute btree could join: max depth * block size + * the inode bmap btree could join or split: max depth * block size + * And the bmap_finish transaction can free the attr blocks freed giving: + * the agf for the ag in which the blocks live: 2 * sector size + * the agfl for the ag in which the blocks live: 2 * sector size + * the superblock for the free block count: sector size + * the allocation btrees: 2 exts * 2 trees * (2 * max depth - 1) * block size + */ +STATIC uint +xfs_calc_attrrm_reservation( + struct xfs_mount *mp) +{ + return XFS_DQUOT_LOGRES(mp) + + MAX((xfs_calc_inode_res(mp, 1) + + xfs_calc_buf_res(XFS_DA_NODE_MAXDEPTH, + XFS_FSB_TO_B(mp, 1)) + + (uint)XFS_FSB_TO_B(mp, + XFS_BM_MAXLEVELS(mp, XFS_ATTR_FORK)) + + xfs_calc_buf_res(XFS_BM_MAXLEVELS(mp, XFS_DATA_FORK), 0)), + (xfs_calc_buf_res(5, mp->m_sb.sb_sectsize) + + xfs_calc_buf_res(XFS_ALLOCFREE_LOG_COUNT(mp, 2), + XFS_FSB_TO_B(mp, 1)))); +} + +/* + * Clearing a bad agino number in an agi hash bucket. + */ +STATIC uint +xfs_calc_clear_agi_bucket_reservation( + struct xfs_mount *mp) +{ + return xfs_calc_buf_res(1, mp->m_sb.sb_sectsize); +} + +/* + * Clearing the quotaflags in the superblock. + * the super block for changing quota flags: sector size + */ +STATIC uint +xfs_calc_qm_sbchange_reservation( + struct xfs_mount *mp) +{ + return xfs_calc_buf_res(1, mp->m_sb.sb_sectsize); +} + +/* + * Adjusting quota limits. + * the xfs_disk_dquot_t: sizeof(struct xfs_disk_dquot) + */ +STATIC uint +xfs_calc_qm_setqlim_reservation( + struct xfs_mount *mp) +{ + return xfs_calc_buf_res(1, sizeof(struct xfs_disk_dquot)); +} + +/* + * Allocating quota on disk if needed. + * the write transaction log space: M_RES(mp)->tr_write.tr_logres + * the unit of quota allocation: one system block size + */ +STATIC uint +xfs_calc_qm_dqalloc_reservation( + struct xfs_mount *mp) +{ + ASSERT(M_RES(mp)->tr_write.tr_logres); + return M_RES(mp)->tr_write.tr_logres + + xfs_calc_buf_res(1, + XFS_FSB_TO_B(mp, XFS_DQUOT_CLUSTER_SIZE_FSB) - 1); +} + +/* + * Turning off quotas. + * the xfs_qoff_logitem_t: sizeof(struct xfs_qoff_logitem) * 2 + * the superblock for the quota flags: sector size + */ +STATIC uint +xfs_calc_qm_quotaoff_reservation( + struct xfs_mount *mp) +{ + return sizeof(struct xfs_qoff_logitem) * 2 + + xfs_calc_buf_res(1, mp->m_sb.sb_sectsize); +} + +/* + * End of turning off quotas. + * the xfs_qoff_logitem_t: sizeof(struct xfs_qoff_logitem) * 2 + */ +STATIC uint +xfs_calc_qm_quotaoff_end_reservation( + struct xfs_mount *mp) +{ + return sizeof(struct xfs_qoff_logitem) * 2; +} + +/* + * Syncing the incore super block changes to disk. + * the super block to reflect the changes: sector size + */ +STATIC uint +xfs_calc_sb_reservation( + struct xfs_mount *mp) +{ + return xfs_calc_buf_res(1, mp->m_sb.sb_sectsize); +} + +void +xfs_trans_resv_calc( + struct xfs_mount *mp, + struct xfs_trans_resv *resp) +{ + /* + * The following transactions are logged in physical format and + * require a permanent reservation on space. + */ + resp->tr_write.tr_logres = xfs_calc_write_reservation(mp); + resp->tr_write.tr_logcount = XFS_WRITE_LOG_COUNT; + resp->tr_write.tr_logflags |= XFS_TRANS_PERM_LOG_RES; + + resp->tr_itruncate.tr_logres = xfs_calc_itruncate_reservation(mp); + resp->tr_itruncate.tr_logcount = XFS_ITRUNCATE_LOG_COUNT; + resp->tr_itruncate.tr_logflags |= XFS_TRANS_PERM_LOG_RES; + + resp->tr_rename.tr_logres = xfs_calc_rename_reservation(mp); + resp->tr_rename.tr_logcount = XFS_RENAME_LOG_COUNT; + resp->tr_rename.tr_logflags |= XFS_TRANS_PERM_LOG_RES; + + resp->tr_link.tr_logres = xfs_calc_link_reservation(mp); + resp->tr_link.tr_logcount = XFS_LINK_LOG_COUNT; + resp->tr_link.tr_logflags |= XFS_TRANS_PERM_LOG_RES; + + resp->tr_remove.tr_logres = xfs_calc_remove_reservation(mp); + resp->tr_remove.tr_logcount = XFS_REMOVE_LOG_COUNT; + resp->tr_remove.tr_logflags |= XFS_TRANS_PERM_LOG_RES; + + resp->tr_symlink.tr_logres = xfs_calc_symlink_reservation(mp); + resp->tr_symlink.tr_logcount = XFS_SYMLINK_LOG_COUNT; + resp->tr_symlink.tr_logflags |= XFS_TRANS_PERM_LOG_RES; + + resp->tr_create.tr_logres = xfs_calc_create_reservation(mp); + resp->tr_create.tr_logcount = XFS_CREATE_LOG_COUNT; + resp->tr_create.tr_logflags |= XFS_TRANS_PERM_LOG_RES; + + resp->tr_mkdir.tr_logres = xfs_calc_mkdir_reservation(mp); + resp->tr_mkdir.tr_logcount = XFS_MKDIR_LOG_COUNT; + resp->tr_mkdir.tr_logflags |= XFS_TRANS_PERM_LOG_RES; + + resp->tr_ifree.tr_logres = xfs_calc_ifree_reservation(mp); + resp->tr_ifree.tr_logcount = XFS_INACTIVE_LOG_COUNT; + resp->tr_ifree.tr_logflags |= XFS_TRANS_PERM_LOG_RES; + + resp->tr_addafork.tr_logres = xfs_calc_addafork_reservation(mp); + resp->tr_addafork.tr_logcount = XFS_ADDAFORK_LOG_COUNT; + resp->tr_addafork.tr_logflags |= XFS_TRANS_PERM_LOG_RES; + + resp->tr_attrinval.tr_logres = xfs_calc_attrinval_reservation(mp); + resp->tr_attrinval.tr_logcount = XFS_ATTRINVAL_LOG_COUNT; + resp->tr_attrinval.tr_logflags |= XFS_TRANS_PERM_LOG_RES; + + resp->tr_attrsetm.tr_logres = xfs_calc_attrsetm_reservation(mp); + resp->tr_attrsetm.tr_logcount = XFS_ATTRSET_LOG_COUNT; + resp->tr_attrsetm.tr_logflags |= XFS_TRANS_PERM_LOG_RES; + + resp->tr_attrrm.tr_logres = xfs_calc_attrrm_reservation(mp); + resp->tr_attrrm.tr_logcount = XFS_ATTRRM_LOG_COUNT; + resp->tr_attrrm.tr_logflags |= XFS_TRANS_PERM_LOG_RES; + + resp->tr_growrtalloc.tr_logres = xfs_calc_growrtalloc_reservation(mp); + resp->tr_growrtalloc.tr_logcount = XFS_DEFAULT_PERM_LOG_COUNT; + resp->tr_growrtalloc.tr_logflags |= XFS_TRANS_PERM_LOG_RES; + + resp->tr_qm_dqalloc.tr_logres = xfs_calc_qm_dqalloc_reservation(mp); + resp->tr_qm_dqalloc.tr_logcount = XFS_WRITE_LOG_COUNT; + resp->tr_qm_dqalloc.tr_logflags |= XFS_TRANS_PERM_LOG_RES; + + /* + * The following transactions are logged in logical format with + * a default log count. + */ + resp->tr_qm_sbchange.tr_logres = xfs_calc_qm_sbchange_reservation(mp); + resp->tr_qm_sbchange.tr_logcount = XFS_DEFAULT_LOG_COUNT; + + resp->tr_qm_setqlim.tr_logres = xfs_calc_qm_setqlim_reservation(mp); + resp->tr_qm_setqlim.tr_logcount = XFS_DEFAULT_LOG_COUNT; + + resp->tr_qm_quotaoff.tr_logres = xfs_calc_qm_quotaoff_reservation(mp); + resp->tr_qm_quotaoff.tr_logcount = XFS_DEFAULT_LOG_COUNT; + + resp->tr_qm_equotaoff.tr_logres = + xfs_calc_qm_quotaoff_end_reservation(mp); + resp->tr_qm_equotaoff.tr_logcount = XFS_DEFAULT_LOG_COUNT; + + resp->tr_sb.tr_logres = xfs_calc_sb_reservation(mp); + resp->tr_sb.tr_logcount = XFS_DEFAULT_LOG_COUNT; + + /* The following transaction are logged in logical format */ + resp->tr_ichange.tr_logres = xfs_calc_ichange_reservation(mp); + resp->tr_growdata.tr_logres = xfs_calc_growdata_reservation(mp); + resp->tr_swrite.tr_logres = xfs_calc_swrite_reservation(mp); + resp->tr_fsyncts.tr_logres = xfs_calc_swrite_reservation(mp); + resp->tr_writeid.tr_logres = xfs_calc_writeid_reservation(mp); + resp->tr_attrsetrt.tr_logres = xfs_calc_attrsetrt_reservation(mp); + resp->tr_clearagi.tr_logres = xfs_calc_clear_agi_bucket_reservation(mp); + resp->tr_growrtzero.tr_logres = xfs_calc_growrtzero_reservation(mp); + resp->tr_growrtfree.tr_logres = xfs_calc_growrtfree_reservation(mp); +} diff --git a/fs/xfs/xfs_trans_resv.h b/fs/xfs/xfs_trans_resv.h new file mode 100644 index 0000000..de7de9a --- /dev/null +++ b/fs/xfs/xfs_trans_resv.h @@ -0,0 +1,116 @@ +/* + * Copyright (c) 2000-2002,2005 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#ifndef __XFS_TRANS_RESV_H__ +#define __XFS_TRANS_RESV_H__ + +struct xfs_mount; + +/* + * structure for maintaining pre-calculated transaction reservations. + */ +struct xfs_trans_res { + uint tr_logres; /* log space unit in bytes per log ticket */ + int tr_logcount; /* number of log operations per log ticket */ + int tr_logflags; /* log flags, currently only used for indicating + * a reservation request is permanent or not */ +}; + +struct xfs_trans_resv { + struct xfs_trans_res tr_write; /* extent alloc trans */ + struct xfs_trans_res tr_itruncate; /* truncate trans */ + struct xfs_trans_res tr_rename; /* rename trans */ + struct xfs_trans_res tr_link; /* link trans */ + struct xfs_trans_res tr_remove; /* unlink trans */ + struct xfs_trans_res tr_symlink; /* symlink trans */ + struct xfs_trans_res tr_create; /* create trans */ + struct xfs_trans_res tr_mkdir; /* mkdir trans */ + struct xfs_trans_res tr_ifree; /* inode free trans */ + struct xfs_trans_res tr_ichange; /* inode update trans */ + struct xfs_trans_res tr_growdata; /* fs data section grow trans */ + struct xfs_trans_res tr_swrite; /* sync write inode trans */ + struct xfs_trans_res tr_addafork; /* add inode attr fork trans */ + struct xfs_trans_res tr_writeid; /* write setuid/setgid file */ + struct xfs_trans_res tr_attrinval; /* attr fork buffer + * invalidation */ + struct xfs_trans_res tr_attrsetm; /* set/create an attribute at + * mount time */ + struct xfs_trans_res tr_attrsetrt; /* set/create an attribute at + * runtime */ + struct xfs_trans_res tr_attrrm; /* remove an attribute */ + struct xfs_trans_res tr_clearagi; /* clear agi unlinked bucket */ + struct xfs_trans_res tr_growrtalloc; /* grow realtime allocations */ + struct xfs_trans_res tr_growrtzero; /* grow realtime zeroing */ + struct xfs_trans_res tr_growrtfree; /* grow realtime freeing */ + struct xfs_trans_res tr_qm_sbchange; /* change quota flags */ + struct xfs_trans_res tr_qm_setqlim; /* adjust quota limits */ + struct xfs_trans_res tr_qm_dqalloc; /* allocate quota on disk */ + struct xfs_trans_res tr_qm_quotaoff; /* turn quota off */ + struct xfs_trans_res tr_qm_equotaoff;/* end of turn quota off */ + struct xfs_trans_res tr_sb; /* modify superblock */ + struct xfs_trans_res tr_fsyncts; /* update timestamps on fsync */ +}; + +/* shorthand way of accessing reservation structure */ +#define M_RES(mp) (&(mp)->m_resv) + +/* + * Per-extent log reservation for the allocation btree changes + * involved in freeing or allocating an extent. + * 2 trees * (2 blocks/level * max depth - 1) * block size + */ +#define XFS_ALLOCFREE_LOG_RES(mp,nx) \ + ((nx) * (2 * XFS_FSB_TO_B((mp), 2 * XFS_AG_MAXLEVELS(mp) - 1))) +#define XFS_ALLOCFREE_LOG_COUNT(mp,nx) \ + ((nx) * (2 * (2 * XFS_AG_MAXLEVELS(mp) - 1))) + +/* + * Per-directory log reservation for any directory change. + * dir blocks: (1 btree block per level + data block + free block) * dblock size + * bmap btree: (levels + 2) * max depth * block size + * v2 directory blocks can be fragmented below the dirblksize down to the fsb + * size, so account for that in the DAENTER macros. + */ +#define XFS_DIROP_LOG_RES(mp) \ + (XFS_FSB_TO_B(mp, XFS_DAENTER_BLOCKS(mp, XFS_DATA_FORK)) + \ + (XFS_FSB_TO_B(mp, XFS_DAENTER_BMAPS(mp, XFS_DATA_FORK) + 1))) +#define XFS_DIROP_LOG_COUNT(mp) \ + (XFS_DAENTER_BLOCKS(mp, XFS_DATA_FORK) + \ + XFS_DAENTER_BMAPS(mp, XFS_DATA_FORK) + 1) + +/* + * Various log count values. + */ +#define XFS_DEFAULT_LOG_COUNT 1 +#define XFS_DEFAULT_PERM_LOG_COUNT 2 +#define XFS_ITRUNCATE_LOG_COUNT 2 +#define XFS_INACTIVE_LOG_COUNT 2 +#define XFS_CREATE_LOG_COUNT 2 +#define XFS_MKDIR_LOG_COUNT 3 +#define XFS_SYMLINK_LOG_COUNT 3 +#define XFS_REMOVE_LOG_COUNT 2 +#define XFS_LINK_LOG_COUNT 2 +#define XFS_RENAME_LOG_COUNT 2 +#define XFS_WRITE_LOG_COUNT 2 +#define XFS_ADDAFORK_LOG_COUNT 2 +#define XFS_ATTRINVAL_LOG_COUNT 1 +#define XFS_ATTRSET_LOG_COUNT 3 +#define XFS_ATTRRM_LOG_COUNT 3 + +void xfs_trans_resv_calc(struct xfs_mount *mp, struct xfs_trans_resv *resp); + +#endif /* __XFS_TRANS_RESV_H__ */ diff --git a/fs/xfs/xfs_types.h b/fs/xfs/xfs_types.h index 61ba1cf..82bbc34 100644 --- a/fs/xfs/xfs_types.h +++ b/fs/xfs/xfs_types.h @@ -18,42 +18,7 @@ #ifndef __XFS_TYPES_H__ #define __XFS_TYPES_H__ -#ifdef __KERNEL__ - -/* - * Additional type declarations for XFS - */ -typedef signed char __int8_t; -typedef unsigned char __uint8_t; -typedef signed short int __int16_t; -typedef unsigned short int __uint16_t; -typedef signed int __int32_t; -typedef unsigned int __uint32_t; -typedef signed long long int __int64_t; -typedef unsigned long long int __uint64_t; - -typedef __uint32_t prid_t; /* project ID */ -typedef __uint32_t inst_t; /* an instruction */ - -typedef __s64 xfs_off_t; /* <file offset> type */ -typedef unsigned long long xfs_ino_t; /* <inode> type */ -typedef __s64 xfs_daddr_t; /* <disk address> type */ -typedef char * xfs_caddr_t; /* <core address> type */ -typedef __u32 xfs_dev_t; -typedef __u32 xfs_nlink_t; - -/* __psint_t is the same size as a pointer */ -#if (BITS_PER_LONG == 32) -typedef __int32_t __psint_t; -typedef __uint32_t __psunsigned_t; -#elif (BITS_PER_LONG == 64) -typedef __int64_t __psint_t; -typedef __uint64_t __psunsigned_t; -#else -#error BITS_PER_LONG must be 32 or 64 -#endif - -#endif /* __KERNEL__ */ +typedef __uint32_t prid_t; /* project ID */ typedef __uint32_t xfs_agblock_t; /* blockno in alloc. group */ typedef __uint32_t xfs_agino_t; /* inode # within allocation grp */ @@ -146,6 +111,12 @@ typedef __uint64_t xfs_filblks_t; /* number of blocks in a file */ #define XFS_MAX_SECTORSIZE (1 << XFS_MAX_SECTORSIZE_LOG) /* + * Inode fork identifiers. + */ +#define XFS_DATA_FORK 0 +#define XFS_ATTR_FORK 1 + +/* * Min numbers of data/attr fork btree root pointers. */ #define MINDBTPTRS 3 @@ -169,6 +140,23 @@ typedef enum { struct xfs_name { const unsigned char *name; int len; + int type; }; +/* + * uid_t and gid_t are hard-coded to 32 bits in the inode. + * Hence, an 'id' in a dquot is 32 bits.. + */ +typedef __uint32_t xfs_dqid_t; + +/* + * Constants for bit manipulations. + */ +#define XFS_NBBYLOG 3 /* log2(NBBY) */ +#define XFS_WORDLOG 2 /* log2(sizeof(xfs_rtword_t)) */ +#define XFS_NBWORDLOG (XFS_NBBYLOG + XFS_WORDLOG) +#define XFS_NBWORD (1 << XFS_NBWORDLOG) +#define XFS_WORDMASK ((1 << XFS_WORDLOG) - 1) + + #endif /* __XFS_TYPES_H__ */ diff --git a/fs/xfs/xfs_utils.c b/fs/xfs/xfs_utils.c deleted file mode 100644 index 0025c78..0000000 --- a/fs/xfs/xfs_utils.c +++ /dev/null @@ -1,314 +0,0 @@ -/* - * Copyright (c) 2000-2002,2005 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#include "xfs.h" -#include "xfs_fs.h" -#include "xfs_types.h" -#include "xfs_log.h" -#include "xfs_trans.h" -#include "xfs_sb.h" -#include "xfs_ag.h" -#include "xfs_dir2.h" -#include "xfs_mount.h" -#include "xfs_bmap_btree.h" -#include "xfs_dinode.h" -#include "xfs_inode.h" -#include "xfs_inode_item.h" -#include "xfs_bmap.h" -#include "xfs_error.h" -#include "xfs_quota.h" -#include "xfs_itable.h" -#include "xfs_utils.h" - - -/* - * Allocates a new inode from disk and return a pointer to the - * incore copy. This routine will internally commit the current - * transaction and allocate a new one if the Space Manager needed - * to do an allocation to replenish the inode free-list. - * - * This routine is designed to be called from xfs_create and - * xfs_create_dir. - * - */ -int -xfs_dir_ialloc( - xfs_trans_t **tpp, /* input: current transaction; - output: may be a new transaction. */ - xfs_inode_t *dp, /* directory within whose allocate - the inode. */ - umode_t mode, - xfs_nlink_t nlink, - xfs_dev_t rdev, - prid_t prid, /* project id */ - int okalloc, /* ok to allocate new space */ - xfs_inode_t **ipp, /* pointer to inode; it will be - locked. */ - int *committed) - -{ - xfs_trans_t *tp; - xfs_trans_t *ntp; - xfs_inode_t *ip; - xfs_buf_t *ialloc_context = NULL; - int code; - uint log_res; - uint log_count; - void *dqinfo; - uint tflags; - - tp = *tpp; - ASSERT(tp->t_flags & XFS_TRANS_PERM_LOG_RES); - - /* - * xfs_ialloc will return a pointer to an incore inode if - * the Space Manager has an available inode on the free - * list. Otherwise, it will do an allocation and replenish - * the freelist. Since we can only do one allocation per - * transaction without deadlocks, we will need to commit the - * current transaction and start a new one. We will then - * need to call xfs_ialloc again to get the inode. - * - * If xfs_ialloc did an allocation to replenish the freelist, - * it returns the bp containing the head of the freelist as - * ialloc_context. We will hold a lock on it across the - * transaction commit so that no other process can steal - * the inode(s) that we've just allocated. - */ - code = xfs_ialloc(tp, dp, mode, nlink, rdev, prid, okalloc, - &ialloc_context, &ip); - - /* - * Return an error if we were unable to allocate a new inode. - * This should only happen if we run out of space on disk or - * encounter a disk error. - */ - if (code) { - *ipp = NULL; - return code; - } - if (!ialloc_context && !ip) { - *ipp = NULL; - return XFS_ERROR(ENOSPC); - } - - /* - * If the AGI buffer is non-NULL, then we were unable to get an - * inode in one operation. We need to commit the current - * transaction and call xfs_ialloc() again. It is guaranteed - * to succeed the second time. - */ - if (ialloc_context) { - /* - * Normally, xfs_trans_commit releases all the locks. - * We call bhold to hang on to the ialloc_context across - * the commit. Holding this buffer prevents any other - * processes from doing any allocations in this - * allocation group. - */ - xfs_trans_bhold(tp, ialloc_context); - /* - * Save the log reservation so we can use - * them in the next transaction. - */ - log_res = xfs_trans_get_log_res(tp); - log_count = xfs_trans_get_log_count(tp); - - /* - * We want the quota changes to be associated with the next - * transaction, NOT this one. So, detach the dqinfo from this - * and attach it to the next transaction. - */ - dqinfo = NULL; - tflags = 0; - if (tp->t_dqinfo) { - dqinfo = (void *)tp->t_dqinfo; - tp->t_dqinfo = NULL; - tflags = tp->t_flags & XFS_TRANS_DQ_DIRTY; - tp->t_flags &= ~(XFS_TRANS_DQ_DIRTY); - } - - ntp = xfs_trans_dup(tp); - code = xfs_trans_commit(tp, 0); - tp = ntp; - if (committed != NULL) { - *committed = 1; - } - /* - * If we get an error during the commit processing, - * release the buffer that is still held and return - * to the caller. - */ - if (code) { - xfs_buf_relse(ialloc_context); - if (dqinfo) { - tp->t_dqinfo = dqinfo; - xfs_trans_free_dqinfo(tp); - } - *tpp = ntp; - *ipp = NULL; - return code; - } - - /* - * transaction commit worked ok so we can drop the extra ticket - * reference that we gained in xfs_trans_dup() - */ - xfs_log_ticket_put(tp->t_ticket); - code = xfs_trans_reserve(tp, 0, log_res, 0, - XFS_TRANS_PERM_LOG_RES, log_count); - /* - * Re-attach the quota info that we detached from prev trx. - */ - if (dqinfo) { - tp->t_dqinfo = dqinfo; - tp->t_flags |= tflags; - } - - if (code) { - xfs_buf_relse(ialloc_context); - *tpp = ntp; - *ipp = NULL; - return code; - } - xfs_trans_bjoin(tp, ialloc_context); - - /* - * Call ialloc again. Since we've locked out all - * other allocations in this allocation group, - * this call should always succeed. - */ - code = xfs_ialloc(tp, dp, mode, nlink, rdev, prid, - okalloc, &ialloc_context, &ip); - - /* - * If we get an error at this point, return to the caller - * so that the current transaction can be aborted. - */ - if (code) { - *tpp = tp; - *ipp = NULL; - return code; - } - ASSERT(!ialloc_context && ip); - - } else { - if (committed != NULL) - *committed = 0; - } - - *ipp = ip; - *tpp = tp; - - return 0; -} - -/* - * Decrement the link count on an inode & log the change. - * If this causes the link count to go to zero, initiate the - * logging activity required to truncate a file. - */ -int /* error */ -xfs_droplink( - xfs_trans_t *tp, - xfs_inode_t *ip) -{ - int error; - - xfs_trans_ichgtime(tp, ip, XFS_ICHGTIME_CHG); - - ASSERT (ip->i_d.di_nlink > 0); - ip->i_d.di_nlink--; - drop_nlink(VFS_I(ip)); - xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); - - error = 0; - if (ip->i_d.di_nlink == 0) { - /* - * We're dropping the last link to this file. - * Move the on-disk inode to the AGI unlinked list. - * From xfs_inactive() we will pull the inode from - * the list and free it. - */ - error = xfs_iunlink(tp, ip); - } - return error; -} - -/* - * This gets called when the inode's version needs to be changed from 1 to 2. - * Currently this happens when the nlink field overflows the old 16-bit value - * or when chproj is called to change the project for the first time. - * As a side effect the superblock version will also get rev'd - * to contain the NLINK bit. - */ -void -xfs_bump_ino_vers2( - xfs_trans_t *tp, - xfs_inode_t *ip) -{ - xfs_mount_t *mp; - - ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); - ASSERT(ip->i_d.di_version == 1); - - ip->i_d.di_version = 2; - ip->i_d.di_onlink = 0; - memset(&(ip->i_d.di_pad[0]), 0, sizeof(ip->i_d.di_pad)); - mp = tp->t_mountp; - if (!xfs_sb_version_hasnlink(&mp->m_sb)) { - spin_lock(&mp->m_sb_lock); - if (!xfs_sb_version_hasnlink(&mp->m_sb)) { - xfs_sb_version_addnlink(&mp->m_sb); - spin_unlock(&mp->m_sb_lock); - xfs_mod_sb(tp, XFS_SB_VERSIONNUM); - } else { - spin_unlock(&mp->m_sb_lock); - } - } - /* Caller must log the inode */ -} - -/* - * Increment the link count on an inode & log the change. - */ -int -xfs_bumplink( - xfs_trans_t *tp, - xfs_inode_t *ip) -{ - xfs_trans_ichgtime(tp, ip, XFS_ICHGTIME_CHG); - - ASSERT(ip->i_d.di_nlink > 0); - ip->i_d.di_nlink++; - inc_nlink(VFS_I(ip)); - if ((ip->i_d.di_version == 1) && - (ip->i_d.di_nlink > XFS_MAXLINK_1)) { - /* - * The inode has increased its number of links beyond - * what can fit in an old format inode. It now needs - * to be converted to a version 2 inode with a 32 bit - * link count. If this is the first inode in the file - * system to do this, then we need to bump the superblock - * version number as well. - */ - xfs_bump_ino_vers2(tp, ip); - } - - xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); - return 0; -} diff --git a/fs/xfs/xfs_utils.h b/fs/xfs/xfs_utils.h deleted file mode 100644 index 5eeab46..0000000 --- a/fs/xfs/xfs_utils.h +++ /dev/null @@ -1,27 +0,0 @@ -/* - * Copyright (c) 2000-2002,2005 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#ifndef __XFS_UTILS_H__ -#define __XFS_UTILS_H__ - -extern int xfs_dir_ialloc(xfs_trans_t **, xfs_inode_t *, umode_t, xfs_nlink_t, - xfs_dev_t, prid_t, int, xfs_inode_t **, int *); -extern int xfs_droplink(xfs_trans_t *, xfs_inode_t *); -extern int xfs_bumplink(xfs_trans_t *, xfs_inode_t *); -extern void xfs_bump_ino_vers2(xfs_trans_t *, xfs_inode_t *); - -#endif /* __XFS_UTILS_H__ */ diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c deleted file mode 100644 index dc730ac..0000000 --- a/fs/xfs/xfs_vnodeops.c +++ /dev/null @@ -1,1870 +0,0 @@ -/* - * Copyright (c) 2000-2006 Silicon Graphics, Inc. - * Copyright (c) 2012 Red Hat, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ - -#include "xfs.h" -#include "xfs_fs.h" -#include "xfs_types.h" -#include "xfs_bit.h" -#include "xfs_log.h" -#include "xfs_trans.h" -#include "xfs_sb.h" -#include "xfs_ag.h" -#include "xfs_dir2.h" -#include "xfs_mount.h" -#include "xfs_da_btree.h" -#include "xfs_bmap_btree.h" -#include "xfs_ialloc_btree.h" -#include "xfs_dinode.h" -#include "xfs_inode.h" -#include "xfs_inode_item.h" -#include "xfs_itable.h" -#include "xfs_ialloc.h" -#include "xfs_alloc.h" -#include "xfs_bmap.h" -#include "xfs_acl.h" -#include "xfs_attr.h" -#include "xfs_error.h" -#include "xfs_quota.h" -#include "xfs_utils.h" -#include "xfs_rtalloc.h" -#include "xfs_trans_space.h" -#include "xfs_log_priv.h" -#include "xfs_filestream.h" -#include "xfs_vnodeops.h" -#include "xfs_trace.h" -#include "xfs_icache.h" -#include "xfs_symlink.h" - - -/* - * This is called by xfs_inactive to free any blocks beyond eof - * when the link count isn't zero and by xfs_dm_punch_hole() when - * punching a hole to EOF. - */ -int -xfs_free_eofblocks( - xfs_mount_t *mp, - xfs_inode_t *ip, - bool need_iolock) -{ - xfs_trans_t *tp; - int error; - xfs_fileoff_t end_fsb; - xfs_fileoff_t last_fsb; - xfs_filblks_t map_len; - int nimaps; - xfs_bmbt_irec_t imap; - - /* - * Figure out if there are any blocks beyond the end - * of the file. If not, then there is nothing to do. - */ - end_fsb = XFS_B_TO_FSB(mp, (xfs_ufsize_t)XFS_ISIZE(ip)); - last_fsb = XFS_B_TO_FSB(mp, mp->m_super->s_maxbytes); - if (last_fsb <= end_fsb) - return 0; - map_len = last_fsb - end_fsb; - - nimaps = 1; - xfs_ilock(ip, XFS_ILOCK_SHARED); - error = xfs_bmapi_read(ip, end_fsb, map_len, &imap, &nimaps, 0); - xfs_iunlock(ip, XFS_ILOCK_SHARED); - - if (!error && (nimaps != 0) && - (imap.br_startblock != HOLESTARTBLOCK || - ip->i_delayed_blks)) { - /* - * Attach the dquots to the inode up front. - */ - error = xfs_qm_dqattach(ip, 0); - if (error) - return error; - - /* - * There are blocks after the end of file. - * Free them up now by truncating the file to - * its current size. - */ - tp = xfs_trans_alloc(mp, XFS_TRANS_INACTIVE); - - if (need_iolock) { - if (!xfs_ilock_nowait(ip, XFS_IOLOCK_EXCL)) { - xfs_trans_cancel(tp, 0); - return EAGAIN; - } - } - - error = xfs_trans_reserve(tp, 0, - XFS_ITRUNCATE_LOG_RES(mp), - 0, XFS_TRANS_PERM_LOG_RES, - XFS_ITRUNCATE_LOG_COUNT); - if (error) { - ASSERT(XFS_FORCED_SHUTDOWN(mp)); - xfs_trans_cancel(tp, 0); - if (need_iolock) - xfs_iunlock(ip, XFS_IOLOCK_EXCL); - return error; - } - - xfs_ilock(ip, XFS_ILOCK_EXCL); - xfs_trans_ijoin(tp, ip, 0); - - /* - * Do not update the on-disk file size. If we update the - * on-disk file size and then the system crashes before the - * contents of the file are flushed to disk then the files - * may be full of holes (ie NULL files bug). - */ - error = xfs_itruncate_extents(&tp, ip, XFS_DATA_FORK, - XFS_ISIZE(ip)); - if (error) { - /* - * If we get an error at this point we simply don't - * bother truncating the file. - */ - xfs_trans_cancel(tp, - (XFS_TRANS_RELEASE_LOG_RES | - XFS_TRANS_ABORT)); - } else { - error = xfs_trans_commit(tp, - XFS_TRANS_RELEASE_LOG_RES); - if (!error) - xfs_inode_clear_eofblocks_tag(ip); - } - - xfs_iunlock(ip, XFS_ILOCK_EXCL); - if (need_iolock) - xfs_iunlock(ip, XFS_IOLOCK_EXCL); - } - return error; -} - -int -xfs_release( - xfs_inode_t *ip) -{ - xfs_mount_t *mp = ip->i_mount; - int error; - - if (!S_ISREG(ip->i_d.di_mode) || (ip->i_d.di_mode == 0)) - return 0; - - /* If this is a read-only mount, don't do this (would generate I/O) */ - if (mp->m_flags & XFS_MOUNT_RDONLY) - return 0; - - if (!XFS_FORCED_SHUTDOWN(mp)) { - int truncated; - - /* - * If we are using filestreams, and we have an unlinked - * file that we are processing the last close on, then nothing - * will be able to reopen and write to this file. Purge this - * inode from the filestreams cache so that it doesn't delay - * teardown of the inode. - */ - if ((ip->i_d.di_nlink == 0) && xfs_inode_is_filestream(ip)) - xfs_filestream_deassociate(ip); - - /* - * If we previously truncated this file and removed old data - * in the process, we want to initiate "early" writeout on - * the last close. This is an attempt to combat the notorious - * NULL files problem which is particularly noticeable from a - * truncate down, buffered (re-)write (delalloc), followed by - * a crash. What we are effectively doing here is - * significantly reducing the time window where we'd otherwise - * be exposed to that problem. - */ - truncated = xfs_iflags_test_and_clear(ip, XFS_ITRUNCATED); - if (truncated) { - xfs_iflags_clear(ip, XFS_IDIRTY_RELEASE); - if (VN_DIRTY(VFS_I(ip)) && ip->i_delayed_blks > 0) { - error = -filemap_flush(VFS_I(ip)->i_mapping); - if (error) - return error; - } - } - } - - if (ip->i_d.di_nlink == 0) - return 0; - - if (xfs_can_free_eofblocks(ip, false)) { - - /* - * If we can't get the iolock just skip truncating the blocks - * past EOF because we could deadlock with the mmap_sem - * otherwise. We'll get another chance to drop them once the - * last reference to the inode is dropped, so we'll never leak - * blocks permanently. - * - * Further, check if the inode is being opened, written and - * closed frequently and we have delayed allocation blocks - * outstanding (e.g. streaming writes from the NFS server), - * truncating the blocks past EOF will cause fragmentation to - * occur. - * - * In this case don't do the truncation, either, but we have to - * be careful how we detect this case. Blocks beyond EOF show - * up as i_delayed_blks even when the inode is clean, so we - * need to truncate them away first before checking for a dirty - * release. Hence on the first dirty close we will still remove - * the speculative allocation, but after that we will leave it - * in place. - */ - if (xfs_iflags_test(ip, XFS_IDIRTY_RELEASE)) - return 0; - - error = xfs_free_eofblocks(mp, ip, true); - if (error && error != EAGAIN) - return error; - - /* delalloc blocks after truncation means it really is dirty */ - if (ip->i_delayed_blks) - xfs_iflags_set(ip, XFS_IDIRTY_RELEASE); - } - return 0; -} - -/* - * xfs_inactive - * - * This is called when the vnode reference count for the vnode - * goes to zero. If the file has been unlinked, then it must - * now be truncated. Also, we clear all of the read-ahead state - * kept for the inode here since the file is now closed. - */ -int -xfs_inactive( - xfs_inode_t *ip) -{ - xfs_bmap_free_t free_list; - xfs_fsblock_t first_block; - int committed; - xfs_trans_t *tp; - xfs_mount_t *mp; - int error; - int truncate = 0; - - /* - * If the inode is already free, then there can be nothing - * to clean up here. - */ - if (ip->i_d.di_mode == 0 || is_bad_inode(VFS_I(ip))) { - ASSERT(ip->i_df.if_real_bytes == 0); - ASSERT(ip->i_df.if_broot_bytes == 0); - return VN_INACTIVE_CACHE; - } - - mp = ip->i_mount; - - error = 0; - - /* If this is a read-only mount, don't do this (would generate I/O) */ - if (mp->m_flags & XFS_MOUNT_RDONLY) - goto out; - - if (ip->i_d.di_nlink != 0) { - /* - * force is true because we are evicting an inode from the - * cache. Post-eof blocks must be freed, lest we end up with - * broken free space accounting. - */ - if (xfs_can_free_eofblocks(ip, true)) { - error = xfs_free_eofblocks(mp, ip, false); - if (error) - return VN_INACTIVE_CACHE; - } - goto out; - } - - if (S_ISREG(ip->i_d.di_mode) && - (ip->i_d.di_size != 0 || XFS_ISIZE(ip) != 0 || - ip->i_d.di_nextents > 0 || ip->i_delayed_blks > 0)) - truncate = 1; - - error = xfs_qm_dqattach(ip, 0); - if (error) - return VN_INACTIVE_CACHE; - - tp = xfs_trans_alloc(mp, XFS_TRANS_INACTIVE); - error = xfs_trans_reserve(tp, 0, - (truncate || S_ISLNK(ip->i_d.di_mode)) ? - XFS_ITRUNCATE_LOG_RES(mp) : - XFS_IFREE_LOG_RES(mp), - 0, - XFS_TRANS_PERM_LOG_RES, - XFS_ITRUNCATE_LOG_COUNT); - if (error) { - ASSERT(XFS_FORCED_SHUTDOWN(mp)); - xfs_trans_cancel(tp, 0); - return VN_INACTIVE_CACHE; - } - - xfs_ilock(ip, XFS_ILOCK_EXCL); - xfs_trans_ijoin(tp, ip, 0); - - if (S_ISLNK(ip->i_d.di_mode)) { - error = xfs_inactive_symlink(ip, &tp); - if (error) - goto out_cancel; - } else if (truncate) { - ip->i_d.di_size = 0; - xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); - - error = xfs_itruncate_extents(&tp, ip, XFS_DATA_FORK, 0); - if (error) - goto out_cancel; - - ASSERT(ip->i_d.di_nextents == 0); - } - - /* - * If there are attributes associated with the file then blow them away - * now. The code calls a routine that recursively deconstructs the - * attribute fork. We need to just commit the current transaction - * because we can't use it for xfs_attr_inactive(). - */ - if (ip->i_d.di_anextents > 0) { - ASSERT(ip->i_d.di_forkoff != 0); - - error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES); - if (error) - goto out_unlock; - - xfs_iunlock(ip, XFS_ILOCK_EXCL); - - error = xfs_attr_inactive(ip); - if (error) - goto out; - - tp = xfs_trans_alloc(mp, XFS_TRANS_INACTIVE); - error = xfs_trans_reserve(tp, 0, - XFS_IFREE_LOG_RES(mp), - 0, XFS_TRANS_PERM_LOG_RES, - XFS_INACTIVE_LOG_COUNT); - if (error) { - xfs_trans_cancel(tp, 0); - goto out; - } - - xfs_ilock(ip, XFS_ILOCK_EXCL); - xfs_trans_ijoin(tp, ip, 0); - } - - if (ip->i_afp) - xfs_idestroy_fork(ip, XFS_ATTR_FORK); - - ASSERT(ip->i_d.di_anextents == 0); - - /* - * Free the inode. - */ - xfs_bmap_init(&free_list, &first_block); - error = xfs_ifree(tp, ip, &free_list); - if (error) { - /* - * If we fail to free the inode, shut down. The cancel - * might do that, we need to make sure. Otherwise the - * inode might be lost for a long time or forever. - */ - if (!XFS_FORCED_SHUTDOWN(mp)) { - xfs_notice(mp, "%s: xfs_ifree returned error %d", - __func__, error); - xfs_force_shutdown(mp, SHUTDOWN_META_IO_ERROR); - } - xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES|XFS_TRANS_ABORT); - } else { - /* - * Credit the quota account(s). The inode is gone. - */ - xfs_trans_mod_dquot_byino(tp, ip, XFS_TRANS_DQ_ICOUNT, -1); - - /* - * Just ignore errors at this point. There is nothing we can - * do except to try to keep going. Make sure it's not a silent - * error. - */ - error = xfs_bmap_finish(&tp, &free_list, &committed); - if (error) - xfs_notice(mp, "%s: xfs_bmap_finish returned error %d", - __func__, error); - error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES); - if (error) - xfs_notice(mp, "%s: xfs_trans_commit returned error %d", - __func__, error); - } - - /* - * Release the dquots held by inode, if any. - */ - xfs_qm_dqdetach(ip); -out_unlock: - xfs_iunlock(ip, XFS_ILOCK_EXCL); -out: - return VN_INACTIVE_CACHE; -out_cancel: - xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES | XFS_TRANS_ABORT); - goto out_unlock; -} - -/* - * Lookups up an inode from "name". If ci_name is not NULL, then a CI match - * is allowed, otherwise it has to be an exact match. If a CI match is found, - * ci_name->name will point to a the actual name (caller must free) or - * will be set to NULL if an exact match is found. - */ -int -xfs_lookup( - xfs_inode_t *dp, - struct xfs_name *name, - xfs_inode_t **ipp, - struct xfs_name *ci_name) -{ - xfs_ino_t inum; - int error; - uint lock_mode; - - trace_xfs_lookup(dp, name); - - if (XFS_FORCED_SHUTDOWN(dp->i_mount)) - return XFS_ERROR(EIO); - - lock_mode = xfs_ilock_map_shared(dp); - error = xfs_dir_lookup(NULL, dp, name, &inum, ci_name); - xfs_iunlock_map_shared(dp, lock_mode); - - if (error) - goto out; - - error = xfs_iget(dp->i_mount, NULL, inum, 0, 0, ipp); - if (error) - goto out_free_name; - - return 0; - -out_free_name: - if (ci_name) - kmem_free(ci_name->name); -out: - *ipp = NULL; - return error; -} - -int -xfs_create( - xfs_inode_t *dp, - struct xfs_name *name, - umode_t mode, - xfs_dev_t rdev, - xfs_inode_t **ipp) -{ - int is_dir = S_ISDIR(mode); - struct xfs_mount *mp = dp->i_mount; - struct xfs_inode *ip = NULL; - struct xfs_trans *tp = NULL; - int error; - xfs_bmap_free_t free_list; - xfs_fsblock_t first_block; - bool unlock_dp_on_error = false; - uint cancel_flags; - int committed; - prid_t prid; - struct xfs_dquot *udqp = NULL; - struct xfs_dquot *gdqp = NULL; - struct xfs_dquot *pdqp = NULL; - uint resblks; - uint log_res; - uint log_count; - - trace_xfs_create(dp, name); - - if (XFS_FORCED_SHUTDOWN(mp)) - return XFS_ERROR(EIO); - - if (dp->i_d.di_flags & XFS_DIFLAG_PROJINHERIT) - prid = xfs_get_projid(dp); - else - prid = XFS_PROJID_DEFAULT; - - /* - * Make sure that we have allocated dquot(s) on disk. - */ - error = xfs_qm_vop_dqalloc(dp, current_fsuid(), current_fsgid(), prid, - XFS_QMOPT_QUOTALL | XFS_QMOPT_INHERIT, - &udqp, &gdqp, &pdqp); - if (error) - return error; - - if (is_dir) { - rdev = 0; - resblks = XFS_MKDIR_SPACE_RES(mp, name->len); - log_res = XFS_MKDIR_LOG_RES(mp); - log_count = XFS_MKDIR_LOG_COUNT; - tp = xfs_trans_alloc(mp, XFS_TRANS_MKDIR); - } else { - resblks = XFS_CREATE_SPACE_RES(mp, name->len); - log_res = XFS_CREATE_LOG_RES(mp); - log_count = XFS_CREATE_LOG_COUNT; - tp = xfs_trans_alloc(mp, XFS_TRANS_CREATE); - } - - cancel_flags = XFS_TRANS_RELEASE_LOG_RES; - - /* - * Initially assume that the file does not exist and - * reserve the resources for that case. If that is not - * the case we'll drop the one we have and get a more - * appropriate transaction later. - */ - error = xfs_trans_reserve(tp, resblks, log_res, 0, - XFS_TRANS_PERM_LOG_RES, log_count); - if (error == ENOSPC) { - /* flush outstanding delalloc blocks and retry */ - xfs_flush_inodes(mp); - error = xfs_trans_reserve(tp, resblks, log_res, 0, - XFS_TRANS_PERM_LOG_RES, log_count); - } - if (error == ENOSPC) { - /* No space at all so try a "no-allocation" reservation */ - resblks = 0; - error = xfs_trans_reserve(tp, 0, log_res, 0, - XFS_TRANS_PERM_LOG_RES, log_count); - } - if (error) { - cancel_flags = 0; - goto out_trans_cancel; - } - - xfs_ilock(dp, XFS_ILOCK_EXCL | XFS_ILOCK_PARENT); - unlock_dp_on_error = true; - - xfs_bmap_init(&free_list, &first_block); - - /* - * Reserve disk quota and the inode. - */ - error = xfs_trans_reserve_quota(tp, mp, udqp, gdqp, - pdqp, resblks, 1, 0); - if (error) - goto out_trans_cancel; - - error = xfs_dir_canenter(tp, dp, name, resblks); - if (error) - goto out_trans_cancel; - - /* - * A newly created regular or special file just has one directory - * entry pointing to them, but a directory also the "." entry - * pointing to itself. - */ - error = xfs_dir_ialloc(&tp, dp, mode, is_dir ? 2 : 1, rdev, - prid, resblks > 0, &ip, &committed); - if (error) { - if (error == ENOSPC) - goto out_trans_cancel; - goto out_trans_abort; - } - - /* - * Now we join the directory inode to the transaction. We do not do it - * earlier because xfs_dir_ialloc might commit the previous transaction - * (and release all the locks). An error from here on will result in - * the transaction cancel unlocking dp so don't do it explicitly in the - * error path. - */ - xfs_trans_ijoin(tp, dp, XFS_ILOCK_EXCL); - unlock_dp_on_error = false; - - error = xfs_dir_createname(tp, dp, name, ip->i_ino, - &first_block, &free_list, resblks ? - resblks - XFS_IALLOC_SPACE_RES(mp) : 0); - if (error) { - ASSERT(error != ENOSPC); - goto out_trans_abort; - } - xfs_trans_ichgtime(tp, dp, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG); - xfs_trans_log_inode(tp, dp, XFS_ILOG_CORE); - - if (is_dir) { - error = xfs_dir_init(tp, ip, dp); - if (error) - goto out_bmap_cancel; - - error = xfs_bumplink(tp, dp); - if (error) - goto out_bmap_cancel; - } - - /* - * If this is a synchronous mount, make sure that the - * create transaction goes to disk before returning to - * the user. - */ - if (mp->m_flags & (XFS_MOUNT_WSYNC|XFS_MOUNT_DIRSYNC)) - xfs_trans_set_sync(tp); - - /* - * Attach the dquot(s) to the inodes and modify them incore. - * These ids of the inode couldn't have changed since the new - * inode has been locked ever since it was created. - */ - xfs_qm_vop_create_dqattach(tp, ip, udqp, gdqp, pdqp); - - error = xfs_bmap_finish(&tp, &free_list, &committed); - if (error) - goto out_bmap_cancel; - - error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES); - if (error) - goto out_release_inode; - - xfs_qm_dqrele(udqp); - xfs_qm_dqrele(gdqp); - xfs_qm_dqrele(pdqp); - - *ipp = ip; - return 0; - - out_bmap_cancel: - xfs_bmap_cancel(&free_list); - out_trans_abort: - cancel_flags |= XFS_TRANS_ABORT; - out_trans_cancel: - xfs_trans_cancel(tp, cancel_flags); - out_release_inode: - /* - * Wait until after the current transaction is aborted to - * release the inode. This prevents recursive transactions - * and deadlocks from xfs_inactive. - */ - if (ip) - IRELE(ip); - - xfs_qm_dqrele(udqp); - xfs_qm_dqrele(gdqp); - xfs_qm_dqrele(pdqp); - - if (unlock_dp_on_error) - xfs_iunlock(dp, XFS_ILOCK_EXCL); - return error; -} - -#ifdef DEBUG -int xfs_locked_n; -int xfs_small_retries; -int xfs_middle_retries; -int xfs_lots_retries; -int xfs_lock_delays; -#endif - -/* - * Bump the subclass so xfs_lock_inodes() acquires each lock with - * a different value - */ -static inline int -xfs_lock_inumorder(int lock_mode, int subclass) -{ - if (lock_mode & (XFS_IOLOCK_SHARED|XFS_IOLOCK_EXCL)) - lock_mode |= (subclass + XFS_LOCK_INUMORDER) << XFS_IOLOCK_SHIFT; - if (lock_mode & (XFS_ILOCK_SHARED|XFS_ILOCK_EXCL)) - lock_mode |= (subclass + XFS_LOCK_INUMORDER) << XFS_ILOCK_SHIFT; - - return lock_mode; -} - -/* - * The following routine will lock n inodes in exclusive mode. - * We assume the caller calls us with the inodes in i_ino order. - * - * We need to detect deadlock where an inode that we lock - * is in the AIL and we start waiting for another inode that is locked - * by a thread in a long running transaction (such as truncate). This can - * result in deadlock since the long running trans might need to wait - * for the inode we just locked in order to push the tail and free space - * in the log. - */ -void -xfs_lock_inodes( - xfs_inode_t **ips, - int inodes, - uint lock_mode) -{ - int attempts = 0, i, j, try_lock; - xfs_log_item_t *lp; - - ASSERT(ips && (inodes >= 2)); /* we need at least two */ - - try_lock = 0; - i = 0; - -again: - for (; i < inodes; i++) { - ASSERT(ips[i]); - - if (i && (ips[i] == ips[i-1])) /* Already locked */ - continue; - - /* - * If try_lock is not set yet, make sure all locked inodes - * are not in the AIL. - * If any are, set try_lock to be used later. - */ - - if (!try_lock) { - for (j = (i - 1); j >= 0 && !try_lock; j--) { - lp = (xfs_log_item_t *)ips[j]->i_itemp; - if (lp && (lp->li_flags & XFS_LI_IN_AIL)) { - try_lock++; - } - } - } - - /* - * If any of the previous locks we have locked is in the AIL, - * we must TRY to get the second and subsequent locks. If - * we can't get any, we must release all we have - * and try again. - */ - - if (try_lock) { - /* try_lock must be 0 if i is 0. */ - /* - * try_lock means we have an inode locked - * that is in the AIL. - */ - ASSERT(i != 0); - if (!xfs_ilock_nowait(ips[i], xfs_lock_inumorder(lock_mode, i))) { - attempts++; - - /* - * Unlock all previous guys and try again. - * xfs_iunlock will try to push the tail - * if the inode is in the AIL. - */ - - for(j = i - 1; j >= 0; j--) { - - /* - * Check to see if we've already - * unlocked this one. - * Not the first one going back, - * and the inode ptr is the same. - */ - if ((j != (i - 1)) && ips[j] == - ips[j+1]) - continue; - - xfs_iunlock(ips[j], lock_mode); - } - - if ((attempts % 5) == 0) { - delay(1); /* Don't just spin the CPU */ -#ifdef DEBUG - xfs_lock_delays++; -#endif - } - i = 0; - try_lock = 0; - goto again; - } - } else { - xfs_ilock(ips[i], xfs_lock_inumorder(lock_mode, i)); - } - } - -#ifdef DEBUG - if (attempts) { - if (attempts < 5) xfs_small_retries++; - else if (attempts < 100) xfs_middle_retries++; - else xfs_lots_retries++; - } else { - xfs_locked_n++; - } -#endif -} - -/* - * xfs_lock_two_inodes() can only be used to lock one type of lock - * at a time - the iolock or the ilock, but not both at once. If - * we lock both at once, lockdep will report false positives saying - * we have violated locking orders. - */ -void -xfs_lock_two_inodes( - xfs_inode_t *ip0, - xfs_inode_t *ip1, - uint lock_mode) -{ - xfs_inode_t *temp; - int attempts = 0; - xfs_log_item_t *lp; - - if (lock_mode & (XFS_IOLOCK_SHARED|XFS_IOLOCK_EXCL)) - ASSERT((lock_mode & (XFS_ILOCK_SHARED|XFS_ILOCK_EXCL)) == 0); - ASSERT(ip0->i_ino != ip1->i_ino); - - if (ip0->i_ino > ip1->i_ino) { - temp = ip0; - ip0 = ip1; - ip1 = temp; - } - - again: - xfs_ilock(ip0, xfs_lock_inumorder(lock_mode, 0)); - - /* - * If the first lock we have locked is in the AIL, we must TRY to get - * the second lock. If we can't get it, we must release the first one - * and try again. - */ - lp = (xfs_log_item_t *)ip0->i_itemp; - if (lp && (lp->li_flags & XFS_LI_IN_AIL)) { - if (!xfs_ilock_nowait(ip1, xfs_lock_inumorder(lock_mode, 1))) { - xfs_iunlock(ip0, lock_mode); - if ((++attempts % 5) == 0) - delay(1); /* Don't just spin the CPU */ - goto again; - } - } else { - xfs_ilock(ip1, xfs_lock_inumorder(lock_mode, 1)); - } -} - -int -xfs_remove( - xfs_inode_t *dp, - struct xfs_name *name, - xfs_inode_t *ip) -{ - xfs_mount_t *mp = dp->i_mount; - xfs_trans_t *tp = NULL; - int is_dir = S_ISDIR(ip->i_d.di_mode); - int error = 0; - xfs_bmap_free_t free_list; - xfs_fsblock_t first_block; - int cancel_flags; - int committed; - int link_zero; - uint resblks; - uint log_count; - - trace_xfs_remove(dp, name); - - if (XFS_FORCED_SHUTDOWN(mp)) - return XFS_ERROR(EIO); - - error = xfs_qm_dqattach(dp, 0); - if (error) - goto std_return; - - error = xfs_qm_dqattach(ip, 0); - if (error) - goto std_return; - - if (is_dir) { - tp = xfs_trans_alloc(mp, XFS_TRANS_RMDIR); - log_count = XFS_DEFAULT_LOG_COUNT; - } else { - tp = xfs_trans_alloc(mp, XFS_TRANS_REMOVE); - log_count = XFS_REMOVE_LOG_COUNT; - } - cancel_flags = XFS_TRANS_RELEASE_LOG_RES; - - /* - * We try to get the real space reservation first, - * allowing for directory btree deletion(s) implying - * possible bmap insert(s). If we can't get the space - * reservation then we use 0 instead, and avoid the bmap - * btree insert(s) in the directory code by, if the bmap - * insert tries to happen, instead trimming the LAST - * block from the directory. - */ - resblks = XFS_REMOVE_SPACE_RES(mp); - error = xfs_trans_reserve(tp, resblks, XFS_REMOVE_LOG_RES(mp), 0, - XFS_TRANS_PERM_LOG_RES, log_count); - if (error == ENOSPC) { - resblks = 0; - error = xfs_trans_reserve(tp, 0, XFS_REMOVE_LOG_RES(mp), 0, - XFS_TRANS_PERM_LOG_RES, log_count); - } - if (error) { - ASSERT(error != ENOSPC); - cancel_flags = 0; - goto out_trans_cancel; - } - - xfs_lock_two_inodes(dp, ip, XFS_ILOCK_EXCL); - - xfs_trans_ijoin(tp, dp, XFS_ILOCK_EXCL); - xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL); - - /* - * If we're removing a directory perform some additional validation. - */ - if (is_dir) { - ASSERT(ip->i_d.di_nlink >= 2); - if (ip->i_d.di_nlink != 2) { - error = XFS_ERROR(ENOTEMPTY); - goto out_trans_cancel; - } - if (!xfs_dir_isempty(ip)) { - error = XFS_ERROR(ENOTEMPTY); - goto out_trans_cancel; - } - } - - xfs_bmap_init(&free_list, &first_block); - error = xfs_dir_removename(tp, dp, name, ip->i_ino, - &first_block, &free_list, resblks); - if (error) { - ASSERT(error != ENOENT); - goto out_bmap_cancel; - } - xfs_trans_ichgtime(tp, dp, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG); - - if (is_dir) { - /* - * Drop the link from ip's "..". - */ - error = xfs_droplink(tp, dp); - if (error) - goto out_bmap_cancel; - - /* - * Drop the "." link from ip to self. - */ - error = xfs_droplink(tp, ip); - if (error) - goto out_bmap_cancel; - } else { - /* - * When removing a non-directory we need to log the parent - * inode here. For a directory this is done implicitly - * by the xfs_droplink call for the ".." entry. - */ - xfs_trans_log_inode(tp, dp, XFS_ILOG_CORE); - } - - /* - * Drop the link from dp to ip. - */ - error = xfs_droplink(tp, ip); - if (error) - goto out_bmap_cancel; - - /* - * Determine if this is the last link while - * we are in the transaction. - */ - link_zero = (ip->i_d.di_nlink == 0); - - /* - * If this is a synchronous mount, make sure that the - * remove transaction goes to disk before returning to - * the user. - */ - if (mp->m_flags & (XFS_MOUNT_WSYNC|XFS_MOUNT_DIRSYNC)) - xfs_trans_set_sync(tp); - - error = xfs_bmap_finish(&tp, &free_list, &committed); - if (error) - goto out_bmap_cancel; - - error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES); - if (error) - goto std_return; - - /* - * If we are using filestreams, kill the stream association. - * If the file is still open it may get a new one but that - * will get killed on last close in xfs_close() so we don't - * have to worry about that. - */ - if (!is_dir && link_zero && xfs_inode_is_filestream(ip)) - xfs_filestream_deassociate(ip); - - return 0; - - out_bmap_cancel: - xfs_bmap_cancel(&free_list); - cancel_flags |= XFS_TRANS_ABORT; - out_trans_cancel: - xfs_trans_cancel(tp, cancel_flags); - std_return: - return error; -} - -int -xfs_link( - xfs_inode_t *tdp, - xfs_inode_t *sip, - struct xfs_name *target_name) -{ - xfs_mount_t *mp = tdp->i_mount; - xfs_trans_t *tp; - int error; - xfs_bmap_free_t free_list; - xfs_fsblock_t first_block; - int cancel_flags; - int committed; - int resblks; - - trace_xfs_link(tdp, target_name); - - ASSERT(!S_ISDIR(sip->i_d.di_mode)); - - if (XFS_FORCED_SHUTDOWN(mp)) - return XFS_ERROR(EIO); - - error = xfs_qm_dqattach(sip, 0); - if (error) - goto std_return; - - error = xfs_qm_dqattach(tdp, 0); - if (error) - goto std_return; - - tp = xfs_trans_alloc(mp, XFS_TRANS_LINK); - cancel_flags = XFS_TRANS_RELEASE_LOG_RES; - resblks = XFS_LINK_SPACE_RES(mp, target_name->len); - error = xfs_trans_reserve(tp, resblks, XFS_LINK_LOG_RES(mp), 0, - XFS_TRANS_PERM_LOG_RES, XFS_LINK_LOG_COUNT); - if (error == ENOSPC) { - resblks = 0; - error = xfs_trans_reserve(tp, 0, XFS_LINK_LOG_RES(mp), 0, - XFS_TRANS_PERM_LOG_RES, XFS_LINK_LOG_COUNT); - } - if (error) { - cancel_flags = 0; - goto error_return; - } - - xfs_lock_two_inodes(sip, tdp, XFS_ILOCK_EXCL); - - xfs_trans_ijoin(tp, sip, XFS_ILOCK_EXCL); - xfs_trans_ijoin(tp, tdp, XFS_ILOCK_EXCL); - - /* - * If we are using project inheritance, we only allow hard link - * creation in our tree when the project IDs are the same; else - * the tree quota mechanism could be circumvented. - */ - if (unlikely((tdp->i_d.di_flags & XFS_DIFLAG_PROJINHERIT) && - (xfs_get_projid(tdp) != xfs_get_projid(sip)))) { - error = XFS_ERROR(EXDEV); - goto error_return; - } - - error = xfs_dir_canenter(tp, tdp, target_name, resblks); - if (error) - goto error_return; - - xfs_bmap_init(&free_list, &first_block); - - error = xfs_dir_createname(tp, tdp, target_name, sip->i_ino, - &first_block, &free_list, resblks); - if (error) - goto abort_return; - xfs_trans_ichgtime(tp, tdp, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG); - xfs_trans_log_inode(tp, tdp, XFS_ILOG_CORE); - - error = xfs_bumplink(tp, sip); - if (error) - goto abort_return; - - /* - * If this is a synchronous mount, make sure that the - * link transaction goes to disk before returning to - * the user. - */ - if (mp->m_flags & (XFS_MOUNT_WSYNC|XFS_MOUNT_DIRSYNC)) { - xfs_trans_set_sync(tp); - } - - error = xfs_bmap_finish (&tp, &free_list, &committed); - if (error) { - xfs_bmap_cancel(&free_list); - goto abort_return; - } - - return xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES); - - abort_return: - cancel_flags |= XFS_TRANS_ABORT; - error_return: - xfs_trans_cancel(tp, cancel_flags); - std_return: - return error; -} - -int -xfs_set_dmattrs( - xfs_inode_t *ip, - u_int evmask, - u_int16_t state) -{ - xfs_mount_t *mp = ip->i_mount; - xfs_trans_t *tp; - int error; - - if (!capable(CAP_SYS_ADMIN)) - return XFS_ERROR(EPERM); - - if (XFS_FORCED_SHUTDOWN(mp)) - return XFS_ERROR(EIO); - - tp = xfs_trans_alloc(mp, XFS_TRANS_SET_DMATTRS); - error = xfs_trans_reserve(tp, 0, XFS_ICHANGE_LOG_RES (mp), 0, 0, 0); - if (error) { - xfs_trans_cancel(tp, 0); - return error; - } - xfs_ilock(ip, XFS_ILOCK_EXCL); - xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL); - - ip->i_d.di_dmevmask = evmask; - ip->i_d.di_dmstate = state; - - xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); - error = xfs_trans_commit(tp, 0); - - return error; -} - -/* - * xfs_alloc_file_space() - * This routine allocates disk space for the given file. - * - * If alloc_type == 0, this request is for an ALLOCSP type - * request which will change the file size. In this case, no - * DMAPI event will be generated by the call. A TRUNCATE event - * will be generated later by xfs_setattr. - * - * If alloc_type != 0, this request is for a RESVSP type - * request, and a DMAPI DM_EVENT_WRITE will be generated if the - * lower block boundary byte address is less than the file's - * length. - * - * RETURNS: - * 0 on success - * errno on error - * - */ -STATIC int -xfs_alloc_file_space( - xfs_inode_t *ip, - xfs_off_t offset, - xfs_off_t len, - int alloc_type, - int attr_flags) -{ - xfs_mount_t *mp = ip->i_mount; - xfs_off_t count; - xfs_filblks_t allocated_fsb; - xfs_filblks_t allocatesize_fsb; - xfs_extlen_t extsz, temp; - xfs_fileoff_t startoffset_fsb; - xfs_fsblock_t firstfsb; - int nimaps; - int quota_flag; - int rt; - xfs_trans_t *tp; - xfs_bmbt_irec_t imaps[1], *imapp; - xfs_bmap_free_t free_list; - uint qblocks, resblks, resrtextents; - int committed; - int error; - - trace_xfs_alloc_file_space(ip); - - if (XFS_FORCED_SHUTDOWN(mp)) - return XFS_ERROR(EIO); - - error = xfs_qm_dqattach(ip, 0); - if (error) - return error; - - if (len <= 0) - return XFS_ERROR(EINVAL); - - rt = XFS_IS_REALTIME_INODE(ip); - extsz = xfs_get_extsz_hint(ip); - - count = len; - imapp = &imaps[0]; - nimaps = 1; - startoffset_fsb = XFS_B_TO_FSBT(mp, offset); - allocatesize_fsb = XFS_B_TO_FSB(mp, count); - - /* - * Allocate file space until done or until there is an error - */ - while (allocatesize_fsb && !error) { - xfs_fileoff_t s, e; - - /* - * Determine space reservations for data/realtime. - */ - if (unlikely(extsz)) { - s = startoffset_fsb; - do_div(s, extsz); - s *= extsz; - e = startoffset_fsb + allocatesize_fsb; - if ((temp = do_mod(startoffset_fsb, extsz))) - e += temp; - if ((temp = do_mod(e, extsz))) - e += extsz - temp; - } else { - s = 0; - e = allocatesize_fsb; - } - - /* - * The transaction reservation is limited to a 32-bit block - * count, hence we need to limit the number of blocks we are - * trying to reserve to avoid an overflow. We can't allocate - * more than @nimaps extents, and an extent is limited on disk - * to MAXEXTLEN (21 bits), so use that to enforce the limit. - */ - resblks = min_t(xfs_fileoff_t, (e - s), (MAXEXTLEN * nimaps)); - if (unlikely(rt)) { - resrtextents = qblocks = resblks; - resrtextents /= mp->m_sb.sb_rextsize; - resblks = XFS_DIOSTRAT_SPACE_RES(mp, 0); - quota_flag = XFS_QMOPT_RES_RTBLKS; - } else { - resrtextents = 0; - resblks = qblocks = XFS_DIOSTRAT_SPACE_RES(mp, resblks); - quota_flag = XFS_QMOPT_RES_REGBLKS; - } - - /* - * Allocate and setup the transaction. - */ - tp = xfs_trans_alloc(mp, XFS_TRANS_DIOSTRAT); - error = xfs_trans_reserve(tp, resblks, - XFS_WRITE_LOG_RES(mp), resrtextents, - XFS_TRANS_PERM_LOG_RES, - XFS_WRITE_LOG_COUNT); - /* - * Check for running out of space - */ - if (error) { - /* - * Free the transaction structure. - */ - ASSERT(error == ENOSPC || XFS_FORCED_SHUTDOWN(mp)); - xfs_trans_cancel(tp, 0); - break; - } - xfs_ilock(ip, XFS_ILOCK_EXCL); - error = xfs_trans_reserve_quota_nblks(tp, ip, qblocks, - 0, quota_flag); - if (error) - goto error1; - - xfs_trans_ijoin(tp, ip, 0); - - xfs_bmap_init(&free_list, &firstfsb); - error = xfs_bmapi_write(tp, ip, startoffset_fsb, - allocatesize_fsb, alloc_type, &firstfsb, - 0, imapp, &nimaps, &free_list); - if (error) { - goto error0; - } - - /* - * Complete the transaction - */ - error = xfs_bmap_finish(&tp, &free_list, &committed); - if (error) { - goto error0; - } - - error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES); - xfs_iunlock(ip, XFS_ILOCK_EXCL); - if (error) { - break; - } - - allocated_fsb = imapp->br_blockcount; - - if (nimaps == 0) { - error = XFS_ERROR(ENOSPC); - break; - } - - startoffset_fsb += allocated_fsb; - allocatesize_fsb -= allocated_fsb; - } - - return error; - -error0: /* Cancel bmap, unlock inode, unreserve quota blocks, cancel trans */ - xfs_bmap_cancel(&free_list); - xfs_trans_unreserve_quota_nblks(tp, ip, (long)qblocks, 0, quota_flag); - -error1: /* Just cancel transaction */ - xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES | XFS_TRANS_ABORT); - xfs_iunlock(ip, XFS_ILOCK_EXCL); - return error; -} - -/* - * Zero file bytes between startoff and endoff inclusive. - * The iolock is held exclusive and no blocks are buffered. - * - * This function is used by xfs_free_file_space() to zero - * partial blocks when the range to free is not block aligned. - * When unreserving space with boundaries that are not block - * aligned we round up the start and round down the end - * boundaries and then use this function to zero the parts of - * the blocks that got dropped during the rounding. - */ -STATIC int -xfs_zero_remaining_bytes( - xfs_inode_t *ip, - xfs_off_t startoff, - xfs_off_t endoff) -{ - xfs_bmbt_irec_t imap; - xfs_fileoff_t offset_fsb; - xfs_off_t lastoffset; - xfs_off_t offset; - xfs_buf_t *bp; - xfs_mount_t *mp = ip->i_mount; - int nimap; - int error = 0; - - /* - * Avoid doing I/O beyond eof - it's not necessary - * since nothing can read beyond eof. The space will - * be zeroed when the file is extended anyway. - */ - if (startoff >= XFS_ISIZE(ip)) - return 0; - - if (endoff > XFS_ISIZE(ip)) - endoff = XFS_ISIZE(ip); - - bp = xfs_buf_get_uncached(XFS_IS_REALTIME_INODE(ip) ? - mp->m_rtdev_targp : mp->m_ddev_targp, - BTOBB(mp->m_sb.sb_blocksize), 0); - if (!bp) - return XFS_ERROR(ENOMEM); - - xfs_buf_unlock(bp); - - for (offset = startoff; offset <= endoff; offset = lastoffset + 1) { - offset_fsb = XFS_B_TO_FSBT(mp, offset); - nimap = 1; - error = xfs_bmapi_read(ip, offset_fsb, 1, &imap, &nimap, 0); - if (error || nimap < 1) - break; - ASSERT(imap.br_blockcount >= 1); - ASSERT(imap.br_startoff == offset_fsb); - lastoffset = XFS_FSB_TO_B(mp, imap.br_startoff + 1) - 1; - if (lastoffset > endoff) - lastoffset = endoff; - if (imap.br_startblock == HOLESTARTBLOCK) - continue; - ASSERT(imap.br_startblock != DELAYSTARTBLOCK); - if (imap.br_state == XFS_EXT_UNWRITTEN) - continue; - XFS_BUF_UNDONE(bp); - XFS_BUF_UNWRITE(bp); - XFS_BUF_READ(bp); - XFS_BUF_SET_ADDR(bp, xfs_fsb_to_db(ip, imap.br_startblock)); - xfsbdstrat(mp, bp); - error = xfs_buf_iowait(bp); - if (error) { - xfs_buf_ioerror_alert(bp, - "xfs_zero_remaining_bytes(read)"); - break; - } - memset(bp->b_addr + - (offset - XFS_FSB_TO_B(mp, imap.br_startoff)), - 0, lastoffset - offset + 1); - XFS_BUF_UNDONE(bp); - XFS_BUF_UNREAD(bp); - XFS_BUF_WRITE(bp); - xfsbdstrat(mp, bp); - error = xfs_buf_iowait(bp); - if (error) { - xfs_buf_ioerror_alert(bp, - "xfs_zero_remaining_bytes(write)"); - break; - } - } - xfs_buf_free(bp); - return error; -} - -/* - * xfs_free_file_space() - * This routine frees disk space for the given file. - * - * This routine is only called by xfs_change_file_space - * for an UNRESVSP type call. - * - * RETURNS: - * 0 on success - * errno on error - * - */ -STATIC int -xfs_free_file_space( - xfs_inode_t *ip, - xfs_off_t offset, - xfs_off_t len, - int attr_flags) -{ - int committed; - int done; - xfs_fileoff_t endoffset_fsb; - int error; - xfs_fsblock_t firstfsb; - xfs_bmap_free_t free_list; - xfs_bmbt_irec_t imap; - xfs_off_t ioffset; - xfs_extlen_t mod=0; - xfs_mount_t *mp; - int nimap; - uint resblks; - xfs_off_t rounding; - int rt; - xfs_fileoff_t startoffset_fsb; - xfs_trans_t *tp; - int need_iolock = 1; - - mp = ip->i_mount; - - trace_xfs_free_file_space(ip); - - error = xfs_qm_dqattach(ip, 0); - if (error) - return error; - - error = 0; - if (len <= 0) /* if nothing being freed */ - return error; - rt = XFS_IS_REALTIME_INODE(ip); - startoffset_fsb = XFS_B_TO_FSB(mp, offset); - endoffset_fsb = XFS_B_TO_FSBT(mp, offset + len); - - if (attr_flags & XFS_ATTR_NOLOCK) - need_iolock = 0; - if (need_iolock) { - xfs_ilock(ip, XFS_IOLOCK_EXCL); - /* wait for the completion of any pending DIOs */ - inode_dio_wait(VFS_I(ip)); - } - - rounding = max_t(xfs_off_t, 1 << mp->m_sb.sb_blocklog, PAGE_CACHE_SIZE); - ioffset = offset & ~(rounding - 1); - error = -filemap_write_and_wait_range(VFS_I(ip)->i_mapping, - ioffset, -1); - if (error) - goto out_unlock_iolock; - truncate_pagecache_range(VFS_I(ip), ioffset, -1); - - /* - * Need to zero the stuff we're not freeing, on disk. - * If it's a realtime file & can't use unwritten extents then we - * actually need to zero the extent edges. Otherwise xfs_bunmapi - * will take care of it for us. - */ - if (rt && !xfs_sb_version_hasextflgbit(&mp->m_sb)) { - nimap = 1; - error = xfs_bmapi_read(ip, startoffset_fsb, 1, - &imap, &nimap, 0); - if (error) - goto out_unlock_iolock; - ASSERT(nimap == 0 || nimap == 1); - if (nimap && imap.br_startblock != HOLESTARTBLOCK) { - xfs_daddr_t block; - - ASSERT(imap.br_startblock != DELAYSTARTBLOCK); - block = imap.br_startblock; - mod = do_div(block, mp->m_sb.sb_rextsize); - if (mod) - startoffset_fsb += mp->m_sb.sb_rextsize - mod; - } - nimap = 1; - error = xfs_bmapi_read(ip, endoffset_fsb - 1, 1, - &imap, &nimap, 0); - if (error) - goto out_unlock_iolock; - ASSERT(nimap == 0 || nimap == 1); - if (nimap && imap.br_startblock != HOLESTARTBLOCK) { - ASSERT(imap.br_startblock != DELAYSTARTBLOCK); - mod++; - if (mod && (mod != mp->m_sb.sb_rextsize)) - endoffset_fsb -= mod; - } - } - if ((done = (endoffset_fsb <= startoffset_fsb))) - /* - * One contiguous piece to clear - */ - error = xfs_zero_remaining_bytes(ip, offset, offset + len - 1); - else { - /* - * Some full blocks, possibly two pieces to clear - */ - if (offset < XFS_FSB_TO_B(mp, startoffset_fsb)) - error = xfs_zero_remaining_bytes(ip, offset, - XFS_FSB_TO_B(mp, startoffset_fsb) - 1); - if (!error && - XFS_FSB_TO_B(mp, endoffset_fsb) < offset + len) - error = xfs_zero_remaining_bytes(ip, - XFS_FSB_TO_B(mp, endoffset_fsb), - offset + len - 1); - } - - /* - * free file space until done or until there is an error - */ - resblks = XFS_DIOSTRAT_SPACE_RES(mp, 0); - while (!error && !done) { - - /* - * allocate and setup the transaction. Allow this - * transaction to dip into the reserve blocks to ensure - * the freeing of the space succeeds at ENOSPC. - */ - tp = xfs_trans_alloc(mp, XFS_TRANS_DIOSTRAT); - tp->t_flags |= XFS_TRANS_RESERVE; - error = xfs_trans_reserve(tp, - resblks, - XFS_WRITE_LOG_RES(mp), - 0, - XFS_TRANS_PERM_LOG_RES, - XFS_WRITE_LOG_COUNT); - - /* - * check for running out of space - */ - if (error) { - /* - * Free the transaction structure. - */ - ASSERT(error == ENOSPC || XFS_FORCED_SHUTDOWN(mp)); - xfs_trans_cancel(tp, 0); - break; - } - xfs_ilock(ip, XFS_ILOCK_EXCL); - error = xfs_trans_reserve_quota(tp, mp, - ip->i_udquot, ip->i_gdquot, ip->i_pdquot, - resblks, 0, XFS_QMOPT_RES_REGBLKS); - if (error) - goto error1; - - xfs_trans_ijoin(tp, ip, 0); - - /* - * issue the bunmapi() call to free the blocks - */ - xfs_bmap_init(&free_list, &firstfsb); - error = xfs_bunmapi(tp, ip, startoffset_fsb, - endoffset_fsb - startoffset_fsb, - 0, 2, &firstfsb, &free_list, &done); - if (error) { - goto error0; - } - - /* - * complete the transaction - */ - error = xfs_bmap_finish(&tp, &free_list, &committed); - if (error) { - goto error0; - } - - error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES); - xfs_iunlock(ip, XFS_ILOCK_EXCL); - } - - out_unlock_iolock: - if (need_iolock) - xfs_iunlock(ip, XFS_IOLOCK_EXCL); - return error; - - error0: - xfs_bmap_cancel(&free_list); - error1: - xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES | XFS_TRANS_ABORT); - xfs_iunlock(ip, need_iolock ? (XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL) : - XFS_ILOCK_EXCL); - return error; -} - - -STATIC int -xfs_zero_file_space( - struct xfs_inode *ip, - xfs_off_t offset, - xfs_off_t len, - int attr_flags) -{ - struct xfs_mount *mp = ip->i_mount; - uint granularity; - xfs_off_t start_boundary; - xfs_off_t end_boundary; - int error; - - granularity = max_t(uint, 1 << mp->m_sb.sb_blocklog, PAGE_CACHE_SIZE); - - /* - * Round the range of extents we are going to convert inwards. If the - * offset is aligned, then it doesn't get changed so we zero from the - * start of the block offset points to. - */ - start_boundary = round_up(offset, granularity); - end_boundary = round_down(offset + len, granularity); - - ASSERT(start_boundary >= offset); - ASSERT(end_boundary <= offset + len); - - if (!(attr_flags & XFS_ATTR_NOLOCK)) - xfs_ilock(ip, XFS_IOLOCK_EXCL); - - if (start_boundary < end_boundary - 1) { - /* punch out the page cache over the conversion range */ - truncate_pagecache_range(VFS_I(ip), start_boundary, - end_boundary - 1); - /* convert the blocks */ - error = xfs_alloc_file_space(ip, start_boundary, - end_boundary - start_boundary - 1, - XFS_BMAPI_PREALLOC | XFS_BMAPI_CONVERT, - attr_flags); - if (error) - goto out_unlock; - - /* We've handled the interior of the range, now for the edges */ - if (start_boundary != offset) - error = xfs_iozero(ip, offset, start_boundary - offset); - if (error) - goto out_unlock; - - if (end_boundary != offset + len) - error = xfs_iozero(ip, end_boundary, - offset + len - end_boundary); - - } else { - /* - * It's either a sub-granularity range or the range spanned lies - * partially across two adjacent blocks. - */ - error = xfs_iozero(ip, offset, len); - } - -out_unlock: - if (!(attr_flags & XFS_ATTR_NOLOCK)) - xfs_iunlock(ip, XFS_IOLOCK_EXCL); - return error; - -} - -/* - * xfs_change_file_space() - * This routine allocates or frees disk space for the given file. - * The user specified parameters are checked for alignment and size - * limitations. - * - * RETURNS: - * 0 on success - * errno on error - * - */ -int -xfs_change_file_space( - xfs_inode_t *ip, - int cmd, - xfs_flock64_t *bf, - xfs_off_t offset, - int attr_flags) -{ - xfs_mount_t *mp = ip->i_mount; - int clrprealloc; - int error; - xfs_fsize_t fsize; - int setprealloc; - xfs_off_t startoffset; - xfs_trans_t *tp; - struct iattr iattr; - - if (!S_ISREG(ip->i_d.di_mode)) - return XFS_ERROR(EINVAL); - - switch (bf->l_whence) { - case 0: /*SEEK_SET*/ - break; - case 1: /*SEEK_CUR*/ - bf->l_start += offset; - break; - case 2: /*SEEK_END*/ - bf->l_start += XFS_ISIZE(ip); - break; - default: - return XFS_ERROR(EINVAL); - } - - /* - * length of <= 0 for resv/unresv/zero is invalid. length for - * alloc/free is ignored completely and we have no idea what userspace - * might have set it to, so set it to zero to allow range - * checks to pass. - */ - switch (cmd) { - case XFS_IOC_ZERO_RANGE: - case XFS_IOC_RESVSP: - case XFS_IOC_RESVSP64: - case XFS_IOC_UNRESVSP: - case XFS_IOC_UNRESVSP64: - if (bf->l_len <= 0) - return XFS_ERROR(EINVAL); - break; - default: - bf->l_len = 0; - break; - } - - if (bf->l_start < 0 || - bf->l_start > mp->m_super->s_maxbytes || - bf->l_start + bf->l_len < 0 || - bf->l_start + bf->l_len >= mp->m_super->s_maxbytes) - return XFS_ERROR(EINVAL); - - bf->l_whence = 0; - - startoffset = bf->l_start; - fsize = XFS_ISIZE(ip); - - setprealloc = clrprealloc = 0; - switch (cmd) { - case XFS_IOC_ZERO_RANGE: - error = xfs_zero_file_space(ip, startoffset, bf->l_len, - attr_flags); - if (error) - return error; - setprealloc = 1; - break; - - case XFS_IOC_RESVSP: - case XFS_IOC_RESVSP64: - error = xfs_alloc_file_space(ip, startoffset, bf->l_len, - XFS_BMAPI_PREALLOC, attr_flags); - if (error) - return error; - setprealloc = 1; - break; - - case XFS_IOC_UNRESVSP: - case XFS_IOC_UNRESVSP64: - if ((error = xfs_free_file_space(ip, startoffset, bf->l_len, - attr_flags))) - return error; - break; - - case XFS_IOC_ALLOCSP: - case XFS_IOC_ALLOCSP64: - case XFS_IOC_FREESP: - case XFS_IOC_FREESP64: - /* - * These operations actually do IO when extending the file, but - * the allocation is done seperately to the zeroing that is - * done. This set of operations need to be serialised against - * other IO operations, such as truncate and buffered IO. We - * need to take the IOLOCK here to serialise the allocation and - * zeroing IO to prevent other IOLOCK holders (e.g. getbmap, - * truncate, direct IO) from racing against the transient - * allocated but not written state we can have here. - */ - xfs_ilock(ip, XFS_IOLOCK_EXCL); - if (startoffset > fsize) { - error = xfs_alloc_file_space(ip, fsize, - startoffset - fsize, 0, - attr_flags | XFS_ATTR_NOLOCK); - if (error) { - xfs_iunlock(ip, XFS_IOLOCK_EXCL); - break; - } - } - - iattr.ia_valid = ATTR_SIZE; - iattr.ia_size = startoffset; - - error = xfs_setattr_size(ip, &iattr, - attr_flags | XFS_ATTR_NOLOCK); - xfs_iunlock(ip, XFS_IOLOCK_EXCL); - - if (error) - return error; - - clrprealloc = 1; - break; - - default: - ASSERT(0); - return XFS_ERROR(EINVAL); - } - - /* - * update the inode timestamp, mode, and prealloc flag bits - */ - tp = xfs_trans_alloc(mp, XFS_TRANS_WRITEID); - - if ((error = xfs_trans_reserve(tp, 0, XFS_WRITEID_LOG_RES(mp), - 0, 0, 0))) { - /* ASSERT(0); */ - xfs_trans_cancel(tp, 0); - return error; - } - - xfs_ilock(ip, XFS_ILOCK_EXCL); - xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL); - - if ((attr_flags & XFS_ATTR_DMI) == 0) { - ip->i_d.di_mode &= ~S_ISUID; - - /* - * Note that we don't have to worry about mandatory - * file locking being disabled here because we only - * clear the S_ISGID bit if the Group execute bit is - * on, but if it was on then mandatory locking wouldn't - * have been enabled. - */ - if (ip->i_d.di_mode & S_IXGRP) - ip->i_d.di_mode &= ~S_ISGID; - - xfs_trans_ichgtime(tp, ip, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG); - } - if (setprealloc) - ip->i_d.di_flags |= XFS_DIFLAG_PREALLOC; - else if (clrprealloc) - ip->i_d.di_flags &= ~XFS_DIFLAG_PREALLOC; - - xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); - if (attr_flags & XFS_ATTR_SYNC) - xfs_trans_set_sync(tp); - return xfs_trans_commit(tp, 0); -} diff --git a/fs/xfs/xfs_vnodeops.h b/fs/xfs/xfs_vnodeops.h deleted file mode 100644 index 38c67c3..0000000 --- a/fs/xfs/xfs_vnodeops.h +++ /dev/null @@ -1,55 +0,0 @@ -#ifndef _XFS_VNODEOPS_H -#define _XFS_VNODEOPS_H 1 - -struct attrlist_cursor_kern; -struct file; -struct iattr; -struct inode; -struct iovec; -struct kiocb; -struct pipe_inode_info; -struct uio; -struct xfs_inode; - - -int xfs_setattr_nonsize(struct xfs_inode *ip, struct iattr *vap, int flags); -int xfs_setattr_size(struct xfs_inode *ip, struct iattr *vap, int flags); -#define XFS_ATTR_DMI 0x01 /* invocation from a DMI function */ -#define XFS_ATTR_NONBLOCK 0x02 /* return EAGAIN if operation would block */ -#define XFS_ATTR_NOLOCK 0x04 /* Don't grab any conflicting locks */ -#define XFS_ATTR_NOACL 0x08 /* Don't call xfs_acl_chmod */ -#define XFS_ATTR_SYNC 0x10 /* synchronous operation required */ - -int xfs_readlink(struct xfs_inode *ip, char *link); -int xfs_release(struct xfs_inode *ip); -int xfs_inactive(struct xfs_inode *ip); -int xfs_lookup(struct xfs_inode *dp, struct xfs_name *name, - struct xfs_inode **ipp, struct xfs_name *ci_name); -int xfs_create(struct xfs_inode *dp, struct xfs_name *name, umode_t mode, - xfs_dev_t rdev, struct xfs_inode **ipp); -int xfs_remove(struct xfs_inode *dp, struct xfs_name *name, - struct xfs_inode *ip); -int xfs_link(struct xfs_inode *tdp, struct xfs_inode *sip, - struct xfs_name *target_name); -int xfs_readdir(struct xfs_inode *dp, struct dir_context *ctx, size_t bufsize); -int xfs_symlink(struct xfs_inode *dp, struct xfs_name *link_name, - const char *target_path, umode_t mode, struct xfs_inode **ipp); -int xfs_set_dmattrs(struct xfs_inode *ip, u_int evmask, u_int16_t state); -int xfs_change_file_space(struct xfs_inode *ip, int cmd, - xfs_flock64_t *bf, xfs_off_t offset, int attr_flags); -int xfs_rename(struct xfs_inode *src_dp, struct xfs_name *src_name, - struct xfs_inode *src_ip, struct xfs_inode *target_dp, - struct xfs_name *target_name, struct xfs_inode *target_ip); -int xfs_attr_get(struct xfs_inode *ip, const unsigned char *name, - unsigned char *value, int *valuelenp, int flags); -int xfs_attr_set(struct xfs_inode *dp, const unsigned char *name, - unsigned char *value, int valuelen, int flags); -int xfs_attr_remove(struct xfs_inode *dp, const unsigned char *name, int flags); -int xfs_attr_list(struct xfs_inode *dp, char *buffer, int bufsize, - int flags, struct attrlist_cursor_kern *cursor); - -int xfs_iozero(struct xfs_inode *, loff_t, size_t); -int xfs_zero_eof(struct xfs_inode *, xfs_off_t, xfs_fsize_t); -int xfs_free_eofblocks(struct xfs_mount *, struct xfs_inode *, bool); - -#endif /* _XFS_VNODEOPS_H */ diff --git a/fs/xfs/xfs_xattr.c b/fs/xfs/xfs_xattr.c index 87d3e03..e01f35e 100644 --- a/fs/xfs/xfs_xattr.c +++ b/fs/xfs/xfs_xattr.c @@ -17,13 +17,13 @@ */ #include "xfs.h" +#include "xfs_log_format.h" #include "xfs_da_btree.h" #include "xfs_bmap_btree.h" #include "xfs_inode.h" #include "xfs_attr.h" #include "xfs_attr_leaf.h" #include "xfs_acl.h" -#include "xfs_vnodeops.h" #include <linux/posix_acl_xattr.h> #include <linux/xattr.h> |