diff options
author | Jens Axboe <jens.axboe@oracle.com> | 2009-05-22 18:25:34 (GMT) |
---|---|---|
committer | Jens Axboe <jens.axboe@oracle.com> | 2009-05-22 18:25:34 (GMT) |
commit | e4b636366c00738b9609cda307014d71b1225b7f (patch) | |
tree | 760b67b3624eda62e943e48ce93635c30a5b47bf /fs | |
parent | b9ed7252d219c1c663944bf03846eabb515dbe75 (diff) | |
parent | 279e677faa775ad16e75c32e1bf4a37f8158bc61 (diff) | |
download | linux-fsl-qoriq-e4b636366c00738b9609cda307014d71b1225b7f.tar.xz |
Merge branch 'master' into for-2.6.31
Conflicts:
drivers/block/hd.c
drivers/block/mg_disk.c
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Diffstat (limited to 'fs')
114 files changed, 2011 insertions, 2728 deletions
diff --git a/fs/9p/vfs_super.c b/fs/9p/vfs_super.c index 5f8ab8a..ab5547f 100644 --- a/fs/9p/vfs_super.c +++ b/fs/9p/vfs_super.c @@ -37,6 +37,7 @@ #include <linux/mount.h> #include <linux/idr.h> #include <linux/sched.h> +#include <linux/smp_lock.h> #include <net/9p/9p.h> #include <net/9p/client.h> @@ -155,6 +156,7 @@ static int v9fs_get_sb(struct file_system_type *fs_type, int flags, root = d_alloc_root(inode); if (!root) { + iput(inode); retval = -ENOMEM; goto release_sb; } @@ -173,10 +175,7 @@ P9_DPRINTK(P9_DEBUG_VFS, " simple set mount, return 0\n"); return 0; release_sb: - if (sb) { - up_write(&sb->s_umount); - deactivate_super(sb); - } + deactivate_locked_super(sb); free_stat: kfree(st); @@ -230,9 +229,12 @@ static int v9fs_show_options(struct seq_file *m, struct vfsmount *mnt) static void v9fs_umount_begin(struct super_block *sb) { - struct v9fs_session_info *v9ses = sb->s_fs_info; + struct v9fs_session_info *v9ses; + lock_kernel(); + v9ses = sb->s_fs_info; v9fs_session_cancel(v9ses); + unlock_kernel(); } static const struct super_operations v9fs_super_ops = { diff --git a/fs/affs/super.c b/fs/affs/super.c index 5ce695e..63f5183 100644 --- a/fs/affs/super.c +++ b/fs/affs/super.c @@ -507,8 +507,7 @@ affs_remount(struct super_block *sb, int *flags, char *data) kfree(new_opts); return -EINVAL; } - kfree(sb->s_options); - sb->s_options = new_opts; + replace_mount_options(sb, new_opts); sbi->s_flags = mount_flags; sbi->s_mode = mode; diff --git a/fs/afs/super.c b/fs/afs/super.c index aee239a..76828e5 100644 --- a/fs/afs/super.c +++ b/fs/afs/super.c @@ -405,21 +405,20 @@ static int afs_get_sb(struct file_system_type *fs_type, sb->s_flags = flags; ret = afs_fill_super(sb, ¶ms); if (ret < 0) { - up_write(&sb->s_umount); - deactivate_super(sb); + deactivate_locked_super(sb); goto error; } - sb->s_options = new_opts; + save_mount_options(sb, new_opts); sb->s_flags |= MS_ACTIVE; } else { _debug("reuse"); - kfree(new_opts); ASSERTCMP(sb->s_flags, &, MS_ACTIVE); } simple_set_mnt(mnt, sb); afs_put_volume(params.volume); afs_put_cell(params.cell); + kfree(new_opts); _leave(" = 0 [%p]", sb); return 0; diff --git a/fs/autofs4/expire.c b/fs/autofs4/expire.c index 75f7dda..3077d8f 100644 --- a/fs/autofs4/expire.c +++ b/fs/autofs4/expire.c @@ -70,8 +70,10 @@ static int autofs4_mount_busy(struct vfsmount *mnt, struct dentry *dentry) * Otherwise it's an offset mount and we need to check * if we can umount its mount, if there is one. */ - if (!d_mountpoint(dentry)) + if (!d_mountpoint(dentry)) { + status = 0; goto done; + } } /* Update the expiry counter if fs is busy */ diff --git a/fs/binfmt_elf_fdpic.c b/fs/binfmt_elf_fdpic.c index 70cfc4b..fdb66fa 100644 --- a/fs/binfmt_elf_fdpic.c +++ b/fs/binfmt_elf_fdpic.c @@ -1388,7 +1388,7 @@ static void fill_prstatus(struct elf_prstatus *prstatus, prstatus->pr_sigpend = p->pending.signal.sig[0]; prstatus->pr_sighold = p->blocked.sig[0]; prstatus->pr_pid = task_pid_vnr(p); - prstatus->pr_ppid = task_pid_vnr(p->parent); + prstatus->pr_ppid = task_pid_vnr(p->real_parent); prstatus->pr_pgrp = task_pgrp_vnr(p); prstatus->pr_sid = task_session_vnr(p); if (thread_group_leader(p)) { @@ -1433,7 +1433,7 @@ static int fill_psinfo(struct elf_prpsinfo *psinfo, struct task_struct *p, psinfo->pr_psargs[len] = 0; psinfo->pr_pid = task_pid_vnr(p); - psinfo->pr_ppid = task_pid_vnr(p->parent); + psinfo->pr_ppid = task_pid_vnr(p->real_parent); psinfo->pr_pgrp = task_pgrp_vnr(p); psinfo->pr_sid = task_session_vnr(p); @@ -817,6 +817,9 @@ struct bio *bio_copy_user_iov(struct request_queue *q, len += iov[i].iov_len; } + if (offset) + nr_pages++; + bmd = bio_alloc_map_data(nr_pages, iov_count, gfp_mask); if (!bmd) return ERR_PTR(-ENOMEM); diff --git a/fs/btrfs/Makefile b/fs/btrfs/Makefile index 9adf5e4..9421284 100644 --- a/fs/btrfs/Makefile +++ b/fs/btrfs/Makefile @@ -1,25 +1,10 @@ -ifneq ($(KERNELRELEASE),) -# kbuild part of makefile obj-$(CONFIG_BTRFS_FS) := btrfs.o -btrfs-y := super.o ctree.o extent-tree.o print-tree.o root-tree.o dir-item.o \ + +btrfs-y += super.o ctree.o extent-tree.o print-tree.o root-tree.o dir-item.o \ file-item.o inode-item.o inode-map.o disk-io.o \ transaction.o inode.o file.o tree-defrag.o \ extent_map.o sysfs.o struct-funcs.o xattr.o ordered-data.o \ extent_io.o volumes.o async-thread.o ioctl.o locking.o orphan.o \ ref-cache.o export.o tree-log.o acl.o free-space-cache.o zlib.o \ compression.o delayed-ref.o -else - -# Normal Makefile - -KERNELDIR := /lib/modules/`uname -r`/build -all: - $(MAKE) -C $(KERNELDIR) M=`pwd` CONFIG_BTRFS_FS=m modules - -modules_install: - $(MAKE) -C $(KERNELDIR) M=`pwd` modules_install -clean: - $(MAKE) -C $(KERNELDIR) M=`pwd` clean - -endif diff --git a/fs/btrfs/acl.c b/fs/btrfs/acl.c index 7fdd184..cbba000 100644 --- a/fs/btrfs/acl.c +++ b/fs/btrfs/acl.c @@ -60,15 +60,20 @@ static struct posix_acl *btrfs_get_acl(struct inode *inode, int type) return ERR_PTR(-EINVAL); } + /* Handle the cached NULL acl case without locking */ + acl = ACCESS_ONCE(*p_acl); + if (!acl) + return acl; + spin_lock(&inode->i_lock); - if (*p_acl != BTRFS_ACL_NOT_CACHED) - acl = posix_acl_dup(*p_acl); + acl = *p_acl; + if (acl != BTRFS_ACL_NOT_CACHED) + acl = posix_acl_dup(acl); spin_unlock(&inode->i_lock); - if (acl) + if (acl != BTRFS_ACL_NOT_CACHED) return acl; - size = __btrfs_getxattr(inode, name, "", 0); if (size > 0) { value = kzalloc(size, GFP_NOFS); @@ -80,9 +85,12 @@ static struct posix_acl *btrfs_get_acl(struct inode *inode, int type) btrfs_update_cached_acl(inode, p_acl, acl); } kfree(value); - } else if (size == -ENOENT) { + } else if (size == -ENOENT || size == -ENODATA || size == 0) { + /* FIXME, who returns -ENOENT? I think nobody */ acl = NULL; btrfs_update_cached_acl(inode, p_acl, acl); + } else { + acl = ERR_PTR(-EIO); } return acl; diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c index a99f1c2..fedf8b9 100644 --- a/fs/btrfs/ctree.c +++ b/fs/btrfs/ctree.c @@ -1469,6 +1469,7 @@ read_block_for_search(struct btrfs_trans_handle *trans, u32 blocksize; struct extent_buffer *b = *eb_ret; struct extent_buffer *tmp; + int ret; blocknr = btrfs_node_blockptr(b, slot); gen = btrfs_node_ptr_generation(b, slot); @@ -1476,6 +1477,10 @@ read_block_for_search(struct btrfs_trans_handle *trans, tmp = btrfs_find_tree_block(root, blocknr, blocksize); if (tmp && btrfs_buffer_uptodate(tmp, gen)) { + /* + * we found an up to date block without sleeping, return + * right away + */ *eb_ret = tmp; return 0; } @@ -1483,7 +1488,9 @@ read_block_for_search(struct btrfs_trans_handle *trans, /* * reduce lock contention at high levels * of the btree by dropping locks before - * we read. + * we read. Don't release the lock on the current + * level because we need to walk this node to figure + * out which blocks to read. */ btrfs_unlock_up_safe(p, level + 1); btrfs_set_path_blocking(p); @@ -1494,10 +1501,21 @@ read_block_for_search(struct btrfs_trans_handle *trans, reada_for_search(root, p, level, slot, key->objectid); btrfs_release_path(NULL, p); + + ret = -EAGAIN; tmp = read_tree_block(root, blocknr, blocksize, gen); - if (tmp) + if (tmp) { + /* + * If the read above didn't mark this buffer up to date, + * it will never end up being up to date. Set ret to EIO now + * and give up so that our caller doesn't loop forever + * on our EAGAINs. + */ + if (!btrfs_buffer_uptodate(tmp, 0)) + ret = -EIO; free_extent_buffer(tmp); - return -EAGAIN; + } + return ret; } /* @@ -1696,6 +1714,9 @@ cow_done: if (ret == -EAGAIN) goto again; + if (ret == -EIO) + goto done; + if (!p->skip_locking) { int lret; @@ -1738,6 +1759,8 @@ done: */ if (!p->leave_spinning) btrfs_set_path_blocking(p); + if (ret < 0) + btrfs_release_path(root, p); return ret; } @@ -4212,6 +4235,11 @@ again: if (ret == -EAGAIN) goto again; + if (ret < 0) { + btrfs_release_path(root, path); + goto done; + } + if (!path->skip_locking) { ret = btrfs_try_spin_lock(next); if (!ret) { @@ -4246,6 +4274,11 @@ again: if (ret == -EAGAIN) goto again; + if (ret < 0) { + btrfs_release_path(root, path); + goto done; + } + if (!path->skip_locking) { btrfs_assert_tree_locked(path->nodes[level]); ret = btrfs_try_spin_lock(next); diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index ad96495..4414a5d 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -881,6 +881,9 @@ struct btrfs_fs_info { u64 metadata_alloc_profile; u64 system_alloc_profile; + unsigned data_chunk_allocations; + unsigned metadata_ratio; + void *bdev_holder; }; @@ -2174,7 +2177,8 @@ int btrfs_check_file(struct btrfs_root *root, struct inode *inode); extern struct file_operations btrfs_file_operations; int btrfs_drop_extents(struct btrfs_trans_handle *trans, struct btrfs_root *root, struct inode *inode, - u64 start, u64 end, u64 inline_limit, u64 *hint_block); + u64 start, u64 end, u64 locked_end, + u64 inline_limit, u64 *hint_block); int btrfs_mark_extent_written(struct btrfs_trans_handle *trans, struct btrfs_root *root, struct inode *inode, u64 start, u64 end); diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index a6b8374..4b0ea0b 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -232,10 +232,14 @@ static int csum_tree_block(struct btrfs_root *root, struct extent_buffer *buf, memcpy(&found, result, csum_size); read_extent_buffer(buf, &val, 0, csum_size); - printk(KERN_INFO "btrfs: %s checksum verify failed " - "on %llu wanted %X found %X level %d\n", - root->fs_info->sb->s_id, - buf->start, val, found, btrfs_header_level(buf)); + if (printk_ratelimit()) { + printk(KERN_INFO "btrfs: %s checksum verify " + "failed on %llu wanted %X found %X " + "level %d\n", + root->fs_info->sb->s_id, + (unsigned long long)buf->start, val, found, + btrfs_header_level(buf)); + } if (result != (char *)&inline_result) kfree(result); return 1; @@ -268,10 +272,13 @@ static int verify_parent_transid(struct extent_io_tree *io_tree, ret = 0; goto out; } - printk("parent transid verify failed on %llu wanted %llu found %llu\n", - (unsigned long long)eb->start, - (unsigned long long)parent_transid, - (unsigned long long)btrfs_header_generation(eb)); + if (printk_ratelimit()) { + printk("parent transid verify failed on %llu wanted %llu " + "found %llu\n", + (unsigned long long)eb->start, + (unsigned long long)parent_transid, + (unsigned long long)btrfs_header_generation(eb)); + } ret = 1; clear_extent_buffer_uptodate(io_tree, eb); out: @@ -415,9 +422,12 @@ static int btree_readpage_end_io_hook(struct page *page, u64 start, u64 end, found_start = btrfs_header_bytenr(eb); if (found_start != start) { - printk(KERN_INFO "btrfs bad tree block start %llu %llu\n", - (unsigned long long)found_start, - (unsigned long long)eb->start); + if (printk_ratelimit()) { + printk(KERN_INFO "btrfs bad tree block start " + "%llu %llu\n", + (unsigned long long)found_start, + (unsigned long long)eb->start); + } ret = -EIO; goto err; } @@ -429,8 +439,10 @@ static int btree_readpage_end_io_hook(struct page *page, u64 start, u64 end, goto err; } if (check_tree_block_fsid(root, eb)) { - printk(KERN_INFO "btrfs bad fsid on block %llu\n", - (unsigned long long)eb->start); + if (printk_ratelimit()) { + printk(KERN_INFO "btrfs bad fsid on block %llu\n", + (unsigned long long)eb->start); + } ret = -EIO; goto err; } @@ -584,18 +596,7 @@ int btrfs_wq_submit_bio(struct btrfs_fs_info *fs_info, struct inode *inode, btrfs_set_work_high_prio(&async->work); btrfs_queue_worker(&fs_info->workers, &async->work); -#if 0 - int limit = btrfs_async_submit_limit(fs_info); - if (atomic_read(&fs_info->nr_async_submits) > limit) { - wait_event_timeout(fs_info->async_submit_wait, - (atomic_read(&fs_info->nr_async_submits) < limit), - HZ/10); - wait_event_timeout(fs_info->async_submit_wait, - (atomic_read(&fs_info->nr_async_bios) < limit), - HZ/10); - } -#endif while (atomic_read(&fs_info->async_submit_draining) && atomic_read(&fs_info->nr_async_submits)) { wait_event(fs_info->async_submit_wait, @@ -770,27 +771,6 @@ static void btree_invalidatepage(struct page *page, unsigned long offset) } } -#if 0 -static int btree_writepage(struct page *page, struct writeback_control *wbc) -{ - struct buffer_head *bh; - struct btrfs_root *root = BTRFS_I(page->mapping->host)->root; - struct buffer_head *head; - if (!page_has_buffers(page)) { - create_empty_buffers(page, root->fs_info->sb->s_blocksize, - (1 << BH_Dirty)|(1 << BH_Uptodate)); - } - head = page_buffers(page); - bh = head; - do { - if (buffer_dirty(bh)) - csum_tree_block(root, bh, 0); - bh = bh->b_this_page; - } while (bh != head); - return block_write_full_page(page, btree_get_block, wbc); -} -#endif - static struct address_space_operations btree_aops = { .readpage = btree_readpage, .writepage = btree_writepage, @@ -868,8 +848,6 @@ struct extent_buffer *read_tree_block(struct btrfs_root *root, u64 bytenr, if (ret == 0) set_bit(EXTENT_BUFFER_UPTODATE, &buf->bflags); - else - WARN_ON(1); return buf; } @@ -1278,11 +1256,7 @@ static int btrfs_congested_fn(void *congested_data, int bdi_bits) int ret = 0; struct btrfs_device *device; struct backing_dev_info *bdi; -#if 0 - if ((bdi_bits & (1 << BDI_write_congested)) && - btrfs_congested_async(info, 0)) - return 1; -#endif + list_for_each_entry(device, &info->fs_devices->devices, dev_list) { if (!device->bdev) continue; @@ -1604,6 +1578,7 @@ struct btrfs_root *open_ctree(struct super_block *sb, fs_info->btree_inode = new_inode(sb); fs_info->btree_inode->i_ino = 1; fs_info->btree_inode->i_nlink = 1; + fs_info->metadata_ratio = 8; fs_info->thread_pool_size = min_t(unsigned long, num_online_cpus() + 2, 8); @@ -1694,7 +1669,7 @@ struct btrfs_root *open_ctree(struct super_block *sb, if (features) { printk(KERN_ERR "BTRFS: couldn't mount because of " "unsupported optional features (%Lx).\n", - features); + (unsigned long long)features); err = -EINVAL; goto fail_iput; } @@ -1704,7 +1679,7 @@ struct btrfs_root *open_ctree(struct super_block *sb, if (!(sb->s_flags & MS_RDONLY) && features) { printk(KERN_ERR "BTRFS: couldn't mount RDWR because of " "unsupported option features (%Lx).\n", - features); + (unsigned long long)features); err = -EINVAL; goto fail_iput; } @@ -2296,7 +2271,7 @@ int close_ctree(struct btrfs_root *root) if (fs_info->delalloc_bytes) { printk(KERN_INFO "btrfs: at unmount delalloc count %llu\n", - fs_info->delalloc_bytes); + (unsigned long long)fs_info->delalloc_bytes); } if (fs_info->total_ref_cache_size) { printk(KERN_INFO "btrfs: at umount reference cache size %llu\n", @@ -2333,16 +2308,6 @@ int close_ctree(struct btrfs_root *root) btrfs_stop_workers(&fs_info->endio_write_workers); btrfs_stop_workers(&fs_info->submit_workers); -#if 0 - while (!list_empty(&fs_info->hashers)) { - struct btrfs_hasher *hasher; - hasher = list_entry(fs_info->hashers.next, struct btrfs_hasher, - hashers); - list_del(&hasher->hashers); - crypto_free_hash(&fs_info->hash_tfm); - kfree(hasher); - } -#endif btrfs_close_devices(fs_info->fs_devices); btrfs_mapping_tree_free(&fs_info->mapping_tree); diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 178df4c..3e2c7c7 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -312,7 +312,7 @@ btrfs_lookup_first_block_group(struct btrfs_fs_info *info, u64 bytenr) } /* - * return the block group that contains teh given bytenr + * return the block group that contains the given bytenr */ struct btrfs_block_group_cache *btrfs_lookup_block_group( struct btrfs_fs_info *info, @@ -1844,10 +1844,14 @@ again: printk(KERN_ERR "no space left, need %llu, %llu delalloc bytes" ", %llu bytes_used, %llu bytes_reserved, " "%llu bytes_pinned, %llu bytes_readonly, %llu may use" - "%llu total\n", bytes, data_sinfo->bytes_delalloc, - data_sinfo->bytes_used, data_sinfo->bytes_reserved, - data_sinfo->bytes_pinned, data_sinfo->bytes_readonly, - data_sinfo->bytes_may_use, data_sinfo->total_bytes); + "%llu total\n", (unsigned long long)bytes, + (unsigned long long)data_sinfo->bytes_delalloc, + (unsigned long long)data_sinfo->bytes_used, + (unsigned long long)data_sinfo->bytes_reserved, + (unsigned long long)data_sinfo->bytes_pinned, + (unsigned long long)data_sinfo->bytes_readonly, + (unsigned long long)data_sinfo->bytes_may_use, + (unsigned long long)data_sinfo->total_bytes); return -ENOSPC; } data_sinfo->bytes_may_use += bytes; @@ -1918,15 +1922,29 @@ void btrfs_delalloc_free_space(struct btrfs_root *root, struct inode *inode, spin_unlock(&info->lock); } +static void force_metadata_allocation(struct btrfs_fs_info *info) +{ + struct list_head *head = &info->space_info; + struct btrfs_space_info *found; + + rcu_read_lock(); + list_for_each_entry_rcu(found, head, list) { + if (found->flags & BTRFS_BLOCK_GROUP_METADATA) + found->force_alloc = 1; + } + rcu_read_unlock(); +} + static int do_chunk_alloc(struct btrfs_trans_handle *trans, struct btrfs_root *extent_root, u64 alloc_bytes, u64 flags, int force) { struct btrfs_space_info *space_info; + struct btrfs_fs_info *fs_info = extent_root->fs_info; u64 thresh; int ret = 0; - mutex_lock(&extent_root->fs_info->chunk_mutex); + mutex_lock(&fs_info->chunk_mutex); flags = btrfs_reduce_alloc_profile(extent_root, flags); @@ -1958,6 +1976,18 @@ static int do_chunk_alloc(struct btrfs_trans_handle *trans, } spin_unlock(&space_info->lock); + /* + * if we're doing a data chunk, go ahead and make sure that + * we keep a reasonable number of metadata chunks allocated in the + * FS as well. + */ + if (flags & BTRFS_BLOCK_GROUP_DATA) { + fs_info->data_chunk_allocations++; + if (!(fs_info->data_chunk_allocations % + fs_info->metadata_ratio)) + force_metadata_allocation(fs_info); + } + ret = btrfs_alloc_chunk(trans, extent_root, flags); if (ret) space_info->full = 1; @@ -2798,9 +2828,12 @@ static void dump_space_info(struct btrfs_space_info *info, u64 bytes) info->bytes_pinned - info->bytes_reserved), (info->full) ? "" : "not "); printk(KERN_INFO "space_info total=%llu, pinned=%llu, delalloc=%llu," - " may_use=%llu, used=%llu\n", info->total_bytes, - info->bytes_pinned, info->bytes_delalloc, info->bytes_may_use, - info->bytes_used); + " may_use=%llu, used=%llu\n", + (unsigned long long)info->total_bytes, + (unsigned long long)info->bytes_pinned, + (unsigned long long)info->bytes_delalloc, + (unsigned long long)info->bytes_may_use, + (unsigned long long)info->bytes_used); down_read(&info->groups_sem); list_for_each_entry(cache, &info->block_groups, list) { diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index 05a1c42..fe9eb99 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -17,12 +17,6 @@ #include "ctree.h" #include "btrfs_inode.h" -/* temporary define until extent_map moves out of btrfs */ -struct kmem_cache *btrfs_cache_create(const char *name, size_t size, - unsigned long extra_flags, - void (*ctor)(void *, struct kmem_cache *, - unsigned long)); - static struct kmem_cache *extent_state_cache; static struct kmem_cache *extent_buffer_cache; @@ -58,15 +52,15 @@ struct extent_page_data { int __init extent_io_init(void) { - extent_state_cache = btrfs_cache_create("extent_state", - sizeof(struct extent_state), 0, - NULL); + extent_state_cache = kmem_cache_create("extent_state", + sizeof(struct extent_state), 0, + SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD, NULL); if (!extent_state_cache) return -ENOMEM; - extent_buffer_cache = btrfs_cache_create("extent_buffers", - sizeof(struct extent_buffer), 0, - NULL); + extent_buffer_cache = kmem_cache_create("extent_buffers", + sizeof(struct extent_buffer), 0, + SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD, NULL); if (!extent_buffer_cache) goto free_state_cache; return 0; @@ -1407,69 +1401,6 @@ out: return total_bytes; } -#if 0 -/* - * helper function to lock both pages and extents in the tree. - * pages must be locked first. - */ -static int lock_range(struct extent_io_tree *tree, u64 start, u64 end) -{ - unsigned long index = start >> PAGE_CACHE_SHIFT; - unsigned long end_index = end >> PAGE_CACHE_SHIFT; - struct page *page; - int err; - - while (index <= end_index) { - page = grab_cache_page(tree->mapping, index); - if (!page) { - err = -ENOMEM; - goto failed; - } - if (IS_ERR(page)) { - err = PTR_ERR(page); - goto failed; - } - index++; - } - lock_extent(tree, start, end, GFP_NOFS); - return 0; - -failed: - /* - * we failed above in getting the page at 'index', so we undo here - * up to but not including the page at 'index' - */ - end_index = index; - index = start >> PAGE_CACHE_SHIFT; - while (index < end_index) { - page = find_get_page(tree->mapping, index); - unlock_page(page); - page_cache_release(page); - index++; - } - return err; -} - -/* - * helper function to unlock both pages and extents in the tree. - */ -static int unlock_range(struct extent_io_tree *tree, u64 start, u64 end) -{ - unsigned long index = start >> PAGE_CACHE_SHIFT; - unsigned long end_index = end >> PAGE_CACHE_SHIFT; - struct page *page; - - while (index <= end_index) { - page = find_get_page(tree->mapping, index); - unlock_page(page); - page_cache_release(page); - index++; - } - unlock_extent(tree, start, end, GFP_NOFS); - return 0; -} -#endif - /* * set the private field for a given byte offset in the tree. If there isn't * an extent_state there already, this does nothing. diff --git a/fs/btrfs/extent_map.c b/fs/btrfs/extent_map.c index b187917..30c9365 100644 --- a/fs/btrfs/extent_map.c +++ b/fs/btrfs/extent_map.c @@ -6,19 +6,14 @@ #include <linux/hardirq.h> #include "extent_map.h" -/* temporary define until extent_map moves out of btrfs */ -struct kmem_cache *btrfs_cache_create(const char *name, size_t size, - unsigned long extra_flags, - void (*ctor)(void *, struct kmem_cache *, - unsigned long)); static struct kmem_cache *extent_map_cache; int __init extent_map_init(void) { - extent_map_cache = btrfs_cache_create("extent_map", - sizeof(struct extent_map), 0, - NULL); + extent_map_cache = kmem_cache_create("extent_map", + sizeof(struct extent_map), 0, + SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD, NULL); if (!extent_map_cache) return -ENOMEM; return 0; @@ -43,7 +38,6 @@ void extent_map_tree_init(struct extent_map_tree *tree, gfp_t mask) tree->map.rb_node = NULL; spin_lock_init(&tree->lock); } -EXPORT_SYMBOL(extent_map_tree_init); /** * alloc_extent_map - allocate new extent map structure @@ -64,7 +58,6 @@ struct extent_map *alloc_extent_map(gfp_t mask) atomic_set(&em->refs, 1); return em; } -EXPORT_SYMBOL(alloc_extent_map); /** * free_extent_map - drop reference count of an extent_map @@ -83,7 +76,6 @@ void free_extent_map(struct extent_map *em) kmem_cache_free(extent_map_cache, em); } } -EXPORT_SYMBOL(free_extent_map); static struct rb_node *tree_insert(struct rb_root *root, u64 offset, struct rb_node *node) @@ -264,7 +256,6 @@ int add_extent_mapping(struct extent_map_tree *tree, out: return ret; } -EXPORT_SYMBOL(add_extent_mapping); /* simple helper to do math around the end of an extent, handling wrap */ static u64 range_end(u64 start, u64 len) @@ -326,7 +317,6 @@ found: out: return em; } -EXPORT_SYMBOL(lookup_extent_mapping); /** * remove_extent_mapping - removes an extent_map from the extent tree @@ -346,4 +336,3 @@ int remove_extent_mapping(struct extent_map_tree *tree, struct extent_map *em) em->in_tree = 0; return ret; } -EXPORT_SYMBOL(remove_extent_mapping); diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index 482f8db..1d51dc3 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -272,83 +272,6 @@ int btrfs_drop_extent_cache(struct inode *inode, u64 start, u64 end, return 0; } -int btrfs_check_file(struct btrfs_root *root, struct inode *inode) -{ - return 0; -#if 0 - struct btrfs_path *path; - struct btrfs_key found_key; - struct extent_buffer *leaf; - struct btrfs_file_extent_item *extent; - u64 last_offset = 0; - int nritems; - int slot; - int found_type; - int ret; - int err = 0; - u64 extent_end = 0; - - path = btrfs_alloc_path(); - ret = btrfs_lookup_file_extent(NULL, root, path, inode->i_ino, - last_offset, 0); - while (1) { - nritems = btrfs_header_nritems(path->nodes[0]); - if (path->slots[0] >= nritems) { - ret = btrfs_next_leaf(root, path); - if (ret) - goto out; - nritems = btrfs_header_nritems(path->nodes[0]); - } - slot = path->slots[0]; - leaf = path->nodes[0]; - btrfs_item_key_to_cpu(leaf, &found_key, slot); - if (found_key.objectid != inode->i_ino) - break; - if (found_key.type != BTRFS_EXTENT_DATA_KEY) - goto out; - - if (found_key.offset < last_offset) { - WARN_ON(1); - btrfs_print_leaf(root, leaf); - printk(KERN_ERR "inode %lu found offset %llu " - "expected %llu\n", inode->i_ino, - (unsigned long long)found_key.offset, - (unsigned long long)last_offset); - err = 1; - goto out; - } - extent = btrfs_item_ptr(leaf, slot, - struct btrfs_file_extent_item); - found_type = btrfs_file_extent_type(leaf, extent); - if (found_type == BTRFS_FILE_EXTENT_REG) { - extent_end = found_key.offset + - btrfs_file_extent_num_bytes(leaf, extent); - } else if (found_type == BTRFS_FILE_EXTENT_INLINE) { - struct btrfs_item *item; - item = btrfs_item_nr(leaf, slot); - extent_end = found_key.offset + - btrfs_file_extent_inline_len(leaf, extent); - extent_end = (extent_end + root->sectorsize - 1) & - ~((u64)root->sectorsize - 1); - } - last_offset = extent_end; - path->slots[0]++; - } - if (0 && last_offset < inode->i_size) { - WARN_ON(1); - btrfs_print_leaf(root, leaf); - printk(KERN_ERR "inode %lu found offset %llu size %llu\n", - inode->i_ino, (unsigned long long)last_offset, - (unsigned long long)inode->i_size); - err = 1; - - } -out: - btrfs_free_path(path); - return err; -#endif -} - /* * this is very complex, but the basic idea is to drop all extents * in the range start - end. hint_block is filled in with a block number @@ -363,15 +286,16 @@ out: */ noinline int btrfs_drop_extents(struct btrfs_trans_handle *trans, struct btrfs_root *root, struct inode *inode, - u64 start, u64 end, u64 inline_limit, u64 *hint_byte) + u64 start, u64 end, u64 locked_end, + u64 inline_limit, u64 *hint_byte) { u64 extent_end = 0; - u64 locked_end = end; u64 search_start = start; u64 leaf_start; u64 ram_bytes = 0; u64 orig_parent = 0; u64 disk_bytenr = 0; + u64 orig_locked_end = locked_end; u8 compression; u8 encryption; u16 other_encoding = 0; @@ -684,11 +608,10 @@ next_slot: } out: btrfs_free_path(path); - if (locked_end > end) { - unlock_extent(&BTRFS_I(inode)->io_tree, end, locked_end - 1, - GFP_NOFS); + if (locked_end > orig_locked_end) { + unlock_extent(&BTRFS_I(inode)->io_tree, orig_locked_end, + locked_end - 1, GFP_NOFS); } - btrfs_check_file(root, inode); return ret; } diff --git a/fs/btrfs/free-space-cache.c b/fs/btrfs/free-space-cache.c index 768b952..0bc9365 100644 --- a/fs/btrfs/free-space-cache.c +++ b/fs/btrfs/free-space-cache.c @@ -332,13 +332,17 @@ int btrfs_remove_free_space(struct btrfs_block_group_cache *block_group, printk(KERN_ERR "couldn't find space %llu to free\n", (unsigned long long)offset); printk(KERN_ERR "cached is %d, offset %llu bytes %llu\n", - block_group->cached, block_group->key.objectid, - block_group->key.offset); + block_group->cached, + (unsigned long long)block_group->key.objectid, + (unsigned long long)block_group->key.offset); btrfs_dump_free_space(block_group, bytes); } else if (info) { printk(KERN_ERR "hmm, found offset=%llu bytes=%llu, " "but wanted offset=%llu bytes=%llu\n", - info->offset, info->bytes, offset, bytes); + (unsigned long long)info->offset, + (unsigned long long)info->bytes, + (unsigned long long)offset, + (unsigned long long)bytes); } WARN_ON(1); } @@ -357,8 +361,9 @@ void btrfs_dump_free_space(struct btrfs_block_group_cache *block_group, info = rb_entry(n, struct btrfs_free_space, offset_index); if (info->bytes >= bytes) count++; - printk(KERN_ERR "entry offset %llu, bytes %llu\n", info->offset, - info->bytes); + printk(KERN_ERR "entry offset %llu, bytes %llu\n", + (unsigned long long)info->offset, + (unsigned long long)info->bytes); } printk(KERN_INFO "%d blocks of free space at or bigger than bytes is" "\n", count); diff --git a/fs/btrfs/inode-map.c b/fs/btrfs/inode-map.c index cc7334d..9abbced 100644 --- a/fs/btrfs/inode-map.c +++ b/fs/btrfs/inode-map.c @@ -79,7 +79,7 @@ int btrfs_find_free_objectid(struct btrfs_trans_handle *trans, } path = btrfs_alloc_path(); BUG_ON(!path); - search_start = max(search_start, BTRFS_FIRST_FREE_OBJECTID); + search_start = max(search_start, (u64)BTRFS_FIRST_FREE_OBJECTID); search_key.objectid = search_start; search_key.type = 0; search_key.offset = 0; diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 65219f6..1c8b019 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -70,7 +70,6 @@ static struct extent_io_ops btrfs_extent_io_ops; static struct kmem_cache *btrfs_inode_cachep; struct kmem_cache *btrfs_trans_handle_cachep; struct kmem_cache *btrfs_transaction_cachep; -struct kmem_cache *btrfs_bit_radix_cachep; struct kmem_cache *btrfs_path_cachep; #define S_SHIFT 12 @@ -234,7 +233,7 @@ static noinline int cow_file_range_inline(struct btrfs_trans_handle *trans, } ret = btrfs_drop_extents(trans, root, inode, start, - aligned_end, start, &hint_byte); + aligned_end, aligned_end, start, &hint_byte); BUG_ON(ret); if (isize > actual_end) @@ -1439,6 +1438,7 @@ static int insert_reserved_file_extent(struct btrfs_trans_handle *trans, struct inode *inode, u64 file_pos, u64 disk_bytenr, u64 disk_num_bytes, u64 num_bytes, u64 ram_bytes, + u64 locked_end, u8 compression, u8 encryption, u16 other_encoding, int extent_type) { @@ -1455,7 +1455,8 @@ static int insert_reserved_file_extent(struct btrfs_trans_handle *trans, path->leave_spinning = 1; ret = btrfs_drop_extents(trans, root, inode, file_pos, - file_pos + num_bytes, file_pos, &hint); + file_pos + num_bytes, locked_end, + file_pos, &hint); BUG_ON(ret); ins.objectid = inode->i_ino; @@ -1590,6 +1591,8 @@ static int btrfs_finish_ordered_io(struct inode *inode, u64 start, u64 end) ordered_extent->disk_len, ordered_extent->len, ordered_extent->len, + ordered_extent->file_offset + + ordered_extent->len, compressed, 0, 0, BTRFS_FILE_EXTENT_REG); BUG_ON(ret); @@ -1819,10 +1822,12 @@ good: return 0; zeroit: - printk(KERN_INFO "btrfs csum failed ino %lu off %llu csum %u " - "private %llu\n", page->mapping->host->i_ino, - (unsigned long long)start, csum, - (unsigned long long)private); + if (printk_ratelimit()) { + printk(KERN_INFO "btrfs csum failed ino %lu off %llu csum %u " + "private %llu\n", page->mapping->host->i_ino, + (unsigned long long)start, csum, + (unsigned long long)private); + } memset(kaddr + offset, 1, end - start + 1); flush_dcache_page(page); kunmap_atomic(kaddr, KM_USER0); @@ -2011,6 +2016,57 @@ void btrfs_orphan_cleanup(struct btrfs_root *root) } /* + * very simple check to peek ahead in the leaf looking for xattrs. If we + * don't find any xattrs, we know there can't be any acls. + * + * slot is the slot the inode is in, objectid is the objectid of the inode + */ +static noinline int acls_after_inode_item(struct extent_buffer *leaf, + int slot, u64 objectid) +{ + u32 nritems = btrfs_header_nritems(leaf); + struct btrfs_key found_key; + int scanned = 0; + + slot++; + while (slot < nritems) { + btrfs_item_key_to_cpu(leaf, &found_key, slot); + + /* we found a different objectid, there must not be acls */ + if (found_key.objectid != objectid) + return 0; + + /* we found an xattr, assume we've got an acl */ + if (found_key.type == BTRFS_XATTR_ITEM_KEY) + return 1; + + /* + * we found a key greater than an xattr key, there can't + * be any acls later on + */ + if (found_key.type > BTRFS_XATTR_ITEM_KEY) + return 0; + + slot++; + scanned++; + + /* + * it goes inode, inode backrefs, xattrs, extents, + * so if there are a ton of hard links to an inode there can + * be a lot of backrefs. Don't waste time searching too hard, + * this is just an optimization + */ + if (scanned >= 8) + break; + } + /* we hit the end of the leaf before we found an xattr or + * something larger than an xattr. We have to assume the inode + * has acls + */ + return 1; +} + +/* * read an inode from the btree into the in-memory inode */ void btrfs_read_locked_inode(struct inode *inode) @@ -2021,6 +2077,7 @@ void btrfs_read_locked_inode(struct inode *inode) struct btrfs_timespec *tspec; struct btrfs_root *root = BTRFS_I(inode)->root; struct btrfs_key location; + int maybe_acls; u64 alloc_group_block; u32 rdev; int ret; @@ -2067,6 +2124,16 @@ void btrfs_read_locked_inode(struct inode *inode) alloc_group_block = btrfs_inode_block_group(leaf, inode_item); + /* + * try to precache a NULL acl entry for files that don't have + * any xattrs or acls + */ + maybe_acls = acls_after_inode_item(leaf, path->slots[0], inode->i_ino); + if (!maybe_acls) { + BTRFS_I(inode)->i_acl = NULL; + BTRFS_I(inode)->i_default_acl = NULL; + } + BTRFS_I(inode)->block_group = btrfs_find_block_group(root, 0, alloc_group_block, 0); btrfs_free_path(path); @@ -2877,6 +2944,7 @@ int btrfs_cont_expand(struct inode *inode, loff_t size) err = btrfs_drop_extents(trans, root, inode, cur_offset, cur_offset + hole_size, + block_end, cur_offset, &hint_byte); if (err) break; @@ -3041,8 +3109,8 @@ static noinline void init_btrfs_i(struct inode *inode) { struct btrfs_inode *bi = BTRFS_I(inode); - bi->i_acl = NULL; - bi->i_default_acl = NULL; + bi->i_acl = BTRFS_ACL_NOT_CACHED; + bi->i_default_acl = BTRFS_ACL_NOT_CACHED; bi->generation = 0; bi->sequence = 0; @@ -3054,6 +3122,7 @@ static noinline void init_btrfs_i(struct inode *inode) bi->flags = 0; bi->index_cnt = (u64)-1; bi->last_unlink_trans = 0; + bi->ordered_data_close = 0; extent_map_tree_init(&BTRFS_I(inode)->extent_tree, GFP_NOFS); extent_io_tree_init(&BTRFS_I(inode)->io_tree, inode->i_mapping, GFP_NOFS); @@ -4227,7 +4296,6 @@ out: } if (err) { free_extent_map(em); - WARN_ON(1); return ERR_PTR(err); } return em; @@ -4634,47 +4702,36 @@ void btrfs_destroy_cachep(void) kmem_cache_destroy(btrfs_trans_handle_cachep); if (btrfs_transaction_cachep) kmem_cache_destroy(btrfs_transaction_cachep); - if (btrfs_bit_radix_cachep) - kmem_cache_destroy(btrfs_bit_radix_cachep); if (btrfs_path_cachep) kmem_cache_destroy(btrfs_path_cachep); } -struct kmem_cache *btrfs_cache_create(const char *name, size_t size, - unsigned long extra_flags, - void (*ctor)(void *)) -{ - return kmem_cache_create(name, size, 0, (SLAB_RECLAIM_ACCOUNT | - SLAB_MEM_SPREAD | extra_flags), ctor); -} - int btrfs_init_cachep(void) { - btrfs_inode_cachep = btrfs_cache_create("btrfs_inode_cache", - sizeof(struct btrfs_inode), - 0, init_once); + btrfs_inode_cachep = kmem_cache_create("btrfs_inode_cache", + sizeof(struct btrfs_inode), 0, + SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD, init_once); if (!btrfs_inode_cachep) goto fail; - btrfs_trans_handle_cachep = - btrfs_cache_create("btrfs_trans_handle_cache", - sizeof(struct btrfs_trans_handle), - 0, NULL); + + btrfs_trans_handle_cachep = kmem_cache_create("btrfs_trans_handle_cache", + sizeof(struct btrfs_trans_handle), 0, + SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD, NULL); if (!btrfs_trans_handle_cachep) goto fail; - btrfs_transaction_cachep = btrfs_cache_create("btrfs_transaction_cache", - sizeof(struct btrfs_transaction), - 0, NULL); + + btrfs_transaction_cachep = kmem_cache_create("btrfs_transaction_cache", + sizeof(struct btrfs_transaction), 0, + SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD, NULL); if (!btrfs_transaction_cachep) goto fail; - btrfs_path_cachep = btrfs_cache_create("btrfs_path_cache", - sizeof(struct btrfs_path), - 0, NULL); + + btrfs_path_cachep = kmem_cache_create("btrfs_path_cache", + sizeof(struct btrfs_path), 0, + SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD, NULL); if (!btrfs_path_cachep) goto fail; - btrfs_bit_radix_cachep = btrfs_cache_create("btrfs_radix", 256, - SLAB_DESTROY_BY_RCU, NULL); - if (!btrfs_bit_radix_cachep) - goto fail; + return 0; fail: btrfs_destroy_cachep(); @@ -4972,7 +5029,7 @@ out_fail: static int prealloc_file_range(struct btrfs_trans_handle *trans, struct inode *inode, u64 start, u64 end, - u64 alloc_hint, int mode) + u64 locked_end, u64 alloc_hint, int mode) { struct btrfs_root *root = BTRFS_I(inode)->root; struct btrfs_key ins; @@ -4993,7 +5050,8 @@ static int prealloc_file_range(struct btrfs_trans_handle *trans, ret = insert_reserved_file_extent(trans, inode, cur_offset, ins.objectid, ins.offset, ins.offset, - ins.offset, 0, 0, 0, + ins.offset, locked_end, + 0, 0, 0, BTRFS_FILE_EXTENT_PREALLOC); BUG_ON(ret); num_bytes -= ins.offset; @@ -5022,6 +5080,7 @@ static long btrfs_fallocate(struct inode *inode, int mode, u64 alloc_start; u64 alloc_end; u64 alloc_hint = 0; + u64 locked_end; u64 mask = BTRFS_I(inode)->root->sectorsize - 1; struct extent_map *em; struct btrfs_trans_handle *trans; @@ -5043,6 +5102,7 @@ static long btrfs_fallocate(struct inode *inode, int mode, goto out; } + locked_end = alloc_end - 1; while (1) { struct btrfs_ordered_extent *ordered; @@ -5055,8 +5115,8 @@ static long btrfs_fallocate(struct inode *inode, int mode, /* the extent lock is ordered inside the running * transaction */ - lock_extent(&BTRFS_I(inode)->io_tree, alloc_start, - alloc_end - 1, GFP_NOFS); + lock_extent(&BTRFS_I(inode)->io_tree, alloc_start, locked_end, + GFP_NOFS); ordered = btrfs_lookup_first_ordered_extent(inode, alloc_end - 1); if (ordered && @@ -5064,7 +5124,7 @@ static long btrfs_fallocate(struct inode *inode, int mode, ordered->file_offset < alloc_end) { btrfs_put_ordered_extent(ordered); unlock_extent(&BTRFS_I(inode)->io_tree, - alloc_start, alloc_end - 1, GFP_NOFS); + alloc_start, locked_end, GFP_NOFS); btrfs_end_transaction(trans, BTRFS_I(inode)->root); /* @@ -5089,7 +5149,8 @@ static long btrfs_fallocate(struct inode *inode, int mode, last_byte = (last_byte + mask) & ~mask; if (em->block_start == EXTENT_MAP_HOLE) { ret = prealloc_file_range(trans, inode, cur_offset, - last_byte, alloc_hint, mode); + last_byte, locked_end + 1, + alloc_hint, mode); if (ret < 0) { free_extent_map(em); break; @@ -5105,7 +5166,7 @@ static long btrfs_fallocate(struct inode *inode, int mode, break; } } - unlock_extent(&BTRFS_I(inode)->io_tree, alloc_start, alloc_end - 1, + unlock_extent(&BTRFS_I(inode)->io_tree, alloc_start, locked_end, GFP_NOFS); btrfs_end_transaction(trans, BTRFS_I(inode)->root); diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index 9f135e8..2624b53 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -437,10 +437,6 @@ out_unlock: return 0; } -/* - * Called inside transaction, so use GFP_NOFS - */ - static int btrfs_ioctl_resize(struct btrfs_root *root, void __user *arg) { u64 new_size; @@ -477,11 +473,13 @@ static int btrfs_ioctl_resize(struct btrfs_root *root, void __user *arg) *devstr = '\0'; devstr = vol_args->name; devid = simple_strtoull(devstr, &end, 10); - printk(KERN_INFO "resizing devid %llu\n", devid); + printk(KERN_INFO "resizing devid %llu\n", + (unsigned long long)devid); } device = btrfs_find_device(root, devid, NULL, NULL); if (!device) { - printk(KERN_INFO "resizer unable to find device %llu\n", devid); + printk(KERN_INFO "resizer unable to find device %llu\n", + (unsigned long long)devid); ret = -EINVAL; goto out_unlock; } @@ -805,7 +803,8 @@ static long btrfs_ioctl_clone(struct file *file, unsigned long srcfd, BUG_ON(!trans); /* punch hole in destination first */ - btrfs_drop_extents(trans, root, inode, off, off+len, 0, &hint_byte); + btrfs_drop_extents(trans, root, inode, off, off + len, + off + len, 0, &hint_byte); /* clone data */ key.objectid = src->i_ino; diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c index a7acfe6..2ff7cd2 100644 --- a/fs/btrfs/super.c +++ b/fs/btrfs/super.c @@ -68,7 +68,7 @@ enum { Opt_degraded, Opt_subvol, Opt_device, Opt_nodatasum, Opt_nodatacow, Opt_max_extent, Opt_max_inline, Opt_alloc_start, Opt_nobarrier, Opt_ssd, Opt_thread_pool, Opt_noacl, Opt_compress, Opt_notreelog, - Opt_flushoncommit, Opt_err, + Opt_ratio, Opt_flushoncommit, Opt_err, }; static match_table_t tokens = { @@ -87,6 +87,7 @@ static match_table_t tokens = { {Opt_noacl, "noacl"}, {Opt_notreelog, "notreelog"}, {Opt_flushoncommit, "flushoncommit"}, + {Opt_ratio, "metadata_ratio=%d"}, {Opt_err, NULL}, }; @@ -195,7 +196,7 @@ int btrfs_parse_options(struct btrfs_root *root, char *options) info->max_extent = max_t(u64, info->max_extent, root->sectorsize); printk(KERN_INFO "btrfs: max_extent at %llu\n", - info->max_extent); + (unsigned long long)info->max_extent); } break; case Opt_max_inline: @@ -210,7 +211,7 @@ int btrfs_parse_options(struct btrfs_root *root, char *options) root->sectorsize); } printk(KERN_INFO "btrfs: max_inline at %llu\n", - info->max_inline); + (unsigned long long)info->max_inline); } break; case Opt_alloc_start: @@ -220,7 +221,7 @@ int btrfs_parse_options(struct btrfs_root *root, char *options) kfree(num); printk(KERN_INFO "btrfs: allocations start at %llu\n", - info->alloc_start); + (unsigned long long)info->alloc_start); } break; case Opt_noacl: @@ -234,6 +235,15 @@ int btrfs_parse_options(struct btrfs_root *root, char *options) printk(KERN_INFO "btrfs: turning on flush-on-commit\n"); btrfs_set_opt(info->mount_opt, FLUSHONCOMMIT); break; + case Opt_ratio: + intarg = 0; + match_int(&args[0], &intarg); + if (intarg) { + info->metadata_ratio = intarg; + printk(KERN_INFO "btrfs: metadata ratio %d\n", + info->metadata_ratio); + } + break; default: break; } @@ -410,11 +420,14 @@ static int btrfs_show_options(struct seq_file *seq, struct vfsmount *vfs) if (btrfs_test_opt(root, NOBARRIER)) seq_puts(seq, ",nobarrier"); if (info->max_extent != (u64)-1) - seq_printf(seq, ",max_extent=%llu", info->max_extent); + seq_printf(seq, ",max_extent=%llu", + (unsigned long long)info->max_extent); if (info->max_inline != 8192 * 1024) - seq_printf(seq, ",max_inline=%llu", info->max_inline); + seq_printf(seq, ",max_inline=%llu", + (unsigned long long)info->max_inline); if (info->alloc_start != 0) - seq_printf(seq, ",alloc_start=%llu", info->alloc_start); + seq_printf(seq, ",alloc_start=%llu", + (unsigned long long)info->alloc_start); if (info->thread_pool_size != min_t(unsigned long, num_online_cpus() + 2, 8)) seq_printf(seq, ",thread_pool=%d", info->thread_pool_size); @@ -423,9 +436,9 @@ static int btrfs_show_options(struct seq_file *seq, struct vfsmount *vfs) if (btrfs_test_opt(root, SSD)) seq_puts(seq, ",ssd"); if (btrfs_test_opt(root, NOTREELOG)) - seq_puts(seq, ",no-treelog"); + seq_puts(seq, ",notreelog"); if (btrfs_test_opt(root, FLUSHONCOMMIT)) - seq_puts(seq, ",flush-on-commit"); + seq_puts(seq, ",flushoncommit"); if (!(root->fs_info->sb->s_flags & MS_POSIXACL)) seq_puts(seq, ",noacl"); return 0; @@ -489,8 +502,7 @@ static int btrfs_get_sb(struct file_system_type *fs_type, int flags, if (s->s_root) { if ((flags ^ s->s_flags) & MS_RDONLY) { - up_write(&s->s_umount); - deactivate_super(s); + deactivate_locked_super(s); error = -EBUSY; goto error_close_devices; } @@ -504,8 +516,7 @@ static int btrfs_get_sb(struct file_system_type *fs_type, int flags, error = btrfs_fill_super(s, fs_devices, data, flags & MS_SILENT ? 1 : 0); if (error) { - up_write(&s->s_umount); - deactivate_super(s); + deactivate_locked_super(s); goto error_free_subvol_name; } @@ -522,15 +533,13 @@ static int btrfs_get_sb(struct file_system_type *fs_type, int flags, mutex_unlock(&s->s_root->d_inode->i_mutex); if (IS_ERR(root)) { - up_write(&s->s_umount); - deactivate_super(s); + deactivate_locked_super(s); error = PTR_ERR(root); goto error_free_subvol_name; } if (!root->d_inode) { dput(root); - up_write(&s->s_umount); - deactivate_super(s); + deactivate_locked_super(s); error = -ENXIO; goto error_free_subvol_name; } diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c index 2869b33..01b1436 100644 --- a/fs/btrfs/transaction.c +++ b/fs/btrfs/transaction.c @@ -687,7 +687,13 @@ static noinline int wait_transaction_pre_flush(struct btrfs_fs_info *info) prepare_to_wait(&info->transaction_wait, &wait, TASK_UNINTERRUPTIBLE); mutex_unlock(&info->trans_mutex); + + atomic_dec(&info->throttles); + wake_up(&info->transaction_throttle); + schedule(); + + atomic_inc(&info->throttles); mutex_lock(&info->trans_mutex); finish_wait(&info->transaction_wait, &wait); } diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c index 25f20ea..db5e212 100644 --- a/fs/btrfs/tree-log.c +++ b/fs/btrfs/tree-log.c @@ -536,7 +536,7 @@ static noinline int replay_one_extent(struct btrfs_trans_handle *trans, saved_nbytes = inode_get_bytes(inode); /* drop any overlapping extents */ ret = btrfs_drop_extents(trans, root, inode, - start, extent_end, start, &alloc_hint); + start, extent_end, extent_end, start, &alloc_hint); BUG_ON(ret); if (found_type == BTRFS_FILE_EXTENT_REG || diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index e53835b..5f01dad 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -1543,7 +1543,7 @@ static noinline int btrfs_update_device(struct btrfs_trans_handle *trans, btrfs_set_device_io_align(leaf, dev_item, device->io_align); btrfs_set_device_io_width(leaf, dev_item, device->io_width); btrfs_set_device_sector_size(leaf, dev_item, device->sector_size); - btrfs_set_device_total_bytes(leaf, dev_item, device->total_bytes); + btrfs_set_device_total_bytes(leaf, dev_item, device->disk_total_bytes); btrfs_set_device_bytes_used(leaf, dev_item, device->bytes_used); btrfs_mark_buffer_dirty(leaf); @@ -1940,14 +1940,6 @@ int btrfs_shrink_device(struct btrfs_device *device, u64 new_size) device->total_bytes = new_size; if (device->writeable) device->fs_devices->total_rw_bytes -= diff; - ret = btrfs_update_device(trans, device); - if (ret) { - unlock_chunks(root); - btrfs_end_transaction(trans, root); - goto done; - } - WARN_ON(diff > old_total); - btrfs_set_super_total_bytes(super_copy, old_total - diff); unlock_chunks(root); btrfs_end_transaction(trans, root); @@ -1979,7 +1971,7 @@ int btrfs_shrink_device(struct btrfs_device *device, u64 new_size) length = btrfs_dev_extent_length(l, dev_extent); if (key.offset + length <= new_size) - goto done; + break; chunk_tree = btrfs_dev_extent_chunk_tree(l, dev_extent); chunk_objectid = btrfs_dev_extent_chunk_objectid(l, dev_extent); @@ -1992,6 +1984,26 @@ int btrfs_shrink_device(struct btrfs_device *device, u64 new_size) goto done; } + /* Shrinking succeeded, else we would be at "done". */ + trans = btrfs_start_transaction(root, 1); + if (!trans) { + ret = -ENOMEM; + goto done; + } + lock_chunks(root); + + device->disk_total_bytes = new_size; + /* Now btrfs_update_device() will change the on-disk size. */ + ret = btrfs_update_device(trans, device); + if (ret) { + unlock_chunks(root); + btrfs_end_transaction(trans, root); + goto done; + } + WARN_ON(diff > old_total); + btrfs_set_super_total_bytes(super_copy, old_total - diff); + unlock_chunks(root); + btrfs_end_transaction(trans, root); done: btrfs_free_path(path); return ret; @@ -3076,7 +3088,8 @@ static int fill_device_from_item(struct extent_buffer *leaf, unsigned long ptr; device->devid = btrfs_device_id(leaf, dev_item); - device->total_bytes = btrfs_device_total_bytes(leaf, dev_item); + device->disk_total_bytes = btrfs_device_total_bytes(leaf, dev_item); + device->total_bytes = device->disk_total_bytes; device->bytes_used = btrfs_device_bytes_used(leaf, dev_item); device->type = btrfs_device_type(leaf, dev_item); device->io_align = btrfs_device_io_align(leaf, dev_item); diff --git a/fs/btrfs/volumes.h b/fs/btrfs/volumes.h index 5836327..5c3ff6d 100644 --- a/fs/btrfs/volumes.h +++ b/fs/btrfs/volumes.h @@ -61,6 +61,9 @@ struct btrfs_device { /* size of the device */ u64 total_bytes; + /* size of the disk */ + u64 disk_total_bytes; + /* bytes used */ u64 bytes_used; diff --git a/fs/buffer.c b/fs/buffer.c index b3e5be7..aed2977 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -2397,7 +2397,8 @@ block_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf, if ((page->mapping != inode->i_mapping) || (page_offset(page) > size)) { /* page got truncated out from underneath us */ - goto out_unlock; + unlock_page(page); + goto out; } /* page is wholly or partially inside EOF */ @@ -2411,14 +2412,15 @@ block_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf, ret = block_commit_write(page, 0, end); if (unlikely(ret)) { + unlock_page(page); if (ret == -ENOMEM) ret = VM_FAULT_OOM; else /* -ENOSPC, -EIO, etc */ ret = VM_FAULT_SIGBUS; - } + } else + ret = VM_FAULT_LOCKED; -out_unlock: - unlock_page(page); +out: return ret; } diff --git a/fs/cifs/CHANGES b/fs/cifs/CHANGES index 9d1fb6e..f20c406 100644 --- a/fs/cifs/CHANGES +++ b/fs/cifs/CHANGES @@ -1,3 +1,16 @@ +Version 1.58 +------------ +Guard against buffer overruns in various UCS-2 to UTF-8 string conversions +when the UTF-8 string is composed of unusually long (more than 4 byte) converted +characters. Add support for mounting root of a share which redirects immediately +to DFS target. Convert string conversion functions from Unicode to more +accurately mark string length before allocating memory (which may help the +rare cases where a UTF-8 string is much larger than the UCS2 string that +we converted from). Fix endianness of the vcnum field used during +session setup to distinguish multiple mounts to same server from different +userids. Raw NTLMSSP fixed (it requires /proc/fs/cifs/experimental +flag to be set to 2, and mount must enable krb5 to turn on extended security). + Version 1.57 ------------ Improve support for multiple security contexts to the same server. We diff --git a/fs/cifs/README b/fs/cifs/README index 0743418..db208ddb 100644 --- a/fs/cifs/README +++ b/fs/cifs/README @@ -651,7 +651,15 @@ Experimental When set to 1 used to enable certain experimental signing turned on in case buffer was modified just before it was sent, also this flag will be used to use the new experimental directory change - notification code). + notification code). When set to 2 enables + an additional experimental feature, "raw ntlmssp" + session establishment support (which allows + specifying "sec=ntlmssp" on mount). The Linux cifs + module will use ntlmv2 authentication encapsulated + in "raw ntlmssp" (not using SPNEGO) when + "sec=ntlmssp" is specified on mount. + This support also requires building cifs with + the CONFIG_CIFS_EXPERIMENTAL configuration flag. These experimental features and tracing can be enabled by changing flags in /proc/fs/cifs (after the cifs module has been installed or built into the diff --git a/fs/cifs/cifs_dfs_ref.c b/fs/cifs/cifs_dfs_ref.c index 5fdbf8a..83d6275 100644 --- a/fs/cifs/cifs_dfs_ref.c +++ b/fs/cifs/cifs_dfs_ref.c @@ -340,28 +340,24 @@ cifs_dfs_follow_mountpoint(struct dentry *dentry, struct nameidata *nd) cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); for (i = 0; i < num_referrals; i++) { + int len; dump_referral(referrals+i); - /* connect to a storage node */ - if (referrals[i].flags & DFSREF_STORAGE_SERVER) { - int len; - len = strlen(referrals[i].node_name); - if (len < 2) { - cERROR(1, ("%s: Net Address path too short: %s", + /* connect to a node */ + len = strlen(referrals[i].node_name); + if (len < 2) { + cERROR(1, ("%s: Net Address path too short: %s", __func__, referrals[i].node_name)); - rc = -EINVAL; - goto out_err; - } - mnt = cifs_dfs_do_refmount(nd->path.mnt, - nd->path.dentry, - referrals + i); - cFYI(1, ("%s: cifs_dfs_do_refmount:%s , mnt:%p", - __func__, + rc = -EINVAL; + goto out_err; + } + mnt = cifs_dfs_do_refmount(nd->path.mnt, + nd->path.dentry, referrals + i); + cFYI(1, ("%s: cifs_dfs_do_refmount:%s , mnt:%p", __func__, referrals[i].node_name, mnt)); - /* complete mount procedure if we accured submount */ - if (!IS_ERR(mnt)) - break; - } + /* complete mount procedure if we accured submount */ + if (!IS_ERR(mnt)) + break; } /* we need it cause for() above could exit without valid submount */ diff --git a/fs/cifs/cifs_unicode.c b/fs/cifs/cifs_unicode.c index 7d75272..60e3c42 100644 --- a/fs/cifs/cifs_unicode.c +++ b/fs/cifs/cifs_unicode.c @@ -1,7 +1,7 @@ /* * fs/cifs/cifs_unicode.c * - * Copyright (c) International Business Machines Corp., 2000,2005 + * Copyright (c) International Business Machines Corp., 2000,2009 * Modified by Steve French (sfrench@us.ibm.com) * * This program is free software; you can redistribute it and/or modify @@ -26,31 +26,157 @@ #include "cifs_debug.h" /* - * NAME: cifs_strfromUCS() - * - * FUNCTION: Convert little-endian unicode string to character string + * cifs_ucs2_bytes - how long will a string be after conversion? + * @ucs - pointer to input string + * @maxbytes - don't go past this many bytes of input string + * @codepage - destination codepage * + * Walk a ucs2le string and return the number of bytes that the string will + * be after being converted to the given charset, not including any null + * termination required. Don't walk past maxbytes in the source buffer. */ int -cifs_strfromUCS_le(char *to, const __le16 *from, - int len, const struct nls_table *codepage) +cifs_ucs2_bytes(const __le16 *from, int maxbytes, + const struct nls_table *codepage) { int i; - int outlen = 0; + int charlen, outlen = 0; + int maxwords = maxbytes / 2; + char tmp[NLS_MAX_CHARSET_SIZE]; - for (i = 0; (i < len) && from[i]; i++) { - int charlen; - /* 2.4.0 kernel or greater */ - charlen = - codepage->uni2char(le16_to_cpu(from[i]), &to[outlen], - NLS_MAX_CHARSET_SIZE); - if (charlen > 0) { + for (i = 0; from[i] && i < maxwords; i++) { + charlen = codepage->uni2char(le16_to_cpu(from[i]), tmp, + NLS_MAX_CHARSET_SIZE); + if (charlen > 0) outlen += charlen; - } else { - to[outlen++] = '?'; + else + outlen++; + } + + return outlen; +} + +/* + * cifs_mapchar - convert a little-endian char to proper char in codepage + * @target - where converted character should be copied + * @src_char - 2 byte little-endian source character + * @cp - codepage to which character should be converted + * @mapchar - should character be mapped according to mapchars mount option? + * + * This function handles the conversion of a single character. It is the + * responsibility of the caller to ensure that the target buffer is large + * enough to hold the result of the conversion (at least NLS_MAX_CHARSET_SIZE). + */ +static int +cifs_mapchar(char *target, const __le16 src_char, const struct nls_table *cp, + bool mapchar) +{ + int len = 1; + + if (!mapchar) + goto cp_convert; + + /* + * BB: Cannot handle remapping UNI_SLASH until all the calls to + * build_path_from_dentry are modified, as they use slash as + * separator. + */ + switch (le16_to_cpu(src_char)) { + case UNI_COLON: + *target = ':'; + break; + case UNI_ASTERIK: + *target = '*'; + break; + case UNI_QUESTION: + *target = '?'; + break; + case UNI_PIPE: + *target = '|'; + break; + case UNI_GRTRTHAN: + *target = '>'; + break; + case UNI_LESSTHAN: + *target = '<'; + break; + default: + goto cp_convert; + } + +out: + return len; + +cp_convert: + len = cp->uni2char(le16_to_cpu(src_char), target, + NLS_MAX_CHARSET_SIZE); + if (len <= 0) { + *target = '?'; + len = 1; + } + goto out; +} + +/* + * cifs_from_ucs2 - convert utf16le string to local charset + * @to - destination buffer + * @from - source buffer + * @tolen - destination buffer size (in bytes) + * @fromlen - source buffer size (in bytes) + * @codepage - codepage to which characters should be converted + * @mapchar - should characters be remapped according to the mapchars option? + * + * Convert a little-endian ucs2le string (as sent by the server) to a string + * in the provided codepage. The tolen and fromlen parameters are to ensure + * that the code doesn't walk off of the end of the buffer (which is always + * a danger if the alignment of the source buffer is off). The destination + * string is always properly null terminated and fits in the destination + * buffer. Returns the length of the destination string in bytes (including + * null terminator). + * + * Note that some windows versions actually send multiword UTF-16 characters + * instead of straight UCS-2. The linux nls routines however aren't able to + * deal with those characters properly. In the event that we get some of + * those characters, they won't be translated properly. + */ +int +cifs_from_ucs2(char *to, const __le16 *from, int tolen, int fromlen, + const struct nls_table *codepage, bool mapchar) +{ + int i, charlen, safelen; + int outlen = 0; + int nullsize = nls_nullsize(codepage); + int fromwords = fromlen / 2; + char tmp[NLS_MAX_CHARSET_SIZE]; + + /* + * because the chars can be of varying widths, we need to take care + * not to overflow the destination buffer when we get close to the + * end of it. Until we get to this offset, we don't need to check + * for overflow however. + */ + safelen = tolen - (NLS_MAX_CHARSET_SIZE + nullsize); + + for (i = 0; i < fromwords && from[i]; i++) { + /* + * check to see if converting this character might make the + * conversion bleed into the null terminator + */ + if (outlen >= safelen) { + charlen = cifs_mapchar(tmp, from[i], codepage, mapchar); + if ((outlen + charlen) > (tolen - nullsize)) + break; } + + /* put converted char into 'to' buffer */ + charlen = cifs_mapchar(&to[outlen], from[i], codepage, mapchar); + outlen += charlen; } - to[outlen] = 0; + + /* properly null-terminate string */ + for (i = 0; i < nullsize; i++) + to[outlen++] = 0; + return outlen; } @@ -88,3 +214,41 @@ cifs_strtoUCS(__le16 *to, const char *from, int len, return i; } +/* + * cifs_strndup_from_ucs - copy a string from wire format to the local codepage + * @src - source string + * @maxlen - don't walk past this many bytes in the source string + * @is_unicode - is this a unicode string? + * @codepage - destination codepage + * + * Take a string given by the server, convert it to the local codepage and + * put it in a new buffer. Returns a pointer to the new string or NULL on + * error. + */ +char * +cifs_strndup_from_ucs(const char *src, const int maxlen, const bool is_unicode, + const struct nls_table *codepage) +{ + int len; + char *dst; + + if (is_unicode) { + len = cifs_ucs2_bytes((__le16 *) src, maxlen, codepage); + len += nls_nullsize(codepage); + dst = kmalloc(len, GFP_KERNEL); + if (!dst) + return NULL; + cifs_from_ucs2(dst, (__le16 *) src, len, maxlen, codepage, + false); + } else { + len = strnlen(src, maxlen); + len++; + dst = kmalloc(len, GFP_KERNEL); + if (!dst) + return NULL; + strlcpy(dst, src, len); + } + + return dst; +} + diff --git a/fs/cifs/cifs_unicode.h b/fs/cifs/cifs_unicode.h index 14eb9a2..6506382 100644 --- a/fs/cifs/cifs_unicode.h +++ b/fs/cifs/cifs_unicode.h @@ -5,7 +5,7 @@ * Convert a unicode character to upper or lower case using * compressed tables. * - * Copyright (c) International Business Machines Corp., 2000,2007 + * Copyright (c) International Business Machines Corp., 2000,2009 * * 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 @@ -37,6 +37,19 @@ #define UNIUPR_NOLOWER /* Example to not expand lower case tables */ +/* + * Windows maps these to the user defined 16 bit Unicode range since they are + * reserved symbols (along with \ and /), otherwise illegal to store + * in filenames in NTFS + */ +#define UNI_ASTERIK (__u16) ('*' + 0xF000) +#define UNI_QUESTION (__u16) ('?' + 0xF000) +#define UNI_COLON (__u16) (':' + 0xF000) +#define UNI_GRTRTHAN (__u16) ('>' + 0xF000) +#define UNI_LESSTHAN (__u16) ('<' + 0xF000) +#define UNI_PIPE (__u16) ('|' + 0xF000) +#define UNI_SLASH (__u16) ('\\' + 0xF000) + /* Just define what we want from uniupr.h. We don't want to define the tables * in each source file. */ @@ -59,8 +72,14 @@ extern struct UniCaseRange UniLowerRange[]; #endif /* UNIUPR_NOLOWER */ #ifdef __KERNEL__ -int cifs_strfromUCS_le(char *, const __le16 *, int, const struct nls_table *); +int cifs_from_ucs2(char *to, const __le16 *from, int tolen, int fromlen, + const struct nls_table *codepage, bool mapchar); +int cifs_ucs2_bytes(const __le16 *from, int maxbytes, + const struct nls_table *codepage); int cifs_strtoUCS(__le16 *, const char *, int, const struct nls_table *); +char *cifs_strndup_from_ucs(const char *src, const int maxlen, + const bool is_unicode, + const struct nls_table *codepage); #endif /* diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 0d6d8b5..5e6d358 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -35,6 +35,7 @@ #include <linux/delay.h> #include <linux/kthread.h> #include <linux/freezer.h> +#include <linux/smp_lock.h> #include "cifsfs.h" #include "cifspdu.h" #define DECLARE_GLOBALS_HERE @@ -530,6 +531,7 @@ static void cifs_umount_begin(struct super_block *sb) if (tcon == NULL) return; + lock_kernel(); read_lock(&cifs_tcp_ses_lock); if (tcon->tc_count == 1) tcon->tidStatus = CifsExiting; @@ -548,6 +550,7 @@ static void cifs_umount_begin(struct super_block *sb) } /* BB FIXME - finish add checks for tidStatus BB */ + unlock_kernel(); return; } @@ -599,8 +602,7 @@ cifs_get_sb(struct file_system_type *fs_type, rc = cifs_read_super(sb, data, dev_name, flags & MS_SILENT ? 1 : 0); if (rc) { - up_write(&sb->s_umount); - deactivate_super(sb); + deactivate_locked_super(sb); return rc; } sb->s_flags |= MS_ACTIVE; diff --git a/fs/cifs/cifsfs.h b/fs/cifs/cifsfs.h index 77e190d..051b71c 100644 --- a/fs/cifs/cifsfs.h +++ b/fs/cifs/cifsfs.h @@ -100,5 +100,5 @@ extern long cifs_ioctl(struct file *filep, unsigned int cmd, unsigned long arg); extern const struct export_operations cifs_export_ops; #endif /* EXPERIMENTAL */ -#define CIFS_VERSION "1.57" +#define CIFS_VERSION "1.58" #endif /* _CIFSFS_H */ diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h index df40ab6..a61ab77 100644 --- a/fs/cifs/cifsglob.h +++ b/fs/cifs/cifsglob.h @@ -82,8 +82,8 @@ enum securityEnum { LANMAN, /* Legacy LANMAN auth */ NTLM, /* Legacy NTLM012 auth with NTLM hash */ NTLMv2, /* Legacy NTLM auth with NTLMv2 hash */ - RawNTLMSSP, /* NTLMSSP without SPNEGO */ - NTLMSSP, /* NTLMSSP via SPNEGO */ + RawNTLMSSP, /* NTLMSSP without SPNEGO, NTLMv2 hash */ + NTLMSSP, /* NTLMSSP via SPNEGO, NTLMv2 hash */ Kerberos, /* Kerberos via SPNEGO */ MSKerberos, /* MS Kerberos via SPNEGO */ }; @@ -531,6 +531,7 @@ static inline void free_dfs_info_array(struct dfs_info3_param *param, #define CIFSSEC_MAY_PLNTXT 0 #endif /* weak passwords */ #define CIFSSEC_MAY_SEAL 0x00040 /* not supported yet */ +#define CIFSSEC_MAY_NTLMSSP 0x00080 /* raw ntlmssp with ntlmv2 */ #define CIFSSEC_MUST_SIGN 0x01001 /* note that only one of the following can be set so the @@ -543,22 +544,23 @@ require use of the stronger protocol */ #define CIFSSEC_MUST_LANMAN 0x10010 #define CIFSSEC_MUST_PLNTXT 0x20020 #ifdef CONFIG_CIFS_UPCALL -#define CIFSSEC_MASK 0x3F03F /* allows weak security but also krb5 */ +#define CIFSSEC_MASK 0xAF0AF /* allows weak security but also krb5 */ #else -#define CIFSSEC_MASK 0x37037 /* current flags supported if weak */ +#define CIFSSEC_MASK 0xA70A7 /* current flags supported if weak */ #endif /* UPCALL */ #else /* do not allow weak pw hash */ #ifdef CONFIG_CIFS_UPCALL -#define CIFSSEC_MASK 0x0F00F /* flags supported if no weak allowed */ +#define CIFSSEC_MASK 0x8F08F /* flags supported if no weak allowed */ #else -#define CIFSSEC_MASK 0x07007 /* flags supported if no weak allowed */ +#define CIFSSEC_MASK 0x87087 /* flags supported if no weak allowed */ #endif /* UPCALL */ #endif /* WEAK_PW_HASH */ #define CIFSSEC_MUST_SEAL 0x40040 /* not supported yet */ +#define CIFSSEC_MUST_NTLMSSP 0x80080 /* raw ntlmssp with ntlmv2 */ #define CIFSSEC_DEF (CIFSSEC_MAY_SIGN | CIFSSEC_MAY_NTLM | CIFSSEC_MAY_NTLMV2) #define CIFSSEC_MAX (CIFSSEC_MUST_SIGN | CIFSSEC_MUST_NTLMV2) -#define CIFSSEC_AUTH_MASK (CIFSSEC_MAY_NTLM | CIFSSEC_MAY_NTLMV2 | CIFSSEC_MAY_LANMAN | CIFSSEC_MAY_PLNTXT | CIFSSEC_MAY_KRB5) +#define CIFSSEC_AUTH_MASK (CIFSSEC_MAY_NTLM | CIFSSEC_MAY_NTLMV2 | CIFSSEC_MAY_LANMAN | CIFSSEC_MAY_PLNTXT | CIFSSEC_MAY_KRB5 | CIFSSEC_MAY_NTLMSSP) /* ***************************************************************** * All constants go here diff --git a/fs/cifs/cifsproto.h b/fs/cifs/cifsproto.h index 4167716..fae0839 100644 --- a/fs/cifs/cifsproto.h +++ b/fs/cifs/cifsproto.h @@ -260,8 +260,7 @@ extern int CIFSUnixCreateSymLink(const int xid, const struct nls_table *nls_codepage); extern int CIFSSMBUnixQuerySymLink(const int xid, struct cifsTconInfo *tcon, - const unsigned char *searchName, - char *syminfo, const int buflen, + const unsigned char *searchName, char **syminfo, const struct nls_table *nls_codepage); extern int CIFSSMBQueryReparseLinkInfo(const int xid, struct cifsTconInfo *tcon, @@ -307,8 +306,6 @@ extern int CIFSGetSrvInodeNumber(const int xid, struct cifsTconInfo *tcon, const unsigned char *searchName, __u64 *inode_number, const struct nls_table *nls_codepage, int remap_special_chars); -extern int cifs_convertUCSpath(char *target, const __le16 *source, int maxlen, - const struct nls_table *codepage); extern int cifsConvertToUCS(__le16 *target, const char *source, int maxlen, const struct nls_table *cp, int mapChars); diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index a0845dc..5759ba5 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -1,7 +1,7 @@ /* * fs/cifs/cifssmb.c * - * Copyright (C) International Business Machines Corp., 2002,2008 + * Copyright (C) International Business Machines Corp., 2002,2009 * Author(s): Steve French (sfrench@us.ibm.com) * * Contains the routines for constructing the SMB PDUs themselves @@ -81,41 +81,6 @@ static struct { #endif /* CONFIG_CIFS_WEAK_PW_HASH */ #endif /* CIFS_POSIX */ -/* Allocates buffer into dst and copies smb string from src to it. - * caller is responsible for freeing dst if function returned 0. - * returns: - * on success - 0 - * on failure - errno - */ -static int -cifs_strncpy_to_host(char **dst, const char *src, const int maxlen, - const bool is_unicode, const struct nls_table *nls_codepage) -{ - int plen; - - if (is_unicode) { - plen = UniStrnlen((wchar_t *)src, maxlen); - *dst = kmalloc(plen + 2, GFP_KERNEL); - if (!*dst) - goto cifs_strncpy_to_host_ErrExit; - cifs_strfromUCS_le(*dst, (__le16 *)src, plen, nls_codepage); - } else { - plen = strnlen(src, maxlen); - *dst = kmalloc(plen + 2, GFP_KERNEL); - if (!*dst) - goto cifs_strncpy_to_host_ErrExit; - strncpy(*dst, src, plen); - } - (*dst)[plen] = 0; - (*dst)[plen+1] = 0; /* harmless for ASCII case, needed for Unicode */ - return 0; - -cifs_strncpy_to_host_ErrExit: - cERROR(1, ("Failed to allocate buffer for string\n")); - return -ENOMEM; -} - - /* Mark as invalid, all open files on tree connections since they were closed when session to server was lost */ static void mark_open_files_invalid(struct cifsTconInfo *pTcon) @@ -484,6 +449,14 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses) cFYI(1, ("Kerberos only mechanism, enable extended security")); pSMB->hdr.Flags2 |= SMBFLG2_EXT_SEC; } +#ifdef CONFIG_CIFS_EXPERIMENTAL + else if ((secFlags & CIFSSEC_MUST_NTLMSSP) == CIFSSEC_MUST_NTLMSSP) + pSMB->hdr.Flags2 |= SMBFLG2_EXT_SEC; + else if ((secFlags & CIFSSEC_AUTH_MASK) == CIFSSEC_MAY_NTLMSSP) { + cFYI(1, ("NTLMSSP only mechanism, enable extended security")); + pSMB->hdr.Flags2 |= SMBFLG2_EXT_SEC; + } +#endif count = 0; for (i = 0; i < CIFS_NUM_PROT; i++) { @@ -620,6 +593,8 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses) server->secType = NTLMv2; else if (secFlags & CIFSSEC_MAY_KRB5) server->secType = Kerberos; + else if (secFlags & CIFSSEC_MAY_NTLMSSP) + server->secType = NTLMSSP; else if (secFlags & CIFSSEC_MAY_LANMAN) server->secType = LANMAN; /* #ifdef CONFIG_CIFS_EXPERIMENTAL @@ -2417,8 +2392,7 @@ winCreateHardLinkRetry: int CIFSSMBUnixQuerySymLink(const int xid, struct cifsTconInfo *tcon, - const unsigned char *searchName, - char *symlinkinfo, const int buflen, + const unsigned char *searchName, char **symlinkinfo, const struct nls_table *nls_codepage) { /* SMB_QUERY_FILE_UNIX_LINK */ @@ -2428,6 +2402,7 @@ CIFSSMBUnixQuerySymLink(const int xid, struct cifsTconInfo *tcon, int bytes_returned; int name_len; __u16 params, byte_count; + char *data_start; cFYI(1, ("In QPathSymLinkInfo (Unix) for path %s", searchName)); @@ -2482,30 +2457,26 @@ querySymLinkRetry: /* decode response */ rc = validate_t2((struct smb_t2_rsp *)pSMBr); - if (rc || (pSMBr->ByteCount < 2)) /* BB also check enough total bytes returned */ - rc = -EIO; /* bad smb */ + if (rc || (pSMBr->ByteCount < 2)) + rc = -EIO; else { - __u16 data_offset = le16_to_cpu(pSMBr->t2.DataOffset); - __u16 count = le16_to_cpu(pSMBr->t2.DataCount); + bool is_unicode; + u16 count = le16_to_cpu(pSMBr->t2.DataCount); + + data_start = ((char *) &pSMBr->hdr.Protocol) + + le16_to_cpu(pSMBr->t2.DataOffset); + + if (pSMBr->hdr.Flags2 & SMBFLG2_UNICODE) + is_unicode = true; + else + is_unicode = false; - if (pSMBr->hdr.Flags2 & SMBFLG2_UNICODE) { - name_len = UniStrnlen((wchar_t *) ((char *) - &pSMBr->hdr.Protocol + data_offset), - min_t(const int, buflen, count) / 2); /* BB FIXME investigate remapping reserved chars here */ - cifs_strfromUCS_le(symlinkinfo, - (__le16 *) ((char *)&pSMBr->hdr.Protocol - + data_offset), - name_len, nls_codepage); - } else { - strncpy(symlinkinfo, - (char *) &pSMBr->hdr.Protocol + - data_offset, - min_t(const int, buflen, count)); - } - symlinkinfo[buflen] = 0; - /* just in case so calling code does not go off the end of buffer */ + *symlinkinfo = cifs_strndup_from_ucs(data_start, count, + is_unicode, nls_codepage); + if (!symlinkinfo) + rc = -ENOMEM; } } cifs_buf_release(pSMB); @@ -2603,7 +2574,6 @@ validate_ntransact(char *buf, char **ppparm, char **ppdata, *pparmlen = parm_count; return 0; } -#endif /* CIFS_EXPERIMENTAL */ int CIFSSMBQueryReparseLinkInfo(const int xid, struct cifsTconInfo *tcon, @@ -2613,7 +2583,6 @@ CIFSSMBQueryReparseLinkInfo(const int xid, struct cifsTconInfo *tcon, { int rc = 0; int bytes_returned; - int name_len; struct smb_com_transaction_ioctl_req *pSMB; struct smb_com_transaction_ioctl_rsp *pSMBr; @@ -2650,59 +2619,55 @@ CIFSSMBQueryReparseLinkInfo(const int xid, struct cifsTconInfo *tcon, } else { /* decode response */ __u32 data_offset = le32_to_cpu(pSMBr->DataOffset); __u32 data_count = le32_to_cpu(pSMBr->DataCount); - if ((pSMBr->ByteCount < 2) || (data_offset > 512)) + if ((pSMBr->ByteCount < 2) || (data_offset > 512)) { /* BB also check enough total bytes returned */ rc = -EIO; /* bad smb */ - else { - if (data_count && (data_count < 2048)) { - char *end_of_smb = 2 /* sizeof byte count */ + - pSMBr->ByteCount + - (char *)&pSMBr->ByteCount; + goto qreparse_out; + } + if (data_count && (data_count < 2048)) { + char *end_of_smb = 2 /* sizeof byte count */ + + pSMBr->ByteCount + (char *)&pSMBr->ByteCount; - struct reparse_data *reparse_buf = + struct reparse_data *reparse_buf = (struct reparse_data *) ((char *)&pSMBr->hdr.Protocol + data_offset); - if ((char *)reparse_buf >= end_of_smb) { - rc = -EIO; - goto qreparse_out; - } - if ((reparse_buf->LinkNamesBuf + - reparse_buf->TargetNameOffset + - reparse_buf->TargetNameLen) > - end_of_smb) { - cFYI(1, ("reparse buf beyond SMB")); - rc = -EIO; - goto qreparse_out; - } + if ((char *)reparse_buf >= end_of_smb) { + rc = -EIO; + goto qreparse_out; + } + if ((reparse_buf->LinkNamesBuf + + reparse_buf->TargetNameOffset + + reparse_buf->TargetNameLen) > end_of_smb) { + cFYI(1, ("reparse buf beyond SMB")); + rc = -EIO; + goto qreparse_out; + } - if (pSMBr->hdr.Flags2 & SMBFLG2_UNICODE) { - name_len = UniStrnlen((wchar_t *) + if (pSMBr->hdr.Flags2 & SMBFLG2_UNICODE) { + cifs_from_ucs2(symlinkinfo, (__le16 *) (reparse_buf->LinkNamesBuf + reparse_buf->TargetNameOffset), - min(buflen/2, - reparse_buf->TargetNameLen / 2)); - cifs_strfromUCS_le(symlinkinfo, - (__le16 *) (reparse_buf->LinkNamesBuf + - reparse_buf->TargetNameOffset), - name_len, nls_codepage); - } else { /* ASCII names */ - strncpy(symlinkinfo, - reparse_buf->LinkNamesBuf + - reparse_buf->TargetNameOffset, - min_t(const int, buflen, - reparse_buf->TargetNameLen)); - } - } else { - rc = -EIO; - cFYI(1, ("Invalid return data count on " - "get reparse info ioctl")); + buflen, + reparse_buf->TargetNameLen, + nls_codepage, 0); + } else { /* ASCII names */ + strncpy(symlinkinfo, + reparse_buf->LinkNamesBuf + + reparse_buf->TargetNameOffset, + min_t(const int, buflen, + reparse_buf->TargetNameLen)); } - symlinkinfo[buflen] = 0; /* just in case so the caller - does not go off the end of the buffer */ - cFYI(1, ("readlink result - %s", symlinkinfo)); + } else { + rc = -EIO; + cFYI(1, ("Invalid return data count on " + "get reparse info ioctl")); } + symlinkinfo[buflen] = 0; /* just in case so the caller + does not go off the end of the buffer */ + cFYI(1, ("readlink result - %s", symlinkinfo)); } + qreparse_out: cifs_buf_release(pSMB); @@ -2711,6 +2676,7 @@ qreparse_out: return rc; } +#endif /* CIFS_EXPERIMENTAL */ #ifdef CONFIG_CIFS_POSIX @@ -3928,27 +3894,6 @@ GetInodeNumOut: return rc; } -/* computes length of UCS string converted to host codepage - * @src: UCS string - * @maxlen: length of the input string in UCS characters - * (not in bytes) - * - * return: size of input string in host codepage - */ -static int hostlen_fromUCS(const __le16 *src, const int maxlen, - const struct nls_table *nls_codepage) { - int i; - int hostlen = 0; - char to[4]; - int charlen; - for (i = 0; (i < maxlen) && src[i]; ++i) { - charlen = nls_codepage->uni2char(le16_to_cpu(src[i]), - to, NLS_MAX_CHARSET_SIZE); - hostlen += charlen > 0 ? charlen : 1; - } - return hostlen; -} - /* parses DFS refferal V3 structure * caller is responsible for freeing target_nodes * returns: @@ -3994,7 +3939,7 @@ parse_DFS_referrals(TRANSACTION2_GET_DFS_REFER_RSP *pSMBr, cFYI(1, ("num_referrals: %d dfs flags: 0x%x ... \n", *num_of_nodes, - le16_to_cpu(pSMBr->DFSFlags))); + le32_to_cpu(pSMBr->DFSFlags))); *target_nodes = kzalloc(sizeof(struct dfs_info3_param) * *num_of_nodes, GFP_KERNEL); @@ -4010,14 +3955,14 @@ parse_DFS_referrals(TRANSACTION2_GET_DFS_REFER_RSP *pSMBr, int max_len; struct dfs_info3_param *node = (*target_nodes)+i; - node->flags = le16_to_cpu(pSMBr->DFSFlags); + node->flags = le32_to_cpu(pSMBr->DFSFlags); if (is_unicode) { __le16 *tmp = kmalloc(strlen(searchName)*2 + 2, GFP_KERNEL); cifsConvertToUCS((__le16 *) tmp, searchName, PATH_MAX, nls_codepage, remap); - node->path_consumed = hostlen_fromUCS(tmp, - le16_to_cpu(pSMBr->PathConsumed)/2, + node->path_consumed = cifs_ucs2_bytes(tmp, + le16_to_cpu(pSMBr->PathConsumed), nls_codepage); kfree(tmp); } else @@ -4029,20 +3974,20 @@ parse_DFS_referrals(TRANSACTION2_GET_DFS_REFER_RSP *pSMBr, /* copy DfsPath */ temp = (char *)ref + le16_to_cpu(ref->DfsPathOffset); max_len = data_end - temp; - rc = cifs_strncpy_to_host(&(node->path_name), temp, - max_len, is_unicode, nls_codepage); - if (rc) + node->path_name = cifs_strndup_from_ucs(temp, max_len, + is_unicode, nls_codepage); + if (!node->path_name) { + rc = -ENOMEM; goto parse_DFS_referrals_exit; + } /* copy link target UNC */ temp = (char *)ref + le16_to_cpu(ref->NetworkAddressOffset); max_len = data_end - temp; - rc = cifs_strncpy_to_host(&(node->node_name), temp, - max_len, is_unicode, nls_codepage); - if (rc) - goto parse_DFS_referrals_exit; - - ref += le16_to_cpu(ref->Size); + node->node_name = cifs_strndup_from_ucs(temp, max_len, + is_unicode, nls_codepage); + if (!node->node_name) + rc = -ENOMEM; } parse_DFS_referrals_exit: diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index bacdef1..4aa81a5 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -1,7 +1,7 @@ /* * fs/cifs/connect.c * - * Copyright (C) International Business Machines Corp., 2002,2008 + * Copyright (C) International Business Machines Corp., 2002,2009 * Author(s): Steve French (sfrench@us.ibm.com) * * This library is free software; you can redistribute it and/or modify @@ -32,6 +32,7 @@ #include <linux/kthread.h> #include <linux/pagevec.h> #include <linux/freezer.h> +#include <linux/namei.h> #include <asm/uaccess.h> #include <asm/processor.h> #include <net/ipv6.h> @@ -978,6 +979,13 @@ cifs_parse_mount_options(char *options, const char *devname, return 1; } else if (strnicmp(value, "krb5", 4) == 0) { vol->secFlg |= CIFSSEC_MAY_KRB5; +#ifdef CONFIG_CIFS_EXPERIMENTAL + } else if (strnicmp(value, "ntlmsspi", 8) == 0) { + vol->secFlg |= CIFSSEC_MAY_NTLMSSP | + CIFSSEC_MUST_SIGN; + } else if (strnicmp(value, "ntlmssp", 7) == 0) { + vol->secFlg |= CIFSSEC_MAY_NTLMSSP; +#endif } else if (strnicmp(value, "ntlmv2i", 7) == 0) { vol->secFlg |= CIFSSEC_MAY_NTLMV2 | CIFSSEC_MUST_SIGN; @@ -2278,6 +2286,7 @@ cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, #ifdef CONFIG_CIFS_DFS_UPCALL struct dfs_info3_param *referrals = NULL; unsigned int num_referrals = 0; + int referral_walks_count = 0; try_mount_again: #endif full_path = NULL; @@ -2525,6 +2534,16 @@ remote_path_check: /* get referral if needed */ if (rc == -EREMOTE) { #ifdef CONFIG_CIFS_DFS_UPCALL + if (referral_walks_count > MAX_NESTED_LINKS) { + /* + * BB: when we implement proper loop detection, + * we will remove this check. But now we need it + * to prevent an indefinite loop if 'DFS tree' is + * misconfigured (i.e. has loops). + */ + rc = -ELOOP; + goto mount_fail_check; + } /* convert forward to back slashes in prepath here if needed */ if ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_POSIX_PATHS) == 0) convert_delimiter(cifs_sb->prepath, @@ -2558,6 +2577,7 @@ remote_path_check: cleanup_volume_info(&volume_info); FreeXid(xid); kfree(full_path); + referral_walks_count++; goto try_mount_again; } #else /* No DFS support, return error on mount */ @@ -2592,1041 +2612,6 @@ out: return rc; } -static int -CIFSSessSetup(unsigned int xid, struct cifsSesInfo *ses, - char session_key[CIFS_SESS_KEY_SIZE], - const struct nls_table *nls_codepage) -{ - struct smb_hdr *smb_buffer; - struct smb_hdr *smb_buffer_response; - SESSION_SETUP_ANDX *pSMB; - SESSION_SETUP_ANDX *pSMBr; - char *bcc_ptr; - char *user; - char *domain; - int rc = 0; - int remaining_words = 0; - int bytes_returned = 0; - int len; - __u32 capabilities; - __u16 count; - - cFYI(1, ("In sesssetup")); - if (ses == NULL) - return -EINVAL; - user = ses->userName; - domain = ses->domainName; - smb_buffer = cifs_buf_get(); - - if (smb_buffer == NULL) - return -ENOMEM; - - smb_buffer_response = smb_buffer; - pSMBr = pSMB = (SESSION_SETUP_ANDX *) smb_buffer; - - /* send SMBsessionSetup here */ - header_assemble(smb_buffer, SMB_COM_SESSION_SETUP_ANDX, - NULL /* no tCon exists yet */ , 13 /* wct */ ); - - smb_buffer->Mid = GetNextMid(ses->server); - pSMB->req_no_secext.AndXCommand = 0xFF; - pSMB->req_no_secext.MaxBufferSize = cpu_to_le16(ses->server->maxBuf); - pSMB->req_no_secext.MaxMpxCount = cpu_to_le16(ses->server->maxReq); - - if (ses->server->secMode & - (SECMODE_SIGN_REQUIRED | SECMODE_SIGN_ENABLED)) - smb_buffer->Flags2 |= SMBFLG2_SECURITY_SIGNATURE; - - capabilities = CAP_LARGE_FILES | CAP_NT_SMBS | CAP_LEVEL_II_OPLOCKS | - CAP_LARGE_WRITE_X | CAP_LARGE_READ_X; - if (ses->capabilities & CAP_UNICODE) { - smb_buffer->Flags2 |= SMBFLG2_UNICODE; - capabilities |= CAP_UNICODE; - } - if (ses->capabilities & CAP_STATUS32) { - smb_buffer->Flags2 |= SMBFLG2_ERR_STATUS; - capabilities |= CAP_STATUS32; - } - if (ses->capabilities & CAP_DFS) { - smb_buffer->Flags2 |= SMBFLG2_DFS; - capabilities |= CAP_DFS; - } - pSMB->req_no_secext.Capabilities = cpu_to_le32(capabilities); - - pSMB->req_no_secext.CaseInsensitivePasswordLength = - cpu_to_le16(CIFS_SESS_KEY_SIZE); - - pSMB->req_no_secext.CaseSensitivePasswordLength = - cpu_to_le16(CIFS_SESS_KEY_SIZE); - bcc_ptr = pByteArea(smb_buffer); - memcpy(bcc_ptr, (char *) session_key, CIFS_SESS_KEY_SIZE); - bcc_ptr += CIFS_SESS_KEY_SIZE; - memcpy(bcc_ptr, (char *) session_key, CIFS_SESS_KEY_SIZE); - bcc_ptr += CIFS_SESS_KEY_SIZE; - - if (ses->capabilities & CAP_UNICODE) { - if ((long) bcc_ptr % 2) { /* must be word aligned for Unicode */ - *bcc_ptr = 0; - bcc_ptr++; - } - if (user == NULL) - bytes_returned = 0; /* skip null user */ - else - bytes_returned = - cifs_strtoUCS((__le16 *) bcc_ptr, user, 100, - nls_codepage); - /* convert number of 16 bit words to bytes */ - bcc_ptr += 2 * bytes_returned; - bcc_ptr += 2; /* trailing null */ - if (domain == NULL) - bytes_returned = - cifs_strtoUCS((__le16 *) bcc_ptr, - "CIFS_LINUX_DOM", 32, nls_codepage); - else - bytes_returned = - cifs_strtoUCS((__le16 *) bcc_ptr, domain, 64, - nls_codepage); - bcc_ptr += 2 * bytes_returned; - bcc_ptr += 2; - bytes_returned = - cifs_strtoUCS((__le16 *) bcc_ptr, "Linux version ", - 32, nls_codepage); - bcc_ptr += 2 * bytes_returned; - bytes_returned = - cifs_strtoUCS((__le16 *) bcc_ptr, utsname()->release, - 32, nls_codepage); - bcc_ptr += 2 * bytes_returned; - bcc_ptr += 2; - bytes_returned = - cifs_strtoUCS((__le16 *) bcc_ptr, CIFS_NETWORK_OPSYS, - 64, nls_codepage); - bcc_ptr += 2 * bytes_returned; - bcc_ptr += 2; - } else { - if (user != NULL) { - strncpy(bcc_ptr, user, 200); - bcc_ptr += strnlen(user, 200); - } - *bcc_ptr = 0; - bcc_ptr++; - if (domain == NULL) { - strcpy(bcc_ptr, "CIFS_LINUX_DOM"); - bcc_ptr += strlen("CIFS_LINUX_DOM") + 1; - } else { - strncpy(bcc_ptr, domain, 64); - bcc_ptr += strnlen(domain, 64); - *bcc_ptr = 0; - bcc_ptr++; - } - strcpy(bcc_ptr, "Linux version "); - bcc_ptr += strlen("Linux version "); - strcpy(bcc_ptr, utsname()->release); - bcc_ptr += strlen(utsname()->release) + 1; - strcpy(bcc_ptr, CIFS_NETWORK_OPSYS); - bcc_ptr += strlen(CIFS_NETWORK_OPSYS) + 1; - } - count = (long) bcc_ptr - (long) pByteArea(smb_buffer); - smb_buffer->smb_buf_length += count; - pSMB->req_no_secext.ByteCount = cpu_to_le16(count); - - rc = SendReceive(xid, ses, smb_buffer, smb_buffer_response, - &bytes_returned, CIFS_LONG_OP); - if (rc) { -/* rc = map_smb_to_linux_error(smb_buffer_response); now done in SendReceive */ - } else if ((smb_buffer_response->WordCount == 3) - || (smb_buffer_response->WordCount == 4)) { - __u16 action = le16_to_cpu(pSMBr->resp.Action); - __u16 blob_len = le16_to_cpu(pSMBr->resp.SecurityBlobLength); - if (action & GUEST_LOGIN) - cFYI(1, ("Guest login")); /* BB mark SesInfo struct? */ - ses->Suid = smb_buffer_response->Uid; /* UID left in wire format - (little endian) */ - cFYI(1, ("UID = %d ", ses->Suid)); - /* response can have either 3 or 4 word count - Samba sends 3 */ - bcc_ptr = pByteArea(smb_buffer_response); - if ((pSMBr->resp.hdr.WordCount == 3) - || ((pSMBr->resp.hdr.WordCount == 4) - && (blob_len < pSMBr->resp.ByteCount))) { - if (pSMBr->resp.hdr.WordCount == 4) - bcc_ptr += blob_len; - - if (smb_buffer->Flags2 & SMBFLG2_UNICODE) { - if ((long) (bcc_ptr) % 2) { - remaining_words = - (BCC(smb_buffer_response) - 1) / 2; - /* Unicode strings must be word - aligned */ - bcc_ptr++; - } else { - remaining_words = - BCC(smb_buffer_response) / 2; - } - len = - UniStrnlen((wchar_t *) bcc_ptr, - remaining_words - 1); -/* We look for obvious messed up bcc or strings in response so we do not go off - the end since (at least) WIN2K and Windows XP have a major bug in not null - terminating last Unicode string in response */ - kfree(ses->serverOS); - ses->serverOS = kzalloc(2 * (len + 1), - GFP_KERNEL); - if (ses->serverOS == NULL) - goto sesssetup_nomem; - cifs_strfromUCS_le(ses->serverOS, - (__le16 *)bcc_ptr, - len, nls_codepage); - bcc_ptr += 2 * (len + 1); - remaining_words -= len + 1; - ses->serverOS[2 * len] = 0; - ses->serverOS[1 + (2 * len)] = 0; - if (remaining_words > 0) { - len = UniStrnlen((wchar_t *)bcc_ptr, - remaining_words-1); - kfree(ses->serverNOS); - ses->serverNOS = kzalloc(2 * (len + 1), - GFP_KERNEL); - if (ses->serverNOS == NULL) - goto sesssetup_nomem; - cifs_strfromUCS_le(ses->serverNOS, - (__le16 *)bcc_ptr, - len, nls_codepage); - bcc_ptr += 2 * (len + 1); - ses->serverNOS[2 * len] = 0; - ses->serverNOS[1 + (2 * len)] = 0; - if (strncmp(ses->serverNOS, - "NT LAN Manager 4", 16) == 0) { - cFYI(1, ("NT4 server")); - ses->flags |= CIFS_SES_NT4; - } - remaining_words -= len + 1; - if (remaining_words > 0) { - len = UniStrnlen((wchar_t *) bcc_ptr, remaining_words); - /* last string is not always null terminated - (for e.g. for Windows XP & 2000) */ - kfree(ses->serverDomain); - ses->serverDomain = - kzalloc(2*(len+1), - GFP_KERNEL); - if (ses->serverDomain == NULL) - goto sesssetup_nomem; - cifs_strfromUCS_le(ses->serverDomain, - (__le16 *)bcc_ptr, - len, nls_codepage); - bcc_ptr += 2 * (len + 1); - ses->serverDomain[2*len] = 0; - ses->serverDomain[1+(2*len)] = 0; - } else { /* else no more room so create - dummy domain string */ - kfree(ses->serverDomain); - ses->serverDomain = - kzalloc(2, GFP_KERNEL); - } - } else { /* no room so create dummy domain - and NOS string */ - - /* if these kcallocs fail not much we - can do, but better to not fail the - sesssetup itself */ - kfree(ses->serverDomain); - ses->serverDomain = - kzalloc(2, GFP_KERNEL); - kfree(ses->serverNOS); - ses->serverNOS = - kzalloc(2, GFP_KERNEL); - } - } else { /* ASCII */ - len = strnlen(bcc_ptr, 1024); - if (((long) bcc_ptr + len) - (long) - pByteArea(smb_buffer_response) - <= BCC(smb_buffer_response)) { - kfree(ses->serverOS); - ses->serverOS = kzalloc(len + 1, - GFP_KERNEL); - if (ses->serverOS == NULL) - goto sesssetup_nomem; - strncpy(ses->serverOS, bcc_ptr, len); - - bcc_ptr += len; - /* null terminate the string */ - bcc_ptr[0] = 0; - bcc_ptr++; - - len = strnlen(bcc_ptr, 1024); - kfree(ses->serverNOS); - ses->serverNOS = kzalloc(len + 1, - GFP_KERNEL); - if (ses->serverNOS == NULL) - goto sesssetup_nomem; - strncpy(ses->serverNOS, bcc_ptr, len); - bcc_ptr += len; - bcc_ptr[0] = 0; - bcc_ptr++; - - len = strnlen(bcc_ptr, 1024); - kfree(ses->serverDomain); - ses->serverDomain = kzalloc(len + 1, - GFP_KERNEL); - if (ses->serverDomain == NULL) - goto sesssetup_nomem; - strncpy(ses->serverDomain, bcc_ptr, - len); - bcc_ptr += len; - bcc_ptr[0] = 0; - bcc_ptr++; - } else - cFYI(1, - ("Variable field of length %d " - "extends beyond end of smb ", - len)); - } - } else { - cERROR(1, ("Security Blob Length extends beyond " - "end of SMB")); - } - } else { - cERROR(1, ("Invalid Word count %d: ", - smb_buffer_response->WordCount)); - rc = -EIO; - } -sesssetup_nomem: /* do not return an error on nomem for the info strings, - since that could make reconnection harder, and - reconnection might be needed to free memory */ - cifs_buf_release(smb_buffer); - - return rc; -} - -static int -CIFSNTLMSSPNegotiateSessSetup(unsigned int xid, - struct cifsSesInfo *ses, bool *pNTLMv2_flag, - const struct nls_table *nls_codepage) -{ - struct smb_hdr *smb_buffer; - struct smb_hdr *smb_buffer_response; - SESSION_SETUP_ANDX *pSMB; - SESSION_SETUP_ANDX *pSMBr; - char *bcc_ptr; - char *domain; - int rc = 0; - int remaining_words = 0; - int bytes_returned = 0; - int len; - int SecurityBlobLength = sizeof(NEGOTIATE_MESSAGE); - PNEGOTIATE_MESSAGE SecurityBlob; - PCHALLENGE_MESSAGE SecurityBlob2; - __u32 negotiate_flags, capabilities; - __u16 count; - - cFYI(1, ("In NTLMSSP sesssetup (negotiate)")); - if (ses == NULL) - return -EINVAL; - domain = ses->domainName; - *pNTLMv2_flag = false; - smb_buffer = cifs_buf_get(); - if (smb_buffer == NULL) { - return -ENOMEM; - } - smb_buffer_response = smb_buffer; - pSMB = (SESSION_SETUP_ANDX *) smb_buffer; - pSMBr = (SESSION_SETUP_ANDX *) smb_buffer_response; - - /* send SMBsessionSetup here */ - header_assemble(smb_buffer, SMB_COM_SESSION_SETUP_ANDX, - NULL /* no tCon exists yet */ , 12 /* wct */ ); - - smb_buffer->Mid = GetNextMid(ses->server); - pSMB->req.hdr.Flags2 |= SMBFLG2_EXT_SEC; - pSMB->req.hdr.Flags |= (SMBFLG_CASELESS | SMBFLG_CANONICAL_PATH_FORMAT); - - pSMB->req.AndXCommand = 0xFF; - pSMB->req.MaxBufferSize = cpu_to_le16(ses->server->maxBuf); - pSMB->req.MaxMpxCount = cpu_to_le16(ses->server->maxReq); - - if (ses->server->secMode & (SECMODE_SIGN_REQUIRED | SECMODE_SIGN_ENABLED)) - smb_buffer->Flags2 |= SMBFLG2_SECURITY_SIGNATURE; - - capabilities = CAP_LARGE_FILES | CAP_NT_SMBS | CAP_LEVEL_II_OPLOCKS | - CAP_EXTENDED_SECURITY; - if (ses->capabilities & CAP_UNICODE) { - smb_buffer->Flags2 |= SMBFLG2_UNICODE; - capabilities |= CAP_UNICODE; - } - if (ses->capabilities & CAP_STATUS32) { - smb_buffer->Flags2 |= SMBFLG2_ERR_STATUS; - capabilities |= CAP_STATUS32; - } - if (ses->capabilities & CAP_DFS) { - smb_buffer->Flags2 |= SMBFLG2_DFS; - capabilities |= CAP_DFS; - } - pSMB->req.Capabilities = cpu_to_le32(capabilities); - - bcc_ptr = (char *) &pSMB->req.SecurityBlob; - SecurityBlob = (PNEGOTIATE_MESSAGE) bcc_ptr; - strncpy(SecurityBlob->Signature, NTLMSSP_SIGNATURE, 8); - SecurityBlob->MessageType = NtLmNegotiate; - negotiate_flags = - NTLMSSP_NEGOTIATE_UNICODE | NTLMSSP_NEGOTIATE_OEM | - NTLMSSP_REQUEST_TARGET | NTLMSSP_NEGOTIATE_NTLM | - NTLMSSP_NEGOTIATE_56 | - /* NTLMSSP_NEGOTIATE_ALWAYS_SIGN | */ NTLMSSP_NEGOTIATE_128; - if (sign_CIFS_PDUs) - negotiate_flags |= NTLMSSP_NEGOTIATE_SIGN; -/* if (ntlmv2_support) - negotiate_flags |= NTLMSSP_NEGOTIATE_NTLMV2;*/ - /* setup pointers to domain name and workstation name */ - bcc_ptr += SecurityBlobLength; - - SecurityBlob->WorkstationName.Buffer = 0; - SecurityBlob->WorkstationName.Length = 0; - SecurityBlob->WorkstationName.MaximumLength = 0; - - /* Domain not sent on first Sesssetup in NTLMSSP, instead it is sent - along with username on auth request (ie the response to challenge) */ - SecurityBlob->DomainName.Buffer = 0; - SecurityBlob->DomainName.Length = 0; - SecurityBlob->DomainName.MaximumLength = 0; - if (ses->capabilities & CAP_UNICODE) { - if ((long) bcc_ptr % 2) { - *bcc_ptr = 0; - bcc_ptr++; - } - - bytes_returned = - cifs_strtoUCS((__le16 *) bcc_ptr, "Linux version ", - 32, nls_codepage); - bcc_ptr += 2 * bytes_returned; - bytes_returned = - cifs_strtoUCS((__le16 *) bcc_ptr, utsname()->release, 32, - nls_codepage); - bcc_ptr += 2 * bytes_returned; - bcc_ptr += 2; /* null terminate Linux version */ - bytes_returned = - cifs_strtoUCS((__le16 *) bcc_ptr, CIFS_NETWORK_OPSYS, - 64, nls_codepage); - bcc_ptr += 2 * bytes_returned; - *(bcc_ptr + 1) = 0; - *(bcc_ptr + 2) = 0; - bcc_ptr += 2; /* null terminate network opsys string */ - *(bcc_ptr + 1) = 0; - *(bcc_ptr + 2) = 0; - bcc_ptr += 2; /* null domain */ - } else { /* ASCII */ - strcpy(bcc_ptr, "Linux version "); - bcc_ptr += strlen("Linux version "); - strcpy(bcc_ptr, utsname()->release); - bcc_ptr += strlen(utsname()->release) + 1; - strcpy(bcc_ptr, CIFS_NETWORK_OPSYS); - bcc_ptr += strlen(CIFS_NETWORK_OPSYS) + 1; - bcc_ptr++; /* empty domain field */ - *bcc_ptr = 0; - } - SecurityBlob->NegotiateFlags = cpu_to_le32(negotiate_flags); - pSMB->req.SecurityBlobLength = cpu_to_le16(SecurityBlobLength); - count = (long) bcc_ptr - (long) pByteArea(smb_buffer); - smb_buffer->smb_buf_length += count; - pSMB->req.ByteCount = cpu_to_le16(count); - - rc = SendReceive(xid, ses, smb_buffer, smb_buffer_response, - &bytes_returned, CIFS_LONG_OP); - - if (smb_buffer_response->Status.CifsError == - cpu_to_le32(NT_STATUS_MORE_PROCESSING_REQUIRED)) - rc = 0; - - if (rc) { -/* rc = map_smb_to_linux_error(smb_buffer_response); *//* done in SendReceive now */ - } else if ((smb_buffer_response->WordCount == 3) - || (smb_buffer_response->WordCount == 4)) { - __u16 action = le16_to_cpu(pSMBr->resp.Action); - __u16 blob_len = le16_to_cpu(pSMBr->resp.SecurityBlobLength); - - if (action & GUEST_LOGIN) - cFYI(1, ("Guest login")); - /* Do we want to set anything in SesInfo struct when guest login? */ - - bcc_ptr = pByteArea(smb_buffer_response); - /* response can have either 3 or 4 word count - Samba sends 3 */ - - SecurityBlob2 = (PCHALLENGE_MESSAGE) bcc_ptr; - if (SecurityBlob2->MessageType != NtLmChallenge) { - cFYI(1, ("Unexpected NTLMSSP message type received %d", - SecurityBlob2->MessageType)); - } else if (ses) { - ses->Suid = smb_buffer_response->Uid; /* UID left in le format */ - cFYI(1, ("UID = %d", ses->Suid)); - if ((pSMBr->resp.hdr.WordCount == 3) - || ((pSMBr->resp.hdr.WordCount == 4) - && (blob_len < - pSMBr->resp.ByteCount))) { - - if (pSMBr->resp.hdr.WordCount == 4) { - bcc_ptr += blob_len; - cFYI(1, ("Security Blob Length %d", - blob_len)); - } - - cFYI(1, ("NTLMSSP Challenge rcvd")); - - memcpy(ses->server->cryptKey, - SecurityBlob2->Challenge, - CIFS_CRYPTO_KEY_SIZE); - if (SecurityBlob2->NegotiateFlags & - cpu_to_le32(NTLMSSP_NEGOTIATE_NTLMV2)) - *pNTLMv2_flag = true; - - if ((SecurityBlob2->NegotiateFlags & - cpu_to_le32(NTLMSSP_NEGOTIATE_ALWAYS_SIGN)) - || (sign_CIFS_PDUs > 1)) - ses->server->secMode |= - SECMODE_SIGN_REQUIRED; - if ((SecurityBlob2->NegotiateFlags & - cpu_to_le32(NTLMSSP_NEGOTIATE_SIGN)) && (sign_CIFS_PDUs)) - ses->server->secMode |= - SECMODE_SIGN_ENABLED; - - if (smb_buffer->Flags2 & SMBFLG2_UNICODE) { - if ((long) (bcc_ptr) % 2) { - remaining_words = - (BCC(smb_buffer_response) - - 1) / 2; - /* Must word align unicode strings */ - bcc_ptr++; - } else { - remaining_words = - BCC - (smb_buffer_response) / 2; - } - len = - UniStrnlen((wchar_t *) bcc_ptr, - remaining_words - 1); -/* We look for obvious messed up bcc or strings in response so we do not go off - the end since (at least) WIN2K and Windows XP have a major bug in not null - terminating last Unicode string in response */ - kfree(ses->serverOS); - ses->serverOS = - kzalloc(2 * (len + 1), GFP_KERNEL); - cifs_strfromUCS_le(ses->serverOS, - (__le16 *) - bcc_ptr, len, - nls_codepage); - bcc_ptr += 2 * (len + 1); - remaining_words -= len + 1; - ses->serverOS[2 * len] = 0; - ses->serverOS[1 + (2 * len)] = 0; - if (remaining_words > 0) { - len = UniStrnlen((wchar_t *) - bcc_ptr, - remaining_words - - 1); - kfree(ses->serverNOS); - ses->serverNOS = - kzalloc(2 * (len + 1), - GFP_KERNEL); - cifs_strfromUCS_le(ses-> - serverNOS, - (__le16 *) - bcc_ptr, - len, - nls_codepage); - bcc_ptr += 2 * (len + 1); - ses->serverNOS[2 * len] = 0; - ses->serverNOS[1 + - (2 * len)] = 0; - remaining_words -= len + 1; - if (remaining_words > 0) { - len = UniStrnlen((wchar_t *) bcc_ptr, remaining_words); - /* last string not always null terminated - (for e.g. for Windows XP & 2000) */ - kfree(ses->serverDomain); - ses->serverDomain = - kzalloc(2 * - (len + - 1), - GFP_KERNEL); - cifs_strfromUCS_le - (ses->serverDomain, - (__le16 *)bcc_ptr, - len, nls_codepage); - bcc_ptr += - 2 * (len + 1); - ses->serverDomain[2*len] - = 0; - ses->serverDomain - [1 + (2 * len)] - = 0; - } /* else no more room so create dummy domain string */ - else { - kfree(ses->serverDomain); - ses->serverDomain = - kzalloc(2, - GFP_KERNEL); - } - } else { /* no room so create dummy domain and NOS string */ - kfree(ses->serverDomain); - ses->serverDomain = - kzalloc(2, GFP_KERNEL); - kfree(ses->serverNOS); - ses->serverNOS = - kzalloc(2, GFP_KERNEL); - } - } else { /* ASCII */ - len = strnlen(bcc_ptr, 1024); - if (((long) bcc_ptr + len) - (long) - pByteArea(smb_buffer_response) - <= BCC(smb_buffer_response)) { - kfree(ses->serverOS); - ses->serverOS = - kzalloc(len + 1, - GFP_KERNEL); - strncpy(ses->serverOS, - bcc_ptr, len); - - bcc_ptr += len; - bcc_ptr[0] = 0; /* null terminate string */ - bcc_ptr++; - - len = strnlen(bcc_ptr, 1024); - kfree(ses->serverNOS); - ses->serverNOS = - kzalloc(len + 1, - GFP_KERNEL); - strncpy(ses->serverNOS, bcc_ptr, len); - bcc_ptr += len; - bcc_ptr[0] = 0; - bcc_ptr++; - - len = strnlen(bcc_ptr, 1024); - kfree(ses->serverDomain); - ses->serverDomain = - kzalloc(len + 1, - GFP_KERNEL); - strncpy(ses->serverDomain, - bcc_ptr, len); - bcc_ptr += len; - bcc_ptr[0] = 0; - bcc_ptr++; - } else - cFYI(1, - ("field of length %d " - "extends beyond end of smb", - len)); - } - } else { - cERROR(1, ("Security Blob Length extends beyond" - " end of SMB")); - } - } else { - cERROR(1, ("No session structure passed in.")); - } - } else { - cERROR(1, ("Invalid Word count %d:", - smb_buffer_response->WordCount)); - rc = -EIO; - } - - cifs_buf_release(smb_buffer); - - return rc; -} -static int -CIFSNTLMSSPAuthSessSetup(unsigned int xid, struct cifsSesInfo *ses, - char *ntlm_session_key, bool ntlmv2_flag, - const struct nls_table *nls_codepage) -{ - struct smb_hdr *smb_buffer; - struct smb_hdr *smb_buffer_response; - SESSION_SETUP_ANDX *pSMB; - SESSION_SETUP_ANDX *pSMBr; - char *bcc_ptr; - char *user; - char *domain; - int rc = 0; - int remaining_words = 0; - int bytes_returned = 0; - int len; - int SecurityBlobLength = sizeof(AUTHENTICATE_MESSAGE); - PAUTHENTICATE_MESSAGE SecurityBlob; - __u32 negotiate_flags, capabilities; - __u16 count; - - cFYI(1, ("In NTLMSSPSessSetup (Authenticate)")); - if (ses == NULL) - return -EINVAL; - user = ses->userName; - domain = ses->domainName; - smb_buffer = cifs_buf_get(); - if (smb_buffer == NULL) { - return -ENOMEM; - } - smb_buffer_response = smb_buffer; - pSMB = (SESSION_SETUP_ANDX *)smb_buffer; - pSMBr = (SESSION_SETUP_ANDX *)smb_buffer_response; - - /* send SMBsessionSetup here */ - header_assemble(smb_buffer, SMB_COM_SESSION_SETUP_ANDX, - NULL /* no tCon exists yet */ , 12 /* wct */ ); - - smb_buffer->Mid = GetNextMid(ses->server); - pSMB->req.hdr.Flags |= (SMBFLG_CASELESS | SMBFLG_CANONICAL_PATH_FORMAT); - pSMB->req.hdr.Flags2 |= SMBFLG2_EXT_SEC; - pSMB->req.AndXCommand = 0xFF; - pSMB->req.MaxBufferSize = cpu_to_le16(ses->server->maxBuf); - pSMB->req.MaxMpxCount = cpu_to_le16(ses->server->maxReq); - - pSMB->req.hdr.Uid = ses->Suid; - - if (ses->server->secMode & (SECMODE_SIGN_REQUIRED | SECMODE_SIGN_ENABLED)) - smb_buffer->Flags2 |= SMBFLG2_SECURITY_SIGNATURE; - - capabilities = CAP_LARGE_FILES | CAP_NT_SMBS | CAP_LEVEL_II_OPLOCKS | - CAP_EXTENDED_SECURITY; - if (ses->capabilities & CAP_UNICODE) { - smb_buffer->Flags2 |= SMBFLG2_UNICODE; - capabilities |= CAP_UNICODE; - } - if (ses->capabilities & CAP_STATUS32) { - smb_buffer->Flags2 |= SMBFLG2_ERR_STATUS; - capabilities |= CAP_STATUS32; - } - if (ses->capabilities & CAP_DFS) { - smb_buffer->Flags2 |= SMBFLG2_DFS; - capabilities |= CAP_DFS; - } - pSMB->req.Capabilities = cpu_to_le32(capabilities); - - bcc_ptr = (char *)&pSMB->req.SecurityBlob; - SecurityBlob = (PAUTHENTICATE_MESSAGE)bcc_ptr; - strncpy(SecurityBlob->Signature, NTLMSSP_SIGNATURE, 8); - SecurityBlob->MessageType = NtLmAuthenticate; - bcc_ptr += SecurityBlobLength; - negotiate_flags = NTLMSSP_NEGOTIATE_UNICODE | NTLMSSP_REQUEST_TARGET | - NTLMSSP_NEGOTIATE_NTLM | NTLMSSP_NEGOTIATE_TARGET_INFO | - 0x80000000 | NTLMSSP_NEGOTIATE_128; - if (sign_CIFS_PDUs) - negotiate_flags |= /* NTLMSSP_NEGOTIATE_ALWAYS_SIGN |*/ NTLMSSP_NEGOTIATE_SIGN; - if (ntlmv2_flag) - negotiate_flags |= NTLMSSP_NEGOTIATE_NTLMV2; - -/* setup pointers to domain name and workstation name */ - - SecurityBlob->WorkstationName.Buffer = 0; - SecurityBlob->WorkstationName.Length = 0; - SecurityBlob->WorkstationName.MaximumLength = 0; - SecurityBlob->SessionKey.Length = 0; - SecurityBlob->SessionKey.MaximumLength = 0; - SecurityBlob->SessionKey.Buffer = 0; - - SecurityBlob->LmChallengeResponse.Length = 0; - SecurityBlob->LmChallengeResponse.MaximumLength = 0; - SecurityBlob->LmChallengeResponse.Buffer = 0; - - SecurityBlob->NtChallengeResponse.Length = - cpu_to_le16(CIFS_SESS_KEY_SIZE); - SecurityBlob->NtChallengeResponse.MaximumLength = - cpu_to_le16(CIFS_SESS_KEY_SIZE); - memcpy(bcc_ptr, ntlm_session_key, CIFS_SESS_KEY_SIZE); - SecurityBlob->NtChallengeResponse.Buffer = - cpu_to_le32(SecurityBlobLength); - SecurityBlobLength += CIFS_SESS_KEY_SIZE; - bcc_ptr += CIFS_SESS_KEY_SIZE; - - if (ses->capabilities & CAP_UNICODE) { - if (domain == NULL) { - SecurityBlob->DomainName.Buffer = 0; - SecurityBlob->DomainName.Length = 0; - SecurityBlob->DomainName.MaximumLength = 0; - } else { - __u16 ln = cifs_strtoUCS((__le16 *) bcc_ptr, domain, 64, - nls_codepage); - ln *= 2; - SecurityBlob->DomainName.MaximumLength = - cpu_to_le16(ln); - SecurityBlob->DomainName.Buffer = - cpu_to_le32(SecurityBlobLength); - bcc_ptr += ln; - SecurityBlobLength += ln; - SecurityBlob->DomainName.Length = cpu_to_le16(ln); - } - if (user == NULL) { - SecurityBlob->UserName.Buffer = 0; - SecurityBlob->UserName.Length = 0; - SecurityBlob->UserName.MaximumLength = 0; - } else { - __u16 ln = cifs_strtoUCS((__le16 *) bcc_ptr, user, 64, - nls_codepage); - ln *= 2; - SecurityBlob->UserName.MaximumLength = - cpu_to_le16(ln); - SecurityBlob->UserName.Buffer = - cpu_to_le32(SecurityBlobLength); - bcc_ptr += ln; - SecurityBlobLength += ln; - SecurityBlob->UserName.Length = cpu_to_le16(ln); - } - - /* SecurityBlob->WorkstationName.Length = - cifs_strtoUCS((__le16 *) bcc_ptr, "AMACHINE",64, nls_codepage); - SecurityBlob->WorkstationName.Length *= 2; - SecurityBlob->WorkstationName.MaximumLength = - cpu_to_le16(SecurityBlob->WorkstationName.Length); - SecurityBlob->WorkstationName.Buffer = - cpu_to_le32(SecurityBlobLength); - bcc_ptr += SecurityBlob->WorkstationName.Length; - SecurityBlobLength += SecurityBlob->WorkstationName.Length; - SecurityBlob->WorkstationName.Length = - cpu_to_le16(SecurityBlob->WorkstationName.Length); */ - - if ((long) bcc_ptr % 2) { - *bcc_ptr = 0; - bcc_ptr++; - } - bytes_returned = - cifs_strtoUCS((__le16 *) bcc_ptr, "Linux version ", - 32, nls_codepage); - bcc_ptr += 2 * bytes_returned; - bytes_returned = - cifs_strtoUCS((__le16 *) bcc_ptr, utsname()->release, 32, - nls_codepage); - bcc_ptr += 2 * bytes_returned; - bcc_ptr += 2; /* null term version string */ - bytes_returned = - cifs_strtoUCS((__le16 *) bcc_ptr, CIFS_NETWORK_OPSYS, - 64, nls_codepage); - bcc_ptr += 2 * bytes_returned; - *(bcc_ptr + 1) = 0; - *(bcc_ptr + 2) = 0; - bcc_ptr += 2; /* null terminate network opsys string */ - *(bcc_ptr + 1) = 0; - *(bcc_ptr + 2) = 0; - bcc_ptr += 2; /* null domain */ - } else { /* ASCII */ - if (domain == NULL) { - SecurityBlob->DomainName.Buffer = 0; - SecurityBlob->DomainName.Length = 0; - SecurityBlob->DomainName.MaximumLength = 0; - } else { - __u16 ln; - negotiate_flags |= NTLMSSP_NEGOTIATE_DOMAIN_SUPPLIED; - strncpy(bcc_ptr, domain, 63); - ln = strnlen(domain, 64); - SecurityBlob->DomainName.MaximumLength = - cpu_to_le16(ln); - SecurityBlob->DomainName.Buffer = - cpu_to_le32(SecurityBlobLength); - bcc_ptr += ln; - SecurityBlobLength += ln; - SecurityBlob->DomainName.Length = cpu_to_le16(ln); - } - if (user == NULL) { - SecurityBlob->UserName.Buffer = 0; - SecurityBlob->UserName.Length = 0; - SecurityBlob->UserName.MaximumLength = 0; - } else { - __u16 ln; - strncpy(bcc_ptr, user, 63); - ln = strnlen(user, 64); - SecurityBlob->UserName.MaximumLength = cpu_to_le16(ln); - SecurityBlob->UserName.Buffer = - cpu_to_le32(SecurityBlobLength); - bcc_ptr += ln; - SecurityBlobLength += ln; - SecurityBlob->UserName.Length = cpu_to_le16(ln); - } - /* BB fill in our workstation name if known BB */ - - strcpy(bcc_ptr, "Linux version "); - bcc_ptr += strlen("Linux version "); - strcpy(bcc_ptr, utsname()->release); - bcc_ptr += strlen(utsname()->release) + 1; - strcpy(bcc_ptr, CIFS_NETWORK_OPSYS); - bcc_ptr += strlen(CIFS_NETWORK_OPSYS) + 1; - bcc_ptr++; /* null domain */ - *bcc_ptr = 0; - } - SecurityBlob->NegotiateFlags = cpu_to_le32(negotiate_flags); - pSMB->req.SecurityBlobLength = cpu_to_le16(SecurityBlobLength); - count = (long) bcc_ptr - (long) pByteArea(smb_buffer); - smb_buffer->smb_buf_length += count; - pSMB->req.ByteCount = cpu_to_le16(count); - - rc = SendReceive(xid, ses, smb_buffer, smb_buffer_response, - &bytes_returned, CIFS_LONG_OP); - if (rc) { -/* rc = map_smb_to_linux_error(smb_buffer_response) done in SendReceive now */ - } else if ((smb_buffer_response->WordCount == 3) || - (smb_buffer_response->WordCount == 4)) { - __u16 action = le16_to_cpu(pSMBr->resp.Action); - __u16 blob_len = le16_to_cpu(pSMBr->resp.SecurityBlobLength); - if (action & GUEST_LOGIN) - cFYI(1, ("Guest login")); /* BB Should we set anything - in SesInfo struct ? */ -/* if (SecurityBlob2->MessageType != NtLm??) { - cFYI("Unexpected message type on auth response is %d")); - } */ - - if (ses) { - cFYI(1, - ("Check challenge UID %d vs auth response UID %d", - ses->Suid, smb_buffer_response->Uid)); - /* UID left in wire format */ - ses->Suid = smb_buffer_response->Uid; - bcc_ptr = pByteArea(smb_buffer_response); - /* response can have either 3 or 4 word count - Samba sends 3 */ - if ((pSMBr->resp.hdr.WordCount == 3) - || ((pSMBr->resp.hdr.WordCount == 4) - && (blob_len < - pSMBr->resp.ByteCount))) { - if (pSMBr->resp.hdr.WordCount == 4) { - bcc_ptr += - blob_len; - cFYI(1, - ("Security Blob Length %d ", - blob_len)); - } - - cFYI(1, - ("NTLMSSP response to Authenticate ")); - - if (smb_buffer->Flags2 & SMBFLG2_UNICODE) { - if ((long) (bcc_ptr) % 2) { - remaining_words = - (BCC(smb_buffer_response) - - 1) / 2; - bcc_ptr++; /* Unicode strings must be word aligned */ - } else { - remaining_words = BCC(smb_buffer_response) / 2; - } - len = UniStrnlen((wchar_t *) bcc_ptr, - remaining_words - 1); -/* We look for obvious messed up bcc or strings in response so we do not go off - the end since (at least) WIN2K and Windows XP have a major bug in not null - terminating last Unicode string in response */ - kfree(ses->serverOS); - ses->serverOS = - kzalloc(2 * (len + 1), GFP_KERNEL); - cifs_strfromUCS_le(ses->serverOS, - (__le16 *) - bcc_ptr, len, - nls_codepage); - bcc_ptr += 2 * (len + 1); - remaining_words -= len + 1; - ses->serverOS[2 * len] = 0; - ses->serverOS[1 + (2 * len)] = 0; - if (remaining_words > 0) { - len = UniStrnlen((wchar_t *) - bcc_ptr, - remaining_words - - 1); - kfree(ses->serverNOS); - ses->serverNOS = - kzalloc(2 * (len + 1), - GFP_KERNEL); - cifs_strfromUCS_le(ses-> - serverNOS, - (__le16 *) - bcc_ptr, - len, - nls_codepage); - bcc_ptr += 2 * (len + 1); - ses->serverNOS[2 * len] = 0; - ses->serverNOS[1+(2*len)] = 0; - remaining_words -= len + 1; - if (remaining_words > 0) { - len = UniStrnlen((wchar_t *) bcc_ptr, remaining_words); - /* last string not always null terminated (e.g. for Windows XP & 2000) */ - kfree(ses->serverDomain); - ses->serverDomain = - kzalloc(2 * - (len + - 1), - GFP_KERNEL); - cifs_strfromUCS_le - (ses-> - serverDomain, - (__le16 *) - bcc_ptr, len, - nls_codepage); - bcc_ptr += - 2 * (len + 1); - ses-> - serverDomain[2 - * len] - = 0; - ses-> - serverDomain[1 - + - (2 - * - len)] - = 0; - } /* else no more room so create dummy domain string */ - else { - kfree(ses->serverDomain); - ses->serverDomain = kzalloc(2,GFP_KERNEL); - } - } else { /* no room so create dummy domain and NOS string */ - kfree(ses->serverDomain); - ses->serverDomain = kzalloc(2, GFP_KERNEL); - kfree(ses->serverNOS); - ses->serverNOS = kzalloc(2, GFP_KERNEL); - } - } else { /* ASCII */ - len = strnlen(bcc_ptr, 1024); - if (((long) bcc_ptr + len) - - (long) pByteArea(smb_buffer_response) - <= BCC(smb_buffer_response)) { - kfree(ses->serverOS); - ses->serverOS = kzalloc(len + 1, GFP_KERNEL); - strncpy(ses->serverOS,bcc_ptr, len); - - bcc_ptr += len; - bcc_ptr[0] = 0; /* null terminate the string */ - bcc_ptr++; - - len = strnlen(bcc_ptr, 1024); - kfree(ses->serverNOS); - ses->serverNOS = kzalloc(len+1, - GFP_KERNEL); - strncpy(ses->serverNOS, - bcc_ptr, len); - bcc_ptr += len; - bcc_ptr[0] = 0; - bcc_ptr++; - - len = strnlen(bcc_ptr, 1024); - kfree(ses->serverDomain); - ses->serverDomain = - kzalloc(len+1, - GFP_KERNEL); - strncpy(ses->serverDomain, - bcc_ptr, len); - bcc_ptr += len; - bcc_ptr[0] = 0; - bcc_ptr++; - } else - cFYI(1, ("field of length %d " - "extends beyond end of smb ", - len)); - } - } else { - cERROR(1, ("Security Blob extends beyond end " - "of SMB")); - } - } else { - cERROR(1, ("No session structure passed in.")); - } - } else { - cERROR(1, ("Invalid Word count %d: ", - smb_buffer_response->WordCount)); - rc = -EIO; - } - - cifs_buf_release(smb_buffer); - - return rc; -} - int CIFSTCon(unsigned int xid, struct cifsSesInfo *ses, const char *tree, struct cifsTconInfo *tcon, @@ -3638,7 +2623,7 @@ CIFSTCon(unsigned int xid, struct cifsSesInfo *ses, TCONX_RSP *pSMBr; unsigned char *bcc_ptr; int rc = 0; - int length; + int length, bytes_left; __u16 count; if (ses == NULL) @@ -3726,14 +2711,22 @@ CIFSTCon(unsigned int xid, struct cifsSesInfo *ses, rc = SendReceive(xid, ses, smb_buffer, smb_buffer_response, &length, CIFS_STD_OP); - /* if (rc) rc = map_smb_to_linux_error(smb_buffer_response); */ /* above now done in SendReceive */ if ((rc == 0) && (tcon != NULL)) { + bool is_unicode; + tcon->tidStatus = CifsGood; tcon->need_reconnect = false; tcon->tid = smb_buffer_response->Tid; bcc_ptr = pByteArea(smb_buffer_response); - length = strnlen(bcc_ptr, BCC(smb_buffer_response) - 2); + bytes_left = BCC(smb_buffer_response); + length = strnlen(bcc_ptr, bytes_left - 2); + if (smb_buffer->Flags2 & SMBFLG2_UNICODE) + is_unicode = true; + else + is_unicode = false; + + /* skip service field (NB: this field is always ASCII) */ if (length == 3) { if ((bcc_ptr[0] == 'I') && (bcc_ptr[1] == 'P') && @@ -3748,39 +2741,16 @@ CIFSTCon(unsigned int xid, struct cifsSesInfo *ses, } } bcc_ptr += length + 1; + bytes_left -= (length + 1); strncpy(tcon->treeName, tree, MAX_TREE_SIZE); - if (smb_buffer->Flags2 & SMBFLG2_UNICODE) { - length = UniStrnlen((wchar_t *) bcc_ptr, 512); - if ((bcc_ptr + (2 * length)) - - pByteArea(smb_buffer_response) <= - BCC(smb_buffer_response)) { - kfree(tcon->nativeFileSystem); - tcon->nativeFileSystem = - kzalloc((4 * length) + 2, GFP_KERNEL); - if (tcon->nativeFileSystem) { - cifs_strfromUCS_le( - tcon->nativeFileSystem, - (__le16 *) bcc_ptr, - length, nls_codepage); - cFYI(1, ("nativeFileSystem=%s", - tcon->nativeFileSystem)); - } - } - /* else do not bother copying these information fields*/ - } else { - length = strnlen(bcc_ptr, 1024); - if ((bcc_ptr + length) - - pByteArea(smb_buffer_response) <= - BCC(smb_buffer_response)) { - kfree(tcon->nativeFileSystem); - tcon->nativeFileSystem = - kzalloc(length + 1, GFP_KERNEL); - if (tcon->nativeFileSystem) - strncpy(tcon->nativeFileSystem, bcc_ptr, - length); - } - /* else do not bother copying these information fields*/ - } + + /* mostly informational -- no need to fail on error here */ + tcon->nativeFileSystem = cifs_strndup_from_ucs(bcc_ptr, + bytes_left, is_unicode, + nls_codepage); + + cFYI(1, ("nativeFileSystem=%s", tcon->nativeFileSystem)); + if ((smb_buffer_response->WordCount == 3) || (smb_buffer_response->WordCount == 7)) /* field is in same location */ @@ -3819,8 +2789,6 @@ int cifs_setup_session(unsigned int xid, struct cifsSesInfo *pSesInfo, struct nls_table *nls_info) { int rc = 0; - char ntlm_session_key[CIFS_SESS_KEY_SIZE]; - bool ntlmv2_flag = false; int first_time = 0; struct TCP_Server_Info *server = pSesInfo->server; @@ -3852,83 +2820,19 @@ int cifs_setup_session(unsigned int xid, struct cifsSesInfo *pSesInfo, pSesInfo->capabilities = server->capabilities; if (linuxExtEnabled == 0) pSesInfo->capabilities &= (~CAP_UNIX); - /* pSesInfo->sequence_number = 0;*/ + cFYI(1, ("Security Mode: 0x%x Capabilities: 0x%x TimeAdjust: %d", server->secMode, server->capabilities, server->timeAdj)); - if (experimEnabled < 2) - rc = CIFS_SessSetup(xid, pSesInfo, first_time, nls_info); - else if (extended_security - && (pSesInfo->capabilities & CAP_EXTENDED_SECURITY) - && (server->secType == NTLMSSP)) { - rc = -EOPNOTSUPP; - } else if (extended_security - && (pSesInfo->capabilities & CAP_EXTENDED_SECURITY) - && (server->secType == RawNTLMSSP)) { - cFYI(1, ("NTLMSSP sesssetup")); - rc = CIFSNTLMSSPNegotiateSessSetup(xid, pSesInfo, &ntlmv2_flag, - nls_info); - if (!rc) { - if (ntlmv2_flag) { - char *v2_response; - cFYI(1, ("more secure NTLM ver2 hash")); - if (CalcNTLMv2_partial_mac_key(pSesInfo, - nls_info)) { - rc = -ENOMEM; - goto ss_err_exit; - } else - v2_response = kmalloc(16 + 64 /* blob*/, - GFP_KERNEL); - if (v2_response) { - CalcNTLMv2_response(pSesInfo, - v2_response); - /* if (first_time) - cifs_calculate_ntlmv2_mac_key */ - kfree(v2_response); - /* BB Put dummy sig in SessSetup PDU? */ - } else { - rc = -ENOMEM; - goto ss_err_exit; - } - - } else { - SMBNTencrypt(pSesInfo->password, - server->cryptKey, - ntlm_session_key); - - if (first_time) - cifs_calculate_mac_key( - &server->mac_signing_key, - ntlm_session_key, - pSesInfo->password); - } - /* for better security the weaker lanman hash not sent - in AuthSessSetup so we no longer calculate it */ - - rc = CIFSNTLMSSPAuthSessSetup(xid, pSesInfo, - ntlm_session_key, - ntlmv2_flag, - nls_info); - } - } else { /* old style NTLM 0.12 session setup */ - SMBNTencrypt(pSesInfo->password, server->cryptKey, - ntlm_session_key); - - if (first_time) - cifs_calculate_mac_key(&server->mac_signing_key, - ntlm_session_key, - pSesInfo->password); - - rc = CIFSSessSetup(xid, pSesInfo, ntlm_session_key, nls_info); - } + rc = CIFS_SessSetup(xid, pSesInfo, first_time, nls_info); if (rc) { cERROR(1, ("Send error in SessSetup = %d", rc)); } else { cFYI(1, ("CIFS Session Established successfully")); - spin_lock(&GlobalMid_Lock); - pSesInfo->status = CifsGood; - pSesInfo->need_reconnect = false; - spin_unlock(&GlobalMid_Lock); + spin_lock(&GlobalMid_Lock); + pSesInfo->status = CifsGood; + pSesInfo->need_reconnect = false; + spin_unlock(&GlobalMid_Lock); } ss_err_exit: diff --git a/fs/cifs/dir.c b/fs/cifs/dir.c index 461750e..11431ed 100644 --- a/fs/cifs/dir.c +++ b/fs/cifs/dir.c @@ -281,6 +281,7 @@ cifs_create(struct inode *inode, struct dentry *direntry, int mode, int create_options = CREATE_NOT_DIR; int oplock = 0; int oflags; + bool posix_create = false; /* * BB below access is probably too much for mknod to request * but we have to do query and setpathinfo so requesting @@ -328,11 +329,13 @@ cifs_create(struct inode *inode, struct dentry *direntry, int mode, negotation. EREMOTE indicates DFS junction, which is not handled in posix open */ - if ((rc == 0) && (newinode == NULL)) - goto cifs_create_get_file_info; /* query inode info */ - else if (rc == 0) /* success, no need to query */ - goto cifs_create_set_dentry; - else if ((rc != -EIO) && (rc != -EREMOTE) && + if (rc == 0) { + posix_create = true; + if (newinode == NULL) /* query inode info */ + goto cifs_create_get_file_info; + else /* success, no need to query */ + goto cifs_create_set_dentry; + } else if ((rc != -EIO) && (rc != -EREMOTE) && (rc != -EOPNOTSUPP)) /* path not found or net err */ goto cifs_create_out; /* else fallthrough to retry, using older open call, this is @@ -464,7 +467,7 @@ cifs_create_set_dentry: if ((nd == NULL) || (!(nd->flags & LOOKUP_OPEN))) { /* mknod case - do not leave file open */ CIFSSMBClose(xid, tcon, fileHandle); - } else if (newinode) { + } else if (!(posix_create) && (newinode)) { cifs_fill_fileinfo(newinode, fileHandle, cifs_sb->tcon, write_only); } diff --git a/fs/cifs/file.c b/fs/cifs/file.c index 50ca088..38c06f8 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -129,15 +129,12 @@ static inline int cifs_posix_open_inode_helper(struct inode *inode, struct file *file, struct cifsInodeInfo *pCifsInode, struct cifsFileInfo *pCifsFile, int oplock, u16 netfid) { - struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); -/* struct timespec temp; */ /* BB REMOVEME BB */ file->private_data = kmalloc(sizeof(struct cifsFileInfo), GFP_KERNEL); if (file->private_data == NULL) return -ENOMEM; pCifsFile = cifs_init_private(file->private_data, inode, file, netfid); write_lock(&GlobalSMBSeslock); - list_add(&pCifsFile->tlist, &cifs_sb->tcon->openFileList); pCifsInode = CIFS_I(file->f_path.dentry->d_inode); if (pCifsInode == NULL) { @@ -145,17 +142,6 @@ static inline int cifs_posix_open_inode_helper(struct inode *inode, return -EINVAL; } - /* want handles we can use to read with first - in the list so we do not have to walk the - list to search for one in write_begin */ - if ((file->f_flags & O_ACCMODE) == O_WRONLY) { - list_add_tail(&pCifsFile->flist, - &pCifsInode->openFileList); - } else { - list_add(&pCifsFile->flist, - &pCifsInode->openFileList); - } - if (pCifsInode->clientCanCacheRead) { /* we have the inode open somewhere else no need to discard cache data */ diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c index f36b4e4..9c869a6 100644 --- a/fs/cifs/inode.c +++ b/fs/cifs/inode.c @@ -962,13 +962,21 @@ undo_setattr: goto out_close; } + +/* + * If dentry->d_inode is null (usually meaning the cached dentry + * is a negative dentry) then we would attempt a standard SMB delete, but + * if that fails we can not attempt the fall back mechanisms on EACESS + * but will return the EACESS to the caller. Note that the VFS does not call + * unlink on negative dentries currently. + */ int cifs_unlink(struct inode *dir, struct dentry *dentry) { int rc = 0; int xid; char *full_path = NULL; struct inode *inode = dentry->d_inode; - struct cifsInodeInfo *cifsInode = CIFS_I(inode); + struct cifsInodeInfo *cifs_inode; struct super_block *sb = dir->i_sb; struct cifs_sb_info *cifs_sb = CIFS_SB(sb); struct cifsTconInfo *tcon = cifs_sb->tcon; @@ -1012,7 +1020,7 @@ psx_del_no_retry: rc = cifs_rename_pending_delete(full_path, dentry, xid); if (rc == 0) drop_nlink(inode); - } else if (rc == -EACCES && dosattr == 0) { + } else if ((rc == -EACCES) && (dosattr == 0) && inode) { attrs = kzalloc(sizeof(*attrs), GFP_KERNEL); if (attrs == NULL) { rc = -ENOMEM; @@ -1020,7 +1028,8 @@ psx_del_no_retry: } /* try to reset dos attributes */ - origattr = cifsInode->cifsAttrs; + cifs_inode = CIFS_I(inode); + origattr = cifs_inode->cifsAttrs; if (origattr == 0) origattr |= ATTR_NORMAL; dosattr = origattr & ~ATTR_READONLY; @@ -1041,13 +1050,13 @@ psx_del_no_retry: out_reval: if (inode) { - cifsInode = CIFS_I(inode); - cifsInode->time = 0; /* will force revalidate to get info + cifs_inode = CIFS_I(inode); + cifs_inode->time = 0; /* will force revalidate to get info when needed */ inode->i_ctime = current_fs_time(sb); } dir->i_ctime = dir->i_mtime = current_fs_time(sb); - cifsInode = CIFS_I(dir); + cifs_inode = CIFS_I(dir); CIFS_I(dir)->time = 0; /* force revalidate of dir as well */ kfree(full_path); diff --git a/fs/cifs/link.c b/fs/cifs/link.c index 63f6440..ea9d11e 100644 --- a/fs/cifs/link.c +++ b/fs/cifs/link.c @@ -119,16 +119,11 @@ cifs_follow_link(struct dentry *direntry, struct nameidata *nd) full_path = build_path_from_dentry(direntry); if (!full_path) - goto out_no_free; + goto out; cFYI(1, ("Full path: %s inode = 0x%p", full_path, inode)); cifs_sb = CIFS_SB(inode->i_sb); pTcon = cifs_sb->tcon; - target_path = kmalloc(PATH_MAX, GFP_KERNEL); - if (!target_path) { - target_path = ERR_PTR(-ENOMEM); - goto out; - } /* We could change this to: if (pTcon->unix_ext) @@ -138,8 +133,7 @@ cifs_follow_link(struct dentry *direntry, struct nameidata *nd) if (pTcon->ses->capabilities & CAP_UNIX) rc = CIFSSMBUnixQuerySymLink(xid, pTcon, full_path, - target_path, - PATH_MAX-1, + &target_path, cifs_sb->local_nls); else { /* BB add read reparse point symlink code here */ @@ -148,22 +142,16 @@ cifs_follow_link(struct dentry *direntry, struct nameidata *nd) /* BB Add MAC style xsymlink check here if enabled */ } - if (rc == 0) { - -/* BB Add special case check for Samba DFS symlinks */ - - target_path[PATH_MAX-1] = 0; - } else { + if (rc != 0) { kfree(target_path); target_path = ERR_PTR(rc); } -out: kfree(full_path); -out_no_free: +out: FreeXid(xid); nd_set_link(nd, target_path); - return NULL; /* No cookie */ + return NULL; } int @@ -224,98 +212,6 @@ cifs_symlink(struct inode *inode, struct dentry *direntry, const char *symname) return rc; } -int -cifs_readlink(struct dentry *direntry, char __user *pBuffer, int buflen) -{ - struct inode *inode = direntry->d_inode; - int rc = -EACCES; - int xid; - int oplock = 0; - struct cifs_sb_info *cifs_sb; - struct cifsTconInfo *pTcon; - char *full_path = NULL; - char *tmpbuffer; - int len; - __u16 fid; - - xid = GetXid(); - cifs_sb = CIFS_SB(inode->i_sb); - pTcon = cifs_sb->tcon; - -/* BB would it be safe against deadlock to grab this sem - even though rename itself grabs the sem and calls lookup? */ -/* mutex_lock(&inode->i_sb->s_vfs_rename_mutex);*/ - full_path = build_path_from_dentry(direntry); -/* mutex_unlock(&inode->i_sb->s_vfs_rename_mutex);*/ - - if (full_path == NULL) { - FreeXid(xid); - return -ENOMEM; - } - - cFYI(1, - ("Full path: %s inode = 0x%p pBuffer = 0x%p buflen = %d", - full_path, inode, pBuffer, buflen)); - if (buflen > PATH_MAX) - len = PATH_MAX; - else - len = buflen; - tmpbuffer = kmalloc(len, GFP_KERNEL); - if (tmpbuffer == NULL) { - kfree(full_path); - FreeXid(xid); - return -ENOMEM; - } - -/* BB add read reparse point symlink code and - Unix extensions symlink code here BB */ -/* We could disable this based on pTcon->unix_ext flag instead ... but why? */ - if (cifs_sb->tcon->ses->capabilities & CAP_UNIX) - rc = CIFSSMBUnixQuerySymLink(xid, pTcon, full_path, - tmpbuffer, - len - 1, - cifs_sb->local_nls); - else if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_UNX_EMUL) { - cERROR(1, ("SFU style symlinks not implemented yet")); - /* add open and read as in fs/cifs/inode.c */ - } else { - rc = CIFSSMBOpen(xid, pTcon, full_path, FILE_OPEN, GENERIC_READ, - OPEN_REPARSE_POINT, &fid, &oplock, NULL, - cifs_sb->local_nls, - cifs_sb->mnt_cifs_flags & - CIFS_MOUNT_MAP_SPECIAL_CHR); - if (!rc) { - rc = CIFSSMBQueryReparseLinkInfo(xid, pTcon, full_path, - tmpbuffer, - len - 1, - fid, - cifs_sb->local_nls); - if (CIFSSMBClose(xid, pTcon, fid)) { - cFYI(1, ("Error closing junction point " - "(open for ioctl)")); - } - /* If it is a DFS junction earlier we would have gotten - PATH_NOT_COVERED returned from server so we do - not need to request the DFS info here */ - } - } - /* BB Anything else to do to handle recursive links? */ - /* BB Should we be using page ops here? */ - - /* BB null terminate returned string in pBuffer? BB */ - if (rc == 0) { - rc = vfs_readlink(direntry, pBuffer, len, tmpbuffer); - cFYI(1, - ("vfs_readlink called from cifs_readlink returned %d", - rc)); - } - - kfree(tmpbuffer); - kfree(full_path); - FreeXid(xid); - return rc; -} - void cifs_put_link(struct dentry *direntry, struct nameidata *nd, void *cookie) { char *p = nd_get_link(nd); diff --git a/fs/cifs/misc.c b/fs/cifs/misc.c index 4c89c57..e079a91 100644 --- a/fs/cifs/misc.c +++ b/fs/cifs/misc.c @@ -635,77 +635,6 @@ dump_smb(struct smb_hdr *smb_buf, int smb_buf_length) return; } -/* Windows maps these to the user defined 16 bit Unicode range since they are - reserved symbols (along with \ and /), otherwise illegal to store - in filenames in NTFS */ -#define UNI_ASTERIK (__u16) ('*' + 0xF000) -#define UNI_QUESTION (__u16) ('?' + 0xF000) -#define UNI_COLON (__u16) (':' + 0xF000) -#define UNI_GRTRTHAN (__u16) ('>' + 0xF000) -#define UNI_LESSTHAN (__u16) ('<' + 0xF000) -#define UNI_PIPE (__u16) ('|' + 0xF000) -#define UNI_SLASH (__u16) ('\\' + 0xF000) - -/* Convert 16 bit Unicode pathname from wire format to string in current code - page. Conversion may involve remapping up the seven characters that are - only legal in POSIX-like OS (if they are present in the string). Path - names are little endian 16 bit Unicode on the wire */ -int -cifs_convertUCSpath(char *target, const __le16 *source, int maxlen, - const struct nls_table *cp) -{ - int i, j, len; - __u16 src_char; - - for (i = 0, j = 0; i < maxlen; i++) { - src_char = le16_to_cpu(source[i]); - switch (src_char) { - case 0: - goto cUCS_out; /* BB check this BB */ - case UNI_COLON: - target[j] = ':'; - break; - case UNI_ASTERIK: - target[j] = '*'; - break; - case UNI_QUESTION: - target[j] = '?'; - break; - /* BB We can not handle remapping slash until - all the calls to build_path_from_dentry - are modified, as they use slash as separator BB */ - /* case UNI_SLASH: - target[j] = '\\'; - break;*/ - case UNI_PIPE: - target[j] = '|'; - break; - case UNI_GRTRTHAN: - target[j] = '>'; - break; - case UNI_LESSTHAN: - target[j] = '<'; - break; - default: - len = cp->uni2char(src_char, &target[j], - NLS_MAX_CHARSET_SIZE); - if (len > 0) { - j += len; - continue; - } else { - target[j] = '?'; - } - } - j++; - /* make sure we do not overrun callers allocated temp buffer */ - if (j >= (2 * NAME_MAX)) - break; - } -cUCS_out: - target[j] = 0; - return j; -} - /* Convert 16 bit Unicode pathname to wire format from string in current code page. Conversion may involve remapping up the seven characters that are only legal in POSIX-like OS (if they are present in the string). Path diff --git a/fs/cifs/netmisc.c b/fs/cifs/netmisc.c index 8703d68..e2fe998 100644 --- a/fs/cifs/netmisc.c +++ b/fs/cifs/netmisc.c @@ -79,6 +79,7 @@ static const struct smb_to_posix_error mapping_table_ERRDOS[] = { {ErrQuota, -EDQUOT}, {ErrNotALink, -ENOLINK}, {ERRnetlogonNotStarted, -ENOPROTOOPT}, + {ERRsymlink, -EOPNOTSUPP}, {ErrTooManyLinks, -EMLINK}, {0, 0} }; @@ -714,6 +715,7 @@ static const struct { ERRDOS, ERRnoaccess, 0xc000028f}, { ERRDOS, ERRnoaccess, 0xc0000290}, { ERRDOS, ERRbadfunc, 0xc000029c}, { + ERRDOS, ERRsymlink, NT_STATUS_STOPPED_ON_SYMLINK}, { ERRDOS, ERRinvlevel, 0x007c0001}, }; /***************************************************************************** diff --git a/fs/cifs/nterr.h b/fs/cifs/nterr.h index 588abbb..25726736 100644 --- a/fs/cifs/nterr.h +++ b/fs/cifs/nterr.h @@ -35,8 +35,6 @@ struct nt_err_code_struct { extern const struct nt_err_code_struct nt_errs[]; /* Win32 Status codes. */ - -#define STATUS_BUFFER_OVERFLOW 0x80000005 #define STATUS_MORE_ENTRIES 0x0105 #define ERROR_INVALID_PARAMETER 0x0057 #define ERROR_INSUFFICIENT_BUFFER 0x007a @@ -50,6 +48,13 @@ extern const struct nt_err_code_struct nt_errs[]; #define STATUS_SOME_UNMAPPED 0x0107 #define STATUS_BUFFER_OVERFLOW 0x80000005 #define NT_STATUS_NO_MORE_ENTRIES 0x8000001a +#define NT_STATUS_MEDIA_CHANGED 0x8000001c +#define NT_STATUS_END_OF_MEDIA 0x8000001e +#define NT_STATUS_MEDIA_CHECK 0x80000020 +#define NT_STATUS_NO_DATA_DETECTED 0x8000001c +#define NT_STATUS_STOPPED_ON_SYMLINK 0x8000002d +#define NT_STATUS_DEVICE_REQUIRES_CLEANING 0x80000288 +#define NT_STATUS_DEVICE_DOOR_OPEN 0x80000288 #define NT_STATUS_UNSUCCESSFUL 0xC0000000 | 0x0001 #define NT_STATUS_NOT_IMPLEMENTED 0xC0000000 | 0x0002 #define NT_STATUS_INVALID_INFO_CLASS 0xC0000000 | 0x0003 diff --git a/fs/cifs/ntlmssp.h b/fs/cifs/ntlmssp.h index c377d80..49c9a4e 100644 --- a/fs/cifs/ntlmssp.h +++ b/fs/cifs/ntlmssp.h @@ -27,29 +27,39 @@ #define UnknownMessage cpu_to_le32(8) /* Negotiate Flags */ -#define NTLMSSP_NEGOTIATE_UNICODE 0x01 /* Text strings are in unicode */ -#define NTLMSSP_NEGOTIATE_OEM 0x02 /* Text strings are in OEM */ -#define NTLMSSP_REQUEST_TARGET 0x04 /* Server return its auth realm */ -#define NTLMSSP_NEGOTIATE_SIGN 0x0010 /* Request signature capability */ -#define NTLMSSP_NEGOTIATE_SEAL 0x0020 /* Request confidentiality */ -#define NTLMSSP_NEGOTIATE_DGRAM 0x0040 -#define NTLMSSP_NEGOTIATE_LM_KEY 0x0080 /* Sign/seal use LM session key */ -#define NTLMSSP_NEGOTIATE_NTLM 0x0200 /* NTLM authentication */ -#define NTLMSSP_NEGOTIATE_DOMAIN_SUPPLIED 0x1000 +#define NTLMSSP_NEGOTIATE_UNICODE 0x01 /* Text strings are unicode */ +#define NTLMSSP_NEGOTIATE_OEM 0x02 /* Text strings are in OEM */ +#define NTLMSSP_REQUEST_TARGET 0x04 /* Srv returns its auth realm */ +/* define reserved9 0x08 */ +#define NTLMSSP_NEGOTIATE_SIGN 0x0010 /* Request signing capability */ +#define NTLMSSP_NEGOTIATE_SEAL 0x0020 /* Request confidentiality */ +#define NTLMSSP_NEGOTIATE_DGRAM 0x0040 +#define NTLMSSP_NEGOTIATE_LM_KEY 0x0080 /* Use LM session key */ +/* defined reserved 8 0x0100 */ +#define NTLMSSP_NEGOTIATE_NTLM 0x0200 /* NTLM authentication */ +#define NTLMSSP_NEGOTIATE_NT_ONLY 0x0400 /* Lanman not allowed */ +#define NTLMSSP_ANONYMOUS 0x0800 +#define NTLMSSP_NEGOTIATE_DOMAIN_SUPPLIED 0x1000 /* reserved6 */ #define NTLMSSP_NEGOTIATE_WORKSTATION_SUPPLIED 0x2000 -#define NTLMSSP_NEGOTIATE_LOCAL_CALL 0x4000 /* client/server on same machine */ -#define NTLMSSP_NEGOTIATE_ALWAYS_SIGN 0x8000 /* Sign for all security levels */ -#define NTLMSSP_TARGET_TYPE_DOMAIN 0x10000 -#define NTLMSSP_TARGET_TYPE_SERVER 0x20000 -#define NTLMSSP_TARGET_TYPE_SHARE 0x40000 -#define NTLMSSP_NEGOTIATE_NTLMV2 0x80000 -#define NTLMSSP_REQUEST_INIT_RESP 0x100000 -#define NTLMSSP_REQUEST_ACCEPT_RESP 0x200000 -#define NTLMSSP_REQUEST_NOT_NT_KEY 0x400000 +#define NTLMSSP_NEGOTIATE_LOCAL_CALL 0x4000 /* client/server same machine */ +#define NTLMSSP_NEGOTIATE_ALWAYS_SIGN 0x8000 /* Sign. All security levels */ +#define NTLMSSP_TARGET_TYPE_DOMAIN 0x10000 +#define NTLMSSP_TARGET_TYPE_SERVER 0x20000 +#define NTLMSSP_TARGET_TYPE_SHARE 0x40000 +#define NTLMSSP_NEGOTIATE_EXTENDED_SEC 0x80000 /* NB:not related to NTLMv2 pwd*/ +/* #define NTLMSSP_REQUEST_INIT_RESP 0x100000 */ +#define NTLMSSP_NEGOTIATE_IDENTIFY 0x100000 +#define NTLMSSP_REQUEST_ACCEPT_RESP 0x200000 /* reserved5 */ +#define NTLMSSP_REQUEST_NON_NT_KEY 0x400000 #define NTLMSSP_NEGOTIATE_TARGET_INFO 0x800000 -#define NTLMSSP_NEGOTIATE_128 0x20000000 -#define NTLMSSP_NEGOTIATE_KEY_XCH 0x40000000 -#define NTLMSSP_NEGOTIATE_56 0x80000000 +/* #define reserved4 0x1000000 */ +#define NTLMSSP_NEGOTIATE_VERSION 0x2000000 /* we do not set */ +/* #define reserved3 0x4000000 */ +/* #define reserved2 0x8000000 */ +/* #define reserved1 0x10000000 */ +#define NTLMSSP_NEGOTIATE_128 0x20000000 +#define NTLMSSP_NEGOTIATE_KEY_XCH 0x40000000 +#define NTLMSSP_NEGOTIATE_56 0x80000000 /* Although typedefs are not commonly used for structure definitions */ /* in the Linux kernel, in this particular case they are useful */ @@ -60,32 +70,36 @@ typedef struct _SECURITY_BUFFER { __le16 Length; __le16 MaximumLength; - __le32 Buffer; /* offset to buffer */ + __le32 BufferOffset; /* offset to buffer */ } __attribute__((packed)) SECURITY_BUFFER; typedef struct _NEGOTIATE_MESSAGE { __u8 Signature[sizeof(NTLMSSP_SIGNATURE)]; - __le32 MessageType; /* 1 */ + __le32 MessageType; /* NtLmNegotiate = 1 */ __le32 NegotiateFlags; SECURITY_BUFFER DomainName; /* RFC 1001 style and ASCII */ SECURITY_BUFFER WorkstationName; /* RFC 1001 and ASCII */ + /* SECURITY_BUFFER for version info not present since we + do not set the version is present flag */ char DomainString[0]; /* followed by WorkstationString */ } __attribute__((packed)) NEGOTIATE_MESSAGE, *PNEGOTIATE_MESSAGE; typedef struct _CHALLENGE_MESSAGE { __u8 Signature[sizeof(NTLMSSP_SIGNATURE)]; - __le32 MessageType; /* 2 */ + __le32 MessageType; /* NtLmChallenge = 2 */ SECURITY_BUFFER TargetName; __le32 NegotiateFlags; __u8 Challenge[CIFS_CRYPTO_KEY_SIZE]; __u8 Reserved[8]; SECURITY_BUFFER TargetInfoArray; + /* SECURITY_BUFFER for version info not present since we + do not set the version is present flag */ } __attribute__((packed)) CHALLENGE_MESSAGE, *PCHALLENGE_MESSAGE; typedef struct _AUTHENTICATE_MESSAGE { - __u8 Signature[sizeof (NTLMSSP_SIGNATURE)]; - __le32 MessageType; /* 3 */ + __u8 Signature[sizeof(NTLMSSP_SIGNATURE)]; + __le32 MessageType; /* NtLmsAuthenticate = 3 */ SECURITY_BUFFER LmChallengeResponse; SECURITY_BUFFER NtChallengeResponse; SECURITY_BUFFER DomainName; @@ -93,5 +107,7 @@ typedef struct _AUTHENTICATE_MESSAGE { SECURITY_BUFFER WorkstationName; SECURITY_BUFFER SessionKey; __le32 NegotiateFlags; + /* SECURITY_BUFFER for version info not present since we + do not set the version is present flag */ char UserString[0]; } __attribute__((packed)) AUTHENTICATE_MESSAGE, *PAUTHENTICATE_MESSAGE; diff --git a/fs/cifs/readdir.c b/fs/cifs/readdir.c index 1a8be62..964e097 100644 --- a/fs/cifs/readdir.c +++ b/fs/cifs/readdir.c @@ -31,6 +31,13 @@ #include "cifs_fs_sb.h" #include "cifsfs.h" +/* + * To be safe - for UCS to UTF-8 with strings loaded with the rare long + * characters alloc more to account for such multibyte target UTF-8 + * characters. + */ +#define UNICODE_NAME_MAX ((4 * NAME_MAX) + 2) + #ifdef CONFIG_CIFS_DEBUG2 static void dump_cifs_file_struct(struct file *file, char *label) { @@ -438,6 +445,38 @@ static void unix_fill_in_inode(struct inode *tmp_inode, } } +/* BB eventually need to add the following helper function to + resolve NT_STATUS_STOPPED_ON_SYMLINK return code when + we try to do FindFirst on (NTFS) directory symlinks */ +/* +int get_symlink_reparse_path(char *full_path, struct cifs_sb_info *cifs_sb, + int xid) +{ + __u16 fid; + int len; + int oplock = 0; + int rc; + struct cifsTconInfo *ptcon = cifs_sb->tcon; + char *tmpbuffer; + + rc = CIFSSMBOpen(xid, ptcon, full_path, FILE_OPEN, GENERIC_READ, + OPEN_REPARSE_POINT, &fid, &oplock, NULL, + cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); + if (!rc) { + tmpbuffer = kmalloc(maxpath); + rc = CIFSSMBQueryReparseLinkInfo(xid, ptcon, full_path, + tmpbuffer, + maxpath -1, + fid, + cifs_sb->local_nls); + if (CIFSSMBClose(xid, ptcon, fid)) { + cFYI(1, ("Error closing temporary reparsepoint open)")); + } + } +} + */ + static int initiate_cifs_search(const int xid, struct file *file) { int rc = 0; @@ -493,7 +532,10 @@ ffirst_retry: CIFS_MOUNT_MAP_SPECIAL_CHR, CIFS_DIR_SEP(cifs_sb)); if (rc == 0) cifsFile->invalidHandle = false; - if ((rc == -EOPNOTSUPP) && + /* BB add following call to handle readdir on new NTFS symlink errors + else if STATUS_STOPPED_ON_SYMLINK + call get_symlink_reparse_path and retry with new path */ + else if ((rc == -EOPNOTSUPP) && (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SERVER_INUM)) { cifs_sb->mnt_cifs_flags &= ~CIFS_MOUNT_SERVER_INUM; goto ffirst_retry; @@ -822,7 +864,7 @@ static int find_cifs_entry(const int xid, struct cifsTconInfo *pTcon, /* inode num, inode type and filename returned */ static int cifs_get_name_from_search_buf(struct qstr *pqst, char *current_entry, __u16 level, unsigned int unicode, - struct cifs_sb_info *cifs_sb, int max_len, __u64 *pinum) + struct cifs_sb_info *cifs_sb, unsigned int max_len, __u64 *pinum) { int rc = 0; unsigned int len = 0; @@ -881,14 +923,12 @@ static int cifs_get_name_from_search_buf(struct qstr *pqst, } if (unicode) { - /* BB fixme - test with long names */ - /* Note converted filename can be longer than in unicode */ - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR) - pqst->len = cifs_convertUCSpath((char *)pqst->name, - (__le16 *)filename, len/2, nlt); - else - pqst->len = cifs_strfromUCS_le((char *)pqst->name, - (__le16 *)filename, len/2, nlt); + pqst->len = cifs_from_ucs2((char *) pqst->name, + (__le16 *) filename, + UNICODE_NAME_MAX, + min(len, max_len), nlt, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); } else { pqst->name = filename; pqst->len = len; @@ -898,8 +938,8 @@ static int cifs_get_name_from_search_buf(struct qstr *pqst, return rc; } -static int cifs_filldir(char *pfindEntry, struct file *file, - filldir_t filldir, void *direntry, char *scratch_buf, int max_len) +static int cifs_filldir(char *pfindEntry, struct file *file, filldir_t filldir, + void *direntry, char *scratch_buf, unsigned int max_len) { int rc = 0; struct qstr qstring; @@ -996,7 +1036,7 @@ int cifs_readdir(struct file *file, void *direntry, filldir_t filldir) int num_to_fill = 0; char *tmp_buf = NULL; char *end_of_smb; - int max_len; + unsigned int max_len; xid = GetXid(); @@ -1070,11 +1110,7 @@ int cifs_readdir(struct file *file, void *direntry, filldir_t filldir) cifsFile->srch_inf.ntwrk_buf_start); end_of_smb = cifsFile->srch_inf.ntwrk_buf_start + max_len; - /* To be safe - for UCS to UTF-8 with strings loaded - with the rare long characters alloc more to account for - such multibyte target UTF-8 characters. cifs_unicode.c, - which actually does the conversion, has the same limit */ - tmp_buf = kmalloc((2 * NAME_MAX) + 4, GFP_KERNEL); + tmp_buf = kmalloc(UNICODE_NAME_MAX, GFP_KERNEL); for (i = 0; (i < num_to_fill) && (rc == 0); i++) { if (current_entry == NULL) { /* evaluate whether this case is an error */ diff --git a/fs/cifs/sess.c b/fs/cifs/sess.c index c652c73..897a052 100644 --- a/fs/cifs/sess.c +++ b/fs/cifs/sess.c @@ -3,7 +3,7 @@ * * SMB/CIFS session setup handling routines * - * Copyright (c) International Business Machines Corp., 2006, 2007 + * Copyright (c) International Business Machines Corp., 2006, 2009 * Author(s): Steve French (sfrench@us.ibm.com) * * This library is free software; you can redistribute it and/or modify @@ -111,7 +111,7 @@ static __le16 get_next_vcnum(struct cifsSesInfo *ses) get_vc_num_exit: write_unlock(&cifs_tcp_ses_lock); - return le16_to_cpu(vcnum); + return cpu_to_le16(vcnum); } static __u32 cifs_ssetup_hdr(struct cifsSesInfo *ses, SESSION_SETUP_ANDX *pSMB) @@ -277,12 +277,11 @@ static void ascii_ssetup_strings(char **pbcc_area, struct cifsSesInfo *ses, *pbcc_area = bcc_ptr; } -static int decode_unicode_ssetup(char **pbcc_area, int bleft, - struct cifsSesInfo *ses, - const struct nls_table *nls_cp) +static void +decode_unicode_ssetup(char **pbcc_area, int bleft, struct cifsSesInfo *ses, + const struct nls_table *nls_cp) { - int rc = 0; - int words_left, len; + int len; char *data = *pbcc_area; cFYI(1, ("bleft %d", bleft)); @@ -300,63 +299,29 @@ static int decode_unicode_ssetup(char **pbcc_area, int bleft, ++bleft; } - words_left = bleft / 2; - - /* save off server operating system */ - len = UniStrnlen((wchar_t *) data, words_left); - - if (len >= words_left) - return rc; - kfree(ses->serverOS); - /* UTF-8 string will not grow more than four times as big as UCS-16 */ - ses->serverOS = kzalloc((4 * len) + 2 /* trailing null */, GFP_KERNEL); - if (ses->serverOS != NULL) { - cifs_strfromUCS_le(ses->serverOS, (__le16 *)data, len, nls_cp); - cFYI(1, ("serverOS=%s", ses->serverOS)); - } - data += 2 * (len + 1); - words_left -= len + 1; - - /* save off server network operating system */ - len = UniStrnlen((wchar_t *) data, words_left); - - if (len >= words_left) - return rc; + ses->serverOS = cifs_strndup_from_ucs(data, bleft, true, nls_cp); + cFYI(1, ("serverOS=%s", ses->serverOS)); + len = (UniStrnlen((wchar_t *) data, bleft / 2) * 2) + 2; + data += len; + bleft -= len; + if (bleft <= 0) + return; kfree(ses->serverNOS); - ses->serverNOS = kzalloc((4 * len) + 2 /* trailing null */, GFP_KERNEL); - if (ses->serverNOS != NULL) { - cifs_strfromUCS_le(ses->serverNOS, (__le16 *)data, len, - nls_cp); - cFYI(1, ("serverNOS=%s", ses->serverNOS)); - if (strncmp(ses->serverNOS, "NT LAN Manager 4", 16) == 0) { - cFYI(1, ("NT4 server")); - ses->flags |= CIFS_SES_NT4; - } - } - data += 2 * (len + 1); - words_left -= len + 1; - - /* save off server domain */ - len = UniStrnlen((wchar_t *) data, words_left); - - if (len > words_left) - return rc; + ses->serverNOS = cifs_strndup_from_ucs(data, bleft, true, nls_cp); + cFYI(1, ("serverNOS=%s", ses->serverNOS)); + len = (UniStrnlen((wchar_t *) data, bleft / 2) * 2) + 2; + data += len; + bleft -= len; + if (bleft <= 0) + return; kfree(ses->serverDomain); - ses->serverDomain = kzalloc((4 * len) + 2, GFP_KERNEL); - if (ses->serverDomain != NULL) { - cifs_strfromUCS_le(ses->serverDomain, (__le16 *)data, len, - nls_cp); - cFYI(1, ("serverDomain=%s", ses->serverDomain)); - } - data += 2 * (len + 1); - words_left -= len + 1; - - cFYI(1, ("words left: %d", words_left)); + ses->serverDomain = cifs_strndup_from_ucs(data, bleft, true, nls_cp); + cFYI(1, ("serverDomain=%s", ses->serverDomain)); - return rc; + return; } static int decode_ascii_ssetup(char **pbcc_area, int bleft, @@ -413,6 +378,186 @@ static int decode_ascii_ssetup(char **pbcc_area, int bleft, return rc; } +static int decode_ntlmssp_challenge(char *bcc_ptr, int blob_len, + struct cifsSesInfo *ses) +{ + CHALLENGE_MESSAGE *pblob = (CHALLENGE_MESSAGE *)bcc_ptr; + + if (blob_len < sizeof(CHALLENGE_MESSAGE)) { + cERROR(1, ("challenge blob len %d too small", blob_len)); + return -EINVAL; + } + + if (memcmp(pblob->Signature, "NTLMSSP", 8)) { + cERROR(1, ("blob signature incorrect %s", pblob->Signature)); + return -EINVAL; + } + if (pblob->MessageType != NtLmChallenge) { + cERROR(1, ("Incorrect message type %d", pblob->MessageType)); + return -EINVAL; + } + + memcpy(ses->server->cryptKey, pblob->Challenge, CIFS_CRYPTO_KEY_SIZE); + /* BB we could decode pblob->NegotiateFlags; some may be useful */ + /* In particular we can examine sign flags */ + /* BB spec says that if AvId field of MsvAvTimestamp is populated then + we must set the MIC field of the AUTHENTICATE_MESSAGE */ + + return 0; +} + +#ifdef CONFIG_CIFS_EXPERIMENTAL +/* BB Move to ntlmssp.c eventually */ + +/* We do not malloc the blob, it is passed in pbuffer, because + it is fixed size, and small, making this approach cleaner */ +static void build_ntlmssp_negotiate_blob(unsigned char *pbuffer, + struct cifsSesInfo *ses) +{ + NEGOTIATE_MESSAGE *sec_blob = (NEGOTIATE_MESSAGE *)pbuffer; + __u32 flags; + + memcpy(sec_blob->Signature, NTLMSSP_SIGNATURE, 8); + sec_blob->MessageType = NtLmNegotiate; + + /* BB is NTLMV2 session security format easier to use here? */ + flags = NTLMSSP_NEGOTIATE_56 | NTLMSSP_REQUEST_TARGET | + NTLMSSP_NEGOTIATE_128 | NTLMSSP_NEGOTIATE_UNICODE | + NTLMSSP_NEGOTIATE_NT_ONLY | NTLMSSP_NEGOTIATE_NTLM; + if (ses->server->secMode & + (SECMODE_SIGN_REQUIRED | SECMODE_SIGN_ENABLED)) + flags |= NTLMSSP_NEGOTIATE_SIGN; + if (ses->server->secMode & SECMODE_SIGN_REQUIRED) + flags |= NTLMSSP_NEGOTIATE_ALWAYS_SIGN; + + sec_blob->NegotiateFlags |= cpu_to_le32(flags); + + sec_blob->WorkstationName.BufferOffset = 0; + sec_blob->WorkstationName.Length = 0; + sec_blob->WorkstationName.MaximumLength = 0; + + /* Domain name is sent on the Challenge not Negotiate NTLMSSP request */ + sec_blob->DomainName.BufferOffset = 0; + sec_blob->DomainName.Length = 0; + sec_blob->DomainName.MaximumLength = 0; +} + +/* We do not malloc the blob, it is passed in pbuffer, because its + maximum possible size is fixed and small, making this approach cleaner. + This function returns the length of the data in the blob */ +static int build_ntlmssp_auth_blob(unsigned char *pbuffer, + struct cifsSesInfo *ses, + const struct nls_table *nls_cp, int first) +{ + AUTHENTICATE_MESSAGE *sec_blob = (AUTHENTICATE_MESSAGE *)pbuffer; + __u32 flags; + unsigned char *tmp; + char ntlm_session_key[CIFS_SESS_KEY_SIZE]; + + memcpy(sec_blob->Signature, NTLMSSP_SIGNATURE, 8); + sec_blob->MessageType = NtLmAuthenticate; + + flags = NTLMSSP_NEGOTIATE_56 | + NTLMSSP_REQUEST_TARGET | NTLMSSP_NEGOTIATE_TARGET_INFO | + NTLMSSP_NEGOTIATE_128 | NTLMSSP_NEGOTIATE_UNICODE | + NTLMSSP_NEGOTIATE_NT_ONLY | NTLMSSP_NEGOTIATE_NTLM; + if (ses->server->secMode & + (SECMODE_SIGN_REQUIRED | SECMODE_SIGN_ENABLED)) + flags |= NTLMSSP_NEGOTIATE_SIGN; + if (ses->server->secMode & SECMODE_SIGN_REQUIRED) + flags |= NTLMSSP_NEGOTIATE_ALWAYS_SIGN; + + tmp = pbuffer + sizeof(AUTHENTICATE_MESSAGE); + sec_blob->NegotiateFlags |= cpu_to_le32(flags); + + sec_blob->LmChallengeResponse.BufferOffset = + cpu_to_le32(sizeof(AUTHENTICATE_MESSAGE)); + sec_blob->LmChallengeResponse.Length = 0; + sec_blob->LmChallengeResponse.MaximumLength = 0; + + /* calculate session key, BB what about adding similar ntlmv2 path? */ + SMBNTencrypt(ses->password, ses->server->cryptKey, ntlm_session_key); + if (first) + cifs_calculate_mac_key(&ses->server->mac_signing_key, + ntlm_session_key, ses->password); + + memcpy(tmp, ntlm_session_key, CIFS_SESS_KEY_SIZE); + sec_blob->NtChallengeResponse.BufferOffset = cpu_to_le32(tmp - pbuffer); + sec_blob->NtChallengeResponse.Length = cpu_to_le16(CIFS_SESS_KEY_SIZE); + sec_blob->NtChallengeResponse.MaximumLength = + cpu_to_le16(CIFS_SESS_KEY_SIZE); + + tmp += CIFS_SESS_KEY_SIZE; + + if (ses->domainName == NULL) { + sec_blob->DomainName.BufferOffset = cpu_to_le32(tmp - pbuffer); + sec_blob->DomainName.Length = 0; + sec_blob->DomainName.MaximumLength = 0; + tmp += 2; + } else { + int len; + len = cifs_strtoUCS((__le16 *)tmp, ses->domainName, + MAX_USERNAME_SIZE, nls_cp); + len *= 2; /* unicode is 2 bytes each */ + len += 2; /* trailing null */ + sec_blob->DomainName.BufferOffset = cpu_to_le32(tmp - pbuffer); + sec_blob->DomainName.Length = cpu_to_le16(len); + sec_blob->DomainName.MaximumLength = cpu_to_le16(len); + tmp += len; + } + + if (ses->userName == NULL) { + sec_blob->UserName.BufferOffset = cpu_to_le32(tmp - pbuffer); + sec_blob->UserName.Length = 0; + sec_blob->UserName.MaximumLength = 0; + tmp += 2; + } else { + int len; + len = cifs_strtoUCS((__le16 *)tmp, ses->userName, + MAX_USERNAME_SIZE, nls_cp); + len *= 2; /* unicode is 2 bytes each */ + len += 2; /* trailing null */ + sec_blob->UserName.BufferOffset = cpu_to_le32(tmp - pbuffer); + sec_blob->UserName.Length = cpu_to_le16(len); + sec_blob->UserName.MaximumLength = cpu_to_le16(len); + tmp += len; + } + + sec_blob->WorkstationName.BufferOffset = cpu_to_le32(tmp - pbuffer); + sec_blob->WorkstationName.Length = 0; + sec_blob->WorkstationName.MaximumLength = 0; + tmp += 2; + + sec_blob->SessionKey.BufferOffset = cpu_to_le32(tmp - pbuffer); + sec_blob->SessionKey.Length = 0; + sec_blob->SessionKey.MaximumLength = 0; + return tmp - pbuffer; +} + + +static void setup_ntlmssp_neg_req(SESSION_SETUP_ANDX *pSMB, + struct cifsSesInfo *ses) +{ + build_ntlmssp_negotiate_blob(&pSMB->req.SecurityBlob[0], ses); + pSMB->req.SecurityBlobLength = cpu_to_le16(sizeof(NEGOTIATE_MESSAGE)); + + return; +} + +static int setup_ntlmssp_auth_req(SESSION_SETUP_ANDX *pSMB, + struct cifsSesInfo *ses, + const struct nls_table *nls, int first_time) +{ + int bloblen; + + bloblen = build_ntlmssp_auth_blob(&pSMB->req.SecurityBlob[0], ses, nls, + first_time); + pSMB->req.SecurityBlobLength = cpu_to_le16(bloblen); + + return bloblen; +} +#endif + int CIFS_SessSetup(unsigned int xid, struct cifsSesInfo *ses, int first_time, const struct nls_table *nls_cp) @@ -431,6 +576,7 @@ CIFS_SessSetup(unsigned int xid, struct cifsSesInfo *ses, int first_time, __u16 action; int bytes_remaining; struct key *spnego_key = NULL; + __le32 phase = NtLmNegotiate; /* NTLMSSP, if needed, is multistage */ if (ses == NULL) return -EINVAL; @@ -438,6 +584,10 @@ CIFS_SessSetup(unsigned int xid, struct cifsSesInfo *ses, int first_time, type = ses->server->secType; cFYI(1, ("sess setup type %d", type)); +ssetup_ntlmssp_authenticate: + if (phase == NtLmChallenge) + phase = NtLmAuthenticate; /* if ntlmssp, now final phase */ + if (type == LANMAN) { #ifndef CONFIG_CIFS_WEAK_PW_HASH /* LANMAN and plaintext are less secure and off by default. @@ -651,9 +801,53 @@ CIFS_SessSetup(unsigned int xid, struct cifsSesInfo *ses, int first_time, goto ssetup_exit; #endif /* CONFIG_CIFS_UPCALL */ } else { +#ifdef CONFIG_CIFS_EXPERIMENTAL + if ((experimEnabled > 1) && (type == RawNTLMSSP)) { + if ((pSMB->req.hdr.Flags2 & SMBFLG2_UNICODE) == 0) { + cERROR(1, ("NTLMSSP requires Unicode support")); + rc = -ENOSYS; + goto ssetup_exit; + } + + cFYI(1, ("ntlmssp session setup phase %d", phase)); + pSMB->req.hdr.Flags2 |= SMBFLG2_EXT_SEC; + capabilities |= CAP_EXTENDED_SECURITY; + pSMB->req.Capabilities |= cpu_to_le32(capabilities); + if (phase == NtLmNegotiate) { + setup_ntlmssp_neg_req(pSMB, ses); + iov[1].iov_len = sizeof(NEGOTIATE_MESSAGE); + } else if (phase == NtLmAuthenticate) { + int blob_len; + blob_len = setup_ntlmssp_auth_req(pSMB, ses, + nls_cp, + first_time); + iov[1].iov_len = blob_len; + /* Make sure that we tell the server that we + are using the uid that it just gave us back + on the response (challenge) */ + smb_buf->Uid = ses->Suid; + } else { + cERROR(1, ("invalid phase %d", phase)); + rc = -ENOSYS; + goto ssetup_exit; + } + iov[1].iov_base = &pSMB->req.SecurityBlob[0]; + /* unicode strings must be word aligned */ + if ((iov[0].iov_len + iov[1].iov_len) % 2) { + *bcc_ptr = 0; + bcc_ptr++; + } + unicode_oslm_strings(&bcc_ptr, nls_cp); + } else { + cERROR(1, ("secType %d not supported!", type)); + rc = -ENOSYS; + goto ssetup_exit; + } +#else cERROR(1, ("secType %d not supported!", type)); rc = -ENOSYS; goto ssetup_exit; +#endif } iov[2].iov_base = str_area; @@ -669,12 +863,23 @@ CIFS_SessSetup(unsigned int xid, struct cifsSesInfo *ses, int first_time, /* SMB request buf freed in SendReceive2 */ cFYI(1, ("ssetup rc from sendrecv2 is %d", rc)); - if (rc) - goto ssetup_exit; pSMB = (SESSION_SETUP_ANDX *)iov[0].iov_base; smb_buf = (struct smb_hdr *)iov[0].iov_base; + if ((type == RawNTLMSSP) && (smb_buf->Status.CifsError == + cpu_to_le32(NT_STATUS_MORE_PROCESSING_REQUIRED))) { + if (phase != NtLmNegotiate) { + cERROR(1, ("Unexpected more processing error")); + goto ssetup_exit; + } + /* NTLMSSP Negotiate sent now processing challenge (response) */ + phase = NtLmChallenge; /* process ntlmssp challenge */ + rc = 0; /* MORE_PROC rc is not an error here, but expected */ + } + if (rc) + goto ssetup_exit; + if ((smb_buf->WordCount != 3) && (smb_buf->WordCount != 4)) { rc = -EIO; cERROR(1, ("bad word count %d", smb_buf->WordCount)); @@ -693,12 +898,18 @@ CIFS_SessSetup(unsigned int xid, struct cifsSesInfo *ses, int first_time, if (smb_buf->WordCount == 4) { __u16 blob_len; blob_len = le16_to_cpu(pSMB->resp.SecurityBlobLength); - bcc_ptr += blob_len; if (blob_len > bytes_remaining) { cERROR(1, ("bad security blob length %d", blob_len)); rc = -EINVAL; goto ssetup_exit; } + if (phase == NtLmChallenge) { + rc = decode_ntlmssp_challenge(bcc_ptr, blob_len, ses); + /* now goto beginning for ntlmssp authenticate phase */ + if (rc) + goto ssetup_exit; + } + bcc_ptr += blob_len; bytes_remaining -= blob_len; } @@ -709,8 +920,7 @@ CIFS_SessSetup(unsigned int xid, struct cifsSesInfo *ses, int first_time, ++bcc_ptr; --bytes_remaining; } - rc = decode_unicode_ssetup(&bcc_ptr, bytes_remaining, - ses, nls_cp); + decode_unicode_ssetup(&bcc_ptr, bytes_remaining, ses, nls_cp); } else { rc = decode_ascii_ssetup(&bcc_ptr, bytes_remaining, ses, nls_cp); @@ -728,5 +938,9 @@ ssetup_exit: } else if (resp_buf_type == CIFS_LARGE_BUFFER) cifs_buf_release(iov[0].iov_base); + /* if ntlmssp, and negotiate succeeded, proceed to authenticate phase */ + if ((phase == NtLmChallenge) && (rc == 0)) + goto ssetup_ntlmssp_authenticate; + return rc; } diff --git a/fs/cifs/smberr.h b/fs/cifs/smberr.h index 7f50e85..c5084d2 100644 --- a/fs/cifs/smberr.h +++ b/fs/cifs/smberr.h @@ -110,6 +110,7 @@ /* Below errors are used internally (do not come over the wire) for passthrough from STATUS codes to POSIX only */ +#define ERRsymlink 0xFFFD #define ErrTooManyLinks 0xFFFE /* Following error codes may be generated with the ERRSRV error class.*/ diff --git a/fs/configfs/symlink.c b/fs/configfs/symlink.c index 932a92b..c8afa6b 100644 --- a/fs/configfs/symlink.c +++ b/fs/configfs/symlink.c @@ -135,7 +135,7 @@ int configfs_symlink(struct inode *dir, struct dentry *dentry, const char *symna struct path path; struct configfs_dirent *sd; struct config_item *parent_item; - struct config_item *target_item; + struct config_item *target_item = NULL; struct config_item_type *type; ret = -EPERM; /* What lack-of-symlink returns */ diff --git a/fs/dcache.c b/fs/dcache.c index 1fcffeb..75659a6 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -481,7 +481,7 @@ restart: if ((flags & DCACHE_REFERENCED) && (dentry->d_flags & DCACHE_REFERENCED)) { dentry->d_flags &= ~DCACHE_REFERENCED; - list_move_tail(&dentry->d_lru, &referenced); + list_move(&dentry->d_lru, &referenced); spin_unlock(&dentry->d_lock); } else { list_move_tail(&dentry->d_lru, &tmp); diff --git a/fs/devpts/inode.c b/fs/devpts/inode.c index 63a4a59..c68edb9 100644 --- a/fs/devpts/inode.c +++ b/fs/devpts/inode.c @@ -90,6 +90,15 @@ static inline struct super_block *pts_sb_from_inode(struct inode *inode) #define PARSE_MOUNT 0 #define PARSE_REMOUNT 1 +/* + * parse_mount_options(): + * Set @opts to mount options specified in @data. If an option is not + * specified in @data, set it to its default value. The exception is + * 'newinstance' option which can only be set/cleared on a mount (i.e. + * cannot be changed during remount). + * + * Note: @data may be NULL (in which case all options are set to default). + */ static int parse_mount_options(char *data, int op, struct pts_mount_opts *opts) { char *p; @@ -355,12 +364,9 @@ static int devpts_get_sb(struct file_system_type *fs_type, struct pts_mount_opts opts; struct super_block *s; - memset(&opts, 0, sizeof(opts)); - if (data) { - error = parse_mount_options(data, PARSE_MOUNT, &opts); - if (error) - return error; - } + error = parse_mount_options(data, PARSE_MOUNT, &opts); + if (error) + return error; if (opts.newinstance) s = sget(fs_type, NULL, set_anon_super, NULL); @@ -389,11 +395,10 @@ static int devpts_get_sb(struct file_system_type *fs_type, return 0; out_dput: - dput(s->s_root); + dput(s->s_root); /* undo dget() in simple_set_mnt() */ out_undo_sget: - up_write(&s->s_umount); - deactivate_super(s); + deactivate_locked_super(s); return error; } diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c index 93bc0f8..2f0945d 100644 --- a/fs/ecryptfs/inode.c +++ b/fs/ecryptfs/inode.c @@ -667,7 +667,7 @@ ecryptfs_readlink(struct dentry *dentry, char __user *buf, int bufsiz) lower_buf = kmalloc(lower_bufsiz, GFP_KERNEL); if (lower_buf == NULL) { printk(KERN_ERR "%s: Out of memory whilst attempting to " - "kmalloc [%d] bytes\n", __func__, lower_bufsiz); + "kmalloc [%zd] bytes\n", __func__, lower_bufsiz); rc = -ENOMEM; goto out; } @@ -690,7 +690,7 @@ ecryptfs_readlink(struct dentry *dentry, char __user *buf, int bufsiz) } /* Check for bufsiz <= 0 done in sys_readlinkat() */ rc = copy_to_user(buf, plaintext_name, - min((unsigned) bufsiz, plaintext_name_size)); + min((size_t) bufsiz, plaintext_name_size)); if (rc) rc = -EFAULT; else diff --git a/fs/ecryptfs/main.c b/fs/ecryptfs/main.c index ccabd5f..9f0aa98 100644 --- a/fs/ecryptfs/main.c +++ b/fs/ecryptfs/main.c @@ -614,9 +614,8 @@ static int ecryptfs_get_sb(struct file_system_type *fs_type, int flags, } goto out; out_abort: - dput(sb->s_root); - up_write(&sb->s_umount); - deactivate_super(sb); + dput(sb->s_root); /* aka mnt->mnt_root, as set by get_sb_nodev() */ + deactivate_locked_super(sb); out: return rc; } diff --git a/fs/eventpoll.c b/fs/eventpoll.c index a89f370..5458e80 100644 --- a/fs/eventpoll.c +++ b/fs/eventpoll.c @@ -1212,7 +1212,7 @@ SYSCALL_DEFINE1(epoll_create1, int, flags) SYSCALL_DEFINE1(epoll_create, int, size) { - if (size < 0) + if (size <= 0) return -EINVAL; return sys_epoll_create1(0); @@ -69,17 +69,18 @@ int suid_dumpable = 0; static LIST_HEAD(formats); static DEFINE_RWLOCK(binfmt_lock); -int register_binfmt(struct linux_binfmt * fmt) +int __register_binfmt(struct linux_binfmt * fmt, int insert) { if (!fmt) return -EINVAL; write_lock(&binfmt_lock); - list_add(&fmt->lh, &formats); + insert ? list_add(&fmt->lh, &formats) : + list_add_tail(&fmt->lh, &formats); write_unlock(&binfmt_lock); return 0; } -EXPORT_SYMBOL(register_binfmt); +EXPORT_SYMBOL(__register_binfmt); void unregister_binfmt(struct linux_binfmt * fmt) { @@ -104,40 +105,28 @@ static inline void put_binfmt(struct linux_binfmt * fmt) SYSCALL_DEFINE1(uselib, const char __user *, library) { struct file *file; - struct nameidata nd; char *tmp = getname(library); int error = PTR_ERR(tmp); - if (!IS_ERR(tmp)) { - error = path_lookup_open(AT_FDCWD, tmp, - LOOKUP_FOLLOW, &nd, - FMODE_READ|FMODE_EXEC); - putname(tmp); - } - if (error) + if (IS_ERR(tmp)) + goto out; + + file = do_filp_open(AT_FDCWD, tmp, + O_LARGEFILE | O_RDONLY | FMODE_EXEC, 0, + MAY_READ | MAY_EXEC | MAY_OPEN); + putname(tmp); + error = PTR_ERR(file); + if (IS_ERR(file)) goto out; error = -EINVAL; - if (!S_ISREG(nd.path.dentry->d_inode->i_mode)) + if (!S_ISREG(file->f_path.dentry->d_inode->i_mode)) goto exit; error = -EACCES; - if (nd.path.mnt->mnt_flags & MNT_NOEXEC) - goto exit; - - error = inode_permission(nd.path.dentry->d_inode, - MAY_READ | MAY_EXEC | MAY_OPEN); - if (error) - goto exit; - error = ima_path_check(&nd.path, MAY_READ | MAY_EXEC | MAY_OPEN); - if (error) + if (file->f_path.mnt->mnt_flags & MNT_NOEXEC) goto exit; - file = nameidata_to_filp(&nd, O_RDONLY|O_LARGEFILE); - error = PTR_ERR(file); - if (IS_ERR(file)) - goto out; - fsnotify_open(file->f_path.dentry); error = -ENOEXEC; @@ -159,13 +148,10 @@ SYSCALL_DEFINE1(uselib, const char __user *, library) } read_unlock(&binfmt_lock); } +exit: fput(file); out: return error; -exit: - release_open_intent(&nd); - path_put(&nd.path); - goto out; } #ifdef CONFIG_MMU @@ -660,47 +646,33 @@ EXPORT_SYMBOL(setup_arg_pages); struct file *open_exec(const char *name) { - struct nameidata nd; struct file *file; int err; - err = path_lookup_open(AT_FDCWD, name, LOOKUP_FOLLOW, &nd, - FMODE_READ|FMODE_EXEC); - if (err) + file = do_filp_open(AT_FDCWD, name, + O_LARGEFILE | O_RDONLY | FMODE_EXEC, 0, + MAY_EXEC | MAY_OPEN); + if (IS_ERR(file)) goto out; err = -EACCES; - if (!S_ISREG(nd.path.dentry->d_inode->i_mode)) - goto out_path_put; - - if (nd.path.mnt->mnt_flags & MNT_NOEXEC) - goto out_path_put; - - err = inode_permission(nd.path.dentry->d_inode, MAY_EXEC | MAY_OPEN); - if (err) - goto out_path_put; - err = ima_path_check(&nd.path, MAY_EXEC | MAY_OPEN); - if (err) - goto out_path_put; + if (!S_ISREG(file->f_path.dentry->d_inode->i_mode)) + goto exit; - file = nameidata_to_filp(&nd, O_RDONLY|O_LARGEFILE); - if (IS_ERR(file)) - return file; + if (file->f_path.mnt->mnt_flags & MNT_NOEXEC) + goto exit; fsnotify_open(file->f_path.dentry); err = deny_write_access(file); - if (err) { - fput(file); - goto out; - } + if (err) + goto exit; +out: return file; - out_path_put: - release_open_intent(&nd); - path_put(&nd.path); - out: +exit: + fput(file); return ERR_PTR(err); } EXPORT_SYMBOL(open_exec); diff --git a/fs/ext2/super.c b/fs/ext2/super.c index f983225..5c4afe6 100644 --- a/fs/ext2/super.c +++ b/fs/ext2/super.c @@ -1395,8 +1395,10 @@ static ssize_t ext2_quota_write(struct super_block *sb, int type, blk++; } out: - if (len == towrite) + if (len == towrite) { + mutex_unlock(&inode->i_mutex); return err; + } if (inode->i_size < off+len-towrite) i_size_write(inode, off+len-towrite); inode->i_version++; diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index e403321..e3a55eb 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -1841,11 +1841,13 @@ ext4_ext_put_in_cache(struct inode *inode, ext4_lblk_t block, { struct ext4_ext_cache *cex; BUG_ON(len == 0); + spin_lock(&EXT4_I(inode)->i_block_reservation_lock); cex = &EXT4_I(inode)->i_cached_extent; cex->ec_type = type; cex->ec_block = block; cex->ec_len = len; cex->ec_start = start; + spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); } /* @@ -1902,12 +1904,17 @@ ext4_ext_in_cache(struct inode *inode, ext4_lblk_t block, struct ext4_extent *ex) { struct ext4_ext_cache *cex; + int ret = EXT4_EXT_CACHE_NO; + /* + * We borrow i_block_reservation_lock to protect i_cached_extent + */ + spin_lock(&EXT4_I(inode)->i_block_reservation_lock); cex = &EXT4_I(inode)->i_cached_extent; /* has cache valid data? */ if (cex->ec_type == EXT4_EXT_CACHE_NO) - return EXT4_EXT_CACHE_NO; + goto errout; BUG_ON(cex->ec_type != EXT4_EXT_CACHE_GAP && cex->ec_type != EXT4_EXT_CACHE_EXTENT); @@ -1918,11 +1925,11 @@ ext4_ext_in_cache(struct inode *inode, ext4_lblk_t block, ext_debug("%u cached by %u:%u:%llu\n", block, cex->ec_block, cex->ec_len, cex->ec_start); - return cex->ec_type; + ret = cex->ec_type; } - - /* not in cache */ - return EXT4_EXT_CACHE_NO; +errout: + spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); + return ret; } /* @@ -2875,6 +2882,8 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, if (allocated > max_blocks) allocated = max_blocks; set_buffer_unwritten(bh_result); + bh_result->b_bdev = inode->i_sb->s_bdev; + bh_result->b_blocknr = newblock; goto out2; } diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index c6bd6ce..2a9ffd5 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -1149,6 +1149,7 @@ int ext4_get_blocks_wrap(handle_t *handle, struct inode *inode, sector_t block, int retval; clear_buffer_mapped(bh); + clear_buffer_unwritten(bh); /* * Try to see if we can get the block without requesting @@ -1179,6 +1180,18 @@ int ext4_get_blocks_wrap(handle_t *handle, struct inode *inode, sector_t block, return retval; /* + * When we call get_blocks without the create flag, the + * BH_Unwritten flag could have gotten set if the blocks + * requested were part of a uninitialized extent. We need to + * clear this flag now that we are committed to convert all or + * part of the uninitialized extent to be an initialized + * extent. This is because we need to avoid the combination + * of BH_Unwritten and BH_Mapped flags being simultaneously + * set on the buffer_head. + */ + clear_buffer_unwritten(bh); + + /* * New blocks allocate and/or writing to uninitialized extent * will possibly result in updating i_data, so we take * the write lock of i_data_sem, and call get_blocks() @@ -2297,6 +2310,10 @@ static int ext4_da_get_block_prep(struct inode *inode, sector_t iblock, struct buffer_head *bh_result, int create) { int ret = 0; + sector_t invalid_block = ~((sector_t) 0xffff); + + if (invalid_block < ext4_blocks_count(EXT4_SB(inode->i_sb)->s_es)) + invalid_block = ~0; BUG_ON(create == 0); BUG_ON(bh_result->b_size != inode->i_sb->s_blocksize); @@ -2318,11 +2335,18 @@ static int ext4_da_get_block_prep(struct inode *inode, sector_t iblock, /* not enough space to reserve */ return ret; - map_bh(bh_result, inode->i_sb, 0); + map_bh(bh_result, inode->i_sb, invalid_block); set_buffer_new(bh_result); set_buffer_delay(bh_result); } else if (ret > 0) { bh_result->b_size = (ret << inode->i_blkbits); + /* + * With sub-block writes into unwritten extents + * we also need to mark the buffer as new so that + * the unwritten parts of the buffer gets correctly zeroed. + */ + if (buffer_unwritten(bh_result)) + set_buffer_new(bh_result); ret = 0; } @@ -4357,11 +4381,9 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino) ei->i_flags = le32_to_cpu(raw_inode->i_flags); inode->i_blocks = ext4_inode_blocks(raw_inode, ei); ei->i_file_acl = le32_to_cpu(raw_inode->i_file_acl_lo); - if (EXT4_SB(inode->i_sb)->s_es->s_creator_os != - cpu_to_le32(EXT4_OS_HURD)) { + if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_64BIT)) ei->i_file_acl |= ((__u64)le16_to_cpu(raw_inode->i_file_acl_high)) << 32; - } inode->i_size = ext4_isize(raw_inode); ei->i_disksize = inode->i_size; inode->i_generation = le32_to_cpu(raw_inode->i_generation); @@ -4409,9 +4431,23 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino) (__u64)(le32_to_cpu(raw_inode->i_version_hi)) << 32; } - if (ei->i_flags & EXT4_EXTENTS_FL) { - /* Validate extent which is part of inode */ - ret = ext4_ext_check_inode(inode); + ret = 0; + if (ei->i_file_acl && + ((ei->i_file_acl < + (le32_to_cpu(EXT4_SB(sb)->s_es->s_first_data_block) + + EXT4_SB(sb)->s_gdb_count)) || + (ei->i_file_acl >= ext4_blocks_count(EXT4_SB(sb)->s_es)))) { + ext4_error(sb, __func__, + "bad extended attribute block %llu in inode #%lu", + ei->i_file_acl, inode->i_ino); + ret = -EIO; + goto bad_inode; + } else if (ei->i_flags & EXT4_EXTENTS_FL) { + if (S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode) || + (S_ISLNK(inode->i_mode) && + !ext4_inode_is_fast_symlink(inode))) + /* Validate extent which is part of inode */ + ret = ext4_ext_check_inode(inode); } else if (S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode) || (S_ISLNK(inode->i_mode) && !ext4_inode_is_fast_symlink(inode))) { @@ -117,11 +117,13 @@ SYSCALL_DEFINE2(dup2, unsigned int, oldfd, unsigned int, newfd) { if (unlikely(newfd == oldfd)) { /* corner case */ struct files_struct *files = current->files; + int retval = oldfd; + rcu_read_lock(); if (!fcheck_files(files, oldfd)) - oldfd = -EBADF; + retval = -EBADF; rcu_read_unlock(); - return oldfd; + return retval; } return sys_dup3(oldfd, newfd, 0); } diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c index 459b73d..91f7c85f 100644 --- a/fs/fuse/inode.c +++ b/fs/fuse/inode.c @@ -19,6 +19,7 @@ #include <linux/random.h> #include <linux/sched.h> #include <linux/exportfs.h> +#include <linux/smp_lock.h> MODULE_AUTHOR("Miklos Szeredi <miklos@szeredi.hu>"); MODULE_DESCRIPTION("Filesystem in Userspace"); @@ -259,7 +260,9 @@ struct inode *fuse_iget(struct super_block *sb, u64 nodeid, static void fuse_umount_begin(struct super_block *sb) { + lock_kernel(); fuse_abort_conn(get_fuse_conn_super(sb)); + unlock_kernel(); } static void fuse_send_destroy(struct fuse_conn *fc) @@ -908,6 +911,7 @@ static int fuse_fill_super(struct super_block *sb, void *data, int silent) err_put_root: dput(root_dentry); err_put_conn: + bdi_destroy(&fc->bdi); fuse_conn_put(fc); err_fput: fput(file); diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index 1afd9f2..ff49810 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -1304,6 +1304,7 @@ static int gfs2_shrink_glock_memory(int nr, gfp_t gfp_mask) nr--; if (queue_delayed_work(glock_workqueue, &gl->gl_work, 0) == 0) gfs2_glock_put(gl); + got_ref = 0; } spin_lock(&lru_lock); if (may_demote) diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c index 650a730..1ff9473 100644 --- a/fs/gfs2/ops_fstype.c +++ b/fs/gfs2/ops_fstype.c @@ -1282,21 +1282,21 @@ static int gfs2_get_sb(struct file_system_type *fs_type, int flags, static struct super_block *get_gfs2_sb(const char *dev_name) { struct super_block *sb; - struct nameidata nd; + struct path path; int error; - error = path_lookup(dev_name, LOOKUP_FOLLOW, &nd); + error = kern_path(dev_name, LOOKUP_FOLLOW, &path); if (error) { printk(KERN_WARNING "GFS2: path_lookup on %s returned error %d\n", dev_name, error); return NULL; } - sb = nd.path.dentry->d_inode->i_sb; + sb = path.dentry->d_inode->i_sb; if (sb && (sb->s_type == &gfs2_fs_type)) atomic_inc(&sb->s_active); else sb = NULL; - path_put(&nd.path); + path_put(&path); return sb; } diff --git a/fs/hpfs/super.c b/fs/hpfs/super.c index fecf402..fc77965 100644 --- a/fs/hpfs/super.c +++ b/fs/hpfs/super.c @@ -423,8 +423,7 @@ static int hpfs_remount_fs(struct super_block *s, int *flags, char *data) if (!(*flags & MS_RDONLY)) mark_dirty(s); - kfree(s->s_options); - s->s_options = new_opts; + replace_mount_options(s, new_opts); return 0; diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 153d968..c1462d4 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -312,16 +312,6 @@ out: return retval; } -/* - * Read a page. Again trivial. If it didn't already exist - * in the page cache, it is zero-filled. - */ -static int hugetlbfs_readpage(struct file *file, struct page * page) -{ - unlock_page(page); - return -EINVAL; -} - static int hugetlbfs_write_begin(struct file *file, struct address_space *mapping, loff_t pos, unsigned len, unsigned flags, @@ -701,7 +691,6 @@ static void hugetlbfs_destroy_inode(struct inode *inode) } static const struct address_space_operations hugetlbfs_aops = { - .readpage = hugetlbfs_readpage, .write_begin = hugetlbfs_write_begin, .write_end = hugetlbfs_write_end, .set_page_dirty = hugetlbfs_set_page_dirty, @@ -99,7 +99,7 @@ static DEFINE_MUTEX(iprune_mutex); */ struct inodes_stat_t inodes_stat; -static struct kmem_cache * inode_cachep __read_mostly; +static struct kmem_cache *inode_cachep __read_mostly; static void wake_up_inode(struct inode *inode) { @@ -124,7 +124,7 @@ struct inode *inode_init_always(struct super_block *sb, struct inode *inode) static struct inode_operations empty_iops; static const struct file_operations empty_fops; - struct address_space * const mapping = &inode->i_data; + struct address_space *const mapping = &inode->i_data; inode->i_sb = sb; inode->i_blkbits = sb->s_blocksize_bits; @@ -216,7 +216,7 @@ static struct inode *alloc_inode(struct super_block *sb) return NULL; } -void destroy_inode(struct inode *inode) +void destroy_inode(struct inode *inode) { BUG_ON(inode_has_buffers(inode)); security_inode_free(inode); @@ -252,12 +252,11 @@ void inode_init_once(struct inode *inode) mutex_init(&inode->inotify_mutex); #endif } - EXPORT_SYMBOL(inode_init_once); static void init_once(void *foo) { - struct inode * inode = (struct inode *) foo; + struct inode *inode = (struct inode *) foo; inode_init_once(inode); } @@ -265,7 +264,7 @@ static void init_once(void *foo) /* * inode_lock must be held */ -void __iget(struct inode * inode) +void __iget(struct inode *inode) { if (atomic_read(&inode->i_count)) { atomic_inc(&inode->i_count); @@ -289,7 +288,7 @@ void clear_inode(struct inode *inode) { might_sleep(); invalidate_inode_buffers(inode); - + BUG_ON(inode->i_data.nrpages); BUG_ON(!(inode->i_state & I_FREEING)); BUG_ON(inode->i_state & I_CLEAR); @@ -303,7 +302,6 @@ void clear_inode(struct inode *inode) cd_forget(inode); inode->i_state = I_CLEAR; } - EXPORT_SYMBOL(clear_inode); /* @@ -351,8 +349,8 @@ static int invalidate_list(struct list_head *head, struct list_head *dispose) next = head->next; for (;;) { - struct list_head * tmp = next; - struct inode * inode; + struct list_head *tmp = next; + struct inode *inode; /* * We can reschedule here without worrying about the list's @@ -391,7 +389,7 @@ static int invalidate_list(struct list_head *head, struct list_head *dispose) * fails because there are busy inodes then a non zero value is returned. * If the discard is successful all the inodes have been discarded. */ -int invalidate_inodes(struct super_block * sb) +int invalidate_inodes(struct super_block *sb) { int busy; LIST_HEAD(throw_away); @@ -407,7 +405,6 @@ int invalidate_inodes(struct super_block * sb) return busy; } - EXPORT_SYMBOL(invalidate_inodes); static int can_unuse(struct inode *inode) @@ -504,7 +501,7 @@ static int shrink_icache_memory(int nr, gfp_t gfp_mask) * Nasty deadlock avoidance. We may hold various FS locks, * and we don't want to recurse into the FS that called us * in clear_inode() and friends.. - */ + */ if (!(gfp_mask & __GFP_FS)) return -1; prune_icache(nr); @@ -524,10 +521,13 @@ static void __wait_on_freeing_inode(struct inode *inode); * by hand after calling find_inode now! This simplifies iunique and won't * add any additional branch in the common code. */ -static struct inode * find_inode(struct super_block * sb, struct hlist_head *head, int (*test)(struct inode *, void *), void *data) +static struct inode *find_inode(struct super_block *sb, + struct hlist_head *head, + int (*test)(struct inode *, void *), + void *data) { struct hlist_node *node; - struct inode * inode = NULL; + struct inode *inode = NULL; repeat: hlist_for_each_entry(inode, node, head, i_hash) { @@ -548,10 +548,11 @@ repeat: * find_inode_fast is the fast path version of find_inode, see the comment at * iget_locked for details. */ -static struct inode * find_inode_fast(struct super_block * sb, struct hlist_head *head, unsigned long ino) +static struct inode *find_inode_fast(struct super_block *sb, + struct hlist_head *head, unsigned long ino) { struct hlist_node *node; - struct inode * inode = NULL; + struct inode *inode = NULL; repeat: hlist_for_each_entry(inode, node, head, i_hash) { @@ -631,10 +632,10 @@ struct inode *new_inode(struct super_block *sb) * here to attempt to avoid that. */ static unsigned int last_ino; - struct inode * inode; + struct inode *inode; spin_lock_prefetch(&inode_lock); - + inode = alloc_inode(sb); if (inode) { spin_lock(&inode_lock); @@ -645,7 +646,6 @@ struct inode *new_inode(struct super_block *sb) } return inode; } - EXPORT_SYMBOL(new_inode); void unlock_new_inode(struct inode *inode) @@ -674,7 +674,6 @@ void unlock_new_inode(struct inode *inode) inode->i_state &= ~(I_LOCK|I_NEW); wake_up_inode(inode); } - EXPORT_SYMBOL(unlock_new_inode); /* @@ -683,13 +682,17 @@ EXPORT_SYMBOL(unlock_new_inode); * We no longer cache the sb_flags in i_flags - see fs.h * -- rmk@arm.uk.linux.org */ -static struct inode * get_new_inode(struct super_block *sb, struct hlist_head *head, int (*test)(struct inode *, void *), int (*set)(struct inode *, void *), void *data) +static struct inode *get_new_inode(struct super_block *sb, + struct hlist_head *head, + int (*test)(struct inode *, void *), + int (*set)(struct inode *, void *), + void *data) { - struct inode * inode; + struct inode *inode; inode = alloc_inode(sb); if (inode) { - struct inode * old; + struct inode *old; spin_lock(&inode_lock); /* We released the lock, so.. */ @@ -731,13 +734,14 @@ set_failed: * get_new_inode_fast is the fast path version of get_new_inode, see the * comment at iget_locked for details. */ -static struct inode * get_new_inode_fast(struct super_block *sb, struct hlist_head *head, unsigned long ino) +static struct inode *get_new_inode_fast(struct super_block *sb, + struct hlist_head *head, unsigned long ino) { - struct inode * inode; + struct inode *inode; inode = alloc_inode(sb); if (inode) { - struct inode * old; + struct inode *old; spin_lock(&inode_lock); /* We released the lock, so.. */ @@ -823,7 +827,6 @@ struct inode *igrab(struct inode *inode) spin_unlock(&inode_lock); return inode; } - EXPORT_SYMBOL(igrab); /** @@ -924,7 +927,6 @@ struct inode *ilookup5_nowait(struct super_block *sb, unsigned long hashval, return ifind(sb, head, test, data, 0); } - EXPORT_SYMBOL(ilookup5_nowait); /** @@ -953,7 +955,6 @@ struct inode *ilookup5(struct super_block *sb, unsigned long hashval, return ifind(sb, head, test, data, 1); } - EXPORT_SYMBOL(ilookup5); /** @@ -976,7 +977,6 @@ struct inode *ilookup(struct super_block *sb, unsigned long ino) return ifind_fast(sb, head, ino); } - EXPORT_SYMBOL(ilookup); /** @@ -1015,7 +1015,6 @@ struct inode *iget5_locked(struct super_block *sb, unsigned long hashval, */ return get_new_inode(sb, head, test, set, data); } - EXPORT_SYMBOL(iget5_locked); /** @@ -1047,7 +1046,6 @@ struct inode *iget_locked(struct super_block *sb, unsigned long ino) */ return get_new_inode_fast(sb, head, ino); } - EXPORT_SYMBOL(iget_locked); int insert_inode_locked(struct inode *inode) @@ -1076,7 +1074,6 @@ int insert_inode_locked(struct inode *inode) iput(old); } } - EXPORT_SYMBOL(insert_inode_locked); int insert_inode_locked4(struct inode *inode, unsigned long hashval, @@ -1106,7 +1103,6 @@ int insert_inode_locked4(struct inode *inode, unsigned long hashval, iput(old); } } - EXPORT_SYMBOL(insert_inode_locked4); /** @@ -1124,7 +1120,6 @@ void __insert_inode_hash(struct inode *inode, unsigned long hashval) hlist_add_head(&inode->i_hash, head); spin_unlock(&inode_lock); } - EXPORT_SYMBOL(__insert_inode_hash); /** @@ -1139,7 +1134,6 @@ void remove_inode_hash(struct inode *inode) hlist_del_init(&inode->i_hash); spin_unlock(&inode_lock); } - EXPORT_SYMBOL(remove_inode_hash); /* @@ -1187,7 +1181,6 @@ void generic_delete_inode(struct inode *inode) BUG_ON(inode->i_state != I_CLEAR); destroy_inode(inode); } - EXPORT_SYMBOL(generic_delete_inode); static void generic_forget_inode(struct inode *inode) @@ -1237,12 +1230,11 @@ void generic_drop_inode(struct inode *inode) else generic_forget_inode(inode); } - EXPORT_SYMBOL_GPL(generic_drop_inode); /* * Called when we're dropping the last reference - * to an inode. + * to an inode. * * Call the FS "drop()" function, defaulting to * the legacy UNIX filesystem behaviour.. @@ -1262,7 +1254,7 @@ static inline void iput_final(struct inode *inode) } /** - * iput - put an inode + * iput - put an inode * @inode: inode to put * * Puts an inode, dropping its usage count. If the inode use count hits @@ -1279,7 +1271,6 @@ void iput(struct inode *inode) iput_final(inode); } } - EXPORT_SYMBOL(iput); /** @@ -1290,10 +1281,10 @@ EXPORT_SYMBOL(iput); * Returns the block number on the device holding the inode that * is the disk block number for the block of the file requested. * That is, asked for block 4 of inode 1 the function will return the - * disk block relative to the disk start that holds that block of the + * disk block relative to the disk start that holds that block of the * file. */ -sector_t bmap(struct inode * inode, sector_t block) +sector_t bmap(struct inode *inode, sector_t block) { sector_t res = 0; if (inode->i_mapping->a_ops->bmap) @@ -1425,7 +1416,6 @@ void file_update_time(struct file *file) mark_inode_dirty_sync(inode); mnt_drop_write(file->f_path.mnt); } - EXPORT_SYMBOL(file_update_time); int inode_needs_sync(struct inode *inode) @@ -1436,7 +1426,6 @@ int inode_needs_sync(struct inode *inode) return 1; return 0; } - EXPORT_SYMBOL(inode_needs_sync); int inode_wait(void *word) @@ -258,7 +258,7 @@ int __generic_block_fiemap(struct inode *inode, long long length = 0, map_len = 0; u64 logical = 0, phys = 0, size = 0; u32 flags = FIEMAP_EXTENT_MERGED; - int ret = 0; + int ret = 0, past_eof = 0, whole_file = 0; if ((ret = fiemap_check_flags(fieinfo, FIEMAP_FLAG_SYNC))) return ret; @@ -266,6 +266,9 @@ int __generic_block_fiemap(struct inode *inode, start_blk = logical_to_blk(inode, start); length = (long long)min_t(u64, len, i_size_read(inode)); + if (length < len) + whole_file = 1; + map_len = length; do { @@ -282,11 +285,26 @@ int __generic_block_fiemap(struct inode *inode, /* HOLE */ if (!buffer_mapped(&tmp)) { + length -= blk_to_logical(inode, 1); + start_blk++; + + /* + * we want to handle the case where there is an + * allocated block at the front of the file, and then + * nothing but holes up to the end of the file properly, + * to make sure that extent at the front gets properly + * marked with FIEMAP_EXTENT_LAST + */ + if (!past_eof && + blk_to_logical(inode, start_blk) >= + blk_to_logical(inode, 0)+i_size_read(inode)) + past_eof = 1; + /* * first hole after going past the EOF, this is our * last extent */ - if (length <= 0) { + if (past_eof && size) { flags = FIEMAP_EXTENT_MERGED|FIEMAP_EXTENT_LAST; ret = fiemap_fill_next_extent(fieinfo, logical, phys, size, @@ -294,15 +312,37 @@ int __generic_block_fiemap(struct inode *inode, break; } - length -= blk_to_logical(inode, 1); - /* if we have holes up to/past EOF then we're done */ - if (length <= 0) + if (length <= 0 || past_eof) break; - - start_blk++; } else { - if (length <= 0 && size) { + /* + * we have gone over the length of what we wanted to + * map, and it wasn't the entire file, so add the extent + * we got last time and exit. + * + * This is for the case where say we want to map all the + * way up to the second to the last block in a file, but + * the last block is a hole, making the second to last + * block FIEMAP_EXTENT_LAST. In this case we want to + * see if there is a hole after the second to last block + * so we can mark it properly. If we found data after + * we exceeded the length we were requesting, then we + * are good to go, just add the extent to the fieinfo + * and break + */ + if (length <= 0 && !whole_file) { + ret = fiemap_fill_next_extent(fieinfo, logical, + phys, size, + flags); + break; + } + + /* + * if size != 0 then we know we already have an extent + * to add, so add it. + */ + if (size) { ret = fiemap_fill_next_extent(fieinfo, logical, phys, size, flags); @@ -319,19 +359,14 @@ int __generic_block_fiemap(struct inode *inode, start_blk += logical_to_blk(inode, size); /* - * if we are past the EOF we need to loop again to see - * if there is a hole so we can mark this extent as the - * last one, and if not keep mapping things until we - * find a hole, or we run out of slots in the extent - * array + * If we are past the EOF, then we need to make sure as + * soon as we find a hole that the last extent we found + * is marked with FIEMAP_EXTENT_LAST */ - if (length <= 0) - continue; - - ret = fiemap_fill_next_extent(fieinfo, logical, phys, - size, flags); - if (ret) - break; + if (!past_eof && + logical+size >= + blk_to_logical(inode, 0)+i_size_read(inode)) + past_eof = 1; } cond_resched(); } while (1); @@ -246,8 +246,7 @@ int get_sb_pseudo(struct file_system_type *fs_type, char *name, return 0; Enomem: - up_write(&s->s_umount); - deactivate_super(s); + deactivate_locked_super(s); return -ENOMEM; } diff --git a/fs/lockd/svc.c b/fs/lockd/svc.c index abf8388..1a54ae1 100644 --- a/fs/lockd/svc.c +++ b/fs/lockd/svc.c @@ -104,6 +104,16 @@ static void set_grace_period(void) schedule_delayed_work(&grace_period_end, grace_period); } +static void restart_grace(void) +{ + if (nlmsvc_ops) { + cancel_delayed_work_sync(&grace_period_end); + locks_end_grace(&lockd_manager); + nlmsvc_invalidate_all(); + set_grace_period(); + } +} + /* * This is the lockd kernel thread */ @@ -149,10 +159,7 @@ lockd(void *vrqstp) if (signalled()) { flush_signals(current); - if (nlmsvc_ops) { - nlmsvc_invalidate_all(); - set_grace_period(); - } + restart_grace(); continue; } @@ -1130,8 +1130,8 @@ int vfs_path_lookup(struct dentry *dentry, struct vfsmount *mnt, * @nd: pointer to nameidata * @open_flags: open intent flags */ -int path_lookup_open(int dfd, const char *name, unsigned int lookup_flags, - struct nameidata *nd, int open_flags) +static int path_lookup_open(int dfd, const char *name, + unsigned int lookup_flags, struct nameidata *nd, int open_flags) { struct file *filp = get_empty_filp(); int err; @@ -1637,18 +1637,19 @@ static int open_will_write_to_fs(int flag, struct inode *inode) * open_to_namei_flags() for more details. */ struct file *do_filp_open(int dfd, const char *pathname, - int open_flag, int mode) + int open_flag, int mode, int acc_mode) { struct file *filp; struct nameidata nd; - int acc_mode, error; + int error; struct path path; struct dentry *dir; int count = 0; int will_write; int flag = open_to_namei_flags(open_flag); - acc_mode = MAY_OPEN | ACC_MODE(flag); + if (!acc_mode) + acc_mode = MAY_OPEN | ACC_MODE(flag); /* O_TRUNC implies we need access checks for write permissions */ if (flag & O_TRUNC) @@ -1869,7 +1870,7 @@ do_link: */ struct file *filp_open(const char *filename, int flags, int mode) { - return do_filp_open(AT_FDCWD, filename, flags, mode); + return do_filp_open(AT_FDCWD, filename, flags, mode, 0); } EXPORT_SYMBOL(filp_open); diff --git a/fs/namespace.c b/fs/namespace.c index 4119620..134d494 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -695,12 +695,16 @@ static inline void mangle(struct seq_file *m, const char *s) */ int generic_show_options(struct seq_file *m, struct vfsmount *mnt) { - const char *options = mnt->mnt_sb->s_options; + const char *options; + + rcu_read_lock(); + options = rcu_dereference(mnt->mnt_sb->s_options); if (options != NULL && options[0]) { seq_putc(m, ','); mangle(m, options); } + rcu_read_unlock(); return 0; } @@ -721,11 +725,22 @@ EXPORT_SYMBOL(generic_show_options); */ void save_mount_options(struct super_block *sb, char *options) { - kfree(sb->s_options); - sb->s_options = kstrdup(options, GFP_KERNEL); + BUG_ON(sb->s_options); + rcu_assign_pointer(sb->s_options, kstrdup(options, GFP_KERNEL)); } EXPORT_SYMBOL(save_mount_options); +void replace_mount_options(struct super_block *sb, char *options) +{ + char *old = sb->s_options; + rcu_assign_pointer(sb->s_options, options); + if (old) { + synchronize_rcu(); + kfree(old); + } +} +EXPORT_SYMBOL(replace_mount_options); + #ifdef CONFIG_PROC_FS /* iterator */ static void *m_start(struct seq_file *m, loff_t *pos) @@ -1073,9 +1088,7 @@ static int do_umount(struct vfsmount *mnt, int flags) */ if (flags & MNT_FORCE && sb->s_op->umount_begin) { - lock_kernel(); sb->s_op->umount_begin(sb); - unlock_kernel(); } /* diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index 370b190..89f98e9 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -1943,7 +1943,8 @@ int nfs_permission(struct inode *inode, int mask) case S_IFREG: /* NFSv4 has atomic_open... */ if (nfs_server_capable(inode, NFS_CAP_ATOMIC_OPEN) - && (mask & MAY_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 5a97bcf..ec7e27d 100644 --- a/fs/nfs/file.c +++ b/fs/nfs/file.c @@ -517,10 +517,10 @@ static int nfs_vm_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf) ret = nfs_updatepage(filp, page, 0, pagelen); out_unlock: + if (!ret) + return VM_FAULT_LOCKED; unlock_page(page); - if (ret) - ret = VM_FAULT_SIGBUS; - return ret; + return VM_FAULT_SIGBUS; } static struct vm_operations_struct nfs_file_vm_ops = { diff --git a/fs/nfs/super.c b/fs/nfs/super.c index 6717200..d2d6778 100644 --- a/fs/nfs/super.c +++ b/fs/nfs/super.c @@ -683,9 +683,12 @@ static int nfs_show_stats(struct seq_file *m, struct vfsmount *mnt) */ static void nfs_umount_begin(struct super_block *sb) { - struct nfs_server *server = NFS_SB(sb); + struct nfs_server *server; struct rpc_clnt *rpc; + lock_kernel(); + + server = NFS_SB(sb); /* -EIO all pending I/O */ rpc = server->client_acl; if (!IS_ERR(rpc)) @@ -693,6 +696,8 @@ static void nfs_umount_begin(struct super_block *sb) rpc = server->client; if (!IS_ERR(rpc)) rpc_killall_tasks(rpc); + + unlock_kernel(); } /* @@ -2106,8 +2111,7 @@ out_err_nosb: error_splat_root: dput(mntroot); error_splat_super: - up_write(&s->s_umount); - deactivate_super(s); + deactivate_locked_super(s); goto out; } @@ -2203,8 +2207,7 @@ out_err_noserver: return error; error_splat_super: - up_write(&s->s_umount); - deactivate_super(s); + deactivate_locked_super(s); dprintk("<-- nfs_xdev_get_sb() = %d [splat]\n", error); return error; } @@ -2464,8 +2467,7 @@ out_free: error_splat_root: dput(mntroot); error_splat_super: - up_write(&s->s_umount); - deactivate_super(s); + deactivate_locked_super(s); goto out; } @@ -2559,8 +2561,7 @@ out_err_noserver: return error; error_splat_super: - up_write(&s->s_umount); - deactivate_super(s); + deactivate_locked_super(s); dprintk("<-- nfs4_xdev_get_sb() = %d [splat]\n", error); return error; } @@ -2644,8 +2645,7 @@ out_err_noserver: return error; error_splat_super: - up_write(&s->s_umount); - deactivate_super(s); + deactivate_locked_super(s); dprintk("<-- nfs4_referral_get_sb() = %d [splat]\n", error); return error; } diff --git a/fs/nfsd/nfs4recover.c b/fs/nfsd/nfs4recover.c index 5275097..b534840 100644 --- a/fs/nfsd/nfs4recover.c +++ b/fs/nfsd/nfs4recover.c @@ -229,7 +229,7 @@ nfsd4_list_rec_dir(struct dentry *dir, recdir_func *f) goto out; status = vfs_readdir(filp, nfsd4_build_namelist, &names); fput(filp); - mutex_lock(&dir->d_inode->i_mutex); + mutex_lock_nested(&dir->d_inode->i_mutex, I_MUTEX_PARENT); while (!list_empty(&names)) { entry = list_entry(names.next, struct name_list, list); @@ -264,7 +264,7 @@ nfsd4_unlink_clid_dir(char *name, int namlen) dprintk("NFSD: nfsd4_unlink_clid_dir. name %.*s\n", namlen, name); - mutex_lock(&rec_dir.dentry->d_inode->i_mutex); + mutex_lock_nested(&rec_dir.dentry->d_inode->i_mutex, I_MUTEX_PARENT); dentry = lookup_one_len(name, rec_dir.dentry, namlen); if (IS_ERR(dentry)) { status = PTR_ERR(dentry); diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index c65a27b..3b711f5 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -580,7 +580,6 @@ free_session(struct kref *kref) struct nfsd4_cache_entry *e = &ses->se_slots[i].sl_cache_entry; nfsd4_release_respages(e->ce_respages, e->ce_resused); } - kfree(ses->se_slots); kfree(ses); } diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c index b820c31..b73549d 100644 --- a/fs/nfsd/nfs4xdr.c +++ b/fs/nfsd/nfs4xdr.c @@ -2214,6 +2214,15 @@ nfsd4_encode_dirent_fattr(struct nfsd4_readdir *cd, dentry = lookup_one_len(name, cd->rd_fhp->fh_dentry, namlen); if (IS_ERR(dentry)) return nfserrno(PTR_ERR(dentry)); + if (!dentry->d_inode) { + /* + * nfsd_buffered_readdir drops the i_mutex between + * readdir and calling this callback, leaving a window + * where this directory entry could have gone away. + */ + dput(dentry); + return nfserr_noent; + } exp_get(exp); /* @@ -2276,6 +2285,7 @@ nfsd4_encode_dirent(void *ccdv, const char *name, int namlen, struct nfsd4_readdir *cd = container_of(ccd, struct nfsd4_readdir, common); int buflen; __be32 *p = cd->buffer; + __be32 *cookiep; __be32 nfserr = nfserr_toosmall; /* In nfsv4, "." and ".." never make it onto the wire.. */ @@ -2292,7 +2302,7 @@ nfsd4_encode_dirent(void *ccdv, const char *name, int namlen, goto fail; *p++ = xdr_one; /* mark entry present */ - cd->offset = p; /* remember pointer */ + cookiep = p; p = xdr_encode_hyper(p, NFS_OFFSET_MAX); /* offset of next entry */ p = xdr_encode_array(p, name, namlen); /* name length & name */ @@ -2306,6 +2316,8 @@ nfsd4_encode_dirent(void *ccdv, const char *name, int namlen, goto fail; case nfserr_dropit: goto fail; + case nfserr_noent: + goto skip_entry; default: /* * If the client requested the RDATTR_ERROR attribute, @@ -2324,6 +2336,8 @@ nfsd4_encode_dirent(void *ccdv, const char *name, int namlen, } cd->buflen -= (p - cd->buffer); cd->buffer = p; + cd->offset = cookiep; +skip_entry: cd->common.err = nfs_ok; return 0; fail: diff --git a/fs/nilfs2/ioctl.c b/fs/nilfs2/ioctl.c index 108d281..50ff3f2 100644 --- a/fs/nilfs2/ioctl.c +++ b/fs/nilfs2/ioctl.c @@ -25,6 +25,7 @@ #include <linux/smp_lock.h> /* lock_kernel(), unlock_kernel() */ #include <linux/capability.h> /* capable() */ #include <linux/uaccess.h> /* copy_from_user(), copy_to_user() */ +#include <linux/vmalloc.h> #include <linux/nilfs2_fs.h> #include "nilfs.h" #include "segment.h" @@ -147,29 +148,12 @@ static ssize_t nilfs_ioctl_do_get_cpinfo(struct the_nilfs *nilfs, __u64 *posp, int flags, void *buf, size_t size, size_t nmembs) { - return nilfs_cpfile_get_cpinfo(nilfs->ns_cpfile, posp, flags, buf, - nmembs); -} - -static int nilfs_ioctl_get_cpinfo(struct inode *inode, struct file *filp, - unsigned int cmd, void __user *argp) -{ - struct the_nilfs *nilfs = NILFS_SB(inode->i_sb)->s_nilfs; - struct nilfs_argv argv; int ret; - if (copy_from_user(&argv, argp, sizeof(argv))) - return -EFAULT; - down_read(&nilfs->ns_segctor_sem); - ret = nilfs_ioctl_wrap_copy(nilfs, &argv, _IOC_DIR(cmd), - nilfs_ioctl_do_get_cpinfo); + ret = nilfs_cpfile_get_cpinfo(nilfs->ns_cpfile, posp, flags, buf, + nmembs); up_read(&nilfs->ns_segctor_sem); - if (ret < 0) - return ret; - - if (copy_to_user(argp, &argv, sizeof(argv))) - ret = -EFAULT; return ret; } @@ -195,28 +179,11 @@ static ssize_t nilfs_ioctl_do_get_suinfo(struct the_nilfs *nilfs, __u64 *posp, int flags, void *buf, size_t size, size_t nmembs) { - return nilfs_sufile_get_suinfo(nilfs->ns_sufile, *posp, buf, nmembs); -} - -static int nilfs_ioctl_get_suinfo(struct inode *inode, struct file *filp, - unsigned int cmd, void __user *argp) -{ - struct the_nilfs *nilfs = NILFS_SB(inode->i_sb)->s_nilfs; - struct nilfs_argv argv; int ret; - if (copy_from_user(&argv, argp, sizeof(argv))) - return -EFAULT; - down_read(&nilfs->ns_segctor_sem); - ret = nilfs_ioctl_wrap_copy(nilfs, &argv, _IOC_DIR(cmd), - nilfs_ioctl_do_get_suinfo); + ret = nilfs_sufile_get_suinfo(nilfs->ns_sufile, *posp, buf, nmembs); up_read(&nilfs->ns_segctor_sem); - if (ret < 0) - return ret; - - if (copy_to_user(argp, &argv, sizeof(argv))) - ret = -EFAULT; return ret; } @@ -242,28 +209,11 @@ static ssize_t nilfs_ioctl_do_get_vinfo(struct the_nilfs *nilfs, __u64 *posp, int flags, void *buf, size_t size, size_t nmembs) { - return nilfs_dat_get_vinfo(nilfs_dat_inode(nilfs), buf, nmembs); -} - -static int nilfs_ioctl_get_vinfo(struct inode *inode, struct file *filp, - unsigned int cmd, void __user *argp) -{ - struct the_nilfs *nilfs = NILFS_SB(inode->i_sb)->s_nilfs; - struct nilfs_argv argv; int ret; - if (copy_from_user(&argv, argp, sizeof(argv))) - return -EFAULT; - down_read(&nilfs->ns_segctor_sem); - ret = nilfs_ioctl_wrap_copy(nilfs, &argv, _IOC_DIR(cmd), - nilfs_ioctl_do_get_vinfo); + ret = nilfs_dat_get_vinfo(nilfs_dat_inode(nilfs), buf, nmembs); up_read(&nilfs->ns_segctor_sem); - if (ret < 0) - return ret; - - if (copy_to_user(argp, &argv, sizeof(argv))) - ret = -EFAULT; return ret; } @@ -276,17 +226,21 @@ nilfs_ioctl_do_get_bdescs(struct the_nilfs *nilfs, __u64 *posp, int flags, struct nilfs_bdesc *bdescs = buf; int ret, i; + down_read(&nilfs->ns_segctor_sem); for (i = 0; i < nmembs; i++) { ret = nilfs_bmap_lookup_at_level(bmap, bdescs[i].bd_offset, bdescs[i].bd_level + 1, &bdescs[i].bd_blocknr); if (ret < 0) { - if (ret != -ENOENT) + if (ret != -ENOENT) { + up_read(&nilfs->ns_segctor_sem); return ret; + } bdescs[i].bd_blocknr = 0; } } + up_read(&nilfs->ns_segctor_sem); return nmembs; } @@ -300,10 +254,11 @@ static int nilfs_ioctl_get_bdescs(struct inode *inode, struct file *filp, if (copy_from_user(&argv, argp, sizeof(argv))) return -EFAULT; - down_read(&nilfs->ns_segctor_sem); + if (argv.v_size != sizeof(struct nilfs_bdesc)) + return -EINVAL; + ret = nilfs_ioctl_wrap_copy(nilfs, &argv, _IOC_DIR(cmd), nilfs_ioctl_do_get_bdescs); - up_read(&nilfs->ns_segctor_sem); if (ret < 0) return ret; @@ -346,10 +301,10 @@ static int nilfs_ioctl_move_inode_block(struct inode *inode, return 0; } -static ssize_t -nilfs_ioctl_do_move_blocks(struct the_nilfs *nilfs, __u64 *posp, int flags, - void *buf, size_t size, size_t nmembs) +static int nilfs_ioctl_move_blocks(struct the_nilfs *nilfs, + struct nilfs_argv *argv, void *buf) { + size_t nmembs = argv->v_nmembs; struct inode *inode; struct nilfs_vdesc *vdesc; struct buffer_head *bh, *n; @@ -410,19 +365,10 @@ nilfs_ioctl_do_move_blocks(struct the_nilfs *nilfs, __u64 *posp, int flags, return ret; } -static inline int nilfs_ioctl_move_blocks(struct the_nilfs *nilfs, - struct nilfs_argv *argv, - int dir) -{ - return nilfs_ioctl_wrap_copy(nilfs, argv, dir, - nilfs_ioctl_do_move_blocks); -} - -static ssize_t -nilfs_ioctl_do_delete_checkpoints(struct the_nilfs *nilfs, __u64 *posp, - int flags, void *buf, size_t size, - size_t nmembs) +static int nilfs_ioctl_delete_checkpoints(struct the_nilfs *nilfs, + struct nilfs_argv *argv, void *buf) { + size_t nmembs = argv->v_nmembs; struct inode *cpfile = nilfs->ns_cpfile; struct nilfs_period *periods = buf; int ret, i; @@ -436,36 +382,21 @@ nilfs_ioctl_do_delete_checkpoints(struct the_nilfs *nilfs, __u64 *posp, return nmembs; } -static inline int nilfs_ioctl_delete_checkpoints(struct the_nilfs *nilfs, - struct nilfs_argv *argv, - int dir) +static int nilfs_ioctl_free_vblocknrs(struct the_nilfs *nilfs, + struct nilfs_argv *argv, void *buf) { - return nilfs_ioctl_wrap_copy(nilfs, argv, dir, - nilfs_ioctl_do_delete_checkpoints); -} + size_t nmembs = argv->v_nmembs; + int ret; -static ssize_t -nilfs_ioctl_do_free_vblocknrs(struct the_nilfs *nilfs, __u64 *posp, int flags, - void *buf, size_t size, size_t nmembs) -{ - int ret = nilfs_dat_freev(nilfs_dat_inode(nilfs), buf, nmembs); + ret = nilfs_dat_freev(nilfs_dat_inode(nilfs), buf, nmembs); return (ret < 0) ? ret : nmembs; } -static inline int nilfs_ioctl_free_vblocknrs(struct the_nilfs *nilfs, - struct nilfs_argv *argv, - int dir) -{ - return nilfs_ioctl_wrap_copy(nilfs, argv, dir, - nilfs_ioctl_do_free_vblocknrs); -} - -static ssize_t -nilfs_ioctl_do_mark_blocks_dirty(struct the_nilfs *nilfs, __u64 *posp, - int flags, void *buf, size_t size, - size_t nmembs) +static int nilfs_ioctl_mark_blocks_dirty(struct the_nilfs *nilfs, + struct nilfs_argv *argv, void *buf) { + size_t nmembs = argv->v_nmembs; struct inode *dat = nilfs_dat_inode(nilfs); struct nilfs_bmap *bmap = NILFS_I(dat)->i_bmap; struct nilfs_bdesc *bdescs = buf; @@ -504,55 +435,37 @@ nilfs_ioctl_do_mark_blocks_dirty(struct the_nilfs *nilfs, __u64 *posp, return nmembs; } -static inline int nilfs_ioctl_mark_blocks_dirty(struct the_nilfs *nilfs, - struct nilfs_argv *argv, - int dir) +static int nilfs_ioctl_free_segments(struct the_nilfs *nilfs, + struct nilfs_argv *argv, void *buf) { - return nilfs_ioctl_wrap_copy(nilfs, argv, dir, - nilfs_ioctl_do_mark_blocks_dirty); -} - -static ssize_t -nilfs_ioctl_do_free_segments(struct the_nilfs *nilfs, __u64 *posp, int flags, - void *buf, size_t size, size_t nmembs) -{ - struct nilfs_sb_info *sbi = nilfs_get_writer(nilfs); + size_t nmembs = argv->v_nmembs; + struct nilfs_sb_info *sbi = nilfs->ns_writer; int ret; - if (unlikely(!sbi)) + if (unlikely(!sbi)) { + /* never happens because called for a writable mount */ + WARN_ON(1); return -EROFS; + } ret = nilfs_segctor_add_segments_to_be_freed( NILFS_SC(sbi), buf, nmembs); - nilfs_put_writer(nilfs); return (ret < 0) ? ret : nmembs; } -static inline int nilfs_ioctl_free_segments(struct the_nilfs *nilfs, - struct nilfs_argv *argv, - int dir) -{ - return nilfs_ioctl_wrap_copy(nilfs, argv, dir, - nilfs_ioctl_do_free_segments); -} - int nilfs_ioctl_prepare_clean_segments(struct the_nilfs *nilfs, - void __user *argp) + struct nilfs_argv *argv, void **kbufs) { - struct nilfs_argv argv[5]; const char *msg; - int dir, ret; - - if (copy_from_user(argv, argp, sizeof(argv))) - return -EFAULT; + int ret; - dir = _IOC_WRITE; - ret = nilfs_ioctl_move_blocks(nilfs, &argv[0], dir); + ret = nilfs_ioctl_move_blocks(nilfs, &argv[0], kbufs[0]); if (ret < 0) { msg = "cannot read source blocks"; goto failed; } - ret = nilfs_ioctl_delete_checkpoints(nilfs, &argv[1], dir); + + ret = nilfs_ioctl_delete_checkpoints(nilfs, &argv[1], kbufs[1]); if (ret < 0) { /* * can safely abort because checkpoints can be removed @@ -561,7 +474,7 @@ int nilfs_ioctl_prepare_clean_segments(struct the_nilfs *nilfs, msg = "cannot delete checkpoints"; goto failed; } - ret = nilfs_ioctl_free_vblocknrs(nilfs, &argv[2], dir); + ret = nilfs_ioctl_free_vblocknrs(nilfs, &argv[2], kbufs[2]); if (ret < 0) { /* * can safely abort because DAT file is updated atomically @@ -570,7 +483,7 @@ int nilfs_ioctl_prepare_clean_segments(struct the_nilfs *nilfs, msg = "cannot delete virtual blocks from DAT file"; goto failed; } - ret = nilfs_ioctl_mark_blocks_dirty(nilfs, &argv[3], dir); + ret = nilfs_ioctl_mark_blocks_dirty(nilfs, &argv[3], kbufs[3]); if (ret < 0) { /* * can safely abort because the operation is nondestructive. @@ -578,7 +491,7 @@ int nilfs_ioctl_prepare_clean_segments(struct the_nilfs *nilfs, msg = "cannot mark copying blocks dirty"; goto failed; } - ret = nilfs_ioctl_free_segments(nilfs, &argv[4], dir); + ret = nilfs_ioctl_free_segments(nilfs, &argv[4], kbufs[4]); if (ret < 0) { /* * can safely abort because this operation is atomic. @@ -598,9 +511,75 @@ int nilfs_ioctl_prepare_clean_segments(struct the_nilfs *nilfs, static int nilfs_ioctl_clean_segments(struct inode *inode, struct file *filp, unsigned int cmd, void __user *argp) { + struct nilfs_argv argv[5]; + const static size_t argsz[5] = { + sizeof(struct nilfs_vdesc), + sizeof(struct nilfs_period), + sizeof(__u64), + sizeof(struct nilfs_bdesc), + sizeof(__u64), + }; + void __user *base; + void *kbufs[5]; + struct the_nilfs *nilfs; + size_t len, nsegs; + int n, ret; + if (!capable(CAP_SYS_ADMIN)) return -EPERM; - return nilfs_clean_segments(inode->i_sb, argp); + + if (copy_from_user(argv, argp, sizeof(argv))) + return -EFAULT; + + nsegs = argv[4].v_nmembs; + if (argv[4].v_size != argsz[4]) + return -EINVAL; + /* + * argv[4] points to segment numbers this ioctl cleans. We + * use kmalloc() for its buffer because memory used for the + * segment numbers is enough small. + */ + kbufs[4] = memdup_user((void __user *)(unsigned long)argv[4].v_base, + nsegs * sizeof(__u64)); + if (IS_ERR(kbufs[4])) + return PTR_ERR(kbufs[4]); + + nilfs = NILFS_SB(inode->i_sb)->s_nilfs; + + for (n = 0; n < 4; n++) { + ret = -EINVAL; + if (argv[n].v_size != argsz[n]) + goto out_free; + + if (argv[n].v_nmembs > nsegs * nilfs->ns_blocks_per_segment) + goto out_free; + + len = argv[n].v_size * argv[n].v_nmembs; + base = (void __user *)(unsigned long)argv[n].v_base; + if (len == 0) { + kbufs[n] = NULL; + continue; + } + + kbufs[n] = vmalloc(len); + if (!kbufs[n]) { + ret = -ENOMEM; + goto out_free; + } + if (copy_from_user(kbufs[n], base, len)) { + ret = -EFAULT; + vfree(kbufs[n]); + goto out_free; + } + } + + ret = nilfs_clean_segments(inode->i_sb, argv, kbufs); + + out_free: + while (--n > 0) + vfree(kbufs[n]); + kfree(kbufs[4]); + return ret; } static int nilfs_ioctl_sync(struct inode *inode, struct file *filp, @@ -621,6 +600,33 @@ static int nilfs_ioctl_sync(struct inode *inode, struct file *filp, return 0; } +static int nilfs_ioctl_get_info(struct inode *inode, struct file *filp, + unsigned int cmd, void __user *argp, + size_t membsz, + ssize_t (*dofunc)(struct the_nilfs *, + __u64 *, int, + void *, size_t, size_t)) + +{ + struct the_nilfs *nilfs = NILFS_SB(inode->i_sb)->s_nilfs; + struct nilfs_argv argv; + int ret; + + if (copy_from_user(&argv, argp, sizeof(argv))) + return -EFAULT; + + if (argv.v_size != membsz) + return -EINVAL; + + ret = nilfs_ioctl_wrap_copy(nilfs, &argv, _IOC_DIR(cmd), dofunc); + if (ret < 0) + return ret; + + if (copy_to_user(argp, &argv, sizeof(argv))) + ret = -EFAULT; + return ret; +} + long nilfs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) { struct inode *inode = filp->f_dentry->d_inode; @@ -632,16 +638,21 @@ long nilfs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) case NILFS_IOCTL_DELETE_CHECKPOINT: return nilfs_ioctl_delete_checkpoint(inode, filp, cmd, argp); case NILFS_IOCTL_GET_CPINFO: - return nilfs_ioctl_get_cpinfo(inode, filp, cmd, argp); + return nilfs_ioctl_get_info(inode, filp, cmd, argp, + sizeof(struct nilfs_cpinfo), + nilfs_ioctl_do_get_cpinfo); case NILFS_IOCTL_GET_CPSTAT: return nilfs_ioctl_get_cpstat(inode, filp, cmd, argp); case NILFS_IOCTL_GET_SUINFO: - return nilfs_ioctl_get_suinfo(inode, filp, cmd, argp); + return nilfs_ioctl_get_info(inode, filp, cmd, argp, + sizeof(struct nilfs_suinfo), + nilfs_ioctl_do_get_suinfo); case NILFS_IOCTL_GET_SUSTAT: return nilfs_ioctl_get_sustat(inode, filp, cmd, argp); case NILFS_IOCTL_GET_VINFO: - /* XXX: rename to ??? */ - return nilfs_ioctl_get_vinfo(inode, filp, cmd, argp); + return nilfs_ioctl_get_info(inode, filp, cmd, argp, + sizeof(struct nilfs_vinfo), + nilfs_ioctl_do_get_vinfo); case NILFS_IOCTL_GET_BDESCS: return nilfs_ioctl_get_bdescs(inode, filp, cmd, argp); case NILFS_IOCTL_CLEAN_SEGMENTS: diff --git a/fs/nilfs2/mdt.c b/fs/nilfs2/mdt.c index 47dd815..bb78745 100644 --- a/fs/nilfs2/mdt.c +++ b/fs/nilfs2/mdt.c @@ -77,19 +77,22 @@ static int nilfs_mdt_create_block(struct inode *inode, unsigned long block, void *)) { struct the_nilfs *nilfs = NILFS_MDT(inode)->mi_nilfs; - struct nilfs_sb_info *writer = NULL; struct super_block *sb = inode->i_sb; struct nilfs_transaction_info ti; struct buffer_head *bh; int err; if (!sb) { - writer = nilfs_get_writer(nilfs); - if (!writer) { + /* + * Make sure this function is not called from any + * read-only context. + */ + if (!nilfs->ns_writer) { + WARN_ON(1); err = -EROFS; goto out; } - sb = writer->s_super; + sb = nilfs->ns_writer->s_super; } nilfs_transaction_begin(sb, &ti, 0); @@ -127,8 +130,6 @@ static int nilfs_mdt_create_block(struct inode *inode, unsigned long block, err = nilfs_transaction_commit(sb); else nilfs_transaction_abort(sb); - if (writer) - nilfs_put_writer(nilfs); out: return err; } @@ -299,7 +300,7 @@ int nilfs_mdt_delete_block(struct inode *inode, unsigned long block) int err; err = nilfs_bmap_delete(ii->i_bmap, block); - if (likely(!err)) { + if (!err || err == -ENOENT) { nilfs_mdt_mark_dirty(inode); nilfs_mdt_forget_block(inode, block); } diff --git a/fs/nilfs2/nilfs.h b/fs/nilfs2/nilfs.h index 3d0c18a..da6fc0b 100644 --- a/fs/nilfs2/nilfs.h +++ b/fs/nilfs2/nilfs.h @@ -236,7 +236,8 @@ extern int nilfs_sync_file(struct file *, struct dentry *, int); /* ioctl.c */ long nilfs_ioctl(struct file *, unsigned int, unsigned long); -int nilfs_ioctl_prepare_clean_segments(struct the_nilfs *, void __user *); +int nilfs_ioctl_prepare_clean_segments(struct the_nilfs *, struct nilfs_argv *, + void **); /* inode.c */ extern struct inode *nilfs_new_inode(struct inode *, int); diff --git a/fs/nilfs2/page.c b/fs/nilfs2/page.c index 1bfbba9..a2692bb 100644 --- a/fs/nilfs2/page.c +++ b/fs/nilfs2/page.c @@ -128,7 +128,8 @@ void nilfs_forget_buffer(struct buffer_head *bh) lock_buffer(bh); clear_buffer_nilfs_volatile(bh); - if (test_clear_buffer_dirty(bh) && nilfs_page_buffers_clean(page)) + clear_buffer_dirty(bh); + if (nilfs_page_buffers_clean(page)) __nilfs_clear_page_dirty(page); clear_buffer_uptodate(bh); diff --git a/fs/nilfs2/recovery.c b/fs/nilfs2/recovery.c index 4fc081e..57afa9d 100644 --- a/fs/nilfs2/recovery.c +++ b/fs/nilfs2/recovery.c @@ -407,6 +407,7 @@ void nilfs_dispose_segment_list(struct list_head *head) } static int nilfs_prepare_segment_for_recovery(struct the_nilfs *nilfs, + struct nilfs_sb_info *sbi, struct nilfs_recovery_info *ri) { struct list_head *head = &ri->ri_used_segments; @@ -421,6 +422,7 @@ static int nilfs_prepare_segment_for_recovery(struct the_nilfs *nilfs, segnum[2] = ri->ri_segnum; segnum[3] = ri->ri_nextnum; + nilfs_attach_writer(nilfs, sbi); /* * Releasing the next segment of the latest super root. * The next segment is invalidated by this recovery. @@ -459,10 +461,10 @@ static int nilfs_prepare_segment_for_recovery(struct the_nilfs *nilfs, nilfs->ns_pseg_offset = 0; nilfs->ns_seg_seq = ri->ri_seq + 2; nilfs->ns_nextnum = nilfs->ns_segnum = segnum[0]; - return 0; failed: /* No need to recover sufile because it will be destroyed on error */ + nilfs_detach_writer(nilfs, sbi); return err; } @@ -728,7 +730,7 @@ int nilfs_recover_logical_segments(struct the_nilfs *nilfs, goto failed; if (ri->ri_need_recovery == NILFS_RECOVERY_ROLLFORWARD_DONE) { - err = nilfs_prepare_segment_for_recovery(nilfs, ri); + err = nilfs_prepare_segment_for_recovery(nilfs, sbi, ri); if (unlikely(err)) { printk(KERN_ERR "NILFS: Error preparing segments for " "recovery.\n"); diff --git a/fs/nilfs2/segment.c b/fs/nilfs2/segment.c index fb70ec3..22c7f65 100644 --- a/fs/nilfs2/segment.c +++ b/fs/nilfs2/segment.c @@ -2589,7 +2589,8 @@ nilfs_remove_written_gcinodes(struct the_nilfs *nilfs, struct list_head *head) } } -int nilfs_clean_segments(struct super_block *sb, void __user *argp) +int nilfs_clean_segments(struct super_block *sb, struct nilfs_argv *argv, + void **kbufs) { struct nilfs_sb_info *sbi = NILFS_SB(sb); struct nilfs_sc_info *sci = NILFS_SC(sbi); @@ -2606,7 +2607,7 @@ int nilfs_clean_segments(struct super_block *sb, void __user *argp) err = nilfs_init_gcdat_inode(nilfs); if (unlikely(err)) goto out_unlock; - err = nilfs_ioctl_prepare_clean_segments(nilfs, argp); + err = nilfs_ioctl_prepare_clean_segments(nilfs, argv, kbufs); if (unlikely(err)) goto out_unlock; diff --git a/fs/nilfs2/segment.h b/fs/nilfs2/segment.h index a98fc1e..476bdd5 100644 --- a/fs/nilfs2/segment.h +++ b/fs/nilfs2/segment.h @@ -222,7 +222,8 @@ extern int nilfs_construct_segment(struct super_block *); extern int nilfs_construct_dsync_segment(struct super_block *, struct inode *, loff_t, loff_t); extern void nilfs_flush_segment(struct super_block *, ino_t); -extern int nilfs_clean_segments(struct super_block *, void __user *); +extern int nilfs_clean_segments(struct super_block *, struct nilfs_argv *, + void **); extern int nilfs_segctor_add_segments_to_be_freed(struct nilfs_sc_info *, __u64 *, size_t); diff --git a/fs/notify/inotify/inotify_user.c b/fs/notify/inotify/inotify_user.c index bed766e..1634319 100644 --- a/fs/notify/inotify/inotify_user.c +++ b/fs/notify/inotify/inotify_user.c @@ -220,7 +220,7 @@ static struct inotify_kernel_event * kernel_event(s32 wd, u32 mask, u32 cookie, rem = 0; } - kevent->name = kmalloc(len + rem, GFP_KERNEL); + kevent->name = kmalloc(len + rem, GFP_NOFS); if (unlikely(!kevent->name)) { kmem_cache_free(event_cachep, kevent); return NULL; diff --git a/fs/ocfs2/dcache.c b/fs/ocfs2/dcache.c index 7d60448..b574431 100644 --- a/fs/ocfs2/dcache.c +++ b/fs/ocfs2/dcache.c @@ -290,6 +290,21 @@ out_attach: else mlog_errno(ret); + /* + * In case of error, manually free the allocation and do the iput(). + * We need to do this because error here means no d_instantiate(), + * which means iput() will not be called during dput(dentry). + */ + if (ret < 0 && !alias) { + ocfs2_lock_res_free(&dl->dl_lockres); + BUG_ON(dl->dl_count != 1); + spin_lock(&dentry_attach_lock); + dentry->d_fsdata = NULL; + spin_unlock(&dentry_attach_lock); + kfree(dl); + iput(inode); + } + dput(alias); return ret; diff --git a/fs/ocfs2/dir.c b/fs/ocfs2/dir.c index e71160c..c575230 100644 --- a/fs/ocfs2/dir.c +++ b/fs/ocfs2/dir.c @@ -2697,7 +2697,7 @@ static int ocfs2_dx_dir_index_block(struct inode *dir, u32 *num_dx_entries, struct buffer_head *dirent_bh) { - int ret, namelen, i; + int ret = 0, namelen, i; char *de_buf, *limit; struct ocfs2_dir_entry *de; struct buffer_head *dx_leaf_bh; @@ -2934,7 +2934,7 @@ static int ocfs2_expand_inline_dir(struct inode *dir, struct buffer_head *di_bh, */ BUG_ON(alloc > 2); - ret = ocfs2_reserve_clusters(osb, alloc, &data_ac); + ret = ocfs2_reserve_clusters(osb, alloc + dx_alloc, &data_ac); if (ret) { mlog_errno(ret); goto out; diff --git a/fs/ocfs2/export.c b/fs/ocfs2/export.c index de3da8e..15713cb 100644 --- a/fs/ocfs2/export.c +++ b/fs/ocfs2/export.c @@ -100,7 +100,8 @@ static struct dentry *ocfs2_get_dentry(struct super_block *sb, /* If the inode allocator bit is clear, this inode must be stale */ if (!set) { - mlog(0, "inode %llu suballoc bit is clear\n", blkno); + mlog(0, "inode %llu suballoc bit is clear\n", + (unsigned long long)blkno); status = -ESTALE; goto unlock_nfs_sync; } @@ -114,7 +115,7 @@ check_err: if (status < 0) { if (status == -ESTALE) { mlog(0, "stale inode ino: %llu generation: %u\n", - blkno, handle->ih_generation); + (unsigned long long)blkno, handle->ih_generation); } result = ERR_PTR(status); goto bail; @@ -129,8 +130,8 @@ check_err: check_gen: if (handle->ih_generation != inode->i_generation) { iput(inode); - mlog(0, "stale inode ino: %llu generation: %u\n", blkno, - handle->ih_generation); + mlog(0, "stale inode ino: %llu generation: %u\n", + (unsigned long long)blkno, handle->ih_generation); result = ERR_PTR(-ESTALE); goto bail; } diff --git a/fs/ocfs2/journal.h b/fs/ocfs2/journal.h index 619dd7f..eb7b763 100644 --- a/fs/ocfs2/journal.h +++ b/fs/ocfs2/journal.h @@ -437,8 +437,9 @@ static inline int ocfs2_unlink_credits(struct super_block *sb) } /* dinode + orphan dir dinode + inode alloc dinode + orphan dir entry + - * inode alloc group descriptor + orphan dir index leaf */ -#define OCFS2_DELETE_INODE_CREDITS (3 * OCFS2_INODE_UPDATE_CREDITS + 3) + * inode alloc group descriptor + orphan dir index root + + * orphan dir index leaf */ +#define OCFS2_DELETE_INODE_CREDITS (3 * OCFS2_INODE_UPDATE_CREDITS + 4) /* dinode update, old dir dinode update, new dir dinode update, old * dir dir entry, new dir dir entry, dir entry update for renaming diff --git a/fs/ocfs2/namei.c b/fs/ocfs2/namei.c index 2220f93..33464c6 100644 --- a/fs/ocfs2/namei.c +++ b/fs/ocfs2/namei.c @@ -1025,10 +1025,8 @@ static int ocfs2_rename(struct inode *old_dir, struct inode *orphan_dir = NULL; struct ocfs2_dinode *newfe = NULL; char orphan_name[OCFS2_ORPHAN_NAMELEN + 1]; - struct buffer_head *orphan_entry_bh = NULL; struct buffer_head *newfe_bh = NULL; struct buffer_head *old_inode_bh = NULL; - struct buffer_head *insert_entry_bh = NULL; struct ocfs2_super *osb = NULL; u64 newfe_blkno, old_de_ino; handle_t *handle = NULL; @@ -1455,8 +1453,6 @@ bail: brelse(old_inode_bh); brelse(old_dir_bh); brelse(new_dir_bh); - brelse(orphan_entry_bh); - brelse(insert_entry_bh); mlog_exit(status); diff --git a/fs/ocfs2/suballoc.c b/fs/ocfs2/suballoc.c index b4ca591..8439f6b 100644 --- a/fs/ocfs2/suballoc.c +++ b/fs/ocfs2/suballoc.c @@ -2197,26 +2197,29 @@ static int ocfs2_get_suballoc_slot_bit(struct ocfs2_super *osb, u64 blkno, struct buffer_head *inode_bh = NULL; struct ocfs2_dinode *inode_fe; - mlog_entry("blkno: %llu\n", blkno); + mlog_entry("blkno: %llu\n", (unsigned long long)blkno); /* dirty read disk */ status = ocfs2_read_blocks_sync(osb, blkno, 1, &inode_bh); if (status < 0) { - mlog(ML_ERROR, "read block %llu failed %d\n", blkno, status); + mlog(ML_ERROR, "read block %llu failed %d\n", + (unsigned long long)blkno, status); goto bail; } inode_fe = (struct ocfs2_dinode *) inode_bh->b_data; if (!OCFS2_IS_VALID_DINODE(inode_fe)) { - mlog(ML_ERROR, "invalid inode %llu requested\n", blkno); + mlog(ML_ERROR, "invalid inode %llu requested\n", + (unsigned long long)blkno); status = -EINVAL; goto bail; } - if (le16_to_cpu(inode_fe->i_suballoc_slot) != OCFS2_INVALID_SLOT && + if (le16_to_cpu(inode_fe->i_suballoc_slot) != (u16)OCFS2_INVALID_SLOT && (u32)le16_to_cpu(inode_fe->i_suballoc_slot) > osb->max_slots - 1) { mlog(ML_ERROR, "inode %llu has invalid suballoc slot %u\n", - blkno, (u32)le16_to_cpu(inode_fe->i_suballoc_slot)); + (unsigned long long)blkno, + (u32)le16_to_cpu(inode_fe->i_suballoc_slot)); status = -EINVAL; goto bail; } @@ -2251,7 +2254,8 @@ static int ocfs2_test_suballoc_bit(struct ocfs2_super *osb, u64 bg_blkno; int status; - mlog_entry("blkno: %llu bit: %u\n", blkno, (unsigned int)bit); + mlog_entry("blkno: %llu bit: %u\n", (unsigned long long)blkno, + (unsigned int)bit); alloc_fe = (struct ocfs2_dinode *)alloc_bh->b_data; if ((bit + 1) > ocfs2_bits_per_group(&alloc_fe->id2.i_chain)) { @@ -2266,7 +2270,8 @@ static int ocfs2_test_suballoc_bit(struct ocfs2_super *osb, status = ocfs2_read_group_descriptor(suballoc, alloc_fe, bg_blkno, &group_bh); if (status < 0) { - mlog(ML_ERROR, "read group %llu failed %d\n", bg_blkno, status); + mlog(ML_ERROR, "read group %llu failed %d\n", + (unsigned long long)bg_blkno, status); goto bail; } @@ -2300,7 +2305,7 @@ int ocfs2_test_inode_bit(struct ocfs2_super *osb, u64 blkno, int *res) struct inode *inode_alloc_inode; struct buffer_head *alloc_bh = NULL; - mlog_entry("blkno: %llu", blkno); + mlog_entry("blkno: %llu", (unsigned long long)blkno); status = ocfs2_get_suballoc_slot_bit(osb, blkno, &suballoc_slot, &suballoc_bit); diff --git a/fs/ocfs2/symlink.c b/fs/ocfs2/symlink.c index ed0a0cf..579dd1b 100644 --- a/fs/ocfs2/symlink.c +++ b/fs/ocfs2/symlink.c @@ -39,6 +39,7 @@ #include <linux/slab.h> #include <linux/pagemap.h> #include <linux/utsname.h> +#include <linux/namei.h> #define MLOG_MASK_PREFIX ML_NAMEI #include <cluster/masklog.h> @@ -54,26 +55,6 @@ #include "buffer_head_io.h" -static char *ocfs2_page_getlink(struct dentry * dentry, - struct page **ppage); -static char *ocfs2_fast_symlink_getlink(struct inode *inode, - struct buffer_head **bh); - -/* get the link contents into pagecache */ -static char *ocfs2_page_getlink(struct dentry * dentry, - struct page **ppage) -{ - struct page * page; - struct address_space *mapping = dentry->d_inode->i_mapping; - page = read_mapping_page(mapping, 0, NULL); - if (IS_ERR(page)) - goto sync_fail; - *ppage = page; - return kmap(page); - -sync_fail: - return (char*)page; -} static char *ocfs2_fast_symlink_getlink(struct inode *inode, struct buffer_head **bh) @@ -128,40 +109,55 @@ out: return ret; } -static void *ocfs2_follow_link(struct dentry *dentry, - struct nameidata *nd) +static void *ocfs2_fast_follow_link(struct dentry *dentry, + struct nameidata *nd) { - int status; - char *link; + int status = 0; + int len; + char *target, *link = ERR_PTR(-ENOMEM); struct inode *inode = dentry->d_inode; - struct page *page = NULL; struct buffer_head *bh = NULL; - - if (ocfs2_inode_is_fast_symlink(inode)) - link = ocfs2_fast_symlink_getlink(inode, &bh); - else - link = ocfs2_page_getlink(dentry, &page); - if (IS_ERR(link)) { - status = PTR_ERR(link); + + mlog_entry_void(); + + BUG_ON(!ocfs2_inode_is_fast_symlink(inode)); + target = ocfs2_fast_symlink_getlink(inode, &bh); + if (IS_ERR(target)) { + status = PTR_ERR(target); mlog_errno(status); goto bail; } - status = vfs_follow_link(nd, link); + /* Fast symlinks can't be large */ + len = strlen(target); + link = kzalloc(len + 1, GFP_NOFS); + if (!link) { + status = -ENOMEM; + mlog_errno(status); + goto bail; + } + + memcpy(link, target, len); + nd_set_link(nd, link); bail: - if (page) { - kunmap(page); - page_cache_release(page); - } brelse(bh); - return ERR_PTR(status); + mlog_exit(status); + return status ? ERR_PTR(status) : link; +} + +static void ocfs2_fast_put_link(struct dentry *dentry, struct nameidata *nd, void *cookie) +{ + char *link = cookie; + + kfree(link); } const struct inode_operations ocfs2_symlink_inode_operations = { .readlink = page_readlink, - .follow_link = ocfs2_follow_link, + .follow_link = page_follow_link_light, + .put_link = page_put_link, .getattr = ocfs2_getattr, .setattr = ocfs2_setattr, .setxattr = generic_setxattr, @@ -171,7 +167,8 @@ const struct inode_operations ocfs2_symlink_inode_operations = { }; const struct inode_operations ocfs2_fast_symlink_inode_operations = { .readlink = ocfs2_readlink, - .follow_link = ocfs2_follow_link, + .follow_link = ocfs2_fast_follow_link, + .put_link = ocfs2_fast_put_link, .getattr = ocfs2_getattr, .setattr = ocfs2_setattr, .setxattr = generic_setxattr, @@ -1033,7 +1033,7 @@ long do_sys_open(int dfd, const char __user *filename, int flags, int mode) if (!IS_ERR(tmp)) { fd = get_unused_fd_flags(flags); if (fd >= 0) { - struct file *f = do_filp_open(dfd, tmp, flags, mode); + struct file *f = do_filp_open(dfd, tmp, flags, mode, 0); if (IS_ERR(f)) { put_unused_fd(fd); fd = PTR_ERR(f); diff --git a/fs/proc/array.c b/fs/proc/array.c index 7e4877d..725a650 100644 --- a/fs/proc/array.c +++ b/fs/proc/array.c @@ -80,6 +80,7 @@ #include <linux/delayacct.h> #include <linux/seq_file.h> #include <linux/pid_namespace.h> +#include <linux/ptrace.h> #include <linux/tracehook.h> #include <asm/pgtable.h> @@ -352,6 +353,7 @@ static int do_task_stat(struct seq_file *m, struct pid_namespace *ns, char state; pid_t ppid = 0, pgid = -1, sid = -1; int num_threads = 0; + int permitted; struct mm_struct *mm; unsigned long long start_time; unsigned long cmin_flt = 0, cmaj_flt = 0; @@ -364,11 +366,14 @@ static int do_task_stat(struct seq_file *m, struct pid_namespace *ns, state = *get_task_state(task); vsize = eip = esp = 0; + permitted = ptrace_may_access(task, PTRACE_MODE_READ); mm = get_task_mm(task); if (mm) { vsize = task_vsize(mm); - eip = KSTK_EIP(task); - esp = KSTK_ESP(task); + if (permitted) { + eip = KSTK_EIP(task); + esp = KSTK_ESP(task); + } } get_task_comm(tcomm, task); @@ -424,7 +429,7 @@ static int do_task_stat(struct seq_file *m, struct pid_namespace *ns, unlock_task_sighand(task, &flags); } - if (!whole || num_threads < 2) + if (permitted && (!whole || num_threads < 2)) wchan = get_wchan(task); if (!whole) { min_flt = task->min_flt; @@ -476,7 +481,7 @@ static int do_task_stat(struct seq_file *m, struct pid_namespace *ns, rsslim, mm ? mm->start_code : 0, mm ? mm->end_code : 0, - mm ? mm->start_stack : 0, + (permitted && mm) ? mm->start_stack : 0, esp, eip, /* The signal information here is obsolete. diff --git a/fs/proc/base.c b/fs/proc/base.c index aa763ab..fb45615 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -322,7 +322,10 @@ static int proc_pid_wchan(struct task_struct *task, char *buffer) wchan = get_wchan(task); if (lookup_symbol_name(wchan, symname) < 0) - return sprintf(buffer, "%lu", wchan); + if (!ptrace_may_access(task, PTRACE_MODE_READ)) + return 0; + else + return sprintf(buffer, "%lu", wchan); else return sprintf(buffer, "%s", symname); } diff --git a/fs/proc/meminfo.c b/fs/proc/meminfo.c index 74ea974..c6b0302 100644 --- a/fs/proc/meminfo.c +++ b/fs/proc/meminfo.c @@ -35,7 +35,7 @@ static int meminfo_proc_show(struct seq_file *m, void *v) #define K(x) ((x) << (PAGE_SHIFT - 10)) si_meminfo(&i); si_swapinfo(&i); - committed = atomic_long_read(&vm_committed_space); + committed = percpu_counter_read_positive(&vm_committed_as); allowed = ((totalram_pages - hugetlb_total_pages()) * sysctl_overcommit_ratio / 100) + total_swap_pages; diff --git a/fs/proc/root.c b/fs/proc/root.c index 1e15a2b..b080b79 100644 --- a/fs/proc/root.c +++ b/fs/proc/root.c @@ -67,8 +67,7 @@ static int proc_get_sb(struct file_system_type *fs_type, sb->s_flags = flags; err = proc_fill_super(sb); if (err) { - up_write(&sb->s_umount); - deactivate_super(sb); + deactivate_locked_super(sb); return err; } diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index 39e4ad4..6f61b7c 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -665,6 +665,10 @@ static ssize_t pagemap_read(struct file *file, char __user *buf, goto out_task; ret = 0; + + if (!count) + goto out_task; + mm = get_task_mm(task); if (!mm) goto out_task; diff --git a/fs/quota/Makefile b/fs/quota/Makefile index 385a083..68d4f6d 100644 --- a/fs/quota/Makefile +++ b/fs/quota/Makefile @@ -1,12 +1,3 @@ -# -# Makefile for the Linux filesystems. -# -# 14 Sep 2000, Christoph Hellwig <hch@infradead.org> -# Rewritten to use lists instead of if-statements. -# - -obj-y := - obj-$(CONFIG_QUOTA) += dquot.o obj-$(CONFIG_QFMT_V1) += quota_v1.o obj-$(CONFIG_QFMT_V2) += quota_v2.o diff --git a/fs/reiserfs/dir.c b/fs/reiserfs/dir.c index 67a80d7..45ee3d3 100644 --- a/fs/reiserfs/dir.c +++ b/fs/reiserfs/dir.c @@ -41,6 +41,18 @@ static int reiserfs_dir_fsync(struct file *filp, struct dentry *dentry, #define store_ih(where,what) copy_item_head (where, what) +static inline bool is_privroot_deh(struct dentry *dir, + struct reiserfs_de_head *deh) +{ + int ret = 0; +#ifdef CONFIG_REISERFS_FS_XATTR + struct dentry *privroot = REISERFS_SB(dir->d_sb)->priv_root; + ret = (dir == dir->d_parent && privroot->d_inode && + deh->deh_objectid == INODE_PKEY(privroot->d_inode)->k_objectid); +#endif + return ret; +} + int reiserfs_readdir_dentry(struct dentry *dentry, void *dirent, filldir_t filldir, loff_t *pos) { @@ -138,18 +150,8 @@ int reiserfs_readdir_dentry(struct dentry *dentry, void *dirent, } /* Ignore the .reiserfs_priv entry */ - if (reiserfs_xattrs(inode->i_sb) && - !old_format_only(inode->i_sb) && - dentry == inode->i_sb->s_root && - REISERFS_SB(inode->i_sb)->priv_root && - REISERFS_SB(inode->i_sb)->priv_root->d_inode - && deh_objectid(deh) == - le32_to_cpu(INODE_PKEY - (REISERFS_SB(inode->i_sb)-> - priv_root->d_inode)-> - k_objectid)) { + if (is_privroot_deh(dentry, deh)) continue; - } d_off = deh_offset(deh); *pos = d_off; diff --git a/fs/reiserfs/namei.c b/fs/reiserfs/namei.c index efd4d72..2715791 100644 --- a/fs/reiserfs/namei.c +++ b/fs/reiserfs/namei.c @@ -338,21 +338,8 @@ static struct dentry *reiserfs_lookup(struct inode *dir, struct dentry *dentry, &path_to_entry, &de); pathrelse(&path_to_entry); if (retval == NAME_FOUND) { - /* Hide the .reiserfs_priv directory */ - if (reiserfs_xattrs(dir->i_sb) && - !old_format_only(dir->i_sb) && - REISERFS_SB(dir->i_sb)->priv_root && - REISERFS_SB(dir->i_sb)->priv_root->d_inode && - de.de_objectid == - le32_to_cpu(INODE_PKEY - (REISERFS_SB(dir->i_sb)->priv_root->d_inode)-> - k_objectid)) { - reiserfs_write_unlock(dir->i_sb); - return ERR_PTR(-EACCES); - } - - inode = - reiserfs_iget(dir->i_sb, (struct cpu_key *)&(de.de_dir_id)); + inode = reiserfs_iget(dir->i_sb, + (struct cpu_key *)&(de.de_dir_id)); if (!inode || IS_ERR(inode)) { reiserfs_write_unlock(dir->i_sb); return ERR_PTR(-EACCES); diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c index 0ae6486..3567fb9 100644 --- a/fs/reiserfs/super.c +++ b/fs/reiserfs/super.c @@ -448,13 +448,11 @@ int remove_save_link(struct inode *inode, int truncate) static void reiserfs_kill_sb(struct super_block *s) { if (REISERFS_SB(s)) { -#ifdef CONFIG_REISERFS_FS_XATTR if (REISERFS_SB(s)->xattr_root) { d_invalidate(REISERFS_SB(s)->xattr_root); dput(REISERFS_SB(s)->xattr_root); REISERFS_SB(s)->xattr_root = NULL; } -#endif if (REISERFS_SB(s)->priv_root) { d_invalidate(REISERFS_SB(s)->priv_root); dput(REISERFS_SB(s)->priv_root); @@ -1316,8 +1314,7 @@ static int reiserfs_remount(struct super_block *s, int *mount_flags, char *arg) } out_ok: - kfree(s->s_options); - s->s_options = new_opts; + replace_mount_options(s, new_opts); return 0; out_err: @@ -1842,7 +1839,8 @@ static int reiserfs_fill_super(struct super_block *s, void *data, int silent) goto error; } - if ((errval = reiserfs_xattr_init(s, s->s_flags))) { + if ((errval = reiserfs_lookup_privroot(s)) || + (errval = reiserfs_xattr_init(s, s->s_flags))) { dput(s->s_root); s->s_root = NULL; goto error; @@ -1855,7 +1853,8 @@ static int reiserfs_fill_super(struct super_block *s, void *data, int silent) reiserfs_info(s, "using 3.5.x disk format\n"); } - if ((errval = reiserfs_xattr_init(s, s->s_flags))) { + if ((errval = reiserfs_lookup_privroot(s)) || + (errval = reiserfs_xattr_init(s, s->s_flags))) { dput(s->s_root); s->s_root = NULL; goto error; diff --git a/fs/reiserfs/xattr.c b/fs/reiserfs/xattr.c index f83f52b..8e7deb0 100644 --- a/fs/reiserfs/xattr.c +++ b/fs/reiserfs/xattr.c @@ -113,41 +113,30 @@ static int xattr_rmdir(struct inode *dir, struct dentry *dentry) #define xattr_may_create(flags) (!flags || flags & XATTR_CREATE) -/* Returns and possibly creates the xattr dir. */ -static struct dentry *lookup_or_create_dir(struct dentry *parent, - const char *name, int flags) +static struct dentry *open_xa_root(struct super_block *sb, int flags) { - struct dentry *dentry; - BUG_ON(!parent); - - dentry = lookup_one_len(name, parent, strlen(name)); - if (IS_ERR(dentry)) - return dentry; - else if (!dentry->d_inode) { - int err = -ENODATA; + struct dentry *privroot = REISERFS_SB(sb)->priv_root; + struct dentry *xaroot; + if (!privroot->d_inode) + return ERR_PTR(-ENODATA); - if (xattr_may_create(flags)) { - mutex_lock_nested(&parent->d_inode->i_mutex, - I_MUTEX_XATTR); - err = xattr_mkdir(parent->d_inode, dentry, 0700); - mutex_unlock(&parent->d_inode->i_mutex); - } + mutex_lock_nested(&privroot->d_inode->i_mutex, I_MUTEX_XATTR); + xaroot = dget(REISERFS_SB(sb)->xattr_root); + if (!xaroot) + xaroot = ERR_PTR(-ENODATA); + else if (!xaroot->d_inode) { + int err = -ENODATA; + if (xattr_may_create(flags)) + err = xattr_mkdir(privroot->d_inode, xaroot, 0700); if (err) { - dput(dentry); - dentry = ERR_PTR(err); + dput(xaroot); + xaroot = ERR_PTR(err); } } - return dentry; -} - -static struct dentry *open_xa_root(struct super_block *sb, int flags) -{ - struct dentry *privroot = REISERFS_SB(sb)->priv_root; - if (!privroot) - return ERR_PTR(-ENODATA); - return lookup_or_create_dir(privroot, XAROOT_NAME, flags); + mutex_unlock(&privroot->d_inode->i_mutex); + return xaroot; } static struct dentry *open_xa_dir(const struct inode *inode, int flags) @@ -163,10 +152,22 @@ static struct dentry *open_xa_dir(const struct inode *inode, int flags) le32_to_cpu(INODE_PKEY(inode)->k_objectid), inode->i_generation); - xadir = lookup_or_create_dir(xaroot, namebuf, flags); + mutex_lock_nested(&xaroot->d_inode->i_mutex, I_MUTEX_XATTR); + + xadir = lookup_one_len(namebuf, xaroot, strlen(namebuf)); + if (!IS_ERR(xadir) && !xadir->d_inode) { + int err = -ENODATA; + if (xattr_may_create(flags)) + err = xattr_mkdir(xaroot->d_inode, xadir, 0700); + if (err) { + dput(xadir); + xadir = ERR_PTR(err); + } + } + + mutex_unlock(&xaroot->d_inode->i_mutex); dput(xaroot); return xadir; - } /* The following are side effects of other operations that aren't explicitly @@ -184,6 +185,7 @@ fill_with_dentries(void *buf, const char *name, int namelen, loff_t offset, { struct reiserfs_dentry_buf *dbuf = buf; struct dentry *dentry; + WARN_ON_ONCE(!mutex_is_locked(&dbuf->xadir->d_inode->i_mutex)); if (dbuf->count == ARRAY_SIZE(dbuf->dentries)) return -ENOSPC; @@ -349,6 +351,7 @@ static struct dentry *xattr_lookup(struct inode *inode, const char *name, if (IS_ERR(xadir)) return ERR_CAST(xadir); + mutex_lock_nested(&xadir->d_inode->i_mutex, I_MUTEX_XATTR); xafile = lookup_one_len(name, xadir, strlen(name)); if (IS_ERR(xafile)) { err = PTR_ERR(xafile); @@ -360,18 +363,15 @@ static struct dentry *xattr_lookup(struct inode *inode, const char *name, if (!xafile->d_inode) { err = -ENODATA; - if (xattr_may_create(flags)) { - mutex_lock_nested(&xadir->d_inode->i_mutex, - I_MUTEX_XATTR); + if (xattr_may_create(flags)) err = xattr_create(xadir->d_inode, xafile, 0700|S_IFREG); - mutex_unlock(&xadir->d_inode->i_mutex); - } } if (err) dput(xafile); out: + mutex_unlock(&xadir->d_inode->i_mutex); dput(xadir); if (err) return ERR_PTR(err); @@ -435,6 +435,7 @@ static int lookup_and_delete_xattr(struct inode *inode, const char *name) if (IS_ERR(xadir)) return PTR_ERR(xadir); + mutex_lock_nested(&xadir->d_inode->i_mutex, I_MUTEX_XATTR); dentry = lookup_one_len(name, xadir, strlen(name)); if (IS_ERR(dentry)) { err = PTR_ERR(dentry); @@ -442,14 +443,13 @@ static int lookup_and_delete_xattr(struct inode *inode, const char *name) } if (dentry->d_inode) { - mutex_lock_nested(&xadir->d_inode->i_mutex, I_MUTEX_XATTR); err = xattr_unlink(xadir->d_inode, dentry); - mutex_unlock(&xadir->d_inode->i_mutex); update_ctime(inode); } dput(dentry); out_dput: + mutex_unlock(&xadir->d_inode->i_mutex); dput(xadir); return err; } @@ -687,20 +687,6 @@ out: return err; } -/* Actual operations that are exported to VFS-land */ -struct xattr_handler *reiserfs_xattr_handlers[] = { - &reiserfs_xattr_user_handler, - &reiserfs_xattr_trusted_handler, -#ifdef CONFIG_REISERFS_FS_SECURITY - &reiserfs_xattr_security_handler, -#endif -#ifdef CONFIG_REISERFS_FS_POSIX_ACL - &reiserfs_posix_acl_access_handler, - &reiserfs_posix_acl_default_handler, -#endif - NULL -}; - /* * In order to implement different sets of xattr operations for each xattr * prefix with the generic xattr API, a filesystem should create a @@ -843,7 +829,7 @@ ssize_t reiserfs_listxattr(struct dentry * dentry, char *buffer, size_t size) if (!dentry->d_inode) return -EINVAL; - if (!reiserfs_xattrs(dentry->d_sb) || + if (!dentry->d_sb->s_xattr || get_inode_sd_version(dentry->d_inode) == STAT_DATA_V1) return -EOPNOTSUPP; @@ -885,42 +871,50 @@ static int reiserfs_check_acl(struct inode *inode, int mask) return error; } -int reiserfs_permission(struct inode *inode, int mask) -{ - /* - * We don't do permission checks on the internal objects. - * Permissions are determined by the "owning" object. - */ - if (IS_PRIVATE(inode)) - return 0; - /* - * Stat data v1 doesn't support ACLs. - */ - if (get_inode_sd_version(inode) == STAT_DATA_V1) - return generic_permission(inode, mask, NULL); - else - return generic_permission(inode, mask, reiserfs_check_acl); -} - static int create_privroot(struct dentry *dentry) { int err; struct inode *inode = dentry->d_parent->d_inode; - mutex_lock_nested(&inode->i_mutex, I_MUTEX_XATTR); + WARN_ON_ONCE(!mutex_is_locked(&inode->i_mutex)); + err = xattr_mkdir(inode, dentry, 0700); - mutex_unlock(&inode->i_mutex); - if (err) { - dput(dentry); - dentry = NULL; + if (err || !dentry->d_inode) { + reiserfs_warning(dentry->d_sb, "jdm-20006", + "xattrs/ACLs enabled and couldn't " + "find/create .reiserfs_priv. " + "Failing mount."); + return -EOPNOTSUPP; } - if (dentry && dentry->d_inode) - reiserfs_info(dentry->d_sb, "Created %s - reserved for xattr " - "storage.\n", PRIVROOT_NAME); + dentry->d_inode->i_flags |= S_PRIVATE; + reiserfs_info(dentry->d_sb, "Created %s - reserved for xattr " + "storage.\n", PRIVROOT_NAME); - return err; + return 0; } +#else +int __init reiserfs_xattr_register_handlers(void) { return 0; } +void reiserfs_xattr_unregister_handlers(void) {} +static int create_privroot(struct dentry *dentry) { return 0; } +#endif + +/* Actual operations that are exported to VFS-land */ +struct xattr_handler *reiserfs_xattr_handlers[] = { +#ifdef CONFIG_REISERFS_FS_XATTR + &reiserfs_xattr_user_handler, + &reiserfs_xattr_trusted_handler, +#endif +#ifdef CONFIG_REISERFS_FS_SECURITY + &reiserfs_xattr_security_handler, +#endif +#ifdef CONFIG_REISERFS_FS_POSIX_ACL + &reiserfs_posix_acl_access_handler, + &reiserfs_posix_acl_default_handler, +#endif + NULL +}; + static int xattr_mount_check(struct super_block *s) { /* We need generation numbers to ensure that the oid mapping is correct @@ -940,21 +934,33 @@ static int xattr_mount_check(struct super_block *s) return 0; } -#else -int __init reiserfs_xattr_register_handlers(void) { return 0; } -void reiserfs_xattr_unregister_handlers(void) {} +int reiserfs_permission(struct inode *inode, int mask) +{ + /* + * We don't do permission checks on the internal objects. + * Permissions are determined by the "owning" object. + */ + if (IS_PRIVATE(inode)) + return 0; + +#ifdef CONFIG_REISERFS_FS_XATTR + /* + * Stat data v1 doesn't support ACLs. + */ + if (get_inode_sd_version(inode) != STAT_DATA_V1) + return generic_permission(inode, mask, reiserfs_check_acl); #endif + return generic_permission(inode, mask, NULL); +} /* This will catch lookups from the fs root to .reiserfs_priv */ static int xattr_lookup_poison(struct dentry *dentry, struct qstr *q1, struct qstr *name) { struct dentry *priv_root = REISERFS_SB(dentry->d_sb)->priv_root; - if (name->len == priv_root->d_name.len && - name->hash == priv_root->d_name.hash && - !memcmp(name->name, priv_root->d_name.name, name->len)) { + if (container_of(q1, struct dentry, d_name) == priv_root) return -ENOENT; - } else if (q1->len == name->len && + if (q1->len == name->len && !memcmp(q1->name, name->name, name->len)) return 0; return 1; @@ -964,73 +970,71 @@ static const struct dentry_operations xattr_lookup_poison_ops = { .d_compare = xattr_lookup_poison, }; +int reiserfs_lookup_privroot(struct super_block *s) +{ + struct dentry *dentry; + int err = 0; + + /* If we don't have the privroot located yet - go find it */ + mutex_lock(&s->s_root->d_inode->i_mutex); + dentry = lookup_one_len(PRIVROOT_NAME, s->s_root, + strlen(PRIVROOT_NAME)); + if (!IS_ERR(dentry)) { + REISERFS_SB(s)->priv_root = dentry; + s->s_root->d_op = &xattr_lookup_poison_ops; + if (dentry->d_inode) + dentry->d_inode->i_flags |= S_PRIVATE; + } else + err = PTR_ERR(dentry); + mutex_unlock(&s->s_root->d_inode->i_mutex); + + return err; +} + /* We need to take a copy of the mount flags since things like * MS_RDONLY don't get set until *after* we're called. * mount_flags != mount_options */ int reiserfs_xattr_init(struct super_block *s, int mount_flags) { int err = 0; + struct dentry *privroot = REISERFS_SB(s)->priv_root; -#ifdef CONFIG_REISERFS_FS_XATTR err = xattr_mount_check(s); if (err) goto error; -#endif - /* If we don't have the privroot located yet - go find it */ - if (!REISERFS_SB(s)->priv_root) { - struct dentry *dentry; - dentry = lookup_one_len(PRIVROOT_NAME, s->s_root, - strlen(PRIVROOT_NAME)); - if (!IS_ERR(dentry)) { -#ifdef CONFIG_REISERFS_FS_XATTR - if (!(mount_flags & MS_RDONLY) && !dentry->d_inode) - err = create_privroot(dentry); -#endif - if (!dentry->d_inode) { - dput(dentry); - dentry = NULL; - } - } else - err = PTR_ERR(dentry); - - if (!err && dentry) { - s->s_root->d_op = &xattr_lookup_poison_ops; - dentry->d_inode->i_flags |= S_PRIVATE; - REISERFS_SB(s)->priv_root = dentry; -#ifdef CONFIG_REISERFS_FS_XATTR - /* xattrs are unavailable */ - } else if (!(mount_flags & MS_RDONLY)) { - /* If we're read-only it just means that the dir - * hasn't been created. Not an error -- just no - * xattrs on the fs. We'll check again if we - * go read-write */ - reiserfs_warning(s, "jdm-20006", - "xattrs/ACLs enabled and couldn't " - "find/create .reiserfs_priv. " - "Failing mount."); - err = -EOPNOTSUPP; -#endif - } + if (!privroot->d_inode && !(mount_flags & MS_RDONLY)) { + mutex_lock(&s->s_root->d_inode->i_mutex); + err = create_privroot(REISERFS_SB(s)->priv_root); + mutex_unlock(&s->s_root->d_inode->i_mutex); } -#ifdef CONFIG_REISERFS_FS_XATTR - if (!err) + if (privroot->d_inode) { s->s_xattr = reiserfs_xattr_handlers; + mutex_lock(&privroot->d_inode->i_mutex); + if (!REISERFS_SB(s)->xattr_root) { + struct dentry *dentry; + dentry = lookup_one_len(XAROOT_NAME, privroot, + strlen(XAROOT_NAME)); + if (!IS_ERR(dentry)) + REISERFS_SB(s)->xattr_root = dentry; + else + err = PTR_ERR(dentry); + } + mutex_unlock(&privroot->d_inode->i_mutex); + } error: if (err) { clear_bit(REISERFS_XATTRS_USER, &(REISERFS_SB(s)->s_mount_opt)); clear_bit(REISERFS_POSIXACL, &(REISERFS_SB(s)->s_mount_opt)); } -#endif /* The super_block MS_POSIXACL must mirror the (no)acl mount option. */ - s->s_flags = s->s_flags & ~MS_POSIXACL; -#ifdef CONFIG_REISERFS_FS_POSIX_ACL if (reiserfs_posixacl(s)) s->s_flags |= MS_POSIXACL; -#endif + else + s->s_flags &= ~MS_POSIXACL; return err; } diff --git a/fs/reiserfs/xattr_security.c b/fs/reiserfs/xattr_security.c index 4d3c20e..a92c879 100644 --- a/fs/reiserfs/xattr_security.c +++ b/fs/reiserfs/xattr_security.c @@ -55,8 +55,16 @@ int reiserfs_security_init(struct inode *dir, struct inode *inode, struct reiserfs_security_handle *sec) { int blocks = 0; - int error = security_inode_init_security(inode, dir, &sec->name, - &sec->value, &sec->length); + int error; + + sec->name = NULL; + + /* Don't add selinux attributes on xattrs - they'll never get used */ + if (IS_PRIVATE(dir)) + return 0; + + error = security_inode_init_security(inode, dir, &sec->name, + &sec->value, &sec->length); if (error) { if (error == -EOPNOTSUPP) error = 0; diff --git a/fs/romfs/super.c b/fs/romfs/super.c index c53b5ef..4ab3c03 100644 --- a/fs/romfs/super.c +++ b/fs/romfs/super.c @@ -298,7 +298,8 @@ static struct inode *romfs_iget(struct super_block *sb, unsigned long pos) struct romfs_inode ri; struct inode *i; unsigned long nlen; - unsigned nextfh, ret; + unsigned nextfh; + int ret; umode_t mode; /* we might have to traverse a chain of "hard link" file entries to get diff --git a/fs/squashfs/Makefile b/fs/squashfs/Makefile index 8258cf9..70e3244 100644 --- a/fs/squashfs/Makefile +++ b/fs/squashfs/Makefile @@ -5,4 +5,3 @@ obj-$(CONFIG_SQUASHFS) += squashfs.o squashfs-y += block.o cache.o dir.o export.o file.o fragment.o id.o inode.o squashfs-y += namei.o super.o symlink.o -#squashfs-y += squashfs2_0.o diff --git a/fs/squashfs/cache.c b/fs/squashfs/cache.c index 1c4739e..40c98fa 100644 --- a/fs/squashfs/cache.c +++ b/fs/squashfs/cache.c @@ -252,6 +252,7 @@ struct squashfs_cache *squashfs_cache_init(char *name, int entries, cache->entries = entries; cache->block_size = block_size; cache->pages = block_size >> PAGE_CACHE_SHIFT; + cache->pages = cache->pages ? cache->pages : 1; cache->name = name; cache->num_waiters = 0; spin_lock_init(&cache->lock); diff --git a/fs/squashfs/super.c b/fs/squashfs/super.c index ffa6edc..0adc624 100644 --- a/fs/squashfs/super.c +++ b/fs/squashfs/super.c @@ -157,6 +157,16 @@ static int squashfs_fill_super(struct super_block *sb, void *data, int silent) if (msblk->block_size > SQUASHFS_FILE_MAX_SIZE) goto failed_mount; + /* + * Check the system page size is not larger than the filesystem + * block size (by default 128K). This is currently not supported. + */ + if (PAGE_CACHE_SIZE > msblk->block_size) { + ERROR("Page size > filesystem block size (%d). This is " + "currently not supported!\n", msblk->block_size); + goto failed_mount; + } + msblk->block_log = le16_to_cpu(sblk->block_log); if (msblk->block_log > SQUASHFS_FILE_MAX_LOG) goto failed_mount; @@ -208,6 +208,34 @@ void deactivate_super(struct super_block *s) EXPORT_SYMBOL(deactivate_super); /** + * deactivate_locked_super - drop an active reference to superblock + * @s: superblock to deactivate + * + * Equivalent of up_write(&s->s_umount); deactivate_super(s);, except that + * it does not unlock it until it's all over. As the result, it's safe to + * use to dispose of new superblock on ->get_sb() failure exits - nobody + * will see the sucker until it's all over. Equivalent using up_write + + * deactivate_super is safe for that purpose only if superblock is either + * safe to use or has NULL ->s_root when we unlock. + */ +void deactivate_locked_super(struct super_block *s) +{ + struct file_system_type *fs = s->s_type; + if (atomic_dec_and_lock(&s->s_active, &sb_lock)) { + s->s_count -= S_BIAS-1; + spin_unlock(&sb_lock); + vfs_dq_off(s, 0); + fs->kill_sb(s); + put_filesystem(fs); + put_super(s); + } else { + up_write(&s->s_umount); + } +} + +EXPORT_SYMBOL(deactivate_locked_super); + +/** * grab_super - acquire an active reference * @s: reference we are trying to make active * @@ -797,8 +825,7 @@ int get_sb_ns(struct file_system_type *fs_type, int flags, void *data, sb->s_flags = flags; err = fill_super(sb, data, flags & MS_SILENT ? 1 : 0); if (err) { - up_write(&sb->s_umount); - deactivate_super(sb); + deactivate_locked_super(sb); return err; } @@ -854,8 +881,7 @@ int get_sb_bdev(struct file_system_type *fs_type, if (s->s_root) { if ((flags ^ s->s_flags) & MS_RDONLY) { - up_write(&s->s_umount); - deactivate_super(s); + deactivate_locked_super(s); error = -EBUSY; goto error_bdev; } @@ -870,8 +896,7 @@ int get_sb_bdev(struct file_system_type *fs_type, sb_set_blocksize(s, block_size(bdev)); error = fill_super(s, data, flags & MS_SILENT ? 1 : 0); if (error) { - up_write(&s->s_umount); - deactivate_super(s); + deactivate_locked_super(s); goto error; } @@ -897,7 +922,7 @@ void kill_block_super(struct super_block *sb) struct block_device *bdev = sb->s_bdev; fmode_t mode = sb->s_mode; - bdev->bd_super = 0; + bdev->bd_super = NULL; generic_shutdown_super(sb); sync_blockdev(bdev); close_bdev_exclusive(bdev, mode); @@ -921,8 +946,7 @@ int get_sb_nodev(struct file_system_type *fs_type, error = fill_super(s, data, flags & MS_SILENT ? 1 : 0); if (error) { - up_write(&s->s_umount); - deactivate_super(s); + deactivate_locked_super(s); return error; } s->s_flags |= MS_ACTIVE; @@ -952,8 +976,7 @@ int get_sb_single(struct file_system_type *fs_type, s->s_flags = flags; error = fill_super(s, data, flags & MS_SILENT ? 1 : 0); if (error) { - up_write(&s->s_umount); - deactivate_super(s); + deactivate_locked_super(s); return error; } s->s_flags |= MS_ACTIVE; @@ -1006,8 +1029,7 @@ vfs_kern_mount(struct file_system_type *type, int flags, const char *name, void return mnt; out_sb: dput(mnt->mnt_root); - up_write(&mnt->mnt_sb->s_umount); - deactivate_super(mnt->mnt_sb); + deactivate_locked_super(mnt->mnt_sb); out_free_secdata: free_secdata(secdata); out_mnt: diff --git a/fs/ubifs/super.c b/fs/ubifs/super.c index faa44f9..e9f7a75 100644 --- a/fs/ubifs/super.c +++ b/fs/ubifs/super.c @@ -2055,8 +2055,7 @@ static int ubifs_get_sb(struct file_system_type *fs_type, int flags, return 0; out_deact: - up_write(&sb->s_umount); - deactivate_super(sb); + deactivate_locked_super(sb); out_close: ubi_close_volume(ubi); return err; diff --git a/fs/ufs/dir.c b/fs/ufs/dir.c index dbbbc46..6321b79 100644 --- a/fs/ufs/dir.c +++ b/fs/ufs/dir.c @@ -666,6 +666,6 @@ not_empty: const struct file_operations ufs_dir_operations = { .read = generic_read_dir, .readdir = ufs_readdir, - .fsync = file_fsync, + .fsync = ufs_sync_file, .llseek = generic_file_llseek, }; diff --git a/fs/ufs/file.c b/fs/ufs/file.c index 625ef17..2bd3a16 100644 --- a/fs/ufs/file.c +++ b/fs/ufs/file.c @@ -30,7 +30,7 @@ #include "ufs.h" -static int ufs_sync_file(struct file *file, struct dentry *dentry, int datasync) +int ufs_sync_file(struct file *file, struct dentry *dentry, int datasync) { struct inode *inode = dentry->d_inode; int err; diff --git a/fs/ufs/ufs.h b/fs/ufs/ufs.h index 69b3427..d0c4acd 100644 --- a/fs/ufs/ufs.h +++ b/fs/ufs/ufs.h @@ -98,8 +98,8 @@ extern void ufs_set_link(struct inode *dir, struct ufs_dir_entry *de, /* file.c */ extern const struct inode_operations ufs_file_inode_operations; extern const struct file_operations ufs_file_operations; - extern const struct address_space_operations ufs_aops; +extern int ufs_sync_file(struct file *, struct dentry *, int); /* ialloc.c */ extern void ufs_free_inode (struct inode *inode); diff --git a/fs/xfs/xfs_bmap.c b/fs/xfs/xfs_bmap.c index 3a6ed42..ca7c600 100644 --- a/fs/xfs/xfs_bmap.c +++ b/fs/xfs/xfs_bmap.c @@ -5880,7 +5880,7 @@ xfs_getbmap( void *arg) /* formatter arg */ { __int64_t bmvend; /* last block requested */ - int error; /* return value */ + int error = 0; /* return value */ __int64_t fixlen; /* length for -1 case */ int i; /* extent number */ int lock; /* lock state */ @@ -5890,39 +5890,18 @@ xfs_getbmap( int nexleft; /* # of user extents left */ int subnex; /* # of bmapi's can do */ int nmap; /* number of map entries */ - struct getbmapx out; /* output structure */ + 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 the BMV_IF_NO_DMAPI_READ interface bit specified, do not - * generate a DMAPI read event. Otherwise, if the DM_EVENT_READ - * bit is set for the file, generate a read event in order - * that the DMAPI application may do its thing before we return - * the extents. Usually this means restoring user file data to - * regions of the file that look like holes. - * - * The "old behavior" (from XFS_IOC_GETBMAP) is to not specify - * BMV_IF_NO_DMAPI_READ so that read events are generated. - * If this were not true, callers of ioctl( XFS_IOC_GETBMAP ) - * could misinterpret holes in a DMAPI file as true holes, - * when in fact they may represent offline user data. - */ - if ((iflags & BMV_IF_NO_DMAPI_READ) == 0 && - DM_EVENT_ENABLED(ip, DM_EVENT_READ) && - whichfork == XFS_DATA_FORK) { - error = XFS_SEND_DATA(mp, DM_EVENT_READ, ip, 0, 0, 0, NULL); - if (error) - return XFS_ERROR(error); - } - if (whichfork == XFS_ATTR_FORK) { if (XFS_IFORK_Q(ip)) { if (ip->i_d.di_aformat != XFS_DINODE_FMT_EXTENTS && @@ -5936,11 +5915,37 @@ xfs_getbmap( ip->i_mount); return XFS_ERROR(EFSCORRUPTED); } - } 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 (whichfork == XFS_DATA_FORK) { + + prealloced = 0; + fixlen = 1LL << 32; + } else { + /* + * If the BMV_IF_NO_DMAPI_READ interface bit specified, do + * not generate a DMAPI read event. Otherwise, if the + * DM_EVENT_READ bit is set for the file, generate a read + * event in order that the DMAPI application may do its thing + * before we return the extents. Usually this means restoring + * user file data to regions of the file that look like holes. + * + * The "old behavior" (from XFS_IOC_GETBMAP) is to not specify + * BMV_IF_NO_DMAPI_READ so that read events are generated. + * If this were not true, callers of ioctl(XFS_IOC_GETBMAP) + * could misinterpret holes in a DMAPI file as true holes, + * when in fact they may represent offline user data. + */ + if (DM_EVENT_ENABLED(ip, DM_EVENT_READ) && + !(iflags & BMV_IF_NO_DMAPI_READ)) { + error = XFS_SEND_DATA(mp, DM_EVENT_READ, ip, + 0, 0, 0, NULL); + if (error) + return XFS_ERROR(error); + } + + 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; @@ -5949,42 +5954,41 @@ xfs_getbmap( prealloced = 0; fixlen = ip->i_size; } - } else { - prealloced = 0; - fixlen = 1LL << 32; } if (bmv->bmv_length == -1) { fixlen = XFS_FSB_TO_BB(mp, XFS_B_TO_FSB(mp, fixlen)); - bmv->bmv_length = MAX( (__int64_t)(fixlen - bmv->bmv_offset), - (__int64_t)0); - } else if (bmv->bmv_length < 0) - return XFS_ERROR(EINVAL); - if (bmv->bmv_length == 0) { + 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; - xfs_ilock(ip, XFS_IOLOCK_SHARED); - if (((iflags & BMV_IF_DELALLOC) == 0) && - (whichfork == XFS_DATA_FORK) && - (ip->i_delayed_blks || ip->i_size > ip->i_d.di_size)) { - /* xfs_fsize_t last_byte = xfs_file_last_byte(ip); */ - error = xfs_flush_pages(ip, (xfs_off_t)0, - -1, 0, FI_REMAPF); - if (error) { - xfs_iunlock(ip, XFS_IOLOCK_SHARED); - return error; + 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) + return XFS_ERROR(ENOMEM); + + xfs_ilock(ip, XFS_IOLOCK_SHARED); + if (whichfork == XFS_DATA_FORK && !(iflags & BMV_IF_DELALLOC)) { + if (ip->i_delayed_blks || ip->i_size > ip->i_d.di_size) { + error = xfs_flush_pages(ip, 0, -1, 0, FI_REMAPF); + if (error) + goto out_unlock_iolock; } - } - ASSERT(whichfork == XFS_ATTR_FORK || (iflags & BMV_IF_DELALLOC) || - ip->i_delayed_blks == 0); + ASSERT(ip->i_delayed_blks == 0); + } lock = xfs_ilock_map_shared(ip); @@ -5995,23 +5999,25 @@ xfs_getbmap( if (nex > XFS_IFORK_NEXTENTS(ip, whichfork) * 2 + 1) nex = XFS_IFORK_NEXTENTS(ip, whichfork) * 2 + 1; - bmapi_flags = xfs_bmapi_aflag(whichfork) | - ((iflags & BMV_IF_PREALLOC) ? 0 : XFS_BMAPI_IGSTATE); + 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_SLEEP); + map = kmem_alloc(subnex * sizeof(*map), KM_MAYFAIL); + if (!map) + goto out_unlock_ilock; bmv->bmv_entries = 0; - if ((XFS_IFORK_NEXTENTS(ip, whichfork) == 0)) { - if (((iflags & BMV_IF_DELALLOC) == 0) || - whichfork == XFS_ATTR_FORK) { - error = 0; - goto unlock_and_return; - } + if (XFS_IFORK_NEXTENTS(ip, whichfork) == 0 && + (whichfork == XFS_ATTR_FORK || !(iflags & BMV_IF_DELALLOC))) { + error = 0; + goto out_free_map; } nexleft = nex; @@ -6023,53 +6029,61 @@ xfs_getbmap( bmapi_flags, NULL, 0, map, &nmap, NULL, NULL); if (error) - goto unlock_and_return; + goto out_free_map; ASSERT(nmap <= subnex); for (i = 0; i < nmap && nexleft && bmv->bmv_length; i++) { - out.bmv_oflags = 0; + out[cur_ext].bmv_oflags = 0; if (map[i].br_state == XFS_EXT_UNWRITTEN) - out.bmv_oflags |= BMV_OF_PREALLOC; + out[cur_ext].bmv_oflags |= BMV_OF_PREALLOC; else if (map[i].br_startblock == DELAYSTARTBLOCK) - out.bmv_oflags |= BMV_OF_DELALLOC; - out.bmv_offset = XFS_FSB_TO_BB(mp, map[i].br_startoff); - out.bmv_length = XFS_FSB_TO_BB(mp, map[i].br_blockcount); - out.bmv_unused1 = out.bmv_unused2 = 0; + 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; ASSERT(((iflags & BMV_IF_DELALLOC) != 0) || (map[i].br_startblock != DELAYSTARTBLOCK)); if (map[i].br_startblock == HOLESTARTBLOCK && whichfork == XFS_ATTR_FORK) { /* came to the end of attribute fork */ - out.bmv_oflags |= BMV_OF_LAST; - goto unlock_and_return; - } else { - int full = 0; /* user array is full */ - - if (!xfs_getbmapx_fix_eof_hole(ip, &out, - prealloced, bmvend, - map[i].br_startblock)) { - goto unlock_and_return; - } - - /* format results & advance arg */ - error = formatter(&arg, &out, &full); - if (error || full) - goto unlock_and_return; - nexleft--; - bmv->bmv_offset = - out.bmv_offset + out.bmv_length; - bmv->bmv_length = MAX((__int64_t)0, - (__int64_t)(bmvend - bmv->bmv_offset)); - bmv->bmv_entries++; + 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; + + nexleft--; + 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); + bmv->bmv_entries++; + cur_ext++; } } while (nmap && nexleft && bmv->bmv_length); -unlock_and_return: + out_free_map: + kmem_free(map); + out_unlock_ilock: xfs_iunlock_map_shared(ip, lock); + out_unlock_iolock: xfs_iunlock(ip, XFS_IOLOCK_SHARED); - kmem_free(map); + 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; + } return error; } diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index e7ae08d..123b20c 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -1258,8 +1258,10 @@ xfs_file_last_byte( * necessary. */ if (ip->i_df.if_flags & XFS_IFEXTENTS) { + xfs_ilock(ip, XFS_ILOCK_SHARED); error = xfs_bmap_last_offset(NULL, ip, &last_block, XFS_DATA_FORK); + xfs_iunlock(ip, XFS_ILOCK_SHARED); if (error) { last_block = 0; } diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c index b101990..65a9972 100644 --- a/fs/xfs/xfs_mount.c +++ b/fs/xfs/xfs_mount.c @@ -291,14 +291,17 @@ xfs_mount_validate_sb( 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) || |