From a69c7c077228b0bef38ccb1d385c099e132fe54b Mon Sep 17 00:00:00 2001 From: Eric Sandeen Date: Wed, 28 Mar 2012 12:21:11 -0500 Subject: xfs: use XFS_BMAP_BMDR_SPACE vs. XFS_BROOT_SIZE_ADJ XFS_BROOT_SIZE_ADJ is an undocumented macro which accounts for the difference in size between the on-disk and in-core btree root. It's much clearer to just use the newly-added XFS_BMAP_BMDR_SPACE macro which gives us the on-disk size directly. In one case, we must test that the if_broot exists before applying the macro, however. Signed-off-by: Eric Sandeen Reviewed-by: Dave Chinner Reviewed-by: Ben Myers Signed-off-by: Ben Myers diff --git a/fs/xfs/xfs_dinode.h b/fs/xfs/xfs_dinode.h index f7a0e95..07d735a 100644 --- a/fs/xfs/xfs_dinode.h +++ b/fs/xfs/xfs_dinode.h @@ -132,9 +132,6 @@ typedef enum xfs_dinode_fmt { #define XFS_LITINO(mp, version) \ ((int)(((mp)->m_sb.sb_inodesize) - xfs_dinode_size(version))) -#define XFS_BROOT_SIZE_ADJ(ip) \ - (XFS_BMBT_BLOCK_LEN((ip)->i_mount) - sizeof(xfs_bmdr_block_t)) - /* * Inode data & attribute fork sizes, per inode. */ diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index 9ecfe1e..b78481f 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -2156,8 +2156,8 @@ xfs_iroot_realloc( np = (char *)XFS_BMAP_BROOT_PTR_ADDR(mp, ifp->if_broot, 1, (int)new_size); ifp->if_broot_bytes = (int)new_size; - ASSERT(ifp->if_broot_bytes <= - XFS_IFORK_SIZE(ip, whichfork) + XFS_BROOT_SIZE_ADJ(ip)); + ASSERT(XFS_BMAP_BMDR_SPACE(ifp->if_broot) <= + XFS_IFORK_SIZE(ip, whichfork)); memmove(np, op, cur_max * (uint)sizeof(xfs_dfsbno_t)); return; } @@ -2210,8 +2210,9 @@ xfs_iroot_realloc( kmem_free(ifp->if_broot); ifp->if_broot = new_broot; ifp->if_broot_bytes = (int)new_size; - ASSERT(ifp->if_broot_bytes <= - XFS_IFORK_SIZE(ip, whichfork) + XFS_BROOT_SIZE_ADJ(ip)); + if (ifp->if_broot) + ASSERT(XFS_BMAP_BMDR_SPACE(ifp->if_broot) <= + XFS_IFORK_SIZE(ip, whichfork)); return; } @@ -2522,9 +2523,8 @@ xfs_iflush_fork( if ((iip->ili_fields & brootflag[whichfork]) && (ifp->if_broot_bytes > 0)) { ASSERT(ifp->if_broot != NULL); - ASSERT(ifp->if_broot_bytes <= - (XFS_IFORK_SIZE(ip, whichfork) + - XFS_BROOT_SIZE_ADJ(ip))); + ASSERT(XFS_BMAP_BMDR_SPACE(ifp->if_broot) <= + XFS_IFORK_SIZE(ip, whichfork)); xfs_bmbt_to_bmdr(mp, ifp->if_broot, ifp->if_broot_bytes, (xfs_bmdr_block_t *)cp, XFS_DFORK_SIZE(dip, mp, whichfork)); -- cgit v0.10.2 From 9cee4c5b7b706157554cb06ca949997240da725a Mon Sep 17 00:00:00 2001 From: Jie Liu Date: Mon, 1 Jul 2013 21:51:32 +0800 Subject: xfs: clean up unused codes at xfs_bulkstat() There are some unused codes at xfs_bulkstat(): - Variable bp is defined to point to the on-disk inode cluster buffer, but it proved to be of no practical help. - We process the chunks of good inodes which were fetched by iterating btree records from an AG. When processing inodes from each chunk, the code recomputing agbno if run into the first inode of a cluster, however, the agbno is not being used thereafter. This patch tries to clean up those things. Signed-off-by: Jie Liu Reviewed-by: Dave Chinner Signed-off-by: Ben Myers diff --git a/fs/xfs/xfs_itable.c b/fs/xfs/xfs_itable.c index bc92c53..b93e14b 100644 --- a/fs/xfs/xfs_itable.c +++ b/fs/xfs/xfs_itable.c @@ -221,7 +221,6 @@ xfs_bulkstat( char __user *ubufp; /* pointer into user's buffer */ int ubelem; /* spaces used in user's buffer */ int ubused; /* bytes used by formatter */ - xfs_buf_t *bp; /* ptr to on-disk inode cluster buf */ /* * Get the last inode value, see if there's nothing to do. @@ -263,7 +262,6 @@ xfs_bulkstat( rval = 0; while (XFS_BULKSTAT_UBLEFT(ubleft) && agno < mp->m_sb.sb_agcount) { cond_resched(); - bp = NULL; error = xfs_ialloc_read_agi(mp, NULL, agno, &agbp); if (error) { /* @@ -436,27 +434,7 @@ xfs_bulkstat( irbp->ir_freecount < XFS_INODES_PER_CHUNK; chunkidx++, clustidx++, agino++) { ASSERT(chunkidx < XFS_INODES_PER_CHUNK); - /* - * Recompute agbno if this is the - * first inode of the cluster. - * - * Careful with clustidx. There can be - * multiple clusters per chunk, a single - * cluster per chunk or a cluster that has - * inodes represented from several different - * chunks (if blocksize is large). - * - * Because of this, the starting clustidx is - * initialized to zero in this loop but must - * later be reset after reading in the cluster - * buffer. - */ - if ((chunkidx & (nicluster - 1)) == 0) { - agbno = XFS_AGINO_TO_AGBNO(mp, - irbp->ir_startino) + - ((chunkidx & nimask) >> - mp->m_sb.sb_inopblog); - } + ino = XFS_AGINO_TO_INO(mp, agno, agino); /* * Skip if this inode is free. @@ -502,10 +480,6 @@ xfs_bulkstat( cond_resched(); } - - if (bp) - xfs_buf_relse(bp); - /* * Set up for the next loop iteration. */ -- cgit v0.10.2 From 862a62937e76a91da218c1ee4dceb7b0700fed67 Mon Sep 17 00:00:00 2001 From: Yann Droneaud Date: Tue, 2 Jul 2013 18:39:34 +0200 Subject: xfs: use get_unused_fd_flags(0) instead of get_unused_fd() Macro get_unused_fd() is used to allocate a file descriptor with default flags. Those default flags (0) can be "unsafe": O_CLOEXEC must be used by default to not leak file descriptor across exec(). Instead of macro get_unused_fd(), functions anon_inode_getfd() or get_unused_fd_flags() should be used with flags given by userspace. If not possible, flags should be set to O_CLOEXEC to provide userspace with a default safe behavor. In a further patch, get_unused_fd() will be removed so that new code start using anon_inode_getfd() or get_unused_fd_flags() with correct flags. This patch replaces calls to get_unused_fd() with equivalent call to get_unused_fd_flags(0) to preserve current behavor for existing code. The hard coded flag value (0) should be reviewed on a per-subsystem basis, and, if possible, set to O_CLOEXEC. Signed-off-by: Yann Droneaud Reviewed-by: Ben Myers Signed-off-by: Ben Myers diff --git a/fs/xfs/xfs_ioctl.c b/fs/xfs/xfs_ioctl.c index 5e99968..dc5b659 100644 --- a/fs/xfs/xfs_ioctl.c +++ b/fs/xfs/xfs_ioctl.c @@ -248,7 +248,7 @@ xfs_open_by_handle( goto out_dput; } - fd = get_unused_fd(); + fd = get_unused_fd_flags(0); if (fd < 0) { error = fd; goto out_dput; -- cgit v0.10.2 From 3e5b7d8b491c3710b7e007eab0a643f923932e3d Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Wed, 10 Jul 2013 07:03:59 +1000 Subject: xfs: update mount options documentation Because it's horribly out of date. And mark various deprecated options as deprecated and give them a removal date. Signed-off-by: Dave Chinner Reviewed-by: Mark Tinguely Signed-off-by: Ben Myers diff --git a/Documentation/filesystems/xfs.txt b/Documentation/filesystems/xfs.txt index 83577f0..12525b1 100644 --- a/Documentation/filesystems/xfs.txt +++ b/Documentation/filesystems/xfs.txt @@ -18,6 +18,8 @@ Mount Options ============= When mounting an XFS filesystem, the following options are accepted. +For boolean mount options, the names with the (*) suffix is the +default behaviour. allocsize=size Sets the buffered I/O end-of-file preallocation size when @@ -25,97 +27,128 @@ When mounting an XFS filesystem, the following options are accepted. Valid values for this option are page size (typically 4KiB) through to 1GiB, inclusive, in power-of-2 increments. - attr2/noattr2 - The options enable/disable (default is disabled for backward - compatibility on-disk) an "opportunistic" improvement to be - made in the way inline extended attributes are stored on-disk. - When the new form is used for the first time (by setting or - removing extended attributes) the on-disk superblock feature - bit field will be updated to reflect this format being in use. + The default behaviour is for dynamic end-of-file + preallocation size, which uses a set of heuristics to + optimise the preallocation size based on the current + allocation patterns within the file and the access patterns + to the file. Specifying a fixed allocsize value turns off + the dynamic behaviour. + + attr2 + noattr2 + The options enable/disable an "opportunistic" improvement to + be made in the way inline extended attributes are stored + on-disk. When the new form is used for the first time when + attr2 is selected (either when setting or removing extended + attributes) the on-disk superblock feature bit field will be + updated to reflect this format being in use. + + The default behaviour is determined by the on-disk feature + bit indicating that attr2 behaviour is active. If either + mount option it set, then that becomes the new default used + by the filesystem. CRC enabled filesystems always use the attr2 format, and so will reject the noattr2 mount option if it is set. - barrier - Enables the use of block layer write barriers for writes into - the journal and unwritten extent conversion. This allows for - drive level write caching to be enabled, for devices that - support write barriers. + barrier (*) + nobarrier + Enables/disables the use of block layer write barriers for + writes into the journal and for data integrity operations. + This allows for drive level write caching to be enabled, for + devices that support write barriers. discard - Issue command to let the block device reclaim space freed by the - filesystem. This is useful for SSD devices, thinly provisioned - LUNs and virtual machine images, but may have a performance - impact. - - dmapi - Enable the DMAPI (Data Management API) event callouts. - Use with the "mtpt" option. - - grpid/bsdgroups and nogrpid/sysvgroups - These options define what group ID a newly created file gets. - When grpid is set, it takes the group ID of the directory in - which it is created; otherwise (the default) it takes the fsgid - of the current process, unless the directory has the setgid bit - set, in which case it takes the gid from the parent directory, - and also gets the setgid bit set if it is a directory itself. - - ihashsize=value - In memory inode hashes have been removed, so this option has - no function as of August 2007. Option is deprecated. - - ikeep/noikeep - When ikeep is specified, XFS does not delete empty inode clusters - and keeps them around on disk. ikeep is the traditional XFS - behaviour. When noikeep is specified, empty inode clusters - are returned to the free space pool. The default is noikeep for - non-DMAPI mounts, while ikeep is the default when DMAPI is in use. - - inode64 - Indicates that XFS is allowed to create inodes at any location - in the filesystem, including those which will result in inode - numbers occupying more than 32 bits of significance. This is - the default allocation option. Applications which do not handle - inode numbers bigger than 32 bits, should use inode32 option. + nodiscard (*) + Enable/disable the issuing of commands to let the block + device reclaim space freed by the filesystem. This is + useful for SSD devices, thinly provisioned LUNs and virtual + machine images, but may have a performance impact. + + Note: It is currently recommended that you use the fstrim + application to discard unused blocks rather than the discard + mount option because the performance impact of this option + is quite severe. + + grpid/bsdgroups + nogrpid/sysvgroups (*) + These options define what group ID a newly created file + gets. When grpid is set, it takes the group ID of the + directory in which it is created; otherwise it takes the + fsgid of the current process, unless the directory has the + setgid bit set, in which case it takes the gid from the + parent directory, and also gets the setgid bit set if it is + a directory itself. + + filestreams + Make the data allocator use the filestreams allocation mode + across the entire filesystem rather than just on directories + configured to use it. + + ikeep + noikeep (*) + When ikeep is specified, XFS does not delete empty inode + clusters and keeps them around on disk. When noikeep is + specified, empty inode clusters are returned to the free + space pool. inode32 - Indicates that XFS is limited to create inodes at locations which - will not result in inode numbers with more than 32 bits of - significance. This is provided for backwards compatibility, since - 64 bits inode numbers might cause problems for some applications - that cannot handle large inode numbers. - - largeio/nolargeio + inode64 (*) + When inode32 is specified, it indicates that XFS limits + inode creation to locations which will not result in inode + numbers with more than 32 bits of significance. + + When inode64 is specified, it indicates that XFS is allowed + to create inodes at any location in the filesystem, + including those which will result in inode numbers occupying + more than 32 bits of significance. + + inode32 is provided for backwards compatibility with older + systems and applications, since 64 bits inode numbers might + cause problems for some applications that cannot handle + large inode numbers. If applications are in use which do + not handle inode numbers bigger than 32 bits, the inode32 + option should be specified. + + + largeio + nolargeio (*) If "nolargeio" is specified, the optimal I/O reported in - st_blksize by stat(2) will be as small as possible to allow user - applications to avoid inefficient read/modify/write I/O. - If "largeio" specified, a filesystem that has a "swidth" specified - will return the "swidth" value (in bytes) in st_blksize. If the - filesystem does not have a "swidth" specified but does specify - an "allocsize" then "allocsize" (in bytes) will be returned - instead. - If neither of these two options are specified, then filesystem - will behave as if "nolargeio" was specified. + st_blksize by stat(2) will be as small as possible to allow + user applications to avoid inefficient read/modify/write + I/O. This is typically the page size of the machine, as + this is the granularity of the page cache. + + If "largeio" specified, a filesystem that was created with a + "swidth" specified will return the "swidth" value (in bytes) + in st_blksize. If the filesystem does not have a "swidth" + specified but does specify an "allocsize" then "allocsize" + (in bytes) will be returned instead. Otherwise the behaviour + is the same as if "nolargeio" was specified. logbufs=value - Set the number of in-memory log buffers. Valid numbers range - from 2-8 inclusive. - The default value is 8 buffers for filesystems with a - blocksize of 64KiB, 4 buffers for filesystems with a blocksize - of 32KiB, 3 buffers for filesystems with a blocksize of 16KiB - and 2 buffers for all other configurations. Increasing the - number of buffers may increase performance on some workloads - at the cost of the memory used for the additional log buffers - and their associated control structures. + Set the number of in-memory log buffers. Valid numbers + range from 2-8 inclusive. + + The default value is 8 buffers. + + If the memory cost of 8 log buffers is too high on small + systems, then it may be reduced at some cost to performance + on metadata intensive workloads. The logbsize option below + controls the size of each buffer and so is also relevent to + this case. logbsize=value - Set the size of each in-memory log buffer. - Size may be specified in bytes, or in kilobytes with a "k" suffix. - Valid sizes for version 1 and version 2 logs are 16384 (16k) and - 32768 (32k). Valid sizes for version 2 logs also include - 65536 (64k), 131072 (128k) and 262144 (256k). - The default value for machines with more than 32MiB of memory - is 32768, machines with less memory use 16384 by default. + Set the size of each in-memory log buffer. The size may be + specified in bytes, or in kilobytes with a "k" suffix. + Valid sizes for version 1 and version 2 logs are 16384 (16k) + and 32768 (32k). Valid sizes for version 2 logs also + include 65536 (64k), 131072 (128k) and 262144 (256k). The + logbsize must be an integer multiple of the log + stripe unit configured at mkfs time. + + The default value for for version 1 logs is 32768, while the + default value for version 2 logs is MAX(32768, log_sunit). logdev=device and rtdev=device Use an external log (metadata journal) and/or real-time device. @@ -124,16 +157,11 @@ When mounting an XFS filesystem, the following options are accepted. optional, and the log section can be separate from the data section or contained within it. - mtpt=mountpoint - Use with the "dmapi" option. The value specified here will be - included in the DMAPI mount event, and should be the path of - the actual mountpoint that is used. - noalign - Data allocations will not be aligned at stripe unit boundaries. - - noatime - Access timestamps are not updated when a file is read. + Data allocations will not be aligned at stripe unit + boundaries. This is only relevant to filesystems created + with non-zero data alignment parameters (sunit, swidth) by + mkfs. norecovery The filesystem will be mounted without running log recovery. @@ -144,8 +172,14 @@ When mounting an XFS filesystem, the following options are accepted. the mount will fail. nouuid - Don't check for double mounted file systems using the file system uuid. - This is useful to mount LVM snapshot volumes. + Don't check for double mounted file systems using the file + system uuid. This is useful to mount LVM snapshot volumes, + and often used in combination with "norecovery" for mounting + read-only snapshots. + + noquota + Forcibly turns off all quota accounting and enforcement + within the filesystem. uquota/usrquota/uqnoenforce/quota User disk quota accounting enabled, and limits (optionally) @@ -160,24 +194,64 @@ When mounting an XFS filesystem, the following options are accepted. enforced. Refer to xfs_quota(8) for further details. sunit=value and swidth=value - Used to specify the stripe unit and width for a RAID device or - a stripe volume. "value" must be specified in 512-byte block - units. - If this option is not specified and the filesystem was made on - a stripe volume or the stripe width or unit were specified for - the RAID device at mkfs time, then the mount system call will - restore the value from the superblock. For filesystems that - are made directly on RAID devices, these options can be used - to override the information in the superblock if the underlying - disk layout changes after the filesystem has been created. - The "swidth" option is required if the "sunit" option has been - specified, and must be a multiple of the "sunit" value. + Used to specify the stripe unit and width for a RAID device + or a stripe volume. "value" must be specified in 512-byte + block units. These options are only relevant to filesystems + that were created with non-zero data alignment parameters. + + The sunit and swidth parameters specified must be compatible + with the existing filesystem alignment characteristics. In + general, that means the only valid changes to sunit are + increasing it by a power-of-2 multiple. Valid swidth values + are any integer multiple of a valid sunit value. + + Typically the only time these mount options are necessary if + after an underlying RAID device has had it's geometry + modified, such as adding a new disk to a RAID5 lun and + reshaping it. swalloc Data allocations will be rounded up to stripe width boundaries when the current end of file is being extended and the file size is larger than the stripe width size. + wsync + When specified, all filesystem namespace operations are + executed synchronously. This ensures that when the namespace + operation (create, unlink, etc) completes, the change to the + namespace is on stable storage. This is useful in HA setups + where failover must not result in clients seeing + inconsistent namespace presentation during or after a + failover event. + + +Deprecated Mount Options +======================== + + delaylog/nodelaylog + Delayed logging is the only logging method that XFS supports + now, so these mount options are now ignored. + + Due for removal in 3.12. + + ihashsize=value + In memory inode hashes have been removed, so this option has + no function as of August 2007. Option is deprecated. + + Due for removal in 3.12. + + irixsgid + This behaviour is now controlled by a sysctl, so the mount + option is ignored. + + Due for removal in 3.12. + + osyncisdsync + osyncisosync + O_SYNC and O_DSYNC are fully supported, so there is no need + for these options any more. + + Due for removal in 3.12. sysctls ======= @@ -189,15 +263,20 @@ The following sysctls are available for the XFS filesystem: in /proc/fs/xfs/stat. It then immediately resets to "0". fs.xfs.xfssyncd_centisecs (Min: 100 Default: 3000 Max: 720000) - The interval at which the xfssyncd thread flushes metadata - out to disk. This thread will flush log activity out, and - do some processing on unlinked inodes. + The interval at which the filesystem flushes metadata + out to disk and runs internal cache cleanup routines. - fs.xfs.xfsbufd_centisecs (Min: 50 Default: 100 Max: 3000) - The interval at which xfsbufd scans the dirty metadata buffers list. + fs.xfs.filestream_centisecs (Min: 1 Default: 3000 Max: 360000) + The interval at which the filesystem ages filestreams cache + references and returns timed-out AGs back to the free stream + pool. - fs.xfs.age_buffer_centisecs (Min: 100 Default: 1500 Max: 720000) - The age at which xfsbufd flushes dirty metadata buffers to disk. + fs.xfs.speculative_prealloc_lifetime + (Units: seconds Min: 1 Default: 300 Max: 86400) + The interval at which the background scanning for inodes + with unused speculative preallocation runs. The scan + removes unused preallocation from clean inodes and releases + the unused space back to the free pool. fs.xfs.error_level (Min: 0 Default: 3 Max: 11) A volume knob for error reporting when internal errors occur. @@ -254,9 +333,31 @@ The following sysctls are available for the XFS filesystem: by the xfs_io(8) chattr command on a directory to be inherited by files in that directory. + fs.xfs.inherit_nodefrag (Min: 0 Default: 1 Max: 1) + Setting this to "1" will cause the "nodefrag" flag set + by the xfs_io(8) chattr command on a directory to be + inherited by files in that directory. + fs.xfs.rotorstep (Min: 1 Default: 1 Max: 256) In "inode32" allocation mode, this option determines how many files the allocator attempts to allocate in the same allocation group before moving to the next allocation group. The intent is to control the rate at which the allocator moves between allocation groups when allocating extents for new files. + +Deprecated Sysctls +================== + + fs.xfs.xfsbufd_centisecs (Min: 50 Default: 100 Max: 3000) + Dirty metadata is now tracked by the log subsystem and + flushing is driven by log space and idling demands. The + xfsbufd no longer exists, so this syctl does nothing. + + Due for removal in 3.14. + + fs.xfs.age_buffer_centisecs (Min: 100 Default: 1500 Max: 720000) + Dirty metadata is now tracked by the log subsystem and + flushing is driven by log space and idling demands. The + xfsbufd no longer exists, so this syctl does nothing. + + Due for removal in 3.14. -- cgit v0.10.2 From f3508bcddf8fae6ebd21000d708cf09e7e77a963 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Wed, 10 Jul 2013 07:04:00 +1000 Subject: xfs: remove local fork format handling from xfs_bmapi_write() The conversion from local format to extent format requires interpretation of the data in the fork being converted, so it cannot be done in a generic way. It is up to the caller to convert the fork format to extent format before calling into xfs_bmapi_write() so format conversion can be done correctly. The code in xfs_bmapi_write() to convert the format is used implicitly by the attribute and directory code, but they specifically zero the fork size so that the conversion does not do any allocation or manipulation. Move this conversion into the shortform to leaf functions for the dir/attr code so the conversions are explicitly controlled by all callers. Now we can remove the conversion code in xfs_bmapi_write. Signed-off-by: Dave Chinner Reviewed-by: Mark Tinguely Signed-off-by: Ben Myers diff --git a/fs/xfs/xfs_attr_leaf.c b/fs/xfs/xfs_attr_leaf.c index 31d3cd1..b800fbc 100644 --- a/fs/xfs/xfs_attr_leaf.c +++ b/fs/xfs/xfs_attr_leaf.c @@ -690,6 +690,8 @@ xfs_attr_shortform_to_leaf(xfs_da_args_t *args) sf = (xfs_attr_shortform_t *)tmpbuffer; xfs_idata_realloc(dp, -size, XFS_ATTR_FORK); + xfs_bmap_local_to_extents_empty(dp, XFS_ATTR_FORK); + bp = NULL; error = xfs_da_grow_inode(args, &blkno); if (error) { diff --git a/fs/xfs/xfs_bmap.c b/fs/xfs/xfs_bmap.c index 8904284..05c698c 100644 --- a/fs/xfs/xfs_bmap.c +++ b/fs/xfs/xfs_bmap.c @@ -1161,6 +1161,24 @@ xfs_bmap_extents_to_btree( * since the file data needs to get logged so things will stay consistent. * (The bmap-level manipulations are ok, though). */ +void +xfs_bmap_local_to_extents_empty( + struct xfs_inode *ip, + int whichfork) +{ + struct xfs_ifork *ifp = XFS_IFORK_PTR(ip, whichfork); + + ASSERT(XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_LOCAL); + ASSERT(ifp->if_bytes == 0); + ASSERT(XFS_IFORK_NEXTENTS(ip, whichfork) == 0); + + xfs_bmap_forkoff_reset(ip->i_mount, ip, whichfork); + ifp->if_flags &= ~XFS_IFINLINE; + ifp->if_flags |= XFS_IFEXTENTS; + XFS_IFORK_FMT_SET(ip, whichfork, XFS_DINODE_FMT_EXTENTS); +} + + STATIC int /* error */ xfs_bmap_local_to_extents( xfs_trans_t *tp, /* transaction pointer */ @@ -1174,9 +1192,12 @@ xfs_bmap_local_to_extents( struct xfs_inode *ip, struct xfs_ifork *ifp)) { - int error; /* error return value */ + int error = 0; int flags; /* logging flags returned */ xfs_ifork_t *ifp; /* inode fork pointer */ + xfs_alloc_arg_t args; /* allocation arguments */ + xfs_buf_t *bp; /* buffer for extent block */ + xfs_bmbt_rec_host_t *ep; /* extent record pointer */ /* * We don't want to deal with the case of keeping inode data inline yet. @@ -1185,68 +1206,65 @@ xfs_bmap_local_to_extents( ASSERT(!(S_ISREG(ip->i_d.di_mode) && whichfork == XFS_DATA_FORK)); ifp = XFS_IFORK_PTR(ip, whichfork); ASSERT(XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_LOCAL); + + if (!ifp->if_bytes) { + xfs_bmap_local_to_extents_empty(ip, whichfork); + flags = XFS_ILOG_CORE; + goto done; + } + flags = 0; error = 0; - if (ifp->if_bytes) { - xfs_alloc_arg_t args; /* allocation arguments */ - xfs_buf_t *bp; /* buffer for extent block */ - xfs_bmbt_rec_host_t *ep;/* extent record pointer */ - - ASSERT((ifp->if_flags & - (XFS_IFINLINE|XFS_IFEXTENTS|XFS_IFEXTIREC)) == XFS_IFINLINE); - memset(&args, 0, sizeof(args)); - args.tp = tp; - args.mp = ip->i_mount; - args.firstblock = *firstblock; - /* - * Allocate a block. We know we need only one, since the - * file currently fits in an inode. - */ - if (*firstblock == NULLFSBLOCK) { - args.fsbno = XFS_INO_TO_FSB(args.mp, ip->i_ino); - args.type = XFS_ALLOCTYPE_START_BNO; - } else { - args.fsbno = *firstblock; - args.type = XFS_ALLOCTYPE_NEAR_BNO; - } - args.total = total; - args.minlen = args.maxlen = args.prod = 1; - error = xfs_alloc_vextent(&args); - if (error) - goto done; - - /* Can't fail, the space was reserved. */ - ASSERT(args.fsbno != NULLFSBLOCK); - ASSERT(args.len == 1); - *firstblock = args.fsbno; - bp = xfs_btree_get_bufl(args.mp, tp, args.fsbno, 0); - - /* initialise the block and copy the data */ - init_fn(tp, bp, ip, ifp); - - /* account for the change in fork size and log everything */ - xfs_trans_log_buf(tp, bp, 0, ifp->if_bytes - 1); - xfs_bmap_forkoff_reset(args.mp, ip, whichfork); - xfs_idata_realloc(ip, -ifp->if_bytes, whichfork); - xfs_iext_add(ifp, 0, 1); - ep = xfs_iext_get_ext(ifp, 0); - xfs_bmbt_set_allf(ep, 0, args.fsbno, 1, XFS_EXT_NORM); - trace_xfs_bmap_post_update(ip, 0, - whichfork == XFS_ATTR_FORK ? BMAP_ATTRFORK : 0, - _THIS_IP_); - XFS_IFORK_NEXT_SET(ip, whichfork, 1); - ip->i_d.di_nblocks = 1; - xfs_trans_mod_dquot_byino(tp, ip, - XFS_TRANS_DQ_BCOUNT, 1L); - flags |= xfs_ilog_fext(whichfork); + ASSERT((ifp->if_flags & (XFS_IFINLINE|XFS_IFEXTENTS|XFS_IFEXTIREC)) == + XFS_IFINLINE); + memset(&args, 0, sizeof(args)); + args.tp = tp; + args.mp = ip->i_mount; + args.firstblock = *firstblock; + /* + * Allocate a block. We know we need only one, since the + * file currently fits in an inode. + */ + if (*firstblock == NULLFSBLOCK) { + args.fsbno = XFS_INO_TO_FSB(args.mp, ip->i_ino); + args.type = XFS_ALLOCTYPE_START_BNO; } else { - ASSERT(XFS_IFORK_NEXTENTS(ip, whichfork) == 0); - xfs_bmap_forkoff_reset(ip->i_mount, ip, whichfork); + args.fsbno = *firstblock; + args.type = XFS_ALLOCTYPE_NEAR_BNO; } - ifp->if_flags &= ~XFS_IFINLINE; - ifp->if_flags |= XFS_IFEXTENTS; - XFS_IFORK_FMT_SET(ip, whichfork, XFS_DINODE_FMT_EXTENTS); + args.total = total; + args.minlen = args.maxlen = args.prod = 1; + error = xfs_alloc_vextent(&args); + if (error) + goto done; + + /* Can't fail, the space was reserved. */ + ASSERT(args.fsbno != NULLFSBLOCK); + ASSERT(args.len == 1); + *firstblock = args.fsbno; + bp = xfs_btree_get_bufl(args.mp, tp, args.fsbno, 0); + + /* initialise the block and copy the data */ + init_fn(tp, bp, ip, ifp); + + /* account for the change in fork size and log everything */ + xfs_trans_log_buf(tp, bp, 0, ifp->if_bytes - 1); + xfs_idata_realloc(ip, -ifp->if_bytes, whichfork); + xfs_bmap_local_to_extents_empty(ip, whichfork); flags |= XFS_ILOG_CORE; + + xfs_iext_add(ifp, 0, 1); + ep = xfs_iext_get_ext(ifp, 0); + xfs_bmbt_set_allf(ep, 0, args.fsbno, 1, XFS_EXT_NORM); + trace_xfs_bmap_post_update(ip, 0, + whichfork == XFS_ATTR_FORK ? BMAP_ATTRFORK : 0, + _THIS_IP_); + XFS_IFORK_NEXT_SET(ip, whichfork, 1); + ip->i_d.di_nblocks = 1; + xfs_trans_mod_dquot_byino(tp, ip, + XFS_TRANS_DQ_BCOUNT, 1L); + flags |= xfs_ilog_fext(whichfork); + done: *logflagsp = flags; return error; @@ -1323,25 +1341,6 @@ xfs_bmap_add_attrfork_extents( } /* - * Block initialisation function for local to extent format conversion. - * - * This shouldn't actually be called by anyone, so make sure debug kernels cause - * a noticable failure. - */ -STATIC void -xfs_bmap_local_to_extents_init_fn( - struct xfs_trans *tp, - struct xfs_buf *bp, - struct xfs_inode *ip, - struct xfs_ifork *ifp) -{ - ASSERT(0); - bp->b_ops = &xfs_bmbt_buf_ops; - memcpy(bp->b_addr, ifp->if_u1.if_data, ifp->if_bytes); - xfs_trans_buf_set_type(tp, bp, XFS_BLFT_BTREE_BUF); -} - -/* * Called from xfs_bmap_add_attrfork to handle local format files. Each * different data fork content type needs a different callout to do the * conversion. Some are basic and only require special block initialisation @@ -1381,9 +1380,9 @@ xfs_bmap_add_attrfork_local( flags, XFS_DATA_FORK, xfs_symlink_local_to_remote); - return xfs_bmap_local_to_extents(tp, ip, firstblock, 1, flags, - XFS_DATA_FORK, - xfs_bmap_local_to_extents_init_fn); + /* should only be called for types that support local format data */ + ASSERT(0); + return EFSCORRUPTED; } /* @@ -4907,20 +4906,19 @@ xfs_bmapi_write( orig_mval = mval; orig_nmap = *nmap; #endif + whichfork = (flags & XFS_BMAPI_ATTRFORK) ? + XFS_ATTR_FORK : XFS_DATA_FORK; ASSERT(*nmap >= 1); ASSERT(*nmap <= XFS_BMAP_MAX_NMAP); ASSERT(!(flags & XFS_BMAPI_IGSTATE)); ASSERT(tp != NULL); ASSERT(len > 0); - - whichfork = (flags & XFS_BMAPI_ATTRFORK) ? - XFS_ATTR_FORK : XFS_DATA_FORK; + ASSERT(XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_LOCAL); if (unlikely(XFS_TEST_ERROR( (XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS && - XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE && - XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_LOCAL), + XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE), mp, XFS_ERRTAG_BMAPIFORMAT, XFS_RANDOM_BMAPIFORMAT))) { XFS_ERROR_REPORT("xfs_bmapi_write", XFS_ERRLEVEL_LOW, mp); return XFS_ERROR(EFSCORRUPTED); @@ -4933,37 +4931,6 @@ xfs_bmapi_write( XFS_STATS_INC(xs_blk_mapw); - if (XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_LOCAL) { - /* - * XXX (dgc): This assumes we are only called for inodes that - * contain content neutral data in local format. Anything that - * contains caller-specific data in local format that needs - * transformation to move to a block format needs to do the - * conversion to extent format itself. - * - * Directory data forks and attribute forks handle this - * themselves, but with the addition of metadata verifiers every - * data fork in local format now contains caller specific data - * and as such conversion through this function is likely to be - * broken. - * - * The only likely user of this branch is for remote symlinks, - * but we cannot overwrite the data fork contents of the symlink - * (EEXIST occurs higher up the stack) and so it will never go - * from local format to extent format here. Hence I don't think - * this branch is ever executed intentionally and we should - * consider removing it and asserting that xfs_bmapi_write() - * cannot be called directly on local format forks. i.e. callers - * are completely responsible for local to extent format - * conversion, not xfs_bmapi_write(). - */ - error = xfs_bmap_local_to_extents(tp, ip, firstblock, total, - &bma.logflags, whichfork, - xfs_bmap_local_to_extents_init_fn); - if (error) - goto error0; - } - if (*firstblock == NULLFSBLOCK) { if (XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_BTREE) bma.minleft = be16_to_cpu(ifp->if_broot->bb_level) + 1; diff --git a/fs/xfs/xfs_bmap.h b/fs/xfs/xfs_bmap.h index 5f469c3..1cf1292 100644 --- a/fs/xfs/xfs_bmap.h +++ b/fs/xfs/xfs_bmap.h @@ -172,6 +172,7 @@ void xfs_bmap_trace_exlist(struct xfs_inode *ip, xfs_extnum_t cnt, #endif int xfs_bmap_add_attrfork(struct xfs_inode *ip, int size, int rsvd); +void xfs_bmap_local_to_extents_empty(struct xfs_inode *ip, int whichfork); void xfs_bmap_add_free(xfs_fsblock_t bno, xfs_filblks_t len, struct xfs_bmap_free *flist, struct xfs_mount *mp); void xfs_bmap_cancel(struct xfs_bmap_free *flist); diff --git a/fs/xfs/xfs_dir2_block.c b/fs/xfs/xfs_dir2_block.c index e59f5fc..53b9aa2 100644 --- a/fs/xfs/xfs_dir2_block.c +++ b/fs/xfs/xfs_dir2_block.c @@ -29,6 +29,7 @@ #include "xfs_dinode.h" #include "xfs_inode.h" #include "xfs_inode_item.h" +#include "xfs_bmap.h" #include "xfs_buf_item.h" #include "xfs_dir2.h" #include "xfs_dir2_format.h" @@ -1167,13 +1168,15 @@ xfs_dir2_sf_to_block( __be16 *tagp; /* end of data entry */ xfs_trans_t *tp; /* transaction pointer */ struct xfs_name name; + struct xfs_ifork *ifp; trace_xfs_dir2_sf_to_block(args); dp = args->dp; tp = args->trans; mp = dp->i_mount; - ASSERT(dp->i_df.if_flags & XFS_IFINLINE); + ifp = XFS_IFORK_PTR(dp, XFS_DATA_FORK); + ASSERT(ifp->if_flags & XFS_IFINLINE); /* * Bomb out if the shortform directory is way too short. */ @@ -1182,22 +1185,23 @@ xfs_dir2_sf_to_block( return XFS_ERROR(EIO); } - oldsfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data; + oldsfp = (xfs_dir2_sf_hdr_t *)ifp->if_u1.if_data; - ASSERT(dp->i_df.if_bytes == dp->i_d.di_size); - ASSERT(dp->i_df.if_u1.if_data != NULL); + ASSERT(ifp->if_bytes == dp->i_d.di_size); + ASSERT(ifp->if_u1.if_data != NULL); ASSERT(dp->i_d.di_size >= xfs_dir2_sf_hdr_size(oldsfp->i8count)); + ASSERT(dp->i_d.di_nextents == 0); /* * Copy the directory into a temporary buffer. * Then pitch the incore inode data so we can make extents. */ - sfp = kmem_alloc(dp->i_df.if_bytes, KM_SLEEP); - memcpy(sfp, oldsfp, dp->i_df.if_bytes); + sfp = kmem_alloc(ifp->if_bytes, KM_SLEEP); + memcpy(sfp, oldsfp, ifp->if_bytes); - xfs_idata_realloc(dp, -dp->i_df.if_bytes, XFS_DATA_FORK); + xfs_idata_realloc(dp, -ifp->if_bytes, XFS_DATA_FORK); + xfs_bmap_local_to_extents_empty(dp, XFS_DATA_FORK); dp->i_d.di_size = 0; - xfs_trans_log_inode(tp, dp, XFS_ILOG_CORE); /* * Add block 0 to the inode. -- cgit v0.10.2 From b0a9dab78aee2a479d7c226e6939d553967e4024 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Wed, 10 Jul 2013 07:04:01 +1000 Subject: xfs: dquot log reservations are too small During review of the separate project quota inode patches, it became obvious that the dquot log reservation calculation underestimated the number dquots that can be modified in a transaction. This has it's roots way back in the Irix quota implementation. That is, when quotas were first implemented in XFS, it only supported user and project quotas as Irix did not have group quotas. Hence the worst case operation involving dquot modification was calculated to involve 2 user dquots and 1 project dquot or 1 user dequot and 2 project dquots. i.e. 3 dquots. This was determined back in 1996, and has remained unchanged ever since. However, back in 2001, the Linux XFS port dropped all support for project quota and implmented group quotas over the top. This was effectively done with a search-and-replace of project with group, and as such the log reservation was not changed. However, with the advent of group quotas, chmod and rename now could modify more than 3 dquots in a single transaction - both could modify 4 dquots. Hence this log reservation has been wrong for a long time. In 2005, project quota support was reintroduced into Linux, but it was implemented to be mutually exclusive to group quotas and so this didn't add any new changes to the dquot log reservation. Hence when project quotas were in use (rather than group quotas) the log reservation was again valid, just like in the Irix days. Now, with the addition of the separate project quota inode, group and project quotas are no longer mutually exclusive, and hence operations can now modify three dquots per inode where previously it was only two. The worst case here is the rename transaction, which can allocate/free space on two different directory inodes, and if they have different uid/gid/prid configurations and are world writeable, then rename can actually modify 6 different dquots now. Further, the dquot log reservation doesn't take into account the space used by the dquot log format structure that precedes the dquot that is logged, and hence further underestimates the worst case log space required by dquots during a transaction. This has been missing since the first commit in 1996. Hence the worst case log reservation needs to be increased from 3 to 6, and it needs to take into account a log format header for each of those dquots. Signed-off-by: Dave Chinner Reviewed-by: Mark Tinguely Signed-off-by: Ben Myers diff --git a/fs/xfs/xfs_quota.h b/fs/xfs/xfs_quota.h index c3483ba..40d508b 100644 --- a/fs/xfs/xfs_quota.h +++ b/fs/xfs/xfs_quota.h @@ -108,11 +108,28 @@ typedef struct xfs_dqblk { { XFS_DQ_FREEING, "FREEING" } /* - * In the worst case, when both user and group quotas are on, - * we can have a max of three dquots changing in a single transaction. + * We have the possibility of all three quota types being active at once, and + * hence free space modification requires modification of all three current + * dquots in a single transaction. For this case we need to have a reservation + * of at least 3 dquots. + * + * However, a chmod operation can change both UID and GID in a single + * transaction, resulting in requiring {old, new} x {uid, gid} dquots to be + * modified. Hence for this case we need to reserve space for at least 4 dquots. + * + * And in the worst case, there's a rename operation that can be modifying up to + * 4 inodes with dquots attached to them. In reality, the only inodes that can + * have their dquots modified are the source and destination directory inodes + * due to directory name creation and removal. That can require space allocation + * and/or freeing on both directory inodes, and hence all three dquots on each + * inode can be modified. And if the directories are world writeable, all the + * dquots can be unique and so 6 dquots can be modified.... + * + * And, of course, we also need to take into account the dquot log format item + * used to describe each dquot. */ -#define XFS_DQUOT_LOGRES(mp) (sizeof(xfs_disk_dquot_t) * 3) - +#define XFS_DQUOT_LOGRES(mp) \ + ((sizeof(struct xfs_dq_logformat) + sizeof(struct xfs_disk_dquot)) * 6) /* * These are the structures used to lay out dquots and quotaoff diff --git a/fs/xfs/xfs_trans_dquot.c b/fs/xfs/xfs_trans_dquot.c index 3ba64d5..db041a5 100644 --- a/fs/xfs/xfs_trans_dquot.c +++ b/fs/xfs/xfs_trans_dquot.c @@ -291,11 +291,10 @@ xfs_trans_mod_dquot( /* - * Given an array of dqtrx structures, lock all the dquots associated - * and join them to the transaction, provided they have been modified. - * We know that the highest number of dquots (of one type - usr OR grp), - * involved in a transaction is 2 and that both usr and grp combined - 3. - * So, we don't attempt to make this very generic. + * Given an array of dqtrx structures, lock all the dquots associated and join + * them to the transaction, provided they have been modified. We know that the + * highest number of dquots of one type - usr, grp OR prj - involved in a + * transaction is 2 so we don't need to make this very generic. */ STATIC void xfs_trans_dqlockedjoin( -- cgit v0.10.2 From 42c49d7f249c2487f36d3314753d5d8ebcee8249 Mon Sep 17 00:00:00 2001 From: Carlos Maiolino Date: Fri, 21 Jun 2013 14:45:53 -0300 Subject: xfs: fix sgid inheritance for subdirectories inheriting default acls [V3] XFS removes sgid bits of subdirectories under a directory containing a default acl. When a default acl is set, it implies xfs to call xfs_setattr_nonsize() in its code path. Such function is shared among mkdir and chmod system calls, and does some checks unneeded by mkdir (calling inode_change_ok()). Such checks remove sgid bit from the inode after it has been granted. With this patch, we extend the meaning of XFS_ATTR_NOACL flag to avoid these checks when acls are being inherited (thanks hch). Also, xfs_setattr_mode, doesn't need to re-check for group id and capabilities permissions, this only implies in another try to remove sgid bit from the directories. Such check is already done either on inode_change_ok() or xfs_setattr_nonsize(). Changelog: V2: Extends the meaning of XFS_ATTR_NOACL instead of wrap the tests into another function V3: Remove S_ISDIR check in xfs_setattr_nonsize() from the patch Signed-off-by: Carlos Maiolino Reviewed-by: Ben Myers Signed-off-by: Ben Myers diff --git a/fs/xfs/xfs_iops.c b/fs/xfs/xfs_iops.c index c69bbc4..8865261 100644 --- a/fs/xfs/xfs_iops.c +++ b/fs/xfs/xfs_iops.c @@ -467,9 +467,6 @@ xfs_setattr_mode( ASSERT(tp); ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); - if (!in_group_p(inode->i_gid) && !capable(CAP_FSETID)) - mode &= ~S_ISGID; - ip->i_d.di_mode &= S_IFMT; ip->i_d.di_mode |= mode & ~S_IFMT; @@ -495,15 +492,18 @@ xfs_setattr_nonsize( trace_xfs_setattr(ip); - if (mp->m_flags & XFS_MOUNT_RDONLY) - return XFS_ERROR(EROFS); + /* If acls are being inherited, we already have this checked */ + if (!(flags & XFS_ATTR_NOACL)) { + if (mp->m_flags & XFS_MOUNT_RDONLY) + return XFS_ERROR(EROFS); - if (XFS_FORCED_SHUTDOWN(mp)) - return XFS_ERROR(EIO); + if (XFS_FORCED_SHUTDOWN(mp)) + return XFS_ERROR(EIO); - error = -inode_change_ok(inode, iattr); - if (error) - return XFS_ERROR(error); + error = -inode_change_ok(inode, iattr); + if (error) + return XFS_ERROR(error); + } ASSERT((mask & ATTR_SIZE) == 0); -- cgit v0.10.2 From 92f8ff73f18672b03ec8b92197cdddf2b5de7ea0 Mon Sep 17 00:00:00 2001 From: Chandra Seetharaman Date: Thu, 11 Jul 2013 00:00:40 -0500 Subject: xfs: Add pquota fields where gquota is used. Add project quota changes to all the places where group quota field is used: * add separate project quota members into various structures * split project quota and group quotas so that instead of overriding the group quota members incore, the new project quota members are used instead * get rid of usage of the OQUOTA flag incore, in favor of separate group and project quota flags. * add a project dquot argument to various functions. Not using the pquotino field from superblock yet. Signed-off-by: Chandra Seetharaman Reviewed-by: Ben Myers Signed-off-by: Ben Myers diff --git a/fs/xfs/xfs_dquot.c b/fs/xfs/xfs_dquot.c index f01012d..0adf27e 100644 --- a/fs/xfs/xfs_dquot.c +++ b/fs/xfs/xfs_dquot.c @@ -936,6 +936,7 @@ xfs_qm_dqput_final( { struct xfs_quotainfo *qi = dqp->q_mount->m_quotainfo; struct xfs_dquot *gdqp; + struct xfs_dquot *pdqp; trace_xfs_dqput_free(dqp); @@ -949,21 +950,29 @@ xfs_qm_dqput_final( /* * If we just added a udquot to the freelist, then we want to release - * the gdquot reference that it (probably) has. Otherwise it'll keep - * the gdquot from getting reclaimed. + * the gdquot/pdquot reference that it (probably) has. Otherwise it'll + * keep the gdquot/pdquot from getting reclaimed. */ gdqp = dqp->q_gdquot; if (gdqp) { xfs_dqlock(gdqp); dqp->q_gdquot = NULL; } + + pdqp = dqp->q_pdquot; + if (pdqp) { + xfs_dqlock(pdqp); + dqp->q_pdquot = NULL; + } xfs_dqunlock(dqp); /* - * If we had a group quota hint, release it now. + * If we had a group/project quota hint, release it now. */ if (gdqp) xfs_qm_dqput(gdqp); + if (pdqp) + xfs_qm_dqput(pdqp); } /* diff --git a/fs/xfs/xfs_dquot.h b/fs/xfs/xfs_dquot.h index b596626..55abbca 100644 --- a/fs/xfs/xfs_dquot.h +++ b/fs/xfs/xfs_dquot.h @@ -53,6 +53,7 @@ typedef struct xfs_dquot { xfs_fileoff_t q_fileoffset; /* offset in quotas file */ struct xfs_dquot*q_gdquot; /* group dquot, hint only */ + struct xfs_dquot*q_pdquot; /* project dquot, hint only */ xfs_disk_dquot_t q_core; /* actual usage & quotas */ xfs_dq_logitem_t q_logitem; /* dquot log item */ xfs_qcnt_t q_res_bcount; /* total regular nblks used+reserved */ @@ -118,8 +119,9 @@ static inline int xfs_this_quota_on(struct xfs_mount *mp, int type) case XFS_DQ_USER: return XFS_IS_UQUOTA_ON(mp); case XFS_DQ_GROUP: + return XFS_IS_GQUOTA_ON(mp); case XFS_DQ_PROJ: - return XFS_IS_OQUOTA_ON(mp); + return XFS_IS_PQUOTA_ON(mp); default: return 0; } @@ -131,8 +133,9 @@ static inline xfs_dquot_t *xfs_inode_dquot(struct xfs_inode *ip, int type) case XFS_DQ_USER: return ip->i_udquot; case XFS_DQ_GROUP: - case XFS_DQ_PROJ: return ip->i_gdquot; + case XFS_DQ_PROJ: + return ip->i_pdquot; default: return NULL; } diff --git a/fs/xfs/xfs_icache.c b/fs/xfs/xfs_icache.c index 9560dc1f..3f90e1c 100644 --- a/fs/xfs/xfs_icache.c +++ b/fs/xfs/xfs_icache.c @@ -337,6 +337,7 @@ xfs_iget_cache_miss( iflags |= XFS_IDONTCACHE; ip->i_udquot = NULL; ip->i_gdquot = NULL; + ip->i_pdquot = NULL; xfs_iflags_set(ip, iflags); /* insert the new inode */ diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h index 9112979..b55fd34 100644 --- a/fs/xfs/xfs_inode.h +++ b/fs/xfs/xfs_inode.h @@ -250,6 +250,7 @@ typedef struct xfs_inode { struct xfs_mount *i_mount; /* fs mount struct ptr */ struct xfs_dquot *i_udquot; /* user dquot */ struct xfs_dquot *i_gdquot; /* group dquot */ + struct xfs_dquot *i_pdquot; /* project dquot */ /* Inode location stuff */ xfs_ino_t i_ino; /* inode number (agno/agino)*/ diff --git a/fs/xfs/xfs_ioctl.c b/fs/xfs/xfs_ioctl.c index dc5b659..6e2bca5 100644 --- a/fs/xfs/xfs_ioctl.c +++ b/fs/xfs/xfs_ioctl.c @@ -928,7 +928,7 @@ xfs_ioctl_setattr( struct xfs_trans *tp; unsigned int lock_flags = 0; struct xfs_dquot *udqp = NULL; - struct xfs_dquot *gdqp = NULL; + struct xfs_dquot *pdqp = NULL; struct xfs_dquot *olddquot = NULL; int code; @@ -957,7 +957,7 @@ xfs_ioctl_setattr( if (XFS_IS_QUOTA_ON(mp) && (mask & FSX_PROJID)) { code = xfs_qm_vop_dqalloc(ip, ip->i_d.di_uid, ip->i_d.di_gid, fa->fsx_projid, - XFS_QMOPT_PQUOTA, &udqp, &gdqp); + XFS_QMOPT_PQUOTA, &udqp, NULL, &pdqp); if (code) return code; } @@ -994,8 +994,8 @@ xfs_ioctl_setattr( XFS_IS_PQUOTA_ON(mp) && xfs_get_projid(ip) != fa->fsx_projid) { ASSERT(tp); - code = xfs_qm_vop_chown_reserve(tp, ip, udqp, gdqp, - capable(CAP_FOWNER) ? + code = xfs_qm_vop_chown_reserve(tp, ip, udqp, NULL, + pdqp, capable(CAP_FOWNER) ? XFS_QMOPT_FORCE_RES : 0); if (code) /* out of quota */ goto error_return; @@ -1113,7 +1113,7 @@ xfs_ioctl_setattr( if (xfs_get_projid(ip) != fa->fsx_projid) { if (XFS_IS_QUOTA_RUNNING(mp) && XFS_IS_PQUOTA_ON(mp)) { olddquot = xfs_qm_vop_chown(tp, ip, - &ip->i_gdquot, gdqp); + &ip->i_pdquot, pdqp); } xfs_set_projid(ip, fa->fsx_projid); @@ -1160,13 +1160,13 @@ xfs_ioctl_setattr( */ xfs_qm_dqrele(olddquot); xfs_qm_dqrele(udqp); - xfs_qm_dqrele(gdqp); + xfs_qm_dqrele(pdqp); return code; error_return: xfs_qm_dqrele(udqp); - xfs_qm_dqrele(gdqp); + xfs_qm_dqrele(pdqp); xfs_trans_cancel(tp, 0); if (lock_flags) xfs_iunlock(ip, lock_flags); diff --git a/fs/xfs/xfs_iops.c b/fs/xfs/xfs_iops.c index 8865261..96dda62 100644 --- a/fs/xfs/xfs_iops.c +++ b/fs/xfs/xfs_iops.c @@ -539,7 +539,7 @@ xfs_setattr_nonsize( ASSERT(udqp == NULL); ASSERT(gdqp == NULL); error = xfs_qm_vop_dqalloc(ip, uid, gid, xfs_get_projid(ip), - qflags, &udqp, &gdqp); + qflags, &udqp, &gdqp, NULL); if (error) return error; } @@ -575,7 +575,7 @@ xfs_setattr_nonsize( (XFS_IS_GQUOTA_ON(mp) && igid != gid))) { ASSERT(tp); error = xfs_qm_vop_chown_reserve(tp, ip, udqp, gdqp, - capable(CAP_FOWNER) ? + NULL, capable(CAP_FOWNER) ? XFS_QMOPT_FORCE_RES : 0); if (error) /* out of quota */ goto out_trans_cancel; diff --git a/fs/xfs/xfs_qm.c b/fs/xfs/xfs_qm.c index 7a3e007..d320794 100644 --- a/fs/xfs/xfs_qm.c +++ b/fs/xfs/xfs_qm.c @@ -137,6 +137,7 @@ xfs_qm_dqpurge( struct xfs_mount *mp = dqp->q_mount; struct xfs_quotainfo *qi = mp->m_quotainfo; struct xfs_dquot *gdqp = NULL; + struct xfs_dquot *pdqp = NULL; xfs_dqlock(dqp); if ((dqp->dq_flags & XFS_DQ_FREEING) || dqp->q_nrefs != 0) { @@ -145,8 +146,7 @@ xfs_qm_dqpurge( } /* - * If this quota has a group hint attached, prepare for releasing it - * now. + * If this quota has a hint attached, prepare for releasing it now. */ gdqp = dqp->q_gdquot; if (gdqp) { @@ -154,6 +154,12 @@ xfs_qm_dqpurge( dqp->q_gdquot = NULL; } + pdqp = dqp->q_pdquot; + if (pdqp) { + xfs_dqlock(pdqp); + dqp->q_pdquot = NULL; + } + dqp->dq_flags |= XFS_DQ_FREEING; xfs_dqflock(dqp); @@ -208,6 +214,8 @@ xfs_qm_dqpurge( if (gdqp) xfs_qm_dqput(gdqp); + if (pdqp) + xfs_qm_dqput(pdqp); return 0; } @@ -364,6 +372,10 @@ xfs_qm_unmount_quotas( IRELE(mp->m_quotainfo->qi_gquotaip); mp->m_quotainfo->qi_gquotaip = NULL; } + if (mp->m_quotainfo->qi_pquotaip) { + IRELE(mp->m_quotainfo->qi_pquotaip); + mp->m_quotainfo->qi_pquotaip = NULL; + } } } @@ -410,7 +422,10 @@ xfs_qm_dqattach_one( * be reclaimed as long as we have a ref from inode and we * hold the ilock. */ - dqp = udqhint->q_gdquot; + if (type == XFS_DQ_GROUP) + dqp = udqhint->q_gdquot; + else + dqp = udqhint->q_pdquot; if (dqp && be32_to_cpu(dqp->q_core.d_id) == id) { ASSERT(*IO_idqpp == NULL); @@ -453,28 +468,42 @@ xfs_qm_dqattach_one( /* - * Given a udquot and gdquot, attach a ptr to the group dquot in the - * udquot as a hint for future lookups. + * Given a udquot and group/project type, attach the group/project + * dquot pointer to the udquot as a hint for future lookups. */ STATIC void -xfs_qm_dqattach_grouphint( - xfs_dquot_t *udq, - xfs_dquot_t *gdq) +xfs_qm_dqattach_hint( + struct xfs_inode *ip, + int type) { - xfs_dquot_t *tmp; + struct xfs_dquot **dqhintp; + struct xfs_dquot *dqp; + struct xfs_dquot *udq = ip->i_udquot; + + ASSERT(type == XFS_DQ_GROUP || type == XFS_DQ_PROJ); xfs_dqlock(udq); - tmp = udq->q_gdquot; - if (tmp) { - if (tmp == gdq) + if (type == XFS_DQ_GROUP) { + dqp = ip->i_gdquot; + dqhintp = &udq->q_gdquot; + } else { + dqp = ip->i_pdquot; + dqhintp = &udq->q_pdquot; + } + + if (*dqhintp) { + struct xfs_dquot *tmp; + + if (*dqhintp == dqp) goto done; - udq->q_gdquot = NULL; + tmp = *dqhintp; + *dqhintp = NULL; xfs_qm_dqrele(tmp); } - udq->q_gdquot = xfs_qm_dqhold(gdq); + *dqhintp = xfs_qm_dqhold(dqp); done: xfs_dqunlock(udq); } @@ -527,12 +556,8 @@ xfs_qm_dqattach_locked( } ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); - if (XFS_IS_OQUOTA_ON(mp)) { - error = XFS_IS_GQUOTA_ON(mp) ? - xfs_qm_dqattach_one(ip, ip->i_d.di_gid, XFS_DQ_GROUP, - flags & XFS_QMOPT_DQALLOC, - ip->i_udquot, &ip->i_gdquot) : - xfs_qm_dqattach_one(ip, xfs_get_projid(ip), XFS_DQ_PROJ, + if (XFS_IS_GQUOTA_ON(mp)) { + error = xfs_qm_dqattach_one(ip, ip->i_d.di_gid, XFS_DQ_GROUP, flags & XFS_QMOPT_DQALLOC, ip->i_udquot, &ip->i_gdquot); /* @@ -544,14 +569,28 @@ xfs_qm_dqattach_locked( nquotas++; } + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); + if (XFS_IS_PQUOTA_ON(mp)) { + error = xfs_qm_dqattach_one(ip, xfs_get_projid(ip), XFS_DQ_PROJ, + flags & XFS_QMOPT_DQALLOC, + ip->i_udquot, &ip->i_pdquot); + /* + * Don't worry about the udquot that we may have + * attached above. It'll get detached, if not already. + */ + if (error) + goto done; + nquotas++; + } + /* - * Attach this group quota to the user quota as a hint. + * Attach this group/project quota to the user quota as a hint. * This WON'T, in general, result in a thrash. */ - if (nquotas == 2) { + if (nquotas > 1 && ip->i_udquot) { ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); - ASSERT(ip->i_udquot); - ASSERT(ip->i_gdquot); + ASSERT(ip->i_gdquot || !XFS_IS_GQUOTA_ON(mp)); + ASSERT(ip->i_pdquot || !XFS_IS_PQUOTA_ON(mp)); /* * We do not have i_udquot locked at this point, but this check @@ -560,7 +599,10 @@ xfs_qm_dqattach_locked( * succeed in general. */ if (ip->i_udquot->q_gdquot != ip->i_gdquot) - xfs_qm_dqattach_grouphint(ip->i_udquot, ip->i_gdquot); + xfs_qm_dqattach_hint(ip, XFS_DQ_GROUP); + + if (ip->i_udquot->q_pdquot != ip->i_pdquot) + xfs_qm_dqattach_hint(ip, XFS_DQ_PROJ); } done: @@ -568,8 +610,10 @@ xfs_qm_dqattach_locked( if (!error) { if (XFS_IS_UQUOTA_ON(mp)) ASSERT(ip->i_udquot); - if (XFS_IS_OQUOTA_ON(mp)) + if (XFS_IS_GQUOTA_ON(mp)) ASSERT(ip->i_gdquot); + if (XFS_IS_PQUOTA_ON(mp)) + ASSERT(ip->i_pdquot); } ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); #endif @@ -602,7 +646,7 @@ void xfs_qm_dqdetach( xfs_inode_t *ip) { - if (!(ip->i_udquot || ip->i_gdquot)) + if (!(ip->i_udquot || ip->i_gdquot || ip->i_pdquot)) return; trace_xfs_dquot_dqdetach(ip); @@ -616,6 +660,10 @@ xfs_qm_dqdetach( xfs_qm_dqrele(ip->i_gdquot); ip->i_gdquot = NULL; } + if (ip->i_pdquot) { + xfs_qm_dqrele(ip->i_pdquot); + ip->i_pdquot = NULL; + } } int @@ -660,6 +708,7 @@ xfs_qm_init_quotainfo( INIT_RADIX_TREE(&qinf->qi_uquota_tree, GFP_NOFS); INIT_RADIX_TREE(&qinf->qi_gquota_tree, GFP_NOFS); + INIT_RADIX_TREE(&qinf->qi_pquota_tree, GFP_NOFS); mutex_init(&qinf->qi_tree_lock); INIT_LIST_HEAD(&qinf->qi_lru_list); @@ -761,6 +810,10 @@ xfs_qm_destroy_quotainfo( IRELE(qi->qi_gquotaip); qi->qi_gquotaip = NULL; } + if (qi->qi_pquotaip) { + IRELE(qi->qi_pquotaip); + qi->qi_pquotaip = NULL; + } mutex_destroy(&qi->qi_quotaofflock); kmem_free(qi); mp->m_quotainfo = NULL; @@ -1269,13 +1322,14 @@ xfs_qm_quotacheck( LIST_HEAD (buffer_list); struct xfs_inode *uip = mp->m_quotainfo->qi_uquotaip; struct xfs_inode *gip = mp->m_quotainfo->qi_gquotaip; + struct xfs_inode *pip = mp->m_quotainfo->qi_pquotaip; count = INT_MAX; structsz = 1; lastino = 0; flags = 0; - ASSERT(uip || gip); + ASSERT(uip || gip || pip); ASSERT(XFS_IS_QUOTA_RUNNING(mp)); xfs_notice(mp, "Quotacheck needed: Please wait."); @@ -1294,13 +1348,19 @@ xfs_qm_quotacheck( } if (gip) { - error = xfs_qm_dqiterate(mp, gip, XFS_IS_GQUOTA_ON(mp) ? - XFS_QMOPT_GQUOTA : XFS_QMOPT_PQUOTA, + error = xfs_qm_dqiterate(mp, gip, XFS_QMOPT_GQUOTA, &buffer_list); if (error) goto error_return; - flags |= XFS_IS_GQUOTA_ON(mp) ? - XFS_GQUOTA_CHKD : XFS_PQUOTA_CHKD; + flags |= XFS_GQUOTA_CHKD; + } + + if (pip) { + error = xfs_qm_dqiterate(mp, pip, XFS_QMOPT_PQUOTA, + &buffer_list); + if (error) + goto error_return; + flags |= XFS_PQUOTA_CHKD; } do { @@ -1397,6 +1457,7 @@ xfs_qm_init_quotainos( { struct xfs_inode *uip = NULL; struct xfs_inode *gip = NULL; + struct xfs_inode *pip = NULL; int error; __int64_t sbflags = 0; uint flags = 0; @@ -1415,7 +1476,7 @@ xfs_qm_init_quotainos( if (error) return XFS_ERROR(error); } - if (XFS_IS_OQUOTA_ON(mp) && + if (XFS_IS_GQUOTA_ON(mp) && mp->m_sb.sb_gquotino != NULLFSINO) { ASSERT(mp->m_sb.sb_gquotino > 0); error = xfs_iget(mp, NULL, mp->m_sb.sb_gquotino, @@ -1423,6 +1484,15 @@ xfs_qm_init_quotainos( if (error) goto error_rele; } + /* XXX: Use gquotino for now */ + if (XFS_IS_PQUOTA_ON(mp) && + mp->m_sb.sb_gquotino != NULLFSINO) { + ASSERT(mp->m_sb.sb_gquotino > 0); + error = xfs_iget(mp, NULL, mp->m_sb.sb_gquotino, + 0, 0, &pip); + if (error) + goto error_rele; + } } else { flags |= XFS_QMOPT_SBVERSION; sbflags |= (XFS_SB_VERSIONNUM | XFS_SB_UQUOTINO | @@ -1430,7 +1500,7 @@ xfs_qm_init_quotainos( } /* - * Create the two inodes, if they don't exist already. The changes + * Create the three inodes, if they don't exist already. The changes * made above will get added to a transaction and logged in one of * the qino_alloc calls below. If the device is readonly, * temporarily switch to read-write to do this. @@ -1444,17 +1514,27 @@ xfs_qm_init_quotainos( flags &= ~XFS_QMOPT_SBVERSION; } - if (XFS_IS_OQUOTA_ON(mp) && gip == NULL) { - flags |= (XFS_IS_GQUOTA_ON(mp) ? - XFS_QMOPT_GQUOTA : XFS_QMOPT_PQUOTA); + if (XFS_IS_GQUOTA_ON(mp) && gip == NULL) { error = xfs_qm_qino_alloc(mp, &gip, - sbflags | XFS_SB_GQUOTINO, flags); + sbflags | XFS_SB_GQUOTINO, + flags | XFS_QMOPT_GQUOTA); + if (error) + goto error_rele; + + flags &= ~XFS_QMOPT_SBVERSION; + } + if (XFS_IS_PQUOTA_ON(mp) && pip == NULL) { + /* XXX: Use XFS_SB_GQUOTINO for now */ + error = xfs_qm_qino_alloc(mp, &pip, + sbflags | XFS_SB_GQUOTINO, + flags | XFS_QMOPT_PQUOTA); if (error) goto error_rele; } mp->m_quotainfo->qi_uquotaip = uip; mp->m_quotainfo->qi_gquotaip = gip; + mp->m_quotainfo->qi_pquotaip = pip; return 0; @@ -1463,6 +1543,8 @@ error_rele: IRELE(uip); if (gip) IRELE(gip); + if (pip) + IRELE(pip); return XFS_ERROR(error); } @@ -1657,11 +1739,13 @@ xfs_qm_vop_dqalloc( prid_t prid, uint flags, struct xfs_dquot **O_udqpp, - struct xfs_dquot **O_gdqpp) + struct xfs_dquot **O_gdqpp, + struct xfs_dquot **O_pdqpp) { struct xfs_mount *mp = ip->i_mount; struct xfs_dquot *uq = NULL; struct xfs_dquot *gq = NULL; + struct xfs_dquot *pq = NULL; int error; uint lockflags; @@ -1741,24 +1825,25 @@ xfs_qm_vop_dqalloc( ASSERT(ip->i_gdquot); gq = xfs_qm_dqhold(ip->i_gdquot); } - } else if ((flags & XFS_QMOPT_PQUOTA) && XFS_IS_PQUOTA_ON(mp)) { + } + if ((flags & XFS_QMOPT_PQUOTA) && XFS_IS_PQUOTA_ON(mp)) { if (xfs_get_projid(ip) != prid) { xfs_iunlock(ip, lockflags); error = xfs_qm_dqget(mp, NULL, (xfs_dqid_t)prid, XFS_DQ_PROJ, XFS_QMOPT_DQALLOC | XFS_QMOPT_DOWARN, - &gq); + &pq); if (error) { ASSERT(error != ENOENT); goto error_rele; } - xfs_dqunlock(gq); + xfs_dqunlock(pq); lockflags = XFS_ILOCK_SHARED; xfs_ilock(ip, lockflags); } else { - ASSERT(ip->i_gdquot); - gq = xfs_qm_dqhold(ip->i_gdquot); + ASSERT(ip->i_pdquot); + pq = xfs_qm_dqhold(ip->i_pdquot); } } if (uq) @@ -1773,9 +1858,15 @@ xfs_qm_vop_dqalloc( *O_gdqpp = gq; else if (gq) xfs_qm_dqrele(gq); + if (O_pdqpp) + *O_pdqpp = pq; + else if (pq) + xfs_qm_dqrele(pq); return 0; error_rele: + if (gq) + xfs_qm_dqrele(gq); if (uq) xfs_qm_dqrele(uq); return error; @@ -1830,14 +1921,17 @@ xfs_qm_vop_chown_reserve( struct xfs_inode *ip, struct xfs_dquot *udqp, struct xfs_dquot *gdqp, + struct xfs_dquot *pdqp, uint flags) { struct xfs_mount *mp = ip->i_mount; uint delblks, blkflags, prjflags = 0; struct xfs_dquot *udq_unres = NULL; struct xfs_dquot *gdq_unres = NULL; + struct xfs_dquot *pdq_unres = NULL; struct xfs_dquot *udq_delblks = NULL; struct xfs_dquot *gdq_delblks = NULL; + struct xfs_dquot *pdq_delblks = NULL; int error; @@ -1861,24 +1955,28 @@ xfs_qm_vop_chown_reserve( udq_unres = ip->i_udquot; } } - if (XFS_IS_OQUOTA_ON(ip->i_mount) && gdqp) { - if (XFS_IS_PQUOTA_ON(ip->i_mount) && - xfs_get_projid(ip) != be32_to_cpu(gdqp->q_core.d_id)) - prjflags = XFS_QMOPT_ENOSPC; - - if (prjflags || - (XFS_IS_GQUOTA_ON(ip->i_mount) && - ip->i_d.di_gid != be32_to_cpu(gdqp->q_core.d_id))) { - gdq_delblks = gdqp; - if (delblks) { - ASSERT(ip->i_gdquot); - gdq_unres = ip->i_gdquot; - } + if (XFS_IS_GQUOTA_ON(ip->i_mount) && gdqp && + ip->i_d.di_gid != be32_to_cpu(gdqp->q_core.d_id)) { + gdq_delblks = gdqp; + if (delblks) { + ASSERT(ip->i_gdquot); + gdq_unres = ip->i_gdquot; + } + } + + if (XFS_IS_PQUOTA_ON(ip->i_mount) && pdqp && + xfs_get_projid(ip) != be32_to_cpu(pdqp->q_core.d_id)) { + prjflags = XFS_QMOPT_ENOSPC; + pdq_delblks = pdqp; + if (delblks) { + ASSERT(ip->i_pdquot); + pdq_unres = ip->i_pdquot; } } error = xfs_trans_reserve_quota_bydquots(tp, ip->i_mount, - udq_delblks, gdq_delblks, ip->i_d.di_nblocks, 1, + udq_delblks, gdq_delblks, pdq_delblks, + ip->i_d.di_nblocks, 1, flags | blkflags | prjflags); if (error) return error; @@ -1893,16 +1991,17 @@ xfs_qm_vop_chown_reserve( /* * Do the reservations first. Unreservation can't fail. */ - ASSERT(udq_delblks || gdq_delblks); - ASSERT(udq_unres || gdq_unres); + ASSERT(udq_delblks || gdq_delblks || pdq_delblks); + ASSERT(udq_unres || gdq_unres || pdq_unres); error = xfs_trans_reserve_quota_bydquots(NULL, ip->i_mount, - udq_delblks, gdq_delblks, (xfs_qcnt_t)delblks, 0, + udq_delblks, gdq_delblks, pdq_delblks, + (xfs_qcnt_t)delblks, 0, flags | blkflags | prjflags); if (error) return error; xfs_trans_reserve_quota_bydquots(NULL, ip->i_mount, - udq_unres, gdq_unres, -((xfs_qcnt_t)delblks), 0, - blkflags); + udq_unres, gdq_unres, pdq_unres, + -((xfs_qcnt_t)delblks), 0, blkflags); } return (0); @@ -1941,7 +2040,8 @@ xfs_qm_vop_create_dqattach( struct xfs_trans *tp, struct xfs_inode *ip, struct xfs_dquot *udqp, - struct xfs_dquot *gdqp) + struct xfs_dquot *gdqp, + struct xfs_dquot *pdqp) { struct xfs_mount *mp = tp->t_mountp; @@ -1961,13 +2061,18 @@ xfs_qm_vop_create_dqattach( } if (gdqp) { ASSERT(ip->i_gdquot == NULL); - ASSERT(XFS_IS_OQUOTA_ON(mp)); - ASSERT((XFS_IS_GQUOTA_ON(mp) ? - ip->i_d.di_gid : xfs_get_projid(ip)) == - be32_to_cpu(gdqp->q_core.d_id)); - + ASSERT(XFS_IS_GQUOTA_ON(mp)); + ASSERT(ip->i_d.di_gid == be32_to_cpu(gdqp->q_core.d_id)); ip->i_gdquot = xfs_qm_dqhold(gdqp); xfs_trans_mod_dquot(tp, gdqp, XFS_TRANS_DQ_ICOUNT, 1); } + if (pdqp) { + ASSERT(ip->i_pdquot == NULL); + ASSERT(XFS_IS_PQUOTA_ON(mp)); + ASSERT(xfs_get_projid(ip) == be32_to_cpu(pdqp->q_core.d_id)); + + ip->i_pdquot = xfs_qm_dqhold(pdqp); + xfs_trans_mod_dquot(tp, pdqp, XFS_TRANS_DQ_ICOUNT, 1); + } } diff --git a/fs/xfs/xfs_qm.h b/fs/xfs/xfs_qm.h index bdb4f8b..579d6a0 100644 --- a/fs/xfs/xfs_qm.h +++ b/fs/xfs/xfs_qm.h @@ -44,9 +44,11 @@ extern struct kmem_zone *xfs_qm_dqtrxzone; typedef struct xfs_quotainfo { struct radix_tree_root qi_uquota_tree; struct radix_tree_root qi_gquota_tree; + struct radix_tree_root qi_pquota_tree; struct mutex qi_tree_lock; - xfs_inode_t *qi_uquotaip; /* user quota inode */ - xfs_inode_t *qi_gquotaip; /* group quota inode */ + struct xfs_inode *qi_uquotaip; /* user quota inode */ + struct xfs_inode *qi_gquotaip; /* group quota inode */ + struct xfs_inode *qi_pquotaip; /* project quota inode */ struct list_head qi_lru_list; struct mutex qi_lru_lock; int qi_lru_count; @@ -78,8 +80,9 @@ xfs_dquot_tree( case XFS_DQ_USER: return &qi->qi_uquota_tree; case XFS_DQ_GROUP: - case XFS_DQ_PROJ: return &qi->qi_gquota_tree; + case XFS_DQ_PROJ: + return &qi->qi_pquota_tree; default: ASSERT(0); } @@ -93,8 +96,9 @@ xfs_dq_to_quota_inode(struct xfs_dquot *dqp) case XFS_DQ_USER: return dqp->q_mount->m_quotainfo->qi_uquotaip; case XFS_DQ_GROUP: - case XFS_DQ_PROJ: return dqp->q_mount->m_quotainfo->qi_gquotaip; + case XFS_DQ_PROJ: + return dqp->q_mount->m_quotainfo->qi_pquotaip; default: ASSERT(0); } @@ -107,18 +111,20 @@ extern void xfs_trans_mod_dquot(struct xfs_trans *, struct xfs_dquot *, uint, long); extern int xfs_trans_reserve_quota_bydquots(struct xfs_trans *, struct xfs_mount *, struct xfs_dquot *, - struct xfs_dquot *, long, long, uint); + struct xfs_dquot *, struct xfs_dquot *, + long, long, uint); extern void xfs_trans_dqjoin(struct xfs_trans *, struct xfs_dquot *); extern void xfs_trans_log_dquot(struct xfs_trans *, struct xfs_dquot *); /* - * We keep the usr and grp dquots separately so that locking will be easier - * to do at commit time. All transactions that we know of at this point + * We keep the usr, grp, and prj dquots separately so that locking will be + * easier to do at commit time. All transactions that we know of at this point * affect no more than two dquots of one type. Hence, the TRANS_MAXDQS value. */ enum { XFS_QM_TRANS_USR = 0, XFS_QM_TRANS_GRP, + XFS_QM_TRANS_PRJ, XFS_QM_TRANS_DQTYPES }; #define XFS_QM_TRANS_MAXDQS 2 diff --git a/fs/xfs/xfs_qm_bhv.c b/fs/xfs/xfs_qm_bhv.c index 2d02eac..437a52d 100644 --- a/fs/xfs/xfs_qm_bhv.c +++ b/fs/xfs/xfs_qm_bhv.c @@ -112,16 +112,16 @@ xfs_qm_newmount( if (((uquotaondisk && !XFS_IS_UQUOTA_ON(mp)) || (!uquotaondisk && XFS_IS_UQUOTA_ON(mp)) || - (pquotaondisk && !XFS_IS_PQUOTA_ON(mp)) || - (!pquotaondisk && XFS_IS_PQUOTA_ON(mp)) || (gquotaondisk && !XFS_IS_GQUOTA_ON(mp)) || - (!gquotaondisk && XFS_IS_OQUOTA_ON(mp))) && + (!gquotaondisk && XFS_IS_GQUOTA_ON(mp)) || + (pquotaondisk && !XFS_IS_PQUOTA_ON(mp)) || + (!pquotaondisk && XFS_IS_PQUOTA_ON(mp))) && xfs_dev_is_read_only(mp, "changing quota state")) { xfs_warn(mp, "please mount with%s%s%s%s.", (!quotaondisk ? "out quota" : ""), (uquotaondisk ? " usrquota" : ""), - (pquotaondisk ? " prjquota" : ""), - (gquotaondisk ? " grpquota" : "")); + (gquotaondisk ? " grpquota" : ""), + (pquotaondisk ? " prjquota" : "")); return XFS_ERROR(EPERM); } diff --git a/fs/xfs/xfs_qm_syscalls.c b/fs/xfs/xfs_qm_syscalls.c index a08801a..b936383 100644 --- a/fs/xfs/xfs_qm_syscalls.c +++ b/fs/xfs/xfs_qm_syscalls.c @@ -119,7 +119,8 @@ xfs_qm_scall_quotaoff( dqtype |= XFS_QMOPT_GQUOTA; flags |= (XFS_GQUOTA_CHKD | XFS_GQUOTA_ENFD); inactivate_flags |= XFS_GQUOTA_ACTIVE; - } else if (flags & XFS_PQUOTA_ACCT) { + } + if (flags & XFS_PQUOTA_ACCT) { dqtype |= XFS_QMOPT_PQUOTA; flags |= (XFS_PQUOTA_CHKD | XFS_PQUOTA_ENFD); inactivate_flags |= XFS_PQUOTA_ACTIVE; @@ -214,10 +215,14 @@ xfs_qm_scall_quotaoff( IRELE(q->qi_uquotaip); q->qi_uquotaip = NULL; } - if ((dqtype & (XFS_QMOPT_GQUOTA|XFS_QMOPT_PQUOTA)) && q->qi_gquotaip) { + if ((dqtype & XFS_QMOPT_GQUOTA) && q->qi_gquotaip) { IRELE(q->qi_gquotaip); q->qi_gquotaip = NULL; } + if ((dqtype & XFS_QMOPT_PQUOTA) && q->qi_pquotaip) { + IRELE(q->qi_pquotaip); + q->qi_pquotaip = NULL; + } out_unlock: mutex_unlock(&q->qi_quotaofflock); @@ -859,9 +864,11 @@ xfs_dqrele_inode( { /* skip quota inodes */ if (ip == ip->i_mount->m_quotainfo->qi_uquotaip || - ip == ip->i_mount->m_quotainfo->qi_gquotaip) { + ip == ip->i_mount->m_quotainfo->qi_gquotaip || + ip == ip->i_mount->m_quotainfo->qi_pquotaip) { ASSERT(ip->i_udquot == NULL); ASSERT(ip->i_gdquot == NULL); + ASSERT(ip->i_pdquot == NULL); return 0; } @@ -870,10 +877,14 @@ xfs_dqrele_inode( xfs_qm_dqrele(ip->i_udquot); ip->i_udquot = NULL; } - if (flags & (XFS_PQUOTA_ACCT|XFS_GQUOTA_ACCT) && ip->i_gdquot) { + if ((flags & XFS_GQUOTA_ACCT) && ip->i_gdquot) { xfs_qm_dqrele(ip->i_gdquot); ip->i_gdquot = NULL; } + if ((flags & XFS_PQUOTA_ACCT) && ip->i_pdquot) { + xfs_qm_dqrele(ip->i_pdquot); + ip->i_pdquot = NULL; + } xfs_iunlock(ip, XFS_ILOCK_EXCL); return 0; } diff --git a/fs/xfs/xfs_quota.h b/fs/xfs/xfs_quota.h index 40d508b..09777c4 100644 --- a/fs/xfs/xfs_quota.h +++ b/fs/xfs/xfs_quota.h @@ -288,10 +288,10 @@ typedef struct xfs_qoff_logformat { * we didn't have the inode locked, the appropriate dquot(s) will be * attached atomically. */ -#define XFS_NOT_DQATTACHED(mp, ip) ((XFS_IS_UQUOTA_ON(mp) &&\ - (ip)->i_udquot == NULL) || \ - (XFS_IS_OQUOTA_ON(mp) && \ - (ip)->i_gdquot == NULL)) +#define XFS_NOT_DQATTACHED(mp, ip) \ + ((XFS_IS_UQUOTA_ON(mp) && (ip)->i_udquot == NULL) || \ + (XFS_IS_GQUOTA_ON(mp) && (ip)->i_gdquot == NULL) || \ + (XFS_IS_PQUOTA_ON(mp) && (ip)->i_pdquot == NULL)) #define XFS_QM_NEED_QUOTACHECK(mp) \ ((XFS_IS_UQUOTA_ON(mp) && \ @@ -346,17 +346,18 @@ extern int xfs_trans_reserve_quota_nblks(struct xfs_trans *, struct xfs_inode *, long, long, uint); extern int xfs_trans_reserve_quota_bydquots(struct xfs_trans *, struct xfs_mount *, struct xfs_dquot *, - struct xfs_dquot *, long, long, uint); + struct xfs_dquot *, struct xfs_dquot *, long, long, uint); extern int xfs_qm_vop_dqalloc(struct xfs_inode *, uid_t, gid_t, prid_t, uint, - struct xfs_dquot **, struct xfs_dquot **); + struct xfs_dquot **, struct xfs_dquot **, struct xfs_dquot **); extern void xfs_qm_vop_create_dqattach(struct xfs_trans *, struct xfs_inode *, - struct xfs_dquot *, struct xfs_dquot *); + struct xfs_dquot *, struct xfs_dquot *, struct xfs_dquot *); extern int xfs_qm_vop_rename_dqattach(struct xfs_inode **); extern struct xfs_dquot *xfs_qm_vop_chown(struct xfs_trans *, struct xfs_inode *, struct xfs_dquot **, struct xfs_dquot *); extern int xfs_qm_vop_chown_reserve(struct xfs_trans *, struct xfs_inode *, - struct xfs_dquot *, struct xfs_dquot *, uint); + struct xfs_dquot *, struct xfs_dquot *, + struct xfs_dquot *, uint); extern int xfs_qm_dqattach(struct xfs_inode *, uint); extern int xfs_qm_dqattach_locked(struct xfs_inode *, uint); extern void xfs_qm_dqdetach(struct xfs_inode *); @@ -370,10 +371,12 @@ extern void xfs_qm_unmount_quotas(struct xfs_mount *); #else static inline int xfs_qm_vop_dqalloc(struct xfs_inode *ip, uid_t uid, gid_t gid, prid_t prid, - uint flags, struct xfs_dquot **udqp, struct xfs_dquot **gdqp) + uint flags, struct xfs_dquot **udqp, struct xfs_dquot **gdqp, + struct xfs_dquot **pdqp) { *udqp = NULL; *gdqp = NULL; + *pdqp = NULL; return 0; } #define xfs_trans_dup_dqinfo(tp, tp2) @@ -388,14 +391,15 @@ static inline int xfs_trans_reserve_quota_nblks(struct xfs_trans *tp, } static inline int xfs_trans_reserve_quota_bydquots(struct xfs_trans *tp, struct xfs_mount *mp, struct xfs_dquot *udqp, - struct xfs_dquot *gdqp, long nblks, long nions, uint flags) + struct xfs_dquot *gdqp, struct xfs_dquot *pdqp, + long nblks, long nions, uint flags) { return 0; } -#define xfs_qm_vop_create_dqattach(tp, ip, u, g) +#define xfs_qm_vop_create_dqattach(tp, ip, u, g, p) #define xfs_qm_vop_rename_dqattach(it) (0) #define xfs_qm_vop_chown(tp, ip, old, new) (NULL) -#define xfs_qm_vop_chown_reserve(tp, ip, u, g, fl) (0) +#define xfs_qm_vop_chown_reserve(tp, ip, u, g, p, fl) (0) #define xfs_qm_dqattach(ip, fl) (0) #define xfs_qm_dqattach_locked(ip, fl) (0) #define xfs_qm_dqdetach(ip) @@ -409,8 +413,8 @@ static inline int xfs_trans_reserve_quota_bydquots(struct xfs_trans *tp, #define xfs_trans_unreserve_quota_nblks(tp, ip, nblks, ninos, flags) \ xfs_trans_reserve_quota_nblks(tp, ip, -(nblks), -(ninos), flags) -#define xfs_trans_reserve_quota(tp, mp, ud, gd, nb, ni, f) \ - xfs_trans_reserve_quota_bydquots(tp, mp, ud, gd, nb, ni, \ +#define xfs_trans_reserve_quota(tp, mp, ud, gd, pd, nb, ni, f) \ + xfs_trans_reserve_quota_bydquots(tp, mp, ud, gd, pd, nb, ni, \ f | XFS_QMOPT_RES_REGBLKS) extern int xfs_qm_dqcheck(struct xfs_mount *, xfs_disk_dquot_t *, diff --git a/fs/xfs/xfs_symlink.c b/fs/xfs/xfs_symlink.c index e830fb5..f4895b6 100644 --- a/fs/xfs/xfs_symlink.c +++ b/fs/xfs/xfs_symlink.c @@ -360,6 +360,7 @@ xfs_symlink( prid_t prid; struct xfs_dquot *udqp = NULL; struct xfs_dquot *gdqp = NULL; + struct xfs_dquot *pdqp = NULL; uint resblks; *ipp = NULL; @@ -386,7 +387,7 @@ xfs_symlink( * Make sure that we have allocated dquot(s) on disk. */ error = xfs_qm_vop_dqalloc(dp, current_fsuid(), current_fsgid(), prid, - XFS_QMOPT_QUOTALL | XFS_QMOPT_INHERIT, &udqp, &gdqp); + XFS_QMOPT_QUOTALL | XFS_QMOPT_INHERIT, &udqp, &gdqp, &pdqp); if (error) goto std_return; @@ -427,7 +428,8 @@ xfs_symlink( /* * Reserve disk quota : blocks and inode. */ - error = xfs_trans_reserve_quota(tp, mp, udqp, gdqp, resblks, 1, 0); + error = xfs_trans_reserve_quota(tp, mp, udqp, gdqp, + pdqp, resblks, 1, 0); if (error) goto error_return; @@ -465,7 +467,7 @@ xfs_symlink( /* * Also attach the dquot(s) to it, if applicable. */ - xfs_qm_vop_create_dqattach(tp, ip, udqp, gdqp); + xfs_qm_vop_create_dqattach(tp, ip, udqp, gdqp, pdqp); if (resblks) resblks -= XFS_IALLOC_SPACE_RES(mp); @@ -563,6 +565,7 @@ xfs_symlink( error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES); xfs_qm_dqrele(udqp); xfs_qm_dqrele(gdqp); + xfs_qm_dqrele(pdqp); *ipp = ip; return 0; @@ -576,6 +579,7 @@ xfs_symlink( xfs_trans_cancel(tp, cancel_flags); xfs_qm_dqrele(udqp); xfs_qm_dqrele(gdqp); + xfs_qm_dqrele(pdqp); if (unlock_dp_on_error) xfs_iunlock(dp, XFS_ILOCK_EXCL); diff --git a/fs/xfs/xfs_trans_dquot.c b/fs/xfs/xfs_trans_dquot.c index db041a5..61407a8 100644 --- a/fs/xfs/xfs_trans_dquot.c +++ b/fs/xfs/xfs_trans_dquot.c @@ -163,8 +163,10 @@ xfs_trans_mod_dquot_byino( if (XFS_IS_UQUOTA_ON(mp) && ip->i_udquot) (void) xfs_trans_mod_dquot(tp, ip->i_udquot, field, delta); - if (XFS_IS_OQUOTA_ON(mp) && ip->i_gdquot) + if (XFS_IS_GQUOTA_ON(mp) && ip->i_gdquot) (void) xfs_trans_mod_dquot(tp, ip->i_gdquot, field, delta); + if (XFS_IS_PQUOTA_ON(mp) && ip->i_pdquot) + (void) xfs_trans_mod_dquot(tp, ip->i_pdquot, field, delta); } STATIC struct xfs_dqtrx * @@ -177,8 +179,12 @@ xfs_trans_get_dqtrx( if (XFS_QM_ISUDQ(dqp)) qa = tp->t_dqinfo->dqs[XFS_QM_TRANS_USR]; - else + else if (XFS_QM_ISGDQ(dqp)) qa = tp->t_dqinfo->dqs[XFS_QM_TRANS_GRP]; + else if (XFS_QM_ISPDQ(dqp)) + qa = tp->t_dqinfo->dqs[XFS_QM_TRANS_PRJ]; + else + return NULL; for (i = 0; i < XFS_QM_TRANS_MAXDQS; i++) { if (qa[i].qt_dquot == NULL || @@ -727,8 +733,8 @@ error_return: /* * Given dquot(s), make disk block and/or inode reservations against them. - * The fact that this does the reservation against both the usr and - * grp/prj quotas is important, because this follows a both-or-nothing + * The fact that this does the reservation against user, group and + * project quotas is important, because this follows a all-or-nothing * approach. * * flags = XFS_QMOPT_FORCE_RES evades limit enforcement. Used by chown. @@ -743,6 +749,7 @@ xfs_trans_reserve_quota_bydquots( struct xfs_mount *mp, struct xfs_dquot *udqp, struct xfs_dquot *gdqp, + struct xfs_dquot *pdqp, long nblks, long ninos, uint flags) @@ -770,11 +777,21 @@ xfs_trans_reserve_quota_bydquots( goto unwind_usr; } + if (pdqp) { + error = xfs_trans_dqresv(tp, mp, pdqp, nblks, ninos, flags); + if (error) + goto unwind_grp; + } + /* * Didn't change anything critical, so, no need to log */ return 0; +unwind_grp: + flags |= XFS_QMOPT_FORCE_RES; + if (gdqp) + xfs_trans_dqresv(tp, mp, gdqp, -nblks, -ninos, flags); unwind_usr: flags |= XFS_QMOPT_FORCE_RES; if (udqp) @@ -816,6 +833,7 @@ xfs_trans_reserve_quota_nblks( */ return xfs_trans_reserve_quota_bydquots(tp, mp, ip->i_udquot, ip->i_gdquot, + ip->i_pdquot, nblks, ninos, flags); } diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c index 42c0ef2..dc730ac 100644 --- a/fs/xfs/xfs_vnodeops.c +++ b/fs/xfs/xfs_vnodeops.c @@ -489,6 +489,7 @@ xfs_create( prid_t prid; struct xfs_dquot *udqp = NULL; struct xfs_dquot *gdqp = NULL; + struct xfs_dquot *pdqp = NULL; uint resblks; uint log_res; uint log_count; @@ -507,7 +508,8 @@ xfs_create( * Make sure that we have allocated dquot(s) on disk. */ error = xfs_qm_vop_dqalloc(dp, current_fsuid(), current_fsgid(), prid, - XFS_QMOPT_QUOTALL | XFS_QMOPT_INHERIT, &udqp, &gdqp); + XFS_QMOPT_QUOTALL | XFS_QMOPT_INHERIT, + &udqp, &gdqp, &pdqp); if (error) return error; @@ -559,7 +561,8 @@ xfs_create( /* * Reserve disk quota and the inode. */ - error = xfs_trans_reserve_quota(tp, mp, udqp, gdqp, resblks, 1, 0); + error = xfs_trans_reserve_quota(tp, mp, udqp, gdqp, + pdqp, resblks, 1, 0); if (error) goto out_trans_cancel; @@ -623,7 +626,7 @@ xfs_create( * These ids of the inode couldn't have changed since the new * inode has been locked ever since it was created. */ - xfs_qm_vop_create_dqattach(tp, ip, udqp, gdqp); + xfs_qm_vop_create_dqattach(tp, ip, udqp, gdqp, pdqp); error = xfs_bmap_finish(&tp, &free_list, &committed); if (error) @@ -635,6 +638,7 @@ xfs_create( xfs_qm_dqrele(udqp); xfs_qm_dqrele(gdqp); + xfs_qm_dqrele(pdqp); *ipp = ip; return 0; @@ -656,6 +660,7 @@ xfs_create( xfs_qm_dqrele(udqp); xfs_qm_dqrele(gdqp); + xfs_qm_dqrele(pdqp); if (unlock_dp_on_error) xfs_iunlock(dp, XFS_ILOCK_EXCL); @@ -1568,7 +1573,7 @@ xfs_free_file_space( } xfs_ilock(ip, XFS_ILOCK_EXCL); error = xfs_trans_reserve_quota(tp, mp, - ip->i_udquot, ip->i_gdquot, + ip->i_udquot, ip->i_gdquot, ip->i_pdquot, resblks, 0, XFS_QMOPT_RES_REGBLKS); if (error) goto error1; -- cgit v0.10.2 From c31ad439e8d111bf911c9cc80619cebde411a44d Mon Sep 17 00:00:00 2001 From: Chandra Seetharaman Date: Wed, 10 Jul 2013 18:00:36 -0500 Subject: xfs: Fix the logic check for all quotas being turned off During the review of seperate pquota inode patches, David noticed that the test to detect all quotas being turned off was incorrect, and hence the block was not freeing all the quota information. The check made sense in Irix, but in Linux, quota is turned off one at a time, which makes the test invalid for Linux. This problem existed since XFS was ported to Linux. David suggested to fix the problem by detecting when all quotas are turned off by checking m_qflags. Signed-off-by: Chandra Seetharaman Reviewed-by: Ben Myers Signed-off-by: Ben Myers diff --git a/fs/xfs/xfs_qm_syscalls.c b/fs/xfs/xfs_qm_syscalls.c index b936383..e4f8b2d 100644 --- a/fs/xfs/xfs_qm_syscalls.c +++ b/fs/xfs/xfs_qm_syscalls.c @@ -199,10 +199,9 @@ xfs_qm_scall_quotaoff( } /* - * If quotas is completely disabled, close shop. + * If all quotas are completely turned off, close shop. */ - if (((flags & XFS_MOUNT_QUOTA_ALL) == XFS_MOUNT_QUOTA_SET1) || - ((flags & XFS_MOUNT_QUOTA_ALL) == XFS_MOUNT_QUOTA_SET2)) { + if (mp->m_qflags == 0) { mutex_unlock(&q->qi_quotaofflock); xfs_qm_destroy_quotainfo(mp); return (0); diff --git a/fs/xfs/xfs_quota.h b/fs/xfs/xfs_quota.h index 09777c4..b14f42c 100644 --- a/fs/xfs/xfs_quota.h +++ b/fs/xfs/xfs_quota.h @@ -301,14 +301,6 @@ typedef struct xfs_qoff_logformat { (XFS_IS_PQUOTA_ON(mp) && \ (mp->m_sb.sb_qflags & XFS_PQUOTA_CHKD) == 0)) -#define XFS_MOUNT_QUOTA_SET1 (XFS_UQUOTA_ACCT|XFS_UQUOTA_ENFD|\ - XFS_UQUOTA_CHKD|XFS_GQUOTA_ACCT|\ - XFS_GQUOTA_ENFD|XFS_GQUOTA_CHKD) - -#define XFS_MOUNT_QUOTA_SET2 (XFS_UQUOTA_ACCT|XFS_UQUOTA_ENFD|\ - XFS_UQUOTA_CHKD|XFS_PQUOTA_ACCT|\ - XFS_PQUOTA_ENFD|XFS_PQUOTA_CHKD) - #define XFS_MOUNT_QUOTA_ALL (XFS_UQUOTA_ACCT|XFS_UQUOTA_ENFD|\ XFS_UQUOTA_CHKD|XFS_GQUOTA_ACCT|\ XFS_GQUOTA_ENFD|XFS_GQUOTA_CHKD|\ -- cgit v0.10.2