summaryrefslogtreecommitdiffstats
path: root/fs
AgeCommit message (Collapse)Author
2011-07-08xfs: improve sync behaviour in the face of aggressive dirtyingChristoph Hellwig
The following script from Wu Fengguang shows very bad behaviour in XFS when aggressively dirtying data during a sync on XFS, with sync times up to almost 10 times as long as ext4. A large part of the issue is that XFS writes data out itself two times in the ->sync_fs method, overriding the livelock protection in the core writeback code, and another issue is the lock-less xfs_ioend_wait call, which doesn't prevent new ioend from being queue up while waiting for the count to reach zero. This patch removes the XFS-internal sync calls and relies on the VFS to do it's work just like all other filesystems do. Note that the i_iocount wait which is rather suboptimal is simply removed here. We already do it in ->write_inode, which keeps the current supoptimal behaviour. We'll eventually need to remove that as well, but that's material for a separate commit. ------------------------------ snip ------------------------------ #!/bin/sh umount /dev/sda7 mkfs.xfs -f /dev/sda7 # mkfs.ext4 /dev/sda7 # mkfs.btrfs /dev/sda7 mount /dev/sda7 /fs echo $((50<<20)) > /proc/sys/vm/dirty_bytes pid= for i in `seq 10` do dd if=/dev/zero of=/fs/zero-$i bs=1M count=1000 & pid="$pid $!" done sleep 1 tic=$(date +'%s') sync tac=$(date +'%s') echo echo sync time: $((tac-tic)) egrep '(Dirty|Writeback|NFS_Unstable)' /proc/meminfo pidof dd > /dev/null && { kill -9 $pid; echo sync NOT livelocked; } ------------------------------ snip ------------------------------ Signed-off-by: Christoph Hellwig <hch@lst.de> Reported-by: Wu Fengguang <fengguang.wu@intel.com> Reviewed-by: Alex Elder <aelder@sgi.com> Reviewed-by: Dave Chinner <dchinner@redhat.com>
2011-07-08xfs: split xfs_itruncate_finishChristoph Hellwig
Split the guts of xfs_itruncate_finish that loop over the existing extents and calls xfs_bunmapi on them into a new helper, xfs_itruncate_externs. Make xfs_attr_inactive call it directly instead of xfs_itruncate_finish, which allows to simplify the latter a lot, by only letting it deal with the data fork. As a result xfs_itruncate_finish is renamed to xfs_itruncate_data to make its use case more obvious. Also remove the sync parameter from xfs_itruncate_data, which has been unessecary since the introduction of the busy extent list in 2002, and completely dead code since 2003 when the XFS_BMAPI_ASYNC parameter was made a no-op. I can't actually see why the xfs_attr_inactive needs to set the transaction sync, but let's keep this patch simple and without changes in behaviour. Also avoid passing a useless argument to xfs_isize_check, and make it private to xfs_inode.c. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Alex Elder <aelder@sgi.com> Reviewed-by: Dave Chinner <dchinner@redhat.com>
2011-07-08xfs: kill xfs_itruncate_startChristoph Hellwig
xfs_itruncate_start is a rather length wrapper that evaluates to a call to xfs_ioend_wait and xfs_tosspages, and only has two callers. Instead of using the complicated checks left over from IRIX where we can to truncate the pagecache just call xfs_tosspages (aka truncate_inode_pages) directly as we want to get rid of all data after i_size, and truncate_inode_pages handles incorrect alignments and too large offsets just fine. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Alex Elder <aelder@sgi.com> Reviewed-by: Dave Chinner <dchinner@redhat.com>
2011-07-08xfs: always log timestamp updates in xfs_setattr_sizeChristoph Hellwig
Get rid of the special case where we use unlogged timestamp updates for a truncate to the current inode size, and just call xfs_setattr_nonsize for it to treat it like a utimes calls. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Alex Elder <aelder@sgi.com> Reviewed-by: Dave Chinner <dchinner@redhat.com>
2011-07-08xfs: split xfs_setattrChristoph Hellwig
Split up xfs_setattr into two functions, one for the complex truncate handling, and one for the trivial attribute updates. Also move both new routines to xfs_iops.c as they are fairly Linux-specific. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Alex Elder <aelder@sgi.com> Reviewed-by: Dave Chinner <dchinner@redhat.com>
2011-07-08xfs: work around bogus gcc warning in xfs_allocbt_init_cursorChristoph Hellwig
GCC 4.6 complains about an array subscript is above array bounds when using the btree index to index into the agf_levels array. The only two indices passed in are 0 and 1, and we have an assert insuring that. Replace the trick of using the array index directly with using constants in the already existing branch for assigning the XFS_BTREE_LASTREC_UPDATE flag. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Alex Elder <aelder@sgi.com> Reviewed-by: Dave Chinner <dchinner@redhat.com>
2011-07-08xfs: re-enable non-blocking behaviour in xfs_map_blocksChristoph Hellwig
The non-blockig behaviour in xfs_vm_writepage currently is conditional on having both the WB_SYNC_NONE sync_mode and the nonblocking flag set. The latter used to be used by both pdflush, kswapd and a few other places in older kernels, but has been fading out starting with the introduction of the per-bdi flusher threads. Enable the non-blocking behaviour for all WB_SYNC_NONE calls to get back the behaviour we want. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Alex Elder <aelder@sgi.com> Reviewed-by: Dave Chinner <dchinner@redhat.com>
2011-07-08xfs: PF_FSTRANS should never be set in ->writepageChristoph Hellwig
Now that we reject direct reclaim in addition to always using GFP_NOFS allocation there's no chance we'll ever end up in ->writepage with PF_FSTRANS set. Add a WARN_ON if we hit this case, and stop checking if we'd actually need to start a transaction. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Alex Elder <aelder@sgi.com> Reviewed-by: Dave Chinner <dchinner@redhat.com>
2011-07-08UBIFS: fix master node recoveryAnatolij Gustschin
When the 1st LEB was unmapped and written but 2nd LEB not, the master node recovery doesn't succeed after power cut. We see following error when mounting UBIFS partition on NOR flash: UBIFS error (pid 1137): ubifs_recover_master_node: failed to recover master node Correct 2nd master node offset check is needed to fix the problem. If the 2nd master node is at the end in the 2nd LEB, first master node is used for recovery. When checking for this condition we should check whether the master node is exactly at the end of the LEB (without remaining empty space) or whether it is followed by an empty space less than the master node size. Artem: when the error happened, offs2 = 261120, sz = 512, c->leb_size = 262016. Signed-off-by: Anatolij Gustschin <agust@denx.de> Signed-off-by: Artem Bityutskiy <dedekind1@gmail.com>
2011-07-08cifs: factor smb_vol allocation out of cifs_setup_volume_infoJeff Layton
Signed-off-by: Jeff Layton <jlayton@redhat.com> Reviewed-by: Pavel Shilovsky <piastryyy@gmail.com> Signed-off-by: Steve French <sfrench@us.ibm.com>
2011-07-07FS-Cache: Add a helper to bulk uncache pages on an inodeDavid Howells
Add an FS-Cache helper to bulk uncache pages on an inode. This will only work for the circumstance where the pages in the cache correspond 1:1 with the pages attached to an inode's page cache. This is required for CIFS and NFS: When disabling inode cookie, we were returning the cookie and setting cifsi->fscache to NULL but failed to invalidate any previously mapped pages. This resulted in "Bad page state" errors and manifested in other kind of errors when running fsstress. Fix it by uncaching mapped pages when we disable the inode cookie. This patch should fix the following oops and "Bad page state" errors seen during fsstress testing. ------------[ cut here ]------------ kernel BUG at fs/cachefiles/namei.c:201! invalid opcode: 0000 [#1] SMP Pid: 5, comm: kworker/u:0 Not tainted 2.6.38.7-30.fc15.x86_64 #1 Bochs Bochs RIP: 0010: cachefiles_walk_to_object+0x436/0x745 [cachefiles] RSP: 0018:ffff88002ce6dd00 EFLAGS: 00010282 RAX: ffff88002ef165f0 RBX: ffff88001811f500 RCX: 0000000000000000 RDX: 0000000000000000 RSI: 0000000000000100 RDI: 0000000000000282 RBP: ffff88002ce6dda0 R08: 0000000000000100 R09: ffffffff81b3a300 R10: 0000ffff00066c0a R11: 0000000000000003 R12: ffff88002ae54840 R13: ffff88002ae54840 R14: ffff880029c29c00 R15: ffff88001811f4b0 FS: 00007f394dd32720(0000) GS:ffff88002ef00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b CR2: 00007fffcb62ddf8 CR3: 000000001825f000 CR4: 00000000000006e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400 Process kworker/u:0 (pid: 5, threadinfo ffff88002ce6c000, task ffff88002ce55cc0) Stack: 0000000000000246 ffff88002ce55cc0 ffff88002ce6dd58 ffff88001815dc00 ffff8800185246c0 ffff88001811f618 ffff880029c29d18 ffff88001811f380 ffff88002ce6dd50 ffffffff814757e4 ffff88002ce6dda0 ffffffff8106ac56 Call Trace: cachefiles_lookup_object+0x78/0xd4 [cachefiles] fscache_lookup_object+0x131/0x16d [fscache] fscache_object_work_func+0x1bc/0x669 [fscache] process_one_work+0x186/0x298 worker_thread+0xda/0x15d kthread+0x84/0x8c kernel_thread_helper+0x4/0x10 RIP cachefiles_walk_to_object+0x436/0x745 [cachefiles] ---[ end trace 1d481c9af1804caa ]--- I tested the uncaching by the following means: (1) Create a big file on my NFS server (104857600 bytes). (2) Read the file into the cache with md5sum on the NFS client. Look in /proc/fs/fscache/stats: Pages : mrk=25601 unc=0 (3) Open the file for read/write ("bash 5<>/warthog/bigfile"). Look in proc again: Pages : mrk=25601 unc=25601 Reported-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: David Howells <dhowells@redhat.com> Reviewed-and-Tested-by: Suresh Jayaraman <sjayaraman@suse.de> cc: stable@kernel.org Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-07-07hfsplus: Add error propagation for hfsplus_ext_write_extent_lockedAlexey Khoroshilov
Implement error propagation through the callers of hfsplus_ext_write_extent_locked(). Signed-off-by: Alexey Khoroshilov <khoroshilov@ispras.ru> Signed-off-by: Christoph Hellwig <hch@lst.de>
2011-07-07hfsplus: add error checking for hfs_find_init()Alexey Khoroshilov
hfs_find_init() may fail with ENOMEM, but there are places, where the returned value is not checked. The consequences can be very unpleasant, e.g. kfree uninitialized pointer and inappropriate mutex unlocking. The patch adds checks for errors in hfs_find_init(). Found by Linux Driver Verification project (linuxtesting.org). Signed-off-by: Alexey Khoroshilov <khoroshilov@ispras.ru> Signed-off-by: Christoph Hellwig <hch@lst.de>
2011-07-06btrfs: fix oops when doing space balanceMiao Xie
We need to make sure the data relocation inode doesn't go through the delayed metadata updates, otherwise we get an oops during balance: kernel BUG at fs/btrfs/relocation.c:4303! [SNIP] Call Trace: [<ffffffffa03143fd>] ? update_ref_for_cow+0x22d/0x330 [btrfs] [<ffffffffa0314951>] __btrfs_cow_block+0x451/0x5e0 [btrfs] [<ffffffffa031355d>] ? read_block_for_search+0x14d/0x4d0 [btrfs] [<ffffffffa0314beb>] btrfs_cow_block+0x10b/0x240 [btrfs] [<ffffffffa031acae>] btrfs_search_slot+0x49e/0x7a0 [btrfs] [<ffffffffa032d8af>] btrfs_lookup_inode+0x2f/0xa0 [btrfs] [<ffffffff8147bf0e>] ? mutex_lock+0x1e/0x50 [<ffffffffa0380cf1>] btrfs_update_delayed_inode+0x71/0x160 [btrfs] [<ffffffffa037ff27>] ? __btrfs_release_delayed_node+0x67/0x190 [btrfs] [<ffffffffa0381cf8>] btrfs_run_delayed_items+0xe8/0x120 [btrfs] [<ffffffffa03365e0>] btrfs_commit_transaction+0x250/0x850 [btrfs] [<ffffffff810f91d9>] ? find_get_pages+0x39/0x130 [<ffffffffa0336cd5>] ? join_transaction+0x25/0x250 [btrfs] [<ffffffff81081de0>] ? wake_up_bit+0x40/0x40 [<ffffffffa03785fa>] prepare_to_relocate+0xda/0xf0 [btrfs] [<ffffffffa037f2bb>] relocate_block_group+0x4b/0x620 [btrfs] [<ffffffffa0334cf5>] ? btrfs_clean_old_snapshots+0x35/0x150 [btrfs] [<ffffffffa037fa43>] btrfs_relocate_block_group+0x1b3/0x2e0 [btrfs] [<ffffffffa0368ec0>] ? btrfs_tree_unlock+0x50/0x50 [btrfs] [<ffffffffa035e39b>] btrfs_relocate_chunk+0x8b/0x670 [btrfs] [<ffffffffa031303d>] ? btrfs_set_path_blocking+0x3d/0x50 [btrfs] [<ffffffffa03577d8>] ? read_extent_buffer+0xd8/0x1d0 [btrfs] [<ffffffffa031bea1>] ? btrfs_previous_item+0xb1/0x150 [btrfs] [<ffffffffa03577d8>] ? read_extent_buffer+0xd8/0x1d0 [btrfs] [<ffffffffa035f5aa>] btrfs_balance+0x21a/0x2b0 [btrfs] [<ffffffffa0368898>] btrfs_ioctl+0x798/0xd20 [btrfs] [<ffffffff8111e358>] ? handle_mm_fault+0x148/0x270 [<ffffffff814809e8>] ? do_page_fault+0x1d8/0x4b0 [<ffffffff81160d6a>] do_vfs_ioctl+0x9a/0x540 [<ffffffff811612b1>] sys_ioctl+0xa1/0xb0 [<ffffffff81484ec2>] system_call_fastpath+0x16/0x1b [SNIP] RIP [<ffffffffa037c1cc>] btrfs_reloc_cow_block+0x22c/0x270 [btrfs] Signed-off-by: Miao Xie <miaox@cn.fujitsu.com> Signed-off-by: Chris Mason <chris.mason@oracle.com>
2011-07-06Btrfs: don't panic if we get an error while balancing V2Josef Bacik
A user reported an error where if we try to balance an fs after a device has been removed it will blow up. This is because we get an EIO back and this is where BUG_ON(ret) bites us in the ass. To fix we just exit. Thanks, Reported-by: Anand Jain <Anand.Jain@oracle.com> Signed-off-by: Josef Bacik <josef@redhat.com> Signed-off-by: Chris Mason <chris.mason@oracle.com>
2011-07-06btrfs: add missing options displayed in mount outputDavid Sterba
There are three missed mount options settable by user which are not currently displayed in mount output. Signed-off-by: David Sterba <dsterba@suse.cz> Signed-off-by: Chris Mason <chris.mason@oracle.com>
2011-07-06dlm: dump address of unknown nodeMasatake YAMATO
When the dlm fails to make a network connection to another node, include the address of the node in the error message. Signed-off-by: Masatake YAMATO <yamato@redhat.com> Signed-off-by: David Teigland <teigland@redhat.com>
2011-07-06xfs: unpin stale inodes directly in IOP_COMMITTEDDave Chinner
When inodes are marked stale in a transaction, they are treated specially when the inode log item is being inserted into the AIL. It tries to avoid moving the log item forward in the AIL due to a race condition with the writing the underlying buffer back to disk. The was "fixed" in commit de25c18 ("xfs: avoid moving stale inodes in the AIL"). To avoid moving the item forward, we return a LSN smaller than the commit_lsn of the completing transaction, thereby trying to trick the commit code into not moving the inode forward at all. I'm not sure this ever worked as intended - it assumes the inode is already in the AIL, but I don't think the returned LSN would have been small enough to prevent moving the inode. It appears that the reason it worked is that the lower LSN of the inodes meant they were inserted into the AIL and flushed before the inode buffer (which was moved to the commit_lsn of the transaction). The big problem is that with delayed logging, the returning of the different LSN means insertion takes the slow, non-bulk path. Worse yet is that insertion is to a position -before- the commit_lsn so it is doing a AIL traversal on every insertion, and has to walk over all the items that have already been inserted into the AIL. It's expensive. To compound the matter further, with delayed logging inodes are likely to go from clean to stale in a single checkpoint, which means they aren't even in the AIL at all when we come across them at AIL insertion time. Hence these were all getting inserted into the AIL when they simply do not need to be as inodes marked XFS_ISTALE are never written back. Transactional/recovery integrity is maintained in this case by the other items in the unlink transaction that were modified (e.g. the AGI btree blocks) and committed in the same checkpoint. So to fix this, simply unpin the stale inodes directly in xfs_inode_item_committed() and return -1 to indicate that the AIL insertion code does not need to do any further processing of these inodes. Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Alex Elder <aelder@sgi.com>
2011-07-06cifs: have cifs_cleanup_volume_info not take a double pointerJeff Layton
...as that makes for a cumbersome interface. Make it take a regular smb_vol pointer and rely on the caller to zero it out if needed. Signed-off-by: Jeff Layton <jlayton@redhat.com> Reviewed-by: Pavel Shilovsky <piastryyy@gmail.com> Signed-off-by: Steve French <sfrench@us.ibm.com>
2011-07-06cifs: fix build_unc_path_to_root to account for a prefixpathJeff Layton
Regression introduced by commit f87d39d9513. Signed-off-by: Jeff Layton <jlayton@redhat.com> Reviewed-by: Pavel Shilovsky <piastryyy@gmail.com> Signed-off-by: Steve French <sfrench@us.ibm.com>
2011-07-06cifs: remove bogus call to cifs_cleanup_volume_infoJeff Layton
This call to cifs_cleanup_volume_info is clearly wrong. As soon as it's called the following call to cifs_get_tcp_session will oops as the volume_info pointer will then be NULL. The caller of cifs_mount should clean up this data since it passed it in. There's no need for us to call this here. Regression introduced by commit 724d9f1cfba. Reported-by: Adam Williamson <awilliam@redhat.com> Cc: Pavel Shilovsky <piastryyy@gmail.com> Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Steve French <sfrench@us.ibm.com>
2011-07-06FDPIC: Fix memory leakDavidlohr Bueso
The shdr4extnum variable isn't being freed in the cleanup process of elf_fdpic_core_dump(). Signed-off-by: Davidlohr Bueso <dave@gnu.org> Signed-off-by: David Howells <dhowells@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-07-06fs: fix lock initializationMiklos Szeredi
locks_alloc_lock() assumed that the allocated struct file_lock is already initialized to zero members. This is only true for the first allocation of the structure, after reuse some of the members will have random values. This will for example result in passing random fl_start values to userspace in fuse for FL_FLOCK locks, which is an information leak at best. Fix by reinitializing those members which may be non-zero after freeing. Signed-off-by: Miklos Szeredi <mszeredi@suse.cz> CC: stable@kernel.org Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-07-05Merge branch 'for-linus' of ↵Linus Torvalds
git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client * 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client: ceph: fix sync and dio writes across stripe boundaries libceph: fix page calculation for non-page-aligned io ceph: fix page alignment corrections
2011-07-05Merge branch 'for-linus' of ↵Linus Torvalds
git://git.kernel.org/pub/scm/linux/kernel/git/hch/hfsplus * 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/hch/hfsplus: hfsplus: Fix double iput of the same inode in hfsplus_fill_super() hfsplus: add missing call to bio_put()
2011-07-04UBIFS: improve power cut emulation testingArtem Bityutskiy
This patch cleans-up and improves the power cut testing: 1. Kill custom 'simple_random()' function and use 'random32()' instead. 2. Make timeout larger 3. When cutting the buffer - fill the end with random data sometimes, not only with 0xFFs. 4. Some times cut in the middle of the buffer, not always at the end. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
2011-07-04UBIFS: rename recovery testing variablesArtem Bityutskiy
Since the recovery testing is effectively about emulating power cuts by UBIFS, use "power cut" as the base term for all the related variables and name them correspondingly. This is just a minor clean-up for the sake of readability. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
2011-07-04UBIFS: remove custom list of superblocksArtem Bityutskiy
This is a clean-up of the power-cut emulation code - remove the custom list of superblocks which we maintained to find the superblock by the UBI volume descriptor. We do not need that crud any longer, because now we can get the superblock as a function argument. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
2011-07-04UBIFS: stop re-defining UBI operationsArtem Bityutskiy
Now when we use UBIFS helpers for all the I/O, we can remove the horrible hack of re-defining UBI I/O functions. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
2011-07-04UBIFS: switch to I/O helpersArtem Bityutskiy
Switch the rest of direct UBI calls to UBIFS helper functions. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
2011-07-04UBIFS: switch to ubifs_leb_writeArtem Bityutskiy
Stop using 'ubi_leb_write()' directly and switch to the 'ubifs_leb_write()' helper. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
2011-07-04UBIFS: switch to ubifs_leb_readArtem Bityutskiy
Instead of using 'ubi_read()' function directly, used the 'ubifs_leb_read()' helper function instead. This allows to get rid of several redundant error messages and make sure that we always have a stack dump on read errors. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
2011-07-04UBIFS: introduce more I/O helpersArtem Bityutskiy
Introduce the following I/O helper functions: 'ubifs_leb_read()', 'ubifs_leb_write()', 'ubifs_leb_change()', 'ubifs_leb_unmap()', 'ubifs_leb_map()', 'ubifs_is_mapped(). The idea is to wrap all UBI I/O functions in order to encapsulate various assertions and error path handling (error message, stack dump, switching to R/O mode). And there are some other benefits of this which will be used in the following patches. This patch does not switch whole UBIFS to use these functions yet. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
2011-07-04UBIFS: always print stacktrace when switching to R/O modeArtem Bityutskiy
When switching to R/O mode due to an I/O error, always dump the stack, not only when debugging is enabled. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
2011-07-04UBIFS: remove unused and unneeded debugging functionArtem Bityutskiy
This patch contains several minor clean-up and preparational cahnges. 1. Remove 'dbg_read()', 'dbg_write()', 'dbg_change()', and 'dbg_leb_erase()' functions as they are not used. 2. Remove 'dbg_leb_read()' and 'dbg_is_mapped()' as they are not really needed, it is fine to let reads go through in failure mode. 3. Rename 'offset' argument to 'offs' to be consistent with the rest of UBIFS code. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
2011-07-04UBIFS: add global debugfs knobsArtem Bityutskiy
Now we have per-FS (superblock) debugfs knobs, but they have one drawback - you have to first mount the FS and only after this you can switch self-checks on/off. But often we want to have the checks enabled during the mount. Introduce global debugging knobs for this purpose. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
2011-07-04UBIFS: introduce debugfs helpersArtem Bityutskiy
Separate out pieces of code from the debugfs file read/write functions and create separate 'interpret_user_input()'/'provide_user_output()' helpers. These helpers will be needed in one of the following patches, so this is just a preparational change. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
2011-07-04UBIFS: re-arrange debugging code a bitArtem Bityutskiy
Move 'dbg_debugfs_init()' and 'dbg_debugfs_exit()' functions which initialize debugfs for whole UBIFS subsystem below the code which initializes debugfs for a particular UBIFS instance. And do the same for 'ubifs_debugging_init()' and 'ubifs_debugging_exit()' functions. This layout is a bit better for the next patches, so this is just a preparation. Also, rename 'open_debugfs_file()' into 'dfs_file_open()' for consistency. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
2011-07-04UBIFS: be more informative in failure modeArtem Bityutskiy
When we are testing UBIFS recovery, it is better to print in which eraseblock we are going to fail. Currently UBIFS prints it only if recovery debugging messages are enabled, but this is not very practical. So change 'dbg_rcvry()' messages to 'ubifs_warn()' messages. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
2011-07-04UBIFS: switch self-check knobs to debugfsArtem Bityutskiy
UBIFS has many built-in self-check functions which can be enabled using the debug_chks module parameter or the corresponding sysfs file (/sys/module/ubifs/parameters/debug_chks). However, this is not flexible enough because it is not per-filesystem. This patch moves this to debugfs interfaces. We already have debugfs support, so this patch just adds more debugfs files. While looking at debugfs support I've noticed that it is racy WRT file-system unmount, and added a TODO entry for that. This problem has been there for long time and it is quite standard debugfs PITA. The plan is to fix this later. This patch is simple, but it is large because it changes many places where we check if a particular type of checks is enabled or disabled. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
2011-07-04UBIFS: lessen amount of debugging check typesArtem Bityutskiy
We have too many different debugging checks - lessen the amount by merging all index-related checks into one. At the same time, move the "force in-the-gap" test to the "index checks" class, because it is too heavy for the "general" class. This patch merges TNC, Old index, and Index size check and calles this just "index checks". Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
2011-07-04UBIFS: introduce helper functions for debugging checks and testsArtem Bityutskiy
This patch introduces helper functions for all debugging checks, so instead of doing if (!(ubifs_chk_flags & UBIFS_CHK_GEN)) we now do if (!dbg_is_chk_gen(c)) This is a preparation to further changes where the flags will go away, and we'll need to only change the helper functions, but the code which utilizes them won't be touched. At the same time this patch removes 'dbg_force_in_the_gaps()', 'dbg_force_in_the_gaps_enabled()', and dbg_failure_mode helpers for consistency. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
2011-07-04UBIFS: amend debugging inode size check function prototypeArtem Bityutskiy
Add 'const struct ubifs_info *c' parameter to 'dbg_check_synced_i_size()' function because we'll need it in the next patch when we switch to debugfs. So this patch is just a preparation. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
2011-07-04UBIFS: amend debugging name check function prototypeArtem Bityutskiy
Add 'struct ubifs_info *c' parameter to the 'dbg_check_name()' debugging function - it will be needed in one of the following commits where we switch to debugfs. So this is just a preparation. Mark parameters as 'const' while on it. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
2011-07-04UBIFS: add few commentaries about TNCArtem Bityutskiy
Add a couple of comments - while looking into TNC I could not easily figure out few facts, so it is a good idea to document them in the code. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
2011-07-04UBIFS: use correct flags in lpropsArtem Bityutskiy
The UBIFS lpt tree is in many aspects similar to the TNC tree, and we have similar flags for these trees. And by mistake we use the COW_ZNODE flag for LPT in some places, instead of the right flag COW_CNODE. And this works only because these two constants have the same value. This patch makes all the LPT code to use COW_CNODE and also changes COW_CNODE constant value to make sure we do not misuse the flags any more. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
2011-07-04UBIFS: harmonize znode flag helpersArtem Bityutskiy
We have 3 znode flags: cow, obsolete, dirty. For the last flag we have a 'ubifs_zn_dirty()' helper function, but for the other 2 flags we use 'test_bit()' directly. This patch makes the situation more consistent and introduces helpers for the other 2 flags: 'ubifs_zn_cow()' and 'ubifs_zn_obsolete()'. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
2011-07-04UBIFS: remove dead codeArtem Bityutskiy
Remove dead pieces of code under "if (c->min_io_size == 1)" statement - we never execute it because in UBIFS 'c->min_io_size' is always at least 8. This are leftovers from old pre-mainline prototype. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
2011-07-04UBIFS: remove unnecessary bracketsArtem Bityutskiy
Remove unnecessary brackets in "inode->i_flags |= (S_NOCMTIME)" statement to make the code not look silly. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
2011-07-04UBIFS: minor cleanup: use S_ISREG helperArtem Bityutskiy
Instead of using long "(inode->i_mode & S_IFMT) != S_IFREG" expression, use shorted "!S_ISREG(inode->i_mode)". Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>