summaryrefslogtreecommitdiffstats
path: root/fs
Commit message (Collapse)AuthorAgeFilesLines
* Merge branch 'for-linus' of ↵Linus Torvalds2009-09-302-0/+2
|\ | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/ryusuke/nilfs2 * 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ryusuke/nilfs2: nilfs2: fix missing initialization of i_dir_start_lookup member nilfs2: fix missing zero-fill initialization of btree node cache
| * nilfs2: fix missing initialization of i_dir_start_lookup memberRyusuke Konishi2009-09-291-0/+1
| | | | | | | | | | | | | | | | | | | | | | The i_dir_start_lookup field in nilfs_inode_info objects should be cleared when the objects are allocated, but the the initialization was missing in case of reading from disk. This adds the initialization. Since the variable just gives a start page on directory lookups, the bug was nonfatal until now. Signed-off-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp>
| * nilfs2: fix missing zero-fill initialization of btree node cacheRyusuke Konishi2009-09-291-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This will fix file system corruption which infrequently happens after mount. The problem was reported from users with the title "[NILFS users] Fail to mount NILFS." (Message-ID: <200908211918.34720.yuri@itinteg.net>), and so forth. I've also experienced the corruption multiple times on kernel 2.6.30 and 2.6.31. The problem turned out to be caused due to discordance between mapping->nrpages of a btree node cache and the actual number of pages hung on the cache; if the mapping->nrpages becomes zero even as it has pages, truncate_inode_pages() returns without doing anything. Usually this is harmless except it may cause page leak, but garbage collection fairly infrequently sees a stale page remained in the btree node cache of DAT (i.e. disk address translation file of nilfs), and induces the corruption. I identified a missing initialization in btree node caches was the root cause. This corrects the bug. I've tested this for kernel 2.6.30 and 2.6.31. Reported-by: Yuri Chislov <yuri@itinteg.net> Signed-off-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp> Cc: stable <stable@kernel.org>
* | Merge branch 'for_linus' of ↵Linus Torvalds2009-09-3015-723/+1093
|\ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4 * 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4: ext4: Fix time encoding with extra epoch bits ext4: Add a stub for mpage_da_data in the trace header jbd2: Use tracepoints for history file ext4: Use tracepoints for mb_history trace file ext4, jbd2: Drop unneeded printks at mount and unmount time ext4: Handle nested ext4_journal_start/stop calls without a journal ext4: Make sure ext4_dirty_inode() updates the inode in no journal mode ext4: Avoid updating the inode table bh twice in no journal mode ext4: EXT4_IOC_MOVE_EXT: Check for different original and donor inodes first ext4: async direct IO for holes and fallocate support ext4: Use end_io callback to avoid direct I/O fallback to buffered I/O ext4: Split uninitialized extents for direct I/O ext4: release reserved quota when block reservation for delalloc retry ext4: Adjust ext4_da_writepages() to write out larger contiguous chunks ext4: Fix hueristic which avoids group preallocation for closed files ext4: Use ext4_msg() for ext4_da_writepage() errors ext4: Update documentation about quota mount options
| * | ext4: Fix time encoding with extra epoch bitsTheodore Ts'o2009-09-301-3/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | "Looking at ext4.h, I think the setting of extra time fields forgets to mask the epoch bits so the epoch part overwrites nsec part. The second change is only for coherency (2 -> EXT4_EPOCH_BITS)." Thanks to Damien Guibouret for pointing out this problem. Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
| * | jbd2: Use tracepoints for history fileTheodore Ts'o2009-09-303-208/+45
| | | | | | | | | | | | | | | | | | | | | | | | | | | The /proc/fs/jbd2/<dev>/history was maintained manually; by using tracepoints, we can get all of the existing functionality of the /proc file plus extra capabilities thanks to the ftrace infrastructure. We save memory as a bonus. Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
| * | ext4: Use tracepoints for mb_history trace fileTheodore Ts'o2009-09-304-347/+19
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The /proc/fs/ext4/<dev>/mb_history was maintained manually, and had a number of problems: it required a largish amount of memory to be allocated for each ext4 filesystem, and the s_mb_history_lock introduced a CPU contention problem. By ripping out the mb_history code and replacing it with ftrace tracepoints, and we get more functionality: timestamps, event filtering, the ability to correlate mballoc history with other ext4 tracepoints, etc. Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
| * | ext4, jbd2: Drop unneeded printks at mount and unmount timeTheodore Ts'o2009-09-295-22/+10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | There are a number of kernel printk's which are printed when an ext4 filesystem is mounted and unmounted. Disable them to economize space in the system logs. In addition, disabling the mballoc stats by default saves a number of unneeded atomic operations for every block allocation or deallocation. Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
| * | ext4: Handle nested ext4_journal_start/stop calls without a journalCurt Wohlgemuth2009-09-293-13/+38
| | | | | | | | | | | | | | | | | | | | | | | | This patch fixes a problem with handling nested calls to ext4_journal_start/ext4_journal_stop, when there is no journal present. Signed-off-by: Curt Wohlgemuth <curtw@google.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
| * | ext4: Make sure ext4_dirty_inode() updates the inode in no journal modeCurt Wohlgemuth2009-09-291-15/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch a problem that ext4_dirty_inode() was not calling ext4_mark_inode_dirty() if the current_handle is not valid, which it is the case in no journal mode. It also removes a test for non-matching transaction which can never happen. Signed-off-by: Curt Wohlgemuth <curtw@google.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
| * | ext4: Avoid updating the inode table bh twice in no journal modeFrank Mayhar2009-09-291-21/+16
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This is a cleanup of commit 91ac6f4. Since ext4_mark_inode_dirty() has already called ext4_mark_iloc_dirty(), which in turn calls ext4_do_update_inode(), it's not necessary to have ext4_write_inode() call ext4_do_update_inode() in no journal mode. Indeed, it would be duplicated work. Reviewed-by: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: Frank Mayhar <fmayhar@google.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
| * | ext4: EXT4_IOC_MOVE_EXT: Check for different original and donor inodes firstTheodore Ts'o2009-09-281-8/+8
| | | | | | | | | | | | | | | | | | | | | | | | Move the check to make sure the original and donor inodes are different earlier, to avoid a potential deadlock by trying to lock the same inode twice. Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
| * | ext4: async direct IO for holes and fallocate supportMingming Cao2009-09-285-41/+234
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | For async direct IO that covers holes or fallocate, the end_io callback function now queued the convertion work on workqueue but don't flush the work rightaway as it might take too long to afford. But when fsync is called after all the data is completed, user expects the metadata also being updated before fsync returns. Thus we need to flush the conversion work when fsync() is called. This patch keep track of a listed of completed async direct io that has a work queued on workqueue. When fsync() is called, it will go through the list and do the conversion. Signed-off-by: Mingming Cao <cmm@us.ibm.com>
| * | ext4: Use end_io callback to avoid direct I/O fallback to buffered I/OMingming Cao2009-09-283-1/+210
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Currently the DIO VFS code passes create = 0 when writing to the middle of file. It does this to avoid block allocation for holes, so as not to expose stale data out when there is a parallel buffered read (which does not hold the i_mutex lock). Direct I/O writes into holes falls back to buffered IO for this reason. Since preallocated extents are treated as holes when doing a get_block() look up (buffer is not mapped), direct IO over fallocate also falls back to buffered IO. Thus ext4 actually silently falls back to buffered IO in above two cases, which is undesirable. To fix this, this patch creates unitialized extents when a direct I/O write into holes in sparse files, and registering an end_io callback which converts the uninitialized extent to an initialized extent after the I/O is completed. Singed-Off-By: Mingming Cao <cmm@us.ibm.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
| * | ext4: Split uninitialized extents for direct I/OMingming Cao2009-09-286-42/+419
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When writing into an unitialized extent via direct I/O, and the direct I/O doesn't exactly cover the unitialized extent, split the extent into uninitialized and initialized extents before submitting the I/O. This avoids needing to deal with an ENOSPC error in the end_io callback that gets used for direct I/O. When the IO is complete, the written extent will be marked as initialized. Singed-Off-By: Mingming Cao <cmm@us.ibm.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
| * | ext4: release reserved quota when block reservation for delalloc retryMingming Cao2009-09-281-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | ext4_da_reserve_space() can reserve quota blocks multiple times if ext4_claim_free_blocks() fail and we retry the allocation. We should release the quota reservation before restarting. Bug found by Jan Kara. Signed-off-by: Mingming Cao <cmm@us.ibm.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
| * | ext4: Adjust ext4_da_writepages() to write out larger contiguous chunksTheodore Ts'o2009-09-293-12/+97
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Work around problems in the writeback code to force out writebacks in larger chunks than just 4mb, which is just too small. This also works around limitations in the ext4 block allocator, which can't allocate more than 2048 blocks at a time. So we need to defeat the round-robin characteristics of the writeback code and try to write out as many blocks in one inode before allowing the writeback code to move on to another inode. We add a a new per-filesystem tunable, max_writeback_mb_bump, which caps this to a default of 128mb per inode. Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
| * | ext4: Fix hueristic which avoids group preallocation for closed filesTheodore Ts'o2009-09-281-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | The hueristic was designed to avoid using locality group preallocation when writing the last segment of a closed file. Fix it by move setting size to the maximum of size and isize until after we check whether size == isize. Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
| * | ext4: Use ext4_msg() for ext4_da_writepage() errorsTheodore Ts'o2009-09-261-25/+25
| | | | | | | | | | | | | | | | | | | | | | | | This allows the user to see what filesystem was involved with a particular ext4_da_writepage() error. Also, use KERN_CRIT which is more appropriate than KERN_EMERG. Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
* | | Merge git://git.kernel.org/pub/scm/linux/kernel/git/hirofumi/fatfs-2.6Linus Torvalds2009-09-305-26/+25
|\ \ \ | |_|/ |/| | | | | | | | | | | | | | * git://git.kernel.org/pub/scm/linux/kernel/git/hirofumi/fatfs-2.6: fat: Check s_dirt in fat_sync_fs() vfat: change the default from shortname=lower to shortname=mixed fat/nls: Fix handling of utf8 invalid char
| * | fat: Check s_dirt in fat_sync_fs()OGAWA Hirofumi2009-09-203-9/+15
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | If we didn't check sb->s_dirt, it will update the FSINFO unconditionally. It will reduce the filetime of flash base device. So, this checks sb->s_dirt. sb->s_dirt is racy, however FSINFO is just hint. So even if there is race, and we hit it, it would not become big problem. And this also is as workaround of suspend problem. Signed-off-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
| * | vfat: change the default from shortname=lower to shortname=mixedPaul Wise2009-08-011-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Because, with "shortname=lower", copying one FAT filesystem tree to another FAT filesystem tree using Linux results in semantically different filesystems. (E.g.: Filenames which were once "all uppercase" are now "all lowercase"). So, this changes the default of "shortname=lower" to "shortname=mixed". Signed-off-by: Paul Wise <pabs3@bonedaddy.net> [change fat_show_options()] Signed-off-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
| * | fat/nls: Fix handling of utf8 invalid charOGAWA Hirofumi2009-08-012-15/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | With utf8 option, vfat allowed the duplicated filenames. Normal nls returns -EINVAL for invalid char. But utf8s_to_utf16s() skipped the invalid char historically. So, this changes the utf8s_to_utf16s() directly to return -EINVAL for invalid char, because vfat is only user of it. mkdir /mnt/fatfs FILENAME=`echo -ne "invalidutf8char_\\0341_endofchar"` echo "Using filename: $FILENAME" dd if=/dev/zero of=fatfs bs=512 count=128 mkdosfs -F 32 fatfs mount -o loop,utf8 fatfs /mnt/fatfs touch "/mnt/fatfs/$FILENAME" umount /mnt/fatfs mount -o loop,utf8 fatfs /mnt/fatfs touch "/mnt/fatfs/$FILENAME" ls -l /mnt/fatfs umount /mnt/fatfs ---- And the output is: Using filename: invalidutf8char_\0341_endofchar 128+0 records in 128+0 records out 65536 bytes (66 kB) copied, 0.000388118 s, 169 MB/s mkdosfs 2.11 (12 Mar 2005) total 0 -rwxr-xr-x 1 root root 0 Jun 28 19:46 invalidutf8char__endofchar -rwxr-xr-x 1 root root 0 Jun 28 19:46 invalidutf8char__endofchar Tested-by: Marton Balint <cus@fazekas.hu> Signed-off-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
* | | const: mark struct vm_struct_operationsAlexey Dobriyan2009-09-2711-14/+14
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * mark struct vm_area_struct::vm_ops as const * mark vm_ops in AGP code But leave TTM code alone, something is fishy there with global vm_ops being used. Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* | | Merge branch 'writeback' of git://git.kernel.dk/linux-2.6-blockLinus Torvalds2009-09-261-2/+4
|\ \ \ | | | | | | | | | | | | | | | | * 'writeback' of git://git.kernel.dk/linux-2.6-block: writeback: pass in super_block to bdi_start_writeback()
| * | | writeback: pass in super_block to bdi_start_writeback()Jens Axboe2009-09-261-2/+4
| | |/ | |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | Sometimes we only want to write pages from a specific super_block, so allow that to be passed in. This fixes a problem with commit 56a131dcf7ed36c3c6e36bea448b674ea85ed5bb causing writeback on all super_blocks on a bdi, where we only really want to sync a specific sb from writeback_inodes_sb(). Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
* | | Merge git://git.kernel.org/pub/scm/linux/kernel/git/sfrench/cifs-2.6Linus Torvalds2009-09-2611-258/+159
|\ \ \ | |/ / |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * git://git.kernel.org/pub/scm/linux/kernel/git/sfrench/cifs-2.6: cifs: fix locking and list handling code in cifs_open and its helper [CIFS] Remove build warning cifs: fix problems with last two commits [CIFS] Fix build break when keys support turned off cifs: eliminate cifs_init_private cifs: convert oplock breaks to use slow_work facility (try #4) cifs: have cifsFileInfo hold an extra inode reference cifs: take read lock on GlobalSMBSes_lock in is_valid_oplock_break cifs: remove cifsInodeInfo.oplockPending flag cifs: fix oplock request handling in posix codepath [CIFS] Re-enable Lanman security
| * | cifs: fix locking and list handling code in cifs_open and its helperJeff Layton2009-09-251-19/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The patch to remove cifs_init_private introduced a locking imbalance. It didn't remove the leftover list addition code and the unlocking in that function. cifs_new_fileinfo does the list addition now, so there should be no need to do it outside of that function. pCifsInode will never be NULL, so we don't need to check for that. This patch also gets rid of the ugly locking and unlocking across function calls. Signed-off-by: Jeff Layton <jlayton@redhat.com> Acked-by: Steve French <sfrench@us.ibm.com> Signed-off-by: Steve French <sfrench@us.ibm.com>
| * | [CIFS] Remove build warningSteve French2009-09-251-2/+2
| | | | | | | | | | | | | | | Acked-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Steve French <sfrench@us.ibm.com>
| * | cifs: fix problems with last two commitsJeff Layton2009-09-251-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Fix problems with commits: 086f68bd97126618ecb2dcff5f766f3a21722df7 3bc303c254335dbd7c7012cc1760b12f1d5514d3 Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Steve French <sfrench@us.ibm.com>
| * | [CIFS] Fix build break when keys support turned offSteve French2009-09-251-1/+1
| | | | | | | | | | | | | | | Acked-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Steve French <sfrench@us.ibm.com>
| * | cifs: eliminate cifs_init_privateJeff Layton2009-09-243-50/+20
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | ...it does the same thing as cifs_fill_fileinfo, but doesn't handle the flist ordering correctly. Also rename cifs_fill_fileinfo to a more descriptive name and have it take an open flags arg instead of just a write_only flag. That makes the logic in the callers a little simpler. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Steve French <sfrench@us.ibm.com>
| * | cifs: convert oplock breaks to use slow_work facility (try #4)Jeff Layton2009-09-2410-175/+119
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This is the fourth respin of the patch to convert oplock breaks to use the slow_work facility. A customer of ours was testing a backport of one of the earlier patchsets, and hit a "Busy inodes after umount..." problem. An oplock break job had raced with a umount, and the superblock got torn down and its memory reused. When the oplock break job tried to dereference the inode->i_sb, the kernel oopsed. This patchset has the oplock break job hold an inode and vfsmount reference until the oplock break completes. With this, there should be no need to take a tcon reference (the vfsmount implicitly holds one already). Currently, when an oplock break comes in there's a chance that the oplock break job won't occur if the allocation of the oplock_q_entry fails. There are also some rather nasty races in the allocation and handling these structs. Rather than allocating oplock queue entries when an oplock break comes in, add a few extra fields to the cifsFileInfo struct. Get rid of the dedicated cifs_oplock_thread as well and queue the oplock break job to the slow_work thread pool. This approach also has the advantage that the oplock break jobs can potentially run in parallel rather than be serialized like they are today. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Steve French <sfrench@us.ibm.com>
| * | cifs: have cifsFileInfo hold an extra inode referenceJeff Layton2009-09-153-3/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | It's possible that this struct will outlive the filp to which it is attached. If it does and it needs to do some work on the inode, then it'll need a reference. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Steve French <sfrench@us.ibm.com>
| * | cifs: take read lock on GlobalSMBSes_lock in is_valid_oplock_breakJeff Layton2009-09-151-3/+3
| | | | | | | | | | | | | | | | | | | | | | | | ...rather than a write lock. It doesn't change the list so a read lock should be sufficient. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Steve French <sfrench@us.ibm.com>
| * | cifs: remove cifsInodeInfo.oplockPending flagJeff Layton2009-09-152-2/+0
| | | | | | | | | | | | | | | | | | | | | It's set on oplock break but nothing ever looks at it. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Steve French <sfrench@us.ibm.com>
| * | cifs: fix oplock request handling in posix codepathJeff Layton2009-09-153-11/+14
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | cifs_posix_open takes a "poplock" argument that's intended to be used in the actual posix open call to set the "Flags" field. It ignores this value however and declares an "oplock" parameter on the stack that it passes uninitialized to the CIFSPOSIXOpen function. Not only does this mean that the oplock request flags are bogus, but the result that's expected to be in that variable is unchanged. Fix this, and also clean up the type of the oplock parameter used. Since it's expected to be __u32, we should use that everywhere and not implicitly cast it from a signed type. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Steve French <sfrench@us.ibm.com>
| * | [CIFS] Re-enable Lanman securityChuck Ebbert2009-09-151-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | commit ac68392460ffefed13020967bae04edc4d3add06 ("[CIFS] Allow raw ntlmssp code to be enabled with sec=ntlmssp") added a new bit to the allowed security flags mask but seems to have inadvertently removed Lanman security from the allowed flags. Add it back. CC: Stable <stable@kernel.org> Signed-off-by: Chuck Ebbert <cebbert@redhat.com> Signed-off-by: Steve French <sfrench@us.ibm.com>
* | | Merge branch 'writeback' of git://git.kernel.dk/linux-2.6-blockLinus Torvalds2009-09-252-53/+118
|\ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * 'writeback' of git://git.kernel.dk/linux-2.6-block: writeback: writeback_inodes_sb() should use bdi_start_writeback() writeback: don't delay inodes redirtied by a fast dirtier writeback: make the super_block pinning more efficient writeback: don't resort for a single super_block in move_expired_inodes() writeback: move inodes from one super_block together writeback: get rid to incorrect references to pdflush in comments writeback: improve readability of the wb_writeback() continue/break logic writeback: cleanup writeback_single_inode() writeback: kupdate writeback shall not stop when more io is possible writeback: stop background writeback when below background threshold writeback: balance_dirty_pages() shall write more than dirtied pages fs: Fix busyloop in wb_writeback()
| * | | writeback: writeback_inodes_sb() should use bdi_start_writeback()Jens Axboe2009-09-251-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Pointless to iterate other devices looking for a super, when we have a bdi mapping. Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
| * | | writeback: don't delay inodes redirtied by a fast dirtierWu Fengguang2009-09-251-3/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Debug traces show that in per-bdi writeback, the inode under writeback almost always get redirtied by a busy dirtier. We used to call redirty_tail() in this case, which could delay inode for up to 30s. This is unacceptable because it now happens so frequently for plain cp/dd, that the accumulated delays could make writeback of big files very slow. So let's distinguish between data redirty and metadata only redirty. The first one is caused by a busy dirtier, while the latter one could happen in XFS, NFS, etc. when they are doing delalloc or updating isize. The inode being busy dirtied will now be requeued for next io, while the inode being redirtied by fs will continue to be delayed to avoid repeated IO. CC: Jan Kara <jack@suse.cz> CC: Theodore Ts'o <tytso@mit.edu> CC: Dave Chinner <david@fromorbit.com> CC: Chris Mason <chris.mason@oracle.com> CC: Christoph Hellwig <hch@infradead.org> Signed-off-by: Wu Fengguang <fengguang.wu@intel.com> Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
| * | | writeback: make the super_block pinning more efficientJens Axboe2009-09-251-17/+29
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Currently we pin the inode->i_sb for every single inode. This increases cache traffic on sb->s_umount sem. Lets instead cache the inode sb pin state and keep the super_block pinned for as long as keep writing out inodes from the same super_block. Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
| * | | writeback: don't resort for a single super_block in move_expired_inodes()Jens Axboe2009-09-251-1/+11
| | | | | | | | | | | | | | | | | | | | | | | | | | | | If we only moved inodes from a single super_block to the temporary list, there's no point in doing a resort for multiple super_blocks. Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
| * | | writeback: move inodes from one super_block togetherShaohua Li2009-09-251-3/+18
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | __mark_inode_dirty adds inode to wb dirty list in random order. If a disk has several partitions, writeback might keep spindle moving between partitions. To reduce the move, better write big chunk of one partition and then move to another. Inodes from one fs usually are in one partion, so idealy move indoes from one fs together should reduce spindle move. This patch tries to address this. Before per-bdi writeback is added, the behavior is write indoes from one fs first and then another, so the patch restores previous behavior. The loop in the patch is a bit ugly, should we add a dirty list for each superblock in bdi_writeback? Test in a two partition disk with attached fio script shows about 3% ~ 6% improvement. Signed-off-by: Shaohua Li <shaohua.li@intel.com> Reviewed-by: Wu Fengguang <fengguang.wu@intel.com> Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
| * | | writeback: get rid to incorrect references to pdflush in commentsJens Axboe2009-09-252-9/+6
| | | | | | | | | | | | | | | | Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
| * | | writeback: improve readability of the wb_writeback() continue/break logicJens Axboe2009-09-251-20/+23
| | | | | | | | | | | | | | | | | | | | | | | | | | | | And throw some comments in there, too. Reviewed-by: Wu Fengguang <fengguang.wu@intel.com> Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
| * | | writeback: cleanup writeback_single_inode()Wu Fengguang2009-09-251-8/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Make the if-else straight in writeback_single_inode(). No behavior change. Cc: Jan Kara <jack@suse.cz> Cc: Michael Rubin <mrubin@google.com> Cc: Peter Zijlstra <peterz@infradead.org> Signed-off-by: Fengguang Wu <wfg@mail.ustc.edu.cn> Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
| * | | writeback: kupdate writeback shall not stop when more io is possibleWu Fengguang2009-09-251-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Fix the kupdate case, which disregards wbc.more_io and stop writeback prematurely even when there are more inodes to be synced. wbc.more_io should always be respected. Also remove the pages_skipped check. It will set when some page(s) of some inode(s) cannot be written for now. Such inodes will be delayed for a while. This variable has nothing to do with whether there are other writeable inodes. CC: Jan Kara <jack@suse.cz> CC: Dave Chinner <david@fromorbit.com> CC: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Wu Fengguang <fengguang.wu@intel.com> Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
| * | | writeback: stop background writeback when below background thresholdWu Fengguang2009-09-251-11/+17
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Treat bdi_start_writeback(0) as a special request to do background write, and stop such work when we are below the background dirty threshold. Also simplify the (nr_pages <= 0) checks. Since we already pass in nr_pages=LONG_MAX for WB_SYNC_ALL and background writes, we don't need to worry about it being decreased to zero. Reported-by: Richard Kennedy <richard@rsk.demon.co.uk> CC: Jan Kara <jack@suse.cz> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Wu Fengguang <fengguang.wu@intel.com> Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
| * | | fs: Fix busyloop in wb_writeback()Jan Kara2009-09-251-1/+18
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | If all inodes are under writeback (e.g. in case when there's only one inode with dirty pages), wb_writeback() with WB_SYNC_NONE work basically degrades to busylooping until I_SYNC flags of the inode is cleared. Fix the problem by waiting on I_SYNC flags of an inode on b_more_io list in case we failed to write anything. Tested-by: Wu Fengguang <fengguang.wu@intel.com> Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
generated by cgit v1.2.3 (git 2.25.1) at 2024-08-20 00:32:58 +0000