summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* Merge tag 'notifications-pipe-prep-20191115' of ↵Linus Torvalds2019-11-3013-334/+532
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-fs Pull pipe rework from David Howells: "This is my set of preparatory patches for building a general notification queue on top of pipes. It makes a number of significant changes: - It removes the nr_exclusive argument from __wake_up_sync_key() as this is always 1. This prepares for the next step: - Adds wake_up_interruptible_sync_poll_locked() so that poll can be woken up from a function that's holding the poll waitqueue spinlock. - Change the pipe buffer ring to be managed in terms of unbounded head and tail indices rather than bounded index and length. This means that reading the pipe only needs to modify one index, not two. - A selection of helper functions are provided to query the state of the pipe buffer, plus a couple to apply updates to the pipe indices. - The pipe ring is allowed to have kernel-reserved slots. This allows many notification messages to be spliced in by the kernel without allowing userspace to pin too many pages if it writes to the same pipe. - Advance the head and tail indices inside the pipe waitqueue lock and use wake_up_interruptible_sync_poll_locked() to poke poll without having to take the lock twice. - Rearrange pipe_write() to preallocate the buffer it is going to write into and then drop the spinlock. This allows kernel notifications to then be added the ring whilst it is filling the buffer it allocated. The read side is stalled because the pipe mutex is still held. - Don't wake up readers on a pipe if there was already data in it when we added more. - Don't wake up writers on a pipe if the ring wasn't full before we removed a buffer" * tag 'notifications-pipe-prep-20191115' of git://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-fs: pipe: Remove sync on wake_ups pipe: Increase the writer-wakeup threshold to reduce context-switch count pipe: Check for ring full inside of the spinlock in pipe_write() pipe: Remove redundant wakeup from pipe_write() pipe: Rearrange sequence in pipe_write() to preallocate slot pipe: Conditionalise wakeup in pipe_read() pipe: Advance tail pointer inside of wait spinlock in pipe_read() pipe: Allow pipes to have kernel-reserved slots pipe: Use head and tail pointers for the ring, not cursor and length Add wake_up_interruptible_sync_poll_locked() Remove the nr_exclusive argument from __wake_up_sync_key() pipe: Reduce #inclusion of pipe_fs_i.h
| * pipe: Remove sync on wake_upsDavid Howells2019-11-151-4/+4
| |
| * pipe: Increase the writer-wakeup threshold to reduce context-switch countDavid Howells2019-11-151-2/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Increase the threshold at which the reader sends a wake event to the writers in the queue such that the queue must be half empty before the wake is issued rather than the wake being issued when just a single slot available. This reduces the number of context switches in the tests significantly, without altering the amount of work achieved. With my pipe-bench program, there's a 20% reduction versus an unpatched kernel. Suggested-by: Konstantin Khlebnikov <khlebnikov@yandex-team.ru> Signed-off-by: David Howells <dhowells@redhat.com>
| * pipe: Check for ring full inside of the spinlock in pipe_write()David Howells2019-11-151-0/+5
| | | | | | | | | | | | | | | | | | | | | | Make pipe_write() check to see if the ring has become full between it taking the pipe mutex, checking the ring status and then taking the spinlock. This can happen if a notification is written into the pipe as that happens without the pipe mutex. Signed-off-by: David Howells <dhowells@redhat.com>
| * pipe: Remove redundant wakeup from pipe_write()David Howells2019-11-151-5/+0
| | | | | | | | | | | | Remove a redundant wakeup from pipe_write(). Signed-off-by: David Howells <dhowells@redhat.com>
| * pipe: Rearrange sequence in pipe_write() to preallocate slotDavid Howells2019-11-151-18/+33
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Rearrange the sequence in pipe_write() so that the allocation of the new buffer, the allocation of a ring slot and the attachment to the ring is done under the pipe wait spinlock and then the lock is dropped and the buffer can be filled. The data copy needs to be done with the spinlock unheld and irqs enabled, so the lock needs to be dropped first. However, the reader can't progress as we're holding pipe->mutex. We also need to drop the lock as that would impact others looking at the pipe waitqueue, such as poll(), the consumer and a future kernel message writer. We just abandon the preallocated slot if we get a copy error. Future writes may continue it and a future read will eventually recycle it. Signed-off-by: David Howells <dhowells@redhat.com>
| * pipe: Conditionalise wakeup in pipe_read()David Howells2019-11-151-9/+6
| | | | | | | | | | | | | | Only do a wakeup in pipe_read() if we made space in a completely full buffer. The producer shouldn't be waiting on pipe->wait otherwise. Signed-off-by: David Howells <dhowells@redhat.com>
| * pipe: Advance tail pointer inside of wait spinlock in pipe_read()David Howells2019-11-151-1/+7
| | | | | | | | | | | | | | | | Advance the pipe ring tail pointer inside of wait spinlock in pipe_read() so that the pipe can be written into with kernel notifications from contexts where pipe->mutex cannot be taken. Signed-off-by: David Howells <dhowells@redhat.com>
| * pipe: Allow pipes to have kernel-reserved slotsDavid Howells2019-11-155-24/+30
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Split pipe->ring_size into two numbers: (1) pipe->ring_size - indicates the hard size of the pipe ring. (2) pipe->max_usage - indicates the maximum number of pipe ring slots that userspace orchestrated events can fill. This allows for a pipe that is both writable by the general kernel notification facility and by userspace, allowing plenty of ring space for notifications to be added whilst preventing userspace from being able to pin too much unswappable kernel space. Signed-off-by: David Howells <dhowells@redhat.com>
| * pipe: Use head and tail pointers for the ring, not cursor and lengthDavid Howells2019-10-317-292/+448
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Convert pipes to use head and tail pointers for the buffer ring rather than pointer and length as the latter requires two atomic ops to update (or a combined op) whereas the former only requires one. (1) The head pointer is the point at which production occurs and points to the slot in which the next buffer will be placed. This is equivalent to pipe->curbuf + pipe->nrbufs. The head pointer belongs to the write-side. (2) The tail pointer is the point at which consumption occurs. It points to the next slot to be consumed. This is equivalent to pipe->curbuf. The tail pointer belongs to the read-side. (3) head and tail are allowed to run to UINT_MAX and wrap naturally. They are only masked off when the array is being accessed, e.g.: pipe->bufs[head & mask] This means that it is not necessary to have a dead slot in the ring as head == tail isn't ambiguous. (4) The ring is empty if "head == tail". A helper, pipe_empty(), is provided for this. (5) The occupancy of the ring is "head - tail". A helper, pipe_occupancy(), is provided for this. (6) The number of free slots in the ring is "pipe->ring_size - occupancy". A helper, pipe_space_for_user() is provided to indicate how many slots userspace may use. (7) The ring is full if "head - tail >= pipe->ring_size". A helper, pipe_full(), is provided for this. Signed-off-by: David Howells <dhowells@redhat.com>
| * Add wake_up_interruptible_sync_poll_locked()David Howells2019-10-312-0/+26
| | | | | | | | | | | | | | | | | | Add a wakeup call for a case whereby the caller already has the waitqueue spinlock held. This can be used by pipes to alter the ring buffer indices and issue a wakeup under the same spinlock. Signed-off-by: David Howells <dhowells@redhat.com> Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
| * Remove the nr_exclusive argument from __wake_up_sync_key()David Howells2019-10-233-15/+9
| | | | | | | | | | | | | | | | | | Remove the nr_exclusive argument from __wake_up_sync_key() and derived functions as everything seems to set it to 1. Note also that if it wasn't set to 1, it would clear WF_SYNC anyway. Signed-off-by: David Howells <dhowells@redhat.com> Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
| * pipe: Reduce #inclusion of pipe_fs_i.hDavid Howells2019-10-233-3/+0
| | | | | | | | | | | | | | Remove some #inclusions of linux/pipe_fs_i.h that don't seem to be necessary any more. Signed-off-by: David Howells <dhowells@redhat.com>
* | Merge tag 'fsnotify_for_v5.5-rc1' of ↵Linus Torvalds2019-11-304-3/+4
|\ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs Pull fsnotify updates from Jan Kara: "Three fsnotify cleanups" * tag 'fsnotify_for_v5.5-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs: fsnotify: Add git tree reference to MAINTAINERS fsnotify/fdinfo: exportfs_encode_inode_fh() takes pointer as 4th argument fsnotify: move declaration of fsnotify_mark_connector_cachep to fsnotify.h
| * | fsnotify: Add git tree reference to MAINTAINERSJan Kara2019-10-301-0/+1
| | | | | | | | | | | | | | | | | | Add reference to git tree with fsnotify changes to MAINTAINERS file. Signed-off-by: Jan Kara <jack@suse.cz>
| * | fsnotify/fdinfo: exportfs_encode_inode_fh() takes pointer as 4th argumentBen Dooks (Codethink)2019-10-171-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The call to exportfs_encode_inode_fh() takes an pointer as the 4th argument, so replace the integer 0 with the NULL pointer. This fixes the following sparse warning: fs/notify/fdinfo.c:53:87: warning: Using plain integer as NULL pointer Link: https://lore.kernel.org/r/20191016095955.3347-1-ben.dooks@codethink.co.uk Signed-off-by: Ben Dooks <ben.dooks@codethink.co.uk> Signed-off-by: Jan Kara <jack@suse.cz>
| * | fsnotify: move declaration of fsnotify_mark_connector_cachep to fsnotify.hBen Dooks2019-10-172-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Move fsnotify_mark_connector_cachep to fsnotify.h to properly share it with the user in mark.c and avoid the following warning from sparse: fs/notify/mark.c:82:19: warning: symbol 'fsnotify_mark_connector_cachep' was not declared. Should it be static? Link: https://lore.kernel.org/r/20191015132518.21819-1-ben.dooks@codethink.co.uk Signed-off-by: Ben Dooks <ben.dooks@codethink.co.uk> Signed-off-by: Jan Kara <jack@suse.cz>
* | | Merge tag 'for_v5.5-rc1' of ↵Linus Torvalds2019-11-3021-243/+248
|\ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs Pull ext2, quota, reiserfs cleanups and fixes from Jan Kara: - Refactor the quota on/off kernel internal interfaces (mostly for ubifs quota support as ubifs does not want to have inodes holding quota information) - A few other small quota fixes and cleanups - Various small ext2 fixes and cleanups - Reiserfs xattr fix and one cleanup * tag 'for_v5.5-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs: (28 commits) ext2: code cleanup for descriptor_loc() fs/quota: handle overflows of sysctl fs.quota.* and report as unsigned long ext2: fix improper function comment ext2: code cleanup for ext2_try_to_allocate() ext2: skip unnecessary operations in ext2_try_to_allocate() ext2: Simplify initialization in ext2_try_to_allocate() ext2: code cleanup by calling ext2_group_last_block_no() ext2: introduce new helper ext2_group_last_block_no() reiserfs: replace open-coded atomic_dec_and_mutex_lock() ext2: check err when partial != NULL quota: Handle quotas without quota inodes in dquot_get_state() quota: Make dquot_disable() work without quota inodes quota: Drop dquot_enable() fs: Use dquot_load_quota_inode() from filesystems quota: Rename vfs_load_quota_inode() to dquot_load_quota_inode() quota: Simplify dquot_resume() quota: Factor out setup of quota inode quota: Check that quota is not dirty before release quota: fix livelock in dquot_writeback_dquots ext2: don't set *count in the case of failure in ext2_try_to_allocate() ...
| * | | ext2: code cleanup for descriptor_loc()Chengguang Xu2019-11-191-4/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Code cleanup by removing unnecessary variable in descriptor_loc(). Link: https://lore.kernel.org/r/20191115224900.2613-1-cgxu519@mykernel.net Signed-off-by: Chengguang Xu <cgxu519@mykernel.net> Signed-off-by: Jan Kara <jack@suse.cz>
| * | | fs/quota: handle overflows of sysctl fs.quota.* and report as unsigned longKonstantin Khlebnikov2019-11-112-13/+18
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Quota statistics counted as 64-bit per-cpu counter. Reading sums per-cpu fractions as signed 64-bit int, filters negative values and then reports lower half as signed 32-bit int. Result may looks like: fs.quota.allocated_dquots = 22327 fs.quota.cache_hits = -489852115 fs.quota.drops = -487288718 fs.quota.free_dquots = 22083 fs.quota.lookups = -486883485 fs.quota.reads = 22327 fs.quota.syncs = 335064 fs.quota.writes = 3088689 Values bigger than 2^31-1 reported as negative. All counters except "allocated_dquots" and "free_dquots" are monotonic, thus they should be reported as is without filtering negative values. Kernel doesn't have generic helper for 64-bit sysctl yet, let's use at least unsigned long. Link: https://lore.kernel.org/r/157337934693.2078.9842146413181153727.stgit@buzz Signed-off-by: Konstantin Khlebnikov <khlebnikov@yandex-team.ru> Signed-off-by: Jan Kara <jack@suse.cz>
| * | | ext2: fix improper function commentChengguang Xu2019-11-071-3/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Just fix a improper function comment. Link: https://lore.kernel.org/r/20191104114036.9893-5-cgxu519@mykernel.net Signed-off-by: Chengguang Xu <cgxu519@mykernel.net> Signed-off-by: Jan Kara <jack@suse.cz>
| * | | ext2: code cleanup for ext2_try_to_allocate()Chengguang Xu2019-11-071-19/+10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Code cleanup by removing duplicated code. Link: https://lore.kernel.org/r/20191104114036.9893-4-cgxu519@mykernel.net Signed-off-by: Chengguang Xu <cgxu519@mykernel.net> Signed-off-by: Jan Kara <jack@suse.cz>
| * | | ext2: skip unnecessary operations in ext2_try_to_allocate()Chengguang Xu2019-11-071-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Move 'repeat' tag to proper place so that we can skip unnecessary operations in ext2_try_to_allocate(). Link: https://lore.kernel.org/r/20191104114036.9893-3-cgxu519@mykernel.net Signed-off-by: Chengguang Xu <cgxu519@mykernel.net> Signed-off-by: Jan Kara <jack@suse.cz>
| * | | ext2: Simplify initialization in ext2_try_to_allocate()Jan Kara2019-11-071-17/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Somewhat simplify the logic initializing search start and end in ext2_try_to_allocate(). No functional change. Signed-off-by: Jan Kara <jack@suse.cz>
| * | | ext2: code cleanup by calling ext2_group_last_block_no()Chengguang Xu2019-11-062-15/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Call common helper ext2_group_last_block_no() to calculate group last block number. Link: https://lore.kernel.org/r/20191104114036.9893-2-cgxu519@mykernel.net Signed-off-by: Chengguang Xu <cgxu519@mykernel.net> Signed-off-by: Jan Kara <jack@suse.cz>
| * | | ext2: introduce new helper ext2_group_last_block_no()Chengguang Xu2019-11-061-0/+12
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Introduce new helper ext2_group_last_block_no() to calculate last block num for specific block group, we can replace open coded logic by calling this common helper. Link: https://lore.kernel.org/r/20191104114036.9893-1-cgxu519@mykernel.net Signed-off-by: Chengguang Xu <cgxu519@mykernel.net> Signed-off-by: Jan Kara <jack@suse.cz>
| * | | Pull series refactoring quota enabling and disabling code.Jan Kara2019-11-06687-6544/+6289
| |\ \ \
| | * | | quota: Handle quotas without quota inodes in dquot_get_state()Jan Kara2019-11-041-2/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Make dquot_get_state() gracefully handle a situation when there are no quota files present even though quotas are enabled. Signed-off-by: Jan Kara <jack@suse.cz>
| | * | | quota: Make dquot_disable() work without quota inodesJan Kara2019-11-041-44/+29
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Quota on and quota off are protected by s_umount semaphore held in exclusive mode since commit 7d6cd73d33b6 "quota: Hold s_umount in exclusive mode when enabling / disabling quotas". This makes it impossible for dquot_disable() to race with other enabling or disabling of quotas. Simplify the cleanup done by dquot_disable() based on this fact and also remove some stale comments. As a bonus this cleanup makes dquot_disable() properly handle a case when there are no quota inodes. Signed-off-by: Jan Kara <jack@suse.cz>
| | * | | quota: Drop dquot_enable()Jan Kara2019-11-042-47/+16
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Now dquot_enable() has only two internal callers and both of them just need to update quota flags and don't need most of checks. Just drop dquot_enable() and fold necessary functionality into the two calling places. Signed-off-by: Jan Kara <jack@suse.cz>
| | * | | fs: Use dquot_load_quota_inode() from filesystemsJan Kara2019-11-043-4/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Use dquot_load_quota_inode from filesystems instead of dquot_enable(). In all three cases we want to load quota inode and never use the function to update quota flags. Signed-off-by: Jan Kara <jack@suse.cz>
| | * | | quota: Rename vfs_load_quota_inode() to dquot_load_quota_inode()Jan Kara2019-11-042-6/+15
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Rename vfs_load_quota_inode() to dquot_load_quota_inode() to be consistent with naming of other functions used for enabling quota accounting from filesystems. Also export the function and add some sanity checks to assure filesystems are calling the function properly. Signed-off-by: Jan Kara <jack@suse.cz>
| | * | | quota: Simplify dquot_resume()Jan Kara2019-11-041-6/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | We already have quota inode loaded when resuming quotas. Use vfs_load_quota() to avoid some pointless churn with the quota inode. Signed-off-by: Jan Kara <jack@suse.cz>
| | * | | quota: Factor out setup of quota inodeJan Kara2019-11-042-41/+69
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Factor out setting up of quota inode and eventual error cleanup from vfs_load_quota_inode(). This will simplify situation for filesystems that don't have any quota inodes. Signed-off-by: Jan Kara <jack@suse.cz>
| * | | | reiserfs: replace open-coded atomic_dec_and_mutex_lock()Nikitas Angelinas2019-11-051-8/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Replace the open-coded logic of atomic_dec_and_mutex_lock() in reiserfs_file_release(). Link: https://lore.kernel.org/r/20191103094431.GA18576-nikitas.angelinas@gmail.com Signed-off-by: Nikitas Angelinas <nikitas.angelinas@gmail.com> Signed-off-by: Jan Kara <jack@suse.cz>
| * | | | ext2: check err when partial != NULLChengguang Xu2019-11-051-2/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Check err when partial == NULL is meaningless because partial == NULL means getting branch successfully without error. CC: stable@vger.kernel.org Link: https://lore.kernel.org/r/20191105045100.7104-1-cgxu519@mykernel.net Signed-off-by: Chengguang Xu <cgxu519@mykernel.net> Signed-off-by: Jan Kara <jack@suse.cz>
| * | | | quota: Check that quota is not dirty before releaseDmitry Monakhov2019-10-313-2/+12
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | There is a race window where quota was redirted once we drop dq_list_lock inside dqput(), but before we grab dquot->dq_lock inside dquot_release() TASK1 TASK2 (chowner) ->dqput() we_slept: spin_lock(&dq_list_lock) if (dquot_dirty(dquot)) { spin_unlock(&dq_list_lock); dquot->dq_sb->dq_op->write_dquot(dquot); goto we_slept if (test_bit(DQ_ACTIVE_B, &dquot->dq_flags)) { spin_unlock(&dq_list_lock); dquot->dq_sb->dq_op->release_dquot(dquot); dqget() mark_dquot_dirty() dqput() goto we_slept; } So dquot dirty quota will be released by TASK1, but on next we_sleept loop we detect this and call ->write_dquot() for it. XFSTEST: https://github.com/dmonakhov/xfstests/commit/440a80d4cbb39e9234df4d7240aee1d551c36107 Link: https://lore.kernel.org/r/20191031103920.3919-2-dmonakhov@openvz.org CC: stable@vger.kernel.org Signed-off-by: Dmitry Monakhov <dmtrmonakhov@yandex-team.ru> Signed-off-by: Jan Kara <jack@suse.cz>
| * | | | quota: fix livelock in dquot_writeback_dquotsDmitry Monakhov2019-10-311-4/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Write only quotas which are dirty at entry. XFSTEST: https://github.com/dmonakhov/xfstests/commit/b10ad23566a5bf75832a6f500e1236084083cddc Link: https://lore.kernel.org/r/20191031103920.3919-1-dmonakhov@openvz.org CC: stable@vger.kernel.org Signed-off-by: Konstantin Khlebnikov <khlebnikov@yandex-team.ru> Signed-off-by: Dmitry Monakhov <dmtrmonakhov@yandex-team.ru> Signed-off-by: Jan Kara <jack@suse.cz>
| * | | | ext2: don't set *count in the case of failure in ext2_try_to_allocate()Chengguang Xu2019-10-311-1/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Currently we set *count to num(value 0) in the failure of block allocation in ext2_try_to_allocate(). Without reservation, we reuse *count(value 0) to retry block allocation and wrong *count will cause only allocating maximum 1 block even though having sufficent free blocks in that block group. Finally, it probably cause significant fragmentation. Link: https://lore.kernel.org/r/20191026090721.23794-1-cgxu519@mykernel.net Signed-off-by: Chengguang Xu <cgxu519@mykernel.net> Signed-off-by: Jan Kara <jack@suse.cz>
| * | | | reiserfs: fix extended attributes on the root directoryJeff Mahoney2019-10-316-14/+32
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Since commit d0a5b995a308 (vfs: Add IOP_XATTR inode operations flag) extended attributes haven't worked on the root directory in reiserfs. This is due to reiserfs conditionally setting the sb->s_xattrs handler array depending on whether it located or create the internal privroot directory. It necessarily does this after the root inode is already read in. The IOP_XATTR flag is set during inode initialization, so it never gets set on the root directory. This commit unconditionally assigns sb->s_xattrs and clears IOP_XATTR on internal inodes. The old return values due to the conditional assignment are handled via open_xa_root, which now returns EOPNOTSUPP as the VFS would have done. Link: https://lore.kernel.org/r/20191024143127.17509-1-jeffm@suse.com CC: stable@vger.kernel.org Fixes: d0a5b995a308 ("vfs: Add IOP_XATTR inode operations flag") Signed-off-by: Jeff Mahoney <jeffm@suse.com> Signed-off-by: Jan Kara <jack@suse.cz>
| * | | | ext2: return error when fail to allocating memory in ioctlChengguang Xu2019-10-241-1/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Currently, we do not check memory allocation result for ei->i_block_alloc_info in ioctl, this patch checks it and returns error in failure case. Link: https://lore.kernel.org/r/20191023135643.28837-1-cgxu519@mykernel.net Signed-off-by: Chengguang Xu <cgxu519@mykernel.net> Signed-off-by: Jan Kara <jack@suse.cz>
| * | | | ext2: add missing brelse in ext2_new_blocks()Chengguang Xu2019-10-221-0/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Fix missing brelse of bitmap_bh in the case of retry. Link: https://lore.kernel.org/r/20191022091738.9160-1-cgxu519@mykernel.net Signed-off-by: Chengguang Xu <cgxu519@mykernel.net> Signed-off-by: Jan Kara <jack@suse.cz>
| * | | | ext2: adjust block num when retry allocationChengguang Xu2019-10-211-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Set block num to original *count in a case of retrying allocation in case num < *count Link: https://lore.kernel.org/r/20191020232326.84881-1-cgxu519@mykernel.net Signed-off-by: Chengguang Xu <cgxu519@mykernel.net> Signed-off-by: Jan Kara <jack@suse.cz>
| * | | | quota: minor code cleanup for v1_format_opsChengguang Xu2019-10-171-1/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | It's not a functinal change, it's just for keeping consistent coding style. Link: https://lore.kernel.org/r/20191010130924.17697-1-cgxu519@mykernel.net Signed-off-by: Chengguang Xu <cgxu519@mykernel.net> Signed-off-by: Jan Kara <jack@suse.cz>
| * | | | quota: check quota type in early stageChengguang Xu2019-10-081-4/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Check quota type in early stage so we can avoid many unncessary operations when the type is wrong. Link: https://lore.kernel.org/r/20191008145059.21402-1-cgxu519@mykernel.net Signed-off-by: Chengguang Xu <cgxu519@mykernel.net> Signed-off-by: Jan Kara <jack@suse.cz>
| * | | | quota: code cleanup for hash bits calculationChengguang Xu2019-10-041-5/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Code cleanup for hash bits calculation by calling ilog2(). Link: https://lore.kernel.org/r/20190923135223.27674-1-cgxu519@zoho.com.cn Signed-off-by: Chengguang Xu <cgxu519@zoho.com.cn> Signed-off-by: Jan Kara <jack@suse.cz>
| * | | | quota: avoid increasing DQST_LOOKUPS when iterating over dirty/inuse listChengguang Xu2019-10-041-2/+0
| | |/ / | |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | It is meaningless to increase DQST_LOOKUPS number while iterating over dirty/inuse list, so just avoid it. Link: https://lore.kernel.org/r/20190926083408.4269-1-cgxu519@zoho.com.cn Signed-off-by: Chengguang Xu <cgxu519@zoho.com.cn> Signed-off-by: Jan Kara <jack@suse.cz>
* | | | Merge tag 'erofs-for-5.5-rc1' of ↵Linus Torvalds2019-11-309-203/+190
|\ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/xiang/erofs Pull erofs updates from Gao Xiang: "No major kernel updates for this round since I'm fully diving into LZMA algorithm internals now to provide high CR XZ algorihm support. That needs more work and time for me to get a better compression time. Summary: - Introduce superblock checksum support - Set iowait when waiting I/O for sync decompression path - Several code cleanups" * tag 'erofs-for-5.5-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/xiang/erofs: erofs: remove unnecessary output in erofs_show_options() erofs: drop all vle annotations for runtime names erofs: support superblock checksum erofs: set iowait for sync decompression erofs: clean up decompress queue stuffs erofs: get rid of __stagingpage_alloc helper erofs: remove dead code since managed cache is now built-in erofs: clean up collection handling routines
| * | | | erofs: remove unnecessary output in erofs_show_options()Chengguang Xu2019-11-241-3/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | We have already handled cache_strategy option carefully, so incorrect setting could not pass option parsing. Meanwhile, print 'cache_strategy=(unknown)' can cause failure on remount. Link: https://lore.kernel.org/r/20191119115049.3401-1-cgxu519@mykernel.net Signed-off-by: Chengguang Xu <cgxu519@mykernel.net> Reviewed-by: Gao Xiang <gaoxiang25@huawei.com> Reviewed-by: Chao Yu <yuchao0@huawei.com> Signed-off-by: Gao Xiang <gaoxiang25@huawei.com>
| * | | | erofs: drop all vle annotations for runtime namesGao Xiang2019-11-243-50/+44
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | VLE was an old informal name of fixed-sized output compression which came from published ATC'19 paper [1]. Drop those old annotations since erofs can handle all encoded clusters in block-aligned basis, which is wider than fixed-sized output compression after larger clustersize feature is fully implemented. Unaligned encoding won't be considered in EROFS since it's not friendly to inplace I/O and perhaps decompression inplace. a) Fixed-sized output compression with 16KB pcluster: ___________________________________ |xxxxxxxx|xxxxxxxx|xxxxxxxx|xxxxxxxx| |___ 0___|___ 1___|___ 2___|___ 3___| physical blocks b) Block-aligned fixed-sized input compression with 16KB pcluster: ___________________________________ |xxxxxxxx|xxxxxxxx|xxxxxxxx|xxx00000| |___ 0___|___ 1___|___ 2___|___ 3___| physical blocks c) Block-unaligned fixed-sized input compression with 16KB compression unit: ____________________________________________ |..xxxxxx|xxxxxxxx|xxxxxxxx|xxxxxxxx|x.......| |___ 0___|___ 1___|___ 2___|___ 3___|___ 4___| physical blocks Refine better names for those as well. [1] https://www.usenix.org/conference/atc19/presentation/gao Link: https://lore.kernel.org/r/20191108033733.63919-1-gaoxiang25@huawei.com Reviewed-by: Chao Yu <yuchao0@huawei.com> Signed-off-by: Gao Xiang <gaoxiang25@huawei.com>
generated by cgit v1.2.3 (git 2.25.1) at 2024-10-09 08:10:41 +0000