summaryrefslogtreecommitdiff
path: root/drivers/md
AgeCommit message (Collapse)Author
2024-03-04dm vdo permassert: audit all of ASSERT to test for VDO_SUCCESSMike Snitzer
Also rename ASSERT to VDO_ASSERT and ASSERT_LOG_ONLY to VDO_ASSERT_LOG_ONLY. But re-introduce ASSERT and ASSERT_LOG_ONLY as a placeholder for the benefit of dm-vdo/indexer. Signed-off-by: Mike Snitzer <snitzer@kernel.org> Signed-off-by: Matthew Sakai <msakai@redhat.com>
2024-03-04dm-vdo funnel-workqueue: return VDO_SUCCESS from make_simple_work_queueMike Snitzer
Signed-off-by: Mike Snitzer <snitzer@kernel.org> Signed-off-by: Matthew Sakai <msakai@redhat.com>
2024-03-04dm vdo thread-utils: return VDO_SUCCESS on vdo_create_thread successMike Snitzer
Update all callers to check for VDO_SUCCESS. Signed-off-by: Mike Snitzer <snitzer@kernel.org> Signed-off-by: Matthew Sakai <msakai@redhat.com>
2024-03-04dm vdo int-map: return VDO_SUCCESS on successMike Snitzer
Update all callers to check for VDO_SUCCESS (most already did). Also fix whitespace for update_mapping() parameters. Signed-off-by: Mike Snitzer <snitzer@kernel.org> Signed-off-by: Matthew Sakai <msakai@redhat.com>
2024-03-04dm vdo: check for VDO_SUCCESS return value from memory-alloc functionsMike Snitzer
VDO_SUCCESS and UDS_SUCCESS were used interchangably, update all callers of VDO's memory-alloc functions to consistently check for VDO_SUCCESS. Signed-off-by: Mike Snitzer <snitzer@kernel.org> Signed-off-by: Matthew Sakai <msakai@redhat.com>
2024-03-04dm vdo memory-alloc: return VDO_SUCCESS on successMike Snitzer
Signed-off-by: Mike Snitzer <snitzer@kernel.org> Signed-off-by: Matthew Sakai <msakai@redhat.com>
2024-03-04dm vdo errors: remove unused error codesMatthew Sakai
Also define VDO_SUCCESS in a more central location, and rename error block constants for clarity. Signed-off-by: Matthew Sakai <msakai@redhat.com> Signed-off-by: Mike Snitzer <snitzer@kernel.org>
2024-03-04dm vdo memory-alloc: rename vdo_do_allocation to __vdo_do_allocationMike Snitzer
__vdo_do_allocation shouldn't be used outside of memory-alloc.h, so add hidden prefix. Also, tabify the vdo_allocate_extended macro. Signed-off-by: Mike Snitzer <snitzer@kernel.org> Signed-off-by: Matthew Sakai <msakai@redhat.com>
2024-03-04dm vdo memory-alloc: change from uds_ to vdo_ namespaceMike Snitzer
Signed-off-by: Mike Snitzer <snitzer@kernel.org> Signed-off-by: Matthew Sakai <msakai@redhat.com>
2024-03-04dm-vdo: change unnamed enums to definesBruce Johnston
Signed-off-by: Bruce Johnston <bjohnsto@redhat.com> Signed-off-by: Matthew Sakai <msakai@redhat.com> Signed-off-by: Mike Snitzer <snitzer@kernel.org>
2024-03-04dm vdo: remove outdated pointer_map referenceMatthew Sakai
Signed-off-by: Matthew Sakai <msakai@redhat.com> Signed-off-by: Mike Snitzer <snitzer@kernel.org>
2024-03-04dm vdo: update module commentsMatthew Sakai
Update outdated comments referring to separate VDO and UDS modules. Signed-off-by: Matthew Sakai <msakai@redhat.com> Signed-off-by: Mike Snitzer <snitzer@kernel.org>
2024-03-04dm vdo indexer delta-index: fix typos in commentsMatthew Sakai
Signed-off-by: Matthew Sakai <msakai@redhat.com> Signed-off-by: Mike Snitzer <snitzer@kernel.org>
2024-03-04dm vdo: fix various function names referenced in comment blocksJiapeng Chong
No functional modification involved. Reported-by: Abaci Robot <abaci@linux.alibaba.com> Signed-off-by: Jiapeng Chong <jiapeng.chong@linux.alibaba.com> Signed-off-by: Matthew Sakai <msakai@redhat.com> Signed-off-by: Mike Snitzer <snitzer@kernel.org>
2024-03-04dm vdo: move indexer files into sub-directoryMike Snitzer
The goal is to assist high-level understanding of which code is conceptually specific to VDO's indexer. Signed-off-by: Mike Snitzer <snitzer@kernel.org> Signed-off-by: Matthew Sakai <msakai@redhat.com>
2024-03-04dm vdo: remove unnecessary indexer.h includesMatthew Sakai
Signed-off-by: Matthew Sakai <msakai@redhat.com> Signed-off-by: Mike Snitzer <snitzer@kernel.org>
2024-03-04dm vdo: clean up scnprintf usageChung Chung
Ignore scnprintf return status since it is not necessary. Change write_* functions type from int to void since we no longer return any result. Also, clean up any code that checks or uses any scnprintf return results. Check uds_allocate return code which was previous ignored, return and log error when uds_allocate failed. Reported-by: Dan Carpenter <dan.carpenter@linaro.org> Signed-off-by: Chung Chung <cchung@redhat.com> Signed-off-by: Matthew Sakai <msakai@redhat.com> Signed-off-by: Mike Snitzer <snitzer@kernel.org>
2024-03-04dm vdo: include <asm/current.h> to resolve current being undeclaredMike Snitzer
Reported when building on loongarch. Reported-by: Randy Dunlap <rdunlap@infradead.org> Signed-off-by: Mike Snitzer <snitzer@kernel.org> Signed-off-by: Bruce Johnston <bjohnsto@redhat.com> Signed-off-by: Matthew Sakai <msakai@redhat.com>
2024-03-04dm vdo indexer-volume: fix missing mutex_lock in process_entryMike Snitzer
Must mutex_lock after dm_bufio_read, before dm_bufio_read error handling, otherwise process_entry error path will return without volume->read_threads_mutex held. This fixes potential double mutex_unlock. Reported-by: Dan Carpenter <dan.carpenter@linaro.org> Signed-off-by: Mike Snitzer <snitzer@kernel.org> Signed-off-by: Susan LeGendre-McGhee <slegendr@redhat.com> Signed-off-by: Matthew Sakai <msakai@redhat.com>
2024-03-04dm vdo flush: initialize return to NULL in allocate_flushMike Snitzer
Otherwise, error path could result in allocate_flush's subsequent check for flush being non-NULL leading to false positive. Reported-by: Dan Carpenter <dan.carpenter@linaro.org> Signed-off-by: Susan LeGendre-McGhee <slegendr@redhat.com> Signed-off-by: Matthew Sakai <msakai@redhat.com> Signed-off-by: Mike Snitzer <snitzer@kernel.org>
2024-03-04dm vdo slab-depot: delete unnecessary check in allocate_componentsDan Carpenter
This is a duplicate check so it can't be true. Delete it. Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org> Signed-off-by: Susan LeGendre-McGhee <slegendr@redhat.com> Signed-off-by: Matthew Sakai <msakai@redhat.com> Signed-off-by: Mike Snitzer <snitzer@kernel.org>
2024-03-04dm vdo memory-alloc: simplify allocations_allowed()Mike Snitzer
Signed-off-by: Mike Snitzer <snitzer@kernel.org> Signed-off-by: Susan LeGendre-McGhee <slegendr@redhat.com> Signed-off-by: Matthew Sakai <msakai@redhat.com>
2024-03-04dm vdo: remove internal ticket referencesSusan LeGendre-McGhee
Signed-off-by: Susan LeGendre-McGhee <slegendr@redhat.com> Signed-off-by: Matthew Sakai <msakai@redhat.com> Signed-off-by: Mike Snitzer <snitzer@kernel.org>
2024-03-02dm-verity: Convert from tasklet to BH workqueueTejun Heo
The only generic interface to execute asynchronously in the BH context is tasklet; however, it's marked deprecated and has some design flaws. To replace tasklets, BH workqueue support was recently added. A BH workqueue behaves similarly to regular workqueues except that the queued work items are executed in the BH context. This commit converts dm-verity from tasklet to BH workqueue. It backfills tasklet code that was removed with commit 0a9bab391e33 ("dm-crypt, dm-verity: disable tasklets") and tweaks to use BH workqueue (and does some renaming). This is a minimal conversion which doesn't rename the related names including the "try_verify_in_tasklet" option. If this patch is applied, a follow-up patch would be necessary. I couldn't decide whether the option name would need to be updated too. Signed-off-by: Tejun Heo <tj@kernel.org> [snitzer: rename 'use_tasklet' to 'use_bh_wq' and 'in_tasklet' to 'in_bh'] Signed-off-by: Mike Snitzer <snitzer@kernel.org>
2024-03-02dm-crypt: Convert from tasklet to BH workqueueTejun Heo
The only generic interface to execute asynchronously in the BH context is tasklet; however, it's marked deprecated and has some design flaws. To replace tasklets, BH workqueue support was recently added. A BH workqueue behaves similarly to regular workqueues except that the queued work items are executed in the BH context. This commit converts dm-crypt from tasklet to BH workqueue. It backfills tasklet code that was removed with commit 0a9bab391e33 ("dm-crypt, dm-verity: disable tasklets") and tweaks to use BH workqueue. Like a regular workqueue, a BH workqueue allows freeing the currently executing work item. Converting from tasklet to BH workqueue removes the need for deferring bio_endio() again to a work item, which was buggy anyway. I tested this lightly with "--perf-no_read_workqueue --perf-no_write_workqueue" + some code modifications, but would really -appreciate if someone who knows the code base better could take a look. Signed-off-by: Tejun Heo <tj@kernel.org> Link: http://lkml.kernel.org/r/82b964f0-c2c8-a2c6-5b1f-f3145dc2c8e5@redhat.com [snitzer: rebase ontop of commit 0a9bab391e33 reduced this commit's changes] Signed-off-by: Mike Snitzer <snitzer@kernel.org>
2024-03-01dm: use queue_limits_setChristoph Hellwig
Use queue_limits_set which validates the limits and takes care of updating the readahead settings instead of directly assigning them to the queue. For that make sure all limits are actually updated before the assignment. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Mike Snitzer <snitzer@kernel.org> Link: https://lore.kernel.org/r/20240228225653.947152-4-hch@lst.de Signed-off-by: Jens Axboe <axboe@kernel.dk>
2024-03-01dm vdo thread-device: rename all methods to reflect vdo-only useMike Snitzer
Also moved vdo_init()'s call to vdo_initialize_thread_device_registry next to other registry initialization. Signed-off-by: Mike Snitzer <snitzer@kernel.org> Signed-off-by: Matthew Sakai <msakai@redhat.com>
2024-03-01dm vdo thread-registry: rename all methods to reflect vdo-only useMike Snitzer
Otherwise, uds_ prefix is misleading (vdo_ is the new catch-all for code that is used by vdo-only or _both_ vdo and the indexer code). Signed-off-by: Mike Snitzer <snitzer@kernel.org> Signed-off-by: Matthew Sakai <msakai@redhat.com>
2024-03-01dm vdo thread-utils: cleanup included headersMike Snitzer
Signed-off-by: Mike Snitzer <snitzer@kernel.org> Signed-off-by: Matthew Sakai <msakai@redhat.com>
2024-03-01dm vdo thread-utils: further cleanup of thread functionsMike Snitzer
Change thread function prefix from "uds_" to "vdo_" and fix vdo_join_threads() to return void. Signed-off-by: Mike Snitzer <snitzer@kernel.org> Signed-off-by: Matthew Sakai <msakai@redhat.com>
2024-03-01dm vdo thread-utils: remove all uds_*_mutex wrappersMike Snitzer
Just use mutex_init, mutex_lock and mutex_unlock. Signed-off-by: Mike Snitzer <snitzer@kernel.org> Signed-off-by: Matthew Sakai <msakai@redhat.com>
2024-03-01dm vdo thread-utils: push uds_*_cond interface down to indexerMike Snitzer
Only used by indexer components. Also return void from uds_init_cond(), remove uds_destroy_cond(), and fix up all callers. Signed-off-by: Mike Snitzer <snitzer@kernel.org> Signed-off-by: Matthew Sakai <msakai@redhat.com>
2024-03-01dm vdo: fold thread-cond-var.c into thread-utilsMike Snitzer
Further cleanup is needed for thread-utils interfaces given many functions should return void or be removed entirely because they amount to obfuscation via wrappers. Signed-off-by: Mike Snitzer <snitzer@kernel.org> Signed-off-by: Matthew Sakai <msakai@redhat.com>
2024-03-01dm vdo indexer: rename uds.h to indexer.hMike Snitzer
Also remove unnecessary include from funnel-queue.c. Signed-off-by: Mike Snitzer <snitzer@kernel.org> Signed-off-by: Matthew Sakai <msakai@redhat.com>
2024-03-01dm vdo: rename uds-threads.[ch] to thread-utils.[ch]Mike Snitzer
Signed-off-by: Mike Snitzer <snitzer@kernel.org> Signed-off-by: Matthew Sakai <msakai@redhat.com>
2024-03-01dm vdo indexer sparse-cache: cleanup threads_barrier codeMike Snitzer
Rename 'barrier' to 'threads_barrier', remove useless uds_destroy_barrier(), return void from remaining methods and clean up uds_make_sparse_cache() accordingly. Also remove uds_ prefix from the 2 remaining threads_barrier functions. Signed-off-by: Mike Snitzer <snitzer@kernel.org> Signed-off-by: Matthew Sakai <msakai@redhat.com>
2024-03-01dm vdo uds-threads: push 'barrier' down to sparse-cacheMike Snitzer
The sparse-cache is the only user of the 'barrier' data structure, so just move it private to it. Signed-off-by: Mike Snitzer <snitzer@kernel.org> Signed-off-by: Matthew Sakai <msakai@redhat.com>
2024-03-01dm vdo uds-threads: eliminate uds_*_semaphore interfacesMike Snitzer
The implementation of thread 'barrier' data structure does not require overdone private semaphore wrappers. Also rename the barrier structure's 'mutex' member (a semaphore) to 'lock'. Signed-off-by: Mike Snitzer <snitzer@kernel.org> Signed-off-by: Matthew Sakai <msakai@redhat.com>
2024-03-01dm vdo: make uds_*_semaphore interface private to uds-threads.cMike Snitzer
Signed-off-by: Mike Snitzer <snitzer@kernel.org> Signed-off-by: Matthew Sakai <msakai@redhat.com>
2024-03-01dm vdo block-map: rename page state name from "UDS_FREE" to "FREE"Mike Snitzer
Only used for log message, but no need for "UDS_" prefix. Signed-off-by: Mike Snitzer <snitzer@kernel.org> Signed-off-by: Susan LeGendre-McGhee <slegendr@redhat.com> Signed-off-by: Matthew Sakai <msakai@redhat.com>
2024-03-01dm vdo volume-index: fix an assert statement in ↵Harshit Mogalapalli
start_restoring_volume_sub_index() Use "==" instead of "=" in ASSERT() statement. Fixes: ef074a31e88e ("dm vdo: implement the volume index") Signed-off-by: Harshit Mogalapalli <harshit.m.mogalapalli@oracle.com> Signed-off-by: Susan LeGendre-McGhee <slegendr@redhat.com> Signed-off-by: Matthew Sakai <msakai@redhat.com> Signed-off-by: Mike Snitzer <snitzer@kernel.org>
2024-02-29md/raid1: factor out helpers to choose the best rdev from read_balance()Yu Kuai
The way that best rdev is chosen: 1) If the read is sequential from one rdev: - if rdev is rotational, use this rdev; - if rdev is non-rotational, use this rdev until total read length exceed disk opt io size; 2) If the read is not sequential: - if there is idle disk, use it, otherwise: - if the array has non-rotational disk, choose the rdev with minimal inflight IO; - if all the underlaying disks are rotational disk, choose the rdev with closest IO; There are no functional changes, just to make code cleaner and prepare for following refactor. Co-developed-by: Paul Luse <paul.e.luse@linux.intel.com> Signed-off-by: Paul Luse <paul.e.luse@linux.intel.com> Signed-off-by: Yu Kuai <yukuai3@huawei.com> Reviewed-by: Xiao Ni <xni@redhat.com> Signed-off-by: Song Liu <song@kernel.org> Link: https://lore.kernel.org/r/20240229095714.926789-12-yukuai1@huaweicloud.com
2024-02-29md/raid1: factor out the code to manage sequential IOYu Kuai
There is no functional change for now, make read_balance() cleaner and prepare to fix problems and refactor the handler of sequential IO. Co-developed-by: Paul Luse <paul.e.luse@linux.intel.com> Signed-off-by: Paul Luse <paul.e.luse@linux.intel.com> Signed-off-by: Yu Kuai <yukuai3@huawei.com> Reviewed-by: Xiao Ni <xni@redhat.com> Signed-off-by: Song Liu <song@kernel.org> Link: https://lore.kernel.org/r/20240229095714.926789-11-yukuai1@huaweicloud.com
2024-02-29md/raid1: factor out choose_bb_rdev() from read_balance()Yu Kuai
read_balance() is hard to understand because there are too many status and branches, and it's overlong. This patch factor out the case to read the rdev with bad blocks from read_balance(), there are no functional changes. Co-developed-by: Paul Luse <paul.e.luse@linux.intel.com> Signed-off-by: Paul Luse <paul.e.luse@linux.intel.com> Signed-off-by: Yu Kuai <yukuai3@huawei.com> Reviewed-by: Xiao Ni <xni@redhat.com> Signed-off-by: Song Liu <song@kernel.org> Link: https://lore.kernel.org/r/20240229095714.926789-10-yukuai1@huaweicloud.com
2024-02-29md/raid1: factor out choose_slow_rdev() from read_balance()Yu Kuai
read_balance() is hard to understand because there are too many status and branches, and it's overlong. This patch factor out the case to read the slow rdev from read_balance(), there are no functional changes. Co-developed-by: Paul Luse <paul.e.luse@linux.intel.com> Signed-off-by: Paul Luse <paul.e.luse@linux.intel.com> Signed-off-by: Yu Kuai <yukuai3@huawei.com> Reviewed-by: Xiao Ni <xni@redhat.com> Signed-off-by: Song Liu <song@kernel.org> Link: https://lore.kernel.org/r/20240229095714.926789-9-yukuai1@huaweicloud.com
2024-02-29md/raid1: factor out read_first_rdev() from read_balance()Yu Kuai
read_balance() is hard to understand because there are too many status and branches, and it's overlong. This patch factor out the case to read the first rdev from read_balance(), there are no functional changes. Co-developed-by: Paul Luse <paul.e.luse@linux.intel.com> Signed-off-by: Paul Luse <paul.e.luse@linux.intel.com> Signed-off-by: Yu Kuai <yukuai3@huawei.com> Reviewed-by: Xiao Ni <xni@redhat.com> Signed-off-by: Song Liu <song@kernel.org> Link: https://lore.kernel.org/r/20240229095714.926789-8-yukuai1@huaweicloud.com
2024-02-29md/raid1-10: factor out a new helper raid1_should_read_first()Yu Kuai
If resync is in progress, read_balance() should find the first usable disk, otherwise, data could be inconsistent after resync is done. raid1 and raid10 implement the same checking, hence factor out the checking to make code cleaner. Noted that raid1 is using 'mddev->recovery_cp', which is updated after all resync IO is done, while raid10 is using 'conf->next_resync', which is inaccurate because raid10 update it before submitting resync IO. Fortunately, raid10 read IO can't concurrent with resync IO, hence there is no problem. And this patch also switch raid10 to use 'mddev->recovery_cp'. Co-developed-by: Paul Luse <paul.e.luse@linux.intel.com> Signed-off-by: Paul Luse <paul.e.luse@linux.intel.com> Signed-off-by: Yu Kuai <yukuai3@huawei.com> Reviewed-by: Xiao Ni <xni@redhat.com> Signed-off-by: Song Liu <song@kernel.org> Link: https://lore.kernel.org/r/20240229095714.926789-7-yukuai1@huaweicloud.com
2024-02-29md/raid1-10: add a helper raid1_check_read_range()Yu Kuai
The checking and handler of bad blocks appear many timers during read_balance() in raid1 and raid10. This helper will be used in later patches to simplify read_balance() a lot. Co-developed-by: Paul Luse <paul.e.luse@linux.intel.com> Signed-off-by: Paul Luse <paul.e.luse@linux.intel.com> Signed-off-by: Yu Kuai <yukuai3@huawei.com> Reviewed-by: Xiao Ni <xni@redhat.com> Signed-off-by: Song Liu <song@kernel.org> Link: https://lore.kernel.org/r/20240229095714.926789-6-yukuai1@huaweicloud.com
2024-02-29md/raid1: fix choose next idle in read_balance()Yu Kuai
Commit 12cee5a8a29e ("md/raid1: prevent merging too large request") add the case choose next idle in read_balance(): read_balance: for_each_rdev if(next_seq_sect == this_sector || dist == 0) -> sequential reads best_disk = disk; if (...) choose_next_idle = 1 continue; for_each_rdev -> iterate next rdev if (pending == 0) best_disk = disk; -> choose the next idle disk break; if (choose_next_idle) -> keep using this rdev if there are no other idle disk contine However, commit 2e52d449bcec ("md/raid1: add failfast handling for reads.") remove the code: - /* If device is idle, use it */ - if (pending == 0) { - best_disk = disk; - break; - } Hence choose next idle will never work now, fix this problem by following: 1) don't set best_disk in this case, read_balance() will choose the best disk after iterating all the disks; 2) add 'pending' so that other idle disk will be chosen; 3) add a new local variable 'sequential_disk' to record the disk, and if there is no other idle disk, 'sequential_disk' will be chosen; Fixes: 2e52d449bcec ("md/raid1: add failfast handling for reads.") Co-developed-by: Paul Luse <paul.e.luse@linux.intel.com> Signed-off-by: Paul Luse <paul.e.luse@linux.intel.com> Signed-off-by: Yu Kuai <yukuai3@huawei.com> Reviewed-by: Xiao Ni <xni@redhat.com> Signed-off-by: Song Liu <song@kernel.org> Link: https://lore.kernel.org/r/20240229095714.926789-5-yukuai1@huaweicloud.com
2024-02-29md/raid1: record nonrot rdevs while adding/removing rdevs to confYu Kuai
For raid1, each read will iterate all the rdevs from conf and check if any rdev is non-rotational, then choose rdev with minimal IO inflight if so, or rdev with closest distance otherwise. Disk nonrot info can be changed through sysfs entry: /sys/block/[disk_name]/queue/rotational However, consider that this should only be used for testing, and user really shouldn't do this in real life. Record the number of non-rotational disks in conf, to avoid checking each rdev in IO fast path and simplify read_balance() a little bit. Co-developed-by: Paul Luse <paul.e.luse@linux.intel.com> Signed-off-by: Paul Luse <paul.e.luse@linux.intel.com> Signed-off-by: Yu Kuai <yukuai3@huawei.com> Signed-off-by: Song Liu <song@kernel.org> Link: https://lore.kernel.org/r/20240229095714.926789-4-yukuai1@huaweicloud.com
generated by cgit (git 2.34.1) at 2025-09-24 10:47:04 +0000