- 04 Nov, 2016 2 commits
-
-
Shaohua Li authored
This is corresponding part for blk-mq. Disk with multiple hardware queues doesn't need this as we only hold 1 request at most. Signed-off-by:
Shaohua Li <shli@fb.com> Signed-off-by:
Jens Axboe <axboe@fb.com>
-
Shaohua Li authored
Currently block plug holds up to 16 non-mergeable requests. This makes sense if the request size is small, eg, reduce lock contention. But if request size is big enough, we don't need to worry about lock contention. Holding such request makes no sense and it lows the disk utilization. In practice, this improves 10% throughput for my raid5 sequential write workload. The size (128k) is arbitrary right now, but it makes sure lock contention is small. This probably could be more intelligent, eg, check average request size holded. Since this is mainly for sequential IO, probably not worthy. V2: check the last request instead of the first request, so as long as there is one big size request we flush the plug. Signed-off-by:
-
- 03 Nov, 2016 1 commit
-
-
Johannes Thumshirn authored
bsg_validate_sgv4_hdr() doesn't care about the request_queue, so drop it from it's arguments. Signed-off-by:
Johannes Thumshirn <jthumshirn@suse.de> Signed-off-by:
-
- 02 Nov, 2016 16 commits
-
-
Bart Van Assche authored
Make nvme_requeue_req() check BLK_MQ_S_STOPPED instead of QUEUE_FLAG_STOPPED. Remove the QUEUE_FLAG_STOPPED manipulations that became superfluous because of this change. Change blk_queue_stopped() tests into blk_mq_queue_stopped(). This patch fixes a race condition: using queue_flag_clear_unlocked() is not safe if any other function that manipulates the queue flags can be called concurrently, e.g. blk_cleanup_queue(). Signed-off-by:
Bart Van Assche <bart.vanassche@sandisk.com> Cc: Keith Busch <keith.busch@intel.com> Cc: Sagi Grimberg <sagi@grimberg.me> Reviewed-by:
Christoph Hellwig <hch@lst.de> Signed-off-by:
-
Bart Van Assche authored
Avoid that nvme_queue_rq() is still running when nvme_stop_queues() returns. Signed-off-by:
Sagi Grimberg <sagi@grimberg.me> Reviewed-by:
-
Bart Van Assche authored
Ensure that all ongoing dm_mq_queue_rq() and dm_mq_requeue_request() calls have stopped before setting the "queue stopped" flag. This allows to remove the "queue stopped" test from dm_mq_queue_rq() and dm_mq_requeue_request(). This patch fixes a race condition because dm_mq_queue_rq() is called without holding the queue lock and hence BLK_MQ_S_STOPPED can be set at any time while dm_mq_queue_rq() is in progress. This patch prevents that the following hang occurs sporadically when using dm-mq: INFO: task systemd-udevd:10111 blocked for more than 480 seconds. Call Trace: [<ffffffff8161f397>] schedule+0x37/0x90 [<ffffffff816239ef>] schedule_timeout+0x27f/0x470 [<ffffffff8161e76f>] io_schedule_timeout+0x9f/0x110 [<ffffffff8161fb36>] bit_wait_io+0x16/0x60 [<ffffffff8161f929>] __wait_on_bit_lock+0x49/0xa0 [<ffffffff8114fe69>] __lock_page+0xb9/0xc0 [<ffffffff81165d90>] truncate_inode_pages_range+0x3e0/0x760 [<ffffffff81166120>] truncate_inode_pages+0x10/0x20 [<ffffffff81212a20>] kill_bdev+0x30/0x40 [<ffffffff81213d41>] __blkdev_put+0x71/0x360 [<ffffffff81214079>] blkdev_put+0x49/0x170 [<ffffffff812141c0>] blkdev_close+0x20/0x30 [<ffffffff811d48e8>] __fput+0xe8/0x1f0 [<ffffffff811d4a29>] ____fput+0x9/0x10 [<ffffffff810842d3>] task_work_run+0x83/0xb0 [<ffffffff8106606e>] do_exit+0x3ee/0xc40 [<ffffffff8106694b>] do_group_exit+0x4b/0xc0 [<ffffffff81073d9a>] get_signal+0x2ca/0x940 [<ffffffff8101bf43>] do_signal+0x23/0x660 [<ffffffff810022b3>] exit_to_usermode_loop+0x73/0xb0 [<ffffffff81002cb0>] syscall_return_slowpath+0xb0/0xc0 [<ffffffff81624e33>] entry_SYSCALL_64_fastpath+0xa6/0xa8 Signed-off-by:
Mike Snitzer <snitzer@redhat.com> Reviewed-by:
Hannes Reinecke <hare@suse.com> Reviewed-by:
-
Bart Van Assche authored
Instead of manipulating both QUEUE_FLAG_STOPPED and BLK_MQ_S_STOPPED in the dm start and stop queue functions, only manipulate the latter flag. Change blk_queue_stopped() tests into blk_mq_queue_stopped(). Signed-off-by:
-
Bart Van Assche authored
Most blk_mq_requeue_request() and blk_mq_add_to_requeue_list() calls are followed by kicking the requeue list. Hence add an argument to these two functions that allows to kick the requeue list. This was proposed by Christoph Hellwig. Signed-off-by:
-
Bart Van Assche authored
blk_mq_quiesce_queue() waits until ongoing .queue_rq() invocations have finished. This function does *not* wait until all outstanding requests have finished (this means invocation of request.end_io()). The algorithm used by blk_mq_quiesce_queue() is as follows: * Hold either an RCU read lock or an SRCU read lock around .queue_rq() calls. The former is used if .queue_rq() does not block and the latter if .queue_rq() may block. * blk_mq_quiesce_queue() first calls blk_mq_stop_hw_queues() followed by synchronize_srcu() or synchronize_rcu(). The latter call waits for .queue_rq() invocations that started before blk_mq_quiesce_queue() was called. * The blk_mq_hctx_stopped() calls that control whether or not .queue_rq() will be called are called with the (S)RCU read lock held. This is necessary to avoid race conditions against blk_mq_quiesce_queue(). Signed-off-by:
Ming Lei <tom.leiming@gmail.com> Reviewed-by:
-
Bart Van Assche authored
Since blk_mq_requeue_work() no longer restarts stopped queues canceling requeue work is no longer needed to prevent that a stopped queue would be restarted. Hence remove this function. Signed-off-by:
-
Bart Van Assche authored
Since blk_mq_requeue_work() starts stopped queues and since execution of this function can be scheduled after a queue has been stopped it is not possible to stop queues without using an additional state variable to track whether or not the queue has been stopped. Hence modify blk_mq_requeue_work() such that it does not start stopped queues. My conclusion after a review of the blk_mq_stop_hw_queues() and blk_mq_{delay_,}kick_requeue_list() callers is as follows: * In the dm driver starting and stopping queues should only happen if __dm_suspend() or __dm_resume() is called and not if the requeue list is processed. * In the SCSI core queue stopping and starting should only be performed by the scsi_internal_device_block() and scsi_internal_device_unblock() functions but not by any other function. Although the blk_mq_stop_hw_queue() call in scsi_queue_rq() may help to reduce CPU load if a LLD queue is full, figuring out whether or not a queue should be restarted when requeueing a command would require to introduce additional locking in scsi_mq_requeue_cmd() to avoid a race with scsi_internal_device_block(). Avoid this complexity by removing the blk_mq_stop_hw_queue() call from scsi_queue_rq(). * In the NVMe core only the functions that call blk_mq_start_stopped_hw_queues() explicitly should start stopped queues. * A blk_mq_start_stopped_hwqueues() call must be added in the xen-blkfront driver in its blkif_recover() function. Signed-off-by: -
Bart Van Assche authored
Move the "hctx stopped" test and the insert request calls into blk_mq_direct_issue_request(). Rename that function into blk_mq_try_issue_directly() to reflect its new semantics. Pass the hctx pointer to that function instead of looking it up a second time. These changes avoid that code has to be duplicated in the next patch. Signed-off-by:
-
Bart Van Assche authored
The function blk_queue_stopped() allows to test whether or not a traditional request queue has been stopped. Introduce a helper function that allows block drivers to query easily whether or not one or more hardware contexts of a blk-mq queue have been stopped. Signed-off-by:
-
Bart Van Assche authored
Multiple functions test the BLK_MQ_S_STOPPED bit so introduce a helper function that performs this test. Signed-off-by:
-
Bart Van Assche authored
The meaning of the BLK_MQ_S_STOPPED flag is "do not call .queue_rq()". Hence modify blk_mq_make_request() such that requests are queued instead of issued if a queue has been stopped. Reported-by:
-
Kent Overstreet authored
This is a helper that pins down a range from an iov_iter and adds it to a bio without requiring a separate memory allocation for the page array. It will be used for upcoming direct I/O implementations for block devices and iomap based file systems. Signed-off-by:
Kent Overstreet <kent.overstreet@gmail.com> [hch: ported to the iov_iter interface, renamed and added comments. All blame should be directed to me and all fame should go to Kent after this!] Signed-off-by:
-
Jens Axboe authored
If we're doing background type writes, then use the appropriate background write flags for that. Signed-off-by:
-
Jens Axboe authored
Add wbc_to_write_flags(), which returns the write modifier flags to use, based on a struct writeback_control. No functional changes in this patch, but it prepares us for factoring other wbc fields for write type. Signed-off-by:
Jan Kara <jack@suse.cz> Reviewed-by:
-
Jens Axboe authored
This adds a new request flag, REQ_BACKGROUND, that callers can use to tell the block layer that this is background (non-urgent) IO. Signed-off-by:
-
- 01 Nov, 2016 16 commits
-
-
Christoph Hellwig authored
Now that we have a separate header for struct bio_vec there is absolutely no excuse for including this header from non-block I/O code. Signed-off-by:
-
Christoph Hellwig authored
It's only needed for the CONFIG_SWAP-only use of bio_end_io_t. Because CONFIG_SWAP implies CONFIG_BLOCK this will allow to drop some ifdefs in blk_types.h. Instead we'll need to add a few explicit includes that were implicit before, though. Signed-off-by:
-
Christoph Hellwig authored
The file only needs the struct bvec_iter delcaration, which is available from bvec.h. Signed-off-by:
-
Christoph Hellwig authored
No need for it - we only use struct bio_vec in prototypes and already have forward declarations for it. Signed-off-by:
-
Christoph Hellwig authored
Nothing in fs.h should require blk_types.h to be included. Signed-off-by:
-
Christoph Hellwig authored
This is where all the other bio operations live, so users must include bio.h anyway. Signed-off-by:
-
Christoph Hellwig authored
Move READ and WRITE to kernel.h and don't define them in terms of block layer ops; they are our generic data direction indicators these days and have no more resemblance with the block layer ops. Signed-off-by:
-
Christoph Hellwig authored
Remove the WRITE_* and READ_SYNC wrappers, and just use the flags directly. Where applicable this also drops usage of the bio_set_op_attrs wrapper. Signed-off-by:
-
Christoph Hellwig authored
Noidle should be the default for writes as seen by all the compounds definitions in fs.h using it. In fact only direct I/O really should be using NODILE, so turn the whole flag around to get the defaults right, which will make our life much easier especially onces the WRITE_* defines go away. This assumes all the existing "raw" users of REQ_SYNC for writes want noidle behavior, which seems to be spot on from a quick audit. Signed-off-by:
-
Christoph Hellwig authored
Instead of requiring everyone to specify the REQ_SYNC flag aѕ well. Signed-off-by:
-
Christoph Hellwig authored
Reads are synchronous per definition, don't add another flag for it. Signed-off-by:
-
Christoph Hellwig authored
Signed-off-by:
-
Christoph Hellwig authored
(and remove one layer of masking for the op_is_write call next to it). Signed-off-by:
-
Christoph Hellwig authored
Signed-off-by:
-
Christoph Hellwig authored
Signed-off-by:
-
Christoph Hellwig authored
Signed-off-by:
-
- 28 Oct, 2016 5 commits
-
-
Christoph Hellwig authored
Currently the block layer op_is_write, bio_data_dir and rq_data_dir helper treat every operation that is not a READ as a data out operation. This worked surprisingly long, but the new REQ_OP_ZONE_REPORT operation actually adds a second operation that reads data from the device. Surprisingly nothing critical relied on this direction, but this might be a good opportunity to properly fix this issue up. We take a little inspiration and use the least significant bit of the operation number to encode the data direction, which just requires us to renumber the operations to fix this scheme. Signed-off-by:
Shaun Tancheff <shaun.tancheff@seagate.com> Signed-off-by:
-
Christoph Hellwig authored
Now that we don't need the common flags to overflow outside the range of a 32-bit type we can encode them the same way for both the bio and request fields. This in addition allows us to place the operation first (and make some room for more ops while we're at it) and to stop having to shift around the operation values. In addition this allows passing around only one value in the block layer instead of two (and eventuall also in the file systems, but we can do that later) and thus clean up a lot of code. Last but not least this allows decreasing the size of the cmd_flags field in struct request to 32-bits. Various functions passing this value could also be updated, but I'd like to avoid the churn for now. Signed-off-by:
-
Christoph Hellwig authored
A lot of the REQ_* flags are only used on struct requests, and only of use to the block layer and a few drivers that dig into struct request internals. This patch adds a new req_flags_t rq_flags field to struct request for them, and thus dramatically shrinks the number of common requests. It also removes the unfortunate situation where we have to fit the fields from the same enum into 32 bits for struct bio and 64 bits for struct request. Signed-off-by:
-
Christoph Hellwig authored
It's the last bio-only REQ_* flag, and we have space for it in the bio bi_flags field. Signed-off-by:
-
Christoph Hellwig authored
The information that am I/O is a read-ahead can be useful for drivers. In fact the NVMe driver already checks it, even if it won't ever be set at the moment. Signed-off-by:
-
