search for: get_request_wait

When user hot-unplug a disk which is busy serving I/O, __blk_run_queue might be unable to drain all the requests. As a result, the blk_drain_queue() would loop forever and blk_cleanup_queue would not return. So hot-unplug will fail. This patch adds a callback in blk_drain_queue() for low lever driver to abort requests. Currently, this is useful for virtio-blk to do cleanup in hot-unplug. Cc:

When user hot-unplug a disk which is busy serving I/O, __blk_run_queue might be unable to drain all the requests. As a result, the blk_drain_queue() would loop forever and blk_cleanup_queue would not return. So hot-unplug will fail. This patch adds a callback in blk_drain_queue() for low lever driver to abort requests. Currently, this is useful for virtio-blk to do cleanup in hot-unplug. Cc:

...But I see still that btrfs_start_delalloc_inodes() handles same inode more than once, and in some cases never returns in 15 minutes or more, delaying all other transactions. And snapshot creation gets stuck for all this time. (The stack I see on kernel 3.6 is like this: [<ffffffff812f26c6>] get_request_wait+0xf6/0x1d0 [<ffffffff812f35df>] blk_queue_bio+0x7f/0x380 [<ffffffff812f0374>] generic_make_request.part.50+0x74/0xb0 [<ffffffff812f0788>] generic_make_request+0x68/0x70 [<ffffffff812f0815>] submit_bio+0x85/0x110 [<ffffffffa034ace5>] btrfs_map_bio+0x165/0x2f0 [btrfs] [&...

...se lock (&(&q->__queue_lock)->rlock) at: [<ffffffff813274d2>] blk_queue_bio+0x2a2/0x380 but there are no more locks to release! other info that might help us debug this: 1 lock held by fio/3587: #0: (&(&vblk->lock)->rlock){......}, at: [<ffffffff8132661a>] get_request_wait+0x19a/0x250 Other drivers use block layer provided spinlock as well, e.g. SCSI. Switching to the block layer provided spinlock saves a bit of memory and does not increase lock contention. Performance test shows no real difference is observed before and after this patch. Changes in v2: Improve co...

...se lock (&(&q->__queue_lock)->rlock) at: [<ffffffff813274d2>] blk_queue_bio+0x2a2/0x380 but there are no more locks to release! other info that might help us debug this: 1 lock held by fio/3587: #0: (&(&vblk->lock)->rlock){......}, at: [<ffffffff8132661a>] get_request_wait+0x19a/0x250 Other drivers use block layer provided spinlock as well, e.g. SCSI. Switching to the block layer provided spinlock saves a bit of memory and does not increase lock contention. Performance test shows no real difference is observed before and after this patch. Changes in v2: Improve co...

This patch set fixes the race when hot-unplug stressed disk. Asias He (3): virtio-blk: Call del_gendisk() before disable guest kick virtio-blk: Reset device after blk_cleanup_queue() virtio-blk: Use block layer provided spinlock drivers/block/virtio_blk.c | 25 ++++++------------------- 1 file changed, 6 insertions(+), 19 deletions(-) -- 1.7.10.2

This patch set fixes the race when hot-unplug stressed disk. Asias He (3): virtio-blk: Call del_gendisk() before disable guest kick virtio-blk: Reset device after blk_cleanup_queue() virtio-blk: Use block layer provided spinlock drivers/block/virtio_blk.c | 25 ++++++------------------- 1 file changed, 6 insertions(+), 19 deletions(-) -- 1.7.10.2

...468 [163440.613503] Call Trace: [163440.613520] [<ffffffff810e7e95>] ? kmem_cache_alloc+0x8c/0xf0 [163440.613528] [<ffffffff8102ddc0>] ? pvclock_clocksource_read+0x3a/0x8b [163440.613537] [<ffffffff8130b5f1>] ? io_schedule+0x73/0xb7 [163440.613546] [<ffffffff8118053b>] ? get_request_wait+0xf0/0x188 [163440.613553] [<ffffffff81065d0a>] ? autoremove_wake_function+0x0/0x2e [163440.613559] [<ffffffff811808ca>] ? __make_request+0x2f7/0x428 [163440.613565] [<ffffffff8117f0a7>] ? generic_make_request+0x299/0x2f9 [163440.613574] [<ffffffff81176f2d>] ? elv_rb_lat...

...warn_slowpath_null+0xd/0x10 [ 288.525321] [<c03ada74>] cfq_cic_lookup+0x61/0xce [ 288.525330] [<c03adc27>] cfq_may_queue+0x1e/0xab [ 288.525338] [<c039d892>] elv_may_queue+0x14/0x1b [ 288.525348] [<c03a291f>] get_request+0x24/0x253 [ 288.525356] [<c03a300e>] get_request_wait+0x18/0x12f [ 288.525366] [<c0239077>] ? hrtimer_wakeup+0x0/0x1c [ 288.525375] [<c0236d61>] ? remove_wait_queue+0x31/0x36 [ 288.525383] [<c03a315c>] blk_get_request+0x37/0x58 [ 288.525394] [<c0428cd1>] scsi_execute+0x23/0x11b [ 288.525403] [<c0428e1b>] scsi_ex...

On 06/21/2011 01:15 PM, Jan Stilow wrote: > Hello, > > Nirbheek Chauhan <nirbheek <at> gentoo.org> writes: >> [...] >> >> Every few minutes, (I guess) when applications do fsync (firefox, >> xchat, vim, etc), all applications that use fsync() hang for several >> seconds, and applications that use general IO suffer extreme >> slowdowns.

...d the nr_batch_requests limit. There will always - * be some limit enforced by BLK_BATCH_TIME. - */ - if (ioc_batching(q, ioc)) - ioc->nr_batch_requests--; - trace_block_getrq(q, bio, rw_flags & 1); + + if (drop_lock) + spin_lock_irq(&ctx->lock); out: return rq; } /** * get_request_wait - get a free request with retry * @q: request_queue to allocate request from * @rw_flags: RW and SYNC flags * @bio: bio to allocate request for (can be %NULL) * * Get a free request from @q. This function keeps retrying under memory * pressure and fails iff @q is dead. * - * Must be...

...d the nr_batch_requests limit. There will always - * be some limit enforced by BLK_BATCH_TIME. - */ - if (ioc_batching(q, ioc)) - ioc->nr_batch_requests--; - trace_block_getrq(q, bio, rw_flags & 1); + + if (drop_lock) + spin_lock_irq(&ctx->lock); out: return rq; } /** * get_request_wait - get a free request with retry * @q: request_queue to allocate request from * @rw_flags: RW and SYNC flags * @bio: bio to allocate request for (can be %NULL) * * Get a free request from @q. This function keeps retrying under memory * pressure and fails iff @q is dead. * - * Must be...

...nction+0x0/0x10 Oct 8 02:40:43 (none) kernel: [<c12e67c2>] wait_for_completion+0x12/0x20 Oct 8 02:40:43 (none) kernel: [<c1176448>] blk_execute_rq+0x78/0xd0 Oct 8 02:40:43 (none) kernel: [<c1176310>] ? blk_end_sync_rq+0x0/0x30 Oct 8 02:40:43 (none) kernel: [<c117295d>] ? get_request_wait+0x1d/0x190 Oct 8 02:40:43 (none) kernel: [<c103570d>] ? wakeup_preempt_entity.clone.94+0x8d/0xa0 Oct 8 02:40:43 (none) kernel: [<c1172b2b>] ? blk_get_request+0x5b/0x80 Oct 8 02:40:43 (none) kernel: [<f84318b7>] scsi_execute+0xc7/0x110 [scsi_mod] Oct 8 02:40:43 (none) kernel: [...