From 0fb5b1fb30fba3671dd5b1489d78e93e08d62e4e Mon Sep 17 00:00:00 2001 From: "Martin K. Petersen" Date: Thu, 4 Feb 2016 00:52:12 -0500 Subject: block/sd: Return -EREMOTEIO when WRITE SAME and DISCARD are disabled When a storage device rejects a WRITE SAME command we will disable write same functionality for the device and return -EREMOTEIO to the block layer. -EREMOTEIO will in turn prevent DM from retrying the I/O and/or failing the path. Yiwen Jiang discovered a small race where WRITE SAME requests issued simultaneously would cause -EIO to be returned. This happened because any requests being prepared after WRITE SAME had been disabled for the device caused us to return BLKPREP_KILL. The latter caused the block layer to return -EIO upon completion. To overcome this we introduce BLKPREP_INVALID which indicates that this is an invalid request for the device. blk_peek_request() is modified to return -EREMOTEIO in that case. Reported-by: Yiwen Jiang Suggested-by: Mike Snitzer Reviewed-by: Hannes Reinicke Reviewed-by: Ewan Milne Reviewed-by: Yiwen Jiang Signed-off-by: Martin K. Petersen --- include/linux/blkdev.h | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) (limited to 'include/linux/blkdev.h') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index d372ea87ead5..a9b643a0647f 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -681,9 +681,12 @@ static inline bool blk_write_same_mergeable(struct bio *a, struct bio *b) /* * q->prep_rq_fn return values */ -#define BLKPREP_OK 0 /* serve it */ -#define BLKPREP_KILL 1 /* fatal error, kill */ -#define BLKPREP_DEFER 2 /* leave on queue */ +enum { + BLKPREP_OK, /* serve it */ + BLKPREP_KILL, /* fatal error, kill, return -EIO */ + BLKPREP_DEFER, /* leave on queue */ + BLKPREP_INVALID, /* invalid command, kill, return -EREMOTEIO */ +}; extern unsigned long blk_max_low_pfn, blk_max_pfn; -- cgit From d07ab6d114774d7fcb53c57d7474aef459713451 Mon Sep 17 00:00:00 2001 From: Mika Westerberg Date: Thu, 18 Feb 2016 10:54:11 +0200 Subject: block: Add blk_set_runtime_active() If block device is left runtime suspended during system suspend, resume hook of the driver typically corrects runtime PM status of the device back to "active" after it is resumed. However, this is not enough as queue's runtime PM status is still "suspended". As long as it is in this state blk_pm_peek_request() returns NULL and thus prevents new requests to be processed. Add new function blk_set_runtime_active() that can be used to force the queue status back to "active" as needed. Signed-off-by: Mika Westerberg Acked-by: Jens Axboe Signed-off-by: Tejun Heo --- block/blk-core.c | 24 ++++++++++++++++++++++++ include/linux/blkdev.h | 2 ++ 2 files changed, 26 insertions(+) (limited to 'include/linux/blkdev.h') diff --git a/block/blk-core.c b/block/blk-core.c index ab51685988c2..11e371e76cb7 100644 --- a/block/blk-core.c +++ b/block/blk-core.c @@ -3527,6 +3527,30 @@ void blk_post_runtime_resume(struct request_queue *q, int err) spin_unlock_irq(q->queue_lock); } EXPORT_SYMBOL(blk_post_runtime_resume); + +/** + * blk_set_runtime_active - Force runtime status of the queue to be active + * @q: the queue of the device + * + * If the device is left runtime suspended during system suspend the resume + * hook typically resumes the device and corrects runtime status + * accordingly. However, that does not affect the queue runtime PM status + * which is still "suspended". This prevents processing requests from the + * queue. + * + * This function can be used in driver's resume hook to correct queue + * runtime PM status and re-enable peeking requests from the queue. It + * should be called before first request is added to the queue. + */ +void blk_set_runtime_active(struct request_queue *q) +{ + spin_lock_irq(q->queue_lock); + q->rpm_status = RPM_ACTIVE; + pm_runtime_mark_last_busy(q->dev); + pm_request_autosuspend(q->dev); + spin_unlock_irq(q->queue_lock); +} +EXPORT_SYMBOL(blk_set_runtime_active); #endif int __init blk_dev_init(void) diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 29189aeace19..8a77c47f8d21 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -1026,6 +1026,7 @@ extern int blk_pre_runtime_suspend(struct request_queue *q); extern void blk_post_runtime_suspend(struct request_queue *q, int err); extern void blk_pre_runtime_resume(struct request_queue *q); extern void blk_post_runtime_resume(struct request_queue *q, int err); +extern void blk_set_runtime_active(struct request_queue *q); #else static inline void blk_pm_runtime_init(struct request_queue *q, struct device *dev) {} @@ -1036,6 +1037,7 @@ static inline int blk_pre_runtime_suspend(struct request_queue *q) static inline void blk_post_runtime_suspend(struct request_queue *q, int err) {} static inline void blk_pre_runtime_resume(struct request_queue *q) {} static inline void blk_post_runtime_resume(struct request_queue *q, int err) {} +extern inline void blk_set_runtime_active(struct request_queue *q) {} #endif /* -- cgit From e0af29171aa8912e1ca95023b75ef336cd70d661 Mon Sep 17 00:00:00 2001 From: Ming Lei Date: Fri, 26 Feb 2016 23:40:51 +0800 Subject: block: check virt boundary in bio_will_gap() In the following patch, the way for figuring out the last bvec will be changed with a bit cost introduced, so return immediately if the queue doesn't have virt boundary limit. Actually most of devices have not this limit. Reviewed-by: Sagi Grimberg Reviewed-by: Christoph Hellwig Signed-off-by: Ming Lei Signed-off-by: Jens Axboe --- include/linux/blkdev.h | 16 +++++++++++----- 1 file changed, 11 insertions(+), 5 deletions(-) (limited to 'include/linux/blkdev.h') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 4571ef1a12a9..cd06a41581b3 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -1372,6 +1372,13 @@ static inline void put_dev_sector(Sector p) page_cache_release(p.v); } +static inline bool __bvec_gap_to_prev(struct request_queue *q, + struct bio_vec *bprv, unsigned int offset) +{ + return offset || + ((bprv->bv_offset + bprv->bv_len) & queue_virt_boundary(q)); +} + /* * Check if adding a bio_vec after bprv with offset would create a gap in * the SG list. Most drivers don't care about this, but some do. @@ -1381,18 +1388,17 @@ static inline bool bvec_gap_to_prev(struct request_queue *q, { if (!queue_virt_boundary(q)) return false; - return offset || - ((bprv->bv_offset + bprv->bv_len) & queue_virt_boundary(q)); + return __bvec_gap_to_prev(q, bprv, offset); } static inline bool bio_will_gap(struct request_queue *q, struct bio *prev, struct bio *next) { - if (!bio_has_data(prev)) + if (!bio_has_data(prev) || !queue_virt_boundary(q)) return false; - return bvec_gap_to_prev(q, &prev->bi_io_vec[prev->bi_vcnt - 1], - next->bi_io_vec[0].bv_offset); + return __bvec_gap_to_prev(q, &prev->bi_io_vec[prev->bi_vcnt - 1], + next->bi_io_vec[0].bv_offset); } static inline bool req_gap_back_merge(struct request *req, struct bio *bio) -- cgit From 25e71a99f10e444cd00bb2ebccb11e1c9fb672b1 Mon Sep 17 00:00:00 2001 From: Ming Lei Date: Fri, 26 Feb 2016 23:40:52 +0800 Subject: block: get the 1st and last bvec via helpers This patch applies the two introduced helpers to figure out the 1st and last bvec, and fixes the original way after bio splitting. Cc: stable@vger.kernel.org Reported-by: Sagi Grimberg Reviewed-by: Sagi Grimberg Reviewed-by: Christoph Hellwig Signed-off-by: Ming Lei Signed-off-by: Jens Axboe --- include/linux/blkdev.h | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-) (limited to 'include/linux/blkdev.h') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index cd06a41581b3..d7f6bca707ef 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -1394,11 +1394,16 @@ static inline bool bvec_gap_to_prev(struct request_queue *q, static inline bool bio_will_gap(struct request_queue *q, struct bio *prev, struct bio *next) { - if (!bio_has_data(prev) || !queue_virt_boundary(q)) - return false; + if (bio_has_data(prev) && queue_virt_boundary(q)) { + struct bio_vec pb, nb; + + bio_get_last_bvec(prev, &pb); + bio_get_first_bvec(next, &nb); - return __bvec_gap_to_prev(q, &prev->bi_io_vec[prev->bi_vcnt - 1], - next->bi_io_vec[0].bv_offset); + return __bvec_gap_to_prev(q, &pb, nb.bv_offset); + } + + return false; } static inline bool req_gap_back_merge(struct request *req, struct bio *bio) -- cgit From f21018427cb007a0894c36ad702990ab639cbbb4 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 3 Mar 2016 14:43:45 -0700 Subject: block: fix blk_rq_get_max_sectors for driver private requests Driver private request types should not get the artifical cap for the FS requests. This is important to use the full device capabilities for internal command or NVMe pass through commands. Signed-off-by: Christoph Hellwig Reported-by: Jeff Lien Tested-by: Jeff Lien Reviewed-by: Keith Busch Updated by me to use an explicit check for the one command type that does support extended checking, instead of relying on the ordering of the enum command values - as suggested by Keith. Signed-off-by: Jens Axboe --- include/linux/blkdev.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux/blkdev.h') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index d7f6bca707ef..413c84fbc4ed 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -895,7 +895,7 @@ static inline unsigned int blk_rq_get_max_sectors(struct request *rq) { struct request_queue *q = rq->q; - if (unlikely(rq->cmd_type == REQ_TYPE_BLOCK_PC)) + if (unlikely(rq->cmd_type != REQ_TYPE_FS)) return q->limits.max_hw_sectors; if (!q->limits.chunk_sectors || (rq->cmd_flags & REQ_DISCARD)) -- cgit From 8e0b60b96ba06d826a2b26e23b1986853a4e5291 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 3 Mar 2016 16:04:03 +0100 Subject: blk-mq: enable polling support by default Now that applications need to explicitly ask for polling we can enable it by default in blk-mq drivers. Note that this will only have an affect on driver that supply a poll function, which currently only includes nvme. Signed-off-by: Christoph Hellwig Reviewed-by: Stephen Bates Tested-by: Stephen Bates Reviewed-by: Jeff Moyer Signed-off-by: Al Viro --- include/linux/blkdev.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include/linux/blkdev.h') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 29189aeace19..ff33a1ab58c9 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -499,7 +499,8 @@ struct request_queue { #define QUEUE_FLAG_MQ_DEFAULT ((1 << QUEUE_FLAG_IO_STAT) | \ (1 << QUEUE_FLAG_STACKABLE) | \ - (1 << QUEUE_FLAG_SAME_COMP)) + (1 << QUEUE_FLAG_SAME_COMP) | \ + (1 << QUEUE_FLAG_POLL)) static inline void queue_lockdep_assert_held(struct request_queue *q) { -- cgit