From 6f491a8d4b92d1a840fd9209cba783c84437d0b7 Mon Sep 17 00:00:00 2001 From: Ming Lei Date: Wed, 27 Nov 2024 21:51:28 +0800 Subject: block: track disk DEAD state automatically for modeling queue freeze lockdep Now we only verify the outmost freeze & unfreeze in current context in case that !q->mq_freeze_depth, so it is reliable to save disk DEAD state when we want to lock the freeze queue since the state is one per-task variable now. Doing this way can kill lots of false positive when freeze queue is called before adding disk[1]. [1] https://lore.kernel.org/linux-block/6741f6b2.050a0220.1cc393.0017.GAE@google.com/ Signed-off-by: Ming Lei Link: https://lore.kernel.org/r/20241127135133.3952153-3-ming.lei@redhat.com Signed-off-by: Jens Axboe --- include/linux/blkdev.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 378d3a1a22fc..522cf8eef66c 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -581,6 +581,8 @@ struct request_queue { #ifdef CONFIG_LOCKDEP struct task_struct *mq_freeze_owner; int mq_freeze_owner_depth; + /* Records disk state in current context, used in unfreeze queue */ + bool mq_freeze_disk_dead; #endif wait_queue_head_t mq_freeze_wq; /* -- cgit v1.2.3 From f6661b1d0525f3764596a1b65eeed9e75aecafa7 Mon Sep 17 00:00:00 2001 From: Ming Lei Date: Wed, 27 Nov 2024 21:51:30 +0800 Subject: block: track queue dying state automatically for modeling queue freeze lockdep Now we only verify the outmost freeze & unfreeze in current context in case that !q->mq_freeze_depth, so it is reliable to save queue lying state when we want to lock the freeze queue since the state is one per-task variable now. Signed-off-by: Ming Lei Link: https://lore.kernel.org/r/20241127135133.3952153-5-ming.lei@redhat.com Signed-off-by: Jens Axboe --- include/linux/blkdev.h | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 522cf8eef66c..5d40af2ef971 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -581,8 +581,12 @@ struct request_queue { #ifdef CONFIG_LOCKDEP struct task_struct *mq_freeze_owner; int mq_freeze_owner_depth; - /* Records disk state in current context, used in unfreeze queue */ + /* + * Records disk & queue state in current context, used in unfreeze + * queue + */ bool mq_freeze_disk_dead; + bool mq_freeze_queue_dying; #endif wait_queue_head_t mq_freeze_wq; /* -- cgit v1.2.3 From 5c292ac6e69f390179b93dc104b40903cddce636 Mon Sep 17 00:00:00 2001 From: John Garry Date: Mon, 2 Dec 2024 11:19:56 +0000 Subject: block: Delete bio_prio() Since commit 43b62ce3ff0a ("block: move bio io prio to a new field"), macro bio_prio() does nothing but return the value in bio->bi_ioprio. Most other places just read bio->bi_ioprio directly, so replace bi_ioprio() callsites with reading bio->bi_ioprio directly and delete that macro. Signed-off-by: John Garry Reviewed-by: Chaitanya Kulkarni Link: https://lore.kernel.org/r/20241202111957.2311683-2-john.g.garry@oracle.com Signed-off-by: Jens Axboe --- include/linux/bio.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include') diff --git a/include/linux/bio.h b/include/linux/bio.h index 7a1b3b1a8fed..99676916f3db 100644 --- a/include/linux/bio.h +++ b/include/linux/bio.h @@ -19,7 +19,6 @@ static inline unsigned int bio_max_segs(unsigned int nr_segs) return min(nr_segs, BIO_MAX_VECS); } -#define bio_prio(bio) (bio)->bi_ioprio #define bio_set_prio(bio, prio) ((bio)->bi_ioprio = prio) #define bio_iter_iovec(bio, iter) \ -- cgit v1.2.3 From 19206d3f5ef7f051056d2fb49203a347e4844e6e Mon Sep 17 00:00:00 2001 From: John Garry Date: Mon, 2 Dec 2024 11:19:57 +0000 Subject: block: Delete bio_set_prio() Since commit 43b62ce3ff0a ("block: move bio io prio to a new field"), macro bio_set_prio() does nothing but set bio->bi_ioprio. All other places just set bio->bi_ioprio directly, so replace bio_set_prio() remaining callsites with setting bio->bi_ioprio directly and delete that macro. Signed-off-by: John Garry Acked-by: Jack Wang Reviewed-by: Chaitanya Kulkarni Link: https://lore.kernel.org/r/20241202111957.2311683-3-john.g.garry@oracle.com Signed-off-by: Jens Axboe --- include/linux/bio.h | 2 -- 1 file changed, 2 deletions(-) (limited to 'include') diff --git a/include/linux/bio.h b/include/linux/bio.h index 99676916f3db..1eec59699100 100644 --- a/include/linux/bio.h +++ b/include/linux/bio.h @@ -19,8 +19,6 @@ static inline unsigned int bio_max_segs(unsigned int nr_segs) return min(nr_segs, BIO_MAX_VECS); } -#define bio_set_prio(bio, prio) ((bio)->bi_ioprio = prio) - #define bio_iter_iovec(bio, iter) \ bvec_iter_bvec((bio)->bi_io_vec, (iter)) -- cgit v1.2.3 From fea4952df0eeec4e1a295ebaac9f61c0065fae87 Mon Sep 17 00:00:00 2001 From: Daniel Wagner Date: Mon, 2 Dec 2024 15:00:09 +0100 Subject: driver core: bus: add irq_get_affinity callback to bus_type Introducing a callback in struct bus_type so that a subsystem can hook up the getters directly. This approach avoids exposing random getters in any subsystems APIs. Acked-by: Bjorn Helgaas Reviewed-by: Christoph Hellwig Reviewed-by: Hannes Reinecke Reviewed-by: Ming Lei Acked-by: Greg Kroah-Hartman Signed-off-by: Daniel Wagner Link: https://lore.kernel.org/r/20241202-refactor-blk-affinity-helpers-v6-1-27211e9c2cd5@kernel.org Signed-off-by: Jens Axboe --- include/linux/device/bus.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include') diff --git a/include/linux/device/bus.h b/include/linux/device/bus.h index cdc4757217f9..b18658bce2c3 100644 --- a/include/linux/device/bus.h +++ b/include/linux/device/bus.h @@ -48,6 +48,7 @@ struct fwnode_handle; * will never get called until they do. * @remove: Called when a device removed from this bus. * @shutdown: Called at shut-down time to quiesce the device. + * @irq_get_affinity: Get IRQ affinity mask for the device on this bus. * * @online: Called to put the device back online (after offlining it). * @offline: Called to put the device offline for hot-removal. May fail. @@ -87,6 +88,8 @@ struct bus_type { void (*sync_state)(struct device *dev); void (*remove)(struct device *dev); void (*shutdown)(struct device *dev); + const struct cpumask *(*irq_get_affinity)(struct device *dev, + unsigned int irq_vec); int (*online)(struct device *dev); int (*offline)(struct device *dev); -- cgit v1.2.3 From 1452e9b470c903fc4137a448e9f5767e92d68229 Mon Sep 17 00:00:00 2001 From: Daniel Wagner Date: Mon, 2 Dec 2024 15:00:12 +0100 Subject: blk-mq: introduce blk_mq_map_hw_queues blk_mq_pci_map_queues and blk_mq_virtio_map_queues will create a CPU to hardware queue mapping based on affinity information. These two function share common code and only differ on how the affinity information is retrieved. Also, those functions are located in the block subsystem where it doesn't really fit in. They are virtio and pci subsystem specific. Thus introduce provide a generic mapping function which uses the irq_get_affinity callback from bus_type. Originally idea from Ming Lei Reviewed-by: Christoph Hellwig Reviewed-by: Hannes Reinecke Reviewed-by: Ming Lei Reviewed-by: John Garry Signed-off-by: Daniel Wagner Link: https://lore.kernel.org/r/20241202-refactor-blk-affinity-helpers-v6-4-27211e9c2cd5@kernel.org Signed-off-by: Jens Axboe --- include/linux/blk-mq.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include') diff --git a/include/linux/blk-mq.h b/include/linux/blk-mq.h index c596e0e4cb75..769eab6247d4 100644 --- a/include/linux/blk-mq.h +++ b/include/linux/blk-mq.h @@ -921,6 +921,8 @@ void blk_mq_unfreeze_queue_non_owner(struct request_queue *q); void blk_freeze_queue_start_non_owner(struct request_queue *q); void blk_mq_map_queues(struct blk_mq_queue_map *qmap); +void blk_mq_map_hw_queues(struct blk_mq_queue_map *qmap, + struct device *dev, unsigned int offset); void blk_mq_update_nr_hw_queues(struct blk_mq_tag_set *set, int nr_hw_queues); void blk_mq_quiesce_queue_nowait(struct request_queue *q); -- cgit v1.2.3 From 9bc1e897a821f19ba3775bb013a8a6fb121c3ca1 Mon Sep 17 00:00:00 2001 From: Daniel Wagner Date: Mon, 2 Dec 2024 15:00:16 +0100 Subject: blk-mq: remove unused queue mapping helpers There are no users left of the pci and virtio queue mapping helpers. Thus remove them. Reviewed-by: Christoph Hellwig Reviewed-by: Hannes Reinecke Reviewed-by: Ming Lei Reviewed-by: John Garry Signed-off-by: Daniel Wagner Link: https://lore.kernel.org/r/20241202-refactor-blk-affinity-helpers-v6-8-27211e9c2cd5@kernel.org Signed-off-by: Jens Axboe --- include/linux/blk-mq-pci.h | 11 ----------- include/linux/blk-mq-virtio.h | 11 ----------- 2 files changed, 22 deletions(-) delete mode 100644 include/linux/blk-mq-pci.h delete mode 100644 include/linux/blk-mq-virtio.h (limited to 'include') diff --git a/include/linux/blk-mq-pci.h b/include/linux/blk-mq-pci.h deleted file mode 100644 index ca544e1d3508..000000000000 --- a/include/linux/blk-mq-pci.h +++ /dev/null @@ -1,11 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0 */ -#ifndef _LINUX_BLK_MQ_PCI_H -#define _LINUX_BLK_MQ_PCI_H - -struct blk_mq_queue_map; -struct pci_dev; - -void blk_mq_pci_map_queues(struct blk_mq_queue_map *qmap, struct pci_dev *pdev, - int offset); - -#endif /* _LINUX_BLK_MQ_PCI_H */ diff --git a/include/linux/blk-mq-virtio.h b/include/linux/blk-mq-virtio.h deleted file mode 100644 index 13226e9b22dd..000000000000 --- a/include/linux/blk-mq-virtio.h +++ /dev/null @@ -1,11 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0 */ -#ifndef _LINUX_BLK_MQ_VIRTIO_H -#define _LINUX_BLK_MQ_VIRTIO_H - -struct blk_mq_queue_map; -struct virtio_device; - -void blk_mq_virtio_map_queues(struct blk_mq_queue_map *qmap, - struct virtio_device *vdev, int first_vec); - -#endif /* _LINUX_BLK_MQ_VIRTIO_H */ -- cgit v1.2.3 From cc76ace465d6977b47daa427379b7be1e0976f12 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 19 Dec 2024 07:01:59 +0100 Subject: block: remove BLK_MQ_F_SHOULD_MERGE BLK_MQ_F_SHOULD_MERGE is set for all tag_sets except those that purely process passthrough commands (bsg-lib, ufs tmf, various nvme admin queues) and thus don't even check the flag. Remove it to simplify the driver interface. Signed-off-by: Christoph Hellwig Link: https://lore.kernel.org/r/20241219060214.1928848-1-hch@lst.de Signed-off-by: Jens Axboe --- include/linux/blk-mq.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include') diff --git a/include/linux/blk-mq.h b/include/linux/blk-mq.h index 769eab6247d4..7f6c482ebf54 100644 --- a/include/linux/blk-mq.h +++ b/include/linux/blk-mq.h @@ -668,7 +668,6 @@ struct blk_mq_ops { /* Keep hctx_flag_name[] in sync with the definitions below */ enum { - BLK_MQ_F_SHOULD_MERGE = 1 << 0, BLK_MQ_F_TAG_QUEUE_SHARED = 1 << 1, /* * Set when this device requires underlying blk-mq device for -- cgit v1.2.3 From 6aeb4f836480617be472de767c4cb09c1060a067 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Fri, 3 Jan 2025 08:33:57 +0100 Subject: block: remove bio_add_pc_page Lift bio_split_rw_at into blk_rq_append_bio so that it validates the hardware limits. With this all passthrough callers can simply add bio_add_page to build the bio and delay checking for exceeding of limits to this point instead of doing it for each page. While this looks like adding a new expensive loop over all bio_vecs, blk_rq_append_bio is already doing that just to counter the number of segments. Signed-off-by: Christoph Hellwig Reviewed-by: Sagi Grimberg Link: https://lore.kernel.org/r/20250103073417.459715-2-hch@lst.de Signed-off-by: Jens Axboe --- include/linux/bio.h | 2 -- 1 file changed, 2 deletions(-) (limited to 'include') diff --git a/include/linux/bio.h b/include/linux/bio.h index 1eec59699100..4b79bf50f4f0 100644 --- a/include/linux/bio.h +++ b/include/linux/bio.h @@ -413,8 +413,6 @@ int __must_check bio_add_page(struct bio *bio, struct page *page, unsigned len, unsigned off); bool __must_check bio_add_folio(struct bio *bio, struct folio *folio, size_t len, size_t off); -extern int bio_add_pc_page(struct request_queue *, struct bio *, struct page *, - unsigned int, unsigned int); void __bio_add_page(struct bio *bio, struct page *page, unsigned int len, unsigned int off); void bio_add_folio_nofail(struct bio *bio, struct folio *folio, size_t len, -- cgit v1.2.3 From 02ee5d69e3baf2796ba75b928fcbc9cf7884c5e9 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Fri, 3 Jan 2025 08:33:58 +0100 Subject: block: remove blk_rq_bio_prep There is not real point in a helper just to assign three values to four fields, especially when the surrounding code is working on the neighbor fields directly. Signed-off-by: Christoph Hellwig Reviewed-by: Sagi Grimberg Link: https://lore.kernel.org/r/20250103073417.459715-3-hch@lst.de Signed-off-by: Jens Axboe --- include/linux/blk-mq.h | 8 -------- 1 file changed, 8 deletions(-) (limited to 'include') diff --git a/include/linux/blk-mq.h b/include/linux/blk-mq.h index 7f6c482ebf54..6340293511c9 100644 --- a/include/linux/blk-mq.h +++ b/include/linux/blk-mq.h @@ -978,14 +978,6 @@ static inline void blk_mq_cleanup_rq(struct request *rq) rq->q->mq_ops->cleanup_rq(rq); } -static inline void blk_rq_bio_prep(struct request *rq, struct bio *bio, - unsigned int nr_segs) -{ - rq->nr_phys_segments = nr_segs; - rq->__data_len = bio->bi_iter.bi_size; - rq->bio = rq->biotail = bio; -} - void blk_mq_hctx_set_fq_lock_class(struct blk_mq_hw_ctx *hctx, struct lock_class_key *key); -- cgit v1.2.3 From 2caca8fc7aad9ea9a6ea3ed26ed146b1e5f06fab Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 6 Jan 2025 09:14:37 +0100 Subject: block: use page_to_phys in bvec_phys Use page_to_phys instead of open coding it now that it is available in an architecture independent way. Signed-off-by: Christoph Hellwig Link: https://lore.kernel.org/r/20250106081437.798213-1-hch@lst.de Signed-off-by: Jens Axboe --- include/linux/bvec.h | 7 +------ 1 file changed, 1 insertion(+), 6 deletions(-) (limited to 'include') diff --git a/include/linux/bvec.h b/include/linux/bvec.h index f41c7f0ef91e..ba8f52d48b94 100644 --- a/include/linux/bvec.h +++ b/include/linux/bvec.h @@ -286,12 +286,7 @@ static inline void *bvec_virt(struct bio_vec *bvec) */ static inline phys_addr_t bvec_phys(const struct bio_vec *bvec) { - /* - * Note this open codes page_to_phys because page_to_phys is defined in - * , which we don't want to pull in here. If it ever moves to - * a sensible place we should start using it. - */ - return PFN_PHYS(page_to_pfn(bvec->bv_page)) + bvec->bv_offset; + return page_to_phys(bvec->bv_page) + bvec->bv_offset; } #endif /* __LINUX_BVEC_H */ -- cgit v1.2.3 From e7602bb4f3a1234df8b75728ac3260bcb8242612 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 6 Jan 2025 09:35:10 +0100 Subject: block: remove BLK_MQ_F_NO_SCHED The only queues that really can't support a scheduler are those that do not have a gendisk associated with them, and thus can't be used for non-passthrough commands. In addition to those null_blk can optionally set the flag, which is a bad odd. Replace the null_blk usage with BLK_MQ_F_NO_SCHED_BY_DEFAULT to keep the expected semantics and then remove BLK_MQ_F_NO_SCHED as the non-disk queues never call into elevator_init_mq or blk_register_queue which adds the sysfs attributes. Signed-off-by: Christoph Hellwig Link: https://lore.kernel.org/r/20250106083531.799976-4-hch@lst.de Signed-off-by: Jens Axboe --- include/linux/blk-mq.h | 2 -- 1 file changed, 2 deletions(-) (limited to 'include') diff --git a/include/linux/blk-mq.h b/include/linux/blk-mq.h index 6340293511c9..f2ff0ffa0535 100644 --- a/include/linux/blk-mq.h +++ b/include/linux/blk-mq.h @@ -676,8 +676,6 @@ enum { BLK_MQ_F_STACKING = 1 << 2, BLK_MQ_F_TAG_HCTX_SHARED = 1 << 3, BLK_MQ_F_BLOCKING = 1 << 4, - /* Do not allow an I/O scheduler to be configured. */ - BLK_MQ_F_NO_SCHED = 1 << 5, /* * Select 'none' during queue registration in case of a single hwq -- cgit v1.2.3 From ce32496ec1abe866225f2e2005ceda68cf4c7bf4 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 6 Jan 2025 09:35:11 +0100 Subject: block: simplify tag allocation policy selection Use a plain BLK_MQ_F_* flag to select the round robin tag selection instead of overlaying an enum with just two possible values into the flags space. Doing so allows adding a BLK_MQ_F_MAX sentinel for simplified overflow checking in the messy debugfs helpers. Signed-off-by: Christoph Hellwig Reviewed-by: John Garry Link: https://lore.kernel.org/r/20250106083531.799976-5-hch@lst.de Signed-off-by: Jens Axboe --- include/linux/blk-mq.h | 22 +++++++--------------- include/linux/libata.h | 4 ++-- include/scsi/scsi_host.h | 6 ++++-- 3 files changed, 13 insertions(+), 19 deletions(-) (limited to 'include') diff --git a/include/linux/blk-mq.h b/include/linux/blk-mq.h index f2ff0ffa0535..a0a9007cc1e3 100644 --- a/include/linux/blk-mq.h +++ b/include/linux/blk-mq.h @@ -296,13 +296,6 @@ enum blk_eh_timer_return { BLK_EH_RESET_TIMER, }; -/* Keep alloc_policy_name[] in sync with the definitions below */ -enum { - BLK_TAG_ALLOC_FIFO, /* allocate starting from 0 */ - BLK_TAG_ALLOC_RR, /* allocate starting from last allocated tag */ - BLK_TAG_ALLOC_MAX -}; - /** * struct blk_mq_hw_ctx - State for a hardware queue facing the hardware * block device @@ -677,20 +670,19 @@ enum { BLK_MQ_F_TAG_HCTX_SHARED = 1 << 3, BLK_MQ_F_BLOCKING = 1 << 4, + /* + * Alloc tags on a round-robin base instead of the first available one. + */ + BLK_MQ_F_TAG_RR = 1 << 5, + /* * Select 'none' during queue registration in case of a single hwq * or shared hwqs instead of 'mq-deadline'. */ BLK_MQ_F_NO_SCHED_BY_DEFAULT = 1 << 6, - BLK_MQ_F_ALLOC_POLICY_START_BIT = 7, - BLK_MQ_F_ALLOC_POLICY_BITS = 1, + + BLK_MQ_F_MAX = 1 << 7, }; -#define BLK_MQ_FLAG_TO_ALLOC_POLICY(flags) \ - ((flags >> BLK_MQ_F_ALLOC_POLICY_START_BIT) & \ - ((1 << BLK_MQ_F_ALLOC_POLICY_BITS) - 1)) -#define BLK_ALLOC_POLICY_TO_MQ_FLAG(policy) \ - ((policy & ((1 << BLK_MQ_F_ALLOC_POLICY_BITS) - 1)) \ - << BLK_MQ_F_ALLOC_POLICY_START_BIT) #define BLK_MQ_MAX_DEPTH (10240) #define BLK_MQ_NO_HCTX_IDX (-1U) diff --git a/include/linux/libata.h b/include/linux/libata.h index c1a85d46eba6..be5183d75736 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -1467,13 +1467,13 @@ extern const struct attribute_group *ata_common_sdev_groups[]; #define ATA_SUBBASE_SHT(drv_name) \ __ATA_BASE_SHT(drv_name), \ .can_queue = ATA_DEF_QUEUE, \ - .tag_alloc_policy = BLK_TAG_ALLOC_RR, \ + .tag_alloc_policy_rr = true, \ .device_configure = ata_scsi_device_configure #define ATA_SUBBASE_SHT_QD(drv_name, drv_qd) \ __ATA_BASE_SHT(drv_name), \ .can_queue = drv_qd, \ - .tag_alloc_policy = BLK_TAG_ALLOC_RR, \ + .tag_alloc_policy_rr = true, \ .device_configure = ata_scsi_device_configure #define ATA_BASE_SHT(drv_name) \ diff --git a/include/scsi/scsi_host.h b/include/scsi/scsi_host.h index 2b4ab0369ffb..02823d6af37d 100644 --- a/include/scsi/scsi_host.h +++ b/include/scsi/scsi_host.h @@ -438,8 +438,10 @@ struct scsi_host_template { */ short cmd_per_lun; - /* If use block layer to manage tags, this is tag allocation policy */ - int tag_alloc_policy; + /* + * Allocate tags starting from last allocated tag. + */ + bool tag_alloc_policy_rr : 1; /* * Track QUEUE_FULL events and reduce queue depth on demand. -- cgit v1.2.3 From 9c96821b44f893fb63f021a28625d3b32c68e8b3 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Fri, 10 Jan 2025 06:47:09 +0100 Subject: block: fix docs for freezing of queue limits updates queue_limits_commit_update is the function that needs to operate on a frozen queue, not queue_limits_start_update. Update the kerneldoc comments to reflect that. Signed-off-by: Christoph Hellwig Reviewed-by: Ming Lei Reviewed-by: Damien Le Moal Reviewed-by: Martin K. Petersen Reviewed-by: Nilay Shroff Reviewed-by: Johannes Thumshirn Reviewed-by: John Garry Link: https://lore.kernel.org/r/20250110054726.1499538-2-hch@lst.de Signed-off-by: Jens Axboe --- include/linux/blkdev.h | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'include') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 5d40af2ef971..e781d4e6f92d 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -944,8 +944,7 @@ static inline unsigned int blk_boundary_sectors_left(sector_t offset, * the caller can modify. The caller must call queue_limits_commit_update() * to finish the update. * - * Context: process context. The caller must have frozen the queue or ensured - * that there is outstanding I/O by other means. + * Context: process context. */ static inline struct queue_limits queue_limits_start_update(struct request_queue *q) -- cgit v1.2.3 From aa427d7b73b196f657d6d2cf0e94eff6b883fdef Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Fri, 10 Jan 2025 06:47:10 +0100 Subject: block: add a queue_limits_commit_update_frozen helper Add a helper that freezes the queue, updates the queue limits and unfreezes the queue and convert all open coded versions of that to the new helper. Signed-off-by: Christoph Hellwig Reviewed-by: John Garry Reviewed-by: Ming Lei Reviewed-by: Damien Le Moal Reviewed-by: Martin K. Petersen Reviewed-by: Nilay Shroff Reviewed-by: Johannes Thumshirn Link: https://lore.kernel.org/r/20250110054726.1499538-3-hch@lst.de Signed-off-by: Jens Axboe --- include/linux/blkdev.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index e781d4e6f92d..13d353351c37 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -952,6 +952,8 @@ queue_limits_start_update(struct request_queue *q) mutex_lock(&q->limits_lock); return q->limits; } +int queue_limits_commit_update_frozen(struct request_queue *q, + struct queue_limits *lim); int queue_limits_commit_update(struct request_queue *q, struct queue_limits *lim); int queue_limits_set(struct request_queue *q, struct queue_limits *lim); -- cgit v1.2.3 From 30e77e0fbec6940ecc5c79ffe0f076c54cf5a8d9 Mon Sep 17 00:00:00 2001 From: Damien Le Moal Date: Sat, 4 Jan 2025 13:59:34 +0900 Subject: nvme: Move opcode string helper functions declarations Move the declaration of all helper functions converting NVMe command opcodes and status codes into strings from drivers/nvme/host/nvme.h into include/linux/nvme.h, together with the commands definitions. This allows NVMe target drivers to call these functions without having to include a host header file. Signed-off-by: Damien Le Moal Reviewed-by: Christoph Hellwig Tested-by: Rick Wertenbroek Tested-by: Manivannan Sadhasivam Signed-off-by: Keith Busch --- include/linux/nvme.h | 40 ++++++++++++++++++++++++++++++++++++++++ 1 file changed, 40 insertions(+) (limited to 'include') diff --git a/include/linux/nvme.h b/include/linux/nvme.h index 13377dde4527..a5a4ee56efcf 100644 --- a/include/linux/nvme.h +++ b/include/linux/nvme.h @@ -1896,6 +1896,46 @@ static inline bool nvme_is_fabrics(const struct nvme_command *cmd) return cmd->common.opcode == nvme_fabrics_command; } +#ifdef CONFIG_NVME_VERBOSE_ERRORS +const char *nvme_get_error_status_str(u16 status); +const char *nvme_get_opcode_str(u8 opcode); +const char *nvme_get_admin_opcode_str(u8 opcode); +const char *nvme_get_fabrics_opcode_str(u8 opcode); +#else /* CONFIG_NVME_VERBOSE_ERRORS */ +static inline const char *nvme_get_error_status_str(u16 status) +{ + return "I/O Error"; +} +static inline const char *nvme_get_opcode_str(u8 opcode) +{ + return "I/O Cmd"; +} +static inline const char *nvme_get_admin_opcode_str(u8 opcode) +{ + return "Admin Cmd"; +} + +static inline const char *nvme_get_fabrics_opcode_str(u8 opcode) +{ + return "Fabrics Cmd"; +} +#endif /* CONFIG_NVME_VERBOSE_ERRORS */ + +static inline const char *nvme_opcode_str(int qid, u8 opcode) +{ + return qid ? nvme_get_opcode_str(opcode) : + nvme_get_admin_opcode_str(opcode); +} + +static inline const char *nvme_fabrics_opcode_str( + int qid, const struct nvme_command *cmd) +{ + if (nvme_is_fabrics(cmd)) + return nvme_get_fabrics_opcode_str(cmd->fabrics.fctype); + + return nvme_opcode_str(qid, cmd->common.opcode); +} + struct nvme_error_slot { __le64 error_count; __le16 sqid; -- cgit v1.2.3 From 200adac75888182c09027e9b7852507dabd87034 Mon Sep 17 00:00:00 2001 From: Damien Le Moal Date: Sat, 4 Jan 2025 13:59:39 +0900 Subject: nvme: Add PCI transport type Define the transport type NVMF_TRTYPE_PCI for PCI endpoint targets. This transport type is defined using the value 0 which is reserved in the NVMe base specifications v2.1 (Figure 294). Given that struct nvmet_port are zeroed out on creation, to avoid having this transsport type becoming the new default, nvmet_referral_make() and nvmet_ports_make() are modified to initialize a port discovery address transport type field (disc_addr.trtype) to NVMF_TRTYPE_MAX. Any port using this transport type is also skipped and not reported in the discovery log page (nvmet_execute_disc_get_log_page()). The helper function nvmet_is_pci_ctrl() is also introduced to check if a target controller uses the PCI transport. Signed-off-by: Damien Le Moal Reviewed-by: Christoph Hellwig Tested-by: Rick Wertenbroek Tested-by: Manivannan Sadhasivam Signed-off-by: Keith Busch --- include/linux/nvme.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include') diff --git a/include/linux/nvme.h b/include/linux/nvme.h index a5a4ee56efcf..42fc00dc494e 100644 --- a/include/linux/nvme.h +++ b/include/linux/nvme.h @@ -64,6 +64,7 @@ enum { /* Transport Type codes for Discovery Log Page entry TRTYPE field */ enum { + NVMF_TRTYPE_PCI = 0, /* PCI */ NVMF_TRTYPE_RDMA = 1, /* RDMA */ NVMF_TRTYPE_FC = 2, /* Fibre Channel */ NVMF_TRTYPE_TCP = 3, /* TCP/IP */ -- cgit v1.2.3 From 2f2b20fad973d00169d24f5338eb1bf0a42e8218 Mon Sep 17 00:00:00 2001 From: Damien Le Moal Date: Sat, 4 Jan 2025 13:59:46 +0900 Subject: nvmet: Implement host identifier set feature support The NVMe specifications mandate support for the host identifier set_features for controllers that also supports reservations. Satisfy this requirement by implementing handling of the NVME_FEAT_HOST_ID feature for the nvme_set_features command. This implementation is for now effective only for PCI target controllers. For other controller types, the set features command is failed with a NVME_SC_CMD_SEQ_ERROR status as before. As noted in the code, 128 bits host identifiers are supported since the NVMe base specifications version 2.1 indicate in section 5.1.25.1.28.1 that "The controller may support a 64-bit Host Identifier...". The RHII (Reservations and Host Identifier Interaction) bit of the controller attribute (ctratt) field of the identify controller data is also set to indicate that a host ID of "0" is supported but that the host ID must be a non-zero value to use reservations. Signed-off-by: Damien Le Moal Reviewed-by: Christoph Hellwig Tested-by: Rick Wertenbroek Tested-by: Manivannan Sadhasivam Signed-off-by: Keith Busch --- include/linux/nvme.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include') diff --git a/include/linux/nvme.h b/include/linux/nvme.h index 42fc00dc494e..fe3b60818fdc 100644 --- a/include/linux/nvme.h +++ b/include/linux/nvme.h @@ -276,6 +276,7 @@ enum nvme_ctrl_attr { NVME_CTRL_ATTR_HID_128_BIT = (1 << 0), NVME_CTRL_ATTR_TBKAS = (1 << 6), NVME_CTRL_ATTR_ELBAS = (1 << 15), + NVME_CTRL_ATTR_RHII = (1 << 18), }; struct nvme_id_ctrl { -- cgit v1.2.3 From 127186cfb184eaccdfe948e6da66940cfa03efc5 Mon Sep 17 00:00:00 2001 From: Yu Kuai Date: Thu, 2 Jan 2025 19:28:41 +0800 Subject: md: reintroduce md-linear THe md-linear is removed by commit 849d18e27be9 ("md: Remove deprecated CONFIG_MD_LINEAR") because it has been marked as deprecated for a long time. However, md-linear is used widely for underlying disks with different size, sadly we didn't know this until now, and it's true useful to create partitions and assemble multiple raid and then append one to the other. People have to use dm-linear in this case now, however, they will prefer to minimize the number of involved modules. Fixes: 849d18e27be9 ("md: Remove deprecated CONFIG_MD_LINEAR") Cc: stable@vger.kernel.org Signed-off-by: Yu Kuai Acked-by: Coly Li Acked-by: Mike Snitzer Link: https://lore.kernel.org/r/20250102112841.1227111-1-yukuai1@huaweicloud.com Signed-off-by: Song Liu --- include/uapi/linux/raid/md_p.h | 2 +- include/uapi/linux/raid/md_u.h | 2 ++ 2 files changed, 3 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/uapi/linux/raid/md_p.h b/include/uapi/linux/raid/md_p.h index 5a43c23f53bf..ff47b6f0ba0f 100644 --- a/include/uapi/linux/raid/md_p.h +++ b/include/uapi/linux/raid/md_p.h @@ -233,7 +233,7 @@ struct mdp_superblock_1 { char set_name[32]; /* set and interpreted by user-space */ __le64 ctime; /* lo 40 bits are seconds, top 24 are microseconds or 0*/ - __le32 level; /* 0,1,4,5 */ + __le32 level; /* 0,1,4,5, -1 (linear) */ __le32 layout; /* only for raid5 and raid10 currently */ __le64 size; /* used size of component devices, in 512byte sectors */ diff --git a/include/uapi/linux/raid/md_u.h b/include/uapi/linux/raid/md_u.h index 7be89a4906e7..a893010735fb 100644 --- a/include/uapi/linux/raid/md_u.h +++ b/include/uapi/linux/raid/md_u.h @@ -103,6 +103,8 @@ typedef struct mdu_array_info_s { } mdu_array_info_t; +#define LEVEL_LINEAR (-1) + /* we need a value for 'no level specified' and 0 * means 'raid0', so we need something else. This is * for internal use only -- cgit v1.2.3 From 6564862d646e7d630929ba1ff330740bb215bdac Mon Sep 17 00:00:00 2001 From: John Garry Date: Thu, 9 Jan 2025 11:39:59 +0000 Subject: block: Ensure start sector is aligned for stacking atomic writes For stacking atomic writes, ensure that the start sector is aligned with the device atomic write unit min and any boundary. Otherwise, we may permit misaligned atomic writes. Rework bdev_can_atomic_write() into a common helper to resuse the alignment check. There also use atomic_write_hw_unit_min, which is more proper (than atomic_write_unit_min). Fixes: d7f36dc446e89 ("block: Support atomic writes limits for stacked devices") Reviewed-by: Christoph Hellwig Signed-off-by: John Garry Reviewed-by: Martin K. Petersen Link: https://lore.kernel.org/r/20250109114000.2299896-2-john.g.garry@oracle.com Signed-off-by: Jens Axboe --- include/linux/blkdev.h | 21 ++++++++++++--------- 1 file changed, 12 insertions(+), 9 deletions(-) (limited to 'include') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 13d353351c37..7ac153e4423a 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -1706,6 +1706,15 @@ struct io_comp_batch { void (*complete)(struct io_comp_batch *); }; +static inline bool blk_atomic_write_start_sect_aligned(sector_t sector, + struct queue_limits *limits) +{ + unsigned int alignment = max(limits->atomic_write_hw_unit_min, + limits->atomic_write_hw_boundary); + + return IS_ALIGNED(sector, alignment >> SECTOR_SHIFT); +} + static inline bool bdev_can_atomic_write(struct block_device *bdev) { struct request_queue *bd_queue = bdev->bd_queue; @@ -1714,15 +1723,9 @@ static inline bool bdev_can_atomic_write(struct block_device *bdev) if (!limits->atomic_write_unit_min) return false; - if (bdev_is_partition(bdev)) { - sector_t bd_start_sect = bdev->bd_start_sect; - unsigned int alignment = - max(limits->atomic_write_unit_min, - limits->atomic_write_hw_boundary); - - if (!IS_ALIGNED(bd_start_sect, alignment >> SECTOR_SHIFT)) - return false; - } + if (bdev_is_partition(bdev)) + return blk_atomic_write_start_sect_aligned(bdev->bd_start_sect, + limits); return true; } -- cgit v1.2.3 From 6a7e17b22062c84a111d7073c67cc677c4190f32 Mon Sep 17 00:00:00 2001 From: John Garry Date: Thu, 16 Jan 2025 17:02:54 +0000 Subject: block: Add common atomic writes enable flag Currently only stacked devices need to explicitly enable atomic writes by setting BLK_FEAT_ATOMIC_WRITES_STACKED flag. This does not work well for device mapper stacking devices, as there many sets of limits are stacked and what is the 'bottom' and 'top' device can swapped. This means that BLK_FEAT_ATOMIC_WRITES_STACKED needs to be set for many queue limits, which is messy. Generalize enabling atomic writes enabling by ensuring that all devices must explicitly set a flag - that includes NVMe, SCSI sd, and md raid. Signed-off-by: John Garry Reviewed-by: Mike Snitzer Link: https://lore.kernel.org/r/20250116170301.474130-2-john.g.garry@oracle.com Signed-off-by: Jens Axboe --- include/linux/blkdev.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 7ac153e4423a..76f0a4e7c2e5 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -331,8 +331,8 @@ typedef unsigned int __bitwise blk_features_t; #define BLK_FEAT_RAID_PARTIAL_STRIPES_EXPENSIVE \ ((__force blk_features_t)(1u << 15)) -/* stacked device can/does support atomic writes */ -#define BLK_FEAT_ATOMIC_WRITES_STACKED \ +/* atomic writes enabled */ +#define BLK_FEAT_ATOMIC_WRITES \ ((__force blk_features_t)(1u << 16)) /* -- cgit v1.2.3