Skip to contentSkip to conversation

block, dm: don't copy bios for request clones

PATCH New Christoph Hellwig 1 participant 1 message Apr 24, 2015

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 block, dm: don't copy bios for request clones

[recent, the dm-devel list server doesn't seem to like mail sent using git-send-email..]

Currently dm-multipath has to clone the bios for every request sent to the lower devices, which wastes cpu cycles and ties down memory.

This patch instead adds a new REQ_CLONE flag that instructs req_bio_endio to not complete bios attached to a request, which we set on clone requests similar to bios in a flush sequence. With this change I/O errors on a path failure only get propagated to dm-multipath, which can then either resubmit the I/O or complete the bios on the original request.

I've done some basic testing of this on a Linux target with ALUA support, and it survives path failures during I/O nicely.

---
 block/blk-core.c          |  92 +++-----------------------
 drivers/md/dm-table.c     |  27 +++++---
 drivers/md/dm.c           | 162 ++++++++++++----------------------------------
 drivers/md/dm.h           |   4 +-
 include/linux/blk_types.h |   2 +
 include/linux/blkdev.h    |   6 +-
 6 files changed, 73 insertions(+), 220 deletions(-)
block/blk-core.c +10 -82
index fd154b9..57da507 100644
@@ -128,7 +128,8 @@ static void req_bio_endio(struct request *rq, struct bio *bio,
 	bio_advance(bio, nbytes);
 
 	/* don't actually finish bio if it's part of flush sequence */
-	if (bio->bi_iter.bi_size == 0 && !(rq->cmd_flags & REQ_FLUSH_SEQ))
+	if (bio->bi_iter.bi_size == 0 &&
+	    !(rq->cmd_flags & (REQ_FLUSH_SEQ|REQ_CLONE)))
 		bio_endio(bio, error);
 }
 
@@ -2901,95 +2902,22 @@ int blk_lld_busy(struct request_queue *q)
 }
 EXPORT_SYMBOL_GPL(blk_lld_busy);
 
-/**
- * blk_rq_unprep_clone - Helper function to free all bios in a cloned request
- * @rq: the clone request to be cleaned up
- *
- * Description:
- *     Free all bios in @rq for a cloned request.
- */
-void blk_rq_unprep_clone(struct request *rq)
-{
-	struct bio *bio;
-
-	while ((bio = rq->bio) != NULL) {
-		rq->bio = bio->bi_next;
-
-		bio_put(bio);
-	}
-}
-EXPORT_SYMBOL_GPL(blk_rq_unprep_clone);
-
-/*
- * Copy attributes of the original request to the clone request.
- * The actual data parts (e.g. ->cmd, ->sense) are not copied.
- */
-static void __blk_rq_prep_clone(struct request *dst, struct request *src)
+void blk_rq_prep_clone(struct request *dst, struct request *src)
 {
 	dst->cpu = src->cpu;
-	dst->cmd_flags |= (src->cmd_flags & REQ_CLONE_MASK) | REQ_NOMERGE;
+	dst->cmd_flags |= (src->cmd_flags & REQ_CLONE_MASK);
+	dst->cmd_flags |= REQ_NOMERGE | REQ_CLONE;
 	dst->cmd_type = src->cmd_type;
 	dst->__sector = blk_rq_pos(src);
 	dst->__data_len = blk_rq_bytes(src);
 	dst->nr_phys_segments = src->nr_phys_segments;
 	dst->ioprio = src->ioprio;
 	dst->extra_len = src->extra_len;
-}
-
-/**
- * blk_rq_prep_clone - Helper function to setup clone request
- * @rq: the request to be setup
- * @rq_src: original request to be cloned
- * @bs: bio_set that bios for clone are allocated from
- * @gfp_mask: memory allocation mask for bio
- * @bio_ctr: setup function to be called for each clone bio.
- *           Returns %0 for success, non %0 for failure.
- * @data: private data to be passed to @bio_ctr
- *
- * Description:
- *     Clones bios in @rq_src to @rq, and copies attributes of @rq_src to @rq.
- *     The actual data parts of @rq_src (e.g. ->cmd, ->sense)
- *     are not copied, and copying such parts is the caller's responsibility.
- *     Also, pages which the original bios are pointing to are not copied
- *     and the cloned bios just point same pages.
- *     So cloned bios must be completed before original bios, which means
- *     the caller must complete @rq before @rq_src.
- */
-int blk_rq_prep_clone(struct request *rq, struct request *rq_src,
-		      struct bio_set *bs, gfp_t gfp_mask,
-		      int (*bio_ctr)(struct bio *, struct bio *, void *),
-		      void *data)
-{
-	struct bio *bio, *bio_src;
-
-	if (!bs)
-		bs = fs_bio_set;
-
-	__rq_for_each_bio(bio_src, rq_src) {
-		bio = bio_clone_fast(bio_src, gfp_mask, bs);
-		if (!bio)
-			goto free_and_out;
-
-		if (bio_ctr && bio_ctr(bio, bio_src, data))
-			goto free_and_out;
-
-		if (rq->bio) {
-			rq->biotail->bi_next = bio;
-			rq->biotail = bio;
-		} else
-			rq->bio = rq->biotail = bio;
-	}
-
-	__blk_rq_prep_clone(rq, rq_src);
-
-	return 0;
-
-free_and_out:
-	if (bio)
-		bio_put(bio);
-	blk_rq_unprep_clone(rq);
-
-	return -ENOMEM;
+	dst->bio = src->bio;
+	dst->biotail = src->biotail;
+	dst->cmd = src->cmd;
+	dst->cmd_len = src->cmd_len;
+	dst->sense = src->sense;
 }
 EXPORT_SYMBOL_GPL(blk_rq_prep_clone);
 
drivers/md/dm-table.c +17 -10
index 6554d91..4ac0a47 100644
@@ -941,23 +941,30 @@ bool dm_table_mq_request_based(struct dm_table *t)
 
 static int dm_table_alloc_md_mempools(struct dm_table *t)
 {
-	unsigned type = dm_table_get_type(t);
+	int type = dm_table_get_type(t);
 	unsigned per_bio_data_size = 0;
-	struct dm_target *tgt;
 	unsigned i;
 
-	if (unlikely(type == DM_TYPE_NONE)) {
+	switch (type) {
+		for (i = 0; i < t->num_targets; i++) {
+			struct dm_target *tgt = t->targets + i;
+
+			per_bio_data_size = max(per_bio_data_size,
+						tgt->per_bio_data_size);
+		}
+
+		t->mempools = dm_alloc_bio_mempools(t->integrity_supported,
+				per_bio_data_size);
+		break;
+	case DM_TYPE_REQUEST_BASED:
+	case DM_TYPE_MQ_REQUEST_BASED:
+		t->mempools = dm_alloc_rq_mempools(type);
+		break;
+	default:
 		DMWARN("no table type is set, can't allocate mempools");
 		return -EINVAL;
 	}
 
-	if (type == DM_TYPE_BIO_BASED)
-		for (i = 0; i < t->num_targets; i++) {
-			tgt = t->targets + i;
-			per_bio_data_size = max(per_bio_data_size, tgt->per_bio_data_size);
-		}
-
-	t->mempools = dm_alloc_md_mempools(type, t->integrity_supported, per_bio_data_size);
 	if (!t->mempools)
 		return -ENOMEM;
 
drivers/md/dm.c +40 -122

222 lines not on this page · raw message

drivers/md/dm.h +3 -1
index 59f53e7..35887b8 100644
@@ -221,7 +221,9 @@ void dm_kcopyd_exit(void);
 /*
  * Mempool operations
  */
-struct dm_md_mempools *dm_alloc_md_mempools(unsigned type, unsigned integrity, unsigned per_bio_data_size);
+struct dm_md_mempools *dm_alloc_bio_mempools(unsigned integrity,
+		unsigned per_bio_data_size);
+struct dm_md_mempools *dm_alloc_rq_mempools(unsigned type);
 void dm_free_md_mempools(struct dm_md_mempools *pools);
 
 /*
include/linux/blk_types.h +2 -0
index 93d2e71..992ef58 100644
@@ -195,6 +195,7 @@ enum rq_flag_bits {
 	__REQ_HASHED,		/* on IO scheduler merge hash */
 	__REQ_MQ_INFLIGHT,	/* track inflight for MQ */
 	__REQ_NO_TIMEOUT,	/* requests may never expire */
+	__REQ_CLONE,		/* cloned bios */
 	__REQ_NR_BITS,		/* stops here */
 };
 
@@ -249,5 +250,6 @@ enum rq_flag_bits {
 #define REQ_HASHED		(1ULL << __REQ_HASHED)
 #define REQ_MQ_INFLIGHT		(1ULL << __REQ_MQ_INFLIGHT)
 #define REQ_NO_TIMEOUT		(1ULL << __REQ_NO_TIMEOUT)
+#define REQ_CLONE		(1ULL << __REQ_CLONE)
 
 #endif /* __LINUX_BLK_TYPES_H */
include/linux/blkdev.h +1 -5
index 7f9a516..9cb8174 100644
@@ -804,11 +804,7 @@ extern void blk_add_request_payload(struct request *rq, struct page *page,
 		unsigned int len);
 extern int blk_rq_check_limits(struct request_queue *q, struct request *rq);
 extern int blk_lld_busy(struct request_queue *q);
-extern int blk_rq_prep_clone(struct request *rq, struct request *rq_src,
-			     struct bio_set *bs, gfp_t gfp_mask,
-			     int (*bio_ctr)(struct bio *, struct bio *, void *),
-			     void *data);
-extern void blk_rq_unprep_clone(struct request *rq);
+extern void blk_rq_prep_clone(struct request *rq, struct request *rq_src);
 extern int blk_insert_cloned_request(struct request_queue *q,
 				     struct request *rq);
 extern void blk_delay_queue(struct request_queue *, unsigned long);
Signature · 1 line
1.9.1