[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <1385513128-5035-11-git-send-email-kmo@daterainc.com>
Date: Tue, 26 Nov 2013 16:45:14 -0800
From: Kent Overstreet <kmo@...erainc.com>
To: linux-kernel@...r.kernel.org, linux-fsdevel@...r.kernel.org,
hch@...radead.org
Cc: Kent Overstreet <kmo@...erainc.com>, Jens Axboe <axboe@...nel.dk>,
Neil Brown <neilb@...e.de>,
Nagalakshmi Nandigama <Nagalakshmi.Nandigama@....com>,
Sreekanth Reddy <Sreekanth.Reddy@....com>,
"James E.J. Bottomley" <JBottomley@...allels.com>
Subject: [PATCH 11/25] block: Kill bio_segments()/bi_vcnt usage
When we start sharing biovecs, keeping bi_vcnt accurate for splits is
going to be error prone - and unnecessary, if we refactor some code.
So bio_segments() has to go - but most of the existing users just needed
to know if the bio had multiple segments, which is easier - add a
bio_multiple_segments() for them.
(Two of the current uses of bio_segments() are going to go away in a
couple patches, but the current implementation of bio_segments() is
unsafe as soon as we start doing driver conversions for immutable
biovecs - so implement a dumb version for bisectability, it'll go away
in a couple patches)
Signed-off-by: Kent Overstreet <kmo@...erainc.com>
Cc: Jens Axboe <axboe@...nel.dk>
Cc: Neil Brown <neilb@...e.de>
Cc: Nagalakshmi Nandigama <Nagalakshmi.Nandigama@....com>
Cc: Sreekanth Reddy <Sreekanth.Reddy@....com>
Cc: "James E.J. Bottomley" <JBottomley@...allels.com>
---
drivers/block/ps3disk.c | 7 ++-
drivers/md/bcache/io.c | 53 +++++++++------------
drivers/md/raid0.c | 2 +-
drivers/md/raid10.c | 2 +-
drivers/message/fusion/mptsas.c | 8 ++--
drivers/scsi/libsas/sas_expander.c | 8 ++--
drivers/scsi/mpt2sas/mpt2sas_transport.c | 10 ++--
drivers/scsi/mpt3sas/mpt3sas_transport.c | 8 ++--
fs/bio.c | 2 +-
include/linux/bio.h | 81 +++++++++++++++++++-------------
10 files changed, 94 insertions(+), 87 deletions(-)
diff --git a/drivers/block/ps3disk.c b/drivers/block/ps3disk.c
index 1c6edb9..c120d70 100644
--- a/drivers/block/ps3disk.c
+++ b/drivers/block/ps3disk.c
@@ -101,10 +101,9 @@ static void ps3disk_scatter_gather(struct ps3_storage_device *dev,
rq_for_each_segment(bvec, req, iter) {
unsigned long flags;
- dev_dbg(&dev->sbd.core,
- "%s:%u: bio %u: %u segs %u sectors from %lu\n",
- __func__, __LINE__, i, bio_segments(iter.bio),
- bio_sectors(iter.bio), iter.bio->bi_iter.bi_sector);
+ dev_dbg(&dev->sbd.core, "%s:%u: bio %u: %u sectors from %lu\n",
+ __func__, __LINE__, i, bio_sectors(iter.bio),
+ iter.bio->bi_iter.bi_sector);
size = bvec.bv_len;
buf = bvec_kmap_irq(&bvec, &flags);
diff --git a/drivers/md/bcache/io.c b/drivers/md/bcache/io.c
index 9b5b6a4..6e04f3b 100644
--- a/drivers/md/bcache/io.c
+++ b/drivers/md/bcache/io.c
@@ -24,7 +24,8 @@ static void bch_generic_make_request_hack(struct bio *bio)
if (bio->bi_iter.bi_idx) {
struct bio_vec bv;
struct bvec_iter iter;
- struct bio *clone = bio_alloc(GFP_NOIO, bio_segments(bio));
+ unsigned segs = bio_segments(bio);
+ struct bio *clone = bio_alloc(GFP_NOIO, segs);
bio_for_each_segment(bv, bio, iter)
clone->bi_io_vec[clone->bi_vcnt++] = bv;
@@ -32,7 +33,7 @@ static void bch_generic_make_request_hack(struct bio *bio)
clone->bi_iter.bi_sector = bio->bi_iter.bi_sector;
clone->bi_bdev = bio->bi_bdev;
clone->bi_rw = bio->bi_rw;
- clone->bi_vcnt = bio_segments(bio);
+ clone->bi_vcnt = segs;
clone->bi_iter.bi_size = bio->bi_iter.bi_size;
clone->bi_private = bio;
@@ -133,40 +134,32 @@ out:
static unsigned bch_bio_max_sectors(struct bio *bio)
{
- unsigned ret = bio_sectors(bio);
struct request_queue *q = bdev_get_queue(bio->bi_bdev);
- unsigned max_segments = min_t(unsigned, BIO_MAX_PAGES,
- queue_max_segments(q));
+ struct bio_vec bv;
+ struct bvec_iter iter;
+ unsigned ret = 0, seg = 0;
if (bio->bi_rw & REQ_DISCARD)
- return min(ret, q->limits.max_discard_sectors);
-
- if (bio_segments(bio) > max_segments ||
- q->merge_bvec_fn) {
- struct bio_vec bv;
- struct bvec_iter iter;
- unsigned seg = 0;
-
- ret = 0;
+ return min(bio_sectors(bio), q->limits.max_discard_sectors);
- bio_for_each_segment(bv, bio, iter) {
- struct bvec_merge_data bvm = {
- .bi_bdev = bio->bi_bdev,
- .bi_sector = bio->bi_iter.bi_sector,
- .bi_size = ret << 9,
- .bi_rw = bio->bi_rw,
- };
-
- if (seg == max_segments)
- break;
+ bio_for_each_segment(bv, bio, iter) {
+ struct bvec_merge_data bvm = {
+ .bi_bdev = bio->bi_bdev,
+ .bi_sector = bio->bi_iter.bi_sector,
+ .bi_size = ret << 9,
+ .bi_rw = bio->bi_rw,
+ };
+
+ if (seg == min_t(unsigned, BIO_MAX_PAGES,
+ queue_max_segments(q)))
+ break;
- if (q->merge_bvec_fn &&
- q->merge_bvec_fn(q, &bvm, &bv) < (int) bv.bv_len)
- break;
+ if (q->merge_bvec_fn &&
+ q->merge_bvec_fn(q, &bvm, &bv) < (int) bv.bv_len)
+ break;
- seg++;
- ret += bv.bv_len >> 9;
- }
+ seg++;
+ ret += bv.bv_len >> 9;
}
ret = min(ret, queue_max_sectors(q));
diff --git a/drivers/md/raid0.c b/drivers/md/raid0.c
index e38d1d3..8ee1a6c 100644
--- a/drivers/md/raid0.c
+++ b/drivers/md/raid0.c
@@ -528,7 +528,7 @@ static void raid0_make_request(struct mddev *mddev, struct bio *bio)
sector_t sector = bio->bi_iter.bi_sector;
struct bio_pair *bp;
/* Sanity check -- queue functions should prevent this happening */
- if (bio_segments(bio) > 1)
+ if (bio_multiple_segments(bio))
goto bad_map;
/* This is a one page bio that upper layers
* refuse to split for us, so we need to split it.
diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c
index dbf3b63..ac4bfa438c 100644
--- a/drivers/md/raid10.c
+++ b/drivers/md/raid10.c
@@ -1188,7 +1188,7 @@ static void make_request(struct mddev *mddev, struct bio * bio)
|| conf->prev.near_copies < conf->prev.raid_disks))) {
struct bio_pair *bp;
/* Sanity check -- queue functions should prevent this happening */
- if (bio_segments(bio) > 1)
+ if (bio_multiple_segments(bio))
goto bad_map;
/* This is a one page bio that upper layers
* refuse to split for us, so we need to split it.
diff --git a/drivers/message/fusion/mptsas.c b/drivers/message/fusion/mptsas.c
index dd239bd..00d339c 100644
--- a/drivers/message/fusion/mptsas.c
+++ b/drivers/message/fusion/mptsas.c
@@ -2235,10 +2235,10 @@ static int mptsas_smp_handler(struct Scsi_Host *shost, struct sas_rphy *rphy,
}
/* do we need to support multiple segments? */
- if (bio_segments(req->bio) > 1 || bio_segments(rsp->bio) > 1) {
- printk(MYIOC_s_ERR_FMT "%s: multiple segments req %u %u, rsp %u %u\n",
- ioc->name, __func__, bio_segments(req->bio), blk_rq_bytes(req),
- bio_segments(rsp->bio), blk_rq_bytes(rsp));
+ if (bio_multiple_segments(req->bio) ||
+ bio_multiple_segments(rsp->bio)) {
+ printk(MYIOC_s_ERR_FMT "%s: multiple segments req %u, rsp %u\n",
+ ioc->name, __func__, blk_rq_bytes(req), blk_rq_bytes(rsp));
return -EINVAL;
}
diff --git a/drivers/scsi/libsas/sas_expander.c b/drivers/scsi/libsas/sas_expander.c
index 446b851..0cac7d8 100644
--- a/drivers/scsi/libsas/sas_expander.c
+++ b/drivers/scsi/libsas/sas_expander.c
@@ -2163,10 +2163,10 @@ int sas_smp_handler(struct Scsi_Host *shost, struct sas_rphy *rphy,
}
/* do we need to support multiple segments? */
- if (bio_segments(req->bio) > 1 || bio_segments(rsp->bio) > 1) {
- printk("%s: multiple segments req %u %u, rsp %u %u\n",
- __func__, bio_segments(req->bio), blk_rq_bytes(req),
- bio_segments(rsp->bio), blk_rq_bytes(rsp));
+ if (bio_multiple_segments(req->bio) ||
+ bio_multiple_segments(rsp->bio)) {
+ printk("%s: multiple segments req %u, rsp %u\n",
+ __func__, blk_rq_bytes(req), blk_rq_bytes(rsp));
return -EINVAL;
}
diff --git a/drivers/scsi/mpt2sas/mpt2sas_transport.c b/drivers/scsi/mpt2sas/mpt2sas_transport.c
index 7143e86..410f4a3 100644
--- a/drivers/scsi/mpt2sas/mpt2sas_transport.c
+++ b/drivers/scsi/mpt2sas/mpt2sas_transport.c
@@ -1943,7 +1943,7 @@ _transport_smp_handler(struct Scsi_Host *shost, struct sas_rphy *rphy,
ioc->transport_cmds.status = MPT2_CMD_PENDING;
/* Check if the request is split across multiple segments */
- if (bio_segments(req->bio) > 1) {
+ if (bio_multiple_segments(req->bio)) {
u32 offset = 0;
/* Allocate memory and copy the request */
@@ -1975,7 +1975,7 @@ _transport_smp_handler(struct Scsi_Host *shost, struct sas_rphy *rphy,
/* Check if the response needs to be populated across
* multiple segments */
- if (bio_segments(rsp->bio) > 1) {
+ if (bio_multiple_segments(rsp->bio)) {
pci_addr_in = pci_alloc_consistent(ioc->pdev, blk_rq_bytes(rsp),
&pci_dma_in);
if (!pci_addr_in) {
@@ -2042,7 +2042,7 @@ _transport_smp_handler(struct Scsi_Host *shost, struct sas_rphy *rphy,
sgl_flags = (MPI2_SGE_FLAGS_SIMPLE_ELEMENT |
MPI2_SGE_FLAGS_END_OF_BUFFER | MPI2_SGE_FLAGS_HOST_TO_IOC);
sgl_flags = sgl_flags << MPI2_SGE_FLAGS_SHIFT;
- if (bio_segments(req->bio) > 1) {
+ if (bio_multiple_segments(req->bio)) {
ioc->base_add_sg_single(psge, sgl_flags |
(blk_rq_bytes(req) - 4), pci_dma_out);
} else {
@@ -2058,7 +2058,7 @@ _transport_smp_handler(struct Scsi_Host *shost, struct sas_rphy *rphy,
MPI2_SGE_FLAGS_LAST_ELEMENT | MPI2_SGE_FLAGS_END_OF_BUFFER |
MPI2_SGE_FLAGS_END_OF_LIST);
sgl_flags = sgl_flags << MPI2_SGE_FLAGS_SHIFT;
- if (bio_segments(rsp->bio) > 1) {
+ if (bio_multiple_segments(rsp->bio)) {
ioc->base_add_sg_single(psge, sgl_flags |
(blk_rq_bytes(rsp) + 4), pci_dma_in);
} else {
@@ -2103,7 +2103,7 @@ _transport_smp_handler(struct Scsi_Host *shost, struct sas_rphy *rphy,
le16_to_cpu(mpi_reply->ResponseDataLength);
/* check if the resp needs to be copied from the allocated
* pci mem */
- if (bio_segments(rsp->bio) > 1) {
+ if (bio_multiple_segments(rsp->bio)) {
u32 offset = 0;
u32 bytes_to_copy =
le16_to_cpu(mpi_reply->ResponseDataLength);
diff --git a/drivers/scsi/mpt3sas/mpt3sas_transport.c b/drivers/scsi/mpt3sas/mpt3sas_transport.c
index 196a67f..65170cb 100644
--- a/drivers/scsi/mpt3sas/mpt3sas_transport.c
+++ b/drivers/scsi/mpt3sas/mpt3sas_transport.c
@@ -1926,7 +1926,7 @@ _transport_smp_handler(struct Scsi_Host *shost, struct sas_rphy *rphy,
ioc->transport_cmds.status = MPT3_CMD_PENDING;
/* Check if the request is split across multiple segments */
- if (req->bio->bi_vcnt > 1) {
+ if (bio_multiple_segments(req->bio)) {
u32 offset = 0;
/* Allocate memory and copy the request */
@@ -1958,7 +1958,7 @@ _transport_smp_handler(struct Scsi_Host *shost, struct sas_rphy *rphy,
/* Check if the response needs to be populated across
* multiple segments */
- if (rsp->bio->bi_vcnt > 1) {
+ if (bio_multiple_segments(rsp->bio)) {
pci_addr_in = pci_alloc_consistent(ioc->pdev, blk_rq_bytes(rsp),
&pci_dma_in);
if (!pci_addr_in) {
@@ -2019,7 +2019,7 @@ _transport_smp_handler(struct Scsi_Host *shost, struct sas_rphy *rphy,
mpi_request->RequestDataLength = cpu_to_le16(blk_rq_bytes(req) - 4);
psge = &mpi_request->SGL;
- if (req->bio->bi_vcnt > 1)
+ if (bio_multiple_segments(req->bio))
ioc->build_sg(ioc, psge, pci_dma_out, (blk_rq_bytes(req) - 4),
pci_dma_in, (blk_rq_bytes(rsp) + 4));
else
@@ -2064,7 +2064,7 @@ _transport_smp_handler(struct Scsi_Host *shost, struct sas_rphy *rphy,
/* check if the resp needs to be copied from the allocated
* pci mem */
- if (rsp->bio->bi_vcnt > 1) {
+ if (bio_multiple_segments(rsp->bio)) {
u32 offset = 0;
u32 bytes_to_copy =
le16_to_cpu(mpi_reply->ResponseDataLength);
diff --git a/fs/bio.c b/fs/bio.c
index f61e59b..e32f2ff 100644
--- a/fs/bio.c
+++ b/fs/bio.c
@@ -1733,7 +1733,7 @@ struct bio_pair *bio_split(struct bio *bi, int first_sectors)
trace_block_split(bdev_get_queue(bi->bi_bdev), bi,
bi->bi_iter.bi_sector + first_sectors);
- BUG_ON(bio_segments(bi) > 1);
+ BUG_ON(bio_multiple_segments(bi));
atomic_set(&bp->cnt, 3);
bp->error = 0;
bp->bio1 = *bi;
diff --git a/include/linux/bio.h b/include/linux/bio.h
index 930cb73..aea9896 100644
--- a/include/linux/bio.h
+++ b/include/linux/bio.h
@@ -97,13 +97,46 @@
#define bio_offset(bio) bio_iter_offset((bio), (bio)->bi_iter)
#define bio_iovec(bio) bio_iter_iovec((bio), (bio)->bi_iter)
-#define bio_segments(bio) ((bio)->bi_vcnt - (bio)->bi_iter.bi_idx)
+#define bio_multiple_segments(bio) \
+ ((bio)->bi_iter.bi_size != bio_iovec(bio).bv_len)
#define bio_sectors(bio) ((bio)->bi_iter.bi_size >> 9)
#define bio_end_sector(bio) ((bio)->bi_iter.bi_sector + bio_sectors((bio)))
+/*
+ * Check whether this bio carries any data or not. A NULL bio is allowed.
+ */
+static inline bool bio_has_data(struct bio *bio)
+{
+ if (bio &&
+ bio->bi_iter.bi_size &&
+ !(bio->bi_rw & REQ_DISCARD))
+ return true;
+
+ return false;
+}
+
+static inline bool bio_is_rw(struct bio *bio)
+{
+ if (!bio_has_data(bio))
+ return false;
+
+ if (bio->bi_rw & BIO_NO_ADVANCE_ITER_MASK)
+ return false;
+
+ return true;
+}
+
+static inline bool bio_mergeable(struct bio *bio)
+{
+ if (bio->bi_rw & REQ_NOMERGE_FLAGS)
+ return false;
+
+ return true;
+}
+
static inline unsigned int bio_cur_bytes(struct bio *bio)
{
- if (bio->bi_vcnt)
+ if (bio_has_data(bio))
return bio_iovec(bio).bv_len;
else /* dataless requests such as discard */
return bio->bi_iter.bi_size;
@@ -111,7 +144,7 @@ static inline unsigned int bio_cur_bytes(struct bio *bio)
static inline void *bio_data(struct bio *bio)
{
- if (bio->bi_vcnt)
+ if (bio_has_data(bio))
return page_address(bio_page(bio)) + bio_offset(bio);
return NULL;
@@ -221,6 +254,18 @@ static inline void bio_advance_iter(struct bio *bio, struct bvec_iter *iter,
#define bio_iter_last(bvec, iter) ((iter).bi_size == (bvec).bv_len)
+static inline unsigned bio_segments(struct bio *bio)
+{
+ unsigned segs = 0;
+ struct bio_vec bv;
+ struct bvec_iter iter;
+
+ bio_for_each_segment(bv, bio, iter)
+ segs++;
+
+ return segs;
+}
+
/*
* get a reference to a bio, so it won't disappear. the intended use is
* something like:
@@ -435,36 +480,6 @@ static inline char *__bio_kmap_irq(struct bio *bio, unsigned short idx,
#define bio_kunmap_irq(buf,flags) __bio_kunmap_irq(buf, flags)
/*
- * Check whether this bio carries any data or not. A NULL bio is allowed.
- */
-static inline bool bio_has_data(struct bio *bio)
-{
- if (bio && bio->bi_vcnt)
- return true;
-
- return false;
-}
-
-static inline bool bio_is_rw(struct bio *bio)
-{
- if (!bio_has_data(bio))
- return false;
-
- if (bio->bi_rw & REQ_WRITE_SAME)
- return false;
-
- return true;
-}
-
-static inline bool bio_mergeable(struct bio *bio)
-{
- if (bio->bi_rw & REQ_NOMERGE_FLAGS)
- return false;
-
- return true;
-}
-
-/*
* BIO list management for use by remapping drivers (e.g. DM or MD) and loop.
*
* A bio_list anchors a singly-linked list of bios chained through the bi_next
--
1.8.4.4
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists