linux-fsdevel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH V2 0/2] block: add bio_iov_iter_nvecs for figuring out nr_vecs
@ 2020-12-03  2:29 Ming Lei
  2020-12-03  2:29 ` [PATCH V2 1/2] " Ming Lei
  2020-12-03  2:29 ` [PATCH V2 2/2] block: rename the local variable for holding return value of bio_iov_iter_nvecs Ming Lei
  0 siblings, 2 replies; 7+ messages in thread
From: Ming Lei @ 2020-12-03  2:29 UTC (permalink / raw)
  To: Jens Axboe
  Cc: linux-block, Ming Lei, Pavel Begunkov, Christoph Hellwig,
	Matthew Wilcox, linux-fsdevel

Hi,

Add add bio_iov_iter_nvecs for figuring out nr_vecs, so that we can
avoid iov_iter_npages() for bvec iter.

V2:
	- split out renaming part into one patch

Ming Lei (2):
  block: add bio_iov_iter_nvecs for figuring out nr_vecs
  block: rename the local variable for holding return value of
    bio_iov_iter_nvecs

 fs/block_dev.c       | 30 +++++++++++++++---------------
 fs/iomap/direct-io.c | 14 +++++++-------
 include/linux/bio.h  | 10 ++++++++++
 3 files changed, 32 insertions(+), 22 deletions(-)

Cc: Pavel Begunkov <asml.silence@gmail.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: linux-fsdevel@vger.kernel.org


-- 
2.28.0


^ permalink raw reply	[flat|nested] 7+ messages in thread

* [PATCH V2 1/2] block: add bio_iov_iter_nvecs for figuring out nr_vecs
  2020-12-03  2:29 [PATCH V2 0/2] block: add bio_iov_iter_nvecs for figuring out nr_vecs Ming Lei
@ 2020-12-03  2:29 ` Ming Lei
  2020-12-03  6:46   ` Hannes Reinecke
  2020-12-09  8:31   ` Christoph Hellwig
  2020-12-03  2:29 ` [PATCH V2 2/2] block: rename the local variable for holding return value of bio_iov_iter_nvecs Ming Lei
  1 sibling, 2 replies; 7+ messages in thread
From: Ming Lei @ 2020-12-03  2:29 UTC (permalink / raw)
  To: Jens Axboe
  Cc: linux-block, Ming Lei, Pavel Begunkov, Christoph Hellwig,
	Matthew Wilcox, linux-fsdevel

Pavel reported that iov_iter_npages is a bit heavy in case of bvec
iter.

Turns out it isn't necessary to iterate every page in the bvec iter,
and we call iov_iter_npages() just for figuring out how many bio
vecs need to be allocated. And we can simply map each vector in bvec iter
to bio's vec, so just return iter->nr_segs from bio_iov_iter_nvecs() for
bvec iter.

This patch is based on Mathew's post:

https://lore.kernel.org/linux-block/20201120123931.GN29991@casper.infradead.org/

Reviewed-by: Pavel Begunkov <asml.silence@gmail.com>
Reviewed-by: Christoph Hellwig <hch@infradead.org>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: linux-fsdevel@vger.kernel.org
Signed-off-by: Ming Lei <ming.lei@redhat.com>
---
 fs/block_dev.c       |  4 ++--
 fs/iomap/direct-io.c |  4 ++--
 include/linux/bio.h  | 10 ++++++++++
 3 files changed, 14 insertions(+), 4 deletions(-)

diff --git a/fs/block_dev.c b/fs/block_dev.c
index 9e56ee1f2652..fbcc900229af 100644
--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -417,7 +417,7 @@ __blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter, int nr_pages)
 		dio->size += bio->bi_iter.bi_size;
 		pos += bio->bi_iter.bi_size;
 
-		nr_pages = iov_iter_npages(iter, BIO_MAX_PAGES);
+		nr_pages = bio_iov_iter_nvecs(iter, BIO_MAX_PAGES);
 		if (!nr_pages) {
 			bool polled = false;
 
@@ -482,7 +482,7 @@ blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter)
 {
 	int nr_pages;
 
-	nr_pages = iov_iter_npages(iter, BIO_MAX_PAGES + 1);
+	nr_pages = bio_iov_iter_nvecs(iter, BIO_MAX_PAGES + 1);
 	if (!nr_pages)
 		return 0;
 	if (is_sync_kiocb(iocb) && nr_pages <= BIO_MAX_PAGES)
diff --git a/fs/iomap/direct-io.c b/fs/iomap/direct-io.c
index 933f234d5bec..0635469e811a 100644
--- a/fs/iomap/direct-io.c
+++ b/fs/iomap/direct-io.c
@@ -250,7 +250,7 @@ iomap_dio_bio_actor(struct inode *inode, loff_t pos, loff_t length,
 	orig_count = iov_iter_count(dio->submit.iter);
 	iov_iter_truncate(dio->submit.iter, length);
 
-	nr_pages = iov_iter_npages(dio->submit.iter, BIO_MAX_PAGES);
+	nr_pages = bio_iov_iter_nvecs(dio->submit.iter, BIO_MAX_PAGES);
 	if (nr_pages <= 0) {
 		ret = nr_pages;
 		goto out;
@@ -308,7 +308,7 @@ iomap_dio_bio_actor(struct inode *inode, loff_t pos, loff_t length,
 		dio->size += n;
 		copied += n;
 
-		nr_pages = iov_iter_npages(dio->submit.iter, BIO_MAX_PAGES);
+		nr_pages = bio_iov_iter_nvecs(dio->submit.iter, BIO_MAX_PAGES);
 		iomap_dio_submit_bio(dio, iomap, bio, pos);
 		pos += n;
 	} while (nr_pages);
diff --git a/include/linux/bio.h b/include/linux/bio.h
index 1edda614f7ce..8a7f44006f65 100644
--- a/include/linux/bio.h
+++ b/include/linux/bio.h
@@ -10,6 +10,7 @@
 #include <linux/ioprio.h>
 /* struct bio, bio_vec and BIO_* flags are defined in blk_types.h */
 #include <linux/blk_types.h>
+#include <linux/uio.h>
 
 #define BIO_DEBUG
 
@@ -820,4 +821,13 @@ static inline void bio_set_polled(struct bio *bio, struct kiocb *kiocb)
 		bio->bi_opf |= REQ_NOWAIT;
 }
 
+static inline int bio_iov_iter_nvecs(const struct iov_iter *i, int maxvecs)
+{
+	if (!iov_iter_count(i))
+		return 0;
+	if (iov_iter_is_bvec(i))
+		return min_t(int, maxvecs, i->nr_segs);
+	return iov_iter_npages(i, maxvecs);
+}
+
 #endif /* __LINUX_BIO_H */
-- 
2.28.0


^ permalink raw reply related	[flat|nested] 7+ messages in thread

* [PATCH V2 2/2] block: rename the local variable for holding return value of bio_iov_iter_nvecs
  2020-12-03  2:29 [PATCH V2 0/2] block: add bio_iov_iter_nvecs for figuring out nr_vecs Ming Lei
  2020-12-03  2:29 ` [PATCH V2 1/2] " Ming Lei
@ 2020-12-03  2:29 ` Ming Lei
  2020-12-03  6:47   ` Hannes Reinecke
  2020-12-09  8:32   ` Christoph Hellwig
  1 sibling, 2 replies; 7+ messages in thread
From: Ming Lei @ 2020-12-03  2:29 UTC (permalink / raw)
  To: Jens Axboe
  Cc: linux-block, Ming Lei, Pavel Begunkov, Christoph Hellwig,
	Matthew Wilcox, linux-fsdevel

Now the local variable for holding return value of bio_iov_iter_nvecs is
'nr_pages', which is a bit misleading, and the actual meaning is number
of bio vectors, and it is also used for this way.

So rename the local variable, and no function change.

Reviewed-by: Pavel Begunkov <asml.silence@gmail.com>
Reviewed-by: Christoph Hellwig <hch@infradead.org>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: linux-fsdevel@vger.kernel.org
Signed-off-by: Ming Lei <ming.lei@redhat.com>
---
 fs/block_dev.c       | 30 +++++++++++++++---------------
 fs/iomap/direct-io.c | 14 +++++++-------
 2 files changed, 22 insertions(+), 22 deletions(-)

diff --git a/fs/block_dev.c b/fs/block_dev.c
index fbcc900229af..d699f3af1a09 100644
--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -215,7 +215,7 @@ static void blkdev_bio_end_io_simple(struct bio *bio)
 
 static ssize_t
 __blkdev_direct_IO_simple(struct kiocb *iocb, struct iov_iter *iter,
-		int nr_pages)
+		int nr_vecs)
 {
 	struct file *file = iocb->ki_filp;
 	struct block_device *bdev = I_BDEV(bdev_file_inode(file));
@@ -230,16 +230,16 @@ __blkdev_direct_IO_simple(struct kiocb *iocb, struct iov_iter *iter,
 	    (bdev_logical_block_size(bdev) - 1))
 		return -EINVAL;
 
-	if (nr_pages <= DIO_INLINE_BIO_VECS)
+	if (nr_vecs <= DIO_INLINE_BIO_VECS)
 		vecs = inline_vecs;
 	else {
-		vecs = kmalloc_array(nr_pages, sizeof(struct bio_vec),
+		vecs = kmalloc_array(nr_vecs, sizeof(struct bio_vec),
 				     GFP_KERNEL);
 		if (!vecs)
 			return -ENOMEM;
 	}
 
-	bio_init(&bio, vecs, nr_pages);
+	bio_init(&bio, vecs, nr_vecs);
 	bio_set_dev(&bio, bdev);
 	bio.bi_iter.bi_sector = pos >> 9;
 	bio.bi_write_hint = iocb->ki_hint;
@@ -350,7 +350,7 @@ static void blkdev_bio_end_io(struct bio *bio)
 }
 
 static ssize_t
-__blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter, int nr_pages)
+__blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter, int nr_vecs)
 {
 	struct file *file = iocb->ki_filp;
 	struct inode *inode = bdev_file_inode(file);
@@ -368,7 +368,7 @@ __blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter, int nr_pages)
 	    (bdev_logical_block_size(bdev) - 1))
 		return -EINVAL;
 
-	bio = bio_alloc_bioset(GFP_KERNEL, nr_pages, &blkdev_dio_pool);
+	bio = bio_alloc_bioset(GFP_KERNEL, nr_vecs, &blkdev_dio_pool);
 
 	dio = container_of(bio, struct blkdev_dio, bio);
 	dio->is_sync = is_sync = is_sync_kiocb(iocb);
@@ -417,8 +417,8 @@ __blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter, int nr_pages)
 		dio->size += bio->bi_iter.bi_size;
 		pos += bio->bi_iter.bi_size;
 
-		nr_pages = bio_iov_iter_nvecs(iter, BIO_MAX_PAGES);
-		if (!nr_pages) {
+		nr_vecs = bio_iov_iter_nvecs(iter, BIO_MAX_PAGES);
+		if (!nr_vecs) {
 			bool polled = false;
 
 			if (iocb->ki_flags & IOCB_HIPRI) {
@@ -448,7 +448,7 @@ __blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter, int nr_pages)
 		}
 
 		submit_bio(bio);
-		bio = bio_alloc(GFP_KERNEL, nr_pages);
+		bio = bio_alloc(GFP_KERNEL, nr_vecs);
 	}
 
 	if (!is_poll)
@@ -480,15 +480,15 @@ __blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter, int nr_pages)
 static ssize_t
 blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter)
 {
-	int nr_pages;
+	int nr_vecs;
 
-	nr_pages = bio_iov_iter_nvecs(iter, BIO_MAX_PAGES + 1);
-	if (!nr_pages)
+	nr_vecs = bio_iov_iter_nvecs(iter, BIO_MAX_PAGES + 1);
+	if (!nr_vecs)
 		return 0;
-	if (is_sync_kiocb(iocb) && nr_pages <= BIO_MAX_PAGES)
-		return __blkdev_direct_IO_simple(iocb, iter, nr_pages);
+	if (is_sync_kiocb(iocb) && nr_vecs <= BIO_MAX_PAGES)
+		return __blkdev_direct_IO_simple(iocb, iter, nr_vecs);
 
-	return __blkdev_direct_IO(iocb, iter, min(nr_pages, BIO_MAX_PAGES));
+	return __blkdev_direct_IO(iocb, iter, min(nr_vecs, BIO_MAX_PAGES));
 }
 
 static __init int blkdev_init(void)
diff --git a/fs/iomap/direct-io.c b/fs/iomap/direct-io.c
index 0635469e811a..cc779ecc8144 100644
--- a/fs/iomap/direct-io.c
+++ b/fs/iomap/direct-io.c
@@ -211,7 +211,7 @@ iomap_dio_bio_actor(struct inode *inode, loff_t pos, loff_t length,
 	struct bio *bio;
 	bool need_zeroout = false;
 	bool use_fua = false;
-	int nr_pages, ret = 0;
+	int nr_vecs, ret = 0;
 	size_t copied = 0;
 	size_t orig_count;
 
@@ -250,9 +250,9 @@ iomap_dio_bio_actor(struct inode *inode, loff_t pos, loff_t length,
 	orig_count = iov_iter_count(dio->submit.iter);
 	iov_iter_truncate(dio->submit.iter, length);
 
-	nr_pages = bio_iov_iter_nvecs(dio->submit.iter, BIO_MAX_PAGES);
-	if (nr_pages <= 0) {
-		ret = nr_pages;
+	nr_vecs = bio_iov_iter_nvecs(dio->submit.iter, BIO_MAX_PAGES);
+	if (nr_vecs <= 0) {
+		ret = nr_vecs;
 		goto out;
 	}
 
@@ -271,7 +271,7 @@ iomap_dio_bio_actor(struct inode *inode, loff_t pos, loff_t length,
 			goto out;
 		}
 
-		bio = bio_alloc(GFP_KERNEL, nr_pages);
+		bio = bio_alloc(GFP_KERNEL, nr_vecs);
 		bio_set_dev(bio, iomap->bdev);
 		bio->bi_iter.bi_sector = iomap_sector(iomap, pos);
 		bio->bi_write_hint = dio->iocb->ki_hint;
@@ -308,10 +308,10 @@ iomap_dio_bio_actor(struct inode *inode, loff_t pos, loff_t length,
 		dio->size += n;
 		copied += n;
 
-		nr_pages = bio_iov_iter_nvecs(dio->submit.iter, BIO_MAX_PAGES);
+		nr_vecs = bio_iov_iter_nvecs(dio->submit.iter, BIO_MAX_PAGES);
 		iomap_dio_submit_bio(dio, iomap, bio, pos);
 		pos += n;
-	} while (nr_pages);
+	} while (nr_vecs);
 
 	/*
 	 * We need to zeroout the tail of a sub-block write if the extent type
-- 
2.28.0


^ permalink raw reply related	[flat|nested] 7+ messages in thread

* Re: [PATCH V2 1/2] block: add bio_iov_iter_nvecs for figuring out nr_vecs
  2020-12-03  2:29 ` [PATCH V2 1/2] " Ming Lei
@ 2020-12-03  6:46   ` Hannes Reinecke
  2020-12-09  8:31   ` Christoph Hellwig
  1 sibling, 0 replies; 7+ messages in thread
From: Hannes Reinecke @ 2020-12-03  6:46 UTC (permalink / raw)
  To: Ming Lei, Jens Axboe
  Cc: linux-block, Pavel Begunkov, Christoph Hellwig, Matthew Wilcox,
	linux-fsdevel

On 12/3/20 3:29 AM, Ming Lei wrote:
> Pavel reported that iov_iter_npages is a bit heavy in case of bvec
> iter.
> 
> Turns out it isn't necessary to iterate every page in the bvec iter,
> and we call iov_iter_npages() just for figuring out how many bio
> vecs need to be allocated. And we can simply map each vector in bvec iter
> to bio's vec, so just return iter->nr_segs from bio_iov_iter_nvecs() for
> bvec iter.
> 
> This patch is based on Mathew's post:
> 
> https://lore.kernel.org/linux-block/20201120123931.GN29991@casper.infradead.org/
> 
> Reviewed-by: Pavel Begunkov <asml.silence@gmail.com>
> Reviewed-by: Christoph Hellwig <hch@infradead.org>
> Cc: Matthew Wilcox <willy@infradead.org>
> Cc: linux-fsdevel@vger.kernel.org
> Signed-off-by: Ming Lei <ming.lei@redhat.com>
> ---
>   fs/block_dev.c       |  4 ++--
>   fs/iomap/direct-io.c |  4 ++--
>   include/linux/bio.h  | 10 ++++++++++
>   3 files changed, 14 insertions(+), 4 deletions(-)
> 
Reviewed-by: Hannes Reinecke <hare@suse.de>

Cheers,

Hannes
-- 
Dr. Hannes Reinecke                Kernel Storage Architect
hare@suse.de                              +49 911 74053 688
SUSE Software Solutions GmbH, Maxfeldstr. 5, 90409 Nürnberg
HRB 36809 (AG Nürnberg), Geschäftsführer: Felix Imendörffer

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: [PATCH V2 2/2] block: rename the local variable for holding return value of bio_iov_iter_nvecs
  2020-12-03  2:29 ` [PATCH V2 2/2] block: rename the local variable for holding return value of bio_iov_iter_nvecs Ming Lei
@ 2020-12-03  6:47   ` Hannes Reinecke
  2020-12-09  8:32   ` Christoph Hellwig
  1 sibling, 0 replies; 7+ messages in thread
From: Hannes Reinecke @ 2020-12-03  6:47 UTC (permalink / raw)
  To: Ming Lei, Jens Axboe
  Cc: linux-block, Pavel Begunkov, Christoph Hellwig, Matthew Wilcox,
	linux-fsdevel

On 12/3/20 3:29 AM, Ming Lei wrote:
> Now the local variable for holding return value of bio_iov_iter_nvecs is
> 'nr_pages', which is a bit misleading, and the actual meaning is number
> of bio vectors, and it is also used for this way.
> 
> So rename the local variable, and no function change.
> 
> Reviewed-by: Pavel Begunkov <asml.silence@gmail.com>
> Reviewed-by: Christoph Hellwig <hch@infradead.org>
> Cc: Matthew Wilcox <willy@infradead.org>
> Cc: linux-fsdevel@vger.kernel.org
> Signed-off-by: Ming Lei <ming.lei@redhat.com>
> ---
>   fs/block_dev.c       | 30 +++++++++++++++---------------
>   fs/iomap/direct-io.c | 14 +++++++-------
>   2 files changed, 22 insertions(+), 22 deletions(-)
> 
Reviewed-by: Hannes Reinecke <hare@suse.de>

Cheers,

Hannes
-- 
Dr. Hannes Reinecke                Kernel Storage Architect
hare@suse.de                              +49 911 74053 688
SUSE Software Solutions GmbH, Maxfeldstr. 5, 90409 Nürnberg
HRB 36809 (AG Nürnberg), Geschäftsführer: Felix Imendörffer

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: [PATCH V2 1/2] block: add bio_iov_iter_nvecs for figuring out nr_vecs
  2020-12-03  2:29 ` [PATCH V2 1/2] " Ming Lei
  2020-12-03  6:46   ` Hannes Reinecke
@ 2020-12-09  8:31   ` Christoph Hellwig
  1 sibling, 0 replies; 7+ messages in thread
From: Christoph Hellwig @ 2020-12-09  8:31 UTC (permalink / raw)
  To: Ming Lei
  Cc: Jens Axboe, linux-block, Pavel Begunkov, Christoph Hellwig,
	Matthew Wilcox, linux-fsdevel

Looks good,

Reviewed-by: Christoph Hellwig <hch@lst.de>

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: [PATCH V2 2/2] block: rename the local variable for holding return value of bio_iov_iter_nvecs
  2020-12-03  2:29 ` [PATCH V2 2/2] block: rename the local variable for holding return value of bio_iov_iter_nvecs Ming Lei
  2020-12-03  6:47   ` Hannes Reinecke
@ 2020-12-09  8:32   ` Christoph Hellwig
  1 sibling, 0 replies; 7+ messages in thread
From: Christoph Hellwig @ 2020-12-09  8:32 UTC (permalink / raw)
  To: Ming Lei
  Cc: Jens Axboe, linux-block, Pavel Begunkov, Christoph Hellwig,
	Matthew Wilcox, linux-fsdevel

Looks good,

Reviewed-by: Christoph Hellwig <hch@lst.de>

^ permalink raw reply	[flat|nested] 7+ messages in thread

end of thread, other threads:[~2020-12-09  8:33 UTC | newest]

Thread overview: 7+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-12-03  2:29 [PATCH V2 0/2] block: add bio_iov_iter_nvecs for figuring out nr_vecs Ming Lei
2020-12-03  2:29 ` [PATCH V2 1/2] " Ming Lei
2020-12-03  6:46   ` Hannes Reinecke
2020-12-09  8:31   ` Christoph Hellwig
2020-12-03  2:29 ` [PATCH V2 2/2] block: rename the local variable for holding return value of bio_iov_iter_nvecs Ming Lei
2020-12-03  6:47   ` Hannes Reinecke
2020-12-09  8:32   ` Christoph Hellwig

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).