block: fix bio_add_page for non trivial merge_bvec_fn case

PATCH In review Dmitry Monakhov 2 participants 5 messages Jan 26, 2010 – Jan 27, 2010 VFS

Share & tools

Dmitry Monakhov

author

2010-01-26 13:15 UTC #0 lore raw

PATCH 1/1 block: fix bio_add_page for non trivial merge_bvec_fn case

Hi, year ago I've sent a patch which fix false bio merge rejects, but seems patch was missed. Currently the issue is still present.

Quoted 1 line

From 92a97ef181e15caa94bd56a1ade5c337db599b79 Mon Sep 17 00:00:00 2001

From: Dmitry Monakhov <[email protected]>
Date: Tue, 26 Jan 2010 16:01:34 +0300
Subject: [PATCH] [PATCH] block: fix bio_add_page for non trivial merge_bvec_fn case

We have to properly decrease bi_size in order to merge_bvec_fn return right result. Otherwise this result in false merge rejects for two absolutely valid bio_vecs. This may cause significant performance penalty for example Itanium: page_size == 16k, fs_block_size == 1k and block device is raid with small chunk_size.

Signed-off-by Dmitry Monakhov <[email protected]>

---
 fs/bio.c |    3 ++-
 1 files changed, 2 insertions(+), 1 deletions(-)

fs/bio.c +2 -1

index 76e6713..9f8e517 100644
@@ -548,7 +548,8 @@ static int __bio_add_page(struct request_queue *q, struct bio *bio, struct page
 				struct bvec_merge_data bvm = {
 					.bi_bdev = bio->bi_bdev,
 					.bi_sector = bio->bi_sector,
-					.bi_size = bio->bi_size,
+					.bi_size = bio->bi_size -
+							(prev->bv_len - len),
 					.bi_rw = bio->bi_rw,
 				};

Signature · 1 line

1.6.3.3

Jens Axboe

2010-01-26 13:29 UTC #1 lore raw

Quoted 25 lines · Dmitry Monakhov

Hi, year ago I've sent a patch which fix false bio merge rejects, but seems patch was missed. Currently the issue is still present.
From 92a97ef181e15caa94bd56a1ade5c337db599b79 Mon Sep 17 00:00:00 2001
From: Dmitry Monakhov <[email protected]>
Date: Tue, 26 Jan 2010 16:01:34 +0300
Subject: [PATCH] [PATCH] block: fix bio_add_page for non trivial merge_bvec_fn case
We have to properly decrease bi_size in order to merge_bvec_fn return right result. Otherwise this result in false merge rejects for two absolutely valid bio_vecs. This may cause significant performance penalty for example Itanium: page_size == 16k, fs_block_size == 1k and block device is raid with small chunk_size.
Signed-off-by Dmitry Monakhov <[email protected]>
---
 fs/bio.c |    3 ++-
 1 files changed, 2 insertions(+), 1 deletions(-)
fs/bio.c +0 -0
index 76e6713..9f8e517 100644

fs/bio.c static int __bio_add_page(struct request_queue *q, struct bio *bio, struct page · line 554

Quoted 8 lines

fs/bio.c +2 -1

@@ -548,7 +548,8 @@ static int __bio_add_page(struct request_queue *q, struct bio *bio, struct page
 				struct bvec_merge_data bvm = {
 					.bi_bdev = bio->bi_bdev,
 					.bi_sector = bio->bi_sector,
-					.bi_size = bio->bi_size,
+					.bi_size = bio->bi_size -
+							(prev->bv_len - len),
 					.bi_rw = bio->bi_rw,
 				};

Hmm confused. why isn't this just bio->bi_size - len?

Signature · 1 line

Jens Axboe

Dmitry Monakhov

author

2010-01-26 14:17 UTC #2 lore raw

Quoted 21 lines · Jens Axboe

Quoted 20 lines · Dmitry Monakhov
Hi, year ago I've sent a patch which fix false bio merge rejects, but seems patch was missed. Currently the issue is still present.
From 92a97ef181e15caa94bd56a1ade5c337db599b79 Mon Sep 17 00:00:00 2001
From: Dmitry Monakhov <[email protected]>
Date: Tue, 26 Jan 2010 16:01:34 +0300
Subject: [PATCH] [PATCH] block: fix bio_add_page for non trivial merge_bvec_fn case
We have to properly decrease bi_size in order to merge_bvec_fn return right result. Otherwise this result in false merge rejects for two absolutely valid bio_vecs. This may cause significant performance penalty for example Itanium: page_size == 16k, fs_block_size == 1k and block device is raid with small chunk_size.
Signed-off-by Dmitry Monakhov <[email protected]>
---
fs/bio.c | 3 ++-
 1 files changed, 2 insertions(+), 1 deletions(-)

Quoted 15 lines

fs/bio.c static int __bio_add_page(struct request_queue *q, struct bio *bio, struct page · line 554

Quoted 13 lines

fs/bio.c +2 -1

index 76e6713..9f8e517 100644
@@ -548,7 +548,8 @@ static int __bio_add_page(struct request_queue *q, struct bio *bio, struct page
                               struct bvec_merge_data bvm = {
                                       .bi_bdev = bio->bi_bdev,
                                       .bi_sector = bio->bi_sector,
-                                     .bi_size = bio->bi_size,
+                                     .bi_size = bio->bi_size -
+                                                     (prev->bv_len - len),
                                       .bi_rw = bio->bi_rw,
                               };

Hmm confused. why isn't this just bio->bi_size - len?

because we have following scenario:
0) old_bv_len = prev->bv_len ;
1) prev->bv_len += len;
2)->merge_bvec_fn()
usually it looks like follows
if (bio->bv_len + bvm->bi_size > max_chunk_size)
goto fail
So formally we detach last bvec increase it size and try to
attach it to bio.

but old_bv_len is already accounted in bi_size, so we have to decrease
bi_size to this value, and old_bv_len = (prev->bv_len - len) at
that moment.
I've used math manipuation in order to avoid temporary variable because stack size is matter on block layer.

Quoted 5 lines

--
Jens Axboe

Dmitry Monakhov

author

2010-01-27 21:11 UTC #3 lore raw

Quoted 22 lines · Dmitry Monakhov

Quoted 21 lines · Jens Axboe
Quoted 20 lines · Dmitry Monakhov
Hi, year ago I've sent a patch which fix false bio merge rejects, but seems patch was missed. Currently the issue is still present.
From 92a97ef181e15caa94bd56a1ade5c337db599b79 Mon Sep 17 00:00:00 2001
From: Dmitry Monakhov <[email protected]>
Date: Tue, 26 Jan 2010 16:01:34 +0300
Subject: [PATCH] [PATCH] block: fix bio_add_page for non trivial merge_bvec_fn case
We have to properly decrease bi_size in order to merge_bvec_fn return right result. Otherwise this result in false merge rejects for two absolutely valid bio_vecs. This may cause significant performance penalty for example Itanium: page_size == 16k, fs_block_size == 1k and block device is raid with small chunk_size.
Signed-off-by Dmitry Monakhov <[email protected]>
---
fs/bio.c | 3 ++-
 1 files changed, 2 insertions(+), 1 deletions(-)

Quoted 15 lines

Quoted 7 lines

fs/bio.c +0 -0

index 76e6713..9f8e517 100644
@@ -548,7 +548,8 @@ static int __bio_add_page(struct request_queue *q, struct bio *bio, struct page
                               struct bvec_merge_data bvm = {
                                       .bi_bdev = bio->bi_bdev,

Quoted 8 lines

fs/bio.c static int __bio_add_page(struct request_queue *q, struct bio *bio, struct page · line 554

Quoted 6 lines

fs/bio.c +2 -1

                                       .bi_sector = bio->bi_sector,
-                                     .bi_size = bio->bi_size,
+                                     .bi_size = bio->bi_size -
+                                                     (prev->bv_len - len),
                                       .bi_rw = bio->bi_rw,
                               };

Hmm confused. why isn't this just bio->bi_size - len?

I've attached more descriptive version of the patch. Jens, please clarify your opinion to the patch( do you like it or not?) I don't want it miss again.

Quoted 1 line

From 65ac8c9a2988310a79215b68e9c93b6dca6c6665 Mon Sep 17 00:00:00 2001

From: Dmitry Monakhov <[email protected]>
Date: Wed, 27 Jan 2010 22:44:36 +0300
Subject: [PATCH] [PATCH] block: fix bio_add_page for non trivial merge_bvec_fn case

We have to properly decrease bi_size in order to merge_bvec_fn return right result. Otherwise this result in false merge rejects for two absolutely valid bio_vecs. This may cause significant performance penalty for example fs_block_size == 1k and block device is raid0 with small chunk_size = 8k. Then it is impossible to merge 7-th fs-block in to bio which already has 6 fs-blocks.

Signed-off-by Dmitry Monakhov <[email protected]>

---
 fs/bio.c |    7 ++++++-
 1 files changed, 6 insertions(+), 1 deletions(-)

fs/bio.c +6 -1

index 76e6713..7448c7d 100644
@@ -542,13 +542,18 @@ static int __bio_add_page(struct request_queue *q, struct bio *bio, struct page
 
 		if (page == prev->bv_page &&
 		    offset == prev->bv_offset + prev->bv_len) {
+			unsigned int prev_bv_len = prev->bv_len;
 			prev->bv_len += len;
 
 			if (q->merge_bvec_fn) {
 				struct bvec_merge_data bvm = {
+					/* prev_bvec is already charged in
+					   bi_size, discharge it in order to
+					   simulate merging updated prev_bvec
+					   as new bvec. */
 					.bi_bdev = bio->bi_bdev,
 					.bi_sector = bio->bi_sector,
-					.bi_size = bio->bi_size,
+					.bi_size = bio->bi_size - prev_bv_len,
 					.bi_rw = bio->bi_rw,
 				};

Signature · 1 line

1.6.3.3

Jens Axboe

2010-01-27 21:14 UTC #4 lore raw

Quoted 39 lines · Dmitry Monakhov

Quoted 22 lines · Dmitry Monakhov
Quoted 21 lines · Jens Axboe
Quoted 20 lines · Dmitry Monakhov
Hi, year ago I've sent a patch which fix false bio merge rejects, but seems patch was missed. Currently the issue is still present.
From 92a97ef181e15caa94bd56a1ade5c337db599b79 Mon Sep 17 00:00:00 2001
From: Dmitry Monakhov <[email protected]>
Date: Tue, 26 Jan 2010 16:01:34 +0300
Subject: [PATCH] [PATCH] block: fix bio_add_page for non trivial merge_bvec_fn case
We have to properly decrease bi_size in order to merge_bvec_fn return right result. Otherwise this result in false merge rejects for two absolutely valid bio_vecs. This may cause significant performance penalty for example Itanium: page_size == 16k, fs_block_size == 1k and block device is raid with small chunk_size.
Signed-off-by Dmitry Monakhov <[email protected]>
---
fs/bio.c | 3 ++-
 1 files changed, 2 insertions(+), 1 deletions(-)
Quoted 15 lines
Quoted 7 lines
Quoted 7 lines
fs/bio.c +0 -0
index 76e6713..9f8e517 100644
@@ -548,7 +548,8 @@ static int __bio_add_page(struct request_queue *q, struct bio *bio, struct page
                               struct bvec_merge_data bvm = {
                                       .bi_bdev = bio->bi_bdev,
Quoted 8 lines
fs/bio.c static int __bio_add_page(struct request_queue *q, struct bio *bio, struct page · line 558
Quoted 6 lines
fs/bio.c +2 -1
                                       .bi_sector = bio->bi_sector,
-                                     .bi_size = bio->bi_size,
+                                     .bi_size = bio->bi_size -
+                                                     (prev->bv_len - len),
                                       .bi_rw = bio->bi_rw,
                               };
Hmm confused. why isn't this just bio->bi_size - len?

Quoted 3 lines

I've attached more descriptive version of the patch. Jens, please clarify your opinion to the patch( do you like it or not?) I don't want it miss again.

Yes thanks, I agree with your evaluation, it's definitely correct. I'll add it.

Signature · 1 line

Jens Axboe