Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.kernel > #1582504
| From | Ming Lei <tom.leiming@gmail.com> |
|---|---|
| Newsgroups | linux.kernel |
| Subject | [PATCH 14/17] md: raid10: avoid direct access to bvec table in sync_request_write() |
| Date | 2017-02-16 12:50 +0100 |
| Message-ID | <tbsJs-5E4-29@gated-at.bofh.it> (permalink) |
| References | <tbsJr-5E4-3@gated-at.bofh.it> |
| Organization | linux.* mail to news gateway |
The cost is 256bytes(8*16*2) stack space, and just use the bio
helper to retrieve pages from bio.
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
---
drivers/md/raid10.c | 15 +++++++++++++--
1 file changed, 13 insertions(+), 2 deletions(-)
diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c
index 5c698f3d3083..69fe2a3cef89 100644
--- a/drivers/md/raid10.c
+++ b/drivers/md/raid10.c
@@ -2036,6 +2036,8 @@ static void sync_request_write(struct mddev *mddev, struct r10bio *r10_bio)
int i, first;
struct bio *tbio, *fbio;
int vcnt;
+ struct bio_vec *bvl;
+ struct page *fbio_pages[RESYNC_PAGES], *tbio_pages[RESYNC_PAGES];
atomic_set(&r10_bio->remaining, 1);
@@ -2052,6 +2054,10 @@ static void sync_request_write(struct mddev *mddev, struct r10bio *r10_bio)
fbio->bi_iter.bi_size = r10_bio->sectors << 9;
fbio->bi_iter.bi_idx = 0;
+ /* the bio has been filled up in raid10_sync_request */
+ bio_for_each_segment_all(bvl, fbio, i)
+ fbio_pages[i] = bvl->bv_page;
+
vcnt = (r10_bio->sectors + (PAGE_SIZE >> 9) - 1) >> (PAGE_SHIFT - 9);
/* now find blocks with errors */
for (i=0 ; i < conf->copies ; i++) {
@@ -2072,12 +2078,17 @@ static void sync_request_write(struct mddev *mddev, struct r10bio *r10_bio)
* All vec entries are PAGE_SIZE;
*/
int sectors = r10_bio->sectors;
+
+ /* the bio has been filled up in raid10_sync_request */
+ bio_for_each_segment_all(bvl, tbio, j)
+ tbio_pages[j] = bvl->bv_page;
+
for (j = 0; j < vcnt; j++) {
int len = PAGE_SIZE;
if (sectors < (len / 512))
len = sectors * 512;
- if (memcmp(page_address(fbio->bi_io_vec[j].bv_page),
- page_address(tbio->bi_io_vec[j].bv_page),
+ if (memcmp(page_address(fbio_pages[j]),
+ page_address(tbio_pages[j]),
len))
break;
sectors -= len/512;
--
2.7.4
Back to linux.kernel | Previous | Next | Find similar | Unroll thread
[PATCH 14/17] md: raid10: avoid direct access to bvec table in sync_request_write() Ming Lei <tom.leiming@gmail.com> - 2017-02-16 12:50 +0100
csiph-web