From: "Jianpeng Ma" <majianpeng@gmail.com>
To: shli <shli@kernel.org>
Cc: Neil Brown <neilb@suse.de>, linux-raid <linux-raid@vger.kernel.org>
Subject: Re: Re: [patch 2/2 v3]raid5: create multiple threads to handle stripes
Date: Mon, 24 Sep 2012 19:15:12 +0800 [thread overview]
Message-ID: <201209241914270784655@gmail.com> (raw)
In-Reply-To: CANejiEU_JeNGEeMO+u8_8pXt6jVfb5SBfnLUJSsSCrfr9bgi0A@mail.gmail.com
Sorry for using this title to send this.
This code maybe error,so it should correct(If multiple-thread dosen't enable,it maybe also error,but no obviously).
In func handle_parity_checks5:
>>conf->mddev->resync_mismatches += STRIPE_SECTORS;
If multi-thread handle stripe on different cpu, there will be raced on resync_mismatches.
So the patch is:
Subject: [PATCH] md:change resync_mismatches to atomic64_t to avoid races
Signed-off-by: Jianpeng Ma <majianpeng@gmail.com>
---
drivers/md/md.c | 7 ++++---
drivers/md/md.h | 2 +-
drivers/md/raid1.c | 2 +-
drivers/md/raid10.c | 2 +-
drivers/md/raid5.c | 4 ++--
5 files changed, 9 insertions(+), 8 deletions(-)
diff --git a/drivers/md/md.c b/drivers/md/md.c
index 308e87b..b0b9474 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -4278,7 +4278,8 @@ static ssize_t
mismatch_cnt_show(struct mddev *mddev, char *page)
{
return sprintf(page, "%llu\n",
- (unsigned long long) mddev->resync_mismatches);
+ (unsigned long long)
+ atomic64_read(&mddev->resync_mismatches));
}
static struct md_sysfs_entry md_scan_mode =
@@ -5245,7 +5246,7 @@ static void md_clean(struct mddev *mddev)
mddev->new_layout = 0;
mddev->new_chunk_sectors = 0;
mddev->curr_resync = 0;
- mddev->resync_mismatches = 0;
+ atomic64_set(&mddev->resync_mismatches, 0);
mddev->suspend_lo = mddev->suspend_hi = 0;
mddev->sync_speed_min = mddev->sync_speed_max = 0;
mddev->recovery = 0;
@@ -7349,7 +7350,7 @@ void md_do_sync(struct mddev *mddev)
* which defaults to physical size, but can be virtual size
*/
max_sectors = mddev->resync_max_sectors;
- mddev->resync_mismatches = 0;
+ atomic64_set(&mddev->resync_mismatches, 0);
/* we don't use the checkpoint if there's a bitmap */
if (test_bit(MD_RECOVERY_REQUESTED, &mddev->recovery))
j = mddev->resync_min;
diff --git a/drivers/md/md.h b/drivers/md/md.h
index f385b03..8f341d8 100644
--- a/drivers/md/md.h
+++ b/drivers/md/md.h
@@ -282,7 +282,7 @@ struct mddev {
sector_t resync_max_sectors; /* may be set by personality */
- sector_t resync_mismatches; /* count of sectors where
+ atomic64_t resync_mismatches; /* count of sectors where
* parity/replica mismatch found
*/
diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c
index 611b5f7..ebceb98 100644
--- a/drivers/md/raid1.c
+++ b/drivers/md/raid1.c
@@ -1867,7 +1867,7 @@ static int process_checks(struct r1bio *r1_bio)
} else
j = 0;
if (j >= 0)
- mddev->resync_mismatches += r1_bio->sectors;
+ atomic64_add(r1_bio->sectors, &mddev->resync_mismatches);
if (j < 0 || (test_bit(MD_RECOVERY_CHECK, &mddev->recovery)
&& test_bit(BIO_UPTODATE, &sbio->bi_flags))) {
/* No need to write to this device. */
diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c
index 1c2eb38..ad0a8b7 100644
--- a/drivers/md/raid10.c
+++ b/drivers/md/raid10.c
@@ -1950,7 +1950,7 @@ static void sync_request_write(struct mddev *mddev, struct r10bio *r10_bio)
break;
if (j == vcnt)
continue;
- mddev->resync_mismatches += r10_bio->sectors;
+ atomic64_add(r10_bio->sectors, &mddev->resync_mismatches);
if (test_bit(MD_RECOVERY_CHECK, &mddev->recovery))
/* Don't fix anything. */
continue;
diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index 7031b86..0a2702f 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -2931,7 +2931,7 @@ static void handle_parity_checks5(struct r5conf *conf, struct stripe_head *sh,
*/
set_bit(STRIPE_INSYNC, &sh->state);
else {
- conf->mddev->resync_mismatches += STRIPE_SECTORS;
+ atomic64_add(STRIPE_SECTORS, &conf->mddev->resync_mismatches);
if (test_bit(MD_RECOVERY_CHECK, &conf->mddev->recovery))
/* don't try to repair!! */
set_bit(STRIPE_INSYNC, &sh->state);
@@ -3083,7 +3083,7 @@ static void handle_parity_checks6(struct r5conf *conf, struct stripe_head *sh,
*/
}
} else {
- conf->mddev->resync_mismatches += STRIPE_SECTORS;
+ atomic64_add(STRIPE_SECTORS, &conf->mddev->resync_mismatches);
if (test_bit(MD_RECOVERY_CHECK, &conf->mddev->recovery))
/* don't try to repair!! */
set_bit(STRIPE_INSYNC, &sh->state);
--
1.7.9.5
next prev parent reply other threads:[~2012-09-24 11:15 UTC|newest]
Thread overview: 29+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-08-09 8:58 [patch 2/2 v3]raid5: create multiple threads to handle stripes Shaohua Li
2012-08-11 8:45 ` Jianpeng Ma
2012-08-13 0:21 ` Shaohua Li
2012-08-13 1:06 ` Jianpeng Ma
2012-08-13 2:13 ` Shaohua Li
2012-08-13 2:20 ` Shaohua Li
2012-08-13 2:25 ` Jianpeng Ma
2012-08-13 4:21 ` NeilBrown
2012-08-14 10:39 ` Jianpeng Ma
2012-08-15 3:51 ` Shaohua Li
2012-08-15 6:21 ` Jianpeng Ma
2012-08-15 8:04 ` Shaohua Li
2012-08-15 8:19 ` Jianpeng Ma
2012-09-24 11:15 ` Jianpeng Ma [this message]
2012-09-26 1:26 ` NeilBrown
2012-08-13 9:11 ` Jianpeng Ma
2012-08-13 4:29 ` NeilBrown
2012-08-13 6:22 ` Shaohua Li
2013-03-07 7:31 ` Shaohua Li
2013-03-12 1:39 ` NeilBrown
2013-03-13 0:44 ` Stan Hoeppner
2013-03-28 6:47 ` NeilBrown
2013-03-28 16:53 ` Stan Hoeppner
2013-03-29 2:34 ` Shaohua Li
2013-03-29 9:36 ` Stan Hoeppner
2013-04-01 1:57 ` Shaohua Li
2013-04-01 19:31 ` Stan Hoeppner
2013-04-02 0:39 ` Shaohua Li
2013-04-02 3:12 ` Stan Hoeppner
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=201209241914270784655@gmail.com \
--to=majianpeng@gmail.com \
--cc=linux-raid@vger.kernel.org \
--cc=neilb@suse.de \
--cc=shli@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
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).