| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-Id: <20250109015145.158868-6-yukuai1@huaweicloud.com>
Date: Thu, 9 Jan 2025 09:51:45 +0800
From: Yu Kuai <yukuai1@...weicloud.com>
To: song@...nel.org,
xni@...hat.com,
yukuai3@...wei.com
Cc: linux-raid@...r.kernel.org,
linux-kernel@...r.kernel.org,
yukuai1@...weicloud.com,
yi.zhang@...wei.com,
yangerkun@...wei.com
Subject: [PATCH v2 md-6.14 5/5] md/md-bitmap: move bitmap_{start, end}write to md upper layer
From: Yu Kuai <yukuai3@...wei.com>
There are two BUG reports that raid5 will hang at
bitmap_startwrite([1],[2]), root cause is that bitmap start write and end
write is unbalanced, it's not quite clear where, and while reviewing raid5
code, it's found that bitmap operations can be optimized. For example,
for a 4 disks raid5, with chunksize=8k, if user issue a IO (0 + 48k) to
the array:
┌────────────────────────────────────────────────────────────┐
│chunk 0 │
│ ┌────────────┬─────────────┬─────────────┬────────────┼
│ sh0 │A0: 0 + 4k │A1: 8k + 4k │A2: 16k + 4k │A3: P │
│ ┼────────────┼─────────────┼─────────────┼────────────┼
│ sh1 │B0: 4k + 4k │B1: 12k + 4k │B2: 20k + 4k │B3: P │
┼──────┴────────────┴─────────────┴─────────────┴────────────┼
│chunk 1 │
│ ┌────────────┬─────────────┬─────────────┬────────────┤
│ sh2 │C0: 24k + 4k│C1: 32k + 4k │C2: P │C3: 40k + 4k│
│ ┼────────────┼─────────────┼─────────────┼────────────┼
│ sh3 │D0: 28k + 4k│D1: 36k + 4k │D2: P │D3: 44k + 4k│
└──────┴────────────┴─────────────┴─────────────┴────────────┘
Before this patch, 4 stripe head will be used, and each sh will attach
bio for 3 disks, and each attached bio will trigger
bitmap_startwrite() once, which means total 12 times.
- 3 times (0 + 4k), for (A0, A1 and A2)
- 3 times (4 + 4k), for (B0, B1 and B2)
- 3 times (8 + 4k), for (C0, C1 and C3)
- 3 times (12 + 4k), for (D0, D1 and D3)
After this patch, md upper layer will calculate that IO range (0 + 48k)
is corresponding to the bitmap (0 + 16k), and call bitmap_startwrite()
just once.
Noted that this patch will align bitmap ranges to the chunks, for example,
if user issue a IO (0 + 4k) to array:
- Before this patch, 1 time (0 + 4k), for A0;
- After this patch, 1 time (0 + 8k) for chunk 0;
Usually, one bitmap bit will represent more than one disk chunk, and this
doesn't have any difference. And even if user really created a array
that one chunk contain multiple bits, the overhead is that more data
will be recovered after power failure.
Also remove STRIPE_BITMAP_PENDING since it's not used anymore.
[1] https://lore.kernel.org/all/CAJpMwyjmHQLvm6zg1cmQErttNNQPDAAXPKM3xgTjMhbfts986Q@mail.gmail.com/
[2] https://lore.kernel.org/all/ADF7D720-5764-4AF3-B68E-1845988737AA@flyingcircus.io/
Signed-off-by: Yu Kuai <yukuai3@...wei.com>
---
drivers/md/md.c | 29 +++++++++++++++++++++++
drivers/md/md.h | 2 ++
drivers/md/raid1.c | 4 ----
drivers/md/raid10.c | 3 ---
drivers/md/raid5-cache.c | 2 --
drivers/md/raid5.c | 50 +++++-----------------------------------
drivers/md/raid5.h | 3 ---
7 files changed, 37 insertions(+), 56 deletions(-)
diff --git a/drivers/md/md.c b/drivers/md/md.c
index 3dd013f82e26..866015b681af 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -8745,12 +8745,32 @@ void md_submit_discard_bio(struct mddev *mddev, struct md_rdev *rdev,
}
EXPORT_SYMBOL_GPL(md_submit_discard_bio);
+static void md_bitmap_start(struct mddev *mddev,
+ struct md_io_clone *md_io_clone)
+{
+ if (mddev->pers->bitmap_sector)
+ mddev->pers->bitmap_sector(mddev, &md_io_clone->offset,
+ &md_io_clone->sectors);
+
+ mddev->bitmap_ops->startwrite(mddev, md_io_clone->offset,
+ md_io_clone->sectors);
+}
+
+static void md_bitmap_end(struct mddev *mddev, struct md_io_clone *md_io_clone)
+{
+ mddev->bitmap_ops->endwrite(mddev, md_io_clone->offset,
+ md_io_clone->sectors);
+}
+
static void md_end_clone_io(struct bio *bio)
{
struct md_io_clone *md_io_clone = bio->bi_private;
struct bio *orig_bio = md_io_clone->orig_bio;
struct mddev *mddev = md_io_clone->mddev;
+ if (bio_data_dir(orig_bio) == WRITE && mddev->bitmap)
+ md_bitmap_end(mddev, md_io_clone);
+
if (bio->bi_status && !orig_bio->bi_status)
orig_bio->bi_status = bio->bi_status;
@@ -8775,6 +8795,12 @@ static void md_clone_bio(struct mddev *mddev, struct bio **bio)
if (blk_queue_io_stat(bdev->bd_disk->queue))
md_io_clone->start_time = bio_start_io_acct(*bio);
+ if (bio_data_dir(*bio) == WRITE && mddev->bitmap) {
+ md_io_clone->offset = (*bio)->bi_iter.bi_sector;
+ md_io_clone->sectors = bio_sectors(*bio);
+ md_bitmap_start(mddev, md_io_clone);
+ }
+
clone->bi_end_io = md_end_clone_io;
clone->bi_private = md_io_clone;
*bio = clone;
@@ -8793,6 +8819,9 @@ void md_free_cloned_bio(struct bio *bio)
struct bio *orig_bio = md_io_clone->orig_bio;
struct mddev *mddev = md_io_clone->mddev;
+ if (bio_data_dir(orig_bio) == WRITE && mddev->bitmap)
+ md_bitmap_end(mddev, md_io_clone);
+
if (bio->bi_status && !orig_bio->bi_status)
orig_bio->bi_status = bio->bi_status;
diff --git a/drivers/md/md.h b/drivers/md/md.h
index de6dadb9a40b..def808064ad8 100644
--- a/drivers/md/md.h
+++ b/drivers/md/md.h
@@ -831,6 +831,8 @@ struct md_io_clone {
struct mddev *mddev;
struct bio *orig_bio;
unsigned long start_time;
+ sector_t offset;
+ unsigned long sectors;
struct bio bio_clone;
};
diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c
index 4d09d85321b6..a5cd6522fc2d 100644
--- a/drivers/md/raid1.c
+++ b/drivers/md/raid1.c
@@ -422,8 +422,6 @@ static void close_write(struct r1bio *r1_bio)
if (test_bit(R1BIO_BehindIO, &r1_bio->state))
mddev->bitmap_ops->end_behind_write(mddev);
- /* clear the bitmap if all writes complete successfully */
- mddev->bitmap_ops->endwrite(mddev, r1_bio->sector, r1_bio->sectors);
md_write_end(mddev);
}
@@ -1632,8 +1630,6 @@ static void raid1_write_request(struct mddev *mddev, struct bio *bio,
if (test_bit(R1BIO_BehindIO, &r1_bio->state))
mddev->bitmap_ops->start_behind_write(mddev);
- mddev->bitmap_ops->startwrite(mddev, r1_bio->sector,
- r1_bio->sectors);
first_clone = 0;
}
diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c
index 340a4710c222..e1e6cd7fb125 100644
--- a/drivers/md/raid10.c
+++ b/drivers/md/raid10.c
@@ -428,8 +428,6 @@ static void close_write(struct r10bio *r10_bio)
{
struct mddev *mddev = r10_bio->mddev;
- /* clear the bitmap if all writes complete successfully */
- mddev->bitmap_ops->endwrite(mddev, r10_bio->sector, r10_bio->sectors);
md_write_end(mddev);
}
@@ -1506,7 +1504,6 @@ static void raid10_write_request(struct mddev *mddev, struct bio *bio,
md_account_bio(mddev, &bio);
r10_bio->master_bio = bio;
atomic_set(&r10_bio->remaining, 1);
- mddev->bitmap_ops->startwrite(mddev, r10_bio->sector, r10_bio->sectors);
for (i = 0; i < conf->copies; i++) {
if (r10_bio->devs[i].bio)
diff --git a/drivers/md/raid5-cache.c b/drivers/md/raid5-cache.c
index ba4f9577c737..011246e16a99 100644
--- a/drivers/md/raid5-cache.c
+++ b/drivers/md/raid5-cache.c
@@ -313,8 +313,6 @@ void r5c_handle_cached_data_endio(struct r5conf *conf,
if (sh->dev[i].written) {
set_bit(R5_UPTODATE, &sh->dev[i].flags);
r5c_return_dev_pending_writes(conf, &sh->dev[i]);
- conf->mddev->bitmap_ops->endwrite(conf->mddev,
- sh->sector, RAID5_STRIPE_SECTORS(conf));
}
}
}
diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index 5377f4c3fffc..c97df8acf0ba 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -913,8 +913,7 @@ static bool stripe_can_batch(struct stripe_head *sh)
if (raid5_has_log(conf) || raid5_has_ppl(conf))
return false;
return test_bit(STRIPE_BATCH_READY, &sh->state) &&
- !test_bit(STRIPE_BITMAP_PENDING, &sh->state) &&
- is_full_stripe_write(sh);
+ is_full_stripe_write(sh);
}
/* we only do back search */
@@ -3640,29 +3639,9 @@ static void __add_stripe_bio(struct stripe_head *sh, struct bio *bi,
(*bip)->bi_iter.bi_sector, sh->sector, dd_idx,
sh->dev[dd_idx].sector);
- if (conf->mddev->bitmap && firstwrite) {
- /* Cannot hold spinlock over bitmap_startwrite,
- * but must ensure this isn't added to a batch until
- * we have added to the bitmap and set bm_seq.
- * So set STRIPE_BITMAP_PENDING to prevent
- * batching.
- * If multiple __add_stripe_bio() calls race here they
- * much all set STRIPE_BITMAP_PENDING. So only the first one
- * to complete "bitmap_startwrite" gets to set
- * STRIPE_BIT_DELAY. This is important as once a stripe
- * is added to a batch, STRIPE_BIT_DELAY cannot be changed
- * any more.
- */
- set_bit(STRIPE_BITMAP_PENDING, &sh->state);
- spin_unlock_irq(&sh->stripe_lock);
- conf->mddev->bitmap_ops->startwrite(conf->mddev, sh->sector,
- RAID5_STRIPE_SECTORS(conf));
- spin_lock_irq(&sh->stripe_lock);
- clear_bit(STRIPE_BITMAP_PENDING, &sh->state);
- if (!sh->batch_head) {
- sh->bm_seq = conf->seq_flush+1;
- set_bit(STRIPE_BIT_DELAY, &sh->state);
- }
+ if (conf->mddev->bitmap && firstwrite && !sh->batch_head) {
+ sh->bm_seq = conf->seq_flush+1;
+ set_bit(STRIPE_BIT_DELAY, &sh->state);
}
}
@@ -3713,7 +3692,6 @@ handle_failed_stripe(struct r5conf *conf, struct stripe_head *sh,
BUG_ON(sh->batch_head);
for (i = disks; i--; ) {
struct bio *bi;
- int bitmap_end = 0;
if (test_bit(R5_ReadError, &sh->dev[i].flags)) {
struct md_rdev *rdev = conf->disks[i].rdev;
@@ -3738,8 +3716,6 @@ handle_failed_stripe(struct r5conf *conf, struct stripe_head *sh,
sh->dev[i].towrite = NULL;
sh->overwrite_disks = 0;
spin_unlock_irq(&sh->stripe_lock);
- if (bi)
- bitmap_end = 1;
log_stripe_write_finished(sh);
@@ -3754,10 +3730,6 @@ handle_failed_stripe(struct r5conf *conf, struct stripe_head *sh,
bio_io_error(bi);
bi = nextbi;
}
- if (bitmap_end)
- conf->mddev->bitmap_ops->endwrite(conf->mddev,
- sh->sector, RAID5_STRIPE_SECTORS(conf));
- bitmap_end = 0;
/* and fail all 'written' */
bi = sh->dev[i].written;
sh->dev[i].written = NULL;
@@ -3766,7 +3738,6 @@ handle_failed_stripe(struct r5conf *conf, struct stripe_head *sh,
sh->dev[i].page = sh->dev[i].orig_page;
}
- if (bi) bitmap_end = 1;
while (bi && bi->bi_iter.bi_sector <
sh->dev[i].sector + RAID5_STRIPE_SECTORS(conf)) {
struct bio *bi2 = r5_next_bio(conf, bi, sh->dev[i].sector);
@@ -3800,9 +3771,6 @@ handle_failed_stripe(struct r5conf *conf, struct stripe_head *sh,
bi = nextbi;
}
}
- if (bitmap_end)
- conf->mddev->bitmap_ops->endwrite(conf->mddev,
- sh->sector, RAID5_STRIPE_SECTORS(conf));
/* If we were in the middle of a write the parity block might
* still be locked - so just clear all R5_LOCKED flags
*/
@@ -4151,8 +4119,7 @@ static void handle_stripe_clean_event(struct r5conf *conf,
bio_endio(wbi);
wbi = wbi2;
}
- conf->mddev->bitmap_ops->endwrite(conf->mddev,
- sh->sector, RAID5_STRIPE_SECTORS(conf));
+
if (head_sh->batch_head) {
sh = list_first_entry(&sh->batch_list,
struct stripe_head,
@@ -4977,8 +4944,7 @@ static void break_stripe_batch_list(struct stripe_head *head_sh,
(1 << STRIPE_COMPUTE_RUN) |
(1 << STRIPE_DISCARD) |
(1 << STRIPE_BATCH_READY) |
- (1 << STRIPE_BATCH_ERR) |
- (1 << STRIPE_BITMAP_PENDING)),
+ (1 << STRIPE_BATCH_ERR)),
"stripe state: %lx\n", sh->state);
WARN_ONCE(head_sh->state & ((1 << STRIPE_DISCARD) |
(1 << STRIPE_REPLACED)),
@@ -5869,10 +5835,6 @@ static void make_discard_request(struct mddev *mddev, struct bio *bi)
}
spin_unlock_irq(&sh->stripe_lock);
if (conf->mddev->bitmap) {
- for (d = 0; d < conf->raid_disks - conf->max_degraded;
- d++)
- mddev->bitmap_ops->startwrite(mddev, sh->sector,
- RAID5_STRIPE_SECTORS(conf));
sh->bm_seq = conf->seq_flush + 1;
set_bit(STRIPE_BIT_DELAY, &sh->state);
}
diff --git a/drivers/md/raid5.h b/drivers/md/raid5.h
index 69000fb90bd5..eafc6e9ed6ee 100644
--- a/drivers/md/raid5.h
+++ b/drivers/md/raid5.h
@@ -371,9 +371,6 @@ enum {
STRIPE_ON_RELEASE_LIST,
STRIPE_BATCH_READY,
STRIPE_BATCH_ERR,
- STRIPE_BITMAP_PENDING, /* Being added to bitmap, don't add
- * to batch yet.
- */
STRIPE_LOG_TRAPPED, /* trapped into log (see raid5-cache.c)
* this bit is used in two scenarios:
*
--
2.39.2
Powered by blists - more mailing lists