xiaomi-sm8450/android_kernel_xiaomi_sm8450
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

md: fix deadlock between mddev_suspend() and md_write_start()

Browse Source 
If mddev_suspend() races with md_write_start() we can deadlock
with mddev_suspend() waiting for the request that is currently
in md_write_start() to complete the ->make_request() call,
and md_write_start() waiting for the metadata to be updated
to mark the array as 'dirty'.
As metadata updates done by md_check_recovery() only happen then
the mddev_lock() can be claimed, and as mddev_suspend() is often
called with the lock held, these threads wait indefinitely for each
other.

We fix this by having md_write_start() abort if mddev_suspend()
is happening, and ->make_request() aborts if md_write_start()
aborted.
md_make_request() can detect this abort, decrease the ->active_io
count, and wait for mddev_suspend().

Reported-by: Nix <nix@esperi.org.uk>
Fix: 68866e425be2(MD: no sync IO while suspended)
Cc: stable@vger.kernel.org
Signed-off-by: NeilBrown <neilb@suse.com>
Signed-off-by: Shaohua Li <shli@fb.com>
This commit is contained in:
NeilBrown
2017-06-05 16:49:39 +10:00
committed by Shaohua Li
parent 63f700aab4
commit cc27b0c78c
9 changed files with 56 additions and 35 deletions
Download Patch File Download Diff File Expand all files Collapse all files

10
drivers/md/raid10.c
View File

@@ -1303,8 +1303,6 @@ static void raid10_write_request(struct mddev *mddev, struct bio *bio,
sector_t sectors;
int max_sectors;
md_write_start(mddev, bio);
/*
* Register the new request and wait if the reconstruction
* thread has put up a bar for new requests.
@@ -1525,7 +1523,7 @@ static void __make_request(struct mddev *mddev, struct bio *bio, int sectors)
raid10_write_request(mddev, bio, r10_bio);
}
static void raid10_make_request(struct mddev *mddev, struct bio *bio)
static bool raid10_make_request(struct mddev *mddev, struct bio *bio)
{
struct r10conf *conf = mddev->private;
sector_t chunk_mask = (conf->geo.chunk_mask & conf->prev.chunk_mask);
@@ -1534,9 +1532,12 @@ static void raid10_make_request(struct mddev *mddev, struct bio *bio)
if (unlikely(bio->bi_opf & REQ_PREFLUSH)) {
md_flush_request(mddev, bio);
return;
return true;
}
if (!md_write_start(mddev, bio))
return false;
/*
* If this request crosses a chunk boundary, we need to split
* it.
@@ -1553,6 +1554,7 @@ static void raid10_make_request(struct mddev *mddev, struct bio *bio)
/* In case raid10d snuck in to freeze_array */
wake_up(&conf->wait_barrier);
return true;
}
static void raid10_status(struct seq_file *seq, struct mddev *mddev)