renegade-project/linux-next
2
0
Fork 0
mirror of https://github.com/edk2-porting/linux-next.git synced 2024-12-18 18:23:53 +08:00
Code

md-cluster/bitmap: don't call md_bitmap_sync_with_cluster during reshaping stage

Browse Source 
When reshape is happening in one node, other nodes could receive
lots of RESYNCING messages, so md_bitmap_sync_with_cluster is called.

Since the resyncing window is typically small in these RESYNCING
messages, so WARN is always triggered, so we should not call the
func when reshape is happening.

Reviewed-by: NeilBrown <neilb@suse.com>
Signed-off-by: Guoqing Jiang <gqjiang@suse.com>
Signed-off-by: Shaohua Li <shli@fb.com>
This commit is contained in:
Guoqing Jiang 2018-10-18 16:37:46 +08:00 committed by Shaohua Li
parent ca1e98e04a
commit cbce6863b6
1 changed files with 20 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

27
drivers/md/md-cluster.c
View File

@ -457,12 +457,13 @@ static void remove_suspend_info(struct mddev *mddev, int slot)
mddev->pers->quiesce(mddev, 0);
}
static void process_suspend_info(struct mddev *mddev,
int slot, sector_t lo, sector_t hi)
{
struct md_cluster_info *cinfo = mddev->cluster_info;
struct suspend_info *s;
struct mdp_superblock_1 *sb = NULL;
struct md_rdev *rdev;
if (!hi) {
/*
@ -476,6 +477,12 @@ static void process_suspend_info(struct mddev *mddev,
return;
}
rdev_for_each(rdev, mddev)
if (rdev->raid_disk > -1 && !test_bit(Faulty, &rdev->flags)) {
sb = page_address(rdev->sb_page);
break;
}
/*
* The bitmaps are not same for different nodes
* if RESYNCING is happening in one node, then
@ -488,12 +495,18 @@ static void process_suspend_info(struct mddev *mddev,
* sync_low/hi is used to record the region which
* arrived in the previous RESYNCING message,
*
* Call bitmap_sync_with_cluster to clear
* NEEDED_MASK and set RESYNC_MASK since
* resync thread is running in another node,
* so we don't need to do the resync again
* with the same section */
md_bitmap_sync_with_cluster(mddev, cinfo->sync_low, cinfo->sync_hi, lo, hi);
* Call md_bitmap_sync_with_cluster to clear NEEDED_MASK
* and set RESYNC_MASK since resync thread is running
* in another node, so we don't need to do the resync
* again with the same section.
*
* Skip md_bitmap_sync_with_cluster in case reshape
* happening, because reshaping region is small and
* we don't want to trigger lots of WARN.
*/
if (sb && !(le32_to_cpu(sb->feature_map) & MD_FEATURE_RESHAPE_ACTIVE))
md_bitmap_sync_with_cluster(mddev, cinfo->sync_low,
cinfo->sync_hi, lo, hi);
cinfo->sync_low = lo;
cinfo->sync_hi = hi;