Neil,
There are two ways I see to fix the problem of the "sync" table argument
not causing the array to re-sync. We could set mddev->events to
UINT_MAX and force an mddev/bitmap event mismatch - causing the bitmap
code to set BITMAP_STALE itself. Otherwise, we can do as I have below
and set BITMAP_STALE ourselves. The real difference is whether this is
done before or after md_run(). The first solution can be done before
md_run(); the second cannot because the bitmap is allocated during
md_run(). Either solution works and correctly sets the bitmap to stale
before bitmap_load() is called. So, rather than having to think about
issues that come with setting events to UINT_MAX, I chose to set the
bitmap stale after md_run().
brassow
DM RAID: Fix for "sync" directive ineffectiveness
There are two table arguments that can be given to a DM RAID target that
control whether the array is forced to (re)synchronize or skip initialization:
"sync" and "nosync". When "sync" is given, we set mddev->recovery_cp to 0
in order to cause the device to resynchronize. This is insufficient if there
is a bitmap in use, because the array will simply look at the bitmap and see
that there is no recovery necessary.
Once md_run() is called and creates the bitmap, we set the bitmap as stale.
When the bitmap is loaded during the resume phase, a full resync will
result.
Signed-off-by: Jonathan Brassow <redacted>
Index: linux-upstream/drivers/md/dm-raid.c
===================================================================
--- linux-upstream.orig/drivers/md/dm-raid.c
+++ linux-upstream/drivers/md/dm-raid.c
@@ -1123,6 +1123,10 @@ static int raid_ctr(struct dm_target *ti
ret = -EINVAL;
goto size_mismatch;
}
+
+ if (rs->print_flags & DMPF_SYNC)
+ set_bit(BITMAP_STALE, &(rs->md.bitmap->flags));
+
rs->callbacks.congested_fn = raid_is_congested;
dm_table_add_target_callbacks(ti->table, &rs->callbacks);