| From | lizf@kernel ... | Subject | [PATCH 3.4 087/146] md/raid5: don't record new size if resize_stripes fails. | Date | Tue, 15 Sep 2015 17:03:22 +0800 |
| |
From: NeilBrown <neilb@suse.de>
3.4.109-rc1 review patch. If anyone has any objections, please let me know.
------------------
commit 6e9eac2dcee5e19f125967dd2be3e36558c42fff upstream.
If any memory allocation in resize_stripes fails we will return -ENOMEM, but in some cases we update conf->pool_size anyway.
This means that if we try again, the allocations will be assumed to be larger than they are, and badness results.
So only update pool_size if there is no error.
This bug was introduced in 2.6.17 and the patch is suitable for -stable.
Fixes: ad01c9e3752f ("[PATCH] md: Allow stripes to be expanded in preparation for expanding an array") Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Zefan Li <lizefan@huawei.com> --- drivers/md/raid5.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index c276ad0..7a218e8 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -1622,7 +1622,8 @@ static int resize_stripes(struct r5conf *conf, int newsize) conf->slab_cache = sc; conf->active_name = 1-conf->active_name; - conf->pool_size = newsize; + if (!err) + conf->pool_size = newsize; return err; } -- 1.9.1
|