diff options
author | Richard Weinberger <richard@nod.at> | 2016-08-24 12:36:14 (GMT) |
---|---|---|
committer | Richard Weinberger <richard@nod.at> | 2016-10-02 20:54:01 (GMT) |
commit | 2e8f08deabbc7eefe4c5838aaa6aa9a23a8acf2e (patch) | |
tree | 978c8095e114898a898df44014996611a39ddfef /drivers/mtd/ubi/fastmap-wl.c | |
parent | 2365418879e9abf12ea9def7f9f3caf0dfa7ffb0 (diff) | |
download | linux-2e8f08deabbc7eefe4c5838aaa6aa9a23a8acf2e.tar.xz |
ubi: Fix races around ubi_refill_pools()
When writing a new Fastmap the first thing that happens
is refilling the pools in memory.
At this stage it is possible that new PEBs from the new pools
get already claimed and written with data.
If this happens before the new Fastmap data structure hits the
flash and we face power cut the freshly written PEB will not
scanned and unnoticed.
Solve the issue by locking the pools until Fastmap is written.
Cc: <stable@vger.kernel.org>
Fixes: dbb7d2a88d ("UBI: Add fastmap core")
Signed-off-by: Richard Weinberger <richard@nod.at>
Diffstat (limited to 'drivers/mtd/ubi/fastmap-wl.c')
-rw-r--r-- | drivers/mtd/ubi/fastmap-wl.c | 6 |
1 files changed, 4 insertions, 2 deletions
diff --git a/drivers/mtd/ubi/fastmap-wl.c b/drivers/mtd/ubi/fastmap-wl.c index 30d3999..4f0bd6b 100644 --- a/drivers/mtd/ubi/fastmap-wl.c +++ b/drivers/mtd/ubi/fastmap-wl.c @@ -262,6 +262,8 @@ static struct ubi_wl_entry *get_peb_for_wl(struct ubi_device *ubi) struct ubi_fm_pool *pool = &ubi->fm_wl_pool; int pnum; + ubi_assert(rwsem_is_locked(&ubi->fm_eba_sem)); + if (pool->used == pool->size) { /* We cannot update the fastmap here because this * function is called in atomic context. @@ -303,7 +305,7 @@ int ubi_ensure_anchor_pebs(struct ubi_device *ubi) wrk->anchor = 1; wrk->func = &wear_leveling_worker; - schedule_ubi_work(ubi, wrk); + __schedule_ubi_work(ubi, wrk); return 0; } @@ -344,7 +346,7 @@ int ubi_wl_put_fm_peb(struct ubi_device *ubi, struct ubi_wl_entry *fm_e, spin_unlock(&ubi->wl_lock); vol_id = lnum ? UBI_FM_DATA_VOLUME_ID : UBI_FM_SB_VOLUME_ID; - return schedule_erase(ubi, e, vol_id, lnum, torture); + return schedule_erase(ubi, e, vol_id, lnum, torture, true); } /** |