This is a note to let you know that I've just added the patch titled dm writecache: flush origin device when writing and cache is full to the 5.10-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary The filename of the patch is: dm-writecache-flush-origin-device-when-writing-and-cache-is-full.patch and it can be found in the queue-5.10 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable@xxxxxxxxxxxxxxx> know about it. >From ee55b92a7391bf871939330f662651b54be51b73 Mon Sep 17 00:00:00 2001 From: Mikulas Patocka <mpatocka@xxxxxxxxxx> Date: Tue, 15 Jun 2021 13:45:55 -0400 Subject: dm writecache: flush origin device when writing and cache is full From: Mikulas Patocka <mpatocka@xxxxxxxxxx> commit ee55b92a7391bf871939330f662651b54be51b73 upstream. Commit d53f1fafec9d086f1c5166436abefdaef30e0363 ("dm writecache: do direct write if the cache is full") changed dm-writecache, so that it writes directly to the origin device if the cache is full. Unfortunately, it doesn't forward flush requests to the origin device, so that there is a bug where flushes are being ignored. Fix this by adding missing flush forwarding. For PMEM mode, we fix this bug by disabling direct writes to the origin device, because it performs better. Signed-off-by: Mikulas Patocka <mpatocka@xxxxxxxxxx> Fixes: d53f1fafec9d ("dm writecache: do direct write if the cache is full") Cc: stable@xxxxxxxxxxxxxxx # v5.7+ Signed-off-by: Mike Snitzer <snitzer@xxxxxxxxxx> Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx> --- drivers/md/dm-writecache.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) --- a/drivers/md/dm-writecache.c +++ b/drivers/md/dm-writecache.c @@ -1297,8 +1297,12 @@ static int writecache_map(struct dm_targ writecache_flush(wc); if (writecache_has_error(wc)) goto unlock_error; + if (unlikely(wc->cleaner)) + goto unlock_remap_origin; goto unlock_submit; } else { + if (dm_bio_get_target_bio_nr(bio)) + goto unlock_remap_origin; writecache_offload_bio(wc, bio); goto unlock_return; } @@ -1377,7 +1381,7 @@ read_next_block: } e = writecache_pop_from_freelist(wc, (sector_t)-1); if (unlikely(!e)) { - if (!found_entry) { + if (!WC_MODE_PMEM(wc) && !found_entry) { direct_write: e = writecache_find_entry(wc, bio->bi_iter.bi_sector, WFE_RETURN_FOLLOWING); if (e) { @@ -2483,7 +2487,7 @@ overflow: goto bad; } - ti->num_flush_bios = 1; + ti->num_flush_bios = WC_MODE_PMEM(wc) ? 1 : 2; ti->flush_supported = true; ti->num_discard_bios = 1; Patches currently in stable-queue which might be from mpatocka@xxxxxxxxxx are queue-5.10/dm-writecache-don-t-split-bios-when-overwriting-cont.patch queue-5.10/dm-writecache-commit-just-one-block-not-a-full-page.patch queue-5.10/dm-writecache-flush-origin-device-when-writing-and-cache-is-full.patch