There are currently any issuer of REQ_OP_ZONE_RESET and
REQ_OP_ZONE_FINISH operations that set REQ_NOWAIT. However, as we cannot
handle this flag correctly due to the potential request allocation
failure that may happen in blk_mq_submit_bio() after blk_zone_plug_bio()
has handled the zone write plug write pointer updates for the targeted
zones, modify blk_zone_wplug_handle_reset_or_finish() to warn if this
flag is set and ignore it.

Fixes: dd291d77cc90 ("block: Introduce zone write plugging")
Cc: sta...@vger.kernel.org
Signed-off-by: Damien Le Moal <dlem...@kernel.org>
Reviewed-by: Christoph Hellwig <h...@lst.de>
---
 block/blk-zoned.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/block/blk-zoned.c b/block/blk-zoned.c
index 7982b9494d63..ee9c67121c6c 100644
--- a/block/blk-zoned.c
+++ b/block/blk-zoned.c
@@ -707,6 +707,15 @@ static bool blk_zone_wplug_handle_reset_or_finish(struct 
bio *bio,
                return true;
        }
 
+       /*
+        * No-wait reset or finish BIOs do not make much sense as the callers
+        * issue these as blocking operations in most cases. To avoid issues
+        * the BIO execution potentially failing with BLK_STS_AGAIN, warn about
+        * REQ_NOWAIT being set and ignore that flag.
+        */
+       if (WARN_ON_ONCE(bio->bi_opf & REQ_NOWAIT))
+               bio->bi_opf &= ~REQ_NOWAIT;
+
        /*
         * If we have a zone write plug, set its write pointer offset to 0
         * (reset case) or to the zone size (finish case). This will abort all
-- 
2.47.1


Reply via email to