[03/18] block: fix elevator init check

Commit Message

Christoph Hellwig Jan. 25, 2017, 5:25 p.m. UTC

We can't initalize the elevator fields for flushes as flush share space
in struct request with the elevator data.  But currently we can't
commnicate that a request is a flush through blk_get_request as we
can only pass READ or WRITE, and the low-level code looks at the
possible NULL bio to check for a flush.

Fix this by allowing to pass any block op and flags, and by checking for
the flush flags in __get_request.

Signed-off-by: Christoph Hellwig <[e-mail]>
Reviewed-by: Hannes Reinecke <[e-mail]>
---
 block/blk-core.c | 26 ++++----------------------
 1 file changed, 4 insertions(+), 22 deletions(-)

Comments

Martin K. Petersen Jan. 26, 2017, 3:01 a.m. UTC | #1

>>>>> "Christoph" == Christoph Hellwig <[e-mail]> writes:

Christoph> We can't initalize the elevator fields for flushes as flush
Christoph> share space in struct request with the elevator data.  But
Christoph> currently we can't commnicate that a request is a flush

communicate

Christoph> through blk_get_request as we can only pass READ or WRITE,
Christoph> and the low-level code looks at the possible NULL bio to
Christoph> check for a flush.

Christoph> Fix this by allowing to pass any block op and flags, and by
Christoph> checking for the flush flags in __get_request.

Reviewed-by: Martin K. Petersen <[e-mail]>

Bart Van Assche Jan. 26, 2017, 11:21 p.m. UTC | #2

On Wed, 2017-01-25 at 18:25 +0100, Christoph Hellwig wrote:
> We can't initalize the elevator fields for flushes as flush share space
> in struct request with the elevator data.  But currently we can't
> commnicate that a request is a flush through blk_get_request as we
> can only pass READ or WRITE, and the low-level code looks at the
> possible NULL bio to check for a flush.
> 
> Fix this by allowing to pass any block op and flags, and by checking for
> the flush flags in __get_request.

Reviewed-by: Bart Van Assche <[e-mail]>--
To unsubscribe from this list: send the line "unsubscribe linux-block" in
the body of a message to [e-mail]
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Patch

diff --git a/block/blk-core.c b/block/blk-core.c
index b830e14..a84c1b9 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -1022,25 +1022,6 @@  int blk_update_nr_requests(struct request_queue *q, unsigned int nr)
 	return 0;
 }
 
-/*
- * Determine if elevator data should be initialized when allocating the
- * request associated with @bio.
- */
-static bool blk_rq_should_init_elevator(struct bio *bio)
-{
-	if (!bio)
-		return true;
-
-	/*
-	 * Flush requests do not use the elevator so skip initialization.
-	 * This allows a request to share the flush and elevator data.
-	 */
-	if (op_is_flush(bio->bi_opf))
-		return false;
-
-	return true;
-}
-
 /**
  * __get_request - get a free request
  * @rl: request list to allocate from
@@ -1119,10 +1100,13 @@  static struct request *__get_request(struct request_list *rl, unsigned int op,
 	 * request is freed.  This guarantees icq's won't be destroyed and
 	 * makes creating new ones safe.
 	 *
+	 * Flush requests do not use the elevator so skip initialization.
+	 * This allows a request to share the flush and elevator data.
+	 *
 	 * Also, lookup icq while holding queue_lock.  If it doesn't exist,
 	 * it will be created after releasing queue_lock.
 	 */
-	if (blk_rq_should_init_elevator(bio) && !blk_queue_bypass(q)) {
+	if (!op_is_flush(op) && !blk_queue_bypass(q)) {
 		rq_flags |= RQF_ELVPRIV;
 		q->nr_rqs_elvpriv++;
 		if (et->icq_cache && ioc)
@@ -1276,8 +1260,6 @@  static struct request *blk_old_get_request(struct request_queue *q, int rw,
 {
 	struct request *rq;
 
-	BUG_ON(rw != READ && rw != WRITE);
-
 	/* create ioc upfront */
 	create_io_context(gfp_mask, q->node);