On 11/07/2024 19:08, Bart Van Assche wrote:
On 7/11/24 1:23 AM, John Garry wrote:
diff --git a/include/linux/blk-mq.h b/include/linux/blk-mq.h
index 89ba6b16fe8b..225e51698470 100644
--- a/include/linux/blk-mq.h
+++ b/include/linux/blk-mq.h
@@ -664,12 +664,14 @@ enum {
BLK_MQ_F_ALLOC_POLICY_START_BIT = 8,
BLK_MQ_F_ALLOC_POLICY_BITS = 1,
+ /* Keep hctx_state_name[] in sync with the definitions below */
BLK_MQ_S_STOPPED = 0,
BLK_MQ_S_TAG_ACTIVE = 1,
BLK_MQ_S_SCHED_RESTART = 2,
/* hw queue is inactive after all its CPUs become offline */
BLK_MQ_S_INACTIVE = 3,
+ BLK_MQ_S_MAX,
Please create a new "enum {" section for the BLK_MQ_S_ constants.
That will make it more clear that these are unrelated to the other
constants defined above and below the BLK_MQ_S_ constants.
ok, fine.
If we are going to start breaking up this misc enum, then where do
BLK_MQ_MAX_DEPTH and BLK_MQ_CPU_WORK_BATCH really belong? I'd be more
inclined to have a separate #define for each of them. I am not sure how
they ever ended up in this same enum.
BTW, BLK_MQ_CPU_WORK_BATCH is only used in block/blk-mq.c, so I don't
see why it is even in a public API.
