[PATCH 4.16 42/68] blk-mq: make sure that correct hctx->next_cpu is set

[Greg Kroah-Hartman]

4.16-stable review patch.  If anyone has any objections, please let me know.

------------------

From: Ming Lei <ming.lei@xxxxxxxxxx>

commit a1c735fb790745f94a359df45c11df4a69760389 upstream.

>From commit 20e4d81393196 (blk-mq: simplify queue mapping & schedule
with each possisble CPU), one hctx can be mapped from all offline CPUs,
then hctx->next_cpu can be set as wrong.

This patch fixes this issue by making hctx->next_cpu pointing to the
first CPU in hctx->cpumask if all CPUs in hctx->cpumask are offline.

Cc: Stefan Haberland <sth@xxxxxxxxxxxxxxxxxx>
Tested-by: Christian Borntraeger <borntraeger@xxxxxxxxxx>
Reviewed-by: Christoph Hellwig <hch@xxxxxx>
Reviewed-by: Sagi Grimberg <sagi@xxxxxxxxxxx>
Fixes: 20e4d81393196 ("blk-mq: simplify queue mapping & schedule with each possisble CPU")
Cc: stable@xxxxxxxxxxxxxxx
Signed-off-by: Ming Lei <ming.lei@xxxxxxxxxx>
Signed-off-by: Jens Axboe <axboe@xxxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>

---
 block/blk-mq.c |    2 ++
 1 file changed, 2 insertions(+)

--- a/block/blk-mq.c
+++ b/block/blk-mq.c
@@ -2439,6 +2439,8 @@ static void blk_mq_map_swqueue(struct re
 		 */
 		hctx->next_cpu = cpumask_first_and(hctx->cpumask,
 				cpu_online_mask);
+		if (hctx->next_cpu >= nr_cpu_ids)
+			hctx->next_cpu = cpumask_first(hctx->cpumask);
 		hctx->next_cpu_batch = BLK_MQ_CPU_WORK_BATCH;
 	}
 }