| Message ID | 20191114230233.3582-1-junxiao.bi@oracle.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | block-mq: ratelimit the warning log | expand |
diff --git a/block/blk-mq.c b/block/blk-mq.c index ec791156e9cc..846f5d26c523 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -1384,7 +1384,7 @@ static void __blk_mq_run_hw_queue(struct blk_mq_hw_ctx *hctx) * handle dispatched requests to this hctx */ if (!cpumask_test_cpu(raw_smp_processor_id(), hctx->cpumask) && - cpu_online(hctx->next_cpu)) { + cpu_online(hctx->next_cpu) && printk_ratelimit()) { printk(KERN_WARNING "run queue from wrong CPU %d, hctx %s\n", raw_smp_processor_id(), cpumask_empty(hctx->cpumask) ? "inactive": "active");
When doing cpu online/offlile, sometimes this warning will be triggered, it's harmless, io will be dispatched. But sometimes it warns too much and even stall the whole system, so ratelimit it. Signed-off-by: Junxiao Bi <junxiao.bi@oracle.com> --- block/blk-mq.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)