blk-mq: insert passthrough request into hctx->dispatch directly

mainline inclusion
from mainline-5.6-rc4
commit 01e99aec
category: bugfix
bugzilla: 42777
CVE: NA

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

For some reason, device may be in one situation which can't handle
FS request, so STS_RESOURCE is always returned and the FS request
will be added to hctx->dispatch. However passthrough request may
be required at that time for fixing the problem. If passthrough
request is added to scheduler queue, there isn't any chance for
blk-mq to dispatch it given we prioritize requests in hctx->dispatch.
Then the FS IO request may never be completed, and IO hang is caused.

So passthrough request has to be added to hctx->dispatch directly
for fixing the IO hang.

Fix this issue by inserting passthrough request into hctx->dispatch
directly together withing adding FS request to the tail of
hctx->dispatch in blk_mq_dispatch_rq_list(). Actually we add FS request
to tail of hctx->dispatch at default, see blk_mq_request_bypass_insert().

Then it becomes consistent with original legacy IO request
path, in which passthrough request is always added to q->queue_head.

Cc: Dongli Zhang <dongli.zhang@oracle.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Ewan D. Milne <emilne@redhat.com>
Signed-off-by: NMing Lei <ming.lei@redhat.com>
Signed-off-by: NJens Axboe <axboe@kernel.dk>

Conflicts:
  block/blk-flush.c
  block/blk-mq.c
  block/blk-mq-sched.c
Signed-off-by: NYu Kuai <yukuai3@huawei.com>
Reviewed-by: NYufen Yu <yuyufen@huawei.com>
Signed-off-by: NYang Yingliang <yangyingliang@huawei.com>

block/blk-flush.c

@@ -495,7 +495,7 @@ void blk_insert_flush(struct request *rq)
 	if ((policy & REQ_FSEQ_DATA) &&
 	    !(policy & (REQ_FSEQ_PREFLUSH | REQ_FSEQ_POSTFLUSH))) {
 		if (q->mq_ops)
-			blk_mq_request_bypass_insert(rq, false);
+			blk_mq_request_bypass_insert(rq, false, false);
 		else
 			list_add_tail(&rq->queuelist, &q->queue_head);
 		return;

block/blk-mq-sched.c

@@ -357,13 +357,19 @@ static bool blk_mq_sched_bypass_insert(struct blk_mq_hw_ctx *hctx,
 				       bool has_sched,
 				       struct request *rq)
 {
-	/* dispatch flush rq directly */
-	if (rq->rq_flags & RQF_FLUSH_SEQ) {
-		spin_lock(&hctx->lock);
-		list_add(&rq->queuelist, &hctx->dispatch);
-		spin_unlock(&hctx->lock);
+	/*
+	 * dispatch flush and passthrough rq directly
+	 *
+	 * passthrough request has to be added to hctx->dispatch directly.
+	 * For some reason, device may be in one situation which can't
+	 * handle FS request, so STS_RESOURCE is always returned and the
+	 * FS request will be added to hctx->dispatch. However passthrough
+	 * request may be required at that time for fixing the problem. If
+	 * passthrough request is added to scheduler queue, there isn't any
+	 * chance to dispatch it given we prioritize requests in hctx->dispatch.
+	 */
+	if ((rq->rq_flags & RQF_FLUSH_SEQ) || blk_rq_is_passthrough(rq))
 		return true;
-	}
 
 	if (has_sched)
 		rq->rq_flags |= RQF_SORTED;
@@ -387,8 +393,10 @@ void blk_mq_sched_insert_request(struct request *rq, bool at_head,
 
 	WARN_ON(e && (rq->tag != -1));
 
-	if (blk_mq_sched_bypass_insert(hctx, !!e, rq))
+	if (blk_mq_sched_bypass_insert(hctx, !!e, rq)) {
+		blk_mq_request_bypass_insert(rq, at_head, false);
 		goto run;
+	}
 
 	if (e && e->type->ops.mq.insert_requests) {
 		LIST_HEAD(list);

block/blk-mq.c

@@ -721,7 +721,7 @@ static void blk_mq_requeue_work(struct work_struct *work)
 		 * merge.
 		 */
 		if (rq->rq_flags & RQF_DONTPREP)
-			blk_mq_request_bypass_insert(rq, false);
+			blk_mq_request_bypass_insert(rq, false, false);
 		else
 			blk_mq_sched_insert_request(rq, true, false, false);
 	}
@@ -1227,7 +1227,7 @@ bool blk_mq_dispatch_rq_list(struct request_queue *q, struct list_head *list,
 		bool needs_restart;
 
 		spin_lock(&hctx->lock);
-		list_splice_init(list, &hctx->dispatch);
+		list_splice_tail_init(list, &hctx->dispatch);
 		spin_unlock(&hctx->lock);
 
 		/*
@@ -1590,13 +1590,17 @@ void __blk_mq_insert_request(struct blk_mq_hw_ctx *hctx, struct request *rq,
  * Should only be used carefully, when the caller knows we want to
  * bypass a potential IO scheduler on the target device.
  */
-void blk_mq_request_bypass_insert(struct request *rq, bool run_queue)
+void blk_mq_request_bypass_insert(struct request *rq, bool at_head,
+				bool run_queue)
 {
 	struct blk_mq_ctx *ctx = rq->mq_ctx;
 	struct blk_mq_hw_ctx *hctx = blk_mq_map_queue(rq->q, ctx->cpu);
 
 	spin_lock(&hctx->lock);
-	list_add_tail(&rq->queuelist, &hctx->dispatch);
+	if (at_head)
+		list_add(&rq->queuelist, &hctx->dispatch);
+	else
+		list_add_tail(&rq->queuelist, &hctx->dispatch);
 	spin_unlock(&hctx->lock);
 
 	if (run_queue)
@@ -1776,7 +1780,7 @@ static blk_status_t __blk_mq_try_issue_directly(struct blk_mq_hw_ctx *hctx,
 	if (bypass_insert)
 		return BLK_STS_RESOURCE;
 
-	blk_mq_request_bypass_insert(rq, run_queue);
+	blk_mq_request_bypass_insert(rq, false, run_queue);
 	return BLK_STS_OK;
 }
 
@@ -1792,7 +1796,7 @@ static void blk_mq_try_issue_directly(struct blk_mq_hw_ctx *hctx,
 
 	ret = __blk_mq_try_issue_directly(hctx, rq, cookie, false);
 	if (ret == BLK_STS_RESOURCE || ret == BLK_STS_DEV_RESOURCE)
-		blk_mq_request_bypass_insert(rq, true);
+		blk_mq_request_bypass_insert(rq, false, true);
 	else if (ret != BLK_STS_OK)
 		blk_mq_end_request(rq, ret);
 
@@ -1827,7 +1831,7 @@ void blk_mq_try_issue_list_directly(struct blk_mq_hw_ctx *hctx,
 		if (ret != BLK_STS_OK) {
 			if (ret == BLK_STS_RESOURCE ||
 					ret == BLK_STS_DEV_RESOURCE) {
-				blk_mq_request_bypass_insert(rq,
+				blk_mq_request_bypass_insert(rq, false,
 							list_empty(list));
 				break;
 			}

block/blk-mq.h

@@ -64,7 +64,8 @@ int blk_mq_alloc_rqs(struct blk_mq_tag_set *set, struct blk_mq_tags *tags,
  */
 void __blk_mq_insert_request(struct blk_mq_hw_ctx *hctx, struct request *rq,
 				bool at_head);
-void blk_mq_request_bypass_insert(struct request *rq, bool run_queue);
+void blk_mq_request_bypass_insert(struct request *rq, bool at_head,
+				  bool run_queue);
 void blk_mq_insert_requests(struct blk_mq_hw_ctx *hctx, struct blk_mq_ctx *ctx,
 				struct list_head *list);