Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1407404

[PATCH] blk-mq: clear q->mq_ops if init fail

From Ming Lin <mlin@kernel.org>
Newsgroups linux.kernel
Subject [PATCH] blk-mq: clear q->mq_ops if init fail
Date 2016-05-26 08:30 +0200
Message-ID <rCWXT-2Qp-19@gated-at.bofh.it> (permalink)
Organization linux.* mail to news gateway

Show all headers | View raw


From: Ming Lin <ming.l@samsung.com>

blk_mq_init_queue() calls blk_mq_init_allocated_queue(), but q->mq_ops
was not cleared when blk_mq_init_allocated_queue() fails.
Then blk_cleanup_queue() calls blk_mq_free_queue() which will crash because:
- q->all_q_node is not added to all_q_list yet
- q->tag_set is NULL
- hctx was not setup yet or already freed

Fixed it by clearing q->mq_ops on error path.

Signed-off-by: Ming Lin <ming.l@samsung.com>
---
 block/blk-mq.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/block/blk-mq.c b/block/blk-mq.c
index 67bf8ed..86f08b1 100644
--- a/block/blk-mq.c
+++ b/block/blk-mq.c
@@ -2054,7 +2054,7 @@ struct request_queue *blk_mq_init_allocated_queue(struct blk_mq_tag_set *set,
 
 	q->queue_ctx = alloc_percpu(struct blk_mq_ctx);
 	if (!q->queue_ctx)
-		return ERR_PTR(-ENOMEM);
+		goto err_exit;
 
 	q->queue_hw_ctx = kzalloc_node(nr_cpu_ids * sizeof(*(q->queue_hw_ctx)),
 						GFP_KERNEL, set->numa_node);
@@ -2118,6 +2118,8 @@ err_map:
 	kfree(q->queue_hw_ctx);
 err_percpu:
 	free_percpu(q->queue_ctx);
+err_exit:
+	q->mq_ops = NULL;
 	return ERR_PTR(-ENOMEM);
 }
 EXPORT_SYMBOL(blk_mq_init_allocated_queue);
-- 
1.9.1

Back to linux.kernel | Previous | NextNext in thread | Find similar | Unroll thread


Thread

[PATCH] blk-mq: clear q->mq_ops if init fail Ming Lin <mlin@kernel.org> - 2016-05-26 08:30 +0200
  Re: [PATCH] blk-mq: clear q->mq_ops if init fail Christoph Hellwig <hch@lst.de> - 2016-05-26 09:10 +0200
  Re: [PATCH] blk-mq: clear q->mq_ops if init fail Jens Axboe <axboe@kernel.dk> - 2016-05-26 17:00 +0200

csiph-web