Received-SPF: pass (google.com: domain of linux-kernel+bounces-196550-linux.lists.archive=gmail.com@vger.kernel.org designates 139.178.88.99 as permitted sender) client-ip=139.178.88.99;
From: hexue <xue01.he@samsung.com>
To: axboe@kernel.dk
Cc: linux-block@vger.kernel.org, linux-kernel@vger.kernel.org, hexue
	<xue01.he@samsung.com>
Subject: [PATCH] block: Avoid polling configuration errors
Date: Fri, 31 May 2024 17:10:15 +0800
Message-Id: <20240531091015.2636025-1-xue01.he@samsung.com>
Precedence: bulk
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
Content-Type: text/plain; charset="utf-8"
CMS-TYPE: 105P
DLP-Filter: Pass
References: <CGME20240531091021epcas5p48fdbd6302bec7a91ff66272c600b0dab@epcas5p4.samsung.com>

Here's a misconfigured if application is doing polled IO
for devices that don't have a poll queue, the process will
continue to do syscall between user space and kernel space,
as in normal poll IO, CPU utilization will be 100%. IO actually
arrives through interruption.

This patch returns a signal that does not support the operation
when the underlying device does not have a poll queue, avoiding
performance and CPU simultaneous loss.

Signed-off-by: hexue <xue01.he@samsung.com>
---
 block/blk-core.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/block/blk-core.c b/block/blk-core.c
index 01186333c88e..0afcd74ae939 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -913,7 +913,7 @@ int bio_poll(struct bio *bio, struct io_comp_batch *iob, unsigned int flags)
 	q = bdev_get_queue(bdev);
 	if (cookie == BLK_QC_T_NONE ||
 	    !test_bit(QUEUE_FLAG_POLL, &q->queue_flags))
-		return 0;
+		return -EOPNOTSUPP;
 
 	blk_flush_plug(current->plug, false);
 
-- 
2.40.1