Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Tue, 26 Oct 2021 09:28:39 +0800
From:   Ming Lei <ming.lei@redhat.com>
To:     Dmitry Osipenko <digetx@gmail.com>
Cc:     Stephen Rothwell <sfr@canb.auug.org.au>,
        Linux Next Mailing List <linux-next@vger.kernel.org>,
        Ulf Hansson <ulf.hansson@linaro.org>,
        Adrian Hunter <adrian.hunter@intel.com>,
        Jens Axboe <axboe@kernel.dk>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        linux-mmc <linux-mmc@vger.kernel.org>,
        linux-block <linux-block@vger.kernel.org>
Subject: Re: linux-next: Tree for Oct 25
Message-ID: <YXdZxzAPejKyE8Oi@T590>
References: <20211025204921.73cb3011@canb.auug.org.au>
 <82bbf33e-918f-da01-95e6-9b2cc1b8b610@gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <82bbf33e-918f-da01-95e6-9b2cc1b8b610@gmail.com>
Precedence: bulk

On Tue, Oct 26, 2021 at 01:11:07AM +0300, Dmitry Osipenko wrote:
> Hello,
> 
> Recent -next has this new warning splat coming from MMC, please take a look.
> 
>  ------------[ cut here ]------------
>  WARNING: CPU: 0 PID: 525 at kernel/sched/core.c:9477 __might_sleep+0x65/0x68
>  do not call blocking ops when !TASK_RUNNING; state=2 set at [<4316eb02>] prepare_to_wait+0x2e/0xb8
>  Modules linked in:
>  CPU: 0 PID: 525 Comm: Xorg Tainted: G        W         5.15.0-rc6-next-20211025-00226-g89ccd6948ec3 #5
>  Hardware name: NVIDIA Tegra SoC (Flattened Device Tree)
>  (unwind_backtrace) from [<c01089f9>] (show_stack+0x11/0x14)
>  (show_stack) from [<c0afacb9>] (dump_stack_lvl+0x2b/0x34)
>  (dump_stack_lvl) from [<c011f689>] (__warn+0xa1/0xe8)
>  (__warn) from [<c0af6729>] (warn_slowpath_fmt+0x65/0x7c)
>  (warn_slowpath_fmt) from [<c01421b9>] (__might_sleep+0x65/0x68)
>  (__might_sleep) from [<c07eb377>] (mmc_blk_rw_wait+0x2f/0x118)
>  (mmc_blk_rw_wait) from [<c07eba11>] (mmc_blk_mq_issue_rq+0x219/0x71c)
>  (mmc_blk_mq_issue_rq) from [<c07ec199>] (mmc_mq_queue_rq+0xf9/0x200)
>  (mmc_mq_queue_rq) from [<c04ad247>] (__blk_mq_try_issue_directly+0xcb/0x100)
>  (__blk_mq_try_issue_directly) from [<c04adb89>] (blk_mq_request_issue_directly+0x2d/0x48)
>  (blk_mq_request_issue_directly) from [<c04adcf3>] (blk_mq_flush_plug_list+0x14f/0x1f4)
>  (blk_mq_flush_plug_list) from [<c04a5313>] (blk_flush_plug+0x83/0xb8)
>  (blk_flush_plug) from [<c0b013cb>] (io_schedule+0x2b/0x3c)
>  (io_schedule) from [<c0b01a17>] (bit_wait_io+0xf/0x48)

The following patch should fix the issue:

diff --git a/block/blk-mq.c b/block/blk-mq.c
index a71aeed7b987..bee9cb2a44cb 100644
--- a/block/blk-mq.c
+++ b/block/blk-mq.c
@@ -2223,7 +2223,7 @@ void blk_mq_flush_plug_list(struct blk_plug *plug, bool from_schedule)
 		return;
 	plug->rq_count = 0;
 
-	if (!plug->multiple_queues && !plug->has_elevator) {
+	if (!plug->multiple_queues && !plug->has_elevator && !from_schedule) {
 		blk_mq_plug_issue_direct(plug, from_schedule);
 		if (rq_list_empty(plug->mq_list))
 			return;

-- 
Ming