Received-SPF: pass (google.com: domain of linux-kernel+bounces-22607-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45e3:2400::1 as permitted sender) client-ip=2604:1380:45e3:2400::1;
Date: Wed, 10 Jan 2024 11:40:28 -0700
From: Keith Busch <kbusch@kernel.org>
To: William Butler <wab@google.com>
Cc: Pavel Emelyanov <xemul@openvz.org>, Christoph Hellwig <hch@lst.de>,
	Sagi Grimberg <sagi@grimberg.me>, Jens Axboe <axboe@kernel.dk>,
	linux-kernel@vger.kernel.org, linux-nvme@lists.infradead.org
Subject: Re: [PATCH] nvme: Inform controller of doorbell config before
 unquiescing IO queues
Message-ID: <ZZ7knJGSVx8hWvaW@kbusch-mbp>
References: <20240110182855.1833512-1-wab@google.com>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20240110182855.1833512-1-wab@google.com>

On Wed, Jan 10, 2024 at 06:28:55PM +0000, William Butler wrote:
> During resets, if queues are unquiesced first, then the host can submit
> IOs to the controller using shadow doorbell logic but the controller
> won't be aware. This can lead to necessary MMIO doorbells from being
> not issued, causing requests to be delayed and timed-out.

Good catch. I guess we wouldn't normally expect any requests to be in
queue at this point, but plenty of reasons it could happen.

Applied to nvme-6.8 with a slightly updated title to match the local
conventions.