Received-SPF: pass (google.com: domain of linux-kernel+bounces-202536-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.199.223 as permitted sender) client-ip=147.75.199.223;
From: Sasha Levin <sashal@kernel.org>
To: linux-kernel@vger.kernel.org,
	stable@vger.kernel.org
Cc: Sagi Grimberg <sagi@grimberg.me>,
	Alex Turin <alex@vastdata.com>,
	Christoph Hellwig <hch@lst.de>,
	Keith Busch <kbusch@kernel.org>,
	Sasha Levin <sashal@kernel.org>,
	kch@nvidia.com,
	linux-nvme@lists.infradead.org
Subject: [PATCH AUTOSEL 5.15 09/12] nvmet: fix a possible leak when destroy a ctrl during qp establishment
Date: Wed,  5 Jun 2024 08:05:19 -0400
Message-ID: <20240605120528.2967750-9-sashal@kernel.org>
In-Reply-To: <20240605120528.2967750-1-sashal@kernel.org>
References: <20240605120528.2967750-1-sashal@kernel.org>
Precedence: bulk
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit

From: Sagi Grimberg <sagi@grimberg.me>

[ Upstream commit c758b77d4a0a0ed3a1292b3fd7a2aeccd1a169a4 ]

In nvmet_sq_destroy we capture sq->ctrl early and if it is non-NULL we
know that a ctrl was allocated (in the admin connect request handler)
and we need to release pending AERs, clear ctrl->sqs and sq->ctrl
(for nvme-loop primarily), and drop the final reference on the ctrl.

However, a small window is possible where nvmet_sq_destroy starts (as
a result of the client giving up and disconnecting) concurrently with
the nvme admin connect cmd (which may be in an early stage). But *before*
kill_and_confirm of sq->ref (i.e. the admin connect managed to get an sq
live reference). In this case, sq->ctrl was allocated however after it was
captured in a local variable in nvmet_sq_destroy.
This prevented the final reference drop on the ctrl.

Solve this by re-capturing the sq->ctrl after all inflight request has
completed, where for sure sq->ctrl reference is final, and move forward
based on that.

This issue was observed in an environment with many hosts connecting
multiple ctrls simoutanuosly, creating a delay in allocating a ctrl
leading up to this race window.

Reported-by: Alex Turin <alex@vastdata.com>
Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Keith Busch <kbusch@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/nvme/target/core.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/drivers/nvme/target/core.c b/drivers/nvme/target/core.c
index 2c44d5a95c8d6..ef2e500bccfdf 100644
--- a/drivers/nvme/target/core.c
+++ b/drivers/nvme/target/core.c
@@ -801,6 +801,15 @@ void nvmet_sq_destroy(struct nvmet_sq *sq)
 	wait_for_completion(&sq->free_done);
 	percpu_ref_exit(&sq->ref);
 
+	/*
+	 * we must reference the ctrl again after waiting for inflight IO
+	 * to complete. Because admin connect may have sneaked in after we
+	 * store sq->ctrl locally, but before we killed the percpu_ref. the
+	 * admin connect allocates and assigns sq->ctrl, which now needs a
+	 * final ref put, as this ctrl is going away.
+	 */
+	ctrl = sq->ctrl;
+
 	if (ctrl) {
 		/*
 		 * The teardown flow may take some time, and the host may not
-- 
2.43.0