dmaengine: fix use-after-free in dma_chan_put() and dma_release_channel()grafted

When dma_device_put() drops the last reference on chan->device->ref,
dma_device_release() runs and may free the dma_device along with its
channels.

dma_chan_put() then still reads chan->device->owner via
dma_chan_to_owner() for the trailing module_put(). KASAN catches it:

	slab-use-after-free in dma_chan_put+0x3e6/0x4c0
	Read of size 8 by task insmod/6319
	Freed by task 6319:
	  kfree+0x225/0x470
	  dma_chan_put+0x395/0x4c0
	  dmaengine_put+0xf8/0x160

Cache the module owner in dma_chan_put() before the put so the trailing
module_put() does not need chan->device.

Fixes: 8ad342a86359 ("dmaengine: Add reference counting to dma_device struct")
Suggested-by: Sashiko <sashiko-bot@kernel.org>
Link: https://sashiko.dev/#/patchset/20260518-dmaengine-kref-fix-v1-1-4d6125048fb7@amd.com
Reviewed-by: Frank Li <Frank.Li@nxp.com>
Reviewed-by: Logan Gunthorpe <logang@deltatee.com>
Signed-off-by: Shivank Garg <shivankg@amd.com>
Link: https://patch.msgid.link/20260822-dmaengine-kref-fix-v5-3-d4a4ee47d927@amd.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

1 files changed, 48 insertions, 0 deletions

diff --git a/include/net/dsa_stubs.h b/include/net/dsa_stubs.h
new file mode 100644
index 000000000..6f384897f
--- /dev/null
+++ b/include/net/dsa_stubs.h
@@ -0,0 +1,48 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ * include/net/dsa_stubs.h - Stubs for the Distributed Switch Architecture framework
+ */
+
+#include <linux/mutex.h>
+#include <linux/netdevice.h>
+#include <linux/net_tstamp.h>
+#include <net/dsa.h>
+
+#if IS_ENABLED(CONFIG_NET_DSA)
+
+extern const struct dsa_stubs *dsa_stubs;
+
+struct dsa_stubs {
+	int (*conduit_hwtstamp_validate)(struct net_device *dev,
+					 const struct kernel_hwtstamp_config *config,
+					 struct netlink_ext_ack *extack);
+};
+
+static inline int dsa_conduit_hwtstamp_validate(struct net_device *dev,
+						const struct kernel_hwtstamp_config *config,
+						struct netlink_ext_ack *extack)
+{
+	if (!netdev_uses_dsa(dev))
+		return 0;
+
+	/* rtnl_lock() is a sufficient guarantee, because as long as
+	 * netdev_uses_dsa() returns true, the dsa_core module is still
+	 * registered, and so, dsa_unregister_stubs() couldn't have run.
+	 * For netdev_uses_dsa() to start returning false, it would imply that
+	 * dsa_conduit_teardown() has executed, which requires rtnl_lock().
+	 */
+	ASSERT_RTNL();
+
+	return dsa_stubs->conduit_hwtstamp_validate(dev, config, extack);
+}
+
+#else
+
+static inline int dsa_conduit_hwtstamp_validate(struct net_device *dev,
+						const struct kernel_hwtstamp_config *config,
+						struct netlink_ext_ack *extack)
+{
+	return 0;
+}
+
+#endif