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, 35 insertions, 0 deletions

diff --git a/rust/syn/verbatim.rs b/rust/syn/verbatim.rs
new file mode 100644
index 000000000..97bc08f70
--- /dev/null
+++ b/rust/syn/verbatim.rs
@@ -0,0 +1,35 @@
+// SPDX-License-Identifier: Apache-2.0 OR MIT
+
+use crate::parse::ParseStream;
+use proc_macro2::{Delimiter, TokenStream};
+use std::cmp::Ordering;
+use std::iter;
+
+pub(crate) fn between<'a>(begin: ParseStream<'a>, end: ParseStream<'a>) -> TokenStream {
+    let end = end.cursor();
+    let mut cursor = begin.cursor();
+    assert!(crate::buffer::same_buffer(end, cursor));
+
+    let mut tokens = TokenStream::new();
+    while cursor != end {
+        let (tt, next) = cursor.token_tree().unwrap();
+
+        if crate::buffer::cmp_assuming_same_buffer(end, next) == Ordering::Less {
+            // A syntax node can cross the boundary of a None-delimited group
+            // due to such groups being transparent to the parser in most cases.
+            // Any time this occurs the group is known to be semantically
+            // irrelevant. https://github.com/dtolnay/syn/issues/1235
+            if let Some((inside, _span, after)) = cursor.group(Delimiter::None) {
+                assert!(next == after);
+                cursor = inside;
+                continue;
+            } else {
+                panic!("verbatim end must not be inside a delimited group");
+            }
+        }
+
+        tokens.extend(iter::once(tt));
+        cursor = next;
+    }
+    tokens
+}