From mboxrd@z Thu Jan 1 00:00:00 1970 From: Claude Code Review Bot To: dri-devel-reviews@example.com Subject: Claude review: udmabuf: fix DMA direction mismatch in release_udmabuf() Date: Mon, 16 Mar 2026 11:52:22 +1000 Message-ID: In-Reply-To: <20260314232722.15555-1-mikhail.v.gavrilov@gmail.com> References: <20260314232722.15555-1-mikhail.v.gavrilov@gmail.com> X-Mailer: Claude Code Patch Reviewer Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Overall Series Review Subject: udmabuf: fix DMA direction mismatch in release_udmabuf() Author: Mikhail Gavrilov Patches: 1 Reviewed: 2026-03-16T11:52:22.902903 --- This is a single-patch fix for a real DMA API direction mismatch bug in `ud= mabuf`. The bug is clearly described: `begin_cpu_udmabuf()` maps the sg_tab= le with the caller-provided direction but `release_udmabuf()` always unmapp= ed with `DMA_BIDIRECTIONAL`, triggering a warning with `CONFIG_DMA_API_DEBU= G=3Dy`. The fix is minimal and correct =E2=80=94 it stores the mapping dire= ction and replays it on unmap. **Verdict: Looks good overall, with one observation about a pre-existing is= sue worth noting.** --- Generated by Claude Code Patch Reviewer