From mboxrd@z Thu Jan 1 00:00:00 1970 From: Claude Code Review Bot To: dri-devel-reviews@example.com Subject: Claude review: mm/memory: Fix spurious warning when unmapping device-private/exclusive pages Date: Tue, 05 May 2026 09:37:51 +1000 Message-ID: In-Reply-To: <20260501065116.2057242-1-apopple@nvidia.com> References: <20260501065116.2057242-1-apopple@nvidia.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: mm/memory: Fix spurious warning when unmapping device-private/excl= usive pages Author: Alistair Popple Patches: 2 Reviewed: 2026-05-05T09:37:51.112715 --- This is a single well-targeted bugfix for a real correctness issue in the M= M unmap path. The analysis is sound: device private/exclusive migration val= idates folios using `folio_test_anon()`, but the unmap path's assertion use= d `vma_is_anonymous()` =E2=80=94 a weaker condition that fails for CoW'd pa= ges in private file-backed mappings (`MAP_PRIVATE` on a file). The fix alig= ns the assertion with the actual invariant, and the selftest exercises the = exact scenario that triggers the spurious warning. The commit message is ex= cellent =E2=80=94 clear problem statement, good explanation of the asymmetr= y between the two predicates, and a full stack trace. **Verdict: Looks good.** One minor nit on the test. --- Generated by Claude Code Patch Reviewer