From mboxrd@z Thu Jan 1 00:00:00 1970 From: Claude Code Review Bot To: dri-devel-reviews@example.com Subject: Claude review: Fix some issues from igt runs. Date: Thu, 04 Jun 2026 13:30:06 +1000 Message-ID: In-Reply-To: <20260602004203.102901-1-maaz.mombasawala@broadcom.com> References: <20260602004203.102901-1-maaz.mombasawala@broadcom.com> X-Mailer: Claude Code Patch Reviewer Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 7bit MIME-Version: 1.0 Overall Series Review Subject: Fix some issues from igt runs. Author: Maaz Mombasawala Patches: 5 Reviewed: 2026-06-04T13:30:06.941577 --- This is a 4-patch series from Maaz Mombasawala (Broadcom/VMware) fixing issues found by running igt-gpu-tools against the vmwgfx driver, now at v5. The series addresses a null deref in cursor updates, missing vrefresh validation, missing TTM reservation locks, and dumb buffer lifetime management. **Patches 1 and 2** are clean and correct. **Patch 3** has a problematic error handling path in `vmw_bo_free` where a failed `ttm_bo_reserve` will still proceed to call `vmw_resource_mob_detach` (which asserts the lock is held) and `ttm_bo_unreserve` (on an unreserved BO). **Patch 4** has the most significant issue: the return value of `ttm_base_object_lookup_for_ref()` is not checked, which can lead to a refcount underflow in `vmw_dumb_surface_unref()` at cleanup time if the lookup fails. Additionally, Patch 3 is missing a Fixes tag even though it addresses specific dma_resv_assert_held warnings. --- --- Generated by Claude Code Patch Reviewer