From mboxrd@z Thu Jan 1 00:00:00 1970 From: Claude Code Review Bot To: dri-devel-reviews@example.com Subject: Claude review: drm/i915/dp: Fix link retrain loop on pre-gen8 platforms with DP adapters Date: Tue, 05 May 2026 10:51:24 +1000 Message-ID: In-Reply-To: <20260430030846.29373-1-s.mavlenkov@mail.ru> References: <20260430030846.29373-1-s.mavlenkov@mail.ru> 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: drm/i915/dp: Fix link retrain loop on pre-gen8 platforms with DP adapters Author: bitey Patches: 2 Reviewed: 2026-05-05T10:51:24.766213 --- This is a single patch attempting to fix an infinite link retrain loop on pre-gen8 i915 platforms when using DP-to-HDMI protocol converters (specifically a Chrontel CH73111). The commit message is well-written and clearly identifies three contributing root causes in the modern DP link training path. However, the patch has a **critical build-breaking error** and several design-level concerns. **Severity: The patch will not compile as submitted.** The approach of gating all three code paths behind `DISPLAY_VER(display) >= 8` is a broad workaround rather than a targeted fix. It effectively disables post-training link monitoring and short-HPD link checks for all pre-gen8 platforms, not just the ones affected by misbehaving adapters. A more surgical fix (e.g., quirking the specific adapter, or using DPCD revision for the `short_pulse` change) would be preferable. Additionally, the `Signed-off-by` uses the handle "bitey" rather than a real name, which does not satisfy the DCO (Developer Certificate of Origin) requirements for kernel contributions. --- --- Generated by Claude Code Patch Reviewer