From mboxrd@z Thu Jan 1 00:00:00 1970 From: Claude Code Review Bot To: dri-devel-reviews@example.com Subject: Claude review: var-som-6ul: improve support for variants Date: Tue, 03 Mar 2026 12:58:05 +1000 Message-ID: In-Reply-To: <20260302190953.669325-1-hugo@hugovil.com> References: <20260302190953.669325-1-hugo@hugovil.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: var-som-6ul: improve support for variants Author: Hugo Villeneuve Patches: 23 Reviewed: 2026-03-03T12:58:05.042886 --- This is a 14-patch series that improves device tree support for the Variscite VAR-SOM-6UL boards. The first few patches fix genuine DT warnings, then the bulk of the series refactors the existing monolithic DTS files into modular DTSI includes (SD, WiFi, ENET1, ENET2, audio, LVDS) and adds imx6ull CPU variant support. The refactoring direction is reasonable, but there are several technical issues: 1. **Critical**: The imx6ull variant files put `MX6ULL_PAD_SNVS_TAMPER*` pads under `&iomuxc` when they must go under `&iomuxc_snvs` on the imx6ull. 2. **Significant**: The ENET1/ENET2 factoring creates a hard dependency where enet1.dtsi can only work if enet2.dtsi is included first, which the author acknowledges but should be resolved. 3. **Significant**: Duplicate `ethphy0` node definitions across enet1.dtsi and enet2.dtsi. 4. The compatible string naming in patch 13 is inconsistent between commit message and code. 5. The LVDS panel node placement in concerto-common means even boards without LVDS get an orphan panel node. The series would benefit from being split: fixes (1-2) separate from the refactoring/new-features (3-14), as the fixes target stable while the refactoring does not. --- --- Generated by Claude Code Patch Reviewer