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: Set dev->registered back to false in case of register failure Date: Mon, 25 May 2026 18:27:44 +1000 Message-ID: In-Reply-To: <20260522154037.46892-3-krzysztof.niemiec@intel.com> References: <20260522154037.46892-3-krzysztof.niemiec@intel.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: drm: Set dev->registered back to false in case of register failure Author: Krzysztof Niemiec Patches: 2 Reviewed: 2026-05-25T18:27:44.910345 --- This is a single-patch fix for a real bug in the `drm_dev_register()` error= path. The analysis is correct and the fix is appropriate. The problem: `dev->registered` is set to `true` at line 1091, before `drive= r->load()` (line 1094) and `drm_modeset_register_all()` (line 1100) are cal= led. If either of these fails, the error path cleans up the registration bu= t never resets `dev->registered` back to `false`. This leaves the device in= an inconsistent state =E2=80=94 all registration side-effects have been un= wound, but the flag still claims the device is registered. The fix is a one-line addition that is both minimal and correct. --- Generated by Claude Code Patch Reviewer