From mboxrd@z Thu Jan 1 00:00:00 1970 From: Claude Code Review Bot To: dri-devel-reviews@example.com Subject: Claude review: bus: mhi: Centralize firmware image table selection at probe time Date: Mon, 09 Mar 2026 08:09:00 +1000 Message-ID: In-Reply-To: <20260307-sahara_protocol_new_v2-v2-4-29dc748b5e9c@oss.qualcomm.com> References: <20260307-sahara_protocol_new_v2-v2-0-29dc748b5e9c@oss.qualcomm.com> <20260307-sahara_protocol_new_v2-v2-4-29dc748b5e9c@oss.qualcomm.com> X-Mailer: Claude Code Patch Reviewer Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Patch Review The variant table approach is reasonable. - **`sahara_select_variant` returns NULL =E2=86=92 `-ENODEV` is now a hard = failure.** Previously, the driver defaulted to AIC100 for any unrecognized = controller name. Now, any device that doesn't match a variant table entry w= ill fail probe. This is a behavioral change that could break existing setup= s if the controller name doesn't match exactly. The commit message claims "= preserving the existing behavior for all supported AIC devices" but doesn't= note this fallback removal. - **Double space typo in commit message:** `"harder to follow"` -- minor n= it. - **`fw_folder` field added to `sahara_context` but not used yet** in this = patch. It's used in patch 6. This is fine for bisectability since it's just= a struct field addition. --- Generated by Claude Code Patch Reviewer