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/imagination: Switch reset_reason fields from enum to u32 Date: Fri, 13 Mar 2026 14:16:58 +1000 Message-ID: In-Reply-To: <20260312-b4-firmware-context-reset-notification-handling-v2-2-aec5a64cb06f@imgtec.com> References: <20260312-b4-firmware-context-reset-notification-handling-v2-0-aec5a64cb06f@imgtec.com> <20260312-b4-firmware-context-reset-notification-handling-v2-2-aec5a64cb06f@imgtec.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 This patch changes `reset_reason` from `enum rogue_context_reset_reason` to= `u32` in two structs that define firmware-shared data structures. The rati= onale is correct =E2=80=94 `enum` size is compiler-dependent and unsuitable= for firmware ABI structures. The `Fixes:` tag referencing the original FWIF headers commit is appropriat= e. **Minor observation:** The patch changes the field type in `rogue_fwif_fwcc= b_cmd_context_reset_data` and `rogue_context_reset_reason_data`, but there = are likely existing users of `reset_reason_data.reset_reason` elsewhere in = the driver that previously relied on the enum type. Let me verify this isn'= t a problem: The existing code in `pvr_context.c` that reads `reset_reason` from `rogue_= context_reset_reason_data` should be unaffected since `u32` is assignment-c= ompatible with the previous enum values. No functional issues. **Looks good.** --- --- Generated by Claude Code Patch Reviewer