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/display: Disable Panel Replay for DP-tunneling without optimization Date: Fri, 13 Mar 2026 14:30:27 +1000 Message-ID: In-Reply-To: <20260312050035.3493690-4-animesh.manna@intel.com> References: <20260312050035.3493690-1-animesh.manna@intel.com> <20260312050035.3493690-4-animesh.manna@intel.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 **Typo in debug message:** ```c "Panel Replay is disabled as DP tunelling enabled without optimization\n" ``` "tunelling" =E2=86=92 "tunneling" (missing 'n'). **Placement of the check:** The check is inserted after computing `link_off= _after_as_sdp_when_pr_active` and `disable_as_sdp_when_pr_active` but befor= e the `!intel_dp_is_edp()` early return. This means these `crtc_state` fiel= ds get set even when Panel Replay will be disabled by this new check. Those= fields are only meaningful when Panel Replay is active, so this is a minor= inefficiency but not a bug =E2=80=94 the state is ignored when `has_panel_= replay` ends up false. **The spec quote is helpful** =E2=80=94 it clearly justifies why Panel Repl= ay must be disabled when the bridge doesn't support PR tunneling optimizati= on. Good commit message. **Ordering concern between patch 2 and patch 3:** Patch 2 enables the optim= ization bit in the activation path, and patch 3 disables Panel Replay entir= ely when the optimization isn't supported. If patch 2 were applied alone (w= ithout patch 3), Panel Replay would still be enabled on tunneled connection= s without optimization support =E2=80=94 just without the optimization bit = set. The DP spec says PR should not be enabled in that case. Consider wheth= er these should be reordered or squashed so there's no window where the dri= ver violates the spec, even bisecting between patches. **Summary of issues to address:** 1. **Typo** in patch 1 doc: "optimiation" =E2=86=92 "optimization" 2. **Typo** in patch 3 debug message: "tunelling" =E2=86=92 "tunneling" =20 3. **Copy-paste error** in patch 1 `intel_dp_tunnel_pr_optimization_support= ed()` doc: says "BW allocation mode" instead of "PR BW optimization" 4. **Missing deactivation path** in patch 2: verify that `TRANS_DP2_PR_TUNN= ELING_ENABLE` is cleared when Panel Replay is disabled 5. **Patch ordering**: patch 3 should logically come before patch 2 to avoi= d a bisect-unsafe spec violation window 6. **Minor**: consider adding `pr_optimization_support` change detection in= `update_tunnel_regs()` for consistency with `bw_alloc_supported` --- Generated by Claude Code Patch Reviewer