From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from gabe.freedesktop.org (gabe.freedesktop.org [131.252.210.177]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature ECDSA (prime256v1) server-digest SHA256) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 036FB606A320 for ; Fri, 20 Nov 2026 14:49:07 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.a=rsa-sha256 header.s=k20260515 header.b=enhTVTbT Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id 9C2BF10E90B; Fri, 20 Nov 2026 14:49:01 +0000 (UTC) Received: from sea.source.kernel.org (sea.source.kernel.org [172.234.252.31]) by gabe.freedesktop.org (Postfix) with ESMTPS id 415B5112AE5; Fri, 20 Nov 2026 14:48:24 +0000 (UTC) Received: from smtp.kernel.org (quasi.space.kernel.org [100.103.45.18]) by sea.source.kernel.org (Postfix) with ESMTP id 2A461412E7; Fri, 20 Nov 2026 14:48:24 +0000 (UTC) Received: by smtp.kernel.org (Postfix) with ESMTPSA id 532841F000CA; Fri, 20 Nov 2026 14:48:23 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=kernel.org; s=k20260515; t=1795186103; bh=U8ePJMeHIjONytavSquAM03LDhtIP3ucu16x1Y7QSk4=; h=From:Subject:Reply-To:To:In-Reply-To:References:Date; b=enhTVTbT37xupdXqNdfvuH7X8C+G9e3rY30MXleaVypdKsTvt7wIrJuuFywOlUdj3 XO5NYDM64bqMmgF0BRahmMufEeoSd/pTYogtIRdgmbzcDOZCWWWVSzBLz1JgwE36Zs UD1x4n2/7kVMFyMBoxqX8Vm6rHv5mnGxzldrjKm8/Zozcm+smydmUBqW+qccqCpFmk 2WzvQQ/uWR7b4aQ4VmKMXUMwbA9emfhXkmJXM/Ii7tnVdRK9Ok1Zf2waFIL29uPhZ3 H2/HQUbxrRmlEDo5jRG34LNDbDFoFUmr+znPqcCPHNTva3Z+DGZMIxwvOD+duFKMJR ydbXcIaKKK0iw== Message-ID: From: sashiko-bot@kernel.org Subject: Re: [PATCH 27/29] drm/i915/dp: Pick the DSC slice count for pass-through within the HDMI limits To: "Ankit Nautiyal" , intel-gfx@lists.freedesktop.org, intel-xe@lists.freedesktop.org In-Reply-To: <20261120140729.466177-28-ankit.k.nautiyal@intel.com> References: <20261120140729.466177-1-ankit.k.nautiyal@intel.com> <20261120140729.466177-28-ankit.k.nautiyal@intel.com> Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Date: Fri, 20 Nov 2026 14:48:22 +0000 X-BeenThere: intel-gfx@lists.freedesktop.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Intel graphics driver community testing & development List-Archive: List-Post: List-Help: List-Subscribe: , Reply-To: sashiko-reviews@lists.linux.dev Sender: "Intel-gfx" X-Spam-Status: No, score=-10.8 required=3.0 tests=BAYES_00,DKIMWL_WL_HIGH, DKIM_SIGNED,DKIM_VALID,DKIM_VALID_AU,HEADER_FROM_DIFFERENT_DOMAINS, INCLUDES_PATCH,MAILING_LIST_MULTI,SPF_HELO_NONE,SPF_PASS autolearn=unavailable autolearn_force=no version=4.0.1 X-Spam-Checker-Version: SpamAssassin 4.0.1 (2024-03-26) on aws-us-west-2-korg-listserv-alma10-1 X-Archived-By: Thank you for your contribution! Sashiko AI review found 1 potential issue to consider: - [Medium] drm/i915/dp: PCON pass-through DSC slice limits ignored in TU and min_hblank -- commit 5df55e454ef968026d14526944396c73e571eb41 Author: Ankit Nautiyal drm/i915/dp: Pick the DSC slice count for pass-through within the HDMI limi= ts This commit populates the downstream DSC slice count limits from the HDMI sink when passing a DSC 1.2a stream through a PCON. > diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i9= 15/display/intel_dp.c > index b4f943d6401d4..3a53d23c4587c 100644 > --- a/drivers/gpu/drm/i915/display/intel_dp.c > +++ b/drivers/gpu/drm/i915/display/intel_dp.c > @@ -2912,7 +2912,8 @@ intel_dp_dsc_compute_pipe_bpp_limits(struct intel_c= onnector *connector, > =20 > /* > * The stream of a PCON passing DSC through is not recompressed, so it h= as to > - * fit the HDMI FRL link as is. Limit the compressed bpp of the DP strea= m. > + * fit the HDMI link and be one the HDMI sink can decode as is. Limit th= e DSC > + * parameters of the DP stream accordingly. > * > * This applies only if the HDMI link needs DSC and the DP stream is a D= SC 1.2a > * stream, as passing it through is the only PCON operation possible the= n. > @@ -2936,6 +2937,13 @@ intel_dp_pcon_dsc_compute_passthrough_limits(struc= t intel_dp *intel_dp, > true) =3D=3D MODE_OK) > return true; > =20 > + if (!intel_dp_pcon_dsc_hdmi_slice_limits(connector, adjusted_mode, > + adjusted_mode->crtc_clock, > + crtc_state->sink_format, > + &limits->dfp_dsc.min_slices, > + &limits->dfp_dsc.max_slices)) > + return false; [Severity: Medium] Do the TU and min_hblank calculations still ignore limits->dfp_dsc when the HDMI sink requires a larger minimum slice count than the DP sink? In intel_dp_dsc_compute_config(), intel_dp_dsc_compute_pipe_bpp() runs before intel_dp_dsc_get_slice_config() populates pipe_config->dsc.slice_config: drivers/gpu/drm/i915/display/intel_dp.c:intel_dp_dsc_compute_config() { if (!is_mst) { ret =3D intel_dp_dsc_compute_pipe_bpp(intel_dp, pipe_config, conn_state, limits); if (ret) { ... return ret; } } if (!intel_dp_dsc_get_slice_config(connector, adjusted_mode->crtc_clock, adjusted_mode->crtc_hdisplay, num_joined_pipes, limits->dfp_dsc.min_slices, limits->dfp_dsc.max_slices, &pipe_config->dsc.slice_config)) return -EINVAL; } On a UHBR link, intel_dp_dsc_compute_pipe_bpp() -> dsc_compute_link_config() -> intel_dp_mtp_tu_compute_config() obtains the slice count via intel_dp_dsc_get_slice_count(), which hardcodes dfp_min_slices =3D 0 and dfp_max_slices =3D INT_MAX (and on non-UHBR links, dsc_compute_link_config() reads intel_dsc_line_slice_count(&pipe_config->dsc.slice_config) while slice_config is still zeroed). Later in intel_dp_compute_config(), intel_dp_compute_min_hblank() also calls intel_dp_dsc_get_slice_count() instead of using intel_dsc_line_slice_count(&crtc_state->dsc.slice_config). As a result, symbol cycles, link M/N, TU, and min_hblank are computed for the unconstrained DP slice count rather than the slice count programmed in pipe_config->dsc.slice_config. Computing pipe_config->dsc.slice_config before intel_dp_dsc_compute_pipe_bpp() and reading intel_dsc_line_slice_count(&crtc_state->dsc.slice_config) in the downstream callers would keep these calculations consistent. --=20 Sashiko AI review =C2=B7 https://sashiko.dev/#/patchset/20261120140729.4661= 77-1-ankit.k.nautiyal@intel.com?part=3D27