From patchwork Sat Dec 2 21:40:07 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Dmitry Baryshkov X-Patchwork-Id: 749826 Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=linaro.org header.i=@linaro.org header.b="q8loNusc" Received: from mail-lf1-x136.google.com (mail-lf1-x136.google.com [IPv6:2a00:1450:4864:20::136]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 018F3E5 for ; Sat, 2 Dec 2023 13:40:22 -0800 (PST) Received: by mail-lf1-x136.google.com with SMTP id 2adb3069b0e04-50be3eed85aso1271191e87.2 for ; Sat, 02 Dec 2023 13:40:21 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=linaro.org; s=google; t=1701553220; x=1702158020; darn=vger.kernel.org; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:from:to:cc:subject:date :message-id:reply-to; bh=fLSVQeJPy9RHvzEIAIGg2GSwp9vuyoSPXrOy+9C++e8=; b=q8loNusclLRCP/0fFx+y4uzb+gKNs9+fZyxZhMBs9eJkNIWVRtnpCRIqTXJrjeHfte upBjlwUOVAdSfJ6fP5bwrmh+eo1mNTx3PBWRHRxdKuy9Ui9Kafc4vKYaFdDI75sRBRDO QODCDCA8NnihO5vLOgEDqubutDGDZdV52pJJTGxX5FmsQFl0qZorxHaSVoO7RH+48JtQ IEbgvapvtfikGP6nAZ5NPMNi2mXN8j/CQYfimjUR8+OcnyxIHdTBdsbT5uH7jCiZx36h IdIlnJznprESTWbOwyILjZJ9Tl4aDvzApjf3H+VNlyzOGcMbD3z/ggxw/j03PCwVh+dK pZzg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1701553220; x=1702158020; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=fLSVQeJPy9RHvzEIAIGg2GSwp9vuyoSPXrOy+9C++e8=; b=Nt5NZY/ecPpQDi5b49gdev7fVHqpuKhqoRB7b7ympxF8WX+sU5cIbugSUhCvDvteev a7Nb9JxcIaT29xaiJTBKqU/ukNGFNC3t/WEnwCvC7iXXCziZA9w1njzrjY4Z1t0nF7/I gpyoM7u/y/WJv4WruQwiLNqGYUgAKWIQIdfqXt5YuPixhJ6dLRhEWp/0r9i7+OTUoKlV hdubCiXirvFMdXJC3aJ9+DhoHj/gMKmoUsI3eXRI5jSD2HFTn2ubeY42432S6iICuhrs oWJ8/BfZYzUhTIbibDIze/BjZQt8sdYfCzFhjbENG6X+SRcQQyNEUOXxYiV8xOWSDjRp XjWA== X-Gm-Message-State: AOJu0YyWezgisNRqwh/3rMG4PAloVlJ/MDIpNPI8IxpaUW6o7oWnGlth Bn9SR6l+c0Qdr0d72scTMnr8Nw== X-Google-Smtp-Source: AGHT+IGRMr5u5wWPSGNm/JAcceii5YQ2zc9nADWu0tcWXTMu9kEFYRpjPv71XPomc80+LwAQAILJxA== X-Received: by 2002:a05:6512:3692:b0:50b:f006:3f92 with SMTP id d18-20020a056512369200b0050bf0063f92mr202746lfs.134.1701553219944; Sat, 02 Dec 2023 13:40:19 -0800 (PST) Received: from umbar.unikie.fi ([192.130.178.91]) by smtp.gmail.com with ESMTPSA id f24-20020ac25338000000b0050be6038170sm282996lfh.48.2023.12.02.13.40.19 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Sat, 02 Dec 2023 13:40:19 -0800 (PST) From: Dmitry Baryshkov To: Rob Clark , Sean Paul , Abhinav Kumar , Marijn Suijten Cc: Stephen Boyd , David Airlie , Daniel Vetter , Bjorn Andersson , linux-arm-msm@vger.kernel.org, dri-devel@lists.freedesktop.org, freedreno@lists.freedesktop.org Subject: [PATCH 03/12] drm/msm/dpu: use format-related definitions from mdp_common.xml.h Date: Sun, 3 Dec 2023 00:40:07 +0300 Message-Id: <20231202214016.1257621-4-dmitry.baryshkov@linaro.org> X-Mailer: git-send-email 2.39.2 In-Reply-To: <20231202214016.1257621-1-dmitry.baryshkov@linaro.org> References: <20231202214016.1257621-1-dmitry.baryshkov@linaro.org> Precedence: bulk X-Mailing-List: linux-arm-msm@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Instead of having DPU-specific defines, switch to the definitions from the mdp_common.xml.h file. This is the preparation for merged of DPU and MDP format tables. Signed-off-by: Dmitry Baryshkov Reviewed-by: Abhinav Kumar --- .../drm/msm/disp/dpu1/dpu_encoder_phys_wb.c | 2 +- drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c | 290 +++++++++--------- drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c | 6 +- drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h | 64 +--- drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c | 12 +- drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c | 4 +- 6 files changed, 164 insertions(+), 214 deletions(-) diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c index 0b6a761d68b7..4fead04d83a0 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c @@ -640,7 +640,7 @@ static void dpu_encoder_phys_wb_prepare_wb_job(struct dpu_encoder_phys *phys_enc wb_cfg->dest.height = job->fb->height; wb_cfg->dest.num_planes = wb_cfg->dest.format->num_planes; - if ((wb_cfg->dest.format->fetch_planes == DPU_PLANE_PLANAR) && + if ((wb_cfg->dest.format->fetch_planes == MDP_PLANE_PLANAR) && (wb_cfg->dest.format->element[0] == C1_B_Cb)) swap(wb_cfg->dest.plane_addr[1], wb_cfg->dest.plane_addr[2]); diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c index e366ab134249..05e82f5dd0e6 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c @@ -35,11 +35,11 @@ bp, flg, fm, np) \ { \ .base.pixel_format = DRM_FORMAT_ ## fmt, \ - .fetch_planes = DPU_PLANE_INTERLEAVED, \ + .fetch_planes = MDP_PLANE_INTERLEAVED, \ .alpha_enable = alpha, \ .element = { (e0), (e1), (e2), (e3) }, \ .bits = { g, b, r, a }, \ - .chroma_sample = DPU_CHROMA_RGB, \ + .chroma_sample = CHROMA_FULL, \ .unpack_align_msb = 0, \ .unpack_tight = 1, \ .unpack_count = uc, \ @@ -54,11 +54,11 @@ bp, flg, fm, np) \ alpha, bp, flg, fm, np, th) \ { \ .base.pixel_format = DRM_FORMAT_ ## fmt, \ - .fetch_planes = DPU_PLANE_INTERLEAVED, \ + .fetch_planes = MDP_PLANE_INTERLEAVED, \ .alpha_enable = alpha, \ .element = { (e0), (e1), (e2), (e3) }, \ .bits = { g, b, r, a }, \ - .chroma_sample = DPU_CHROMA_RGB, \ + .chroma_sample = CHROMA_FULL, \ .unpack_align_msb = 0, \ .unpack_tight = 1, \ .unpack_count = uc, \ @@ -74,7 +74,7 @@ alpha, bp, flg, fm, np, th) \ alpha, chroma, count, bp, flg, fm, np) \ { \ .base.pixel_format = DRM_FORMAT_ ## fmt, \ - .fetch_planes = DPU_PLANE_INTERLEAVED, \ + .fetch_planes = MDP_PLANE_INTERLEAVED, \ .alpha_enable = alpha, \ .element = { (e0), (e1), (e2), (e3)}, \ .bits = { g, b, r, a }, \ @@ -92,7 +92,7 @@ alpha, chroma, count, bp, flg, fm, np) \ #define PSEUDO_YUV_FMT(fmt, a, r, g, b, e0, e1, chroma, flg, fm, np) \ { \ .base.pixel_format = DRM_FORMAT_ ## fmt, \ - .fetch_planes = DPU_PLANE_PSEUDO_PLANAR, \ + .fetch_planes = MDP_PLANE_PSEUDO_PLANAR, \ .alpha_enable = false, \ .element = { (e0), (e1), 0, 0 }, \ .bits = { g, b, r, a }, \ @@ -111,7 +111,7 @@ alpha, chroma, count, bp, flg, fm, np) \ flg, fm, np, th) \ { \ .base.pixel_format = DRM_FORMAT_ ## fmt, \ - .fetch_planes = DPU_PLANE_PSEUDO_PLANAR, \ + .fetch_planes = MDP_PLANE_PSEUDO_PLANAR, \ .alpha_enable = false, \ .element = { (e0), (e1), 0, 0 }, \ .bits = { g, b, r, a }, \ @@ -129,7 +129,7 @@ flg, fm, np, th) \ #define PSEUDO_YUV_FMT_LOOSE(fmt, a, r, g, b, e0, e1, chroma, flg, fm, np)\ { \ .base.pixel_format = DRM_FORMAT_ ## fmt, \ - .fetch_planes = DPU_PLANE_PSEUDO_PLANAR, \ + .fetch_planes = MDP_PLANE_PSEUDO_PLANAR, \ .alpha_enable = false, \ .element = { (e0), (e1), 0, 0 }, \ .bits = { g, b, r, a }, \ @@ -148,7 +148,7 @@ flg, fm, np, th) \ flg, fm, np, th) \ { \ .base.pixel_format = DRM_FORMAT_ ## fmt, \ - .fetch_planes = DPU_PLANE_PSEUDO_PLANAR, \ + .fetch_planes = MDP_PLANE_PSEUDO_PLANAR, \ .alpha_enable = false, \ .element = { (e0), (e1), 0, 0 }, \ .bits = { g, b, r, a }, \ @@ -168,7 +168,7 @@ flg, fm, np, th) \ flg, fm, np) \ { \ .base.pixel_format = DRM_FORMAT_ ## fmt, \ - .fetch_planes = DPU_PLANE_PLANAR, \ + .fetch_planes = MDP_PLANE_PLANAR, \ .alpha_enable = alpha, \ .element = { (e0), (e1), (e2), 0 }, \ .bits = { g, b, r, a }, \ @@ -195,286 +195,286 @@ struct dpu_media_color_map { static const struct dpu_format dpu_format_map[] = { INTERLEAVED_RGB_FMT(ARGB8888, - COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, + BPC8A, BPC8, BPC8, BPC8, C1_B_Cb, C0_G_Y, C2_R_Cr, C3_ALPHA, 4, true, 4, 0, - DPU_FETCH_LINEAR, 1), + MDP_FETCH_LINEAR, 1), INTERLEAVED_RGB_FMT(ABGR8888, - COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, + BPC8A, BPC8, BPC8, BPC8, C2_R_Cr, C0_G_Y, C1_B_Cb, C3_ALPHA, 4, true, 4, 0, - DPU_FETCH_LINEAR, 1), + MDP_FETCH_LINEAR, 1), INTERLEAVED_RGB_FMT(XBGR8888, - COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, + BPC8A, BPC8, BPC8, BPC8, C2_R_Cr, C0_G_Y, C1_B_Cb, C3_ALPHA, 4, false, 4, 0, - DPU_FETCH_LINEAR, 1), + MDP_FETCH_LINEAR, 1), INTERLEAVED_RGB_FMT(RGBA8888, - COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, + BPC8A, BPC8, BPC8, BPC8, C3_ALPHA, C1_B_Cb, C0_G_Y, C2_R_Cr, 4, true, 4, 0, - DPU_FETCH_LINEAR, 1), + MDP_FETCH_LINEAR, 1), INTERLEAVED_RGB_FMT(BGRA8888, - COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, + BPC8A, BPC8, BPC8, BPC8, C3_ALPHA, C2_R_Cr, C0_G_Y, C1_B_Cb, 4, true, 4, 0, - DPU_FETCH_LINEAR, 1), + MDP_FETCH_LINEAR, 1), INTERLEAVED_RGB_FMT(BGRX8888, - COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, + BPC8A, BPC8, BPC8, BPC8, C3_ALPHA, C2_R_Cr, C0_G_Y, C1_B_Cb, 4, false, 4, 0, - DPU_FETCH_LINEAR, 1), + MDP_FETCH_LINEAR, 1), INTERLEAVED_RGB_FMT(XRGB8888, - COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, + BPC8A, BPC8, BPC8, BPC8, C1_B_Cb, C0_G_Y, C2_R_Cr, C3_ALPHA, 4, false, 4, 0, - DPU_FETCH_LINEAR, 1), + MDP_FETCH_LINEAR, 1), INTERLEAVED_RGB_FMT(RGBX8888, - COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, + BPC8A, BPC8, BPC8, BPC8, C3_ALPHA, C1_B_Cb, C0_G_Y, C2_R_Cr, 4, false, 4, 0, - DPU_FETCH_LINEAR, 1), + MDP_FETCH_LINEAR, 1), INTERLEAVED_RGB_FMT(RGB888, - 0, COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, + 0, BPC8, BPC8, BPC8, C1_B_Cb, C0_G_Y, C2_R_Cr, 0, 3, false, 3, 0, - DPU_FETCH_LINEAR, 1), + MDP_FETCH_LINEAR, 1), INTERLEAVED_RGB_FMT(BGR888, - 0, COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, + 0, BPC8, BPC8, BPC8, C2_R_Cr, C0_G_Y, C1_B_Cb, 0, 3, false, 3, 0, - DPU_FETCH_LINEAR, 1), + MDP_FETCH_LINEAR, 1), INTERLEAVED_RGB_FMT(RGB565, - 0, COLOR_5BIT, COLOR_6BIT, COLOR_5BIT, + 0, BPC5, BPC6, BPC5, C1_B_Cb, C0_G_Y, C2_R_Cr, 0, 3, false, 2, 0, - DPU_FETCH_LINEAR, 1), + MDP_FETCH_LINEAR, 1), INTERLEAVED_RGB_FMT(BGR565, - 0, COLOR_5BIT, COLOR_6BIT, COLOR_5BIT, + 0, BPC5, BPC6, BPC5, C2_R_Cr, C0_G_Y, C1_B_Cb, 0, 3, false, 2, 0, - DPU_FETCH_LINEAR, 1), + MDP_FETCH_LINEAR, 1), INTERLEAVED_RGB_FMT(ARGB1555, - COLOR_ALPHA_1BIT, COLOR_5BIT, COLOR_5BIT, COLOR_5BIT, + BPC1A, BPC5, BPC5, BPC5, C1_B_Cb, C0_G_Y, C2_R_Cr, C3_ALPHA, 4, true, 2, 0, - DPU_FETCH_LINEAR, 1), + MDP_FETCH_LINEAR, 1), INTERLEAVED_RGB_FMT(ABGR1555, - COLOR_ALPHA_1BIT, COLOR_5BIT, COLOR_5BIT, COLOR_5BIT, + BPC1A, BPC5, BPC5, BPC5, C2_R_Cr, C0_G_Y, C1_B_Cb, C3_ALPHA, 4, true, 2, 0, - DPU_FETCH_LINEAR, 1), + MDP_FETCH_LINEAR, 1), INTERLEAVED_RGB_FMT(RGBA5551, - COLOR_ALPHA_1BIT, COLOR_5BIT, COLOR_5BIT, COLOR_5BIT, + BPC1A, BPC5, BPC5, BPC5, C3_ALPHA, C1_B_Cb, C0_G_Y, C2_R_Cr, 4, true, 2, 0, - DPU_FETCH_LINEAR, 1), + MDP_FETCH_LINEAR, 1), INTERLEAVED_RGB_FMT(BGRA5551, - COLOR_ALPHA_1BIT, COLOR_5BIT, COLOR_5BIT, COLOR_5BIT, + BPC1A, BPC5, BPC5, BPC5, C3_ALPHA, C2_R_Cr, C0_G_Y, C1_B_Cb, 4, true, 2, 0, - DPU_FETCH_LINEAR, 1), + MDP_FETCH_LINEAR, 1), INTERLEAVED_RGB_FMT(XRGB1555, - COLOR_ALPHA_1BIT, COLOR_5BIT, COLOR_5BIT, COLOR_5BIT, + BPC1A, BPC5, BPC5, BPC5, C1_B_Cb, C0_G_Y, C2_R_Cr, C3_ALPHA, 4, false, 2, 0, - DPU_FETCH_LINEAR, 1), + MDP_FETCH_LINEAR, 1), INTERLEAVED_RGB_FMT(XBGR1555, - COLOR_ALPHA_1BIT, COLOR_5BIT, COLOR_5BIT, COLOR_5BIT, + BPC1A, BPC5, BPC5, BPC5, C2_R_Cr, C0_G_Y, C1_B_Cb, C3_ALPHA, 4, false, 2, 0, - DPU_FETCH_LINEAR, 1), + MDP_FETCH_LINEAR, 1), INTERLEAVED_RGB_FMT(RGBX5551, - COLOR_ALPHA_1BIT, COLOR_5BIT, COLOR_5BIT, COLOR_5BIT, + BPC1A, BPC5, BPC5, BPC5, C3_ALPHA, C1_B_Cb, C0_G_Y, C2_R_Cr, 4, false, 2, 0, - DPU_FETCH_LINEAR, 1), + MDP_FETCH_LINEAR, 1), INTERLEAVED_RGB_FMT(BGRX5551, - COLOR_ALPHA_1BIT, COLOR_5BIT, COLOR_5BIT, COLOR_5BIT, + BPC1A, BPC5, BPC5, BPC5, C3_ALPHA, C2_R_Cr, C0_G_Y, C1_B_Cb, 4, false, 2, 0, - DPU_FETCH_LINEAR, 1), + MDP_FETCH_LINEAR, 1), INTERLEAVED_RGB_FMT(ARGB4444, - COLOR_ALPHA_4BIT, COLOR_4BIT, COLOR_4BIT, COLOR_4BIT, + BPC4A, BPC4, BPC4, BPC4, C1_B_Cb, C0_G_Y, C2_R_Cr, C3_ALPHA, 4, true, 2, 0, - DPU_FETCH_LINEAR, 1), + MDP_FETCH_LINEAR, 1), INTERLEAVED_RGB_FMT(ABGR4444, - COLOR_ALPHA_4BIT, COLOR_4BIT, COLOR_4BIT, COLOR_4BIT, + BPC4A, BPC4, BPC4, BPC4, C2_R_Cr, C0_G_Y, C1_B_Cb, C3_ALPHA, 4, true, 2, 0, - DPU_FETCH_LINEAR, 1), + MDP_FETCH_LINEAR, 1), INTERLEAVED_RGB_FMT(RGBA4444, - COLOR_ALPHA_4BIT, COLOR_4BIT, COLOR_4BIT, COLOR_4BIT, + BPC4A, BPC4, BPC4, BPC4, C3_ALPHA, C1_B_Cb, C0_G_Y, C2_R_Cr, 4, true, 2, 0, - DPU_FETCH_LINEAR, 1), + MDP_FETCH_LINEAR, 1), INTERLEAVED_RGB_FMT(BGRA4444, - COLOR_ALPHA_4BIT, COLOR_4BIT, COLOR_4BIT, COLOR_4BIT, + BPC4A, BPC4, BPC4, BPC4, C3_ALPHA, C2_R_Cr, C0_G_Y, C1_B_Cb, 4, true, 2, 0, - DPU_FETCH_LINEAR, 1), + MDP_FETCH_LINEAR, 1), INTERLEAVED_RGB_FMT(XRGB4444, - COLOR_ALPHA_4BIT, COLOR_4BIT, COLOR_4BIT, COLOR_4BIT, + BPC4A, BPC4, BPC4, BPC4, C1_B_Cb, C0_G_Y, C2_R_Cr, C3_ALPHA, 4, false, 2, 0, - DPU_FETCH_LINEAR, 1), + MDP_FETCH_LINEAR, 1), INTERLEAVED_RGB_FMT(XBGR4444, - COLOR_ALPHA_4BIT, COLOR_4BIT, COLOR_4BIT, COLOR_4BIT, + BPC4A, BPC4, BPC4, BPC4, C2_R_Cr, C0_G_Y, C1_B_Cb, C3_ALPHA, 4, false, 2, 0, - DPU_FETCH_LINEAR, 1), + MDP_FETCH_LINEAR, 1), INTERLEAVED_RGB_FMT(RGBX4444, - COLOR_ALPHA_4BIT, COLOR_4BIT, COLOR_4BIT, COLOR_4BIT, + BPC4A, BPC4, BPC4, BPC4, C3_ALPHA, C1_B_Cb, C0_G_Y, C2_R_Cr, 4, false, 2, 0, - DPU_FETCH_LINEAR, 1), + MDP_FETCH_LINEAR, 1), INTERLEAVED_RGB_FMT(BGRX4444, - COLOR_ALPHA_4BIT, COLOR_4BIT, COLOR_4BIT, COLOR_4BIT, + BPC4A, BPC4, BPC4, BPC4, C3_ALPHA, C2_R_Cr, C0_G_Y, C1_B_Cb, 4, false, 2, 0, - DPU_FETCH_LINEAR, 1), + MDP_FETCH_LINEAR, 1), INTERLEAVED_RGB_FMT(BGRA1010102, - COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, + BPC8A, BPC8, BPC8, BPC8, C3_ALPHA, C2_R_Cr, C0_G_Y, C1_B_Cb, 4, true, 4, DPU_FORMAT_FLAG_DX, - DPU_FETCH_LINEAR, 1), + MDP_FETCH_LINEAR, 1), INTERLEAVED_RGB_FMT(RGBA1010102, - COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, + BPC8A, BPC8, BPC8, BPC8, C3_ALPHA, C1_B_Cb, C0_G_Y, C2_R_Cr, 4, true, 4, DPU_FORMAT_FLAG_DX, - DPU_FETCH_LINEAR, 1), + MDP_FETCH_LINEAR, 1), INTERLEAVED_RGB_FMT(ABGR2101010, - COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, + BPC8A, BPC8, BPC8, BPC8, C2_R_Cr, C0_G_Y, C1_B_Cb, C3_ALPHA, 4, true, 4, DPU_FORMAT_FLAG_DX, - DPU_FETCH_LINEAR, 1), + MDP_FETCH_LINEAR, 1), INTERLEAVED_RGB_FMT(ARGB2101010, - COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, + BPC8A, BPC8, BPC8, BPC8, C1_B_Cb, C0_G_Y, C2_R_Cr, C3_ALPHA, 4, true, 4, DPU_FORMAT_FLAG_DX, - DPU_FETCH_LINEAR, 1), + MDP_FETCH_LINEAR, 1), INTERLEAVED_RGB_FMT(XRGB2101010, - COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, + BPC8A, BPC8, BPC8, BPC8, C1_B_Cb, C0_G_Y, C2_R_Cr, C3_ALPHA, 4, false, 4, DPU_FORMAT_FLAG_DX, - DPU_FETCH_LINEAR, 1), + MDP_FETCH_LINEAR, 1), INTERLEAVED_RGB_FMT(BGRX1010102, - COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, + BPC8A, BPC8, BPC8, BPC8, C3_ALPHA, C2_R_Cr, C0_G_Y, C1_B_Cb, 4, false, 4, DPU_FORMAT_FLAG_DX, - DPU_FETCH_LINEAR, 1), + MDP_FETCH_LINEAR, 1), INTERLEAVED_RGB_FMT(XBGR2101010, - COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, + BPC8A, BPC8, BPC8, BPC8, C2_R_Cr, C0_G_Y, C1_B_Cb, C3_ALPHA, 4, false, 4, DPU_FORMAT_FLAG_DX, - DPU_FETCH_LINEAR, 1), + MDP_FETCH_LINEAR, 1), INTERLEAVED_RGB_FMT(RGBX1010102, - COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, + BPC8A, BPC8, BPC8, BPC8, C3_ALPHA, C1_B_Cb, C0_G_Y, C2_R_Cr, 4, false, 4, DPU_FORMAT_FLAG_DX, - DPU_FETCH_LINEAR, 1), + MDP_FETCH_LINEAR, 1), PSEUDO_YUV_FMT(NV12, - 0, COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, + 0, BPC8, BPC8, BPC8, C1_B_Cb, C2_R_Cr, - DPU_CHROMA_420, DPU_FORMAT_FLAG_YUV, - DPU_FETCH_LINEAR, 2), + CHROMA_420, DPU_FORMAT_FLAG_YUV, + MDP_FETCH_LINEAR, 2), PSEUDO_YUV_FMT(NV21, - 0, COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, + 0, BPC8, BPC8, BPC8, C2_R_Cr, C1_B_Cb, - DPU_CHROMA_420, DPU_FORMAT_FLAG_YUV, - DPU_FETCH_LINEAR, 2), + CHROMA_420, DPU_FORMAT_FLAG_YUV, + MDP_FETCH_LINEAR, 2), PSEUDO_YUV_FMT(NV16, - 0, COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, + 0, BPC8, BPC8, BPC8, C1_B_Cb, C2_R_Cr, - DPU_CHROMA_H2V1, DPU_FORMAT_FLAG_YUV, - DPU_FETCH_LINEAR, 2), + CHROMA_H2V1, DPU_FORMAT_FLAG_YUV, + MDP_FETCH_LINEAR, 2), PSEUDO_YUV_FMT(NV61, - 0, COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, + 0, BPC8, BPC8, BPC8, C2_R_Cr, C1_B_Cb, - DPU_CHROMA_H2V1, DPU_FORMAT_FLAG_YUV, - DPU_FETCH_LINEAR, 2), + CHROMA_H2V1, DPU_FORMAT_FLAG_YUV, + MDP_FETCH_LINEAR, 2), PSEUDO_YUV_FMT_LOOSE(P010, - 0, COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, + 0, BPC8, BPC8, BPC8, C1_B_Cb, C2_R_Cr, - DPU_CHROMA_420, DPU_FORMAT_FLAG_DX | DPU_FORMAT_FLAG_YUV, - DPU_FETCH_LINEAR, 2), + CHROMA_420, DPU_FORMAT_FLAG_DX | DPU_FORMAT_FLAG_YUV, + MDP_FETCH_LINEAR, 2), INTERLEAVED_YUV_FMT(VYUY, - 0, COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, + 0, BPC8, BPC8, BPC8, C2_R_Cr, C0_G_Y, C1_B_Cb, C0_G_Y, - false, DPU_CHROMA_H2V1, 4, 2, DPU_FORMAT_FLAG_YUV, - DPU_FETCH_LINEAR, 2), + false, CHROMA_H2V1, 4, 2, DPU_FORMAT_FLAG_YUV, + MDP_FETCH_LINEAR, 2), INTERLEAVED_YUV_FMT(UYVY, - 0, COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, + 0, BPC8, BPC8, BPC8, C1_B_Cb, C0_G_Y, C2_R_Cr, C0_G_Y, - false, DPU_CHROMA_H2V1, 4, 2, DPU_FORMAT_FLAG_YUV, - DPU_FETCH_LINEAR, 2), + false, CHROMA_H2V1, 4, 2, DPU_FORMAT_FLAG_YUV, + MDP_FETCH_LINEAR, 2), INTERLEAVED_YUV_FMT(YUYV, - 0, COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, + 0, BPC8, BPC8, BPC8, C0_G_Y, C1_B_Cb, C0_G_Y, C2_R_Cr, - false, DPU_CHROMA_H2V1, 4, 2, DPU_FORMAT_FLAG_YUV, - DPU_FETCH_LINEAR, 2), + false, CHROMA_H2V1, 4, 2, DPU_FORMAT_FLAG_YUV, + MDP_FETCH_LINEAR, 2), INTERLEAVED_YUV_FMT(YVYU, - 0, COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, + 0, BPC8, BPC8, BPC8, C0_G_Y, C2_R_Cr, C0_G_Y, C1_B_Cb, - false, DPU_CHROMA_H2V1, 4, 2, DPU_FORMAT_FLAG_YUV, - DPU_FETCH_LINEAR, 2), + false, CHROMA_H2V1, 4, 2, DPU_FORMAT_FLAG_YUV, + MDP_FETCH_LINEAR, 2), PLANAR_YUV_FMT(YUV420, - 0, COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, + 0, BPC8, BPC8, BPC8, C2_R_Cr, C1_B_Cb, C0_G_Y, - false, DPU_CHROMA_420, 1, DPU_FORMAT_FLAG_YUV, - DPU_FETCH_LINEAR, 3), + false, CHROMA_420, 1, DPU_FORMAT_FLAG_YUV, + MDP_FETCH_LINEAR, 3), PLANAR_YUV_FMT(YVU420, - 0, COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, + 0, BPC8, BPC8, BPC8, C1_B_Cb, C2_R_Cr, C0_G_Y, - false, DPU_CHROMA_420, 1, DPU_FORMAT_FLAG_YUV, - DPU_FETCH_LINEAR, 3), + false, CHROMA_420, 1, DPU_FORMAT_FLAG_YUV, + MDP_FETCH_LINEAR, 3), }; /* @@ -485,88 +485,88 @@ static const struct dpu_format dpu_format_map[] = { */ static const struct dpu_format dpu_format_map_ubwc[] = { INTERLEAVED_RGB_FMT_TILED(BGR565, - 0, COLOR_5BIT, COLOR_6BIT, COLOR_5BIT, + 0, BPC5, BPC6, BPC5, C2_R_Cr, C0_G_Y, C1_B_Cb, 0, 3, false, 2, DPU_FORMAT_FLAG_COMPRESSED, - DPU_FETCH_UBWC, 2, DPU_TILE_HEIGHT_UBWC), + MDP_FETCH_UBWC, 2, DPU_TILE_HEIGHT_UBWC), INTERLEAVED_RGB_FMT_TILED(ABGR8888, - COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, + BPC8A, BPC8, BPC8, BPC8, C2_R_Cr, C0_G_Y, C1_B_Cb, C3_ALPHA, 4, true, 4, DPU_FORMAT_FLAG_COMPRESSED, - DPU_FETCH_UBWC, 2, DPU_TILE_HEIGHT_UBWC), + MDP_FETCH_UBWC, 2, DPU_TILE_HEIGHT_UBWC), /* ARGB8888 and ABGR8888 purposely have the same color * ordering. The hardware only supports ABGR8888 UBWC * natively. */ INTERLEAVED_RGB_FMT_TILED(ARGB8888, - COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, + BPC8A, BPC8, BPC8, BPC8, C2_R_Cr, C0_G_Y, C1_B_Cb, C3_ALPHA, 4, true, 4, DPU_FORMAT_FLAG_COMPRESSED, - DPU_FETCH_UBWC, 2, DPU_TILE_HEIGHT_UBWC), + MDP_FETCH_UBWC, 2, DPU_TILE_HEIGHT_UBWC), INTERLEAVED_RGB_FMT_TILED(XBGR8888, - COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, + BPC8A, BPC8, BPC8, BPC8, C2_R_Cr, C0_G_Y, C1_B_Cb, C3_ALPHA, 4, false, 4, DPU_FORMAT_FLAG_COMPRESSED, - DPU_FETCH_UBWC, 2, DPU_TILE_HEIGHT_UBWC), + MDP_FETCH_UBWC, 2, DPU_TILE_HEIGHT_UBWC), INTERLEAVED_RGB_FMT_TILED(XRGB8888, - COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, + BPC8A, BPC8, BPC8, BPC8, C2_R_Cr, C0_G_Y, C1_B_Cb, C3_ALPHA, 4, false, 4, DPU_FORMAT_FLAG_COMPRESSED, - DPU_FETCH_UBWC, 2, DPU_TILE_HEIGHT_UBWC), + MDP_FETCH_UBWC, 2, DPU_TILE_HEIGHT_UBWC), INTERLEAVED_RGB_FMT_TILED(ABGR2101010, - COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, + BPC8A, BPC8, BPC8, BPC8, C2_R_Cr, C0_G_Y, C1_B_Cb, C3_ALPHA, 4, true, 4, DPU_FORMAT_FLAG_DX | DPU_FORMAT_FLAG_COMPRESSED, - DPU_FETCH_UBWC, 2, DPU_TILE_HEIGHT_UBWC), + MDP_FETCH_UBWC, 2, DPU_TILE_HEIGHT_UBWC), INTERLEAVED_RGB_FMT_TILED(XBGR2101010, - COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, + BPC8A, BPC8, BPC8, BPC8, C2_R_Cr, C0_G_Y, C1_B_Cb, C3_ALPHA, 4, true, 4, DPU_FORMAT_FLAG_DX | DPU_FORMAT_FLAG_COMPRESSED, - DPU_FETCH_UBWC, 2, DPU_TILE_HEIGHT_UBWC), + MDP_FETCH_UBWC, 2, DPU_TILE_HEIGHT_UBWC), INTERLEAVED_RGB_FMT_TILED(XRGB2101010, - COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, + BPC8A, BPC8, BPC8, BPC8, C2_R_Cr, C0_G_Y, C1_B_Cb, C3_ALPHA, 4, true, 4, DPU_FORMAT_FLAG_DX | DPU_FORMAT_FLAG_COMPRESSED, - DPU_FETCH_UBWC, 2, DPU_TILE_HEIGHT_UBWC), + MDP_FETCH_UBWC, 2, DPU_TILE_HEIGHT_UBWC), /* XRGB2101010 and ARGB2101010 purposely have the same color * ordering. The hardware only supports ARGB2101010 UBWC * natively. */ INTERLEAVED_RGB_FMT_TILED(ARGB2101010, - COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, + BPC8A, BPC8, BPC8, BPC8, C2_R_Cr, C0_G_Y, C1_B_Cb, C3_ALPHA, 4, true, 4, DPU_FORMAT_FLAG_DX | DPU_FORMAT_FLAG_COMPRESSED, - DPU_FETCH_UBWC, 2, DPU_TILE_HEIGHT_UBWC), + MDP_FETCH_UBWC, 2, DPU_TILE_HEIGHT_UBWC), PSEUDO_YUV_FMT_TILED(NV12, - 0, COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, + 0, BPC8, BPC8, BPC8, C1_B_Cb, C2_R_Cr, - DPU_CHROMA_420, DPU_FORMAT_FLAG_YUV | + CHROMA_420, DPU_FORMAT_FLAG_YUV | DPU_FORMAT_FLAG_COMPRESSED, - DPU_FETCH_UBWC, 4, DPU_TILE_HEIGHT_NV12), + MDP_FETCH_UBWC, 4, DPU_TILE_HEIGHT_NV12), PSEUDO_YUV_FMT_TILED(P010, - 0, COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, + 0, BPC8, BPC8, BPC8, C1_B_Cb, C2_R_Cr, - DPU_CHROMA_420, DPU_FORMAT_FLAG_DX | + CHROMA_420, DPU_FORMAT_FLAG_DX | DPU_FORMAT_FLAG_YUV | DPU_FORMAT_FLAG_COMPRESSED, - DPU_FETCH_UBWC, 4, DPU_TILE_HEIGHT_UBWC), + MDP_FETCH_UBWC, 4, DPU_TILE_HEIGHT_UBWC), }; /* _dpu_get_v_h_subsample_rate - Get subsample rates for all formats we support * Note: Not using the drm_format_*_subsampling since we have formats */ static void _dpu_get_v_h_subsample_rate( - enum dpu_chroma_samp_type chroma_sample, + enum mdp_chroma_samp_type chroma_sample, uint32_t *v_sample, uint32_t *h_sample) { @@ -574,15 +574,15 @@ static void _dpu_get_v_h_subsample_rate( return; switch (chroma_sample) { - case DPU_CHROMA_H2V1: + case CHROMA_H2V1: *v_sample = 1; *h_sample = 2; break; - case DPU_CHROMA_H1V2: + case CHROMA_H1V2: *v_sample = 2; *h_sample = 1; break; - case DPU_CHROMA_420: + case CHROMA_420: *v_sample = 2; *h_sample = 2; break; @@ -724,7 +724,7 @@ static int _dpu_format_get_plane_sizes_linear( layout->num_planes = fmt->num_planes; /* Due to memset above, only need to set planes of interest */ - if (fmt->fetch_planes == DPU_PLANE_INTERLEAVED) { + if (fmt->fetch_planes == MDP_PLANE_INTERLEAVED) { layout->num_planes = 1; layout->plane_size[0] = width * height * layout->format->bpp; layout->plane_pitch[0] = width * layout->format->bpp; @@ -751,7 +751,7 @@ static int _dpu_format_get_plane_sizes_linear( layout->plane_size[1] = layout->plane_pitch[1] * (height / v_subsample); - if (fmt->fetch_planes == DPU_PLANE_PSEUDO_PLANAR) { + if (fmt->fetch_planes == MDP_PLANE_PSEUDO_PLANAR) { layout->num_planes = 2; layout->plane_size[1] *= 2; layout->plane_pitch[1] *= 2; diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c index e8b8908d3e12..06c6001b0626 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c @@ -204,9 +204,9 @@ static void dpu_hw_intf_setup_timing_engine(struct dpu_hw_intf *ctx, (0x21 << 8)); else /* Interface treats all the pixel data in RGB888 format */ - panel_format = (COLOR_8BIT | - (COLOR_8BIT << 2) | - (COLOR_8BIT << 4) | + panel_format = (BPC8 | + (BPC8 << 2) | + (BPC8 << 4) | (0x21 << 8)); DPU_REG_WRITE(c, INTF_HSYNC_CTL, hsync_ctl); diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h index d85157acfbf8..ebb3f9f493c3 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h @@ -9,6 +9,7 @@ #include #include "msm_drv.h" +#include "disp/mdp_common.xml.h" #define DPU_DBG_NAME "dpu" @@ -49,12 +50,12 @@ enum dpu_format_flags { (test_bit(DPU_FORMAT_FLAG_YUV_BIT, (X)->flag)) #define DPU_FORMAT_IS_DX(X) \ (test_bit(DPU_FORMAT_FLAG_DX_BIT, (X)->flag)) -#define DPU_FORMAT_IS_LINEAR(X) ((X)->fetch_mode == DPU_FETCH_LINEAR) +#define DPU_FORMAT_IS_LINEAR(X) ((X)->fetch_mode == MDP_FETCH_LINEAR) #define DPU_FORMAT_IS_TILE(X) \ - (((X)->fetch_mode == DPU_FETCH_UBWC) && \ + (((X)->fetch_mode == MDP_FETCH_UBWC) && \ !test_bit(DPU_FORMAT_FLAG_COMPRESSED_BIT, (X)->flag)) #define DPU_FORMAT_IS_UBWC(X) \ - (((X)->fetch_mode == DPU_FETCH_UBWC) && \ + (((X)->fetch_mode == MDP_FETCH_UBWC) && \ test_bit(DPU_FORMAT_FLAG_COMPRESSED_BIT, (X)->flag)) #define DPU_BLEND_FG_ALPHA_FG_CONST (0 << 0) @@ -291,57 +292,6 @@ enum { C3_ALPHA = 3 }; -/** - * enum dpu_plane_type - defines how the color component pixel packing - * @DPU_PLANE_INTERLEAVED : Color components in single plane - * @DPU_PLANE_PLANAR : Color component in separate planes - * @DPU_PLANE_PSEUDO_PLANAR : Chroma components interleaved in separate plane - */ -enum dpu_plane_type { - DPU_PLANE_INTERLEAVED, - DPU_PLANE_PLANAR, - DPU_PLANE_PSEUDO_PLANAR, -}; - -/** - * enum dpu_chroma_samp_type - chroma sub-samplng type - * @DPU_CHROMA_RGB : No chroma subsampling - * @DPU_CHROMA_H2V1 : Chroma pixels are horizontally subsampled - * @DPU_CHROMA_H1V2 : Chroma pixels are vertically subsampled - * @DPU_CHROMA_420 : 420 subsampling - */ -enum dpu_chroma_samp_type { - DPU_CHROMA_RGB, - DPU_CHROMA_H2V1, - DPU_CHROMA_H1V2, - DPU_CHROMA_420 -}; - -/** - * dpu_fetch_type - Defines How DPU HW fetches data - * @DPU_FETCH_LINEAR : fetch is line by line - * @DPU_FETCH_TILE : fetches data in Z order from a tile - * @DPU_FETCH_UBWC : fetch and decompress data - */ -enum dpu_fetch_type { - DPU_FETCH_LINEAR, - DPU_FETCH_TILE, - DPU_FETCH_UBWC -}; - -/** - * Value of enum chosen to fit the number of bits - * expected by the HW programming. - */ -enum { - COLOR_ALPHA_1BIT = 0, - COLOR_ALPHA_4BIT = 1, - COLOR_4BIT = 0, - COLOR_5BIT = 1, /* No 5-bit Alpha */ - COLOR_6BIT = 2, /* 6-Bit Alpha also = 2 */ - COLOR_8BIT = 3, /* 8-Bit Alpha also = 3 */ -}; - /** * enum dpu_3d_blend_mode * Desribes how the 3d data is blended @@ -381,17 +331,17 @@ enum dpu_3d_blend_mode { */ struct dpu_format { struct msm_format base; - enum dpu_plane_type fetch_planes; + enum mdp_fetch_type fetch_planes; u8 element[DPU_MAX_PLANES]; u8 bits[DPU_MAX_PLANES]; - enum dpu_chroma_samp_type chroma_sample; + enum mdp_chroma_samp_type chroma_sample; u8 unpack_align_msb; u8 unpack_tight; u8 unpack_count; u8 bpp; u8 alpha_enable; u8 num_planes; - enum dpu_fetch_type fetch_mode; + enum mdp_fetch_mode fetch_mode; DECLARE_BITMAP(flag, DPU_FORMAT_FLAG_BIT_MAX); u16 tile_width; u16 tile_height; diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c index 8e3c65989c49..d6530b11fc82 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c @@ -240,10 +240,10 @@ static void dpu_hw_sspp_setup_format(struct dpu_sw_pipe *pipe, chroma_samp = fmt->chroma_sample; if (flags & DPU_SSPP_SOURCE_ROTATED_90) { - if (chroma_samp == DPU_CHROMA_H2V1) - chroma_samp = DPU_CHROMA_H1V2; - else if (chroma_samp == DPU_CHROMA_H1V2) - chroma_samp = DPU_CHROMA_H2V1; + if (chroma_samp == CHROMA_H2V1) + chroma_samp = CHROMA_H1V2; + else if (chroma_samp == CHROMA_H1V2) + chroma_samp = CHROMA_H2V1; } src_format = (chroma_samp << 23) | (fmt->fetch_planes << 19) | @@ -253,7 +253,7 @@ static void dpu_hw_sspp_setup_format(struct dpu_sw_pipe *pipe, if (flags & DPU_SSPP_ROT_90) src_format |= BIT(11); /* ROT90 */ - if (fmt->alpha_enable && fmt->fetch_planes == DPU_PLANE_INTERLEAVED) + if (fmt->alpha_enable && fmt->fetch_planes == MDP_PLANE_INTERLEAVED) src_format |= BIT(8); /* SRCC3_EN */ if (flags & DPU_SSPP_SOLID_FILL) @@ -266,7 +266,7 @@ static void dpu_hw_sspp_setup_format(struct dpu_sw_pipe *pipe, (fmt->unpack_align_msb << 18) | ((fmt->bpp - 1) << 9); - if (fmt->fetch_mode != DPU_FETCH_LINEAR) { + if (fmt->fetch_mode != MDP_FETCH_LINEAR) { if (DPU_FORMAT_IS_UBWC(fmt)) opmode |= MDSS_MDP_OP_BWC_EN; src_format |= (fmt->fetch_mode & 3) << 30; /*FRAME_FORMAT */ diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c index 1e0da38c6f2a..17d0b4bcdc89 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c @@ -215,8 +215,8 @@ static int _dpu_plane_calc_fill_level(struct drm_plane *plane, /* FIXME: in multirect case account for the src_width of all the planes */ - if (fmt->fetch_planes == DPU_PLANE_PSEUDO_PLANAR) { - if (fmt->chroma_sample == DPU_CHROMA_420) { + if (fmt->fetch_planes == MDP_PLANE_PSEUDO_PLANAR) { + if (fmt->chroma_sample == CHROMA_420) { /* NV12 */ total_fl = (fixed_buff_size / 2) / ((src_width + 32) * fmt->bpp);