From: Konrad Dybcio <konradybcio@kernel.org>
To: Bjorn Andersson <andersson@kernel.org>,
Konrad Dybcio <konradybcio@kernel.org>,
Rob Clark <robdclark@gmail.com>,
Abhinav Kumar <quic_abhinavk@quicinc.com>,
Dmitry Baryshkov <lumag@kernel.org>,
Akhil P Oommen <quic_akhilpo@quicinc.com>,
Sean Paul <sean@poorly.run>, David Airlie <airlied@gmail.com>,
Simona Vetter <simona@ffwll.ch>
Cc: Marijn Suijten <marijn.suijten@somainline.org>,
linux-kernel@vger.kernel.org, linux-arm-msm@vger.kernel.org,
dri-devel@lists.freedesktop.org,
freedreno@lists.freedesktop.org,
Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Subject: [PATCH RFT 10/14] drm/msm/a6xx: Stop tracking macrotile_mode (again)
Date: Thu, 08 May 2025 20:12:42 +0200 [thread overview]
Message-ID: <20250508-topic-ubwc_central-v1-10-035c4c5cbe50@oss.qualcomm.com> (raw)
In-Reply-To: <20250508-topic-ubwc_central-v1-0-035c4c5cbe50@oss.qualcomm.com>
From: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
SC8180X (A680) and SA8775P (A663) require a write to that register,
while other SKUs are fine with the default value. Don't overwrite it
needlessly, requiring the developer to read the value back from
hardware just to put it in the driver again, introducing much more room
for error.
Signed-off-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
---
drivers/gpu/drm/msm/adreno/a6xx_gpu.c | 19 +++++--------------
1 file changed, 5 insertions(+), 14 deletions(-)
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
index 60f89a2d851a5c383fc14cce4c483f630132a9a6..bee7e9685aa3ea282fb20ef479e4d243d28418f7 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
@@ -594,7 +594,6 @@ static int a6xx_calc_ubwc_config(struct adreno_gpu *gpu)
gpu->ubwc_config.min_acc_len = 0;
gpu->ubwc_config.ubwc_swizzle = 0x6;
- gpu->ubwc_config.macrotile_mode = 0;
gpu->ubwc_config.highest_bank_bit = 2;
if (adreno_is_a610(gpu)) {
@@ -616,13 +615,8 @@ static int a6xx_calc_ubwc_config(struct adreno_gpu *gpu)
if (adreno_is_a621(gpu))
gpu->ubwc_config.highest_bank_bit = 0;
- if (adreno_is_a623(gpu)) {
+ if (adreno_is_a623(gpu))
gpu->ubwc_config.highest_bank_bit = 3;
- gpu->ubwc_config.macrotile_mode = 1;
- }
-
- if (adreno_is_a680(gpu))
- gpu->ubwc_config.macrotile_mode = 1;
if (adreno_is_a650(gpu) ||
adreno_is_a660(gpu) ||
@@ -631,19 +625,15 @@ static int a6xx_calc_ubwc_config(struct adreno_gpu *gpu)
adreno_is_a740_family(gpu)) {
/* TODO: get ddr type from bootloader and use 2 for LPDDR4 */
gpu->ubwc_config.highest_bank_bit = 3;
- gpu->ubwc_config.macrotile_mode = 1;
}
if (adreno_is_a663(gpu)) {
gpu->ubwc_config.highest_bank_bit = 0;
- gpu->ubwc_config.macrotile_mode = 1;
gpu->ubwc_config.ubwc_swizzle = 0x4;
}
- if (adreno_is_7c3(gpu)) {
+ if (adreno_is_7c3(gpu))
gpu->ubwc_config.highest_bank_bit = 1;
- gpu->ubwc_config.macrotile_mode = 1;
- }
if (adreno_is_a702(gpu)) {
gpu->ubwc_config.highest_bank_bit = 1;
@@ -691,8 +681,9 @@ static void a6xx_set_ubwc_config(struct msm_gpu *gpu)
gpu_write(gpu, REG_A6XX_UCHE_MODE_CNTL,
adreno_gpu->ubwc_config.min_acc_len << 23 | hbb_lo << 21);
- gpu_write(gpu, REG_A6XX_RBBM_NC_MODE_CNTL,
- adreno_gpu->ubwc_config.macrotile_mode);
+ /* The reset value only needs altering in some cases */
+ if (adreno_is_a680(adreno_gpu) || adreno_is_a663(adreno_gpu))
+ gpu_write(gpu, REG_A6XX_RBBM_NC_MODE_CNTL, BIT(0));
}
static void a7xx_patch_pwrup_reglist(struct msm_gpu *gpu)
--
2.49.0
next prev parent reply other threads:[~2025-05-08 18:14 UTC|newest]
Thread overview: 43+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-05-08 18:12 [RFT PATCH 00/14] Add a single source of truth for UBWC configuration data Konrad Dybcio
2025-05-08 18:12 ` [PATCH RFT 01/14] soc: qcom: Add UBWC config provider Konrad Dybcio
2025-05-08 20:12 ` Connor Abbott
2025-05-09 12:28 ` Konrad Dybcio
2025-05-11 21:13 ` Dmitry Baryshkov
2025-05-12 8:57 ` Dmitry Baryshkov
2025-05-11 22:21 ` Bjorn Andersson
2025-05-08 18:12 ` [PATCH RFT 02/14] drm/msm: Use the central UBWC config database Konrad Dybcio
2025-05-08 18:12 ` [PATCH RFT 03/14] drm/msm/adreno: Offset the HBB value by 13 Konrad Dybcio
2025-05-08 19:03 ` Connor Abbott
2025-05-09 12:30 ` Konrad Dybcio
2025-05-08 18:12 ` [PATCH RFT 04/14] drm/msm/a6xx: Get a handle to the common UBWC config Konrad Dybcio
2025-05-08 18:41 ` Rob Clark
2025-05-09 12:31 ` Konrad Dybcio
2025-05-09 13:52 ` Rob Clark
2025-05-09 17:00 ` Konrad Dybcio
2025-05-13 22:06 ` Rob Clark
2025-05-13 22:11 ` Konrad Dybcio
2025-05-08 18:12 ` [PATCH RFT 05/14] drm/msm/a6xx: Resolve the meaning of AMSBC Konrad Dybcio
2025-05-08 19:16 ` Connor Abbott
2025-05-09 12:31 ` Konrad Dybcio
2025-05-08 18:12 ` [PATCH RFT 06/14] drm/msm/a6xx: Simplify uavflagprd_inv detection Konrad Dybcio
2025-05-08 19:05 ` Connor Abbott
2025-05-09 12:36 ` Konrad Dybcio
2025-05-08 18:12 ` [PATCH RFT 07/14] drm/msm/a6xx: Resolve the meaning of UBWC_MODE Konrad Dybcio
2025-05-08 18:25 ` Connor Abbott
2025-05-09 12:37 ` Konrad Dybcio
2025-05-09 12:39 ` Konrad Dybcio
2025-05-08 18:12 ` [PATCH RFT 08/14] drm/msm/a6xx: Replace '2' with BIT(1) in level2_swizzling_dis calc Konrad Dybcio
2025-05-08 18:12 ` [PATCH RFT 09/14] drm/msm/a6xx: Resolve the meaning of rgb565_predicator Konrad Dybcio
2025-05-08 18:12 ` Konrad Dybcio [this message]
2025-05-08 18:33 ` [PATCH RFT 10/14] drm/msm/a6xx: Stop tracking macrotile_mode (again) Connor Abbott
2025-05-09 12:45 ` Konrad Dybcio
2025-05-09 14:52 ` Connor Abbott
2025-05-08 18:12 ` [PATCH RFT 11/14] drm/msm/a6xx: Simplify min_acc_len calculation Konrad Dybcio
2025-05-08 18:12 ` [PATCH RFT 12/14] drm/msm/adreno: Switch to the common UBWC config struct Konrad Dybcio
2025-05-08 18:12 ` [PATCH RFT 13/14] drm/msm/a6xx: Drop cfg->ubwc_swizzle override Konrad Dybcio
2025-05-08 19:26 ` Connor Abbott
2025-05-09 13:17 ` Konrad Dybcio
2025-05-09 13:37 ` Konrad Dybcio
2025-05-09 14:48 ` Connor Abbott
2025-05-09 15:52 ` Konrad Dybcio
2025-05-08 18:12 ` [PATCH RFT 14/14] drm/msm/a5xx: Use UBWC data from the common UBWC config struct Konrad Dybcio
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20250508-topic-ubwc_central-v1-10-035c4c5cbe50@oss.qualcomm.com \
--to=konradybcio@kernel.org \
--cc=airlied@gmail.com \
--cc=andersson@kernel.org \
--cc=dri-devel@lists.freedesktop.org \
--cc=freedreno@lists.freedesktop.org \
--cc=konrad.dybcio@oss.qualcomm.com \
--cc=linux-arm-msm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=lumag@kernel.org \
--cc=marijn.suijten@somainline.org \
--cc=quic_abhinavk@quicinc.com \
--cc=quic_akhilpo@quicinc.com \
--cc=robdclark@gmail.com \
--cc=sean@poorly.run \
--cc=simona@ffwll.ch \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®