* [PATCH v2 0/2] arm64: dts: qcom: sm8[56]50: correct MDSS interconnects
@ 2024-10-26 17:59 Dmitry Baryshkov
2024-10-26 17:59 ` [PATCH v2 1/2] arm64: dts: qcom: sm8550: " Dmitry Baryshkov
` (2 more replies)
0 siblings, 3 replies; 6+ messages in thread
From: Dmitry Baryshkov @ 2024-10-26 17:59 UTC (permalink / raw)
To: Bjorn Andersson, Konrad Dybcio, Rob Herring, Krzysztof Kozlowski,
Conor Dooley, Neil Armstrong
Cc: Abhinav Kumar, Rob Clark, Marijn Suijten, linux-arm-msm,
devicetree, linux-kernel, freedreno, stable
Both SM8550 and SM8650 misuse mdp1-mem interconnect path for the
LLCC->EBI path, while it should only be used for the MDP->EBI paths.
This kind of misuse can result in bandwidth underflows, possibly
degrading picture quality as the required memory bandwidth is divided
between all mdpN-mem paths (and LLCC-EBI should not be a part of such
division).
Drop mdp1-mem paths and use MDP-EBI path directly.
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
Changes in v2:
- Fixed type in 'degrading' in the commit messages (Konrad)
- Link to v1: https://lore.kernel.org/r/20241008-fix-sm8x50-mdp-icc-v1-0-77ffd361b8de@linaro.org
---
Dmitry Baryshkov (2):
arm64: dts: qcom: sm8550: correct MDSS interconnects
arm64: dts: qcom: sm8650: correct MDSS interconnects
arch/arm64/boot/dts/qcom/sm8550.dtsi | 5 ++---
arch/arm64/boot/dts/qcom/sm8650.dtsi | 5 +----
2 files changed, 3 insertions(+), 7 deletions(-)
---
base-commit: 7f773fd61baa9b136faa5c4e6555aa64c758d07c
change-id: 20241006-fix-sm8x50-mdp-icc-138afd0e083a
Best regards,
--
Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
^ permalink raw reply [flat|nested] 6+ messages in thread
* [PATCH v2 1/2] arm64: dts: qcom: sm8550: correct MDSS interconnects
2024-10-26 17:59 [PATCH v2 0/2] arm64: dts: qcom: sm8[56]50: correct MDSS interconnects Dmitry Baryshkov
@ 2024-10-26 17:59 ` Dmitry Baryshkov
2024-12-02 14:39 ` Konrad Dybcio
2024-10-26 17:59 ` [PATCH v2 2/2] arm64: dts: qcom: sm8650: " Dmitry Baryshkov
2024-12-27 4:15 ` [PATCH v2 0/2] arm64: dts: qcom: sm8[56]50: " Bjorn Andersson
2 siblings, 1 reply; 6+ messages in thread
From: Dmitry Baryshkov @ 2024-10-26 17:59 UTC (permalink / raw)
To: Bjorn Andersson, Konrad Dybcio, Rob Herring, Krzysztof Kozlowski,
Conor Dooley, Neil Armstrong
Cc: Abhinav Kumar, Rob Clark, Marijn Suijten, linux-arm-msm,
devicetree, linux-kernel, freedreno, stable
SM8550 lists two interconnects for the display subsystem, mdp0-mem
(between MDP and LLCC) and mdp1-mem (between LLCC and EBI, memory).
The second interconnect is a misuse. mdpN-mem paths should be used for
several outboud MDP interconnects rather than the path between LLCC and
memory. This kind of misuse can result in bandwidth underflows, possibly
degrading picture quality as the required memory bandwidth is divided
between all mdpN-mem paths (and LLCC-EBI should not be a part of such
division).
Drop the second path and use direct MDP-EBI path for mdp0-mem until we
support separate MDP-LLCC and LLCC-EBI paths.
Fixes: d7da51db5b81 ("arm64: dts: qcom: sm8550: add display hardware devices")
Cc: stable@kernel.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
arch/arm64/boot/dts/qcom/sm8550.dtsi | 5 ++---
1 file changed, 2 insertions(+), 3 deletions(-)
diff --git a/arch/arm64/boot/dts/qcom/sm8550.dtsi b/arch/arm64/boot/dts/qcom/sm8550.dtsi
index 9dc0ee3eb98f..cca10f9faa14 100644
--- a/arch/arm64/boot/dts/qcom/sm8550.dtsi
+++ b/arch/arm64/boot/dts/qcom/sm8550.dtsi
@@ -2871,9 +2871,8 @@ mdss: display-subsystem@ae00000 {
power-domains = <&dispcc MDSS_GDSC>;
- interconnects = <&mmss_noc MASTER_MDP 0 &gem_noc SLAVE_LLCC 0>,
- <&mc_virt MASTER_LLCC 0 &mc_virt SLAVE_EBI1 0>;
- interconnect-names = "mdp0-mem", "mdp1-mem";
+ interconnects = <&mmss_noc MASTER_MDP 0 &mc_virt SLAVE_EBI1 0>;
+ interconnect-names = "mdp0-mem";
iommus = <&apps_smmu 0x1c00 0x2>;
--
2.39.5
^ permalink raw reply related [flat|nested] 6+ messages in thread
* [PATCH v2 2/2] arm64: dts: qcom: sm8650: correct MDSS interconnects
2024-10-26 17:59 [PATCH v2 0/2] arm64: dts: qcom: sm8[56]50: correct MDSS interconnects Dmitry Baryshkov
2024-10-26 17:59 ` [PATCH v2 1/2] arm64: dts: qcom: sm8550: " Dmitry Baryshkov
@ 2024-10-26 17:59 ` Dmitry Baryshkov
2024-12-30 9:42 ` Neil Armstrong
2024-12-27 4:15 ` [PATCH v2 0/2] arm64: dts: qcom: sm8[56]50: " Bjorn Andersson
2 siblings, 1 reply; 6+ messages in thread
From: Dmitry Baryshkov @ 2024-10-26 17:59 UTC (permalink / raw)
To: Bjorn Andersson, Konrad Dybcio, Rob Herring, Krzysztof Kozlowski,
Conor Dooley, Neil Armstrong
Cc: Abhinav Kumar, Rob Clark, Marijn Suijten, linux-arm-msm,
devicetree, linux-kernel, freedreno, stable
SM8650 lists two interconnects for the display subsystem, mdp0-mem
(between MDP and LLCC) and mdp1-mem (between LLCC and EBI, memory).
The second interconnect is a misuse. mdpN-mem paths should be used for
several outboud MDP interconnects rather than the path between LLCC and
memory. This kind of misuse can result in bandwidth underflows, possibly
degrading picture quality as the required memory bandwidth is divided
between all mdpN-mem paths (and LLCC-EBI should not be a part of such
division).
Drop the second path and use direct MDP-EBI path for mdp0-mem until we
support separate MDP-LLCC and LLCC-EBI paths.
Fixes: 10e024671295 ("arm64: dts: qcom: sm8650: add interconnect dependent device nodes")
Cc: stable@kernel.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
arch/arm64/boot/dts/qcom/sm8650.dtsi | 5 +----
1 file changed, 1 insertion(+), 4 deletions(-)
diff --git a/arch/arm64/boot/dts/qcom/sm8650.dtsi b/arch/arm64/boot/dts/qcom/sm8650.dtsi
index 173e092b15e2..4b54063c0e0c 100644
--- a/arch/arm64/boot/dts/qcom/sm8650.dtsi
+++ b/arch/arm64/boot/dts/qcom/sm8650.dtsi
@@ -3455,11 +3455,8 @@ mdss: display-subsystem@ae00000 {
resets = <&dispcc DISP_CC_MDSS_CORE_BCR>;
interconnects = <&mmss_noc MASTER_MDP QCOM_ICC_TAG_ALWAYS
- &gem_noc SLAVE_LLCC QCOM_ICC_TAG_ALWAYS>,
- <&mc_virt MASTER_LLCC QCOM_ICC_TAG_ALWAYS
&mc_virt SLAVE_EBI1 QCOM_ICC_TAG_ALWAYS>;
- interconnect-names = "mdp0-mem",
- "mdp1-mem";
+ interconnect-names = "mdp0-mem";
power-domains = <&dispcc MDSS_GDSC>;
--
2.39.5
^ permalink raw reply related [flat|nested] 6+ messages in thread
* Re: [PATCH v2 1/2] arm64: dts: qcom: sm8550: correct MDSS interconnects
2024-10-26 17:59 ` [PATCH v2 1/2] arm64: dts: qcom: sm8550: " Dmitry Baryshkov
@ 2024-12-02 14:39 ` Konrad Dybcio
0 siblings, 0 replies; 6+ messages in thread
From: Konrad Dybcio @ 2024-12-02 14:39 UTC (permalink / raw)
To: Dmitry Baryshkov, Bjorn Andersson, Konrad Dybcio, Rob Herring,
Krzysztof Kozlowski, Conor Dooley, Neil Armstrong
Cc: Abhinav Kumar, Rob Clark, Marijn Suijten, linux-arm-msm,
devicetree, linux-kernel, freedreno, stable
On 26.10.2024 7:59 PM, Dmitry Baryshkov wrote:
> SM8550 lists two interconnects for the display subsystem, mdp0-mem
> (between MDP and LLCC) and mdp1-mem (between LLCC and EBI, memory).
> The second interconnect is a misuse. mdpN-mem paths should be used for
> several outboud MDP interconnects rather than the path between LLCC and
> memory. This kind of misuse can result in bandwidth underflows, possibly
> degrading picture quality as the required memory bandwidth is divided
> between all mdpN-mem paths (and LLCC-EBI should not be a part of such
> division).
>
> Drop the second path and use direct MDP-EBI path for mdp0-mem until we
> support separate MDP-LLCC and LLCC-EBI paths.
>
> Fixes: d7da51db5b81 ("arm64: dts: qcom: sm8550: add display hardware devices")
> Cc: stable@kernel.org
> Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
> ---
Feel free to use QCOM_ICC_TAG_ALWAYS, for both patches:
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Konrad
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH v2 0/2] arm64: dts: qcom: sm8[56]50: correct MDSS interconnects
2024-10-26 17:59 [PATCH v2 0/2] arm64: dts: qcom: sm8[56]50: correct MDSS interconnects Dmitry Baryshkov
2024-10-26 17:59 ` [PATCH v2 1/2] arm64: dts: qcom: sm8550: " Dmitry Baryshkov
2024-10-26 17:59 ` [PATCH v2 2/2] arm64: dts: qcom: sm8650: " Dmitry Baryshkov
@ 2024-12-27 4:15 ` Bjorn Andersson
2 siblings, 0 replies; 6+ messages in thread
From: Bjorn Andersson @ 2024-12-27 4:15 UTC (permalink / raw)
To: Konrad Dybcio, Rob Herring, Krzysztof Kozlowski, Conor Dooley,
Neil Armstrong, Dmitry Baryshkov
Cc: Abhinav Kumar, Rob Clark, Marijn Suijten, linux-arm-msm,
devicetree, linux-kernel, freedreno, stable
On Sat, 26 Oct 2024 20:59:39 +0300, Dmitry Baryshkov wrote:
> Both SM8550 and SM8650 misuse mdp1-mem interconnect path for the
> LLCC->EBI path, while it should only be used for the MDP->EBI paths.
>
> This kind of misuse can result in bandwidth underflows, possibly
> degrading picture quality as the required memory bandwidth is divided
> between all mdpN-mem paths (and LLCC-EBI should not be a part of such
> division).
>
> [...]
Applied, thanks!
[1/2] arm64: dts: qcom: sm8550: correct MDSS interconnects
commit: b8591df49cde459e3b84cdc0517d7bf92053d244
[2/2] arm64: dts: qcom: sm8650: correct MDSS interconnects
commit: 9fa33cbca3d2842f1f47ed4e5f6574e611dae32b
Best regards,
--
Bjorn Andersson <andersson@kernel.org>
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH v2 2/2] arm64: dts: qcom: sm8650: correct MDSS interconnects
2024-10-26 17:59 ` [PATCH v2 2/2] arm64: dts: qcom: sm8650: " Dmitry Baryshkov
@ 2024-12-30 9:42 ` Neil Armstrong
0 siblings, 0 replies; 6+ messages in thread
From: Neil Armstrong @ 2024-12-30 9:42 UTC (permalink / raw)
To: Dmitry Baryshkov, Bjorn Andersson, Konrad Dybcio, Rob Herring,
Krzysztof Kozlowski, Conor Dooley
Cc: Abhinav Kumar, Rob Clark, Marijn Suijten, linux-arm-msm,
devicetree, linux-kernel, freedreno, stable
On 26/10/2024 19:59, Dmitry Baryshkov wrote:
> SM8650 lists two interconnects for the display subsystem, mdp0-mem
> (between MDP and LLCC) and mdp1-mem (between LLCC and EBI, memory).
> The second interconnect is a misuse. mdpN-mem paths should be used for
> several outboud MDP interconnects rather than the path between LLCC and
> memory. This kind of misuse can result in bandwidth underflows, possibly
> degrading picture quality as the required memory bandwidth is divided
> between all mdpN-mem paths (and LLCC-EBI should not be a part of such
> division).
>
> Drop the second path and use direct MDP-EBI path for mdp0-mem until we
> support separate MDP-LLCC and LLCC-EBI paths.
>
> Fixes: 10e024671295 ("arm64: dts: qcom: sm8650: add interconnect dependent device nodes")
> Cc: stable@kernel.org
> Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
> ---
> arch/arm64/boot/dts/qcom/sm8650.dtsi | 5 +----
> 1 file changed, 1 insertion(+), 4 deletions(-)
>
> diff --git a/arch/arm64/boot/dts/qcom/sm8650.dtsi b/arch/arm64/boot/dts/qcom/sm8650.dtsi
> index 173e092b15e2..4b54063c0e0c 100644
> --- a/arch/arm64/boot/dts/qcom/sm8650.dtsi
> +++ b/arch/arm64/boot/dts/qcom/sm8650.dtsi
> @@ -3455,11 +3455,8 @@ mdss: display-subsystem@ae00000 {
> resets = <&dispcc DISP_CC_MDSS_CORE_BCR>;
>
> interconnects = <&mmss_noc MASTER_MDP QCOM_ICC_TAG_ALWAYS
> - &gem_noc SLAVE_LLCC QCOM_ICC_TAG_ALWAYS>,
> - <&mc_virt MASTER_LLCC QCOM_ICC_TAG_ALWAYS
> &mc_virt SLAVE_EBI1 QCOM_ICC_TAG_ALWAYS>;
> - interconnect-names = "mdp0-mem",
> - "mdp1-mem";
> + interconnect-names = "mdp0-mem";
>
> power-domains = <&dispcc MDSS_GDSC>;
>
>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2024-12-30 9:42 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2024-10-26 17:59 [PATCH v2 0/2] arm64: dts: qcom: sm8[56]50: correct MDSS interconnects Dmitry Baryshkov
2024-10-26 17:59 ` [PATCH v2 1/2] arm64: dts: qcom: sm8550: " Dmitry Baryshkov
2024-12-02 14:39 ` Konrad Dybcio
2024-10-26 17:59 ` [PATCH v2 2/2] arm64: dts: qcom: sm8650: " Dmitry Baryshkov
2024-12-30 9:42 ` Neil Armstrong
2024-12-27 4:15 ` [PATCH v2 0/2] arm64: dts: qcom: sm8[56]50: " Bjorn Andersson
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox