| Message ID | 20250212-mst_qcs8300-v1-0-38a8aa08394b@quicinc.com |
|---|---|
| Headers | show |
| Series | Add MST support for qcs8300 platform | expand |
On 2025/2/12 20:26, Dmitry Baryshkov wrote: > On Wed, 12 Feb 2025 at 12:54, Krzysztof Kozlowski <krzk@kernel.org> wrote: >> >> On 12/02/2025 11:41, Dmitry Baryshkov wrote: >>> On Wed, Feb 12, 2025 at 03:12:24PM +0800, Yongxing Mou wrote: >>>> We need to enable mst for qcs8300, dp0 controller will support 2 streams >>>> output. So not reuse sm8650 dp controller driver and will add a new driver >>>> patch for qcs8300 mst feature. Modify the corresponding dt-bingding file >>>> to compatible with the qcs8300-dp. >>> >>> NAK for a different reason: QCS8300 is still compatible with SM8650. >>> Enable features instead of randomly reshuffle compats. In this case, >>> enable MST for both architectures. >>> >> So the original patch was probably correct... > > I have no idea. I'd let Yongxing Mou to comment on this. It would be > nice instead of getting a lengthy explanation of obvious items to get > an answer to an actual question: is QCS8300 compatible with SM8650 or > not. In other words whether they can support the same number of MST > streams on each controller or not. > Hi, in hardware, the SM8650's DP controller supports 2 INTFs, while the QCS8300's DP0 controller supports 4 INTFs.In sst mode, only one INTF will be used, they are same, but for MST, sm8650 supports 2 streams while qcs8300 support 4 streams. Thanks.
On 2025/2/12 17:06, Krzysztof Kozlowski wrote: > On 12/02/2025 08:12, Yongxing Mou wrote: >> This series of patches introduces how to enable MST functionality on >> the qcs8300 platform. The qcs8300 platform uses dpu_8_4 hardware, which >> is the same as the sa8775p, but it only has one DPU. So it only has one >> DP0 controller, supporting 4-stream MST. This patch only enables >> 2-stream MST, using intf0 and intf3. The first and second patches are >> modifications to the correspond dt-bindings, third patch is the dp >> controller driver after not reuse sm8650, fourth patch is the qcs8300 >> dts modification which add the clk support for stream 1. >> >> Signed-off-by: Yongxing Mou <quic_yongmou@quicinc.com> >> --- >> This patch depends on following series: >> https://lore.kernel.org/all/20250120-mdssdt_qcs8300-v4-0-1687e7842125@quicinc.com/ >> https://lore.kernel.org/all/20250114-dts_qcs8300-v3-0-d114cc5e4af9@quicinc.com/ >> https://lore.kernel.org/all/20241202-dp_mst_bindings-v1-0-9a9a43b0624a@quicinc.com >> https://lore.kernel.org/all/20241205-dp_mst-v1-0-f8618d42a99a@quicinc.com/ > > Buggy patch cannot be the dependency. You need to fix the original patch > instead. You are not supposed to wait till knowingly incorrect patches > get merged so you can send a fix. You must respond to that patch > stopping its merging process. > > Best regards, > Krzysztof Thanks for the reminder. I hadn't noticed that before. We will wait for our dependencies to be ready before sending the new patchset.
On Wed, Feb 19, 2025 at 06:02:27PM +0800, Yongxing Mou wrote: > > > On 2025/2/12 17:06, Krzysztof Kozlowski wrote: > > On 12/02/2025 08:12, Yongxing Mou wrote: > > > This series of patches introduces how to enable MST functionality on > > > the qcs8300 platform. The qcs8300 platform uses dpu_8_4 hardware, which > > > is the same as the sa8775p, but it only has one DPU. So it only has one > > > DP0 controller, supporting 4-stream MST. This patch only enables > > > 2-stream MST, using intf0 and intf3. The first and second patches are > > > modifications to the correspond dt-bindings, third patch is the dp > > > controller driver after not reuse sm8650, fourth patch is the qcs8300 > > > dts modification which add the clk support for stream 1. > > > > > > Signed-off-by: Yongxing Mou <quic_yongmou@quicinc.com> > > > --- > > > This patch depends on following series: > > > https://lore.kernel.org/all/20250120-mdssdt_qcs8300-v4-0-1687e7842125@quicinc.com/ > > > https://lore.kernel.org/all/20250114-dts_qcs8300-v3-0-d114cc5e4af9@quicinc.com/ > > > https://lore.kernel.org/all/20241202-dp_mst_bindings-v1-0-9a9a43b0624a@quicinc.com > > > https://lore.kernel.org/all/20241205-dp_mst-v1-0-f8618d42a99a@quicinc.com/ > > > > Buggy patch cannot be the dependency. You need to fix the original patch > > instead. You are not supposed to wait till knowingly incorrect patches > > get merged so you can send a fix. You must respond to that patch > > stopping its merging process. > > > > Best regards, > > Krzysztof > Thanks for the reminder. I hadn't noticed that before. We will wait for our > dependencies to be ready before sending the new patchset. Hmm, no. Work with Abhinav and send the bindings (including all 4 stream clocks) and platform enablement. Then MST support can come up together with the rest of MST patchset.
On 19/02/2025 11:02, Yongxing Mou wrote: > > > On 2025/2/12 17:06, Krzysztof Kozlowski wrote: >> On 12/02/2025 08:12, Yongxing Mou wrote: >>> This series of patches introduces how to enable MST functionality on >>> the qcs8300 platform. The qcs8300 platform uses dpu_8_4 hardware, which >>> is the same as the sa8775p, but it only has one DPU. So it only has one >>> DP0 controller, supporting 4-stream MST. This patch only enables >>> 2-stream MST, using intf0 and intf3. The first and second patches are >>> modifications to the correspond dt-bindings, third patch is the dp >>> controller driver after not reuse sm8650, fourth patch is the qcs8300 >>> dts modification which add the clk support for stream 1. >>> >>> Signed-off-by: Yongxing Mou <quic_yongmou@quicinc.com> >>> --- >>> This patch depends on following series: >>> https://lore.kernel.org/all/20250120-mdssdt_qcs8300-v4-0-1687e7842125@quicinc.com/ >>> https://lore.kernel.org/all/20250114-dts_qcs8300-v3-0-d114cc5e4af9@quicinc.com/ >>> https://lore.kernel.org/all/20241202-dp_mst_bindings-v1-0-9a9a43b0624a@quicinc.com >>> https://lore.kernel.org/all/20241205-dp_mst-v1-0-f8618d42a99a@quicinc.com/ >> >> Buggy patch cannot be the dependency. You need to fix the original patch >> instead. You are not supposed to wait till knowingly incorrect patches >> get merged so you can send a fix. You must respond to that patch >> stopping its merging process. >> >> Best regards, >> Krzysztof > Thanks for the reminder. I hadn't noticed that before. We will wait for > our dependencies to be ready before sending the new patchset. So this means you intend the knowingly incorrect patch to be applied? If so, all this should be just NAKed. Best regards, Krzysztof
On Wed, Feb 19, 2025 at 05:56:14PM +0800, Yongxing Mou wrote: > > > On 2025/2/12 20:26, Dmitry Baryshkov wrote: > > On Wed, 12 Feb 2025 at 12:54, Krzysztof Kozlowski <krzk@kernel.org> wrote: > > > > > > On 12/02/2025 11:41, Dmitry Baryshkov wrote: > > > > On Wed, Feb 12, 2025 at 03:12:24PM +0800, Yongxing Mou wrote: > > > > > We need to enable mst for qcs8300, dp0 controller will support 2 streams > > > > > output. So not reuse sm8650 dp controller driver and will add a new driver > > > > > patch for qcs8300 mst feature. Modify the corresponding dt-bingding file > > > > > to compatible with the qcs8300-dp. > > > > > > > > NAK for a different reason: QCS8300 is still compatible with SM8650. > > > > Enable features instead of randomly reshuffle compats. In this case, > > > > enable MST for both architectures. > > > > > > > So the original patch was probably correct... > > > > I have no idea. I'd let Yongxing Mou to comment on this. It would be > > nice instead of getting a lengthy explanation of obvious items to get > > an answer to an actual question: is QCS8300 compatible with SM8650 or > > not. In other words whether they can support the same number of MST > > streams on each controller or not. > > > Hi, in hardware, the SM8650's DP controller supports 2 INTFs, while the > QCS8300's DP0 controller supports 4 INTFs.In sst mode, only one INTF will be > used, they are same, but for MST, sm8650 supports 2 streams while qcs8300 > support 4 streams. Thanks. So, they are not compatible. Please use separate compatible.
On 2025/2/19 20:07, Krzysztof Kozlowski wrote: > On 19/02/2025 11:02, Yongxing Mou wrote: >> >> >> On 2025/2/12 17:06, Krzysztof Kozlowski wrote: >>> On 12/02/2025 08:12, Yongxing Mou wrote: >>>> This series of patches introduces how to enable MST functionality on >>>> the qcs8300 platform. The qcs8300 platform uses dpu_8_4 hardware, which >>>> is the same as the sa8775p, but it only has one DPU. So it only has one >>>> DP0 controller, supporting 4-stream MST. This patch only enables >>>> 2-stream MST, using intf0 and intf3. The first and second patches are >>>> modifications to the correspond dt-bindings, third patch is the dp >>>> controller driver after not reuse sm8650, fourth patch is the qcs8300 >>>> dts modification which add the clk support for stream 1. >>>> >>>> Signed-off-by: Yongxing Mou <quic_yongmou@quicinc.com> >>>> --- >>>> This patch depends on following series: >>>> https://lore.kernel.org/all/20250120-mdssdt_qcs8300-v4-0-1687e7842125@quicinc.com/ >>>> https://lore.kernel.org/all/20250114-dts_qcs8300-v3-0-d114cc5e4af9@quicinc.com/ >>>> https://lore.kernel.org/all/20241202-dp_mst_bindings-v1-0-9a9a43b0624a@quicinc.com >>>> https://lore.kernel.org/all/20241205-dp_mst-v1-0-f8618d42a99a@quicinc.com/ >>> >>> Buggy patch cannot be the dependency. You need to fix the original patch >>> instead. You are not supposed to wait till knowingly incorrect patches >>> get merged so you can send a fix. You must respond to that patch >>> stopping its merging process. >>> >>> Best regards, >>> Krzysztof >> Thanks for the reminder. I hadn't noticed that before. We will wait for >> our dependencies to be ready before sending the new patchset. > > So this means you intend the knowingly incorrect patch to be applied? If > so, all this should be just NAKed. > > Best regards, > Krzysztof Hi, that is not my intention. I will correct the patch after i know it is incorrect. Thanks.
This series of patches introduces how to enable MST functionality on the qcs8300 platform. The qcs8300 platform uses dpu_8_4 hardware, which is the same as the sa8775p, but it only has one DPU. So it only has one DP0 controller, supporting 4-stream MST. This patch only enables 2-stream MST, using intf0 and intf3. The first and second patches are modifications to the correspond dt-bindings, third patch is the dp controller driver after not reuse sm8650, fourth patch is the qcs8300 dts modification which add the clk support for stream 1. Signed-off-by: Yongxing Mou <quic_yongmou@quicinc.com> --- This patch depends on following series: https://lore.kernel.org/all/20250120-mdssdt_qcs8300-v4-0-1687e7842125@quicinc.com/ https://lore.kernel.org/all/20250114-dts_qcs8300-v3-0-d114cc5e4af9@quicinc.com/ https://lore.kernel.org/all/20241202-dp_mst_bindings-v1-0-9a9a43b0624a@quicinc.com https://lore.kernel.org/all/20241205-dp_mst-v1-0-f8618d42a99a@quicinc.com/ --- Yongxing Mou (4): dt-bindings: display/msm: Redocument the dp-controller for QCS8300 dt-bindings: display/msm: Add stream 1 pixel clock for QCS8300 drm/msm/dp: Populate the max_streams for qcs8300 mst controller arm64: dts: qcom: qcs8300: Add support for stream 1 clk for DP MST .../devicetree/bindings/display/msm/dp-controller.yaml | 5 +---- .../devicetree/bindings/display/msm/qcom,qcs8300-mdss.yaml | 14 ++++++++------ arch/arm64/boot/dts/qcom/qcs8300.dtsi | 12 ++++++++---- drivers/gpu/drm/msm/dp/dp_display.c | 8 ++++++++ 4 files changed, 25 insertions(+), 14 deletions(-) --- base-commit: 7ba9bcc5090556c007d9a718d7176e097fe54f19 change-id: 20250211-mst_qcs8300-4c18a5179165 Best regards,