mbox series

[v2,0/8] MSM8998 clk cleanups and fixups

Message ID 20230622-topic-8998clk-v2-0-6222fbc2916b@linaro.org
Headers
Series MSM8998 clk cleanups and fixups |

Message

Konrad Dybcio July 3, 2023, 6:20 p.m. UTC 
  The MSM8998 clock controller drivers have some rough edges around whether
and how Linux should touch them, which this series tries to sand down
a bit.

MSM8998 maple seems not to explode, please give it a spin on your boards.

Signed-off-by: Konrad Dybcio <konrad.dybcio@linaro.org>
---
Changes in v2:
- Pick up tags (thanks a lot Jeffrey for testing)
- Drop patch "Don't poke at some BIMC GPU clocks"
- Link to v1: https://lore.kernel.org/r/20230622-topic-8998clk-v1-0-5b7a0d6e98b1@linaro.org

---
Konrad Dybcio (8):
      dt-bindings: clk: qcom,gcc-msm8998: Add missing GPU/MMSS GPLL0 legs
      dt-bindings: clock: qcom,mmcc: Add GPLL0_DIV for MSM8998
      clk: qcom: gcc-msm8998: Control MMSS and GPUSS GPLL0 outputs properly
      clk: qcom: mmcc-msm8998: Properly consume GPLL0 inputs
      clk: qcom: gpucc-msm8998: Use the correct GPLL0 leg with old DTs
      clk: qcom: gcc-msm8998: Don't check halt bit on some branch clks
      arm64: dts: qcom: msm8998: Use the correct GPLL0 leg for GPUCC
      arm64: dts: qcom: msm8998: Use the correct GPLL0_DIV leg for MMCC

 .../devicetree/bindings/clock/qcom,mmcc.yaml       |  2 +
 arch/arm64/boot/dts/qcom/msm8998.dtsi              |  8 ++-
 drivers/clk/qcom/gcc-msm8998.c                     | 64 +++++++++++++++++++++-
 drivers/clk/qcom/gpucc-msm8998.c                   |  2 +-
 drivers/clk/qcom/mmcc-msm8998.c                    | 35 +++---------
 include/dt-bindings/clock/qcom,gcc-msm8998.h       |  3 +
 6 files changed, 80 insertions(+), 34 deletions(-)
---
base-commit: 296d53d8f84ce50ffaee7d575487058c8d437335
change-id: 20230622-topic-8998clk-4317986f3008

Best regards,

Comments

Bjorn Andersson Aug. 14, 2023, 3:27 a.m. UTC | #1 
On Mon, 03 Jul 2023 20:20:04 +0200, Konrad Dybcio wrote:
> The MSM8998 clock controller drivers have some rough edges around whether
> and how Linux should touch them, which this series tries to sand down
> a bit.
> 
> MSM8998 maple seems not to explode, please give it a spin on your boards.
> 
> 
> [...]

Applied, thanks!

[7/8] arm64: dts: qcom: msm8998: Use the correct GPLL0 leg for GPUCC
      commit: 00ada6afea88006187b38bd96b6d1b7d8e3d90cb
[8/8] arm64: dts: qcom: msm8998: Use the correct GPLL0_DIV leg for MMCC
      commit: 63f4e4b447c50ba7e5fc3929644d2d152acb6117

Best regards,