| Message ID | 20230902-msm8226-ocmemcx_ahb_clk-remove-v1-1-8124dbde83b9@z3ntu.xyz (mailing list archive) |
|---|---|
| State | Awaiting Upstream, archived |
| Headers | show |
| Series | clk: qcom: mmcc-msm8974: remove ocmemcx_ahb_clk | expand |
On 2.09.2023 19:34, Luca Weiss wrote: > According to a commit in the 3.4 vendor kernel sources[0] the > ocmemcx_ahb_clk clock "is controlled by RPM and should not be touched by > APPS.". > > [0] https://git.codelinaro.org/clo/la/kernel/msm/-/commit/37df5f2d91b4d5768b37fcaacaeea958dd683ebc > > And indeed, when using MDSS+GPU+OCMEM on MSM8226 and not using > clk_ignore_unused, when Linux tries to disable the clock the device > crashes and reboots. > > And since there's also no evidence of this clock in msm8974 vendor > kernel sources, remove the clock for msm8226 and msm8974. Going over the downstream clock driver for 26 and 74 and comparing the registered clocks with mainline may be useful. Older clock drivers were likely written by hand and people doing that likely didn't have a great reference of which clocks should be skipped due to having been moved to firmware somewhere in the bringup process. Reviewed-by: Konrad Dybcio <konrad.dybcio@linaro.org> Konrad
On Sat, 02 Sep 2023 19:34:23 +0200, Luca Weiss wrote: > According to a commit in the 3.4 vendor kernel sources[0] the > ocmemcx_ahb_clk clock "is controlled by RPM and should not be touched by > APPS.". > > [0] https://git.codelinaro.org/clo/la/kernel/msm/-/commit/37df5f2d91b4d5768b37fcaacaeea958dd683ebc > > And indeed, when using MDSS+GPU+OCMEM on MSM8226 and not using > clk_ignore_unused, when Linux tries to disable the clock the device > crashes and reboots. > > [...] Applied, thanks! [1/1] clk: qcom: mmcc-msm8974: remove ocmemcx_ahb_clk commit: 471e2875f8904539985e62220afd6c88e70779fa Best regards,
diff --git a/drivers/clk/qcom/mmcc-msm8974.c b/drivers/clk/qcom/mmcc-msm8974.c index c37524d55d8e..3ce40c80666b 100644 --- a/drivers/clk/qcom/mmcc-msm8974.c +++ b/drivers/clk/qcom/mmcc-msm8974.c @@ -2171,22 +2171,6 @@ static struct clk_branch mmss_s0_axi_clk = { }, }; -static struct clk_branch ocmemcx_ahb_clk = { - .halt_reg = 0x405c, - .clkr = { - .enable_reg = 0x405c, - .enable_mask = BIT(0), - .hw.init = &(struct clk_init_data){ - .name = "ocmemcx_ahb_clk", - .parent_hws = (const struct clk_hw*[]){ - &mmss_ahb_clk_src.clkr.hw - }, - .num_parents = 1, - .ops = &clk_branch2_ops, - }, - }, -}; - static struct clk_branch ocmemcx_ocmemnoc_clk = { .halt_reg = 0x4058, .clkr = { @@ -2504,7 +2488,6 @@ static struct clk_regmap *mmcc_msm8226_clocks[] = { [MMSS_MMSSNOC_BTO_AHB_CLK] = &mmss_mmssnoc_bto_ahb_clk.clkr, [MMSS_MMSSNOC_AXI_CLK] = &mmss_mmssnoc_axi_clk.clkr, [MMSS_S0_AXI_CLK] = &mmss_s0_axi_clk.clkr, - [OCMEMCX_AHB_CLK] = &ocmemcx_ahb_clk.clkr, [OXILI_GFX3D_CLK] = &oxili_gfx3d_clk.clkr, [OXILICX_AHB_CLK] = &oxilicx_ahb_clk.clkr, [OXILICX_AXI_CLK] = &oxilicx_axi_clk.clkr, @@ -2661,7 +2644,6 @@ static struct clk_regmap *mmcc_msm8974_clocks[] = { [MMSS_MMSSNOC_BTO_AHB_CLK] = &mmss_mmssnoc_bto_ahb_clk.clkr, [MMSS_MMSSNOC_AXI_CLK] = &mmss_mmssnoc_axi_clk.clkr, [MMSS_S0_AXI_CLK] = &mmss_s0_axi_clk.clkr, - [OCMEMCX_AHB_CLK] = &ocmemcx_ahb_clk.clkr, [OCMEMCX_OCMEMNOC_CLK] = &ocmemcx_ocmemnoc_clk.clkr, [OCMEMNOC_CLK] = &ocmemnoc_clk.clkr, [OXILI_GFX3D_CLK] = &oxili_gfx3d_clk.clkr, diff --git a/include/dt-bindings/clock/qcom,mmcc-msm8974.h b/include/dt-bindings/clock/qcom,mmcc-msm8974.h index a62cb0629a7a..743ee60632eb 100644 --- a/include/dt-bindings/clock/qcom,mmcc-msm8974.h +++ b/include/dt-bindings/clock/qcom,mmcc-msm8974.h @@ -121,7 +121,6 @@ #define MMSS_MMSSNOC_BTO_AHB_CLK 112 #define MMSS_MMSSNOC_AXI_CLK 113 #define MMSS_S0_AXI_CLK 114 -#define OCMEMCX_AHB_CLK 115 #define OCMEMCX_OCMEMNOC_CLK 116 #define OXILI_OCMEMGX_CLK 117 #define OCMEMNOC_CLK 118
According to a commit in the 3.4 vendor kernel sources[0] the ocmemcx_ahb_clk clock "is controlled by RPM and should not be touched by APPS.". [0] https://git.codelinaro.org/clo/la/kernel/msm/-/commit/37df5f2d91b4d5768b37fcaacaeea958dd683ebc And indeed, when using MDSS+GPU+OCMEM on MSM8226 and not using clk_ignore_unused, when Linux tries to disable the clock the device crashes and reboots. And since there's also no evidence of this clock in msm8974 vendor kernel sources, remove the clock for msm8226 and msm8974. Fixes: d8b212014e69 ("clk: qcom: Add support for MSM8974's multimedia clock controller (MMCC)") Signed-off-by: Luca Weiss <luca@z3ntu.xyz> --- drivers/clk/qcom/mmcc-msm8974.c | 18 ------------------ include/dt-bindings/clock/qcom,mmcc-msm8974.h | 1 - 2 files changed, 19 deletions(-) --- base-commit: 7c2878be573282a9961c359b806ccf70afe1a6b6 change-id: 20230902-msm8226-ocmemcx_ahb_clk-remove-85dc043ccf12 Best regards,