The aim of this series is to get the dispcc and gpucc in a workable shape upstream for sc7180. I personally wasn't focusing on (and didn't test much) videocc but pulled it along for the ride. Most of the work in this series deals with the fact that the parenting info for these clock controllers was in a bad shape. It looks like it was half transitioned from the old way of doing things (relying on global names) to the new way of doing things (putting the linkage in the device tree). This should fully transition us. As part of this transition I update the sdm845.dtsi file to specify the info as per the new way of doing things. Although I've now put the linkage info in the sdm845.dtsi file, though, I haven't updated the sdm845 clock drivers in Linux so they still work via the global name matching. It's left as an exercise to the reader to update the sdm845 clock drivers in Linux. This series passes these things for me on linux-next (next-20200129) after picking the recent gcc fix I posted [1]: for f in \ Documentation/devicetree/bindings/clock/qcom,msm8998-gpucc.yaml \ Documentation/devicetree/bindings/clock/qcom,sc7180-dispcc.yaml \ Documentation/devicetree/bindings/clock/qcom,sc7180-gpucc.yaml \ Documentation/devicetree/bindings/clock/qcom,sc7180-videocc.yaml \ Documentation/devicetree/bindings/clock/qcom,sdm845-dispcc.yaml \ Documentation/devicetree/bindings/clock/qcom,sdm845-gpucc.yaml \ Documentation/devicetree/bindings/clock/qcom,sdm845-videocc.yaml; do \ ARCH=arm64 make dtbs_check DT_SCHEMA_FILES=$f; \ done I also tried this: # Delete broken yaml: rm Documentation/devicetree/bindings/pci/intel-gw-pcie.yaml ARCH=arm64 make dt_binding_check | grep 'clock/qcom' ...and that didn't seem to indicate problems. I also tried this (make sure you don't run w/ -j64 or diff is hard): # Delete broken yaml: rm Documentation/devicetree/bindings/pci/intel-gw-pcie.yaml git checkout beforeMyCode ARCH=arm64 make dt_binding_check > old.txt 2>&1 git checkout myCode ARCH=arm64 make dt_binding_check > new.txt 2>&1 diff old.txt new.txt ...and that didn't seem to indicate problems. I have confirmed that (with extra patches) the display/gpu come up on sc7180 and sdm845-cheza. You can find the top of my downstream tree at: https://crrev.com/c/2017976/4 I have confirmed that sdm845-cheza display / GPU come up atop next-20200129, which is what this series is posted against. From v2 to v3, this series has quite a few changes. Mostly it's: - Always split into multiple files (Stephen). - Use internal names, not purist names (Taniya). - I realized that I forgot to update the sc7180 video clock controller driver in v2. - A few other misc cleanups / fixes, see each patch for details. From v3 to v4 this just removes the "bindings/" from the schema ID and adds Rob's tag to two of the bindings patches. The third binding patch didn't get Rob's Ack because he suggested combining two files might be a good idea. I'm assuming he'll be OK with leaving them split since Stephen prefers it that way, but if I have to do a v5 I can do that. It feels like with this many patches there's very little chance I didn't do something stupid like make a tpyo or a paste-o paste-o, though I tried to cross-check as much as I could. I apologize in advance for the stupid things I did that I should have known better about. [1] https://lore.kernel.org/r/20200129152458.v2.1.I4452dc951d7556ede422835268742b25a18b356b@changeid Changes in v4: - (non-change): Didn't combine sdm845 & sc7180 gpucc as per Stephen. - Added Rob's review tag. - Fixed schema id to not have "bindings/" as per Rob. Changes in v3: - Add Matthias tag. - Added include file to description. - Added pointer to inlude file in description. - Added videocc include file. - Discovered / added new gcc input clock on sdm845. - Everyone but msm8998 now uses internal QC names. - Fixed typo grpahics => graphics - Newly discovered gcc_disp_gpll0_div_clk_src added. - Patch ("clk: qcom: Get rid of fallback...dispcc-sc7180") split out for v3. - Patch ("clk: qcom: Get rid of the test...dispcc-sc7180") split out for v3. - Patch ("clk: qcom: Get rid of the test...gpucc-sc7180") split out for v3. - Patch ("clk: qcom: Get rid of the test...videocc-sc7180") new for v3. - Patch ("clk: qcom: Use ARRAY_SIZE in dispcc-sc7180...") split out for v3. - Patch ("clk: qcom: Use ARRAY_SIZE in gpucc-sc7180...") split out for v3. - Patch ("clk: qcom: Use ARRAY_SIZE in videocc-sc7180...") new for v3. - Split bindings into 3 files. - Split sc7180 and sdm845 into two files. - Split videocc bindings into 2 files. - Switched names to internal QC names rather than logical ones. - Unlike in v2, use internal name instead of purist name. - Updated commit description. Changes in v2: - Added includes - Changed various parent names to match bindings / driver - Patch ("arm64: dts: qcom: sdm845: Add...dispcc") new for v2. - Patch ("arm64: dts: qcom: sdm845: Add...gpucc") new for v2. - Patch ("arm64: dts: qcom: sdm845: Add...videocc") new for v2. - Patch ("clk: qcom: rcg2: Don't crash...") new for v2. - Patch ("dt-bindings: clock: Cleanup qcom,videocc") new for v2. - Patch ("dt-bindings: clock: Fix qcom,dispcc...") new for v2. - Patch ("dt-bindings: clock: Fix qcom,gpucc...") new for v2. Douglas Anderson (14): clk: qcom: rcg2: Don't crash if our parent can't be found; return an error dt-bindings: clock: Fix qcom,dispcc bindings for sdm845/sc7180 arm64: dts: qcom: sdm845: Add the missing clocks on the dispcc clk: qcom: Get rid of fallback global names for dispcc-sc7180 clk: qcom: Get rid of the test clock for dispcc-sc7180 clk: qcom: Use ARRAY_SIZE in dispcc-sc7180 for parent clocks dt-bindings: clock: Fix qcom,gpucc bindings for sdm845/sc7180/msm8998 arm64: dts: qcom: sdm845: Add missing clocks / fix names on the gpucc clk: qcom: Get rid of the test clock for gpucc-sc7180 clk: qcom: Use ARRAY_SIZE in gpucc-sc7180 for parent clocks dt-bindings: clock: Cleanup qcom,videocc bindings for sdm845/sc7180 clk: qcom: Get rid of the test clock for videocc-sc7180 clk: qcom: Use ARRAY_SIZE in videocc-sc7180 for parent clocks arm64: dts: qcom: sdm845: Add the missing clock on the videocc Taniya Das (1): arm64: dts: sc7180: Add clock controller nodes .../devicetree/bindings/clock/qcom,gpucc.yaml | 72 -------------- ...om,dispcc.yaml => qcom,msm8998-gpucc.yaml} | 33 +++---- .../bindings/clock/qcom,sc7180-dispcc.yaml | 84 ++++++++++++++++ .../bindings/clock/qcom,sc7180-gpucc.yaml | 72 ++++++++++++++ .../bindings/clock/qcom,sc7180-videocc.yaml | 63 ++++++++++++ .../bindings/clock/qcom,sdm845-dispcc.yaml | 99 +++++++++++++++++++ .../bindings/clock/qcom,sdm845-gpucc.yaml | 72 ++++++++++++++ ...,videocc.yaml => qcom,sdm845-videocc.yaml} | 27 ++--- arch/arm64/boot/dts/qcom/sc7180.dtsi | 47 +++++++++ arch/arm64/boot/dts/qcom/sdm845.dtsi | 28 +++++- drivers/clk/qcom/clk-rcg2.c | 3 + drivers/clk/qcom/dispcc-sc7180.c | 45 +++------ drivers/clk/qcom/gpucc-sc7180.c | 4 +- drivers/clk/qcom/videocc-sc7180.c | 4 +- 14 files changed, 513 insertions(+), 140 deletions(-) delete mode 100644 Documentation/devicetree/bindings/clock/qcom,gpucc.yaml rename Documentation/devicetree/bindings/clock/{qcom,dispcc.yaml => qcom,msm8998-gpucc.yaml} (51%) create mode 100644 Documentation/devicetree/bindings/clock/qcom,sc7180-dispcc.yaml create mode 100644 Documentation/devicetree/bindings/clock/qcom,sc7180-gpucc.yaml create mode 100644 Documentation/devicetree/bindings/clock/qcom,sc7180-videocc.yaml create mode 100644 Documentation/devicetree/bindings/clock/qcom,sdm845-dispcc.yaml create mode 100644 Documentation/devicetree/bindings/clock/qcom,sdm845-gpucc.yaml rename Documentation/devicetree/bindings/clock/{qcom,videocc.yaml => qcom,sdm845-videocc.yaml} (61%) -- 2.25.0.341.g760bfbb309-goog
When I got my clock parenting slightly wrong I ended up with a crash that looked like this: Unable to handle kernel NULL pointer dereference at virtual address 0000000000000000 ... pc : clk_hw_get_rate+0x14/0x44 ... Call trace: clk_hw_get_rate+0x14/0x44 _freq_tbl_determine_rate+0x94/0xfc clk_rcg2_determine_rate+0x2c/0x38 clk_core_determine_round_nolock+0x4c/0x88 clk_core_round_rate_nolock+0x6c/0xa8 clk_core_round_rate_nolock+0x9c/0xa8 clk_core_set_rate_nolock+0x70/0x180 clk_set_rate+0x3c/0x6c of_clk_set_defaults+0x254/0x360 platform_drv_probe+0x28/0xb0 really_probe+0x120/0x2dc driver_probe_device+0x64/0xfc device_driver_attach+0x4c/0x6c __driver_attach+0xac/0xc0 bus_for_each_dev+0x84/0xcc driver_attach+0x2c/0x38 bus_add_driver+0xfc/0x1d0 driver_register+0x64/0xf8 __platform_driver_register+0x4c/0x58 msm_drm_register+0x5c/0x60 ... It turned out that clk_hw_get_parent_by_index() was returning NULL and we weren't checking. Let's check it so that we don't crash. Fixes: ac269395cdd8 ("clk: qcom: Convert to clk_hw based provider APIs") Signed-off-by: Douglas Anderson <dianders@chromium.org> Reviewed-by: Matthias Kaehlcke <mka@chromium.org> --- I haven't gone back and tried to reproduce this same crash on older kernels, but I'll put the blame on commit ac269395cdd8 ("clk: qcom: Convert to clk_hw based provider APIs"). Before that if we got a NULL parent back it was fine and dandy since a NULL "struct clk" is valid to use but a NULL "struct clk_hw" is not. Changes in v4: None Changes in v3: - Add Matthias tag. Changes in v2: - Patch ("clk: qcom: rcg2: Don't crash...") new for v2. drivers/clk/qcom/clk-rcg2.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/clk/qcom/clk-rcg2.c b/drivers/clk/qcom/clk-rcg2.c index da045b200def..9098001ac805 100644 --- a/drivers/clk/qcom/clk-rcg2.c +++ b/drivers/clk/qcom/clk-rcg2.c @@ -218,6 +218,9 @@ static int _freq_tbl_determine_rate(struct clk_hw *hw, const struct freq_tbl *f, clk_flags = clk_hw_get_flags(hw); p = clk_hw_get_parent_by_index(hw, index); + if (!p) + return -EINVAL; + if (clk_flags & CLK_SET_RATE_PARENT) { rate = f->freq; if (f->pre_div) { -- 2.25.0.341.g760bfbb309-goog
The qcom,dispcc bindings had a few problems with them: 1. They didn't specify all the clocks that dispcc is a client of. Specifically on sc7180 there are two clocks from the DSI PHY and two from the DP PHY. On sdm845 there are actually two DSI PHYs (each of which has two clocks) and an extra clock from the gcc. These all need to be specified. 2. The sdm845.dtsi has existed for quite some time without specifying the clocks. The Linux driver was relying on global names to match things up. While we should transition things, it should be noted in the bindings. 3. The names used the bindings for "xo" and "gpll0" didn't match the names that QC used for these clocks internally and this was causing confusion / difficulty with their code generation tools. Switched to the internal names to simplify everyone's lives. It's not quite as clean in a purist sense but it should avoid headaches. This officially changes the binding, but that seems OK in this case. Also note that I updated the example. Fixes: 5d28e44ba630 ("dt-bindings: clock: Add YAML schemas for the QCOM DISPCC clock bindings") Signed-off-by: Douglas Anderson <dianders@chromium.org> --- Changes in v4: - Added Rob's review tag. - Fixed schema id to not have "bindings/" as per Rob. Changes in v3: - Added include file to description. - Discovered / added new gcc input clock on sdm845. - Split sc7180 and sdm845 into two files. - Switched names to internal QC names rather than logical ones. - Updated commit description. Changes in v2: - Patch ("dt-bindings: clock: Fix qcom,dispcc...") new for v2. .../bindings/clock/qcom,dispcc.yaml | 67 ------------- .../bindings/clock/qcom,sc7180-dispcc.yaml | 84 ++++++++++++++++ .../bindings/clock/qcom,sdm845-dispcc.yaml | 99 +++++++++++++++++++ 3 files changed, 183 insertions(+), 67 deletions(-) delete mode 100644 Documentation/devicetree/bindings/clock/qcom,dispcc.yaml create mode 100644 Documentation/devicetree/bindings/clock/qcom,sc7180-dispcc.yaml create mode 100644 Documentation/devicetree/bindings/clock/qcom,sdm845-dispcc.yaml diff --git a/Documentation/devicetree/bindings/clock/qcom,dispcc.yaml b/Documentation/devicetree/bindings/clock/qcom,dispcc.yaml deleted file mode 100644 index 9c58e02a1de1..000000000000 --- a/Documentation/devicetree/bindings/clock/qcom,dispcc.yaml +++ /dev/null @@ -1,67 +0,0 @@ -# SPDX-License-Identifier: GPL-2.0-only -%YAML 1.2 ---- -$id: http://devicetree.org/schemas/bindings/clock/qcom,dispcc.yaml# -$schema: http://devicetree.org/meta-schemas/core.yaml# - -title: Qualcomm Display Clock & Reset Controller Binding - -maintainers: - - Taniya Das <tdas@codeaurora.org> - -description: | - Qualcomm display clock control module which supports the clocks, resets and - power domains. - -properties: - compatible: - enum: - - qcom,sc7180-dispcc - - qcom,sdm845-dispcc - - clocks: - minItems: 1 - maxItems: 2 - items: - - description: Board XO source - - description: GPLL0 source from GCC - - clock-names: - items: - - const: xo - - const: gpll0 - - '#clock-cells': - const: 1 - - '#reset-cells': - const: 1 - - '#power-domain-cells': - const: 1 - - reg: - maxItems: 1 - -required: - - compatible - - reg - - clocks - - clock-names - - '#clock-cells' - - '#reset-cells' - - '#power-domain-cells' - -examples: - # Example of DISPCC with clock node properties for SDM845: - - | - clock-controller@af00000 { - compatible = "qcom,sdm845-dispcc"; - reg = <0xaf00000 0x10000>; - clocks = <&rpmhcc 0>, <&gcc 24>; - clock-names = "xo", "gpll0"; - #clock-cells = <1>; - #reset-cells = <1>; - #power-domain-cells = <1>; - }; -... diff --git a/Documentation/devicetree/bindings/clock/qcom,sc7180-dispcc.yaml b/Documentation/devicetree/bindings/clock/qcom,sc7180-dispcc.yaml new file mode 100644 index 000000000000..0429062f1585 --- /dev/null +++ b/Documentation/devicetree/bindings/clock/qcom,sc7180-dispcc.yaml @@ -0,0 +1,84 @@ +# SPDX-License-Identifier: GPL-2.0-only +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/clock/qcom,sc7180-dispcc.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Qualcomm Display Clock & Reset Controller Binding for SC7180 + +maintainers: + - Taniya Das <tdas@codeaurora.org> + +description: | + Qualcomm display clock control module which supports the clocks, resets and + power domains on SC7180. + + See also dt-bindings/clock/qcom,dispcc-sc7180.h. + +properties: + compatible: + const: qcom,sc7180-dispcc + + clocks: + items: + - description: Board XO source + - description: GPLL0 source from GCC + - description: Byte clock from DSI PHY + - description: Pixel clock from DSI PHY + - description: Link clock from DP PHY + - description: VCO DIV clock from DP PHY + + clock-names: + items: + - const: bi_tcxo + - const: gcc_disp_gpll0_clk_src + - const: dsi0_phy_pll_out_byteclk + - const: dsi0_phy_pll_out_dsiclk + - const: dp_phy_pll_link_clk + - const: dp_phy_pll_vco_div_clk + + '#clock-cells': + const: 1 + + '#reset-cells': + const: 1 + + '#power-domain-cells': + const: 1 + + reg: + maxItems: 1 + +required: + - compatible + - reg + - clocks + - clock-names + - '#clock-cells' + - '#reset-cells' + - '#power-domain-cells' + +examples: + - | + #include <dt-bindings/clock/qcom,gcc-sc7180.h> + #include <dt-bindings/clock/qcom,rpmh.h> + clock-controller@af00000 { + compatible = "qcom,sc7180-dispcc"; + reg = <0 0x0af00000 0 0x200000>; + clocks = <&rpmhcc RPMH_CXO_CLK>, + <&gcc GCC_DISP_GPLL0_CLK_SRC>, + <&dsi_phy 0>, + <&dsi_phy 1>, + <&dp_phy 0>, + <&dp_phy 1>; + clock-names = "bi_tcxo", + "gcc_disp_gpll0_clk_src", + "dsi0_phy_pll_out_byteclk", + "dsi0_phy_pll_out_dsiclk", + "dp_phy_pll_link_clk", + "dp_phy_pll_vco_div_clk"; + #clock-cells = <1>; + #reset-cells = <1>; + #power-domain-cells = <1>; + }; +... diff --git a/Documentation/devicetree/bindings/clock/qcom,sdm845-dispcc.yaml b/Documentation/devicetree/bindings/clock/qcom,sdm845-dispcc.yaml new file mode 100644 index 000000000000..89269ddfbdcd --- /dev/null +++ b/Documentation/devicetree/bindings/clock/qcom,sdm845-dispcc.yaml @@ -0,0 +1,99 @@ +# SPDX-License-Identifier: GPL-2.0-only +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/clock/qcom,sdm845-dispcc.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Qualcomm Display Clock & Reset Controller Binding for SDM845 + +maintainers: + - Taniya Das <tdas@codeaurora.org> + +description: | + Qualcomm display clock control module which supports the clocks, resets and + power domains on SDM845. + + See also dt-bindings/clock/qcom,dispcc-sdm845.h. + +properties: + compatible: + const: qcom,sdm845-dispcc + + # NOTE: sdm845.dtsi existed for quite some time and specified no clocks. + # The code had to use hardcoded mechanisms to find the input clocks. + # New dts files should have these clocks. + clocks: + items: + - description: Board XO source + - description: GPLL0 source from GCC + - description: GPLL0 div source from GCC + - description: Byte clock from DSI PHY0 + - description: Pixel clock from DSI PHY0 + - description: Byte clock from DSI PHY1 + - description: Pixel clock from DSI PHY1 + - description: Link clock from DP PHY + - description: VCO DIV clock from DP PHY + + clock-names: + items: + - const: bi_tcxo + - const: gcc_disp_gpll0_clk_src + - const: gcc_disp_gpll0_div_clk_src + - const: dsi0_phy_pll_out_byteclk + - const: dsi0_phy_pll_out_dsiclk + - const: dsi1_phy_pll_out_byteclk + - const: dsi1_phy_pll_out_dsiclk + - const: dp_link_clk_divsel_ten + - const: dp_vco_divided_clk_src_mux + + '#clock-cells': + const: 1 + + '#reset-cells': + const: 1 + + '#power-domain-cells': + const: 1 + + reg: + maxItems: 1 + +required: + - compatible + - reg + - clocks + - clock-names + - '#clock-cells' + - '#reset-cells' + - '#power-domain-cells' + +examples: + - | + #include <dt-bindings/clock/qcom,gcc-sdm845.h> + #include <dt-bindings/clock/qcom,rpmh.h> + clock-controller@af00000 { + compatible = "qcom,sdm845-dispcc"; + reg = <0 0x0af00000 0 0x10000>; + clocks = <&rpmhcc RPMH_CXO_CLK>, + <&gcc GCC_DISP_GPLL0_CLK_SRC>, + <&gcc GCC_DISP_GPLL0_DIV_CLK_SRC>, + <&dsi0_phy 0>, + <&dsi0_phy 1>, + <&dsi1_phy 0>, + <&dsi1_phy 1>, + <&dp_phy 0>, + <&dp_phy 1>; + clock-names = "bi_tcxo", + "gcc_disp_gpll0_clk_src", + "gcc_disp_gpll0_div_clk_src", + "dsi0_phy_pll_out_byteclk", + "dsi0_phy_pll_out_dsiclk", + "dsi1_phy_pll_out_byteclk", + "dsi1_phy_pll_out_dsiclk", + "dp_link_clk_divsel_ten", + "dp_vco_divided_clk_src_mux"; + #clock-cells = <1>; + #reset-cells = <1>; + #power-domain-cells = <1>; + }; +... -- 2.25.0.341.g760bfbb309-goog
We're transitioning over to requiring the Qualcomm Display Clock Controller to specify all the input clocks. Let's add them for sdm845. NOTES: - Until the Linux driver for sdm845's dispcc is updated, these clocks will not actually be used in Linux. It will continue to use global clock names to match things up. - Although the clocks from the DP PHY are required, the DP PHY isn't represented in the dts yet. Apparently the magic for this is just to use <0>. Signed-off-by: Douglas Anderson <dianders@chromium.org> --- Changes in v4: None Changes in v3: - Newly discovered gcc_disp_gpll0_div_clk_src added. - Unlike in v2, use internal name instead of purist name. Changes in v2: - Patch ("arm64: dts: qcom: sdm845: Add...dispcc") new for v2. arch/arm64/boot/dts/qcom/sdm845.dtsi | 18 ++++++++++++++++++ 1 file changed, 18 insertions(+) diff --git a/arch/arm64/boot/dts/qcom/sdm845.dtsi b/arch/arm64/boot/dts/qcom/sdm845.dtsi index d42302b8889b..0985813fee50 100644 --- a/arch/arm64/boot/dts/qcom/sdm845.dtsi +++ b/arch/arm64/boot/dts/qcom/sdm845.dtsi @@ -2933,6 +2933,24 @@ opp-200000000 { dispcc: clock-controller@af00000 { compatible = "qcom,sdm845-dispcc"; reg = <0 0x0af00000 0 0x10000>; + clocks = <&rpmhcc RPMH_CXO_CLK>, + <&gcc GCC_DISP_GPLL0_CLK_SRC>, + <&gcc GCC_DISP_GPLL0_DIV_CLK_SRC>, + <&dsi0_phy 0>, + <&dsi0_phy 1>, + <&dsi1_phy 0>, + <&dsi1_phy 1>, + <0>, + <0>; + clock-names = "bi_tcxo", + "gcc_disp_gpll0_clk_src", + "gcc_disp_gpll0_div_clk_src", + "dsi0_phy_pll_out_byteclk", + "dsi0_phy_pll_out_dsiclk", + "dsi1_phy_pll_out_byteclk", + "dsi1_phy_pll_out_dsiclk", + "dp_link_clk_divsel_ten", + "dp_vco_divided_clk_src_mux"; #clock-cells = <1>; #reset-cells = <1>; #power-domain-cells = <1>; -- 2.25.0.341.g760bfbb309-goog
In the new world input clocks should be matched by ".fw_name". sc7180 is new enough that no backward compatibility use of global names should be needed. Remove it. With a proper device tree and downstream display patches I have verified booting a sc7180 up and seeing the display after this patch. Fixes: dd3d06622138 ("clk: qcom: Add display clock controller driver for SC7180") Signed-off-by: Douglas Anderson <dianders@chromium.org> --- Changes in v4: None Changes in v3: - Patch ("clk: qcom: Get rid of fallback...dispcc-sc7180") split out for v3. - Unlike in v2, use internal name instead of purist name. Changes in v2: None drivers/clk/qcom/dispcc-sc7180.c | 23 ++++++++++------------- 1 file changed, 10 insertions(+), 13 deletions(-) diff --git a/drivers/clk/qcom/dispcc-sc7180.c b/drivers/clk/qcom/dispcc-sc7180.c index 30c1e25d3edb..a820e1558677 100644 --- a/drivers/clk/qcom/dispcc-sc7180.c +++ b/drivers/clk/qcom/dispcc-sc7180.c @@ -81,7 +81,7 @@ static const struct parent_map disp_cc_parent_map_0[] = { static const struct clk_parent_data disp_cc_parent_data_0[] = { { .fw_name = "bi_tcxo" }, - { .fw_name = "core_bi_pll_test_se", .name = "core_bi_pll_test_se" }, + { .fw_name = "core_bi_pll_test_se" }, }; static const struct parent_map disp_cc_parent_map_1[] = { @@ -93,10 +93,9 @@ static const struct parent_map disp_cc_parent_map_1[] = { static const struct clk_parent_data disp_cc_parent_data_1[] = { { .fw_name = "bi_tcxo" }, - { .fw_name = "dp_phy_pll_link_clk", .name = "dp_phy_pll_link_clk" }, - { .fw_name = "dp_phy_pll_vco_div_clk", - .name = "dp_phy_pll_vco_div_clk"}, - { .fw_name = "core_bi_pll_test_se", .name = "core_bi_pll_test_se" }, + { .fw_name = "dp_phy_pll_link_clk" }, + { .fw_name = "dp_phy_pll_vco_div_clk" }, + { .fw_name = "core_bi_pll_test_se" }, }; static const struct parent_map disp_cc_parent_map_2[] = { @@ -107,9 +106,8 @@ static const struct parent_map disp_cc_parent_map_2[] = { static const struct clk_parent_data disp_cc_parent_data_2[] = { { .fw_name = "bi_tcxo" }, - { .fw_name = "dsi0_phy_pll_out_byteclk", - .name = "dsi0_phy_pll_out_byteclk" }, - { .fw_name = "core_bi_pll_test_se", .name = "core_bi_pll_test_se" }, + { .fw_name = "dsi0_phy_pll_out_byteclk" }, + { .fw_name = "core_bi_pll_test_se" }, }; static const struct parent_map disp_cc_parent_map_3[] = { @@ -125,7 +123,7 @@ static const struct clk_parent_data disp_cc_parent_data_3[] = { { .hw = &disp_cc_pll0.clkr.hw }, { .fw_name = "gcc_disp_gpll0_clk_src" }, { .hw = &disp_cc_pll0_out_even.clkr.hw }, - { .fw_name = "core_bi_pll_test_se", .name = "core_bi_pll_test_se" }, + { .fw_name = "core_bi_pll_test_se" }, }; static const struct parent_map disp_cc_parent_map_4[] = { @@ -137,7 +135,7 @@ static const struct parent_map disp_cc_parent_map_4[] = { static const struct clk_parent_data disp_cc_parent_data_4[] = { { .fw_name = "bi_tcxo" }, { .fw_name = "gcc_disp_gpll0_clk_src" }, - { .fw_name = "core_bi_pll_test_se", .name = "core_bi_pll_test_se" }, + { .fw_name = "core_bi_pll_test_se" }, }; static const struct parent_map disp_cc_parent_map_5[] = { @@ -148,9 +146,8 @@ static const struct parent_map disp_cc_parent_map_5[] = { static const struct clk_parent_data disp_cc_parent_data_5[] = { { .fw_name = "bi_tcxo" }, - { .fw_name = "dsi0_phy_pll_out_dsiclk", - .name = "dsi0_phy_pll_out_dsiclk" }, - { .fw_name = "core_bi_pll_test_se", .name = "core_bi_pll_test_se" }, + { .fw_name = "dsi0_phy_pll_out_dsiclk" }, + { .fw_name = "core_bi_pll_test_se" }, }; static const struct freq_tbl ftbl_disp_cc_mdss_ahb_clk_src[] = { -- 2.25.0.341.g760bfbb309-goog
The test clock isn't in the bindings and apparently it's not used by anyone upstream. Remove it. Suggested-by: Stephen Boyd <swboyd@chromium.org> Signed-off-by: Douglas Anderson <dianders@chromium.org> --- Changes in v4: None Changes in v3: - Patch ("clk: qcom: Get rid of the test...dispcc-sc7180") split out for v3. Changes in v2: None drivers/clk/qcom/dispcc-sc7180.c | 32 ++++++++++---------------------- 1 file changed, 10 insertions(+), 22 deletions(-) diff --git a/drivers/clk/qcom/dispcc-sc7180.c b/drivers/clk/qcom/dispcc-sc7180.c index a820e1558677..397f5d9dafc8 100644 --- a/drivers/clk/qcom/dispcc-sc7180.c +++ b/drivers/clk/qcom/dispcc-sc7180.c @@ -76,38 +76,32 @@ static struct clk_alpha_pll_postdiv disp_cc_pll0_out_even = { static const struct parent_map disp_cc_parent_map_0[] = { { P_BI_TCXO, 0 }, - { P_CORE_BI_PLL_TEST_SE, 7 }, }; static const struct clk_parent_data disp_cc_parent_data_0[] = { { .fw_name = "bi_tcxo" }, - { .fw_name = "core_bi_pll_test_se" }, }; static const struct parent_map disp_cc_parent_map_1[] = { { P_BI_TCXO, 0 }, { P_DP_PHY_PLL_LINK_CLK, 1 }, { P_DP_PHY_PLL_VCO_DIV_CLK, 2 }, - { P_CORE_BI_PLL_TEST_SE, 7 }, }; static const struct clk_parent_data disp_cc_parent_data_1[] = { { .fw_name = "bi_tcxo" }, { .fw_name = "dp_phy_pll_link_clk" }, { .fw_name = "dp_phy_pll_vco_div_clk" }, - { .fw_name = "core_bi_pll_test_se" }, }; static const struct parent_map disp_cc_parent_map_2[] = { { P_BI_TCXO, 0 }, { P_DSI0_PHY_PLL_OUT_BYTECLK, 1 }, - { P_CORE_BI_PLL_TEST_SE, 7 }, }; static const struct clk_parent_data disp_cc_parent_data_2[] = { { .fw_name = "bi_tcxo" }, { .fw_name = "dsi0_phy_pll_out_byteclk" }, - { .fw_name = "core_bi_pll_test_se" }, }; static const struct parent_map disp_cc_parent_map_3[] = { @@ -115,7 +109,6 @@ static const struct parent_map disp_cc_parent_map_3[] = { { P_DISP_CC_PLL0_OUT_MAIN, 1 }, { P_GPLL0_OUT_MAIN, 4 }, { P_DISP_CC_PLL0_OUT_EVEN, 5 }, - { P_CORE_BI_PLL_TEST_SE, 7 }, }; static const struct clk_parent_data disp_cc_parent_data_3[] = { @@ -123,31 +116,26 @@ static const struct clk_parent_data disp_cc_parent_data_3[] = { { .hw = &disp_cc_pll0.clkr.hw }, { .fw_name = "gcc_disp_gpll0_clk_src" }, { .hw = &disp_cc_pll0_out_even.clkr.hw }, - { .fw_name = "core_bi_pll_test_se" }, }; static const struct parent_map disp_cc_parent_map_4[] = { { P_BI_TCXO, 0 }, { P_GPLL0_OUT_MAIN, 4 }, - { P_CORE_BI_PLL_TEST_SE, 7 }, }; static const struct clk_parent_data disp_cc_parent_data_4[] = { { .fw_name = "bi_tcxo" }, { .fw_name = "gcc_disp_gpll0_clk_src" }, - { .fw_name = "core_bi_pll_test_se" }, }; static const struct parent_map disp_cc_parent_map_5[] = { { P_BI_TCXO, 0 }, { P_DSI0_PHY_PLL_OUT_DSICLK, 1 }, - { P_CORE_BI_PLL_TEST_SE, 7 }, }; static const struct clk_parent_data disp_cc_parent_data_5[] = { { .fw_name = "bi_tcxo" }, { .fw_name = "dsi0_phy_pll_out_dsiclk" }, - { .fw_name = "core_bi_pll_test_se" }, }; static const struct freq_tbl ftbl_disp_cc_mdss_ahb_clk_src[] = { @@ -166,7 +154,7 @@ static struct clk_rcg2 disp_cc_mdss_ahb_clk_src = { .clkr.hw.init = &(struct clk_init_data){ .name = "disp_cc_mdss_ahb_clk_src", .parent_data = disp_cc_parent_data_4, - .num_parents = 3, + .num_parents = 2, .flags = CLK_SET_RATE_PARENT, .ops = &clk_rcg2_shared_ops, }, @@ -180,7 +168,7 @@ static struct clk_rcg2 disp_cc_mdss_byte0_clk_src = { .clkr.hw.init = &(struct clk_init_data){ .name = "disp_cc_mdss_byte0_clk_src", .parent_data = disp_cc_parent_data_2, - .num_parents = 3, + .num_parents = 2, .flags = CLK_SET_RATE_PARENT, .ops = &clk_byte2_ops, }, @@ -213,7 +201,7 @@ static struct clk_rcg2 disp_cc_mdss_dp_crypto_clk_src = { .clkr.hw.init = &(struct clk_init_data){ .name = "disp_cc_mdss_dp_crypto_clk_src", .parent_data = disp_cc_parent_data_1, - .num_parents = 4, + .num_parents = 3, .flags = CLK_SET_RATE_PARENT, .ops = &clk_byte2_ops, }, @@ -227,7 +215,7 @@ static struct clk_rcg2 disp_cc_mdss_dp_link_clk_src = { .clkr.hw.init = &(struct clk_init_data){ .name = "disp_cc_mdss_dp_link_clk_src", .parent_data = disp_cc_parent_data_1, - .num_parents = 4, + .num_parents = 3, .flags = CLK_SET_RATE_PARENT, .ops = &clk_byte2_ops, }, @@ -241,7 +229,7 @@ static struct clk_rcg2 disp_cc_mdss_dp_pixel_clk_src = { .clkr.hw.init = &(struct clk_init_data){ .name = "disp_cc_mdss_dp_pixel_clk_src", .parent_data = disp_cc_parent_data_1, - .num_parents = 4, + .num_parents = 3, .flags = CLK_SET_RATE_PARENT, .ops = &clk_dp_ops, }, @@ -256,7 +244,7 @@ static struct clk_rcg2 disp_cc_mdss_esc0_clk_src = { .clkr.hw.init = &(struct clk_init_data){ .name = "disp_cc_mdss_esc0_clk_src", .parent_data = disp_cc_parent_data_2, - .num_parents = 3, + .num_parents = 2, .ops = &clk_rcg2_ops, }, }; @@ -279,7 +267,7 @@ static struct clk_rcg2 disp_cc_mdss_mdp_clk_src = { .clkr.hw.init = &(struct clk_init_data){ .name = "disp_cc_mdss_mdp_clk_src", .parent_data = disp_cc_parent_data_3, - .num_parents = 5, + .num_parents = 4, .ops = &clk_rcg2_shared_ops, }, }; @@ -292,7 +280,7 @@ static struct clk_rcg2 disp_cc_mdss_pclk0_clk_src = { .clkr.hw.init = &(struct clk_init_data){ .name = "disp_cc_mdss_pclk0_clk_src", .parent_data = disp_cc_parent_data_5, - .num_parents = 3, + .num_parents = 2, .flags = CLK_SET_RATE_PARENT, .ops = &clk_pixel_ops, }, @@ -307,7 +295,7 @@ static struct clk_rcg2 disp_cc_mdss_rot_clk_src = { .clkr.hw.init = &(struct clk_init_data){ .name = "disp_cc_mdss_rot_clk_src", .parent_data = disp_cc_parent_data_3, - .num_parents = 5, + .num_parents = 4, .ops = &clk_rcg2_shared_ops, }, }; @@ -321,7 +309,7 @@ static struct clk_rcg2 disp_cc_mdss_vsync_clk_src = { .clkr.hw.init = &(struct clk_init_data){ .name = "disp_cc_mdss_vsync_clk_src", .parent_data = disp_cc_parent_data_0, - .num_parents = 2, + .num_parents = 1, .ops = &clk_rcg2_shared_ops, }, }; -- 2.25.0.341.g760bfbb309-goog
It's nicer to use ARRAY_SIZE instead of hardcoding. Had we always been doing this it would have prevented a previous bug. See commit 74c31ff9c84a ("clk: qcom: gpu_cc_gmu_clk_src has 5 parents, not 6"). Signed-off-by: Douglas Anderson <dianders@chromium.org> --- Changes in v4: None Changes in v3: - Patch ("clk: qcom: Use ARRAY_SIZE in dispcc-sc7180...") split out for v3. Changes in v2: None drivers/clk/qcom/dispcc-sc7180.c | 22 +++++++++++----------- 1 file changed, 11 insertions(+), 11 deletions(-) diff --git a/drivers/clk/qcom/dispcc-sc7180.c b/drivers/clk/qcom/dispcc-sc7180.c index 397f5d9dafc8..dd7af41e47eb 100644 --- a/drivers/clk/qcom/dispcc-sc7180.c +++ b/drivers/clk/qcom/dispcc-sc7180.c @@ -154,7 +154,7 @@ static struct clk_rcg2 disp_cc_mdss_ahb_clk_src = { .clkr.hw.init = &(struct clk_init_data){ .name = "disp_cc_mdss_ahb_clk_src", .parent_data = disp_cc_parent_data_4, - .num_parents = 2, + .num_parents = ARRAY_SIZE(disp_cc_parent_data_4), .flags = CLK_SET_RATE_PARENT, .ops = &clk_rcg2_shared_ops, }, @@ -168,7 +168,7 @@ static struct clk_rcg2 disp_cc_mdss_byte0_clk_src = { .clkr.hw.init = &(struct clk_init_data){ .name = "disp_cc_mdss_byte0_clk_src", .parent_data = disp_cc_parent_data_2, - .num_parents = 2, + .num_parents = ARRAY_SIZE(disp_cc_parent_data_2), .flags = CLK_SET_RATE_PARENT, .ops = &clk_byte2_ops, }, @@ -188,7 +188,7 @@ static struct clk_rcg2 disp_cc_mdss_dp_aux_clk_src = { .clkr.hw.init = &(struct clk_init_data){ .name = "disp_cc_mdss_dp_aux_clk_src", .parent_data = disp_cc_parent_data_0, - .num_parents = 2, + .num_parents = ARRAY_SIZE(disp_cc_parent_data_0), .ops = &clk_rcg2_ops, }, }; @@ -201,7 +201,7 @@ static struct clk_rcg2 disp_cc_mdss_dp_crypto_clk_src = { .clkr.hw.init = &(struct clk_init_data){ .name = "disp_cc_mdss_dp_crypto_clk_src", .parent_data = disp_cc_parent_data_1, - .num_parents = 3, + .num_parents = ARRAY_SIZE(disp_cc_parent_data_1), .flags = CLK_SET_RATE_PARENT, .ops = &clk_byte2_ops, }, @@ -215,7 +215,7 @@ static struct clk_rcg2 disp_cc_mdss_dp_link_clk_src = { .clkr.hw.init = &(struct clk_init_data){ .name = "disp_cc_mdss_dp_link_clk_src", .parent_data = disp_cc_parent_data_1, - .num_parents = 3, + .num_parents = ARRAY_SIZE(disp_cc_parent_data_1), .flags = CLK_SET_RATE_PARENT, .ops = &clk_byte2_ops, }, @@ -229,7 +229,7 @@ static struct clk_rcg2 disp_cc_mdss_dp_pixel_clk_src = { .clkr.hw.init = &(struct clk_init_data){ .name = "disp_cc_mdss_dp_pixel_clk_src", .parent_data = disp_cc_parent_data_1, - .num_parents = 3, + .num_parents = ARRAY_SIZE(disp_cc_parent_data_1), .flags = CLK_SET_RATE_PARENT, .ops = &clk_dp_ops, }, @@ -244,7 +244,7 @@ static struct clk_rcg2 disp_cc_mdss_esc0_clk_src = { .clkr.hw.init = &(struct clk_init_data){ .name = "disp_cc_mdss_esc0_clk_src", .parent_data = disp_cc_parent_data_2, - .num_parents = 2, + .num_parents = ARRAY_SIZE(disp_cc_parent_data_2), .ops = &clk_rcg2_ops, }, }; @@ -267,7 +267,7 @@ static struct clk_rcg2 disp_cc_mdss_mdp_clk_src = { .clkr.hw.init = &(struct clk_init_data){ .name = "disp_cc_mdss_mdp_clk_src", .parent_data = disp_cc_parent_data_3, - .num_parents = 4, + .num_parents = ARRAY_SIZE(disp_cc_parent_data_3), .ops = &clk_rcg2_shared_ops, }, }; @@ -280,7 +280,7 @@ static struct clk_rcg2 disp_cc_mdss_pclk0_clk_src = { .clkr.hw.init = &(struct clk_init_data){ .name = "disp_cc_mdss_pclk0_clk_src", .parent_data = disp_cc_parent_data_5, - .num_parents = 2, + .num_parents = ARRAY_SIZE(disp_cc_parent_data_5), .flags = CLK_SET_RATE_PARENT, .ops = &clk_pixel_ops, }, @@ -295,7 +295,7 @@ static struct clk_rcg2 disp_cc_mdss_rot_clk_src = { .clkr.hw.init = &(struct clk_init_data){ .name = "disp_cc_mdss_rot_clk_src", .parent_data = disp_cc_parent_data_3, - .num_parents = 4, + .num_parents = ARRAY_SIZE(disp_cc_parent_data_3), .ops = &clk_rcg2_shared_ops, }, }; @@ -309,7 +309,7 @@ static struct clk_rcg2 disp_cc_mdss_vsync_clk_src = { .clkr.hw.init = &(struct clk_init_data){ .name = "disp_cc_mdss_vsync_clk_src", .parent_data = disp_cc_parent_data_0, - .num_parents = 1, + .num_parents = ARRAY_SIZE(disp_cc_parent_data_0), .ops = &clk_rcg2_shared_ops, }, }; -- 2.25.0.341.g760bfbb309-goog
The qcom,gpucc bindings had a few problems with them: 1. When things were converted to yaml the name of the "gpll0 main" clock got changed from "gpll0" to "gpll0_main". Change it back for msm8998. 2. Apparently there is a push not to use purist aliases for clocks but instead to just use the internal Qualcomm names. For sdm845 and sc7180 (where the drivers haven't already been changed) move in this direction. Things were also getting complicated harder to deal with by jamming several SoCs into one file. Splitting simplifies things. Fixes: 5c6f3a36b913 ("dt-bindings: clock: Add YAML schemas for the QCOM GPUCC clock bindings") Signed-off-by: Douglas Anderson <dianders@chromium.org> --- Changes in v4: - (non-change): Didn't combine sdm845 & sc7180 gpucc as per Stephen. - Fixed schema id to not have "bindings/" as per Rob. Changes in v3: - Added pointer to inlude file in description. - Everyone but msm8998 now uses internal QC names. - Fixed typo grpahics => graphics - Split bindings into 3 files. Changes in v2: - Patch ("dt-bindings: clock: Fix qcom,gpucc...") new for v2. .../devicetree/bindings/clock/qcom,gpucc.yaml | 72 ------------------- .../bindings/clock/qcom,msm8998-gpucc.yaml | 66 +++++++++++++++++ .../bindings/clock/qcom,sc7180-gpucc.yaml | 72 +++++++++++++++++++ .../bindings/clock/qcom,sdm845-gpucc.yaml | 72 +++++++++++++++++++ 4 files changed, 210 insertions(+), 72 deletions(-) delete mode 100644 Documentation/devicetree/bindings/clock/qcom,gpucc.yaml create mode 100644 Documentation/devicetree/bindings/clock/qcom,msm8998-gpucc.yaml create mode 100644 Documentation/devicetree/bindings/clock/qcom,sc7180-gpucc.yaml create mode 100644 Documentation/devicetree/bindings/clock/qcom,sdm845-gpucc.yaml diff --git a/Documentation/devicetree/bindings/clock/qcom,gpucc.yaml b/Documentation/devicetree/bindings/clock/qcom,gpucc.yaml deleted file mode 100644 index 622845aa643f..000000000000 --- a/Documentation/devicetree/bindings/clock/qcom,gpucc.yaml +++ /dev/null @@ -1,72 +0,0 @@ -# SPDX-License-Identifier: GPL-2.0-only -%YAML 1.2 ---- -$id: http://devicetree.org/schemas/bindings/clock/qcom,gpucc.yaml# -$schema: http://devicetree.org/meta-schemas/core.yaml# - -title: Qualcomm Graphics Clock & Reset Controller Binding - -maintainers: - - Taniya Das <tdas@codeaurora.org> - -description: | - Qualcomm grpahics clock control module which supports the clocks, resets and - power domains. - -properties: - compatible: - enum: - - qcom,msm8998-gpucc - - qcom,sc7180-gpucc - - qcom,sdm845-gpucc - - clocks: - minItems: 1 - maxItems: 3 - items: - - description: Board XO source - - description: GPLL0 main branch source from GCC(gcc_gpu_gpll0_clk_src) - - description: GPLL0 div branch source from GCC(gcc_gpu_gpll0_div_clk_src) - - clock-names: - minItems: 1 - maxItems: 3 - items: - - const: xo - - const: gpll0_main - - const: gpll0_div - - '#clock-cells': - const: 1 - - '#reset-cells': - const: 1 - - '#power-domain-cells': - const: 1 - - reg: - maxItems: 1 - -required: - - compatible - - reg - - clocks - - clock-names - - '#clock-cells' - - '#reset-cells' - - '#power-domain-cells' - -examples: - # Example of GPUCC with clock node properties for SDM845: - - | - clock-controller@5090000 { - compatible = "qcom,sdm845-gpucc"; - reg = <0x5090000 0x9000>; - clocks = <&rpmhcc 0>, <&gcc 31>, <&gcc 32>; - clock-names = "xo", "gpll0_main", "gpll0_div"; - #clock-cells = <1>; - #reset-cells = <1>; - #power-domain-cells = <1>; - }; -... diff --git a/Documentation/devicetree/bindings/clock/qcom,msm8998-gpucc.yaml b/Documentation/devicetree/bindings/clock/qcom,msm8998-gpucc.yaml new file mode 100644 index 000000000000..7d853c1a85e5 --- /dev/null +++ b/Documentation/devicetree/bindings/clock/qcom,msm8998-gpucc.yaml @@ -0,0 +1,66 @@ +# SPDX-License-Identifier: GPL-2.0-only +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/clock/qcom,msm8998-gpucc.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Qualcomm Graphics Clock & Reset Controller Binding for MSM8998 + +maintainers: + - Taniya Das <tdas@codeaurora.org> + +description: | + Qualcomm graphics clock control module which supports the clocks, resets and + power domains on MSM8998. + + See also dt-bindings/clock/qcom,gpucc-msm8998.h. + +properties: + compatible: + const: qcom,msm8998-gpucc + + clocks: + items: + - description: Board XO source + - description: GPLL0 main branch source (gcc_gpu_gpll0_clk_src) + + clock-names: + items: + - const: xo + - const: gpll0 + + '#clock-cells': + const: 1 + + '#reset-cells': + const: 1 + + '#power-domain-cells': + const: 1 + + reg: + maxItems: 1 + +required: + - compatible + - reg + - clocks + - clock-names + - '#clock-cells' + - '#reset-cells' + - '#power-domain-cells' + +examples: + - | + #include <dt-bindings/clock/qcom,gcc-msm8998.h> + #include <dt-bindings/clock/qcom,rpmcc.h> + clock-controller@5065000 { + compatible = "qcom,msm8998-gpucc"; + #clock-cells = <1>; + #reset-cells = <1>; + #power-domain-cells = <1>; + reg = <0x05065000 0x9000>; + clocks = <&rpmcc RPM_SMD_XO_CLK_SRC>, <&gcc GPLL0_OUT_MAIN>; + clock-names = "xo", "gpll0"; + }; +... diff --git a/Documentation/devicetree/bindings/clock/qcom,sc7180-gpucc.yaml b/Documentation/devicetree/bindings/clock/qcom,sc7180-gpucc.yaml new file mode 100644 index 000000000000..5785192cc4be --- /dev/null +++ b/Documentation/devicetree/bindings/clock/qcom,sc7180-gpucc.yaml @@ -0,0 +1,72 @@ +# SPDX-License-Identifier: GPL-2.0-only +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/clock/qcom,sc7180-gpucc.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Qualcomm Graphics Clock & Reset Controller Binding for SC7180 + +maintainers: + - Taniya Das <tdas@codeaurora.org> + +description: | + Qualcomm graphics clock control module which supports the clocks, resets and + power domains on SC7180. + + See also dt-bindings/clock/qcom,gpucc-sc7180.h. + +properties: + compatible: + const: qcom,sc7180-gpucc + + clocks: + items: + - description: Board XO source + - description: GPLL0 main branch source + - description: GPLL0 div branch source + + clock-names: + items: + - const: bi_tcxo + - const: gcc_gpu_gpll0_clk_src + - const: gcc_gpu_gpll0_div_clk_src + + '#clock-cells': + const: 1 + + '#reset-cells': + const: 1 + + '#power-domain-cells': + const: 1 + + reg: + maxItems: 1 + +required: + - compatible + - reg + - clocks + - clock-names + - '#clock-cells' + - '#reset-cells' + - '#power-domain-cells' + +examples: + - | + #include <dt-bindings/clock/qcom,gcc-sc7180.h> + #include <dt-bindings/clock/qcom,rpmh.h> + clock-controller@5090000 { + compatible = "qcom,sc7180-gpucc"; + reg = <0 0x05090000 0 0x9000>; + clocks = <&rpmhcc RPMH_CXO_CLK>, + <&gcc GCC_GPU_GPLL0_CLK_SRC>, + <&gcc GCC_GPU_GPLL0_DIV_CLK_SRC>; + clock-names = "bi_tcxo", + "gcc_gpu_gpll0_clk_src", + "gcc_gpu_gpll0_div_clk_src"; + #clock-cells = <1>; + #reset-cells = <1>; + #power-domain-cells = <1>; + }; +... diff --git a/Documentation/devicetree/bindings/clock/qcom,sdm845-gpucc.yaml b/Documentation/devicetree/bindings/clock/qcom,sdm845-gpucc.yaml new file mode 100644 index 000000000000..bac04f1c5d79 --- /dev/null +++ b/Documentation/devicetree/bindings/clock/qcom,sdm845-gpucc.yaml @@ -0,0 +1,72 @@ +# SPDX-License-Identifier: GPL-2.0-only +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/clock/qcom,sdm845-gpucc.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Qualcomm Graphics Clock & Reset Controller Binding for SDM845 + +maintainers: + - Taniya Das <tdas@codeaurora.org> + +description: | + Qualcomm graphics clock control module which supports the clocks, resets and + power domains on SDM845. + + See also dt-bindings/clock/qcom,gpucc-sdm845.h. + +properties: + compatible: + const: qcom,sdm845-gpucc + + clocks: + items: + - description: Board XO source + - description: GPLL0 main branch source + - description: GPLL0 div branch source + + clock-names: + items: + - const: bi_tcxo + - const: gcc_gpu_gpll0_clk_src + - const: gcc_gpu_gpll0_div_clk_src + + '#clock-cells': + const: 1 + + '#reset-cells': + const: 1 + + '#power-domain-cells': + const: 1 + + reg: + maxItems: 1 + +required: + - compatible + - reg + - clocks + - clock-names + - '#clock-cells' + - '#reset-cells' + - '#power-domain-cells' + +examples: + - | + #include <dt-bindings/clock/qcom,gcc-sdm845.h> + #include <dt-bindings/clock/qcom,rpmh.h> + clock-controller@5090000 { + compatible = "qcom,sdm845-gpucc"; + reg = <0 0x05090000 0 0x9000>; + clocks = <&rpmhcc RPMH_CXO_CLK>, + <&gcc GCC_GPU_GPLL0_CLK_SRC>, + <&gcc GCC_GPU_GPLL0_DIV_CLK_SRC>; + clock-names = "bi_tcxo", + "gcc_gpu_gpll0_clk_src", + "gcc_gpu_gpll0_div_clk_src"; + #clock-cells = <1>; + #reset-cells = <1>; + #power-domain-cells = <1>; + }; +... -- 2.25.0.341.g760bfbb309-goog
We're transitioning over to requiring the Qualcomm GPU Clock Controller to specify all the input clocks. Let's add them for sdm845. As part of this we've decided that the xo clock should be referred to in the bindings as "bi_tcxo". Change the dts. NOTE: Until the Linux driver for sdm845's gpucc is updated, these clocks will not actually be used in Linux. It will continue to use global clock names to match things up. Of course, Linux didn't use the old "xo" clock anyway. Signed-off-by: Douglas Anderson <dianders@chromium.org> --- Changes in v4: None Changes in v3: - Unlike in v2, use internal name instead of purist name. Changes in v2: - Patch ("arm64: dts: qcom: sdm845: Add...gpucc") new for v2. arch/arm64/boot/dts/qcom/sdm845.dtsi | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/arch/arm64/boot/dts/qcom/sdm845.dtsi b/arch/arm64/boot/dts/qcom/sdm845.dtsi index 0985813fee50..35d7fcbda43c 100644 --- a/arch/arm64/boot/dts/qcom/sdm845.dtsi +++ b/arch/arm64/boot/dts/qcom/sdm845.dtsi @@ -1903,8 +1903,12 @@ gpucc: clock-controller@5090000 { #clock-cells = <1>; #reset-cells = <1>; #power-domain-cells = <1>; - clocks = <&rpmhcc RPMH_CXO_CLK>; - clock-names = "xo"; + clocks = <&rpmhcc RPMH_CXO_CLK>, + <&gcc GCC_GPU_GPLL0_CLK_SRC>, + <&gcc GCC_GPU_GPLL0_DIV_CLK_SRC>; + clock-names = "bi_tcxo", + "gcc_gpu_gpll0_clk_src", + "gcc_gpu_gpll0_div_clk_src"; }; stm@6002000 { -- 2.25.0.341.g760bfbb309-goog
The test clock isn't in the bindings and apparently it's not used by anyone upstream. Remove it. Suggested-by: Stephen Boyd <swboyd@chromium.org> Signed-off-by: Douglas Anderson <dianders@chromium.org> --- Changes in v4: None Changes in v3: - Patch ("clk: qcom: Get rid of the test...gpucc-sc7180") split out for v3. Changes in v2: None drivers/clk/qcom/gpucc-sc7180.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/drivers/clk/qcom/gpucc-sc7180.c b/drivers/clk/qcom/gpucc-sc7180.c index ec61194cceaf..c88f00125775 100644 --- a/drivers/clk/qcom/gpucc-sc7180.c +++ b/drivers/clk/qcom/gpucc-sc7180.c @@ -60,7 +60,6 @@ static const struct parent_map gpu_cc_parent_map_0[] = { { P_GPU_CC_PLL1_OUT_MAIN, 3 }, { P_GPLL0_OUT_MAIN, 5 }, { P_GPLL0_OUT_MAIN_DIV, 6 }, - { P_CORE_BI_PLL_TEST_SE, 7 }, }; static const struct clk_parent_data gpu_cc_parent_data_0[] = { @@ -68,7 +67,6 @@ static const struct clk_parent_data gpu_cc_parent_data_0[] = { { .hw = &gpu_cc_pll1.clkr.hw }, { .fw_name = "gcc_gpu_gpll0_clk_src" }, { .fw_name = "gcc_gpu_gpll0_div_clk_src" }, - { .fw_name = "core_bi_pll_test_se", .name = "core_bi_pll_test_se" }, }; static const struct freq_tbl ftbl_gpu_cc_gmu_clk_src[] = { @@ -86,7 +84,7 @@ static struct clk_rcg2 gpu_cc_gmu_clk_src = { .clkr.hw.init = &(struct clk_init_data){ .name = "gpu_cc_gmu_clk_src", .parent_data = gpu_cc_parent_data_0, - .num_parents = 5, + .num_parents = 4, .flags = CLK_SET_RATE_PARENT, .ops = &clk_rcg2_shared_ops, }, -- 2.25.0.341.g760bfbb309-goog
It's nicer to use ARRAY_SIZE instead of hardcoding. Had we always been doing this it would have prevented a previous bug. See commit 74c31ff9c84a ("clk: qcom: gpu_cc_gmu_clk_src has 5 parents, not 6"). Signed-off-by: Douglas Anderson <dianders@chromium.org> --- Changes in v4: None Changes in v3: - Patch ("clk: qcom: Use ARRAY_SIZE in gpucc-sc7180...") split out for v3. Changes in v2: None drivers/clk/qcom/gpucc-sc7180.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/clk/qcom/gpucc-sc7180.c b/drivers/clk/qcom/gpucc-sc7180.c index c88f00125775..a96c0b945de2 100644 --- a/drivers/clk/qcom/gpucc-sc7180.c +++ b/drivers/clk/qcom/gpucc-sc7180.c @@ -84,7 +84,7 @@ static struct clk_rcg2 gpu_cc_gmu_clk_src = { .clkr.hw.init = &(struct clk_init_data){ .name = "gpu_cc_gmu_clk_src", .parent_data = gpu_cc_parent_data_0, - .num_parents = 4, + .num_parents = ARRAY_SIZE(gpu_cc_parent_data_0), .flags = CLK_SET_RATE_PARENT, .ops = &clk_rcg2_shared_ops, }, -- 2.25.0.341.g760bfbb309-goog
This makes the qcom,videocc bindings match the recent changes to the dispcc and gpucc. 1. Switched to using "bi_tcxo" instead of "xo". 2. Adds a description for the XO clock. Not terribly important but nice if it cleanly matches its cousins. 3. Updates the example to use the symbolic name for the RPMH clock and also show that the real devices are currently using 2 address cells / size cells and fixes the spacing on the closing brace. 4. Split into 2 files. In this case they could probably share one file, but let's be consistent. Signed-off-by: Douglas Anderson <dianders@chromium.org> --- Changes in v4: - Added Rob's review tag. - Fixed schema id to not have "bindings/" as per Rob. Changes in v3: - Added include file to description. - Split videocc bindings into 2 files. - Unlike in v2, use internal name instead of purist name. Changes in v2: - Patch ("dt-bindings: clock: Cleanup qcom,videocc") new for v2. .../bindings/clock/qcom,sc7180-videocc.yaml | 63 +++++++++++++++++++ ...,videocc.yaml => qcom,sdm845-videocc.yaml} | 27 ++++---- 2 files changed, 77 insertions(+), 13 deletions(-) create mode 100644 Documentation/devicetree/bindings/clock/qcom,sc7180-videocc.yaml rename Documentation/devicetree/bindings/clock/{qcom,videocc.yaml => qcom,sdm845-videocc.yaml} (61%) diff --git a/Documentation/devicetree/bindings/clock/qcom,sc7180-videocc.yaml b/Documentation/devicetree/bindings/clock/qcom,sc7180-videocc.yaml new file mode 100644 index 000000000000..31df901884ac --- /dev/null +++ b/Documentation/devicetree/bindings/clock/qcom,sc7180-videocc.yaml @@ -0,0 +1,63 @@ +# SPDX-License-Identifier: GPL-2.0-only +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/clock/qcom,sc7180-videocc.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Qualcomm Video Clock & Reset Controller Binding for SC7180 + +maintainers: + - Taniya Das <tdas@codeaurora.org> + +description: | + Qualcomm video clock control module which supports the clocks, resets and + power domains on SC7180. + + See also dt-bindings/clock/qcom,videocc-sc7180.h. + +properties: + compatible: + const: qcom,sc7180-videocc + + clocks: + items: + - description: Board XO source + + clock-names: + items: + - const: bi_tcxo + + '#clock-cells': + const: 1 + + '#reset-cells': + const: 1 + + '#power-domain-cells': + const: 1 + + reg: + maxItems: 1 + +required: + - compatible + - reg + - clocks + - clock-names + - '#clock-cells' + - '#reset-cells' + - '#power-domain-cells' + +examples: + - | + #include <dt-bindings/clock/qcom,rpmh.h> + clock-controller@ab00000 { + compatible = "qcom,sc7180-videocc"; + reg = <0 0x0ab00000 0 0x10000>; + clocks = <&rpmhcc RPMH_CXO_CLK>; + clock-names = "bi_tcxo"; + #clock-cells = <1>; + #reset-cells = <1>; + #power-domain-cells = <1>; + }; +... diff --git a/Documentation/devicetree/bindings/clock/qcom,videocc.yaml b/Documentation/devicetree/bindings/clock/qcom,sdm845-videocc.yaml similarity index 61% rename from Documentation/devicetree/bindings/clock/qcom,videocc.yaml rename to Documentation/devicetree/bindings/clock/qcom,sdm845-videocc.yaml index 43cfc893a8d1..9d216c0f11d4 100644 --- a/Documentation/devicetree/bindings/clock/qcom,videocc.yaml +++ b/Documentation/devicetree/bindings/clock/qcom,sdm845-videocc.yaml @@ -1,30 +1,31 @@ # SPDX-License-Identifier: GPL-2.0-only %YAML 1.2 --- -$id: http://devicetree.org/schemas/bindings/clock/qcom,videocc.yaml# +$id: http://devicetree.org/schemas/clock/qcom,sdm845-videocc.yaml# $schema: http://devicetree.org/meta-schemas/core.yaml# -title: Qualcomm Video Clock & Reset Controller Binding +title: Qualcomm Video Clock & Reset Controller Binding for SDM845 maintainers: - Taniya Das <tdas@codeaurora.org> description: | Qualcomm video clock control module which supports the clocks, resets and - power domains. + power domains on SDM845. + + See also dt-bindings/clock/qcom,videocc-sdm845.h. properties: compatible: - enum: - - qcom,sc7180-videocc - - qcom,sdm845-videocc + const: qcom,sdm845-videocc clocks: - maxItems: 1 + items: + - description: Board XO source clock-names: items: - - const: xo + - const: bi_tcxo '#clock-cells': const: 1 @@ -48,15 +49,15 @@ required: - '#power-domain-cells' examples: - # Example of VIDEOCC with clock node properties for SDM845: - | + #include <dt-bindings/clock/qcom,rpmh.h> clock-controller@ab00000 { compatible = "qcom,sdm845-videocc"; - reg = <0xab00000 0x10000>; - clocks = <&rpmhcc 0>; - clock-names = "xo"; + reg = <0 0x0ab00000 0 0x10000>; + clocks = <&rpmhcc RPMH_CXO_CLK>; + clock-names = "bi_tcxo"; #clock-cells = <1>; #reset-cells = <1>; #power-domain-cells = <1>; - }; + }; ... -- 2.25.0.341.g760bfbb309-goog
The test clock isn't in the bindings and apparently it's not used by anyone upstream. Remove it. Suggested-by: Stephen Boyd <swboyd@chromium.org> Signed-off-by: Douglas Anderson <dianders@chromium.org> --- Changes in v4: None Changes in v3: - Patch ("clk: qcom: Get rid of the test...videocc-sc7180") new for v3. Changes in v2: None drivers/clk/qcom/videocc-sc7180.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/drivers/clk/qcom/videocc-sc7180.c b/drivers/clk/qcom/videocc-sc7180.c index 76add30024aa..653fc4e6bb6f 100644 --- a/drivers/clk/qcom/videocc-sc7180.c +++ b/drivers/clk/qcom/videocc-sc7180.c @@ -50,13 +50,11 @@ static struct clk_alpha_pll video_pll0 = { static const struct parent_map video_cc_parent_map_1[] = { { P_BI_TCXO, 0 }, { P_VIDEO_PLL0_OUT_MAIN, 1 }, - { P_CORE_BI_PLL_TEST_SE, 7 }, }; static const struct clk_parent_data video_cc_parent_data_1[] = { { .fw_name = "bi_tcxo" }, { .hw = &video_pll0.clkr.hw }, - { .fw_name = "core_bi_pll_test_se", .name = "core_bi_pll_test_se" }, }; static const struct freq_tbl ftbl_video_cc_venus_clk_src[] = { @@ -78,7 +76,7 @@ static struct clk_rcg2 video_cc_venus_clk_src = { .clkr.hw.init = &(struct clk_init_data){ .name = "video_cc_venus_clk_src", .parent_data = video_cc_parent_data_1, - .num_parents = 3, + .num_parents = 2, .flags = CLK_SET_RATE_PARENT, .ops = &clk_rcg2_shared_ops, }, -- 2.25.0.341.g760bfbb309-goog
It's nicer to use ARRAY_SIZE instead of hardcoding. Had we always been doing this it would have prevented a previous bug. See commit 74c31ff9c84a ("clk: qcom: gpu_cc_gmu_clk_src has 5 parents, not 6"). Signed-off-by: Douglas Anderson <dianders@chromium.org> --- Changes in v4: None Changes in v3: - Patch ("clk: qcom: Use ARRAY_SIZE in videocc-sc7180...") new for v3. Changes in v2: None drivers/clk/qcom/videocc-sc7180.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/clk/qcom/videocc-sc7180.c b/drivers/clk/qcom/videocc-sc7180.c index 653fc4e6bb6f..c363c3cc544e 100644 --- a/drivers/clk/qcom/videocc-sc7180.c +++ b/drivers/clk/qcom/videocc-sc7180.c @@ -76,7 +76,7 @@ static struct clk_rcg2 video_cc_venus_clk_src = { .clkr.hw.init = &(struct clk_init_data){ .name = "video_cc_venus_clk_src", .parent_data = video_cc_parent_data_1, - .num_parents = 2, + .num_parents = ARRAY_SIZE(video_cc_parent_data_1), .flags = CLK_SET_RATE_PARENT, .ops = &clk_rcg2_shared_ops, }, -- 2.25.0.341.g760bfbb309-goog
We're transitioning over to requiring the Qualcomm Video Clock Controller to specify all the input clocks. Let's add the one input clock for the videocc for sdm845. NOTE: Until the Linux driver for sdm845's video is updated, this clock will not actually be used in Linux. It will continue to use global clock names to match things up. Signed-off-by: Douglas Anderson <dianders@chromium.org> --- Changes in v4: None Changes in v3: - Unlike in v2, use internal name instead of purist name. Changes in v2: - Patch ("arm64: dts: qcom: sdm845: Add...videocc") new for v2. arch/arm64/boot/dts/qcom/sdm845.dtsi | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/arm64/boot/dts/qcom/sdm845.dtsi b/arch/arm64/boot/dts/qcom/sdm845.dtsi index 35d7fcbda43c..3ad08d9deb54 100644 --- a/arch/arm64/boot/dts/qcom/sdm845.dtsi +++ b/arch/arm64/boot/dts/qcom/sdm845.dtsi @@ -2607,6 +2607,8 @@ video-core1 { videocc: clock-controller@ab00000 { compatible = "qcom,sdm845-videocc"; reg = <0 0x0ab00000 0 0x10000>; + clocks = <&rpmhcc RPMH_CXO_CLK>; + clock-names = "bi_tcxo"; #clock-cells = <1>; #power-domain-cells = <1>; #reset-cells = <1>; -- 2.25.0.341.g760bfbb309-goog
From: Taniya Das <tdas@codeaurora.org> Add the display, video & graphics clock controller nodes supported on SC7180. NOTE: the dispcc needs input clocks from various PHYs that aren't in the device tree yet. For now we'll leave these stubbed out with <0>, which is apparently the magic way to do this. These clocks aren't really "optional" and this stubbing out method is apparently the best way to handle it. Signed-off-by: Taniya Das <tdas@codeaurora.org> Signed-off-by: Douglas Anderson <dianders@chromium.org> --- Changes in v4: None Changes in v3: - Added videocc include file. - Unlike in v2, use internal name instead of purist name. Changes in v2: - Added includes - Changed various parent names to match bindings / driver arch/arm64/boot/dts/qcom/sc7180.dtsi | 47 ++++++++++++++++++++++++++++ 1 file changed, 47 insertions(+) diff --git a/arch/arm64/boot/dts/qcom/sc7180.dtsi b/arch/arm64/boot/dts/qcom/sc7180.dtsi index 8011c5fe2a31..57ff5e0f7ae6 100644 --- a/arch/arm64/boot/dts/qcom/sc7180.dtsi +++ b/arch/arm64/boot/dts/qcom/sc7180.dtsi @@ -5,8 +5,11 @@ * Copyright (c) 2019, The Linux Foundation. All rights reserved. */ +#include <dt-bindings/clock/qcom,dispcc-sc7180.h> #include <dt-bindings/clock/qcom,gcc-sc7180.h> +#include <dt-bindings/clock/qcom,gpucc-sc7180.h> #include <dt-bindings/clock/qcom,rpmh.h> +#include <dt-bindings/clock/qcom,videocc-sc7180.h> #include <dt-bindings/interrupt-controller/arm-gic.h> #include <dt-bindings/phy/phy-qcom-qusb2.h> #include <dt-bindings/power/qcom-aoss-qmp.h> @@ -1039,6 +1042,20 @@ pinmux { }; }; + gpucc: clock-controller@5090000 { + compatible = "qcom,sc7180-gpucc"; + reg = <0 0x05090000 0 0x9000>; + clocks = <&rpmhcc RPMH_CXO_CLK>, + <&gcc GCC_GPU_GPLL0_CLK_SRC>, + <&gcc GCC_GPU_GPLL0_DIV_CLK_SRC>; + clock-names = "bi_tcxo", + "gcc_gpu_gpll0_clk_src", + "gcc_gpu_gpll0_div_clk_src"; + #clock-cells = <1>; + #reset-cells = <1>; + #power-domain-cells = <1>; + }; + qspi: spi@88dc000 { compatible = "qcom,qspi-v1"; reg = <0 0x088dc000 0 0x600>; @@ -1151,6 +1168,36 @@ usb_1_dwc3: dwc3@a600000 { }; }; + videocc: clock-controller@ab00000 { + compatible = "qcom,sc7180-videocc"; + reg = <0 0x0ab00000 0 0x10000>; + clocks = <&rpmhcc RPMH_CXO_CLK>; + clock-names = "bi_tcxo"; + #clock-cells = <1>; + #reset-cells = <1>; + #power-domain-cells = <1>; + }; + + dispcc: clock-controller@af00000 { + compatible = "qcom,sc7180-dispcc"; + reg = <0 0x0af00000 0 0x200000>; + clocks = <&rpmhcc RPMH_CXO_CLK>, + <&gcc GCC_DISP_GPLL0_CLK_SRC>, + <0>, + <0>, + <0>, + <0>; + clock-names = "bi_tcxo", + "gcc_disp_gpll0_clk_src", + "dsi0_phy_pll_out_byteclk", + "dsi0_phy_pll_out_dsiclk", + "dp_phy_pll_link_clk", + "dp_phy_pll_vco_div_clk"; + #clock-cells = <1>; + #reset-cells = <1>; + #power-domain-cells = <1>; + }; + pdc: interrupt-controller@b220000 { compatible = "qcom,sc7180-pdc", "qcom,pdc"; reg = <0 0x0b220000 0 0x30000>; -- 2.25.0.341.g760bfbb309-goog
Quoting Douglas Anderson (2020-02-03 10:31:33)
>
> .../devicetree/bindings/clock/qcom,gpucc.yaml | 72 --------------
> ...om,dispcc.yaml => qcom,msm8998-gpucc.yaml} | 33 +++----
> .../bindings/clock/qcom,sc7180-dispcc.yaml | 84 ++++++++++++++++
> .../bindings/clock/qcom,sc7180-gpucc.yaml | 72 ++++++++++++++
> .../bindings/clock/qcom,sc7180-videocc.yaml | 63 ++++++++++++
> .../bindings/clock/qcom,sdm845-dispcc.yaml | 99 +++++++++++++++++++
> .../bindings/clock/qcom,sdm845-gpucc.yaml | 72 ++++++++++++++
> ...,videocc.yaml => qcom,sdm845-videocc.yaml} | 27 ++---
> arch/arm64/boot/dts/qcom/sc7180.dtsi | 47 +++++++++
> arch/arm64/boot/dts/qcom/sdm845.dtsi | 28 +++++-
I don't want to take patches touching dts/qcom/. These aren't necessary
to merge right now, correct? Or at least, they can go via arm-soc tree?
Hi,
On Mon, Feb 3, 2020 at 11:30 AM Stephen Boyd <sboyd@kernel.org> wrote:
>
> Quoting Douglas Anderson (2020-02-03 10:31:33)
> >
> > .../devicetree/bindings/clock/qcom,gpucc.yaml | 72 --------------
> > ...om,dispcc.yaml => qcom,msm8998-gpucc.yaml} | 33 +++----
> > .../bindings/clock/qcom,sc7180-dispcc.yaml | 84 ++++++++++++++++
> > .../bindings/clock/qcom,sc7180-gpucc.yaml | 72 ++++++++++++++
> > .../bindings/clock/qcom,sc7180-videocc.yaml | 63 ++++++++++++
> > .../bindings/clock/qcom,sdm845-dispcc.yaml | 99 +++++++++++++++++++
> > .../bindings/clock/qcom,sdm845-gpucc.yaml | 72 ++++++++++++++
> > ...,videocc.yaml => qcom,sdm845-videocc.yaml} | 27 ++---
> > arch/arm64/boot/dts/qcom/sc7180.dtsi | 47 +++++++++
> > arch/arm64/boot/dts/qcom/sdm845.dtsi | 28 +++++-
>
> I don't want to take patches touching dts/qcom/. These aren't necessary
> to merge right now, correct? Or at least, they can go via arm-soc tree?
Right. They can go later.
Specifically for sdm845 until the sdm845 patches lands the old dts
trees will yell about the missing clocks, but it's not like they will
compile fail. Also the bindings themselves will validate fine.
There's no other way forward, though, and the old bindings caused
similar yells.
For sc7180 there's no usage of any of these clocks and this adds the
first usage, so definitely no problem there.
Once you've landed then Bjorn or Andy can pick up the dts.
-Doug
On Mon 03 Feb 11:41 PST 2020, Doug Anderson wrote: > Hi, > > On Mon, Feb 3, 2020 at 11:30 AM Stephen Boyd <sboyd@kernel.org> wrote: > > > > Quoting Douglas Anderson (2020-02-03 10:31:33) > > > > > > .../devicetree/bindings/clock/qcom,gpucc.yaml | 72 -------------- > > > ...om,dispcc.yaml => qcom,msm8998-gpucc.yaml} | 33 +++---- > > > .../bindings/clock/qcom,sc7180-dispcc.yaml | 84 ++++++++++++++++ > > > .../bindings/clock/qcom,sc7180-gpucc.yaml | 72 ++++++++++++++ > > > .../bindings/clock/qcom,sc7180-videocc.yaml | 63 ++++++++++++ > > > .../bindings/clock/qcom,sdm845-dispcc.yaml | 99 +++++++++++++++++++ > > > .../bindings/clock/qcom,sdm845-gpucc.yaml | 72 ++++++++++++++ > > > ...,videocc.yaml => qcom,sdm845-videocc.yaml} | 27 ++--- > > > arch/arm64/boot/dts/qcom/sc7180.dtsi | 47 +++++++++ > > > arch/arm64/boot/dts/qcom/sdm845.dtsi | 28 +++++- > > > > I don't want to take patches touching dts/qcom/. These aren't necessary > > to merge right now, correct? Or at least, they can go via arm-soc tree? > > Right. They can go later. > > Specifically for sdm845 until the sdm845 patches lands the old dts > trees will yell about the missing clocks, but it's not like they will > compile fail. Also the bindings themselves will validate fine. > There's no other way forward, though, and the old bindings caused > similar yells. > Can you please help me parse this, will old DT cause complaints or will it fail to boot? > For sc7180 there's no usage of any of these clocks and this adds the > first usage, so definitely no problem there. > > Once you've landed then Bjorn or Andy can pick up the dts. > Do I need to apply these after Stephen picks the driver patches? Or are they simply nop until the driver patches lands? Regards, Bjorn
Hi, On Mon, Feb 3, 2020 at 12:04 PM Bjorn Andersson <bjorn.andersson@linaro.org> wrote: > > On Mon 03 Feb 11:41 PST 2020, Doug Anderson wrote: > > > Hi, > > > > On Mon, Feb 3, 2020 at 11:30 AM Stephen Boyd <sboyd@kernel.org> wrote: > > > > > > Quoting Douglas Anderson (2020-02-03 10:31:33) > > > > > > > > .../devicetree/bindings/clock/qcom,gpucc.yaml | 72 -------------- > > > > ...om,dispcc.yaml => qcom,msm8998-gpucc.yaml} | 33 +++---- > > > > .../bindings/clock/qcom,sc7180-dispcc.yaml | 84 ++++++++++++++++ > > > > .../bindings/clock/qcom,sc7180-gpucc.yaml | 72 ++++++++++++++ > > > > .../bindings/clock/qcom,sc7180-videocc.yaml | 63 ++++++++++++ > > > > .../bindings/clock/qcom,sdm845-dispcc.yaml | 99 +++++++++++++++++++ > > > > .../bindings/clock/qcom,sdm845-gpucc.yaml | 72 ++++++++++++++ > > > > ...,videocc.yaml => qcom,sdm845-videocc.yaml} | 27 ++--- > > > > arch/arm64/boot/dts/qcom/sc7180.dtsi | 47 +++++++++ > > > > arch/arm64/boot/dts/qcom/sdm845.dtsi | 28 +++++- > > > > > > I don't want to take patches touching dts/qcom/. These aren't necessary > > > to merge right now, correct? Or at least, they can go via arm-soc tree? > > > > Right. They can go later. > > > > Specifically for sdm845 until the sdm845 patches lands the old dts > > trees will yell about the missing clocks, but it's not like they will > > compile fail. Also the bindings themselves will validate fine. > > There's no other way forward, though, and the old bindings caused > > similar yells. > > > > Can you please help me parse this, will old DT cause complaints or will > it fail to boot? Sorry, let me try to be more clear: a) Without my series: "make dtbs_check" limited to the bindings files I'm touching yells. Examples: .../msm8998-mtp.dt.yaml: clock-controller@5065000: clock-names:1: 'gpll0_main' was expected .../sdm845-mtp.dt.yaml: clock-controller@af00000: 'clocks' is a required property .../sdm845-mtp.dt.yaml: clock-controller@af00000: 'clock-names' is a required property .../sdm845-mtp.dt.yaml: clock-controller@ab00000: 'clocks' is a required property .../sdm845-mtp.dt.yaml: clock-controller@ab00000: 'clock-names' is a required property b) With just the bindings from my series, "make dtbs_check" yells, but yells about different things. The "gpll0_main" one goes away but this one is new: .../sdm845-mtp.dt.yaml: clock-controller@5090000: clock-names:0: 'bi_tcxo' was expected .../sdm845-mtp.dt.yaml: clock-controller@5090000: clock-names: ['xo'] is too short .../sdm845-mtp.dt.yaml: clock-controller@5090000: clocks: [[26, 0]] is too short c) With just the "dts" from my series, you'll again get different yells from "make dtbs_check". I won't bother listing them, but they are similar to the above. d) With everything from my series, the "make dtbs_check" limited to the bindings files I'm touching is clean. --- * sdm845's ability to boot is unaffected by this series. I have tested booting sdm845 after this series and it's fine. Since nothing in this series touches the sdm845 clock drivers (only the bindings and the dts) that means that the new dts must work with the existing drivers. You can land the sdm845 dts any time after Stephen and Rob are happy with the bindings. * It should be fine to land the sc7180 dts file without waiting for the driver change. The dts here will work with both the dispcc/gpucc/videocc that's in clk-next and also the one that results from applying all of my patches. > > For sc7180 there's no usage of any of these clocks and this adds the > > first usage, so definitely no problem there. > > > > Once you've landed then Bjorn or Andy can pick up the dts. > > > > Do I need to apply these after Stephen picks the driver patches? Or are > they simply nop until the driver patches lands? The sdm845 patches are a nop until some future patch changes the Linux driver to start using them. I don't know of anyone planning to do that. The only real result from an sdm845 perspective will be making things "more correct" from a dts standpoint and keeping "make dtbs_check" happier. The sc7180 patches are OK to land even without the driver. They're not a nop and I've actually validated that I can bring the display/gpu up with them (even without the driver changes) on my downstream kernel. --- Sorry it's so confusing. Happy to try to clarify more if the above is still too hard to follow. -Doug
On Mon 03 Feb 12:48 PST 2020, Doug Anderson wrote:
[..]
> Sorry it's so confusing. Happy to try to clarify more if the above is
> still too hard to follow.
>
Thanks for the clarification! I will pick up the dts patches
Regards,
Bjorn
Quoting Douglas Anderson (2020-02-03 10:31:34)
> When I got my clock parenting slightly wrong I ended up with a crash
> that looked like this:
>
> Unable to handle kernel NULL pointer dereference at virtual
> address 0000000000000000
> ...
> pc : clk_hw_get_rate+0x14/0x44
> ...
> Call trace:
> clk_hw_get_rate+0x14/0x44
> _freq_tbl_determine_rate+0x94/0xfc
> clk_rcg2_determine_rate+0x2c/0x38
> clk_core_determine_round_nolock+0x4c/0x88
> clk_core_round_rate_nolock+0x6c/0xa8
> clk_core_round_rate_nolock+0x9c/0xa8
> clk_core_set_rate_nolock+0x70/0x180
> clk_set_rate+0x3c/0x6c
> of_clk_set_defaults+0x254/0x360
> platform_drv_probe+0x28/0xb0
> really_probe+0x120/0x2dc
> driver_probe_device+0x64/0xfc
> device_driver_attach+0x4c/0x6c
> __driver_attach+0xac/0xc0
> bus_for_each_dev+0x84/0xcc
> driver_attach+0x2c/0x38
> bus_add_driver+0xfc/0x1d0
> driver_register+0x64/0xf8
> __platform_driver_register+0x4c/0x58
> msm_drm_register+0x5c/0x60
> ...
>
> It turned out that clk_hw_get_parent_by_index() was returning NULL and
> we weren't checking. Let's check it so that we don't crash.
>
> Fixes: ac269395cdd8 ("clk: qcom: Convert to clk_hw based provider APIs")
> Signed-off-by: Douglas Anderson <dianders@chromium.org>
> Reviewed-by: Matthias Kaehlcke <mka@chromium.org>
> ---
Applied to clk-next
Quoting Douglas Anderson (2020-02-03 10:31:35)
> The qcom,dispcc bindings had a few problems with them:
>
> 1. They didn't specify all the clocks that dispcc is a client of.
> Specifically on sc7180 there are two clocks from the DSI PHY and
> two from the DP PHY. On sdm845 there are actually two DSI PHYs
> (each of which has two clocks) and an extra clock from the gcc.
> These all need to be specified.
>
> 2. The sdm845.dtsi has existed for quite some time without specifying
> the clocks. The Linux driver was relying on global names to match
> things up. While we should transition things, it should be noted
> in the bindings.
>
> 3. The names used the bindings for "xo" and "gpll0" didn't match the
> names that QC used for these clocks internally and this was causing
> confusion / difficulty with their code generation tools. Switched
> to the internal names to simplify everyone's lives. It's not quite
> as clean in a purist sense but it should avoid headaches. This
> officially changes the binding, but that seems OK in this case.
>
> Also note that I updated the example.
>
> Fixes: 5d28e44ba630 ("dt-bindings: clock: Add YAML schemas for the QCOM DISPCC clock bindings")
> Signed-off-by: Douglas Anderson <dianders@chromium.org>
> ---
Applied to clk-next
Quoting Douglas Anderson (2020-02-03 10:31:37)
> In the new world input clocks should be matched by ".fw_name". sc7180
> is new enough that no backward compatibility use of global names
> should be needed. Remove it.
>
> With a proper device tree and downstream display patches I have
> verified booting a sc7180 up and seeing the display after this patch.
>
> Fixes: dd3d06622138 ("clk: qcom: Add display clock controller driver for SC7180")
> Signed-off-by: Douglas Anderson <dianders@chromium.org>
> ---
Applied to clk-next
Quoting Douglas Anderson (2020-02-03 10:31:38)
> The test clock isn't in the bindings and apparently it's not used by
> anyone upstream. Remove it.
>
> Suggested-by: Stephen Boyd <swboyd@chromium.org>
> Signed-off-by: Douglas Anderson <dianders@chromium.org>
> ---
Applied to clk-next
Quoting Douglas Anderson (2020-02-03 10:31:39)
> It's nicer to use ARRAY_SIZE instead of hardcoding. Had we always
> been doing this it would have prevented a previous bug. See commit
> 74c31ff9c84a ("clk: qcom: gpu_cc_gmu_clk_src has 5 parents, not 6").
>
> Signed-off-by: Douglas Anderson <dianders@chromium.org>
> ---
Applied to clk-next
Quoting Douglas Anderson (2020-02-03 10:31:40)
> The qcom,gpucc bindings had a few problems with them:
>
> 1. When things were converted to yaml the name of the "gpll0 main"
> clock got changed from "gpll0" to "gpll0_main". Change it back for
> msm8998.
>
> 2. Apparently there is a push not to use purist aliases for clocks but
> instead to just use the internal Qualcomm names. For sdm845 and
> sc7180 (where the drivers haven't already been changed) move in
> this direction.
>
> Things were also getting complicated harder to deal with by jamming
> several SoCs into one file. Splitting simplifies things.
>
> Fixes: 5c6f3a36b913 ("dt-bindings: clock: Add YAML schemas for the QCOM GPUCC clock bindings")
> Signed-off-by: Douglas Anderson <dianders@chromium.org>
> ---
Applied to clk-next
Quoting Douglas Anderson (2020-02-03 10:31:42)
> The test clock isn't in the bindings and apparently it's not used by
> anyone upstream. Remove it.
>
> Suggested-by: Stephen Boyd <swboyd@chromium.org>
> Signed-off-by: Douglas Anderson <dianders@chromium.org>
> ---
Applied to clk-next
Quoting Douglas Anderson (2020-02-03 10:31:43)
> It's nicer to use ARRAY_SIZE instead of hardcoding. Had we always
> been doing this it would have prevented a previous bug. See commit
> 74c31ff9c84a ("clk: qcom: gpu_cc_gmu_clk_src has 5 parents, not 6").
>
> Signed-off-by: Douglas Anderson <dianders@chromium.org>
> ---
Applied to clk-next
Quoting Douglas Anderson (2020-02-03 10:31:44)
> This makes the qcom,videocc bindings match the recent changes to the
> dispcc and gpucc.
>
> 1. Switched to using "bi_tcxo" instead of "xo".
>
> 2. Adds a description for the XO clock. Not terribly important but
> nice if it cleanly matches its cousins.
>
> 3. Updates the example to use the symbolic name for the RPMH clock and
> also show that the real devices are currently using 2 address cells
> / size cells and fixes the spacing on the closing brace.
>
> 4. Split into 2 files. In this case they could probably share one
> file, but let's be consistent.
>
> Signed-off-by: Douglas Anderson <dianders@chromium.org>
> ---
>
> Changes in v4:
> - Added Rob's review tag.
I don't see it. I guess I should add it?
Quoting Douglas Anderson (2020-02-03 10:31:45)
> The test clock isn't in the bindings and apparently it's not used by
> anyone upstream. Remove it.
>
> Suggested-by: Stephen Boyd <swboyd@chromium.org>
> Signed-off-by: Douglas Anderson <dianders@chromium.org>
> ---
Applied to clk-next
Quoting Douglas Anderson (2020-02-03 10:31:46)
> It's nicer to use ARRAY_SIZE instead of hardcoding. Had we always
> been doing this it would have prevented a previous bug. See commit
> 74c31ff9c84a ("clk: qcom: gpu_cc_gmu_clk_src has 5 parents, not 6").
>
> Signed-off-by: Douglas Anderson <dianders@chromium.org>
> ---
Applied to clk-next
Hi, On Tue, Feb 4, 2020 at 9:49 AM Stephen Boyd <sboyd@kernel.org> wrote: > > Quoting Douglas Anderson (2020-02-03 10:31:44) > > This makes the qcom,videocc bindings match the recent changes to the > > dispcc and gpucc. > > > > 1. Switched to using "bi_tcxo" instead of "xo". > > > > 2. Adds a description for the XO clock. Not terribly important but > > nice if it cleanly matches its cousins. > > > > 3. Updates the example to use the symbolic name for the RPMH clock and > > also show that the real devices are currently using 2 address cells > > / size cells and fixes the spacing on the closing brace. > > > > 4. Split into 2 files. In this case they could probably share one > > file, but let's be consistent. > > > > Signed-off-by: Douglas Anderson <dianders@chromium.org> > > --- > > > > Changes in v4: > > - Added Rob's review tag. > > I don't see it. I guess I should add it? Whoops. I knew I'd mess something like this up. Looks like it got pushed without his tag but I don't think it's a big deal. For reference, he added his tag in: https://lore.kernel.org/r/CAL_Jsq+_2E-bAbP9F6VYkWRp0crEyRGa5peuwP58-PZniVny7w@mail.gmail.com I did manage to remove the "/bindings" from the ID as he requested, so at least I didn't mess up the important part... -Doug -Doug