arm64: dts: qcom: sc7x80: A smattering of misc dts cleanups + herobrine-rev1
Subsystems (4)
v3 cover arm64: dts: qcom: sc7x80: A smattering of misc dts cleanups + herobrine-rev1
This series is "v2" of my "smattering of misc dts cleanups" series plus v3 of the tail end of the series adding herobrine-rev1. I've set the version number to the larger of the two to (I hope) help allevitate confusion.
For the cleanups, there's not a lot holding this series together except that it fixes a smattering of random dts stuff that I noticed recently. There are not a lot of dependencies and some of the patches could be reordered if desired.
Hopefully these look OK and can be applied quickly to avoid conflicts with other work going on.
For herobrine-rev1, it can be noted that it's likely that with the introduction of -rev1 we can drop -rev0 support, but we'll keep it for now (though we won't try to "fit it in" and share code with it). This series is confirmed to boot atop the top of the linux qualcomm tree, commit a5ee6b7720cb ("Merge branches 'arm64-defconfig-for-5.18', 'arm64-for-5.18', 'dts-for-5.18', 'arm64-fixes-for-5.17' and 'dts-fixes-for-5.17' into for-next")
Changes in v3:
- Removed extra blank lines
- ("Fix sort order of dp_hot_plug_det") new for v3.
- ("Add edp_out port and HPD lines") new for v3.
- ("Move pcie1_clkreq pull / drive str to boards") new for v3.
- ("sc7280-idp: Disable pull from pcie1_clkreq") new for v3.
- ("Remove dp_hot_plug_det pull from SoC dtsi file") new for v3.
- ("Add a blank line in the dp node") new for v3.
- Rebased atop dts cleanup patches.
- Add regulator suffix as per dts cleanup patches.
- Set PCIe bias / pull as per dts cleanup patches.
- Add dp_hot_plug_det pull as per dts cleanup patches.
- Setup SD card same as dts cleanup patches.
- ("sc7280: Add the CPU compatible to the soc@0 node") new for v3.
- ("Remove "qcom,sc7280" from top-level") patch new for v3.
Changes in v2:
- Herobrine compatible on one line, not two
- Wording change in comments for components enabled per-board
- Always sort "bias" above "drive-strength" in pinctrl.
- Properly sort "hub_en" pinctrl.
- Two comments moved from multiline to single line.
- Space after "/delete-property/"
Douglas Anderson (14):
arm64: dts: qcom: sc7180-trogdor: Add "-regulator" suffix to
pp3300_hub
arm64: dts: qcom: sc7280-herobrine: Consistently add "-regulator"
suffix
arm64: dts: qcom: sc7280: Properly sort sdc pinctrl lines
arm64: dts: qcom: sc7280: Clean up sdc1 / sdc2 pinctrl
arm64: dts: qcom: sc7280-idp: No need for "input-enable" on sw_ctrl
arm64: dts: qcom: sc7280: Fix sort order of dp_hot_plug_det /
pcie1_clkreq_n
arm64: dts: qcom: sc7280: Add edp_out port and HPD lines
arm64: dts: qcom: sc7280: Move pcie1_clkreq pull / drive str to boards
arm64: dts: qcom: sc7280: Disable pull from pcie1_clkreq
arm64: dts: qcom: sc7280: Move dp_hot_plug_det pull from SoC dtsi file
arm64: dts: qcom: sc7280: Add a blank line in the dp node
arm64: dts: qcom: sc7280: Add herobrine-r1
arm64: dts: qcom: sc7280: Add the CPU compatible to the soc@0 node
arm64: dts: qcom: sc7280: Remove "qcom,sc7280" from top-level of
boards
arch/arm64/boot/dts/qcom/Makefile | 1 + arch/arm64/boot/dts/qcom/sc7180-trogdor.dtsi | 2 +- arch/arm64/boot/dts/qcom/sc7280-crd.dts | 2 +- .../qcom/sc7280-herobrine-herobrine-r0.dts | 97 +-- .../qcom/sc7280-herobrine-herobrine-r1.dts | 313 +++++++ .../arm64/boot/dts/qcom/sc7280-herobrine.dtsi | 785 ++++++++++++++++++ arch/arm64/boot/dts/qcom/sc7280-idp.dts | 2 +- arch/arm64/boot/dts/qcom/sc7280-idp.dtsi | 99 +-- arch/arm64/boot/dts/qcom/sc7280-idp2.dts | 2 +- arch/arm64/boot/dts/qcom/sc7280-qcard.dtsi | 547 ++++++++++++ arch/arm64/boot/dts/qcom/sc7280.dtsi | 182 ++-- 11 files changed, 1845 insertions(+), 187 deletions(-) create mode 100644 arch/arm64/boot/dts/qcom/sc7280-herobrine-herobrine-r1.dts create mode 100644 arch/arm64/boot/dts/qcom/sc7280-herobrine.dtsi create mode 100644 arch/arm64/boot/dts/qcom/sc7280-qcard.dtsi
Signature · 1 line
2.35.0.rc2.247.g8bbb082509-goog
PATCH v3 1/14 arm64: dts: qcom: sc7180-trogdor: Add "-regulator" suffix to pp3300_hub
All of the other fixed regulators have the "-regulator" suffix. Add it to pp3300_hub to match.
---
(no changes since v1)
arch/arm64/boot/dts/qcom/sc7180-trogdor.dtsi | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
arch/arm64/boot/dts/qcom/sc7180-trogdor.dtsi +1 -1
@@ -284,7 +284,7 @@ pp3300_fp_tp: pp3300-fp-tp-regulator { vin-supply = <&pp3300_a>; }; - pp3300_hub: pp3300-hub { + pp3300_hub: pp3300-hub-regulator { compatible = "regulator-fixed"; regulator-name = "pp3300_hub";
Signature · 1 line
2.35.0.rc2.247.g8bbb082509-goog
Re: [PATCH v3 01/14] arm64: dts: qcom: sc7180-trogdor: Add "-regulator" suffix to pp3300_hub
Quoting Douglas Anderson (2022-02-02 13:23:35)
Quoted 6 lines
All of the other fixed regulators have the "-regulator" suffix. Add it to
pp3300_hubto match.Signed-off-by Douglas Anderson <[email protected]>Reviewed-by Matthias Kaehlcke <[email protected]>---
PATCH v3 2/14 arm64: dts: qcom: sc7280-herobrine: Consistently add "-regulator" suffix
In reply to Douglas AndersonSome of the fixed regulators were missing the "-regulator" suffix. Add it to be consistent within the file and consistent with the fixed regulators in sc7180-trogdor.
---
(no changes since v1)
.../boot/dts/qcom/sc7280-herobrine-herobrine-r0.dts | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-)
arch/arm64/boot/dts/qcom/sc7280-herobrine-herobrine-r0.dts +6 -6
@@ -177,7 +177,7 @@ pp3300_tp: pp3300-tp-regulator { vin-supply = <&pp3300_z1>; }; - pp2850_uf_cam: pp2850-uf-cam { + pp2850_uf_cam: pp2850-uf-cam-regulator { compatible = "regulator-fixed"; regulator-name = "pp2850_uf_cam";
@@ -192,7 +192,7 @@ pp2850_uf_cam: pp2850-uf-cam { vin-supply = <&pp3300_cam>; }; - pp2850_vcm_wf_cam: pp2850-vcm-wf-cam { + pp2850_vcm_wf_cam: pp2850-vcm-wf-cam-regulator { compatible = "regulator-fixed"; regulator-name = "pp2850_vcm_wf_cam";
@@ -207,7 +207,7 @@ pp2850_vcm_wf_cam: pp2850-vcm-wf-cam { vin-supply = <&pp3300_cam>; }; - pp2850_wf_cam: pp2850-wf-cam { + pp2850_wf_cam: pp2850-wf-cam-regulator { compatible = "regulator-fixed"; regulator-name = "pp2850_wf_cam";
@@ -251,7 +251,7 @@ pp1800_fp: pp1800-fp-regulator { status = "disabled"; }; - pp1800_uf_cam: pp1800-uf-cam { + pp1800_uf_cam: pp1800-uf-cam-regulator { compatible = "regulator-fixed"; regulator-name = "pp1800_uf_cam";
@@ -271,7 +271,7 @@ pp1800_uf_cam: pp1800-uf-cam { vin-supply = <&pp1800_l19b>; }; - pp1800_wf_cam: pp1800-wf-cam { + pp1800_wf_cam: pp1800-wf-cam-regulator { compatible = "regulator-fixed"; regulator-name = "pp1800_wf_cam";
@@ -291,7 +291,7 @@ pp1800_wf_cam: pp1800-wf-cam { vin-supply = <&pp1800_l19b>; }; - pp1200_wf_cam: pp1200-wf-cam { + pp1200_wf_cam: pp1200-wf-cam-regulator { compatible = "regulator-fixed"; regulator-name = "pp1200_wf_cam";
Signature · 1 line
2.35.0.rc2.247.g8bbb082509-goog
Re: [PATCH v3 02/14] arm64: dts: qcom: sc7280-herobrine: Consistently add "-regulator" suffix
Quoting Douglas Anderson (2022-02-02 13:23:36)
Quoted 7 lines
Some of the fixed regulators were missing the "-regulator" suffix. Add it to be consistent within the file and consistent with the fixed regulators in sc7180-trogdor.
Signed-off-by Douglas Anderson <[email protected]>Reviewed-by Matthias Kaehlcke <[email protected]>---
PATCH v3 3/14 arm64: dts: qcom: sc7280: Properly sort sdc pinctrl lines
In reply to Douglas AndersonThe sdc1 / sdc2 pinctrl lines were randomly stuffed in the middle of the qup pinctrl lines. Sort them properly. This is a no-op change. Just code movement.
---
(no changes since v1)
arch/arm64/boot/dts/qcom/sc7280.dtsi | 154 +++++++++++++-------------- 1 file changed, 77 insertions(+), 77 deletions(-)
arch/arm64/boot/dts/qcom/sc7280.dtsi +77 -77
@@ -3783,83 +3783,6 @@ qup_uart7_rx: qup-uart7-rx { function = "qup07"; }; - sdc1_on: sdc1-on { - clk { - pins = "sdc1_clk"; - }; - - cmd { - pins = "sdc1_cmd"; - }; - - data { - pins = "sdc1_data"; - }; - - rclk { - pins = "sdc1_rclk"; - }; - }; - - sdc1_off: sdc1-off { - clk { - pins = "sdc1_clk"; - drive-strength = <2>; - bias-bus-hold; - }; - - cmd { - pins = "sdc1_cmd"; - drive-strength = <2>; - bias-bus-hold; - }; - - data { - pins = "sdc1_data"; - drive-strength = <2>; - bias-bus-hold; - }; - - rclk { - pins = "sdc1_rclk"; - bias-bus-hold; - }; - }; - - sdc2_on: sdc2-on { - clk { - pins = "sdc2_clk"; - }; - - cmd { - pins = "sdc2_cmd"; - }; - - data { - pins = "sdc2_data"; - }; - }; - - sdc2_off: sdc2-off { - clk { - pins = "sdc2_clk"; - drive-strength = <2>; - bias-bus-hold; - }; - - cmd { - pins ="sdc2_cmd"; - drive-strength = <2>; - bias-bus-hold; - }; - - data { - pins ="sdc2_data"; - drive-strength = <2>; - bias-bus-hold; - }; - }; - qup_uart8_cts: qup-uart8-cts { pins = "gpio32"; function = "qup10";
@@ -4019,6 +3942,83 @@ qup_uart15_rx: qup-uart15-rx { pins = "gpio63"; function = "qup17"; }; + + sdc1_on: sdc1-on { + clk { + pins = "sdc1_clk"; + }; + + cmd { + pins = "sdc1_cmd"; + }; + + data { + pins = "sdc1_data"; + }; + + rclk { + pins = "sdc1_rclk"; + }; + }; + + sdc1_off: sdc1-off { + clk { + pins = "sdc1_clk"; + drive-strength = <2>; + bias-bus-hold; + }; + + cmd { + pins = "sdc1_cmd"; + drive-strength = <2>; + bias-bus-hold; + }; + + data { + pins = "sdc1_data"; + drive-strength = <2>; + bias-bus-hold; + }; + + rclk { + pins = "sdc1_rclk"; + bias-bus-hold; + }; + }; + + sdc2_on: sdc2-on { + clk { + pins = "sdc2_clk"; + }; + + cmd { + pins = "sdc2_cmd"; + }; + + data { + pins = "sdc2_data"; + }; + }; + + sdc2_off: sdc2-off { + clk { + pins = "sdc2_clk"; + drive-strength = <2>; + bias-bus-hold; + }; + + cmd { + pins ="sdc2_cmd"; + drive-strength = <2>; + bias-bus-hold; + }; + + data { + pins ="sdc2_data"; + drive-strength = <2>; + bias-bus-hold; + }; + }; }; imem@146a5000 {
Signature · 1 line
2.35.0.rc2.247.g8bbb082509-goog
Re: [PATCH v3 03/14] arm64: dts: qcom: sc7280: Properly sort sdc pinctrl lines
Quoting Douglas Anderson (2022-02-02 13:23:37)
Quoted 7 lines
The sdc1 / sdc2 pinctrl lines were randomly stuffed in the middle of the qup pinctrl lines. Sort them properly. This is a no-op change. Just code movement.
Signed-off-by Douglas Anderson <[email protected]>Reviewed-by Matthias Kaehlcke <[email protected]>---
PATCH v3 4/14 arm64: dts: qcom: sc7280: Clean up sdc1 / sdc2 pinctrl
In reply to Douglas AndersonThis patch makes a few improvements to the way that sdc1 / sdc2 pinctrl is specified on sc7280:
1. There's no reason to "group" the sdc pins into one overarching node and there's a downside: we have to replicate the hierarchy in the board device tree files. Let's clean this up.
2. There's really not a lot of reason not to list the "pinctrl" for sdc1 (eMMC) in the SoC dtsi file. These aren't GPIO pins and everyone's going to specify the same pins.
3. Even though it's likely that boards will need to override pinctrl for sdc2 (SD card) to add the card detect GPIO, we can be symmetric and add it to the SoC dsti file.
4. Let's get rid of the word "on" from the normal config and add a "sleep" suffix to the sleep config. This looks cleaner to me.
This is intended to be a no-op change but it could plausibly change behavior depending on how the pinctrl code parses things. One thing to note is that "SD card detect" is explicitly listed now as keeping its pull enabled in sleep since we still want to detect card insertions even if the controller is suspended (because no card is inserted). The pinctrl framework likely did this anyway, but it's nice to see it explicit.
---
Changes in v3:
- Removed extra blank lines
.../qcom/sc7280-herobrine-herobrine-r0.dts | 73 +++++------ arch/arm64/boot/dts/qcom/sc7280-idp.dtsi | 91 +++++++------- arch/arm64/boot/dts/qcom/sc7280.dtsi | 117 +++++++++--------- 3 files changed, 133 insertions(+), 148 deletions(-)
arch/arm64/boot/dts/qcom/sc7280-herobrine-herobrine-r0.dts +33 -40
@@ -676,9 +676,6 @@ &qupv3_id_1 { &sdhc_1 { status = "okay"; - pinctrl-names = "default", "sleep"; - pinctrl-0 = <&sdc1_on>; - pinctrl-1 = <&sdc1_off>; vmmc-supply = <&pp2950_l7b>; vqmmc-supply = <&pp1800_l19b>; };
@@ -686,9 +683,8 @@ &sdhc_1 { &sdhc_2 { status = "okay"; - pinctrl-names = "default", "sleep"; - pinctrl-0 = <&sdc2_on>; - pinctrl-1 = <&sdc2_off>; + pinctrl-0 = <&sdc2_clk>, <&sdc2_cmd>, <&sdc2_data>, <&sd_cd>; + pinctrl-1 = <&sdc2_clk_sleep>, <&sdc2_cmd_sleep>, <&sdc2_data_sleep>, <&sd_cd>; vmmc-supply = <&pp2950_l9c>; vqmmc-supply = <&ppvar_l6c>;
@@ -883,47 +879,38 @@ &qup_uart7_rx { bias-pull-up; }; -&sdc1_on { - clk { - bias-disable; - drive-strength = <16>; - }; - - cmd { - bias-pull-up; - drive-strength = <10>; - }; +&sdc1_clk { + bias-disable; + drive-strength = <16>; +}; - data { - bias-pull-up; - drive-strength = <10>; - }; +&sdc1_cmd { + bias-pull-up; + drive-strength = <10>; +}; - rclk { - bias-pull-down; - }; +&sdc1_data { + bias-pull-up; + drive-strength = <10>; }; -&sdc2_on { - clk { - bias-disable; - drive-strength = <16>; - }; +&sdc1_rclk { + bias-pull-down; +}; - cmd { - bias-pull-up; - drive-strength = <10>; - }; +&sdc2_clk { + bias-disable; + drive-strength = <16>; +}; - data { - bias-pull-up; - drive-strength = <10>; - }; +&sdc2_cmd { + bias-pull-up; + drive-strength = <10>; +}; - sd-cd { - pins = "gpio91"; - bias-pull-up; - }; +&sdc2_data { + bias-pull-up; + drive-strength = <10>; }; /* PINCTRL - board-specific pinctrl */
@@ -1311,6 +1298,12 @@ qup_uart7_sleep_tx: qup-uart7-sleep-tx { bias-pull-up; }; + sd_cd: sd-cd { + pins = "gpio91"; + function = "gpio"; + bias-pull-up; + }; + tp_int_odl: tp-int-odl { pins = "gpio102"; function = "gpio";
arch/arm64/boot/dts/qcom/sc7280-idp.dtsi +42 -49
120 lines not on this page · raw message
arch/arm64/boot/dts/qcom/sc7280.dtsi +58 -59
151 lines not on this page · raw message
Signature · 1 line
2.35.0.rc2.247.g8bbb082509-goog
Re: [PATCH v3 04/14] arm64: dts: qcom: sc7280: Clean up sdc1 / sdc2 pinctrl
Quoted 18 lines · Douglas Anderson
This patch makes a few improvements to the way that sdc1 / sdc2 pinctrl is specified on sc7280:
1. There's no reason to "group" the sdc pins into one overarching node and there's a downside: we have to replicate the hierarchy in the board device tree files. Let's clean this up.
2. There's really not a lot of reason not to list the "pinctrl" for sdc1 (eMMC) in the SoC dtsi file. These aren't GPIO pins and everyone's going to specify the same pins.
3. Even though it's likely that boards will need to override pinctrl for sdc2 (SD card) to add the card detect GPIO, we can be symmetric and add it to the SoC dsti file.
4. Let's get rid of the word "on" from the normal config and add a "sleep" suffix to the sleep config. This looks cleaner to me.
Quoted 7 lines
This is intended to be a no-op change but it could plausibly change behavior depending on how the pinctrl code parses things. One thing to note is that "SD card detect" is explicitly listed now as keeping its pull enabled in sleep since we still want to detect card insertions even if the controller is suspended (because no card is inserted). The pinctrl framework likely did this anyway, but it's nice to see it explicit.
Re: [PATCH v3 04/14] arm64: dts: qcom: sc7280: Clean up sdc1 / sdc2 pinctrl
In reply to Douglas AndersonQuoting Douglas Anderson (2022-02-02 13:23:38)
Quoted 26 lines
This patch makes a few improvements to the way that sdc1 / sdc2 pinctrl is specified on sc7280:
1. There's no reason to "group" the sdc pins into one overarching node and there's a downside: we have to replicate the hierarchy in the board device tree files. Let's clean this up.
2. There's really not a lot of reason not to list the "pinctrl" for sdc1 (eMMC) in the SoC dtsi file. These aren't GPIO pins and everyone's going to specify the same pins.
3. Even though it's likely that boards will need to override pinctrl for sdc2 (SD card) to add the card detect GPIO, we can be symmetric and add it to the SoC dsti file.
4. Let's get rid of the word "on" from the normal config and add a "sleep" suffix to the sleep config. This looks cleaner to me.
This is intended to be a no-op change but it could plausibly change behavior depending on how the pinctrl code parses things. One thing to note is that "SD card detect" is explicitly listed now as keeping its pull enabled in sleep since we still want to detect card insertions even if the controller is suspended (because no card is inserted). The pinctrl framework likely did this anyway, but it's nice to see it explicit.
Quoted 2 lines
Signed-off-by Douglas Anderson <[email protected]>---
PATCH v3 5/14 arm64: dts: qcom: sc7280-idp: No need for "input-enable" on sw_ctrl
In reply to Douglas AndersonSpecifying "input-enable" on a MSM GPIO is a no-op for the most part. The only thing it really does is to explicitly force the output of a GPIO to be disabled right at the point of a pinctrl transition. We don't need to do this and we don't typically specify "input-enable" unless there's a good reason to. Remove it.
---
(no changes since v1)
arch/arm64/boot/dts/qcom/sc7280-idp.dtsi | 1 - 1 file changed, 1 deletion(-)
arch/arm64/boot/dts/qcom/sc7280-idp.dtsi +0 -1
@@ -534,7 +534,6 @@ sd_cd: sd-cd { sw_ctrl: sw-ctrl { pins = "gpio86"; function = "gpio"; - input-enable; bias-pull-down; }; };
Signature · 1 line
2.35.0.rc2.247.g8bbb082509-goog
Re: [PATCH v3 05/14] arm64: dts: qcom: sc7280-idp: No need for "input-enable" on sw_ctrl
Quoting Douglas Anderson (2022-02-02 13:23:39)
Quoted 9 lines
Specifying "input-enable" on a MSM GPIO is a no-op for the most part. The only thing it really does is to explicitly force the output of a GPIO to be disabled right at the point of a pinctrl transition. We don't need to do this and we don't typically specify "input-enable" unless there's a good reason to. Remove it.
Signed-off-by Douglas Anderson <[email protected]>Reviewed-by Matthias Kaehlcke <[email protected]>---
PATCH v3 6/14 arm64: dts: qcom: sc7280: Fix sort order of dp_hot_plug_det / pcie1_clkreq_n
In reply to Douglas AndersonThe two nodes were mis-sorted. Reorder. This is a no-op change.
---
Changes in v3:
- ("Fix sort order of dp_hot_plug_det") new for v3.
arch/arm64/boot/dts/qcom/sc7280.dtsi | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-)
arch/arm64/boot/dts/qcom/sc7280.dtsi +6 -6
@@ -3271,6 +3271,12 @@ tlmm: pinctrl@f100000 { gpio-ranges = <&tlmm 0 0 175>; wakeup-parent = <&pdc>; + dp_hot_plug_det: dp-hot-plug-det { + pins = "gpio47"; + function = "dp_hot"; + bias-disable; + }; + pcie1_clkreq_n: pcie1-clkreq-n { pins = "gpio79"; function = "pcie1_clkreqn";
@@ -3278,12 +3284,6 @@ pcie1_clkreq_n: pcie1-clkreq-n { bias-pull-up; }; - dp_hot_plug_det: dp-hot-plug-det { - pins = "gpio47"; - function = "dp_hot"; - bias-disable; - }; - qspi_clk: qspi-clk { pins = "gpio14"; function = "qspi_clk";
Signature · 1 line
2.35.0.rc2.247.g8bbb082509-goog
Re: [PATCH v3 06/14] arm64: dts: qcom: sc7280: Fix sort order of dp_hot_plug_det / pcie1_clkreq_n
Quoted 3 lines · Douglas Anderson
The two nodes were mis-sorted. Reorder. This is a no-op change.
Signed-off-by Douglas Anderson <[email protected]>
Re: [PATCH v3 06/14] arm64: dts: qcom: sc7280: Fix sort order of dp_hot_plug_det / pcie1_clkreq_n
In reply to Douglas AndersonQuoting Douglas Anderson (2022-02-02 13:23:40)
Quoted 4 lines
The two nodes were mis-sorted. Reorder. This is a no-op change.
Signed-off-by Douglas Anderson <[email protected]>---
PATCH v3 7/14 arm64: dts: qcom: sc7280: Add edp_out port and HPD lines
In reply to Douglas AndersonLike dp_out, we should have defined edp_out in sc7280.dtsi so we don't need to do this in the board files.
Like dp_hot_plug_det, we should define edp_hot_plug_det in sc7280.dtsi.
We should set the default pinctrl for edp_hot_plug_det in sc7280.dtsi. NOTE: this is _unlike_ the dp_hot_plug_det. It is reasonable that in some boards the dedicated DP Hot Plug Detect will not be hooked up in favor of Type C mechanisms. This is unlike eDP where the Hot Plug Detect line (which functions as "panel ready" in eDP) is highly likely to be used by boards.
---
Changes in v3:
- ("Add edp_out port and HPD lines") new for v3.
arch/arm64/boot/dts/qcom/sc7280.dtsi | 13 +++++++++++++ 1 file changed, 13 insertions(+)
arch/arm64/boot/dts/qcom/sc7280.dtsi +13 -0
@@ -3010,6 +3010,8 @@ mdss_dsi_phy: phy@ae94400 { mdss_edp: edp@aea0000 { compatible = "qcom,sc7280-edp"; + pinctrl-names = "default"; + pinctrl-0 = <&edp_hot_plug_det>; reg = <0 0xaea0000 0 0x200>, <0 0xaea0200 0 0x200>,
@@ -3052,12 +3054,18 @@ mdss_edp: edp@aea0000 { ports { #address-cells = <1>; #size-cells = <0>; + port@0 { reg = <0>; edp_in: endpoint { remote-endpoint = <&dpu_intf5_out>; }; }; + + port@1 { + reg = <1>; + edp_out: endpoint { }; + }; }; edp_opp_table: opp-table {
@@ -3277,6 +3285,11 @@ dp_hot_plug_det: dp-hot-plug-det { bias-disable; }; + edp_hot_plug_det: edp-hot-plug-det { + pins = "gpio60"; + function = "edp_hot"; + }; + pcie1_clkreq_n: pcie1-clkreq-n { pins = "gpio79"; function = "pcie1_clkreqn";
Signature · 1 line
2.35.0.rc2.247.g8bbb082509-goog
Re: [PATCH v3 07/14] arm64: dts: qcom: sc7280: Add edp_out port and HPD lines
Quoted 6 lines · Douglas Anderson
Like
dp_out, we should have definededp_outinsc7280.dtsiso we don't need to do this in the board files.Like
dp_hot_plug_det, we should defineedp_hot_plug_detinsc7280.dtsi.
Quoted 8 lines
We should set the default pinctrl for
edp_hot_plug_detinsc7280.dtsi. NOTE: this is _unlike_ thedp_hot_plug_det. It is reasonable that in some boards the dedicated DP Hot Plug Detect will not be hooked up in favor of Type C mechanisms. This is unlike eDP where the Hot Plug Detect line (which functions as "panel ready" in eDP) is highly likely to be used by boards.Signed-off-by Douglas Anderson <[email protected]>
Re: [PATCH v3 07/14] arm64: dts: qcom: sc7280: Add edp_out port and HPD lines
In reply to Douglas AndersonQuoting Douglas Anderson (2022-02-02 13:23:41)
Quoted 13 lines
Like
dp_out, we should have definededp_outinsc7280.dtsiso we don't need to do this in the board files.Like
dp_hot_plug_det, we should defineedp_hot_plug_detinsc7280.dtsi.We should set the default pinctrl for
edp_hot_plug_detinsc7280.dtsi. NOTE: this is _unlike_ thedp_hot_plug_det. It is reasonable that in some boards the dedicated DP Hot Plug Detect will not be hooked up in favor of Type C mechanisms. This is unlike eDP where the Hot Plug Detect line (which functions as "panel ready" in eDP) is highly likely to be used by boards.
Quoted 2 lines
Signed-off-by Douglas Anderson <[email protected]>---
PATCH v3 8/14 arm64: dts: qcom: sc7280: Move pcie1_clkreq pull / drive str to boards
In reply to Douglas AndersonPullups and drive strength don't belong in the SoC dtsi file. Move to the board file.
---
Changes in v3:
- ("Move pcie1_clkreq pull / drive str to boards") new for v3.
arch/arm64/boot/dts/qcom/sc7280-herobrine-herobrine-r0.dts | 5 +++++ arch/arm64/boot/dts/qcom/sc7280-idp.dtsi | 5 +++++ arch/arm64/boot/dts/qcom/sc7280.dtsi | 2 -- 3 files changed, 10 insertions(+), 2 deletions(-)
arch/arm64/boot/dts/qcom/sc7280-herobrine-herobrine-r0.dts +5 -0
@@ -826,6 +826,11 @@ &usb_2_hsphy { /* PINCTRL - additions to nodes defined in sc7280.dtsi */ +&pcie1_clkreq_n { + bias-pull-up; + drive-strength = <2>; +}; + &qspi_cs0 { bias-disable; };
arch/arm64/boot/dts/qcom/sc7280-idp.dtsi +5 -0
@@ -366,6 +366,11 @@ key_vol_up_default: key-vol-up-default { }; }; +&pcie1_clkreq_n { + bias-pull-up; + drive-strength = <2>; +}; + &qspi_cs0 { bias-disable; };
arch/arm64/boot/dts/qcom/sc7280.dtsi +0 -2
@@ -3293,8 +3293,6 @@ edp_hot_plug_det: edp-hot-plug-det { pcie1_clkreq_n: pcie1-clkreq-n { pins = "gpio79"; function = "pcie1_clkreqn"; - drive-strength = <2>; - bias-pull-up; }; qspi_clk: qspi-clk {
Signature · 1 line
2.35.0.rc2.247.g8bbb082509-goog
Re: [PATCH v3 08/14] arm64: dts: qcom: sc7280: Move pcie1_clkreq pull / drive str to boards
Quoted 4 lines · Douglas Anderson
Pullups and drive strength don't belong in the SoC dtsi file. Move to the board file.
Signed-off-by Douglas Anderson <[email protected]>
Re: [PATCH v3 08/14] arm64: dts: qcom: sc7280: Move pcie1_clkreq pull / drive str to boards
In reply to Douglas AndersonQuoting Douglas Anderson (2022-02-02 13:23:42)
Quoted 5 lines
Pullups and drive strength don't belong in the SoC dtsi file. Move to the board file.
Signed-off-by Douglas Anderson <[email protected]>---
PATCH v3 9/14 arm64: dts: qcom: sc7280: Disable pull from pcie1_clkreq
In reply to Douglas AndersonI believe that the PCIe clkreq pin is an output. That means we shouldn't have a pull enabled for it. Turn it off.
---
Changes in v3:
- ("sc7280-idp: Disable pull from pcie1_clkreq") new for v3.
arch/arm64/boot/dts/qcom/sc7280-herobrine-herobrine-r0.dts | 2 +- arch/arm64/boot/dts/qcom/sc7280-idp.dtsi | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-)
arch/arm64/boot/dts/qcom/sc7280-herobrine-herobrine-r0.dts +1 -1
@@ -827,7 +827,7 @@ &usb_2_hsphy { /* PINCTRL - additions to nodes defined in sc7280.dtsi */ &pcie1_clkreq_n { - bias-pull-up; + bias-disable; drive-strength = <2>; };
arch/arm64/boot/dts/qcom/sc7280-idp.dtsi +1 -1
@@ -367,7 +367,7 @@ key_vol_up_default: key-vol-up-default { }; &pcie1_clkreq_n { - bias-pull-up; + bias-disable; drive-strength = <2>; };
Signature · 1 line
2.35.0.rc2.247.g8bbb082509-goog
Re: [PATCH v3 09/14] arm64: dts: qcom: sc7280: Disable pull from pcie1_clkreq
Quoted 4 lines · Douglas Anderson
I believe that the PCIe clkreq pin is an output. That means we shouldn't have a pull enabled for it. Turn it off.
Signed-off-by Douglas Anderson <[email protected]>
Re: [PATCH v3 09/14] arm64: dts: qcom: sc7280: Disable pull from pcie1_clkreq
In reply to Douglas AndersonQuoting Douglas Anderson (2022-02-02 13:23:43)
Quoted 2 lines
I believe that the PCIe clkreq pin is an output. That means we shouldn't have a pull enabled for it. Turn it off.
It sounds like it's a request from the PCI device to the PCI phy that the clk should be on. I googled pcie clkreq open drain and this pdf[1] says
"The CLKREQ# signal is an open drain, active low signal that is driven low by the PCI Express M.2 add-I Card function to request that the PCI Express reference clock be available (active clock state) in order to allow the PCI Express interface to send/receive data"
so presumably if there isn't an external pull on the signal the open drain feature will not work and the PCIe device won't be able to drive it low.
Re: [PATCH v3 09/14] arm64: dts: qcom: sc7280: Disable pull from pcie1_clkreq
Hi,
Quoted 14 lines · Stephen Boyd
Quoting Douglas Anderson (2022-02-02 13:23:43)
Quoted 2 lines
I believe that the PCIe clkreq pin is an output. That means we shouldn't have a pull enabled for it. Turn it off.
It sounds like it's a request from the PCI device to the PCI phy that the clk should be on. I googled pcie clkreq open drain and this pdf[1] says
"The CLKREQ# signal is an open drain, active low signal that is driven low by the PCI Express M.2 add-I Card function to request that the PCI Express reference clock be available (active clock state) in order to allow the PCI Express interface to send/receive data"
Quoted 5 lines
so presumably if there isn't an external pull on the signal the open drain feature will not work and the PCIe device won't be able to drive it low.
Yeah, I had some trouble figuring this out too, so if someone knows better than me then I'm more than happy to take advice here. I thought I had found something claiming that "clkreq" was an output and on the schematic I have from Qualcomm it shows an arrow going out from the SoC for this signal indicating that it's an output from the SoC. Of course, those arrows are notoriously wrong but at least it's one piece of evidence that someone thought this was an output from the SoC.
Hrm, but I just checked the sc7280 "datasheet" which claims that this is an input. Sigh.
I guess the options are:
* If we're sure this is an input to the SoC then I think we should remove the drive-strength, right?
* If we don't know then I guess we can leave both?
In any case, for now we can just drop this patch?
-Doug
Re: [PATCH v3 09/14] arm64: dts: qcom: sc7280: Disable pull from pcie1_clkreq
Quoting Doug Anderson (2022-02-03 13:53:09)
Quoted 31 lines
Hi,
Quoted 14 lines · Stephen Boyd
Quoting Douglas Anderson (2022-02-02 13:23:43)
Quoted 2 lines
I believe that the PCIe clkreq pin is an output. That means we shouldn't have a pull enabled for it. Turn it off.
It sounds like it's a request from the PCI device to the PCI phy that the clk should be on. I googled pcie clkreq open drain and this pdf[1] says
"The CLKREQ# signal is an open drain, active low signal that is driven low by the PCI Express M.2 add-I Card function to request that the PCI Express reference clock be available (active clock state) in order to allow the PCI Express interface to send/receive data"
Quoted 5 lines
so presumably if there isn't an external pull on the signal the open drain feature will not work and the PCIe device won't be able to drive it low.
Yeah, I had some trouble figuring this out too, so if someone knows better than me then I'm more than happy to take advice here. I thought I had found something claiming that "clkreq" was an output and on the schematic I have from Qualcomm it shows an arrow going out from the SoC for this signal indicating that it's an output from the SoC. Of course, those arrows are notoriously wrong but at least it's one piece of evidence that someone thought this was an output from the SoC.
Quoted 7 lines
Hrm, but I just checked the sc7280 "datasheet" which claims that this is an input. Sigh.
I guess the options are:
* If we're sure this is an input to the SoC then I think we should remove the drive-strength, right?
* If we don't know then I guess we can leave both?
I'll wait for qcom folks to confirm. Maybe it's bidirectional because it is an open drain signal. I'm showing my cards that I'm no PCIe expert :)
Quoted 4 lines
In any case, for now we can just drop this patch?
Sounds good to me. It needs to be resolved through for herobrine-r1?
Re: [PATCH v3 09/14] arm64: dts: qcom: sc7280: Disable pull from pcie1_clkreq
Hi,
Quoted 36 lines · Stephen Boyd
Quoting Doug Anderson (2022-02-03 13:53:09)
Quoted 34 lines
Hi,
Quoted 14 lines · Stephen Boyd
Quoting Douglas Anderson (2022-02-02 13:23:43)
Quoted 2 lines
I believe that the PCIe clkreq pin is an output. That means we shouldn't have a pull enabled for it. Turn it off.
It sounds like it's a request from the PCI device to the PCI phy that the clk should be on. I googled pcie clkreq open drain and this pdf[1] says
"The CLKREQ# signal is an open drain, active low signal that is driven low by the PCI Express M.2 add-I Card function to request that the PCI Express reference clock be available (active clock state) in order to allow the PCI Express interface to send/receive data"
Quoted 5 lines
so presumably if there isn't an external pull on the signal the open drain feature will not work and the PCIe device won't be able to drive it low.
Yeah, I had some trouble figuring this out too, so if someone knows better than me then I'm more than happy to take advice here. I thought I had found something claiming that "clkreq" was an output and on the schematic I have from Qualcomm it shows an arrow going out from the SoC for this signal indicating that it's an output from the SoC. Of course, those arrows are notoriously wrong but at least it's one piece of evidence that someone thought this was an output from the SoC.
Hrm, but I just checked the sc7280 "datasheet" which claims that this is an input. Sigh.
Quoted 7 lines
Quoted 4 lines
I guess the options are:
* If we're sure this is an input to the SoC then I think we should remove the drive-strength, right?
* If we don't know then I guess we can leave both?I'll wait for qcom folks to confirm. Maybe it's bidirectional because it is an open drain signal. I'm showing my cards that I'm no PCIe expert :)
Ah ha! I searched some more and found a Qualcomm PCIe user guide on this!
CLKREQ# signal properties – Bi-directional clock request signals whether the RC or AP requires control
So it sounds as if leaving it as pull-up and having drive-strength as 2 is right. tl;dr: drop this patch from the series...
Quoted 4 lines
Quoted 2 lines
In any case, for now we can just drop this patch?
Sounds good to me. It needs to be resolved through for herobrine-r1?
Yup. I responded to that patch and for now I'll wait for Bjorn to give me direction on how to handle it.
-Doug
PATCH v3 10/14 arm64: dts: qcom: sc7280: Move dp_hot_plug_det pull from SoC dtsi file
In reply to Douglas AndersonPulls should be in the board files, not in the SoC dtsi file. Remove. Even though the sc7280 boards don't currently refer to dp_hot_plug_det, let's re-add the pulls there just to keep this as a no-op change. If boards don't need this / don't want it later then we can remove it from them.
---
Changes in v3:
- ("Remove dp_hot_plug_det pull from SoC dtsi file") new for v3.
arch/arm64/boot/dts/qcom/sc7280-herobrine-herobrine-r0.dts | 4 ++++ arch/arm64/boot/dts/qcom/sc7280-idp.dtsi | 4 ++++ arch/arm64/boot/dts/qcom/sc7280.dtsi | 1 - 3 files changed, 8 insertions(+), 1 deletion(-)
arch/arm64/boot/dts/qcom/sc7280-herobrine-herobrine-r0.dts +4 -0
@@ -826,6 +826,10 @@ &usb_2_hsphy { /* PINCTRL - additions to nodes defined in sc7280.dtsi */ +&dp_hot_plug_det { + bias-disable; +}; + &pcie1_clkreq_n { bias-disable; drive-strength = <2>;
arch/arm64/boot/dts/qcom/sc7280-idp.dtsi +4 -0
@@ -355,6 +355,10 @@ bluetooth: bluetooth { /* PINCTRL - additions to nodes defined in sc7280.dtsi */ +&dp_hot_plug_det { + bias-disable; +}; + &pm7325_gpios { key_vol_up_default: key-vol-up-default { pins = "gpio6";
arch/arm64/boot/dts/qcom/sc7280.dtsi +0 -1
@@ -3282,7 +3282,6 @@ tlmm: pinctrl@f100000 { dp_hot_plug_det: dp-hot-plug-det { pins = "gpio47"; function = "dp_hot"; - bias-disable; }; edp_hot_plug_det: edp-hot-plug-det {
Signature · 1 line
2.35.0.rc2.247.g8bbb082509-goog
Re: [PATCH v3 10/14] arm64: dts: qcom: sc7280: Move dp_hot_plug_det pull from SoC dtsi file
Quoted 7 lines · Douglas Anderson
Pulls should be in the board files, not in the SoC dtsi file. Remove. Even though the sc7280 boards don't currently refer to
dp_hot_plug_det, let's re-add the pulls there just to keep this as a no-op change. If boards don't need this / don't want it later then we can remove it from them.Signed-off-by Douglas Anderson <[email protected]>
Re: [PATCH v3 10/14] arm64: dts: qcom: sc7280: Move dp_hot_plug_det pull from SoC dtsi file
In reply to Douglas AndersonQuoting Douglas Anderson (2022-02-02 13:23:44)
Quoted 8 lines
Pulls should be in the board files, not in the SoC dtsi file. Remove. Even though the sc7280 boards don't currently refer to
dp_hot_plug_det, let's re-add the pulls there just to keep this as a no-op change. If boards don't need this / don't want it later then we can remove it from them.Signed-off-by Douglas Anderson <[email protected]>---
PATCH v3 11/14 arm64: dts: qcom: sc7280: Add a blank line in the dp node
In reply to Douglas AndersonIt's weird that there's a blank line between the two port nodes but not between the attributes and the first port node. Add an extra blank line to make it look right.
---
Changes in v3:
- ("Add a blank line in the dp node") new for v3.
arch/arm64/boot/dts/qcom/sc7280.dtsi | 1 + 1 file changed, 1 insertion(+)
arch/arm64/boot/dts/qcom/sc7280.dtsi +1 -0
@@ -3147,6 +3147,7 @@ mdss_dp: displayport-controller@ae90000 { ports { #address-cells = <1>; #size-cells = <0>; + port@0 { reg = <0>; dp_in: endpoint {
Signature · 1 line
2.35.0.rc2.247.g8bbb082509-goog
Re: [PATCH v3 11/14] arm64: dts: qcom: sc7280: Add a blank line in the dp node
Quoted 5 lines · Douglas Anderson
It's weird that there's a blank line between the two port nodes but not between the attributes and the first port node. Add an extra blank line to make it look right.
Signed-off-by Douglas Anderson <[email protected]>
Re: [PATCH v3 11/14] arm64: dts: qcom: sc7280: Add a blank line in the dp node
In reply to Douglas AndersonQuoting Douglas Anderson (2022-02-02 13:23:45)
Quoted 6 lines
It's weird that there's a blank line between the two port nodes but not between the attributes and the first port node. Add an extra blank line to make it look right.
Signed-off-by Douglas Anderson <[email protected]>---
PATCH v3 12/14 arm64: dts: qcom: sc7280: Add herobrine-r1
In reply to Douglas AndersonAdd the new herobrine-r1. Note that this is pretty much a re-design compared to herobrine-r0 so we don't attempt any dtsi to share stuff between them.
This patch attempts to define things at 3 levels:
1. The Qcard level. Herobrine includes a Qcard PCB and the Qcard PCB is supposed to be the same (modulo stuffing options) across multiple boards, so trying to define what's there hopefully makes sense. NOTE that newer "CRD" boards from Qualcomm also use Qcard. When support for CRD3 is added hopefully it can use the Qcard include (and perhaps we should even evaluate it using herobrine.dtsi?)
2. The herobrine "baseboard" level. Right now most stuff is here with the exception of things that we _know_ will be different per board. We know that not all boards will have the same set of eMMC, nvme, and SD. We also know that the exact pin names are likely to be different.
3. The actual "board" level, AKA herobrine-rev1.
NOTES:
- This boots to command prompt. We're still waiting on the PWM driver.
- This assumes LTE for now. Once it's clear how WiFi-only SKUs will work we expect some small changes.
---
Removed Reviewed-by tags in v3 since it felt like there were enough changes that people should re-confirm that they're happy.
Changes in v3:
- Rebased atop dts cleanup patches.
- Add regulator suffix as per dts cleanup patches.
- Set PCIe bias / pull as per dts cleanup patches.
- Add dp_hot_plug_det pull as per dts cleanup patches.
- Setup SD card same as dts cleanup patches.
Changes in v2:
- Herobrine compatible on one line, not two
- Wording change in comments for components enabled per-board
- Always sort "bias" above "drive-strength" in pinctrl.
- Properly sort "hub_en" pinctrl.
- Two comments moved from multiline to single line.
- Space after "/delete-property/"
arch/arm64/boot/dts/qcom/Makefile | 1 + .../qcom/sc7280-herobrine-herobrine-r0.dts | 3 +- .../qcom/sc7280-herobrine-herobrine-r1.dts | 313 +++++++ .../arm64/boot/dts/qcom/sc7280-herobrine.dtsi | 785 ++++++++++++++++++ arch/arm64/boot/dts/qcom/sc7280-qcard.dtsi | 547 ++++++++++++ 5 files changed, 1647 insertions(+), 2 deletions(-) create mode 100644 arch/arm64/boot/dts/qcom/sc7280-herobrine-herobrine-r1.dts create mode 100644 arch/arm64/boot/dts/qcom/sc7280-herobrine.dtsi create mode 100644 arch/arm64/boot/dts/qcom/sc7280-qcard.dtsi
arch/arm64/boot/dts/qcom/Makefile +1 -0
@@ -84,6 +84,7 @@ dtb-$(CONFIG_ARCH_QCOM) += sc7180-trogdor-pompom-r3-lte.dtb dtb-$(CONFIG_ARCH_QCOM) += sc7180-trogdor-r1.dtb dtb-$(CONFIG_ARCH_QCOM) += sc7180-trogdor-r1-lte.dtb dtb-$(CONFIG_ARCH_QCOM) += sc7280-herobrine-herobrine-r0.dtb +dtb-$(CONFIG_ARCH_QCOM) += sc7280-herobrine-herobrine-r1.dtb dtb-$(CONFIG_ARCH_QCOM) += sc7280-idp.dtb dtb-$(CONFIG_ARCH_QCOM) += sc7280-idp2.dtb dtb-$(CONFIG_ARCH_QCOM) += sc7280-crd.dtb
arch/arm64/boot/dts/qcom/sc7280-herobrine-herobrine-r0.dts +1 -2
@@ -25,8 +25,7 @@ / { model = "Google Herobrine (rev0)"; - compatible = "google,herobrine", - "qcom,sc7280"; + compatible = "google,herobrine-rev0", "qcom,sc7280"; }; / {
arch/arm64/boot/dts/qcom/sc7280-herobrine-herobrine-r1.dts new file +313 -0
313 lines not on this page · raw message
arch/arm64/boot/dts/qcom/sc7280-herobrine.dtsi new file +785 -0
785 lines not on this page · raw message
arch/arm64/boot/dts/qcom/sc7280-qcard.dtsi new file +547 -0
547 lines not on this page · raw message
Signature · 1 line
2.35.0.rc2.247.g8bbb082509-goog
Re: [PATCH v3 12/14] arm64: dts: qcom: sc7280: Add herobrine-r1
Quoted 20 lines · Douglas Anderson
Add the new herobrine-r1. Note that this is pretty much a re-design compared to herobrine-r0 so we don't attempt any dtsi to share stuff between them.
This patch attempts to define things at 3 levels:
1. The Qcard level. Herobrine includes a Qcard PCB and the Qcard PCB is supposed to be the same (modulo stuffing options) across multiple boards, so trying to define what's there hopefully makes sense. NOTE that newer "CRD" boards from Qualcomm also use Qcard. When support for CRD3 is added hopefully it can use the Qcard include (and perhaps we should even evaluate it using
herobrine.dtsi?)
2. The herobrine "baseboard" level. Right now most stuff is here with the exception of things that we _know_ will be different per board. We know that not all boards will have the same set of eMMC, nvme, and SD. We also know that the exact pin names are likely to be different.
3. The actual "board" level, AKA herobrine-rev1.
Quoted 6 lines
NOTES:
- This boots to command prompt. We're still waiting on the PWM driver.
- This assumes LTE for now. Once it's clear how WiFi-only SKUs will work we expect some small changes.Signed-off-by Douglas Anderson <[email protected]>
Re: [PATCH v3 12/14] arm64: dts: qcom: sc7280: Add herobrine-r1
In reply to Douglas AndersonHi,
arch/arm64/boot/dts/qcom/sc7280-herobrine.dtsi line 531
Quoted 5 lines · Douglas Anderson
+&pcie1_clkreq_n {
arch/arm64/boot/dts/qcom/sc7280-herobrine.dtsi new file +3
-0+ bias-disable; + drive-strength = <2>; +};
As per the discussion [1] then maybe the above should be bias-pull-up instead of bias-disable. I'm happy to spin this, have it fixed by the maintainer when applied, or do a follow-up patch to fix this. Please let me know.
-Doug
[1] https://lore.kernel.org/r/CAD=FV=UKKZaHHz3-idahLg-ey3xmSZWKeTVVipzpZNQAkUVKmQ@mail.gmail.com
Re: [PATCH v3 12/14] arm64: dts: qcom: sc7280: Add herobrine-r1
Hi,
Quoted 10 lines · Doug Anderson
Hi,
arch/arm64/boot/dts/qcom/sc7280-herobrine.dtsi line 531
Quoted 5 lines · Douglas Anderson
+&pcie1_clkreq_n {
arch/arm64/boot/dts/qcom/sc7280-herobrine.dtsi new file +3
-0+ bias-disable; + drive-strength = <2>; +};
Quoted 4 lines
As per the discussion [1] then maybe the above should be
bias-pull-upinstead ofbias-disable. I'm happy to spin this, have it fixed by the maintainer when applied, or do a follow-up patch to fix this. Please let me know.
Breadcrumbs: since most of this series landed but not this patch, I've sent a new version with the fixup.
https://lore.kernel.org/r/20220204140550.v4.1.I5604b7af908e8bbe709ac037a6a8a6ba8a2bfa94@changeid/
PATCH v3 13/14 arm64: dts: qcom: sc7280: Add the CPU compatible to the soc@0 node
In reply to Douglas AndersonWe'd like to start including the CPU name as the compatible under the "soc" node so that we can get rid of it from the top-level compatible string.
---
Probably needs a .yaml file somewhere?
Changes in v3:
- ("sc7280: Add the CPU compatible to the soc@0 node") new for v3.
arch/arm64/boot/dts/qcom/sc7280.dtsi | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
arch/arm64/boot/dts/qcom/sc7280.dtsi +1 -1
@@ -573,7 +573,7 @@ soc: soc@0 { #size-cells = <2>; ranges = <0 0 0 0 0x10 0>; dma-ranges = <0 0 0 0 0x10 0>; - compatible = "simple-bus"; + compatible = "qcom,sc7280", "simple-bus"; gcc: clock-controller@100000 { compatible = "qcom,gcc-sc7280";
Signature · 1 line
2.35.0.rc2.247.g8bbb082509-goog
Re: [PATCH v3 13/14] arm64: dts: qcom: sc7280: Add the CPU compatible to the soc@0 node
Quoting Douglas Anderson (2022-02-02 13:23:47)
Quoted 7 lines
We'd like to start including the CPU name as the compatible under the "soc" node so that we can get rid of it from the top-level compatible string.
Suggested-by Stephen Boyd <[email protected]>Signed-off-by Douglas Anderson <[email protected]>---
Re: [PATCH v3 13/14] arm64: dts: qcom: sc7280: Add the CPU compatible to the soc@0 node
In reply to Douglas AndersonQuoted 20 lines · Douglas Anderson
We'd like to start including the CPU name as the compatible under the "soc" node so that we can get rid of it from the top-level compatible string.
Suggested-by Stephen Boyd <[email protected]>Signed-off-by Douglas Anderson <[email protected]>---
Probably needs a .yaml file somewhere?Changes in v3:
- ("sc7280: Add the CPU compatible to the soc@0 node") new for v3.arch/arm64/boot/dts/qcom/sc7280.dtsi | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)arch/arm64/boot/dts/qcom/sc7280.dtsi +0
-0
arch/arm64/boot/dts/qcom/sc7280.dtsi soc: soc@0 { · line 576
Quoted 5 lines
arch/arm64/boot/dts/qcom/sc7280.dtsi +1
-1@@ -573,7 +573,7 @@ soc: soc@0 { #size-cells = <2>; ranges = <0 0 0 0 0x10 0>; dma-ranges = <0 0 0 0 0x10 0>; - compatible = "simple-bus"; + compatible = "qcom,sc7280", "simple-bus";
To me this implies that /soc represents the sc7280, but as noted earlier I don't think that's accurate. E.g. if this node represents the sc7280, why are the cpus described outside this node?
Further more, if we look at the reg nodes on this bus it's clear that this is some mmio bus, which per the ranges has 36 bit address width. But not all buses in the sc7280 has 36 bit address width, so it's not inconceivable that one would actually have to split /soc into more than one entity with different dma-ranges. Perhaps not today, but I don't like the precedence it sets.
Regards,
Bjorn
Quoted 6 lines
arch/arm64/boot/dts/qcom/sc7280.dtsi +0
-0gcc: clock-controller@100000 { compatible = "qcom,gcc-sc7280";Signature · 1 line
2.35.0.rc2.247.g8bbb082509-goog
Re: [PATCH v3 13/14] arm64: dts: qcom: sc7280: Add the CPU compatible to the soc@0 node
Quoting Bjorn Andersson (2022-02-04 13:51:44)
Quoted 17 lines · Douglas Anderson
Quoted 16 lines
We'd like to start including the CPU name as the compatible under the "soc" node so that we can get rid of it from the top-level compatible string.
Suggested-by Stephen Boyd <[email protected]>Signed-off-by Douglas Anderson <[email protected]>---
Probably needs a .yaml file somewhere?Changes in v3:
- ("sc7280: Add the CPU compatible to the soc@0 node") new for v3.arch/arm64/boot/dts/qcom/sc7280.dtsi | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
Quoted 14 lines
arch/arm64/boot/dts/qcom/sc7280.dtsi soc: soc@0 { · line 576
Quoted 10 lines
arch/arm64/boot/dts/qcom/sc7280.dtsi +1
-1@@ -573,7 +573,7 @@ soc: soc@0 { #size-cells = <2>; ranges = <0 0 0 0 0x10 0>; dma-ranges = <0 0 0 0 0x10 0>; - compatible = "simple-bus"; + compatible = "qcom,sc7280", "simple-bus";To me this implies that /soc represents the sc7280, but as noted earlier I don't think that's accurate. E.g. if this node represents the sc7280, why are the cpus described outside this node?
They're outside the soc node because cpus have historically been described at the root of the DT. The concept of an 'soc' node came after the cpu nodes.
Quoted 8 lines
Further more, if we look at the reg nodes on this bus it's clear that this is some mmio bus, which per the ranges has 36 bit address width. But not all buses in the sc7280 has 36 bit address width, so it's not inconceivable that one would actually have to split /soc into more than one entity with different dma-ranges. Perhaps not today, but I don't like the precedence it sets.
That should be fine. We can separate the nodes inside the soc node if we need to by having different bus nodes underneath the soc node and then change the dma-ranges within those bus nodes accordingly. We wouldn't introduce another soc node to solve this problem.
PATCH v3 14/14 arm64: dts: qcom: sc7280: Remove "qcom,sc7280" from top-level of boards
In reply to Douglas AndersonThere's a proposal to take the SoC name out of the top-level compatible and move it under the "soc@0" node. Building on the patch ("arm64: dts: qcom: sc7280: Add the CPU compatible to the soc@0 node"), which added this to the soc@0 node, this removes it from the top-level node.
NOTE: while the previous patch could land at any time without any compatibility issues, this patch will cause problems without a code change to the cpufreq driver [1].
[1] https://lore.kernel.org/r/CAE-0n50sX9-0MxcpF+3Rwqm75jSw5=aNwdsitLwE2sEA69jLJw@mail.gmail.com
---
Not for commiting right now since we'd need the corresponding code change.
Changes in v3:
- ("Remove "qcom,sc7280" from top-level") patch new for v3.
arch/arm64/boot/dts/qcom/sc7280-crd.dts | 2 +- arch/arm64/boot/dts/qcom/sc7280-herobrine-herobrine-r0.dts | 2 +- arch/arm64/boot/dts/qcom/sc7280-herobrine-herobrine-r1.dts | 2 +- arch/arm64/boot/dts/qcom/sc7280-idp.dts | 2 +- arch/arm64/boot/dts/qcom/sc7280-idp2.dts | 2 +- 5 files changed, 5 insertions(+), 5 deletions(-)
arch/arm64/boot/dts/qcom/sc7280-crd.dts +1 -1
@@ -12,7 +12,7 @@ / { model = "Qualcomm Technologies, Inc. sc7280 CRD platform"; - compatible = "qcom,sc7280-crd", "google,hoglin", "qcom,sc7280"; + compatible = "qcom,sc7280-crd", "google,hoglin"; aliases { serial0 = &uart5;
arch/arm64/boot/dts/qcom/sc7280-herobrine-herobrine-r0.dts +1 -1
@@ -25,7 +25,7 @@ / { model = "Google Herobrine (rev0)"; - compatible = "google,herobrine-rev0", "qcom,sc7280"; + compatible = "google,herobrine-rev0"; }; / {
arch/arm64/boot/dts/qcom/sc7280-herobrine-herobrine-r1.dts +1 -1
@@ -11,7 +11,7 @@ / { model = "Google Herobrine (rev1+)"; - compatible = "google,herobrine", "qcom,sc7280"; + compatible = "google,herobrine"; }; /* ADDITIONS TO NODES DEFINED IN PARENT DEVICE TREE FILES */
arch/arm64/boot/dts/qcom/sc7280-idp.dts +1 -1
@@ -13,7 +13,7 @@ / { model = "Qualcomm Technologies, Inc. sc7280 IDP SKU1 platform"; - compatible = "qcom,sc7280-idp", "google,senor", "qcom,sc7280"; + compatible = "qcom,sc7280-idp", "google,senor"; aliases { serial0 = &uart5;
arch/arm64/boot/dts/qcom/sc7280-idp2.dts +1 -1
@@ -12,7 +12,7 @@ / { model = "Qualcomm Technologies, Inc. sc7280 IDP SKU2 platform"; - compatible = "qcom,sc7280-idp2", "google,piglin", "qcom,sc7280"; + compatible = "qcom,sc7280-idp2", "google,piglin"; aliases { serial0 = &uart5;
Signature · 1 line
2.35.0.rc2.247.g8bbb082509-goog
Re: [PATCH v3 14/14] arm64: dts: qcom: sc7280: Remove "qcom,sc7280" from top-level of boards
Quoting Douglas Anderson (2022-02-02 13:23:48)
Quoted 10 lines
There's a proposal to take the SoC name out of the top-level compatible and move it under the "soc@0" node. Building on the patch ("arm64: dts: qcom: sc7280: Add the CPU compatible to the soc@0 node"), which added this to the soc@0 node, this removes it from the top-level node.
NOTE: while the previous patch could land at any time without any compatibility issues, this patch will cause problems without a code change to the cpufreq driver [1].
Quoted 5 lines
[1] https://lore.kernel.org/r/CAE-0n50sX9-0MxcpF+3Rwqm75jSw5=aNwdsitLwE2sEA69jLJw@mail.gmail.com
Suggested-by Stephen Boyd <[email protected]>Signed-off-by Douglas Anderson <[email protected]>---
It needs to wait though, as stated above.
Bjorn Andersson
AppliedRe: (subset) [PATCH v3 00/14] arm64: dts: qcom: sc7x80: A smattering of misc dts cleanups + herobrine-rev1
In reply to Douglas AndersonQuoted 11 lines · Douglas Anderson
This series is "v2" of my "smattering of misc dts cleanups" series plus v3 of the tail end of the series adding herobrine-rev1. I've set the version number to the larger of the two to (I hope) help allevitate confusion.
For the cleanups, there's not a lot holding this series together except that it fixes a smattering of random dts stuff that I noticed recently. There are not a lot of dependencies and some of the patches could be reordered if desired.
[...]
Applied, thanks!
[01/14] arm64: dts: qcom: sc7180-trogdor: Add "-regulator" suffix to pp3300_hub
commit: 171bac46700fcdb2310209dffb382533fe54522a
[02/14] arm64: dts: qcom: sc7280-herobrine: Consistently add "-regulator" suffix
commit: 7a86ac04056569bf5ec663fbb02d79c5e304545a
[03/14] arm64: dts: qcom: sc7280: Properly sort sdc pinctrl lines
commit: b1969bc522187dc6f436301eb71051b24135b607
[04/14] arm64: dts: qcom: sc7280: Clean up sdc1 / sdc2 pinctrl
commit: f9800dde34e678d7ed1de9e95b4b65a257fd0f93
[05/14] arm64: dts: qcom: sc7280-idp: No need for "input-enable" on sw_ctrl
commit: 8fdedd6c64643884dc6bbf6d9a7aabda1713354f
[06/14] arm64: dts: qcom: sc7280: Fix sort order of dp_hot_plug_det / pcie1_clkreq_n
commit: bbef2a9ca08749c89925d2bb49f4ce1c945acc90
[07/14] arm64: dts: qcom: sc7280: Add edp_out port and HPD lines
commit: 118cd3b8ec0db02eb7306c30c1551ef9af885689
[08/14] arm64: dts: qcom: sc7280: Move pcie1_clkreq pull / drive str to boards
commit: 376e9183c1d1dde6972257a823cf484cc5124b7b
[10/14] arm64: dts: qcom: sc7280: Move dp_hot_plug_det pull from SoC dtsi file
commit: ad4152d6e2599c62ef012e528acc5e77ca6765c1
[11/14] arm64: dts: qcom: sc7280: Add a blank line in the dp node
commit: 96b34a6ea7d03876fb9b82ac8db5648a24fc7b2eBest regards,
Signature · 1 line
Bjorn Andersson <[email protected]>