[PATCH v2 0/5] arm64: dts: qcom: sdm845-mtp: misc improvements

[PATCH v2 0/5] arm64: dts: qcom: sdm845-mtp: misc improvements

[Dmitry Baryshkov]

Several misc improvements, take away from testing recent kernel on
SDM845-MTP board.

Changes since v1:
 - Reordered PCIe properties according to Konrad's suggestions
 - Fixed pcie1 perst to have bias-pull-down instead of bias-disable

Dmitry Baryshkov (5):
  arm64: dts: qcom: sdm845-mtp: fix WiFi configuration
  arm64: dts: qcom: sdm845-mtp: specify wifi variant
  arm64: dts: qcom: sdm845-mtp: switch to mbn firmware
  arm64: dts: qcom: sdm845-mtp: enable Vol-/reset button
  arm64: dts: qcom: sdm845-mtp: enable PCIe support

 arch/arm64/boot/dts/qcom/sdm845-mtp.dts | 90 ++++++++++++++++++++++++-
 1 file changed, 88 insertions(+), 2 deletions(-)

-- 
2.39.2

[PATCH v2 1/5] arm64: dts: qcom: sdm845-mtp: fix WiFi configuration

[Dmitry Baryshkov]

Enable the host-cap-8bit quirk on this device. It is required for the
WiFi to function properly.

Fixes: 022bccb840b7 ("arm64: dts: sdm845: Add WCN3990 WLAN module device node")
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 arch/arm64/boot/dts/qcom/sdm845-mtp.dts | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/arch/arm64/boot/dts/qcom/sdm845-mtp.dts b/arch/arm64/boot/dts/qcom/sdm845-mtp.dts
index b3c27a524742..1516113391ed 100644
--- a/arch/arm64/boot/dts/qcom/sdm845-mtp.dts
+++ b/arch/arm64/boot/dts/qcom/sdm845-mtp.dts
@@ -716,6 +716,8 @@ &wifi {
 	vdd-1.8-xo-supply = <&vreg_l7a_1p8>;
 	vdd-1.3-rfa-supply = <&vreg_l17a_1p3>;
 	vdd-3.3-ch0-supply = <&vreg_l25a_3p3>;
+
+	qcom,snoc-host-cap-8bit-quirk;
 };
 
 /* PINCTRL - additions to nodes defined in sdm845.dtsi */
-- 
2.39.2

[PATCH v2 2/5] arm64: dts: qcom: sdm845-mtp: specify wifi variant

[Dmitry Baryshkov]

Specify the variant for the WiFi BDF file, "Qualcomm_sdm845mtp" to ease
distinguishing from other (possible) devices using the same board id.

For the reference:

ath10k_snoc 18800000.wifi: qmi chip_id 0x30214 chip_family 0x4001 board_id 0x3f soc_id 0x40030001
ath10k_snoc 18800000.wifi: qmi fw_version 0x2009856b fw_build_timestamp 2018-07-19 12:28 fw_build_id QC_IMAGE_VERSION_STRING=WLAN.HL.2.0-01387-QCAHLSWMTPLZ-1
ath10k_snoc 18800000.wifi: failed to fetch board data for bus=snoc,qmi-board-id=3f,qmi-chip-id=30214,variant=Qualcomm_sdm845mtp from ath10k/WCN3990/hw1.0/board-2.bin
ath10k_snoc 18800000.wifi: wcn3990 hw1.0 target 0x00000008 chip_id 0x00000000 sub 0000:0000
ath10k_snoc 18800000.wifi: kconfig debug 1 debugfs 0 tracing 0 dfs 0 testmode 0
ath10k_snoc 18800000.wifi: firmware ver  api 5 features wowlan,mgmt-tx-by-reference,non-bmi crc32 b3d4b790
ath10k_snoc 18800000.wifi: htt-ver 3.53 wmi-op 4 htt-op 3 cal file max-sta 32 raw 0 hwcrypto 1
ath10k_snoc 18800000.wifi: invalid MAC address; choosing random

Cc: Kalle Valo <kvalo@kernel.org>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 arch/arm64/boot/dts/qcom/sdm845-mtp.dts | 1 +
 1 file changed, 1 insertion(+)

diff --git a/arch/arm64/boot/dts/qcom/sdm845-mtp.dts b/arch/arm64/boot/dts/qcom/sdm845-mtp.dts
index 1516113391ed..56f5bf52d9bd 100644
--- a/arch/arm64/boot/dts/qcom/sdm845-mtp.dts
+++ b/arch/arm64/boot/dts/qcom/sdm845-mtp.dts
@@ -718,6 +718,7 @@ &wifi {
 	vdd-3.3-ch0-supply = <&vreg_l25a_3p3>;
 
 	qcom,snoc-host-cap-8bit-quirk;
+	qcom,ath10k-calibration-variant = "Qualcomm_sdm845mtp";
 };
 
 /* PINCTRL - additions to nodes defined in sdm845.dtsi */
-- 
2.39.2

[PATCH v2 3/5] arm64: dts: qcom: sdm845-mtp: switch to mbn firmware

[Dmitry Baryshkov]

We have switched most of devices to use mbn (squashed) firmware files
instead of spit mdt+bNN. Even this DT uses modem.mbn and a630_zap.mbn.
Let's switch adsp and cdsp firmware files to use .mbn format too.

Reviewed-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 arch/arm64/boot/dts/qcom/sdm845-mtp.dts | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/arm64/boot/dts/qcom/sdm845-mtp.dts b/arch/arm64/boot/dts/qcom/sdm845-mtp.dts
index 56f5bf52d9bd..7579ba66c6d8 100644
--- a/arch/arm64/boot/dts/qcom/sdm845-mtp.dts
+++ b/arch/arm64/boot/dts/qcom/sdm845-mtp.dts
@@ -114,7 +114,7 @@ trip-point {
 
 &adsp_pas {
 	status = "okay";
-	firmware-name = "qcom/sdm845/adsp.mdt";
+	firmware-name = "qcom/sdm845/adsp.mbn";
 };
 
 &apps_rsc {
@@ -415,7 +415,7 @@ vreg_s3c_0p6: smps3 {
 
 &cdsp_pas {
 	status = "okay";
-	firmware-name = "qcom/sdm845/cdsp.mdt";
+	firmware-name = "qcom/sdm845/cdsp.mbn";
 };
 
 &gcc {
-- 
2.39.2

Re: [PATCH v2 3/5] arm64: dts: qcom: sdm845-mtp: switch to mbn firmware

[Arnd Bergmann]

On Sun, Aug 27, 2023, at 00:19, Dmitry Baryshkov wrote:
> We have switched most of devices to use mbn (squashed) firmware files
> instead of spit mdt+bNN. Even this DT uses modem.mbn and a630_zap.mbn.
> Let's switch adsp and cdsp firmware files to use .mbn format too.
>
> Reviewed-by: Konrad Dybcio <konrad.dybcio@linaro.org>
> Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>

I saw this pop up in the dts pull request today, and this needs
some more clarification I think: Why does the dts have to know
about which format the files are in? I would expect that to be
handled by the driver internally.

From what I can tell, either the two files are compatible,
and the driver can just try both names by manipulating that
string, or the two are incompatible and changing the firmware
name would also mean that the dts node and driver would
have to be aware of the difference.

      Arnd

Re: [PATCH v2 3/5] arm64: dts: qcom: sdm845-mtp: switch to mbn firmware

[Bjorn Andersson]

On Mon, Oct 16, 2023 at 03:39:01PM +0200, Arnd Bergmann wrote:
> On Sun, Aug 27, 2023, at 00:19, Dmitry Baryshkov wrote:
> > We have switched most of devices to use mbn (squashed) firmware files
> > instead of spit mdt+bNN. Even this DT uses modem.mbn and a630_zap.mbn.
> > Let's switch adsp and cdsp firmware files to use .mbn format too.
> >
> > Reviewed-by: Konrad Dybcio <konrad.dybcio@linaro.org>
> > Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
> 
> I saw this pop up in the dts pull request today, and this needs
> some more clarification I think: Why does the dts have to know
> about which format the files are in? I would expect that to be
> handled by the driver internally.
> 
> From what I can tell, either the two files are compatible,
> and the driver can just try both names by manipulating that
> string, or the two are incompatible and changing the firmware
> name would also mean that the dts node and driver would
> have to be aware of the difference.
> 

The implementation does not care about the file name, it will detect
which of the two formats is being provided and load the content
accordingly.

So what this patch does is to align the firmware file name to match what
we have in linux-firmware, and what the few of us booting this device
have hacked around in /lib/firmware for (quite) a while now.

Regards,
Bjorn

[PATCH v2 4/5] arm64: dts: qcom: sdm845-mtp: enable Vol-/reset button

[Dmitry Baryshkov]

Wire up the Vol- / reset button on Qualcomm SDM845 MTP board.

Reviewed-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 arch/arm64/boot/dts/qcom/sdm845-mtp.dts | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/arch/arm64/boot/dts/qcom/sdm845-mtp.dts b/arch/arm64/boot/dts/qcom/sdm845-mtp.dts
index 7579ba66c6d8..aec3f358d426 100644
--- a/arch/arm64/boot/dts/qcom/sdm845-mtp.dts
+++ b/arch/arm64/boot/dts/qcom/sdm845-mtp.dts
@@ -609,6 +609,11 @@ quiet-thermistor@4 {
 	};
 };
 
+&pm8998_resin {
+	linux,code = <KEY_VOLUMEDOWN>;
+	status = "okay";
+};
+
 &qupv3_id_1 {
 	status = "okay";
 };
-- 
2.39.2

[PATCH v2 5/5] arm64: dts: qcom: sdm845-mtp: enable PCIe support

[Dmitry Baryshkov]

Enable two PCIe hosts support on Qualcomm SDM845 MTP board.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 arch/arm64/boot/dts/qcom/sdm845-mtp.dts | 78 +++++++++++++++++++++++++
 1 file changed, 78 insertions(+)

diff --git a/arch/arm64/boot/dts/qcom/sdm845-mtp.dts b/arch/arm64/boot/dts/qcom/sdm845-mtp.dts
index aec3f358d426..76bfa786612c 100644
--- a/arch/arm64/boot/dts/qcom/sdm845-mtp.dts
+++ b/arch/arm64/boot/dts/qcom/sdm845-mtp.dts
@@ -533,6 +533,38 @@ &mss_pil {
 	firmware-name = "qcom/sdm845/mba.mbn", "qcom/sdm845/modem.mbn";
 };
 
+&pcie0 {
+	perst-gpios = <&tlmm 35 GPIO_ACTIVE_LOW>;
+
+	pinctrl-0 = <&pcie0_default_state>;
+	pinctrl-names = "default";
+
+	status = "okay";
+};
+
+&pcie0_phy {
+	vdda-phy-supply = <&vreg_l1a_0p875>;
+	vdda-pll-supply = <&vreg_l26a_1p2>;
+
+	status = "okay";
+};
+
+&pcie1 {
+	perst-gpios = <&tlmm 102 GPIO_ACTIVE_LOW>;
+
+	pinctrl-names = "default";
+	pinctrl-0 = <&pcie1_default_state>;
+
+	status = "okay";
+};
+
+&pcie1_phy {
+	status = "okay";
+
+	vdda-phy-supply = <&vreg_l1a_0p875>;
+	vdda-pll-supply = <&vreg_l26a_1p2>;
+};
+
 &pm8998_adc {
 	channel@4c {
 		reg = <ADC5_XO_THERM_100K_PU>;
@@ -630,6 +662,52 @@ &sdhc_2 {
 	cd-gpios = <&tlmm 126 GPIO_ACTIVE_LOW>;
 };
 
+&tlmm {
+	pcie0_default_state: pcie0-default-state {
+		clkreq-pins {
+			pins = "gpio36";
+			function = "pci_e0";
+			bias-pull-up;
+		};
+
+		perst-n-pins {
+			pins = "gpio35";
+			function = "gpio";
+			drive-strength = <2>;
+			bias-pull-down;
+		};
+
+		wake-n-pins {
+			pins = "gpio37";
+			function = "gpio";
+			drive-strength = <2>;
+			bias-pull-up;
+		};
+	};
+
+	pcie1_default_state: pcie1-default-state {
+		clkreq-pins {
+			pins = "gpio103";
+			function = "pci_e1";
+			bias-pull-up;
+		};
+
+		perst-n-pins {
+			pins = "gpio102";
+			function = "gpio";
+			drive-strength = <16>;
+			bias-pull-down;
+		};
+
+		wake-n-pins {
+			pins = "gpio104";
+			function = "gpio";
+			drive-strength = <2>;
+			bias-pull-up;
+		};
+	};
+};
+
 &uart9 {
 	status = "okay";
 };
-- 
2.39.2

Re: [PATCH v2 0/5] arm64: dts: qcom: sdm845-mtp: misc improvements

[Bjorn Andersson]

On Sun, 27 Aug 2023 01:19:10 +0300, Dmitry Baryshkov wrote:
> Several misc improvements, take away from testing recent kernel on
> SDM845-MTP board.
> 
> Changes since v1:
>  - Reordered PCIe properties according to Konrad's suggestions
>  - Fixed pcie1 perst to have bias-pull-down instead of bias-disable
> 
> [...]

Applied, thanks!

[1/5] arm64: dts: qcom: sdm845-mtp: fix WiFi configuration
      commit: b33868a52f342d9b1f20aa5bffe40cbd69bd0a4b
[2/5] arm64: dts: qcom: sdm845-mtp: specify wifi variant
      commit: 84c7786499880d47fbcf0cee9661fd96e026ab2a
[3/5] arm64: dts: qcom: sdm845-mtp: switch to mbn firmware
      commit: 37857110db63783bac36960aa3bd335d4adea9d7
[4/5] arm64: dts: qcom: sdm845-mtp: enable Vol-/reset button
      commit: be30dc31823d46dca3a184896433ebf8beffa1a0
[5/5] arm64: dts: qcom: sdm845-mtp: enable PCIe support
      commit: da3620d7c7d78a375b21ccf046b6617598b7c824

Best regards,
-- 
Bjorn Andersson <andersson@kernel.org>