On 1/10/2023 6:20 PM, Konrad Dybcio wrote:
On 10.01.2023 13:13, devi priya wrote:
From: POOVENDHAN SELVARAJ <quic_poovendh@xxxxxxxxxxx>
Add initial device tree support for Qualcomm IPQ9574 SoC
and AL02 board
Co-developed-by: Anusha Rao <quic_anusha@xxxxxxxxxxx>
Signed-off-by: Anusha Rao <quic_anusha@xxxxxxxxxxx>
Co-developed-by: devi priya <quic_devipriy@xxxxxxxxxxx>
Signed-off-by: devi priya <quic_devipriy@xxxxxxxxxxx>
Signed-off-by: POOVENDHAN SELVARAJ <quic_poovendh@xxxxxxxxxxx>
---
arch/arm64/boot/dts/qcom/Makefile | 1 +
arch/arm64/boot/dts/qcom/ipq9574-al02-c7.dts | 69 ++++
arch/arm64/boot/dts/qcom/ipq9574.dtsi | 318 +++++++++++++++++++
3 files changed, 388 insertions(+)
create mode 100644 arch/arm64/boot/dts/qcom/ipq9574-al02-c7.dts
create mode 100644 arch/arm64/boot/dts/qcom/ipq9574.dtsi
diff --git a/arch/arm64/boot/dts/qcom/Makefile b/arch/arm64/boot/dts/qcom/Makefile
index 3e79496292e7..872c62028a0b 100644
--- a/arch/arm64/boot/dts/qcom/Makefile
+++ b/arch/arm64/boot/dts/qcom/Makefile
@@ -7,6 +7,7 @@ dtb-$(CONFIG_ARCH_QCOM) += ipq6018-cp01-c1.dtb
dtb-$(CONFIG_ARCH_QCOM) += ipq8074-hk01.dtb
dtb-$(CONFIG_ARCH_QCOM) += ipq8074-hk10-c1.dtb
dtb-$(CONFIG_ARCH_QCOM) += ipq8074-hk10-c2.dtb
+dtb-$(CONFIG_ARCH_QCOM) += ipq9574-al02-c7.dtb
dtb-$(CONFIG_ARCH_QCOM) += msm8916-alcatel-idol347.dtb
dtb-$(CONFIG_ARCH_QCOM) += msm8916-asus-z00l.dtb
dtb-$(CONFIG_ARCH_QCOM) += msm8916-huawei-g7.dtb
diff --git a/arch/arm64/boot/dts/qcom/ipq9574-al02-c7.dts b/arch/arm64/boot/dts/qcom/ipq9574-al02-c7.dts
new file mode 100644
index 000000000000..ae3c32f3e16a
--- /dev/null
+++ b/arch/arm64/boot/dts/qcom/ipq9574-al02-c7.dts
@@ -0,0 +1,69 @@
+// SPDX-License-Identifier: GPL-2.0-only
BSD3?
Sure, will add dual license
+/*
+ * IPQ9574 AL02-C7 board device tree source
+ *
+ * Copyright (c) 2020-2021 The Linux Foundation. All rights reserved.
+ * Copyright (c) 2022 Qualcomm Innovation Center, Inc. All rights reserved.
Happy new year!
Sure, will update!
+ */
+
+/dts-v1/;
+
+#include "ipq9574.dtsi"
+
+/ {
+ model = "Qualcomm Technologies, Inc. IPQ9574/AP-AL02-C7";
+ compatible = "qcom,ipq9574-ap-al02-c7", "qcom,ipq9574";
+ interrupt-parent = <&intc>;
+
+ aliases {
+ serial0 = &blsp1_uart2;
+ };
+
+ chosen {
+ stdout-path = "serial0:115200n8";
+ };
+};
+
+&blsp1_uart2 {
+ pinctrl-0 = <&uart2_pins>;
+ pinctrl-names = "default";
+ status = "okay";
+};
+
+&sdhc_1 {
+ pinctrl-0 = <&emmc_pins>;
+ pinctrl-names = "default";
+ status = "okay";
+};
+
+&tlmm {
+ emmc_pins: emmc-state {
sdc_default? I suppose you'll introduce a corresponding sleep state
later on, so that'll be easier to distinguish.
Sure, will update the node name. But, we don't have power management
states. It would always be in the running state.
+ emmc-clk-pins {
+ pins = "gpio5";
+ function = "sdc_clk";
+ drive-strength = <8>;
+ bias-disable;
+ };
Please add a newline between subsequent nodes.
Sure, okay
+ emmc-cmd-pins {
+ pins = "gpio4";
+ function = "sdc_cmd";
+ drive-strength = <8>;
+ bias-pull-up;
+ };
+ emmc-data-pins {
+ pins = "gpio0", "gpio1", "gpio2",
+ "gpio3", "gpio6", "gpio7",
+ "gpio8", "gpio9";
The indentation here is wrong.
Sure, will update
+ function = "sdc_data";
+ drive-strength = <8>;
+ bias-pull-up;
+ };
+ emmc-rclk-pins {
+ pins = "gpio10";
+ function = "sdc_rclk";
+ drive-strength = <8>;
+ bias-pull-down;
+ };
+ };
+
+};
diff --git a/arch/arm64/boot/dts/qcom/ipq9574.dtsi b/arch/arm64/boot/dts/qcom/ipq9574.dtsi
new file mode 100644
index 000000000000..188d18688a77
--- /dev/null
+++ b/arch/arm64/boot/dts/qcom/ipq9574.dtsi
@@ -0,0 +1,318 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * IPQ9574 SoC device tree source
+ *
+ * Copyright (c) 2020-2021 The Linux Foundation. All rights reserved.
+ * Copyright (c) 2022, Qualcomm Innovation Center, Inc. All rights reserved.
+ */
+
+#include <dt-bindings/interrupt-controller/arm-gic.h>
+#include <dt-bindings/clock/qcom,gcc-ipq9574.h>
+#include <dt-bindings/reset/qcom,gcc-ipq9574.h>
+
+/ {
+ interrupt-parent = <&intc>;
+ #address-cells = <2>;
+ #size-cells = <2>;
+
+ clocks {
+ bias_pll_ubi_nc_clk: bias_pll_ubi_nc_clk {
Could you tell us something about the purpose of this clock? First
time seeing it, your gcc driver reveals it's connected to at least
PCIe.
Bias_pll_ubi_nc_clk (353MHz) is a backup source for
Q6_AXIM2_CLK/PCIE2_AXIM_CLK/PCIE3_AXIM_CLK/SNOC-CLK.
+ compatible = "fixed-clock";
+ clock-frequency = <353000000>;
+ #clock-cells = <0>;
+ };
+
+ pcie30_phy0_pipe_clk: pcie30_phy0_pipe_clk {
+ compatible = "fixed-clock";
+ clock-frequency = <250000000>;
+ #clock-cells = <0>;
+ };
+
+ pcie30_phy1_pipe_clk: pcie30_phy1_pipe_clk {
+ compatible = "fixed-clock";
+ clock-frequency = <250000000>;
+ #clock-cells = <0>;
+ };
+
+ pcie30_phy2_pipe_clk: pcie30_phy2_pipe_clk {
+ compatible = "fixed-clock";
+ clock-frequency = <250000000>;
+ #clock-cells = <0>;
+ };
+
+ pcie30_phy3_pipe_clk: pcie30_phy3_pipe_clk {
+ compatible = "fixed-clock";
+ clock-frequency = <250000000>;
+ #clock-cells = <0>;
+ };
+
+ usb3phy_0_cc_pipe_clk: usb3phy_0_cc_pipe_clk {
+ compatible = "fixed-clock";
+ clock-frequency = <125000000>;
+ #clock-cells = <0>;
+ };
Do not define these pipe clocks. You can leave the GCC entries as <0>
until you introduce the QMPPHY support, which then you can feed as it
provides these clocks.
Sure, okay
+
+ sleep_clk: sleep-clk {
+ compatible = "fixed-clock";
+ clock-frequency = <32000>;
+ #clock-cells = <0>;
+ };
+
+ xo_board_clk: xo-board-clk {
+ compatible = "fixed-clock";
+ clock-frequency = <24000000>;
The clock frequency should be moved to the device DT, because the
clock is on the board and not on the SoC.
Okay, will move to device DT
+ #clock-cells = <0>;
+ };
+ };
+
+ cpus {
+ #address-cells = <1>;
+ #size-cells = <0>;
+
+ CPU0: cpu@0 {
+ device_type = "cpu";
+ compatible = "arm,cortex-a73";
+ reg = <0x0>;
+ enable-method = "psci";
+ next-level-cache = <&L2_0>;
+ };
+
+ CPU1: cpu@1 {
+ device_type = "cpu";
+ compatible = "arm,cortex-a73";
+ reg = <0x1>;
+ enable-method = "psci";
+ next-level-cache = <&L2_0>;
+ };
+
+ CPU2: cpu@2 {
+ device_type = "cpu";
+ compatible = "arm,cortex-a73";
+ reg = <0x2>;
+ enable-method = "psci";
+ next-level-cache = <&L2_0>;
+ };
+
+ CPU3: cpu@3 {
+ device_type = "cpu";
+ compatible = "arm,cortex-a73";
+ reg = <0x3>;
+ enable-method = "psci";
+ next-level-cache = <&L2_0>;
+ };
+
+ L2_0: l2-cache {
+ compatible = "cache";
+ cache-level = <2>;
+ };
+ };
+
+ memory@40000000 {
+ device_type = "memory";
+ /* We expect the bootloader to fill in the size */
+ reg = <0x0 0x40000000 0x0 0x0>;
+ };
+
+ pmu {
+ compatible = "arm,cortex-a73-pmu";
+ interrupts = <GIC_PPI 7 (GIC_CPU_MASK_SIMPLE(4) | IRQ_TYPE_LEVEL_HIGH)>;
+ };
+
+ psci {
+ compatible = "arm,psci-1.0";
+ method = "smc";
+ };
+
+ reserved-memory {
+ #address-cells = <2>;
+ #size-cells = <2>;
+ ranges;
+
+ tz_region: memory@4a600000 {
+ reg = <0x0 0x4a600000 0x0 0x400000>;
+ no-map;
+ };
That's.. surprisingly little reserved memory.. No hyp? No PIL regions
that make the board explode when something touches them?
The reserved memory for TZ region is 4Mb. yes,we have hypervisor and PIL
regions.
The memory for PIL region will be separately reserved by the remoteproc
+ };
+
+ soc: soc@0 {
+ #address-cells = <1>;
+ #size-cells = <1>;
+ ranges = <0 0 0 0xffffffff>;
Is 32 bits enough for this SoC's bus? Newer ones use 36 or more..
Yes, as the maximum range supported by ipq9574 is 4GB, 32 bits would be
sufficient
+ compatible = "simple-bus";
+
+ tlmm: pinctrl@1000000 {
+ compatible = "qcom,ipq9574-tlmm";
+ reg = <0x01000000 0x300000>;
+ interrupts = <GIC_SPI 208 IRQ_TYPE_LEVEL_HIGH>;
+ gpio-controller;
+ #gpio-cells = <2>;
+ gpio-ranges = <&tlmm 0 0 65>;
+ gpio-reserved-ranges = <59 1>;
I see it's assigned to [rx0, pwm23, qdss_tracedata_a].. Is this
board-specific or is this pin supposed to be forbidden on all IPQ9574
boards?
This pin is reserved by TZ and is forbidden on all IPQ9574 boards.
+ interrupt-controller;
+ #interrupt-cells = <2>;
+
+ uart2_pins: uart2-state {
+ pins = "gpio34", "gpio35";
+ function = "blsp2_uart";
+ drive-strength = <8>;
+ bias-disable;
+ };
+ };
+
+ gcc: clock-controller@1800000 {
+ compatible = "qcom,gcc-ipq9574";
+ reg = <0x1800000 0x80000>;
+ clocks = <&xo_board_clk>,
+ <&sleep_clk>,
+ <&bias_pll_ubi_nc_clk>,
+ <&pcie30_phy0_pipe_clk>,
+ <&pcie30_phy1_pipe_clk>,
+ <&pcie30_phy2_pipe_clk>,
+ <&pcie30_phy3_pipe_clk>,
+ <&usb3phy_0_cc_pipe_clk>;
+ clock-names = "xo",
+ "sleep_clk",
+ "bias_pll_ubi_nc_clk",
+ "pcie30_phy0_pipe_clk",
+ "pcie30_phy1_pipe_clk",
+ "pcie30_phy2_pipe_clk",
+ "pcie30_phy3_pipe_clk",
+ "usb3phy_0_cc_pipe_clk";
Please touch up the indentation.
Sure okay
+ #clock-cells = <1>;
+ #reset-cells = <1>;
+ };
+
+ sdhc_1: sdhci@7804000 {
+ compatible = "qcom,sdhci-msm-v5";
+ reg = <0x7804000 0x1000>, <0x7805000 0x1000>;
+ reg-names = "hc_mem", "cmdq_mem";
+
+ interrupts = <GIC_SPI 123 IRQ_TYPE_LEVEL_HIGH>,
+ <GIC_SPI 138 IRQ_TYPE_LEVEL_HIGH>;
Please touch up the indentation.
Okay
+ interrupt-names = "hc_irq", "pwr_irq";
+
+ clocks = <&xo_board_clk>,
+ <&gcc GCC_SDCC1_AHB_CLK>,
+ <&gcc GCC_SDCC1_APPS_CLK>;
Please touch up the indentation.
Sure, will do
+ clock-names = "xo", "iface", "core";
The order should be "iface", "core", "xo" as per Documentation/devicetree/bindings/mmc/sdhci-msm.yaml
Got it, will update
+ mmc-ddr-1_8v;
+ mmc-hs200-1_8v;
+ mmc-hs400-1_8v;
+ mmc-hs400-enhanced-strobe;
Are these the limitations of the controller? Otherwise they should
probably be moved to the device-specific DT.
Yes, these are specific to the controller, hence added them in SoC
+ max-frequency = <384000000>;
+ bus-width = <8>;
+ non-removable;
And this property too.
Yes, it is specific to the controller
+ status = "disabled";
+ };
+
+ blsp1_uart2: serial@78b1000 {
+ compatible = "qcom,msm-uartdm-v1.4", "qcom,msm-uartdm";
+ reg = <0x078b1000 0x200>;
+ interrupts = <GIC_SPI 306 IRQ_TYPE_LEVEL_HIGH>;
+ clocks = <&gcc GCC_BLSP1_UART3_APPS_CLK>,
+ <&gcc GCC_BLSP1_AHB_CLK>;
+ clock-names = "core", "iface";
+ status = "disabled";
+ };
+
+ intc: interrupt-controller@b000000 {
+ compatible = "qcom,msm-qgic2";
+ reg = <0x0b000000 0x1000>, /* GICD */
+ <0x0b002000 0x1000>, /* GICC */
+ <0x0b001000 0x1000>, /* GICH */
+ <0x0b004000 0x1000>; /* GICV */
+ #address-cells = <1>;
+ #size-cells = <1>;
+ interrupt-controller;
+ #interrupt-cells = <3>;
+ interrupts = <GIC_PPI 9 IRQ_TYPE_LEVEL_HIGH>;
+ ranges = <0 0x0b00c000 0x3000>;
+
+ v2m0: v2m@0 {
+ compatible = "arm,gic-v2m-frame";
+ reg = <0x0 0xffd>;
+ msi-controller;
+ };
+
+ v2m1: v2m@1 {
+ compatible = "arm,gic-v2m-frame";
+ reg = <0x1000 0xffd>;
+ msi-controller;
+ };
+
+ v2m2: v2m@2 {
+ compatible = "arm,gic-v2m-frame";
+ reg = <0x2000 0xffd>;
+ msi-controller;
+ };
+ };
+
+ timer@b120000 {
+ compatible = "arm,armv7-timer-mem";
+ reg = <0xb120000 0x1000>;
+ #address-cells = <1>;
+ #size-cells = <1>;
+ ranges;
+ clock-frequency = <24000000>;
Drop, something earlier in the boot chain already writes to CNTFRQ_ELn.
Sure, okay
+
+ frame@b120000 {
+ reg = <0xb121000 0x1000>,
+ <0xb122000 0x1000>;
+ frame-number = <0>;
+ interrupts = <GIC_SPI 8 IRQ_TYPE_LEVEL_HIGH>,
+ <GIC_SPI 7 IRQ_TYPE_LEVEL_HIGH>;
+ };
+
+ frame@b123000 {
+ reg = <0xb123000 0x1000>;
+ frame-number = <1>;
+ interrupts = <GIC_SPI 9 IRQ_TYPE_LEVEL_HIGH>;
+ status = "disabled";
+ };
+
+ frame@b124000 {
+ reg = <0xb124000 0x1000>;
+ frame-number = <2>;
+ interrupts = <GIC_SPI 10 IRQ_TYPE_LEVEL_HIGH>;
+ status = "disabled";
+ };
+
+ frame@b125000 {
+ reg = <0xb125000 0x1000>;
+ frame-number = <3>;
+ interrupts = <GIC_SPI 11 IRQ_TYPE_LEVEL_HIGH>;
+ status = "disabled";
+ };
+
+ frame@b126000 {
+ reg = <0xb126000 0x1000>;
+ frame-number = <4>;
+ interrupts = <GIC_SPI 12 IRQ_TYPE_LEVEL_HIGH>;
+ status = "disabled";
+ };
+
+ frame@b127000 {
+ reg = <0xb127000 0x1000>;
+ frame-number = <5>;
+ interrupts = <GIC_SPI 13 IRQ_TYPE_LEVEL_HIGH>;
+ status = "disabled";
+ };
+
+ frame@b128000 {
+ reg = <0xb128000 0x1000>;
+ frame-number = <6>;
+ interrupts = <GIC_SPI 14 IRQ_TYPE_LEVEL_HIGH>;
+ status = "disabled";
+ };
+ };
+ };
+
+ timer {
+ compatible = "arm,armv8-timer";
+ interrupts = <GIC_PPI 2 (GIC_CPU_MASK_SIMPLE(4) | IRQ_TYPE_LEVEL_LOW)>,
+ <GIC_PPI 3 (GIC_CPU_MASK_SIMPLE(4) | IRQ_TYPE_LEVEL_LOW)>,
+ <GIC_PPI 4 (GIC_CPU_MASK_SIMPLE(4) | IRQ_TYPE_LEVEL_LOW)>,
+ <GIC_PPI 1 (GIC_CPU_MASK_SIMPLE(4) | IRQ_TYPE_LEVEL_LOW)>;
+ clock-frequency = <24000000>;
Drop, something earlier in the boot chain already writes to CNTFRQ_ELn.
Okay
Konrad
+ };
+};
Best Regards,
Devi Priya
