Skip to contentSkip to conversation

Tegra: MMC: Add DT support for MMC to T20 boards

PATCH Superseded Tom Warren 4 participants 29 messages Feb 11, 2013 – Feb 13, 2013

Subsystems (4)

v2 of 7 Superseded Latest is v7 →

lore
Share & tools
lore

Tom Warren

author
#0 lore raw

v2 cover Tegra: MMC: Add DT support for MMC to T20 boards

This patchset adds device-tree support to the Tegra MMC driver. All device config is done via properties in the DT files instead of hard-coded config options/function arguments.

I've tested this on my Seaboard and everything works fine, including card detect. For the other T20 boards, I've used the Linux kernel DTS files for the sdhci nodes where there wasn't one already, or expanded the info that was already there. Everything builds fine, but I haven't tested anything but Seaboard.

v2: Changes as per Stephen's & Simon's review

Tom Warren (2):
Tegra: fdt: Add/enhance sdhci (mmc) nodes for all T20 DT files
Tegra: MMC: Add DT support to MMC driver for all T20 boards

 arch/arm/dts/tegra20-tamonten.dtsi                |  489 +++++++++++++++++++++
 arch/arm/dts/tegra20.dtsi                         |   16 +-
 arch/arm/include/asm/arch-tegra/mmc.h             |    2 +-
 arch/arm/include/asm/arch-tegra/tegra_mmc.h       |   13 +-
 board/avionic-design/common/tamonten.c            |    4 +-
 board/avionic-design/dts/tegra20-medcom-wide.dts  |    1 +
 board/avionic-design/dts/tegra20-plutux.dts       |    1 +
 board/avionic-design/dts/tegra20-tec.dts          |    1 +
 board/compal/dts/tegra20-paz00.dts                |   19 +-
 board/compal/paz00/paz00.c                        |   11 +-
 board/compulab/dts/tegra20-trimslice.dts          |   14 +
 board/compulab/trimslice/trimslice.c              |    9 +-
 board/nvidia/dts/tegra20-harmony.dts              |   18 +
 board/nvidia/dts/tegra20-seaboard.dts             |   12 +-
 board/nvidia/dts/tegra20-ventana.dts              |   15 +
 board/nvidia/dts/tegra20-whistler.dts             |   13 +
 board/nvidia/harmony/harmony.c                    |   11 +-
 board/nvidia/seaboard/seaboard.c                  |   11 +-
 board/nvidia/whistler/whistler.c                  |    7 +-
 board/toradex/colibri_t20_iris/colibri_t20_iris.c |    2 +-
 board/toradex/dts/tegra20-colibri_t20_iris.dts    |    6 +
 drivers/mmc/tegra_mmc.c                           |  259 +++++++----
 include/configs/medcom-wide.h                     |    2 +
 include/configs/plutux.h                          |    2 +
 include/configs/tec.h                             |    2 +
 include/fdtdec.h                                  |    1 +
 lib/fdtdec.c                                      |    1 +
 27 files changed, 792 insertions(+), 150 deletions(-)
 create mode 100644 arch/arm/dts/tegra20-tamonten.dtsi
author
#1 lore raw

PATCH v2 1/2 Tegra: fdt: Add/enhance sdhci (mmc) nodes for all T20 DT files

Linux dts files were used for those boards that didn't already have sdhci info populated. Tamonten has their own dtsi file with common sdhci nodes (sourced from Linux).

---
v2:
- cleanup comments in dts files/match w/kernel files
- add sdhci aliases in all dts files
- use tegra20-tamonten.dtsi from the kernel for AD boards

 arch/arm/dts/tegra20-tamonten.dtsi               |  489 ++++++++++++++++++++++
 arch/arm/dts/tegra20.dtsi                        |   16 +-
 board/avionic-design/dts/tegra20-medcom-wide.dts |    1 +
 board/avionic-design/dts/tegra20-plutux.dts      |    1 +
 board/avionic-design/dts/tegra20-tec.dts         |    1 +
 board/compal/dts/tegra20-paz00.dts               |   19 +-
 board/compulab/dts/tegra20-trimslice.dts         |   14 +
 board/nvidia/dts/tegra20-harmony.dts             |   18 +
 board/nvidia/dts/tegra20-seaboard.dts            |   12 +-
 board/nvidia/dts/tegra20-ventana.dts             |   15 +
 board/nvidia/dts/tegra20-whistler.dts            |   13 +
 board/toradex/dts/tegra20-colibri_t20_iris.dts   |    6 +
 12 files changed, 595 insertions(+), 10 deletions(-)
 create mode 100644 arch/arm/dts/tegra20-tamonten.dtsi
arch/arm/dts/tegra20-tamonten.dtsi new file +489 -0
new file mode 100644
index 0000000..4766aba
@@ -0,0 +1,489 @@
+/include/ "tegra20.dtsi"
+
+/ {
+	model = "Avionic Design Tamonten SOM";
+	compatible = "ad,tamonten", "nvidia,tegra20";
+
+	memory {
+		reg = <0x00000000 0x20000000>;
+	};
+
+	host1x {
+		hdmi {
+			vdd-supply = <&hdmi_vdd_reg>;
+			pll-supply = <&hdmi_pll_reg>;
+
+			nvidia,ddc-i2c-bus = <&hdmi_ddc>;
+			nvidia,hpd-gpio = <&gpio 111 0>; /* PN7 */
+		};
+	};
+
+	pinmux {
+		pinctrl-names = "default";
+		pinctrl-0 = <&state_default>;
+
+		state_default: pinmux {
+			ata {
+				nvidia,pins = "ata";
+				nvidia,function = "ide";
+			};
+			atb {
+				nvidia,pins = "atb", "gma", "gme";
+				nvidia,function = "sdio4";
+			};
+			atc {
+				nvidia,pins = "atc";
+				nvidia,function = "nand";
+			};
+			atd {
+				nvidia,pins = "atd", "ate", "gmb", "gmd", "gpu",
+					"spia", "spib", "spic";
+				nvidia,function = "gmi";
+			};
+			cdev1 {
+				nvidia,pins = "cdev1";
+				nvidia,function = "plla_out";
+			};
+			cdev2 {
+				nvidia,pins = "cdev2";
+				nvidia,function = "pllp_out4";
+			};
+			crtp {
+				nvidia,pins = "crtp";
+				nvidia,function = "crt";
+			};
+			csus {
+				nvidia,pins = "csus";
+				nvidia,function = "vi_sensor_clk";
+			};
+			dap1 {
+				nvidia,pins = "dap1";
+				nvidia,function = "dap1";
+			};
+			dap2 {
+				nvidia,pins = "dap2";
+				nvidia,function = "dap2";
+			};
+			dap3 {
+				nvidia,pins = "dap3";
+				nvidia,function = "dap3";
+			};
+			dap4 {
+				nvidia,pins = "dap4";
+				nvidia,function = "dap4";
+			};
+			dta {
+				nvidia,pins = "dta", "dtd";
+				nvidia,function = "sdio2";
+			};
+			dtb {
+				nvidia,pins = "dtb", "dtc", "dte";
+				nvidia,function = "rsvd1";
+			};
+			dtf {
+				nvidia,pins = "dtf";
+				nvidia,function = "i2c3";
+			};
+			gmc {
+				nvidia,pins = "gmc";
+				nvidia,function = "uartd";
+			};
+			gpu7 {
+				nvidia,pins = "gpu7";
+				nvidia,function = "rtck";
+			};
+			gpv {
+				nvidia,pins = "gpv", "slxa", "slxk";
+				nvidia,function = "pcie";
+			};
+			hdint {
+				nvidia,pins = "hdint";
+				nvidia,function = "hdmi";
+			};
+			i2cp {
+				nvidia,pins = "i2cp";
+				nvidia,function = "i2cp";
+			};
+			irrx {
+				nvidia,pins = "irrx", "irtx";
+				nvidia,function = "uarta";
+			};
+			kbca {
+				nvidia,pins = "kbca", "kbcb", "kbcc", "kbcd",
+					"kbce", "kbcf";
+				nvidia,function = "kbc";
+			};
+			lcsn {
+				nvidia,pins = "lcsn", "ld0", "ld1", "ld2",
+					"ld3", "ld4", "ld5", "ld6", "ld7",
+					"ld8", "ld9", "ld10", "ld11", "ld12",
+					"ld13", "ld14", "ld15", "ld16", "ld17",
+					"ldc", "ldi", "lhp0", "lhp1", "lhp2",
+					"lhs", "lm0", "lm1", "lpp", "lpw0",
+					"lpw1", "lpw2", "lsc0", "lsc1", "lsck",
+					"lsda", "lsdi", "lspi", "lvp0", "lvp1",
+					"lvs";
+				nvidia,function = "displaya";
+			};
+			owc {
+				nvidia,pins = "owc", "spdi", "spdo", "uac";
+				nvidia,function = "rsvd2";
+			};
+			pmc {
+				nvidia,pins = "pmc";
+				nvidia,function = "pwr_on";
+			};
+			rm {
+				nvidia,pins = "rm";
+				nvidia,function = "i2c1";
+			};
+			sdb {
+				nvidia,pins = "sdb", "sdc", "sdd";
+				nvidia,function = "pwm";
+			};
+			sdio1 {
+				nvidia,pins = "sdio1";
+				nvidia,function = "sdio1";
+			};
+			slxc {
+				nvidia,pins = "slxc", "slxd";
+				nvidia,function = "spdif";
+			};
+			spid {
+				nvidia,pins = "spid", "spie", "spif";
+				nvidia,function = "spi1";
+			};
+			spig {
+				nvidia,pins = "spig", "spih";
+				nvidia,function = "spi2_alt";
+			};
+			uaa {
+				nvidia,pins = "uaa", "uab", "uda";
+				nvidia,function = "ulpi";
+			};
+			uad {
+				nvidia,pins = "uad";
+				nvidia,function = "irda";
+			};
+			uca {
+				nvidia,pins = "uca", "ucb";
+				nvidia,function = "uartc";
+			};
+			conf_ata {
+				nvidia,pins = "ata", "atb", "atc", "atd", "ate",
+					"cdev1", "cdev2", "dap1", "dtb", "gma",
+					"gmb", "gmc", "gmd", "gme", "gpu7",
+					"gpv", "i2cp", "pta", "rm", "slxa",
+					"slxk", "spia", "spib", "uac";
+				nvidia,pull = <0>;
+				nvidia,tristate = <0>;
+			};
+			conf_ck32 {
+				nvidia,pins = "ck32", "ddrc", "pmca", "pmcb",
+					"pmcc", "pmcd", "pmce", "xm2c", "xm2d";
+				nvidia,pull = <0>;
+			};
+			conf_csus {
+				nvidia,pins = "csus", "spid", "spif";
+				nvidia,pull = <1>;
+				nvidia,tristate = <1>;
+			};
+			conf_crtp {
+				nvidia,pins = "crtp", "dap2", "dap3", "dap4",
+					"dtc", "dte", "dtf", "gpu", "sdio1",
+					"slxc", "slxd", "spdi", "spdo", "spig",
+					"uda";
+				nvidia,pull = <0>;
+				nvidia,tristate = <1>;
+			};
+			conf_ddc {
+				nvidia,pins = "ddc", "dta", "dtd", "kbca",
+					"kbcb", "kbcc", "kbcd", "kbce", "kbcf",
+					"sdc";
+				nvidia,pull = <2>;
+				nvidia,tristate = <0>;
+			};
+			conf_hdint {
+				nvidia,pins = "hdint", "lcsn", "ldc", "lm1",
+					"lpw1", "lsc1", "lsck", "lsda", "lsdi",
+					"lvp0", "owc", "sdb";
+				nvidia,tristate = <1>;
+			};
+			conf_irrx {
+				nvidia,pins = "irrx", "irtx", "sdd", "spic",
+					"spie", "spih", "uaa", "uab", "uad",
+					"uca", "ucb";
+				nvidia,pull = <2>;
+				nvidia,tristate = <1>;
+			};
+			conf_lc {
+				nvidia,pins = "lc", "ls";
+				nvidia,pull = <2>;
+			};
+			conf_ld0 {
+				nvidia,pins = "ld0", "ld1", "ld2", "ld3", "ld4",
+					"ld5", "ld6", "ld7", "ld8", "ld9",
+					"ld10", "ld11", "ld12", "ld13", "ld14",
+					"ld15", "ld16", "ld17", "ldi", "lhp0",
+					"lhp1", "lhp2", "lhs", "lm0", "lpp",
+					"lpw0", "lpw2", "lsc0", "lspi", "lvp1",
+					"lvs", "pmc";
+				nvidia,tristate = <0>;
+			};
+			conf_ld17_0 {
+				nvidia,pins = "ld17_0", "ld19_18", "ld21_20",
+					"ld23_22";
+				nvidia,pull = <1>;
+			};
+		};
+
+		state_i2cmux_ddc: pinmux_i2cmux_ddc {
+			ddc {
+				nvidia,pins = "ddc";
+				nvidia,function = "i2c2";
+			};
+			pta {
+				nvidia,pins = "pta";
+				nvidia,function = "rsvd4";
+			};
+		};
+
+		state_i2cmux_pta: pinmux_i2cmux_pta {
+			ddc {
+				nvidia,pins = "ddc";
+				nvidia,function = "rsvd4";
+			};
+			pta {
+				nvidia,pins = "pta";
+				nvidia,function = "i2c2";
+			};
+		};
+
+		state_i2cmux_idle: pinmux_i2cmux_idle {
+			ddc {
+				nvidia,pins = "ddc";
+				nvidia,function = "rsvd4";
+			};
+			pta {
+				nvidia,pins = "pta";
+				nvidia,function = "rsvd4";
+			};
+		};
+	};
+
+	i2s at 70002800 {
+		status = "okay";
+	};
+
+	serial at 70006300 {
+		status = "okay";
+	};
+
+	i2c at 7000c000 {
+		clock-frequency = <400000>;
+		status = "okay";
+	};
+
+	i2c at 7000c400 {
+		clock-frequency = <100000>;
+		status = "okay";
+	};
+
+	i2cmux {
+		compatible = "i2c-mux-pinctrl";
+		#address-cells = <1>;
+		#size-cells = <0>;
+
+		i2c-parent = <&{/i2c@7000c400}>;
+
+		pinctrl-names = "ddc", "pta", "idle";
+		pinctrl-0 = <&state_i2cmux_ddc>;
+		pinctrl-1 = <&state_i2cmux_pta>;
+		pinctrl-2 = <&state_i2cmux_idle>;
+
+		hdmi_ddc: i2c at 0 {
+			reg = <0>;
+			#address-cells = <1>;
+			#size-cells = <0>;
+		};
+
+		i2c at 1 {
+			reg = <1>;
+			#address-cells = <1>;
+			#size-cells = <0>;
+		};
+	};
+
+	i2c at 7000d000 {
+		clock-frequency = <400000>;
+		status = "okay";
+
+		pmic: tps6586x at 34 {
+			compatible = "ti,tps6586x";
+			reg = <0x34>;
+			interrupts = <0 86 0x4>;
+
+			ti,system-power-controller;
+
+			#gpio-cells = <2>;
+			gpio-controller;
+
+			sys-supply = <&vdd_5v0_reg>;
+			vin-sm0-supply = <&sys_reg>;
+			vin-sm1-supply = <&sys_reg>;
+			vin-sm2-supply = <&sys_reg>;
+			vinldo01-supply = <&sm2_reg>;
+			vinldo23-supply = <&sm2_reg>;
+			vinldo4-supply = <&sm2_reg>;
+			vinldo678-supply = <&sm2_reg>;
+			vinldo9-supply = <&sm2_reg>;
+
+			regulators {
+				sys_reg: sys {
+					regulator-name = "vdd_sys";
+					regulator-always-on;
+				};
+
+				sm0 {
+					regulator-name = "vdd_sys_sm0,vdd_core";
+					regulator-min-microvolt = <1200000>;
+					regulator-max-microvolt = <1200000>;
+					regulator-always-on;
+				};
+
+				sm1 {
+					regulator-name = "vdd_sys_sm1,vdd_cpu";
+					regulator-min-microvolt = <1000000>;
+					regulator-max-microvolt = <1000000>;
+					regulator-always-on;
+				};
+
+				sm2_reg: sm2 {
+					regulator-name = "vdd_sys_sm2,vin_ldo*";
+					regulator-min-microvolt = <3700000>;
+					regulator-max-microvolt = <3700000>;
+					regulator-always-on;
+				};
+
+				ldo0 {
+					regulator-name = "vdd_ldo0,vddio_pex_clk";
+					regulator-min-microvolt = <3300000>;
+					regulator-max-microvolt = <3300000>;
+				};
+
+				ldo1 {
+					regulator-name = "vdd_ldo1,avdd_pll*";
+					regulator-min-microvolt = <1100000>;
+					regulator-max-microvolt = <1100000>;
+					regulator-always-on;
+				};
+
+				ldo2 {
+					regulator-name = "vdd_ldo2,vdd_rtc";
+					regulator-min-microvolt = <1200000>;
+					regulator-max-microvolt = <1200000>;
+				};
+
+				ldo3 {
+					regulator-name = "vdd_ldo3,avdd_usb*";
+					regulator-min-microvolt = <3300000>;
+					regulator-max-microvolt = <3300000>;
+					regulator-always-on;
+				};
+
+				ldo4 {
+					regulator-name = "vdd_ldo4,avdd_osc,vddio_sys";
+					regulator-min-microvolt = <1800000>;
+					regulator-max-microvolt = <1800000>;
+					regulator-always-on;
+				};
+
+				ldo5 {
+					regulator-name = "vdd_ldo5,vcore_mmc";
+					regulator-min-microvolt = <2850000>;
+					regulator-max-microvolt = <2850000>;
+				};
+
+				ldo6 {
+					regulator-name = "vdd_ldo6,avdd_vdac";
+					/*
+					 * According to the Tegra 2 Automotive
+					 * DataSheet, a typical value for this
+					 * would be 2.8V, but the PMIC only
+					 * supports 2.85V.
+					 */
+					regulator-min-microvolt = <2850000>;
+					regulator-max-microvolt = <2850000>;
+				};
+
+				hdmi_vdd_reg: ldo7 {
+					regulator-name = "vdd_ldo7,avdd_hdmi";
+					regulator-min-microvolt = <3300000>;
+					regulator-max-microvolt = <3300000>;
+				};
+
+				hdmi_pll_reg: ldo8 {
+					regulator-name = "vdd_ldo8,avdd_hdmi_pll";
+					regulator-min-microvolt = <1800000>;
+					regulator-max-microvolt = <1800000>;
+				};
+
+				ldo9 {
+					regulator-name = "vdd_ldo9,vdd_ddr_rx,avdd_cam";
+					/*
+					 * According to the Tegra 2 Automotive
+					 * DataSheet, a typical value for this
+					 * would be 2.8V, but the PMIC only
+					 * supports 2.85V.
+					 */
+					regulator-min-microvolt = <2850000>;
+					regulator-max-microvolt = <2850000>;
+					regulator-always-on;
+				};
+
+				ldo_rtc {
+					regulator-name = "vdd_rtc_out";
+					regulator-min-microvolt = <3300000>;
+					regulator-max-microvolt = <3300000>;
+					regulator-always-on;
+				};
+			};
+		};
+
+		temperature-sensor at 4c {
+			compatible = "onnn,nct1008";
+			reg = <0x4c>;
+		};
+	};
+
+	pmc {
+		nvidia,invert-interrupt;
+	};
+
+	usb at c5008000 {
+		status = "okay";
+	};
+
+	sdhci at c8000600 {
+		cd-gpios = <&gpio 58 0>; /* gpio PH2 */
+		wp-gpios = <&gpio 59 0>; /* gpio PH3 */
+		bus-width = <4>;
+		status = "okay";
+	};
+
+	regulators {
+		compatible = "simple-bus";
+
+		#address-cells = <1>;
+		#size-cells = <0>;
+
+		vdd_5v0_reg: regulator at 0 {
+			compatible = "regulator-fixed";
+			reg = <0>;
+			regulator-name = "vdd_5v0";
+			regulator-min-microvolt = <5000000>;
+			regulator-max-microvolt = <5000000>;
+			regulator-always-on;
+		};
+	};
+};
arch/arm/dts/tegra20.dtsi +12 -4
index 9a89685..982a070 100644
@@ -318,24 +318,32 @@
 	sdhci at c8000000 {
 		compatible = "nvidia,tegra20-sdhci";
 		reg = <0xc8000000 0x200>;
-		interrupts = < 46 >;
+		interrupts = <0 14 0x04>;
+		clocks = <&tegra_car 14>;
+		status = "disabled";
 	};
 
 	sdhci at c8000200 {
 		compatible = "nvidia,tegra20-sdhci";
 		reg = <0xc8000200 0x200>;
-		interrupts = < 47 >;
+		interrupts = <0 15 0x04>;
+		clocks = <&tegra_car 9>;
+		status = "disabled";
 	};
 
 	sdhci at c8000400 {
 		compatible = "nvidia,tegra20-sdhci";
 		reg = <0xc8000400 0x200>;
-		interrupts = < 51 >;
+		interrupts = <0 19 0x04>;
+		clocks = <&tegra_car 69>;
+		status = "disabled";
 	};
 
 	sdhci at c8000600 {
 		compatible = "nvidia,tegra20-sdhci";
 		reg = <0xc8000600 0x200>;
-		interrupts = < 63 >;
+		interrupts = <0 31 0x04>;
+		clocks = <&tegra_car 15>;
+		status = "disabled";
 	};
 };
board/avionic-design/dts/tegra20-medcom-wide.dts +1 -0
index e46afbe..2444f67 100644
@@ -8,6 +8,7 @@
 
 	aliases {
 		usb0 = "/usb at c5008000";
+		sdhci0 = "/sdhci at c8000600";
 	};
 
 	memory {
board/avionic-design/dts/tegra20-plutux.dts +1 -0
index 3e6cce0..a9dbdc3 100644
@@ -8,6 +8,7 @@
 
 	aliases {
 		usb0 = "/usb at c5008000";
+		sdhci0 = "/sdhci at c8000600";
 	};
 
 	memory {
board/avionic-design/dts/tegra20-tec.dts +1 -0
index bf3ff1d..6afdd5b 100644
@@ -8,6 +8,7 @@
 
 	aliases {
 		usb0 = "/usb at c5008000";
+		sdhci0 = "/sdhci at c8000600";
 	};
 
 	memory {
board/compal/dts/tegra20-paz00.dts +17 -2
index 31b064d..6476940 100644
@@ -3,11 +3,13 @@
 /include/ ARCH_CPU_DTS
 
 / {
-        model = "Toshiba AC100 / Dynabook AZ";
-        compatible = "compal,paz00", "nvidia,tegra20";
+	model = "Toshiba AC100 / Dynabook AZ";
+	compatible = "compal,paz00", "nvidia,tegra20";
 
 	aliases {
 		usb0 = "/usb at c5008000";
+		sdhci0 = "/sdhci at c8000600";
+		sdhci1 = "/sdhci at c8000000";
 	};
 
 	memory {
@@ -53,6 +55,19 @@
 		status = "disabled";
 	};
 
+	sdhci at c8000000 {
+		status = "okay";
+		cd-gpios = <&gpio 173 0>; /* gpio PV5 */
+		wp-gpios = <&gpio 57 0>; /* gpio PH1 */
+		power-gpios = <&gpio 169 0>; /* gpio PV1 */
+		bus-width = <4>;
+	};
+
+	sdhci at c8000600 {
+		status = "okay";
+		bus-width = <8>;
+	};
+
 	lcd_panel: panel {
 		/* PAZ00 has 1024x600 */
 		clock = <54030000>;
board/compulab/dts/tegra20-trimslice.dts +14 -0
index 7aeed67..2957f45 100644
@@ -9,6 +9,8 @@
 	aliases {
 		usb0 = "/usb at c5008000";
 		usb1 = "/usb at c5000000";
+		sdhci0 = "/sdhci at c8000600";
+		sdhci1 = "/sdhci at c8000000";
 	};
 
 	memory {
@@ -47,4 +49,16 @@
 	usb at c5004000 {
 		status = "disabled";
 	};
+
+	sdhci at c8000000 {
+		status = "okay";
+		bus-width = <4>;
+	};
+
+	sdhci at c8000600 {
+		status = "okay";
+		cd-gpios = <&gpio 121 0>; /* gpio PP1 */
+		wp-gpios = <&gpio 122 0>; /* gpio PP2 */
+		bus-width = <4>;
+	};
 };
board/nvidia/dts/tegra20-harmony.dts +18 -0
index aeda3a1..593885d 100644
@@ -9,6 +9,8 @@
 	aliases {
 		usb0 = "/usb at c5008000";
 		usb1 = "/usb at c5004000";
+		sdhci0 = "/sdhci at c8000600";
+		sdhci1 = "/sdhci at c8000200";
 	};
 
 	memory {
@@ -52,4 +54,20 @@
 	usb at c5004000 {
 		nvidia,phy-reset-gpio = <&gpio 169 0>; /* gpio PV1 */
 	};
+
+	sdhci at c8000200 {
+		status = "okay";
+		cd-gpios = <&gpio 69 0>; /* gpio PI5 */
+		wp-gpios = <&gpio 57 0>; /* gpio PH1 */
+		power-gpios = <&gpio 155 0>; /* gpio PT3 */
+		bus-width = <4>;
+	};
+
+	sdhci at c8000600 {
+		status = "okay";
+		cd-gpios = <&gpio 58 0>; /* gpio PH2 */
+		wp-gpios = <&gpio 59 0>; /* gpio PH3 */
+		power-gpios = <&gpio 70 0>; /* gpio PI6 */
+		bus-width = <8>;
+	};
 };
board/nvidia/dts/tegra20-seaboard.dts +8 -4
index 527a296..dccb65e 100644
@@ -12,14 +12,15 @@
 	};
 
 	aliases {
-		/* This defines the order of our USB ports */
+		/* This defines the order of our ports */
 		usb0 = "/usb at c5008000";
 		usb1 = "/usb at c5000000";
-
 		i2c0 = "/i2c at 7000d000";
 		i2c1 = "/i2c at 7000c000";
 		i2c2 = "/i2c at 7000c400";
 		i2c3 = "/i2c at 7000c500";
+		sdhci0 = "/sdhci at c8000600";
+		sdhci1 = "/sdhci at c8000400";
 	};
 
 	memory {
@@ -156,13 +157,16 @@
 	};
 
 	sdhci at c8000400 {
+		status = "okay";
 		cd-gpios = <&gpio 69 0>; /* gpio PI5 */
 		wp-gpios = <&gpio 57 0>; /* gpio PH1 */
-		power-gpios = <&gpio 70 0>; /* gpio PI6 */
+		power-gpios = <&gpio 70 3>; /* gpio PI6 */
+		bus-width = <4>;
 	};
 
 	sdhci at c8000600 {
-		support-8bit;
+		status = "okay";
+		bus-width = <8>;
 	};
 
 	lcd_panel: panel {
board/nvidia/dts/tegra20-ventana.dts +15 -0
index 3e5e39d..0423595 100644
@@ -8,6 +8,8 @@
 
 	aliases {
 		usb0 = "/usb at c5008000";
+		sdhci0 = "/sdhci at c8000600";
+		sdhci1 = "/sdhci at c8000400";
 	};
 
 	memory {
@@ -41,4 +43,17 @@
 	usb at c5004000 {
 		status = "disabled";
 	};
+
+	sdhci at c8000400 {
+		status = "okay";
+		cd-gpios = <&gpio 69 0>; /* gpio PI5 */
+		wp-gpios = <&gpio 57 0>; /* gpio PH1 */
+		power-gpios = <&gpio 70 0>; /* gpio PI6 */
+		bus-width = <4>;
+	};
+
+	sdhci at c8000600 {
+		status = "okay";
+		bus-width = <8>;
+	};
 };
board/nvidia/dts/tegra20-whistler.dts +13 -0
index 4579557..2d0a2ab 100644
@@ -9,6 +9,8 @@
 	aliases {
 		i2c0 = "/i2c at 7000d000";
 		usb0 = "/usb at c5008000";
+		sdhci0 = "/sdhci at c8000600";
+		sdhci1 = "/sdhci at c8000400";
 	};
 
 	memory {
@@ -57,4 +59,15 @@
 	usb at c5004000 {
 		status = "disabled";
 	};
+
+	sdhci at c8000400 {
+		status = "okay";
+		wp-gpios = <&gpio 173 0>; /* gpio PV5 */
+		bus-width = <8>;
+	};
+
+	sdhci at c8000600 {
+		status = "okay";
+		bus-width = <8>;
+	};
 };
board/toradex/dts/tegra20-colibri_t20_iris.dts +6 -0
index c29b43a..0dd4c43 100644
@@ -10,6 +10,7 @@
 		usb0 = "/usb at c5008000";
 		usb1 = "/usb at c5000000";
 		usb2 = "/usb at c5004000";
+		sdhci0 = "/sdhci at c8000600";
 	};
 
 	usb at c5000000 {
@@ -35,4 +36,9 @@
 			compatible = "nand-flash";
 		};
 	};
+
+	sdhci at c8000600 {
+		status = "okay";
+		bus-width = <4>;
+	};
 };
Signature · 1 line
1.7.0.4
#2 lore raw

PATCH v2 1/2 Tegra: fdt: Add/enhance sdhci (mmc) nodes for all T20 DT files

Hi Tom,

Am Montag, den 11.02.2013, 10:17 -0700 schrieb Tom Warren:

Quoted 4 lines

Linux dts files were used for those boards that didn't already have sdhci info populated. Tamonten has their own dtsi file with common sdhci nodes (sourced from Linux).

Quoted 8 lines

---
v2:
- cleanup comments in dts files/match w/kernel files
- add sdhci aliases in all dts files
- use tegra20-tamonten.dtsi from the kernel for AD boards

 arch/arm/dts/tegra20-tamonten.dtsi               |  489 ++++++++++++++++++++++

I'm not sure if pushing the whole file in this patch is the right thing to do.

Quoted 14 lines
 arch/arm/dts/tegra20.dtsi                        |   16 +-
 board/avionic-design/dts/tegra20-medcom-wide.dts |    1 +
 board/avionic-design/dts/tegra20-plutux.dts      |    1 +
 board/avionic-design/dts/tegra20-tec.dts         |    1 +
 board/compal/dts/tegra20-paz00.dts               |   19 +-
 board/compulab/dts/tegra20-trimslice.dts         |   14 +
 board/nvidia/dts/tegra20-harmony.dts             |   18 +
 board/nvidia/dts/tegra20-seaboard.dts            |   12 +-
 board/nvidia/dts/tegra20-ventana.dts             |   15 +
 board/nvidia/dts/tegra20-whistler.dts            |   13 +
 board/toradex/dts/tegra20-colibri_t20_iris.dts   |    6 +
 12 files changed, 595 insertions(+), 10 deletions(-)
 create mode 100644 arch/arm/dts/tegra20-tamonten.dtsi

[...]

Quoted 13 lines
board/toradex/dts/tegra20-colibri_t20_iris.dts +1 -0
index c29b43a..0dd4c43 100644
@@ -10,6 +10,7 @@
 		usb0 = "/usb at c5008000";
 		usb1 = "/usb at c5000000";
 		usb2 = "/usb at c5004000";
+		sdhci0 = "/sdhci at c8000600";
 	};
 
 	usb at c5000000 {

board/toradex/dts/tegra20-colibri_t20_iris.dts line 42

Quoted 7 lines
board/toradex/dts/tegra20-colibri_t20_iris.dts +4 -0
@@ -35,4 +36,9 @@
 			compatible = "nand-flash";
 		};
 	};
+
+	sdhci at c8000600 {
+		status = "okay";
+		bus-width = <4>;

You are missing the CD GPIO here. It's PC7.

board/toradex/dts/tegra20-colibri_t20_iris.dts line 44

Quoted 2 lines
board/toradex/dts/tegra20-colibri_t20_iris.dts +1 -0
+	};
 };

Regards,
Lucas

author
#3 lore raw

PATCH v2 1/2 Tegra: fdt: Add/enhance sdhci (mmc) nodes for all T20 DT files

Lucas,

Quoted 14 lines · Lucas Stach

Hi Tom,

Am Montag, den 11.02.2013, 10:17 -0700 schrieb Tom Warren:

Quoted 11 lines

Linux dts files were used for those boards that didn't already have sdhci info populated. Tamonten has their own dtsi file with common sdhci nodes (sourced from Linux).

---
v2:
- cleanup comments in dts files/match w/kernel files
- add sdhci aliases in all dts files
- use tegra20-tamonten.dtsi from the kernel for AD boards

Quoted 4 lines
Quoted 1 line
 arch/arm/dts/tegra20-tamonten.dtsi               |  489 ++++++++++++++++++++++

I'm not sure if pushing the whole file in this patch is the right thing to do.

I didn't want to edit it since we seem to be moving towards using the Linux DTS files in toto in U-Boot (as per Stephen & Simon). Does it do any harm to have the whole thing here? Saves some work later. Thierry
- what do you think?

Quoted 27 lines
Quoted 15 lines
 arch/arm/dts/tegra20.dtsi                        |   16 +-
 board/avionic-design/dts/tegra20-medcom-wide.dts |    1 +
 board/avionic-design/dts/tegra20-plutux.dts      |    1 +
 board/avionic-design/dts/tegra20-tec.dts         |    1 +
 board/compal/dts/tegra20-paz00.dts               |   19 +-
 board/compulab/dts/tegra20-trimslice.dts         |   14 +
 board/nvidia/dts/tegra20-harmony.dts             |   18 +
 board/nvidia/dts/tegra20-seaboard.dts            |   12 +-
 board/nvidia/dts/tegra20-ventana.dts             |   15 +
 board/nvidia/dts/tegra20-whistler.dts            |   13 +
 board/toradex/dts/tegra20-colibri_t20_iris.dts   |    6 +
 12 files changed, 595 insertions(+), 10 deletions(-)
 create mode 100644 arch/arm/dts/tegra20-tamonten.dtsi

[...]

Quoted 11 lines
board/toradex/dts/tegra20-colibri_t20_iris.dts +1 -0
index c29b43a..0dd4c43 100644
@@ -10,6 +10,7 @@
              usb0 = "/usb at c5008000";
              usb1 = "/usb at c5000000";
              usb2 = "/usb at c5004000";
+             sdhci0 = "/sdhci at c8000600";
      };
 
Quoted 10 lines

board/toradex/dts/tegra20-colibri_t20_iris.dts line 42

Quoted 9 lines
board/toradex/dts/tegra20-colibri_t20_iris.dts +4 -0
      usb at c5000000 {
@@ -35,4 +36,9 @@
                      compatible = "nand-flash";
              };
      };
+
+     sdhci at c8000600 {
+             status = "okay";
+             bus-width = <4>;

You are missing the CD GPIO here. It's PC7.

Thanks, I'll add it.

Tom

Quoted 8 lines

board/toradex/dts/tegra20-colibri_t20_iris.dts line 44

Quoted 3 lines
board/toradex/dts/tegra20-colibri_t20_iris.dts +1 -0
+     };
 };

Regards,
Lucas

#4 lore raw

PATCH v2 1/2 Tegra: fdt: Add/enhance sdhci (mmc) nodes for all T20 DT files

Am Montag, den 11.02.2013, 10:56 -0700 schrieb Tom Warren:

Quoted 19 lines

Lucas,

Quoted 16 lines · Lucas Stach

Hi Tom,

Am Montag, den 11.02.2013, 10:17 -0700 schrieb Tom Warren:

Quoted 12 lines

Linux dts files were used for those boards that didn't already have sdhci info populated. Tamonten has their own dtsi file with common sdhci nodes (sourced from Linux).

---
v2:
- cleanup comments in dts files/match w/kernel files
- add sdhci aliases in all dts files
- use tegra20-tamonten.dtsi from the kernel for AD boards

 arch/arm/dts/tegra20-tamonten.dtsi               |  489 ++++++++++++++++++++++
Quoted 8 lines
Quoted 2 lines

I'm not sure if pushing the whole file in this patch is the right thing to do.

I didn't want to edit it since we seem to be moving towards using the Linux DTS files in toto in U-Boot (as per Stephen & Simon). Does it do any harm to have the whole thing here? Saves some work later. Thierry
- what do you think?

I'm not objecting against pulling in the kernel dtsi in general, as I think most of the things in there like pinmux will become handy in a bit. I would just like to see this huge chunk in a separate patch and this one would then just add the aliases.

Regards,
Lucas

#5 lore raw

PATCH v2 1/2 Tegra: fdt: Add/enhance sdhci (mmc) nodes for all T20 DT files

In reply to Tom Warren
Quoted 19 lines · Tom Warren

Lucas,

Quoted 16 lines · Lucas Stach

Hi Tom,

Am Montag, den 11.02.2013, 10:17 -0700 schrieb Tom Warren:

Quoted 12 lines

Linux dts files were used for those boards that didn't already have sdhci info populated. Tamonten has their own dtsi file with common sdhci nodes (sourced from Linux).

---
v2:
- cleanup comments in dts files/match w/kernel files
- add sdhci aliases in all dts files
- use tegra20-tamonten.dtsi from the kernel for AD boards

 arch/arm/dts/tegra20-tamonten.dtsi               |  489 ++++++++++++++++++++++
Quoted 7 lines
Quoted 2 lines

I'm not sure if pushing the whole file in this patch is the right thing to do.

I didn't want to edit it since we seem to be moving towards using the Linux DTS files in toto in U-Boot (as per Stephen & Simon). Does it do any harm to have the whole thing here? Saves some work later. Thierry
- what do you think?

Given that it isn't used anywhere I don't think we really need it right now. We can always add it later when we can make better use of it.

Thierry
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 836 bytes
Desc: not available
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20130211/833e9773/attachment.pgp>

author
#6 lore raw

PATCH v2 1/2 Tegra: fdt: Add/enhance sdhci (mmc) nodes for all T20 DT files

Thierry/Lucas,

Quoted 23 lines · Thierry Reding
Quoted 22 lines · Tom Warren

Lucas,

Quoted 18 lines · Lucas Stach

Hi Tom,

Am Montag, den 11.02.2013, 10:17 -0700 schrieb Tom Warren:

Quoted 4 lines

Linux dts files were used for those boards that didn't already have sdhci info populated. Tamonten has their own dtsi file with common sdhci nodes (sourced from Linux).

Quoted 8 lines

---
v2:
- cleanup comments in dts files/match w/kernel files
- add sdhci aliases in all dts files
- use tegra20-tamonten.dtsi from the kernel for AD boards

 arch/arm/dts/tegra20-tamonten.dtsi               |  489 ++++++++++++++++++++++

I'm not sure if pushing the whole file in this patch is the right thing to do.

Quoted 7 lines
Quoted 4 lines

I didn't want to edit it since we seem to be moving towards using the Linux DTS files in toto in U-Boot (as per Stephen & Simon). Does it do any harm to have the whole thing here? Saves some work later. Thierry
- what do you think?

Given that it isn't used anywhere I don't think we really need it right now. We can always add it later when we can make better use of it.

It actually is used (for SDMMC/sdhci) now, Thierry. That's why it's in this patchset. I had originally put the sdhci node for Avionic Design boards in their respective .dts files, but Stephen pointed out that the kernel had a tegra20-tamonten.dtsi file with common info, which included the sdhci node, and asked that I use it, instead, so we echo the kernel layout. So I pulled that file into my MMC DT patchset, and used it in all AD board builds (medcom/tec/plutux) - it's pulled in via an override of CONFIG_ARCH_DEVICE_TREE in the config files.

So the options seem to be:

a) Don't use the tamonton dtsi file, and put the sdhci nodes in the AD dts files, just like all other boards (this was my V1 approach). Vetoed by Stephen.
b) Use tegra20-tamonten.dtsi as is, identical to the kernel file. If necessary, I can move it's inclusion to a separate patch, independent of the MMC DT patchset, as suggested by Lucas.
c) Use tegra20-tamonten.dtsi, but just with the sdhci node (is this what you're suggesting, Thierry?). I'd still pull it in via a CONFIG_ARCH_DEVICE_TREE #define in the AD config files.

Let me know ASAP - I'd like to get V3 upstreamed soon so I can move on to work on the T30/T114 MMC patches.

Tom

Quoted 2 lines

Thierry

#7 lore raw

PATCH v2 1/2 Tegra: fdt: Add/enhance sdhci (mmc) nodes for all T20 DT files

Quoted 32 lines · Tom Warren

Thierry/Lucas,

Quoted 28 lines · Thierry Reding
Quoted 26 lines · Tom Warren

Lucas,

Quoted 14 lines · Lucas Stach

Hi Tom,

Am Montag, den 11.02.2013, 10:17 -0700 schrieb Tom Warren:

Quoted 11 lines

Linux dts files were used for those boards that didn't already have sdhci info populated. Tamonten has their own dtsi file with common sdhci nodes (sourced from Linux).

---
v2:
- cleanup comments in dts files/match w/kernel files
- add sdhci aliases in all dts files
- use tegra20-tamonten.dtsi from the kernel for AD boards

Quoted 4 lines
Quoted 1 line
 arch/arm/dts/tegra20-tamonten.dtsi               |  489 ++++++++++++++++++++++

I'm not sure if pushing the whole file in this patch is the right thing to do.

I didn't want to edit it since we seem to be moving towards using the Linux DTS files in toto in U-Boot (as per Stephen & Simon). Does it do any harm to have the whole thing here? Saves some work later. Thierry
- what do you think?

Quoted 5 lines
Quoted 2 lines

Given that it isn't used anywhere I don't think we really need it right now. We can always add it later when we can make better use of it.

It actually is used (for SDMMC/sdhci) now, Thierry. That's why it's in this patchset.

Right, I hadn't looked at that patch yet.

Quoted 20 lines

I had originally put the sdhci node for Avionic Design boards in their respective .dts files, but Stephen pointed out that the kernel had a tegra20-tamonten.dtsi file with common info, which included the sdhci node, and asked that I use it, instead, so we echo the kernel layout. So I pulled that file into my MMC DT patchset, and used it in all AD board builds (medcom/tec/plutux) - it's pulled in via an override of CONFIG_ARCH_DEVICE_TREE in the config files.

So the options seem to be:

a) Don't use the tamonton dtsi file, and put the sdhci nodes in the AD dts files, just like all other boards (this was my V1 approach). Vetoed by Stephen.
b) Use tegra20-tamonten.dtsi as is, identical to the kernel file. If necessary, I can move it's inclusion to a separate patch, independent of the MMC DT patchset, as suggested by Lucas.
c) Use tegra20-tamonten.dtsi, but just with the sdhci node (is this what you're suggesting, Thierry?). I'd still pull it in via a CONFIG_ARCH_DEVICE_TREE #define in the AD config files.

Quoted 2 lines

Let me know ASAP - I'd like to get V3 upstreamed soon so I can move on to work on the T30/T114 MMC patches.

I think option b) sounds fine given that we want to move to the same DTS as the kernel eventually anyway. So for the Tamonten (and AD board) pieces, consider this:

I can't give you a Tested-by because I have a bunch of other things to take care of and I probably won't get to testing this for a few days.

Thierry
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 836 bytes
Desc: not available
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20130212/186540ad/attachment.pgp>

#8 lore raw

PATCH v2 1/2 Tegra: fdt: Add/enhance sdhci (mmc) nodes for all T20 DT files

Quoted 64 lines · Thierry Reding
Quoted 32 lines · Tom Warren

Thierry/Lucas,

Quoted 28 lines · Thierry Reding
Quoted 26 lines · Tom Warren

Lucas,

Quoted 14 lines · Lucas Stach

Hi Tom,

Am Montag, den 11.02.2013, 10:17 -0700 schrieb Tom Warren:

Quoted 11 lines

Linux dts files were used for those boards that didn't already have sdhci info populated. Tamonten has their own dtsi file with common sdhci nodes (sourced from Linux).

---
v2:
- cleanup comments in dts files/match w/kernel files
- add sdhci aliases in all dts files
- use tegra20-tamonten.dtsi from the kernel for AD boards

Quoted 4 lines
Quoted 1 line
 arch/arm/dts/tegra20-tamonten.dtsi               |  489 ++++++++++++++++++++++

I'm not sure if pushing the whole file in this patch is the right thing to do.

I didn't want to edit it since we seem to be moving towards using the Linux DTS files in toto in U-Boot (as per Stephen & Simon). Does it do any harm to have the whole thing here? Saves some work later. Thierry
- what do you think?

Quoted 5 lines
Quoted 2 lines

Given that it isn't used anywhere I don't think we really need it right now. We can always add it later when we can make better use of it.

It actually is used (for SDMMC/sdhci) now, Thierry. That's why it's in this patchset.

Right, I hadn't looked at that patch yet.

Quoted 20 lines

I had originally put the sdhci node for Avionic Design boards in their respective .dts files, but Stephen pointed out that the kernel had a tegra20-tamonten.dtsi file with common info, which included the sdhci node, and asked that I use it, instead, so we echo the kernel layout. So I pulled that file into my MMC DT patchset, and used it in all AD board builds (medcom/tec/plutux) - it's pulled in via an override of CONFIG_ARCH_DEVICE_TREE in the config files.

So the options seem to be:

a) Don't use the tamonton dtsi file, and put the sdhci nodes in the AD dts files, just like all other boards (this was my V1 approach). Vetoed by Stephen.
b) Use tegra20-tamonten.dtsi as is, identical to the kernel file. If necessary, I can move it's inclusion to a separate patch, independent of the MMC DT patchset, as suggested by Lucas.
c) Use tegra20-tamonten.dtsi, but just with the sdhci node (is this what you're suggesting, Thierry?). I'd still pull it in via a CONFIG_ARCH_DEVICE_TREE #define in the AD config files.

Quoted 2 lines

Let me know ASAP - I'd like to get V3 upstreamed soon so I can move on to work on the T30/T114 MMC patches.

Quoted 8 lines

I think option b) sounds fine given that we want to move to the same DTS as the kernel eventually anyway. So for the Tamonten (and AD board) pieces, consider this:

I can't give you a Tested-by because I have a bunch of other things to take care of and I probably won't get to testing this for a few days.

So it turned out that I need to touch U-Boot anyway, so I decided to give this a spin. I noticed that overriding CONFIG_ARCH_DEVICE_TREE from the board configuration file doesn't work currently. What happens is that the autoconf.mk (which is derived from the board configuration) is included before the CPU config.mk which sets CONFIG_ARCH_DEVICE_TREE to tegra20 (or tegra30, tegra114). I came up with the attached patch to set the variable if not set previously (by the board configuration file).

Feel free to squash that in your patch series if you deem it a proper solution. I can also provide a proper separate patch if you prefer.

Thierry
-------------- next part --------------

arch/arm/cpu/armv7/tegra114/config.mk +1 -1
index cb1a19d..e7c22c0 100644
@@ -16,4 +16,4 @@
 # You should have received a copy of the GNU General Public License
 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
 #
-CONFIG_ARCH_DEVICE_TREE := tegra114
+CONFIG_ARCH_DEVICE_TREE ?= tegra114
arch/arm/cpu/armv7/tegra20/config.mk +1 -1
index 6432e75..9042664 100644
@@ -23,4 +23,4 @@
 # Foundation, Inc., 59 Temple Place, Suite 330, Boston,
 # MA 02111-1307 USA
 #
-CONFIG_ARCH_DEVICE_TREE := tegra20
+CONFIG_ARCH_DEVICE_TREE ?= tegra20
arch/arm/cpu/armv7/tegra30/config.mk +1 -1
index 719ca81..0035bc5 100644
@@ -16,4 +16,4 @@
 # You should have received a copy of the GNU General Public License
 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
 #
-CONFIG_ARCH_DEVICE_TREE := tegra30
+CONFIG_ARCH_DEVICE_TREE ?= tegra30

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 836 bytes
Desc: not available
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20130212/66ee8d36/attachment.pgp>

#9 lore raw

PATCH v2 1/2 Tegra: fdt: Add/enhance sdhci (mmc) nodes for all T20 DT files

Quoted 112 lines · Thierry Reding
Quoted 64 lines · Thierry Reding
Quoted 32 lines · Tom Warren

Thierry/Lucas,

Quoted 28 lines · Thierry Reding
Quoted 26 lines · Tom Warren

Lucas,

Quoted 14 lines · Lucas Stach

Hi Tom,

Am Montag, den 11.02.2013, 10:17 -0700 schrieb Tom Warren:

Quoted 11 lines

Linux dts files were used for those boards that didn't already have sdhci info populated. Tamonten has their own dtsi file with common sdhci nodes (sourced from Linux).

---
v2:
- cleanup comments in dts files/match w/kernel files
- add sdhci aliases in all dts files
- use tegra20-tamonten.dtsi from the kernel for AD boards

Quoted 4 lines
Quoted 1 line
 arch/arm/dts/tegra20-tamonten.dtsi               |  489 ++++++++++++++++++++++

I'm not sure if pushing the whole file in this patch is the right thing to do.

I didn't want to edit it since we seem to be moving towards using the Linux DTS files in toto in U-Boot (as per Stephen & Simon). Does it do any harm to have the whole thing here? Saves some work later. Thierry
- what do you think?

Quoted 5 lines
Quoted 2 lines

Given that it isn't used anywhere I don't think we really need it right now. We can always add it later when we can make better use of it.

It actually is used (for SDMMC/sdhci) now, Thierry. That's why it's in this patchset.

Right, I hadn't looked at that patch yet.

Quoted 20 lines

I had originally put the sdhci node for Avionic Design boards in their respective .dts files, but Stephen pointed out that the kernel had a tegra20-tamonten.dtsi file with common info, which included the sdhci node, and asked that I use it, instead, so we echo the kernel layout. So I pulled that file into my MMC DT patchset, and used it in all AD board builds (medcom/tec/plutux) - it's pulled in via an override of CONFIG_ARCH_DEVICE_TREE in the config files.

So the options seem to be:

a) Don't use the tamonton dtsi file, and put the sdhci nodes in the AD dts files, just like all other boards (this was my V1 approach). Vetoed by Stephen.
b) Use tegra20-tamonten.dtsi as is, identical to the kernel file. If necessary, I can move it's inclusion to a separate patch, independent of the MMC DT patchset, as suggested by Lucas.
c) Use tegra20-tamonten.dtsi, but just with the sdhci node (is this what you're suggesting, Thierry?). I'd still pull it in via a CONFIG_ARCH_DEVICE_TREE #define in the AD config files.

Quoted 2 lines

Let me know ASAP - I'd like to get V3 upstreamed soon so I can move on to work on the T30/T114 MMC patches.

Quoted 8 lines

I think option b) sounds fine given that we want to move to the same DTS as the kernel eventually anyway. So for the Tamonten (and AD board) pieces, consider this:

I can't give you a Tested-by because I have a bunch of other things to take care of and I probably won't get to testing this for a few days.

So it turned out that I need to touch U-Boot anyway, so I decided to give this a spin. I noticed that overriding CONFIG_ARCH_DEVICE_TREE from the board configuration file doesn't work currently. What happens is that the autoconf.mk (which is derived from the board configuration) is included before the CPU config.mk which sets CONFIG_ARCH_DEVICE_TREE to tegra20 (or tegra30, tegra114). I came up with the attached patch to set the variable if not set previously (by the board configuration file).

Feel free to squash that in your patch series if you deem it a proper solution. I can also provide a proper separate patch if you prefer.

Thierry

arch/arm/cpu/armv7/tegra114/config.mk +1 -1
index cb1a19d..e7c22c0 100644
@@ -16,4 +16,4 @@
 # You should have received a copy of the GNU General Public License
 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
 #
-CONFIG_ARCH_DEVICE_TREE := tegra114
+CONFIG_ARCH_DEVICE_TREE ?= tegra114
arch/arm/cpu/armv7/tegra20/config.mk +1 -1
index 6432e75..9042664 100644
@@ -23,4 +23,4 @@
 # Foundation, Inc., 59 Temple Place, Suite 330, Boston,
 # MA 02111-1307 USA
 #
-CONFIG_ARCH_DEVICE_TREE := tegra20
+CONFIG_ARCH_DEVICE_TREE ?= tegra20
arch/arm/cpu/armv7/tegra30/config.mk +0 -0
index 719ca81..0035bc5 100644

arch/arm/cpu/armv7/tegra30/config.mk line 19

Quoted 5 lines
arch/arm/cpu/armv7/tegra30/config.mk +1 -1
@@ -16,4 +16,4 @@
 # You should have received a copy of the GNU General Public License
 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
 #
-CONFIG_ARCH_DEVICE_TREE := tegra30
+CONFIG_ARCH_DEVICE_TREE ?= tegra30

I forgot: with that patch applied:

Thierry
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 836 bytes
Desc: not available
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20130212/1cf2fa02/attachment.pgp>

author
#10 lore raw

PATCH v2 1/2 Tegra: fdt: Add/enhance sdhci (mmc) nodes for all T20 DT files

Thierry,

Quoted 121 lines · Thierry Reding
Quoted 112 lines · Thierry Reding
Quoted 64 lines · Thierry Reding
Quoted 32 lines · Tom Warren

Thierry/Lucas,

Quoted 28 lines · Thierry Reding
Quoted 26 lines · Tom Warren

Lucas,

Quoted 14 lines · Lucas Stach

Hi Tom,

Am Montag, den 11.02.2013, 10:17 -0700 schrieb Tom Warren:

Quoted 11 lines

Linux dts files were used for those boards that didn't already have sdhci info populated. Tamonten has their own dtsi file with common sdhci nodes (sourced from Linux).

---
v2:
- cleanup comments in dts files/match w/kernel files
- add sdhci aliases in all dts files
- use tegra20-tamonten.dtsi from the kernel for AD boards

Quoted 4 lines
Quoted 1 line
 arch/arm/dts/tegra20-tamonten.dtsi               |  489 ++++++++++++++++++++++

I'm not sure if pushing the whole file in this patch is the right thing to do.

I didn't want to edit it since we seem to be moving towards using the Linux DTS files in toto in U-Boot (as per Stephen & Simon). Does it do any harm to have the whole thing here? Saves some work later. Thierry
- what do you think?

Quoted 5 lines
Quoted 2 lines

Given that it isn't used anywhere I don't think we really need it right now. We can always add it later when we can make better use of it.

It actually is used (for SDMMC/sdhci) now, Thierry. That's why it's in this patchset.

Right, I hadn't looked at that patch yet.

Quoted 20 lines

I had originally put the sdhci node for Avionic Design boards in their respective .dts files, but Stephen pointed out that the kernel had a tegra20-tamonten.dtsi file with common info, which included the sdhci node, and asked that I use it, instead, so we echo the kernel layout. So I pulled that file into my MMC DT patchset, and used it in all AD board builds (medcom/tec/plutux) - it's pulled in via an override of CONFIG_ARCH_DEVICE_TREE in the config files.

So the options seem to be:

a) Don't use the tamonton dtsi file, and put the sdhci nodes in the AD dts files, just like all other boards (this was my V1 approach). Vetoed by Stephen.
b) Use tegra20-tamonten.dtsi as is, identical to the kernel file. If necessary, I can move it's inclusion to a separate patch, independent of the MMC DT patchset, as suggested by Lucas.
c) Use tegra20-tamonten.dtsi, but just with the sdhci node (is this what you're suggesting, Thierry?). I'd still pull it in via a CONFIG_ARCH_DEVICE_TREE #define in the AD config files.

Quoted 2 lines

Let me know ASAP - I'd like to get V3 upstreamed soon so I can move on to work on the T30/T114 MMC patches.

Quoted 8 lines

I think option b) sounds fine given that we want to move to the same DTS as the kernel eventually anyway. So for the Tamonten (and AD board) pieces, consider this:

I can't give you a Tested-by because I have a bunch of other things to take care of and I probably won't get to testing this for a few days.

So it turned out that I need to touch U-Boot anyway, so I decided to give this a spin. I noticed that overriding CONFIG_ARCH_DEVICE_TREE from the board configuration file doesn't work currently. What happens is that the autoconf.mk (which is derived from the board configuration) is included before the CPU config.mk which sets CONFIG_ARCH_DEVICE_TREE to tegra20 (or tegra30, tegra114). I came up with the attached patch to set the variable if not set previously (by the board configuration file).

Feel free to squash that in your patch series if you deem it a proper solution. I can also provide a proper separate patch if you prefer.

Thierry

arch/arm/cpu/armv7/tegra114/config.mk +1 -1
index cb1a19d..e7c22c0 100644
@@ -16,4 +16,4 @@
 # You should have received a copy of the GNU General Public License
 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
 #
-CONFIG_ARCH_DEVICE_TREE := tegra114
+CONFIG_ARCH_DEVICE_TREE ?= tegra114
arch/arm/cpu/armv7/tegra20/config.mk +1 -1
index 6432e75..9042664 100644
@@ -23,4 +23,4 @@
 # Foundation, Inc., 59 Temple Place, Suite 330, Boston,
 # MA 02111-1307 USA
 #
-CONFIG_ARCH_DEVICE_TREE := tegra20
+CONFIG_ARCH_DEVICE_TREE ?= tegra20
arch/arm/cpu/armv7/tegra30/config.mk +0 -0
index 719ca81..0035bc5 100644

arch/arm/cpu/armv7/tegra30/config.mk line 19

Quoted 5 lines
arch/arm/cpu/armv7/tegra30/config.mk +1 -1
@@ -16,4 +16,4 @@
 # You should have received a copy of the GNU General Public License
 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
 #
-CONFIG_ARCH_DEVICE_TREE := tegra30
+CONFIG_ARCH_DEVICE_TREE ?= tegra30

I forgot: with that patch applied:

Quoted 3 lines

Thierry

Thanks for the patch - I never would have seen that, since it all built OK w/MAKEALL.

Tom

#11 lore raw

PATCH v2 1/2 Tegra: fdt: Add/enhance sdhci (mmc) nodes for all T20 DT files

In reply to Thierry Reding

On 02/12/2013 03:41 AM, Thierry Reding wrote:
...

Quoted 17 lines

So it turned out that I need to touch U-Boot anyway, so I decided to give this a spin. I noticed that overriding CONFIG_ARCH_DEVICE_TREE from the board configuration file doesn't work currently. What happens is that the autoconf.mk (which is derived from the board configuration) is included before the CPU config.mk which sets CONFIG_ARCH_DEVICE_TREE to tegra20 (or tegra30, tegra114). I came up with the attached patch to set the variable if not set previously (by the board configuration file).

Feel free to squash that in your patch series if you deem it a proper solution. I can also provide a proper separate patch if you prefer.

arch/arm/cpu/armv7/tegra114/config.mk +0 -1
 
-CONFIG_ARCH_DEVICE_TREE := tegra114 +CONFIG_ARCH_DEVICE_TREE ?=
Quoted 1 line

tegra114

That looks very odd. What value is CONFIG_ARCH_DEVICE_TREE before that assignment, and why exactly is it wrong?

#12 lore raw

PATCH v2 1/2 Tegra: fdt: Add/enhance sdhci (mmc) nodes for all T20 DT files

Quoted 15 lines · Stephen Warren

On 02/12/2013 03:41 AM, Thierry Reding wrote:
...

Quoted 12 lines

So it turned out that I need to touch U-Boot anyway, so I decided to give this a spin. I noticed that overriding CONFIG_ARCH_DEVICE_TREE from the board configuration file doesn't work currently. What happens is that the autoconf.mk (which is derived from the board configuration) is included before the CPU config.mk which sets CONFIG_ARCH_DEVICE_TREE to tegra20 (or tegra30, tegra114). I came up with the attached patch to set the variable if not set previously (by the board configuration file).

Feel free to squash that in your patch series if you deem it a proper solution. I can also provide a proper separate patch if you prefer.

Quoted 8 lines
Quoted 5 lines
arch/arm/cpu/armv7/tegra114/config.mk +0 -1
 
-CONFIG_ARCH_DEVICE_TREE := tegra114 +CONFIG_ARCH_DEVICE_TREE ?=

tegra114

That looks very odd. What value is CONFIG_ARCH_DEVICE_TREE before that assignment, and why exactly is it wrong?

Tom's patches add a

	#define CONFIG_ARCH_DEVICE_TREE tegra20-tamonten

to the configuration files of Tamonten-derived boards so that the proper DTSI is picked up. However, that line causes the make variable to be defined in autoconf.mk, which is included before the CPU config.mk, so if the config.mk has

CONFIG_ARCH_DEVICE_TREE := tegra20

it will overwrite the value set by autoconf.mk.

Thierry
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 836 bytes
Desc: not available
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20130212/b9aa3c8f/attachment.pgp>

#13 lore raw

PATCH v2 1/2 Tegra: fdt: Add/enhance sdhci (mmc) nodes for all T20 DT files

In reply to Tom Warren

On 02/11/2013 12:21 PM, Tom Warren wrote:
...

Quoted 8 lines

So the options seem to be:

a) Don't use the tamonton dtsi file, and put the sdhci nodes in the AD dts files, just like all other boards (this was my V1 approach). Vetoed by Stephen.
b) Use tegra20-tamonten.dtsi as is, identical to the kernel file. If necessary, I can move it's inclusion to a separate patch, independent of the MMC DT patchset, as suggested by Lucas.

That option looks fine to me. The initial check-in of the .dts file should definitely be a separate patch.

Does U-boot actually work correctly if you give it the kernel DT? There are quite a few quirks that U-Boot relies on IIRC...

#14 lore raw

PATCH v2 1/2 Tegra: fdt: Add/enhance sdhci (mmc) nodes for all T20 DT files

In reply to Tom Warren
Quoted 5 lines · Tom Warren

Linux dts files were used for those boards that didn't already have sdhci info populated. Tamonten has their own dtsi file with common sdhci nodes (sourced from Linux).

board/nvidia/dts/tegra20-seaboard.dts line 165

Quoted 9 lines
board/nvidia/dts/tegra20-seaboard.dts +3 -1
 
 	sdhci at c8000400 {
+		status = "okay";
 		cd-gpios = <&gpio 69 0>; /* gpio PI5 */
 		wp-gpios = <&gpio 57 0>; /* gpio PH1 */
-		power-gpios = <&gpio 70 0>; /* gpio PI6 */
+		power-gpios = <&gpio 70 3>; /* gpio PI6 */
+		bus-width = <4>;
 	};

The "3" for the power-gpios flags looks odd, and doesn't match the kernel. The binding only defines bit 0 for inverted, and doesn't define bit 1.

author
#15 lore raw

PATCH v2 1/2 Tegra: fdt: Add/enhance sdhci (mmc) nodes for all T20 DT files

Stephen,

Quoted 16 lines · Stephen Warren
Quoted 5 lines · Tom Warren

Linux dts files were used for those boards that didn't already have sdhci info populated. Tamonten has their own dtsi file with common sdhci nodes (sourced from Linux).

board/nvidia/dts/tegra20-seaboard.dts line 165

Quoted 9 lines
board/nvidia/dts/tegra20-seaboard.dts +3 -1
 
      sdhci at c8000400 {
+             status = "okay";
              cd-gpios = <&gpio 69 0>; /* gpio PI5 */
              wp-gpios = <&gpio 57 0>; /* gpio PH1 */
-             power-gpios = <&gpio 70 0>; /* gpio PI6 */
+             power-gpios = <&gpio 70 3>; /* gpio PI6 */
+             bus-width = <4>;
      };
Quoted 3 lines

The "3" for the power-gpios flags looks odd, and doesn't match the kernel. The binding only defines bit 0 for inverted, and doesn't define bit 1.

That came from our internal U-Boot repo - don't know why it's setting it as an output and asserted and no other Seaboard DT file does. I'll look into it.

Thanks

author
#16 lore raw

PATCH v2 2/2 Tegra: MMC: Add DT support to MMC driver for all T20 boards

In reply to Tom Warren

tegra_mmc_init() now parses the DT info for bus width, WP/CD GPIOs, etc. Tested on Seaboard, fully functional.

Tamonten boards (medcom-wide, plutux, and tec) use a different/new dtsi file w/common settings.

---
v2:
- all boards now call tegra_mmc_init once, w/no params
- count MMC controllers, not aliases
- AD boards (medcom/plutux/tec) use common tegra20-tamonten.dtsi

 arch/arm/include/asm/arch-tegra/mmc.h             |    2 +-
 arch/arm/include/asm/arch-tegra/tegra_mmc.h       |   13 +-
 board/avionic-design/common/tamonten.c            |    4 +-
 board/compal/paz00/paz00.c                        |   11 +-
 board/compulab/trimslice/trimslice.c              |    9 +-
 board/nvidia/harmony/harmony.c                    |   11 +-
 board/nvidia/seaboard/seaboard.c                  |   11 +-
 board/nvidia/whistler/whistler.c                  |    7 +-
 board/toradex/colibri_t20_iris/colibri_t20_iris.c |    2 +-
 drivers/mmc/tegra_mmc.c                           |  259 +++++++++++++--------
 include/configs/medcom-wide.h                     |    2 +
 include/configs/plutux.h                          |    2 +
 include/configs/tec.h                             |    2 +
 include/fdtdec.h                                  |    1 +
 lib/fdtdec.c                                      |    1 +
 15 files changed, 197 insertions(+), 140 deletions(-)
arch/arm/include/asm/arch-tegra/mmc.h +1 -1
index 5c95047..71ad407 100644
@@ -22,6 +22,6 @@
 #ifndef _TEGRA_MMC_H_
 #define _TEGRA_MMC_H_
 
-int tegra_mmc_init(int dev_index, int bus_width, int pwr_gpio, int cd_gpio);
+void tegra_mmc_init(void);
 
 #endif /* _TEGRA_MMC_H_ */
arch/arm/include/asm/arch-tegra/tegra_mmc.h +9 -4
index dd746ca..bd18f5f 100644
@@ -27,6 +27,8 @@
 #define TEGRA_SDMMC3_BASE	0xC8000400
 #define TEGRA_SDMMC4_BASE	0xC8000600
 
+#define MAX_HOSTS		4	/* Max number of 'hosts'/controllers */
+
 #ifndef __ASSEMBLY__
 struct tegra_mmc {
 	unsigned int	sysad;		/* _SYSTEM_ADDRESS_0 */
@@ -119,12 +121,15 @@ struct tegra_mmc {
 
 struct mmc_host {
 	struct tegra_mmc *reg;
+	int id;			/* device id/number, 0-3 */
+	int enabled;		/* 1 to enable, 0 to disable */
+	int width;		/* Bus Width, 1, 4 or 8 */
+	enum periph_id mmc_id;	/* Peripheral ID: PERIPH_ID_... */
+	struct fdt_gpio_state cd_gpio;		/* Change Detect GPIO */
+	struct fdt_gpio_state pwr_gpio;		/* Power GPIO */
+	struct fdt_gpio_state wp_gpio;		/* Write Protect GPIO */
 	unsigned int version;	/* SDHCI spec. version */
 	unsigned int clock;	/* Current clock (MHz) */
-	unsigned int base;	/* Base address, SDMMC1/2/3/4 */
-	enum periph_id mmc_id;	/* Peripheral ID: PERIPH_ID_... */
-	int pwr_gpio;		/* Power GPIO */
-	int cd_gpio;		/* Change Detect GPIO */
 };
 
 #endif	/* __ASSEMBLY__ */
board/avionic-design/common/tamonten.c +2 -2
index e6a932e..a7dd598 100644
@@ -69,8 +69,8 @@ int board_mmc_init(bd_t *bd)
 	/* Enable muxes, etc. for SDMMC controllers */
 	pin_mux_mmc();
 
-	/* init dev 0, SD slot, with 4-bit bus */
-	tegra_mmc_init(0, 4, GPIO_PI6, GPIO_PH2);
+	/* init mmc devs */
+	tegra_mmc_init();
 
 	return 0;
 }
board/compal/paz00/paz00.c +3 -8
index 1447f47..5bd72bb 100644
@@ -55,18 +55,13 @@ static void pin_mux_mmc(void)
 /* this is a weak define that we are overriding */
 int board_mmc_init(bd_t *bd)
 {
-	debug("board_mmc_init called\n");
+	debug("%s called\n", __func__);
 
 	/* Enable muxes, etc. for SDMMC controllers */
 	pin_mux_mmc();
 
-	debug("board_mmc_init: init eMMC\n");
-	/* init dev 0, eMMC chip, with 8-bit bus */
-	tegra_mmc_init(0, 8, -1, -1);
-
-	debug("board_mmc_init: init SD slot\n");
-	/* init dev 3, SD slot, with 4-bit bus */
-	tegra_mmc_init(3, 4, GPIO_PV1, GPIO_PV5);
+	debug("%s: init MMC\n", __func__);
+	tegra_mmc_init();
 
 	return 0;
 }
board/compulab/trimslice/trimslice.c +3 -6
index 8f4dd09..020283c 100644
@@ -64,16 +64,13 @@ static void pin_mux_mmc(void)
 /* this is a weak define that we are overriding */
 int board_mmc_init(bd_t *bd)
 {
-	debug("board_mmc_init called\n");
+	debug("%s called\n", __func__);
 
 	/* Enable muxes, etc. for SDMMC controllers */
 	pin_mux_mmc();
 
-	/* init dev 0 (SDMMC4), (micro-SD slot) with 4-bit bus */
-	tegra_mmc_init(0, 4, -1, GPIO_PP1);
-
-	/* init dev 3 (SDMMC1), (SD slot) with 4-bit bus */
-	tegra_mmc_init(3, 4, -1, -1);
+	/* init mmc devs */
+	tegra_mmc_init();
 
 	return 0;
 }
board/nvidia/harmony/harmony.c +3 -8
index 93430ed..fba06c2 100644
@@ -58,18 +58,13 @@ static void pin_mux_mmc(void)
 /* this is a weak define that we are overriding */
 int board_mmc_init(bd_t *bd)
 {
-	debug("board_mmc_init called\n");
+	debug("%s called\n", __func__);
 
 	/* Enable muxes, etc. for SDMMC controllers */
 	pin_mux_mmc();
 
-	debug("board_mmc_init: init SD slot J26\n");
-	/* init dev 0, SD slot J26, with 8-bit bus */
-	tegra_mmc_init(0, 8, GPIO_PI6, GPIO_PH2);
-
-	debug("board_mmc_init: init SD slot J5\n");
-	/* init dev 2, SD slot J5, with 4-bit bus */
-	tegra_mmc_init(2, 4, GPIO_PT3, GPIO_PI5);
+	debug("%s: init MMC devs\n", __func__);
+	tegra_mmc_init();
 
 	return 0;
 }
board/nvidia/seaboard/seaboard.c +3 -8
index 3e33da0..fbc9d2e 100644
@@ -65,18 +65,13 @@ static void pin_mux_mmc(void)
 /* this is a weak define that we are overriding */
 int board_mmc_init(bd_t *bd)
 {
-	debug("board_mmc_init called\n");
+	debug("%s called\n", __func__);
 
 	/* Enable muxes, etc. for SDMMC controllers */
 	pin_mux_mmc();
 
-	debug("board_mmc_init: init eMMC\n");
-	/* init dev 0, eMMC chip, with 8-bit bus */
-	tegra_mmc_init(0, 8, -1, -1);
-
-	debug("board_mmc_init: init SD slot\n");
-	/* init dev 1, SD slot, with 4-bit bus */
-	tegra_mmc_init(1, 4, GPIO_PI6, GPIO_PI5);
+	debug("%s: init MMC\n", __func__);
+	tegra_mmc_init();
 
 	return 0;
 }
board/nvidia/whistler/whistler.c +2 -5
index 592cd6b..79fdbda 100644
@@ -73,11 +73,8 @@ int board_mmc_init(bd_t *bd)
 	/* Enable muxes, etc. for SDMMC controllers */
 	pin_mux_mmc();
 
-	/* init dev 0 (SDMMC4), (J29 "HSMMC") with 8-bit bus */
-	tegra_mmc_init(0, 8, -1, -1);
-
-	/* init dev 1 (SDMMC3), (J40 "SDIO3") with 8-bit bus */
-	tegra_mmc_init(1, 8, -1, -1);
+	/* init MMC devs */
+	tegra_mmc_init();
 
 	return 0;
 }
board/toradex/colibri_t20_iris/colibri_t20_iris.c +1 -1
index e40a986..9cc7a87 100644
@@ -39,7 +39,7 @@ int board_mmc_init(bd_t *bd)
 	funcmux_select(PERIPH_ID_SDMMC4, FUNCMUX_SDMMC4_ATB_GMA_4_BIT);
 	pinmux_tristate_disable(PINGRP_GMB);
 
-	tegra_mmc_init(0, 4, -1, GPIO_PC7);
+	tegra_mmc_init();
 
 	return 0;
 }
drivers/mmc/tegra_mmc.c +162 -97

457 lines not on this page · raw message

include/configs/medcom-wide.h +2 -0
index bae4ba0..c355993 100644
@@ -32,6 +32,8 @@
 #define CONFIG_DEFAULT_DEVICE_TREE	tegra20-medcom-wide
 #define CONFIG_OF_CONTROL
 #define CONFIG_OF_SEPARATE
+#undef CONFIG_ARCH_DEVICE_TREE
+#define CONFIG_ARCH_DEVICE_TREE		tegra20-tamonten
 
 /* High-level configuration options */
 #define V_PROMPT			"Tegra20 (Medcom-Wide) # "
include/configs/plutux.h +2 -0
index deee237..928a2ea 100644
@@ -32,6 +32,8 @@
 #define CONFIG_DEFAULT_DEVICE_TREE	tegra20-plutux
 #define CONFIG_OF_CONTROL
 #define CONFIG_OF_SEPARATE
+#undef CONFIG_ARCH_DEVICE_TREE
+#define CONFIG_ARCH_DEVICE_TREE		tegra20-tamonten
 
 /* High-level configuration options */
 #define V_PROMPT			"Tegra20 (Plutux) # "
include/configs/tec.h +2 -0
index caeb9cd..9a340e3 100644
@@ -32,6 +32,8 @@
 #define CONFIG_DEFAULT_DEVICE_TREE	tegra20-tec
 #define CONFIG_OF_CONTROL
 #define CONFIG_OF_SEPARATE
+#undef CONFIG_ARCH_DEVICE_TREE
+#define CONFIG_ARCH_DEVICE_TREE		tegra20-tamonten
 
 /* High-level configuration options */
 #define V_PROMPT			"Tegra20 (TEC) # "
include/fdtdec.h +1 -0
index c7a92d5..32730d2 100644
@@ -71,6 +71,7 @@ enum fdt_compat_id {
 	COMPAT_NVIDIA_TEGRA20_NAND,	/* Tegra2 NAND controller */
 	COMPAT_NVIDIA_TEGRA20_PWM,	/* Tegra 2 PWM controller */
 	COMPAT_NVIDIA_TEGRA20_DC,	/* Tegra 2 Display controller */
+	COMPAT_NVIDIA_TEGRA20_SDMMC,	/* Tegra SDMMC controller */
 	COMPAT_NVIDIA_TEGRA20_SFLASH,	/* Tegra 2 SPI flash controller */
 	COMPAT_NVIDIA_TEGRA20_SLINK,	/* Tegra 2 SPI SLINK controller */
 	COMPAT_SMSC_LAN9215,		/* SMSC 10/100 Ethernet LAN9215 */
lib/fdtdec.c +1 -0
index 7c7fc83..000130f 100644
@@ -46,6 +46,7 @@ static const char * const compat_names[COMPAT_COUNT] = {
 	COMPAT(NVIDIA_TEGRA20_NAND, "nvidia,tegra20-nand"),
 	COMPAT(NVIDIA_TEGRA20_PWM, "nvidia,tegra20-pwm"),
 	COMPAT(NVIDIA_TEGRA20_DC, "nvidia,tegra20-dc"),
+	COMPAT(NVIDIA_TEGRA20_SDMMC, "nvidia,tegra20-sdhci"),
 	COMPAT(NVIDIA_TEGRA20_SFLASH, "nvidia,tegra20-sflash"),
 	COMPAT(NVIDIA_TEGRA20_SLINK, "nvidia,tegra20-slink"),
 	COMPAT(SMSC_LAN9215, "smsc,lan9215"),
Signature · 1 line
1.7.0.4
#17 lore raw

PATCH v2 2/2 Tegra: MMC: Add DT support to MMC driver for all T20 boards

Am Montag, den 11.02.2013, 10:17 -0700 schrieb Tom Warren:

Quoted 30 lines

tegra_mmc_init() now parses the DT info for bus width, WP/CD GPIOs, etc. Tested on Seaboard, fully functional.

Tamonten boards (medcom-wide, plutux, and tec) use a different/new dtsi file w/common settings.

---
v2:
- all boards now call tegra_mmc_init once, w/no params
- count MMC controllers, not aliases
- AD boards (medcom/plutux/tec) use common tegra20-tamonten.dtsi

 arch/arm/include/asm/arch-tegra/mmc.h             |    2 +-
 arch/arm/include/asm/arch-tegra/tegra_mmc.h       |   13 +-
 board/avionic-design/common/tamonten.c            |    4 +-
 board/compal/paz00/paz00.c                        |   11 +-
 board/compulab/trimslice/trimslice.c              |    9 +-
 board/nvidia/harmony/harmony.c                    |   11 +-
 board/nvidia/seaboard/seaboard.c                  |   11 +-
 board/nvidia/whistler/whistler.c                  |    7 +-
 board/toradex/colibri_t20_iris/colibri_t20_iris.c |    2 +-
 drivers/mmc/tegra_mmc.c                           |  259 +++++++++++++--------
 include/configs/medcom-wide.h                     |    2 +
 include/configs/plutux.h                          |    2 +
 include/configs/tec.h                             |    2 +
 include/fdtdec.h                                  |    1 +
 lib/fdtdec.c                                      |    1 +
 15 files changed, 197 insertions(+), 140 deletions(-)

[...]

Quoted 14 lines
board/nvidia/harmony/harmony.c +1 -1
index 93430ed..fba06c2 100644
@@ -58,18 +58,13 @@ static void pin_mux_mmc(void)
 /* this is a weak define that we are overriding */
 int board_mmc_init(bd_t *bd)
 {
-	debug("board_mmc_init called\n");
+	debug("%s called\n", __func__);
 
 	/* Enable muxes, etc. for SDMMC controllers */
 	pin_mux_mmc();
 

board/nvidia/harmony/harmony.c static void pin_mux_mmc(void) · line 68

Quoted 10 lines
board/nvidia/harmony/harmony.c +2 -7
-	debug("board_mmc_init: init SD slot J26\n");
-	/* init dev 0, SD slot J26, with 8-bit bus */
-	tegra_mmc_init(0, 8, GPIO_PI6, GPIO_PH2);
-
-	debug("board_mmc_init: init SD slot J5\n");
-	/* init dev 2, SD slot J5, with 4-bit bus */
-	tegra_mmc_init(2, 4, GPIO_PT3, GPIO_PI5);
+	debug("%s: init MMC devs\n", __func__);
+	tegra_mmc_init();
 

tegra_mmc_init should not be called from every individual board file, but from the common nvidia tegra board file. Only the pinmux should stay in the individual board code, same thing as was done to all the other functions like NAND and USB.

board/nvidia/harmony/harmony.c static void pin_mux_mmc(void) · line 70

Quoted 2 lines
board/nvidia/harmony/harmony.c +0 -0
 	return 0;
 }

[...]

Regards,
Lucas

author
#18 lore raw

PATCH v2 2/2 Tegra: MMC: Add DT support to MMC driver for all T20 boards

Lucas,

Quoted 46 lines · Lucas Stach

Am Montag, den 11.02.2013, 10:17 -0700 schrieb Tom Warren:

Quoted 30 lines

tegra_mmc_init() now parses the DT info for bus width, WP/CD GPIOs, etc. Tested on Seaboard, fully functional.

Tamonten boards (medcom-wide, plutux, and tec) use a different/new dtsi file w/common settings.

---
v2:
- all boards now call tegra_mmc_init once, w/no params
- count MMC controllers, not aliases
- AD boards (medcom/plutux/tec) use common tegra20-tamonten.dtsi

 arch/arm/include/asm/arch-tegra/mmc.h             |    2 +-
 arch/arm/include/asm/arch-tegra/tegra_mmc.h       |   13 +-
 board/avionic-design/common/tamonten.c            |    4 +-
 board/compal/paz00/paz00.c                        |   11 +-
 board/compulab/trimslice/trimslice.c              |    9 +-
 board/nvidia/harmony/harmony.c                    |   11 +-
 board/nvidia/seaboard/seaboard.c                  |   11 +-
 board/nvidia/whistler/whistler.c                  |    7 +-
 board/toradex/colibri_t20_iris/colibri_t20_iris.c |    2 +-
 drivers/mmc/tegra_mmc.c                           |  259 +++++++++++++--------
 include/configs/medcom-wide.h                     |    2 +
 include/configs/plutux.h                          |    2 +
 include/configs/tec.h                             |    2 +
 include/fdtdec.h                                  |    1 +
 lib/fdtdec.c                                      |    1 +
 15 files changed, 197 insertions(+), 140 deletions(-)

[...]

Quoted 14 lines
board/nvidia/harmony/harmony.c +1 -1
index 93430ed..fba06c2 100644
@@ -58,18 +58,13 @@ static void pin_mux_mmc(void)
 /* this is a weak define that we are overriding */
 int board_mmc_init(bd_t *bd)
 {
-     debug("board_mmc_init called\n");
+     debug("%s called\n", __func__);
 
      /* Enable muxes, etc. for SDMMC controllers */
      pin_mux_mmc();
 
Quoted 14 lines

board/nvidia/harmony/harmony.c static void pin_mux_mmc(void) · line 68

Quoted 10 lines
board/nvidia/harmony/harmony.c +2 -7
-     debug("board_mmc_init: init SD slot J26\n");
-     /* init dev 0, SD slot J26, with 8-bit bus */
-     tegra_mmc_init(0, 8, GPIO_PI6, GPIO_PH2);
-
-     debug("board_mmc_init: init SD slot J5\n");
-     /* init dev 2, SD slot J5, with 4-bit bus */
-     tegra_mmc_init(2, 4, GPIO_PT3, GPIO_PI5);
+     debug("%s: init MMC devs\n", __func__);
+     tegra_mmc_init();
 

tegra_mmc_init should not be called from every individual board file, but from the common nvidia tegra board file. Only the pinmux should stay in the individual board code, same thing as was done to all the other functions like NAND and USB.

True. I was originally just adapting the current config-file driven MMC to DT step-by-step, but you're right - it should be called just once for all boards in the common board file. I'll change it in V3.

Thanks,

Tom

Quoted 7 lines

board/nvidia/harmony/harmony.c static void pin_mux_mmc(void) · line 70

Quoted 2 lines
board/nvidia/harmony/harmony.c +0 -0
      return 0;
 }

[...]

Regards,
Lucas

author
#19 lore raw

PATCH v2 2/2 Tegra: MMC: Add DT support to MMC driver for all T20 boards

Lucas,

Quoted 64 lines · Tom Warren

Lucas,

Quoted 46 lines · Lucas Stach

Am Montag, den 11.02.2013, 10:17 -0700 schrieb Tom Warren:

Quoted 30 lines

tegra_mmc_init() now parses the DT info for bus width, WP/CD GPIOs, etc. Tested on Seaboard, fully functional.

Tamonten boards (medcom-wide, plutux, and tec) use a different/new dtsi file w/common settings.

---
v2:
- all boards now call tegra_mmc_init once, w/no params
- count MMC controllers, not aliases
- AD boards (medcom/plutux/tec) use common tegra20-tamonten.dtsi

 arch/arm/include/asm/arch-tegra/mmc.h             |    2 +-
 arch/arm/include/asm/arch-tegra/tegra_mmc.h       |   13 +-
 board/avionic-design/common/tamonten.c            |    4 +-
 board/compal/paz00/paz00.c                        |   11 +-
 board/compulab/trimslice/trimslice.c              |    9 +-
 board/nvidia/harmony/harmony.c                    |   11 +-
 board/nvidia/seaboard/seaboard.c                  |   11 +-
 board/nvidia/whistler/whistler.c                  |    7 +-
 board/toradex/colibri_t20_iris/colibri_t20_iris.c |    2 +-
 drivers/mmc/tegra_mmc.c                           |  259 +++++++++++++--------
 include/configs/medcom-wide.h                     |    2 +
 include/configs/plutux.h                          |    2 +
 include/configs/tec.h                             |    2 +
 include/fdtdec.h                                  |    1 +
 lib/fdtdec.c                                      |    1 +
 15 files changed, 197 insertions(+), 140 deletions(-)

[...]

Quoted 14 lines
board/nvidia/harmony/harmony.c +1 -1
index 93430ed..fba06c2 100644
@@ -58,18 +58,13 @@ static void pin_mux_mmc(void)
 /* this is a weak define that we are overriding */
 int board_mmc_init(bd_t *bd)
 {
-     debug("board_mmc_init called\n");
+     debug("%s called\n", __func__);
 
      /* Enable muxes, etc. for SDMMC controllers */
      pin_mux_mmc();
 
Quoted 14 lines

board/nvidia/harmony/harmony.c static void pin_mux_mmc(void) · line 68

Quoted 10 lines
board/nvidia/harmony/harmony.c +2 -7
-     debug("board_mmc_init: init SD slot J26\n");
-     /* init dev 0, SD slot J26, with 8-bit bus */
-     tegra_mmc_init(0, 8, GPIO_PI6, GPIO_PH2);
-
-     debug("board_mmc_init: init SD slot J5\n");
-     /* init dev 2, SD slot J5, with 4-bit bus */
-     tegra_mmc_init(2, 4, GPIO_PT3, GPIO_PI5);
+     debug("%s: init MMC devs\n", __func__);
+     tegra_mmc_init();
 

tegra_mmc_init should not be called from every individual board file, but from the common nvidia tegra board file. Only the pinmux should stay in the individual board code, same thing as was done to all the other functions like NAND and USB.

Quoted 3 lines

True. I was originally just adapting the current config-file driven MMC to DT step-by-step, but you're right - it should be called just once for all boards in the common board file. I'll change it in V3.

I've looked into this some more, and it appears that I can't just add a call to tegra_mmc_init() from board_init() in boards/nvidia/common/board.c.

board_init() is where the other periphs do their pin_mux and xxx_init() calls (USB, SPI, etc.). board_init() is called early in board_init_r(), before mmc_initialize() is called. mmc_initialize() is needed before tegra_mmc_init() can use the mmc_device struct, etc. So tegra_mmc_init() needs to be called after mmc_initialize(), and right now that's in each board's board_mmc_init(). In board_mmc_init(), each board sets up any power rails needed for SD-card or eMMC access, sets up it's pin muxes for MMC, and then calls tegra_mmc_init() to parse the DT file and populate the mmc structs.

I could move the pin_mux_mmc() function calls from each board file into nvidia/common/board.c's board_init(), but it wouldn't really change much. So I'll leave it as it is for now, with pin_mux_mmc() and tegra_mmc_init() being called from each board's 'board' file (seaboard.c, colibri_t20_iris.c, etc).

Let me know if you see another way to move Tegra MMC init to a common board file that doesn't break the MMC driver flow.

Thanks,

Tom

Quoted 11 lines

Thanks,

Tom

Quoted 7 lines

board/nvidia/harmony/harmony.c static void pin_mux_mmc(void) · line 70

Quoted 2 lines
board/nvidia/harmony/harmony.c +0 -0
      return 0;
 }

[...]

Regards,
Lucas

#20 lore raw

PATCH v2 2/2 Tegra: MMC: Add DT support to MMC driver for all T20 boards

Hello Tom,

Am Dienstag, den 12.02.2013, 12:24 -0700 schrieb Tom Warren:

Quoted 26 lines

Lucas,

Quoted 9 lines
Quoted 5 lines
Quoted 1 line

tegra_mmc_init should not be called from every individual board file, but from the common nvidia tegra board file. Only the pinmux should stay in the individual board code, same thing as was done to all the other functions like NAND and USB.

True. I was originally just adapting the current config-file driven MMC to DT step-by-step, but you're right - it should be called just once for all boards in the common board file. I'll change it in V3.

I've looked into this some more, and it appears that I can't just add a call to tegra_mmc_init() from board_init() in boards/nvidia/common/board.c.

board_init() is where the other periphs do their pin_mux and xxx_init() calls (USB, SPI, etc.). board_init() is called early in board_init_r(), before mmc_initialize() is called. mmc_initialize() is needed before tegra_mmc_init() can use the mmc_device struct, etc. So tegra_mmc_init() needs to be called after mmc_initialize(), and right now that's in each board's board_mmc_init(). In board_mmc_init(), each board sets up any power rails needed for SD-card or eMMC access, sets up it's pin muxes for MMC, and then calls tegra_mmc_init() to parse the DT file and populate the mmc structs.

Quoted 9 lines

I could move the pin_mux_mmc() function calls from each board file into nvidia/common/board.c's board_init(), but it wouldn't really change much. So I'll leave it as it is for now, with pin_mux_mmc() and tegra_mmc_init() being called from each board's 'board' file (seaboard.c, colibri_t20_iris.c, etc).

Let me know if you see another way to move Tegra MMC init to a common board file that doesn't break the MMC driver flow.

I didn't look up the flow myself, as I don't have time for that right now, but I think I've got a pretty good picture from your description.

I think we should really try to make the Tegra MMC init flow as similar as possible to the other peripherals, so I suggest doing the following:
1. Provide a pin_mux_mmc() (possibly with a weak define as done with other peripherals). Boards should do pinmux and rail enabling within this function.
2. Move board_mmc_init() into nvidia/common/board.c, this function should call into the board specific pinmux function and then call tegra_mmc_init().

So even while we don't get the complete same flow as for other peripherals as the board_mmc_init() still has to be a freestanding function, instead of those things being folded into board_init(), we at least gain a clear distinction between the board specific parts and Tegra common code.

Regards,
Lucas

author
#21 lore raw

PATCH v2 2/2 Tegra: MMC: Add DT support to MMC driver for all T20 boards

Lucas,

Quoted 35 lines · Lucas Stach

Hello Tom,

Am Dienstag, den 12.02.2013, 12:24 -0700 schrieb Tom Warren:

Quoted 32 lines

Lucas,

Quoted 9 lines
Quoted 5 lines
Quoted 1 line

tegra_mmc_init should not be called from every individual board file, but from the common nvidia tegra board file. Only the pinmux should stay in the individual board code, same thing as was done to all the other functions like NAND and USB.

True. I was originally just adapting the current config-file driven MMC to DT step-by-step, but you're right - it should be called just once for all boards in the common board file. I'll change it in V3.

I've looked into this some more, and it appears that I can't just add a call to tegra_mmc_init() from board_init() in boards/nvidia/common/board.c.

board_init() is where the other periphs do their pin_mux and xxx_init() calls (USB, SPI, etc.). board_init() is called early in board_init_r(), before mmc_initialize() is called. mmc_initialize() is needed before tegra_mmc_init() can use the mmc_device struct, etc. So tegra_mmc_init() needs to be called after mmc_initialize(), and right now that's in each board's board_mmc_init(). In board_mmc_init(), each board sets up any power rails needed for SD-card or eMMC access, sets up it's pin muxes for MMC, and then calls tegra_mmc_init() to parse the DT file and populate the mmc structs.

I could move the pin_mux_mmc() function calls from each board file into nvidia/common/board.c's board_init(), but it wouldn't really change much. So I'll leave it as it is for now, with pin_mux_mmc() and tegra_mmc_init() being called from each board's 'board' file (seaboard.c, colibri_t20_iris.c, etc).

Quoted 6 lines
Quoted 3 lines

Let me know if you see another way to move Tegra MMC init to a common board file that doesn't break the MMC driver flow.

I didn't look up the flow myself, as I don't have time for that right now, but I think I've got a pretty good picture from your description.

Thanks for the quick response.

Quoted 6 lines

I think we should really try to make the Tegra MMC init flow as similar as possible to the other peripherals, so I suggest doing the following:
1. Provide a pin_mux_mmc() (possibly with a weak define as done with other peripherals). Boards should do pinmux and rail enabling within this function.

(Almost) every Tegra board already has a pin_mux_mmc(). Few boards do any power-rail enabling, but I can move those that do into pin_mux_mmc().

Quoted 3 lines

2. Move board_mmc_init() into nvidia/common/board.c, this function should call into the board specific pinmux function and then call tegra_mmc_init().

Current board_mmc_init() does call pin_mux_mmc() and then tegra_mmc_init(). Moving it from each board file into common/board.c is a good idea. I'll try that.

Thanks

Quoted 10 lines

So even while we don't get the complete same flow as for other peripherals as the board_mmc_init() still has to be a freestanding function, instead of those things being folded into board_init(), we at least gain a clear distinction between the board specific parts and Tegra common code.

Regards,
Lucas

#22 lore raw

PATCH v2 2/2 Tegra: MMC: Add DT support to MMC driver for all T20 boards

In reply to Tom Warren
Quoted 9 lines · Tom Warren

tegra_mmc_init() now parses the DT info for bus width, WP/CD GPIOs, etc. Tested on Seaboard, fully functional.

Tamonten boards (medcom-wide, plutux, and tec) use a different/new dtsi file w/common settings.

drivers/mmc/tegra_mmc.c +1 -0
 
+static int process_nodes(const void *blob, int node_list[], int count)

...

Quoted 2 lines
drivers/mmc/tegra_mmc.c +2 -0
+	/* build mmc_host[] for each controller */
+	for (i = 0; i < count; i++) {

...

drivers/mmc/tegra_mmc.c int tegra_mmc_init(int dev_index, int bus_width, int pwr_gpio, int cd_gpio) · line 635

Quoted 2 lines
drivers/mmc/tegra_mmc.c +2 -0
+		/* Mark position as used */
+		node_list[i] = -1;

Is that needed? Does anything use that array after this function?

include/configs/tec.h

Quoted 3 lines
include/configs/medcom-wide.h +0 -0
include/configs/plutux.h +0 -0
include/configs/tec.h +0 -0

In all 3 of those files ...

Quoted 1 line

#define CONFIG_DEFAULT_DEVICE_TREE tegra20-medcom-wide

Why not change that define ...

include/configs/tec.h

Quoted 4 lines

#define CONFIG_OF_CONTROL
#define CONFIG_OF_SEPARATE

include/configs/tec.h +2 -0
+#undef CONFIG_ARCH_DEVICE_TREE
+#define CONFIG_ARCH_DEVICE_TREE		tegra20-tamonten

rather than adding that one? All the other Tegra boards only set CONFIG_DEFAULT_DEVICE_TREE.

Aside from the few comments I and others have made, this series looks good.

author
#23 lore raw

PATCH v2 2/2 Tegra: MMC: Add DT support to MMC driver for all T20 boards

Stephen,

Quoted 10 lines · Stephen Warren

drivers/mmc/tegra_mmc.c int tegra_mmc_init(int dev_index, int bus_width, int pwr_gpio, int cd_gpio) · line 612

Quoted 9 lines · Tom Warren

tegra_mmc_init() now parses the DT info for bus width, WP/CD GPIOs, etc. Tested on Seaboard, fully functional.

Tamonten boards (medcom-wide, plutux, and tec) use a different/new dtsi file w/common settings.

drivers/mmc/tegra_mmc.c +1 -0
 
+static int process_nodes(const void *blob, int node_list[], int count)
Quoted 8 lines

...

Quoted 2 lines
drivers/mmc/tegra_mmc.c +2 -0
+     /* build mmc_host[] for each controller */
+     for (i = 0; i < count; i++) {

...

drivers/mmc/tegra_mmc.c int tegra_mmc_init(int dev_index, int bus_width, int pwr_gpio, int cd_gpio) · line 635

Quoted 2 lines
drivers/mmc/tegra_mmc.c +2 -0
+             /* Mark position as used */
+             node_list[i] = -1;

Is that needed? Does anything use that array after this function?

No idea. I took this code from the tegra_i2c.c driver.

Quoted 11 lines

include/configs/tec.h

Quoted 4 lines
include/configs/medcom-wide.h +0 -0
include/configs/plutux.h +0 -0
include/configs/tec.h +0 -0

In all 3 of those files ...

Quoted 1 line

#define CONFIG_DEFAULT_DEVICE_TREE tegra20-medcom-wide

Why not change that define ...

Quoted 7 lines

include/configs/tec.h

Quoted 4 lines

#define CONFIG_OF_CONTROL
#define CONFIG_OF_SEPARATE

include/configs/tec.h +2 -0
+#undef CONFIG_ARCH_DEVICE_TREE
+#define CONFIG_ARCH_DEVICE_TREE              tegra20-tamonten

rather than adding that one? All the other Tegra boards only set CONFIG_DEFAULT_DEVICE_TREE.

CONFIG_DEFAULT_DEVICE_TREE is the .dts file (board/nvidia/dts). CONFIG_ARCH_DEVICE_TREE is the .dtsi file (arch/arm/dts). See Thierry's explanation, also.

Quoted 2 lines

Aside from the few comments I and others have made, this series looks good.

Thanks. Working on V3.

#24 lore raw

PATCH v2 2/2 Tegra: MMC: Add DT support to MMC driver for all T20 boards

Quoted 28 lines · Tom Warren

Stephen,

Quoted 17 lines · Stephen Warren

include/configs/tec.h

Quoted 9 lines · Tom Warren

tegra_mmc_init() now parses the DT info for bus width, WP/CD GPIOs, etc. Tested on Seaboard, fully functional.

Tamonten boards (medcom-wide, plutux, and tec) use a different/new dtsi file w/common settings.

include/configs/medcom-wide.h +0 -0
include/configs/plutux.h +0 -0
include/configs/tec.h +0 -0

In all 3 of those files ...

Quoted 1 line

#define CONFIG_DEFAULT_DEVICE_TREE tegra20-medcom-wide

Why not change that define ...

Quoted 7 lines

include/configs/tec.h

Quoted 4 lines

#define CONFIG_OF_CONTROL
#define CONFIG_OF_SEPARATE

include/configs/tec.h +2 -0
+#undef CONFIG_ARCH_DEVICE_TREE
+#define CONFIG_ARCH_DEVICE_TREE              tegra20-tamonten

rather than adding that one? All the other Tegra boards only set CONFIG_DEFAULT_DEVICE_TREE.

Quoted 3 lines

CONFIG_DEFAULT_DEVICE_TREE is the .dts file (board/nvidia/dts). CONFIG_ARCH_DEVICE_TREE is the .dtsi file (arch/arm/dts). See Thierry's explanation, also.

So why set CONFIG_ARCH_DEVICE_TREE to tegra20-tamonten here; if that variable is supposed to point at the SoC .dtsi file, that value is wrong; it *should* be tegra20.dtsi.

Oh yuck. I see what's going on now. e.g. tegra20-medcom-wide.dts is including ARCH_CPU_DTS. It should be including tegra20-tamonten.dtsi, which then includes ARCH_CPU_DTS. The ARCH_CPU_DTS variable is supposed to only ever point at the SoC .dtsi file, not any "intermediate" .dtsi file... You probably need to put tegra20-tamonten.dtsi into board/avionic-design/dts rather than arch/arm/dts to make that work.

author
#25 lore raw

PATCH v2 2/2 Tegra: MMC: Add DT support to MMC driver for all T20 boards

Stephen,

Quoted 37 lines · Stephen Warren
Quoted 28 lines · Tom Warren

Stephen,

Quoted 17 lines · Stephen Warren

include/configs/tec.h

Quoted 9 lines · Tom Warren

tegra_mmc_init() now parses the DT info for bus width, WP/CD GPIOs, etc. Tested on Seaboard, fully functional.

Tamonten boards (medcom-wide, plutux, and tec) use a different/new dtsi file w/common settings.

include/configs/medcom-wide.h +0 -0
include/configs/plutux.h +0 -0
include/configs/tec.h +0 -0

In all 3 of those files ...

Quoted 1 line

#define CONFIG_DEFAULT_DEVICE_TREE tegra20-medcom-wide

Why not change that define ...

Quoted 7 lines

include/configs/tec.h

Quoted 4 lines

#define CONFIG_OF_CONTROL
#define CONFIG_OF_SEPARATE

include/configs/tec.h +2 -0
+#undef CONFIG_ARCH_DEVICE_TREE
+#define CONFIG_ARCH_DEVICE_TREE              tegra20-tamonten

rather than adding that one? All the other Tegra boards only set CONFIG_DEFAULT_DEVICE_TREE.

Quoted 3 lines

CONFIG_DEFAULT_DEVICE_TREE is the .dts file (board/nvidia/dts). CONFIG_ARCH_DEVICE_TREE is the .dtsi file (arch/arm/dts). See Thierry's explanation, also.

So why set CONFIG_ARCH_DEVICE_TREE to tegra20-tamonten here; if that variable is supposed to point at the SoC .dtsi file, that value is wrong; it *should* be tegra20.dtsi.

Quoted 6 lines

Oh yuck. I see what's going on now. e.g. tegra20-medcom-wide.dts is including ARCH_CPU_DTS. It should be including tegra20-tamonten.dtsi, which then includes ARCH_CPU_DTS. The ARCH_CPU_DTS variable is supposed to only ever point at the SoC .dtsi file, not any "intermediate" .dtsi file... You probably need to put tegra20-tamonten.dtsi into board/avionic-design/dts rather than arch/arm/dts to make that work.

OK, I've removed the overrides for CONFIG_ARCH_DEVICE_TREE in the AD config files, and added 'tegra20-tamonten.dtsi' to the include line in the AD .dts files. I have to add an include path to the DTC command line in dts/Makefile so it can find tegra20-tamonten.dtsi in arch/$(ARCH)/dts, but it all builds fine. I'll have to rely on Thierry for testing. This will all be in v3, going up today or tomorrow.

Thanks,

Tom

#26 lore raw

PATCH v2 2/2 Tegra: MMC: Add DT support to MMC driver for all T20 boards

Quoted 47 lines · Tom Warren

Stephen,

Quoted 37 lines · Stephen Warren
Quoted 28 lines · Tom Warren

Stephen,

Quoted 17 lines · Stephen Warren

include/configs/tec.h

Quoted 9 lines · Tom Warren

tegra_mmc_init() now parses the DT info for bus width, WP/CD GPIOs, etc. Tested on Seaboard, fully functional.

Tamonten boards (medcom-wide, plutux, and tec) use a different/new dtsi file w/common settings.

include/configs/medcom-wide.h +0 -0
include/configs/plutux.h +0 -0
include/configs/tec.h +0 -0

In all 3 of those files ...

Quoted 1 line

#define CONFIG_DEFAULT_DEVICE_TREE tegra20-medcom-wide

Why not change that define ...

Quoted 7 lines

include/configs/tec.h

Quoted 4 lines

#define CONFIG_OF_CONTROL
#define CONFIG_OF_SEPARATE

include/configs/tec.h +2 -0
+#undef CONFIG_ARCH_DEVICE_TREE
+#define CONFIG_ARCH_DEVICE_TREE              tegra20-tamonten

rather than adding that one? All the other Tegra boards only set CONFIG_DEFAULT_DEVICE_TREE.

Quoted 3 lines

CONFIG_DEFAULT_DEVICE_TREE is the .dts file (board/nvidia/dts). CONFIG_ARCH_DEVICE_TREE is the .dtsi file (arch/arm/dts). See Thierry's explanation, also.

So why set CONFIG_ARCH_DEVICE_TREE to tegra20-tamonten here; if that variable is supposed to point at the SoC .dtsi file, that value is wrong; it *should* be tegra20.dtsi.

Quoted 6 lines

Oh yuck. I see what's going on now. e.g. tegra20-medcom-wide.dts is including ARCH_CPU_DTS. It should be including tegra20-tamonten.dtsi, which then includes ARCH_CPU_DTS. The ARCH_CPU_DTS variable is supposed to only ever point at the SoC .dtsi file, not any "intermediate" .dtsi file... You probably need to put tegra20-tamonten.dtsi into board/avionic-design/dts rather than arch/arm/dts to make that work.

Quoted 3 lines

OK, I've removed the overrides for CONFIG_ARCH_DEVICE_TREE in the AD config files, and added 'tegra20-tamonten.dtsi' to the include line in the AD .dts files.

Great.

Quoted 3 lines

I have to add an include path to the DTC command line in dts/Makefile so it can find tegra20-tamonten.dtsi in arch/$(ARCH)/dts, but it all builds fine.

If you move tegra20-tamonten.dtsi to boards/avionic-design/dts, do you still need an include path? Since that file is board-/vendor-specific, it really shouldn't be in arch/arm/dts if that directory typically only contains SoC .dtsi files.

author
#27 lore raw

PATCH v2 2/2 Tegra: MMC: Add DT support to MMC driver for all T20 boards

Stephen,

Quoted 54 lines · Stephen Warren
Quoted 47 lines · Tom Warren

Stephen,

Quoted 37 lines · Stephen Warren
Quoted 28 lines · Tom Warren

Stephen,

Quoted 17 lines · Stephen Warren

include/configs/tec.h

Quoted 9 lines · Tom Warren

tegra_mmc_init() now parses the DT info for bus width, WP/CD GPIOs, etc. Tested on Seaboard, fully functional.

Tamonten boards (medcom-wide, plutux, and tec) use a different/new dtsi file w/common settings.

include/configs/medcom-wide.h +0 -0
include/configs/plutux.h +0 -0
include/configs/tec.h +0 -0

In all 3 of those files ...

Quoted 1 line

#define CONFIG_DEFAULT_DEVICE_TREE tegra20-medcom-wide

Why not change that define ...

Quoted 7 lines

include/configs/tec.h

Quoted 4 lines

#define CONFIG_OF_CONTROL
#define CONFIG_OF_SEPARATE

include/configs/tec.h +2 -0
+#undef CONFIG_ARCH_DEVICE_TREE
+#define CONFIG_ARCH_DEVICE_TREE              tegra20-tamonten

rather than adding that one? All the other Tegra boards only set CONFIG_DEFAULT_DEVICE_TREE.

Quoted 3 lines

CONFIG_DEFAULT_DEVICE_TREE is the .dts file (board/nvidia/dts). CONFIG_ARCH_DEVICE_TREE is the .dtsi file (arch/arm/dts). See Thierry's explanation, also.

So why set CONFIG_ARCH_DEVICE_TREE to tegra20-tamonten here; if that variable is supposed to point at the SoC .dtsi file, that value is wrong; it *should* be tegra20.dtsi.

Quoted 6 lines

Oh yuck. I see what's going on now. e.g. tegra20-medcom-wide.dts is including ARCH_CPU_DTS. It should be including tegra20-tamonten.dtsi, which then includes ARCH_CPU_DTS. The ARCH_CPU_DTS variable is supposed to only ever point at the SoC .dtsi file, not any "intermediate" .dtsi file... You probably need to put tegra20-tamonten.dtsi into board/avionic-design/dts rather than arch/arm/dts to make that work.

Quoted 3 lines

OK, I've removed the overrides for CONFIG_ARCH_DEVICE_TREE in the AD config files, and added 'tegra20-tamonten.dtsi' to the include line in the AD .dts files.

Great.

Quoted 8 lines
Quoted 3 lines

I have to add an include path to the DTC command line in dts/Makefile so it can find tegra20-tamonten.dtsi in arch/$(ARCH)/dts, but it all builds fine.

If you move tegra20-tamonten.dtsi to boards/avionic-design/dts, do you still need an include path? Since that file is board-/vendor-specific, it really shouldn't be in arch/arm/dts if that directory typically only contains SoC .dtsi files.

I tried both places, and had to add the include path to get it to build.

I'll try it from board/avionic-design/dts with a new include path.

Tom

#28 lore raw

PATCH v2 2/2 Tegra: MMC: Add DT support to MMC driver for all T20 boards

Quoted 61 lines · Tom Warren

Stephen,

Quoted 58 lines · Stephen Warren
Quoted 47 lines · Tom Warren

Stephen,

Quoted 37 lines · Stephen Warren
Quoted 28 lines · Tom Warren

Stephen,

Quoted 17 lines · Stephen Warren

include/configs/tec.h

Quoted 9 lines · Tom Warren

tegra_mmc_init() now parses the DT info for bus width, WP/CD GPIOs, etc. Tested on Seaboard, fully functional.

Tamonten boards (medcom-wide, plutux, and tec) use a different/new dtsi file w/common settings.

include/configs/medcom-wide.h +0 -0
include/configs/plutux.h +0 -0
include/configs/tec.h +0 -0

In all 3 of those files ...

Quoted 1 line

#define CONFIG_DEFAULT_DEVICE_TREE tegra20-medcom-wide

Why not change that define ...

Quoted 7 lines

include/configs/tec.h

Quoted 4 lines

#define CONFIG_OF_CONTROL
#define CONFIG_OF_SEPARATE

include/configs/tec.h +2 -0
+#undef CONFIG_ARCH_DEVICE_TREE
+#define CONFIG_ARCH_DEVICE_TREE              tegra20-tamonten

rather than adding that one? All the other Tegra boards only set CONFIG_DEFAULT_DEVICE_TREE.

Quoted 3 lines

CONFIG_DEFAULT_DEVICE_TREE is the .dts file (board/nvidia/dts). CONFIG_ARCH_DEVICE_TREE is the .dtsi file (arch/arm/dts). See Thierry's explanation, also.

So why set CONFIG_ARCH_DEVICE_TREE to tegra20-tamonten here; if that variable is supposed to point at the SoC .dtsi file, that value is wrong; it *should* be tegra20.dtsi.

Quoted 6 lines

Oh yuck. I see what's going on now. e.g. tegra20-medcom-wide.dts is including ARCH_CPU_DTS. It should be including tegra20-tamonten.dtsi, which then includes ARCH_CPU_DTS. The ARCH_CPU_DTS variable is supposed to only ever point at the SoC .dtsi file, not any "intermediate" .dtsi file... You probably need to put tegra20-tamonten.dtsi into board/avionic-design/dts rather than arch/arm/dts to make that work.

Quoted 3 lines

OK, I've removed the overrides for CONFIG_ARCH_DEVICE_TREE in the AD config files, and added 'tegra20-tamonten.dtsi' to the include line in the AD .dts files.

Great.

Quoted 3 lines

I have to add an include path to the DTC command line in dts/Makefile so it can find tegra20-tamonten.dtsi in arch/$(ARCH)/dts, but it all builds fine.

Quoted 8 lines
Quoted 4 lines

If you move tegra20-tamonten.dtsi to boards/avionic-design/dts, do you still need an include path? Since that file is board-/vendor-specific, it really shouldn't be in arch/arm/dts if that directory typically only contains SoC .dtsi files.

I tried both places, and had to add the include path to get it to build.

I'll try it from board/avionic-design/dts with a new include path.

Hmmm. Odd.

Since IIRC U-Boot is already running the .dts files through the C pre-processor, you could try:

#include "tegra20-tamonten.dtsi"

rather than:

/include/ "tegra20-tamonten.dtsi";

The kernel is likely moving to this scheme, so U-Boot might want to follow this, even pre-emptively, in order to re-use DT files sometime anyway.