ARM: stm32: Fix Avenger96
Subsystems (3)
v3 cover ARM: stm32: Fix Avenger96
This series fixes the Avenger96 board. Since there are way too many patches floating around on the ML and because I found various other details that needed fixing, I decided to bundle the whole set of fixes into this series.
This should go into current release, otherwise the board doesn't even boot. This V2 is also roughly synchronized with the Linux V2 series fixing very much the same issues.
Marek Vasut (14):
ARM: dts: stm32: Repair SD1 pre-reloc pinmux DT node on AV96
ARM: dts: stm32: Add alternate pinmux for SDMMC1 direction pins
ARM: dts: stm32: Repair SDMMC1 operation on AV96
ARM: dts: stm32: Add alternate pinmux for SDMMC2 pins 4-7
ARM: dts: stm32: Repair SDMMC2 operation
ARM: dts: stm32: Add QSPI NOR on AV96
ARM: dts: stm32: Use DT alias for the configuration EEPROM
ARM: dts: stm32: Add configuration EEPROM on AV96
ARM: dts: stm32: Add alternate pinmux for ethernet RGMII
ARM: dts: stm32: Repair ethernet operation on AV96
ARM: dts: stm32: Add missing ethernet PHY reset on AV96
ARM: dts: stm32: Repair PMIC configuration on AV96
ARM: dts: stm32: Adjust PLL4 settings on AV96
ARM: dts: stm32: Split AV96 into DHCOR SoM and AV96 board
arch/arm/dts/Makefile | 3 +-
arch/arm/dts/stm32mp157-pinctrl.dtsi | 96 +++++
arch/arm/dts/stm32mp157a-avenger96.dts | 365 +-----------------
arch/arm/dts/stm32mp15xx-dhcom.dtsi | 6 +-
.../stm32mp15xx-dhcor-avenger96-u-boot.dtsi | 80 ++++
arch/arm/dts/stm32mp15xx-dhcor-avenger96.dts | 211 ++++++++++
...oot.dtsi => stm32mp15xx-dhcor-u-boot.dtsi} | 80 +---
arch/arm/dts/stm32mp15xx-dhcor.dtsi | 231 +++++++++++
board/dhelectronics/dh_stm32mp1/board.c | 15 +-
doc/board/st/stm32mp1.rst | 8 +-
10 files changed, 645 insertions(+), 450 deletions(-)
create mode 100644 arch/arm/dts/stm32mp15xx-dhcor-avenger96-u-boot.dtsi
create mode 100644 arch/arm/dts/stm32mp15xx-dhcor-avenger96.dts
rename arch/arm/dts/{stm32mp157a-avenger96-u-boot.dtsi => stm32mp15xx-dhcor-u-boot.dtsi} (73%)
create mode 100644 arch/arm/dts/stm32mp15xx-dhcor.dtsiSignature · 1 line
2.25.1
PATCH v3 1/14 ARM: dts: stm32: Repair SD1 pre-reloc pinmux DT node on AV96
The sdmmc1_dir_pins_a: sdmmc1-dir-0 layout changed in commit 35a54d41d9d4 ("ARM: dts: stm32mp1: sync device tree with v5.2-rc4") such that pins{}; became pins1{};pins2{};, however the SPL extras were not updated to reflect that change. Fix this.
This fixes booting from SD1 X9 slot on the AV96 board.
---
V2: Adjust patch subject
V3: No change
--- arch/arm/dts/stm32mp157a-avenger96-u-boot.dtsi | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)
arch/arm/dts/stm32mp157a-avenger96-u-boot.dtsi +4 -1
@@ -152,7 +152,10 @@ &sdmmc1_dir_pins_a { u-boot,dm-spl; - pins { + pins1 { + u-boot,dm-spl; + }; + pins2 { u-boot,dm-spl; }; };
Signature · 1 line
2.25.1
PATCH v3 1/14 ARM: dts: stm32: Repair SD1 pre-reloc pinmux DT node on AV96
Hi,
Quoted 17 lines
From: Marek Vasut <[email protected]>
Sent: mardi 31 mars 2020 19:51The
sdmmc1_dir_pins_a: sdmmc1-dir-0 layout changed in commit35a54d41d9d4("ARM: dts: stm32mp1: sync device tree with v5.2-rc4") such that pins{}; became pins1{};pins2{};, however the SPL extras were not updated to reflect that change. Fix this.This fixes booting from SD1 X9 slot on the AV96 board.
Reviewed-by Patrice Chotard <[email protected]>Reviewed-by Patrick Delaunay <[email protected]>Fixes 35a54d41d9d4 ("ARM: dts: stm32mp1: sync device tree with v5.2-rc4")Signed-off-by Marek Vasut <[email protected]>Cc Manivannan Sadhasivam <[email protected]>Cc Patrick Delaunay <[email protected]>Cc Patrice Chotard <[email protected]>
Quoted 4 lines
---
V2: Adjust patch subject
V3: No change
---
Applied to u-boot-stm/master, thanks!
Regards
Patrick
PATCH v3 2/14 ARM: dts: stm32: Add alternate pinmux for SDMMC1 direction pins
In reply to Marek VasutAdd another mux option for SDMMC1 direction pins, in particular SDMMC1_D123DIR, this is used on AV96 board.
---
V2: New patch
V3: No change
--- arch/arm/dts/stm32mp157-pinctrl.dtsi | 24 ++++++++++++++++++++++++ 1 file changed, 24 insertions(+)
arch/arm/dts/stm32mp157-pinctrl.dtsi +24 -0
@@ -832,6 +832,30 @@ }; }; + sdmmc1_dir_pins_b: sdmmc1-dir-1 { + pins1 { + pinmux = <STM32_PINMUX('F', 2, AF11)>, /* SDMMC1_D0DIR */ + <STM32_PINMUX('E', 14, AF8)>, /* SDMMC1_D123DIR */ + <STM32_PINMUX('B', 9, AF11)>; /* SDMMC1_CDIR */ + slew-rate = <1>; + drive-push-pull; + bias-pull-up; + }; + pins2{ + pinmux = <STM32_PINMUX('E', 4, AF8)>; /* SDMMC1_CKIN */ + bias-pull-up; + }; + }; + + sdmmc1_dir_sleep_pins_b: sdmmc1-dir-sleep-1 { + pins { + pinmux = <STM32_PINMUX('F', 2, ANALOG)>, /* SDMMC1_D0DIR */ + <STM32_PINMUX('E', 14, ANALOG)>, /* SDMMC1_D123DIR */ + <STM32_PINMUX('B', 9, ANALOG)>, /* SDMMC1_CDIR */ + <STM32_PINMUX('E', 4, ANALOG)>; /* SDMMC1_CKIN */ + }; + }; + sdmmc2_b4_pins_a: sdmmc2-b4-0 { pins1 { pinmux = <STM32_PINMUX('B', 14, AF9)>, /* SDMMC2_D0 */
Signature · 1 line
2.25.1
PATCH v3 2/14 ARM: dts: stm32: Add alternate pinmux for SDMMC1 direction pins
Hi,
Quoted 12 lines
From: Marek Vasut <[email protected]>
Sent: mardi 31 mars 2020 19:51Add another mux option for SDMMC1 direction pins, in particular
SDMMC1_D123DIR, this is used on AV96 board.Reviewed-by Patrice Chotard <[email protected]>Reviewed-by Patrick Delaunay <[email protected]>Signed-off-by Marek Vasut <[email protected]>Cc Manivannan Sadhasivam <[email protected]>Cc Patrick Delaunay <[email protected]>Cc Patrice Chotard <[email protected]>
Quoted 4 lines
---
V2: New patch
V3: No change
---
Applied to u-boot-stm/master, thanks!
Regards
Patrick
PATCH v3 3/14 ARM: dts: stm32: Repair SDMMC1 operation on AV96
In reply to Marek VasutThe SD uses different pinmux for the D123DIRline, use such a pinmux, otherwise there is a pinmux collision on the AV96. Add missing SD voltage regulator switch and enable SDR104 operation.
---
V2: New patch
V3: No change
--- .../arm/dts/stm32mp157a-avenger96-u-boot.dtsi | 2 +- arch/arm/dts/stm32mp157a-avenger96.dts | 25 ++++++++++++++++--- 2 files changed, 22 insertions(+), 5 deletions(-)
arch/arm/dts/stm32mp157a-avenger96-u-boot.dtsi +1 -1
@@ -150,7 +150,7 @@ }; }; -&sdmmc1_dir_pins_a { +&sdmmc1_dir_pins_b { u-boot,dm-spl; pins1 { u-boot,dm-spl;
arch/arm/dts/stm32mp157a-avenger96.dts +21 -4
@@ -76,6 +76,20 @@ default-state = "off"; }; }; + + sd_switch: regulator-sd_switch { + compatible = "regulator-gpio"; + regulator-name = "sd_switch"; + regulator-min-microvolt = <1800000>; + regulator-max-microvolt = <2900000>; + regulator-type = "voltage"; + regulator-always-on; + + gpios = <&gpioi 5 GPIO_ACTIVE_HIGH>; + gpios-states = <0>; + states = <1800000 0x1>, + <2900000 0x0>; + }; }; ðernet0 {
@@ -296,15 +310,18 @@ &sdmmc1 { pinctrl-names = "default", "opendrain", "sleep"; - pinctrl-0 = <&sdmmc1_b4_pins_a &sdmmc1_dir_pins_a>; - pinctrl-1 = <&sdmmc1_b4_od_pins_a>; - pinctrl-2 = <&sdmmc1_b4_sleep_pins_a>; - broken-cd; + pinctrl-0 = <&sdmmc1_b4_pins_a &sdmmc1_dir_pins_b>; + pinctrl-1 = <&sdmmc1_b4_od_pins_a &sdmmc1_dir_pins_b>; + pinctrl-2 = <&sdmmc1_b4_sleep_pins_a &sdmmc1_dir_sleep_pins_b>; + cd-gpios = <&gpioi 8 GPIO_ACTIVE_LOW>; + disable-wp; st,sig-dir; st,neg-edge; st,use-ckin; + sd-uhs-sdr104; bus-width = <4>; vmmc-supply = <&vdd_sd>; + vqmmc-supply = <&sd_switch>; status = "okay"; };
Signature · 1 line
2.25.1
PATCH v3 3/14 ARM: dts: stm32: Repair SDMMC1 operation on AV96
Hi Marek,
Quoted 13 lines
From: Marek Vasut <[email protected]>
Sent: mardi 31 mars 2020 19:51The SD uses different pinmux for the D123DIRline, use such a pinmux, otherwise there is a pinmux collision on the AV96. Add missing SD voltage regulator switch and enable SDR104 operation.
Reviewed-by Patrice Chotard <[email protected]>Reviewed-by Patrick Delaunay <[email protected]>Signed-off-by Marek Vasut <[email protected]>Cc Manivannan Sadhasivam <[email protected]>Cc Patrick Delaunay <[email protected]>Cc Patrice Chotard <[email protected]>
Quoted 4 lines
---
V2: New patch
V3: No change
---
Applied to u-boot-stm/master, thanks!
Regards
Patrick
PATCH v3 4/14 ARM: dts: stm32: Add alternate pinmux for SDMMC2 pins 4-7
In reply to Marek VasutAdd another mux option for SDMMC2 pins 4..7, this is used on AV96 board.
---
V2: Use correct pin AFs
V3: No change
--- arch/arm/dts/stm32mp157-pinctrl.dtsi | 21 +++++++++++++++++++++ 1 file changed, 21 insertions(+)
arch/arm/dts/stm32mp157-pinctrl.dtsi +21 -0
@@ -931,6 +931,27 @@ }; }; + sdmmc2_d47_pins_b: sdmmc2-d47-1 { + pins { + pinmux = <STM32_PINMUX('A', 8, AF9)>, /* SDMMC2_D4 */ + <STM32_PINMUX('A', 15, AF9)>, /* SDMMC2_D5 */ + <STM32_PINMUX('C', 6, AF10)>, /* SDMMC2_D6 */ + <STM32_PINMUX('C', 7, AF10)>; /* SDMMC2_D7 */ + slew-rate = <1>; + drive-push-pull; + bias-pull-up; + }; + }; + + sdmmc2_d47_sleep_pins_b: sdmmc2-d47-sleep-1 { + pins { + pinmux = <STM32_PINMUX('A', 8, ANALOG)>, /* SDMMC2_D4 */ + <STM32_PINMUX('A', 15, ANALOG)>, /* SDMMC2_D5 */ + <STM32_PINMUX('C', 6, ANALOG)>, /* SDMMC2_D6 */ + <STM32_PINMUX('C', 7, ANALOG)>; /* SDMMC2_D7 */ + }; + }; + spdifrx_pins_a: spdifrx-0 { pins { pinmux = <STM32_PINMUX('G', 12, AF8)>; /* SPDIF_IN1 */
Signature · 1 line
2.25.1
PATCH v3 4/14 ARM: dts: stm32: Add alternate pinmux for SDMMC2 pins 4-7
Hi,
Quoted 11 lines
From: Marek Vasut <[email protected]>
Sent: mardi 31 mars 2020 19:51Add another mux option for SDMMC2 pins 4..7, this is used on AV96 board.
Reviewed-by Patrice Chotard <[email protected]>Reviewed-by Patrick Delaunay <[email protected]>Signed-off-by Marek Vasut <[email protected]>Cc Manivannan Sadhasivam <[email protected]>Cc Patrick Delaunay <[email protected]>Cc Patrice Chotard <[email protected]>
Quoted 4 lines
---
V2: Use correct pin AFs
V3: No change
---
Applied to u-boot-stm/master, thanks!
Regards
Patrick
PATCH v3 5/14 ARM: dts: stm32: Repair SDMMC2 operation
In reply to Marek VasutThe eMMC uses different pinmux for the top four data lines, use such a pinmux, otherwise it takes a very long time until the test for 8bit operation times out. And this is the correct pinmux per schematic too.
---
V2: Update also the -u-boot.dtsi to match this change
V3: No change
--- arch/arm/dts/stm32mp157a-avenger96-u-boot.dtsi | 2 +- arch/arm/dts/stm32mp157a-avenger96.dts | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-)
arch/arm/dts/stm32mp157a-avenger96-u-boot.dtsi +1 -1
@@ -174,7 +174,7 @@ }; }; -&sdmmc2_d47_pins_a { +&sdmmc2_d47_pins_b { u-boot,dm-spl; pins { u-boot,dm-spl;
arch/arm/dts/stm32mp157a-avenger96.dts +1 -1
@@ -327,7 +327,7 @@ &sdmmc2 { pinctrl-names = "default"; - pinctrl-0 = <&sdmmc2_b4_pins_a &sdmmc2_d47_pins_a>; + pinctrl-0 = <&sdmmc2_b4_pins_a &sdmmc2_d47_pins_b>; non-removable; no-sd; no-sdio;
Signature · 1 line
2.25.1
PATCH v3 5/14 ARM: dts: stm32: Repair SDMMC2 operation
Hi Marek,
Quoted 13 lines
From: Marek Vasut <[email protected]>
Sent: mardi 31 mars 2020 19:51The eMMC uses different pinmux for the top four data lines, use such a pinmux, otherwise it takes a very long time until the test for 8bit operation times out. And this is the correct pinmux per schematic too.
Reviewed-by Patrice Chotard <[email protected]>Reviewed-by Patrick Delaunay <[email protected]>Signed-off-by Marek Vasut <[email protected]>Cc Manivannan Sadhasivam <[email protected]>Cc Patrick Delaunay <[email protected]>Cc Patrice Chotard <[email protected]>
Quoted 4 lines
---
V2: Update also the-u-boot.dtsito match this change
V3: No change
---
Applied to u-boot-stm/master, thanks!
Regards
Patrick
PATCH v3 6/14 ARM: dts: stm32: Add QSPI NOR on AV96
In reply to Marek VasutThe DH Electronics DHCOR SOM has QSPI NOR on the SoM itself, add it into the DT.
---
V2: Drop the explicit flash type in DT node, use spi-flash
V3: Reduce the SPI NOR window to 2 MiB
--- arch/arm/dts/stm32mp157a-avenger96.dts | 20 ++++++++++++++++++++ 1 file changed, 20 insertions(+)
arch/arm/dts/stm32mp157a-avenger96.dts +20 -0
@@ -20,6 +20,7 @@ mmc0 = &sdmmc1; serial0 = &uart4; serial1 = &uart7; + spi0 = &qspi; }; chosen {
@@ -300,6 +301,25 @@ vdd_3v3_usbfs-supply = <&vdd_usb>; }; +&qspi { + pinctrl-names = "default", "sleep"; + pinctrl-0 = <&qspi_clk_pins_a &qspi_bk1_pins_a>; + pinctrl-1 = <&qspi_clk_sleep_pins_a &qspi_bk1_sleep_pins_a>; + reg = <0x58003000 0x1000>, <0x70000000 0x200000>; + #address-cells = <1>; + #size-cells = <0>; + status = "okay"; + + flash0: spi-flash at 0 { + compatible = "jedec,spi-nor"; + reg = <0>; + spi-rx-bus-width = <4>; + spi-max-frequency = <108000000>; + #address-cells = <1>; + #size-cells = <1>; + }; +}; + &rng1 { status = "okay"; };
Signature · 1 line
2.25.1
PATCH v3 6/14 ARM: dts: stm32: Add QSPI NOR on AV96
Hi Marek,
Quoted 6 lines
From: Marek Vasut <[email protected]>
Sent: mardi 31 mars 2020 19:51The DH Electronics DHCOR SOM has QSPI NOR on the SoM itself, add it into the DT.
Quoted 8 lines
Signed-off-by Marek Vasut <[email protected]>Cc Manivannan Sadhasivam <[email protected]>Cc Patrick Delaunay <[email protected]>Cc Patrice Chotard <[email protected]>---
V2: Drop the explicit flash type in DT node, use spi-flash
V3: Reduce the SPI NOR window to 2 MiB
---
Thanks
Patrick
PATCH v3 6/14 ARM: dts: stm32: Add QSPI NOR on AV96
In reply to Marek VasutHi,
Quoted 13 lines
-----Original Message-----
From: Marek Vasut <[email protected]>
Sent: mardi 31 mars 2020 19:51
To: u-boot at lists.denx.de
Cc: Marek Vasut <[email protected]>; Manivannan Sadhasivam <[email protected]>; Patrick DELAUNAY <[email protected]>; Patrice CHOTARD <[email protected]>
Subject: [PATCH V3 06/14] ARM: dts: stm32: Add QSPI NOR on AV96
Importance: HighThe DH Electronics DHCOR SOM has QSPI NOR on the SoM itself, add it into the DT.
Quoted 8 lines
Signed-off-by Marek Vasut <[email protected]>Cc Manivannan Sadhasivam <[email protected]>Cc Patrick Delaunay <[email protected]>Cc Patrice Chotard <[email protected]>---
V2: Drop the explicit flash type in DT node, use spi-flash
V3: Reduce the SPI NOR window to 2 MiB
---
Applied to u-boot-stm/master, thanks!
Regards
Patrick
PATCH v3 7/14 ARM: dts: stm32: Use DT alias for the configuration EEPROM
In reply to Marek VasutUse DT /aliases node to establish a stable phandle to the configuration EEPROM. This permits the configuration EEPROM to be moved e.g. to a different address or a different bus. Adjust the board code to handle new phandle lookup.
---
V2: No change
V3: No change
--- arch/arm/dts/stm32mp15xx-dhcom.dtsi | 6 +++++- board/dhelectronics/dh_stm32mp1/board.c | 15 +++++++-------- 2 files changed, 12 insertions(+), 9 deletions(-)
arch/arm/dts/stm32mp15xx-dhcom.dtsi +5 -1
@@ -10,6 +10,10 @@ #include <dt-bindings/mfd/st,stpmic1.h> / { + aliases { + eeprom0 = &eeprom0; + }; + memory at c0000000 { device_type = "memory"; reg = <0xC0000000 0x40000000>;
@@ -187,7 +191,7 @@ }; }; - eeprom at 50 { + eeprom0: eeprom at 50 { compatible = "atmel,24c02"; reg = <0x50>; pagesize = <16>;
board/dhelectronics/dh_stm32mp1/board.c +7 -8
@@ -78,22 +78,21 @@ DECLARE_GLOBAL_DATA_PTR;int setup_mac_address (void ) { -struct udevice *dev; -ofnode eeprom;unsigned char enetaddr[6 ]; -int ret; +struct udevice *dev; +int off, ret; ret =eth_env_get_enetaddr ("ethaddr" , enetaddr);if (ret)/* ethaddr is already set */ return 0 ; - eeprom =ofnode_path ("/soc/i2c at 5c002000/eeprom at 50" ); -if (!ofnode_valid (eeprom)) { -printf ("Invalid hardware path to EEPROM!\n" ); -return -ENODEV ; + off =fdt_path_offset (gd->fdt_blob,"eeprom0" ); +if (off <0 ) { +printf ("%s: No eeprom0 path offset\n" , __func__); +return off; } - ret =uclass_get_device_by_ofnode (UCLASS_I2C_EEPROM , eeprom, &dev); + ret =uclass_get_device_by_of_offset (UCLASS_I2C_EEPROM , off, &dev);if (ret) {printf ("Cannot find EEPROM!\n" );return ret;
Signature · 1 line
2.25.1
PATCH v3 7/14 ARM: dts: stm32: Use DT alias for the configuration EEPROM
Hi,
Quoted 7 lines
From: Marek Vasut <[email protected]>
Sent: mardi 31 mars 2020 19:51Use DT /aliases node to establish a stable phandle to the configuration EEPROM. This permits the configuration EEPROM to be moved e.g. to a different address or a different bus. Adjust the board code to handle new phandle lookup.
Quoted 7 lines
Reviewed-by Patrice Chotard <[email protected]>Reviewed-by Patrick Delaunay <[email protected]>Signed-off-by Marek Vasut <[email protected]>Cc Manivannan Sadhasivam <[email protected]>Cc Patrick Delaunay <[email protected]>Cc Patrice Chotard <[email protected]>---
Applied to u-boot-stm/master, thanks!
Regards
Patrick
PATCH v3 8/14 ARM: dts: stm32: Add configuration EEPROM on AV96
In reply to Marek VasutThe board has an EEPROM on the same I2C bus as PMIC, at address 0x53. The EEPROM contains the board MAC address.
---
V2: No change
V3: No change
--- arch/arm/dts/stm32mp157a-avenger96.dts | 6 ++++++ 1 file changed, 6 insertions(+)
arch/arm/dts/stm32mp157a-avenger96.dts +6 -0
@@ -289,6 +289,12 @@ status = "disabled"; }; }; + + eeprom at 53 { + compatible = "atmel,24c02"; + reg = <0x53>; + pagesize = <16>; + }; }; &iwdg2 {
Signature · 1 line
2.25.1
PATCH v3 8/14 ARM: dts: stm32: Add configuration EEPROM on AV96
Hi,
Quoted 6 lines
From: Marek Vasut <[email protected]>
Sent: mardi 31 mars 2020 19:52The board has an EEPROM on the same I2C bus as PMIC, at address
0x53. The EEPROM contains the board MAC address.
Quoted 7 lines
Reviewed-by Patrice Chotard <[email protected]>Reviewed-by Patrick Delaunay <[email protected]>Signed-off-by Marek Vasut <[email protected]>Cc Manivannan Sadhasivam <[email protected]>Cc Patrick Delaunay <[email protected]>Cc Patrice Chotard <[email protected]>---
Applied to u-boot-stm/master, with one minor modification
I also add the alias eeprom0 = &eeprom0 (with eeprom0: eeprom at 53) to be compatible with previous patch "ARM: dts: stm32: Use DT alias for the configuration EEPROM"
Regards
Patrick
PATCH v3 8/14 ARM: dts: stm32: Add configuration EEPROM on AV96
Quoted 1 line · Patrick DELAUNAY
Hi,
Hi,
Quoted 14 lines
Quoted 6 lines
From: Marek Vasut <[email protected]>
Sent: mardi 31 mars 2020 19:52The board has an EEPROM on the same I2C bus as PMIC, at address
0x53. The EEPROM contains the board MAC address.Quoted 7 lines
Reviewed-by Patrice Chotard <[email protected]>Reviewed-by Patrick Delaunay <[email protected]>Signed-off-by Marek Vasut <[email protected]>Cc Manivannan Sadhasivam <[email protected]>Cc Patrick Delaunay <[email protected]>Cc Patrice Chotard <[email protected]>---
Quoted 4 lines
Applied to u-boot-stm/master, with one minor modification
I also add the alias eeprom0 = &eeprom0 (with eeprom0: eeprom at 53) to be compatible with previous patch "ARM: dts: stm32: Use DT alias for the configuration EEPROM"
Thanks, this was in 14/14, but got omitted here.
PATCH v3 9/14 ARM: dts: stm32: Add alternate pinmux for ethernet RGMII
In reply to Marek VasutAdd another mux option for DWMAC RGMII, this is used on AV96 board.
---
V2: No change
V3: No change
--- arch/arm/dts/stm32mp157-pinctrl.dtsi | 51 ++++++++++++++++++++++++++++ 1 file changed, 51 insertions(+)
arch/arm/dts/stm32mp157-pinctrl.dtsi +51 -0
@@ -288,6 +288,57 @@ }; }; + ethernet0_rgmii_pins_b: rgmii-1 { + pins1 { + pinmux = <STM32_PINMUX('G', 5, AF11)>, /* ETH_RGMII_CLK125 */ + <STM32_PINMUX('G', 4, AF11)>, /* ETH_RGMII_GTX_CLK */ + <STM32_PINMUX('B', 12, AF11)>, /* ETH_RGMII_TXD0 */ + <STM32_PINMUX('G', 14, AF11)>, /* ETH_RGMII_TXD1 */ + <STM32_PINMUX('C', 2, AF11)>, /* ETH_RGMII_TXD2 */ + <STM32_PINMUX('E', 2, AF11)>, /* ETH_RGMII_TXD3 */ + <STM32_PINMUX('G', 11, AF11)>, /* ETH_RGMII_TX_CTL */ + <STM32_PINMUX('C', 1, AF11)>; /* ETH_MDC */ + bias-disable; + drive-push-pull; + slew-rate = <2>; + }; + pins2 { + pinmux = <STM32_PINMUX('A', 2, AF11)>; /* ETH_MDIO */ + bias-disable; + drive-push-pull; + slew-rate = <0>; + }; + pins3 { + pinmux = <STM32_PINMUX('C', 4, AF11)>, /* ETH_RGMII_RXD0 */ + <STM32_PINMUX('C', 5, AF11)>, /* ETH_RGMII_RXD1 */ + <STM32_PINMUX('H', 6, AF11)>, /* ETH_RGMII_RXD2 */ + <STM32_PINMUX('B', 1, AF11)>, /* ETH_RGMII_RXD3 */ + <STM32_PINMUX('A', 1, AF11)>, /* ETH_RGMII_RX_CLK */ + <STM32_PINMUX('A', 7, AF11)>; /* ETH_RGMII_RX_CTL */ + bias-disable; + }; + }; + + ethernet0_rgmii_pins_sleep_b: rgmii-sleep-1 { + pins1 { + pinmux = <STM32_PINMUX('G', 5, ANALOG)>, /* ETH_RGMII_CLK125 */ + <STM32_PINMUX('G', 4, ANALOG)>, /* ETH_RGMII_GTX_CLK */ + <STM32_PINMUX('B', 12, ANALOG)>, /* ETH_RGMII_TXD0 */ + <STM32_PINMUX('G', 14, ANALOG)>, /* ETH_RGMII_TXD1 */ + <STM32_PINMUX('C', 2, ANALOG)>, /* ETH_RGMII_TXD2 */ + <STM32_PINMUX('E', 2, ANALOG)>, /* ETH_RGMII_TXD3 */ + <STM32_PINMUX('G', 11, ANALOG)>, /* ETH_RGMII_TX_CTL */ + <STM32_PINMUX('A', 2, ANALOG)>, /* ETH_MDIO */ + <STM32_PINMUX('C', 1, ANALOG)>, /* ETH_MDC */ + <STM32_PINMUX('C', 4, ANALOG)>, /* ETH_RGMII_RXD0 */ + <STM32_PINMUX('C', 5, ANALOG)>, /* ETH_RGMII_RXD1 */ + <STM32_PINMUX('H', 6, ANALOG)>, /* ETH_RGMII_RXD2 */ + <STM32_PINMUX('B', 1, ANALOG)>, /* ETH_RGMII_RXD3 */ + <STM32_PINMUX('A', 1, ANALOG)>, /* ETH_RGMII_RX_CLK */ + <STM32_PINMUX('A', 7, ANALOG)>; /* ETH_RGMII_RX_CTL */ + }; + }; + fmc_pins_a: fmc-0 { pins1 { pinmux = <STM32_PINMUX('D', 4, AF12)>, /* FMC_NOE */
Signature · 1 line
2.25.1
PATCH v3 9/14 ARM: dts: stm32: Add alternate pinmux for ethernet RGMII
Hi,
Quoted 5 lines
From: Marek Vasut <[email protected]>
Sent: mardi 31 mars 2020 19:52Add another mux option for DWMAC RGMII, this is used on AV96 board.
Quoted 7 lines
Reviewed-by Patrice Chotard <[email protected]>Reviewed-by Patrick Delaunay <[email protected]>Signed-off-by Marek Vasut <[email protected]>Cc Manivannan Sadhasivam <[email protected]>Cc Patrick Delaunay <[email protected]>Cc Patrice Chotard <[email protected]>---
Applied to u-boot-stm/master, thanks!
Regards
Patrick
PATCH v3 10/14 ARM: dts: stm32: Repair ethernet operation on AV96
In reply to Marek VasutThe AV96 RGMII uses different pinmux for ETH_RGMII_TXD0, ETH_RGMII_RXD2 and ETH_RGMII_TX_CTL. Use the correct pinmux to make ethernet operational.
---
V2: No change
V3: No change
--- arch/arm/dts/stm32mp157a-avenger96.dts | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
arch/arm/dts/stm32mp157a-avenger96.dts +2 -2
@@ -95,8 +95,8 @@ ðernet0 { status = "okay"; - pinctrl-0 = <ðernet0_rgmii_pins_a>; - pinctrl-1 = <ðernet0_rgmii_pins_sleep_a>; + pinctrl-0 = <ðernet0_rgmii_pins_b>; + pinctrl-1 = <ðernet0_rgmii_pins_sleep_b>; pinctrl-names = "default", "sleep"; phy-mode = "rgmii"; max-speed = <1000>;
Signature · 1 line
2.25.1
PATCH v3 10/14 ARM: dts: stm32: Repair ethernet operation on AV96
Hi,
Quoted 7 lines
From: Marek Vasut <[email protected]>
Sent: mardi 31 mars 2020 19:52The AV96 RGMII uses different pinmux for
ETH_RGMII_TXD0,ETH_RGMII_RXD2andETH_RGMII_TX_CTL. Use the correct pinmux to make ethernet operational.
Quoted 7 lines
Reviewed-by Patrice Chotard <[email protected]>Reviewed-by Patrick Delaunay <[email protected]>Signed-off-by Marek Vasut <[email protected]>Cc Manivannan Sadhasivam <[email protected]>Cc Patrick Delaunay <[email protected]>Cc Patrice Chotard <[email protected]>---
Applied to u-boot-stm/master, thanks!
Regards
Patrick
PATCH v3 11/14 ARM: dts: stm32: Add missing ethernet PHY reset on AV96
In reply to Marek VasutAdd PHY reset GPIO on AV96 ethernet PHY.
---
V2: No change
V3: No change
--- arch/arm/dts/stm32mp157a-avenger96.dts | 1 + 1 file changed, 1 insertion(+)
arch/arm/dts/stm32mp157a-avenger96.dts +1 -0
@@ -101,6 +101,7 @@ phy-mode = "rgmii"; max-speed = <1000>; phy-handle = <&phy0>; + phy-reset-gpios = <&gpioz 2 GPIO_ACTIVE_LOW>; mdio0 { #address-cells = <1>;
Signature · 1 line
2.25.1
PATCH v3 11/14 ARM: dts: stm32: Add missing ethernet PHY reset on AV96
Hi,
Quoted 10 lines
From: Marek Vasut <[email protected]>
Sent: mardi 31 mars 2020 19:52Add PHY reset GPIO on AV96 ethernet PHY.
Signed-off-by Marek Vasut <[email protected]>Cc Manivannan Sadhasivam <[email protected]>Cc Patrick Delaunay <[email protected]>Cc Patrice Chotard <[email protected]>---
Applied to u-boot-stm/master, thanks!
Regards
Patrick
PATCH v3 12/14 ARM: dts: stm32: Repair PMIC configuration on AV96
In reply to Marek VasutThe core and vdd PMIC buck regulators were misconfigured, which caused instability of the board and malfunction of high-speed interfaces, like the RGMII. Configure the PMIC correctly to repair these problems. Also, model the missing Enpirion EP53A8LQI on the DHCOR SoM as a fixed regulator.
---
V2: - Model the Enpirion EP53A8LQI on the DHCOR SoM as a fixed regulator
- Adjust the PMIC voltages further
V3: No change
--- arch/arm/dts/stm32mp157a-avenger96.dts | 20 ++++++++++++++++---- 1 file changed, 16 insertions(+), 4 deletions(-)
arch/arm/dts/stm32mp157a-avenger96.dts +16 -4
@@ -91,6 +91,17 @@ states = <1800000 0x1>, <2900000 0x0>; }; + + /* Enpirion EP3A8LQI U2 on the DHCOR */ + vdd_io: regulator-buck-io { + compatible = "regulator-fixed"; + regulator-name = "buck-io"; + regulator-min-microvolt = <1800000>; + regulator-max-microvolt = <1800000>; + regulator-always-on; + regulator-boot-on; + vin-supply = <&vdd>; + }; }; ðernet0 {
@@ -167,7 +178,7 @@ vddcore: buck1 { regulator-name = "vddcore"; - regulator-min-microvolt = <1200000>; + regulator-min-microvolt = <800000>; regulator-max-microvolt = <1350000>; regulator-always-on; regulator-initial-mode = <0>;
@@ -185,8 +196,8 @@ vdd: buck3 { regulator-name = "vdd"; - regulator-min-microvolt = <3300000>; - regulator-max-microvolt = <3300000>; + regulator-min-microvolt = <2900000>; + regulator-max-microvolt = <2900000>; regulator-always-on; st,mask_reset; regulator-initial-mode = <0>;
@@ -268,6 +279,7 @@ regulator-name = "vbus_otg"; interrupts = <IT_OCP_OTG 0>; interrupt-parent = <&pmic>; + regulator-active-discharge = <1>; }; vbus_sw: pwr_sw2 {
@@ -304,7 +316,7 @@ }; &pwr_regulators { - vdd-supply = <&vdd>; + vdd-supply = <&vdd_io>; vdd_3v3_usbfs-supply = <&vdd_usb>; };
Signature · 1 line
2.25.1
PATCH v3 12/14 ARM: dts: stm32: Repair PMIC configuration on AV96
Hi Marek and Pascal,
Quoted 45 lines
From: Marek Vasut <[email protected]>
Sent: mardi 31 mars 2020 19:52The core and vdd PMIC buck regulators were misconfigured, which caused instability of the board and malfunction of high-speed interfaces, like the RGMII. Configure the PMIC correctly to repair these problems. Also, model the missing Enpirion EP53A8LQI on the DHCOR SoM as a fixed regulator.
Reviewed-by Patrice Chotard <[email protected]>Signed-off-by Marek Vasut <[email protected]>Cc Manivannan Sadhasivam <[email protected]>Cc Patrick Delaunay <[email protected]>Cc Patrice Chotard <[email protected]>---
V2: - Model the Enpirion EP53A8LQI on the DHCOR SoM as a fixed regulator
- Adjust the PMIC voltages further
V3: No change--- arch/arm/dts/stm32mp157a-avenger96.dts | 20 ++++++++++++++++---- 1 file changed, 16 insertions(+), 4 deletions(-)arch/arm/dts/stm32mp157a-avenger96.dts +11
-0@@ -91,6 +91,17 @@ states = <1800000 0x1>, <2900000 0x0>; }; + + /* Enpirion EP3A8LQI U2 on the DHCOR */ + vdd_io: regulator-buck-io { + compatible = "regulator-fixed"; + regulator-name = "buck-io"; + regulator-min-microvolt = <1800000>; + regulator-max-microvolt = <1800000>; + regulator-always-on; + regulator-boot-on; + vin-supply = <&vdd>; + }; }; ðernet0 {
arch/arm/dts/stm32mp157a-avenger96.dts line 181
Quoted 5 lines
arch/arm/dts/stm32mp157a-avenger96.dts +1
-1@@ -167,7 +178,7 @@ vddcore: buck1 { regulator-name = "vddcore"; - regulator-min-microvolt = <1200000>; + regulator-min-microvolt = <800000>;
I don't understood this modification, If I correctly remember, on STM32MP15x On ST boards, we set this value to 1200000 to protect VDDCORE on SOC (it is not allowed to be reduce lower than 1,2V.
0.8V is the minimal supported voltage by STPMIC but not the minal expected VDDCORE.
Quoted 15 lines
arch/arm/dts/stm32mp157a-avenger96.dts +2
-2regulator-max-microvolt = <1350000>; regulator-always-on; regulator-initial-mode = <0>;@@ -185,8 +196,8 @@ vdd: buck3 { regulator-name = "vdd"; - regulator-min-microvolt = <3300000>; - regulator-max-microvolt = <3300000>; + regulator-min-microvolt = <2900000>; + regulator-max-microvolt = <2900000>; regulator-always-on; st,mask_reset; regulator-initial-mode = <0>;
arch/arm/dts/stm32mp157a-avenger96.dts line 282
Quoted 4 lines
arch/arm/dts/stm32mp157a-avenger96.dts +1
-0@@ -268,6 +279,7 @@ regulator-name = "vbus_otg"; interrupts = <IT_OCP_OTG 0>; interrupt-parent = <&pmic>; + regulator-active-discharge = <1>;
It seems this setting introduced other-consumption, it should not be used.
Quoted 13 lines
arch/arm/dts/stm32mp157a-avenger96.dts +1
-2}; vbus_sw: pwr_sw2 {@@ -304,7 +316,7 @@ }; &pwr_regulators { - vdd-supply = <&vdd>; + vdd-supply = <&vdd_io>; vdd_3v3_usbfs-supply = <&vdd_usb>; }; --
Quoted 1 line
2.25.1
I let Pascal complete my first feedback or react on the Linux upstream.
Patrick
PATCH v3 12/14 ARM: dts: stm32: Repair PMIC configuration on AV96
Quoted 1 line · Patrick DELAUNAY
Hi Marek and Pascal,
Hi,
[...]
Quoted 11 lines
arch/arm/dts/stm32mp157a-avenger96.dts line 181
Quoted 6 lines
arch/arm/dts/stm32mp157a-avenger96.dts +1
-1@@ -167,7 +178,7 @@ vddcore: buck1 { regulator-name = "vddcore"; - regulator-min-microvolt = <1200000>; + regulator-min-microvolt = <800000>;I don't understood this modification, If I correctly remember, on STM32MP15x On ST boards, we set this value to 1200000 to protect VDDCORE on SOC (it is not allowed to be reduce lower than 1,2V.
Quoted 2 lines
0.8V is the minimal supported voltage by STPMIC but not the minal expected VDDCORE.
Then we should likely drop this part.
That said, is the core voltage supposed to be 1V2 even in sleep states ?
Quoted 18 lines
Quoted 12 lines
arch/arm/dts/stm32mp157a-avenger96.dts +2
-2@@ -185,8 +196,8 @@ vdd: buck3 { regulator-name = "vdd"; - regulator-min-microvolt = <3300000>; - regulator-max-microvolt = <3300000>; + regulator-min-microvolt = <2900000>; + regulator-max-microvolt = <2900000>; regulator-always-on; st,mask_reset; regulator-initial-mode = <0>;arch/arm/dts/stm32mp157a-avenger96.dts line 282
Quoted 4 lines
arch/arm/dts/stm32mp157a-avenger96.dts +1
-0@@ -268,6 +279,7 @@ regulator-name = "vbus_otg"; interrupts = <IT_OCP_OTG 0>; interrupt-parent = <&pmic>; + regulator-active-discharge = <1>;It seems this setting introduced other-consumption, it should not be used.
Other-consumption ? What does that mean ?
PATCH v3 12/14 ARM: dts: stm32: Repair PMIC configuration on AV96
Quoted 39 lines · Marek Vasut
Quoted 1 line · Patrick DELAUNAY
Hi Marek and Pascal,
Hi,
[...]
Quoted 10 lines
arch/arm/dts/stm32mp157a-avenger96.dts line 181
Quoted 6 lines
arch/arm/dts/stm32mp157a-avenger96.dts +1
-1@@ -167,7 +178,7 @@ vddcore: buck1 { regulator-name = "vddcore"; - regulator-min-microvolt = <1200000>; + regulator-min-microvolt = <800000>;I don't understood this modification, If I correctly remember, on STM32MP15x On ST boards, we set this value to 1200000 to protect VDDCORE on SOC (it is not allowed to be reduce lower than 1,2V.
Quoted 2 lines
0.8V is the minimal supported voltage by STPMIC but not the minal expected VDDCORE.
Then we should likely drop this part.
That said, is the core voltage supposed to be 1V2 even in sleep states ?
Quoted 17 lines
Quoted 12 lines
arch/arm/dts/stm32mp157a-avenger96.dts +2
-2@@ -185,8 +196,8 @@ vdd: buck3 { regulator-name = "vdd"; - regulator-min-microvolt = <3300000>; - regulator-max-microvolt = <3300000>; + regulator-min-microvolt = <2900000>; + regulator-max-microvolt = <2900000>; regulator-always-on; st,mask_reset; regulator-initial-mode = <0>;arch/arm/dts/stm32mp157a-avenger96.dts line 282
Quoted 4 lines
arch/arm/dts/stm32mp157a-avenger96.dts +1
-0@@ -268,6 +279,7 @@ regulator-name = "vbus_otg"; interrupts = <IT_OCP_OTG 0>; interrupt-parent = <&pmic>; + regulator-active-discharge = <1>;It seems this setting introduced other-consumption, it should not be used.
Quoted 1 line
Other-consumption ? What does that mean ?
It's a typo, Patrick means over-consumption.
Patrice
PATCH v3 12/14 ARM: dts: stm32: Repair PMIC configuration on AV96
Quoted 41 lines · Patrice CHOTARD
Quoted 39 lines · Marek Vasut
Quoted 1 line · Patrick DELAUNAY
Hi Marek and Pascal,
Hi,
[...]
Quoted 10 lines
arch/arm/dts/stm32mp157a-avenger96.dts line 181
Quoted 6 lines
arch/arm/dts/stm32mp157a-avenger96.dts +1
-1@@ -167,7 +178,7 @@ vddcore: buck1 { regulator-name = "vddcore"; - regulator-min-microvolt = <1200000>; + regulator-min-microvolt = <800000>;I don't understood this modification, If I correctly remember, on STM32MP15x On ST boards, we set this value to 1200000 to protect VDDCORE on SOC (it is not allowed to be reduce lower than 1,2V.
Quoted 2 lines
0.8V is the minimal supported voltage by STPMIC but not the minal expected VDDCORE.
Then we should likely drop this part.
That said, is the core voltage supposed to be 1V2 even in sleep states ?
Quoted 17 lines
Quoted 12 lines
arch/arm/dts/stm32mp157a-avenger96.dts +2
-2@@ -185,8 +196,8 @@ vdd: buck3 { regulator-name = "vdd"; - regulator-min-microvolt = <3300000>; - regulator-max-microvolt = <3300000>; + regulator-min-microvolt = <2900000>; + regulator-max-microvolt = <2900000>; regulator-always-on; st,mask_reset; regulator-initial-mode = <0>;arch/arm/dts/stm32mp157a-avenger96.dts line 282
Quoted 4 lines
arch/arm/dts/stm32mp157a-avenger96.dts +1
-0@@ -268,6 +279,7 @@ regulator-name = "vbus_otg"; interrupts = <IT_OCP_OTG 0>; interrupt-parent = <&pmic>; + regulator-active-discharge = <1>;It seems this setting introduced other-consumption, it should not be used.
Quoted 3 lines
Quoted 1 line
Other-consumption ? What does that mean ?
It's a typo, Patrick means over-consumption.
Ah. Do you have any details on that ?
PATCH v3 12/14 ARM: dts: stm32: Repair PMIC configuration on AV96
Hi,
Quoted 19 lines
From: Marek Vasut <[email protected]>
Sent: mercredi 1 avril 2020 14:53
To: Patrice CHOTARD <[email protected]>; Patrick DELAUNAY <[email protected]>; u-boot at lists.denx.de; Pascal PAILLET-LME <[email protected]>
Cc: Manivannan Sadhasivam <[email protected]>; Alexandre TORGUE <[email protected]>
Subject: Re: [PATCH V3 12/14] ARM: dts: stm32: Repair PMIC configuration on AV96
Importance: HighQuoted 7 lines · Patrice CHOTARD
Quoted 5 lines · Marek Vasut
Quoted 1 line · Patrick DELAUNAY
Hi Marek and Pascal,
Hi,
[...]
[...]
Quoted 15 lines
Quoted 15 lines
Quoted 15 lines
Quoted 15 lines
Quoted 10 lines
arch/arm/dts/stm32mp157a-avenger96.dts +2
-2vdd: buck3 { regulator-name = "vdd"; - regulator-min-microvolt = <3300000>; - regulator-max-microvolt = <3300000>; + regulator-min-microvolt = <2900000>; + regulator-max-microvolt = <2900000>; regulator-always-on; st,mask_reset; regulator-initial-mode = <0>;arch/arm/dts/stm32mp157a-avenger96.dts line 282
Quoted 4 lines
arch/arm/dts/stm32mp157a-avenger96.dts +1
-0@@ -268,6 +279,7 @@ regulator-name = "vbus_otg"; interrupts = <IT_OCP_OTG 0>; interrupt-parent = <&pmic>; + regulator-active-discharge = <1>;It seems this setting introduced other-consumption, it should not be used.
Quoted 5 lines
Quoted 3 lines
Quoted 1 line
Other-consumption ? What does that mean ?
It's a typo, Patrick means over-consumption.
Ah. Do you have any details on that ?
Sorry for typo.
No details on my side, it was a Pascal Paillet's direct feedback (Maintainer of STPMIC1 driver in Linux).
Anyway he will review your patch for Linux, with more details, and I merge this patch without modification for U-Boot.
Patrick
PATCH v3 12/14 ARM: dts: stm32: Repair PMIC configuration on AV96
Quoted 43 lines · Patrick DELAUNAY
Hi,
Quoted 19 lines
From: Marek Vasut <[email protected]>
Sent: mercredi 1 avril 2020 14:53
To: Patrice CHOTARD <[email protected]>; Patrick DELAUNAY <[email protected]>; u-boot at lists.denx.de; Pascal PAILLET-LME <[email protected]>
Cc: Manivannan Sadhasivam <[email protected]>; Alexandre TORGUE <[email protected]>
Subject: Re: [PATCH V3 12/14] ARM: dts: stm32: Repair PMIC configuration on AV96
Importance: HighQuoted 7 lines · Patrice CHOTARD
Quoted 5 lines · Marek Vasut
Quoted 1 line · Patrick DELAUNAY
Hi Marek and Pascal,
Hi,
[...]
[...]
Quoted 15 lines
Quoted 15 lines
Quoted 15 lines
Quoted 15 lines
Quoted 10 lines
arch/arm/dts/stm32mp157a-avenger96.dts +2
-2vdd: buck3 { regulator-name = "vdd"; - regulator-min-microvolt = <3300000>; - regulator-max-microvolt = <3300000>; + regulator-min-microvolt = <2900000>; + regulator-max-microvolt = <2900000>; regulator-always-on; st,mask_reset; regulator-initial-mode = <0>;arch/arm/dts/stm32mp157a-avenger96.dts line 282
Quoted 4 lines
arch/arm/dts/stm32mp157a-avenger96.dts +1
-0@@ -268,6 +279,7 @@ regulator-name = "vbus_otg"; interrupts = <IT_OCP_OTG 0>; interrupt-parent = <&pmic>; + regulator-active-discharge = <1>;It seems this setting introduced other-consumption, it should not be used.
Quoted 5 lines
Quoted 3 lines
Quoted 1 line
Other-consumption ? What does that mean ?
It's a typo, Patrick means over-consumption.
Ah. Do you have any details on that ?
Quoted 7 lines
Sorry for typo.
No details on my side, it was a Pascal Paillet's direct feedback (Maintainer of STPMIC1 driver in Linux).
Anyway he will review your patch for Linux, with more details, and I merge this patch without modification for U-Boot.
OK
PATCH v3 12/14 ARM: dts: stm32: Repair PMIC configuration on AV96
In reply to Patrick DELAUNAYHi Marek,
Quoted 52 lines
-----Original Message-----
From: Patrick DELAUNAY <[email protected]>
Sent: mercredi 1 avril 2020 15:52
To: Marek Vasut <[email protected]>; Patrice CHOTARD <[email protected]>; u-boot at lists.denx.de; Pascal PAILLET-LME <[email protected]>
Cc: Manivannan Sadhasivam <[email protected]>; Alexandre TORGUE <[email protected]>
Subject: RE: [PATCH V3 12/14] ARM: dts: stm32: Repair PMIC configuration on AV96Hi,
Quoted 20 lines
From: Marek Vasut <[email protected]>
Sent: mercredi 1 avril 2020 14:53
To: Patrice CHOTARD <[email protected]>; Patrick DELAUNAY <[email protected]>; u-boot at lists.denx.de; Pascal PAILLET-LME <[email protected]>
Cc: Manivannan Sadhasivam <[email protected]>; Alexandre TORGUE <[email protected]>
Subject: Re: [PATCH V3 12/14] ARM: dts: stm32: Repair PMIC configuration on
AV96
Importance: HighQuoted 7 lines · Patrice CHOTARD
Quoted 5 lines · Marek Vasut
Quoted 1 line · Patrick DELAUNAY
Hi Marek and Pascal,
Hi,
[...]
[...]
Quoted 9 lines
Quoted 9 lines
Quoted 9 lines
Quoted 9 lines
arch/arm/dts/stm32mp157a-avenger96.dts
Quoted 9 lines
arch/arm/dts/stm32mp157a-avenger96.dts +2
-2vdd: buck3 { regulator-name = "vdd"; - regulator-min-microvolt = <3300000>; - regulator-max-microvolt = <3300000>; + regulator-min-microvolt = <2900000>; + regulator-max-microvolt = <2900000>; regulator-always-on; st,mask_reset; regulator-initial-mode = <0>; @@ -268,6 +279,7 @@
Quoted 5 lines
Quoted 5 lines
Quoted 5 lines
Quoted 5 lines
arch/arm/dts/stm32mp157a-avenger96.dts line 282
Quoted 4 lines
arch/arm/dts/stm32mp157a-avenger96.dts +1
-0regulator-name = "vbus_otg"; interrupts = <IT_OCP_OTG 0>; interrupt-parent = <&pmic>; + regulator-active-discharge = <1>;It seems this setting introduced other-consumption, it should not be
used.
Quoted 2 lines
Quoted 2 lines
Quoted 1 line
Other-consumption ? What does that mean ?
Quoted 8 lines
Quoted 3 lines
Quoted 1 line
It's a typo, Patrick means over-consumption.
Ah. Do you have any details on that ?
Sorry for typo.
No details on my side, it was a Pascal Paillet's direct feedback (Maintainer of STPMIC1 driver in Linux).
You can enable the active discharge on the STPMIC1's VBUSOTG. But, as soon as a USB host peripheral (such as laptop) will attach and provide VBUS, there will be a ~8mA continuous power consumption on VBUS through this OTG active discharge. There is no functional issue behind this 8mA; but you need to know that 8mA is not compliant with the USB suspend constraint (2.5mA max).
Quoted 5 lines
Anyway he will review your patch for Linux, with more details, and I merge this patch without modification for U-Boot.
Patrick
Pascal
PATCH v3 12/14 ARM: dts: stm32: Repair PMIC configuration on AV96
Quoted 1 line · Pascal PAILLET-LME
Hi Marek,
Hi,
[...]
Quoted 20 lines
Quoted 20 lines
Quoted 9 lines
Quoted 9 lines
Quoted 9 lines
Quoted 9 lines
Review
Quoted 9 lines
(quoted diff) +2
-2vdd: buck3 { regulator-name = "vdd"; - regulator-min-microvolt = <3300000>; - regulator-max-microvolt = <3300000>; + regulator-min-microvolt = <2900000>; + regulator-max-microvolt = <2900000>; regulator-always-on; st,mask_reset; regulator-initial-mode = <0>; @@ -268,6 +279,7 @@
Quoted 5 lines
Quoted 5 lines
Quoted 5 lines
Quoted 5 lines
Review
Quoted 4 lines
(quoted diff) +1
-0regulator-name = "vbus_otg"; interrupts = <IT_OCP_OTG 0>; interrupt-parent = <&pmic>; + regulator-active-discharge = <1>;It seems this setting introduced other-consumption, it should not be
used.
Quoted 4 lines
Quoted 3 lines
Quoted 1 line
Other-consumption ? What does that mean ?
It's a typo, Patrick means over-consumption.
Quoted 7 lines
Quoted 6 lines
Quoted 1 line
Ah. Do you have any details on that ?
Sorry for typo.
No details on my side, it was a Pascal Paillet's direct feedback (Maintainer of STPMIC1 driver in Linux).
You can enable the active discharge on the STPMIC1's VBUSOTG. But, as soon as a USB host peripheral (such as laptop) will attach and provide VBUS, there will be a ~8mA continuous power consumption on VBUS through this OTG active discharge. There is no functional issue behind this 8mA; but you need to know that 8mA is not compliant with the USB suspend constraint (2.5mA max).
So why is it in there ?
PATCH v3 12/14 ARM: dts: stm32: Repair PMIC configuration on AV96
Quoted 44 lines
-----Original Message-----
From: Marek Vasut <[email protected]>
Sent: jeudi 2 avril 2020 15:07
To: Pascal PAILLET-LME <[email protected]>; Patrick DELAUNAY <[email protected]>; Patrice CHOTARD <[email protected]>; u- boot at lists.denx.de
Cc: Manivannan Sadhasivam <[email protected]>; Alexandre TORGUE <[email protected]>
Subject: Re: [PATCH V3 12/14] ARM: dts: stm32: Repair PMIC configuration on AV96Quoted 1 line · Pascal PAILLET-LME
Hi Marek,
Hi,
[...]Quoted 18 lines
Quoted 18 lines
Quoted 9 lines
Quoted 9 lines
Quoted 9 lines
Quoted 9 lines
Review
Quoted 9 lines
(quoted diff) +2
-2vdd: buck3 { regulator-name = "vdd"; - regulator-min-microvolt = <3300000>; - regulator-max-microvolt = <3300000>; + regulator-min-microvolt = <2900000>; + regulator-max-microvolt = <2900000>; regulator-always-on; st,mask_reset; regulator-initial-mode = <0>; @@ -268,6 +279,7 @@
Quoted 5 lines
Quoted 5 lines
Quoted 5 lines
Quoted 5 lines
Review
Quoted 4 lines
(quoted diff) +1
-0regulator-name = "vbus_otg"; interrupts = <IT_OCP_OTG 0>; interrupt-parent = <&pmic>; + regulator-active-discharge = <1>;It seems this setting introduced other-consumption, it should not be
used.
Quoted 2 lines
Quoted 2 lines
Quoted 1 line
Other-consumption ? What does that mean ?
Quoted 7 lines
Quoted 7 lines
Quoted 3 lines
Quoted 1 line
It's a typo, Patrick means over-consumption.
Ah. Do you have any details on that ?
Sorry for typo.
No details on my side, it was a Pascal Paillet's direct feedback (Maintainer
of
Quoted 1 line
Quoted 1 line
STPMIC1 driver in Linux).
Quoted 8 lines
Quoted 1 line
You can enable the active discharge on the STPMIC1's VBUSOTG. But, as
soon as a USB host peripheral (such as laptop) will attach and provide VBUS, there will be a ~8mA continuous power consumption on VBUS through this OTG active discharge. There is no functional issue behind this 8mA; but you need to know that 8mA is not compliant with the USB suspend constraint (2.5mA max).
So why is it in there ?
Sorry, I'm not sure to catch your point... The property is available because the stpmic1 Propose this feature (enable a pull down resistor when VBUS_OTG is disabled). We have Found during some test that it is better to not enable active discharge on vbus_otg. I hope it answers,
pascal
PATCH v3 12/14 ARM: dts: stm32: Repair PMIC configuration on AV96
On 4/2/20 3:43 PM, Pascal PAILLET-LME wrote:
Hi,
[...]
Quoted 10 lines
Quoted 10 lines
Quoted 1 line
Quoted 1 line
No details on my side, it was a Pascal Paillet's direct feedback (Maintainer
of
Quoted 2 lines
Quoted 1 line
STPMIC1 driver in Linux).
You can enable the active discharge on the STPMIC1's VBUSOTG. But, as
soon as a USB host peripheral (such as laptop) will attach and provide VBUS, there will be a ~8mA continuous power consumption on VBUS through this OTG active discharge. There is no functional issue behind this 8mA; but you need to know that 8mA is not compliant with the USB suspend constraint (2.5mA max).
Quoted 6 lines
Quoted 1 line
So why is it in there ?
Sorry, I'm not sure to catch your point... The property is available because the stpmic1 Propose this feature (enable a pull down resistor when
VBUS_OTGis disabled). We have Found during some test that it is better to not enable active discharge onvbus_otg. I hope it answers,
pascal
Can you go into a bit more detail on what you found during your tests and why it was a problem ?
PATCH v3 12/14 ARM: dts: stm32: Repair PMIC configuration on AV96
In reply to Marek VasutHi,
Quoted 8 lines
From: Marek Vasut <[email protected]>
Sent: mardi 31 mars 2020 19:52The core and vdd PMIC buck regulators were misconfigured, which caused instability of the board and malfunction of high-speed interfaces, like the RGMII. Configure the PMIC correctly to repair these problems. Also, model the missing Enpirion EP53A8LQI on the DHCOR SoM as a fixed regulator.
Quoted 6 lines
Reviewed-by Patrice Chotard <[email protected]>Signed-off-by Marek Vasut <[email protected]>Cc Manivannan Sadhasivam <[email protected]>Cc Patrick Delaunay <[email protected]>Cc Patrice Chotard <[email protected]>---
Applied to u-boot-stm/master, thanks!
Regards
Patrick
PATCH v3 13/14 ARM: dts: stm32: Adjust PLL4 settings on AV96
In reply to Marek VasutThe PLL4 is supplying SDMMC12, SDMMC3 and SPDIF with 120 MHz and FDCAN with 96 MHz. This isn't good for the SDMMC interfaces, which can not easily divide the clock down to e.g. 50 MHz for high speed SD and eMMC devices, so those devices end up running at 30 MHz as that is 120 MHz / 4. Adjust the PLL4 settings such that both PLL4P and PLL4R run at 100 MHz instead, which is easy to divide to 50MHz for optimal operation of both SD and eMMC, SPDIF clock are not that much slower and FDCAN is also unaffected.
---
V2: Move this patch before the split of AV96 into SoM and carrier
V3: No change
--- arch/arm/dts/stm32mp157a-avenger96-u-boot.dtsi | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
arch/arm/dts/stm32mp157a-avenger96-u-boot.dtsi +2 -2
@@ -130,11 +130,11 @@ u-boot,dm-pre-reloc; }; - /* VCO = 480.0 MHz => P = 120, Q = 40, R = 96 */ + /* VCO = 600.0 MHz => P = 100, Q = 50, R = 100 */ pll4: st,pll at 3 { compatible = "st,stm32mp1-pll"; reg = <3>; - cfg = < 1 39 3 11 4 PQR(1,1,1) >; + cfg = < 1 49 5 11 5 PQR(1,1,1) >; u-boot,dm-pre-reloc; }; };
Signature · 1 line
2.25.1
PATCH v3 13/14 ARM: dts: stm32: Adjust PLL4 settings on AV96
Hi Gerald and Manivannan,
Quoted 17 lines
From: Marek Vasut <[email protected]>
Sent: mardi 31 mars 2020 19:52The PLL4 is supplying SDMMC12, SDMMC3 and SPDIF with 120 MHz and FDCAN with 96 MHz. This isn't good for the SDMMC interfaces, which can not easily divide the clock down to e.g. 50 MHz for high speed SD and eMMC devices, so those devices end up running at 30 MHz as that is 120 MHz / 4. Adjust the PLL4 settings such that both PLL4P and PLL4R run at 100 MHz instead, which is easy to divide to 50MHz for optimal operation of both SD and eMMC, SPDIF clock are not that much slower and FDCAN is also unaffected.
Reviewed-by Patrice Chotard <[email protected]>Reviewed-by Patrick Delaunay <[email protected]>Signed-off-by Marek Vasut <[email protected]>Cc Manivannan Sadhasivam <[email protected]>Cc Patrick Delaunay <[email protected]>Cc Patrice Chotard <[email protected]>
Quoted 4 lines
---
V2: Move this patch before the split of AV96 into SoM and carrier
V3: No change
---
This patch update the PLL4 frequency used on AV96 board, with different of reference clock tree used on ST board, this new setting allow to optimize the SDMMC frequency (50MHz vs 30Mz).
I don't know why the previous PLL4 frequency was chosen as a compromise on reference clock-tree (PLL4 is used by mostly all the peripheral, with display and audio requirements).
Can you cross check the proposed clock tree and ack this patch if these ST requirements are not applicable on AV96 board.
Anyway the code is correct.
Thanks
Patrick
PATCH v3 13/14 ARM: dts: stm32: Adjust PLL4 settings on AV96
Quoted 1 line · Patrick DELAUNAY
Hi Gerald and Manivannan,
Hi,
Quoted 26 lines
Quoted 17 lines
From: Marek Vasut <[email protected]>
Sent: mardi 31 mars 2020 19:52The PLL4 is supplying SDMMC12, SDMMC3 and SPDIF with 120 MHz and FDCAN with 96 MHz. This isn't good for the SDMMC interfaces, which can not easily divide the clock down to e.g. 50 MHz for high speed SD and eMMC devices, so those devices end up running at 30 MHz as that is 120 MHz / 4. Adjust the PLL4 settings such that both PLL4P and PLL4R run at 100 MHz instead, which is easy to divide to 50MHz for optimal operation of both SD and eMMC, SPDIF clock are not that much slower and FDCAN is also unaffected.
Reviewed-by Patrice Chotard <[email protected]>Reviewed-by Patrick Delaunay <[email protected]>Signed-off-by Marek Vasut <[email protected]>Cc Manivannan Sadhasivam <[email protected]>Cc Patrick Delaunay <[email protected]>Cc Patrice Chotard <[email protected]>Quoted 4 lines
---
V2: Move this patch before the split of AV96 into SoM and carrier
V3: No change
---This patch update the PLL4 frequency used on AV96 board, with different of reference clock tree used on ST board, this new setting allow to optimize the SDMMC frequency (50MHz vs 30Mz).
Quoted 8 lines
I don't know why the previous PLL4 frequency was chosen as a compromise on reference clock-tree (PLL4 is used by mostly all the peripheral, with display and audio requirements).
Can you cross check the proposed clock tree and ack this patch if these ST requirements are not applicable on AV96 board.
Anyway the code is correct.
Likely because these PLL settings are being copied from reference platform to other platforms etc.
But I did notice one odd thing, which is when running the SD1 in SDR104, the read data transfers can be unstable, which I suspect is because the bus runs at actual 100 MHz instead of some 60 MHz. I need to look at that with a scope, so that's to be checked. For now I turned the SDR104 off.
PATCH v3 13/14 ARM: dts: stm32: Adjust PLL4 settings on AV96
Hi Marek and Patrick
Quoted 15 lines
From: Marek Vasut <[email protected]>
Sent: mercredi 1 avril 2020 13:09Quoted 1 line · Patrick DELAUNAY
Hi Gerald and Manivannan,
Hi,
Quoted 7 lines
Quoted 7 lines
From: Marek Vasut <[email protected]>
Sent: mardi 31 mars 2020 19:52The PLL4 is supplying SDMMC12, SDMMC3 and SPDIF with 120 MHz and FDCAN with 96 MHz. This isn't good for the SDMMC interfaces, which can not easily divide the clock down to e.g. 50 MHz for high speed SD and eMMC devices, so those devices end up running at 30 MHz as that is
Quoted 5 lines
120 MHz / 4.
Quoted 3 lines
Quoted 3 lines
Adjust the PLL4 settings such that both PLL4P and PLL4R run at 100 MHz instead, which is easy to divide to 50MHz for optimal operation of both SD and eMMC, SPDIF clock are not that much slower and FDCAN is
also unaffected.
As far as I see, SPDIF is not supported on Avenger board so we don't care for this one.
FDCAN can be supported via the expansion connector, so better to keep it high (and < 100 MHz).
Be careful because the LTDC pixel clock also comes from the PLL4Q and it is used for the HDMI on Avenger.
The pixel clock is very constraint and I am surprised by the initial 40 MHz configuration (that becomes 50 MHz with your patch).
I would recommend to align the Avenger configuration to ST boards one, that is the best compromise found so far (99 MHz for SDMMC and 74.250 MHz for HDMI):
https://wiki.st.com/stm32mpu/wiki/STM32MP15_clock_tree /* VCO = 594.0 MHz => P = 99, Q = 74, R = 74 */ pll4: st,pll at 3 { cfg = < 3 98 5 7 7 PQR(1,1,1) >; u-boot,dm-pre-reloc; };
Quoted 19 lines
Quoted 19 lines
Quoted 10 lines
Reviewed-by Patrice Chotard <[email protected]>Reviewed-by Patrick Delaunay <[email protected]>Signed-off-by Marek Vasut <[email protected]>Cc Manivannan Sadhasivam <[email protected]>Cc Patrick Delaunay <[email protected]>Cc Patrice Chotard <[email protected]>---
V2: Move this patch before the split of AV96 into SoM and carrier
V3: No change
---This patch update the PLL4 frequency used on AV96 board, with different of reference clock tree used on ST board, this new setting allow to optimize the SDMMC frequency (50MHz vs 30Mz).
I don't know why the previous PLL4 frequency was chosen as a compromise on reference clock-tree (PLL4 is used by mostly all the peripheral, with display and audio requirements).
Quoted 7 lines
Quoted 4 lines
Can you cross check the proposed clock tree and ack this patch if these ST requirements are not applicable on AV96 board.
Anyway the code is correct.
Likely because these PLL settings are being copied from reference platform to other platforms etc.
As far as I remember, we never had this configuration for PLL4 on ST boards, so the copy certainly comes from somewhere else.
Quoted 4 lines
But I did notice one odd thing, which is when running the SD1 in SDR104, the read data transfers can be unstable, which I suspect is because the bus runs at actual 100 MHz instead of some 60 MHz. I need to look at that with a scope, so that's to be checked. For now I turned the SDR104 off.
PATCH v3 13/14 ARM: dts: stm32: Adjust PLL4 settings on AV96
Hi,
Quoted 28 lines · Gerald BAEZA
Hi Marek and Patrick
Quoted 15 lines
From: Marek Vasut <[email protected]>
Sent: mercredi 1 avril 2020 13:09Quoted 1 line · Patrick DELAUNAY
Hi Gerald and Manivannan,
Hi,
Quoted 7 lines
Quoted 7 lines
From: Marek Vasut <[email protected]>
Sent: mardi 31 mars 2020 19:52The PLL4 is supplying SDMMC12, SDMMC3 and SPDIF with 120 MHz and FDCAN with 96 MHz. This isn't good for the SDMMC interfaces, which can not easily divide the clock down to e.g. 50 MHz for high speed SD and eMMC devices, so those devices end up running at 30 MHz as that is
Quoted 5 lines
120 MHz / 4.
Quoted 3 lines
Quoted 3 lines
Adjust the PLL4 settings such that both PLL4P and PLL4R run at 100 MHz instead, which is easy to divide to 50MHz for optimal operation of both SD and eMMC, SPDIF clock are not that much slower and FDCAN is
also unaffected.
As far as I see, SPDIF is not supported on Avenger board so we don't care for this one.
FDCAN can be supported via the expansion connector, so better to keep it high (and < 100 MHz).
Be careful because the LTDC pixel clock also comes from the PLL4Q and it is used for the HDMI on Avenger.
The pixel clock is very constraint and I am surprised by the initial 40 MHz configuration (that becomes 50 MHz with your patch).
I would recommend to align the Avenger configuration to ST boards one, that is the best compromise found so far (99 MHz for SDMMC and 74.250 MHz for HDMI):
Quoted 7 lines
https://wiki.st.com/stm32mpu/wiki/STM32MP15_clock_tree /* VCO = 594.0 MHz => P = 99, Q = 74, R = 74 */ pll4: st,pll at 3 { cfg = < 3 98 5 7 7 PQR(1,1,1) >; u-boot,dm-pre-reloc; };
This looks good to me.
Quoted 22 lines
Quoted 22 lines
Quoted 22 lines
Quoted 10 lines
Reviewed-by Patrice Chotard <[email protected]>Reviewed-by Patrick Delaunay <[email protected]>Signed-off-by Marek Vasut <[email protected]>Cc Manivannan Sadhasivam <[email protected]>Cc Patrick Delaunay <[email protected]>Cc Patrice Chotard <[email protected]>---
V2: Move this patch before the split of AV96 into SoM and carrier
V3: No change
---This patch update the PLL4 frequency used on AV96 board, with different of reference clock tree used on ST board, this new setting allow to optimize the SDMMC frequency (50MHz vs 30Mz).
I don't know why the previous PLL4 frequency was chosen as a compromise on reference clock-tree (PLL4 is used by mostly all the peripheral, with display and audio requirements).
Can you cross check the proposed clock tree and ack this patch if these ST requirements are not applicable on AV96 board.
Quoted 7 lines
Quoted 4 lines
Quoted 1 line
Anyway the code is correct.
Likely because these PLL settings are being copied from reference platform to other platforms etc.
As far as I remember, we never had this configuration for PLL4 on ST boards, so the copy certainly comes from somewhere else.
I took the PLL settings from Downstream code done by Arrow at that time. Since I never had access to PMIC info, I just used them blindly since that worked for initial upstream bringup on old board. We had several discussions about PMIC at that time but never touched PLL4.
So feel free to modify it as Gerald suggested.
Thanks,
Mani
Quoted 5 lines
Quoted 4 lines
But I did notice one odd thing, which is when running the SD1 in SDR104, the read data transfers can be unstable, which I suspect is because the bus runs at actual 100 MHz instead of some 60 MHz. I need to look at that with a scope, so that's to be checked. For now I turned the SDR104 off.
PATCH v3 13/14 ARM: dts: stm32: Adjust PLL4 settings on AV96
In reply to Gerald BAEZAQuoted 1 line · Gerald BAEZA
Hi Marek and Patrick
Hi,
Quoted 15 lines
Quoted 15 lines
From: Marek Vasut <[email protected]>
Sent: mercredi 1 avril 2020 13:09Quoted 1 line · Patrick DELAUNAY
Hi Gerald and Manivannan,
Hi,
Quoted 7 lines
Quoted 7 lines
From: Marek Vasut <[email protected]>
Sent: mardi 31 mars 2020 19:52The PLL4 is supplying SDMMC12, SDMMC3 and SPDIF with 120 MHz and FDCAN with 96 MHz. This isn't good for the SDMMC interfaces, which can not easily divide the clock down to e.g. 50 MHz for high speed SD and eMMC devices, so those devices end up running at 30 MHz as that is
Quoted 8 lines
Quoted 5 lines
120 MHz / 4.
Quoted 3 lines
Quoted 3 lines
Adjust the PLL4 settings such that both PLL4P and PLL4R run at 100 MHz instead, which is easy to divide to 50MHz for optimal operation of both SD and eMMC, SPDIF clock are not that much slower and FDCAN is
also unaffected.
As far as I see, SPDIF is not supported on Avenger board so we don't care for this one. FDCAN can be supported via the expansion connector, so better to keep it high (and < 100 MHz).
Why < 100 MHz and not <= 100 MHz ?
Quoted 2 lines
Be careful because the LTDC pixel clock also comes from the PLL4Q and it is used for the HDMI on Avenger. The pixel clock is very constraint and I am surprised by the initial 40 MHz configuration (that becomes 50 MHz with your patch).
Is that a problem that the LTDC pixel clock are 50 MHz ? The kernel will reconfigure them anyway, so the 50 MHz is not the final value.
Quoted 1 line
I would recommend to align the Avenger configuration to ST boards one, that is the best compromise found so far (99 MHz for SDMMC and 74.250 MHz for HDMI):
Why is this better than 100/50/100 ?
Quoted 6 lines
https://wiki.st.com/stm32mpu/wiki/STM32MP15_clock_tree /* VCO = 594.0 MHz => P = 99, Q = 74, R = 74 */ pll4: st,pll at 3 { cfg = < 3 98 5 7 7 PQR(1,1,1) >; u-boot,dm-pre-reloc; };
[...]
PATCH v3 13/14 ARM: dts: stm32: Adjust PLL4 settings on AV96
Hi Marek
Quoted 30 lines
From: Marek Vasut <[email protected]>
Sent: mercredi 1 avril 2020 20:49Quoted 1 line · Gerald BAEZA
Hi Marek and Patrick
Hi,
Quoted 16 lines
Quoted 16 lines
From: Marek Vasut <[email protected]>
Sent: mercredi 1 avril 2020 13:09Quoted 1 line · Patrick DELAUNAY
Hi Gerald and Manivannan,
Hi,
Quoted 8 lines
Quoted 8 lines
From: Marek Vasut <[email protected]>
Sent: mardi 31 mars 2020 19:52The PLL4 is supplying SDMMC12, SDMMC3 and SPDIF with 120 MHz and FDCAN with 96 MHz. This isn't good for the SDMMC interfaces, which can not easily divide the clock down to e.g. 50 MHz for high speed SD and eMMC devices, so those devices end up running at 30 MHz as that is
Quoted 4 lines
Quoted 4 lines
120 MHz / 4.
Quoted 3 lines
Quoted 3 lines
Adjust the PLL4 settings such that both PLL4P and PLL4R run at 100 MHz instead, which is easy to divide to 50MHz for optimal operation of both SD and eMMC, SPDIF clock are not that much slower and
FDCAN
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
is
Quoted 8 lines
Quoted 3 lines
Quoted 1 line
also unaffected.
As far as I see, SPDIF is not supported on Avenger board so we don't care
for this one.
Quoted 1 line
FDCAN can be supported via the expansion connector, so better to keep it
high (and < 100 MHz).
Why < 100 MHz and not <= 100 MHz ?
You're right, it is <= 100 MHz and it is important to be precise since you were exactly on 100 MHz :)
Quoted 7 lines
Quoted 1 line
Be careful because the LTDC pixel clock also comes from the PLL4Q and it is
used for the HDMI on Avenger.
Quoted 1 line
The pixel clock is very constraint and I am surprised by the initial 40 MHz
configuration (that becomes 50 MHz with your patch).
Is that a problem that the LTDC pixel clock are 50 MHz ? The kernel will reconfigure them anyway, so the 50 MHz is not the final value.
The kernel set_rate() changes the PLL output divisor, so it will indeed be able to switch back to (600/15=) 40 MHz from an initial (600/12=) 50 MHz.
Quoted 6 lines
Quoted 1 line
I would recommend to align the Avenger configuration to ST boards one,
that is the best compromise found so far (99 MHz for SDMMC and 74.250 MHz for HDMI):
Why is this better than 100/50/100 ?
Quoted 7 lines
Quoted 6 lines
https://wiki.st.com/stm32mpu/wiki/STM32MP15_clock_tree /* VCO = 594.0 MHz => P = 99, Q = 74, R = 74 */ pll4: st,pll at 3 { cfg = < 3 98 5 7 7 PQR(1,1,1) >; u-boot,dm-pre-reloc; };[...]
The simplest explanation I found is here:
https://timetoexplore.net/blog/video-timings-vga-720p-1080p
(you can also look for "74.25" in the HDMI specification but there is more text to read)
So 74.250 MHz is the needed pixel clock for 720p resolution in HDMI, that is quite common, so this frequency is wished for interoperability with a maximum of TVs. This frequency cannot be reached from the initial or your proposed PLL4 configuration, that is why I proposed to change this and align on ST board clock tree.
For the other displays we are supporting on ST boards, the kernel is tuning (down) this 74.250 MHz frequency via the set_rate (that changes the PLL output divisor).
Best regards
G?rald
PATCH v3 13/14 ARM: dts: stm32: Adjust PLL4 settings on AV96
Quoted 67 lines · Gerald BAEZA
Hi Marek
Quoted 30 lines
From: Marek Vasut <[email protected]>
Sent: mercredi 1 avril 2020 20:49Quoted 1 line · Gerald BAEZA
Hi Marek and Patrick
Hi,
Quoted 16 lines
Quoted 16 lines
From: Marek Vasut <[email protected]>
Sent: mercredi 1 avril 2020 13:09Quoted 1 line · Patrick DELAUNAY
Hi Gerald and Manivannan,
Hi,
Quoted 8 lines
Quoted 8 lines
From: Marek Vasut <[email protected]>
Sent: mardi 31 mars 2020 19:52The PLL4 is supplying SDMMC12, SDMMC3 and SPDIF with 120 MHz and FDCAN with 96 MHz. This isn't good for the SDMMC interfaces, which can not easily divide the clock down to e.g. 50 MHz for high speed SD and eMMC devices, so those devices end up running at 30 MHz as that is
Quoted 4 lines
Quoted 4 lines
120 MHz / 4.
Quoted 3 lines
Quoted 3 lines
Adjust the PLL4 settings such that both PLL4P and PLL4R run at 100 MHz instead, which is easy to divide to 50MHz for optimal operation of both SD and eMMC, SPDIF clock are not that much slower and
FDCAN
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
is
Quoted 8 lines
Quoted 3 lines
Quoted 1 line
also unaffected.
As far as I see, SPDIF is not supported on Avenger board so we don't care
for this one.
Quoted 1 line
FDCAN can be supported via the expansion connector, so better to keep it
high (and < 100 MHz).
Why < 100 MHz and not <= 100 MHz ?
You're right, it is <= 100 MHz and it is important to be precise since you were exactly on 100 MHz :)
Quoted 7 lines
Quoted 1 line
Be careful because the LTDC pixel clock also comes from the PLL4Q and it is
used for the HDMI on Avenger.
Quoted 1 line
The pixel clock is very constraint and I am surprised by the initial 40 MHz
configuration (that becomes 50 MHz with your patch).
Is that a problem that the LTDC pixel clock are 50 MHz ? The kernel will reconfigure them anyway, so the 50 MHz is not the final value.
The kernel
set_rate()changes the PLL output divisor, so it will indeed be able to switch back to (600/15=) 40 MHz from an initial (600/12=) 50 MHz.Quoted 6 lines
Quoted 1 line
I would recommend to align the Avenger configuration to ST boards one,
that is the best compromise found so far (99 MHz for SDMMC and 74.250 MHz for HDMI):
Why is this better than 100/50/100 ?
Quoted 7 lines
Quoted 6 lines
https://wiki.st.com/stm32mpu/wiki/STM32MP15_clock_tree /* VCO = 594.0 MHz => P = 99, Q = 74, R = 74 */ pll4: st,pll at 3 { cfg = < 3 98 5 7 7 PQR(1,1,1) >; u-boot,dm-pre-reloc; };[...]
Quoted 8 lines
The simplest explanation I found is here:
https://timetoexplore.net/blog/video-timings-vga-720p-1080p
(you can also look for "74.25" in the HDMI specification but there is more text to read)So 74.250 MHz is the needed pixel clock for 720p resolution in HDMI, that is quite common, so this frequency is wished for interoperability with a maximum of TVs. This frequency cannot be reached from the initial or your proposed PLL4 configuration, that is why I proposed to change this and align on ST board clock tree.
For the other displays we are supporting on ST boards, the kernel is tuning (down) this 74.250 MHz frequency via the
set_rate(that changes the PLL output divisor).
The article you found is focused on FPGA, where the FPGA is connected directly to the HDMI TMDS signals without transceiver, so the FPGA has to generate very precise clock. I don't think this is our case here.
The ADV7513 datasheet says nothing about such specific clock rate, it only mentions maximum clock per resolution to be 82.5 MHz and that it can do pixel repeat to meet the timing requirements.
(Note to self, PLL4R should run at 99 MHz too, to run FDCAN fast enough.)
PATCH v3 13/14 ARM: dts: stm32: Adjust PLL4 settings on AV96
Hi,
Quoted 100 lines
From: Marek Vasut <[email protected]>
Sent: jeudi 2 avril 2020 15:05Quoted 22 lines · Gerald BAEZA
Hi Marek
Quoted 20 lines
From: Marek Vasut <[email protected]>
Sent: mercredi 1 avril 2020 20:49Quoted 1 line · Gerald BAEZA
Hi Marek and Patrick
Hi,
Quoted 12 lines
Quoted 12 lines
From: Marek Vasut <[email protected]>
Sent: mercredi 1 avril 2020 13:09Quoted 1 line · Patrick DELAUNAY
Hi Gerald and Manivannan,
Hi,
Quoted 4 lines
Quoted 4 lines
From: Marek Vasut <[email protected]>
Sent: mardi 31 mars 2020 19:52The PLL4 is supplying SDMMC12, SDMMC3 and SPDIF with 120 MHz
and
Quoted 28 lines
Quoted 13 lines
Quoted 9 lines
Quoted 9 lines
Quoted 4 lines
Quoted 4 lines
FDCAN with 96 MHz. This isn't good for the SDMMC interfaces, which can not easily divide the clock down to e.g. 50 MHz for high speed SD and eMMC devices, so those devices end up running at 30 MHz as that is
120 MHz / 4.
Quoted 4 lines
Quoted 4 lines
Adjust the PLL4 settings such that both PLL4P and PLL4R run at 100 MHz instead, which is easy to divide to 50MHz for optimal operation of both SD and eMMC, SPDIF clock are not that much slower and
FDCAN
Quoted 3 lines
Quoted 2 lines
Quoted 1 line
Quoted 1 line
is
also unaffected.
Quoted 8 lines
Quoted 2 lines
As far as I see, SPDIF is not supported on Avenger board so we don't care
for this one.
Quoted 2 lines
FDCAN can be supported via the expansion connector, so better to keep it
high (and < 100 MHz).
Why < 100 MHz and not <= 100 MHz ?
You're right, it is <= 100 MHz and it is important to be precise since you were exactly on 100 MHz :)
Quoted 3 lines
Quoted 2 lines
Be careful because the LTDC pixel clock also comes from the PLL4Q and it is
used for the HDMI on Avenger.
Quoted 8 lines
Quoted 6 lines
Quoted 2 lines
The pixel clock is very constraint and I am surprised by the initial 40 MHz
configuration (that becomes 50 MHz with your patch).
Is that a problem that the LTDC pixel clock are 50 MHz ? The kernel will reconfigure them anyway, so the 50 MHz is not the final value.
The kernel
set_rate()changes the PLL output divisor, so it will indeed be able toswitch back to (600/15=) 40 MHz from an initial (600/12=) 50 MHz.
Quoted 14 lines
Quoted 14 lines
Quoted 3 lines
I would recommend to align the Avenger configuration to ST boards one,
that is the best compromise found so far (99 MHz for SDMMC and 74.250 MHz for HDMI):
Why is this better than 100/50/100 ?
Quoted 6 lines
https://wiki.st.com/stm32mpu/wiki/STM32MP15_clock_tree /* VCO = 594.0 MHz => P = 99, Q = 74, R = 74 */ pll4: st,pll at 3 { cfg = < 3 98 5 7 7 PQR(1,1,1) >; u-boot,dm-pre-reloc; };Quoted 8 lines
Quoted 1 line
[...]
The simplest explanation I found is here:
https://timetoexplore.net/blog/video-timings-vga-720p-1080p
(you can also look for "74.25" in the HDMI specification but there is more text to read)So 74.250 MHz is the needed pixel clock for 720p resolution in HDMI, that is
quite common, so this frequency is wished for interoperability with a maximum of TVs.
Quoted 1 line
This frequency cannot be reached from the initial or your proposed PLL4
configuration, that is why I proposed to change this and align on ST board clock tree.
Quoted 2 lines
For the other displays we are supporting on ST boards, the kernel is tuning
(down) this 74.250 MHz frequency via the
set_rate(that changes the PLL output divisor).The article you found is focused on FPGA, where the FPGA is connected directly to the HDMI TMDS signals without transceiver, so the FPGA has to generate very precise clock. I don't think this is our case here.
Quoted 5 lines
The ADV7513 datasheet says nothing about such specific clock rate, it only mentions maximum clock per resolution to be 82.5 MHz and that it can do pixel repeat to meet the timing requirements.
(Note to self, PLL4R should run at 99 MHz too, to run FDCAN fast enough.)
Without more requirement on the clock tree, the new PLL4 is acceptable.
Applied to u-boot-stm/master, thanks!
Regards
Patrick
PATCH v3 14/14 ARM: dts: stm32: Split AV96 into DHCOR SoM and AV96 board
In reply to Marek VasutThe AV96 is in fact an assembly of DH Electronics DHCOR SoM on top of an AV96 reference board. Split the DTs to reflect that and make sure to DHCOR SoM can be reused on other boards easily.
It is also highly recommended to configure the board for the DHCOM make stm32mp15_dhcom_basic_defconfig
make DEVICE_TREE=stm32mp15xx-dhcor-avenger96
as that permits reusing the board code for the DH components, like accessing and reading out the ethernet MAC from EEPROM.
---
V2: No change
V3: No change
---
arch/arm/dts/Makefile | 3 +-
arch/arm/dts/stm32mp157a-avenger96.dts | 421 +-----------------
.../stm32mp15xx-dhcor-avenger96-u-boot.dtsi | 80 ++++
arch/arm/dts/stm32mp15xx-dhcor-avenger96.dts | 211 +++++++++
...oot.dtsi => stm32mp15xx-dhcor-u-boot.dtsi} | 79 +---
arch/arm/dts/stm32mp15xx-dhcor.dtsi | 231 ++++++++++
doc/board/st/stm32mp1.rst | 8 +-
7 files changed, 535 insertions(+), 498 deletions(-)
create mode 100644 arch/arm/dts/stm32mp15xx-dhcor-avenger96-u-boot.dtsi
create mode 100644 arch/arm/dts/stm32mp15xx-dhcor-avenger96.dts
rename arch/arm/dts/{stm32mp157a-avenger96-u-boot.dtsi => stm32mp15xx-dhcor-u-boot.dtsi} (75%)
create mode 100644 arch/arm/dts/stm32mp15xx-dhcor.dtsiarch/arm/dts/Makefile +2 -1
@@ -884,7 +884,8 @@ dtb-$(CONFIG_STM32MP15x) += \ stm32mp157c-dk2.dtb \ stm32mp157c-ed1.dtb \ stm32mp157c-ev1.dtb \ - stm32mp15xx-dhcom-pdk2.dtb + stm32mp15xx-dhcom-pdk2.dtb \ + stm32mp15xx-dhcor-avenger96.dtb dtb-$(CONFIG_SOC_K3_AM6) += k3-am654-base-board.dtb k3-am654-r5-base-board.dtb dtb-$(CONFIG_SOC_K3_J721E) += k3-j721e-common-proc-board.dtb \
arch/arm/dts/stm32mp157a-avenger96.dts +2 -419
424 lines not on this page · raw message
arch/arm/dts/stm32mp15xx-dhcor-avenger96-u-boot.dtsi new file +80 -0
80 lines not on this page · raw message
arch/arm/dts/stm32mp15xx-dhcor-avenger96.dts new file +211 -0
211 lines not on this page · raw message
arch/arm/dts/stm32mp15xx-dhcor-u-boot.dtsi renamed from arch/arm/dts/stm32mp157a-avenger96-u-boot.dtsi +5 -74
103 lines not on this page · raw message
arch/arm/dts/stm32mp15xx-dhcor.dtsi new file +231 -0
231 lines not on this page · raw message
doc/board/st/stm32mp1.rst +4 -4
@@ -43,11 +43,11 @@ And the necessary drivers Currently the following boards are supported: - + stm32mp157a-avenger96.dts + stm32mp157a-dk1.dts + stm32mp157c-dk2.dts + stm32mp157c-ed1.dts + stm32mp157c-ev1.dts + + stm32mp15xx-dhcor-avenger96.dts Boot Sequences --------------
@@ -145,9 +145,9 @@ the supported device trees for STM32MP15x are: + stm32mp157c-dk2 -+ avenger96: Avenger96 board from Arrow Electronics ++ avenger96: Avenger96 board from Arrow Electronics based on DH Elec. DHCOR SoM - + stm32mp157a-avenger96 + + stm32mp15xx-dhcor-avenger96 Build Procedure ---------------
@@ -229,7 +229,7 @@ Build Procedure # export KBUILD_OUTPUT=stm32mp15_basic # make stm32mp15_basic_defconfig - # make DEVICE_TREE=stm32mp157a-avenger96 all + # make DEVICE_TREE=stm32mp15xx-dhcor-avenger96 all 6. Output files
Signature · 1 line
2.25.1
PATCH v3 14/14 ARM: dts: stm32: Split AV96 into DHCOR SoM and AV96 board
Dear Marek,
Quoted 13 lines
From: Marek Vasut <[email protected]>
Sent: mardi 31 mars 2020 19:52The AV96 is in fact an assembly of DH Electronics DHCOR SoM on top of an AV96 reference board. Split the DTs to reflect that and make sure to DHCOR SoM can be reused on other boards easily.
It is also highly recommended to configure the board for the DHCOM make
stm32mp15_dhcom_basic_defconfigmake DEVICE_TREE=stm32mp15xx-dhcor- avenger96
as that permits reusing the board code for the DH components, like accessing and reading out the ethernet MAC from EEPROM.
Quoted 8 lines
Signed-off-by Marek Vasut <[email protected]>Cc Manivannan Sadhasivam <[email protected]>Cc Patrick Delaunay <[email protected]>Cc Patrice Chotard <[email protected]>---
V2: No change
V3: No change
---
Applied to u-boot-stm/master, with few modifications:
arch/arm/dts/stm32mp15xx-dhcor-u-boot.dtsi
-
-&v3v3 {
- regulator-always-on;
-};
Not needed as already present in arch/arm/dts/stm32mp15xx-dhcor.dtsi
I solve rebase conflict on arch/arm/dts/stm32mp15xx-dhcor.dtsi
- reg = <0x58003000 0x1000>, <0x70000000 0x4000000>;
+ reg = <0x58003000 0x1000>, <0x70000000 0x200000>;
And issue after kernel device tree alignment :
arch/arm/dts/stm32mp15xx-dhcor-avenger96-u-boot.dtsi
&sdmmc1_b4_pins_a {
u-boot,dm-spl;
- pins {
+ pins1 {
+ u-boot,dm-spl;
+ };
+ pins2 {
u-boot,dm-spl;
};
};
Regards
Patrick
PATCH v3 14/14 ARM: dts: stm32: Split AV96 into DHCOR SoM and AV96 board
Quoted 28 lines · Patrick DELAUNAY
Dear Marek,
Quoted 13 lines
From: Marek Vasut <[email protected]>
Sent: mardi 31 mars 2020 19:52The AV96 is in fact an assembly of DH Electronics DHCOR SoM on top of an AV96 reference board. Split the DTs to reflect that and make sure to DHCOR SoM can be reused on other boards easily.
It is also highly recommended to configure the board for the DHCOM make
stm32mp15_dhcom_basic_defconfigmake DEVICE_TREE=stm32mp15xx-dhcor- avenger96
as that permits reusing the board code for the DH components, like accessing and reading out the ethernet MAC from EEPROM.Quoted 8 lines
Signed-off-by Marek Vasut <[email protected]>Cc Manivannan Sadhasivam <[email protected]>Cc Patrick Delaunay <[email protected]>Cc Patrice Chotard <[email protected]>---
V2: No change
V3: No change
---Applied to u-boot-stm/master, with few modifications:
arch/arm/dts/stm32mp15xx-dhcor-u-boot.dtsi
arch/arm/dts/stm32mp157a-avenger96.dts line 93
Quoted 4 lines
arch/arm/dts/stm32mp157a-avenger96.dts +0
-4- -&v3v3 { - regulator-always-on; -};
It seems this patch was applied incorrectly and the 1V8 and 3V3 differentiation between the SoMs is completely missing from mainline. I will send a subsequent patch to correct that.
PATCH v3 14/14 ARM: dts: stm32: Split AV96 into DHCOR SoM and AV96 board
Dear Marek,
Quoted 34 lines
From: Marek Vasut <[email protected]>
Sent: lundi 27 avril 2020 12:30Quoted 30 lines · Patrick DELAUNAY
Dear Marek,
Quoted 15 lines
From: Marek Vasut <[email protected]>
Sent: mardi 31 mars 2020 19:52The AV96 is in fact an assembly of DH Electronics DHCOR SoM on top of an
AV96 reference board. Split the DTs to reflect that and make sure to DHCOR SoM can be reused on other boards easily.It is also highly recommended to configure the board for the DHCOM make
stm32mp15_dhcom_basic_defconfigmake DEVICE_TREE=stm32mp15xx-dhcor-
avenger96
as that permits reusing the board code for the DH components, like accessing and reading out the ethernet MAC from EEPROM.Quoted 8 lines
Signed-off-by Marek Vasut <[email protected]>Cc Manivannan Sadhasivam <[email protected]>Cc Patrick Delaunay <[email protected]>Cc Patrice Chotard <[email protected]>---
V2: No change
V3: No change
---Applied to u-boot-stm/master, with few modifications:
arch/arm/dts/stm32mp15xx-dhcor-u-boot.dtsi
Quoted 8 lines
arch/arm/dts/stm32mp157a-avenger96.dts line 93
Quoted 4 lines
arch/arm/dts/stm32mp157a-avenger96.dts +0
-4- -&v3v3 { - regulator-always-on; -};It seems this patch was applied incorrectly and the 1V8 and 3V3 differentiation between the SoMs is completely missing from mainline. I will send a subsequent patch to correct that.
Strange, I don't remember conflict on this patch....
And 1.8V was present for the dhcor.dtsi in initial patch, I miss something I thing when I apply the patch:
+++ b/arch/arm/dts/stm32mp15xx-dhcor.dtsi
@@ -0,0 +1,231 @@
+
+ /* Enpirion EP3A8LQI U2 on the DHCOR */
+ vdd_io: regulator-buck-io {
+ compatible = "regulator-fixed";
+ regulator-name = "buck-io";
+ regulator-min-microvolt = <1800000>;
+ regulator-max-microvolt = <1800000>;
+ regulator-always-on;
+ regulator-boot-on;
+ vin-supply = <&vdd>;
+ };
[...]
+
+&pwr_regulators {
+ vdd-supply = <&vdd_io>;
+ vdd_3v3_usbfs-supply = <&vdd_usb>;
+};
Anyway thanks to the check and the correction.
Regards
Patrick
PATCH v3 14/14 ARM: dts: stm32: Split AV96 into DHCOR SoM and AV96 board
Quoted 70 lines · Patrick DELAUNAY
Dear Marek,
Quoted 34 lines
From: Marek Vasut <[email protected]>
Sent: lundi 27 avril 2020 12:30Quoted 30 lines · Patrick DELAUNAY
Dear Marek,
Quoted 15 lines
From: Marek Vasut <[email protected]>
Sent: mardi 31 mars 2020 19:52The AV96 is in fact an assembly of DH Electronics DHCOR SoM on top of an
AV96 reference board. Split the DTs to reflect that and make sure to DHCOR SoM can be reused on other boards easily.It is also highly recommended to configure the board for the DHCOM make
stm32mp15_dhcom_basic_defconfigmake DEVICE_TREE=stm32mp15xx-dhcor-
avenger96
as that permits reusing the board code for the DH components, like accessing and reading out the ethernet MAC from EEPROM.Quoted 8 lines
Signed-off-by Marek Vasut <[email protected]>Cc Manivannan Sadhasivam <[email protected]>Cc Patrick Delaunay <[email protected]>Cc Patrice Chotard <[email protected]>---
V2: No change
V3: No change
---Applied to u-boot-stm/master, with few modifications:
arch/arm/dts/stm32mp15xx-dhcor-u-boot.dtsiQuoted 8 lines
arch/arm/dts/stm32mp157a-avenger96.dts line 93
Quoted 4 lines
arch/arm/dts/stm32mp157a-avenger96.dts +0
-4- -&v3v3 { - regulator-always-on; -};It seems this patch was applied incorrectly and the 1V8 and 3V3 differentiation between the SoMs is completely missing from mainline. I will send a subsequent patch to correct that.
Strange, I don't remember conflict on this patch....
And 1.8V was present for the
dhcor.dtsiin initial patch, I miss something I thing when I apply the patch:+++
b/arch/arm/dts/stm32mp15xx-dhcor.dtsiarch/arm/dts/stm32mp15xx-dhcor.dtsi new file +16
-0@@ -0,0 +1,231 @@ + + /* Enpirion EP3A8LQI U2 on the DHCOR */ + vdd_io: regulator-buck-io { + compatible = "regulator-fixed"; + regulator-name = "buck-io"; + regulator-min-microvolt = <1800000>; + regulator-max-microvolt = <1800000>; + regulator-always-on; + regulator-boot-on; + vin-supply = <&vdd>; + }; + +&pwr_regulators { + vdd-supply = <&vdd_io>; + vdd_3v3_usbfs-supply = <&vdd_usb>; +};
Quoted 1 line
Anyway thanks to the check and the correction.
The thing is, there are two versions of the SoM that's soldered on the bottom of the AV96. The one on AV96 has that regulator populated, but that is optional, so for other devices you can just opt for 3V3 IO and the regulator won't be there. That's why there are the two DTs (io1v8 and io3v3).