September 2024 - U-Boot

[PATCH] clk: renesas: Fix missing unbind on driver look up failure
by Marek Vasut 05 Sep '24

05 Sep '24

In case lists_driver_lookup_name("rst_gen3") fails, the clk_gen3 driver has to be unbound before erroring out. Reorder the function such that both clk_gen3 and rst_gen3 drivers are looked up first, if either look up fails then the function errors out right away. Second, the clk_gen3 is bound, if binding fails, the function still errors out right away because there is nothing to undo after either lists_driver_lookup_name() call. Finally, rst_gen3 is bound and if this binding fails, then that is the only place where clk_gen3 has to be unbound before erroring out. Signed-off-by: Marek Vasut <marek.vasut+renesas(a)mailbox.org> --- drivers/clk/renesas/clk-rcar-gen3.c | 19 ++++++++++--------- 1 file changed, 10 insertions(+), 9 deletions(-) diff --git a/drivers/clk/renesas/clk-rcar-gen3.c b/drivers/clk/renesas/clk-rcar-gen3.c index b84024266f4..ac05788ccc5 100644 --- a/drivers/clk/renesas/clk-rcar-gen3.c +++ b/drivers/clk/renesas/clk-rcar-gen3.c @@ -579,23 +579,24 @@ int gen3_cpg_bind(struct udevice *parent) struct cpg_mssr_info *info = (struct cpg_mssr_info *)dev_get_driver_data(parent); struct udevice *cdev, *rdev; - struct driver *drv; + struct driver *cdrv, *rdrv; int ret; - drv = lists_driver_lookup_name("clk_gen3"); - if (!drv) + cdrv = lists_driver_lookup_name("clk_gen3"); + if (!cdrv) return -ENOENT; - ret = device_bind_with_driver_data(parent, drv, "clk_gen3", (ulong)info, + + rdrv = lists_driver_lookup_name("rst_gen3"); + if (!rdrv) + return -ENOENT; + + ret = device_bind_with_driver_data(parent, cdrv, "clk_gen3", (ulong)info, dev_ofnode(parent), &cdev); if (ret) return ret; - drv = lists_driver_lookup_name("rst_gen3"); - if (!drv) - return -ENOENT; - - ret = device_bind_with_driver_data(parent, drv, "rst_gen3", (ulong)cdev, + ret = device_bind_with_driver_data(parent, rdrv, "rst_gen3", (ulong)cdev, dev_ofnode(parent), &rdev); if (ret) device_unbind(cdev); -- 2.45.2

1 0

[PATCH 00/20] arm: Initial support for Analog Devices SC5xx boards
by Oliver Gaskell via B4 Relay 04 Sep '24

04 Sep '24

ADSP-SC5xx is a series of ARM-based DSPs. This comprises the armv7 based SC57x, SC58x and SC594 series, and the armv8 based SC598. This patch series includes configurations, init code, and minimal DTs to enable Analog Devices' evaluation boards for these SoCs to boot through SPL and into U-Boot Proper, as well as devicetree schemas for the added DTs. This patch series depends on ("arm: Add Analog Devices SC5xx Machine Type") (https://lists.denx.de/pipermail/u-boot/2024-April/552043.html) Signed-off-by: Oliver Gaskell <Oliver.Gaskell(a)analog.com> --- Oliver Gaskell (20): arm: dts: Support SC573-EZKIT arm: dts: Support SC584-EZKIT arm: dts: Support SC589-MINI arm: dts: Support SC589-EZKIT arm: dts: Support SC594-SOM-EZKIT arm: dts: Support SC594-SOM-EZLITE arm: dts: Support SC598-SOM-EZKIT arm: dts: Support SC598-SOM-EZLITE dt-bindings: arm: Add SC5xx Series binding dt-bindings: clock: Add SC5xx clock tree bindings dt-bindings: timer: Add SC5xx Timer bindings arm: mach-sc5xx: clean up Kconfig arm: SC598-SOM-EZKIT initial support arm: SC598-SOM-EZLITE initial support arm: SC594-SOM-EZKIT initial support arm: SC594-SOM-EZLITE initial support arm: SC584-EZKIT initial support arm: SC589-EZKIT initial support arm: SC589-MINI initial support arm: SC573-EZKIT initial support MAINTAINERS | 7 + arch/arm/Kconfig | 14 ++ arch/arm/dts/Makefile | 9 + arch/arm/dts/sc573-ezkit.dts | 13 ++ arch/arm/dts/sc57x.dtsi | 21 +++ arch/arm/dts/sc584-ezkit.dts | 13 ++ arch/arm/dts/sc589-ezkit.dts | 13 ++ arch/arm/dts/sc589-mini.dts | 13 ++ arch/arm/dts/sc58x.dtsi | 23 +++ arch/arm/dts/sc594-som-ezkit.dts | 13 ++ arch/arm/dts/sc594-som-ezlite.dts | 13 ++ arch/arm/dts/sc594-som.dtsi | 19 +++ arch/arm/dts/sc598-som-ezkit.dts | 13 ++ arch/arm/dts/sc598-som-ezlite.dts | 13 ++ arch/arm/dts/sc598-som.dtsi | 31 ++++ arch/arm/dts/sc5xx.dtsi | 54 ++++++ arch/arm/mach-sc5xx/Kconfig | 134 ++++++++++++--- arch/arm/mach-sc5xx/Makefile | 4 + arch/arm/mach-sc5xx/sc57x-spl.c | 26 +++ arch/arm/mach-sc5xx/sc58x-spl.c | 26 +++ arch/arm/mach-sc5xx/sc59x-spl.c | 26 +++ arch/arm/mach-sc5xx/sc59x_64-spl.c | 26 +++ arch/arm/mach-sc5xx/sc59x_64.c | 25 +++ arch/arm/mach-sc5xx/soc.c | 44 +++++ board/adi/sc573-ezkit/Kconfig | 160 ++++++++++++++++++ board/adi/sc573-ezkit/sc573-ezkit.env | 13 ++ board/adi/sc584-ezkit/Kconfig | 160 ++++++++++++++++++ board/adi/sc584-ezkit/sc584-ezkit.env | 13 ++ board/adi/sc589-ezkit/Kconfig | 160 ++++++++++++++++++ board/adi/sc589-ezkit/sc589-ezkit.env | 13 ++ board/adi/sc589-mini/Kconfig | 160 ++++++++++++++++++ board/adi/sc589-mini/sc589-mini.env | 13 ++ board/adi/sc594-som-ezkit/Kconfig | 181 +++++++++++++++++++++ board/adi/sc594-som-ezkit/sc594-som-ezkit.env | 13 ++ board/adi/sc594-som-ezlite/Kconfig | 181 +++++++++++++++++++++ board/adi/sc594-som-ezlite/sc594-som-ezlite.env | 13 ++ board/adi/sc598-som-ezkit/Kconfig | 150 +++++++++++++++++ board/adi/sc598-som-ezkit/sc598-som-ezkit.env | 13 ++ board/adi/sc598-som-ezlite/Kconfig | 150 +++++++++++++++++ board/adi/sc598-som-ezlite/sc598-som-ezlite.env | 13 ++ doc/device-tree-bindings/arm/adi/adi,sc5xx.yaml | 46 ++++++ .../clock/adi,sc5xx-clocks.yaml | 112 +++++++++++++ .../timer/adi,sc5xx-gptimer.yaml | 42 +++++ include/configs/sc573-ezkit.h | 18 ++ include/configs/sc584-ezkit.h | 18 ++ include/configs/sc589.h | 19 +++ include/configs/sc594-som.h | 19 +++ include/configs/sc598-som.h | 23 +++ 48 files changed, 2274 insertions(+), 22 deletions(-) --- base-commit: 34377bc0b0dc446853c7831efaccef8b50c22841 change-id: 20240821-sc5xx-upstreaming-patches-c048ba9f979a Best regards, -- Oliver Gaskell <Oliver.Gaskell(a)analog.com>

3 27

Please pull u-boot-marvell/master
by Stefan Roese 04 Sep '24

04 Sep '24

Hi Tom, I know it's late in the release process. But please consider pulling these Turris Omnia DDR related patches / fixes, which Marek would like to see in the upcoming release: ---------------------------------------------------------------- - Turris Omnia DDR speed change with reset button (Marek) ---------------------------------------------------------------- Here the Azure build, without any issues: https://dev.azure.com/sr0718/u-boot/_build/results?buildId=374&view=results Thanks, Stefan The following changes since commit 1312faac5f52d27cfb45dfe1a5a93a2944ca5c21: Prepare v2024.10-rc4 (2024-09-02 14:07:53 -0600) are available in the Git repository at: git@source.denx.de:u-boot/custodians/u-boot-marvell.git for you to fetch changes up to bc8664c9f31195a0249dd4eb0794f4a8625ad864: arm: mvebu: turris_omnia: Switch DDR speed to 1333H when reset 9 is selected (2024-09-03 14:24:32 +0200) ---------------------------------------------------------------- Marek Behún (3): arm: mvebu: turris_omnia: Rename variable holding EEPROM udevice arm: mvebu: turris_omnia: Use the i2c_eeprom misc driver for EEPROM reading in U-Boot proper arm: mvebu: turris_omnia: Switch DDR speed to 1333H when reset 9 is selected arch/arm/mach-mvebu/Kconfig | 1 + board/CZ.NIC/turris_omnia/turris_omnia.c | 101 ++++++++++++++++++++++++++----- configs/turris_omnia_defconfig | 1 - 3 files changed, 86 insertions(+), 17 deletions(-)

2 1

[PATCH V3 0/9] Add Anbernic RG35XX-2024
by Chris Morgan 04 Sep '24

04 Sep '24

From: Chris Morgan <macromorgan(a)hotmail.com> Add support for the Anbernic RG35XX-2024. This device is a handheld gaming console (series) based around the Anbernic H700 SoC. It comes with 1GB of RAM and multiple face buttons for input. Add support for the simplest model for now, and then iterate later to add board detection logic, support the entire device lineup, and move to using the mainline device tree. Changes since V2: - Moved back to using the U-Boot specific device tree, for now. - Changed the cherry picked device tree commits to update both the U-Boot device tree and the mainline device tree, so either could be used. - Made the DRAM PHY pinconf compile time selectable instead of runtime selectable to decrease SPL memory usage. Changes since V1: - Removed DDR3 and LPDDR3 patches from Jernej Skrabec, as they could not be tested by me and were not required to bring up this device. - Changed the default TPR6 parameter based on a suggestion from Mikhail Kalashnikov. - Cherry picked 2 commits from linux-next and converted devicetree to upstream. Chris Morgan (5): sunxi: H616: DRAM: Add alternative pin mapping sunxi: Correct TPR6 parameter for H616 DRAM driver arm64: dts: allwinner: h616: Add r_i2c pinctrl nodes arm64: dts: allwinner: h616: Change RG35XX PMIC to r_i2c sunxi: Add support for Anbernic RG35XX-2024 Jernej Skrabec (4): sunxi: H616: dram: LPDDR4: adjust settings sunxi: H616: DRAM: Adjust configuration procedure sunxi: H616: DRAM: Adjust size scan procedure sunxi: H616: dram: Update mbus priorities arch/arm/dts/sun50i-h616.dtsi | 2 + .../dts/sun50i-h700-anbernic-rg35xx-2024.dts | 6 +- arch/arm/mach-sunxi/Kconfig | 19 ++- arch/arm/mach-sunxi/dram_sun50i_h616.c | 137 ++++++++++++------ .../dram_timings/h616_lpddr4_2133.c | 2 +- board/sunxi/MAINTAINERS | 5 + configs/anbernic_rg35xx_h700_defconfig | 28 ++++ .../src/arm64/allwinner/sun50i-h616.dtsi | 2 + .../sun50i-h700-anbernic-rg35xx-2024.dts | 6 +- 9 files changed, 154 insertions(+), 53 deletions(-) create mode 100644 configs/anbernic_rg35xx_h700_defconfig -- 2.34.1

3 13

[PATCH u-boot-marvell 0/3] Turris Omnia DDR speed change with reset button (for 2024.10)
by Marek Behún 04 Sep '24

04 Sep '24

Hello Stefan, this series adds the ability to select 1333H DDR speed mode for DDR training (instead of the default 1600K) with the factory reset button. If possible, I would like if this series went for U-Boot 2024.10 release. The heavy stuff to support DDR speed change is already in, this just adds the ability to do it with reset button instead of console command (which is inconvenient for most users). On Omnia, the user can select factory reset type by pressing the reset button on the rear panel and holding it. The front LEDs light on in sequence, representing a counter. Every ~2.5 second another LED lights on. When only the first LED is on, the board will do a standard reboot. Then different kinds of factory reboot will happen for reset types 1 to 7 (when 1, 2, ... or 7 LEDs are on) [1]. This series adds another reset type, number 10, which will make the board boot in 1333H DDR speed mode, and also write this setting to EEPROM, so that subsequent boots will also train in 1333H mode. The user can set back the default mode with the `eeprom` U-Boot command, or with the `omnia-eeprom` Linux command. Marek [1] https://docs.turris.cz/hw/omnia/rescue-modes/ Marek Behún (3): arm: mvebu: turris_omnia: Rename variable holding EEPROM udevice arm: mvebu: turris_omnia: Use the i2c_eeprom misc driver for EEPROM reading in U-Boot proper arm: mvebu: turris_omnia: Switch DDR speed to 1333H when reset 9 is selected arch/arm/mach-mvebu/Kconfig | 1 + board/CZ.NIC/turris_omnia/turris_omnia.c | 101 +++++++++++++++++++---- configs/turris_omnia_defconfig | 1 - 3 files changed, 86 insertions(+), 17 deletions(-) -- 2.44.2

2 9

[PATCH v5 0/6] FUSB302 USB-C controller support
by Sebastian Reichel 04 Sep '24

04 Sep '24

Hi, On ROCK 5B power is usually supplied via it's USB-C port. This port has the data lines connected to RK3588, VBUS connected to the input regulator and CC pins connected to FUSB302. FUSB302 is a USB-C controller, which can be accessed via I2C from RK3588. The USB-C controller is needed to figure out the USB-C cable orientation, but also to do USB PD communication. Thus it would be great to enable support for it in the operating system. But the USB-PD specification requires, that a device reacts to USB-PD messages send by the power-supply within around 5 seconds. If that does not happen the power-supply assumes, that the device does not support USB-PD. If a device later starts sending USB-PD messages it is considered an error, which is solved by doing a hard reset. A USB-PD hard reset means, that all supply voltages are removed for a short period of time. For boards, which are solely powered through their USB-C port, like the Radxa Rock 5B, this results in an machine reset. This is currently worked around by not describing the FUSB302 in the kernel DT, so nothing will ever speak USB-PD on the Rock 5B. This means 1. the USB-C port cannot be used at all 2. the board will be running via fallback supply, which provides limited power capabilities In order to avoid the hard reset, this adds FUSB302 support to U-Boot, so that we react to the power-supply's queries in time. The code, which is originally from the Linux kernel, consists of two parts: 1. the tcpm state machine, which implements the Type C port manager state machine as described in the USB PD specification 2. the fusb302 driver, which knows about specific registers Especially the first part has been heavily modified compared to the kernel, which makes use of multiple delayed works and threads. For this I used a priorly ported version from Rockchip, removed their hacks and any states not necessary in U-Boot (e.g. audio accessory support). This version has been tested on Radxa Rock 5B using the open source TF-A (patches recently got merged into master branch) using the following power supplies: * non PD capable (reports 5V 0A) * RavPower 90W * UGREEN 100W * Anker 45W * RavPower PB Changes since PATCHv4: * Rebase to latest master (v2024.10-rc4) * Fix regression introduced in PATCHv3: The exit code for fusb302_set_vconn() when vconn stays the same should be 0 instead of undefined * Fix regression introduced in PATCHv3: tcpm_set_polarity() needs to update port->polarity when there is no driver specific set_polarity function Changes since PATCHv3: * Rebase to latest master (57949a99b7bd) * Rework autoprobing; tcpm_post_bind tries to check if a probe is needed based on DT properties and then sets the DM_FLAG_PROBE_AFTER_BIND flag. If this accidently increases boot time on some boards they are probably missing a 'self-powered;' flag in their USB-C connector node. * Rock 5B ft_board_setup(): check for CONFIG_TYPEC_FUSB302 instead of CONFIG_MISC_INIT_R to reflect the above change * Drop adding 'CONFIG_MISC_INIT_R' to rock5b-rk3588_defconfig because of the reworked autoprobing * Log an error message if tcpm_send_queued_message() exits early * Add R-b for the rockchip specific changes from Kever Yang * Increase the timeout in tcpm_pd_transmit(), which seems to be the issue Sören Moch ran into. I could not really reproduce it. Please test if this helps with your supply (and share the logs if it does not help). I noticed this version has issues with an old PD capable powerbank I own. I'm still looking into that but wanted to share a new version so that the other changes can get reviewed. Changes since PATCHv2: * Rebase to latest master (a70d991212c9) * Drop Wang Jie from Cc, since his address bounces * I did not add a test suite. I fully agree, that it would be nice to have one, but it is also a lot of work. Just emulating a FUSB302 is not enough, we also need to emulate the remote USB-PD device to make this useful. Without that we would only test what happens when a non-responsive USB-PD supply is attached and that is the boring, very simple path in this code. On the other hand implementing an emulated USB-C PD remote side more or less requires implementing another big state machine. * Add documentation for the 'tcpm' command (Simon Glass) * I also got asked to add more documentation for the feature in general. I'm not really sure what is needed. At least Tim Harvey managed to use it considering he got stusb4500 working on top of PATCHV2. If there are more specific requests I can write something. * Just like PATCHv1, PATCHv2 received a 'Tested-by: Soeren Moch <smoch(a)web.de>', which I did not take over. This time the changes are a lot smaller, but the different handling of timeouts might have broken something. (On my end things still behave correctly of course, otherwise I wouldn't have sent this series) * Address feedback from Marek Vasut - TCPM core - drop useless "failure" function - replace printf with log in TCPM command code - invert some conditions to reduce indent - use read_poll_timeout() in tcpm_pd_transmit() - did not add any more 'static' keywords to functions, the non-static TCPM functions are exported and called from the fusb302 driver. - add a safety guard to tcpm_send_queued_message(), which stops the loop if more than 100 messages are send directly after each other. This should never happen anyways, but it's better to exit cleanly if the code runs astray. - FUSB302 driver - Drop default initialization for most local ret variables - Make some local variables constant - Make set_polarity callback optional on the uclass level and drop the empty FUSB302 implementation - Change buffer initialization in send message function - Create define for max. message length magic value - Rock 5B board code - Add comment why misc_init_r explicitly initializes PD - Split defconfig and DT into separate patches (Kever Yang) Changes since PATCHv1: * tcpm: split uclass specific code to tcpm-uclass * tcpm_print_info: move printing part to cmd/tcpm.c * tcpm_print_info: report more information - PD revision - Cable orientation - Power role - Data role - Print TCPM state based on connection status * tcpm: use "struct udevice *dev" instead of "struct tcpm_port *port" as function argument in most places and drop dev from the tcpm_port struct * tcpm: avoid useless kzalloc + copy + free for incoming events * tcpm: use dev_get_uclass_plat() for tcpm_port * tcpm: run tcpm_port_init() and tcpm_poll_event() from UCLASS post_probe() * tcpm/fusb302: get rid of tcpc_dev by using dm_tcpm_ops() for the function pointers and introducing get_connector_node() for the ofnode * fusb302: use "struct udevice *dev" instead of "struct fusb302_chip *chip" as function argument and drop dev from the fusb302_chip struct * fusb302: drop multiple unused members from fusb302_chip * fusb302: directly use udelay instead of usleep_range define * fusb302: use fusb302_ prefix for all functions. Afterwards tcpm_ prefix is only used for the tcpm code itself * fusb302: move fusb302_poll_event() to avoid forward defintion * fusb302: drop probe function * fusb302: drop unused LOG_BUFFER defines * roughly 20% of all lines of the series changed between v1 and v2, so I did not collect the Tested-by from Soeren Moch Greetings, -- Sebastian Sebastian Reichel (6): usb: tcpm: add core framework usb: tcpm: fusb302: add driver board: rock5b-rk3588: enable USB-C in operating system rockchip: rk3588-rock-5b: Add USB-C controller to u-boot.dtsi rockchip: rock5b-rk3588: Enable USB-C PD support MAINTAINERS: add TCPM section MAINTAINERS | 9 + Makefile | 1 + arch/arm/dts/rk3588-rock-5b-u-boot.dtsi | 28 + board/radxa/rock5b-rk3588/Makefile | 6 + board/radxa/rock5b-rk3588/rock5b-rk3588.c | 16 + cmd/Kconfig | 7 + cmd/Makefile | 1 + cmd/tcpm.c | 136 ++ configs/rock5b-rk3588_defconfig | 4 + doc/usage/cmd/tcpm.rst | 66 + doc/usage/index.rst | 1 + drivers/usb/Kconfig | 2 + drivers/usb/tcpm/Kconfig | 16 + drivers/usb/tcpm/Makefile | 4 + drivers/usb/tcpm/fusb302.c | 1323 ++++++++++++ drivers/usb/tcpm/fusb302_reg.h | 177 ++ drivers/usb/tcpm/tcpm-internal.h | 173 ++ drivers/usb/tcpm/tcpm-uclass.c | 145 ++ drivers/usb/tcpm/tcpm.c | 2288 +++++++++++++++++++++ include/dm/uclass-id.h | 1 + include/usb/pd.h | 516 +++++ include/usb/tcpm.h | 99 + 22 files changed, 5019 insertions(+) create mode 100644 board/radxa/rock5b-rk3588/Makefile create mode 100644 board/radxa/rock5b-rk3588/rock5b-rk3588.c create mode 100644 cmd/tcpm.c create mode 100644 doc/usage/cmd/tcpm.rst create mode 100644 drivers/usb/tcpm/Kconfig create mode 100644 drivers/usb/tcpm/Makefile create mode 100644 drivers/usb/tcpm/fusb302.c create mode 100644 drivers/usb/tcpm/fusb302_reg.h create mode 100644 drivers/usb/tcpm/tcpm-internal.h create mode 100644 drivers/usb/tcpm/tcpm-uclass.c create mode 100644 drivers/usb/tcpm/tcpm.c create mode 100644 include/usb/pd.h create mode 100644 include/usb/tcpm.h -- 2.45.2

3 10

[PATCH v1 0/3] Arm: npcm: modify npcm8xx boot setting
by Jim Liu 04 Sep '24

04 Sep '24

Modify npcm8xx new boot design. Correct memory setting and set gpio default value. Jim Liu (2): board: arbel: correct the dram bank size configs: arbel_evb: change env offset and boot address Stanley Chu (1): pinctrl: npcm8xx: clear all gpio events board/nuvoton/arbel_evb/arbel_evb.c | 8 ++++++-- configs/arbel_evb_defconfig | 4 ++-- drivers/pinctrl/nuvoton/pinctrl-npcm8xx.c | 18 ++++++++++++++++++ include/configs/arbel.h | 2 +- 4 files changed, 27 insertions(+), 5 deletions(-) -- 2.34.1

1 0

[PATCH 0/2] fdtoverlay related changes
by Prasad Kummari 03 Sep '24

03 Sep '24

Prasad Kummari (2): kbuild: Add generic rule to apply fdtoverlay arm64: zynqmp: Remove overlays and add new dtb entries for ZynqMP arch/arm/dts/Makefile | 78 +++++++++++++++++++++++++++++++++--------- scripts/Kbuild.include | 4 +++ scripts/Makefile.build | 1 + scripts/Makefile.lib | 27 +++++++++++++++ 4 files changed, 93 insertions(+), 17 deletions(-) -- 2.25.1

2 3

[PATCH v4 0/5] net: dwc_eth_qos: Add glue driver for Intel MAC
by Philip Oberfichtner 03 Sep '24

03 Sep '24

This patch series implements the dwc_eth_qos glue driver for Intel SOCs. Before doing that, a few general adaptions to the dwc_eth_qos.c main driver are required. Most notably, the preparation for PCI based driver instances, which do not necessarily use a device tree. Changes in V4: - Fix printf format string Changes in V3: - Replace mfence() with mb() - Clean-up eqos_get_base_addr() - Several style fixes for dwc_eth_qos_intel Philip Oberfichtner (5): x86: provide mb() macro net: dwc_eth_qos: Fix header to be self-contained net: dwc_eth_qos: Adapt probe() for PCI devices net: dwc_eth_qos: Implement bind() for PCI devices net: dwc_eth_qos: Add glue driver for Intel MAC arch/x86/cpu/mp_init.c | 10 +- arch/x86/include/asm/cpu.h | 5 - arch/x86/include/asm/io.h | 1 + drivers/net/Kconfig | 7 + drivers/net/Makefile | 1 + drivers/net/dwc_eth_qos.c | 83 +++++- drivers/net/dwc_eth_qos.h | 8 +- drivers/net/dwc_eth_qos_imx.c | 6 + drivers/net/dwc_eth_qos_intel.c | 449 +++++++++++++++++++++++++++++ drivers/net/dwc_eth_qos_intel.h | 57 ++++ drivers/net/dwc_eth_qos_qcom.c | 6 + drivers/net/dwc_eth_qos_rockchip.c | 6 + drivers/net/dwc_eth_qos_starfive.c | 6 + drivers/net/dwc_eth_qos_stm32.c | 6 + include/pci_ids.h | 9 + 15 files changed, 641 insertions(+), 19 deletions(-) create mode 100644 drivers/net/dwc_eth_qos_intel.c create mode 100644 drivers/net/dwc_eth_qos_intel.h -- 2.39.2

3 8

[PATCH v4 0/6] FUSB302 USB-C controller support
by Sebastian Reichel 03 Sep '24

03 Sep '24

Hi, On ROCK 5B power is usually supplied via it's USB-C port. This port has the data lines connected to RK3588, VBUS connected to the input regulator and CC pins connected to FUSB302. FUSB302 is a USB-C controller, which can be accessed via I2C from RK3588. The USB-C controller is needed to figure out the USB-C cable orientation, but also to do USB PD communication. Thus it would be great to enable support for it in the operating system. But the USB-PD specification requires, that a device reacts to USB-PD messages send by the power-supply within around 5 seconds. If that does not happen the power-supply assumes, that the device does not support USB-PD. If a device later starts sending USB-PD messages it is considered an error, which is solved by doing a hard reset. A USB-PD hard reset means, that all supply voltages are removed for a short period of time. For boards, which are solely powered through their USB-C port, like the Radxa Rock 5B, this results in an machine reset. This is currently worked around by not describing the FUSB302 in the kernel DT, so nothing will ever speak USB-PD on the Rock 5B. This means 1. the USB-C port cannot be used at all 2. the board will be running via fallback supply, which provides limited power capabilities In order to avoid the hard reset, this adds FUSB302 support to U-Boot, so that we react to the power-supply's queries in time. The code, which is originally from the Linux kernel, consists of two parts: 1. the tcpm state machine, which implements the Type C port manager state machine as described in the USB PD specification 2. the fusb302 driver, which knows about specific registers Especially the first part has been heavily modified compared to the kernel, which makes use of multiple delayed works and threads. For this I used a priorly ported version from Rockchip, removed their hacks and any states not necessary in U-Boot (e.g. audio accessory support). This version has been tested on Radxa Rock 5B using the open source TF-A (patches recently got merged into master branch) using the following power supplies: * non PD capable (reports 5V 0A) * RavPower 90W (ok) * UGREEN 100W (ok) * Anker 45W (ok) * RavPower PB (hard resets in U-Boot, but succeeds at some point, I still need to investigate) Changes since PATCHv3: * Rebase to latest master (57949a99b7bd) * Rework autoprobing; tcpm_post_bind tries to check if a probe is needed based on DT properties and then sets the DM_FLAG_PROBE_AFTER_BIND flag. If this accidently increases boot time on some boards they are probably missing a 'self-powered;' flag in their USB-C connector node. * Rock 5B ft_board_setup(): check for CONFIG_TYPEC_FUSB302 instead of CONFIG_MISC_INIT_R to reflect the above change * Drop adding 'CONFIG_MISC_INIT_R' to rock5b-rk3588_defconfig because of the reworked autoprobing * Log an error message if tcpm_send_queued_message() exits early * Add R-b for the rockchip specific changes from Kever Yang * Increase the timeout in tcpm_pd_transmit(), which seems to be the issue Sören Moch ran into. I could not really reproduce it. Please test if this helps with your supply (and share the logs if it does not help). I noticed this version has issues with an old PD capable powerbank I own. I'm still looking into that but wanted to share a new version so that the other changes can get reviewed. Changes since PATCHv2: * Rebase to latest master (a70d991212c9) * Drop Wang Jie from Cc, since his address bounces * I did not add a test suite. I fully agree, that it would be nice to have one, but it is also a lot of work. Just emulating a FUSB302 is not enough, we also need to emulate the remote USB-PD device to make this useful. Without that we would only test what happens when a non-responsive USB-PD supply is attached and that is the boring, very simple path in this code. On the other hand implementing an emulated USB-C PD remote side more or less requires implementing another big state machine. * Add documentation for the 'tcpm' command (Simon Glass) * I also got asked to add more documentation for the feature in general. I'm not really sure what is needed. At least Tim Harvey managed to use it considering he got stusb4500 working on top of PATCHV2. If there are more specific requests I can write something. * Just like PATCHv1, PATCHv2 received a 'Tested-by: Soeren Moch <smoch(a)web.de>', which I did not take over. This time the changes are a lot smaller, but the different handling of timeouts might have broken something. (On my end things still behave correctly of course, otherwise I wouldn't have sent this series) * Address feedback from Marek Vasut - TCPM core - drop useless "failure" function - replace printf with log in TCPM command code - invert some conditions to reduce indent - use read_poll_timeout() in tcpm_pd_transmit() - did not add any more 'static' keywords to functions, the non-static TCPM functions are exported and called from the fusb302 driver. - add a safety guard to tcpm_send_queued_message(), which stops the loop if more than 100 messages are send directly after each other. This should never happen anyways, but it's better to exit cleanly if the code runs astray. - FUSB302 driver - Drop default initialization for most local ret variables - Make some local variables constant - Make set_polarity callback optional on the uclass level and drop the empty FUSB302 implementation - Change buffer initialization in send message function - Create define for max. message length magic value - Rock 5B board code - Add comment why misc_init_r explicitly initializes PD - Split defconfig and DT into separate patches (Kever Yang) Changes since PATCHv1: * tcpm: split uclass specific code to tcpm-uclass * tcpm_print_info: move printing part to cmd/tcpm.c * tcpm_print_info: report more information - PD revision - Cable orientation - Power role - Data role - Print TCPM state based on connection status * tcpm: use "struct udevice *dev" instead of "struct tcpm_port *port" as function argument in most places and drop dev from the tcpm_port struct * tcpm: avoid useless kzalloc + copy + free for incoming events * tcpm: use dev_get_uclass_plat() for tcpm_port * tcpm: run tcpm_port_init() and tcpm_poll_event() from UCLASS post_probe() * tcpm/fusb302: get rid of tcpc_dev by using dm_tcpm_ops() for the function pointers and introducing get_connector_node() for the ofnode * fusb302: use "struct udevice *dev" instead of "struct fusb302_chip *chip" as function argument and drop dev from the fusb302_chip struct * fusb302: drop multiple unused members from fusb302_chip * fusb302: directly use udelay instead of usleep_range define * fusb302: use fusb302_ prefix for all functions. Afterwards tcpm_ prefix is only used for the tcpm code itself * fusb302: move fusb302_poll_event() to avoid forward defintion * fusb302: drop probe function * fusb302: drop unused LOG_BUFFER defines * roughly 20% of all lines of the series changed between v1 and v2, so I did not collect the Tested-by from Soeren Moch Greetings, -- Sebastian Sebastian Reichel (6): usb: tcpm: add core framework usb: tcpm: fusb302: add driver board: rock5b-rk3588: enable USB-C in operating system rockchip: rk3588-rock-5b: Add USB-C controller to u-boot.dtsi rockchip: rock5b-rk3588: Enable USB-C PD support MAINTAINERS: add TCPM section MAINTAINERS | 9 + Makefile | 1 + arch/arm/dts/rk3588-rock-5b-u-boot.dtsi | 28 + board/radxa/rock5b-rk3588/Makefile | 6 + board/radxa/rock5b-rk3588/rock5b-rk3588.c | 16 + cmd/Kconfig | 7 + cmd/Makefile | 1 + cmd/tcpm.c | 136 ++ configs/rock5b-rk3588_defconfig | 4 + doc/usage/cmd/tcpm.rst | 66 + doc/usage/index.rst | 1 + drivers/usb/Kconfig | 2 + drivers/usb/tcpm/Kconfig | 16 + drivers/usb/tcpm/Makefile | 4 + drivers/usb/tcpm/fusb302.c | 1322 ++++++++++++ drivers/usb/tcpm/fusb302_reg.h | 177 ++ drivers/usb/tcpm/tcpm-internal.h | 173 ++ drivers/usb/tcpm/tcpm-uclass.c | 145 ++ drivers/usb/tcpm/tcpm.c | 2289 +++++++++++++++++++++ include/dm/uclass-id.h | 1 + include/usb/pd.h | 516 +++++ include/usb/tcpm.h | 99 + 22 files changed, 5019 insertions(+) create mode 100644 board/radxa/rock5b-rk3588/Makefile create mode 100644 board/radxa/rock5b-rk3588/rock5b-rk3588.c create mode 100644 cmd/tcpm.c create mode 100644 doc/usage/cmd/tcpm.rst create mode 100644 drivers/usb/tcpm/Kconfig create mode 100644 drivers/usb/tcpm/Makefile create mode 100644 drivers/usb/tcpm/fusb302.c create mode 100644 drivers/usb/tcpm/fusb302_reg.h create mode 100644 drivers/usb/tcpm/tcpm-internal.h create mode 100644 drivers/usb/tcpm/tcpm-uclass.c create mode 100644 drivers/usb/tcpm/tcpm.c create mode 100644 include/usb/pd.h create mode 100644 include/usb/tcpm.h -- 2.45.2

5 13