September 2022 - U-Boot

[PATCH v2 0/7] SPL fixes for STM32F7 MCUs
by Patrice Chotard 26 Sep '22

26 Sep '22

This series is fixing issues in SPL boot mode: - SPL binary size over 0x8000 bytes for STM32F746-disco, STM32F769-DISCO and STM32746G-EVAL boards - fix embedded flash size for STM32F769-DISCO - fix CONFIG_SYS_SPL_ARGS_ADDR for STM32F746-DISCO and STM32746G-EVAL Changes in v2: - Remove SYS_UBOOT_FLAG from stm32f746-disco.h Patrice Chotard (7): configs: stm32f746-disco: Fix SPL boot configs: stm32f769-disco: Fix SPL boot configs: stm32746g-eval: Fix SPL boot configs: stm32f746-disco: Fix CONFIG_SYS_SPL_ARGS_ADDR configs: stm32746g-eval: Fix CONFIG_SYS_SPL_ARGS_ADDR config: stm32f769-disco: Fix internal flash size configs: stm32f746-disco: Remove CONFIG_SYS_UBOOT_START flag arch/arm/include/asm/arch-stm32f7/stm32.h | 6 +++--- configs/stm32746g-eval_spl_defconfig | 9 +++++---- configs/stm32f746-disco_spl_defconfig | 9 +++++---- configs/stm32f769-disco_defconfig | 2 +- configs/stm32f769-disco_spl_defconfig | 9 +++++---- include/configs/stm32f746-disco.h | 1 - 6 files changed, 19 insertions(+), 17 deletions(-) -- 2.25.1

4 22

[PATCH 1/1] sandbox: unblock signal before calling execv()
by Heinrich Schuchardt 26 Sep '22

26 Sep '22

The following faulty behavior was observed. The sandbox configured with CONFIG_SANDBOX_CRASH_RESET=y was invoked with ./u-boot -T -S After executing `exception undefined' the sandbox reboots. When executing `exception undefined' the sandbox exits with SIGSEGV. The expected behavior is that the sandbox should reboot again. If we are relaunching the sandbox in a signal handler, we have to unblock the respective signal before calling execv(). See signal(7) man-page. Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt(a)canonical.com> --- arch/sandbox/cpu/os.c | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) diff --git a/arch/sandbox/cpu/os.c b/arch/sandbox/cpu/os.c index f937991139..2c7aeea070 100644 --- a/arch/sandbox/cpu/os.c +++ b/arch/sandbox/cpu/os.c @@ -1012,8 +1012,24 @@ void *os_find_text_base(void) return base; } +/** + * os_unblock_signals() - unblock all signals + * + * If we are relaunching the sandbox in a signal handler, we have to unblock + * the respective signal before calling execv(). See signal(7) man-page. + */ +static void os_unblock_signals(void) +{ + sigset_t sigs; + + sigfillset(&sigs); + sigprocmask(SIG_UNBLOCK, &sigs, NULL); +} + void os_relaunch(char *argv[]) { + os_unblock_signals(); + execv(argv[0], argv); os_exit(1); } -- 2.37.2

2 2

[PATCH v4 0/1] blk: Rationalise the block interface
by Simon Glass 26 Sep '22

26 Sep '22

The block interface has two separate implementations, one using driver model and one not. The latter is really only needed for SPL, where size constraints allegedly don't allow use of driver model. Of course we still need space for filesystems and other code, so it isn't clear that driver model is anything more than the straw that breaks the camel's back. The driver model version uses a uclass ID for the interface time, but converts back and forth between that and if_type, which is the legacy type. The HAVE_BLOCK_DEVICE define is mostly a hangover from the old days. At present its main purpose is to enable the legacy block implementation in SPL. Finally the use of 'select' to enable BLK does not work very well. It causes kconfig errors when another option depends on BLK and it is not recommended by the kconfig style guide. This series aims to clean things up: - Enable BLK based on whether different media types are used, but still allow boards to disable it - Rename HAVE_BLOCK_DEVICE to indicates its real purpose - Drop if_type and use the uclass instead - Drop some obsolete if_type values An issue not resolved by this series is that the sandbox host interface does not actually have a device. At present it uses the root device, which was convenience for the driver model conversion but not really correct. It should be possible to clean this up, in a future series. Another minor issue is the use of UCLASS_USB for a mass-storage device. This has been the case for a while and is not addresed by this series, other than to add a comment. Note that this test relies on Tom Rini's series to drop various boards including warp and cm_t335 Finally, a patch is included to make binman put fake files in a subdirectory, since repeated runs of certain boards can cause unrelated failues (e.g. chromebook_coral) when fake files are left around. Changes in v4: - Fix comments from Takahiro Akashi missed on the previous revision Changes in v3: - Drop previously applied patches - Rebase on -next after EFI changes Changes in v2: - Use conv_uclass_id() instead of the confusing uclass_id_to_uclass_id() Simon Glass (1): blk: Rename if_type to uclass_id board/st/common/stm32mp_dfu.c | 2 +- cmd/bcb.c | 4 +- cmd/blk_common.c | 18 ++--- cmd/lsblk.c | 2 +- cmd/mmc.c | 4 +- common/spl/spl_sata.c | 2 +- common/spl/spl_usb.c | 2 +- common/usb_storage.c | 6 +- disk/disk-uclass.c | 8 +- disk/part.c | 14 ++-- doc/usage/partitions.rst | 2 +- drivers/ata/sata.c | 2 +- drivers/block/blk-uclass.c | 145 +++++++++++++++++----------------- drivers/block/blk_legacy.c | 58 +++++++------- drivers/block/ide.c | 8 +- drivers/block/sandbox.c | 8 +- drivers/mmc/mmc-uclass.c | 2 +- drivers/mmc/mmc_legacy.c | 8 +- drivers/net/fsl_enetc.c | 12 +-- drivers/net/fsl_enetc.h | 2 +- drivers/scsi/scsi.c | 6 +- drivers/virtio/virtio_blk.c | 2 +- drivers/xen/pvblock.c | 2 +- fs/fat/fat.c | 2 +- include/blk.h | 122 ++++++++++++++-------------- include/efi_loader.h | 4 +- lib/efi_loader/efi_disk.c | 12 +-- lib/efi_loader/efi_var_file.c | 4 +- 28 files changed, 231 insertions(+), 232 deletions(-) -- 2.37.3.968.ga6b4b080e4-goog

1 2

[PATCH v2 00/15] scsi: Convert sandbox SCSI to driver model
by Simon Glass 26 Sep '22

26 Sep '22

This series moves sandbox over to use driver model for SCSI. It takes the opportunity to use common code with the USB flash driver. This is mostly refactoring, but there is a small amount of new code, notably a test which checks that data can be read from a SCSI device. Changes in v2: - Change the comments to modern style - Add a comment for the enum - Add documentation for the structs - Drop the extra blank line - Fix 'start' type Simon Glass (15): scsi: Tidy up comments for struct scsi_cmd sandbox: usb: Rename transfer_len in protocol struct scsi: Move cmd_phase enum to the header scsi: Move core emulation state into a new struct sandbox: Move buffer to scsi_emul_info scsi: Move vendor/product info into the shared struct sandbox: scsi: Move block size into shared struct sandbox: scsi: Move file size into shared struct sandbox: scsi: Move reply setup out of helper sandbox: scsi: Remove setup calls from handle_read() sandbox: scsi: Move structs to header file sandbox: Enable SCSI for all builds sandbox: scsi: Move request-handling code to scsi_emul sandbox: Convert to use driver model for SCSI sandbox: Add a test for SCSI arch/Kconfig | 1 - arch/sandbox/dts/sandbox.dtsi | 4 + arch/sandbox/dts/test.dts | 5 + configs/sandbox64_defconfig | 6 + configs/sandbox_defconfig | 1 + configs/sandbox_flattree_defconfig | 5 + configs/sandbox_noinst_defconfig | 5 + configs/sandbox_spl_defconfig | 7 +- configs/sandbox_vpl_defconfig | 5 + drivers/scsi/Makefile | 1 + drivers/scsi/sandbox_scsi.c | 132 +++++++++++++++++- drivers/scsi/scsi_emul.c | 74 +++++++++++ drivers/usb/emul/sandbox_flash.c | 207 ++++++++++------------------- include/scsi.h | 126 +++++++++++++++--- include/scsi_emul.h | 70 ++++++++++ test/dm/Makefile | 1 + test/dm/scsi.c | 39 ++++++ test/py/tests/test_ut.py | 9 ++ 18 files changed, 535 insertions(+), 163 deletions(-) create mode 100644 drivers/scsi/scsi_emul.c create mode 100644 include/scsi_emul.h create mode 100644 test/dm/scsi.c -- 2.37.3.968.ga6b4b080e4-goog

1 22

[PATCH] board: altera: c5/a10/a5/stratix10: remove Chin Liang as maintainer
by Dinh Nguyen 26 Sep '22

26 Sep '22

Chin Liang is no longer actively maintaining this project. Signed-off-by: Dinh Nguyen <dinguyen(a)kernel.org> --- board/altera/arria10-socdk/MAINTAINERS | 1 - board/altera/arria5-socdk/MAINTAINERS | 1 - board/altera/cyclone5-socdk/MAINTAINERS | 1 - board/altera/stratix10-socdk/MAINTAINERS | 1 - 4 files changed, 4 deletions(-) diff --git a/board/altera/arria10-socdk/MAINTAINERS b/board/altera/arria10-socdk/MAINTAINERS index 5a76efb54b..e288584362 100644 --- a/board/altera/arria10-socdk/MAINTAINERS +++ b/board/altera/arria10-socdk/MAINTAINERS @@ -1,6 +1,5 @@ SOCFPGA BOARD M: Dinh Nguyen <dinguyen(a)kernel.org> -M: Chin-Liang See <clsee(a)altera.com> S: Maintained F: board/altera/arria10-socdk/ F: include/configs/socfpga_arria10_socdk.h diff --git a/board/altera/arria5-socdk/MAINTAINERS b/board/altera/arria5-socdk/MAINTAINERS index 873ec2be2d..a6c5bc8024 100644 --- a/board/altera/arria5-socdk/MAINTAINERS +++ b/board/altera/arria5-socdk/MAINTAINERS @@ -1,6 +1,5 @@ SOCFPGA BOARD M: Dinh Nguyen <dinguyen(a)kernel.org> -M: Chin-Liang See <clsee(a)altera.com> S: Maintained F: board/altera/arria5-socdk/ F: include/configs/socfpga_arria5_socdk.h diff --git a/board/altera/cyclone5-socdk/MAINTAINERS b/board/altera/cyclone5-socdk/MAINTAINERS index ecf1d04f76..6a8dfdd286 100644 --- a/board/altera/cyclone5-socdk/MAINTAINERS +++ b/board/altera/cyclone5-socdk/MAINTAINERS @@ -1,6 +1,5 @@ SOCFPGA BOARD M: Dinh Nguyen <dinguyen(a)kernel.org> -M: Chin-Liang See <clsee(a)altera.com> S: Maintained F: board/altera/cyclone5-socdk/ F: include/configs/socfpga_cyclone5_socdk.h diff --git a/board/altera/stratix10-socdk/MAINTAINERS b/board/altera/stratix10-socdk/MAINTAINERS index 6efd0cfc05..0b475980dd 100644 --- a/board/altera/stratix10-socdk/MAINTAINERS +++ b/board/altera/stratix10-socdk/MAINTAINERS @@ -1,6 +1,5 @@ SOCFPGA BOARD M: Chin-Liang See <chin.liang.see(a)intel.com> -M: Dinh Nguyen <dinh.nguyen(a)intel.com> S: Maintained F: board/altera/stratix10-socdk/ F: include/configs/socfpga_stratix10_socdk.h -- 2.25.1

1 0

[PATCH] configs: socfpga: Add CONFIG_NET_RANDOM_ETHADDR=y to SOCFPGA defconfig
by teik.heng.chong＠intel.com 26 Sep '22

26 Sep '22

From: Tien Fong Chee <tien.fong.chee(a)intel.com> Ethernet initialization is only work with properly set MAC addresses. Hence, this config is required to create the random MAC addresses for Ethernet initialization. Signed-off-by: Tien Fong Chee <tien.fong.chee(a)intel.com> Signed-off-by: Teik Heng Chong <teik.heng.chong(a)intel.com> --- configs/socfpga_arria10_defconfig | 1 + configs/socfpga_arria5_defconfig | 1 + configs/socfpga_cyclone5_defconfig | 1 + 3 files changed, 3 insertions(+) diff --git a/configs/socfpga_arria10_defconfig b/configs/socfpga_arria10_defconfig index 3eac3dfa5d..53c01ff659 100644 --- a/configs/socfpga_arria10_defconfig +++ b/configs/socfpga_arria10_defconfig @@ -60,6 +60,7 @@ CONFIG_PHY_MICREL=y CONFIG_PHY_MICREL_KSZ90X1=y CONFIG_ETH_DESIGNWARE=y CONFIG_MII=y +CONFIG_NET_RANDOM_ETHADDR=y CONFIG_SPI=y CONFIG_TIMER=y CONFIG_SPL_TIMER=y diff --git a/configs/socfpga_arria5_defconfig b/configs/socfpga_arria5_defconfig index dbadb3d49a..da25aa9154 100644 --- a/configs/socfpga_arria5_defconfig +++ b/configs/socfpga_arria5_defconfig @@ -63,6 +63,7 @@ CONFIG_PHY_MICREL=y CONFIG_PHY_MICREL_KSZ90X1=y CONFIG_ETH_DESIGNWARE=y CONFIG_MII=y +CONFIG_NET_RANDOM_ETHADDR=y CONFIG_DM_RESET=y CONFIG_SPI=y CONFIG_CADENCE_QSPI=y diff --git a/configs/socfpga_cyclone5_defconfig b/configs/socfpga_cyclone5_defconfig index 4fd14d2e2c..508bfe31fc 100644 --- a/configs/socfpga_cyclone5_defconfig +++ b/configs/socfpga_cyclone5_defconfig @@ -64,6 +64,7 @@ CONFIG_PHY_MICREL=y CONFIG_PHY_MICREL_KSZ90X1=y CONFIG_ETH_DESIGNWARE=y CONFIG_MII=y +CONFIG_NET_RANDOM_ETHADDR=y CONFIG_DM_RESET=y CONFIG_SPI=y CONFIG_CADENCE_QSPI=y -- 2.25.1

2 1

Please pull u-boot-marvell/master
by Stefan Roese 26 Sep '22

26 Sep '22

Hi Tom, please pull this Turris GPIO fix from Pali: ---------------------------------------------------------------- - gpio: turris_omnia_mcu: Fix registering gpios (Pali) ---------------------------------------------------------------- Here the Azure build, without any issues: https://dev.azure.com/sr0718/u-boot/_build/results?buildId=269 Thanks, Stefan The following changes since commit 435596d57f8beedf36b5dc858fe7ba9d6c03334b: Merge tag 'u-boot-imx-20220922' of https://gitlab.denx.de/u-boot/custodians/u-boot-imx (2022-09-22 10:29:29 -0400) 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 54bcd84adf2dd2a42552143549541b2c809869a6: gpio: turris_omnia_mcu: Fix registering gpios (2022-09-26 10:04:37 +0200) ---------------------------------------------------------------- Pali Rohár (1): gpio: turris_omnia_mcu: Fix registering gpios drivers/gpio/turris_omnia_mcu.c | 19 +++++++++++++++---- 1 file changed, 15 insertions(+), 4 deletions(-)

2 1

[PULL] u-boot-usb/master
by Marek Vasut 26 Sep '22

26 Sep '22

One outstanding driver and one bugfix. The following changes since commit 435596d57f8beedf36b5dc858fe7ba9d6c03334b: Merge tag 'u-boot-imx-20220922' of https://gitlab.denx.de/u-boot/custodians/u-boot-imx (2022-09-22 10:29:29 -0400) are available in the Git repository at: git://source.denx.de/u-boot-usb.git master for you to fetch changes up to 206af3dec00071ec96c0fdec6c8dda3a039c07cf: Fix out of bound access of ep array. (2022-09-26 01:49:36 +0200) ---------------------------------------------------------------- Alison Huffman (1): Fix out of bound access of ep array. Jim Liu (1): usb: host: nuvoton: Add nuvoton NPCM7xx ehci/ohci driver drivers/usb/gadget/dwc2_udc_otg_xfer_dma.c | 6 ++--- drivers/usb/host/Kconfig | 16 ++++++++++++ drivers/usb/host/Makefile | 2 ++ drivers/usb/host/ehci-npcm.c | 119 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ drivers/usb/host/ohci-npcm.c | 108 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 5 files changed, 248 insertions(+), 3 deletions(-) create mode 100644 drivers/usb/host/ehci-npcm.c create mode 100644 drivers/usb/host/ohci-npcm.c

2 1

[PATCH v2 0/9] xilinx: versal-net: Add support for new Versal NET SoC
by Michal Simek 26 Sep '22

26 Sep '22

Hi, I am sending support for new Xilinx/AMD SoC called Versal NET. Versal NET is very similar to origin Versal SOC. There is different register layout, some IPs have been upgraded like i3c and some other changes in different location. Thanks, Michal Changes in v2: - Move symbols from .h to defconfig, enable i2c mux, dm_eth_phy, squashfs and grepenv - Remove CONFIG_SYS_CBSIZE, CONFIG_BOOTP_BOOTFILESIZE, CONFIG_BOOTP_MAY_FAIL from .h file - Setup HAS_CUSTOM_SYS_INIT_SP_ADDR Jay Buddhabhatti (4): clk: versal: Enable clock driver for Versal NET firmware: zynqmp: Add Versal NET compatible string mailbox: zynqmp: Enable ipi mailbox driver for Versal NET reset: zynqmp: Enable reset driver for Versal NET Michal Simek (5): arm64: versal-net: Add support for Versal NET platform spi: cadence_qspi: Add support for Versal NET platform spi: zynqmp_gqspi: Add support for Versal NET arm64: versal-net: Add defconfig for Versal NET arm64: versal-net: Add support for mini configuration Kconfig | 2 +- MAINTAINERS | 7 + arch/arm/Kconfig | 14 ++ arch/arm/Makefile | 1 + arch/arm/dts/Makefile | 3 + arch/arm/dts/versal-net-mini.dts | 67 +++++++ arch/arm/dts/xilinx-versal-net-virt.dts | 11 ++ arch/arm/mach-versal-net/Kconfig | 43 +++++ arch/arm/mach-versal-net/Makefile | 10 ++ arch/arm/mach-versal-net/clk.c | 35 ++++ arch/arm/mach-versal-net/cpu.c | 89 +++++++++ .../mach-versal-net/include/mach/hardware.h | 31 ++++ .../mach-versal-net/include/mach/sys_proto.h | 16 ++ board/xilinx/Kconfig | 6 +- board/xilinx/versal-net/Kconfig | 9 + board/xilinx/versal-net/MAINTAINERS | 8 + board/xilinx/versal-net/Makefile | 9 + board/xilinx/versal-net/board.c | 170 ++++++++++++++++++ configs/xilinx_versal_net_mini_defconfig | 72 ++++++++ configs/xilinx_versal_net_virt_defconfig | 131 ++++++++++++++ drivers/clk/Kconfig | 2 +- drivers/clk/clk_versal.c | 1 + drivers/firmware/firmware-zynqmp.c | 1 + drivers/mailbox/Kconfig | 2 +- drivers/reset/reset-zynqmp.c | 1 + drivers/spi/Kconfig | 2 +- drivers/spi/cadence_ospi_versal.c | 3 +- drivers/spi/zynqmp_gqspi.c | 3 +- env/Kconfig | 6 +- include/configs/xilinx_versal_net.h | 134 ++++++++++++++ include/configs/xilinx_versal_net_mini.h | 21 +++ 31 files changed, 898 insertions(+), 12 deletions(-) create mode 100644 arch/arm/dts/versal-net-mini.dts create mode 100644 arch/arm/dts/xilinx-versal-net-virt.dts create mode 100644 arch/arm/mach-versal-net/Kconfig create mode 100644 arch/arm/mach-versal-net/Makefile create mode 100644 arch/arm/mach-versal-net/clk.c create mode 100644 arch/arm/mach-versal-net/cpu.c create mode 100644 arch/arm/mach-versal-net/include/mach/hardware.h create mode 100644 arch/arm/mach-versal-net/include/mach/sys_proto.h create mode 100644 board/xilinx/versal-net/Kconfig create mode 100644 board/xilinx/versal-net/MAINTAINERS create mode 100644 board/xilinx/versal-net/Makefile create mode 100644 board/xilinx/versal-net/board.c create mode 100644 configs/xilinx_versal_net_mini_defconfig create mode 100644 configs/xilinx_versal_net_virt_defconfig create mode 100644 include/configs/xilinx_versal_net.h create mode 100644 include/configs/xilinx_versal_net_mini.h -- 2.36.1

1 10

[PATCH] gpio: turris_omnia_mcu: Fix registering gpios
by Pali Rohár 26 Sep '22

26 Sep '22

Currently all GPIOs supported by CMD_EXT_CONTROL/CMD_GET_EXT_CONTROL_STATUS commands (last 16 GPIOs) are available only when FEAT_PERIPH_MCU feature bit is set. So do not register these GPIOs by U-Boot driver when this feature bit is not set, so U-Boot 'gpio' command would see only GPIOs which really exists. Fixes: 5e4d24ccc115 ("gpio: Add Turris Omnia MCU driver") Signed-off-by: Pali Rohár <pali(a)kernel.org> --- drivers/gpio/turris_omnia_mcu.c | 19 +++++++++++++++---- 1 file changed, 15 insertions(+), 4 deletions(-) diff --git a/drivers/gpio/turris_omnia_mcu.c b/drivers/gpio/turris_omnia_mcu.c index 986ccde6bc70..2d2bf2d1dd69 100644 --- a/drivers/gpio/turris_omnia_mcu.c +++ b/drivers/gpio/turris_omnia_mcu.c @@ -16,6 +16,8 @@ enum commands_e { /* available if FEAT_EXT_CMDS bit is set in features */ CMD_GET_EXT_STATUS_DWORD = 0x11, + + /* available if FEAT_EXT_CMDS and FEAT_PERIPH_MCU bits are set in featurs */ CMD_EXT_CONTROL = 0x12, CMD_GET_EXT_CONTROL_STATUS = 0x13, }; @@ -54,6 +56,7 @@ enum ctl_byte_e { /* CMD_GET_FEATURES */ enum features_e { + FEAT_PERIPH_MCU = BIT(0), FEAT_EXT_CMDS = BIT(1), }; @@ -84,10 +87,12 @@ static int turris_omnia_mcu_get_function(struct udevice *dev, uint offset) return -EINVAL; return GPIOF_INPUT; - /* bank 2 - supported only when FEAT_EXT_CMDS is set */ + /* bank 2 - supported only when FEAT_EXT_CMDS and FEAT_PERIPH_MCU is set */ case (16 + 32 + 0) ... (16 + 32 + 15): if (!(info->features & FEAT_EXT_CMDS)) return -EINVAL; + if (!(info->features & FEAT_PERIPH_MCU)) + return -EINVAL; return GPIOF_OUTPUT; default: @@ -120,10 +125,12 @@ static int turris_omnia_mcu_get_value(struct udevice *dev, uint offset) return ((((u32)val32[3] << 24) | ((u32)val32[2] << 16) | ((u32)val32[1] << 8) | val32[0]) >> (offset - 16)) & 0x1; - /* bank 2 - supported only when FEAT_EXT_CMDS is set */ + /* bank 2 - supported only when FEAT_EXT_CMDS and FEAT_PERIPH_MCU is set */ case (16 + 32 + 0) ... (16 + 32 + 15): if (!(info->features & FEAT_EXT_CMDS)) return -EINVAL; + if (!(info->features & FEAT_PERIPH_MCU)) + return -EINVAL; ret = dm_i2c_read(dev, CMD_GET_EXT_CONTROL_STATUS, val16, 2); if (ret) return ret; @@ -162,10 +169,12 @@ static int turris_omnia_mcu_set_value(struct udevice *dev, uint offset, int valu val16[0] = value ? val16[1] : 0; return dm_i2c_write(dev, CMD_GENERAL_CONTROL, val16, sizeof(val16)); - /* bank 2 - supported only when FEAT_EXT_CMDS is set */ + /* bank 2 - supported only when FEAT_EXT_CMDS and FEAT_PERIPH_MCU is set */ case (16 + 32 + 0) ... (16 + 32 + 15): if (!(info->features & FEAT_EXT_CMDS)) return -EINVAL; + if (!(info->features & FEAT_PERIPH_MCU)) + return -EINVAL; val32[3] = BIT(offset - 16 - 32) >> 8; val32[2] = BIT(offset - 16 - 32) & 0xff; val32[1] = value ? val32[3] : 0; @@ -282,8 +291,10 @@ static int turris_omnia_mcu_probe(struct udevice *dev) uc_priv->bank_name = "mcu_"; - if (info->features & FEAT_EXT_CMDS) + if ((info->features & FEAT_EXT_CMDS) && (info->features & FEAT_PERIPH_MCU)) uc_priv->gpio_count = 16 + 32 + 16; + else if (info->features & FEAT_EXT_CMDS) + uc_priv->gpio_count = 16 + 32; else uc_priv->gpio_count = 16; -- 2.20.1

2 4