October 2024 - U-Boot

[PATCH] arm64: zynqmp: Fix pwm-fan polarity
by Michal Simek 06 Nov '24

06 Nov '24

From: Saeed Nowshadi <saeed.nowshadi(a)amd.com> In previous version of pwm driver, the polarity of pwm were implemented in reverse. In recent release, that issue in the driver is fixed, therefore, correctly set the polarity in the device tree. Signed-off-by: Saeed Nowshadi <saeed.nowshadi(a)amd.com> Signed-off-by: Michal Simek <michal.simek(a)amd.com> --- arch/arm/dts/zynqmp-sc-revB.dts | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/arm/dts/zynqmp-sc-revB.dts b/arch/arm/dts/zynqmp-sc-revB.dts index 1af3f643567e..c4f70581695d 100644 --- a/arch/arm/dts/zynqmp-sc-revB.dts +++ b/arch/arm/dts/zynqmp-sc-revB.dts @@ -3,7 +3,7 @@ * dts file for Xilinx ZynqMP Generic System Controller * * (C) Copyright 2021 - 2022, Xilinx, Inc. - * (C) Copyright 2022 - 2023, Advanced Micro Devices, Inc. + * (C) Copyright 2022 - 2024, Advanced Micro Devices, Inc. * * Michal Simek <michal.simek(a)amd.com> */ @@ -80,7 +80,7 @@ pwm-fan { compatible = "pwm-fan"; status = "okay"; - pwms = <&ttc0 2 40000 1>; + pwms = <&ttc0 2 40000 0>; }; }; -- 2.43.0

1 1

[PATCH] arm64: zynqmp: Configure SoC RTC on SOM
by Michal Simek 06 Nov '24

06 Nov '24

Use RTC available in HW on Kria SOM without using emulation that's why configure it properly and disable emulated one. Power on reset value of RTC Calibration register without battery backup is not matching with crystal frequency which leads to RTC time drift. That's why write CALIB_WRITE register with crystal frequency (0x7FFF). There is also an option to write zero so that Linux will set default value (0x7FFF) in driver probe but calibration 0 is not permited by DT schema. Co-developed-by: Srinivas Goud <srinivas.goud(a)amd.com> Signed-off-by: Srinivas Goud <srinivas.goud(a)amd.com> Signed-off-by: Michal Simek <michal.simek(a)amd.com> --- arch/arm/dts/zynqmp-sm-k26-revA.dts | 1 + configs/xilinx_zynqmp_kria_defconfig | 1 - 2 files changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm/dts/zynqmp-sm-k26-revA.dts b/arch/arm/dts/zynqmp-sm-k26-revA.dts index 8056f6b176ed..8c43ade94053 100644 --- a/arch/arm/dts/zynqmp-sm-k26-revA.dts +++ b/arch/arm/dts/zynqmp-sm-k26-revA.dts @@ -387,6 +387,7 @@ &rtc { status = "okay"; + calibration = <0x7fff>; }; &lpd_dma_chan1 { diff --git a/configs/xilinx_zynqmp_kria_defconfig b/configs/xilinx_zynqmp_kria_defconfig index dd4df0b2da17..0dddf69c5d05 100644 --- a/configs/xilinx_zynqmp_kria_defconfig +++ b/configs/xilinx_zynqmp_kria_defconfig @@ -187,7 +187,6 @@ CONFIG_DM_PWM=y CONFIG_PWM_CADENCE_TTC=y CONFIG_RESET_ZYNQMP=y CONFIG_DM_RTC=y -CONFIG_RTC_EMULATION=y CONFIG_RTC_ZYNQMP=y CONFIG_SCSI=y CONFIG_ARM_DCC=y -- 2.43.0

1 1

[PATCH] libfdt: Fix build with swig 4.3.0
by Rudi Heitbaum 06 Nov '24

06 Nov '24

Call SWIG_AppendOutput instead of SWIG_Python_AppendOutput so that is_void is handled within swig. Link: https://github.com/swig/swig/commit/cd39cf132c96a0887be07c826b80804d7677a701 Signed-off-by: Rudi Heitbaum <rudi(a)heitbaum.com> --- scripts/dtc/pylibfdt/libfdt.i_shipped | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/scripts/dtc/pylibfdt/libfdt.i_shipped b/scripts/dtc/pylibfdt/libfdt.i_shipped index 56cc5d48f4..e4659489a9 100644 --- a/scripts/dtc/pylibfdt/libfdt.i_shipped +++ b/scripts/dtc/pylibfdt/libfdt.i_shipped @@ -1037,7 +1037,7 @@ typedef uint32_t fdt32_t; fdt_string(fdt1, fdt32_to_cpu($1->nameoff))); buff = PyByteArray_FromStringAndSize( (const char *)($1 + 1), fdt32_to_cpu($1->len)); - resultobj = SWIG_Python_AppendOutput(resultobj, buff); + resultobj = SWIG_AppendOutput(resultobj, buff); } } @@ -1076,7 +1076,7 @@ typedef uint32_t fdt32_t; %typemap(argout) int *depth { PyObject *val = Py_BuildValue("i", *arg$argnum); - resultobj = SWIG_Python_AppendOutput(resultobj, val); + resultobj = SWIG_AppendOutput(resultobj, val); } %apply int *depth { int *depth }; @@ -1092,7 +1092,7 @@ typedef uint32_t fdt32_t; if (PyTuple_GET_SIZE(resultobj) == 0) resultobj = val; else - resultobj = SWIG_Python_AppendOutput(resultobj, val); + resultobj = SWIG_AppendOutput(resultobj, val); } } -- 2.43.0

4 4

[PATCH] armv8: cpu: Implement allow_unaligned()
by Sam Protsenko 06 Nov '24

06 Nov '24

Usually ARMv8 platforms allow unaligned access for Normal memory. But some chips might not allow it by default, having SCTLR.A bit set to 1 before U-Boot execution. One such example is Exynos850 SoC. As allow_unaligned() is not implemented for ARMv8 at the moment, its __weak implementation is used, which does nothing. That might lead to unaligned access abort, for example when running EFI selftest. Fix that by implementing allow_unaligned() for ARMv8. The issue was found when running EFI selftest on E850-96 board (Exynos850 based): => bootefi selftest $fdtcontroladdr ... Executing 'HII database protocols' "Synchronous Abort" handler, esr 0x96000021, far 0xbaac0991 ... resetting ... Unaligned abort happens in u16_strnlen(), which is called from efi_hii_sibt_string_ucs2_block_next(): u16_strlen(blk->string_text) where 'blk' type is struct efi_hii_sibt_string_ucs2_block. Because this struct is packed, doing "->string_text" makes 'blk' address incremented by 1 byte, which makes it unaligned. Although allow_unaligned() was called in efi_init_early() before EFI selftest execution, it wasn't implemented for ARMv8 CPUs, so data abort happened. Signed-off-by: Sam Protsenko <semen.protsenko(a)linaro.org> --- arch/arm/cpu/armv8/cpu.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/arch/arm/cpu/armv8/cpu.c b/arch/arm/cpu/armv8/cpu.c index d568efa427ab..82ecf02f4b03 100644 --- a/arch/arm/cpu/armv8/cpu.c +++ b/arch/arm/cpu/armv8/cpu.c @@ -94,3 +94,8 @@ void armv8_setup_psci(void) secure_ram_addr(psci_arch_init)(); } #endif + +void allow_unaligned(void) +{ + set_sctlr(get_sctlr() & ~CR_A); +} -- 2.39.5

2 2

[PATCH] imx9: Improve boot device auto-selection
by egyszeregy＠freemail.hu 05 Nov '24

05 Nov '24

From: Benjamin Szőke <egyszeregy(a)freemail.hu> Improve "mmcautodetect=yes" boot device auto-selection to able to use it if CONFIG_ENV_IS_NOWHERE is enabled for i.MX9 SoC and i.MX93 EVK board. Signed-off-by: Benjamin Szőke <egyszeregy(a)freemail.hu> --- arch/arm/mach-imx/imx9/soc.c | 6 +++++- board/freescale/imx93_evk/imx93_evk.c | 2 +- include/configs/imx93_evk.h | 6 ++++++ 3 files changed, 12 insertions(+), 2 deletions(-) diff --git a/arch/arm/mach-imx/imx9/soc.c b/arch/arm/mach-imx/imx9/soc.c index 7c28fa39e1..79ddbc7241 100644 --- a/arch/arm/mach-imx/imx9/soc.c +++ b/arch/arm/mach-imx/imx9/soc.c @@ -42,12 +42,16 @@ DECLARE_GLOBAL_DATA_PTR; struct rom_api *g_rom_api = (struct rom_api *)0x1980; -#ifdef CONFIG_ENV_IS_IN_MMC +#if CONFIG_IS_ENABLED(ENV_IS_IN_MMC) || CONFIG_IS_ENABLED(ENV_IS_NOWHERE) __weak int board_mmc_get_env_dev(int devno) { return devno; } +#ifndef CONFIG_SYS_MMC_ENV_DEV +#define CONFIG_SYS_MMC_ENV_DEV 0 +#endif + int mmc_get_env_dev(void) { int ret; diff --git a/board/freescale/imx93_evk/imx93_evk.c b/board/freescale/imx93_evk/imx93_evk.c index 341831a7d3..c9171df330 100644 --- a/board/freescale/imx93_evk/imx93_evk.c +++ b/board/freescale/imx93_evk/imx93_evk.c @@ -58,7 +58,7 @@ int board_init(void) int board_late_init(void) { -#ifdef CONFIG_ENV_IS_IN_MMC +#if CONFIG_IS_ENABLED(ENV_IS_IN_MMC) || CONFIG_IS_ENABLED(ENV_IS_NOWHERE) board_late_mmc_env_init(); #endif diff --git a/include/configs/imx93_evk.h b/include/configs/imx93_evk.h index 53fb8c9b1b..2a9595a261 100644 --- a/include/configs/imx93_evk.h +++ b/include/configs/imx93_evk.h @@ -27,6 +27,12 @@ #define BOOTENV #endif +#if CONFIG_IS_ENABLED(ENV_IS_NOWHERE) +#ifndef CONFIG_SYS_MMC_ENV_DEV +#define CONFIG_SYS_MMC_ENV_DEV 0 +#endif +#endif + /* Initial environment variables */ #define CFG_EXTRA_ENV_SETTINGS \ BOOTENV \ -- 2.47.0.windows.1

4 11

[PATCH 1/4] rockchip: rk3399-roc-pc: Hook sysreset gpio to enable full reset
by Paul Kocialkowski 05 Nov '24

05 Nov '24

From: Paul Kocialkowski <contact(a)paulk.fr> The reset mechanism used by Linux to reset the SoC is known to only partially reset the logic. A mechanism is implemented in rk3399_force_power_on_reset to use a GPIO connected to the PMIC's over-temperature (OTP) reset pin, which fully resets all logic. Hook the associated GPIO where the function expects it to enable this reset mechanism and avoid any possible side-effect of partially-reset units. Without this patch, reading from the micro sd slot fails after a reset. With this mechanism, U-Boot is able to boot from it reliably. Signed-off-by: Paul Kocialkowski <contact(a)paulk.fr> --- arch/arm/dts/rk3399-roc-pc-u-boot.dtsi | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/arch/arm/dts/rk3399-roc-pc-u-boot.dtsi b/arch/arm/dts/rk3399-roc-pc-u-boot.dtsi index aecf7dbe383c..883d399a06a3 100644 --- a/arch/arm/dts/rk3399-roc-pc-u-boot.dtsi +++ b/arch/arm/dts/rk3399-roc-pc-u-boot.dtsi @@ -7,6 +7,10 @@ #include "rk3399-sdram-lpddr4-100.dtsi" / { + config { + sysreset-gpio = <&gpio1 RK_PA6 GPIO_ACTIVE_HIGH>; + }; + vcc_hub_en: vcc_hub_en-regulator { compatible = "regulator-fixed"; enable-active-high; @@ -36,6 +40,10 @@ bootph-pre-ram; }; +&gpio1 { + bootph-pre-ram; +}; + &spi1 { flash@0 { bootph-pre-ram; -- 2.46.2

5 21

[PATCH] boot/image-board.c: boot_get_fpga(): pass compatible flag to fpga_load()
by Peter Korsgaard 05 Nov '24

05 Nov '24

For E.G. signed FPGA bitstreams, similar to how it is done for the FPGA loading from SPL since commit 282eed50ecb7 ("fpga: pass compatible flags to fpga_load()"). Signed-off-by: Peter Korsgaard <peter(a)korsgaard.com> --- boot/image-board.c | 16 +++++++++++++--- 1 file changed, 13 insertions(+), 3 deletions(-) diff --git a/boot/image-board.c b/boot/image-board.c index 1757e5816d8..b726bd6b303 100644 --- a/boot/image-board.c +++ b/boot/image-board.c @@ -624,9 +624,10 @@ int boot_get_fpga(struct bootm_headers *images) void *buf; int conf_noffset; int fit_img_result; - const char *uname, *name; + const char *uname, *name, *compatible; int err; int devnum = 0; /* TODO support multi fpga platforms */ + int flags = 0; if (!IS_ENABLED(CONFIG_FPGA)) return -ENOSYS; @@ -674,20 +675,29 @@ int boot_get_fpga(struct bootm_headers *images) return fit_img_result; } + conf_noffset = fit_image_get_node(buf, uname); + compatible = fdt_getprop(buf, conf_noffset, "compatible", NULL); + if (!compatible) { + printf("'fpga' image without 'compatible' property\n"); + } else { + if (CONFIG_IS_ENABLED(FPGA_LOAD_SECURE)) + flags = fpga_compatible2flag(devnum, compatible); + } + if (!fpga_is_partial_data(devnum, img_len)) { name = "full"; err = fpga_loadbitstream(devnum, (char *)img_data, img_len, BIT_FULL); if (err) err = fpga_load(devnum, (const void *)img_data, - img_len, BIT_FULL, 0); + img_len, BIT_FULL, flags); } else { name = "partial"; err = fpga_loadbitstream(devnum, (char *)img_data, img_len, BIT_PARTIAL); if (err) err = fpga_load(devnum, (const void *)img_data, - img_len, BIT_PARTIAL, 0); + img_len, BIT_PARTIAL, flags); } if (err) -- 2.39.5

1 1

[PATCH v5 00/11] led: introduce LED boot and activity function
by Christian Marangi 05 Nov '24

05 Nov '24

This series is a reworked version of the previous seried: misc: introduce STATUS LED activity function This series port and expand the legacy concept of LED boot from the legacy Status LED API to new LED API. One thing that many device need is a way to communicate to the user that the device is actually doing something. This is especially useful for recovery steps where an user (for example) insert an USB drive, keep a button pressed and the device autorecover. There is currently no way to signal the user externally that the bootloader is processing/recoverying aside from setting a LED on. A solid LED on is not enough and won't actually signal any kind of progress. Solution is the good old blinking LED but uboot doesn't suggest (and support) interrupts and almost all the LED are usually GPIO LED that doesn't support HW blink. Additional Kconfg are also introduced to set the LED boot and activity. Those are referenced by label. A documentation for old and these new LED API is created. (world tested with the azure pipeline) Changes v5: - Use led_get_by_label and don't cache led dev - Improve Kconfig for /options location - Add Reviewed-by tags - Enable blink also on MTD read - Drop additional get_state in led_sw_on_state_change from v4 Changes v4: - Drop led_set_state/period_by_label - Switch to /options/u-boot - Rework to cache label and dev in led uclass - Add multiple patch for additional helper - Rework patches to limit ifdef in some place Changes v3: - Switch to /config property Changes v2: - Drop GPIO SW implementation - Add fix for new LED SW BLINK Christian Marangi (11): led: toggle LED on initial SW blink dm: core: implement ofnode_options helpers led: implement LED boot API common: board_r: rework BOOT LED handling led: implement LED activity API tftp: implement support for LED activity mtd: implement support for LED activity ubi: implement support for LED activity doc: introduce led.rst documentation test: dm: Add tests for LED boot and activity test: dm: Expand ofnode options test with new helper arch/sandbox/dts/test.dts | 5 ++ cmd/mtd.c | 9 +++ cmd/ubi.c | 13 ++- common/board_r.c | 28 +++++-- doc/api/index.rst | 1 + doc/api/led.rst | 10 +++ drivers/core/ofnode.c | 33 ++++++++ drivers/led/Kconfig | 24 ++++++ drivers/led/led-uclass.c | 162 +++++++++++++++++++++++++++++++++++++ drivers/led/led_sw_blink.c | 17 +++- include/dm/ofnode.h | 41 ++++++++++ include/led.h | 147 ++++++++++++++++++++++++++++++++- include/status_led.h | 13 +++ net/net.c | 4 + net/tftp.c | 5 ++ test/dm/led.c | 72 +++++++++++++++++ test/dm/ofnode.c | 9 +++ 17 files changed, 579 insertions(+), 14 deletions(-) create mode 100644 doc/api/led.rst -- 2.45.2

4 29

[PATCH] net: wget: check strict_strtoul() return value
by Jerome Forissier 05 Nov '24

05 Nov '24

Check the return value of strict_strtoul() when processing the Content-Length header as recommended by Coverity [1]. [1] https://lists.denx.de/pipermail/u-boot/2024-October/567050.html Reported-by: Coverity (CID 510464) Signed-off-by: Jerome Forissier <jerome.forissier(a)linaro.org> --- net/wget.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/net/wget.c b/net/wget.c index b4251e0f293..a3821495e03 100644 --- a/net/wget.c +++ b/net/wget.c @@ -256,7 +256,12 @@ static void wget_connected(uchar *pkt, unsigned int tcp_seq_num, content_length = -1; } else { pos += sizeof(content_len) + 2; - strict_strtoul(pos, 10, &content_length); + if (strict_strtoul(pos, 10, &content_length) < 0) { + wget_loop_state = NETLOOP_FAIL; + wget_fail("wget: bad Content-Length\n", tcp_seq_num, tcp_ack_num, action); + net_set_state(NETLOOP_FAIL); + return; + } debug_cond(DEBUG_WGET, "wget: Connected Len %lu\n", content_length); -- 2.40.1

3 3

[PATCH] arm: dts: e850-96: Remove TRNG node from appended dts
by Sam Protsenko 05 Nov '24

05 Nov '24

Commit 136b7b6d2e98 ("Subtree merge tag 'v6.11-dts' of dts repo [1] into dts/upstream") updates the upstream dts for E850-96 board, bringing upstream TRNG node. There is no need to keep TRNG node in appended dts anymore, so remove it. Tested on E850-96 board by running 'rng' command in U-Boot shell. Signed-off-by: Sam Protsenko <semen.protsenko(a)linaro.org> --- arch/arm/dts/exynos850-e850-96-u-boot.dtsi | 11 ----------- 1 file changed, 11 deletions(-) diff --git a/arch/arm/dts/exynos850-e850-96-u-boot.dtsi b/arch/arm/dts/exynos850-e850-96-u-boot.dtsi index 3aa5d8bb10d0..6d7148f7264a 100644 --- a/arch/arm/dts/exynos850-e850-96-u-boot.dtsi +++ b/arch/arm/dts/exynos850-e850-96-u-boot.dtsi @@ -3,17 +3,6 @@ * Copyright (c) 2023 Linaro Ltd. */ -&soc { - /* TODO: Remove this node once it appears in upstream dts */ - trng: rng@12081400 { - compatible = "samsung,exynos850-trng"; - reg = <0x12081400 0x100>; - clocks = <&cmu_core CLK_GOUT_SSS_ACLK>, - <&cmu_core CLK_GOUT_SSS_PCLK>; - clock-names = "secss", "pclk"; - }; -}; - &pmu_system_controller { bootph-all; samsung,uart-debug-1; -- 2.39.5

2 1