July 2022 - U-Boot

Error: cmd/libkiosk_crypto.c:5:10: fatal error: unistd.h: No such file or directory
by Thu Ra 08 Jul '22

08 Jul '22

Dear Dev of denx.de I'm trying to create a custom os which is based on Alpine linux 3.16 with your u-boot-2022.04 for my raspberry pi 4. Unfortunately, I got the following error, please see the attached file(u-boot-bootloader-error.txt). And I attached 3 Makefiles. main-makefile, common-makefile and bootloader-makefile. and also the Version of components as common.mk. Can you help me to solve this problem? And please let me know if you need further information. Your Sincerely, T Ra

2 1

[PATCH] ARM: layerscape: Use ARCH_LS104?A insead of TARGET_LS104?ARDB
by Sean Anderson 08 Jul '22

08 Jul '22

These frequency calculations depend on the RCW format, which is not dependent on any particular board. Switch to using ARCH symbols instead of TARGET. This whole function could probably use less ifdefs, but for now just do a minimal conversion. Fixes: 24cb6f2295 ("fsl-layerscape: Add fsl_esdhc peripheral clock support") Signed-off-by: Sean Anderson <sean.anderson(a)seco.com> --- arch/arm/cpu/armv8/fsl-layerscape/fsl_lsch2_speed.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/arch/arm/cpu/armv8/fsl-layerscape/fsl_lsch2_speed.c b/arch/arm/cpu/armv8/fsl-layerscape/fsl_lsch2_speed.c index 570105a75e..840e6d412b 100644 --- a/arch/arm/cpu/armv8/fsl-layerscape/fsl_lsch2_speed.c +++ b/arch/arm/cpu/armv8/fsl-layerscape/fsl_lsch2_speed.c @@ -29,8 +29,8 @@ void get_sys_info(struct sys_info *sys_info) * mux 2 clock for LS1043A/LS1046A. */ #if defined(CONFIG_SYS_DPAA_FMAN) || \ - defined(CONFIG_TARGET_LS1046ARDB) || \ - defined(CONFIG_TARGET_LS1043ARDB) + defined(CONFIG_ARCH_LS1046A) || \ + defined(CONFIG_ARCH_LS1043A) u32 rcw_tmp; #endif struct ccsr_clk *clk = (void *)(CONFIG_SYS_FSL_CLK_ADDR); @@ -129,13 +129,13 @@ void get_sys_info(struct sys_info *sys_info) #define HWA_CGA_M2_CLK_SEL 0x00000007 #define HWA_CGA_M2_CLK_SHIFT 0 -#if defined(CONFIG_TARGET_LS1046ARDB) || defined(CONFIG_TARGET_LS1043ARDB) +#if defined(CONFIG_ARCH_LS1046A) || defined(CONFIG_ARCH_LS1043A) rcw_tmp = in_be32(&gur->rcwsr[15]); switch ((rcw_tmp & HWA_CGA_M2_CLK_SEL) >> HWA_CGA_M2_CLK_SHIFT) { case 1: sys_info->freq_cga_m2 = freq_c_pll[1]; break; -#if defined(CONFIG_TARGET_LS1046ARDB) +#if defined(CONFIG_ARCH_LS1046A) case 2: sys_info->freq_cga_m2 = freq_c_pll[1] / 2; break; @@ -143,7 +143,7 @@ void get_sys_info(struct sys_info *sys_info) case 3: sys_info->freq_cga_m2 = freq_c_pll[1] / 3; break; -#if defined(CONFIG_TARGET_LS1046ARDB) +#if defined(CONFIG_ARCH_LS1046A) case 6: sys_info->freq_cga_m2 = freq_c_pll[0] / 2; break; -- 2.35.1.1320.gc452695387.dirty

1 2

[PATCH] regmap: fix some comments
by Ralph Siemsen 08 Jul '22

08 Jul '22

Correct spelling and copy/paste errors in comments. Fixes 1c4db59d9b ("regmap: Add support for regmap fields") Signed-off-by: Ralph Siemsen <ralph.siemsen(a)linaro.org> --- include/regmap.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/include/regmap.h b/include/regmap.h index 8216de015d..e81a3602ae 100644 --- a/include/regmap.h +++ b/include/regmap.h @@ -460,7 +460,7 @@ struct reg_field { struct regmap_field; /** - * REG_FIELD() - A convenient way to initialize a 'struct reg_feild'. + * REG_FIELD() - A convenient way to initialize a 'struct reg_field'. * * @_reg: Offset of the register within the regmap bank * @_lsb: lsb of the register field. @@ -519,9 +519,9 @@ void devm_regmap_field_free(struct udevice *dev, struct regmap_field *field); int regmap_field_write(struct regmap_field *field, unsigned int val); /** - * regmap_read() - Read a 32-bit value from a regmap + * regmap_field_read() - Read a 32-bit value from a regmap * - * @field: Regmap field to write to + * @field: Regmap field to read from * @valp: Pointer to the buffer to receive the data read from the regmap * field * -- 2.25.1

3 2

[PATCH] serial: ns16550: Wait in debug_uart_init until tx buffer is empty
by Pali Rohár 08 Jul '22

08 Jul '22

Commit d293759d55cc ("serial: ns16550: Add support for SPL_DEBUG_UART_BASE") fixed support for setting correct early debug UART base address in SPL. But after this commit, output from Marvell A385 BootROM is truncated or lost and not fully present on serial console. Debugging this issue showed that BootROM just put bytes into UART HW output buffer and does not wait until UART HW transmit all characters. U-Boot ns16550 early debug is initialized very early and during its initialization is resetting UART HW and flushing remaining transmit buffer (which still contains BootROM output). Fix this issue by waiting in init function prior resetting UART HW until TxEmpty bit in UART Line Status Register is set. TxEmpty is set when all remaining bytes from HW buffer are transmitted. Signed-off-by: Pali Rohár <pali(a)kernel.org> --- drivers/serial/ns16550.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/serial/ns16550.c b/drivers/serial/ns16550.c index 78bfe6281ce3..13418004e2b0 100644 --- a/drivers/serial/ns16550.c +++ b/drivers/serial/ns16550.c @@ -328,6 +328,8 @@ static inline void _debug_uart_init(void) struct ns16550 *com_port = (struct ns16550 *)CONFIG_VAL(DEBUG_UART_BASE); int baud_divisor; + while (!(serial_din(&com_port->lsr) & UART_LSR_TEMT)); + /* * We copy the code from above because it is already horribly messy. * Trying to refactor to nicely remove the duplication doesn't seem -- 2.20.1

3 3

[PATCH v2] led: pwm: Use NOP uclass driver for top-level node
by Stefan Herbrechtsmeier 08 Jul '22

08 Jul '22

From: Stefan Herbrechtsmeier <stefan.herbrechtsmeier(a)weidmueller.com> The top level DT node of pwm-leds is not a LED itself, bind NOP uclass driver to it, and bind different LED uclass driver to its subnodes which represent the actual LEDs. This change removes the top-level node from the 'led list' command output and is based on the commit 01074697801b ("led: gpio: Use NOP uclass driver for top-level node"). Signed-off-by: Stefan Herbrechtsmeier <stefan.herbrechtsmeier(a)weidmueller.com> --- Changes in v2: - Remove unused uc_plat variable drivers/led/led_pwm.c | 19 +++++++------------ 1 file changed, 7 insertions(+), 12 deletions(-) diff --git a/drivers/led/led_pwm.c b/drivers/led/led_pwm.c index 10bd1636c3..0ebae358eb 100644 --- a/drivers/led/led_pwm.c +++ b/drivers/led/led_pwm.c @@ -95,27 +95,17 @@ static enum led_state_t led_pwm_get_state(struct udevice *dev) static int led_pwm_probe(struct udevice *dev) { struct led_pwm_priv *priv = dev_get_priv(dev); - struct led_uc_plat *uc_plat = dev_get_uclass_plat(dev); - - /* Ignore the top-level LED node */ - if (!uc_plat->label) - return 0; return led_pwm_set_state(dev, (priv->enabled) ? LEDST_ON : LEDST_OFF); } static int led_pwm_of_to_plat(struct udevice *dev) { - struct led_uc_plat *uc_plat = dev_get_uclass_plat(dev); struct led_pwm_priv *priv = dev_get_priv(dev); struct ofnode_phandle_args args; uint def_brightness, max_brightness; int ret; - /* Ignore the top-level LED node */ - if (!uc_plat->label) - return 0; - ret = dev_read_phandle_with_args(dev, "pwms", "#pwm-cells", 0, 0, &args); if (ret) return ret; @@ -173,10 +163,15 @@ static const struct udevice_id led_pwm_ids[] = { U_BOOT_DRIVER(led_pwm) = { .name = LEDS_PWM_DRIVER_NAME, .id = UCLASS_LED, - .of_match = led_pwm_ids, .ops = &led_pwm_ops, .priv_auto = sizeof(struct led_pwm_priv), - .bind = led_pwm_bind, .probe = led_pwm_probe, .of_to_plat = led_pwm_of_to_plat, }; + +U_BOOT_DRIVER(led_pwm_wrap) = { + .name = LEDS_PWM_DRIVER_NAME "_wrap", + .id = UCLASS_NOP, + .of_match = led_pwm_ids, + .bind = led_pwm_bind, +}; -- 2.30.2

2 1

[PATCH v2] mtd: rawnand: Add support to dedicated function to set timings
by kory.maincent＠bootlin.com 08 Jul '22

08 Jul '22

From: Kory Maincent <kory.maincent(a)bootlin.com> With the current code if the board has an ONFI compliant NAND without support to the get and set features, U-boot returns an ENOTSUP error when trying to tune the timings which prevents the probe of the device. Indeed onfi_set_features() return ENOTSUP error if set/get features is not supported. In the case of timings we should not return ENOTSUP because we can use the default timings. The NAND is already capable of listening at its highest supported rate, so we assume in this case that it is fine to skip the operation. Fix it by adding an intermediate nand_onfi_set_timings() function which does not error out if set/get feature is not supported. Signed-off-by: Kory Maincent <kory.maincent(a)bootlin.com> --- Change since v1: - Update commit message drivers/mtd/nand/raw/nand_base.c | 30 +++++++++++++++++++----------- 1 file changed, 19 insertions(+), 11 deletions(-) diff --git a/drivers/mtd/nand/raw/nand_base.c b/drivers/mtd/nand/raw/nand_base.c index 6f81257cf1..e8ece0a4a0 100644 --- a/drivers/mtd/nand/raw/nand_base.c +++ b/drivers/mtd/nand/raw/nand_base.c @@ -974,6 +974,22 @@ static int nand_reset_data_interface(struct nand_chip *chip, int chipnr) return ret; } +static int nand_onfi_set_timings(struct mtd_info *mtd, struct nand_chip *chip) +{ + if (!chip->onfi_version || + !(le16_to_cpu(chip->onfi_params.opt_cmd) + & ONFI_OPT_CMD_SET_GET_FEATURES)) + return 0; + + u8 tmode_param[ONFI_SUBFEATURE_PARAM_LEN] = { + chip->onfi_timing_mode_default, + }; + + return chip->onfi_set_features(mtd, chip, + ONFI_FEATURE_ADDR_TIMING_MODE, + tmode_param); +} + /** * nand_setup_data_interface - Setup the best data interface and timings * @chip: The NAND chip @@ -999,17 +1015,9 @@ static int nand_setup_data_interface(struct nand_chip *chip, int chipnr) * Ensure the timing mode has been changed on the chip side * before changing timings on the controller side. */ - if (chip->onfi_version) { - u8 tmode_param[ONFI_SUBFEATURE_PARAM_LEN] = { - chip->onfi_timing_mode_default, - }; - - ret = chip->onfi_set_features(mtd, chip, - ONFI_FEATURE_ADDR_TIMING_MODE, - tmode_param); - if (ret) - goto err; - } + ret = nand_onfi_set_timings(mtd, chip); + if (ret) + goto err; ret = chip->setup_data_interface(mtd, chipnr, chip->data_interface); err: -- 2.25.1

3 2

[PATCH 1/1] Makefile: respect CONFIG_CC_OPTIMIZE_FOR_DEBUG for host tools
by Heinrich Schuchardt 08 Jul '22

08 Jul '22

If CONFIG_CC_OPTIMIZE_FOR_DEBUG=y, the host tools should be built with debug symbols and with reduced optimization. Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt(a)canonical.com> --- Makefile | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/Makefile b/Makefile index 2fa3a3b488..faa01eadd7 100644 --- a/Makefile +++ b/Makefile @@ -673,6 +673,12 @@ else include/config/auto.conf: ; endif # $(dot-config) +ifdef CONFIG_CC_OPTIMIZE_FOR_DEBUG +KBUILD_HOSTCFLAGS := -Wall -Wstrict-prototypes -Og -g -fomit-frame-pointer \ + $(HOST_LFS_CFLAGS) $(HOSTCFLAGS) +KBUILD_HOSTCXXFLAGS := -Og -g $(HOST_LFS_CFLAGS) $(HOSTCXXFLAGS) +endif + # # Xtensa linker script cannot be preprocessed with -ansi because of # preprocessor operations on strings that don't make C identifiers. -- 2.36.1

2 1

[PATCH 1/1] sound: enable building DA7219 driver with ACPIGEN=n
by Heinrich Schuchardt 08 Jul '22

08 Jul '22

sandbox_defconfig builds the DA7219 driver. It should be possible to build the sandbox without ACPI support. ACPI support in the DA7219 driver is only needed when creating an ACPI table. Fix building with ACPIGEN=n. Fixes: 0324b7123e22 ("sound: Add an ACPI driver for Dialog Semicondutor da7219") Signed-off-by: Heinrich Schuchardt <xypron.glpk(a)gmx.de> --- drivers/sound/da7219.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/sound/da7219.c b/drivers/sound/da7219.c index 8d674bcb4f..c1edef4436 100644 --- a/drivers/sound/da7219.c +++ b/drivers/sound/da7219.c @@ -23,6 +23,7 @@ #define DA7219_ACPI_HID "DLGS7219" +__maybe_unused static int da7219_acpi_fill_ssdt(const struct udevice *dev, struct acpi_ctx *ctx) { @@ -171,10 +172,12 @@ static int da7219_acpi_setup_nhlt(const struct udevice *dev, #endif struct acpi_ops da7219_acpi_ops = { +#ifdef CONFIG_ACPIGEN .fill_ssdt = da7219_acpi_fill_ssdt, #ifdef CONFIG_X86 .setup_nhlt = da7219_acpi_setup_nhlt, #endif +#endif }; static const struct udevice_id da7219_ids[] = { -- 2.36.1

2 1

[PATCH 1/1] snd: enable building max98357a driver with ACPIGEN=n
by Heinrich Schuchardt 08 Jul '22

08 Jul '22

sandbox_defconfig builds the max98357a driver. It should be possible to build the sandbox without ACPI support. ACPI support in the max98357a driver is only needed when creating an ACPI table. Fix building with ACPIGEN=n. Fixes: 54bcca29737f ("sound: Add an ACPI driver for Maxim MAX98357ac") Signed-off-by: Heinrich Schuchardt <xypron.glpk(a)gmx.de> --- drivers/sound/max98357a.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/sound/max98357a.c b/drivers/sound/max98357a.c index a2088f0301..bdf6dc236e 100644 --- a/drivers/sound/max98357a.c +++ b/drivers/sound/max98357a.c @@ -38,6 +38,7 @@ static int max98357a_of_to_plat(struct udevice *dev) return 0; } +__maybe_unused static int max98357a_acpi_fill_ssdt(const struct udevice *dev, struct acpi_ctx *ctx) { @@ -137,10 +138,12 @@ static int max98357a_acpi_setup_nhlt(const struct udevice *dev, #endif struct acpi_ops max98357a_acpi_ops = { +#ifdef CONFIG_ACPIGEN .fill_ssdt = max98357a_acpi_fill_ssdt, #ifdef CONFIG_X86 .setup_nhlt = max98357a_acpi_setup_nhlt, #endif +#endif }; static const struct audio_codec_ops max98357a_ops = { -- 2.36.1

2 1

[PATCH 1/1] test: fix CONFIG_ACPIGEN dependencies
by Heinrich Schuchardt 08 Jul '22

08 Jul '22

Some tests cannot be built with CONFIG_ACPIGEN=n. Consider this in the Makefile. Signed-off-by: Heinrich Schuchardt <xypron.glpk(a)gmx.de> --- test/dm/Makefile | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) diff --git a/test/dm/Makefile b/test/dm/Makefile index f0a7c97e3d..c53e2dea91 100644 --- a/test/dm/Makefile +++ b/test/dm/Makefile @@ -18,9 +18,13 @@ obj-$(CONFIG_UT_DM) += core.o obj-$(CONFIG_UT_DM) += read.o obj-$(CONFIG_UT_DM) += phys2bus.o ifneq ($(CONFIG_SANDBOX),) -obj-$(CONFIG_ACPIGEN) += acpi.o -obj-$(CONFIG_ACPIGEN) += acpigen.o -obj-$(CONFIG_ACPIGEN) += acpi_dp.o +ifeq ($(CONFIG_ACPIGEN),y) +obj-y += acpi.o +obj-y += acpigen.o +obj-y += acpi_dp.o +obj-(CONFIG_DM_GPIO) += gpio.o +obj-y += irq.o +endif obj-$(CONFIG_ADC) += adc.o obj-$(CONFIG_SOUND) += audio.o obj-$(CONFIG_AXI) += axi.o @@ -43,11 +47,9 @@ ifneq ($(CONFIG_EFI_PARTITION),) obj-$(CONFIG_FASTBOOT_FLASH_MMC) += fastboot.o endif obj-$(CONFIG_FIRMWARE) += firmware.o -obj-$(CONFIG_DM_GPIO) += gpio.o obj-$(CONFIG_DM_HWSPINLOCK) += hwspinlock.o obj-$(CONFIG_DM_I2C) += i2c.o obj-$(CONFIG_SOUND) += i2s.o -obj-y += irq.o obj-$(CONFIG_CLK_K210_SET_RATE) += k210_pll.o obj-$(CONFIG_IOMMU) += iommu.o obj-$(CONFIG_LED) += led.o -- 2.36.1

2 1