August 2022 - U-Boot

[PATCH 0/3] gpio: update gpio_get_status()
by Patrice Chotard 03 Aug '22

03 Aug '22

Currently, if pin's function is GPIOF_FUNC, only "func" if displayed without any other information. It would be interesting, if information is available, to indicate which pinmuxing's name is used. Patrice Chotard (3): gpio: Allow to print pin's label even for pin with GPIOF_FUNC function gpio: Fix pin's status display for pin with GPIOF_UNUSED function pinctrl: pinctrl_stm32: Populate uc_priv->name[] with pinmux node's name drivers/gpio/gpio-uclass.c | 18 ++++++++++++------ drivers/pinctrl/pinctrl_stm32.c | 8 ++++++-- 2 files changed, 18 insertions(+), 8 deletions(-) -- 2.25.1

4 8

[PATCH v1] configs: colibri-imx7: Enable bootd command
by Francesco Dolcini 03 Aug '22

03 Aug '22

Enable boot/bootd command in toradex colibri-imx7 defconfig, it's just convenient to have it in and every other toradex board already includes it. Signed-off-by: Francesco Dolcini <francesco.dolcini(a)toradex.com> --- configs/colibri_imx7_defconfig | 1 - configs/colibri_imx7_emmc_defconfig | 1 - 2 files changed, 2 deletions(-) diff --git a/configs/colibri_imx7_defconfig b/configs/colibri_imx7_defconfig index fbc660eb4e94..5ca2b9f40d71 100644 --- a/configs/colibri_imx7_defconfig +++ b/configs/colibri_imx7_defconfig @@ -27,7 +27,6 @@ CONFIG_BOARD_LATE_INIT=y CONFIG_SYS_PROMPT="Colibri iMX7 # " CONFIG_SYS_MAXARGS=32 CONFIG_SYS_PBSIZE=544 -# CONFIG_CMD_BOOTD is not set # CONFIG_BOOTM_PLAN9 is not set # CONFIG_BOOTM_RTEMS is not set # CONFIG_CMD_IMI is not set diff --git a/configs/colibri_imx7_emmc_defconfig b/configs/colibri_imx7_emmc_defconfig index ea1aa4c3a642..b2150f0e63c5 100644 --- a/configs/colibri_imx7_emmc_defconfig +++ b/configs/colibri_imx7_emmc_defconfig @@ -26,7 +26,6 @@ CONFIG_BOARD_LATE_INIT=y CONFIG_SYS_PROMPT="Colibri iMX7 # " CONFIG_SYS_MAXARGS=32 CONFIG_SYS_PBSIZE=544 -# CONFIG_CMD_BOOTD is not set # CONFIG_BOOTM_PLAN9 is not set # CONFIG_BOOTM_RTEMS is not set # CONFIG_CMD_IMI is not set -- 2.25.1

2 1

[PATCH] Makefile: Reduce usage of custom mpc85xx u-boot.bin target
by Pali Rohár 03 Aug '22

03 Aug '22

Building of final u-boot.bin binary for mpc85xx via binman is needed only when inserting DTB binary in the middle of the u-boot ELF binary (before .bootpg and .resetvec ELF sections). These requirements are met when CONFIG_MPC85XX_HAVE_RESET_VECTOR is enabled (= generating .bootpg/.resetvec sections) and CONFIG_OF_SEPARATE is enabled (= inserting DTB binary). So in all other cases use standard build procedure instead of custom mpc85xx u-boot.bin Makefile target via binman. Signed-off-by: Pali Rohár <pali(a)kernel.org> --- This patch depends on patch series: https://patchwork.ozlabs.org/project/uboot/list/?series=312130 --- Makefile | 11 ++++------- arch/powerpc/Kconfig | 2 +- arch/powerpc/dts/u-boot.dtsi | 8 ++++---- 3 files changed, 9 insertions(+), 12 deletions(-) diff --git a/Makefile b/Makefile index b1b08695cc47..d659c8118cba 100644 --- a/Makefile +++ b/Makefile @@ -1214,12 +1214,12 @@ else ifeq ($(CONFIG_OF_SEPARATE).$(CONFIG_OF_OMIT_DTB),y.) u-boot-dtb.bin: u-boot-nodtb.bin dts/dt.dtb FORCE $(call if_changed,cat) -ifneq ($(CONFIG_MPC85xx)$(CONFIG_OF_SEPARATE),yy) +ifneq ($(CONFIG_MPC85XX_HAVE_RESET_VECTOR)$(CONFIG_OF_SEPARATE),yy) u-boot.bin: u-boot-dtb.bin FORCE $(call if_changed,copy) endif -else ifneq ($(CONFIG_MPC85xx)$(CONFIG_OF_SEPARATE),yy) +else ifneq ($(CONFIG_MPC85XX_HAVE_RESET_VECTOR)$(CONFIG_OF_SEPARATE),yy) u-boot.bin: u-boot-nodtb.bin FORCE $(call if_changed,copy) endif @@ -1622,17 +1622,14 @@ u-boot-with-nand-spl.sfp: u-boot-spl-padx4.sfp u-boot.img FORCE endif -ifeq ($(CONFIG_MPC85xx)$(CONFIG_OF_SEPARATE),yy) -u-boot.bin: u-boot-nodtb.bin u-boot.dtb \ - $(if $(CONFIG_MPC85XX_HAVE_RESET_VECTOR), u-boot-br.bin) FORCE +ifeq ($(CONFIG_MPC85XX_HAVE_RESET_VECTOR)$(CONFIG_OF_SEPARATE),yy) +u-boot.bin: u-boot-nodtb.bin u-boot.dtb u-boot-br.bin FORCE $(call if_changed,binman) -ifeq ($(CONFIG_MPC85XX_HAVE_RESET_VECTOR),y) OBJCOPYFLAGS_u-boot-br.bin := -O binary -j .bootpg -j .resetvec u-boot-br.bin: u-boot FORCE $(call if_changed,objcopy) endif -endif quiet_cmd_ldr = LD $@ cmd_ldr = $(LD) $(LDFLAGS_$(@F)) \ diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig index 737bdd8edb41..e0e2662f4c84 100644 --- a/arch/powerpc/Kconfig +++ b/arch/powerpc/Kconfig @@ -20,7 +20,7 @@ config MPC85xx select CREATE_ARCH_SYMLINK select SYS_FSL_DDR select SYS_FSL_DDR_BE - select BINMAN if OF_SEPARATE + select BINMAN if MPC85XX_HAVE_RESET_VECTOR && OF_SEPARATE imply CMD_HASH imply CMD_IRQ imply USB_EHCI_HCD if USB diff --git a/arch/powerpc/dts/u-boot.dtsi b/arch/powerpc/dts/u-boot.dtsi index 0251afddca82..6588bb7fa5e2 100644 --- a/arch/powerpc/dts/u-boot.dtsi +++ b/arch/powerpc/dts/u-boot.dtsi @@ -5,6 +5,8 @@ #include <config.h> +#if defined(CONFIG_MPC85XX_HAVE_RESET_VECTOR) && defined(CONFIG_OF_SEPARATE) + / { binman { filename = "u-boot.bin"; @@ -19,17 +21,15 @@ }; u-boot-dtb-with-ucode { -#ifdef CONFIG_MPC85xx align = <4>; -#endif }; -#ifdef CONFIG_MPC85XX_HAVE_RESET_VECTOR #ifndef CONFIG_RESET_VECTOR_ADDRESS #define CONFIG_RESET_VECTOR_ADDRESS 0xfffffffc #endif powerpc-mpc85xx-bootpg-resetvec { offset = <(CONFIG_RESET_VECTOR_ADDRESS - 0xffc)>; }; -#endif }; }; + +#endif -- 2.20.1

1 0

[PATCH] riscv: cpu: set gp before board_init_f_init_reserve
by Nikita Shubin 03 Aug '22

03 Aug '22

From: Nikita Shubin <n.shubin(a)yadro.com> Restore global pointer before board_init_f_init_reserve call, as "a0" can be set in harts_early_init call and we end up with invalid global pointer. Signed-off-by: Nikita Shubin <n.shubin(a)yadro.com> --- arch/riscv/cpu/start.S | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/riscv/cpu/start.S b/arch/riscv/cpu/start.S index 76850ec9be..623de57551 100644 --- a/arch/riscv/cpu/start.S +++ b/arch/riscv/cpu/start.S @@ -142,6 +142,7 @@ call_harts_early_init: bnez tp, secondary_hart_loop #endif + mv a0, s0 jal board_init_f_init_reserve SREG s1, GD_FIRMWARE_FDT_ADDR(gp) -- 2.35.1

2 3

Building documentation on Windows
by Martin Bonner 03 Aug '22

03 Aug '22

I am trying to build documentation on Windows with `make htmldocs`. In additions to the modules I have installed for building tools (such as fit-check-sign), I have installed pacman -S mingw-w64-x86_64-python-sphinx pacman -S mingw-w64-x86_64-fontconfig pacman -S mingw-w64-x86_64-python-sphinx_rtd_theme in response to errors. My latest error is: Extension error: Could not import extension kfigure (exception: No module named 'six') make[1]: *** [doc/Makefile:68: htmldocs] Error 2 make: *** [Makefile:2412: htmldocs] Error 2 so my natural instinct was `pip install six` - but this failed with Requirement already satisfied: six in /usr/lib/python3.10/site-packages (1.16.0) Does anyone have any suggestion how to fix this? Martin

1 0

[PATCH v2 1/1] doc: add more details for crash dump analysis
by Heinrich Schuchardt 03 Aug '22

03 Aug '22

* describe crashs in UEFI binaries * provide architechture specific information for the sandbox and RISC-V Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt(a)canonical.com> --- v2: describe CONFIG_ARMV8_SPL_EXCEPTION_VECTORS --- doc/develop/crash_dumps.rst | 60 +++++++++++++++++++++++++++++++++++++ 1 file changed, 60 insertions(+) diff --git a/doc/develop/crash_dumps.rst b/doc/develop/crash_dumps.rst index 85030b4e36..b7b81bdce5 100644 --- a/doc/develop/crash_dumps.rst +++ b/doc/develop/crash_dumps.rst @@ -122,3 +122,63 @@ If we want to dive deeper, we can disassemble the U-Boot binary:: This example is based on the ARMv8 architecture but the same procedures can be used on other architectures as well. + +Crashs in UEFI binaries +----------------------- + +If UEFI images are loaded when a crash occurs, their load addresses are +displayed. If the process counter points to an address in a loaded UEFI +binary, the relative process counter position is indicated. Here is an +example executed on the U-Boot sandbox:: + + => load host 0:1 $kernel_addr_r buggy.efi + 5632 bytes read in 0 ms + => bootefi $kernel_addr_r + Booting /buggy.efi + Buggy world! + + Segmentation violation + pc = 0x19fc264c, pc_reloc = 0xffffaa4688b1664c + + UEFI image [0x0000000019fc0000:0x0000000019fc6137] pc=0x264c '/buggy.efi' + +The crash occured in UEFI binary buggy.efi at relative position 0x264c. +Disassembly may be used to find the actual source code location:: + + $ x86_64-linux-gnu-objdump -S -D buggy_efi.so + + 0000000000002640 <memset>: + 2640: f3 0f 1e fa endbr64 + 2644: 48 89 f8 mov %rdi,%rax + 2647: 48 89 f9 mov %rdi,%rcx + 264a: eb 0b jmp 2657 <memset+0x17> + 264c: 40 88 31 mov %sil,(%rcx) + +Architecture specific details +----------------------------- + +ARMv8 +~~~~~ + +On the ARM 64-bit architecture CONFIG_ARMV8_SPL_EXCEPTION_VECTORS controls +if the exception vector tables are set up in the Secondary Program Loader (SPL). +Without initialization of the tables crash dumps cannot be shown. The feature is +disabled by default on most boards to reduce the size of the SPL. + +RISC-V +~~~~~~ + +On the RISC-V architecture CONFIG_SHOW_REGS=y has to be specified to show +all registers in crash dumps. + +Sandbox +~~~~~~~ + +The sandbox U-Boot binary must be invoked with parameter *-S* to display crash +dumps: + +.. code-block:: bash + + ./u-boot -S -T + +Only with CONFIG_SANDBOX_CRASH_RESET=y the sandbox reboots after a crash. -- 2.36.1

1 0

[PATCH 1/1] doc: add more details for crash dump analysis
by Heinrich Schuchardt 02 Aug '22

02 Aug '22

* describe crashs in UEFI binaries * provide architechture specific information for the sandbox and RISC-V Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt(a)canonical.com> --- doc/develop/crash_dumps.rst | 52 +++++++++++++++++++++++++++++++++++++ 1 file changed, 52 insertions(+) diff --git a/doc/develop/crash_dumps.rst b/doc/develop/crash_dumps.rst index 85030b4e36..9e7ae3d751 100644 --- a/doc/develop/crash_dumps.rst +++ b/doc/develop/crash_dumps.rst @@ -122,3 +122,55 @@ If we want to dive deeper, we can disassemble the U-Boot binary:: This example is based on the ARMv8 architecture but the same procedures can be used on other architectures as well. + +Crashs in UEFI binaries +----------------------- + +If UEFI images are loaded when a crash occurs, their load addresses are +displayed. If the process counter points to an address in a loaded UEFI +binary, the relative process counter position is indicated. Here is an +example executed on the U-Boot sandbox:: + + => load host 0:1 $kernel_addr_r buggy.efi + 5632 bytes read in 0 ms + => bootefi $kernel_addr_r + Booting /buggy.efi + Buggy world! + + Segmentation violation + pc = 0x19fc264c, pc_reloc = 0xffffaa4688b1664c + + UEFI image [0x0000000019fc0000:0x0000000019fc6137] pc=0x264c '/buggy.efi' + +The crash occured in UEFI binary buggy.efi at relative position 0x264c. +Disassembly may be used to find the actual source code location:: + + $ x86_64-linux-gnu-objdump -S -D buggy_efi.so + + 0000000000002640 <memset>: + 2640: f3 0f 1e fa endbr64 + 2644: 48 89 f8 mov %rdi,%rax + 2647: 48 89 f9 mov %rdi,%rcx + 264a: eb 0b jmp 2657 <memset+0x17> + 264c: 40 88 31 mov %sil,(%rcx) + +Architecture specific details +----------------------------- + +RISC-V +~~~~~~ + +On the RISC-V architecture CONFIG_SHOW_REGS=y has to be specified to show +all registers in crash dumps. + +Sandbox +~~~~~~~ + +The sandbox U-Boot binary must be invoked with parameter *-S* to display crash +dumps: + +.. code-block:: bash + + ./u-boot -S -T + +Only with CONFIG_SANDBOX_CRASH_RESET=y the sandbox reboots after a crash. -- 2.36.1

2 1

[PATCH] dm: rtc: Avoid a race in the rtc_reset test
by Simon Glass 02 Aug '22

02 Aug '22

Since resetting the RTC on sandbox causes it to read the base time from the system, we cannot rely on this being unchanged since it was last read. Allow for a one-second delay. Fixes: https://source.denx.de/u-boot/u-boot/-/issues/4 Reported-by: Bin Meng <bmeng.cn(a)gmail.com> Reported-by: Tom Rini <trini(a)konsulko.com> Suggested-by: Rasmus Villemoes <rasmus.villemoes(a)prevas.dk> Signed-off-by: Simon Glass <sjg(a)chromium.org> --- test/dm/rtc.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) diff --git a/test/dm/rtc.c b/test/dm/rtc.c index c7f9f8f0ce7..403bf5c640a 100644 --- a/test/dm/rtc.c +++ b/test/dm/rtc.c @@ -245,16 +245,21 @@ static int dm_test_rtc_reset(struct unit_test_state *uts) ut_assertok(dm_rtc_get(dev, &now)); ut_assertok(i2c_emul_find(dev, &emul)); - ut_assert(emul != NULL); + ut_assertnonnull(emul); old_base_time = sandbox_i2c_rtc_get_set_base_time(emul, 0); ut_asserteq(0, sandbox_i2c_rtc_get_set_base_time(emul, -1)); - /* Resetting the RTC should put he base time back to normal */ + /* + * Resetting the RTC should put the base time back to normal. Allow for + * a one-second adjustment in case the time flips over while this + * test process is pre-empted, since reset_time() in i2c_rtc_emul.c + * reads the time from the OS. + */ ut_assertok(dm_rtc_reset(dev)); base_time = sandbox_i2c_rtc_get_set_base_time(emul, -1); - ut_asserteq(old_base_time, base_time); + ut_assert(base_time - old_base_time <= 1); return 0; } -- 2.37.1.455.g008518b4e5-goog

3 8

[PATCH 1/1] doc: crash_dumps.rst missing documentation link
by Heinrich Schuchardt 02 Aug '22

02 Aug '22

Add link to usage/cmd/exception.rst. Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt(a)canonical.com> --- doc/develop/crash_dumps.rst | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/doc/develop/crash_dumps.rst b/doc/develop/crash_dumps.rst index 9e7ae3d751..86e13a8085 100644 --- a/doc/develop/crash_dumps.rst +++ b/doc/develop/crash_dumps.rst @@ -12,8 +12,8 @@ Creating a crash dump voluntarily --------------------------------- For describing the analysis of a crash dump we need an example. U-Boot comes -with a command 'exception' that comes in handy here. The command is enabled -by:: +with a command :doc:`exception <../usage/cmd/exception>` that comes in handy +here. The command is enabled by:: CONFIG_CMD_EXCEPTION=y -- 2.36.1

1 0

Build with CONFIG_TOOLS_LIBCRYPTO disabled is broken
by Alexander Dahl 02 Aug '22

02 Aug '22

Hei hei, once again I tried building U-Boot with CONFIG_TOOLS_LIBCRYPTO disabled and with no libssl-dev installed on the build machine. It does not work. Even porting the patch disabling the build with kwbimage (see links below) does not help anymore, because since commit 6e052d1cbafb ("mkimage: add public key for image pre-load stage") there's a new hard dependency on host openssl in tools/image-host.c and the corresponding kconfig symbol CONFIG_TOOLS_FIT can not be disabled through menuconfig. Just read those discussions again, but not sure what can be done to proceed here? Can I help somehow? (Board to be built here is based on at91 sama5d27 SoC, no fit image required.) Greets Alex Link: https://patchwork.ozlabs.org/project/uboot/patch/ 20220111153120.1276641-1-marex(a)denx.de/ Link: https://patchwork.ozlabs.org/project/uboot/patch/20211021093304.25399-1-pal…

3 3