April 2022 - U-Boot

[PATCH v3 00/29] arm: semihosting: Cleanups and new features
by Sean Anderson 04 Apr '22

04 Apr '22

This cleans up the semihosting code and adds the following new features: - hostfs support (like sandbox) - support for being used as a SPL boot device - serial device support - falling back to normal drivers if semihosting is disabled The main device affected by these changes is vexpress64, so I'd appreciate if Andre (or anyone else) could try booting. These changes are motivated by bringup for ls1046a. When forcing JTAG boot, this device disables most communication peripherals, including serial and ethernet devices. This appears to be fixed in later generation devices, but we are stuck with it for now. Semihosting provides an easy way to run a few console commands. The patches in this series are organized as follows: 0-4: rST conversions and other documentation updates 5-9: Semihosting cleanups 10-14: Filesystem support (including SPL boot device) 15-16: Serial support 16: Documentation update 17: JTAG boot support for LS1046A 19-25: Semihosting fallback 26-29: DM puts support The last two groups of patches are "bonus;" the first 17 patches stand on their own. The last two groups could be broken out as separate series, but I have kept them in this one to help with my sanity (and not have to deal with too many outstanding series). Patch 14 depends on [1] to apply cleanly. Patch 17 depends on [2] for correctness. This series should be applied to u-boot/next (in particular, the EROFS series must have been applied). [1] https://lore.kernel.org/u-boot/CACRpkdZ+9fmNjC_mvrbPa9-iuTQVd8UkJ7Zpe7cL0c5… [2] https://lore.kernel.org/u-boot/20220222183840.1355337-2-sean.anderson@seco.… Changes in v3: - Add a config for puts to reduce the impact on the (vast majority) of boards which don't need it. - Add additional documentation for QEMU and OpenOCD - Add dummy device number for hostfs - Add test for putc/puts - Fix loading the wrong EL's SPSR - Fix null pointer dereference in _serial_puts caused by a missing return. - Make FALLBACK depend on ARM64 - Make puts return the number of characters written on success, instead of reusing the len parameter. - Reorder Kconfigs to group SEMIHOSTING_FALLBACK after SEMIHOSTING (instead of after SPL_SEMIHOSTING_FALLBACK) - Update commit message to describe an alternate implementation - Use SYS_ERRNO instead of SYS_ISERROR, since the latter is not implemented in older QEMUs. Changes in v2: - Add implementation of puts for DM - Add migration instructions for smhload - Add myself as a reviewer - Add semihosting fallback implementation - Compile arch/arm/lib/semihosting.o in SPL - Document debug uart - Fix baud numbers being off by 10 - Fix typos in commit message - Make CONFIG_SPL_SEMIHOSTING depend on SPL - Rebase on Andre's series - Rename non-DM driver struct to match format of other drivers Sean Anderson (29): doc: Convert semihosting readme to rST nxp: ls1046ardb: Convert README to rST doc: ls1046ardb: Expand boot mode section doc: ls1046ardb: Document debug uart arm: smh: Add semihosting entry to MAINTAINERS arm: smh: Export semihosting functions arm: smh: Use numeric modes for smh_open arm: smh: Return errno on error arm: smh: Document functions in header arm: smh: Add some file manipulation commands spl: Add semihosting boot method fs: Add semihosting filesystem cmd: fdt: Use start/size for chosen instead of start/end arm: smh: Remove smhload command arm: smh: Add some functions for working with the host console serial: Add semihosting driver doc: smh: Update semihosting documentation ls1046ardb: Add support for JTAG boot arm64: Save esr in pt_regs arm64: Save spsr in pt_regs arm64: Import some ESR and SPSR defines from Linux arm: smh: Add option to detect semihosting arm64: Catch non-emulated semihosting calls serial: smh: Initialize serial only if semihosting is enabled arm64: ls1046a: Support semihosting fallback serial: dm: Add support for puts serial: sandbox: Implement puts test: serial: Add test for putc/puts serial: smh: Implement puts for DM MAINTAINERS | 5 + arch/arm/Kconfig | 47 +++- arch/arm/cpu/armv8/exceptions.S | 9 +- arch/arm/cpu/armv8/fsl-layerscape/spl.c | 3 + arch/arm/include/asm/esr.h | 343 ++++++++++++++++++++++++ arch/arm/include/asm/proc-armv/ptrace.h | 75 ++++++ arch/arm/include/asm/spl.h | 1 + arch/arm/include/asm/u-boot-arm.h | 7 +- arch/arm/lib/Makefile | 2 +- arch/arm/lib/interrupts_64.c | 80 ++++-- arch/arm/lib/semihosting.c | 230 ++++++++-------- arch/arm/mach-imx/imx8m/soc.c | 4 +- arch/sandbox/include/asm/serial.h | 6 + board/freescale/ls1046ardb/MAINTAINERS | 1 + board/freescale/ls1046ardb/README | 76 ------ board/freescale/ls1046ardb/ls1046ardb.c | 11 + cmd/fdt.c | 6 +- common/spl/Makefile | 1 + common/spl/spl_semihosting.c | 71 +++++ disk/part.c | 4 +- doc/README.semihosting | 38 --- doc/board/nxp/index.rst | 1 + doc/board/nxp/ls1046ardb.rst | 191 +++++++++++++ doc/usage/index.rst | 1 + doc/usage/semihosting.rst | 107 ++++++++ drivers/serial/Kconfig | 37 +++ drivers/serial/Makefile | 1 + drivers/serial/sandbox.c | 38 ++- drivers/serial/serial-uclass.c | 26 +- drivers/serial/serial.c | 2 + drivers/serial/serial_semihosting.c | 187 +++++++++++++ fs/Makefile | 1 + fs/fs.c | 20 ++ fs/semihostingfs.c | 115 ++++++++ include/configs/ls1046ardb.h | 2 + include/configs/vexpress_aemv8.h | 10 +- include/fs.h | 1 + include/semihosting.h | 149 ++++++++++ include/semihostingfs.h | 21 ++ include/serial.h | 19 ++ test/dm/serial.c | 19 ++ 41 files changed, 1689 insertions(+), 279 deletions(-) create mode 100644 arch/arm/include/asm/esr.h delete mode 100644 board/freescale/ls1046ardb/README create mode 100644 common/spl/spl_semihosting.c delete mode 100644 doc/README.semihosting create mode 100644 doc/board/nxp/ls1046ardb.rst create mode 100644 doc/usage/semihosting.rst create mode 100644 drivers/serial/serial_semihosting.c create mode 100644 fs/semihostingfs.c create mode 100644 include/semihosting.h create mode 100644 include/semihostingfs.h -- 2.25.1

4 68

The contradictory nature of spl_nand_fit_read
by Sean Anderson 04 Apr '22

04 Apr '22

Hi all, I don't understand how spl_nand_fit_read is supposed to work. This function has been seemingly hacked together by multiple people over the years, and it (presumably) (somehow) works despite mass confusion of units. I tried to do some refactoring, but the contradictions make it very difficult to determine what is a bug and what is intentional. Lets start with the basics. spl_nand_fit_read is used to implement spl_load_info.load. I've reproduced the documentation for this function below: > read() - Read from device > > @load: Information about the load state > @sector: Sector number to read from (each @load->bl_len bytes) > @count: Number of sectors to read > @buf: Buffer to read into > @return number of sectors read, 0 on error In particular, both @sector and @count are in units of load.bl_len. In addition, the return value should be @count on success. > static ulong spl_nand_fit_read(struct spl_load_info *load, ulong offs, > ulong size, void *dst) > { > int err; > #ifdef CONFIG_SYS_NAND_BLOCK_SIZE The following logic was introduced in commit 9f6a14c47f ("spl: fit: nand: fix fit loading in case of bad blocks"). However, at the time it was not guarded by the above ifdef. nand_spl_adjust_offset is not defined for all nand drivers. It is defined for at least mxs, am335x, atmel, and simple. Additionally, some drivers (such as rockchip) implement bad block skipping in nand_spl_load_image. It's not at all clear to me that there is common config which determines whether nand_spl_adjust_offset is implemented. Nevertheless, in commit aa0032f672 ("spl: fit: nand: allow for non-page-aligned elements"), the above directive selects different code if CONFIG_SYS_NAND_BLOCK_SIZE is defined. > ulong sector; > > sector = *(int *)load->priv; This is the offset passed to nand_spl_load_image. This offset is in units of bytes, and is aligned to the block size. However, it is *not* set if CONFIG_SPL_LOAD_IMX_CONTAINER is enabled. Oops. > offs = sector + nand_spl_adjust_offset(sector, offs - sector); The documentation for this this function is as follows > nand_spl_adjust_offset - Adjust offset from a starting sector > @sector: Address of the sector > @offs: Offset starting from @sector > > If one or more bad blocks are in the address space between @sector > and @sector + @offs, @offs is increased by the NAND block size for > each bad block found. In particular, note that offs is in units of bytes. However, we know from above that at this point in the function, offs is in units of bl_len. Oops. > #else > offs *= load->bl_len; > size *= load->bl_len; This code path adjusts both offs and size to be in units of > #endif > err = nand_spl_load_image(offs, size, dst); So by the time we get here, one code path is in units of bytes, and the other is in units of bl_len. nand_spl_load_image seems to expect bytes (though there is no documentation, so I can't be sure). Which of course begs the question: how did the code introduced in 9f6a14c47f ever work? > if (err) > return 0; > > return size / load->bl_len; And of course, this should only be done if size is in units of bytes. > } I have some patches for what I think are bugs, but the logic of this function is so confused that I am unable to determine if they actually are bugs. I would greatly appreciate if some of the authors of the mentioned commits could comment on their intent, and what they thought the units of offs and size were. --Sean

4 5

[PATCH] mmc: fsl_esdhc: Define macro ESDHCCTL_SNOOP for Snoop attribute
by Pali Rohár 04 Apr '22

04 Apr '22

Signed-off-by: Pali Rohár <pali(a)kernel.org> --- drivers/mmc/fsl_esdhc.c | 2 +- include/fsl_esdhc.h | 1 + 2 files changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/mmc/fsl_esdhc.c b/drivers/mmc/fsl_esdhc.c index 05a6d0ce1562..fdf2cc290e06 100644 --- a/drivers/mmc/fsl_esdhc.c +++ b/drivers/mmc/fsl_esdhc.c @@ -724,7 +724,7 @@ static void esdhc_enable_cache_snooping(struct fsl_esdhc *regs) setbits_be32(&sysconf->sdhccr, 0x02000000); #else - esdhc_write32(&regs->esdhcctl, 0x00000040); + esdhc_write32(&regs->esdhcctl, ESDHCCTL_SNOOP); #endif } diff --git a/include/fsl_esdhc.h b/include/fsl_esdhc.h index f86afe5dad86..7ab1460abc6e 100644 --- a/include/fsl_esdhc.h +++ b/include/fsl_esdhc.h @@ -76,6 +76,7 @@ /* eSDHC control register */ #define ESDHCCTL 0x0002e40c +#define ESDHCCTL_SNOOP (0x00000040) #define ESDHCCTL_PCS (0x00080000) #define ESDHCCTL_FAF (0x00040000) -- 2.20.1

1 0

[PATCH 0/3] efi_loader: setting boot device
by Heinrich Schuchardt 04 Apr '22

04 Apr '22

Up to now the bootefi command uses the last file loaded to determine the boot partition. This has led to user irritation when the fdt has been loaded from another partition after the EFI binary. Before setting the boot device from a loaded file check if it is a PE-COFF image. The first patch carves out the test function. The second make use of the PE-COFF check. The third add the display of the boot device and file path for easier testing. Heinrich Schuchardt (3): efi_loader: print boot device and file path in helloworld efi_loader: carve out efi_check_pe() efi_loader: setting boot device cmd/bootefi.c | 14 ++- doc/uefi/uefi.rst | 11 +- fs/fs.c | 3 +- include/efi_loader.h | 8 +- lib/efi_loader/efi_image_loader.c | 80 ++++++++------ lib/efi_loader/helloworld.c | 167 ++++++++++++++++++++++++------ net/tftp.c | 9 +- 7 files changed, 211 insertions(+), 81 deletions(-) -- 2.29.2

4 18

[ANN] U-Boot v2022.04 released
by Tom Rini 04 Apr '22

04 Apr '22

Hey all, It is release day and here is v2022.04. As a general reminder, there's still many DM migrations that need to be done and while we try and have a non-fatal warning message printed out when building a given platform, some things can slip through the cracks. Checking out your boards config file manually is probably a good idea and seeing if there's any CONFIG_DM_xxx options that should get enabled. Since the last time I iterated on new migration deadlines, there's now CONFIG_TIMER is deprecated, and DM_SERIAL and DM_SCSI need to be migrated. I've been making more progress on CONFIG migration but more help here is always appreciated. And related, buildman needs a little help moving away from needing to use tools/genboards.py to make the old old old style boards.cfg file and instead use kconfiglib directly. In terms of a changelog, git log --merges v2022.04-rc5..v2022.04 contains what I've pulled since the last RC or: git log --merges v2022.01..v2022.04 for changes since the last full release. As always, more details in pull requests (or the tags referenced by them) will result in more details here. With this, the merge window is now open and I'll be merging in the next branch shortly. v2022.07 is scheduled for release on Monday, July 4th 2022 and the merge window closes on April 25th. Thanks all! -- Tom

1 0

[PATCH] env: do not fail on non-existing env.in file
by Andrey Zhizhikin 04 Apr '22

04 Apr '22

If include/generated/env.in does not exist, which is a typical case for clean build, quiet_cmd_gen_envp command tries to delete this file unconditionally. This produces following warning during the build: ENVP include/generated/env.in rm: cannot remove 'include/generated/env.in': No such file or directory Add '-f' option to the `rm` command to not complain if file does not exist. Fixes: f432eb6d8a9d ("env: Avoid using a leftover text-environment file") Signed-off-by: Andrey Zhizhikin <andrey.zhizhikin(a)leica-geosystems.com> --- Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Makefile b/Makefile index 4672147318..868445bf33 100644 --- a/Makefile +++ b/Makefile @@ -1846,7 +1846,7 @@ quiet_cmd_gen_envp = ENVP $@ -I$(srctree)/arch/$(ARCH)/include \ $< -o $@; \ else \ - rm $@; \ + rm -f $@; \ touch $@ ; \ fi include/generated/env.in: include/generated/env.txt FORCE -- 2.25.1

2 1

[PULL next] u-boot-at91-2022.07-a
by Eugen.Hristev＠microchip.com 04 Apr '22

04 Apr '22

Hello Tom, Please pull tag u-boot-at91-2022.07-a , the first set of new at91 features for the next cycle 2022.07 . This feature set includes the new driver for the Atmel TCB timer, alignment in DT for sama7g5 and sama7g5ek board, one Kconfig conversion for external reset, and the usage of Galois tables from ROM for sama5d2 device. Thanks, Eugen The following changes since commit d2e5250be49fce4653689c41a5dc7e2d7e7ecf33: Merge tag 'next-20220328' of https://source.denx.de/u-boot/custodians/u-boot-video into next (2022-03-28 17:04:45 -0400) are available in the Git repository at: https://source.denx.de/u-boot/custodians/u-boot-at91.git tags/u-boot-at91-2022.07-a for you to fetch changes up to ef9d9b2501ad9e0601bd292c4a6c8a7ba8f3a909: configs: sama5d2: enable option CONFIG_ATMEL_TCB_TIMER (2022-04-01 15:42:47 +0300) ---------------------------------------------------------------- First set of u-boot-at91 features for the 2022.07 cycle ---------------------------------------------------------------- Clément Léger (5): timer: atmel_tcb_timer: add atmel_tcb driver ARM: dts: at91: sama5d2: add AIC node ARM: dts: at91: sama5d2: add TCB node ARM: dts: at91: sama5d2: add u-boot, dm-pre-reloc property for TCB configs: sama5d2: enable option CONFIG_ATMEL_TCB_TIMER Eugen Hristev (2): ARM: dts: at91: rename sama7g5ek.dts to at91-sama7g5ek.dts ARM: dts: at91: sama7g5/sama7g5ek: align with Linux DT Mihai Sain (1): board: at91: sama5d2: set blue led on at boot time Tudor Ambarus (3): ARM: at91: sama5d2: Enable the use of Galois Tables from ROM configs: Convert AT91RESET_EXTRST to Kconfig ARM: at91: sama7g5: Reset external devices at software reset MAINTAINERS | 1 + arch/arm/dts/Makefile | 2 +- ...g5ek-u-boot.dtsi => at91-sama7g5ek-u-boot.dtsi} | 27 +- arch/arm/dts/at91-sama7g5ek.dts | 804 ++++++++++++++++++ arch/arm/dts/sama5d2.dtsi | 28 + arch/arm/dts/sama7g5.dtsi | 942 +++++++++++++++++---- arch/arm/dts/sama7g5ek.dts | 275 ------ arch/arm/mach-at91/Kconfig | 4 + arch/arm/mach-at91/armv7/Makefile | 2 + arch/arm/mach-at91/include/mach/sama5d2.h | 10 +- board/atmel/sama5d27_som1_ek/sama5d27_som1_ek.c | 9 + .../atmel/sama5d27_wlsom1_ek/sama5d27_wlsom1_ek.c | 9 + board/atmel/sama5d2_icp/sama5d2_icp.c | 9 + board/atmel/sama5d2_ptc_ek/sama5d2_ptc_ek.c | 9 + board/atmel/sama5d2_xplained/sama5d2_xplained.c | 9 + configs/sama5d2_icp_mmc_defconfig | 2 +- configs/sama5d2_ptc_ek_mmc_defconfig | 2 +- configs/sama5d2_ptc_ek_nandflash_defconfig | 2 +- configs/sama5d2_xplained_emmc_defconfig | 2 +- configs/sama5d2_xplained_mmc_defconfig | 2 +- configs/sama5d2_xplained_qspiflash_defconfig | 2 +- configs/sama5d2_xplained_spiflash_defconfig | 2 +- configs/sama7g5ek_mmc1_defconfig | 2 +- configs/sama7g5ek_mmc_defconfig | 2 +- drivers/timer/Kconfig | 8 + drivers/timer/Makefile | 1 + drivers/timer/atmel_tcb_timer.c | 161 ++++ 27 files changed, 1874 insertions(+), 454 deletions(-) rename arch/arm/dts/{sama7g5ek-u-boot.dtsi => at91-sama7g5ek-u-boot.dtsi} (57%) create mode 100644 arch/arm/dts/at91-sama7g5ek.dts delete mode 100644 arch/arm/dts/sama7g5ek.dts create mode 100644 drivers/timer/atmel_tcb_timer.c

2 1

[PATCH 0/2] sf: dataflash: add support for AT45DB641E
by Luca Ellero 04 Apr '22

04 Apr '22

Since AT45DB641E uses "extended device information" for reading JEDEC ID we must first introduce support for this feature. All code is borrowed from Linux kernel source code: drivers/mtd/devices/mtd_dataflash.c Luca Ellero (2): sf: dataflash: add support for "extended ID" sf: dataflash: add support for AT45DB641E drivers/mtd/spi/sf_dataflash.c | 101 ++++++++++++++++++++++----------- 1 file changed, 67 insertions(+), 34 deletions(-) -- 2.17.1

2 4

Does enabling the Driver Model makes the boot slower ?
by Abder 04 Apr '22

04 Apr '22

Hi, Recently I have worked on porting an imx6 based board's u-boot2014 to u-boot2020 and I enabled all the used drivers to use DM: GPIO, ETH, USB, SPI, MMC, WDT, MTD and REGULATOR. However, I noticed that the boot of the new u-boot is somehow slower than the old one (not by much though) ! Personally, I wouldn't expect it to go as fast as the old one nor faster, as the driver model does more device management than legacy drivers and also each DM initialization requires some fdt nodes searching. I would like to know your opinion on this, if this is mainly due to using the driver model ? And eventually, if by using legacy drivers could it make the boot faster than what we've got right now ? Thank you. Best regards -- Abder

1 0

[PATCH v2 00/23] sunxi: Add and use a pinctrl driver
by Samuel Holland 04 Apr '22

04 Apr '22

This series resolves some longstanding TODOs by implementing a pinctrl driver for sunxi platforms and converting DM drivers to use it. Changes in v2: - Merge all SoC drivers into one file and one U_BOOT_DRIVER. - Add a consumer for the APB bus clock - Add support for the F1C100s SoC Samuel Holland (23): sunxi: pinctrl: Create the driver skeleton sunxi: pinctrl: Implement pin muxing functions sunxi: pinctrl: Implement get_pin_muxing function sunxi: pinctrl: Implement pin configuration pinctrl: sunxi: Add UART pinmuxes sunxi: Skip non-DM UART pin setup when PINCTRL=y pinctrl: sunxi: Add sun4i EMAC pinmuxes net: sunxi_emac: Remove non-DM pin setup pinctrl: sunxi: Add sunxi GMAC pinmuxes sunxi: Remove non-DM GMAC pin setup pinctrl: sunxi: Add sun8i EMAC pinmuxes net: sun8i_emac: Remove non-DM pin setup pinctrl: sunxi: Add I2C pinmuxes sunxi: Remove options and setup code for I2C2-I2C4 sunxi: Remove non-DM I2C clock/pin setup from U-Boot i2c: sun6i_p2wi: Only do non-DM pin setup for non-DM I2C i2c: sun8i_rsb: Only do non-DM pin setup for non-DM I2C pinctrl: sunxi: Add MMC pinmuxes sunxi: Remove non-DM MMC pin setup pinctrl: sunxi: Add the A64 PWM pinmux pwm: sunxi: Remove non-DM pin setup pinctrl: sunxi: Add SPI0 pinmuxes spi: sun4i_spi: Remove non-DM pin setup MAINTAINERS | 1 + arch/arm/Kconfig | 1 + arch/arm/include/asm/arch-sunxi/gpio.h | 17 +- arch/arm/include/asm/arch-sunxi/i2c.h | 11 +- arch/arm/mach-sunxi/Kconfig | 22 - arch/arm/mach-sunxi/board.c | 2 + board/sunxi/board.c | 67 -- board/sunxi/gmac.c | 55 -- drivers/gpio/sunxi_gpio.c | 130 +--- drivers/i2c/sun6i_p2wi.c | 12 +- drivers/i2c/sun8i_rsb.c | 46 +- drivers/net/sun8i_emac.c | 90 --- drivers/net/sunxi_emac.c | 7 +- drivers/pinctrl/Kconfig | 1 + drivers/pinctrl/Makefile | 2 +- drivers/pinctrl/sunxi/Kconfig | 127 ++++ drivers/pinctrl/sunxi/Makefile | 3 + drivers/pinctrl/sunxi/pinctrl-sunxi.c | 887 +++++++++++++++++++++++++ drivers/pwm/sunxi_pwm.c | 11 - drivers/spi/spi-sunxi.c | 84 --- 20 files changed, 1057 insertions(+), 519 deletions(-) create mode 100644 drivers/pinctrl/sunxi/Kconfig create mode 100644 drivers/pinctrl/sunxi/Makefile create mode 100644 drivers/pinctrl/sunxi/pinctrl-sunxi.c -- 2.33.1

3 50