August 2022 - U-Boot

[PATCH 1/5] Nokia RX-51: Fix invalidating zImage kernel format
by Pali Rohár 21 Aug '22

21 Aug '22

Prior starting copy of kernel image to target location, invalidate also zImage magic header. This ensures that on target location would be image with valid header only in the case valid header was also in the source location and copy from source to target finished successfully. Copy is always skipped when kernel image in source location is invalid. Add also comment to the code which explain what is the code doing. Fixes: cc434fccba4c ("Nokia RX-51: Add support for booting kernel in zImage format") Signed-off-by: Pali Rohár <pali(a)kernel.org> --- board/nokia/rx51/lowlevel_init.S | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/board/nokia/rx51/lowlevel_init.S b/board/nokia/rx51/lowlevel_init.S index 11c2cbef89a6..632595f83785 100644 --- a/board/nokia/rx51/lowlevel_init.S +++ b/board/nokia/rx51/lowlevel_init.S @@ -72,7 +72,8 @@ copy_kernel_start: /* remove header in target kernel */ mov r5, #0 - str r5, [r3] + str r5, [r3] /* remove 4 bytes header of kernel uImage */ + str r5, [r3, #36] /* remove 4 bytes header of kernel zImage */ /* check for valid kernel uImage */ ldr r4, [r0] /* r4 - 4 bytes header of kernel */ @@ -93,6 +94,8 @@ copy_kernel_loop: bhi copy_kernel_loop copy_kernel_end: + + /* remove header in source kernel image */ mov r5, #0 str r5, [r0] /* remove 4 bytes header of kernel uImage */ str r5, [r0, #36] /* remove 4 bytes header of kernel zImage */ -- 2.20.1

2 9

Pull request for efi-2022-10-rc3-2
by Heinrich Schuchardt 21 Aug '22

21 Aug '22

Dear Tom, The following changes since commit 1ea6966687e51cee6e273d170ba391dfe7d7f765: Merge tag 'fsl-qoriq-2022-8-17' of https://source.denx.de/u-boot/custodians/u-boot-fsl-qoriq (2022-08-17 12:10:34 -0400) are available in the Git repository at: https://source.denx.de/u-boot/custodians/u-boot-efi.git tags/efi-2022-10-rc3-2 for you to fetch changes up to 2a4fb47533318f26702a8e8a38a002433e75202b: tools: mkeficapsule: use pkg-config for each lib separat (2022-08-20 08:38:12 +0200) Gitlab reported no issues: https://source.denx.de/u-boot/custodians/u-boot-efi/-/pipelines/13176 ---------------------------------------------------------------- Pull request for efi-2022-10-rc3-2 Documentation: * improve description of device probing * describe booting RISC-V with KVM and QEMU UEFI * fix Makefile for mkeficapsule ---------------------------------------------------------------- AKASHI Takahiro (1): efi_loader: disk: remove unused field Heiko Thiery (1): tools: mkeficapsule: use pkg-config for each lib separat Heinrich Schuchardt (1): doc: qemu-riscv: describe booting with QEMU and KVM Michal Suchanek (1): doc: dm: clarify activation. doc/board/emulation/qemu-riscv.rst | 13 ++++++++++ doc/develop/driver-model/design.rst | 47 ++++++++++++++++++++++++++----------- lib/efi_loader/efi_disk.c | 13 ++-------- tools/Makefile | 8 +++++-- 4 files changed, 54 insertions(+), 27 deletions(-)

2 1

[PATCH] doc: dm: clarify activation.
by Michal Suchanek 20 Aug '22

20 Aug '22

Explain when devices should get activated. Signed-off-by: Michal Suchanek <msuchanek(a)suse.de> --- doc/develop/driver-model/design.rst | 22 ++++++++++++++++++++-- 1 file changed, 20 insertions(+), 2 deletions(-) diff --git a/doc/develop/driver-model/design.rst b/doc/develop/driver-model/design.rst index 5f33f9fbb3..c925d21b24 100644 --- a/doc/develop/driver-model/design.rst +++ b/doc/develop/driver-model/design.rst @@ -794,8 +794,26 @@ fall afoul of this rule. Activation/probe ^^^^^^^^^^^^^^^^ -When a device needs to be used, U-Boot activates it, by first reading ofdata -as above and then following these steps (see device_probe()): +To save resources devices in U-Boot are probed lazily. U-Boot is a bootloader, +not an operating system. Many devices are never used during an U-Boot run, and +probing them takes time, requires memory, may add delays to the main loop, etc. + +The device should be probed by the uclass code. Different uclasses are +different but two common use cases can be seen: + + 1. The uclass is asked to look up a specific device, such as SPI bus 0, + first chip select - in this case the returned device should be + activated. + + 2. The uclass is asked to perform a specific function on any device that + supports it, eg. reset the board using any sysreset that can be found - + for this case the core uclass code provides iterators that activate + each device before returning it, and the uclass typically implements a + walk function that iterates over all devices of the uclass and tries + to perform the requested function on each in turn until succesful. + +To activate a device U-Boot first reads ofdata as above and then follows these +steps (see device_probe()): 1. All parent devices are probed. It is not possible to activate a device unless its predecessors (all the way up to the root device) are activated. -- 2.37.1

4 5

[PATCH] tools: mkeficapsule: use pkg-config for each lib separat
by Heiko Thiery 20 Aug '22

20 Aug '22

Signed-off-by: Heiko Thiery <heiko.thiery(a)gmail.com> --- As suggested by AKASHI Takahiro but unfortunatly missed for respin. Thus I do this commit afterwards. tools/Makefile | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/tools/Makefile b/tools/Makefile index 005e7362a3..3626919633 100644 --- a/tools/Makefile +++ b/tools/Makefile @@ -245,9 +245,13 @@ hostprogs-$(CONFIG_ASN1_COMPILER) += asn1_compiler HOSTCFLAGS_asn1_compiler.o = -idirafter $(srctree)/include HOSTCFLAGS_mkeficapsule.o += \ - $(shell pkg-config --cflags gnutls uuid 2> /dev/null || echo "") + $(shell pkg-config --cflags gnutls 2> /dev/null || echo "") +HOSTCFLAGS_mkeficapsule.o += \ + $(shell pkg-config --cflags uuid 2> /dev/null || echo "") +HOSTLDLIBS_mkeficapsule += \ + $(shell pkg-config --libs gnutls 2> /dev/null || echo "-lgnutls") HOSTLDLIBS_mkeficapsule += \ - $(shell pkg-config --libs gnutls uuid 2> /dev/null || echo "-lgnutls -luuid") + $(shell pkg-config --libs uuid 2> /dev/null || echo "-luuid") hostprogs-$(CONFIG_TOOLS_MKEFICAPSULE) += mkeficapsule # We build some files with extra pedantic flags to try to minimize things -- 2.20.1

2 1

removal of MPC8309 support
by Holger Brunck 19 Aug '22

19 Aug '22

Hi, I am going to remove our board support for KMTEGR1 as it is obsolete. It seems it is the only board using MPC8309 CPU and the support for this CPU was introduced due to this board. Should I remove the MPC8309 support completely then? Best regards Holger Brunck

2 1

[PATCH v5 00/21] binman: Rework compression support
by Stefan Herbrechtsmeier 19 Aug '22

19 Aug '22

From: Stefan Herbrechtsmeier <stefan.herbrechtsmeier(a)weidmueller.com> Rework the compression support of binman. Add a length header attribute to dtb entry, manage the compression bintool in the entry class and add support for several compression tools. Changes in v5: - Add commit to avoid duplicates in bintool lists - Add commit to forward AddBintools calls - Add commit to forward AddBintools calls to base class - Add commit to collect bintools before usage - Add commit to check only section data in multi section test - Add commit to add DecompressData function to entry class - Fix decompress - Disable header in testInvalidCompress function - Remove header value in testInvalidCompress - Add commit to move compression bintools creation into test setup - Add commit to select compression bintools in cbfs_util class - Add commit to move compression bintool management into entry class - Use record_missing_bintool function - Use tools.get_bytes function - Add support to DecompressData function - Reuse 85_compress_section.dts file - Remove 236_compress_dtb_missing_bintool.dts file - Use self.comp_bintools dict - Use _CheckBintool(bintool) function - Remove self.fetch_package not None check in fetch function - Add algorithm name to entry.py - Add tool name to ftest.py - Add algorithm name to entry.py - Add tool name to ftest.py - Add algorithm name to entry.py - Add tool name to entry.py and ftest.py - Add algorithm name to entry.py - Add tool name to ftest.py - Test that zstd compress of device-tree files failed - Add algorithm name to entry.py - Add tool name to ftest.py Changes in v4: - Add missing 236_compress_dtb_missing_bintool.dts file Changes in v3: - Move comp_util.py changes (drop with_header) into separate commits. - Add commit to disable compressed data header - Add commit to remove obsolete compressed data header handling - Add commit to support missing compression tools - Use 'tools.get_bytes(0, 64)' instead of 'bytes([0]) * 64' - Check if tool is present - Rename tests - Document class properties Changes in v2: - Add commit to skip elf tests - Reworked to make the feature optional. - Add commit to add compression tests - Add commit to add BintoolPacker class to bintool - Add commit to add bzip2 bintool - Add commit to add gzip bintool - Add commit to add lzop bintool - Add commit to add xz bintool - Add commit to add zstd bintool Stefan Herbrechtsmeier (21): binman: Skip elf tests if python elftools is not available binman: Avoid duplicates in bintool lists binman: Forward AddBintools calls to sub entries in cbfs_util binman: Forward AddBintools calls to base class binman: Collect bintools before usage binman: Check only section data in multi section test binman: Add DecompressData function to entry class binman: Add length header attribute to dtb entry binman: Disable compressed data header binman: Remove obsolete compressed data header handling binman: Move compression bintools creation into test setup binman: Select compression bintools in cbfs_util class binman: Move compression bintool management into entry class binman: Support missing compression tools binman: Add compression tests binman: Add BintoolPacker class to bintool binman: Add bzip2 bintool binman: Add gzip bintool binman: Add lzop bintool binman: Add xz bintool binman: Add zstd bintool tools/binman/bintool.py | 106 ++++++++++++ tools/binman/btool/bzip2.py | 30 ++++ tools/binman/btool/gzip.py | 31 ++++ tools/binman/btool/lzop.py | 30 ++++ tools/binman/btool/xz.py | 31 ++++ tools/binman/btool/zstd.py | 30 ++++ tools/binman/cbfs_util.py | 20 +-- tools/binman/cbfs_util_test.py | 4 +- tools/binman/comp_util.py | 76 --------- tools/binman/control.py | 5 +- tools/binman/elf_test.py | 14 ++ tools/binman/entries.rst | 3 + tools/binman/entry.py | 54 ++++++- tools/binman/entry_test.py | 9 ++ tools/binman/etype/blob_dtb.py | 31 ++++ tools/binman/etype/cbfs.py | 5 + tools/binman/etype/gbb.py | 1 + tools/binman/etype/intel_ifwi.py | 1 + tools/binman/etype/mkimage.py | 1 + tools/binman/etype/section.py | 4 +- tools/binman/etype/vblock.py | 1 + tools/binman/ftest.py | 152 +++++++++++++++--- .../test/235_compress_dtb_prepend_invalid.dts | 17 ++ .../test/236_compress_dtb_prepend_length.dts | 19 +++ .../binman/test/237_compress_dtb_invalid.dts | 16 ++ tools/binman/test/238_compress_dtb_zstd.dts | 16 ++ 26 files changed, 590 insertions(+), 117 deletions(-) create mode 100644 tools/binman/btool/bzip2.py create mode 100644 tools/binman/btool/gzip.py create mode 100644 tools/binman/btool/lzop.py create mode 100644 tools/binman/btool/xz.py create mode 100644 tools/binman/btool/zstd.py delete mode 100644 tools/binman/comp_util.py create mode 100644 tools/binman/test/235_compress_dtb_prepend_invalid.dts create mode 100644 tools/binman/test/236_compress_dtb_prepend_length.dts create mode 100644 tools/binman/test/237_compress_dtb_invalid.dts create mode 100644 tools/binman/test/238_compress_dtb_zstd.dts -- 2.30.2

1 21

[GIT PULL] please pull fsl-qoriq-2022-8-17
by Peng Fan (OSS) 19 Aug '22

19 Aug '22

Hi Tom, Please pull fsl-qoriq-2022-8-17 CI: https://source.denx.de/u-boot/custodians/u-boot-fsl-qoriq/-/pipelines/13155 ---------------------------------------- Enable SPL authentication for ls1021atwr Fdt fixups for ls1043ardb v7.0 board ---------------------------------------- Thanks, Peng. The following changes since commit 20d4c6052fe5826b3421e86b2f0e76a6c22581a7: Merge tag 'efi-2022-10-rc3' of https://source.denx.de/u-boot/custodians/u-boot-efi (2022-08-13 07:37:48 -0400) are available in the Git repository at: https://source.denx.de/u-boot/custodians/u-boot-fsl-qoriq.git tags/fsl-qoriq-2022-8-17 for you to fetch changes up to dfea459f20133c89d666f3778e551b50c7e966f0: board: ls1043ardb: fdt fixups for revision v7.0 boards (2022-08-16 17:07:33 +0800) ---------------------------------------------------------------- Camelia Groza (1): board: ls1043ardb: fdt fixups for revision v7.0 boards Gaurav Jain (1): ls1021atwr: caam: Enable Uboot validaion in SPL. MAINTAINERS | 1 + arch/arm/dts/ls1021a-twr-u-boot.dtsi | 29 +++++++++++++++++++++++++++++ arch/arm/dts/ls1021a-twr.dtsi | 1 + board/freescale/common/fsl_chain_of_trust.c | 6 +----- board/freescale/common/fsl_validate.c | 10 +++++++++- board/freescale/ls1021atwr/ls1021atwr.c | 13 +++++++++++-- board/freescale/ls1043ardb/ls1043ardb.c | 44 +++++++++++++++++++++++++++++++++++++++++++- configs/ls1043ardb_SECURE_BOOT_defconfig | 1 + configs/ls1043ardb_defconfig | 1 + configs/ls1043ardb_nand_SECURE_BOOT_defconfig | 1 + configs/ls1043ardb_nand_defconfig | 1 + configs/ls1043ardb_sdcard_SECURE_BOOT_defconfig | 1 + configs/ls1043ardb_sdcard_defconfig | 1 + configs/ls1043ardb_tfa_SECURE_BOOT_defconfig | 1 + configs/ls1043ardb_tfa_defconfig | 1 + include/configs/ls1043ardb.h | 5 ++++- 16 files changed, 107 insertions(+), 10 deletions(-) create mode 100644 arch/arm/dts/ls1021a-twr-u-boot.dtsi

4 3

[PATCH] riscv: dts: Sync important Unmatched pmic and qspi0 changes from Linux
by Jessica Clarke 19 Aug '22

19 Aug '22

This adds the onkey, RTC and watchdog children to the DA9063 PMIC node, fixes the compatible for qspi0's flash node to match the official DT schema (it being an is25wp256 is discoverable, hence jedec,spi-nor is the only compatible that should be present) and exposes the card detect GPIO. Note that the device trees still diverge in some places (including important things like the PCIe controller's clock name) and should be cleaned up so that a common device tree is used in both projects rather than having different bindings. This patch does not attempt to do that, merely expose important functionality present in Linux's that is not in U-Boot's so that it can be used without the OS providing its own bundled copy. Signed-off-by: Jessica Clarke <jrtc27(a)jrtc27.com> --- arch/riscv/dts/hifive-unmatched-a00.dts | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-) diff --git a/arch/riscv/dts/hifive-unmatched-a00.dts b/arch/riscv/dts/hifive-unmatched-a00.dts index b44e8c160d..ea32ae72e8 100644 --- a/arch/riscv/dts/hifive-unmatched-a00.dts +++ b/arch/riscv/dts/hifive-unmatched-a00.dts @@ -76,6 +76,18 @@ interrupts = <1 IRQ_TYPE_LEVEL_LOW>; interrupt-controller; + onkey { + compatible = "dlg,da9063-onkey"; + }; + + rtc { + compatible = "dlg,da9063-rtc"; + }; + + wdt { + compatible = "dlg,da9063-watchdog"; + }; + regulators { vdd_bcore1: bcore1 { regulator-min-microvolt = <1050000>; @@ -217,7 +229,7 @@ &qspi0 { status = "okay"; flash@0 { - compatible = "issi,is25wp256", "jedec,spi-nor"; + compatible = "jedec,spi-nor"; reg = <0>; spi-max-frequency = <50000000>; m25p,fast-read; @@ -234,6 +246,7 @@ spi-max-frequency = <20000000>; voltage-ranges = <3300 3300>; disable-wp; + gpios = <&gpio 15 GPIO_ACTIVE_LOW>; }; }; -- 2.31.1

2 1

[PATCH 0/5] arm: mvebu: turris_omnia: Fix hangup in debug UART
by Pali Rohár 19 Aug '22

19 Aug '22

If proper U-Boot on Turris Omnia tries to print something on debug UART then CPU hangs. Reason is that debug UART in proper U-Boot for Turris Omnia has incorrect configuration of base register. Base register is different in SPL and also in different stages of proper U-Boot. Fix all these issues. Introduce new arch_very_early_init() ARM function which will be called at very early stage prior initialization of debug UART. Move code which moves internal mvebu registers to this function. And split configuration of DEBUG_UART_BASE into SPL and proper U-Boot to allow setting different values which matches current HW status. With all these changes, it is possible call "printf" function in Turris Omnia's board_early_init_f() function in both SPL and proper U-Boot. Pali Rohár (5): arm: Add new config option ARCH_VERY_EARLY_INIT arm: mvebu: Move internal registers in arch_very_early_init() function serial: Add new config option SPL_DEBUG_UART_BASE serial: ns16550: Add support for SPL_DEBUG_UART_BASE arm: mvebu: turris_omnia: Fix DEBUG_UART_BASE arch/arm/Kconfig | 6 ++++++ arch/arm/lib/crt0.S | 5 +++++ arch/arm/mach-mvebu/Kconfig | 1 + arch/arm/mach-mvebu/Makefile | 1 + arch/arm/mach-mvebu/cpu.c | 31 ------------------------------- arch/arm/mach-mvebu/lowlevel.S | 27 +++++++++++++++++++++++++++ configs/turris_omnia_defconfig | 3 ++- drivers/serial/Kconfig | 7 +++++++ drivers/serial/ns16550.c | 4 ++-- 9 files changed, 51 insertions(+), 34 deletions(-) create mode 100644 arch/arm/mach-mvebu/lowlevel.S -- 2.20.1

2 31

[PATCH] efi_loader: disk: remove unused field
by AKASHI Takahiro 19 Aug '22

19 Aug '22

The field, ifname, in efi_disk_obj is set but never used anywhere. Just remove it. Signed-off-by: AKASHI Takahiro <takahiro.akashi(a)linaro.org> --- lib/efi_loader/efi_disk.c | 13 ++----------- 1 file changed, 2 insertions(+), 11 deletions(-) diff --git a/lib/efi_loader/efi_disk.c b/lib/efi_loader/efi_disk.c index 1d700b2a6be3..7d9e973e5672 100644 --- a/lib/efi_loader/efi_disk.c +++ b/lib/efi_loader/efi_disk.c @@ -29,7 +29,6 @@ const efi_guid_t efi_system_partition_guid = PARTITION_SYSTEM_GUID; * * @header: EFI object header * @ops: EFI disk I/O protocol interface - * @ifname: interface name for block device * @dev_index: device index of block device * @media: block I/O media information * @dp: device path to the block device @@ -40,7 +39,6 @@ const efi_guid_t efi_system_partition_guid = PARTITION_SYSTEM_GUID; struct efi_disk_obj { struct efi_object header; struct efi_block_io ops; - const char *ifname; int dev_index; struct efi_block_io_media media; struct efi_device_path *dp; @@ -380,7 +378,6 @@ static int efi_fs_exists(struct blk_desc *desc, int part) * * @parent: parent handle * @dp_parent: parent device path - * @if_typename: interface name for block device * @desc: internal block device * @dev_index: device index for block device * @part_info: partition info @@ -391,7 +388,6 @@ static int efi_fs_exists(struct blk_desc *desc, int part) static efi_status_t efi_disk_add_dev( efi_handle_t parent, struct efi_device_path *dp_parent, - const char *if_typename, struct blk_desc *desc, int dev_index, struct disk_partition *part_info, @@ -476,7 +472,6 @@ static efi_status_t efi_disk_add_dev( return ret; } diskobj->ops = block_io_disk_template; - diskobj->ifname = if_typename; diskobj->dev_index = dev_index; /* Fill in EFI IO Media info (for read/write callbacks) */ @@ -534,15 +529,13 @@ static int efi_disk_create_raw(struct udevice *dev) { struct efi_disk_obj *disk; struct blk_desc *desc; - const char *if_typename; int diskid; efi_status_t ret; desc = dev_get_uclass_plat(dev); - if_typename = blk_get_if_type_name(desc->if_type); diskid = desc->devnum; - ret = efi_disk_add_dev(NULL, NULL, if_typename, desc, + ret = efi_disk_add_dev(NULL, NULL, desc, diskid, NULL, 0, &disk); if (ret != EFI_SUCCESS) { if (ret == EFI_NOT_READY) @@ -577,7 +570,6 @@ static int efi_disk_create_part(struct udevice *dev) { efi_handle_t parent; struct blk_desc *desc; - const char *if_typename; struct disk_part *part_data; struct disk_partition *info; unsigned int part; @@ -591,7 +583,6 @@ static int efi_disk_create_part(struct udevice *dev) return -1; desc = dev_get_uclass_plat(dev_get_parent(dev)); - if_typename = blk_get_if_type_name(desc->if_type); diskid = desc->devnum; part_data = dev_get_uclass_plat(dev); @@ -603,7 +594,7 @@ static int efi_disk_create_part(struct udevice *dev) return -1; dp_parent = (struct efi_device_path *)handler->protocol_interface; - ret = efi_disk_add_dev(parent, dp_parent, if_typename, desc, diskid, + ret = efi_disk_add_dev(parent, dp_parent, desc, diskid, info, part, &disk); if (ret != EFI_SUCCESS) { log_err("Adding partition for %s failed\n", dev->name); -- 2.36.1

2 1