x86: efi: Fixes and improvements for coreboot
Subsystems (4)
v3 cover x86: efi: Fixes and improvements for coreboot
This little series fixes various bugs and annoyances in coreboot and coreboot64.
With this both coreboot and coreboot64 start up and work reasonably well on Brya (x86 Chromebook) and U-Boot can boot common Linux distros.
- Make coreboot64 debug UART start reliably
- Avoid the long USB-init delay on startup
- Correct the timer speed on coreboo64
- Fix a bootstd cros bug (will likely be squashed into another patch)
- Fix the terribly slow console scrolling
With v2 I have also brought in some lost x86 patches so they are all in one series.
Changes in v3:
- Allow SMBIOS if EFI_LOADER is enabled
- Reword the help
- Use log_debug() to show the message
- Squash in the next patch
Changes in v2:
- Update the malloc size too
- Add new patch
- Add new patch
- Add new patch
- Add new patch
- Add new patch
- Add new patch
- Add new patch
- Add new patch
- Add new patch
Simon Glass (19):
x86: coreboot: Add IDE and SATA
x86: coreboot: Enable standard boot
x86: coreboot: Rearrange arch_cpu_init()
x86: Set the CPU vendor in SPL
x86: Allow APCI in SPL
x86: coreboot: Look for DBG2 UART in SPL too
x86: coreboot: Enable CONFIG_SYS_NS16550_MEM32
x86: coreboot: Drop USB init on startup
x86: coreboot: Align options between coreboot and coreboot64
x86: coreboot: Enable VIDEO_COPY
efi: x86: Correct the condition for installing ACPI tables
x86: smbios: Add a Kconfig indicating SMBIOS-table presence
x86: Record the position of the SMBIOS tables
efi: x86: Use the installed SMBIOS tables
x86: coreboot: Record the position of the SMBIOS tables
x86: doc: Move into its own directory
x86: doc: Update summaries and add links
x86: doc: Split out manual booting into its own file
x86: doc: coreboot: Mention 64-bit Linux distros
arch/sandbox/include/asm/global_data.h | 1 +
arch/x86/cpu/coreboot/coreboot.c | 16 +-
arch/x86/cpu/x86_64/cpu.c | 7 +
arch/x86/dts/coreboot.dts | 1 +
arch/x86/include/asm/global_data.h | 1 +
arch/x86/lib/coreboot/cb_sysinfo.c | 1 +
arch/x86/lib/tables.c | 3 +
configs/coreboot64_defconfig | 25 ++-
configs/coreboot_defconfig | 21 +-
doc/arch/index.rst | 2 +-
doc/arch/x86/index.rst | 12 +
doc/arch/x86/manual_boot.rst | 276 +++++++++++++++++++++++
doc/arch/{ => x86}/x86.rst | 295 ++-----------------------
doc/board/coreboot/coreboot.rst | 38 +++-
drivers/serial/Kconfig | 2 +-
drivers/video/coreboot.c | 12 +
include/asm-generic/global_data.h | 8 +
lib/Kconfig | 23 +-
lib/efi_loader/Makefile | 4 +-
lib/efi_loader/efi_setup.c | 20 +-
lib/efi_loader/efi_smbios.c | 37 +---
21 files changed, 451 insertions(+), 354 deletions(-)
create mode 100644 doc/arch/x86/index.rst
create mode 100644 doc/arch/x86/manual_boot.rst
rename doc/arch/{ => x86}/x86.rst (64%)Signature · 1 line
2.42.0.rc1.204.g551eb34607-goog
PATCH v3 1/19 x86: coreboot: Add IDE and SATA
Add these options to permit access to more disk types.
Add some documentation as well.
---
(no changes since v1)
configs/coreboot64_defconfig | 1 + configs/coreboot_defconfig | 9 +++++++++ doc/board/coreboot/coreboot.rst | 20 ++++++++++++++++++++ 3 files changed, 30 insertions(+)
configs/coreboot64_defconfig +1 -0
@@ -26,6 +26,7 @@ CONFIG_SYS_PBSIZE=532 CONFIG_CMD_IDE=y CONFIG_CMD_MMC=y CONFIG_CMD_PART=y +CONFIG_CMD_SATA=y CONFIG_CMD_USB=y # CONFIG_CMD_SETEXPR is not set CONFIG_CMD_DHCP=y
configs/coreboot_defconfig +9 -0
@@ -22,8 +22,10 @@ CONFIG_LOGF_FUNC=y CONFIG_DISPLAY_BOARDINFO_LATE=y CONFIG_LAST_STAGE_INIT=y CONFIG_PCI_INIT_R=y +CONFIG_CMD_IDE=y CONFIG_CMD_MMC=y CONFIG_CMD_PART=y +CONFIG_CMD_SATA=y CONFIG_CMD_USB=y # CONFIG_CMD_SETEXPR is not set CONFIG_CMD_DHCP=y
@@ -48,6 +50,13 @@ CONFIG_USE_ROOTPATH=y CONFIG_REGMAP=y CONFIG_SYSCON=y # CONFIG_ACPIGEN is not set +CONFIG_SYS_IDE_MAXDEVICE=4 +CONFIG_SYS_ATA_DATA_OFFSET=0 +CONFIG_SYS_ATA_REG_OFFSET=0 +CONFIG_SYS_ATA_ALT_OFFSET=0 +CONFIG_ATAPI=y +CONFIG_LBA48=y +CONFIG_SYS_64BIT_LBA=y CONFIG_NVME_PCI=y # CONFIG_PCI_PNP is not set CONFIG_SOUND=y
doc/board/coreboot/coreboot.rst +20 -0
@@ -41,6 +41,26 @@ At present it seems that for Minnowboard Max, coreboot does not pass through the video information correctly (it always says the resolution is 0x0). This works correctly for link though. +You can run via QEMU using:: + + qemu-system-x86_64 -bios build/coreboot.rom -serial mon:stdio + +The `-serial mon:stdio` part shows both output in the display and on the +console. It is optional. You can add `nographic` as well to *only* get console +output. + +To run with a SATA drive called `$DISK`:: + + qemu-system-x86_64 -bios build/coreboot.rom -serial mon:stdio \ + -drive id=disk,file=$DISK,if=none \ + -device ahci,id=ahci \ + -device ide-hd,drive=disk,bus=ahci.0 + +Then you can scan it with `scsi scan` and access it normally. + +To use 4GB of memory, typically necessary for booting Linux distros, add +`-m 4GB`. + 64-bit U-Boot -------------
Signature · 1 line
2.42.0.rc1.204.g551eb34607-goog
PATCH v3 2/19 x86: coreboot: Enable standard boot
In reply to Simon GlassEnable bootstd options and provide instructions on how to boot a linux distro using coreboot.
---
(no changes since v1)
configs/coreboot64_defconfig | 14 ++------------ configs/coreboot_defconfig | 1 + doc/board/coreboot/coreboot.rst | 16 ++++++++++++++-- 3 files changed, 17 insertions(+), 14 deletions(-)
configs/coreboot64_defconfig +2 -12
@@ -10,40 +10,30 @@ CONFIG_VENDOR_COREBOOT=y CONFIG_TARGET_COREBOOT=y CONFIG_FIT=y CONFIG_FIT_SIGNATURE=y +CONFIG_BOOTSTD_FULL=y +CONFIG_BOOTSTD_DEFAULTS=y CONFIG_SYS_MONITOR_BASE=0x01120000 CONFIG_SHOW_BOOT_PROGRESS=y CONFIG_USE_BOOTARGS=y CONFIG_BOOTARGS="root=/dev/sdb3 init=/sbin/init rootwait ro" -CONFIG_USE_BOOTCOMMAND=y -CONFIG_BOOTCOMMAND="ext2load scsi 0:3 01000000 /boot/vmlinuz; zboot 01000000" CONFIG_PRE_CONSOLE_BUFFER=y CONFIG_SYS_CONSOLE_INFO_QUIET=y CONFIG_DISPLAY_BOARDINFO_LATE=y CONFIG_LAST_STAGE_INIT=y CONFIG_SPL_NO_BSS_LIMIT=y -CONFIG_HUSH_PARSER=y CONFIG_SYS_PBSIZE=532 CONFIG_CMD_IDE=y CONFIG_CMD_MMC=y -CONFIG_CMD_PART=y CONFIG_CMD_SATA=y CONFIG_CMD_USB=y # CONFIG_CMD_SETEXPR is not set -CONFIG_CMD_DHCP=y CONFIG_BOOTP_BOOTFILESIZE=y -CONFIG_CMD_PING=y CONFIG_CMD_TIME=y CONFIG_CMD_SOUND=y -CONFIG_CMD_EXT2=y -CONFIG_CMD_EXT4=y CONFIG_CMD_EXT4_WRITE=y -CONFIG_CMD_FAT=y -CONFIG_CMD_FS_GENERIC=y CONFIG_MAC_PARTITION=y # CONFIG_SPL_MAC_PARTITION is not set # CONFIG_SPL_DOS_PARTITION is not set -CONFIG_ISO_PARTITION=y -CONFIG_EFI_PARTITION=y # CONFIG_SPL_EFI_PARTITION is not set CONFIG_ENV_OVERWRITE=y CONFIG_SYS_RELOC_GD_ENV_ADDR=y
configs/coreboot_defconfig +1 -0
@@ -10,6 +10,7 @@ CONFIG_TARGET_COREBOOT=y CONFIG_FIT=y CONFIG_FIT_SIGNATURE=y CONFIG_BOOTSTD_FULL=y +CONFIG_BOOTSTD_DEFAULTS=y CONFIG_SYS_MONITOR_BASE=0x01110000 CONFIG_SHOW_BOOT_PROGRESS=y CONFIG_USE_BOOTARGS=y
doc/board/coreboot/coreboot.rst +14 -2
@@ -67,9 +67,21 @@ To use 4GB of memory, typically necessary for booting Linux distros, add In addition to the 32-bit 'coreboot' build there is a 'coreboot64' build. This produces an image which can be booted from coreboot (32-bit). Internally it works by using a 32-bit SPL binary to switch to 64-bit for running U-Boot. It -can be useful for running UEFI applications, for example. +can be useful for running UEFI applications, for example with the coreboot +build in `$CBDIR`:: + + DISK=ubuntu-23.04-desktop-amd64.iso + CBDIR=~/coreboot/build + + cp $CBDIR/coreboot.rom.in coreboot.rom + cbfstool coreboot.rom add-flat-binary -f u-boot-x86-with-spl.bin \ + -n fallback/payload -c LZMA -l 0x1110000 -e 0x1110000 + + qemu-system-x86_64 -m 2G -smp 4 -bios coreboot.rom \ + -drive id=disk,file=$DISK,if=none \ + -device ahci,id=ahci \ + -device ide-hd,drive=disk,bus=ahci.0 \ -This has only been lightly tested. CBFS access -----------
Signature · 1 line
2.42.0.rc1.204.g551eb34607-goog
PATCH v3 3/19 x86: coreboot: Rearrange arch_cpu_init()
In reply to Simon GlassInit errors in SPL are currently ignored by this function.
Change the code to init the CPU, reporting an error if something is wrong. After that, look for the coreboot table.
---
(no changes since v1)
arch/x86/cpu/coreboot/coreboot.c | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-)
arch/x86/cpu/coreboot/coreboot.c +9 -3
@@ -20,7 +20,14 @@int arch_cpu_init (void ) { -int ret =get_coreboot_info (&lib_sysinfo); +int ret; + + ret =IS_ENABLED (CONFIG_X86_RUN_64BIT ) ?x86_cpu_reinit_f () : +x86_cpu_init_f (); +if (ret) +return ret; + + ret =get_coreboot_info (&lib_sysinfo);if (ret !=0 ) {printf ("Failed to parse coreboot tables.\n" );return ret;
@@ -28,8 +35,7 @@ int arch_cpu_init(void)timestamp_init (); -return IS_ENABLED (CONFIG_X86_RUN_64BIT ) ?x86_cpu_reinit_f () : -x86_cpu_init_f (); +return 0 ; }int checkcpu (void )
Signature · 1 line
2.42.0.rc1.204.g551eb34607-goog
PATCH v3 4/19 x86: Set the CPU vendor in SPL
In reply to Simon GlassWe don't read this information in 64-bit mode, since we don't have the macros for doing it. Set it to Intel by default. This allows the TSC timer to work correctly.
---
(no changes since v1)
arch/x86/cpu/x86_64/cpu.c | 7 +++++++ 1 file changed, 7 insertions(+)
arch/x86/cpu/x86_64/cpu.c +7 -0
@@ -8,8 +8,11 @@#include <cpu_func.h> #include <debug_uart.h> #include <init.h> +#include <asm/cpu.h> #include <asm/global_data.h> +DECLARE_GLOBAL_DATA_PTR ; +int cpu_has_64bit (void ) {return true;
@@ -38,6 +41,10 @@ int x86_mp_init(void)int x86_cpu_reinit_f (void ) { +/* set the vendor to Intel so that native_calibrate_tsc() works */ + gd->arch.x86_vendor =X86_VENDOR_INTEL ; + gd->arch.has_mtrr = true; +return 0 ; }
Signature · 1 line
2.42.0.rc1.204.g551eb34607-goog
PATCH v3 5/19 x86: Allow APCI in SPL
In reply to Simon GlassThis is needed so we can find the DBG2 table provided by coreboot. Add a Kconfig so it can be enabled.
---
(no changes since v1)
lib/Kconfig | 8 ++++++++ 1 file changed, 8 insertions(+)
lib/Kconfig +8 -0
@@ -289,6 +289,14 @@ config ACPI not necessarily include generation of tables (see GENERATE_ACPI_TABLE), but allows for tables to be located. +config SPL_ACPI + bool "Enable support for ACPI libraries in SPL" + depends on SPL && SUPPORT_ACPI + help + Provides library functions for dealing with ACPI tables in SPL. This + does not necessarily include generation of tables + (see GENERATE_ACPI_TABLE), but allows for tables to be located. + config GENERATE_ACPI_TABLE bool "Generate an ACPI (Advanced Configuration and Power Interface) table" depends on ACPI
Signature · 1 line
2.42.0.rc1.204.g551eb34607-goog
PATCH v3 6/19 x86: coreboot: Look for DBG2 UART in SPL too
In reply to Simon GlassIf coreboot does not set up sysinfo for the UART, SPL currently hangs. Use the DBG2 teechnique there as well. This allows coreboot64 to boot from coreboot even if the console info is missing from sysinfo
---
(no changes since v1)
configs/coreboot64_defconfig | 1 + drivers/serial/Kconfig | 2 +- 2 files changed, 2 insertions(+), 1 deletion(-)
configs/coreboot64_defconfig +1 -0
@@ -55,4 +55,5 @@ CONFIG_SYS_64BIT_LBA=y CONFIG_SOUND=y CONFIG_SOUND_I8254=y CONFIG_CONSOLE_SCROLL_LINES=5 +CONFIG_SPL_ACPI=y # CONFIG_GZIP is not set
drivers/serial/Kconfig +1 -1
@@ -672,7 +672,7 @@ config COREBOOT_SERIAL config COREBOOT_SERIAL_FROM_DBG2 bool "Obtain UART from ACPI tables" depends on COREBOOT_SERIAL - default y if !SPL + default y help Select this to try to find a DBG2 record in the ACPI tables, in the event that coreboot does not provide information about the UART in the
Signature · 1 line
2.42.0.rc1.204.g551eb34607-goog
PATCH v3 7/19 x86: coreboot: Enable CONFIG_SYS_NS16550_MEM32
In reply to Simon GlassThe debug UART on modern machines uses a 32-bit wide transfer. Without this, setting debug output causes a hang or no output. It is not obvious (when enabling CONFIG_DEBUG_UART) that this is needed.
Enable 32-bit access to avoid this trap.
---
(no changes since v1)
configs/coreboot64_defconfig | 1 + configs/coreboot_defconfig | 1 + 2 files changed, 2 insertions(+)
configs/coreboot64_defconfig +1 -0
@@ -52,6 +52,7 @@ CONFIG_ATAPI=y CONFIG_LBA48=y CONFIG_SYS_64BIT_LBA=y # CONFIG_PCI_PNP is not set +CONFIG_SYS_NS16550_MEM32=y CONFIG_SOUND=y CONFIG_SOUND_I8254=y CONFIG_CONSOLE_SCROLL_LINES=5
configs/coreboot_defconfig +1 -0
@@ -60,6 +60,7 @@ CONFIG_LBA48=y CONFIG_SYS_64BIT_LBA=y CONFIG_NVME_PCI=y # CONFIG_PCI_PNP is not set +CONFIG_SYS_NS16550_MEM32=y CONFIG_SOUND=y CONFIG_SOUND_I8254=y CONFIG_CONSOLE_SCROLL_LINES=5
Signature · 1 line
2.42.0.rc1.204.g551eb34607-goog
PATCH v3 8/19 x86: coreboot: Drop USB init on startup
In reply to Simon GlassThis is very annoying as it is quite slow on many machines. Also, U-Boot has an existing 'preboot' mechanism to enable this feature if desired.
Drop this code so that it is possible to choose whether to init USB or not.
---
(no changes since v1)
arch/x86/cpu/coreboot/coreboot.c | 4 ---- 1 file changed, 4 deletions(-)
arch/x86/cpu/coreboot/coreboot.c +0 -4
@@ -82,10 +82,6 @@ static void board_final_init(void)int last_stage_init (void ) { -/* start usb so that usb keyboard can be used as input device */ -if (IS_ENABLED (CONFIG_USB_KEYBOARD )) -usb_init (); -board_final_init ();return 0 ;
Signature · 1 line
2.42.0.rc1.204.g551eb34607-goog
PATCH v3 9/19 x86: coreboot: Align options between coreboot and coreboot64
In reply to Simon GlassThese two builds are similar but have some different options for not good reason. Line them up to be as similar as possible.
---
(no changes since v2)
Changes in v2:
- Update the malloc size too
configs/coreboot64_defconfig | 7 +++++++ configs/coreboot_defconfig | 9 --------- 2 files changed, 7 insertions(+), 9 deletions(-)
configs/coreboot64_defconfig +7 -0
@@ -1,5 +1,6 @@ CONFIG_X86=y CONFIG_TEXT_BASE=0x1120000 +CONFIG_SYS_MALLOC_LEN=0x2000000 CONFIG_NR_DRAM_BANKS=8 CONFIG_ENV_SIZE=0x1000 CONFIG_DEFAULT_DEVICE_TREE="coreboot"
@@ -18,6 +19,9 @@ CONFIG_USE_BOOTARGS=y CONFIG_BOOTARGS="root=/dev/sdb3 init=/sbin/init rootwait ro" CONFIG_PRE_CONSOLE_BUFFER=y CONFIG_SYS_CONSOLE_INFO_QUIET=y +CONFIG_LOG=y +CONFIG_LOGF_LINE=y +CONFIG_LOGF_FUNC=y CONFIG_DISPLAY_BOARDINFO_LATE=y CONFIG_LAST_STAGE_INIT=y CONFIG_SPL_NO_BSS_LIMIT=y
@@ -51,10 +55,13 @@ CONFIG_SYS_ATA_ALT_OFFSET=0 CONFIG_ATAPI=y CONFIG_LBA48=y CONFIG_SYS_64BIT_LBA=y +CONFIG_NVME_PCI=y # CONFIG_PCI_PNP is not set CONFIG_SYS_NS16550_MEM32=y CONFIG_SOUND=y CONFIG_SOUND_I8254=y CONFIG_CONSOLE_SCROLL_LINES=5 CONFIG_SPL_ACPI=y +CONFIG_CMD_DHRYSTONE=y # CONFIG_GZIP is not set +CONFIG_SMBIOS_PARSER=y
configs/coreboot_defconfig +0 -9
@@ -25,23 +25,14 @@ CONFIG_LAST_STAGE_INIT=y CONFIG_PCI_INIT_R=y CONFIG_CMD_IDE=y CONFIG_CMD_MMC=y -CONFIG_CMD_PART=y CONFIG_CMD_SATA=y CONFIG_CMD_USB=y # CONFIG_CMD_SETEXPR is not set -CONFIG_CMD_DHCP=y CONFIG_BOOTP_BOOTFILESIZE=y -CONFIG_CMD_PING=y CONFIG_CMD_TIME=y CONFIG_CMD_SOUND=y -CONFIG_CMD_EXT2=y -CONFIG_CMD_EXT4=y CONFIG_CMD_EXT4_WRITE=y -CONFIG_CMD_FAT=y -CONFIG_CMD_FS_GENERIC=y CONFIG_MAC_PARTITION=y -CONFIG_ISO_PARTITION=y -CONFIG_EFI_PARTITION=y CONFIG_ENV_OVERWRITE=y CONFIG_SYS_RELOC_GD_ENV_ADDR=y CONFIG_USE_BOOTFILE=y
Signature · 1 line
2.42.0.rc1.204.g551eb34607-goog
PATCH v3 10/19 x86: coreboot: Enable VIDEO_COPY
In reply to Simon GlassAt least on modern machines the write-back mechanism for the frame buffer is quite slow when scrolling, since it must read the entire frame buffer and write it back.
Enable the VIDEO_COPY feature to resolve this problem.
---
(no changes since v1)
arch/x86/dts/coreboot.dts | 1 + configs/coreboot64_defconfig | 1 + configs/coreboot_defconfig | 1 + drivers/video/coreboot.c | 12 ++++++++++++ 4 files changed, 15 insertions(+)
arch/x86/dts/coreboot.dts +1 -0
@@ -42,6 +42,7 @@ }; coreboot-fb { + bootph-some-ram; compatible = "coreboot-fb"; }; };
configs/coreboot64_defconfig +1 -0
@@ -60,6 +60,7 @@ CONFIG_NVME_PCI=y CONFIG_SYS_NS16550_MEM32=y CONFIG_SOUND=y CONFIG_SOUND_I8254=y +CONFIG_VIDEO_COPY=y CONFIG_CONSOLE_SCROLL_LINES=5 CONFIG_SPL_ACPI=y CONFIG_CMD_DHRYSTONE=y
configs/coreboot_defconfig +1 -0
@@ -54,6 +54,7 @@ CONFIG_NVME_PCI=y CONFIG_SYS_NS16550_MEM32=y CONFIG_SOUND=y CONFIG_SOUND_I8254=y +CONFIG_VIDEO_COPY=y CONFIG_CONSOLE_SCROLL_LINES=5 CONFIG_CMD_DHRYSTONE=y # CONFIG_GZIP is not set
drivers/video/coreboot.c +12 -0
@@ -73,6 +73,17 @@ err:return ret; } +static int coreboot_video_bind (struct udevice *dev) +{ +struct video_uc_plat *uc_plat =dev_get_uclass_plat (dev); + +/* Set the maximum supported resolution */ + uc_plat->size =4096 *2160 *4 ; +log_debug ("%s: Frame buffer size %x\n" , __func__, uc_plat->size); + +return 0 ; +} +static const struct udevice_id coreboot_video_ids[] = { { .compatible ="coreboot-fb" }, { }
@@ -82,5 +93,6 @@ U_BOOT_DRIVER(coreboot_video) = { .name = "coreboot_video", .id = UCLASS_VIDEO, .of_match = coreboot_video_ids, + .bind = coreboot_video_bind, .probe = coreboot_video_probe, };
Signature · 1 line
2.42.0.rc1.204.g551eb34607-goog
PATCH v3 11/19 efi: x86: Correct the condition for installing ACPI tables
In reply to Simon GlassIt is not always the case that U-Boot builds the ACPI tables itself. For example, when booting from coreboot, the ACPI tables are built by coreboot.
Correct the Makefile condition so that U-Boot can pass on tables built by a previous firmware stage.
Tidy up the installation-condition code while we are here.
---
(no changes since v2)
Changes in v2:
- Add new patch
lib/efi_loader/Makefile | 2 +- lib/efi_loader/efi_setup.c | 10 +++++----- 2 files changed, 6 insertions(+), 6 deletions(-)
lib/efi_loader/Makefile +1 -1
@@ -78,7 +78,7 @@ obj-$(CONFIG_EFI_ESRT) += efi_esrt.o obj-$(CONFIG_VIDEO) += efi_gop.o obj-$(CONFIG_BLK) += efi_disk.o obj-$(CONFIG_NETDEVICES) += efi_net.o -obj-$(CONFIG_GENERATE_ACPI_TABLE) += efi_acpi.o +obj-$(CONFIG_ACPI) += efi_acpi.o obj-$(CONFIG_GENERATE_SMBIOS_TABLE) += efi_smbios.o obj-$(CONFIG_EFI_RNG_PROTOCOL) += efi_rng.o obj-$(CONFIG_EFI_TCG2_PROTOCOL) += efi_tcg2.o
lib/efi_loader/efi_setup.c +5 -5
@@ -321,11 +321,11 @@ efi_status_t efi_init_obj_list(void)if (ret !=EFI_SUCCESS ) goto out;#endif -#ifdef CONFIG_GENERATE_ACPI_TABLE - ret =efi_acpi_register (); -if (ret !=EFI_SUCCESS ) - goto out; -#endif +if (IS_ENABLED (CONFIG_ACPI )) { + ret =efi_acpi_register (); +if (ret !=EFI_SUCCESS ) + goto out; + }#ifdef CONFIG_GENERATE_SMBIOS_TABLE ret =efi_smbios_register ();if (ret !=EFI_SUCCESS )
Signature · 1 line
2.42.0.rc1.204.g551eb34607-goog
PATCH v3 12/19 x86: smbios: Add a Kconfig indicating SMBIOS-table presence
In reply to Simon GlassWhen booted from coreboot, U-Boot does not build the SMBIOS tables, but it should still pass them on to the OS. Add a new option which indicates whether SMBIOS tables are present, however they were built.
Flip the ordering so that the dependency is listed first, which is less confusing.
Adjust GENERATE_SMBIOS_TABLE to depend on this new symbol.
---
Changes in v3:
- Allow SMBIOS if EFI_LOADER is enabled
- Reword the help
Changes in v2:
- Add new patch
lib/Kconfig | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-)
lib/Kconfig +14 -1
@@ -984,8 +984,8 @@ config BLOBLIST_TABLES config GENERATE_SMBIOS_TABLE bool "Generate an SMBIOS (System Management BIOS) table" + depends on SMBIOS default y - depends on X86 || EFI_LOADER help The System Management BIOS (SMBIOS) specification addresses how motherboard and system vendors present management information about
@@ -1054,6 +1054,19 @@ config SPL_OID_REGISTRY unambiguous persistent name (https://en.wikipedia.org/wiki/Object_identifier). Enable fast lookup object identifier registry in the SPL. +config SMBIOS + bool "SMBIOS support" + depends on X86 || EFI_LOADER + default y + help + Indicates that this platform can support System Management BIOS + (SMBIOS) tables. These provide various pieces of information about + the board, such as the manufacturer and the model name. + + See GENERATE_SMBIOS_TABLE which controls whether U-Boot actually + creates these tables, rather than them coming from a previous firmware + stage. + config SMBIOS_PARSER bool "SMBIOS parser" help
Signature · 1 line
2.42.0.rc1.204.g551eb34607-goog
Re: [PATCH v3 12/19] x86: smbios: Add a Kconfig indicating SMBIOS-table presence
Quoted 37 lines · Simon Glass
When booted from coreboot, U-Boot does not build the SMBIOS tables, but it should still pass them on to the OS. Add a new option which indicates whether SMBIOS tables are present, however they were built.
Flip the ordering so that the dependency is listed first, which is less confusing.
Adjust
GENERATE_SMBIOS_TABLEto depend on this new symbol.Signed-off-by Simon Glass <[email protected]>---
Changes in v3:
- Allow SMBIOS ifEFI_LOADERis enabled
- Reword the helpChanges in v2:
- Add new patchlib/Kconfig | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-)lib/Kconfig +1
-1@@ -984,8 +984,8 @@ config BLOBLIST_TABLES config GENERATE_SMBIOS_TABLE bool "Generate an SMBIOS (System Management BIOS) table" + depends on SMBIOS default y - depends on X86 || EFI_LOADER help The System Management BIOS (SMBIOS) specification addresses how motherboard and system vendors present management information about
lib/Kconfig config SPL_OID_REGISTRY · line 1059
Quoted 6 lines
lib/Kconfig +3
-0@@ -1054,6 +1054,19 @@ config SPL_OID_REGISTRY unambiguous persistent name (https://en.wikipedia.org/wiki/Object_identifier). Enable fast lookup object identifier registry in the SPL. +config SMBIOS + bool "SMBIOS support" + depends on X86 || EFI_LOADER
Is there a calling convention for the legacy entry point of Linux to pass SMBIOS tables on arm64 and riscv64?
lib/Kconfig config SPL_OID_REGISTRY · line 1067
Quoted 8 lines
lib/Kconfig +8
-0+ default y + help + Indicates that this platform can support System Management BIOS + (SMBIOS) tables. These provide various pieces of information about + the board, such as the manufacturer and the model name. + + See GENERATE_SMBIOS_TABLE which controls whether U-Boot actually + creates these tables, rather than them coming from a previous firmware
Thanks for updating the help text.
This implies that CONFIG_SMBIOS must be used to decide if efi_smbios_register() is called in efi_init_obj_list(). This change is in patch 14/19.
Quoted 5 lines
lib/Kconfig +2
-0+ stage. + config SMBIOS_PARSER bool "SMBIOS parser" hel
Re: [PATCH v3 12/19] x86: smbios: Add a Kconfig indicating SMBIOS-table presence
Hi Heinrich,
Quoted 58 lines · Heinrich Schuchardt
Quoted 37 lines · Simon Glass
When booted from coreboot, U-Boot does not build the SMBIOS tables, but it should still pass them on to the OS. Add a new option which indicates whether SMBIOS tables are present, however they were built.
Flip the ordering so that the dependency is listed first, which is less confusing.
Adjust
GENERATE_SMBIOS_TABLEto depend on this new symbol.Signed-off-by Simon Glass <[email protected]>---
Changes in v3:
- Allow SMBIOS ifEFI_LOADERis enabled
- Reword the helpChanges in v2:
- Add new patchlib/Kconfig | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-)lib/Kconfig +1
-1@@ -984,8 +984,8 @@ config BLOBLIST_TABLES config GENERATE_SMBIOS_TABLE bool "Generate an SMBIOS (System Management BIOS) table" + depends on SMBIOS default y - depends on X86 || EFI_LOADER help The System Management BIOS (SMBIOS) specification addresses how motherboard and system vendors present management information aboutlib/Kconfig config SPL_OID_REGISTRY · line 1059
Quoted 6 lines
lib/Kconfig +3
-0@@ -1054,6 +1054,19 @@ config SPL_OID_REGISTRY unambiguous persistent name (https://en.wikipedia.org/wiki/Object_identifier). Enable fast lookup object identifier registry in the SPL. +config SMBIOS + bool "SMBIOS support" + depends on X86 || EFI_LOADERIs there a calling convention for the legacy entry point of Linux to pass SMBIOS tables on arm64 and riscv64?
lib/Kconfig config SPL_OID_REGISTRY · line 1067
Quoted 8 lines
lib/Kconfig +8
-0+ default y + help + Indicates that this platform can support System Management BIOS + (SMBIOS) tables. These provide various pieces of information about + the board, such as the manufacturer and the model name. + + See GENERATE_SMBIOS_TABLE which controls whether U-Boot actually + creates these tables, rather than them coming from a previous firmware
Quoted 5 lines
Thanks for updating the help text.
This implies that
CONFIG_SMBIOSmust be used to decide ifefi_smbios_register()is called inefi_init_obj_list(). This change is in patch 14/19.
Yes that's right, since we still need to install it if the platform uses SMBIOS.
Quoted 3 lines
Reviewed-by Heinrich Schuchardt <[email protected]>
Thanks for the review..unfortunately I found that on ARM (and probably RISC-V) U-Boot does not create this table. On x86 last_state_init() calls write_tables() to do this.
So I will have to adjust this...
Regards,
Simon
PATCH v3 13/19 x86: Record the position of the SMBIOS tables
In reply to Simon GlassRemember where these end up so that we can pass this information on to the EFI layer.
---
(no changes since v2)
Changes in v2:
- Add new patch
arch/sandbox/include/asm/global_data.h | 1 + arch/x86/include/asm/global_data.h | 1 + arch/x86/lib/tables.c | 3 +++ include/asm-generic/global_data.h | 8 ++++++++ 4 files changed, 13 insertions(+)
arch/sandbox/include/asm/global_data.h +1 -0
@@ -17,6 +17,7 @@ struct arch_global_data {ulong table_end;/* End address of x86 tables */ ulong table_start_high;/* Start address of high x86 tables */ ulong table_end_high;/* End address of high x86 tables */ +ulong smbios_start;/* Start address of SMBIOS table */ };#include <asm-generic/global_data.h>
arch/x86/include/asm/global_data.h +1 -0
@@ -127,6 +127,7 @@ struct arch_global_data {ulong table_end;/* End address of x86 tables */ ulong table_start_high;/* Start address of high x86 tables */ ulong table_end_high;/* End address of high x86 tables */ + ulong smbios_start; /* Start address of SMBIOS table */ };#endif
arch/x86/lib/tables.c +3 -0
@@ -97,6 +97,9 @@ int write_tables(void)int size = table->size ? :CONFIG_ROM_TABLE_SIZE ;u32 rom_table_end; +if (!strcmp ("smbios" , table->name)) + gd->arch.smbios_start = rom_addr; +if (IS_ENABLED (CONFIG_BLOBLIST_TABLES ) && table->tag) {if (!gd->arch.table_end) gd->arch.table_end = rom_addr;
include/asm-generic/global_data.h +8 -0
@@ -552,6 +552,14 @@ static_assert(sizeof(struct global_data) == GD_SIZE);#define gd_set_acpi_start (addr)#endif +#ifdef CONFIG_SMBIOS +#define gd_smbios_start () gd->smbios_start +#define gd_set_smbios_start (addr) gd->arch.smbios_start = addr +#else +#define gd_smbios_start () 0UL +#define gd_set_smbios_start (addr) +#endif +#if CONFIG_IS_ENABLED (MULTI_DTB_FIT )#define gd_multi_dtb_fit () gd->multi_dtb_fit#define gd_set_multi_dtb_fit (_dtb) gd->multi_dtb_fit = _dtb
Signature · 1 line
2.42.0.rc1.204.g551eb34607-goog
PATCH v3 14/19 efi: x86: Use the installed SMBIOS tables
In reply to Simon GlassU-Boot sets up the SMBIOS tables during startup. Rather than creating a new set, install the existing ones. Rely on the ACPI table's memory-map record to cover the tables.
Tidy up the installation-condition code while we are here.
---
Changes in v3:
- Use log_debug() to show the message
- Squash in the next patch
Changes in v2:
- Add new patch
lib/efi_loader/Makefile | 2 +- lib/efi_loader/efi_setup.c | 10 +++++----- lib/efi_loader/efi_smbios.c | 37 +++++++------------------------------ 3 files changed, 13 insertions(+), 36 deletions(-)
lib/efi_loader/Makefile +1 -1
@@ -79,7 +79,7 @@ obj-$(CONFIG_VIDEO) += efi_gop.o obj-$(CONFIG_BLK) += efi_disk.o obj-$(CONFIG_NETDEVICES) += efi_net.o obj-$(CONFIG_ACPI) += efi_acpi.o -obj-$(CONFIG_GENERATE_SMBIOS_TABLE) += efi_smbios.o +obj-$(CONFIG_SMBIOS) += efi_smbios.o obj-$(CONFIG_EFI_RNG_PROTOCOL) += efi_rng.o obj-$(CONFIG_EFI_TCG2_PROTOCOL) += efi_tcg2.o obj-$(CONFIG_EFI_RISCV_BOOT_PROTOCOL) += efi_riscv.o
lib/efi_loader/efi_setup.c +5 -5
@@ -326,11 +326,11 @@ efi_status_t efi_init_obj_list(void)if (ret !=EFI_SUCCESS ) goto out; } -#ifdef CONFIG_GENERATE_SMBIOS_TABLE - ret =efi_smbios_register (); -if (ret !=EFI_SUCCESS ) - goto out; -#endif +if (IS_ENABLED (CONFIG_SMBIOS )) { + ret =efi_smbios_register (); +if (ret !=EFI_SUCCESS ) + goto out; + } ret =efi_watchdog_register ();if (ret !=EFI_SUCCESS ) goto out;
lib/efi_loader/efi_smbios.c +7 -30
@@ -20,36 +20,13 @@*/ efi_status_t efi_smbios_register (void ) { -/* Map within the low 32 bits, to allow for 32bit SMBIOS tables */ -u64 dmi_addr =U32_MAX ; -efi_status_t ret; -void *dmi; +ulong addr; -/* Reserve 4kiB page for SMBIOS */ - ret =efi_allocate_pages (EFI_ALLOCATE_MAX_ADDRESS , -EFI_RUNTIME_SERVICES_DATA ,1 , &dmi_addr); +/* space for all tables is marked in efi_acpi_register() */ + addr = gd->arch.smbios_start; +log_debug ("EFI using SMBIOS tables at %lx\n" , addr); -if (ret !=EFI_SUCCESS ) { -/* Could not find space in lowmem, use highmem instead */ - ret =efi_allocate_pages (EFI_ALLOCATE_ANY_PAGES , -EFI_RUNTIME_SERVICES_DATA ,1 , - &dmi_addr); - -if (ret !=EFI_SUCCESS ) -return ret; - } - -/* -* Generate SMBIOS tables - we know that efi_allocate_pages() returns -* a 4k-aligned address, so it is safe to assume that -* write_smbios_table() will write the table at that address. -*/ -assert (!(dmi_addr &0xf )); - dmi = (void *)(uintptr_t )dmi_addr; -if (write_smbios_table (map_to_sysmem (dmi))) -/* Install SMBIOS information as configuration table */ -return efi_install_configuration_table (&smbios_guid, dmi); -efi_free_pages (dmi_addr,1 ); -log_err ("Cannot create SMBIOS table\n" ); -return EFI_SUCCESS ; +/* Install SMBIOS information as configuration table */ +return efi_install_configuration_table (&smbios_guid, +map_sysmem (addr,0 )); }
Signature · 1 line
2.42.0.rc1.204.g551eb34607-goog
PATCH v3 15/19 x86: coreboot: Record the position of the SMBIOS tables
In reply to Simon GlassMake a note of where coreboot installed the SMBIOS tables so that we can pass this on to EFI.
---
(no changes since v2)
Changes in v2:
- Add new patch
arch/x86/lib/coreboot/cb_sysinfo.c | 1 + 1 file changed, 1 insertion(+)
arch/x86/lib/coreboot/cb_sysinfo.c +1 -0
@@ -471,6 +471,7 @@ int get_coreboot_info(struct sysinfo_t *info)return -ENOENT ; gd->arch.coreboot_table = addr;gd_set_acpi_start (map_to_sysmem (info->rsdp)); +gd_set_smbios_start (info->smbios_start); gd->flags |=GD_FLG_SKIP_LL_INIT ;return 0 ;
Signature · 1 line
2.42.0.rc1.204.g551eb34607-goog
PATCH v3 16/19 x86: doc: Move into its own directory
In reply to Simon GlassThere is enough material that it makes sense to split this up into several files. Create an x86/ directory for this purpose.
---
(no changes since v2)
Changes in v2:
- Add new patch
doc/arch/index.rst | 2 +-
doc/arch/x86/index.rst | 11 +++++++++++
doc/arch/{ => x86}/x86.rst | 6 +++---
3 files changed, 15 insertions(+), 4 deletions(-)
create mode 100644 doc/arch/x86/index.rst
rename doc/arch/{ => x86}/x86.rst (99%)doc/arch/index.rst +1 -1
@@ -15,5 +15,5 @@ Architecture-specific doc riscv sandbox/index sh - x86 + x86/index xtensa
doc/arch/x86/index.rst new file +11 -0
@@ -0,0 +1,11 @@ +.. SPDX-License-Identifier: GPL-2.0+ */ +.. Copyright 2023 Google LLC +.. sectionauthor:: Simon Glass <[email protected]> + +x86 +=== + +.. toctree:: + :maxdepth: 2 + + x86
doc/arch/x86/x86.rst renamed from doc/arch/x86.rst +3 -3
@@ -695,8 +695,8 @@ to load a 'u-boot-payload.efi', see below test logs on QEMU. No controllers found Hit any key to stop autoboot: 0 -See :doc:`../develop/uefi/u-boot_on_efi` and :doc:`../develop/uefi/uefi` for -details of EFI support in U-Boot. +See :doc:`../../develop/uefi/u-boot_on_efi` and :doc:`../../develop/uefi/uefi` +for details of EFI support in U-Boot. Chain-loading -------------
@@ -732,7 +732,7 @@ SMBIOS tables To generate SMBIOS tables in U-Boot, for use by the OS, enable the CONFIG_GENERATE_SMBIOS_TABLE option. The easiest way to provide the values to use is via the device tree. For details see -:download:`smbios.txt <../device-tree-bindings/sysinfo/smbios.txt>`. +:download:`smbios.txt <../../device-tree-bindings/sysinfo/smbios.txt>`. TODO List ---------
Signature · 1 line
2.42.0.rc1.204.g551eb34607-goog
PATCH v3 17/19 x86: doc: Update summaries and add links
In reply to Simon GlassRefresh the summary information so it is more up-to-date. Add links to the coreboot and slimbootloader docs.
---
(no changes since v2)
Changes in v2:
- Add new patch
doc/arch/x86/x86.rst | 17 ++++++++++------- 1 file changed, 10 insertions(+), 7 deletions(-)
doc/arch/x86/x86.rst +10 -7
@@ -11,9 +11,9 @@ including supported boards, build instructions, todo list, etc. Status ------ U-Boot supports running as a `coreboot`_ payload on x86. So far only Link -(Chromebook Pixel) and `QEMU`_ x86 targets have been tested, but it should -work with minimal adjustments on other x86 boards since coreboot deals with -most of the low-level details. +(Chromebook Pixel), Brya (Alder Lake Chromebook) and `QEMU`_ x86 targets have +been tested, but it should work with minimal adjustments on other x86 boards +since coreboot deals with most of the low-level details. U-Boot is a main bootloader on Intel Edison board.
@@ -31,12 +31,14 @@ are supported: - Link (Chromebook Pixel) - Minnowboard MAX - Samus (Chromebook Pixel 2015) + - Coral (Apollo Lake Chromebooks circa 2017) - QEMU x86 (32-bit & 64-bit) As for loading an OS, U-Boot supports directly booting a 32-bit or 64-bit Linux kernel as part of a FIT image. It also supports a compressed zImage. U-Boot supports loading an x86 VxWorks kernel. Please check README.vxworks -for more details. +for more details. Finally, U-Boot can boot Linux distributions with a UEFI +interface. Build Instructions for U-Boot as BIOS replacement (bare mode) -------------------------------------------------------------
@@ -700,9 +702,10 @@ for details of EFI support in U-Boot. Chain-loading ------------- -U-Boot can be chain-loaded from another bootloader, such as coreboot or -Slim Bootloader. Typically this is done by building for targets 'coreboot' or -'slimbootloader'. +U-Boot can be chain-loaded from another bootloader, such as +:doc:`../../board/coreboot/index` coreboot or +:doc:`../../board/intel/slimbootloader`. Typically this is done by building for +targets 'coreboot' or 'slimbootloader'. For example, at present we have a 'coreboot' target but this runs very different code from the bare-metal targets, such as coral. There is very little
Signature · 1 line
2.42.0.rc1.204.g551eb34607-goog
PATCH v3 18/19 x86: doc: Split out manual booting into its own file
In reply to Simon GlassMove this out of the main file since for simple users it is easier to rely on standard boot.
---
(no changes since v2)
Changes in v2:
- Add new patch
doc/arch/x86/index.rst | 1 + doc/arch/x86/manual_boot.rst | 276 +++++++++++++++++++++++++++++++++++ doc/arch/x86/x86.rst | 272 +--------------------------------- 3 files changed, 280 insertions(+), 269 deletions(-) create mode 100644 doc/arch/x86/manual_boot.rst
doc/arch/x86/index.rst +1 -0
@@ -9,3 +9,4 @@ x86 :maxdepth: 2 x86 + manual_boot
doc/arch/x86/manual_boot.rst new file +276 -0
276 lines not on this page · raw message
doc/arch/x86/x86.rst +3 -269
283 lines not on this page · raw message
Signature · 1 line
2.42.0.rc1.204.g551eb34607-goog
PATCH v3 19/19 x86: doc: coreboot: Mention 64-bit Linux distros
In reply to Simon GlassAdd a little more detail as to why coreboot64 is preferred for booting Linux distros.
---
(no changes since v2)
Changes in v2:
- Add new patch
doc/board/coreboot/coreboot.rst | 2 ++ 1 file changed, 2 insertions(+)
doc/board/coreboot/coreboot.rst +2 -0
@@ -82,6 +82,8 @@ build in `$CBDIR`:: -device ahci,id=ahci \ -device ide-hd,drive=disk,bus=ahci.0 \ +This allows booting and installing various distros, many of which are +64-bit-only, so cannot work with the 32-bit 'coreboot' build. CBFS access -----------
Signature · 1 line
2.42.0.rc1.204.g551eb34607-goog