Debian's arm64 UEFI Secure Boot shim makes the EFI variable store run
out of space while mirroring its MOK database to variables. This can be
observed in QEMU like so:
$ tools/buildman/buildman -o build/qemu_arm64 --boards=qemu_arm64 -w
$ cd build/qemu_arm64
$ curl -L -o debian.iso \
https://cdimage.debian.org/debian-cd/current/arm64/iso-cd/debian-12.0.0-arm64-netinst.iso
$ qemu-system-aarch64 \
-nographic -bios u-boot.bin \
-machine virt -cpu cortex-a53 -m 1G -smp 2 \
-drive if=virtio,file=debian.iso,index=0,format=raw,readonly=on,media=cdrom
[...]
=> # interrupt autoboot
=> env set -e -bs -nv -rt -guid 605dab50-e046-4300-abb6-3dd810dd8b23 SHIM_VERBOSE 1
=> boot
[...]
mok.c:296:mirror_one_esl() SetVariable("MokListXRT43", ... varsz=0x4C) = Out of Resources
mok.c:452:mirror_mok_db() esd:0x7DB92D20 adj:0x30
Failed to set MokListXRT: Out of Resources
mok.c:767:mirror_one_mok_variable() mirror_mok_db("MokListXRT", datasz=17328) returned Out of Resources
mok.c:812:mirror_one_mok_variable() returning Out of Resources
Could not create MokListXRT: Out of Resources
[...]
Welcome to GRUB!
This would normally be fine as shim would continue to run grubaa64.efi,
but shim's error handling code for this case has a bug [1] that causes a
synchronous abort on at least chromebook_kevin (but apparently not on
QEMU arm64).
Double the default variable store size so the variables fit. There is a
note about this value matching PcdFlashNvStorageVariableSize when
EFI_MM_COMM_TEE is enabled, so keep the old default in that case.
[1] https://github.com/rhboot/shim/pull/577
Signed-off-by: Alper Nebi Yasak <alpernebiyasak@gmail.com>
Reviewed-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
The efi_var_mem_free() function calculates the available size for a new
EFI variable by subtracting the occupied buffer size and the overhead
for a new variable from the maximum buffer size set in Kconfig. This
is then returned as QueryVariableInfo()'s RemainingVariableStorageSize
output.
This can underflow as the calculation is done in and processed as
unsigned integer types. Check for underflow before doing the subtraction
and return zero if there's no space.
Fixes: f1f990a8c9 ("efi_loader: memory buffer for variables")
Signed-off-by: Alper Nebi Yasak <alpernebiyasak@gmail.com>
Reviewed-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
After Linux commit ff61f0791ce9, x86 documentation was moved to
arch/x86 and the link in bootparam.h was broken.
Signed-off-by: Paul Barker <paul.barker.ct@bp.renesas.com>
Reviewed-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
DM_GPIO depends on GPIO to be enabled but select will cause that DM_GPIO is
selected without GPIO which ends up in compilation error:
undefined reference to `dm_gpio_set_value'
undefined reference to `dm_gpio_get_value'
undefined reference to `dm_gpio_free'
undefined reference to `gpio_request_by_name'
Signed-off-by: Michal Simek <michal.simek@amd.com>
[trini: Fix configs which had relied on these select's]
Signed-off-by: Tom Rini <trini@konsulko.com>
As implemented in the arch/arm/mach-rockchip/tpl.c file,
the CONFIG_TPL_BANNER_PRINT option will not work
if either of these options is not enabled.
Add dependency constraints to the CONFIG_TPL_BANNER_PRINT option
definition to prevent configuration problems
where option is enabled but do not take effect.
Suggested-by: Yanjie Ren <renyanjie01@gmail.com>
Signed-off-by: Ying Sun <sunying@nj.iscas.ac.cn>
CONFIG_SYS_CONSOLE_ENV_OVERWRITE is implemented in common/console.c
when "#if CONFIG_IS_ENABLED(SYS_CONSOLE_IS_IN_ENV)" is met.
It is recommended to add dependency constraints to its definition.
Suggested-by: Yanjie Ren <renyanjie01@gmail.com>
Signed-off-by: Ying Sun <sunying@nj.iscas.ac.cn>
CONFIG_CMD_SAVES is used to enable support for the "saveenv" command
and is only implemented in cmd/load.c
when "#if defined(CONFIG_CMD_LOADS)" is met.
It is recommended to add dependency constraints to its definition.
Prevents "saveenv" command from not being supported
when "CONFIG_CMD_SAVES=y CONFIG_CMD_LOADS=n".
Suggested-by: Yanjie Ren <renyanjie01@gmail.com>
Signed-off-by: Ying Sun <sunying@nj.iscas.ac.cn>
Reviewed-by: Simon Glass <sjg@chromium.org>
Docs describe 'make clean' to delete most generated files, 'make
mrproper' to delete current configuration and all generated files. This
test tries to assert it.
Idea is to search remaining files by patterns in copies of the initial
out-of-source build, which has two advantages:
- looking in an out-of-source build dir allows to tell generated source
code from committed source code
- copying is fast (compared to rebuilding each time) which allows to do
a "world clean"
Signed-off-by: Tobias Deiminger <tdmg@linutronix.de>
sandbox can generate DT overlays, but they were not cleaned.
Extend the explicit clean-files list accordingly.
Fixes: 95300f203f ("pytest: add sandbox test for "extension" command")
Signed-off-by: Tobias Deiminger <tdmg@linutronix.de>
'make clean' did not descend into arch/$ARCH/dts for arc, m68k, nios2,
sh, xtensa.
Fix it by adding the missing archs to the explicit clean-dirs list.
Signed-off-by: Tobias Deiminger <tdmg@linutronix.de>
VPL artifacts like example vpl/u-boot-vpl are currently not removed by
'make clean'.
We can clean them just as it's already done for SPL and TPL.
Fixes: f86ca5ad8f ("Introduce Verifying Program Loader (VPL)")
Signed-off-by: Tobias Deiminger <tdmg@linutronix.de>
On 'make clean', generated C files in tools/env/ and tools/boot/ are
currently not removed, but they should.
Auto-generation for shared sources was first introduced with
ad80c4a322 ("kbuild, tools: generate wrapper C sources automatically
by Makefile"). Cleanup later regressed (see Fixes:), because shared
files were moved out of lib/ and common/, but 'clean-dirs := lib common'
was not adjusted accordingly. Further, the generated
tools/env/embedded.c became a sibling to project files, which prevents
directory-wise cleanup at all.
To solve it, we establishe tools/generated/ as the sole place for
generated sources. Wrappers are now generated as
tools/generated/<orig_dirname>/<orig_filename>, and 'make clean' can
remove tools/generated/ as a whole (Linux Makefile.asm-generic headers
are cleaned similarly). This way we don't have to maintain separate
clean-files or clean-dirs entries for each single added or moved wrapper
file.
Fixes: 0649cd0d49 ("Move environment files from common/ to env/")
Fixes: 19a91f2464 ("Create a new boot/ directory")
Signed-off-by: Tobias Deiminger <tdmg@linutronix.de>
[trini: Correct mkfwupdate case]
Signed-off-by: Tom Rini <trini@konsulko.com>
The 'booti' command is unable to boot Image.gz due to the absence
of required environment variables 'kernel_comp_addr_r' and
'kernel_comp_size'.
This commit adds these variables and reorganizes the memory layout
to prevent any overlap between binaries and files.
Signed-off-by: Yu Chien Peter Lin <peterlin@andestech.com>
Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
On some AE350 boards, we need to explicitly initialize the priority
registers to a non-zero value so the boot hart can instruct secondary
harts to jump to OpenSBI.
This patch also updates the information about PLICSW.
Signed-off-by: Yu Chien Peter Lin <peterlin@andestech.com>
Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
Describe which numeric values can be used for as scratch options for
OpenSBI.
Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
Reviewed-by: Anup Patel <anup@brainfault.org>
There is a problem that the rates of PLL0 and PLL1 are set incorrectly
because the postdiv1_mask value is incorrectly entered when setting
the pll clk reg. Modify postdiv1's mask value to be put correctly.
Signed-off-by: Hoegeun Kwon <hoegeun.kwon@samsung.com>
Reviewed-by: Minkyu Kang <mk7.kang@samsung.com>
Use the latest OpenSBI v1.2 release binaries for the RISC-V CI.
Signed-off-by: Bin Meng <bmeng@tinylab.org>
Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
The dts sync from Linux leaves mac0/ethernet1 enabled on icicle, but
U-Boot does not currently set a mac address for it. Expand on the code
which currently sets the mac for mac1/ethernet0 to optionally set the
mac address for the second ethernet.
Reviewed-by: Padmarao Begari <padmarao.begari@microchip.com>
Tested-by: Padmarao Begari <padmarao.begari@microchip.com>
Signed-off-by: Conor Dooley <conor.dooley@microchip.com>
The "notable" disappearances are:
- the pac193x stanza - there's nothing in mainline linux w.r.t. bindings
for this & what is going to appear in mainline linux is going to be
incompatible with what is currently in U-Boot.
- operating points - these operating points should not be set at the
soc.dtsi level as they may not be possible depending on the design
programmed to the FPGA
- clock output names - there are defines for the clock indices, these
should not be needed
- the dt maintainers in linux NAKed using defines for IRQ numbers
- the qspi nand, which is not part of the icicle's default configuration
is removed.
Reviewed-by: Padmarao Begari <padmarao.begari@microchip.com>
Tested-by: Padmarao Begari <padmarao.begari@microchip.com>
Signed-off-by: Conor Dooley <conor.dooley@microchip.com>
Reviewed-by: Rick Chen <rick@andestech.com>
The original names picked for the DT doesn't match Linux's naming scheme
and it was renamed there a while ago. Rename it in U-Boot to allow
easily syncing dts between the two projects.
Reviewed-by: Rick Chen <rick@andestech.com>
Reviewed-by: Padmarao Begari <padmarao.begari@microchip.com>
Signed-off-by: Conor Dooley <conor.dooley@microchip.com>
If we're building non FU540/FU740 SoC drivers, then the sifive-prci.o
is not needed. Only build this when CONFIG_CLK_SIFIVE_PRCI is selected.
Signed-off-by: Ben Dooks <ben.dooks@sifive.com>
Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
These seem to be missing, and trying to build ubifs without them
is causing errors due to these being missing.
Signed-off-by: Ben Dooks <ben.dooks@sifive.com>
Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
Add implementations of the local_irq_{save,restore} macros so that
<asm/atomic.h> can be used with riscv.
Signed-off-by: Ben Dooks <ben.dooks@sifive.com>
Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
Add a link from <asm/atomic.h> to the generic one to allow
things like ubifs to be built. This can be extended with
riscv AMO ops at a later date.
Signed-off-by: Ben Dooks <ben.dooks@sifive.com>
Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
OpenSBI already implements some extensions that are not ratified yet:
* Debug Console Extension (DBCN)
* System Suspend Extension (SUSP)
* Collaborative Processor Performance Control Extension (CPPC)
Allow the sbi command to display these.
Provide the FID definitions of the Debug Console Extension. We can use that
extension for an early debug console driver.
Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
pimg64 image pointer is dependent on ESBC_ADDR_64BIT config, which is
getting disabled, due to dependency on ESBC_HDR_LS.
ESBC_HDR_LS is required for LS-CH3 platforms.
So, removing the dependency on ESBC_HDR_LS.
Signed-off-by: Kshitiz Varshney <kshitiz.varshney@nxp.com>
Acked-by: Peng Fan <peng.fan@nxp.com>
Reviewed-by: Gaurav Jain <gaurav.jain@nxp.com>
As the serial devices are configured in the device tree, enable
DM_SERIAL in the ls1046afrwy defconfigs.
Signed-off-by: Camelia Groza <camelia.groza@nxp.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
Reviewed-by: Peng Fan <peng.fan@nxp.com>
As the serial devices are configured in the device tree, enable
DM_SERIAL in the ls1046ardb defconfigs.
Signed-off-by: Camelia Groza <camelia.groza@nxp.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
Reviewed-by: Peng Fan <peng.fan@nxp.com>
Make sure the serial driver is initialized before relocation by tagging
the serial nodes with "bootph-all".
In order to keep the serial nodes in sync with their representation in
the Linux dts, add these u-boot specific properties to *-u-boot.dtsi
files.
Signed-off-by: Camelia Groza <camelia.groza@nxp.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
Reviewed-by: Peng Fan <peng.fan@nxp.com>
Pick up the serial node descriptions from Linux v6.3 for the ls1046ardb
and ls1046afrwy boards and their dependencies. Including the
fsl,qoriq-clockgen.h and arm-gic.h headers forces us to change the include
directives to explicitly go through the C preprocessor for all boards in
the ls1046a SoC family.
Signed-off-by: Camelia Groza <camelia.groza@nxp.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
Reviewed-by: Peng Fan <peng.fan@nxp.com>
As the serial devices are configured in the device tree, enable
DM_SERIAL in the ls1043ardb defconfigs.
Signed-off-by: Camelia Groza <camelia.groza@nxp.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
Reviewed-by: Peng Fan <peng.fan@nxp.com>
Make sure the serial driver is initialized before relocation by tagging
the serial nodes with "bootph-all".
In order to keep the serial nodes in sync with their representation in
the Linux dts, add these u-boot specific properties to *-u-boot.dtsi
files.
Signed-off-by: Camelia Groza <camelia.groza@nxp.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
Reviewed-by: Peng Fan <peng.fan@nxp.com>
Pick up the serial node descriptions from Linux v6.3 for the ls1043ardb
board and its dependencies. Including the fsl,qoriq-clockgen.h and
arm-gic.h headers forces us to change the include directives to explicitly
go through the C preprocessor for all boards in the ls1043a SoC family.
Signed-off-by: Camelia Groza <camelia.groza@nxp.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
Reviewed-by: Peng Fan <peng.fan@nxp.com>
Compiling on armv7 results in:
tools/renesas_spkgimage.c: In function ‘spkgimage_parse_config_line’:
tools/renesas_spkgimage.c:76:66: warning: format ‘%ld’ expects
argument of type ‘long int’, but argument 3 has type ‘size_t’
{aka ‘unsigned int’} [-Wformat=]
76 | "config error: unknown keyword on line %ld\n",
| ~~^
| |
| long int
| %d
77 | line_num);
| ~~~~~~~~
| |
| size_t {aka unsigned int}
The correct printf specifier for size_t is '%zu'.
Fixes: afdfcb11f9 ("tools: spkgimage: add Renesas SPKG format")
Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
Reviewed-by: Marek Vasut <marek.vasut+renesas@mailbox.org>
For changing the USB OTG node status from "okay" to "disabled" more
space is needed, so call fdt_increase_size() to avoid the following
error:
```
U-Boot 2023.07-rc5-0.0.0-devel+git.580eb31199be (Jun 27 2023 - 13:39:58 +0000)
CPU: Freescale i.MX7S rev1.2 800 MHz (running at 792 MHz)
CPU: Extended Commercial temperature grade (-20C to 105C) at 30C
Reset cause: POR
DRAM: initcall sequence 8786eafc failed at call 8781b351 (err=-3)
### ERROR ### Please RESET the board ###
```
Ideally, fdt_status_disabled() should call fdt_increase_size() internally,
so that there would be no need for manually calling it in board code.
Do it manually for now to fix the regression.
Based on the code from board/CZ.NIC/turris_omnia/turris_omnia.c.
Reported-by: Francesco Dolcini <francesco.dolcini@toradex.com>
Signed-off-by: Fabio Estevam <festevam@denx.de>
Reviewed-by: Francesco Dolcini <francesco.dolcini@toradex.com>
Tested-by: Francesco Dolcini <francesco.dolcini@toradex.com> # Toradex Colibri iMX7S
The regulator core can return different codes which are not considered
a real error for this function.
Return success in such cases.
Signed-off-by: Eugen Hristev <eugen.hristev@collabora.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Patrice Chotard <patrice.chotard@foss.st.com>
Simplify the subsystem by renaming `dev_pdata` to just `plat`.
No functional change, just trivial renaming.
Suggested-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Eugen Hristev <eugen.hristev@collabora.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Patrice Chotard <patrice.chotard@foss.st.com>
Some devices share a regulator supply, when the first one will request
regulator disable, the second device will have it's supply cut off before
graciously shutting down. Hence there will be timeouts and other failed
operations.
Implement a reference counter mechanism similar with what is done in
Linux, to keep track of enable and disable requests, and only disable the
regulator when the last of the consumers has requested shutdown.
Signed-off-by: Eugen Hristev <eugen.hristev@collabora.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Patrice Chotard <patrice.chotard@foss.st.com>
Currently, spl_end points to the __bss_end address, which
is an external RAM address instead of the end of the SPL text
section in the internal RAM.
This causes boot failures on imx6-colibri, for example:
```
Trying to boot from MMC1
SPL: Image overlaps SPL
resetting ...
```
Fix this problem by assigning spl_end to _image_binary_end, as this
symbol properly represents the end of the SPL text section.
From u-boot-spl.map:
.end
*(.__end)
0x00000000009121a4 _image_binary_end = .
Fixes: 77aed22b48 ("spl: spl_legacy: Add extra address checks")
Reported-by: Francesco Dolcini <francesco.dolcini@toradex.com>
Signed-off-by: Fabio Estevam <festevam@denx.de>
Tested-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Marek Vasut <marex@denx.de>
Tested-by: Marek Vasut <marex@denx.de> # DH i.MX6Q DHCOM PDK2
Pass _image_binary_end to make a standard way to indicate the end
of the text section in SPL.
The motivation for this is to have a uniform way to handle
the SPL boundary checks.
Signed-off-by: Fabio Estevam <festevam@denx.de>
Reviewed-by: Marek Vasut <marex@denx.de>
Pass _image_binary_end to make a standard way to indicate the end
of the text section in SPL.
The motivation for this is to have a uniform way to handle
the SPL boundary checks.
Signed-off-by: Fabio Estevam <festevam@denx.de>
Reviewed-by: Marek Vasut <marex@denx.de>
