When CONFIG_OF_HAS_PRIOR_STAGE is enabled and fdtfile is not set,
efi_get_distro_fdt_name() returns -EALREADY to indicate the prior-stage
FDT should be used. The block-device EFI path handles this by setting
BOOTFLOWF_USE_PRIOR_FDT, but the network path treats it as an error,
causing the bootflow to stay in 'base' state with a -EALREADY error.
This also means fdt_addr_r is required even when no FDT download is
needed, giving a spurious -EINVAL error.
Fix this by calling efi_get_distro_fdt_name() before checking
fdt_addr_r, and handling -EALREADY by setting BOOTFLOWF_USE_PRIOR_FDT
to skip the FDT download, matching the block-device behaviour.
THere is no test for this at present, since sandbox does not enable
CONFIG_OF_HAS_PRIOR_STAGE and lacks infra for network-based EFI boot.
Signed-off-by: Simon Glass <sjg@chromium.org>
Commit 4cb7243640 ("efi_loader: Disable ANSI output for tests")
introduced efi_console_set_ansi() to suppress ANSI escape sequences
during unit tests. Extend this mechanism to be configurable via a new
Kconfig option CONFIG_EFI_CONSOLE_DISABLE_ANSI.
When CONFIG_EFI_CONSOLE_DISABLE_ANSI is enabled,
efi_console_set_ansi(false) is called at the start of
efi_setup_console_size(). This prevents query_console_serial() from
sending ANSI escape sequences to the terminal, using default 25x80
dimensions instead. This is useful for platforms where the serial
console cannot handle ANSI queries.
Signed-off-by: Michal Simek <michal.simek@amd.com>
Reviewed-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Reviewed-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
Switch fwu_boottime_checks() from EVT_MAIN_LOOP to EVT_POST_PREBOOT
because there is no reason to call FWU so early. FWU triggers EFI
stack initialization before all devices are visible which prevents
the EFI stack from scanning these devices and adding them to EFI
variables.
Signed-off-by: Michal Simek <michal.simek@amd.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Add a new EVT_POST_PREBOOT event type which is fired in main_loop()
after the preboot command has been executed.
Signed-off-by: Michal Simek <michal.simek@amd.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Acked-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
event_notify_null() returns int but its return value is not
checked in run_main_loop() and in fwu_mdata tests.
Add proper error checking to all unchecked call sites.
Signed-off-by: Michal Simek <michal.simek@amd.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Acked-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
When SPI Flash is blank (first boot or erased), efi_var_from_storage()
returns EFI_DEVICE_ERROR because efi_var_restore() fails on invalid
magic/CRC. This prevents the EFI subsystem from initializing.
Check the magic value before attempting to restore variables. If
the magic does not match EFI_VAR_FILE_MAGIC, treat it as an empty
store and return EFI_SUCCESS, matching the behavior of the file-based
efi_var_file.c which deliberately returns EFI_SUCCESS on missing or
corrupted variable files to avoid blocking the boot process.
Similarly, if the magic matches but efi_var_restore() fails (e.g.
corrupted CRC), log the error but still return EFI_SUCCESS.
Suggested-by: John Toomey <john.toomey@amd.com>
Signed-off-by: Michal Simek <michal.simek@amd.com>
Reviewed-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Reviewed-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
These files are now auto-generated by the binman_docs Sphinx extension
during the doc build, so the committed copies and their symlinks in
doc/develop/package/ are no longer needed.
Update binman.rst to reference the Sphinx extension instead of the
manual generation commands.
Signed-off-by: Simon Glass <sjg@chromium.org>
Rebased to apply file deletions cleanly.
Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
Currently entries.rst and bintools.rst are generated manually by running
'binman entry-docs' and 'binman bintool-docs', then committed to the
repo. This means the docs can drift out of date when docstrings are
updated but the RST files are not regenerated.
Add a Sphinx extension (binman_docs) that provides two custom
directives:
.. binman-entry-docs::
.. binman-bintool-docs::
These parse the etype and btool source files using the ast module to
extract class docstrings, then insert the documentation directly into
the document tree. This avoids the need to import binman modules (which
have dependencies like libfdt that are not available in the ReadTheDocs
build environment) and avoids writing any intermediate files.
Signed-off-by: Simon Glass <sjg@chromium.org>
Remove the reference to the non-existent symbol. As ROMs typically
come as powers of two there seems no need for this value.
Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
Fixes: 64542f4616 ("x86: Make ROM_SIZE configurable in Kconfig")
Reviewed-by: Tom Rini <trini@konsulko.com>
Add a paragraph to explain how to get the address of the goldfish
tty and use it as the debug uart. I think the address is actually
fixed right now but it might change in the future.
Signed-off-by: Daniel Palmer <daniel@0x0f.com>
Once the stack is ready we can init the debug uart to help
with debugging so do that.
Signed-off-by: Daniel Palmer <daniel@0x0f.com>
Acked-by: Angelo Dureghello <angelo@kernel-space.org>
Add debug support for the goldfish tty so it can be used for
early debugging. This will be really useful when adding support
for relocation to the m68k qemu virt machine.
Signed-off-by: Daniel Palmer <daniel@0x0f.com>
Reviewed-by: Kuan-Wei Chiu <visitorckw@gmail.com>
Tested-by: Kuan-Wei Chiu <visitorckw@gmail.com>
The extern declaration in ll_entry_get() lacks the __aligned(4)
attribute present in ll_entry_declare(). When the compiler sees an
unaligned extern reference to a linker list entry in the same
compilation unit as its definition, it may increase the section
alignment beyond the expected struct size. This causes gaps in the
linker list array, which the alignment checker reports as failures.
For example, sandbox_dir is both defined and referenced via
DM_DRIVER_GET() in sandboxfs.c. The compiler applies 32-byte
alignment to its section instead of the 4-byte alignment from the
definition, creating an 8-byte gap before it in the driver list.
Add __aligned(4) to the extern declaration in ll_entry_get() to
match ll_entry_declare()
Signed-off-by: Simon Glass <simon.glass@canonical.com>
Change the alignment of end markers in ll_entry_end() and ll_end_decl()
from __aligned(4) and __aligned(CONFIG_LINKER_LIST_ALIGN) respectively
to __aligned(1).
The linker places zero-size end markers at aligned boundaries based on
what follows them. When the next list's start marker has a high alignment
requirement (e.g., 32 bytes), padding gets inserted before the end
marker. This causes the byte span (end - start) to not be an exact
multiple of the struct size.
The compiler optimises pointer subtraction (end - start) using
magic-number multiplication for division. This optimisation only produces
correct results when the byte span is an exact multiple of the struct
size. With padding, the result is garbage (e.g., -858993444 instead of
15).
By using __aligned(1), the end marker is placed immediately after the
last entry with no padding, ensuring (end - start) equals exactly
(n * sizeof) where n is the number of entries. This makes
ll_entry_count() and direct pointer arithmetic work correctly.
Fixes: 0b2fa98aa5 ("linker_lists: Fix alignment issue")
Signed-off-by: Simon Glass <simon.glass@canonical.com>
Replace unbounded sprintf() with snprintf() using sizeof(message)
as the bound for all prompt string assignments in
get_cfgblock_interactive(), get_cfgblock_carrier_interactive(),
do_cfgblock_carrier_create() and do_cfgblock_create(). The
previous calls had no size limit and could overflow the
CONFIG_SYS_CBSIZE-sized stack buffer if SYS_CBSIZE was configured
smaller than the longest prompt string (71 bytes).
Fixes: 8b6dc5d394 ("toradex: tdx-cfg-block: Cleanup interactive cfg block creation")
Signed-off-by: Ngo Luong Thanh Tra <S4210155@student.rmit.edu.au>
To: u-boot@lists.denx.de
Replace unbounded sprintf() with snprintf() using sizeof(info) as
the bound when constructing the board_name string from bdname and
bdtype. The previous call had no size limit and could overflow the
64-byte stack buffer if the concatenated string exceeded 63 bytes.
Fixes: c9c36bf56e ("samsung: misc: use board specific functions to set env board info")
Signed-off-by: Ngo Luong Thanh Tra <S4210155@student.rmit.edu.au>
To: u-boot@lists.denx.de
For some projects, designing the active bootable partition is done by
setting distro_bootpart in the environment instead of changing the bootable
flag on the active partition.
This patch makes it possible to support both models by scanning boot
partitions only if distro_bootpart is not defined in the environment.
Signed-off-by: Hugo Villeneuve <hvilleneuve@dimonoff.com>
Replace unbounded strcpy()/sprintf() calls with snprintf() and
check the return value against remaining buffer capacity at each
append step. The previous size guard did not account for
subsequent dpi suffix, remaining bootargs tail, and bootdev
token appends, allowing overflow when those later writes exceed
the remaining space.
Fixes: d1611086e0 ("arm: add support for SoC s5p4418 (cpu) / nanopi2 board")
To: u-boot@lists.denx.de
Signed-off-by: Ngo Luong Thanh Tra <S4210155@student.rmit.edu.au>
Tested-by: Stefan Bosch <stefan_b@posteo.net>
Add configuration for ARMv8-M aarch32 core, which are currently
Cortex-M23/M33 cores. These cores are treated similar to ARMv7-M
cores, except the code has to be compiled with matching compiler
-march=armv8-m.main flag . These cores have no MMU, they have MPU,
which is currently not configured.
Unlike ARMv7-M, these cores have 512 interrupt vectors. While the
SYS_ARM_ARCH should be set to 8, it is set to 7 because all of the
initialization code is built from arch/arm/cpu/armv7m and not armv8.
Furthermore, CONFIG_ARM64 must be disabled, although DTs for devices
using these cores do come from arch/arm64/boot/dts.
To avoid excess duplication in Makefiles, introduce one new Kconfig
symbol, CPU_V7M_V8M. The CPU_V7M_V8M cover both ARMv7-M and ARMv8-M
cores.
Signed-off-by: Marek Vasut <marek.vasut+renesas@mailbox.org>
Acked-by: Udit Kumar <u-kumar1@ti.com>
The current code sets RV3028_STATUS_PORF instead of clearing it, so the
flag remains asserted. Use dm_i2c_reg_clrset() to clear the bit.
Signed-off-by: Javier Viguera <javier.viguera@digi.com>
Add setuptools package configuration to pyproject.toml so that
u_boot_pylib is installed as a proper Python package without changing
the existing flat directory structure and making sure all modules are
installed.
Signed-off-by: Paul HENRYS <paul.henrys_ext@softathome.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
CI: https://source.denx.de/u-boot/custodians/u-boot-fsl-qoriq/-/pipelines/29808
- Add env variables to assist boot for various LS boards
- Add gpio scmi driver
- Fix setting the function for scmi pinctrl
- Use standard device tree pin muxing format for scmi pinctrl
- Fix protocol version fetch for non-CCF platforms in scmi clk
Add LS1043ARDB-specific variables to assist the boot process,
and update the related common and LS1043AQDS settings accordingly.
Signed-off-by: Chunguang Li <chunguang.li_2@nxp.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
Update the console baudrate to 115200 as default.
Also add env variables to assist boot process.
Signed-off-by: Chunguang Li <chunguang.li_2@nxp.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
Add LS1046ARDB-specific variables to assist the boot flow,
and update the related common and other LS1046A board settings
accordingly.
Signed-off-by: Chunguang Li <chunguang.li_2@nxp.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
Add image, extra_bootargs, othbootargs and console_dbg to assist boot.
Signed-off-by: Chunguang Li <chunguang.li_2@nxp.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
The process through which the MC firmware parses the DPL and initializes
all the requested DPAA2 objects is a complex one which can take quite a
bit of time. For the those circumstances in which a fast boot is
required on DPAA2 based SoCs, add the 'nowait' optional parameter for
the fsl_mc [lazy]apply dpl command.
When this option is used, the Linux kernel fsl-mc bus must wait for
the firmware to finish parsing the DPL before proceeding with probing
all the DPAA2 objects.
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
All the parameters that can be currently passed to the fsl_mc command
are positional arguments which are mandatory. This is not perfectly
clear when reading the help text because of the use of square brackets.
Fix this by changing the square brackets, which are commonly used for
optional parameters, with < .. >.
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
The first parameter of the wait_for_mc() function - booting_mc - is not
used. Remove it.
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
In the original code, I wrote a custom pin muxing parser but the
upstream device trees wouldn't accept something like that so it would
have complicated mergine the device tree files.
Use the standard device tree format with function and groups:
pinmux1: pinmux1 {
function = "f_gpio1";
groups = "grp_1", "grp_3";
};
Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
Acked-by: Linus Walleij <linusw@kernel.org>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
Set BIT(10) when the function needs to be set, otherwise the setting is
ignored.
Fixes: 0cb160f1b6 ("scmi: pinctrl: add pinctrl driver for SCMI")
Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
This provides GPIO support over SCMI. It is built on top of the
pinctrl-scmi driver. A typical device tree entry might look like
this:
gpio1 {
compatible = "scmi-pinctrl-gpio";
gpio-controller;
#gpio-cells = <2>;
ngpios = <10>;
gpio-ranges = <&scmi_pinctrl 0 8 4>,
<&scmi_pinctrl 4 12 1>,
<&scmi_pinctrl 5 15 1>,
<&scmi_pinctrl 6 17 4>;
pinctrl-names = "default";
pinctrl-0 = <&i2c2_pins>;
};
In this GPIO driver the one thing which is different is that in the
gpio-ranges the first numbers which represent how the pins are exposed
to the users have to start at zero and it can't have gaps.
Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
The SCMI clock protocol version was only being fetched when CLK_CCF
was enabled. On non-CCF platforms, the probe function returned early
without fetching the version, leaving priv->version as 0.
This caused issues because code paths like scmi_clk_gate() and
scmi_clk_get_permissions() depend on priv->version to determine
which protocol message format to use, even in non-CCF mode.
Fix this by moving the scmi_generic_protocol_version() call before
the CLK_CCF check, ensuring the version is fetched for both CCF and
non-CCF platforms.
Tested on am62lx_evm.
Fixes: ae7e0330ce ("clk: scmi: add compatibility with clock protocol 2.0")
Signed-off-by: Kamlesh Gurudasani <kamlesh@ti.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
Use dev_phys_to_bus() to convert CPU addresses of DMA descriptors
into bus addresses of DMA descriptors. This is necessary on hardware
which does not have 1:1 mapping between CPU and memory addressed by
the DMA. This has no impact on other hardware which does not need
this conversion.
Signed-off-by: Marek Vasut <marek.vasut+renesas@mailbox.org>
The HSCIF variant present on Renesas R-Car Gen5 SoC is compatible
with the HSCIF variant present on Renesas R-Car Gen4 SoC. Enable
HSSRR register programming for HSCIF present on all 64-bit R-Car
SoCs, which covers R-Car Gen3, Gen4 and newly also Gen5.
Signed-off-by: Marek Vasut <marek.vasut+renesas@mailbox.org>
Fix typo in Kconfig symbol help text, change incorrect A52 to correct R52.
No functional change.
Signed-off-by: Marek Vasut <marek.vasut+renesas@mailbox.org>
Acked-by: Peng Fan <peng.fan@nxp.com>
Use DRAM_CH_CNT macro in renesas_dbsc5_board_config channel count.
The macro is defined in the same file a few lines above. This way,
the maximum channel count in the structure and the macro can not
diverge.
Signed-off-by: Marek Vasut <marek.vasut+renesas@mailbox.org>
Conditionally compile code that is only compatible with 64bit ARMv8
on 64bit R-Car Gen3/4/5 SoCs. Protect such code with CONFIG_ARM64.
This split is implemented in preparation for build of Cortex-M33
version of R-Car Gen5 U-Boot for its RSIPM core.
Signed-off-by: Marek Vasut <marek.vasut+renesas@mailbox.org>
The 64-bit R-Car board code is currently unbuildable on 32-bit ARM
cores. Skip DRAM initialization bits to make the code compatible with
these cores, because those cores do not use the DRAM, they only use
SRAM.
Signed-off-by: Marek Vasut <marek.vasut+renesas@mailbox.org>
Enable the 'smc' command on all R-Car 64-bit systems. This command is
useful for interacting with EL3 firmware, testing interaction with the
PSCI provider or OPTEE-OS.
Signed-off-by: Marek Vasut <marek.vasut+renesas@mailbox.org>
Replace s_init() early initialization at the end of lowlevel_init
by invoking the same code in mach_cpu_init(). The mach_cpu_init()
is called a bit later, but as the code initializes timer and no
code uses timer until mach_cpu_init(), this does not pose a problem.
Signed-off-by: Marek Vasut <marek.vasut+renesas@mailbox.org>
Enable the 'preboot' variable on all R-Car systems. This variable can
optionally be set and contain a script which is executed before the
autoboot timeout starts. This can be used to run critical scripts or
similar tasks. By default, the "preboot" variable is empty.
Signed-off-by: Marek Vasut <marek.vasut+renesas@mailbox.org>
Enable modern HUSH parser on all Renesas R-Car systems. This replaces
the old HUSH parser and includes matching updates to the command line
parser. No functional change expected.
Signed-off-by: Marek Vasut <marek.vasut+renesas@mailbox.org>
David Lechner <dlechner@baylibre.com> says:
This is a series adding support for reading U-Boot env directly from
SCSI devices that do not have a partition table, similar to how we can
already do this for MMC devices.
The motivation behind this is that MediaTek's BSP is already using the
same disk images for both MMC and UFS devices, so we need to be able to
read the env from SCSI devices without requiring a partition UUID.
The series starts with cleaning up a few oddities we noticed in the
existing code. Then some refactoring so that the env code manages
calling scsi_scan() so that we don't have to duplicate that for the
new code path. Then finally, the last few patches add and document the
new functionality.
Link: https://lore.kernel.org/r/20260326-env-scsi-hw-part-support-v1-0-55c9dd07a2cb@baylibre.com
Expand the Kconfig help for ENV_IS_IN_SCSI to explain the other
required config options when this option is enabled.
Signed-off-by: David Lechner <dlechner@baylibre.com>
