When probing a device fails NULL pointer is returned, and following
devices in uclass list cannot be iterated. Skip to next device on error
instead.
With that the only condition under which these simple iteration
functions return error is when the dm is not initialized at uclass_get
time. This is not all that interesting, change return type to void.
Fixes: 6494d708bf ("dm: Add base driver model support")
Signed-off-by: Michal Suchanek <msuchanek@suse.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
uclass_first_device/uclass_next_device return value will be removed,
don't use it.
With the current implementation dev is equivalent to !ret. It is
redundant to check both, ret check can be replaced with dev check, and
ret check inside the iteration is dead code.
Signed-off-by: Michal Suchanek <msuchanek@suse.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
device_probe returns early when the device is already activated.
Add a note to the documentation that it can be used on already activated
devices.
Signed-off-by: Michal Suchanek <msuchanek@suse.de>
The distutils package is deprecated. The upstream libfdt repo uses
setuptools for building the pylibfdt module, so bring in that code,
suitably modified for U-Boot. Also bring in the README.
The modifications include setting the version correctly, making use of
the environment variables provided by the Makefile and various tweaks
to the directories.
Note that the version omits the minus character at the start of
EXTRAVERSION, since this creates a warning. The build is really just used
within U-Boot itself, so it doesn't matter too much if the version matches
upstream, or exactly matches U-Boot.
Signed-off-by: Simon Glass <sjg@chromium.org>
The sandbox64 variant is currently building the sandbox thermal driver
but not the corresponding uclass driver. This results in the sandbox64
variant not booting with the test device tree. Enable building the
thermal uclass for the sandbox64 variant as well.
Also enable the temperature command to allow the test to be run on the
sandbox64 variant.
Signed-off-by: Sughosh Ganu <sughosh.ganu@linaro.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
Adding support to symlink an image packaged using binman.
Signed-off-by: Neha Malcom Francis <n-francis@ti.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Currently sandbox configuration defautls to 64bit and there is no
automation for building 32bit sandbox on 32bit hosts.
Use _LP64 macro as heuristic for detecting 64bit targets.
Signed-off-by: Michal Suchanek <msuchanek@suse.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
This is needed in the platforms that use "ranges" node property for
address translation in their dts for TPL.
Signed-off-by: William Zhang <william.zhang@broadcom.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Python 3.10 requires defining PY_SSIZE_T_CLEAN. This will be fixed in
swig 4.10 but it is not clear when it will be released. There was a
warning since python 3.8.
Link: https://github.com/swig/swig/pull/2277
Signed-off-by: Michal Suchanek <msuchanek@suse.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
python does not like the u-boot- prefix in the version, drop it.
/usr/lib/python3.10/site-packages/setuptools/dist.py:544: UserWarning:
The version specified ('u-boot-2022.10') is an invalid version, this may
not work as expected with newer versions of setuptools, pip, and PyPI.
Please see PEP 440 for more details.
Signed-off-by: Michal Suchanek <msuchanek@suse.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
This path does not seem to be present in clang-14 for some reason. Relax
the regular expression so that the test works, at least for non-LTO.
Signed-off-by: Simon Glass <sjg@chromium.org>
The call to device_bind_with_driver_data() passes id->data but if
the entry has no of_match then the id has not been set by the selected
driver.
Normally this passes unnoticed since a previous driver likely had an
of_match value, so the id is set to that. Of course it is not correct
to pass the id->data from a different driver.
With clang-14 the driver ordering is such that the id is never actually
set in the 'bind /usb@1 usb_ether' line in test_bind_unbind_with_node()
thus causing a crash.
Fix this by passing 0 if the of_match for a driver does not exist.
Signed-off-by: Simon Glass <sjg@chromium.org>
Matching anything that just happens to have the sought-for label as a
prefix is wrong. For example, if the board designer has designated 10
lines for debug purposes, named "debug1" through "debug10", and we are
looking up "debug1", if debug10 happens to be met first during the
iteration we'd wrongly return that.
In theory, this can break existing users that could rely on this
quirk, but OTOH keeping the current broken semantics can cause a lot
of grief for people hitting this in the future and not understanding
why they don't find the line they expect. Considering how few in-tree
defconfigs currently set DM_GPIO_LOOKUP_LABEL (ignoring sandbox, only
four "real" boards), let's fix it before the use becomes more
widespread.
Signed-off-by: Rasmus Villemoes <rasmus.villemoes@prevas.dk>
Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Heiko Schocher <hs@denx.de>
The jmp_buf type is required by the C99 specification.
Defining it for x86 fixes building the longjmp unit test.
Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
If longjmp(jmp_buf env, int val) is called with val = 0, the setjmp()
macro must return 1.
Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
pci_find_first_device description says it can be used for iteration with
itself but it should really be with pci_find_next_device
Signed-off-by: Michal Suchanek <msuchanek@suse.de>
Remove prototype for the removed function fdt_fixup_nor_flash_size.
This patch has no impact as the function is never used.
Fixes: 98f705c9ce ("powerpc: remove 4xx support")
Signed-off-by: Patrick Delaunay <patrick.delaunay@foss.st.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
In a couple of places the document says u-boot,pre-reloc but all
examples show u-boot,dm-pre-reloc, use the latter consistently.
Signed-off-by: Michal Suchanek <msuchanek@suse.de>
Reviewed-by: Fabio Estevam <festevam@denx.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
There are two problems with both strcmp and strncmp:
(1) The C standard is clear that the contents should be compared as
"unsigned char":
The sign of a nonzero value returned by the comparison functions
memcmp, strcmp, and strncmp is determined by the sign of the
difference between the values of the first pair of characters (both
interpreted as unsigned char) that differ in the objects being
compared.
(2) The difference between two char (or unsigned char) values can
range from -255 to +255; so that's (due to integer promotion) the
range of values we could get in the *cs-*ct expressions, but when that
is then shoe-horned into an 8-bit quantity the sign may of course
change.
The impact is somewhat limited by the way these functions
are used in practice:
- Most of the time, one is only interested in equality (or for
strncmp, "starts with"), and the existing functions do correctly
return 0 if and only if the strings are equal [for strncmp, up to
the given bound].
- Also most of the time, the strings being compared only consist of
ASCII characters, i.e. have values in the range [0, 127], and in
that case it doesn't matter if they are interpreted as signed or
unsigned char, and the possible difference range is bounded to
[-127, 127] which does fit the signed char.
For size, one could implement strcmp() in terms of strncmp() - just
make it "return strncmp(a, b, (size_t)-1);". However, performance of
strcmp() does matter somewhat, since it is used all over when parsing
and matching DT nodes and properties, so let's find some other place
to save those ~30 bytes.
Signed-off-by: Rasmus Villemoes <rasmus.villemoes@prevas.dk>
- Reduce memory usage in SPL in some cases, clarify some standalone API
license issues, fix a Kconfig dependency, pin to a specific version of
python setuptools for now, fix a signing problem in mkimage and add a
memory uclass.
The GPMC is a unified memory controller dedicated for interfacing
with external memory devices like
- Asynchronous SRAM-like memories and ASICs
- Asynchronous, synchronous, and page mode burst NOR flash
- NAND flash
- Pseudo-SRAM devices
This driver will take care of setting up the GPMC based on
the settings specified in the Device tree and then
probe its children.
Signed-off-by: Roger Quadros <rogerq@kernel.org>
GPMC stands for General Purpose Memory Controller and it is
present on many Texas Instruments SoCs.
It supports a number of Asynchronous and Synchronous interfaces
and has various settings to configure the bus interface.
The DT bindings define all the various GPMC settings.
As the GPMC supports multiple devices on the bus, each
device is represented as a child and the respective
GPMC settings are situated there. (see ti,gpmc-child.yaml)
These binding docs are picked up from the Linux kernel.
Signed-off-by: Roger Quadros <rogerq@kernel.org>
Introduce CONFIG_SPL_MEMORY to allow Memory drivers to
be built for SPL.
Signed-off-by: Roger Quadros <rogerq@kernel.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
The version used on Ubuntu 2022.04 produces a number of warnings:
/usr/lib/python3/dist-packages/pkg_resources/__init__.py:116:
PkgResourcesDeprecationWarning: 1.16.0-unknown is an invalid version
and will not be supported in a future release
Same with: 0.1.43ubuntu1 11.4.1ubuntu1 2.22.1ubuntu1 1.1build1
According to [1] this is a bug in setuptools. Employ the workaround for
now.
[1] https://askubuntu.com/questions/1406952/what-is-the-meaning-of-this-
pkgresourcesdeprecationwarning-warning-from-pipenv
Signed-off-by: Simon Glass <sjg@chromium.org>
The "common.h" header is not covered by the licensing exception for
standalone applications. Let's drop inclusion of this header from the
hello_world example to prove that a standalone app can be built without
it.
Signed-off-by: Paul Barker <paul.barker@sancloud.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
On 2010-01-27, an email [1] was sent to the mailing list by Wolfgang
Denk which clarified the intended licensing exceptions for standalone
applications. As the "export.h" header and the "stubs.c" source files
are required to implement a standalone application, the intention was
that these files be covered by the licensing exception. This is made
clear in the following quotes from that email:
"exports.h" should be added to the "allowed" file list; there should
be no need to include "common.h". Eventually this needs fixing.
Patches are welcome.
"examples/standalone/stubs.c" should be added to the "allowed" file
list (the ppc_*jmp.S files are LGPLed).
There should be no doubts - the intention is clear, the current state
may need improvement. Help (read: patches) welcome.
[1]: https://lists.denx.de/pipermail/u-boot/2010-January/067174.html
Signed-off-by: Paul Barker <paul.barker@sancloud.com>
Cc: Wolfgang Denk <wd@denx.de>
Acked-by: Wolfgang Denk <wd@denx.de>
OMAP3 BeagleBoard NAND boot hangs when spl_load_legacy_img() tries
to read the header into 'struct hdr' which is allocated on the
stack.
As the header has already been read once before by spl_nand.c,
we can avoid the extra header allocation and read here by
simply passing around the pointer to the header.
This fixes NAND boot on OMAP3 BeagleBoard.
Signed-off-by: Roger Quadros <rogerq@kernel.org>
Reviewed-By: Michael Trimarchi <michael@amarulasolutions.com>
Just like we exclude data-size, data-position, and data-offset from
fit_config_check_sig, we must exclude them while signing as well.
While we're at it, use the FIT_DATA_* defines for fit_config_check_sig
as welll.
Fixes: 8edecd3110 ("fit: Fix verification of images with external data")
Fixes: c522949a29 ("rsa: sig: fix config signature check for fit with padding")
Signed-off-by: Sean Anderson <sean.anderson@seco.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
- nuvoton: add expire function for generic reset (Jim)
- handle watchdogs during keyed autoboot (Rasmus)
- cyclic: Don't disable cylic function upon exceeding CPU time (Stefan)
- ulp wdog: Updates to support iMX93 and DM (Alice)
Avoid concatenation with following message.
Signed-off-by: Michal Suchanek <msuchanek@suse.de>
Reviewed-by: Peng Fan <peng.fan@nxp.com>
Reviewed-by: Jaehoon Chung <jh80.chung@samsung.com>
If a regulator does not support .set_suspend_enable or
.set_suspend_value then ret is set to ENOSYS early in the function.
The most serious impact of this is that when no automatic setting of
voltage is needed then the final regulator_set_enable() is skipped
because ret has not been cleared.
It seems that the error handling for regulator_set_suspend_value() is
also wrong as if this succeeds then the normal boot-on checks are still
required, and again ENOSYS needs special treatment here.
Fixes: 11406b8f7e ("dm: regulator: support regulator more state")
Signed-off-by: John Keeping <john@metanate.com>
Currently the 'sf update' command fails in case the 'start' offset is
not aligned to SPI NOR erase block size. Add the missing alignment
calculation. In case the start offset is in the middle of erase block,
round start address down to the nearest aligned one, compare only the
updated data between what is in the SPI NOR and what is being written,
copy new data at offset of the compare buffer, and write back the entire
erase block.
This is useful e.g. on i.MX6Q where the u-boot-with-spl.imx is at
offset 0x400 in the SPI NOR, while the SPI NOR may have erase block
size e.g. 0x1000 bytes.
Signed-off-by: Marek Vasut <marex@denx.de>
Reviewed-by: Jagan Teki <jagan@amarulasolutions.com>
Layerscape update
- support sysreset,
- de-select FSL_IFC when booting from SD
- disable unused parts of ICID tables
- reduce ns_dev size for csu
- enable dma snooping for ls104x
- nand driver fixups for ls1043ardb rev 7.0 boards.
Enable boot/bootd command in toradex colibri-imx7 defconfig,
it's just convenient to have it in and every other toradex board
already includes it.
Signed-off-by: Francesco Dolcini <francesco.dolcini@toradex.com>
Acked-by: Marcel Ziswiler <marcel.ziswiler@toradex.com>
Avoid the following build errors after the device tree sync:
drivers/spi/mxs_spi.c: In function ‘mxs_spi_probe’:
drivers/spi/mxs_spi.c:327:25: error: ‘struct dtd_fsl_imx23_spi’ has no
member named ‘spi_max_frequency’
327 | priv->max_freq = dtplat->spi_max_frequency;
| ^~
drivers/spi/mxs_spi.c:328:23: error: ‘struct dtd_fsl_imx23_spi’ has no
member named ‘num_cs’
328 | plat->num_cs = dtplat->num_cs;
| ^~
Signed-off-by: Marcel Ziswiler <marcel.ziswiler@toradex.com>
