| Age | Commit message (Collapse) | Author |
|---|
|
Rework dev_phys_to_bus() and dev_bus_to_phys() to respect the size
of the area specified in dma-ranges DT property. The area outside
of ranges is remapped 1:1, while the area in the ranges is remapped
according to the description in the dma-ranges property.
Adjust the test to test the area within the remapped range, not area
outside the remapped range, which was incorrect.
Signed-off-by: Marek Vasut <[email protected]>
Reviewed-by: Markus Schneider-Pargmann <[email protected]>
|
|
The virtio spec clearly states in "3.1.1 Driver Requirements: Device
Initialization" the sequence a client has to follow after device reset.
Because u-boot resets here again, it also needs to set the "acknowledge"
bit again even if this was done in virtio_uclass_child_post_bind already
once before.
Signed-off-by: Christian Pötzsch <[email protected]>
Signed-off-by: Adam Lackorzynski <[email protected]>
[trini: Add VIRTIO_CONFIG_S_ACKNOWLEDGE flag check to the test]
Signed-off-by: Tom Rini <[email protected]>
|
|
Meet Patel <[email protected]> says:
The virtio-rng test to verify effective handling of oversized return
buffers checks that an (undocumented) error is raised, instead of the
real concern, which is the surrounding buffer integrity following a rng
function call.
Update the test to check that the other contents of a buffer remain
unchanged instead of looking for an error code.
Link: https://lore.kernel.org/r/[email protected]
|
|
The virtio-rng test to verify effective handling of oversized return
buffers checks that an (undocumented) error is raised, instead of the
real concern, which is the surrounding buffer integrity following a rng
function call.
Update the test to check that the other contents of a buffer remain
unchanged instead of looking for an error code.
Signed-off-by: Kavin Gunasekara <[email protected]>
Signed-off-by: Meet Patel <[email protected]>
Reviewed-by: Andre Przywara <[email protected]>
|
|
Remove DECLARE_GLOBAL_DATA_PTR from files where gd is not used, and
drop the unnecessary inclusion of asm/global_data.h.
Headers should be included directly by the files that need them,
rather than indirectly via global_data.h.
Reviewed-by: Patrice Chotard <[email protected]> #STMicroelectronics boards and STM32MP1 ram test driver
Tested-by: Anshul Dalal <[email protected]> #TI boards
Acked-by: Yao Zi <[email protected]> #TH1520
Signed-off-by: Peng Fan <[email protected]>
|
|
Add a test for binding of multiple drivers with the same compatible. If
one of the drivers returns -ENODEV the other one needs to be bound.
Reviewed-by: Simon Glass <[email protected]>
Reviewed-by: Mattijs Korpershoek <[email protected]>
Signed-off-by: Markus Schneider-Pargmann (TI.com) <[email protected]>
|
|
* Expected values must always be the first arguments.
* Long values on 64 bit systems require ut_asserteq_64() for checking
Signed-off-by: Heinrich Schuchardt <[email protected]>
|
|
https://source.denx.de/u-boot/custodians/u-boot-socfpga into next
This pull request brings together a set of fixes and enhancements across
the SoCFPGA platform family, with a focus on MMC/SPL robustness, EFI
boot enablement, and Agilex5 SD/eMMC support.
CI: https://source.denx.de/u-boot/custodians/u-boot-socfpga/-/pipelines/28776
Highlights:
*
SPL / MMC:
o
Fix Kconfig handling for
SYS_MMCSD_RAW_MODE_U_BOOT_USE_PARTITION_TYPE
o
Correct raw sector calculations and respect explicit sector values
when loading U-Boot from MMC in SPL
o
Adjust raw MMC loading logic for SoCFPGA platforms
*
EFI boot:
o
Permit EFI booting on SoCFPGA platforms
o
Disable mkeficapsule tool build for Arria 10 where unsupported
*
Agilex5:
o
Upgrade SDHCI controller from SD4HC to SD6HC
o
Enable MMC and Cadence SDHCI support in defconfig
o
Add dedicated eMMC device tree and defconfig for Agilex5 SoCDK
o
Revert incorrect GPIO configuration for SDIO_SEL
o
Refine U-Boot DT handling for SD and eMMC boot variants
*
SPI:
o
Allow disabling the DesignWare SPI driver in SPL via Kconfig
*
Board / configuration fixes:
o
Enable random MAC address generation for Cyclone V
o
Fix DE0-Nano-SoC boot configuration
o
Remove obsolete or conflicting options from multiple legacy
SoCFPGA defconfigs
|
|
While this change was intended to fix a mistake in the code, of calling
the ERR_PTR macro but not making use of the result, it seems that
functionally platforms depend on the loop not existing here. The TI K3
families of platforms for example were broken by this commit.
This reverts commit fe780310cfa8bf5a093894b5cd7fe85c6b02fd91.
Reported-by: Nishanth Menon <[email protected]>
Reviewed-by: Andrew Goodbody <[email protected]>
Reviewed-by: Nishanth Menon <[email protected]>
Signed-off-by: Tom Rini <[email protected]>
|
|
The current FDT decoding tests calculate the memory required
for FDT manipulation by directly adding a fixed margin to
fdt_totalsize(gd->fdt_blob). The static analyzer flagged
"gd->fdt_blob->totalsize" as a tainted value being passed
to fdt_open_into().
Ensure the size is validated by checking that the total size
is within a reasonable maximum FDT limit for unit tests.
Signed-off-by: Adriana Nicolae <[email protected]>
|
|
Prepare v2026.01-rc4
|
|
Andrew Goodbody <[email protected]> says:
Smatch reported an error where a value calculated by ERR_PTR was not
used. Fixing this to return the generated value led to a test failure
which meant updating the sandbox clock code so that it would still cause
the tests to pass with the above correction.
Debugging this problem led to a SIGSEGV which is addressed in 1/3.
Possible memory leaks noticed are addressed in 3/3.
Link: https://lore.kernel.org/r/[email protected]
|
|
In clk_set_default_get_by_id ret is passed to ERR_PTR but nothing is
done with the value that this calculates which is obviously not the
intention of the code. This is confirmed by the code around where this
function is called.
Instead return the value from ERR_PTR.
Then fixup the sandbox code so that the test dm_test_clk does not fail
as it relied on the broken behaviour.
Finally disable part of the test that does not work correctly with
CLK_AUTO_ID
This issue found by Smatch.
Signed-off-by: Andrew Goodbody <[email protected]>
|
|
The Linux kernel can discover SMBIOS tables through two primary methods:
1. Via EFI tables, when using EFI boot;
2. Via the 'smbios3-entrypoint' property in the /chosen node of the
device tree.
When U-Boot boots a Linux kernel using a non-EFI command ("bootm",
"bootz", or "booti"), the kernel relies on the device tree to detect
the hardware. If SMBIOS tables are available in U-Boot, they should
be passed to the kernel via this device tree property.
This patch modifies boot_fdt_prepare(), to inject the SMBIOSv3 table
address into the device tree if there is a table generated by U-boot.
The "board_fdt_chosen_smbios" is weak in order to leave the possibilty
for specific boards to select custom SMBIOS addresses.
The changes in this patch are added in the context of supporting this
device tree property in linux kernel:
https://lkml.org/lkml/2025/10/24/1393
Device tree schema was updated to include the "smbios3-entrypoint" node
in pull request: https://github.com/devicetree-org/dt-schema/pull/177
Signed-off-by: Adriana Nicolae <[email protected]>
|
|
Trivially fix the following warnings in sandbox DTs, which show up with
DTC 1.7.2. Fill in the missing address and adjust emulated I2C address
to fit the 7bit address limit:
"
arch/sandbox/dts/sandbox.dtsi:138.30-140.5: Warning (i2c_bus_reg): /i2c@0/sandbox_pmic: I2C bus unit address format error, expected "40"
arch/sandbox/dts/sandbox.dtsi:146.18-161.5: Warning (i2c_bus_reg): /i2c@0/emul: I2C bus unit address format error, expected "ff"
arch/sandbox/dts/sandbox.dtsi:148.4-17: Warning (i2c_bus_reg): /i2c@0/emul:reg: I2C address must be less than 7-bits, got "0xff". Set I2C_TEN_BIT_ADDRESS for 10 bit addresses or fix the property
"
"
arch/sandbox/dts/.test.dtb.pre.tmp:912.18-926.5: Warning (i2c_bus_reg): /i2c@0/emul: I2C bus unit address format error, expected "ff"
arch/sandbox/dts/.test.dtb.pre.tmp:913.4-17: Warning (i2c_bus_reg): /i2c@0/emul:reg: I2C address must be less than 7-bits, got "0xff". Set I2C_TEN_BIT_ADDRESS for 10 bit addresses or fix the property
arch/sandbox/dts/.test.dtb.pre.tmp:928.30-931.5: Warning (i2c_bus_reg): /i2c@0/sandbox_pmic: I2C bus unit address format error, expected "40"
"
Fix up pmic test to match.
Signed-off-by: Marek Vasut <[email protected]>
Reviewed-by: Heiko Schocher <[email protected]>
Reviewed-by: Mattijs Korpershoek <[email protected]>
|
|
Add a test suite exercising the whole lifetime and callbacks
of interconnect with a fake 5 providers with a split node graph.
The test suite checks the calculus are right and goes to the correct
nodes, and the lifetime of the node is correct.
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Neil Armstrong <[email protected]>
|
|
lookup"
Rasmus Villemoes <[email protected]> says:
Hopefully third time's the charm.
I merely wanted to add support (mostly for use by the 'gpio' shell
command) for looking up a gpio via the gpio-line-names DT property. We
already have a "gpio_request_by_line_name()", but cmd/gpio.c does a
separate "lookup + request", so it felt more natural to teach the
lookup machinery this as well. That ran into
OF_CONTROL-but-not-OF_LIBFDT being a thing for SPL, so here's yet
another attempt.
Now, when trying to do my civic duty and add tests for this, I found
that test/dm/gpio.c has been defunct for a couple of years, and
reinstating it is not entirely trivial.
After a couple of rounds CI is now happy with this:
https://github.com/u-boot/u-boot/pull/828
Link: https://lore.kernel.org/r/[email protected]
|
|
Signed-off-by: Rasmus Villemoes <[email protected]>
Reviewed-by: Heiko Schocher <[email protected]>
|
|
Commit ebaa3d053e5 ("test: fix CONFIG_ACPIGEN dependencies"), which
got into v2022.10-rc1, accidentally left out a $
before (CONFIG_DM_GPIO), with the effect that test/dm/gpio.c has not
been built for three years.
Unsurprisingly, the code in there has bit-rotted.
- There's a missing ; causing plain build fail.
That code was added in 9bf87e256c2 ("test: dm: update test for
open-drain/open-source emulation in gpio-uclass"), which was part of
v2020.07-rc3, i.e. long before the commit causing gpio.c to not be
built at all. It did build at that time, but also, the missing
semicolon wasn't found when fa847bb409d ("test: Wrap assert macros
in ({ ... }) and fix missing semicolons") happened in 2023.
- Commit 592b6f394ae ("led: add function naming option from linux")
bumped sandbox,gpio-count for bank gpio_a in test.dts to 25, but
didn't update the expected global gpio numbers accordingly.
- The "lookup by label" test likely worked when it was added, but then I
inadvertently broke it when I noticed that dm_gpio_lookup_label()
seemed to be broken in commit 10e66449d7e ("gpio-uclass: fix gpio
lookup by label") - which landed in v2023.01-rc1, so after gpio.c
was no longer being built.
The "label" (which is a u-boot concept) that a "hogged gpio" gets is
<gpio hog node name>.gpio-hog, which is why it used to work with the
strncmp() but doesn't with strcmp().
We can either revert 10e66449d7e or append the ".gpio-hog" suffix as
done below. I don't really have a dog in that race; when I did
10e66449d7e, it was because I thought the "lookup by label" was
actually about the standardized gpio-line-names property, but then I
learnt it was not, so is not at all useful to me.
- The leak check now fails.
Test: gpio_leak: gpio.c
test/dm/core.c:112, dm_leak_check_end(): uts->start.uordblks == end.uordblks: Expected 0x2a95b0 (2790832), got 0x2a9650 (2790992)
test/dm/gpio.c:328, dm_test_gpio_leak(): 0 == dm_leak_check_end(uts): Expected 0x0 (0), got 0x1 (1)
Test: gpio_leak: gpio.c (flat tree)
test/dm/core.c:112, dm_leak_check_end(): uts->start.uordblks == end.uordblks: Expected 0x2a9650 (2790992), got 0x2a9700 (2791168)
test/dm/gpio.c:328, dm_test_gpio_leak(): 0 == dm_leak_check_end(uts): Expected 0x0 (0), got 0x1 (1)
And it fails with the same differences (160/176) even if I
remove the three lines that actually exercise any of the gpio code,
i.e. make the whole function amount to
ut_assertok(dm_leak_check_end(uts));
Test: gpio_leak: gpio.c
test/dm/core.c:112, dm_leak_check_end(): uts->start.uordblks == end.uordblks: Expected 0x2a95b0 (2790832), got 0x2a9650 (2790992)
test/dm/gpio.c:325, dm_test_gpio_leak(): 0 == dm_leak_check_end(uts): Expected 0x0 (0), got 0x1 (1)
Test: gpio_leak: gpio.c (flat tree)
test/dm/core.c:112, dm_leak_check_end(): uts->start.uordblks == end.uordblks: Expected 0x2a9650 (2790992), got 0x2a9700 (2791168)
test/dm/gpio.c:325, dm_test_gpio_leak(): 0 == dm_leak_check_end(uts): Expected 0x0 (0), got 0x1 (1)
So I suspect that the leak is somewhere in the test framework
setup/teardown code - dm_leack_check_end() isn't really used
anywhere else except in a dm/core test. Bisecting to figure out
where that was introduced is somewhat of a hassle because of the
other bitrot, and because of the SWIG failure that makes it very
hard to build older U-Boots.
So since it's better to have most of the gpio tests actually
working instead of leaving all of gpio.c as dead code, #if 0 that
part out and leave it as an archeological exercise.
Signed-off-by: Rasmus Villemoes <[email protected]>
Reviewed-by: Heiko Schocher <[email protected]>
|
|
The U-Boot project has been assigned the vendor ID 'UBOO' [1]. Use this
vendor ID and our release version in the ACPI table headers.
[1] ACPI ID Registry
https://uefi.org/ACPI_ID_List
Signed-off-by: Heinrich Schuchardt <[email protected]>
Reviewed-by: Bin Meng <[email protected]>
|
|
Some tests here had not been compile tested before submission and were
missing a ';' on the end of declaring struct udevice *dev. Add it.
Fixes: 9046279d92bb ("test: dm: Add tests for LED boot and activity")
Signed-off-by: Tom Rini <[email protected]>
|
|
The video test here is specific to the sandbox SDL video driver, so only
build it when that is enabled rather than VIDEO is enabled.
Reported-by: Alison Chaiken <[email protected]>
Signed-off-by: Tom Rini <[email protected]>
|
|
Add s simple test for the I3C uclass in sandbox.
Signed-off-by: Dinesh Maniyam <[email protected]>
|
|
uclass_find_next_device always returns 0, so instead make it a void and
update calling sites.
Signed-off-by: Andrew Goodbody <[email protected]>
|
|
In [0], Andrew noted a code quality issue in the implementation of
blk_find_first and blk_find_next. This led to the observation that the
logic of these functions was also likely incorrect, and based on a quick
check it seemed the functions were unused outside of test code, which
did not exercise the potential failure case, so we felt they should be
removed. In [1], a test patch which illustrates the failure in sandbox
is provided for reference.
Because a more thorough check agrees that these functions are unused,
they are currently incorrect, and fixed/removable flags on block devices
prior to probe are unreliable, just remove these functions instead of
fixing them. All potential users should have used blk_first_device_err
instead anyway.
CI results at [2].
[0] https://patchwork.ozlabs.org/project/uboot/patch/[email protected]/
[1] https://gist.github.com/gmalysa/b05e73a5c14bc18c5741a0e0e06a2992
[2] https://gitlab.com/gmalysa/lnxdsp-u-boot/-/pipelines/1931210857
Signed-off-by: Greg Malysa <[email protected]>
Reviewed-by: Andrew Goodbody <[email protected]>
|
|
Create a basic test suit for AES DM uclass that covers all available
operations.
Signed-off-by: Svyatoslav Ryhel <[email protected]>
|
|
https://source.denx.de/u-boot/custodians/u-boot-stm into next
CI: https://source.denx.de/u-boot/custodians/u-boot-stm/-/pipelines/26607
- Add clock and reset drivers support for STM32MP25
- Add STM32H747-Discovery board support
- Add tamp_nvram driver
- Add SPL support and clock tree init to STM32MP13 RCC driver
- Add STM32MP13xx ram support
- Add support for STM32 Image V2.0 for STM32MP13xx
- Fix SYSRAM size on STM32MP13xx
- Fix DBGMCU macro on STM32MP13xx
- Auto-detect ROM API table on STM32MP15xx
|
|
Update the existing test dm_test_clk_ccf() with new CLK_ID macro.
Signed-off-by: Patrick Delaunay <[email protected]>
Signed-off-by: Patrice Chotard <[email protected]>
Cc: Lukasz Majewski <[email protected]>
Cc: Sean Anderson <[email protected]>
Reviewed-by: Patrick Delaunay <[email protected]>
Reviewed-by: Patrice Chotard <[email protected]>
|
|
Some of the Python tests are a pain because they don't reset the TPM
state before each test. Driver model tests do this, so convert the
tests to C.
This means that these tests won't run on real hardware, but we have
tests which do TPM init, so there is still enough coverage.
Rename and update the Python tpm_init test to use 'tpm autostart',
since this fully initializes the TPM and performs the self tests.
Signed-off-by: Simon Glass <[email protected]>
Reviewed-by: Ilias Apalodimas <[email protected]>
Signed-off-by: Ilias Apalodimas <[email protected]>
|
|
Now that env_get_ip() has been removed, the include file <net.h> does
not need anything from <env.h>. Furthermore, include/env.h itself
includes other headers which can lead to longer indirect inclusion
paths. To prepare to remove <env.h> from <net.h> fix all of the
remaining places which had relied on this indirect inclusion to instead
include <env.h> directly.
Reviewed-by: Jerome Forissier <[email protected]> # net/lwip
Acked-by: Ilias Apalodimas <[email protected]>
Reviewed-by: Wolfgang Wallner <[email protected]>
Reviewed-by: Martyn Welch <[email protected]>
Signed-off-by: Tom Rini <[email protected]>
|
|
The include file <command.h> does not need anything from <env.h>.
Furthermore, include/env.h itself includes other headers which can lead
to longer indirect inclusion paths. To prepare to remove <env.h> from
<command.h> fix all of the places which had relied on this indirect
inclusion to instead include <env.h> directly.
Signed-off-by: Tom Rini <[email protected]>
---
Cc: Anatolij Gustschin <[email protected]>
Cc: Bin Meng <[email protected]>
Cc: Heinrich Schuchardt <[email protected]>
Cc: Ilias Apalodimas <[email protected]>
Cc: Jaehoon Chung <[email protected]>
Cc: Jerome Forissier <[email protected]>
Cc: Linus Walleij <[email protected]>
Cc: Lukasz Majewski <[email protected]>
Cc: Marek Vasut <[email protected]>
Cc: Mark Kettenis <[email protected]>
Cc: Masahisa Kojima <[email protected]>
Cc: Mattijs Korpershoek <[email protected]>
Cc: Patrice Chotard <[email protected]>
Cc: Patrick Delaunay <[email protected]>
Cc: Rayagonda Kokatanur <[email protected]>
Cc: Simon Glass <[email protected]>
Cc: Simon Goldschmidt <[email protected]>
Cc: Stefan Bosch <[email protected]>
Cc: Tien Fong Chee <[email protected]>
Cc: Tingting Meng <[email protected]>
Cc: Tobias Waldekranz <[email protected]>
|
|
Provide a way to draw an unfilled box of a certain width. This is useful
for grouping menu items together.
Add a comment showing how to see the copy-framebuffer, for testing.
Signed-off-by: Simon Glass <[email protected]>
|
|
When using expo we want to be able to control the information on the
display and avoid other messages (such as USB scanning) appearing.
Add a 'quiet' flag for the console, to help with this.
The test is a little messy since stdio is still using the original
vidconsole create on start-up. So take care to use the same.
Signed-off-by: Simon Glass <[email protected]>
|
|
We want to check the display contents in expo tests, so move the two
needed functions to a new header file.
Rename them to have a video_ prefix.
Signed-off-by: Simon Glass <[email protected]>
|
|
When writing multiple lines of text we need to be able to control which
text goes on each line. Add a new vidconsole_put_stringn() function to
help with this.
Signed-off-by: Simon Glass <[email protected]>
|
|
Expo needs to be able to word-wrap lines so that they are displayed as
the user expects. Add a limit on the width of each line and support this
in the measurement algorithm.
Add a log category to truetype while we are here.
Signed-off-by: Simon Glass <[email protected]>
|
|
It is useful to be able to embed newline characters in the string and
have the text measured into multiple lines. Add support for this.
Signed-off-by: Simon Glass <[email protected]>
|
|
Create a measured line for the (single) line of text.
Signed-off-by: Simon Glass <[email protected]>
|
|
Update the vidconsole API so that measure() can measure multiple lines
of text. This will make it easier to implement multi-line fields in
expo.
Tidy up the function comments while we are here.
Signed-off-by: Simon Glass <[email protected]>
|
|
Add a simple test which measures a line of text using a Truetype font.
Signed-off-by: Simon Glass <[email protected]>
|
|
CONFIG_VIDEO_COPY implemented a range-based copying mechanism: If we
print a single character, it will always copy the full range of bytes
from the top left corner of the character to the lower right onto the
uncached frame buffer. This includes pretty much the full line contents
of the printed character.
Since we now have proper damage tracking, let's make use of that to reduce
the amount of data we need to copy. With this patch applied, we will only
copy the tiny rectangle surrounding characters when we print them,
speeding up the video console.
After this, changes to the main frame buffer are not immediately copied
to the copy frame buffer, but postponed until the next video device
sync. So issue an explicit sync before inspecting the copy frame buffer
contents for the video tests.
Signed-off-by: Alexander Graf <[email protected]>
[Alper: Rebase for fontdata->height/w, fill_part(), fix memmove(dev),
drop from defconfig, use damage.xstart/yend, use IS_ENABLED(),
call video_sync() before copy_fb check, update video_copy test]
Co-developed-by: Alper Nebi Yasak <[email protected]>
Signed-off-by: Alper Nebi Yasak <[email protected]>
Link: https://lore.kernel.org/u-boot/[email protected]/
|
|
With VIDEO_DAMAGE, the video uclass tracks updated regions of the frame
buffer in order to avoid unnecessary work during a video sync. Enable
the config in sandbox and add a test for it, by printing strings at a
few locations and checking the tracked region.
Signed-off-by: Alper Nebi Yasak <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
Adjust test avoid temporary failures in this patch:
Signed-off-by: Simon Glass <[email protected]>
Link: https://lore.kernel.org/u-boot/[email protected]/
|
|
With VIDEO_COPY enabled, only the modified parts of the frame buffer are
intended to be copied to the hardware. Add a test that checks this, by
overwriting contents we prepared without telling the video uclass and
then checking if the overwritten contents have been redrawn on the next
sync.
Signed-off-by: Alper Nebi Yasak <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
Link: https://lore.kernel.org/u-boot/[email protected]/
|
|
The video tests have a helper function to generate a pseudo-digest of
frame buffer contents, but it only does so for the main one. There is
another check that the copy frame buffer is the same as that. But
neither is enough to test if only the modified regions are copied to the
copy frame buffer, since we will want the two to be different in very
specific ways.
Add a boolean argument to the existing helper function to indicate which
frame buffer we want to inspect, and update the existing callers.
Signed-off-by: Alper Nebi Yasak <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
Link: https://lore.kernel.org/u-boot/[email protected]/
|
|
While checking frame buffer contents, the video tests also check if the
copy frame buffer contents match the main frame buffer. To test if only
the modified regions are updated after a sync, we will need to create
situations where the two are mismatched. Split this check into another
function that we can skip calling, since we won't want it to error on
those mismatched cases.
Signed-off-by: Alper Nebi Yasak <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
Link: https://lore.kernel.org/u-boot/[email protected]/
|
|
It is very surprising that such an uclass, specifically designed to
handle resources that may be shared by different devices, is not keeping
the count of the number of times a power domain has been
enabled/disabled to avoid shutting it down unexpectedly or disabling it
several times.
Doing this causes troubles on eg. i.MX8MP because disabling power
domains can be done in recursive loops were the same power domain
disabled up to 4 times in a row. PGCs seem to have tight FSM internal
timings to respect and it is easy to produce a race condition that puts
the power domains in an unstable state, leading to ADB400 errors and
later crashes in Linux.
Some drivers implement their own mechanism for that, but it is probably
best to add this feature in the uclass and share the common code across
drivers. In order to avoid breaking existing drivers, refcounting is
only enabled if the number of subdomains a device node supports is
explicitly set in the probe function. ->xlate() callbacks will return
the power domain ID which is then being used as the array index to reach
the correct refcounter.
As we do not want to break existing users while stile getting
interesting error codes, the implementation is split between:
- a low-level helper reporting error codes if the requested transition
could not be operated,
- a higher-level helper ignoring the "non error" codes, like EALREADY and
EBUSY.
CI tests using power domains are slightly updated to make sure the count
of on/off calls is even and the results match what we *now* expect. They
are also extended to test the low-level functions.
Signed-off-by: Miquel Raynal <[email protected]>
|
|
Convert the tests to use the do_ping() interface which is now
common to NET and NET_LWIP. This allows running most network test with
SANDBOX and NET_LWIP. A few things to note though:
1. The ARP and IPv6 tests are enabled for NET only
2. The net_retry test is modified to use eth0 (eth@10002000) as the
active (but disabled) interface, and therefore we expect eth1
(eth@10003000) to be the fallback when "netretry" is "yes". This is in
replacement of eth7 (lan1) and eth0 (eth@10002000) respectively.
Indeed, it seems eth7 works with NET by chance and it certainly does not
work with NET_LWIP. I observed that even with NET,
sandbox_eth_disable_response(1, true) has no effect: remove it and
the test still passes. The interface ID is not correct to begin with; 1
corresponds to eth1 (eth@10003000) as shown by debug traces, it is not
eth7 (lan1). And using index 7 causes a SEGV. In fact, it is not the
call to sandbox_eth_disable_response() that prevents the stack from
processing the ICMP reply but the timeout caused by the call to
sandbox_eth_skip_timeout(). Here is what happens when trying to ping
using the eth7 (lan1) interface with NET:
do_ping(...)
net_loop(PING)
ping_start()
eth_rx()
sb_eth_recv()
time_test_add_offset(11000UL);
if (get_timer(0) - time_start > time_delta)
ping_timeout_handler() // ping error, as expected
And the same with NET_LWIP:
do_ping(...)
ping_loop(...)
sys_check_timeouts()
net_lwip_rx(...)
sb_eth_recv()
time_test_add_offset(11000UL);
netif->input(...) // the packet is processed succesfully
By choosing eth0 and sandbox_eth_disable_response(0, true), the incoming
packet is indeed discarded and things work as expected with both network
stacks.
Signed-off-by: Jerome Forissier <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
|
|
Unfortunately this change breaks boot on K3 platform.
U-Boot will hang after:
U-Boot SPL 2025.04-01050-ga40fc5afaec0 (Apr 14 2025 - 07:31:32 +0000)
SYSFW ABI: 3.1 (firmware rev 0x0009 '9.2.7--v09.02.07 (Kool Koala)')
This reverts commit 197376fbf300e92afa0a1583815d9c9eb52d613a as
suggested in [1].
[1] https://lists.denx.de/pipermail/u-boot/2025-April/587032.html
Signed-off-by: Wadim Egorov <[email protected]>
Acked-by: Miquel Raynal <[email protected]>
|
|
It is very surprising that such an uclass, specifically designed to
handle resources that may be shared by different devices, is not keeping
the count of the number of times a power domain has been
enabled/disabled to avoid shutting it down unexpectedly or disabling it
several times.
Doing this causes troubles on eg. i.MX8MP because disabling power
domains can be done in recursive loops were the same power domain
disabled up to 4 times in a row. PGCs seem to have tight FSM internal
timings to respect and it is easy to produce a race condition that puts
the power domains in an unstable state, leading to ADB400 errors and
later crashes in Linux.
CI tests using power domains are slightly updated to make sure the count
of on/off calls is even and the results match what we *now* expect.
As we do not want to break existing users while stile getting
interesting error codes, the implementation is split between:
- a low-level helper reporting error codes if the requested transition
could not be operated,
- a higher-level helper ignoring the "non error" codes, like EALREADY and
EBUSY.
Signed-off-by: Miquel Raynal <[email protected]>
|
|
This is a new DM core helper. There is now a graph endpoint
representation in the sandbox test DTS, so we can just use it to verify
the helper proper behavior.
Suggested-by: Simon Glass <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
Signed-off-by: Miquel Raynal <[email protected]>
|
