The SCSI layer now issues a SYNC_CACHE command after every write to
ensure there is no data loss due to a board reset after write.
Implement support for this command and remove the same logic from the
ATA write path to be consistent with other SCSI backends.
Ranges are not supported and the whole cache will be flushed in all
cases.
This was done per iteration in ata_scsiop_read_write(), but it's not
clear why this was the case, calling it once for the entire write ought
to achieve the same result.
Signed-off-by: Caleb Connolly <caleb.connolly@linaro.org>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
We don't have a mechanism to safely shutdown block devices prior to a
baord reset or driver removal. Prevent data loss by synchronizing the
SCSI cache after every write.
In particular this solves the issue of capsule updates looping on some
devices because the board resets immediately after deleting the capsule
file and this write wouldn't be flushed in time.
This may impact NAND wear, but should be negligible given the usecases
for disk write in U-Boot.
Signed-off-by: Caleb Connolly <caleb.connolly@linaro.org>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
This clears the 6th byte of cmd twice rather than setting the 9th byte
to 0. Fix it.
The only other command that sets the 9th byte is the 64-bit read, so
this likely never caused issues in practise.
Signed-off-by: Caleb Connolly <caleb.connolly@linaro.org>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Since introduction of OF_UPSTREAM flag, U-Boot's dtc must be able
to compile Kernel's device tree.
Since kernel commit 7de129f5389b ("ARM: dts: stm32: stm32mp151a-prtt1l:
Fix QSPI configuration"), label relative path references has been
introduced. These label relative path references is not supported
by current U-Boot dtc version 1.5.0: (see mailing list discussion [1]).
In order to support such label relative patch references
adds following commit from upstream DTC tree:
commit 651410e54cb9 ("util: introduce xstrndup helper")
commit ec7986e682cf ("dtc: introduce label relative path references")
[1] https://lore.kernel.org/all/20250115144428.GZ3476@bill-the-cat/T/
Signed-off-by: Patrice Chotard <patrice.chotard@foss.st.com>
Cc: Tom Rini <trini@konsulko.com>
Cc: Simon Glass <sjg@chromium.org>
Reviewed-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
The symbol "ENV_IS_EMBEDDED" is an environment internal define and not a
real CONFIG symbol. The IS_ENABLED() macro is still valid to use here,
so update the check.
Signed-off-by: Tom Rini <trini@konsulko.com>
The cflags and ldflags of preload_check_sign depend on the openssl
package thus pkg-config is needed to get the location where openssl
is installed.
This fix a potential build failure when openssl is not from the
distro and installed in a varied place.
Signed-off-by: Raymond Mao <raymond.mao@linaro.org>
When using non-trivial values for parameters for this test it
will cause a spurious failure as the test passes a decimal value
to the mtest command which will interpret it as hexadecimal and
result in failure as below.
test/py/tests/test_memtest.py:66: in test_memtest_ddr
assert expected_response in response
E AssertionError: assert 'Tested 16 iteration(s) with 0 errors.' in 'Refusing to do empty test\r\nmtest - simple RAM read/write test\r\n\r\nUsage:\r\nmtest [start [end [pattern [iterations]]]]'
----------------------------- Captured stdout call -----------------------------
U-Boot> mtest 134217728 0x8001000 90 0x10
Refusing to do empty test
mtest - simple RAM read/write test
Usage:
mtest [start [end [pattern [iterations]]]]
The fix is to ensure that all the parameters to the mtest command are
passed as hexadecimal values.
Fixes: 22efc1cf27 ("test/py: memtest: Add tests for mtest command")
Signed-off-by: Andrew Goodbody <andrew.goodbody@linaro.org>
Reviewed-by: Love Kumar <love.kumar@amd.com>
If a series is sent without a cover letter, there is no indication of
the base commit. Add support for this, since single patches of small
series may not always have a cover letter.
Signed-off-by: Simon Glass <sjg@chromium.org>
It is helpful to know which commit patches are based on, even if that
commit might not be available to readers. Add a tag for this in the
cover letter.
Also add the local-branch name since that may be useful to the writer.
Signed-off-by: Simon Glass <sjg@chromium.org>
We had previously gotten this package through a chain of dependencies
with guestfs-tools. Now that we no longer install that package, install
fdisk (for sfdisk) directly.
Fixes: eb1b90ec57 ("Dockerfile: Update to drop virt-make-fs packages")
Signed-off-by: Tom Rini <trini@konsulko.com>
There are some reference platforms from Arm which are not found in QEMU
but instead in the FVP tool. As we can make use of this in CI later on,
download and extract it in our Dockerfile today.
Signed-off-by: Tom Rini <trini@konsulko.com>
Outside of changing versions here the other visible change is that we
tell grub that riscv64 does not have "large model" support. Without this
change the resulting mkimage is non-functional. This is known upstream
already.
Link: https://savannah.gnu.org/bugs/?65909
Signed-off-by: Tom Rini <trini@konsulko.com>
u-boot-dfu-20250410
CI:
- https://source.denx.de/u-boot/custodians/u-boot-dfu/-/pipelines/25615
Usb gadget:
- Add SAM9X60 support to atmel driver
- Fix memory leaks in f_mass_storage gadget driver
- Fix comment typo in dwc3 gadget driver
Fastboot:
- Lift restrictions on !NET_LWIP for USB
Android:
- Fix possible NULL ptr when AVB is out of memory
Recent addition of support for SDM660 inadvertently broke USB PHY power
on sequence on RB1/RB2 and others with following error:
starting USB...
Bus usb@4e00000: QUSB2PHY pll lock failed: status reg = 0
qcom-qusb2-phy phy@1613000: PHY: Failed to power on phy@1613000: -16.
Can't power on PHY0
probe failed, error -16
No USB controllers found
The root cause was the addition of flag se_clk_scheme_default which was
configured correctly for SDM660 but incorrect for all other supported
SoC. Fix that by properly assignment as per upstream Linux driver.
Fixes: 475497dc3c15 ("phy: Add SDM660 support to Qualcomm QUSB2 phy")
Signed-off-by: Sumit Garg <sumit.garg@oss.qualcomm.com>
Reviewed-by:
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20250410080027.208674-3-sumit.garg@kernel.org
Signed-off-by: Caleb Connolly <caleb.connolly@linaro.org>
Currently the msm_sdhci doesn't yet support DLL configurations which are
required to enable bus speeds greater that 100MHz. So disable HS200 mode
support as of now as it requires bus speeds of 200MHz.
This should fix eMMC issues reported on RB1/RB2 although it should fix
issues for all Qcom platforms but it's not seen there as mostly SD cards
available don't support HS200 mode. The SD cards usually works in high
speed mode whose performance remains unaffected by this change. It only
affects RB1/RB2 as eMMC flash on these support HS200 mode but the U-Boot
driver currently is incapable of supporting that.
Reviewed-by: Caleb Connolly <caleb.connolly@linaro.org>
Signed-off-by: Sumit Garg <sumit.garg@oss.qualcomm.com>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20250410080027.208674-2-sumit.garg@kernel.org
Signed-off-by: Caleb Connolly <caleb.connolly@linaro.org>
On v5 & v7 controllers, multiple mapping for different
Execution Environment exists, if the mapping owner is for
a different Execution Environment we can only read and
not write any data.
To allow us to find a Write mapping for our Execution
Environment, we can overwritte a mapping if we encounter
a new one which we own.
Implement this logic, the result is the same mapping
table as in Linux.
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Tested-by: caleb.connolly@linaro.org # sdm845
Link: https://lore.kernel.org/r/20250328-topic-sm8x50-spmi-fix-v1-4-a7548d3aef0d@linaro.org
Signed-off-by: Caleb Connolly <caleb.connolly@linaro.org>
Enable fastboot support over USB, using MMC as the backend. This will be
the internal eMMC on devices that have it, or the sdcard slot on devices
with UFS (if available).
We don't use a fixed address for the fastboot buffer because it's
allocated at runtime per-board. Entering fastboot mode should be done by
executing "run fastboot" or manually running:
fastboot -l $fastboot_addr_r usb 0
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20250324-sdm845-fixes-fastboot-v1-4-d177a10f336d@linaro.org
Signed-off-by: Caleb Connolly <caleb.connolly@linaro.org>
In fsg_common_init, we allocate some buffers via memalign().
However, these buffers are never freed.
Because of that, we cannot call => ums command multiple times on boards
with low memory (CONFIG_SYS_MALLOC_LEN=0x81000):
=> ums 0 mmc 2
UMS: LUN 0, dev mmc 2, hwpart 0, sector 0x0, count 0x3a3e000
|crq->brequest:0x0
CTRL+C - Operation aborted
=> ums 0 mmc 2
UMS: LUN 0, dev mmc 2, hwpart 0, sector 0x0, count 0x3a3e000
failed to start <NULL>: -12
g_dnl_register: failed!, error: -12
g_dnl_register failed
Make sure the fsg buffers are freed when the gadget is unbound by
calling fsg_common_release() in fsg_unbind().
Reported-by: Zixun LI <admin@hifiphile.com>
Signed-off-by: Mattijs Korpershoek <mkorpershoek@baylibre.com>
Tested-by: Zixun LI <admin@hifiphile.com> # on SAM9X60
Link: https://lore.kernel.org/r/20250328-ums-gadget-leak-v1-4-3b677db99bde@baylibre.com
Signed-off-by: Mattijs Korpershoek <mkorpershoek@kernel.org>
