| Age | Commit message (Collapse) | Author |
|---|
|
Disable UBI fastmap support which is not supported by original Maemo 5
kernel and explicitly set UBI BEB limit to 10%, which is the value used by
original Maemo 5 kernel. U-Boot default value is 20%.
Signed-off-by: Pali Rohár <[email protected]>
|
|
When reading of image fails then do not call bootm. This prevents false
positive test result in case something bootable is present in memory.
Signed-off-by: Pali Rohár <[email protected]>
|
|
Original Nokia UBIFS system image has 1870 LEBs, so set UBIFS volume size
in test script to the same value. Number of 1870 LEBs corresponds to 230MiB
(LEB size * num of LEBs = 126KiB * 1870 = 230MiB).
Signed-off-by: Pali Rohár <[email protected]>
|
|
At address 0x80000100 are stored initial atags passed to U-Boot by NOLO.
So do not overwrite them when using $loadaddr variable which value is set
from CONFIG_SYS_LOAD_ADDR option.
Signed-off-by: Pali Rohár <[email protected]>
|
|
If kernel image in uImage or zImage format is not detected by
lowlevel_init.S code then do not clear memory location where image was
expected. If image is not detected then this memory region is unknown.
So do not unconditionally clear it.
Signed-off-by: Pali Rohár <[email protected]>
|
|
Label copy_kernel_start is now unused. Remove it.
Signed-off-by: Pali Rohár <[email protected]>
|
|
Add j721s2 High Security EVM defconfig.
These configs are same as for the non-secure part, except for:
CONFIG_TI_SECURE_DEVICE option set to 'y'
CONFIG_FIT_IMAGE_POST_PROCESS option set to 'y'
CONFIG_SPL_FIT_IMAGE_POST_PROCESS option set to 'y'
CONFIG_BOOTCOMMAND uses FIT images for booting
Signed-off-by: Jayesh Choudhary <[email protected]>
|
|
Add J7200 High Security EVM defconfig.
These defconfigs are the same as for the non-secure part, except for:
CONFIG_TI_SECURE_DEVICE option set to 'y'
CONFIG_BOOTCOMMAND uses FIT images for booting
Signed-off-by: Andrew Davis <[email protected]>
[[email protected]: add few configs from GP variant which were missing]
Signed-off-by: Jayesh Choudhary <[email protected]>
|
|
* Drop legacy /boot/PPTImage.md5 check
* Update device tree naming
* Update t30args#0 root cmd line property to support latest kernel
versions (root=/dev/mmcblk0p2 for linux < 4 and
root=/dev/mmcblk1p2 for linux >= 4)
* Add custom bootloader version string
* Destroy invalid dtb at ${dtbaddr} and configuration script at
${cfgaddr} to ensure proper boot in warm restart case.
Signed-off-by: Bernhard Messerklinger <[email protected]>
Reviewed-by: Wolfgang Wallner <[email protected]>
|
|
Commit 0ea4fc4dcf90 ("board/BuR: invalidate ${dtbaddr} before cfgscr")
destroys the boot targets b_t30lgcy#0 and b_t30lgcy#1. The reason behind
this is, that b_t30lgcy#0 and b_t30lgcy#1 both load the for booting
needed device trees from mmc and the cfgscr script patches those. Because
of this, cfgscr is not allowed to destroy the previously loaded device
tree otherwise cfgscr will fail.
This patch moves the device trees invalidation on warm restart to the
PREBOOT cmd to fix that issue.
Fixes: 0ea4fc4dcf90 ("board/BuR: invalidate ${dtbaddr} before cfgscr")
Signed-off-by: Bernhard Messerklinger <[email protected]>
Reviewed-by: Wolfgang Wallner <[email protected]>
|
|
* Remove unnecessary device tree nodes which are not needed in
U-Boot directly.
* Move all U-Boot specific device tree properties to u-boot dtsi.
Signed-off-by: Bernhard Messerklinger <[email protected]>
Reviewed-by: Wolfgang Wallner <[email protected]>
|
|
Commit 6337d53fdf45 ("arm: dts: sync am33xx with Linux 5.9-rc7") syncs
the am335x device tree with the latest linux kernel am335x device tree.
That causes problems with device tree in SPL stage.
To fix the issues CONFIG_SPL_OF_TRANSLATE must be set to handle the
synced bus addresses correctly.
A custom U-Boot device tree is also needed since the SPL build removes
bus properties from bus nodes which are not explicitly marked with the
u-boot,dm-spl or u-boot,dm-pre-reloc flag. Therefore all parent buses of
the in the SPL needed devices must be marked with u-boot,dm-pre-reloc.
Also since there is no driver for "ti,sysc" compatible property in SPL
the buses marked with this compatible string must also be marked with
compatible = "simple-bus" to make the underlying devices visible in
SPL. Otherwise the matching device drivers aren't found and the uclass
drivers are dropped.
Signed-off-by: Bernhard Messerklinger <[email protected]>
Reviewed-by: Wolfgang Wallner <[email protected]>
|
|
The SPI and NAND board variants never went into production.
Drop those board variants.
Signed-off-by: Bernhard Messerklinger <[email protected]>
Reviewed-by: Wolfgang Wallner <[email protected]>
|
|
https://source.denx.de/u-boot/custodians/u-boot-stm
- Fixes on STM32 I2C drivers
- Activate SCMI regulator for STM32MP15 defconfig, fix the usb start command
for scmi device tree
|
|
- Add uncompressed kernel image support to falcon mode, TEE
improvements, make xyz-modem timeout configurable, gpio updates and
other assorted improvements.
|
|
Check the that the hash length is shorter than the message length. This
avoids:
./tools/../lib/rsa/rsa-verify.c:275:11: warning:
‘*db’ may be used uninitialized [-Wmaybe-uninitialized]
275 | db[0] &= 0xff >> leftmost_bits;
Fixes: 061daa0b61f0 ("rsa: add support of padding pss")
Signed-off-by: Heinrich Schuchardt <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
|
|
Normally, readline is not used int SPL. However, it may be useful to
enable the Freescale DDR interactive mode in SPL, while U-Boot is still
executing from SRAM. The default settings for readline result in a large
buffer being allocated. Reduce the size of the maximum input line, and
the number of lines of scrollback when building for SPL.
Signed-off-by: Sean Anderson <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
|
|
This allows to test if a pin's label if displayed using gpio_get_status()
when this pin is configured in alternate function.
Signed-off-by: Patrice Chotard <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
Reviewed-by: Patrick Delaunay <[email protected]>
|
|
Populate uc_priv->name[] with pinmux node's name in order to indicate
the pinmuxing's name in case GPIO is configured in alternate.
For example, for STM32 SoC's based platform, "gpio status" command
output :
before
Bank GPIOZ:
GPIOZ0: unused : 0 [ ]
GPIOZ1: unused : 0 [ ]
GPIOZ2: unused : 0 [ ]
GPIOZ3: unused : 0 [ ]
GPIOZ4: func
GPIOZ5: func
GPIOZ6: unused : 0 [ ]
GPIOZ7: unused : 0 [ ]
GPIOZ8: unknown
GPIOZ9: unknown
GPIOZ10: unknown
GPIOZ11: unknown
GPIOZ12: unknown
GPIOZ13: unknown
GPIOZ14: unknown
GPIOZ15: unknown
After
Bank GPIOZ:
GPIOZ0: unused : 0 [ ]
GPIOZ1: unused : 0 [ ]
GPIOZ2: unused : 0 [ ]
GPIOZ3: unused : 0 [ ]
GPIOZ4: func i2c4-0
GPIOZ5: func i2c4-0
GPIOZ6: unused : 0 [ ]
GPIOZ7: unused : 0 [ ]
GPIOZ8: unknown
GPIOZ9: unknown
GPIOZ10: unknown
GPIOZ11: unknown
GPIOZ12: unknown
GPIOZ13: unknown
GPIOZ14: unknown
GPIOZ15: unknown
Signed-off-by: Patrice Chotard <[email protected]>
Reviewed-by: Patrick Delaunay <[email protected]>
|
|
Even pin with GPIOF_UNUSED function can have a label.
The criteria to add or not a space character is linked to label not to
the used/unused status.
Signed-off-by: Patrice Chotard <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
Reviewed-by: Patrick Delaunay <[email protected]>
|
|
Currently, if pin's function is GPIOF_FUNC, only "func" if displayed
without any other information. It would be interesting, if information is
available, to indicate which pinmuxing's name is used.
For example, for STM32 SoC's based platform, "gpio status" command
output :
before
Bank GPIOZ:
GPIOZ0: unused : 0 [ ]
GPIOZ1: unused : 0 [ ]
GPIOZ2: unused : 0 [ ]
GPIOZ3: unused : 0 [ ]
GPIOZ4: func
GPIOZ5: func
GPIOZ6: unused : 0 [ ]
GPIOZ7: unused : 0 [ ]
GPIOZ8: unknown
GPIOZ9: unknown
GPIOZ10: unknown
GPIOZ11: unknown
GPIOZ12: unknown
GPIOZ13: unknown
GPIOZ14: unknown
GPIOZ15: unknown
After
Bank GPIOZ:
GPIOZ0: unused : 0 [ ]
GPIOZ1: unused : 0 [ ]
GPIOZ2: unused : 0 [ ]
GPIOZ3: unused : 0 [ ]
GPIOZ4: func i2c4-0
GPIOZ5: func i2c4-0
GPIOZ6: unused : 0 [ ]
GPIOZ7: unused : 0 [ ]
GPIOZ8: unknown
GPIOZ9: unknown
GPIOZ10: unknown
GPIOZ11: unknown
GPIOZ12: unknown
GPIOZ13: unknown
GPIOZ14: unknown
GPIOZ15: unknown
Signed-off-by: Patrice Chotard <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
Reviewed-by: Patrick Delaunay <[email protected]>
|
|
The bar of the structure virtio_pci_cap is the index, and each base
address occupies 4 bytes, so it needs to be multiplied by 4.
This patch fixes a bug reported by Felix Yan
https://lists.denx.de/pipermail/u-boot/2022-August/492779.html
Signed-off-by: Xiang W <[email protected]>
Tested-by: Felix Yan <[email protected]>
|
|
The nvme driver falsely assumed that the last entry on a page
of the prp-list always points to the next page of the prp-list.
This potentially can lead to the illegal creation of pages on
the prp-list with only a single entry. This change now ensures
that splitting the prp-list into multiple pages, behaves now as
required by the NVME-Spec.
Related to this, also the size of the memory allocation is adjusted
accordingly.
Signed-off-by: Alexander Sowarka <[email protected]>
|
|
Signed-off-by: Pali Rohár <[email protected]>
|
|
Now when loadx and loady commands could be aborted / cancelled by CTRL+C,
allow to configure timeout for initial x/y-modem packet via env variable
$loadxy_timeout and by default use value from new compile-time config
option CONFIG_CMD_LOADXY_TIMEOUT. Value is in seconds and zero value means
infinite timeout. Default value is 90s which is the value used before this
change for loadx command.
Other load commands loadb and loads already waits infinitely. Same behavior
for loadx and loady commands can be achieved by setting $loadxy_timeout or
CONFIG_CMD_LOADXY_TIMEOUT to 0.
Signed-off-by: Pali Rohár <[email protected]>
|
|
This fixes numerous cases of format strings not matching their
arguments. Also keep the format strings on one line for easier grepping.
Signed-off-by: Sean Anderson <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
|
|
These two device tree properties were not being applied.
Fixes: 1fd9eb68d6 ("i2c: stm32f7: move driver data of each instance in a privdata")
Signed-off-by: Jorge Ramirez-Ortiz <[email protected]>
Reviewed-by: Alain Volmat <[email protected]>
Reviewed-by: Patrice Chotard <[email protected]>
Reviewed-by: Heiko Schocher <[email protected]>
Reviewed-by: Patrick Delaunay <[email protected]>
Tested-by: Patrick Delaunay <[email protected]>
|
|
Current function stm32_i2c_message_xfer is sending a STOP
whatever the result of the transaction is. This can cause issues
such as making the bus busy since the controller itself is already
sending automatically a STOP when a NACK is generated.
Thanks to Jorge Ramirez-Ortiz for diagnosing and proposing a first
fix for this. [1]
[1] https://lore.kernel.org/u-boot/[email protected]/
Reported-by: Jorge Ramirez-Ortiz, Foundries <[email protected]>
Signed-off-by: Jorge Ramirez-Ortiz <[email protected]>
Signed-off-by: Alain Volmat <[email protected]>
Reviewed-by: Patrice Chotard <[email protected]>
Reviewed-by: Heiko Schocher <[email protected]>
Reviewed-by: Patrick Delaunay <[email protected]>
Tested-by: Patrick Delaunay <[email protected]>
|
|
Functions stm32_i2c_message_start and stm32_i2c_handle_reload
both get a stop boolean indicating if the transfer should end with
a STOP or not. However no specific handling is needed in those
functions hence remove the parameter.
Signed-off-by: Alain Volmat <[email protected]>
Reviewed-by: Patrick Delaunay <[email protected]>
Reviewed-by: Heiko Schocher <[email protected]>
Reviewed-by: Patrice Chotard <[email protected]>
|
|
Comment within stm32_i2c_message_start is misleading, indicating
that AUTOEND bit is setted while it is actually cleared.
Moreover, the bit is actually never setted so there is no need
to clear it hence get rid of this bit clear and the bit macro
as well.
Signed-off-by: Alain Volmat <[email protected]>
Reviewed-by: Patrick Delaunay <[email protected]>
Reviewed-by: Heiko Schocher <[email protected]>
Reviewed-by: Patrice Chotard <[email protected]>
|
|
Bits should be set to 0, not 1.
Signed-off-by: Jorge Ramirez-Ortiz <[email protected]>
Reviewed-by: Patrice Chotard <[email protected]>
Reviewed-by: Patrick Delaunay <[email protected]>
|
|
Activate the support of SCMI regulator to support the scmi_reg11,
scmi_reg18 and scmi_usb33 regulators present in the scmi device tree of
STMicroelectronics boards with stm32mp15-scmi.dtsi
Fixes: 6cccc8d396bf ("ARM: dts: stm32: add SCMI version of STM32 boards (DK1/DK2/ED1/EV1)")
Signed-off-by: Patrick Delaunay <[email protected]>
Reviewed-by: Patrice Chotard <[email protected]>
|
|
Changes optee_rng driver to register itself has a OP-TEE service so
that a device is bound for the driver when OP-TEE enumerates the
PTA RNG service.
Cc: Sughosh Ganu <[email protected]>
Cc: Patrick Delaunay <[email protected]>
Signed-off-by: Etienne Carriere <[email protected]>
Reviewed-by: Patrick Delaunay <[email protected]>
|
|
This change defines resources for OP-TEE service drivers to register
themselves for being bound to when OP-TEE firmware reports the related
service is supported. OP-TEE services are discovered during optee
driver probe sequence which mandates optee driver is always probe once
bound.
Discovery of optee services and binding to related U-Boot drivers is
embedded upon configuration switch CONFIG_OPTEE_SERVICE_DISCOVERY.
Cc: Jens Wiklander <[email protected]>
Cc: Patrick Delaunay <[email protected]>
Signed-off-by: Etienne Carriere <[email protected]>
Reviewed-by: Patrick Delaunay <[email protected]>
|
|
Removes local variable child in optee_probe() that is not used.
Cc: Patrick Delaunay <[email protected]>
Reviewed-by: Patrick Delaunay <[email protected]>
Signed-off-by: Etienne Carriere <[email protected]>
|
|
Allocate memory for buffers at a cache-line boundary to avoid
misaligned buffer address for subsequent reads. This avoids an
additional sector-based memory copy in the fat file system driver:
FAT: Misaligned buffer address (...)
Signed-off-by: Stefan Herbrechtsmeier <[email protected]>
|
|
The device tree binding [1] specify the vendor-id, product-id, device-id
and language-id as 16 bit values and the linux driver reads the boost-up
value as 8 bit value.
[1] https://www.kernel.org/doc/Documentation/devicetree/bindings/usb/usb251xb.txt
Signed-off-by: Stefan Herbrechtsmeier <[email protected]>
Reviewed-by: Marek Vasut <[email protected]>
|
|
Add functions to read 8/16-bit integers like the existing functions for
32/64-bit to simplify read of 8/16-bit integers from device tree
properties.
Signed-off-by: Stefan Herbrechtsmeier <[email protected]>
Reviewed-by: Marek Vasut <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
|
|
Falcon Mode boot sequence.
This is required for architectures which do not support compressed kernel images (i.e. ARM64). This is only used while not booting via FIT image.
Signed-off-by: Nathan Barrett-Morrison <[email protected]>
Cc: Tom Rini <[email protected]>
Reviewed-by: Tom Rini <[email protected]>
|
|
https://source.denx.de/u-boot/custodians/u-boot-efi into next
Pull request for efi next
UEFI:
Implement a command eficonfig to maintain Load Options and boot order via
menus.
|
|
To quote the author:
The previous attempt at this[1] broke a board and was reverted in [2].
This series adopts a slightly different approach, splitting the changes
into many commits.
[1] f33a2c1bd0f ("image: Remove #ifdefs from select_ramdisk()")
[2] 621158d106f ("Revert "image: Remove #ifdefs from select_ramdisk()"")
|
|
Remove all but a few that are difficult, relying on legacy CONFIG options
or optional global_data fields.
Drop the duplicate function name in the comment for boot_get_cmdline().
Signed-off-by: Simon Glass <[email protected]>
|
|
Finish off the refactoring by correcting the indent levels. Note that this
does not include any functional changes.
Signed-off-by: Simon Glass <[email protected]>
|
|
Drop the last one of these, by using a done_select variable to control
whether to fall back to using 'select' as a hex value.
Note that the indentation is not adjusted, to make this easier to review.
Signed-off-by: Simon Glass <[email protected]>
|
|
Drop the prenultimate one of these from select_ramdisk().
Signed-off-by: Simon Glass <[email protected]>
|
|
Drop the #ifdef from near the end of select_ramdisk(). Move some variables
to the top of the function to make this work.
Signed-off-by: Simon Glass <[email protected]>
|
|
Use if() instead of the #ifdef in select_ramdisk().
Signed-off-by: Simon Glass <[email protected]>
|
|
The current switch default is tricky since it relies on #ifdefs to work.
Use a bool instead.
Also fix the comment on @select, since it has a dual purpose.
Signed-off-by: Simon Glass <[email protected]>
|
|
Convert this to an if(), fix the cast from an address to a pointer and
make sure that any error is returned correctly.
Signed-off-by: Simon Glass <[email protected]>
|
|
To quote the author:
This patchset adds the basic infrastructure to periodically execute
code, e.g. all 100ms. Examples for such functions might be LED blinking
etc. The functions that are hooked into this cyclic list should be
small timewise as otherwise the execution of the other code that relies
on a high frequent polling (e.g. UART rx char ready check) might be
delayed too much. This patch also adds the Kconfig option
CONFIG_CYCLIC_MAX_CPU_TIME_US, which configures the max allowed time
for such a cyclic function. If it's execution time exceeds this time,
this cyclic function will get removed from the cyclic list.
How is this cyclic functionality executed?
This patchset integrates the main function responsible for calling all
registered cyclic functions cyclic_run() into the common WATCHDOG_RESET
macro. This guarantees that cyclic_run() is executed very often, which
is necessary for the cyclic functions to get scheduled and executed at
their configured periods.
This cyclic infrastructure will be used by a board specific function on
the NIC23 MIPS Octeon board, which needs to check periodically, if a
PCIe FLR has occurred.
Ideas how to continue:
One idea is to rename WATCHDOG_RESET to something like SCHEDULE and
move the watchdog_reset call into this cyclic infrastructure as well.
Or to perhaps move the shell UART RX ready polling to a cyclic
function.
It's also possible to extend the "cyclic" command, to support the
creation of periodically executed shell commands (for testing etc).
|
