Refactor common QoS code into a new common header file, and the soc
specific setup_qos functions into a common API.
Rename $(soc)_qos_count and $(soc)_qos_data variables to qos_count and
qos_data. When QoS settings of more SoCs are added, only one pair will
be defined at a time, based on the config SOC_K3_$(soc).
This refactoring has been done for 2 major purposes.
- The auto-generated $(soc)_qos_data.c and $(soc)_qos.h files cannot
have any code that is specific to any bootloader. Those files have to
remain agnostic of different bootloader implementations and their
header files.
- The existing implementation was less than ideal and would have enabled
multiple $(soc)_qos_count and $(soc)_qos_data variables for all SoC
variants.
Signed-off-by: Aradhya Bhatia <a-bhatia1@ti.com>
The same U-Boot binary is compatible with multiple Verdin AM62 board
variants. However, some of the SoC models can only operate at a maximum
speed of 1 GHz.
Previously, the boards with lower-speed grades were running at
overclocked speeds, leading to kernel complaints about unsupported
configurations.
To resolve this issue, the operating speed has been decreased to the
maximum allowable value across all Verdin AM62 board variants. As a
result, there is a regression in overall boot time, increasing by around
200 milliseconds for the faster SoC variant.
Signed-off-by: Vitor Soares <vitor.soares@toradex.com>
Reviewed-by: Francesco Dolcini <francesco.dolcini@toradex.com>
Francesco Dolcini <francesco.dolcini@toradex.com> says:
This series fixes two regressions affecting multiple Toradex boards
(i.MX, TI and TEGRA based) and targets the current master, e.g. v2024.04
release.
U-Boot `reset` command is broken on all Tordex i.MX6* based SoMs since
v2023.07, this series fixes it enabling the `wdt-reboot` driver.
Since v2024.04-rc1 reading the Toradex configuration block is not working
properly anymore, the serial number and the hardware version are not read
correctly, preventing the board from functioning correctly (wrong mac address,
wrong DT, ...). This is fixed by reading the config block in `EVT_SETTINGS_R`
and adding a toradex sysinfo driver. In addition to that, we now use a random
mac address in case the config block is invalid.
Reported-by: Sahaj Sarup <sahaj.sarup@linaro.org>
Closes: https://lore.kernel.org/all/CAKZ1LvM45MB8N0CqgU+C7i9=Bjb6kqNTxWo2Tv407HKLWtCMkA@mail.gmail.com/
Reported-by: Marcel Ziswiler <marcel.ziswiler@toradex.com>
Closes: https://lore.kernel.org/all/e40ed93bd8f371ec56b8fc451dcb458f3ce6dcba.camel@toradex.com/
Reported-by: Francesco Dolcini <francesco@dolcini.it>
Closes: https://lore.kernel.org/all/ZY2ZDWAQuTlRjV9H@francesco-nb/
commit 68dcbdd594 ("ARM: imx: Add weak default reset_cpu()")
introduced a regression that 'reset' command unable to reset
imx6 and imx6ull based Toradex's modules in the u-boot.
This change enables a watchdog-based sysreset driver for
Toradex's colibri-imx6, colibri-imx6ull (with nand and emmc)
and apalis-imx6 which solves the reset command regression on
these modules.
Acked-by: Marcel Ziswiler <marcel.ziswiler@toradex.com>
Fixes: 68dcbdd594 ("ARM: imx: Add weak default reset_cpu()")
Signed-off-by: Parth Pancholi <parth.pancholi@toradex.com>
Signed-off-by: Francesco Dolcini <francesco.dolcini@toradex.com>
Reviewed-by: Fabio Estevam <festevam@gmail.com>
Set CONFIG_NET_RANDOM_ETHADDR=y, which sets random eth address in
case there is no toradex config block programmed.
In case of missing toradex config block the generated mac address was
always the same, linked to serial number 0.
Use random generated mac address instead.
Acked-by: Marcel Ziswiler <marcel.ziswiler@toradex.com>
Tested-by: Marcel Ziswiler <marcel.ziswiler@toradex.com> # Verdin iMX8M Plus
Signed-off-by: Emanuele Ghidoli <emanuele.ghidoli@toradex.com>
Signed-off-by: Francesco Dolcini <francesco.dolcini@toradex.com>
Reviewed-by: Fabio Estevam <festevam@gmail.com>
This commit introduces support for the Toradex sysinfo driver in U-Boot,
which uses information from Toradex config block to print correct
board model.
In case the Toradex config block is not present sysinfo prints the model
of the board provided by device tree removing per board specific prints.
Acked-by: Marcel Ziswiler <marcel.ziswiler@toradex.com>
Tested-by: Marcel Ziswiler <marcel.ziswiler@toradex.com> # Verdin iMX8M Plus
Signed-off-by: Emanuele Ghidoli <emanuele.ghidoli@toradex.com>
Signed-off-by: Francesco Dolcini <francesco.dolcini@toradex.com>
Reviewed-by: Fabio Estevam <festevam@gmail.com>
Sysinfo prints the model obtained from device tree, checkboard()
take info from hardware and tdx_checkboard() use the model retrieved by
toradex config block.
Remove the print from checkboard() function because the model obtained
from toradex config block is the most complete.
If toradex config block is missing the model info from device tree is
enough.
Acked-by: Marcel Ziswiler <marcel.ziswiler@toradex.com>
Tested-by: Marcel Ziswiler <marcel.ziswiler@toradex.com> # Verdin iMX8M Plus
Signed-off-by: Emanuele Ghidoli <emanuele.ghidoli@toradex.com>
Signed-off-by: Francesco Dolcini <francesco.dolcini@toradex.com>
Reviewed-by: Fabio Estevam <festevam@gmail.com>
Since checkboard() is used instead of show_board_info(), in case toradex
config block is missing or malformed, checkboard is recursively called.
It prints a long list of "MISSING TORADEX CONFIG BLOCK" till the stack
is full.
Fixes: edb0ecd187 ("toradex: Use checkboard() instead of show_board_info()")
Acked-by: Marcel Ziswiler <marcel.ziswiler@toradex.com>
Tested-by: Marcel Ziswiler <marcel.ziswiler@toradex.com> # Verdin iMX8M Plus
Signed-off-by: Emanuele Ghidoli <emanuele.ghidoli@toradex.com>
Signed-off-by: Francesco Dolcini <francesco.dolcini@toradex.com>
Reviewed-by: Fabio Estevam <festevam@gmail.com>
Instead of downloading coreboot binaries from a Google drive location,
use the ones we have built ourselves.
Signed-off-by: Tom Rini <trini@konsulko.com>
To make CI runs rely less on external servers, build a coreboot release
from source and populate /opt/coreboot with the output.
Signed-off-by: Tom Rini <trini@konsulko.com>
- Enable the thermal driver for the imx8m phycore boards.
- Convert imx53-qsb to watchdog driver to fix the 'reset' command.
- Remove multiline string from imx6dl-sielaff.
- Add SPI boot support for imxrt1050-evk.
- Convert opos6uldev to watchdog driver to fix the 'reset' command.
Commit 68dcbdd594 ("ARM: imx: Add weak default reset_cpu()") caused
the 'reset' command in U-Boot to not cause a board reset.
Fix it by switching to the watchdog driver model via sysreset, which
is the preferred method for implementing the watchdog reset.
Signed-off-by: Sébastien Szymanski <sebastien.szymanski@armadeus.com>
Reviewed-by: Fabio Estevam <festevam@gmail.com>
Commit 68dcbdd594 ("ARM: imx: Add weak default reset_cpu()") caused
the 'reset' command in U-Boot to not cause a board reset.
Fix it by switching to the watchdog driver model via sysreset, which
is the preferred method for implementing the watchdog reset.
Signed-off-by: Fabio Estevam <festevam@gmail.com>
Remove the malformed multiline string and fix the checkpatch warning.
Reported-by: Dan Carpenter <dan.carpenter@linaro.org>
Signed-off-by: Frieder Schrempf <frieder.schrempf@kontron.de>
Add support for booting the imxrt1050-evk from spi.
Add imximage config and the ability for SPL to boot from NOR.
Enable binman in Kconfig and device tree for imxrt* as it is used to
prepend fspi_header.bin to SPL and u-boot.img.
Signed-off-by: Jesse Taube <Mr.Bossman075@gmail.com>
When booting Android with AVB enabled, an OOM is observed:
=> avb init ${mmcdev}
=> avb verify _a
## Android Verified Boot 2.0 version 1.1.0
read_is_device_unlocked not supported yet
read_rollback_index not supported yet
avb_util.c:182: ERROR: Failed to allocate memory.
OOM error occurred during verification
A custom malloc length of 128MB is required as documented in
commit 285a83b12b ("configs: meson64_android: increase SYS_MALLOC_LEN to 128M for AVB")
However, this 128M custom malloc length was not ported to Kconfig in
commit 7cfbba36e9 ("Convert CONFIG_SYS_MALLOC_LEN to Kconfig")
Add it back to fix AVB verification on VIM3/VIM3L.
Fixes: 7cfbba36e9 ("Convert CONFIG_SYS_MALLOC_LEN to Kconfig")
Co-developed-by: Guillaume La Roque <glaroque@baylibre.com>
Signed-off-by: Guillaume La Roque <glaroque@baylibre.com>
Signed-off-by: Mattijs Korpershoek <mkorpershoek@baylibre.com>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Reviewed-by: Igor Opaniuk <igor.opaniuk@gmail.com>
Link: https://lore.kernel.org/r/20240209-vim3-avb-malloc-v1-1-91427d8c19ab@baylibre.com
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Nishanth Menon <nm@ti.com> says:
This is a wide cleanup to switch to setting fdtfile using env_set
instead of scripted magic. 'fdtfile' is expected to be set by default.
This allows the stdboot triggered efi loaders to find the correct OS
device tree file even if regular boot process is interrupted by user
intervention.
We shouldn't need finfdt anymore. Drop the env script.
Reviewed-by: Jonathan Humphreys <j-humphreys@ti.com>
Reviewed-by: Roger Quadros <rogerq@kernel.org>
Signed-off-by: Nishanth Menon <nm@ti.com>
Stop using the findfdt script and switch to setting the fdtfile from C
code.
Reviewed-by: Jonathan Humphreys <j-humphreys@ti.com>
Reviewed-by: Roger Quadros <rogerq@kernel.org>
Signed-off-by: Nishanth Menon <nm@ti.com>
Stop using the findfdt script and switch to setting the fdtfile from C
code.
Reviewed-by: Jonathan Humphreys <j-humphreys@ti.com>
Reviewed-by: Roger Quadros <rogerq@kernel.org>
Signed-off-by: Nishanth Menon <nm@ti.com>
We now can provide a map and have the standard fdtfile variable set from
code itself. This allows for bootstd to "just work".
While at this, replace findfdt in environment with a warning as it is no
longer needed.
Reviewed-by: Jonathan Humphreys <j-humphreys@ti.com>
Reviewed-by: Roger Quadros <rogerq@kernel.org>
Signed-off-by: Nishanth Menon <nm@ti.com>
We now can provide a map and have the standard fdtfile variable set from
code itself. This allows for bootstd to "just work".
While at this, replace findfdt in environment with a warning as it is no
longer needed.
Reviewed-by: Jonathan Humphreys <j-humphreys@ti.com>
Reviewed-by: Roger Quadros <rogerq@kernel.org>
Signed-off-by: Nishanth Menon <nm@ti.com>
We now can provide a map and have the standard fdtfile variable set from
code itself. This allows for bootstd to "just work".
While at this, replace findfdt in environment with a warning as it is no
longer needed.
Reviewed-by: Jonathan Humphreys <j-humphreys@ti.com>
Reviewed-by: Roger Quadros <rogerq@kernel.org>
Signed-off-by: Nishanth Menon <nm@ti.com>
We now can provide a map and have the standard fdtfile variable set from
code itself. This allows for bootstd to "just work".
While at this, replace findfdt in environment with a warning as it is no
longer needed.
Reviewed-by: Jonathan Humphreys <j-humphreys@ti.com>
Reviewed-by: Roger Quadros <rogerq@kernel.org>
Signed-off-by: Nishanth Menon <nm@ti.com>
Stop using the findfdt script and switch to setting the fdtfile from
C code.
While at this, replace findfdt in environment with a warning as it is
no longer needed
Reviewed-by: Jonathan Humphreys <j-humphreys@ti.com>
Reviewed-by: Roger Quadros <rogerq@kernel.org>
Signed-off-by: Nishanth Menon <nm@ti.com>
Stop using the findfdt script and switch to setting the fdtfile from
C code.
While at this, replace findfdt in environment with a warning as it is
no longer needed
Reviewed-by: Jonathan Humphreys <j-humphreys@ti.com>
Reviewed-by: Roger Quadros <rogerq@kernel.org>
Signed-off-by: Nishanth Menon <nm@ti.com>
Introduce a common fdt operations library for basic device tree
operations that are common between various boards.
The first library to introduce here is the capability to set up
fdtfile as a standard variable as part of board identification rather
than depend on scripted ifdeffery.
Reviewed-by: Jonathan Humphreys <j-humphreys@ti.com>
Reviewed-by: Roger Quadros <rogerq@kernel.org>
Signed-off-by: Nishanth Menon <nm@ti.com>
Add missing board/ti/common/Kconfig references for the platforms that
missed it. The intent is for the common Kconfig to be usable across TI
reference boards as required.
Reported-by: Tom Rini <trini@konsulko.com>
Signed-off-by: Nishanth Menon <nm@ti.com>
i.MX93 uses the same USB IP as i.MX8MM. It can then reuse the ehci-mx6
driver.
Reviewed-by: Marek Vasut <marex@denx.de>
Signed-off-by: Mathieu Othacehe <othacehe@gnu.org>
Reviewed-by: Fabio Estevam <festevam@gmail.com>
Additional testing of the changes introduced in commit 33e06dcbe5 "common:
usb-hub: Reset hub port before scanning") revealed that some USB 2.0 and 3.0
flash drives didn't work in U-Boot on some Allwinner SoCs that support USB
2.0 interfaces only. More precisely, some of the tested USB 2.0 and 3.0
flash drives failed to be detected and work on an OrangePi Zero 3, based on
the Allwinner H616 SoC that supports USB 2.0 only, while the same USB flash
drives worked just fine on a Pine64 H64, based on the Allwinner H6 SoC that
supports both USB 2.0 and USB 3.0 interfaces.
The USB ID of the above-mentioned USB 3.0 flash drive that failed to work is
1f75:0917 (Innostor Technology Corporation IS917 Mass storage), it is 32 GB
in size and sold under the PNY brand. The mentioned USB 2.0 drive is some
inexpensive no-name drive with an invalid USB ID.
Resetting USB 3.0 hubs only, which this patch introduces to the USB hub
resets, has been tested to work as expected, resolving the identified issues
on the Allwinner H616, while not introducing any new issues on other tested
Allwinner SoCs. Thus, let's fix it that way.
According to the USB 3.0 specification, resetting a USB 3.0 port is required
when an attached USB device transitions between different states, such as
when it resumes from suspend. Though, the Linux kernel performs additional
USB 3.0 port resets upon initial USB device attachment, as visible in commit
07194ab7be63 ("USB: Reset USB 3.0 devices on (re)discovery") in the kernel
source, to ensure proper state of the USB 3.0 hub port and proper USB mode
negotiation during the initial USB device attachment and enumeration.
These additional types of USB port resets don't exist for USB 2.0 hubs,
according the USB 2.0 specification. The resets seem to be added to the USB
3.0 specification as part of the port and device mode negotiation.
The Linux kernel resets USB 3.0 (i.e. SuperSpeed) hubs only, as visible in
commit 10d674a82e55 ("USB: When hot reset for USB3 fails, try warm reset.")
in the kernel source. The check for SuperSpeed hubs is performed in a way
that also applies to newer SuperSpeed Plus (USB 3.1 or 3.2) hubs as well,
which hopefully makes it future proof.
Fixes: 33e06dcbe5 ("common: usb-hub: Reset hub port before scanning")
Link:
https://lore.kernel.org/u-boot/20240207102327.35125-1-i@shantur.com/T/#u
Link:
https://lore.kernel.org/u-boot/20240201164604.13315fa6@donnerap.manchester.arm.com/T/#u
Signed-off-by: Shantur Rathore <i@shantur.com>
Helped-by: Dragan Simic <dsimic@manjaro.org>
Tested-by: Andre Przywara <andre.przywara@arm.com>
Reviewed-by: Dragan Simic <dsimic@manjaro.org>
Reviewed-by: Marek Vasut <marex@denx.de>
CI: https://source.denx.de/u-boot/custodians/u-boot-imx/-/pipelines/19683
- Convert msc_sm2s_imx8mp to DM_SERIAL.
- Make Ethernel functional on msc_sm2s_imx8mp.
- General improvements for msc_sm2s_imx8mp.
- Add suport for the Sielaff i.MX6 Solo board.
- Update GE HealthCare maitainers' e-mail addresses.
