| Age | Commit message (Collapse) | Author |
|---|
|
Origin DT binding just specify driver but wasn't aligned with DT binding
which came later. Extend description for zcu102 and zcu106 to cover latest
binding.
Signed-off-by: Michal Simek <[email protected]>
Signed-off-by: Michal Simek <[email protected]>
Link: https://lore.kernel.org/r/db66a4bb501183ffbd033da4dd263afdb214f8ec.1652262769.git.michal.simek@amd.com
|
|
BTN_MISC looks like the most reasonable option for this button.
Button is used by firmware to indicate (after reset, power up) that user
wants to do firmware upgrade via firmware update utility.
For bootloader or OS is this just user button which is worth to have it
mapped.
Also button can be used as a wakeup source and pressing it for more time
can generate more chars that's why also adding wakeup-source and autorepeat
properties.
Signed-off-by: Michal Simek <[email protected]>
Reviewed-by: Radhey Shyam Pandey <[email protected]>
Link: https://lore.kernel.org/r/7f6d627473632c3c3036ec9f6aaa36e00f4615e2.1652262769.git.michal.simek@amd.com
|
|
SGMII requires phy to be configured. The support for this has been added to
Linux and U-Boot already that's why also describe the phy via DT. Clock is
coming from si5332 chip (output 1) 125MHz which is only one GT line use on
this board.
Signed-off-by: Michal Simek <[email protected]>
Signed-off-by: Michal Simek <[email protected]>
Link: https://lore.kernel.org/r/8ad8d0c2fc9690cc90f95feddf87b0e94a685a43.1652262769.git.michal.simek@amd.com
|
|
This reverts commit 50a6bd000f94832658f42fb01b9aaf9e39a52004.
As zynqmp mini emmc does not rely on firmware, remove firmware related
device tree modes from zynqmp mini emmc dts files.
Signed-off-by: T Karthik Reddy <[email protected]>
Acked-by: Ashok Reddy Soma <[email protected]>
Signed-off-by: Michal Simek <[email protected]>
Link: https://lore.kernel.org/r/e69b30d82b0307c563fe72630d9172e53964aeda.1652262769.git.michal.simek@amd.com
|
|
There are couple of IPs which are enabled in origin HW design which are
missing in SOM dt. Add them to match default setup.
Signed-off-by: Michal Simek <[email protected]>
Signed-off-by: Michal Simek <[email protected]>
Link: https://lore.kernel.org/r/901401beacbea5931fc18cde20c157e5978a7023.1652262769.git.michal.simek@amd.com
|
|
Add pwm-fan node to control fan through hwmon and change
pwm-cells property to 3 to allow fancontrol utility to
function correctly.
Signed-off-by: Vishal Patel <[email protected]>
Signed-off-by: Michal Simek <[email protected]>
Link: https://lore.kernel.org/r/21b4dfce3e45136a468974ea3dedca03320e27b8.1652262769.git.michal.simek@amd.com
|
|
PL has own power domain which is not described in DT. That's why add it
there by default.
Signed-off-by: Michal Simek <[email protected]>
Signed-off-by: Michal Simek <[email protected]>
Link: https://lore.kernel.org/r/b38e2ea95dab434bc007f9ed6c438c68149744bf.1652262769.git.michal.simek@amd.com
|
|
OPP table name now should start with "opp-table" and OPP entries
shouldn't contain commas and @ signs in accordance to the new schema
requirement.
The same change was done in Linux by commit c6d4a8977598 ("ARM: tegra:
Rename CPU and EMC OPP table device-tree nodes"), commit ffbe853a3f5a
("ARM: dts: sunxi: Fix OPPs node name") or commit b7072cc5704d ("arm64:
dts: qcom: qcs404: Rename CPU and CPR OPP tables").
Signed-off-by: Michal Simek <[email protected]>
Link: https://lore.kernel.org/r/a1176349448df35127dbac15c1eeb2229505bae7.1652262769.git.michal.simek@amd.com
|
|
Using labels helps with better identifications of chips.
Signed-off-by: Michal Simek <[email protected]>
Signed-off-by: Michal Simek <[email protected]>
Link: https://lore.kernel.org/r/15b0f68077fb3c86d438caf8562de87367361c60.1652262769.git.michal.simek@amd.com
|
|
Add mode-pin GPIO controller DT node in zynqmp.dtsi and also setup default
reset-gpios property for usb which is default Xilinx setup.
Signed-off-by: Piyush Mehta <[email protected]>
Signed-off-by: Michal Simek <[email protected]>
Link: https://lore.kernel.org/r/f2a1f6f541c41075ea36062857031bfc28d6d303.1652262769.git.michal.simek@amd.com
|
|
This patch fixes the DP audio and video PLL configurations for the zynqmp-sm-k26-revA som.
The Linux DP driver expects the DP to be using the following PLL config:
- DP video PLL should use the VPLL (0x0)
- DP audio PLL should use the RPLL (0x3)
- DP system time clock PLL should use RPLL (0x3)
Register 0xFD1A0070 configures the DP video PLL.
Register 0xFD1A0074 configures the DP audio PLL.
Register 0xFD1A007C configures the DP system time clock PLL.
Signed-off-by: Neal Frager <[email protected]>
Signed-off-by: Michal Simek <[email protected]>
Link: https://lore.kernel.org/r/fa7e9abc419c9d7648405d1c62367dbe701d09b8.1652709736.git.michal.simek@amd.com
|
|
Observing psgtr pll timeouts with some usb hubs and devices behind it.
Increase timeout to 10ms to take care of it.
Signed-off-by: Ashok Reddy Soma <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Michal Simek <[email protected]>
|
|
In all the ZynqMP boards dts files tx-buswidth is by default set to 1. Due
to this the framework only issues 1-1-1 write commands to the GQSPI driver.
But the GQSPI controller is capable of handling 1-4-4 write commands, so
updated the tx-buswidth to 4 in ZynqMP boards dts files. This would enable
the spi-nor framework to issue 1-4-4 write commands instead of 1-1-1. This
will increase the tx data transfer rate, as now the tx data will be
transferred on four lines instead on single line.
Signed-off-by: Amit Kumar Mahapatra <[email protected]>
Signed-off-by: Ashok Reddy Soma <[email protected]>
Signed-off-by: Michal Simek <[email protected]>
Link: https://lore.kernel.org/r/ad61199f55e5e00f29de6206d9d1872a52a7657e.1652193179.git.michal.simek@amd.com
|
|
CONFIG_SPL_FS_LOAD_PAYLOAD_NAME is used by set_dfu_alt_info() for string
generation. It doesn't depend on SDHCI because the same file can be stored
to other non volatile memories like qspi.
Signed-off-by: Michal Simek <[email protected]>
Link: https://lore.kernel.org/r/97434e122c6f8f042330b60d3e8de4c31f3e3f93.1652192580.git.michal.simek@amd.com
Link: https://lore.kernel.org/r/718de136c68c9a76fc7b4e536a727f401b05bfb9.1652702625.git.michal.simek@amd.com
|
|
This patch fixes two issues in the set_r5_reset function.
1. When in split mode, the lpd_amba_rst bit should only be set when
both r5 cpu cores are in reset. Otherwise, if one r5 core is still
running, setting the lpd_amba_rst bit will cause an error for the
running core. The set_r5_reset function has been modified to check
if the other r5 core is still running before setting the lpd_amba_rst
bit.
2. The cpu_disable function was always assuming that the r5 cores
are in split mode when resetting either core 4 or 5. This is
incorrect for lockstep functionality. This patch adds a function
check_r5_mode to handle the cpu_disable function correctly for
the r5 cores by checking the mode and handling the reset appropriately.
Signed-off-by: Neal Frager <[email protected]>
Signed-off-by: Michal Simek <[email protected]>
Link: https://lore.kernel.org/r/d99cbd7f2394ac055ef27457298f554ff0747ba7.1651648344.git.michal.simek@amd.com
|
|
This patch fixes the DP audio and video PLL configurations for the
zynqmp-zcu102-revA evaluation board.
The Linux DP driver expects the DP to be using the following PLL config:
- DP video PLL should use the VPLL (0x0)
- DP audio PLL should use the RPLL (0x3)
Register 0xFD1A0070 configures the DP video PLL.
Register 0xFD1A0074 configures the DP audio PLL.
Signed-off-by: Neal Frager <[email protected]>
Signed-off-by: Michal Simek <[email protected]>
Link: https://lore.kernel.org/r/b2eb87758e0cd4844e1754da8c58fce58d9cf683.1651740949.git.michal.simek@amd.com
|
|
This patch fixes the DP audio and video PLL configurations
for the zynqmp-zcu106-revA evaluation board.
The Linux DP driver expects the DP to be using the following PLL config:
- DP video PLL should use the VPLL (0x0)
- DP audio PLL should use the RPLL (0x3)
Register 0xFD1A0070 configures the DP video PLL.
Register 0xFD1A0074 configures the DP audio PLL.
Signed-off-by: Neal Frager <[email protected]>
Signed-off-by: Michal Simek <[email protected]>
Link: https://lore.kernel.org/r/62538b4a04dee28a6fc8ac5b85f8c845a5a76aa4.1651740988.git.michal.simek@amd.com
|
|
This patch fixes the DP audio and video PLL configurations for the zynqmp-zcu106-rev1.0 evaluation board.
The Linux DP driver expects the DP to be using the following PLL config:
- DP video PLL should use the VPLL (0x0)
- DP audio PLL should use the RPLL (0x3)
Register 0xFD1A0070 configures the DP video PLL.
Register 0xFD1A0074 configures the DP audio PLL.
Signed-off-by: Neal Frager <[email protected]>
Signed-off-by: Michal Simek <[email protected]>
Link: https://lore.kernel.org/r/ae42ad6185418713a473660c8d15903299af7764.1652192319.git.michal.simek@amd.com
|
|
Currently, pinctrl drivers only get probed if pinconf is actually being
used, however on SoC-s like Armada 3720 pinctrl driver is a also the GPIO
driver.
So, if the pinctrl driver doesn't get probed GPIO-s won't get registered
and thus they cannot be used.
This is a problem on the Methode eDPU as it just uses SB pins as GPIO-s
and without them being registered networking won't work as it only has
one SFP slot and the TX disable GPIO is on the SB controller.
So, probe the pinctrl drivers using DM_FLAG_PROBE_AFTER_BIND like LED
uclass does.
Signed-off-by: Robert Marko <[email protected]>
Reviewed-by: Pali Rohár <[email protected]>
Reviewed-by: Stefan Roese <[email protected]>
|
|
uDPU relies on using fixed-phy for the SFP support, and since the
fixed-phy parsing was moved to the generic driver instead of mvneta
networking stopped working on uDPU with:
uDPU>> dhcp
dm_eth_phy_connect failed
This is due to the conversion commit not enabling fixed-phy support
in defconfig like it did for other boards.
Fixes: 77fcf3cf1251 ("net: mvneta: Convert to use PHY_FIXED for fixed-link")
Signed-off-by: Robert Marko <[email protected]>
Reviewed-by: Stefan Roese <[email protected]>
|
|
This allows to use U-Boot console on Turris devices via network.
Signed-off-by: Pali Rohár <[email protected]>
Reviewed-by: Stefan Roese <[email protected]>
|
|
UART base address is located in internal registers.
Internal registers for 32-bit mvebu boards in SPL are at address 0xd0000000
and in proper U-Boot at address 0xf1000000.
Fix DEBUG_UART_BASE option for all 32-bit mvebu boards.
Signed-off-by: Pali Rohár <[email protected]>
Reviewed-by: Stefan Roese <[email protected]>
|
|
Internal registers in SPL are at address 0xd0000000 and in proper U-Boot at
address 0xf1000000. UART base address is located in internal registers.
Fix DEBUG_UART_BASE option to correct value for both SPL and proper U-Boot.
This change fixes hangup of proper U-Boot when it is trying to print
something via debug UART.
Signed-off-by: Pali Rohár <[email protected]>
Reviewed-by: Stefan Roese <[email protected]>
|
|
Use CONFIG_VAL(DEBUG_UART_BASE) instead of CONFIG_DEBUG_UART_BASE, so
proper config value (CONFIG_DEBUG_UART_BASE or CONFIG_SPL_DEBUG_UART_BASE)
is used based on building target.
Signed-off-by: Pali Rohár <[email protected]>
Reviewed-by: Stefan Roese <[email protected]>
|
|
TPL_DEBUG_UART_BASE is same as DEBUG_UART_BASE, but applies only for TPL.
Signed-off-by: Pali Rohár <[email protected]>
Signed-off-by: Stefan Roese <[email protected]>
|
|
SPL_DEBUG_UART_BASE is same as DEBUG_UART_BASE, but applies only for SPL.
In some cases base address of UART is different in SPL and proper U-Boot.
Signed-off-by: Pali Rohár <[email protected]>
Reviewed-by: Stefan Roese <[email protected]>
|
|
Moving of internal registers from INTREG_BASE_ADDR_REG to SOC_REGS_PHY_BASE
needs to be done very early, prior calling any function which may touch
internal registers, like debug_uart_init().
So do it earlier in arch_very_early_init() instead of arch_cpu_init().
Movement is done in proper U-Boot, not in SPL. SPL may return to bootrom
and bootrom requires internal registers at (old) expected location.
Signed-off-by: Pali Rohár <[email protected]>
Reviewed-by: Stefan Roese <[email protected]>
|
|
When this option is set then ARM _main() function would call
arch_very_early_init() function at the beginning. It would be before
calling any other functions like debug_uart_init() and also before
initializing C runtime environment.
Signed-off-by: Pali Rohár <[email protected]>
Reviewed-by: Stefan Roese <[email protected]>
|
|
Nothing selects ARMADA_64BIT. Instead the 64-bit SoCs just select ARM64
directly. Remove the unused config item.
Signed-off-by: Chris Packham <[email protected]>
Reviewed-by: Stefan Roese <[email protected]>
|
|
CONFIG_MVEBU_NAND_BOOT, CONFIG_MVEBU_SPI_BOOT, CONFIG_MVEBU_MMC_BOOT and
CONFIG_MVEBU_UBOOT_DFLT_NAME are unused when CONFIG_CMD_MVEBU_BUBT is not
enabled. So hide them.
Signed-off-by: Pali Rohár <[email protected]>
Reviewed-by: Stefan Roese <[email protected]>
|
|
zynqmp_pm_is_function_supported() which checks feature support on som,
which is implemented in firmware_zynqmp.c driver. As mini configuration
does not use firmware driver, so create a weak function to avoid
compilation error on zynqmp mini configuration.
Signed-off-by: T Karthik Reddy <[email protected]>
Acked-by: Ashok Reddy Soma <[email protected]>
Reviewed-by: Jaehoon Chung <[email protected]>
Signed-off-by: Michal Simek <[email protected]>
Link: https://lore.kernel.org/r/c60655a509956b8fc3a81671a7dc51157f3973db.1651048030.git.michal.simek@xilinx.com
|
|
This reverts commit 122ca834f2f4a9d70abeece3d1ff200a3556ab24.
Disable CONFIG_ZYNQMP_FIRMWARE config from zynqmp & versal mini emmc
defconfig files, as mini emmc does not use any firmware.
Signed-off-by: T Karthik Reddy <[email protected]>
Acked-by: Ashok Reddy Soma <[email protected]>
Signed-off-by: Michal Simek <[email protected]>
Link: https://lore.kernel.org/r/f73a50980d3af894f1e86cfc98742dbdec573760.1651048242.git.michal.simek@xilinx.com
|
|
board_get_usable_ram_top() was designed for getting the top most location
for U-Boot allocation that's why function itself supports via total_size
parameter to find out where the right location for U-Boot is.
But function itself is also reused by different (EFI) which is passing
total_size as 0 to find out where the usable ram top is. For this case
doesn't make sense (a waste time) to call any lmb functions.
That's why simply return gd->ram_top.
And gd->ram_top is filled already based on previous call for U-Boot iself.
The same solution is also used by stm32mp by commit 92b611e8b003 ("stm32mp:
correctly handle board_get_usable_ram_top(0)") and commit c8510e397fad
("stm32mp: Fix board_get_usable_ram_top()").
Signed-off-by: Michal Simek <[email protected]>
Link: https://lore.kernel.org/r/44470619e74f3e480b70deac24578e3e0d5c907e.1651225945.git.michal.simek@amd.com
|
|
The call to xilinx_pm_request requires an array of a larger size.
Signed-off-by: Michal Simek <[email protected]>
Link: https://lore.kernel.org/r/5f1409de285d7454af171a54e5f115da9d82c44e.1650440343.git.michal.simek@xilinx.com
|
|
The call to xilinx_pm_request requires an array of a larger size.
Signed-off-by: Jorge Ramirez-Ortiz <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Michal Simek <[email protected]>
|
|
- Migrate CONFIG_MTD_CONCAT to Kconfig, use CONFIG_VAL/IS_ENABLED in
more places, rename SPL_LEGACY_IMAGE_SUPPORT to
SPL_LEGACY_IMAGE_FORMAT and update some related dependencies for TI
platforms.
|
|
Update the diagnostic message with revised location of document, which
changed in 3e9fddfc4f1 ("doc: Move devicetree control doc to rST")
Signed-off-by: Ralph Siemsen <[email protected]>
|
|
Drop CONFIG_NEEDS_MANUAL_RELOC ifdefs in board_init_r() and use
IS_ENABLED() instead. Also, use the MANUAL_RELOC() macro to update the
initcall pointers.
Signed-off-by: Ovidiu Panait <[email protected]>
|
|
This converts the following to Kconfig:
CONFIG_MTD_CONCAT
Signed-off-by: Chris Packham <[email protected]>
Reviewed-by: Tom Rini <[email protected]>
Reviewed-by: Stefan Roese <[email protected]>
|
|
There is currently no support for PRE_CONSOLE_BUFFER in SPL, but if
and when that gets implemented, one would almost certainly want to use
a different address and/or size for the buffer (e.g., U-Boot proper
might specify an address in DRAM and a generous buffer, while SPL
would be much more constrained).
So a prerequisite for adding SPL_PRE_CONSOLE_BUFFER is to make the
code use SPL_-specific values. No functional change.
Signed-off-by: Rasmus Villemoes <[email protected]>
|
|
TI secure devices chain-of-trust depends on FIT image processing,
enable it by default on these devices. This also reduces the delta
between the secure and non-secure defconfig files.
Signed-off-by: Andrew Davis <[email protected]>
Reviewed-by: Tom Rini <[email protected]>
|
|
Non-FIT image loading support should be disabled for TI secure
devices as the image handlers for those image types do not follow
our secure boot checks.
Signed-off-by: Andrew Davis <[email protected]>
Reviewed-by: Tom Rini <[email protected]>
|
|
Booting of non-FIT images bypass our chain-of-trust boot flow,
these options should not be allowed when high security is set.
Signed-off-by: Andrew Davis <[email protected]>
Reviewed-by: Tom Rini <[email protected]>
|
|
This matches what this support is called in the non-SPL case. The postfix
_SUPPORT is redundant as enabling Kconfig options implies support.
With this we can use CONFIG_IS_ENABLED() as needed.
Signed-off-by: Andrew Davis <[email protected]>
Reviewed-by: Tom Rini <[email protected]>
|
|
https://source.denx.de/u-boot/custodians/u-boot-stm
Add new STM32 MCU boards and Documentation
STM32 programmer improvements
video: support several LTDC HW versions and fix data enable polarity
board: fix stboard error message, consider USB cable connected when boot device is USB
configs: stm32mp1: set console variable for extlinux.conf
configs: stm32mp1: add support for baudrate higher than 115200 for ST-Link
ARM: stm32mp: Fix Silicon version handling and ft_system_setup()
phy: stm32-usbphyc: Add DT phy tuning support
arm: dts: stm32mp15: alignment with v5.18
ram: Conditionally enable ASR
mach-stm32mp: psci: retain MCUDIVR, PLL3CR, PLL4CR, MSSCKSELR across suspend
configs: Use TFTP_TSIZE on DHSOM and STMicroelectronics boards
ARM: stm32: Use default CONFIG_TFTP_BLOCKSIZE on DHSOM
pinctrl: stm32: rework GPIO holes management
|
|
i2c changes for 2022.07
- i2c: ihs: intel: Fix typo in comments
Patch from Michal
- misc: atsha204a: Add support for atsha204 chip
from Pali
|
|
We should only build support for misc if the appropriate SPL/TPL symbol
is defined. To ease the transition, make SPL/TPL_MISC default to MISC.
This is necessary because many drivers don't specify their dependencies
properly. These defaults can be removed once all drivers depend on the
appropriate config.
Fixes: aaba703fd0 ("spl: misc: Allow misc drivers in SPL and TPL")
Signed-off-by: Sean Anderson <[email protected]>
[trini: Add VPL_MISC symbol, handle like SPL/TPL_MISC]
Signed-off-by: Tom Rini <[email protected]>
|
|
Add documentation for STM32 MCUs (F4, F7 and H7 series).
Signed-off-by: Patrice Chotard <[email protected]>
Reviewed-by: Patrick Delaunay <[email protected]>
|
|
Add stm32746g-eval_spl_defconfig for stm32746g evaluation board to
build SPL.
Signed-off-by: Patrice Chotard <[email protected]>
Reviewed-by: Patrick Delaunay <[email protected]>
|
|
Add stm32746g-eval_defconfig for stm32746g evaluation board to
build U-Boot proper.
Full board description can be found here :
https://www.st.com/en/evaluation-tools/stm32746g-eval.html
Signed-off-by: Patrice Chotard <[email protected]>
Reviewed-by: Patrick Delaunay <[email protected]>
|
