Since commit 27cc5951c8 ("include: env: ti: add default for
do_main_cpsw0_qsgmii_phyinit"), the value of the environment variable
do_main_cpsw0_qsgmii_phyinit happened to remain '0' and couldn't be
changed without user intervention. This behavior is due to the following
cyclic dependency:
A) ti_common.env sets do_main_cpsw0_qsgmii_phyinit to '0' and its value
can only be updated automatically by main_cpsw0_qsgmii_phyinit.
B) main_cpsw0_qsgmii_phyinit is defined in j721e.env and it can run only
if 'do_main_cpsw0_qsgmii_phyinit' is already '1' which isn't possible
unless the user manually assigns the value.
Fix the aforementioned cyclic dependency by using board_late_init() to
detect the QSGMII Daughtercard and set do_main_cpsw0_qsgmii_phyinit.
Additionally, to address the issue of do_main_cpsw0_qsgmii_phyinit being
'undefined' for other platforms, replace:
if test ${do_main_cpsw0_qsgmii_phyinit} -eq 1;
with:
if env exists do_main_cpsw0_qsgmii_phyinit;
in ti_common.env.
Fixes: 27cc5951c8 ("include: env: ti: add default for do_main_cpsw0_qsgmii_phyinit")
Signed-off-by: Siddharth Vadapalli <s-vadapalli@ti.com>
Reviewed-by: Anshul Dalal <anshuld@ti.com>
`dram_init()` is called by R5 SPL and U-Boot, both. It starts by
computing the size of the RAM. In verdin-am62(p), it does so by calling
`get_ram_size()`. This function computes the size of the RAM by writing
over the RAM.
When R5 computes the size of the RAM, it does not update the DT with
this size. As a result, when A53 invokes `dram_init()` again, it has to
compute the size through `get_ram_size()` again.
Commit 13c54cf588 and 0c3a6f748c add firewall over ATF's and OPTEE's
regions. This firewall is added during the R5 SPL stage of boot. So when
A53 attempts to write over RAM in `get_ram_size()`, it writes over the
protected region. Since A53 is a non-secure core, this is blocked by the
firewall.
To fix this, do the following:
* Implement `spl_perform_board_fixups()` function for verdin-am62
and verdin-am62p. Make this function call `fixup_memory_node()`,
which updates the DT.
* Add an if-block in `dram_init()`, to ensure that only R5 is able
to call `get_ram_size()`, and that A53 reads this size from the
DT.
Signed-off-by: Suhaas Joshi <s-joshi@ti.com>
Reviewed-by: Francesco Dolcini <francesco.dolcini@toradex.com>
Move DRAM bank detection from fdtdec to custom implementation to
ensure memory banks are populated before get_page_table_size() is
called during MMU initialization.
The current fdtdec-based approach populates gd->bd->bi_dram[] too
late in the boot sequence, causing get_page_table_size() to be
called with unpopulated DRAM information. This prevents dynamic
page table sizing based on actual memory configuration.
Parse /memory nodes in dram_init() to fill versal2_mem_map[]
early enough for MMU setup. Supports up to
CONFIG_NR_DRAM_BANKS (36) non-contiguous banks with high memory
regions (>4GB) and use __weak get_page_table_size implementation
to estimate page table size based on the populated DRAM banks.
Signed-off-by: Pranav Sanwal <pranav.sanwal@amd.com>
Signed-off-by: Michal Simek <michal.simek@amd.com>
Link: https://lore.kernel.org/r/20260129120021.1328653-3-pranav.sanwal@amd.com
`dram_init()` is called by R5 SPL and U-Boot, both. It starts by
computing the size of the RAM. In verdin-am62(p), it does so by calling
`get_ram_size()`. This function computes the size of the RAM by writing
over the RAM.
When R5 computes the size of the RAM, it does not update the DT with
this size. As a result, when A53 invokes `dram_init()` again, it has to
compute the size through `get_ram_size()` again.
Commit 13c54cf588 and 0c3a6f748c add firewall over ATF's and OPTEE's
regions. This firewall is added during the R5 SPL stage of boot. So when
A53 attempts to write over RAM in `get_ram_size()`, it writes over the
protected region. Since A53 is a non-secure core, this is blocked by the
firewall.
To fix this, do the following:
* Implement `spl_perform_board_fixups()` function for verdin-am62
and verdin-am62p. Make this function call `fixup_memory_node()`,
which updates the DT.
* Add an if-block in `dram_init()`, to ensure that only R5 is able
to call `get_ram_size()`, and that A53 reads this size from the
DT.
Signed-off-by: Suhaas Joshi <s-joshi@ti.com>
Reviewed-by: Francesco Dolcini <francesco.dolcini@toradex.com>
The fdtdfile variable contains quotes:
printenv fdtfile
fdtfile="sifive/hifive-unmatched-a00.dtb"
But this leads to issues which booting with an extlinux.conf format
file failing to find the .dtb file:
Retrieving file: /usr/lib/linux-image-6.12.63+deb13-riscv64/"sifive/hifive-unmatched-a00.dtb"
Skipping fdtdir /usr/lib/linux-image-6.12.63+deb13-riscv64/ for failure retrieving dts
Use the DEFAULT_FDT_FILE variable which has the quotes removed.
Signed-off-by: Vagrant Cascadian <vagrant@debian.org>
First set of u-boot-at91 features for the 2026.04 cycle:
This small fixes set includes fixing 64 bit builds and some warnings for
the at91 serial driver, and some cleanup on the nand driver.
Rely on serial driver calling clk_enable to enable the lpuart clk, no
need to do init_uart_clk in board_early_init_f().
Also remove board_early_init_f(), because it is empty now.
Signed-off-by: Peng Fan <peng.fan@nxp.com>
Acked-by: Francesco Dolcini <francesco.dolcini@toradex.com>
Rely on serial driver calling clk_enable to enable the lpuart clk, no
need to do init_uart_clk in board_early_init_f().
Also remove board_early_init_f(), because it is empty now.
Signed-off-by: Peng Fan <peng.fan@nxp.com>
Rely on serial driver calling clk_enable to enable the lpuart clk, no
need to do init_uart_clk in board_early_init_f().
Also remove board_early_init_f(), because it is empty now.
Signed-off-by: Peng Fan <peng.fan@nxp.com>
If i2c_eeprom_size fails, the error value is stored in
variable size and not ret.
Also, this commit fixes printing the error value.
Signed-off-by: Francois Berder <fberder@outlook.fr>
Move the arch specific atmel_serial.h header from AT91 to the generic
include/dm/platform_data.
This will be used for support on Microchip LAN969x.
Signed-off-by: Robert Marko <robert.marko@sartura.hr>
CI: https://source.denx.de/u-boot/custodians/u-boot-nand-flash/-/pipelines/29183
This series provides a comprehensive cleanup of the Allwinner (sunxi)
NAND controller drivers and introduces full support for the H6 and H616
SoCs in both the main U-Boot driver and the SPL.
The series successfully deduplicates register maps between sunxi_nand.c
and sunxi_nand_spl.c while migrating to a capability-based architecture.
This approach allows the driver to handle the H616's specific
requirements—such as shifted register offsets for ECC/OOB, the removal
of 512B ECC block support, and mandatory MBUS clock gating—without
breaking compatibility for legacy A10/A23 devices.
This adds support for the MT8390 EVK board with the following
features enabled/tested: Boot, UART, Watchdog and MMC.
MT8390 is based on MT8188.
Signed-off-by: Julien Masson <jmasson@baylibre.com>
Signed-off-by: Julien Stephan <jstephan@baylibre.com>
Reviewed-by: Mattijs Korpershoek <mkorpershoek@kernel.org>
Ernest Van Hoecke <ernest.vanhoecke@toradex.com> says:
This is a small, board-specific series for Aquila AM69.
Patch 1 fixes intermittent SPL USB DFU gadget enumeration.
Patch 2 drops a stale MCU_CLKOUT0 enable for ETH_1. V1.1 hardware uses
an external 25 MHz crystal, and support for earlier revisions was
already removed from the DT before upstreaming.
Link: https://lore.kernel.org/r/20260127101413.2812815-1-ernestvanhoecke@gmail.com
On the Toradex Aquila AM69 V1.1, the on-module ETH_1 relies on an
external 25 MHz crystal oscillator. On the V1.0, we needed to enable
MCU_CLKOUT0, but support for this was already dropped from the device
tree before being sent to U-Boot.
Remove this obsolete enabling of MCU_CLKOUT0.
Fixes: 3f0528882c ("board: toradex: add aquila am69 support")
Signed-off-by: Ernest Van Hoecke <ernest.vanhoecke@toradex.com>
Reviewed-by: Francesco Dolcini <francesco.dolcini@toradex.com>
Introduce H6/H616 NAND controller support for SPL
The H616 NAND controller has the same base as A10/A23, with some
differences:
- MDMA is based on chained buffers
- its ECC supports up to 80bit per 1024bytes
- some registers layouts are a bit different, mainly due do the stronger
ECC.
- it uses USER_DATA_LEN registers along USER_DATA registers.
- it needs a specific clock for ECC and MBUS.
For SPL, most of the work was setting the clocks, adding the new
capability structure for H616 and supporting the new USER_DATA_LEN
registers.
Tested on Whatsminer H616 board (with and without scrambling, ECC)
Signed-off-by: Richard Genoud <richard.genoud@bootlin.com>
Signed-off-by: Michael Trimarchi <michael@amarulasolutions.com>
The sunxi_ccm_reg is legacy, drop its usage from nand related code
For that, CCU_NAND0_CLK_CFG and CCU_AHB_GATE1 are added to the clock
files when missing.
And clock code in sunxi_nand{,_spl}.c and board.c are changed to use the
new scheme.
Moreover, drop AHB_DIV_1 in favor of the more readable CCM_NAND_CTRL_M/N
Suggested-by: Andre Przywara <andre.przywara@arm.com>
Signed-off-by: Richard Genoud <richard.genoud@bootlin.com>
Signed-off-by: Michael Trimarchi <michael@amarulasolutions.com>
Kuan-Wei Chiu <visitorckw@gmail.com> says:
Add support for the QEMU 'virt' machine on the m68k architecture. The
QEMU virt machine models a generic system utilizing Goldfish virtual
peripherals and is capable of emulating various classic 68k CPUs.
Currently, U-Boot's m68k architecture support focuses on ColdFire
variants. This series expands support to include the classic M680x0
architecture, implementing the necessary exception vectors, startup
code, and a bootinfo parser compatible with the QEMU interface.
Drivers for Goldfish peripherals (TTY, Timer, RTC) and the QEMU
Virtual System Controller (sysreset) are also added to enable serial
console, timekeeping, and system reset functionality.
The implementation has been verified on QEMU targeting the M68040 CPU,
confirming successful hardware initialization and boot to the U-Boot
command shell. Additionally, the CI configuration was verified locally
using gitlab-ci-local "qemu_m68k_virt test.py", resulting in
PASS qemu_m68k_virt test.py.
Link: https://lore.kernel.org/r/20260107201838.3448806-1-visitorckw@gmail.com
[trini: Re-sort MAINTAINERS entries]
Signed-off-by: Tom Rini <trini@konsulko.com>
Add support for the QEMU 'virt' machine on the m68k architecture. This
board emulates a generic machine based on the Motorola 68040 CPU
equipped with Goldfish virtual peripherals.
Introduce the necessary board configuration and initialization
infrastructure. The implementation includes logic to parse the QEMU
bootinfo interface, enabling dynamic detection of system RAM size to
adapt to the virtual machine's configuration.
Enable the Goldfish TTY driver for serial console output. Additionally,
enable Goldfish RTC and timer drivers to support real-time clock
functionality and nanosecond-resolution delays. Include comprehensive
documentation covering build instructions and usage examples.
Signed-off-by: Kuan-Wei Chiu <visitorckw@gmail.com>
Tested-by: Daniel Palmer <daniel@0x0f.com>
Reviewed-by: Simon Glass <simon.glass@canonical.com>
Following the 'commit 95b5a7de30 ("FWU: STM32MP1: Add support to
read boot index from backup register")', this patch enables reading
the boot index from backup registers on STM32MP2 platforms.
Signed-off-by: Dario Binacchi <dario.binacchi@amarulasolutions.com>
Reviewed-by: Patrice Chotard <patrice.chotard@foss.st.com>
Use plain CONFIG_STM32MP15X to discern code which is specific to
STM32MP15xx in DH STM32MP1 DHSOM board files.
Signed-off-by: Marek Vasut <marex@nabladev.com>
Reviewed-by: Patrice Chotard <patrice.chotard@foss.st.com>
Enable 32k crystal on the board. If external 32k source is not
used, 32k rc-osc comes into play, which is accurate to +-20%.
Signed-off-by: Vishal Mahaveer <vishalm@ti.com>
Enable 32k crystal on the board. If external 32k source is not
used, 32k rc-osc comes into play, which is accurate to +-20%.
Signed-off-by: Vishal Mahaveer <vishalm@ti.com>
Enable 32k crystal on the board. If external 32k source is not
used, 32k rc-osc comes into play, which is accurate to +-20%.
Signed-off-by: Vishal Mahaveer <vishalm@ti.com>
Add a common helper function for doing the basic configuration
required for enabling the 32k crystal on some of the TI boards.
Signed-off-by: Vishal Mahaveer <vishalm@ti.com>
As of v2026.01, no platforms contain any rel.dyn sections in their xPL
phase images. Their inclusion in linker scripts initially was an
oversight as part of taking the full U-Boot linker scripts and modifying
them down. Then in commit 8b0ebe054b ("arm: Update linker scripts to
ensure appended device tree is aligned") these sections were used to
force correct alignment for the device tree. This however, lead to a
different problem.
That problem is that when we do not have a separate BSS section in SPL
we instead would overlay the BSS with the rel.dyn section, in the common
linker script case. This in turn lead to creating an incorrectly sized
BSS "pad" file sometimes (depending on arbitrary changes within the rest
of the binary itself). This in turn lead to the dtb being in the wrong
location in the binary and not found at run time.
This commit fixes a few things:
- Remove the rel.dyn section from all ARM SPL linker scripts.
- In turn, this moves the dtb alignment statement in to another section.
- For ast2600 which uses CONFIG_POSITION_INDEPENDENT we need to keep the
symbols however.
Tested-by: Fabio Estevam <festevam@gmail.com>
Reported-by: Fabio Estevam <festevam@gmail.com>
Co-developed-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Reviewed-by: Chia-Wei, Wang <chiawei_wang@aspeedtech.com>
Signed-off-by: Tom Rini <trini@konsulko.com>
