spl_perform_fixups is redundant in the current implementation. SPLs call
dram_init during spl_enable_cache in arch/arm/mach-k3/common.c. In
U-Boot, dram_init and dram_init_banksize are automatically called in
init_sequence_f. Therefore, SPLs and U-Boot always determine the correct
RAM size. During Linux boot, fdt_fixup_memory_banks adjusts the RAM size
in the Linux device tree, ensuring correct RAM usage.
dram_init() calls get_ram_size() for the Verdin AM62, which determines
the correct RAM size through read/write operations. However, due to the
K3 bootflow placing ATF in RAM and requiring cache coherence, calling
get_ram_size() with caches enabled prevents ATF boot. Removing this call
restores ATF boot functionality.
This fixes a regression introduced in commit 4164289db8 ("board:
verdin-am62: fix missing memory fixup call") which prevents some of our
Verdin AM62 modules from booting.
Fixes: 4164289db8 ("board: verdin-am62: fix missing memory fixup call")
Signed-off-by: Stefan Eichenberger <stefan.eichenberger@toradex.com>
Apalis/Colibri iMX6 V1.2 will replace the STMPE811 ADC/Touch controller
which is EOL by the TLA2024 ADC and AD7879 touch controller.
To support this new version, we detect the presence of the TLA2024
during boot and set a new environment variable named "variant". This
will allow us and users to select the correct DT easily.
By probing via I2C we have a robust detection method instead of relying
on the existing "board_rev" environment variable which is set by the
config block. Users can use "variant" in their DT selection and do not
have to map the board revision to a device tree.
"variant" environment variable behaviour:
* Empty or absent for all versions below v1.2 (STMPE811)
* "-v1.2" for all versions starting from v1.2 (TLA2024 + AD7879)
Usage example:
setenv fdtfile imx6q-apalis${variant}-${fdt_board}.dtb
Signed-off-by: Ernest Van Hoecke <ernest.vanhoecke@toradex.com>
Reviewed-by: Francesco Dolcini <francesco.dolcini@toradex.com>
The commit bc07851897 ("board: ti: Pull redundant DDR functions to a
common location and Fixup DDR size when ECC is enabled") broke DRAM
support for the Verdin AM62. This was partially fixed with commit
3f866c47b5 ("board: verdin-am62: add dram_init_banksize"). However,
because fixup_memory_node was not called, the Linux kernel was started
with the wrong memory size on modules with less memory available. This
resulted in boot failures. Fix this issue by calling fixup_memory_node
in the board file.
spl_perform_fixups will be called in the SPL and now sets the correct
memory size in the device tree of U-Boot by calling fixup_memory_node.
U-Boot will then adjust the memory sizes of Linux during bootm/booti in
fdt_fixup_memory_banks. This chain ensures that U-Boot and Linux only
use RAM that is actually available.
Fixes: 3f866c47b5 ("board: verdin-am62: add dram_init_banksize")
Fixes: bc07851897 ("board: ti: Pull redundant DDR functions to a common location and Fixup DDR size when ECC is enabled")
Signed-off-by: Stefan Eichenberger <stefan.eichenberger@toradex.com>
Acked-by: Francesco Dolcini <francesco.dolcini@toradex.com>
Add the dram_init_banksize function to the board file to properly set
DRAM memory sizes during boot.
The commit bc07851897 ("board: ti: Pull redundant DDR functions to a
common location and Fixup DDR size when ECC is enabled") relocated the
dram_init_banksize function from architecture specific initialization to
the TI board initialization code. As a result, boards relying on the
previous setup now require this function to be defined within their
board file to handle DRAM sizing correctly.
Without this function defined the following error appears during boot:
ERROR: Failed to allocate 0x1000 bytes below 0x0.
Fixes: bc07851897 ("board: ti: Pull redundant DDR functions to a common location and Fixup DDR size when ECC is enabled")
Signed-off-by: Stefan Eichenberger <stefan.eichenberger@toradex.com>
Acked-by: Francesco Dolcini <francesco.dolcini@toradex.com>
The Toradex configuration block is stored in the first boot partition of
the eMMC. After reading the configuration block, U-Boot switches back to
the user partition. Currently, this operation always targets mmc device
0, even when the configuration block is stored on mmc device 2.
This patch addresses the issue by switching the mmc device set in
CONFIG_TDX_CFG_BLOCK_DEV to the user partition, rather than using the
hardcoded device 0.
Fixes: a2777ecb9d ("toradex: config block handling")
Signed-off-by: Stefan Eichenberger <stefan.eichenberger@toradex.com>
Reviewed-by: Francesco Dolcini <francesco.dolcini@toradex.com>
Simplify module version handling by removing the special case for the
Apalis iMX8QM.
The Apalis iMX8QM has been treated as a special case in module version
handling, but was always the default. By removing this special handling,
the code is simplified and easier to maintain.
We will not print the message "Unknown Apalis iMX8 module" anymore.
However, we still handle this because if the config block is missing we
show "MISSING TORADEX CONFIG BLOCK" and if the serial number is unknown
we show "Model: Toradex 0000 UNKNOWN MODULE V1.1A". Therefore, it is
still possible to detect such issues.
Signed-off-by: Stefan Eichenberger <stefan.eichenberger@toradex.com>
Add new PIDs for the Apalis iMX8 modules featuring the industrial
variant of the SoC, which supports a maximum CPU frequency of 1300 MHz.
Signed-off-by: Stefan Eichenberger <stefan.eichenberger@toradex.com>
This preparation step ensures that the first column has sufficient space
to handle longer defines. This is necessary for the new Apalis iMX8
defines.
Signed-off-by: Stefan Eichenberger <stefan.eichenberger@toradex.com>
Apalis iMX8DXP 1GB is currently set to enable when compiling for Apalis
iMX8. This is a mistake the Apalis iMX8DXP 1GB was never released and is
not compatible to the Apalis iMX8 series.
Signed-off-by: Stefan Eichenberger <stefan.eichenberger@toradex.com>
The module pid4 currently corresponds to the index of the toradex_module
array. If a new pid4 is introduced that does not follow the sequence of
the previous entries, it will create a gap in the array.
To address this, embed pid4 within the toradex_som structure and
implement a function to retrieve the index corresponding to pid4.
Signed-off-by: Vitor Soares <vitor.soares@toradex.com>
Acked-by: Francesco Dolcini <francesco.dolcini@toradex.com>
We are about to add a large number of new entries. Update the prefix to
be a little shorter.
For SMBIOS items, use SYSID_SM_ (for System Management) which is enough
to distinguish it. For now at least, it seems that most items will be
for SMBIOS.
Signed-off-by: Simon Glass <sjg@chromium.org>
Acked-by: Raymond Mao <raymond.mao@linaro.org>
Assorted Tegra enhancements. Merged with the recent XPL_BUILD changes,
resolve some whitespace issues and fix the name of the new apalis-tk1
env file by Tom.
Signed-off-by: Tom Rini <trini@konsulko.com>
Convert boards to use text based env. This is the first stage of
conversion, common inclusions should be converted next.
Acked-by: Francesco Dolcini <francesco.dolcini@toradex.com> # Toradex Apalis TK1
Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Svyatoslav Ryhel <clamor95@gmail.com>
The older i.MX8M Mini Verdin SoMs may came with 20 MHz SPI CAN controller
oscillator, the newer SoMs always use 40 MHz oscillator. Handle both by
overriding the oscillator frequency just before booting the kernel.
These are the known variants with 20 MHz oscillator:
- 0055, V1.1A, V1.1B, V1.1C and V1.1D, use a 20MHz oscillator
- 0059, V1.1A and V1.1B, use a 20MHz oscillator
Signed-off-by: Marek Vasut <marex@denx.de>
Reviewed-by: Francesco Dolcini <francesco.dolcini@toradex.com>
Add support for MT53E512M32D1ZW-046 IT:C memory.
This 4 GB memory has 17 row bits instead of 16 and requires 380 ns of
tRFC (tRFCab) instead of 280 ns due to increased channel density to 16 Gb.
Both modifications are retro-compatible with previous memories.
Signed-off-by: Emanuele Ghidoli <emanuele.ghidoli@toradex.com>
As part of bringing the master branch back in to next, we need to allow
for all of these changes to exist here.
Reported-by: Jonas Karlman <jonas@kwiboo.se>
Signed-off-by: Tom Rini <trini@konsulko.com>
When bringing in the series 'arm: dts: am62-beagleplay: Fix Beagleplay
Ethernet"' I failed to notice that b4 noticed it was based on next and
so took that as the base commit and merged that part of next to master.
This reverts commit c8ffd1356d, reversing
changes made to 2ee6f3a5f7.
Reported-by: Jonas Karlman <jonas@kwiboo.se>
Signed-off-by: Tom Rini <trini@konsulko.com>
Move verdin-am62 to OF_UPSTREAM:
- handle the fact that dtbs now have a 'ti/' prefix
- imply OF_UPSTREAM
- remove redundant files from arch/arm/dts leaving only the
*-u-boot.dtsi files
- update MAINTAINERS file
Signed-off-by: Marcel Ziswiler <marcel.ziswiler@toradex.com>
Reviewed-by: Dhruva Gole <d-gole@ti.com>
Reviewed-by: Neha Malcom Francis <n-francis@ti.com>
Reviewed-by: Sumit Garg <sumit.garg@linaro.org>
Move verdin-imx8mm and verdin-imx8mp to OF_UPSTREAM:
- handle the fact that dtbs now have a 'freescale/' prefix
- imply OF_UPSTREAM
- remove redundant files from arch/arm/dts leaving only the
*-u-boot.dtsi files
- update MAINTAINERS files
Signed-off-by: Marcel Ziswiler <marcel.ziswiler@toradex.com>
Reviewed-by: Sumit Garg <sumit.garg@linaro.org>
On Colibri iMX8X, the gpio initialization and usage should be handled by
the device tree, not by the colibri-imx8x.c code. Therefore, remove it.
Signed-off-by: Hiago De Franco <hiago.franco@toradex.com>
Add a weak default reset_cpu() implementation just like
it is done on arch/arm/mach-imx/cpu.c.
This allows the removal of the empty reset_cpu() in several
board files.
Signed-off-by: Fabio Estevam <festevam@denx.de>
Tested-by: Hiago De Franco <hiago.franco@toradex.com> # Toradex Colibri iMX8X
Commit bfb3409d67 ("imx: toradex/apalis-imx8: correct SCU API usage")
made an incorrect logic change in the error code check of
sc_misc_otp_fuse_read():
- if (scierr == SC_ERR_NONE) {
+ if (scierr) {
/* QP has one A72 core disabled */
is_quadplus = ((val >> 4) & 0x3) != 0x0;
}
The other changes in this commit are correct.
sc_misc_otp_fuse_read() returns 0 on a successful fuse read.
This inversion causes board_mem_get_layout() to report incorrect RAM size.
Go back the original error check logic to fix the problem.
Fixes: bfb3409d67 ("imx: toradex/apalis-imx8: correct SCU API usage")
Signed-off-by: Fabio Estevam <festevam@gmail.com>
Reviewed-by: Peng Fan <peng.fan@nxp.com>
Acked-by: Marcel Ziswiler <marcel.ziswiler@toradex.com>
Commit aa6e698a7a ("imx: toradex/colibri-imx8x: correct SCU API usage")
made an incorrect logic change in the error code check of
sc_misc_otp_fuse_read():
- if (sc_err == SC_ERR_NONE) {
+ if (sc_err) {
/* DX has two A35 cores disabled */
return (val & 0xf) != 0x0;
}
The other changes in this commit are correct.
sc_misc_otp_fuse_read() returns 0 on a successful fuse read.
This inversion causes board_mem_get_layout() to report incorrect RAM size.
Go back the original error check logic to fix the problem.
Fixes: aa6e698a7a ("imx: toradex/colibri-imx8x: correct SCU API usage")
Reported-by: Hiago De Franco <hiago.franco@toradex.com>
Signed-off-by: Fabio Estevam <festevam@gmail.com>
Reviewed-by: Peng Fan <peng.fan@nxp.com>
Acked-by: Marcel Ziswiler <marcel.ziswiler@toradex.com>
Tested-by: Hiago De Franco <hiago.franco@toradex.com> # Toradex Colibri iMX8X 1GB
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>
Speed grade T requires the VDD_CORE voltage to be 0.85V if using
the maximum core frequency.
Speed grades G, K, S allow the VDD_CORE voltage to be 0.75V up to the
maximum core frequency but allow to run at 0.85V.
For efficiency in manufacturing and code maintenance we use 0.85V for
the PMIC defaults and device tree settings and dynamically adjust the
voltage in the PMIC and device tree to 0.75V for lower speed SKU to
gain more than 100mW power consumption reduction.
Signed-off-by: Max Krummenacher <max.krummenacher@toradex.com>
Clean up all configuration files to adhere to yamllint rules.
Signed-off-by: Neha Malcom Francis <n-francis@ti.com>
[trini: Update more yaml files added since this was posted]
Signed-off-by: Tom Rini <trini@konsulko.com>
Suggested-by: Nishanth Menon <nm@ti.com>
These files include <common.h> twice. Start by removing the second
inclusion of the file.
Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Tom Rini <trini@konsulko.com>
We need to include <config.h> directly when a file needs to have
something such as CFG_SYS_SDRAM_SIZE referenced as this file is not
automatically globally included and is most commonly indirectly included
via common.h. Remove most cases of arm including config.h directly, but
add it where needed. This includes a few board-specific fixes.
Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Tom Rini <trini@konsulko.com>
This commit adds support for Fastboot boot commands by checking both
CONFIG_CMD_USB_SDP and CONFIG_USB_FUNCTION_FASTBOOT. If either of these
configurations is set, it indicates that the board is in recovery mode
and can use either SDP or Fastboot.
The default option remains the SDP command, but if
CONFIG_CMD_FASTBOOT is set, it changes to 'fastboot usb 0' as
the boot command.
Signed-off-by: Hiago De Franco <hiago.franco@toradex.com>
Reviewed-by: Fabio Estevam <festevam@gmail.com>
