Hi Mathieu,
On 30. 01. 24 10:43, Mathieu Othacehe wrote:
Hello,
Please remove findfdt script.
Not needed when you set fdt_file to CONFIG_DEFAULT_FDT_FILE above.
The reason I used findfdt is that CONFIG_DEFAULT_FDT_FILE is a string so, we end up with:
--8<---------------cut here---------------start------------->8--- fdt_file="oftree" --8<---------------cut here---------------end--------------->8---
which gives:
--8<---------------cut here---------------start------------->8--- u-boot=> run loadfdt Failed to load '"oftree"' --8<---------------cut here---------------end--------------->8---
whereas, if I set:
--8<---------------cut here---------------start------------->8--- u-boot=> setenv fdt_file oftree u-boot=> run loadfdt 24579 bytes read in 5 ms (4.7 MiB/s) --8<---------------cut here---------------end--------------->8---
I then see three options: introduce an unstrigify macro, hardcode fdt_file=oftree in the env file, or use the findfdt hack.
On my side there is no problem, see my output here:
U-Boot 2024.04-rc1-00004-g686cb1ca1391-dirty (Jan 30 2024 - 10:58:24 +0100)
CPU: i.MX93 rev1.1 Model: PHYTEC phyBOARD-Segin-i.MX93 DRAM: 1 GiB Core: 184 devices, 22 uclasses, devicetree: separate WDT: Started wdog@42490000 with servicing every 1000ms (40s timeout) MMC: FSL_SDHC: 0, FSL_SDHC: 1 Loading Environment from MMC... OK In: serial@44380000 Out: serial@44380000 Err: serial@44380000 Net: eth0: ethernet@42890000 Hit any key to stop autoboot: 0 u-boot=> u-boot=> env default -a ## Resetting to default environment u-boot=> saveenv; saveenv Saving Environment to MMC... Writing to redundant MMC(1)... OK Saving Environment to MMC... Writing to MMC(1)... OK u-boot=> printenv fdt_file fdt_file="oftree" u-boot=> run loadfdt 48532 bytes read in 5 ms (9.3 MiB/s) u-boot=> boot switch to partitions #0, OK mmc1 is current device 32688640 bytes read in 411 ms (75.8 MiB/s) Booting from mmc ... 83 bytes read in 1 ms (81.1 KiB/s) 48532 bytes read in 4 ms (11.6 MiB/s) Working FDT set to 83000000 3615 bytes read in 3 ms (1.1 MiB/s) 1209 bytes read in 2 ms (589.8 KiB/s) ## Flattened Device Tree blob at 83000000 Booting using the fdt blob at 0x83000000 Working FDT set to 83000000 Loading Device Tree to 00000000bcf0a000, end 00000000bcf19fff ... OK Working FDT set to bcf0a000
Starting kernel ...
Here is my diff compared to your v3:
--- a/board/phytec/phycore_imx93/phycore_imx93.env +++ b/board/phytec/phycore_imx93/phycore_imx93.env @@ -5,7 +5,7 @@ console=ttyLP0 fdt_addr=0x83000000 fdto_addr=0x830c0000 bootenv_addr=0x83500000 -fdt_file=undefined +fdt_file=CONFIG_DEFAULT_FDT_FILE ipaddr=192.168.3.11 serverip=192.168.3.10 netmask=255.255.255.0 @@ -22,11 +22,7 @@ mmcautodetect=yes mmcargs=setenv bootargs ${mcore_clk} console=${console},${baudrate} earlycon root=/dev/mmcblk${mmcdev}p${mmcroot} ${raucargs} rootwait rw loadimage=fatload mmc ${mmcdev}:${mmcpart} ${loadaddr} ${image} -findfdt=if test $fdt_file = undefined; then - setenv fdt_file CONFIG_DEFAULT_FDT_FILE ; - fi; - echo fdt_file=${fdt_file}; -loadfdt=run findfdt;fatload mmc ${mmcdev}:${mmcpart} ${fdt_addr} ${fdt_file} +loadfdt=fatload mmc ${mmcdev}:${mmcpart} ${fdt_addr} ${fdt_file} mmc_load_overlay=fatload mmc ${mmcdev}:${mmcpart} ${fdto_addr} ${overlay} mmc_apply_overlays=
So this should work. We have the same in the downstream u-boot-imx.
Also, in-tree, phycore_am62x.env has the same:
fdtfile=CONFIG_DEFAULT_FDT_FILE loadfdt=load mmc ${mmcdev}:${mmcpart} ${fdtaddr} ${fdtfile}
Can you re-check, please?
Maybe check your include/generated/environment.h?
I'm not sure which one is the better?
I will fix all the other remarks in the v4.
OK, thanks.
I spotted additional remark that was not fixed in v3:
In file doc/board/phytec/imx93-phyboard-segin.rst, can you please change title from:
phyCORE-i.MX 93 on a phyBOARD-Segin ===================================
to:
phyBOARD-Segin-i.MX93 =====================
and this text from:
U-Boot for the phyCORE-i.MX 93 on a phyBOARD-Segin.
to:
U-Boot for the phyBOARD-Segin-i.MX93.
Last but not least, please remove:
#define EEPROM_ADDR 0x50
from board/phytec/phycore_imx93/spl.c
As support for PHYTEC_EEPROM will come later and that define will be part of that series.
Thanks and best regards, Primoz
Thanks,
Mathieu