Hi Ahmad.
On Mon, Apr 01, 2019 at 12:18:22PM +0200, Ahmad Fatoum wrote:
> The newly added device tree based first stage fails to load the second
> stage from MMC, which might be in relation to a preceding atmel_mci
> "command/data timeout" message.
I had similar troubles with MMC on at91sam9263.
After some digging the fix was simple:
See 5feabc1e6737742f1cf6a1c41921f68b4dbb5c10 ("arm: at91: fix clock to mci1 for at91sam9263")
Maybe you are hit by something as simple as this?
> +
> +static const struct devfs_partition sama5d3_xplained_nand0_partitions[] = {
> + {
> + .offset = 0x00000,
> + .size = SZ_256K,
> + .flags = DEVFS_PARTITION_FIXED,
> + .name = "at91bootstrap_raw",
> + .bbname = "at91bootstrap",
> + },
Strange partition name now we use barebox as first stage.
> +++ b/arch/arm/configs/microchip_ksz9477_evb_bootstrap_mmc_defconfig
Naming confuses me.
Do we stick to the name "first stage" or do we use "bootstrap"?
For me they are the same, but I get confused when we use both.
Sam
_______________________________________________
barebox mailing list
barebox@xxxxxxxxxxxxxxxxxxx
http://lists.infradead.org/mailman/listinfo/barebox
