03:28 <saintdev> full mainline on my C1 :D

03:29 <saintdev> thanks for all your hard work everyone!

15:26 <afaerber> repk, narmstrong: in my testing it looks like the first 512 bytes of the gxl .sd.bin file are random - I verified that I could boot LePotato by using /dev/zero for 444 first bytes

15:37 <repk> afaerber: yes IIRC my reverse of aml_encrypt_gxl the 512 first bytes are filled with random numbers

15:49 <afaerber> thanks for confirming, in that case we could simplify the U-Boot instructions to only dd the second part

15:52 <repk> yes instructions in gxlimg readme only dd the second part, and I'm not aware of any problem with that so far

16:00 <afaerber> repk, I also took a shortcut of using unmodified upstream fiptool with --align 0x4000, dd'ing the first 0x3e00 bytes into a file that I then run through the encryption - avoided needing to reimplement fip support

16:00 <afaerber> --scp-fw for bl30, --soc-fw for bl31 and --nt-fw for bl33

19:03 <lvrp16> Orly)