08:12 <narmstrong> ldevulder: uboot 2019.01 rc1 is out with your change, are you able to run a test on your boards ?

08:54 <ldevulder> narmstrong, I saw this yesterday, yes I will try to test rc1 today or tomorrow

09:08 <narmstrong> ldevulder: thanks !!! I’m glad to have someone who can test !

10:17 <AntonioND> adema_, I won't have time to do anything until tuesday, so you can try it if you want: https://github.com/AntonioND/arm-trusted-firmware/tree/an/gxl it gets to the u-boot prompt, but I have no idea if the secondary cores will boot when linux tries to power them on. it should work...

10:18 <adema_> Ok i'll try it later in the week, maybe tomorrow! Thanks :) poulecaca might also wanna try

10:21 <AntonioND> most of the code is exactly the same as gxbb, but there are a few things I have seen in the original bl31.bin that I haven't implemented. i know that there is code that checks the hw revision but I don't know what that is used for yet

10:23 <poulecaca> AntonioND: does gxbb also use a SCP ?

10:23 <AntonioND> yes

10:29 <poulecaca> you need to communicate with it, to boot a secondary CPU right ?

10:30 <poulecaca> ah yes seems so. https://github.com/AntonioND/arm-trusted-firmware/blob/an/gxl/plat/meson/gxl/gxl_scpi.c

10:30 <AntonioND> yes

10:31 <poulecaca> Is it standardized or have you reversed the SCP ?

10:31 <AntonioND> but then again, all the scpi code of gxl is untested... I think that at least the address is right

10:31 <AntonioND> the scpi protocol is standardised, yes

10:31 <AntonioND> however, the communication channel is implementation defined

10:31 <AntonioND> only the message itself is standard

10:31 <AntonioND> and the problem is that there is a range of implementation defined commands

10:32 <poulecaca> I see

10:32 <AntonioND> and there seem to be quite a few of them in gxl

10:32 <AntonioND> and that's just guesswork

10:32 <AntonioND> however, cpu on, cpu off and similar stuff is standard, so it should be enough

10:32 <poulecaca> I fear that gxl and gxbb does not have same implementation defined commands

10:33 <poulecaca> oh if cpu on/off is standard then there is better chances to work

10:34 <poulecaca> I'll definitly give it a try. Thanks again AntonioND

10:36 <AntonioND> I've seen enough of the code of bl31 to know that the registers and everything are mostly the same. however, psci as a whole is different, and that's why I'm not sure if it will work. I wanted to test things yesterday but I was quite busy in preparation for the next few days that I'm going to be busy...

10:37 <AntonioND> taking as base gxbb i've only changed the base address of bl31, the addresses of the shared mem between TF and u-boot

10:37 <AntonioND> but there is quite a bit of code in bl31 that I have to understand before I'm confident in the port