11:31 <f_> hi

12:12 <tolszak> narmstrong: f_: Hello tobetter provided small snipped to dts to make i2c work. I think it was by hand so I needed to actually modify it a bit but I have i2c working on C4? Do you need that in meta-meson?

12:13 <f_> tolszak: Perhaps you pinged the wrong person :) I'm no expert on ODROID-C4 (in fact, I don't even have the board) and never touched meta-meson

12:13 <f_> But nice that you got it working!

12:14 <f_> (ask tobetter to) send a patch to mainline so that others can benefit from that!

12:17 <tolszak> I pinged you because you took part in the discussion on 16th of August. That's the patch.

12:17 <tolszak> https://dpaste.com//G9UFYH4EN

12:17 <f_> Yeah sure.

12:17 <f_> chewitt also took part in that discussion.

12:17 <tolszak> I think narmstrong: need to decide if it should go in such state to mainline or to be bbappend in meta-meson.

12:18 <tolszak> f_: I know but he is not online. I'm pinging chewitt: too just in case

12:18 <f_> chewitt checks public logs daily.

12:18 <narmstrong> nah it should be in a dtbo we could apply on mainline dtb, but there's no central storage for dtbos....

12:19 <narmstrong> and I didn't add dtbo support in meta-meson, so there's no place to add it either

12:19 <f_> It's always better to send to mainline.

12:19 <tolszak> narmstrong: So: Ask tobetter or ...?

12:19 <narmstrong> well no, because we don't want to force a function on SBCs header pins

12:19 <f_> narmstrong: as a dtbo?

12:19 <narmstrong> tolszak: yeah probably, perhaps they have mainline compatible dtbos

12:19 <tolszak> narmstrong: initial patch had status=disabled

12:20 <narmstrong> yeah but it's dead code, we could have zillions of disable DT nodes, so the right place in in dtbos

12:20 <f_> Yeah I guess.

12:24 <tolszak> Tobetter replied on Odroid forum and that's where I asked him to submit it to mainline. I recall times like 6-7 years ago patching their ubuntu to have anything bootable in reasonable amount of time. Now with Yocto and meta-meson world is a lot simplier :-)

13:12 <f_> Nice, U-Boot SPL linking errors reduced!

13:51 <f_> Oh, by the way, anyone interested in U-Boot SPL supporting booting from USB?

14:00 <f_> I don't think I'll send that to upstream because I don't think it'll get accepted upstream

14:01 <f_> and AFAICT U-Boot SPL doesn't even support getting loaded via USB like this

14:01 <f_> and so, implementing support for that for just one SoC family would be a little pointless tbh

14:02 <f_> For now I made it panic() when it's loaded from USB.

14:07 <narmstrong> f_: until there's a real usage for SPL, like replacing amlogic's bl2, there's no such point

14:08 <narmstrong> but if you manage to make it work, then yeah you can send all your SPL preparation patches asap

14:08 <f_> I'll keep all that in a repo branch `wip/spl-aml` or similar

14:08 <f_> patches for master will go in `for-master/spl-aml`.

14:09 <f_> And I'll also send them with `git send-email`.

14:09 <f_> Honestly I don't think there's too much to do with SPL.

14:10 <f_> But we'll see when I'll test!

14:10 <f_> I don't see any binaries generated by binman though so I think my sleepy brain forgot stuff.

14:22 <f_> narmstrong: hope it will work though :)

14:33 <narmstrong> But at some point I wanted to try and enable SPL to enable falcon boot mode to make a really fast boot to Linux, so it’s not necessarily useless

14:38 <f_> You wanted to, but you didn't do it, right?

14:43 <narmstrong> Yep never did it

14:44 <f_> Probably expect that dream to come true at some point ;)