15:53 <sjg1> Tartarus: I got alex running, but it fails on xtensa since there is no toolchain. What should we do about that?

15:54 <Tartarus> sjg1: yes, I sorted that out in my series. I'll post it next week since it looks like apalos is likely too busy to dig in to what he was telling me is probably a linker issue wrt sandbox + arm64 hosts, and just exclude sandbox there still.

15:58 <apalos> Tartarus: I think it is, I'll be able to look at it next week

15:58 <apalos> IS it too painful to enable the arm64 for !sandbox till I fix it ?

15:58 <apalos> I mean the runners

15:58 <Tartarus> No, it's not bad really, I was just hoping to show it working there too since all the sandbox jobs are the big slow down of that stage

15:58 <Tartarus> So more runners is better

15:59 <apalos> ok do than then if it's not a pain, and I'll have a look once I can,

15:59 <apalos> I sent some patches to make lmb a bit easier to read

15:59 <apalos> But figured out i was gonna end up with too many patches, so i sent a cleanup for now and the rest can wait

16:00 <apalos> xypron already sent some correct feedback, I'll wait a few days in case people care to comment and i'll send a non rfc version

16:02 <Tartarus> OK. And I've fired off a rebase of things now, I'll post it either today or tomorrow

16:03 <apalos> great, thanks

16:05 <Tartarus> And ah yes, the only thing I wasn't sure about is if we should have "arm64" and "fast arm64" tags or just "arm64" tags, as if we allow not-fast arm64 runners, we can get more going, both things like Oracle "free" tier hosts I think (still waiting on mine) to maybe even dedicating current gen ref platforms / etc as they're good enough for running qemu and the other non-big build jobs

16:06 <Tartarus> And if we had 2 dozen of those, that stage of the pipeline would take only as long as the sandbox jobs do

16:18 <apalos> well you can always add the current as 'fast'

16:18 <apalos> and the worst thing that happens is that we only use 'fast'

16:18 <Tartarus> Yeah, it's just a little rework now before I post, or just doing a follow up later

16:19 <Tartarus> And since docker is now being annoying, I'm rebuilding all of the containers and will kick the jobs again in 20min or soonce that's done

16:19 <apalos> ok, I might be able to take a look at the sandbox error later tonight

16:19 <apalos> as I am just over xypron's comments on that RFC for lmb

22:25 <shadow> I think Hal's (v8) JH7110 OF_UPSTREAM series is ready for merge to -next ; is there a delegate that rubber stamps it or does Tom pick it up when it looks like there's consensus?

22:35 <shadow> oh patchwork shows Andes as the delegate.