iooi has quit [Read error: Connection reset by peer]
iooi has joined #fedora-riscv
iooi has quit [Read error: Connection reset by peer]
iooi has joined #fedora-riscv
iooi has quit [Read error: Connection reset by peer]
iooi has joined #fedora-riscv
iooi has quit [Ping timeout: 248 seconds]
iooi has joined #fedora-riscv
iooi has quit [Read error: Connection reset by peer]
iooi has joined #fedora-riscv
iooi has quit [Read error: Connection reset by peer]
iooi has joined #fedora-riscv
King_InuYasha has quit [Read error: Connection reset by peer]
King_InuYasha has joined #fedora-riscv
jcajka has joined #fedora-riscv
esv has quit [Remote host closed the connection]
esv has joined #fedora-riscv
<rwmjones>
yay vision five delivered
<davidlt[m]>
Cool
omac777_2022 has joined #fedora-riscv
iooi has quit [Read error: Connection reset by peer]
iooi has joined #fedora-riscv
davidlt has joined #fedora-riscv
jcajka has quit [Quit: Leaving]
<davidlt[m]>
Long, long time ago.. GCC 11 used to compile in ~60 hours on physical hardware :)
<davidlt[m]>
GCC 13 will probably be 160 hours if not more :)
<davidlt[m]>
nirik: something happened and GCC 13 build on your board got restarted (1-2 days ago).
<davidlt[m]>
Not sure if that's some glitch with Koji, or board got restarted, or something else.
<nirik[m]>
we lost power and I think I might not have put that box on UPS. ;( Will go check it later today. /me adds to todo list.
<nirik[m]>
likely will have to power it off for a sec to move it.
<davidlt[m]>
I am running another two builds (at rwmjones and djdelorie places). Those are all good so far.
unlord has quit [Ping timeout: 268 seconds]
unlord has joined #fedora-riscv
<omac777_2022>
@davidlt do you want other vf2 board for the gcc?
<davidlt[m]>
omac777_2022: nah, we aren't ready to build disk images for that board yet, and I don't really want to hack one (yet).
<davidlt[m]>
I want to see more things getting into a proper shape to get them backported to cook an image.
<omac777_2022>
ok
<davidlt[m]>
Anyways VF2 might not even have enough RAM for GCC compile (TBD once we test that in the future).
<omac777_2022>
ok.
<davidlt[m]>
We just need to get that initial GCC 13 build out to unblock the rest.
<davidlt[m]>
We could fire up 32-core VM. I just used it some days ago to verify that everything is OK now to build, but I don't want to use QEMU too.
<davidlt[m]>
A bit of a pain, but it's temporary thing.
<omac777_2022>
ok
<omac777_2022>
what about a distcc on each vf2 that everybody volunteers, would that help or work?
<davidlt[m]>
I cannot wait to see TH1520 boards from Sipeed too. Those should be way faster than JH7110 (TBD).
<davidlt[m]>
and gives us 16GiB of RAM!
<davidlt[m]>
Nah, I don't want do use that. I want a clean proper Fedora build.
<davidlt[m]>
I do plan to use VF2 later on for build farm. It's good for majority of builds.
<omac777_2022>
Are you saying a clean 160 hours build on one machine rather than a 1 hour build delegated to a bunch of vf2 distcc nodes?
<davidlt[m]>
Yes.
<omac777_2022>
Thanks.
<davidlt[m]>
Unless nirik states that distcc in Koji infra is something that is officially blessed by infra / engineering folks I don't plan to use it.
<davidlt[m]>
Not to mention I don't trust VF2 kernel stuff until it's close to landing in upstream.
<davidlt[m]>
QEMU, and distcc is all great for bootstrap, but afterwards we still need to rebuild it.