ChanServ changed the topic of #asahi-alt to: Asahi Linux: porting Linux to Apple Silicon macs | User-contributed/unofficial distribution ports | Logs: https://alx.sh/l/asahi-alt
KxCORP5 has quit [Quit: Bye!]
KxCORP5 has joined #asahi-alt
minecrell has quit [Read error: Connection reset by peer]
minecrell has joined #asahi-alt
possiblemeatball has quit [Quit: Quit]
kdb4247 has joined #asahi-alt
kdb424 has quit [Read error: Connection reset by peer]
kdb4247 is now known as kdb424
JayBeeFOSS has quit [Ping timeout: 480 seconds]
JayBeeFOSS has joined #asahi-alt
tobhe has joined #asahi-alt
tobhe_ has quit [Ping timeout: 480 seconds]
xal has quit []
chadmed_ has quit [Remote host closed the connection]
jeisom has quit [Ping timeout: 480 seconds]
klausvalka has joined #asahi-alt
xal has joined #asahi-alt
klausvalka has quit [Remote host closed the connection]
chadmed_ has joined #asahi-alt
<chadmed_>
leio: whats going to be the best way to deal with mesa and its unstable deps, assuming we get rid of the recommendation to just ACCEPT_KEYWORDS="~arm64"
<chadmed_>
i dont quite fancy having a 100 billion line package.accept_keywords in asahi-gentoosupport
<chadmed_>
hmmm pkgcheck reckons libclc-17.0.6 has no unstable deps
<chadmed_>
maybe we just send it with mesa and libclc
<leio>
chadmed_: there is no problem with filing stablereqs like I just did. Me and sam are on arm64 project as well, though I personally don't have a stable system to handle the stablereqs; I could get access to the devbox, but arm64 is usually the first to stable something anyhow (it is the fastest dev machine used for keywording work iirc)
<leio>
chadmed_: libclc stabilization in ::gentoo is one of the above linked gentoo bugs I filed
<leio>
spirv-llvm-translator was already stabled (dep of libclc)
<leio>
chadmed_: lsp-plugins is stable too now.
<leio>
just waiting on spirv now, would be done already had I put them in the same bug instead of dependent ones (didn't know they are connected before it)
<leio>
err, waiting on libclc
<chadmed_>
yeah noticed those at work thanks for that :)
<chadmed_>
i do want to get rid of the recommendation to add ~arm64 to global keywords asap though because as you said earlier it is really stupid
<chadmed_>
looks like just adding mesa and libclc to package.accept_keywords is good enough
<leio>
yeah, and you can locally stabilize what's needed in overlay and see what pkgcheck complains and file stablereqs
<leio>
if there's anything else
<chadmed_>
mesa and asahi-meta are the only two packages currently not stable
<chadmed_>
the latter of which is only unstable because of a dependency on mesa :)
<leio>
there really is no need for any keywords on arm64 to be behind amd64, it just needs dealing with
<chadmed_>
we'
<leio>
there's a lot of package.use.mask and package.use.stable.mask type of things to sort out as well in gentoo arm64
<chadmed_>
..re still pretty much ahead of almost every other distro in terms of first-class arm64 support
<leio>
that kind of such stuck around since the early days when I did an initial push years ago
<leio>
(on arm64 in general)
<leio>
s/such/just/
<chadmed_>
i got an rpi 4 during the pandemic lockdowns with the intention of helping to get some of the desktop-ish stuff sorted out but it was still too shit to use as a desktop so i put it on a shelf and left it there :p
<chadmed_>
immolo has way too much patience for these slow SBCs
<leio>
apparently it was already back in 2017 we got sponsored access (WorkOnArm stuff via packet.net) to beefy arm64 server and then eventually in 8 January 2018 I split any initial arm64 into a separate project