Welcome to the main channel on the development of MoarVM, a virtual machine for NQP and Rakudo (moarvm.org). This channel is being logged for historical purposes. Set by lizmat on 24 May 2021. |
|||
01:13
guifa left
01:26
guifa joined
02:09
guifa left
08:54
sena_kun joined
|
|||
patrickb | Did the pre release revert get reapplied again? | 10:02 | |
tellable6 | 2025-01-26T01:40:03Z #raku <[Coke]> patrickb 2025.01 releases are on github | ||
10:17
sena_kun left
|
|||
lizmat | yes | 10:19 | |
I cherry-picked the original commit | |||
patrickb | 👍🏼 | 10:27 | |
lizmat | and yet another Rakudo Weekly News hits the Net: rakudoweekly.blog/2025/01/27/2025-04-the-first/ | 14:08 | |
looks like El_Che has problems building the 2025.01 release: | 16:17 | ||
github.com/nxadm/rakudo-pkg/action...#step:3:75 | |||
anything wrong with this? github.com/nxadm/rakudo-pkg/blob/m...oad.sh#L11 | 16:20 | ||
patrickb [Coke] ^^ | 16:23 | ||
ugexe | 3rdparty/libuv/src/unix/udp.c:1351:10: error: ISO C90 forbids array 'addrs' whose size cannot be evaluated [-Werror=vla] | 16:30 | |
vla was disabled in a recentish commit | |||
Geth | MoarVM/coke/md: 2debb76b97 | (Will Coleda)++ | 18 files Standardize markdown file extension |
16:32 | |
MoarVM: coke++ created pull request #1904: Standardize markdown file extension |
|||
lizmat | ugexe: yeah, but that was post 2025.01 | 16:33 | |
[Coke] | there was some wonkiness with the moarvm tarballs | 16:45 | |
but everyone who tested was able to extract them. | |||
.. it's dying on the git clone?? | |||
ah, no it is dying on that - so the issue maybe that we bumped the 3rdparty code but didn't apply the error relaxing | 16:47 | ||
? | |||
ugexe | those are in the same PR/commit | ||
"Turn off vla errors and bump libuv" | |||
lizmat | and both were *post* 2025.01 | ||
[Coke] | ... so not sure where the errors in the nxadm link above are coming from then. | 16:48 | |
let me back up - what's the error with the 2025.01 build? | |||
lizmat | git checkout 2025.01 | 16:49 | |
warning: unable to rmdir '3rdparty/rapidhash': Directory not empty | |||
Note: switching to '2025.01'. | |||
github.com/nxadm/rakudo-pkg/action...#step:3:75 | |||
[Coke] | do then need a fresh checkout because of submodule cruft in main? | ||
s/checkout/clone | |||
ugexe | the error is libuv + vla | 16:50 | |
[Coke] | that's a warning and there's no failure in that page, though? | 16:51 | |
ugexe | the libuv update shouldn't be in the release, but even if it is then vla should be disabled in the same update | ||
which i dont understand | |||
the warning probably isnt a big deal since the actual error is not referencing it | |||
lizmat | well, El_Che is currently not well, so I guess we'll wait until they can give more feedback | 16:52 | |
ugexe | hmm | ||
[Coke] | on the page with the error, I see 'tar xzf moarvm.tar.gz ' - where is that tgz coming from, is that the problem? | ||
ugexe | the rapidhash thing was added after the release as well i think, and that warning references that | ||
[Coke] | Yup, confusing. Hopefully worst case we can do a .1 tarball if needed. | 16:53 | |
but it really reads like he's building HEAD? | |||
lizmat | but even building HEAD should be ok ? | 16:54 | |
maybe this is just a long standing issue | |||
before, there had not been any commits to MoarVM to build these packages maybe ? | |||
so the fact that HEAD was being built, wasn't notice | |||
d | |||
[Coke] | oh, that would be... bad | 16:55 | |
lizmat | yet, otoh, HEAD should also just build | ||
[Coke] | well, let's hope he feels better and we can get to the bottom of it. | ||
ugexe | github.com/nxadm/rakudo-pkg/action...#step:3:94 | 16:57 | |
it is showing libuv/ has modifications | |||
they should probably add a check to make sure their checkout is clean or whatever | 16:59 | ||
[Coke] | +1 | 17:02 | |
18:23
[Coke]_ joined
18:24
[Coke] left
18:51
[Coke]_ is now known as [Coke]
19:17
sena_kun joined
22:23
sena_kun left
23:45
guifa joined
|