00:45
lopsided98_ has joined #nixos-aarch64
00:47
lopsided98 has quit [Ping timeout: 260 seconds]
01:11
tomberek has quit [Ping timeout: 260 seconds]
01:58
tomberek has joined #nixos-aarch64
02:19
lopsided98_ is now known as lopsided98
02:21
lopsided98 is now known as lopsided98_
02:21
lopsided98_ is now known as lopsided98
02:21
lopsided98 is now known as lopsided
05:53
tomberek has quit [Ping timeout: 260 seconds]
09:14
s33se has quit [Ping timeout: 256 seconds]
09:34
orivej has joined #nixos-aarch64
10:50
orivej has quit [Ping timeout: 256 seconds]
10:50
orivej has joined #nixos-aarch64
13:32
<
Mic92 >
the newest rust fails on aarch64: test [debuginfo-gdb] debuginfo/by-value-self-argument-in-trait-impl.rs ... FAILED
13:34
<
Dezgeg >
I wonder if they just added new tests upstream (or is this whole debuginfo thing new in latest rust)
13:34
<
Mic92 >
it is a new debuginfo thing
13:34
<
Mic92 >
the last one was related to our old gdb
13:35
<
Dezgeg >
yeah, let's open a upstream bug and disable that test
13:36
<
Mic92 >
stage is yours. I will earliest look at that in the evening
13:37
<
Dezgeg >
just removing that file disables it without breaking something else?
13:37
<
Dezgeg >
I can give that a shot if so
13:37
<
Mic92 >
this is what we do for other tests.
13:38
<
Mic92 >
I have not seen interdependencies between test modules so far.
13:38
<
Dezgeg >
ok, I have a recollection I disabled some other test via some other way
13:38
<
Dezgeg >
but maybe it was a test file containing multiple tests
13:39
<
Mic92 >
I think we should require too build on all architectures at least for rust.
13:39
<
Mic92 >
Given that we have LnL for macOS and a open-access box for aarch64
13:41
<
Mic92 >
no problem. I will improve test process for future rust upgrades
13:41
<
LnL >
I thought about testing on the aarch64 box but didn't want to wait for a super long build
13:42
<
LnL >
on darwin we just disabled all of the debuginfo tests
13:47
<
Dezgeg >
I start a test build now
13:50
<
LnL >
if we skip those everywhere we could also get rid of the gdb dependency
13:51
<
Mic92 >
LnL: I would like to keep this for linux-x86_64 or at least know the reason why tests fails
13:53
<
Dezgeg >
yeah I conditionalized this for aarch64 for now
14:21
<
LnL >
I thought we also patched a bunch of those tests to make them work
14:21
<
Dezgeg >
this one looks like a legit upstream bug
14:21
<
LnL >
but I don't see anything so maybe they are not as much of a problem
14:33
<
Dezgeg >
yeah, I noticed by googling
14:37
<
bkchr[m] >
I'm also for removing the file, until upstream has some ideas
15:06
orivej has quit [Ping timeout: 260 seconds]
17:42
orivej has joined #nixos-aarch64
21:20
<
shad >
What is the best way to get an aarch64 build (output of sd-image-aarch64.nix)? qemu based aarch64 VM, access to the build box mentioned on the topic, or a public build somewhere (I have yet to find that URL)?
21:23
<
shad >
samueldr: thank you
21:24
<
shad >
I somehow missed it last time I checked this URL
22:27
orivej_ has joined #nixos-aarch64
22:27
orivej has quit [Ping timeout: 268 seconds]
23:21
orivej_ has quit [Ping timeout: 256 seconds]