💾 Archived View for gemini.ctrl-c.club › ~phoebos › logs › kisslinux-2022-03-21.txt captured on 2024-05-12 at 16:03:04.
⬅️ Previous capture (2022-04-28)
-=-=-=-=-=-=-
[2022-03-21T00:24:15Z] <noocsharp> hi sad_plan
[2022-03-21T00:27:58Z] <sad_plan> how you doing? C:
[2022-03-21T00:51:32Z] <noocsharp> pretty good, hbu?
[2022-03-21T01:36:33Z] <sad_plan> good good. Im good :D
[2022-03-21T04:06:10Z] <noocsharp> phoebos: did you mention using lilypond at some point or am i thinking of someone else?
[2022-03-21T04:28:56Z] <testuser[m]> Hi
[2022-03-21T04:29:24Z] <testuser[m]> @Guest6647:libera.chat: lol chromium wayland is broken on nvidia
[2022-03-21T04:29:32Z] <testuser[m]> It flickers a lot unless u --disable-gpu
[2022-03-21T04:31:19Z] <testuser[m]> What's wrong with firefox? Everything works properly there with hw accel and better than chromium
[2022-03-21T04:38:58Z] <illiliti_> ipv6
[2022-03-21T04:39:25Z] <testuser[m]> i mean for his use case
[2022-03-21T04:40:02Z] <illiliti_> ah
[2022-03-21T05:20:16Z] <testuser[m]> Guest6647: https://termbin.com/xn1e
[2022-03-21T05:21:19Z] <Guest69> testuser[m]: Got it, thanks
[2022-03-21T05:21:43Z] <Guest69> Oh
[2022-03-21T05:21:53Z] <Guest69> Works
[2022-03-21T05:21:58Z] <Guest69> Hm
[2022-03-21T05:22:05Z] <testuser[m]> show ur exact output
[2022-03-21T05:22:09Z] <testuser[m]> what gpu
[2022-03-21T05:22:53Z] <testuser[m]> https://termbin.com/u6ev
[2022-03-21T05:25:22Z] <Guest69> I'll write later, I'm just in college right now C:
[2022-03-21T05:25:29Z] <testuser[m]> ok
[2022-03-21T07:25:32Z] <testuser[m]> illiliti i ran the sandbox unit tests
[2022-03-21T07:25:55Z] <testuser[m]> musl https://termbin.com/ejjd
[2022-03-21T07:25:56Z] <testuser[m]> glibc https://termbin.com/hzs7
[2022-03-21T07:26:15Z] <testuser[m]> 11 extra failures on musl + 1 timeout
[2022-03-21T07:26:25Z] <testuser[m]> the trap message shows up in dmesg
[2022-03-21T07:26:32Z] <testuser[m]> need to check which exact test causes that
[2022-03-21T07:46:40Z] <illiliti> can you attach gdb into it?
[2022-03-21T07:46:46Z] <illiliti> gdb shall catch that sigtrap
[2022-03-21T07:54:23Z] <testuser[m]> yea it does but idk which exact test to look for, since some tests trap on glibc too
[2022-03-21T07:59:06Z] <testuser[m]> wait
[2022-03-21T07:59:08Z] <testuser[m]> ill write a script
[2022-03-21T08:19:40Z] <testuser[m]> o there are some tests where its supposed to die with a sigtrap so idk how tf to differentiate between an actual trap and an intended one
[2022-03-21T08:19:41Z] <testuser[m]> bruh
[2022-03-21T08:51:44Z] <illiliti> i ran chromium with --disable-seccomp-filter-sandbox and it crashes on medium.com almost immediately
[2022-03-21T08:51:50Z] <illiliti> so the bug is not in seccomp if i understand correctly
[2022-03-21T08:59:59Z] <testuser[m]> bruh there are no extra traps in musl tests, same tests trap on glibc and same on musl (intended)
[2022-03-21T09:00:10Z] <testuser[m]> the extra failing tests on musl are realted to filesystem permission stuff
[2022-03-21T09:00:13Z] <testuser[m]> weird
[2022-03-21T09:00:38Z] <testuser[m]> i used this garbage script https://termbin.com/jpuq
[2022-03-21T09:01:36Z] <testuser[m]> https://termbin.com/op0xx these failures are unique to musl
[2022-03-21T09:01:46Z] <testuser[m]> only the ForkInNewPidNamespace.ExitWithSignal seems sus cuz it gets stuck
[2022-03-21T09:08:11Z] <illiliti> i see
[2022-03-21T09:08:18Z] <illiliti> is there a way to enable verbose logs?
[2022-03-21T09:08:31Z] <illiliti> like what assert fails
[2022-03-21T09:10:25Z] <testuser[m]> yea they use gtest so it tells u what value mismatched
[2022-03-21T09:10:30Z] <testuser[m]> but this test just gets stuck so it doesnt know that
[2022-03-21T09:10:33Z] <testuser[m]> ill have to put some printfs
[2022-03-21T09:10:57Z] <illiliti> can you disable that test?
[2022-03-21T09:11:12Z] <testuser[m]> wdym
[2022-03-21T09:11:15Z] <testuser[m]> im running all of them one by one
[2022-03-21T09:12:39Z] <illiliti> ah
[2022-03-21T09:12:43Z] <illiliti> i thought that test prevents other tests from printing logs
[2022-03-21T09:18:49Z] <testuser[m]> btw that threadpoolforeg thing belongs to glib
[2022-03-21T09:18:54Z] <testuser[m]> the one that keeps segfaulting
[2022-03-21T09:20:02Z] <testuser[m]> hmm no
[2022-03-21T09:42:15Z] <testuser[m]> bruh im trying to build with symbol_level set but the obj files dont seem to have any debug info
[2022-03-21T09:42:28Z] <testuser[m]> ig ill have to pass -g manually
[2022-03-21T09:44:15Z] <illiliti> what does is_debug do?
[2022-03-21T09:45:37Z] <illiliti> also set this: dcheck_always_on = true
[2022-03-21T09:47:20Z] <testuser[m]> non debug build just enables more optimization
[2022-03-21T09:47:25Z] <testuser[m]> nothing with symbols
[2022-03-21T09:49:32Z] <illiliti> btw you should use -ggdb rather than -g
[2022-03-21T09:51:21Z] <illiliti> or even -ggdb3, since chromium is highly polluted with macros
[2022-03-21T09:52:17Z] <testuser[m]> it'll probably take a whole day to build with that
[2022-03-21T09:52:31Z] <illiliti> yeah
[2022-03-21T09:52:37Z] <illiliti> whole week
[2022-03-21T09:58:33Z] <testuser[m]> https://termbin.com/serb i checked the args and symbol_level does seem to be set properly, no idea how tf they're passing it to the compiler..
[2022-03-21T09:58:50Z] <testuser[m]> nothing in ninja file or in /proc/pid/cmdline of gcc
[2022-03-21T10:01:07Z] <illiliti> have you set is_debug=true?
[2022-03-21T10:07:48Z] <testuser[m]> doesn't seem to be, it's not present in alpine's debug build file either
[2022-03-21T10:07:57Z] <testuser[m]> They just set symbol level
[2022-03-21T10:08:01Z] <testuser[m]> Seem to be required*
[2022-03-21T10:10:35Z] <illiliti> blink_symbol_level=2
[2022-03-21T10:11:14Z] <testuser[m]> Yeah
[2022-03-21T10:11:28Z] <illiliti> v8_symbol_level=2
[2022-03-21T10:11:30Z] <testuser[m]> Both that and symbol_level without blink_ prefix
[2022-03-21T10:20:56Z] <illiliti> is_component_build=true
[2022-03-21T10:33:09Z] <testuser[m]> Wat
[2022-03-21T10:55:57Z] <phoebos> noocsharp: not me
[2022-03-21T13:22:08Z] <testuser[m]> http://0x0.st/oNO5.png
[2022-03-21T13:22:20Z] <testuser[m]> illiliti: looks like the debug symbols were enabled after all
[2022-03-21T13:27:03Z] <testuser[m]> Lol the final link took up everything including swap (24gb) and crashed
[2022-03-21T13:51:10Z] <illiliti> mold?
[2022-03-21T13:54:20Z] <testuser[m]> gnoo ld
[2022-03-21T13:54:38Z] <testuser[m]> i dont think ram usage will be reduced unless mold uses mmap or someth
[2022-03-21T13:54:49Z] <testuser[m]> 30gb seems to be enough, been linking for 20 mins now..
[2022-03-21T13:56:08Z] <testuser[m]> oh god i forgot i would have to relink if i tried to fix the bug
[2022-03-21T13:56:21Z] <testuser[m]> u can mix linkers right?
[2022-03-21T13:58:12Z] <illiliti> should be possible if lto/pgo isn't used
[2022-03-21T13:58:23Z] <testuser[m]> oh i forgot its just a collection of obj files in static libs before final link, so it shouldnt matter
[2022-03-21T13:58:33Z] <testuser[m]> ill wait 10 more mins else switch to mold
[2022-03-21T14:05:45Z] <testuser[m]> lol mold finished it in under a minute and didnt even touch swap
[2022-03-21T14:08:43Z] <testuser[m]> the exe is 2gb it'll take some time to upload it
[2022-03-21T14:08:50Z] <testuser[m]> good luck opening it under gdb lol
[2022-03-21T14:21:32Z] <illiliti> 2gb omg
[2022-03-21T14:21:47Z] <testuser[m]> is there a way to disable crashing on this fucking DCHECK
[2022-03-21T14:22:09Z] <testuser[m]> the check is in a public header so it'll need a ton of recompilation to comment it out
[2022-03-21T14:22:27Z] <illiliti> why it crashes?
[2022-03-21T14:23:26Z] <testuser[m]> https://github.com/chromium/chromium/blob/75c2ebf35d3ae28207dfb8f3f679cde9d2800ff5/mojo/public/cpp/bindings/remote.h#L261
[2022-03-21T14:23:27Z] <testuser[m]> god knows
[2022-03-21T14:23:42Z] <illiliti> --disable-dcheck ?
[2022-03-21T14:24:38Z] <testuser[m]> doesnt do anything
[2022-03-21T14:31:39Z] <illiliti> well, you have to recompile then
[2022-03-21T14:32:06Z] <illiliti> but, iirc gdb can import debug symbols from other files
[2022-03-21T14:33:04Z] <illiliti> so you can run gdb chromium_without_dcheck and then import debug symbols from chromium_with_dcheck
[2022-03-21T14:33:30Z] <testuser[m]> https://termbin.com/3z88
[2022-03-21T14:33:44Z] <testuser[m]> bruh
[2022-03-21T14:33:53Z] <testuser[m]> i took a snapshot of this tree, lets see
[2022-03-21T14:35:01Z] <testuser[m]> [584/9304]
[2022-03-21T14:35:02Z] <testuser[m]> no
[2022-03-21T14:35:10Z] <testuser[m]> 1/3rd of the full build
[2022-03-21T14:35:11Z] <testuser[m]> no
[2022-03-21T14:37:40Z] <illiliti_> how many cores do you have?
[2022-03-21T14:38:33Z] <testuser[m]> 6
[2022-03-21T14:40:41Z] <testuser[m]> Tbh our time would be better spent figuring out broken ipv6 in Firefox
[2022-03-21T14:51:08Z] <illiliti_> i agree
[2022-03-21T14:51:23Z] <illiliti_> but let's deal with chromium first
[2022-03-21T14:55:19Z] <testuser[m]> I grepped for DCHECK in sandbox/linux there's barely 5-10 of them so if more dchecks get triggered after rebuild im just gonna rebuild with dcheck disabled
[2022-03-21T15:53:35Z] <testuser[m]> Fuck i could've just ld preloaded the abort function
[2022-03-21T16:13:32Z] <testuser[m]> nvm `#define TRAP_SEQUENCE1_() asm volatile("int3")`
[2022-03-21T16:16:30Z] <testuser[m]> i can still do that lol i forgot the macro is called from a function
[2022-03-21T16:16:44Z] <testuser[m]> im going crazy
[2022-03-21T17:07:52Z] <testuser[m]> Sha256 80e211ba5678c4f3c4a296e5a5816a5f2322a8efa40a3a8628d11090f4767afd
[2022-03-21T17:07:54Z] <testuser[m]> https://bin.git-bruh.duckdns.org/chromium.tar.xz
[2022-03-21T17:08:08Z] <testuser[m]> illiliti: it won't trap for any DCHECK failure now
[2022-03-21T17:09:08Z] <testuser[m]> Btw i saw a tab crash and some dcheck was failing for nodejs mutex stuff
[2022-03-21T19:34:29Z] <rohan> hey guys
[2022-03-21T19:34:40Z] <rohan> does mpv in repo support lua scripts?
[2022-03-21T19:38:16Z] <acheam> if you install lua before mpv
[2022-03-21T19:38:55Z] <acheam> because then the mpv build will detect lua is installed and build it with lua suppost
[2022-03-21T19:40:12Z] <Guest69> Hi all again
[2022-03-21T19:40:32Z] <Guest69> acheam: I understand that my mpv does not open videos via youtube-dl, could this be because\ I did not install lua before building mpv?
[2022-03-21T19:45:09Z] <rohan> to only use youtubedl just install luajit in community
[2022-03-21T19:45:24Z] <rohan> and rebuild mpv
[2022-03-21T19:47:11Z] <Guest69> Thanks
[2022-03-21T20:07:45Z] <rohan> acheam: just luajit do the work
[2022-03-21T20:07:57Z] <rohan> i put the scripts in the wrong location
[2022-03-21T20:08:57Z] <acheam> Guest69: yes, as rohan said
[2022-03-21T20:09:10Z] <acheam> rohan: not sure what you mean by "just luajit do the work"
[2022-03-21T20:09:25Z] <rohan> you said to install lua
[2022-03-21T20:09:33Z] <rohan> community have lua and luajit
[2022-03-21T20:10:01Z] <rohan> just using luajit can make the scripts run
[2022-03-21T20:10:11Z] <acheam> ah
[2022-03-21T20:10:22Z] <acheam> mpv works with lua 5.1, 5.2, and luajit
[2022-03-21T20:10:50Z] <acheam> idk what version is in community, I used my own lua5.2 package
[2022-03-21T20:29:49Z] <Guest69> acheam: ok, i'll try it