💾 Archived View for gemini.ctrl-c.club › ~phoebos › logs › kisslinux-2022-05-13.txt captured on 2023-01-29 at 03:47:53.
⬅️ Previous capture (2022-06-04)
-=-=-=-=-=-=-
[2022-05-13T01:24:55Z] <noocsharp> didn't they only publish the kernel modules?
[2022-05-13T01:25:08Z] <noocsharp> i think userspace is still proprietary
[2022-05-13T01:30:01Z] <illiliti> seems so
[2022-05-13T01:37:18Z] <illiliti> well, it looks pretty useless then
[2022-05-13T01:37:34Z] <illiliti> why do i need bike without wheels lol
[2022-05-13T01:40:02Z] <noocsharp> a step forward is a step forward
[2022-05-13T01:53:14Z] <noocsharp> in theory mesa can target it once the interface is stabilized
[2022-05-13T02:39:25Z] <acheam> testuser[m]: kavin.rocks is crimeflared btw
[2022-05-13T02:53:03Z] <acheam> phoebos: i'm honored to be the first bookmark on your website
[2022-05-13T02:53:10Z] <acheam> "unordered" be dammed
[2022-05-13T03:19:43Z] <noocsharp> sotd: https://www.youtube.com/watch?v=F8dKVbP1Nzo
[2022-05-13T06:33:18Z] <testuser[m]> illiliti: noocsharp it'll be forever useless cuz it uses hardware stuff introduced in the 30xx series
[2022-05-13T06:34:00Z] <testuser[m]> acheam: ik but it really doesn't matter tbh
[2022-05-13T06:34:06Z] <testuser[m]> Hi
[2022-05-13T06:39:16Z] <testuser[m]> https://github.com/NVIDIA/open-gpu-kernel-modules/issues/19#issuecomment-1124360887
[2022-05-13T06:45:31Z] <testuser[m]> And, to be clear, while we do share a lot of the code here with the proprietary driver, this is a significantly different driver architecture. There are currently no plans to either drop support, or to open source the proprietary driver which supports Maxwell and later
[2022-05-13T06:45:32Z] <testuser[m]> So bad
[2022-05-13T06:46:47Z] <testuser[m]> Anyway the userspace being opem source is more important cuz the proprietary kernel driver works on musl just fine
[2022-05-13T06:46:47Z] <illiliti> i wonder what nouveau devs think about this driver
[2022-05-13T06:47:02Z] <testuser[m]> U can use nvidia-smi with gcompat
[2022-05-13T06:47:24Z] <testuser[m]> I was able to run sway too with gcompat hack but it stopped working a few driver versions later
[2022-05-13T06:51:34Z] <illiliti> > Will the source for user-mode drivers such as CUDA be published?
[2022-05-13T06:51:40Z] <illiliti> > These changes are for the kernel modules, while the user-mode components are untouched. The user-mode remains closed source and is published with prebuilt binaries in the driver and the CUDA toolkit.
[2022-05-13T06:51:53Z] <illiliti> garbage
[2022-05-13T06:53:37Z] <illiliti> it's yet another gpl condom guys
[2022-05-13T07:12:14Z] <testuser[m]> https://nitter.kavin.rocks/marcan42/status/1524615058688724992
[2022-05-13T07:44:41Z] <testuser[m]> https://github.com/advisories/GHSA-h8c3-8522-vxc6
[2022-05-13T07:45:02Z] <testuser[m]> How does this work
[2022-05-13T07:45:21Z] <testuser[m]> U can reset terminal and do whatever u can with escape sequences but how arbitrary code execution
[2022-05-13T08:42:55Z] <illiliti> very old and buggy terminals allow injection commands into the terminal input buffer
[2022-05-13T08:43:28Z] <testuser[m]> lol
[2022-05-13T08:45:23Z] <illiliti> today terminals sanitize input, so in order to achieve rce user interaction is needed
[2022-05-13T08:46:07Z] <illiliti> unless attacker found a way to bypass sanitizer
[2022-05-13T08:48:29Z] <illiliti> you can do many things with escape sequences without needing to execute commands
[2022-05-13T08:48:56Z] <illiliti> for example, you can stole clipboard via OSC 52
[2022-05-13T08:52:41Z] <illiliti> not directly related to this but don't copy and paste from internet to terminal: https://vid.puffyan.us/watch?v=LFXZqQL4vTY
[2022-05-13T09:58:31Z] <phoebos> acheam: hehe
[2022-05-13T09:58:47Z] <phoebos> it's unordered, it's a <ul>!
[2022-05-13T13:01:51Z] <testuser[m]> illiliti: is there an alternative to pipefail for posix sh
[2022-05-13T13:09:42Z] <illiliti> yes, wrap commands in {} and check their status
[2022-05-13T13:10:39Z] <illiliti> also pipefail will be available in the next posix issue
[2022-05-13T13:11:25Z] <testuser[m]> illiliti: can u send example
[2022-05-13T13:11:30Z] <testuser[m]> wouldnt checking status discard output
[2022-05-13T13:25:28Z] <illiliti> what are you trying to achieve?
[2022-05-13T13:25:50Z] <illiliti> it depends
[2022-05-13T13:27:26Z] <testuser[m]> jq -r .[].file compile_commands.json | while read -r file; do echo $file; done
[2022-05-13T13:39:52Z] <illiliti> i don't see the need for pipefail here
[2022-05-13T13:42:18Z] <testuser[m]> illiliti: https://termbin.com/wvbu
[2022-05-13T13:42:22Z] <testuser[m]> i want the shell to exit
[2022-05-13T13:51:43Z] <illiliti> var=$(jq ..) || exit; for file ....
[2022-05-13T13:51:52Z] <illiliti> or heredoc
[2022-05-13T13:56:13Z] <illiliti> ^ this is much better than trying to emulate pipefail
[2022-05-13T13:57:12Z] <testuser[m]> how can u use a heredoc here
[2022-05-13T14:05:24Z] <illiliti> set -e; while ... done <<EOF\n$(...)\nEOF\n ....
[2022-05-13T14:51:28Z] <phoebos> testuser[m]: { jq -r .[].file "$1" || kill 0; } | while ....
[2022-05-13T14:53:08Z] <testuser[m]> o i tried that but was not sure how to kill the shell that the sub command ran in
[2022-05-13T14:53:08Z] <testuser[m]> thanks
[2022-05-13T14:54:25Z] <phoebos> yeah kill 0 is like kill $ i think
[2022-05-13T14:54:41Z] <illiliti> hacky hack, not needed here
[2022-05-13T14:55:41Z] <phoebos> illiliti: how can you do it without a subshell
[2022-05-13T14:55:53Z] <testuser[m]> illiliti: btw var=$(...) makes an extra copy right
[2022-05-13T14:56:01Z] <testuser[m]> but piping directly feeds stdout into program
[2022-05-13T14:56:06Z] <phoebos> ^
[2022-05-13T14:58:21Z] <illiliti> | makes extra syscalls
[2022-05-13T14:59:17Z] <phoebos> pipe? rather than a new shell?
[2022-05-13T14:59:21Z] <phoebos> *sub
[2022-05-13T15:03:11Z] <phoebos> actually pipes create subshells too
[2022-05-13T15:03:38Z] <testuser[m]> ye
[2022-05-13T15:09:19Z] <phoebos> ig var=$(jq...) || exit 1; while ... done <<EOF\n$var\nEOF\n is best
[2022-05-13T15:09:47Z] <phoebos> or while ... done << EOF\n$(jq ... || kill 0)\nEOF\n