!hzgkCxLtCOTmFXGauj:nixos.org

NixOS Gaming

579 Members
Gaming things, my hands are typing words.156 Servers

Load older messages


SenderMessageTime
19 Jun 2026
@elvishjerricco:matrix.orgElvishJerricco so you can definitely just gain CAP_SYS_NICE 16:21:28
@elvishjerricco:matrix.orgElvishJerricco but for that to be useful, the kernel has to have some internal logic about things your userns owns that CAP_SYS_NICE is allowed to operate on 16:21:57
@elvishjerricco:matrix.orgElvishJerriccoIIUC it's pretty normal for linux caps to have no such logic and just reduce to "after scoping back to the init namespace, what cap remains?"16:22:51
@elvishjerricco:matrix.orgElvishJerricco(oh also mounting additionally has the constraint that you can only make mounts for allowed file systems in a non-init-userns, which currently only includes things like tmpfs and overlayfs)16:24:16
@elvishjerricco:matrix.orgElvishJerricco* (oh also mounting additionally has the constraint that a non-init-uersns can only make mounts for allowed file systems, which currently only includes things like tmpfs and overlayfs)16:24:59
@magic_rb:matrix.redalder.orgmagic_rbJfc this is complicated, but a patch for cap_sys_nice could then be made, if upstream wanted it and i knew how right16:27:27
@elvishjerricco:matrix.orgElvishJerricco you'd have to define (or maybe find documentation on how it's defined) how CAP_SYS_NICE plays together with userns. Like what does the userns own that CAP_SYS_NICE can operate on, because that criteria is how you make it safe 16:28:43
@magic_rb:matrix.redalder.orgmagic_rbI mean id guess it would be "the userns must have created its own pid namespace. Any pid originating in that namespace is fair game. But obviously i know jack shit about this. Ill look at the rtkit way. Doesnt seem that hard16:30:17
@magic_rb:matrix.redalder.orgmagic_rbIt would be nice to have in general and probably required on the frame. Otherwise we'll have frame timing issued16:30:39
@elvishjerricco:matrix.orgElvishJerricco yea I'm only explaining my knowledge of userns and caps in general, I have absolutely no clue about this RT / NICE stuff :P 16:31:00
@magic_rb:matrix.redalder.orgmagic_rbYeah same, probably less than you :P16:31:31
@elvishjerricco:matrix.orgElvishJerricco

oh, this reminded me of something fun:

touch foo
chmod 0400 foo
echo fails > foo # Permission denied
echo works | unshare -c --keep-caps tee foo

You can just write to readonly files unprivileged because you have CAP_DAC_OVERRIDE :)

16:48:35
@elvishjerricco:matrix.orgElvishJerricco (I'm pretty sure the reason this is allowed is because the owner of the userns would have been allowed to just chmod the file back to writable, but it still feels cursed) 16:49:11
@magic_rb:matrix.redalder.orgmagic_rbOh lmao16:55:50
@atemu12:matrix.orgAtemu Oooh wait, that's a neat trick! This might solve an annoyance we have in #Robotnix in that we need to patch calls to cp in the AOSP build system because it defaults to copying permissions of the sources – which are in the nix store of course – and sometimes those files are meant written to somehow. If we could give the processes DAC_OVERRIDE, it might just make those writes work transparently! 17:20:03
@atemu12:matrix.orgAtemu Oooh wait, that's a neat trick! This might solve an annoyance we have in #Robotnix in that we need to patch calls to cp in the AOSP build system because it defaults to copying permissions of the sources – which are in the nix store of course – and sometimes those files are meant written to be written to for godknows what reason. If we could give the processes DAC_OVERRIDE, it might just make those writes work transparently! 17:20:34
@atemu12:matrix.orgAtemu See my reply above; Monado is also an option and is something you can realistically actually use productively these days. Its performance is much superior to SteamVR's vrcompositor IME and having a socket-activated OXR runtime where you don't have to faff with GUIs is really nice.

Note that you only need to patch AMDGPU, which is a module. Much cheaper to build than a full kernel.
17:39:25
@atemu12:matrix.orgAtemu *

See my reply above; Monado is also an option and is something you can realistically actually use productively these days. Its performance is much superior to SteamVR's vrcompositor IME and having a socket-activated OXR runtime where you don't have to faff with a shitty proprietary GUI app that breaks every few weeks is really nice.

Note that you only need to patch AMDGPU, which is a module. Much cheaper to build than a full kernel.

17:42:44
@eyjhb:eyjhb.dkeyJhb Atemu: how do you stream to your headset? Ie. I use ALVR + SteamVR, but with Monado, how would a basic setup for HLA look? 18:17:00
@eyjhb:eyjhb.dkeyJhb(Half-Life Alyx)18:17:05
@atemu12:matrix.orgAtemuALVR reportedly works and AFAUI it actually uses Monado but I don't have any experience with that software as I use an Index. I don't stream and don't see why you'd want to as latency is about the last thing you want in VR. Ig warping makes rotation tolerable insofar as to not make you immediately puke from the delay? Never tried tbh because I'd rather strap a laptop on my back.18:36:42
@marie:marie.cologneMarieI'm waiting for the Frame, not sure how long it will take until it works with non SteamVR stuff18:40:37
@marie:marie.cologneMariei had a quest 3 a while ago, but returned that one. The experience for PCVR was really bad, couldn't get ALVR to not look like shit18:42:34
@marie:marie.cologneMarieand ofc meta trying to make the worst possible experience in general18:42:56
@magic_rb:matrix.redalder.orgmagic_rbLatency is fine with warping and stuff. If i move my head quickly in pcvr i sometimes see blackness in the corner of my eye, but thats like, "flicking" speed. The game itself also was running at only like 40-50fps (cyberpunk2077)19:13:00
@soispha:vhack.euBenedikt joined the room.19:50:57

There are no newer messages yet.


Back to Room ListRoom Version: 10