1 |
W dniu śro, 09.05.2018 o godzinie 09∶56 +0200, użytkownik Dennis |
2 |
Schridde napisał: |
3 |
> On Wednesday, 9 May 2018 09:34:43 CEST Michał Górny wrote: |
4 |
> > W dniu śro, 09.05.2018 o godzinie 08∶51 +0200, użytkownik Dennis |
5 |
> > |
6 |
> > Schridde napisał: |
7 |
> > > I see sandbox violations similar to "ACCESS DENIED: open_wr: /dev/dri/ |
8 |
> > > renderD128" pop up for more and more packages, probably since OpenCL |
9 |
> > > becomes used more widely. Hence I would like to ask: Could we in Gentoo |
10 |
> > > treat GPUs just like CPUs and allow any process to access render nodes |
11 |
> > > (i.e. the GPUs compute capabilities via the specific interface the Linux |
12 |
> > > kernel's DRM offers for that purpose) without sandbox restrictions? |
13 |
> > > |
14 |
> > > See-Also: https://bugs.gentoo.org/654216 |
15 |
> > |
16 |
> > Doesn't accessing those nodes involve a risk of programs being able to |
17 |
> > crash your system (or xorg)? Or cause on-screen artifacts? |
18 |
> |
19 |
> Well, in the presence of Linux kernel bugs, programs could crash the whole |
20 |
> system. But I believe this is also true for all other drivers and compute |
21 |
> devices, including CPUs. |
22 |
> |
23 |
> I assume an application using render nodes could crash X by e.g. consuming all |
24 |
> memory. But then this is also true for all applications using the CPU and its |
25 |
> attached memory for computations. |
26 |
> |
27 |
> On-screen artifacts as in resolution switching and other KMS operations is |
28 |
> explicitly prohibited. Insecure access (using GEM FLINK) to the memory of |
29 |
> other applications (which could cause broken textures / windows with broken |
30 |
> content) is also explicitly prohibited. So my understanding is that the |
31 |
> answer is: No, using render nodes cannot cause on-screen artifacts (modulo the |
32 |
> presence of Linux kernel bugs, s.a.). |
33 |
> |
34 |
> DRM render nodes were specifically introduced to allow GPGPU applications to |
35 |
> run without impacting the security of the system (and without X). |
36 |
> |
37 |
|
38 |
Ok then. I suppose I was confusing it with /dev/nvidia*. |
39 |
|
40 |
-- |
41 |
Best regards, |
42 |
Michał Górny |