summary refs log tree commit diff stats
path: root/results/classifier/zero-shot/118/graphic/1717
blob: 9bf8f5331994c8bd70be1c65466369f81e08cea1 (plain) (blame)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
graphic: 0.973
x86: 0.973
kernel: 0.961
device: 0.922
architecture: 0.903
debug: 0.896
PID: 0.880
peripherals: 0.833
semantic: 0.825
performance: 0.801
virtual: 0.694
permissions: 0.676
mistranslation: 0.674
vnc: 0.635
hypervisor: 0.608
user-level: 0.568
register: 0.563
risc-v: 0.560
VMM: 0.547
ppc: 0.492
boot: 0.468
i386: 0.409
network: 0.388
socket: 0.386
KVM: 0.378
TCG: 0.376
assembly: 0.309
files: 0.287
arm: 0.281

GPU passthrough (NV h100)case vfio Error
Description of problem:
GPU passthrough (NV h100) will case a error 


qemu-system-x86_64: vfio_err_notifier_handler(0000:17:00.0) Unrecoverable error detected. Please collect any data possible and then kill the guest


this error happen in centos, redhat linux,ubuntu with some kernel i have try( 5.19.0,6.0,6.2)
The same server insert L4,L40 GPU, will not happen. Only happen on H100 GPU
The same server install esxios. everything is normal. GPU work fine

With vfio error. there is some idrac log error on my dell server

```
A bus fatal error was detected on a component at slot 2.	Tue Jun 20 2023 05:51:51
A fatal error was detected on a component at bus 23 device 0 function 0.	Tue Jun 20 2023 05:51:51
A fatal error was detected on a component at bus 22 device 2 function 0.	Tue Jun 20 2023 05:51:51
```

Otherwise, I have try to passthrough gpu on dell amd and intel server both. 
With AMD CPU , gpu not working in vm. but will not case vfio error
With INTEL CPU, will case vfio error.
Steps to reproduce:
1. Set GPU passthrought
2. Start VM
3. Do something in vm
Additional information: