summary refs log tree commit diff stats
path: root/results/classifier/zero-shot/118/graphic/2073
blob: c71b0f7ab86daa788582b1fb50d5af9b50965159 (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
graphic: 0.885
permissions: 0.821
device: 0.809
semantic: 0.544
register: 0.522
performance: 0.459
kernel: 0.423
mistranslation: 0.423
PID: 0.393
ppc: 0.387
vnc: 0.381
arm: 0.357
VMM: 0.335
user-level: 0.329
boot: 0.323
x86: 0.299
i386: 0.296
architecture: 0.277
socket: 0.268
KVM: 0.266
risc-v: 0.266
debug: 0.246
network: 0.237
TCG: 0.188
virtual: 0.188
peripherals: 0.172
assembly: 0.168
hypervisor: 0.124
files: 0.100

Audio: missing ability to disable microphone input from host?
Description of problem:
**It appears there is no way to disable the microphone / input to the audio backend device(s).**


There are at least two cases where this matters:
1. The host has no microphone input (e.g. only HDMI audio output with video).
2. The host has a microphone input, but the user doesn't want the guest VM to have access to the microphone/input.

I tried the option in.channels=0, as that seemed the most obvious way, though that doesn't work.

For -audio dsound, it appears that CLSID_DirectSoundCapture is unconditionally acquired.

There will also be later periodic warning/text outputs from QEMU "Could not create a backend for voice virtio.in", if you're running on a host system with no audio input device.

Adding a couple backend checks for channels > 0 may work well.  Not sure if it matters that audio front end device in the VM still thinks there is an audio input.