summary refs log tree commit diff stats
path: root/results/classifier/zero-shot/105/assembly/1772262
blob: 8e29a24b29322f9d34e4343f954e1986cc78420e (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
60
61
62
63
64
65
66
67
assembly: 0.849
device: 0.843
instruction: 0.838
other: 0.835
semantic: 0.830
graphic: 0.825
network: 0.816
boot: 0.804
KVM: 0.795
vnc: 0.784
mistranslation: 0.781
socket: 0.762

Adding -spice doesn't respect environment variable QEMU_AUDIO_DRV

When -spice is added to the commandline, QEMU_AUDIO_DRV=alsa is not respected and I receive no audio from the guest using the alsa driver.  When -spice options are omitted, audio works as usual.

I want to channel mouse and keyboard events to the guest using spice (with looking-glass) instead of a third-party product over a network interface (with synergy), but audio ends up over the spice protocol as well instead of using alsa despite asking otherwise.  For example, one will only hear audio when a program like spicy is running.  Naturally it would be nice if this were not needed, since spice audio is gappier than regular audio on my machine.

Perhaps if -spice added an option similar to agent-mouse=off but for audio, (say audio-pipe=off or so for example,) this might mitigate the issue in a more naive fashion UX-wise, without having to force the issue with environment variables that unfortunately for me don't seem to work. :(

QEMU emulator version 2.12.0

Commandline

        qemu-system-x86_64 \
          -ctrl-grab \
          -enable-kvm \
          -cpu host,hv-time,kvm=off \
          -smp cores=4 \
          -m 8G \
          -M q35 \
          -vga none \
          -netdev tap,id=hostnet,ifname=tap1,script=no,downscript=no \
          -net nic,model=virtio,macaddr=52:54:FD:BF:F7:9A,netdev=hostnet \
          -netdev user,id=usernet,smb=/media/DRIVE-C/tux/vms/share \
          -net nic,model=virtio,macaddr=52:54:2E:40:4F:C8,netdev=usernet \
          -usb \
          -device usb-ehci,id=ehci \
          -device ioh3420,bus=pcie.0,addr=1c,multifunction=on,port=1,chassis=1,id=root.1 \
          -device vfio-pci,host=07:00.0,bus=root.1,addr=00.0,multifunction=on,x-vga=on \
          -device ich9-intel-hda \
          -device hda-micro \
          -drive if=virtio,file=vm/win10/disk.img,media=disk \
          -boot menu=on,splash=splash/boot.jpg,splash-time=5000 \
          -name win10 \
          -device ivshmem-plain,memdev=ivshmem \
          -object memory-backend-file,id=ivshmem,share=on,mem-path=/dev/shm/looking-glass,size=32M \
          -device virtio-serial-pci \
          -device virtserialport,chardev=spicechannel0,name=com.redhat.spice.0 \
          -chardev spicevmc,id=spicechannel0,name=vdagent \
          -spice addr=127.0.0.1,port=5900,disable-ticketing

When the last four options are not present, audio works as expected.  This is with both a windows 10 guest and a windows 7 guest.

Doesn't reproduce.  The *default* driver is changed to spice when spice is enabled, but overriding using QEMU_AUDIO_DRV is not disabled.  Any chance you didn't export QEMU_AUDIO_DRV?

>Any chance you didn't export QEMU_AUDIO_DRV?

Okay, I just checked by calling env in the same fashion.

Well lash me bootlaces, it turns out that sudo needs to be invoked with the -E flag to preserve the environment.  It's funny, the examples I read up on never mentioned this.

Sorry for littering the bug tracker with a silly end-user problem.

Cheers.