aconnect occasionally causes kernel oopses
folkert
folkert at vanheusden.com
Mon Aug 2 17:21:17 CEST 2021
[ kernel bug if repeatingly aconnect'ing midi devices ]
> Does this happen if you do reconnect of kernel sequencer client?
> You can use snd-virmidi as well as snd-dummy.
> I'm asking it because it'll simplify the test a lot, which will be
> almost self-contained.
Like this?
root at lappiemctopface:~# aplaymidi -l
Port Client name Port name
14:0 Midi Through Midi Through Port-0
20:0 Virtual Raw MIDI 1-0 VirMIDI 1-0
21:0 Virtual Raw MIDI 1-1 VirMIDI 1-1
22:0 Virtual Raw MIDI 1-2 VirMIDI 1-2
23:0 Virtual Raw MIDI 1-3 VirMIDI 1-3
128:0 rtpmidi lappiemctopface Network
128:1 rtpmidi lappiemctopface metronoom
128:2 rtpmidi lappiemctopface AppleMidi2IPMidiBridge
128:3 rtpmidi lappiemctopface oensoens
130:0 FLUID Synth (11462) Synth input port (11462:0)
and then:
root at lappiemctopface:~# cat test.sh
while true
do
aconnect 20:0 21:0
aconnect -d 20:0 21:0
done
root at lappiemctopface:~# for i in `seq 0 3` ; do (./test.sh &) ; done
This hard locks-up my laptop: it doesn't even respond to capslock (led
on/off) anymore nor the ctrl+prtscr+alt+b combination.
More information about the Alsa-devel
mailing list