Hello,
Second attempt (with minor modifications), as the
first one did not arrive.
I would like to share with you a QEMU problem I've
encountered,
and hope to get pointers where to look for it.
I am running CVS QEMU, x86_64-softmmu, with vmdk image
and redirected monitor
(unix socket).
If the guest is busy and I rapidly send a LOT of
commands to QEMU monitor,
then eventually QEMU temporarily hangs (vnc and
monitor are not responsive)
At that state:
- QEMU process's cpu usage is around 100%
- strace shows ONLY clock_gettime(CLOCK_MONOTONIC,
{time changing}) = 0
- pstack shows:
#0 0x0000003603e0459b in clock_gettime () from
/lib64/librt.so.1
#1 0x0000000000405c9d in get_clock ()
#2 0x0000000000405d65 in cpu_get_clock ()
#3 0x000000000044f660 in pit_get_count ()
#4 0x000000000044f805 in pit_latch_count ()
#5 0x0000000002450ec8 in code_gen_buffer ()
#6 0xbe26f6118053c79f in ?? ()
#7 0x000000000047df6e in cpu_x86_exec ()
#8 0x000000000040c11a in main_loop ()
#9 0x000000000040da69 in main ()
After a while (few or more minutes), it comes back to
life.
I can shorten my waiting time by sending a SIGUSR2
signal to QEMU (SIGIO and SIGALRM work too).
Why doesn't host_alarm_handler() get called/bring QEMU
back to life ?
Any comment would be appreciated,
Thanks,
Uri.
____________________________________________________________________________________
Finding fabulous fares is fun.
Let Yahoo! FareChase search your favorite travel sites to find flight and hotel
bargains.
http://farechase.yahoo.com/promo-generic-14795097
