qemu coredumps on RELENG_7
Oliver Fromme
olli at lurza.secnetix.de
Mon Mar 17 08:14:19 UTC 2008
Hi,
I'm trying to use the 7.0-RELEASE (i386) ISO in qemu
on a RELENG_7 host (i386, too) from March 5th.
It always coredumps with "invalid system call"
shortly before starting sysinstall, unless I boot
with "safe mode" from the loader menu _or_ without
a vortual harddisk attached to qemu.
These are the last kernel messages:
Timecounter "TSC" frequency 1597573877 Hz quality 800
Timecounters tick every 1.000 msec
hptrr: no controller detected.
md0: Preloaded image </boot/mfsroot> 4423680 bytes at 0xc0d06518
acd0: CDROM <QEMU CD-ROM/0.91> at ata1-master PIO3
ad3: 2048MB <QEMU HARDDISK 0.9.1> at ata1-slave WDMA2
GEOM_LABEL: Label for provider acd0 is iso9660/FreeBSD_bootonly.
Sometimes the last line is the "Timecounter" one,
and sometimes it crashes even in the middle of
printing a line, e.g. "md0: Preloade" is the last
thing printed.
I run qemu like this:
qemu -m 256 -hdd hda.img -cdrom 7.0-RELEASE-i386-bootonly.iso -boot d
where hda.img is a 2 GB empty file created with dd(1)
from /dev/zero. I do _not_ use the kqemu module.
I've compiled qemu without CDROM DMA support (i.e.
with WITHOUT_CDROM_DMA) which supposedly causes
problems sometimes.
Interestingly, the crash does _not_ happen when I omit
the -hd* option. Then sysinstall will come up fine.
These are the next messages in that case:
acd0: CDROM <QEMU CD-ROM/0.91> at ata1-master PIO3
GEOM_LABEL: Label for provider acd0 is iso9660/FreeBSD_bootonly.
Trying to mount root from ufs:/dev/md0
/stand/sysinstall running as init on vty0
acpi0_check: nexus attached
But of course I cannot install because there is no
virtual harddisk. It doesn't matter whether I use
-hda or -hdd ... As soon as there is a disk, qemu
crashes as described above. Also it doesn't matter
whether ACPI in qemu is enabled or disabled.
However, when I select "safe mode" from the loader
menu, then the crash does not happen.
My questions:
1. Is that a known problem?
2. Can anybody else reproduce that problem?
3. Is there a quick workaround, other than "safe mode"?
Thank you very much!
Best regards
Oliver
--
Oliver Fromme, secnetix GmbH & Co. KG, Marktplatz 29, 85567 Grafing b. M.
Handelsregister: Registergericht Muenchen, HRA 74606, Geschäftsfuehrung:
secnetix Verwaltungsgesellsch. mbH, Handelsregister: Registergericht Mün-
chen, HRB 125758, Geschäftsführer: Maik Bachmann, Olaf Erb, Ralf Gebhart
FreeBSD-Dienstleistungen, -Produkte und mehr: http://www.secnetix.de/bsd
"One of the main causes of the fall of the Roman Empire was that,
lacking zero, they had no way to indicate successful termination
of their C programs."
-- Robert Firth
More information about the freebsd-emulation
mailing list