Hi, I am running a Ubuntu 10.04 x86-64 on an ordinary x86 Nehalem machine. After running the following commands; $ wget http://www.yoctoproject.org/downloads/poky/poky-laverne-4.0.tar.bz2 $ tar xjf poky-laverne-4.0.tar.bz2 $ source poky-laverne-4.0/poky-init-build-env poky-4.0-build $ poky-qemu qemux86 matias@matias-desktop:/local/yocto/poky-4.0-build$ poky-qemu qemux86 Continuing with the following parameters: KERNEL: [/local/yocto/poky-4.0-build/tmp/deploy/images/bzImage-qemux86.bin] ROOTFS: [/local/yocto/poky-4.0-build/tmp/deploy/images/poky-image-sato-qemux86.ext3] FSTYPE: [ext3] Setting up tap interface under sudo Acquiring lockfile for tap0... Warning: distccd not present, no distcc support loaded. Running qemu... /local/yocto/poky-4.0-build/tmp/sysroots/x86_64-linux/usr/bin/qemu -kernel /local/yocto/poky-4.0-build/tmp/deploy/images/bzImage-qemux86.bin -net nic,vlan=0 -net tap,vlan=0,ifname=tap0,script=no,downscript=no -hda /local/yocto/poky-4.0-build/tmp/deploy/images/poky-image-sato-qemux86.ext3 -show-cursor -usb -usbdevice wacom-tablet -vga vmware -enable-gl -no-reboot --append "vga=0 root=/dev/hda mem=128M ip=192.168.7.2::192.168.7.1:255.255.255.0 " Enabling opengl /local/yocto/poky-laverne-4.0/scripts/..//scripts/poky-qemu-internal: line 445: 16569 Segmentation fault $QEMUBIN -kernel $KERNEL $QEMUOPTIONS $SERIALOPTS -no-reboot $SCRIPT_QEMU_OPT --append "$KERNCMDLINE $SCRIPT_KERNEL_OPT" Set 'tap0' nonpersistent Releasing lockfile of preconfigured tap device 'tap0' matias@matias-desktop:/local/yocto/poky-4.0-build$ The qemu windows comes up for a moment but then dies due to the line 445: 16569 Segmentation fault. Regards // Matias
Are you using the proprietary NVIDIA Xorg driver? There is a known issue with the openGL library it ships which can cause this behavior when qemu-native is compiled against it.
This can be seen in the kernel log, Feb 3 20:33:17 matias-desktop kernel: [43914.197990] qemu[18471]: segfault at 329bd4e ip 000000000329bd4e sp 00007fff4a6dfa00 error 15 I've tried to start both qemux86-64 and qemux86 and both fails with the same fault.
(In reply to comment #1) > Are you using the proprietary NVIDIA Xorg driver? There is a known issue with > the openGL library it ships which can cause this behavior when qemu-native is > compiled against it. Yes I am , how can I resolve it?
I'm afraid we don't have a workaround for this yet. I was able to verify the crashes no longer happen when I uninstalled the NVIDIA driver and ensured that the OpenGL libraries no longer existed in /usr/lib/nvidia-current. Then when I rebuilt qemu-native (bitbake qemu-native -c clean; bitbake qemu-native), qemu would no longer crash. But this also means you'd lose the performance of the proprietary driver.
(In reply to comment #4) > I'm afraid we don't have a workaround for this yet. I was able to verify the > crashes no longer happen when I uninstalled the NVIDIA driver and ensured that > the OpenGL libraries no longer existed in /usr/lib/nvidia-current. Then when I > rebuilt qemu-native (bitbake qemu-native -c clean; bitbake qemu-native), qemu > would no longer crash. But this also means you'd lose the performance of the > proprietary driver. Aah, Ok -- I have two machines and both have the Nvidia driver since I am also doing some graphics intensive stuff. Have to think about this - perhaps a KVM, Vmware instance but then I loose some of the CPU horse power ... :-) Anyways, thanks for your prompt assistance! Cheers // Matias
One more thing, I just started a Common_PC WindRiver project with; # make start-target TOPTS="-t" and that QEMU instance started nicely - perhaps something there to look at :-) Cheers // Matias
See 649 also
(In reply to comment #1) > Are you using the proprietary NVIDIA Xorg driver? There is a known issue with > the openGL library it ships which can cause this behavior when qemu-native is > compiled against it. Scott, Can you give me some info of this known issue? I suspect nvidia GL library conflict with some system library. Thanks, edwin
it's the same bug as 649 caused by nVidia proprietary driver. *** This bug has been marked as a duplicate of bug 649 ***