Date: Sun, 25 Nov 2012 17:48:54 -0800 From: "sree.openwrk" <sree.openwrk@gmail.com> To: Neel Natu <neelnatu@gmail.com> Cc: freebsd-virtualization@freebsd.org Subject: Re: Bhyve: panic: "Unregistered use of FPU in kernel" when starting guest Message-ID: <CADYnSLCNb5CcLXi%2B_XkJ7Wb-zf5iyzMJCNgms=poXuDDKA32Yg@mail.gmail.com> In-Reply-To: <CADYnSLBN7HhFaZ3yXRkoAvAQ2oDgig=XUpf31T-qHscf8b-0hg@mail.gmail.com> References: <CADYnSLDW492JKvQXN0wTJzV6VdCF_ZxRg4OnaTxqKdWjGzQk_w@mail.gmail.com> <CAFgRE9Ec74HQ9TDs_BwmYwLCebZVg2tQbyxNxJfuvsSX1qQFhQ@mail.gmail.com> <CADYnSLAHf01ZzEn7tQiR3dv2v63mDMj0B3=X5DDsbkS-dG46MQ@mail.gmail.com> <CAFgRE9GpJ8EXJ2GyTcnKeRd8VKmy5v0BfK_P2mHNnFkJQ_Ba0g@mail.gmail.com> <CADYnSLBN7HhFaZ3yXRkoAvAQ2oDgig=XUpf31T-qHscf8b-0hg@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Update- After enabling hw watchdog- by loading the ichwd kernel module and starting the watchdog daemon I am not getting any clue. After enabling hw-watchdog and starting the guest, the system becomes unresponsive. Neither does the watchdogd trigger a system reboot. I followed this- http://koitsu.wordpress.com/2010/07/13/freebsd-and-hardwaresoftware-watchdogs/- to enable hw-watchdog. On Sun, Nov 25, 2012 at 3:46 PM, sree.openwrk <sree.openwrk@gmail.com>wrote: > Hey Neel > I tried to see if the problem is reproducible after the fix went in. > Looks like I have landed in to another issue. Now when I start the guest > vm, the host hangs and could only be rebooted. Since its not a crash I am > not able to get a crash dump. Is this a known issue? Also is there a way I > can config and get to kick off a trace using something lie a hw-watchdog or > something? > > Reproduction steps- > 1. I have 8gb memory and I have limited the host to see only 4gb. > svas-fbsd# sysctl -a | grep hw.physmem > hw.physmem: 3116384256 > > output from dmesg- > real memory = 8589934592 (8192 MB) > avail memory = 2999402496 (2860 MB) > > 2. I checkedout the latest code from svn:// > svn.freebsd.org/base/projects/bhyve. The revision number is 243326. The > host has this image built out of this. > > 3. These are the steps I followed, before starting the guest. Then it > hangs. > a) kldload vmm > b) kldload if_tap > c) ifconfig tap0 create > > 4. I use the prebuilt tarball given at > http://people.freebsd.org/~neel/bhyve/vm1.tar.gz. > > 5. I start the vm as ./vmrun.sh vm1 and it hangs after displaying the > message "Launching virtual machine "vm1" with 768MB memory below 4GB and > 2048MB memory above 4GB ..." > > Thanks > > > > > On Thu, Oct 25, 2012 at 8:16 PM, Neel Natu <neelnatu@gmail.com> wrote: > >> Hi svas, >> >> I just submitted a fix for this: >> http://svnweb.freebsd.org/base?view=revision&revision=242122 >> >> Please let me know if this fixes the panic you reported. >> >> best >> Neel >> >> On Sun, Oct 7, 2012 at 11:12 AM, sree.openwrk <sree.openwrk@gmail.com> >> wrote: >> > Hey Neel >> > Sure. I will reproduce the issue and try to narrow down the problem. >> > Will update you, if I am able to narrow down the problem. >> > >> > Thanks >> > >> > >> > On Thu, Oct 4, 2012 at 9:07 PM, Neel Natu <neelnatu@gmail.com> wrote: >> >> >> >> Hi svas, >> >> >> >> On Thu, Oct 4, 2012 at 5:58 PM, sree.openwrk <sree.openwrk@gmail.com> >> >> wrote: >> >> > Hi >> >> > I am seeing frequent kernel panic when I try to run the guest OS. I >> have >> >> > the >> >> > latest Freebsd (I compiled the latest code from >> >> > svn://svn.freebsd.org/base/projects/bhyve and installed it in my >> >> > machine). >> >> > While trying to run the pre-built guest provided at >> >> > http://people.freebsd.org/~neel/bhyve/vm1.tar.gz I get the kernel >> panic >> >> > (panic: Unregistered use of FPU in kernel) >> >> > I am attaching the core.txt. >> >> > I also want to mention that I was successfully able to boot in to the >> >> > guest >> >> > OS couple of times. But around 7/10 times I get thsi host kernel >> panic. >> >> > >> >> > stack trace: >> >> > vm_run() at vm_run+0xec >> >> > vmmdev_ioctl() at vmmdev_ioctl+0x4ea >> >> > devfs_ioctl_f() at devfs_ioctl_f+0x7a >> >> > kern_ioctl() at kern_ioctl+0xcd >> >> > sys_ioctl() at sys_ioctl+0xfd >> >> > amd64_syscall() at amd64_syscall+0x304 >> >> > Xfast_syscall() at Xfast_syscall+0xf7 >> >> > >> >> >> >> I am seeing this too but nowhere as frequently as you are. I hit this >> >> last night for the first time and not since. >> >> >> >> There hasn't been any change in this part of the code for quite some >> time >> >> now. >> >> >> >> Since you can trigger this much more frequently that me, would it be >> >> possible for you to narrow down which change caused this? It would be >> >> hugely helpful. >> >> >> >> best >> >> Neel >> >> >> >> > >> >> > regards >> >> > svas >> >> > >> >> > >> > >> > >> > >
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CADYnSLCNb5CcLXi%2B_XkJ7Wb-zf5iyzMJCNgms=poXuDDKA32Yg>