[PlanetCCRMA] RT Kernel Crash

[Jonathan E. Brickman]

I'm seeing exactly the same thing.  It happened to me under Fedora 12,
but only when I was either (a) using virt/KVM, (b) VirtualBox, or (c)
Google Earth.  Google Earth and virt/KVM were still usable, VirtualBox
was not.

Here's the text of one of the crashes from the Automatic Bug Reporting
Tool:

------------[ cut here ]------------
WARNING: at fs/namespace.c:648 commit_tree+0xda/0xf5()
Hardware name: MS-7309
Modules linked in: snd_seq_midi snd_seq_midi_event snd_usb_audio
snd_hwdep snd_usb_lib snd_seq_dummy fuse ip6t_REJECT nf_conntrack_ipv6
ip6table_filter ip6_tables ipv6 uinput snd_ice1724 snd_rawmidi
snd_ice17xx_ak4xxx snd_ac97_codec ac97_bus snd_ak4xxx_adda snd_ak4114
snd_pt2258 snd_i2c snd_ak4113 snd_seq snd_seq_device snd_pcm snd_timer
snd soundcore edac_core snd_page_alloc i2c_nforce2 edac_mce_amd
forcedeth k10temp serio_raw microcode pata_acpi ata_generic pata_amd
sata_nv nouveau ttm drm_kms_helper drm i2c_algo_bit video output
i2c_core [last unloaded: freq_table]
Pid: 26896, comm: fusermount Not tainted
2.6.33.4-106.rt20.1.fc13.ccrma.x86_64.rt #1
Call Trace:
[<ffffffff8104bee4>] warn_slowpath_common+0x77/0x8f
[<ffffffff8104bf0b>] warn_slowpath_null+0xf/0x11
[<ffffffff8111a42c>] commit_tree+0xda/0xf5
[<ffffffff8111b27d>] attach_recursive_mnt+0xed/0x183
[<ffffffff8111b400>] graft_tree+0xed/0xef
[<ffffffff8111be27>] do_mount+0x389/0x79f
[<ffffffff810d682e>] ? strndup_user+0x58/0x80
[<ffffffff8111c2c0>] sys_mount+0x83/0xbd
[<ffffffff81009c02>] system_call_fastpath+0x16/0x1b

And another:

BUG: sleeping function called from invalid context at
kernel/rtmutex.c:684
pcnt: 0 0 in_atomic(): 0, irqs_disabled(): 1, pid: 6729, name:
pm-hibernate
Pid: 6729, comm: pm-hibernate Not tainted
2.6.33.4-106.rt20.1.fc13.ccrma.x86_64.rt #1
Call Trace:
[<ffffffff81039297>] __might_sleep+0xf7/0xfe
[<ffffffff81078818>] rt_spin_lock_fastlock.clone.1+0x31/0x77
[<ffffffff81440bce>] rt_spin_lock+0x9/0xb
[<ffffffff81010909>] read_persistent_clock+0x16/0x3b
[<ffffffff8106f063>] timekeeping_suspend+0x15/0x75
[<ffffffff812b2fea>] sysdev_suspend+0x18e/0x404
[<ffffffff8108083b>] hibernation_snapshot+0xf8/0x227
[<ffffffff81080a4b>] hibernate+0xe1/0x1a6
[<ffffffff8107f692>] state_store+0x57/0xce
[<ffffffff81206ccf>] kobj_attr_store+0x17/0x19
[<ffffffff8115a2dc>] sysfs_write_file+0x105/0x141
[<ffffffff8110380c>] vfs_write+0xa9/0x106
[<ffffffff8110391f>] sys_write+0x45/0x69
[<ffffffff81009c02>] system_call_fastpath+0x16/0x1b




On Thu, 2010-07-01 at 14:06 -0400, Donald Steven wrote:
>  It's intermittent and does not occur during the boot process.  BTW,
> it has never occurred with previous Fedora/CCRMA Rt kernels, or with
> Centos/CCRMA.  Any ideas?  If it occurs at other-than-boot-time, what
> log(s) might I look in?
> 
> Don
> 
> Tracey Hytry wrote: 
> > On Wed, 30 Jun 2010 21:54:58 -0400
> > Donald Steven <t6sn7gt at aim.com> wrote:
> > 
> >   
> > > I'm getting kernel crashes with the rt kernel.  It seems to  recover and 
> > > the system does not crash, but is there something I'm doing wrong?
> > >     
> > 
> > I doubt that you're wrong, probably a hardware thing?
> > 
> > Could you post the ending of the dmesg where it shows what the kernel's complaing about?
> > I don't know if we can help, but you know what they say about more eyes.
> > 
> > _______________________________________________
> > PlanetCCRMA mailing list
> > PlanetCCRMA at ccrma.stanford.edu
> > http://ccrma-mail.stanford.edu/mailman/listinfo/planetccrma
> >   
> _______________________________________________
> PlanetCCRMA mailing list
> PlanetCCRMA at ccrma.stanford.edu
> http://ccrma-mail.stanford.edu/mailman/listinfo/planetccrma