crash booting rawhide xen0

[Andy Burns]

Hi there,

I've not been updating my xen kernels to *every* recent rawhide release, 
but did so today, to 2.6.15-1.2041_FC5xen0, my first two attempts to 
boot this xen0 crashed and stackdumped :-(

Having seen it fail twice, I attached a serial console and was preparing 
to capture that with console=ttyS0,115200, but on the next two boots the 
crash didn't happen, however my console output still went to the 
vgaconsole, does xen have a different way to use the serial console?

I'll gladly log a bug if I can make it happen again ...

Also I've just tidied up my grub.conf, /boot/ and /lib/modules/ to get 
rid of heaps of old kernels that the switch from guest/hypervisor to 
xen0/xenU had left around, is there anywhere I can get back a few 
slightly older kernels to try and find out what version caused this 
breakage? I think my last known good version was kernel 2032

[Andy Burns]

Andy Burns wrote:

>  2.6.15-1.2041_FC5xen0, my first two attempts to 
>  boot this xen0 crashed and stackdumped :-( 
OK, with Yongkang's info I've captured this happening

https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=185235
Full serial log attached there, but the offending bit seems to be

Kernel BUG at arch/x86_64/mm/fault-xen.c:292
invalid opcode: 0000 [1] SMP
CPU 0
Modules linked in: snd_mixer_oss snd_pcm snd_timer snd soundcore 
snd_page_alloc hw_random i2c_i801 i2c_core dm_snapshot dm_zero dm_mirror 
dm_mod ext3 jbd ahci libata sd_mod scsi_mod
Pid: 615, comm: udevd Not tainted 2.6.15-1.2041_FC5xen0 #1
RIP: e030:[<ffffffff801197df>] <ffffffff801197df>{do_page_fault+1690}
RSP: e02b:ffffffff80418978  EFLAGS: 00010287
RAX: ffff800000000000 RBX: ffff880000ed7000 RCX: 00003ffffffff000
RDX: 0000000000050204 RSI: 000000000003ac4e RDI: 0000000000000c00
RBP: ffffffff80101c20 R08: 000000000003ac4e R09: ffff880000ae8070
R10: ffff880000ae9000 R11: ffff880000ae9000 R12: ffffc2000000e008
R13: ffffffff80418a58 R14: ffff880038d8f100 R15: 0000000000000000
FS:  00002b07a11e66d0(0000) GS:ffffffff80499000(0000) knlGS:0000000000000000
CS:  e033 DS: 0000 ES: 0000
Process udevd (pid: 615, threadinfo ffff880038cc4000, task ffff880038d8f100)
Stack: 0000000000000001 0000000000030001 ffffffff804189c8 ffffffff801f2163
        ffff880000636048 ffff88000230e800 ffff880000636048 ffff880000636048
        ffff880000636240 ffffffff801e727d
Call Trace: <IRQ> <ffffffff801f2163>{cfq_dispatch_requests+326}
        <ffffffff801e727d>{elv_next_request+321} 
<ffffffff801f49fa>{kobject_get+18}
        <ffffffff880064f6>{:scsi_mod:scsi_request_fn+843} 
<ffffffff8010b503>{error_exit+0}
        <ffffffff8803d447>{:ahci:ahci_interrupt+33} 
<ffffffff80151e15>{handle_IRQ_event+78}
        <ffffffff80151efb>{__do_IRQ+156} <ffffffff8010d4c2>{do_IRQ+65}
        <ffffffff8025eeed>{evtchn_do_upcall+132} 
<ffffffff8010b5b5>{do_hypervisor_callback+37} <EOI>
        <ffffffff8010722a>{hypercall_page+554} 
<ffffffff8010722a>{hypercall_page+554}
        <ffffffff8018e463>{mntput_no_expire+25} 
<ffffffff8025e8d5>{force_evtchn_callback+10}
        <ffffffff80119209>{do_page_fault+196} 
<ffffffff801845a5>{sys_mknodat+371}
        <ffffffff8010b503>{error_exit+0}

[Aleksander Adamowski]

Andy Burns wrote:

>  Andy Burns wrote:
>
> > 2.6.15-1.2041_FC5xen0, my first two attempts to boot this xen0 
> > crashed and stackdumped :-(
>
>
>  OK, with Yongkang's info I've captured this happening
>
>  https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=185235
>  Full serial log attached there, but the offending bit seems to be
>
>  Kernel BUG at arch/x86_64/mm/fault-xen.c:292
>  invalid opcode: 0000 [1] SMP
>  CPU 0
>  Modules linked in: snd_mixer_oss snd_pcm snd_timer snd soundcore 
>  snd_page_alloc hw_random i2c_i801 i2c_core dm_snapshot dm_zero 
>  dm_mirror dm_mod ext3 jbd ahci libata sd_mod scsi_mod
>  Pid: 615, comm: udevd Not tainted 2.6.15-1.2041_FC5xen0 #1
>  RIP: e030:[<ffffffff801197df>] <ffffffff801197df>{do_page_fault+1690}
>  RSP: e02b:ffffffff80418978  EFLAGS: 00010287
>  RAX: ffff800000000000 RBX: ffff880000ed7000 RCX: 00003ffffffff000
>  RDX: 0000000000050204 RSI: 000000000003ac4e RDI: 0000000000000c00
>  RBP: ffffffff80101c20 R08: 000000000003ac4e R09: ffff880000ae8070
>  R10: ffff880000ae9000 R11: ffff880000ae9000 R12: ffffc2000000e008
>  R13: ffffffff80418a58 R14: ffff880038d8f100 R15: 0000000000000000
>  FS:  00002b07a11e66d0(0000) GS:ffffffff80499000(0000) 
>  knlGS:0000000000000000
>  CS:  e033 DS: 0000 ES: 0000
>  Process udevd (pid: 615, threadinfo ffff880038cc4000, task 
>  ffff880038d8f100)
>  Stack: 0000000000000001 0000000000030001 ffffffff804189c8 
>  ffffffff801f2163
>         ffff880000636048 ffff88000230e800 ffff880000636048 
>  ffff880000636048
>         ffff880000636240 ffffffff801e727d
>  Call Trace: <IRQ> <ffffffff801f2163>{cfq_dispatch_requests+326}
>         <ffffffff801e727d>{elv_next_request+321} 
>  <ffffffff801f49fa>{kobject_get+18}
>         <ffffffff880064f6>{:scsi_mod:scsi_request_fn+843} 
>  <ffffffff8010b503>{error_exit+0}
>         <ffffffff8803d447>{:ahci:ahci_interrupt+33} 
>  <ffffffff80151e15>{handle_IRQ_event+78}
>         <ffffffff80151efb>{__do_IRQ+156} <ffffffff8010d4c2>{do_IRQ+65}
>         <ffffffff8025eeed>{evtchn_do_upcall+132} 
>  <ffffffff8010b5b5>{do_hypervisor_callback+37} <EOI>
>         <ffffffff8010722a>{hypercall_page+554} 
>  <ffffffff8010722a>{hypercall_page+554}
>         <ffffffff8018e463>{mntput_no_expire+25} 
>  <ffffffff8025e8d5>{force_evtchn_callback+10}
>         <ffffffff80119209>{do_page_fault+196} 
>  <ffffffff801845a5>{sys_mknodat+371}
>         <ffffffff8010b503>{error_exit+0} 
I've reported this bug that I had with  2.6.15-1.2025_FC5xen0 to Bugzilla:

https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=184378

The error is the same, the stacktrace is different. Please add your
error to that bug and CC yourself in the Bugzilla.

-- 
Best Regards,
     Aleksander Adamowski
         GG#: 274614
         ICQ UIN: 19780575
	http://olo.ab.altkom.pl