Hi,
After an uptime of, say, 4 hours I got an oops on my P4 with HT and 2GB ram.
It has dynamic ticks enabled as well has timer stats. I cannot reboot the
machine to give more details right now (19:00 gmt+1 i can).
For now, this is the oops:
[13092.479019] BUG: unable to handle kernel paging request at virtual
address b9d0458f
[13092.479115] printing eip:
[13092.479160] c10388a8
[13092.479202] *pde = 00000000
[13092.479246] Oops: 0000 [#1]
[13092.479290] SMP
[13092.479401] Modules linked in: wcfxo zaptel tuner tvaudio bttv
video_buf ir_common i2c_algo_bit btcx_risc tveeprom pl2303 pwc usbserial
nfs plusb w83627hf hwmon_vid usbnet snd_pcm_oss eeprom snd_mixer_oss
i2c_isa i2c_core snd_intel8x0 snd_ac97_codec snd_pcm sd_mod snd_timer
scsi_mod snd ide_cd soundcore snd_page_alloc cdrom ac97_bus parport_pc
parport microcode firmware_class netconsole nfsd exportfs lockd sunrpc
ipt_owner ip6table_filter ip6_tables ipv6 ipt_recent xt_limit xt_state
act_police sch_ingress cls_u32 sch_sfq sch_cbq ipt_REJECT ipt_MASQUERADE
ipt_TOS xt_tcpudp iptable_mangle iptable_filter rtl8150 e1000 3c59x mii
iptable_nat ip_tables nf_nat nf_conntrack_ipv4 nf_conntrack nfnetlink
x_tables capability commoncap ppp_deflate zlib_deflate zlib_inflate
ppp_async crc_ccitt ppp_generic slip slhc genrtc rd
[13092.483367] CPU: 1
[13092.483368] EIP: 0060:[<c10388a8>] Not tainted VLI
[13092.483370] EFLAGS: 00010046 (2.6.21.1test #1)
[13092.483511] EIP is at match_entries+0xb/0x38
[13092.483558] eax: c25f9a80 ebx: 00000000 ecx: b9d0458b edx: f039aec8
[13092.483610] esi: f039aec8 edi: c25f99a0 ebp: f039aea0 esp: f039ae9c
[13092.483660] ds: 007b es: 007b fs: 00d8 gs: 0033 ss: 0068
[13092.483712] Process einstein_S5R2_4 (pid: 12710, ti=f039a000
task=f7c32a70 task.ti=f039a000)
[13092.483760] Stack: b9d0458b f039aec0 c1038911 c12109e9 c12ffe10
c25f9ab0 00000096 c25f9b04
[13092.484181] c25f99a0 f039af50 c10389ec c120e6eb c25f9a80
c10383ec c10385f0 00000000
[13092.484581] 00000000 c1495c78 00000000 00000000 c13f7230
00000001 f7c32fb8 c13f87c0
[13092.484988] Call Trace:
[13092.485105] [<c1004d53>] show_trace_log_lvl+0x1a/0x30
[13092.485220] [<c1004e0a>] show_stack_log_lvl+0x8d/0xaa
[13092.485336] [<c1005044>] show_registers+0x1cd/0x2cb
[13092.485457] [<c10052a1>] die+0x11b/0x227
[13092.485578] [<c1212aef>] do_page_fault+0x319/0x57a
[13092.485698] [<c12110cc>] error_code+0x7c/0x84
[13092.485813] [<c1038911>] tstat_lookup+0x3c/0xc9
[13092.485928] [<c10389ec>] timer_stats_update_stats+0x4e/0x6f
[13092.486041] [<c1033d5d>] hrtimer_interrupt+0x184/0x1e3
[13092.486156] [<c100e9e2>] local_apic_timer_interrupt+0x55/0x5b
[13092.486272] [<c100ea12>] smp_apic_timer_interrupt+0x2a/0x39
[13092.486390] [<c1004a3f>] apic_timer_interrupt+0x33/0x38
[13092.486510] =======================
[13092.486575] Code: 3d ff 03 00 00 89 e5 77 13 89 c2 83 c0 01 c1 e2 07 a3
98 2c 37 c1 81 c2 00 2d 37 c1 89 d0 5d c3 55 89 c1 8b 42 04 89 e5 53 31 db
<39> 41 04 74 05 89 d8 5b 5d c3 8b 42 08 39 41 08 75 f3 8b 42 0c
[13092.489766] EIP: [<c10388a8>] match_entries+0xb/0x38 SS:ESP 0068:f039ae9c
[13092.489901] Kernel panic - not syncing: Fatal exception in interrupt
[13092.489951] BUG: at arch/i386/kernel/smp.c:546 smp_call_function()
[13092.490003] [<c1004d53>] show_trace_log_lvl+0x1a/0x30
[13092.490091] [<c1004d7b>] show_trace+0x12/0x14
[13092.490178] [<c1004e75>] dump_stack+0x16/0x18
[13092.490267] [<c100df63>] smp_call_function+0x10f/0x114
[13092.490367] [<c100dfb9>] smp_send_stop+0x1e/0x31
[13092.490454] [<c101ca83>] panic+0x50/0xf5
[13092.490548] [<c100539e>] die+0x218/0x227
[13092.490640] [<c1212aef>] do_page_fault+0x319/0x57a
[13092.490738] [<c12110cc>] error_code+0x7c/0x84
[13092.490827] [<c1038911>] tstat_lookup+0x3c/0xc9
[13092.490914] [<c10389ec>] timer_stats_update_stats+0x4e/0x6f
[13092.491012] [<c1033d5d>] hrtimer_interrupt+0x184/0x1e3
[13092.491113] [<c100e9e2>] local_apic_timer_interrupt+0x55/0x5b
[13092.491201] [<c100ea12>] smp_apic_timer_interrupt+0x2a/0x39
[13092.491289] [<c1004a3f>] apic_timer_interrupt+0x33/0x38
[13092.491381] =======================
[Adding Thomas to CC]
On 15/05/07, Folkert van Heusden <[email protected]> wrote:
> Hi,
>
> After an uptime of, say, 4 hours I got an oops on my P4 with HT and 2GB ram.
> It has dynamic ticks enabled as well has timer stats. I cannot reboot the
> machine to give more details right now (19:00 gmt+1 i can).
>
> For now, this is the oops:
>
> [13092.479019] BUG: unable to handle kernel paging request at virtual
> address b9d0458f
> [13092.479115] printing eip:
> [13092.479160] c10388a8
> [13092.479202] *pde = 00000000
> [13092.479246] Oops: 0000 [#1]
> [13092.479290] SMP
> [13092.479401] Modules linked in: wcfxo zaptel tuner tvaudio bttv
> video_buf ir_common i2c_algo_bit btcx_risc tveeprom pl2303 pwc usbserial
> nfs plusb w83627hf hwmon_vid usbnet snd_pcm_oss eeprom snd_mixer_oss
> i2c_isa i2c_core snd_intel8x0 snd_ac97_codec snd_pcm sd_mod snd_timer
> scsi_mod snd ide_cd soundcore snd_page_alloc cdrom ac97_bus parport_pc
> parport microcode firmware_class netconsole nfsd exportfs lockd sunrpc
> ipt_owner ip6table_filter ip6_tables ipv6 ipt_recent xt_limit xt_state
> act_police sch_ingress cls_u32 sch_sfq sch_cbq ipt_REJECT ipt_MASQUERADE
> ipt_TOS xt_tcpudp iptable_mangle iptable_filter rtl8150 e1000 3c59x mii
> iptable_nat ip_tables nf_nat nf_conntrack_ipv4 nf_conntrack nfnetlink
> x_tables capability commoncap ppp_deflate zlib_deflate zlib_inflate
> ppp_async crc_ccitt ppp_generic slip slhc genrtc rd
> [13092.483367] CPU: 1
> [13092.483368] EIP: 0060:[<c10388a8>] Not tainted VLI
> [13092.483370] EFLAGS: 00010046 (2.6.21.1test #1)
> [13092.483511] EIP is at match_entries+0xb/0x38
> [13092.483558] eax: c25f9a80 ebx: 00000000 ecx: b9d0458b edx: f039aec8
> [13092.483610] esi: f039aec8 edi: c25f99a0 ebp: f039aea0 esp: f039ae9c
> [13092.483660] ds: 007b es: 007b fs: 00d8 gs: 0033 ss: 0068
> [13092.483712] Process einstein_S5R2_4 (pid: 12710, ti=f039a000
> task=f7c32a70 task.ti=f039a000)
> [13092.483760] Stack: b9d0458b f039aec0 c1038911 c12109e9 c12ffe10
> c25f9ab0 00000096 c25f9b04
> [13092.484181] c25f99a0 f039af50 c10389ec c120e6eb c25f9a80
> c10383ec c10385f0 00000000
> [13092.484581] 00000000 c1495c78 00000000 00000000 c13f7230
> 00000001 f7c32fb8 c13f87c0
> [13092.484988] Call Trace:
> [13092.485105] [<c1004d53>] show_trace_log_lvl+0x1a/0x30
> [13092.485220] [<c1004e0a>] show_stack_log_lvl+0x8d/0xaa
> [13092.485336] [<c1005044>] show_registers+0x1cd/0x2cb
> [13092.485457] [<c10052a1>] die+0x11b/0x227
> [13092.485578] [<c1212aef>] do_page_fault+0x319/0x57a
> [13092.485698] [<c12110cc>] error_code+0x7c/0x84
> [13092.485813] [<c1038911>] tstat_lookup+0x3c/0xc9
> [13092.485928] [<c10389ec>] timer_stats_update_stats+0x4e/0x6f
> [13092.486041] [<c1033d5d>] hrtimer_interrupt+0x184/0x1e3
> [13092.486156] [<c100e9e2>] local_apic_timer_interrupt+0x55/0x5b
> [13092.486272] [<c100ea12>] smp_apic_timer_interrupt+0x2a/0x39
> [13092.486390] [<c1004a3f>] apic_timer_interrupt+0x33/0x38
> [13092.486510] =======================
> [13092.486575] Code: 3d ff 03 00 00 89 e5 77 13 89 c2 83 c0 01 c1 e2 07 a3
> 98 2c 37 c1 81 c2 00 2d 37 c1 89 d0 5d c3 55 89 c1 8b 42 04 89 e5 53 31 db
> <39> 41 04 74 05 89 d8 5b 5d c3 8b 42 08 39 41 08 75 f3 8b 42 0c
> [13092.489766] EIP: [<c10388a8>] match_entries+0xb/0x38 SS:ESP 0068:f039ae9c
> [13092.489901] Kernel panic - not syncing: Fatal exception in interrupt
> [13092.489951] BUG: at arch/i386/kernel/smp.c:546 smp_call_function()
> [13092.490003] [<c1004d53>] show_trace_log_lvl+0x1a/0x30
> [13092.490091] [<c1004d7b>] show_trace+0x12/0x14
> [13092.490178] [<c1004e75>] dump_stack+0x16/0x18
> [13092.490267] [<c100df63>] smp_call_function+0x10f/0x114
> [13092.490367] [<c100dfb9>] smp_send_stop+0x1e/0x31
> [13092.490454] [<c101ca83>] panic+0x50/0xf5
> [13092.490548] [<c100539e>] die+0x218/0x227
> [13092.490640] [<c1212aef>] do_page_fault+0x319/0x57a
> [13092.490738] [<c12110cc>] error_code+0x7c/0x84
> [13092.490827] [<c1038911>] tstat_lookup+0x3c/0xc9
> [13092.490914] [<c10389ec>] timer_stats_update_stats+0x4e/0x6f
> [13092.491012] [<c1033d5d>] hrtimer_interrupt+0x184/0x1e3
> [13092.491113] [<c100e9e2>] local_apic_timer_interrupt+0x55/0x5b
> [13092.491201] [<c100ea12>] smp_apic_timer_interrupt+0x2a/0x39
> [13092.491289] [<c1004a3f>] apic_timer_interrupt+0x33/0x38
> [13092.491381] =======================
>
Added to KR as "OOPS in timer stats code"
http://kernelnewbies.org/known_regressions
Regards,
Michal
--
Michal K. K. Piotrowski
Kernel Monkeys
(http://kernel.wikidot.com/start)
Michal Piotrowski wrote:
> [Adding Thomas to CC]
>
> On 15/05/07, Folkert van Heusden <[email protected]> wrote:
>> Hi,
>>
>> After an uptime of, say, 4 hours I got an oops on my P4 with HT and
>> 2GB ram.
>> It has dynamic ticks enabled as well has timer stats. I cannot reboot the
>> machine to give more details right now (19:00 gmt+1 i can).
>>
>> For now, this is the oops:
>>
>> [13092.479019] BUG: unable to handle kernel paging request at virtual
>> address b9d0458f
Timer hash list contains a user address.
On Tue, 2007-05-15 at 14:37 -0400, Chuck Ebbert wrote:
> Michal Piotrowski wrote:
> > [Adding Thomas to CC]
Added Ingo as well :)
> > On 15/05/07, Folkert van Heusden <[email protected]> wrote:
> >> Hi,
> >>
> >> After an uptime of, say, 4 hours I got an oops on my P4 with HT and
> >> 2GB ram.
> >> It has dynamic ticks enabled as well has timer stats. I cannot reboot the
> >> machine to give more details right now (19:00 gmt+1 i can).
> >>
> >> For now, this is the oops:
> >>
> >> [13092.479019] BUG: unable to handle kernel paging request at virtual
> >> address b9d0458f
>
> Timer hash list contains a user address.
Dammit. It's a HT enabled box again. I don't have one.
Folkert, does the problem go away when you disable HT
(CONFIG_SCHED_SMT) ?
tglx
> > >> Hi,
> > >> After an uptime of, say, 4 hours I got an oops on my P4 with HT and
> > >> 2GB ram.
> > >> It has dynamic ticks enabled as well has timer stats. I cannot reboot the
> > >> machine to give more details right now (19:00 gmt+1 i can).
> > >> For now, this is the oops:
> > >> [13092.479019] BUG: unable to handle kernel paging request at virtual
> > >> address b9d0458f
> > Timer hash list contains a user address.
>
> Dammit. It's a HT enabled box again. I don't have one.
> Folkert, does the problem go away when you disable HT
> (CONFIG_SCHED_SMT) ?
Don't know but the last patch I got works fine it seems (10 hours up).
Folkert van Heusden
--
MultiTail is a versatile tool for watching logfiles and output of
commands. Filtering, coloring, merging, diff-view, etc.
http://www.vanheusden.com/multitail/
----------------------------------------------------------------------
Phone: +31-6-41278122, PGP-key: 1F28D8AE, http://www.vanheusden.com
On Wed, 2007-05-16 at 10:00 +0200, Folkert van Heusden wrote:
> > > >> Hi,
> > > >> After an uptime of, say, 4 hours I got an oops on my P4 with HT and
> > > >> 2GB ram.
> > > >> It has dynamic ticks enabled as well has timer stats. I cannot reboot the
> > > >> machine to give more details right now (19:00 gmt+1 i can).
> > > >> For now, this is the oops:
> > > >> [13092.479019] BUG: unable to handle kernel paging request at virtual
> > > >> address b9d0458f
> > > Timer hash list contains a user address.
> >
> > Dammit. It's a HT enabled box again. I don't have one.
> > Folkert, does the problem go away when you disable HT
> > (CONFIG_SCHED_SMT) ?
>
> Don't know but the last patch I got works fine it seems (10 hours up).
/me lost context. Which patch ?
tglx
Thomas Gleixner wrote:
> Dammit. It's a HT enabled box again. I don't have one.
>
> Folkert, does the problem go away when you disable HT
> (CONFIG_SCHED_SMT) ?
Why would that make a difference? If you do suspect it is the problem,
then I think you could just do something like substitute cpu_online_map
for the sibling map in the SMT domain setup in order to reproduce it
locally?
--
SUSE Labs, Novell Inc.
> > > > >> After an uptime of, say, 4 hours I got an oops on my P4 with HT and
> > > > >> 2GB ram.
> > > > >> It has dynamic ticks enabled as well has timer stats. I cannot reboot the
> > > > >> machine to give more details right now (19:00 gmt+1 i can).
> > > > >> For now, this is the oops:
> > > > >> [13092.479019] BUG: unable to handle kernel paging request at virtual
> > > > >> address b9d0458f
> > > > Timer hash list contains a user address.
> > > Dammit. It's a HT enabled box again. I don't have one.
> > > Folkert, does the problem go away when you disable HT
> > > (CONFIG_SCHED_SMT) ?
> > Don't know but the last patch I got works fine it seems (10 hours up).
> /me lost context. Which patch ?
The one by Jan Kara with Message-ID: <[email protected]>
Folkert van Heusden
--
http://www.vanheusden.com/multitail - multitail is tail on steroids. multiple
windows, filtering, coloring, anything you can think of
----------------------------------------------------------------------
Phone: +31-6-41278122, PGP-key: 1F28D8AE, http://www.vanheusden.com