Skip to contentSkip to conversation

2.6.30.1-4 kernels freezes dual pentium-3 system

DISCUSSION [email protected] 2 participants 2 messages Aug 8, 2009

lore
Share & tools
lore
#0 lore raw
Quoted 1 line

Roland, you seem to reply to a mix of different mails.

i replied to this thread:
http://marc.info/?t=124928942600006&r=1&w=2

Quoted 3 lines

I skip the comments
unrelated to my error report. Btw, since then there have been two more reports on lkml of dual P2 systems locking up with 2.6.30++.

could you send pointers?
any existing bugzilla bugentry for these or for your`s ?

Quoted 3 lines
Quoted 1 line

afaik, you can also bisect on a different (i.e. faster) box.

Since we are talking about dual-P3, no you can't.

ok, i missed your comment regarding duration of bisect iteration.
but despite that, compiling on another/faster box and copy the kernel over is possible, afaik (and won`t help here)

Quoted 3 lines
Quoted 1 line

i think it would be more interesting if there is anything in dmesg on such

I posted dmesg etc. yesterday

yes, but that mostly contain bootup messages and theres absolutely no trace from the freeze. its quite common, that after a lockup/crash some messages are lost, that`s why i asked if there is a chance to have serial console or netconsole attached.

Quoted 1 line

Network, sysrq and numlock are all unresponsive.

ok, that`s why we need to have the last informations from before the dying breath...

btw, i one of your dmesg i found these:

Aug  1 19:42:57 frya kernel: [    0.000000] Linux version 2.6.30-gentoo-r4 \
Aug  1 19:42:57 frya kernel: [    4.394755] ------------[ cut here ]------------
Aug  1 19:42:57 frya kernel: [    4.399247] WARNING: at arch/x86/mm/ioremap.c:603 \
                check_early_ioremap_leak+0x53/0x73()
Aug  1 19:42:57 frya kernel: [    4.408763] Hardware name: System Name
Aug  1 19:42:57 frya kernel: [    4.413574] Debug warning: early ioremap leak of 1 \
                areas detected.
Aug  1 19:42:57 frya kernel: [    4.418561] Pid: 1, comm: swapper Not tainted \
                2.6.30-gentoo-r4 #1
Aug  1 19:42:57 frya kernel: [    4.423552] Call Trace:
Aug  1 19:42:57 frya kernel: [    4.428379]  [<c07c90b2>] ? \
                check_early_ioremap_leak+0x53/0x73
Aug  1 19:42:57 frya kernel: [    4.433238]  [<c07c90b2>] ? \
                check_early_ioremap_leak+0x53/0x73
Aug  1 19:42:57 frya kernel: [    4.437821]  [<c01316ac>] \
                warn_slowpath_common+0x6c/0xc0
Aug  1 19:42:57 frya kernel: [    4.442266]  [<c07c90b2>] ? \
                check_early_ioremap_leak+0x53/0x73
Aug  1 19:42:57 frya kernel: [    4.446658]  [<c07c905f>] ? \
                check_early_ioremap_leak+0x0/0x73
Aug  1 19:42:57 frya kernel: [    4.450880]  [<c013174b>] warn_slowpath_fmt+0x2b/0x30
Aug  1 19:42:57 frya kernel: [    4.455062]  [<c07c90b2>] \
                check_early_ioremap_leak+0x53/0x73
Aug  1 19:42:57 frya kernel: [    4.459288]  [<c010103f>] do_one_initcall+0x2f/0x160
Aug  1 19:42:57 frya kernel: [    4.463528]  [<c01e3d3a>] ? \
                create_proc_entry+0x5a/0xb0
Aug  1 19:42:57 frya kernel: [    4.467751]  [<c015cfe5>] ? \
                register_irq_proc+0xa5/0xc0
Aug  1 19:42:57 frya kernel: [    4.471924]  [<c015a1a8>] ? irq_to_desc+0x8/0x40
Aug  1 19:42:57 frya kernel: [    4.476110]  [<c015d075>] ? init_irq_proc+0x75/0x90
Aug  1 19:42:57 frya kernel: [    4.480343]  [<c07b95b5>] kernel_init+0x122/0x179
Aug  1 19:42:57 frya kernel: [    4.484616]  [<c07b9493>] ? kernel_init+0x0/0x179
Aug  1 19:42:57 frya kernel: [    4.488860]  [<c010397f>] \
                kernel_thread_helper+0x7/0x10
Aug  1 19:42:57 frya kernel: [    4.493137] ---[ end trace d2a1e0f2aa8609d9 ]---
Aug  1 19:42:57 frya kernel: [    4.497524] please boot with early_ioremap_debug and \
                report the dmesg.

and:

Aug  1 19:42:57 frya kernel: [    0.121015] * The chipset may have PM-Timer Bug. Due \
                to workarounds for a bug,
Aug  1 19:42:57 frya kernel: [    0.121020] * this clock source is slow. If you are \
                sure your timer does not have
Aug  1 19:42:57 frya kernel: [    0.121024] * this bug, please use "acpi_pm_good" to \
                disable the workaround
Aug  1 19:42:57 frya kernel: [    0.121563] * The chipset may have PM-Timer Bug. Due \
                to workarounds for a bug,
Aug  1 19:42:57 frya kernel: [    0.121568] * this clock source is slow. If you are \
                sure your timer does not have
Aug  1 19:42:57 frya kernel: [    0.121572] * this bug, please use "acpi_pm_good" to \
                disable the workaround

maybe it`s worth having a look at (on 2.6.29 and 2.6.30 system)

/sys/devices/system/clocksource/clocksource0/current_clocksource /sys/devices/system/clocksource/clocksource0/available_clocksource

and/or try specific clocksources with "clocksource=...." on boot (to see if it makes a difference). I have seen system freeze when the kernel detected an unstable clocksource and dynamically switched to another (more stable) one.

roland
________________________________________________________________
Neu: WEB.DE Doppel-FLAT mit Internet-Flatrate + Telefon-Flatrate für nur 19,99 Euro/mtl.!* http://produkte.web.de/go/02/

#1 lore raw
Quoted 4 lines · On Saturday 08 August 2009 11:49:50 [email protected] wrote:
Quoted 2 lines

Btw, since then there have been two more
reports on lkml of dual P2 systems locking up with 2.6.30++.

could you send pointers?

Since I am not subscribed to the LKML (I only read it via a NNTP relay) I have some difficulties providing IDs. Anyway, the subject of the reports I have in mind is "PROBLEM: crash / hang with linux-2.6.30.x on PII-350 SMP system".

Quoted 1 line

any existing bugzilla bugentry for these or for your`s ?

Earlier today I filed bug 13933 on kernel.bugzilla. Unfortunately I had some problems attaching all the files, so my earlier LKML post is still the best source of information.

Quoted 4 lines
Quoted 1 line

Network, sysrq and numlock are all unresponsive.

ok, that`s why we need to have the last informations from before the dying breath...

agreed.

Quoted 4 lines

btw, i one of your dmesg i found these:

Aug  1 19:42:57 frya kernel: [...]

Again, that's someone else's report. Regarding clocksource, while I was trying to hunt down this bug I once booted with "clocksource=acpi_pm" but it made no difference. Truth be told, at that time I was still troubleshooting the nvidia.ko driver.

Martin