Unlucky 3.0-rt7

Unlucky 3.0-rt7

[Tracey Hytry]

I managed to download the new build about an hour and a half after it finished.
Installed it on the fedora 15 partition and the kernel fell on it's face!

It hangs up right after the start of boot, with just a few(maybe 50) lines of text.
I wrote down the last few lines, but don't have much time for setting things up for a serial console(don't know if it would help much here(but I should have the capability if needed)).

Anyway, here's the end of the text.  I'll drop it off on the list this time.

Memory:8089012k/9437184k ...more mem stuff
Preemptible hierarchical RCU implementation
RCU lockdep checking enabled
Verbose stalled-CPUs detection is disabled
NR_IRQS:33024 nr_irqs:512 16
Extended CMOS year: 2000
Console: colour VGA+ 80x25

That's it, and the computer would sit there lost in it's own nightmare getting very warm.
I tried putting "nomodeset rdblacklist=nouveau" on the kernel boot line, but got the same hang/message.
This is the same computer that was mentioned in the message from Fernanado  a few back.

Tracey.

Re: Unlucky 3.0-rt7

[Fernando Lopez-Lezcano]

On 08/05/2011 05:40 AM, Tracey Hytry wrote:
> I managed to download the new build about an hour and a half after it finished.
> Installed it on the fedora 15 partition and the kernel fell on it's face!
>
> It hangs up right after the start of boot, with just a few(maybe 50) lines of text.
> I wrote down the last few lines, but don't have much time for setting things up for a serial console(don't know if it would help much here(but I should have the capability if needed)).
>
> Anyway, here's the end of the text.  I'll drop it off on the list this time.
>
> Memory:8089012k/9437184k ...more mem stuff
> Preemptible hierarchical RCU implementation
> RCU lockdep checking enabled
> Verbose stalled-CPUs detection is disabled
> NR_IRQS:33024 nr_irqs:512 16
> Extended CMOS year: 2000
> Console: colour VGA+ 80x25
>
> That's it, and the computer would sit there lost in it's own nightmare getting very warm.
> I tried putting "nomodeset rdblacklist=nouveau" on the kernel boot line, but got the same hang/message.
> This is the same computer that was mentioned in the message from Fernanado  a few back.

I'll try it out on my test laptop (i386) when I get to ccrma.
(I just left the build going overnight and did not test at all)

-- Fernando

Re: Unlucky 3.0-rt7

[Fernando Lopez-Lezcano]

[-- Attachment #1: Type: text/plain, Size: 4125 bytes --]

On 08/05/2011 09:04 AM, Fernando Lopez-Lezcano wrote:
> On 08/05/2011 05:40 AM, Tracey Hytry wrote:
>> I managed to download the new build about an hour and a half after it
>> finished.
>> Installed it on the fedora 15 partition and the kernel fell on it's face!
>>
>> It hangs up right after the start of boot, with just a few(maybe 50)
>> lines of text.
>> I wrote down the last few lines, but don't have much time for setting
>> things up for a serial console(don't know if it would help much
>> here(but I should have the capability if needed)).
>>
>> Anyway, here's the end of the text. I'll drop it off on the list this
>> time.
>>
>> Memory:8089012k/9437184k ...more mem stuff
>> Preemptible hierarchical RCU implementation
>> RCU lockdep checking enabled
>> Verbose stalled-CPUs detection is disabled
>> NR_IRQS:33024 nr_irqs:512 16
>> Extended CMOS year: 2000
>> Console: colour VGA+ 80x25
>>
>> That's it, and the computer would sit there lost in it's own nightmare
>> getting very warm.
>> I tried putting "nomodeset rdblacklist=nouveau" on the kernel boot
>> line, but got the same hang/message.
>> This is the same computer that was mentioned in the message from
>> Fernanado a few back.
>
> I'll try it out on my test laptop (i386) when I get to ccrma.
> (I just left the build going overnight and did not test at all)

It (rt7 build) booted in my i386/fc15 laptop. But not clean:

[    0.000000] =============================================
[    0.000000] [ INFO: possible recursive locking detected ]
[    0.000000] 3.0.0-1.rt7.1.fc15.ccrma.i686.rtPAE #1
[    0.000000] ---------------------------------------------
[    0.000000] swapper/0 is trying to acquire lock:
[    0.000000]  (&parent->list_lock){+.+...}, at: [<c04fcba8>] 
__cache_free+0x43/0xc3
[    0.000000]
[    0.000000] but task is already holding lock:
[    0.000000]  (&parent->list_lock){+.+...}, at: [<c04fdcd1>] 
do_tune_cpucache+0xf2/0x2bb
[    0.000000]
[    0.000000] other info that might help us debug this:
[    0.000000]  Possible unsafe locking scenario:
[    0.000000]
[    0.000000]        CPU0
[    0.000000]        ----
[    0.000000]   lock(&parent->list_lock);
[    0.000000]   lock(&parent->list_lock);
[    0.000000]
[    0.000000]  *** DEADLOCK ***
[    0.000000]
[    0.000000]  May be due to missing lock nesting notation
[    0.000000]
[    0.000000] 3 locks held by swapper/0:
[    0.000000]  #0:  (cache_chain_mutex){+.+...}, at: [<c0bdd344>] 
kmem_cache_init_late+0x1e/0x6a
[    0.000000]  #1:  (&per_cpu(slab_lock, __cpu).lock){+.+...}, at: 
[<c04fc201>] __local_lock_irq+0x1e/0x5b
[    0.000000]  #2:  (&parent->list_lock){+.+...}, at: [<c04fdcd1>] 
do_tune_cpucache+0xf2/0x2bb
[    0.000000]
[    0.000000] stack backtrace:
[    0.000000] Pid: 0, comm: swapper Not tainted 
3.0.0-1.rt7.1.fc15.ccrma.i686.rtPAE #1
[    0.000000] Call Trace:
[    0.000000]  [<c0858e8b>] ? printk+0x2d/0x2f
[    0.000000]  [<c0475367>] __lock_acquire+0x805/0xb57
[    0.000000]  [<c0472f20>] ? lock_release_holdtime.part.10+0x4b/0x51
[    0.000000]  [<c08617ec>] ? _raw_spin_unlock+0x31/0x3d
[    0.000000]  [<c08606fd>] ? rt_spin_lock_slowlock+0x75/0x190
[    0.000000]  [<c04729df>] ? trace_hardirqs_off+0xb/0xd
[    0.000000]  [<c04fcba8>] ? __cache_free+0x43/0xc3
[    0.000000]  [<c0475b31>] lock_acquire+0xde/0x11d
[    0.000000]  [<c04fcba8>] ? __cache_free+0x43/0xc3
[    0.000000]  [<c0860d87>] rt_spin_lock+0x50/0x56
[    0.000000]  [<c04fcba8>] ? __cache_free+0x43/0xc3
[    0.000000]  [<c04fcba8>] __cache_free+0x43/0xc3
[    0.000000]  [<c0436d31>] ? test_ti_thread_flag+0x8/0x10
[    0.000000]  [<c04fc9e1>] kmem_cache_free+0x6c/0xdc
[    0.000000]  [<c04fcaa0>] slab_destroy+0x4f/0x53
[    0.000000]  [<c04fcb38>] free_block+0x94/0xc1
[    0.000000]  [<c04fdcea>] do_tune_cpucache+0x10b/0x2bb
[    0.000000]  [<c04fe075>] enable_cpucache+0x7b/0xa7
[    0.000000]  [<c0bdd355>] kmem_cache_init_late+0x2f/0x6a
[    0.000000]  [<c0bbd68c>] start_kernel+0x251/0x368
[    0.000000]  [<c0bbd1c4>] ? loglevel+0x18/0x18
[    0.000000]  [<c0bbd0ba>] i386_start_kernel+0xa9/0xaf

Full dmesg and config attached...
-- Fernando

[-- Attachment #2: dmesg.rt7.bz2 --]
[-- Type: application/x-bzip, Size: 20021 bytes --]

[-- Attachment #3: config.rt7.bz2 --]
[-- Type: application/x-bzip, Size: 28862 bytes --]

Re: Unlucky 3.0-rt7

[Peter Zijlstra]

On Fri, 2011-08-05 at 10:03 -0700, Fernando Lopez-Lezcano wrote:
> [    0.000000]  [<c04fc9e1>] kmem_cache_free+0x6c/0xdc
> [    0.000000]  [<c04fcaa0>] slab_destroy+0x4f/0x53
> [    0.000000]  [<c04fcb38>] free_block+0x94/0xc1
> [    0.000000]  [<c04fdcea>] do_tune_cpucache+0x10b/0x2bb
> [    0.000000]  [<c04fe075>] enable_cpucache+0x7b/0xa7
> [    0.000000]  [<c0bdd355>] kmem_cache_init_late+0x2f/0x6a
> [    0.000000]  [<c0bbd68c>] start_kernel+0x251/0x368
> [    0.000000]  [<c0bbd1c4>] ? loglevel+0x18/0x18
> [    0.000000]  [<c0bbd0ba>] i386_start_kernel+0xa9/0xaf 

Ah, I actually have a patch for that but didn't include it in -rt, only
pushed it upstream.. silly me. Let me include all those lockdep fixes.