mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Jeff Lessem <Jeff.Lessem@Colorado.EDU>
To: andrewm@uow.edu.au (Andrew Morton)
Cc: linux-kernel@vger.kernel.org
Subject: Re: Too much memory causes crash when reading/writing to disk
Date: Tue, 17 Jul 2001 07:22:42 -0600	[thread overview]
Message-ID: <200107171322.HAA245907@ibg.colorado.edu> (raw)

Andrew Morton wrote:
>> 
>> I have done a bit more work on the problem I reported in my message
>> "Crashes reading and writing to disk".  To recap, on a machine with
>> 8GB of RAM, either
>> 
>> dd if=/dev/zero bs=1G count=10 | split -b 1073741824
>> 
>> or
>> 
>> find /bigfulldisk -type f -exec cat {} \; > /dev/null
>> 
>> can reliably cause a crash.
>
>It seems that one of your CPUs is stuck in an interrupt
>routine.  Could you please try running with the below
>patch?  Feed the output through ksymoops.

I tried the patch, but the machine came up in a very confused state.
It couldn't mount proc, and other badness.  I made the patch against
2.4.6, because 2.4.7-pre6 doesn't boot at all (I guess I should send
another message about that problem).

>Also (but separately) try enabling the NMI watchdog with
>the `nmi_watchdog=1' kernel boot parameter.

This worked, and I recreated the crash:

ksymoops 2.4.1 on i686 2.4.6.  Options used
     -V (default)
     -k /proc/ksyms (default)
     -l /proc/modules (default)
     -o /lib/modules/2.4.6/ (default)
     -m /boot/System.map-2.4.6 (default)

Warning: You did not tell me where to find symbol information.  I will
assume that the log matches the kernel and modules that are running
right now and I'll use the default options above for symbol resolution.
If the current kernel and/or modules do not match the log, you can get
more accurate output by telling me the kernel version and where to find
map, modules, ksyms etc.  ksymoops -h explains the options.

7552MB HIGHMEM available.
activating NMI Watchdog ... done.
cpu: 0, clocks: 999966, slice: 111107
cpu: 5, clocks: 999966, slice: 111107
cpu: 6, clocks: 999966, slice: 111107
cpu: 7, clocks: 999966, slice: 111107
cpu: 3, clocks: 999966, slice: 111107
cpu: 2, clocks: 999966, slice: 111107
cpu: 1, clocks: 999966, slice: 111107
cpu: 4, clocks: 999966, slice: 111107
NMI Watchdog detected LOCKUP on CPU5, registers:
CPU:    5
EIP:    0010:[<c010b50f>]
Using defaults from ksymoops -t elf32-i386 -a i386
EFLAGS: 00000046
eax: 00000080   ebx: c9c93f7c   ecx: 04b058ee   edx: 000019c0
esi: 20000001   edi: 000000a0   ebp: 00000000   esp: c9c93f2c
ds: 0018   es: 0018   ss: 0018
Process swapper (pid: 0, stackpage=c9c93000)
Stack: c0247e28 c01083b1 00000000 00000000 c9c93f7c c02af980 c0291800 00000000 
       c9c93f74 c0108596 00000000 c9c93f7c c0247e28 c0105180 c9c92000 c0105180 
       000000a0 c0247e28 00000000 c0106d04 c0105180 00000020 c9c92000 c9c92000 
Call Trace: [<c01083b1>] [<c0108596>] [<c0105180>] [<c0105180>] [<c0106d04>] [<c0105180>] [<c0105180>] 
       [<c01051ac>] [<c0105212>] [<c0114e66>] 
Code: c6 05 00 79 24 c0 01 53 e8 80 09 01 00 83 c4 04 83 3d 88 f2 

>>EIP; c010b50f <timer_interrupt+9b/130>   <=====
Trace; c01083b1 <handle_IRQ_event+4d/78>
Trace; c0108596 <do_IRQ+a6/ec>
Trace; c0105180 <default_idle+0/34>
Trace; c0105180 <default_idle+0/34>
Trace; c0106d04 <ret_from_intr+0/7>
Trace; c0105180 <default_idle+0/34>
Trace; c0105180 <default_idle+0/34>
Trace; c01051ac <default_idle+2c/34>
Trace; c0105212 <cpu_idle+3e/54>
Trace; c0114e66 <printk+16e/17c>
Code;  c010b50f <timer_interrupt+9b/130>
00000000 <_EIP>:
Code;  c010b50f <timer_interrupt+9b/130>   <=====
   0:   c6 05 00 79 24 c0 01      movb   $0x1,0xc0247900   <=====
Code;  c010b516 <timer_interrupt+a2/130>
   7:   53                        push   %ebx
Code;  c010b517 <timer_interrupt+a3/130>
   8:   e8 80 09 01 00            call   1098d <_EIP+0x1098d> c011be9c <do_timer+0/30>
Code;  c010b51c <timer_interrupt+a8/130>
   d:   83 c4 04                  add    $0x4,%esp
Code;  c010b51f <timer_interrupt+ab/130>
  10:   83 3d 88 f2 00 00 00      cmpl   $0x0,0xf288


1 warning issued.  Results may not be reliable.

             reply	other threads:[~2001-07-17 13:23 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2001-07-17 13:22 Jeff Lessem [this message]
2001-07-17 14:00 ` Andrew Morton
2001-07-17 16:15   ` Jeff Lessem
2001-07-18  0:31     ` Andrew Morton
  -- strict thread matches above, loose matches on Subject: below --
2001-07-20 16:17 Matt_Domsch
2001-07-19 12:36 Jeff Lessem
2001-07-18 16:17 Jeff Lessem
2001-07-19  6:33 ` Andrew Morton
2001-07-16 15:39 Jeff Lessem
2001-07-17  1:07 ` Andrew Morton

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=200107171322.HAA245907@ibg.colorado.edu \
    --to=jeff.lessem@colorado.edu \
    --cc=andrewm@uow.edu.au \
    --cc=linux-kernel@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®