Re: 2.6.17-rc4-mm1

[Eric Dumazet <dada1@cosmosbay.com>]

Hi Andrew

It seems latest kernels have a problem in kmem_cache_destroy()

On a dual opteron machine (NUMA), its quite easy do trigger the bug : 
doing a oprofile session like :

opcontrol --setup --event=CPU_CLK_UNHALTED:100000 
--vmlinux=/usr/src/linux-2.6.17-rc4-mm1/vmlinux
...
opcontrol --dump
...
opcontrol --deinit    <<<-- Triggers the bug

slab error in kmem_cache_destroy(): cache `dcookie_cache': Can't free 
all objects

Call Trace: <ffffffff80278aa4>{kmem_cache_destroy+212}
      <ffffffff802bc559>{dcookie_unregister+345} 
<ffffffff803bbe7a>{event_buffer_release+26}
      <ffffffff8027cee2>{__fput+98} <ffffffff80279ecd>{filp_close+93}
      <ffffffff8023003e>{put_files_struct+110} 
<ffffffff8023143a>{do_exit+650}
      <ffffffff802f58f1>{__up_write+33} 
<ffffffff80231bb8>{do_group_exit+200}
      <ffffffff802097ce>{system_call+126}

# grep dcookie_cache /proc/slabinfo
dcookie_cache          0      0     32  101    1 : tunables  120   60    
8 : slabdata      0      0      0


This problem is annoying, because in the oprofile case, we must reboot 
the machine to be able to start a new profile session.

Eric



-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/