Re: slap corruption

2009-05-19 Thread Felipe Contreras 
On Tue, May 19, 2009 at 4:56 AM, Guzman Lugo, Fernando x0095...@ti.com wrote:



 Hi everyone,

 We have been looking into the Slab Corruption error and we have found the 
 problem occurs after an application has been terminated abnormally, e.g. 
 Using CTL-C or Segmentation Fault or kill.

 The issue is observed not only for DMM test cases but for others test cases 
 that don't use DMM.

 The problem is reported by 'slab.c' when a slab object which is inactive, it 
 supposed to be filled with POISON_FREE (6b) and has some 00 within the 
 object, this seems to be corrupted.

 Slab corruption: size-64 start=c6bc71e0, len=64
 3Redzone: 0x9f911029d74e35b/0x9f911029d74e35b.
 Redzone: 0x9f911029d74e35b/0x9f911029d74e35b.
 3Last user: [bf063464]Last user: [bf063464](MEM_Free+0x9c/0xe0 
 [bridgedriv
 er])(MEM_Free+0x9c/0xe0 [bridgedriver])

 3010:010: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 
 6b
  6b 00 00 00 00 00 00 00 00


 it seems it is not being freed correctly, because instead of being filled 
 only with 6b it has 00 maybe other process is overwriting it.

 Does anyone have any idea of what can be happening?

Looping Hiroshi who is working on a similar issue... perhaps the same?

-- 
Felipe Contreras
--
To unsubscribe from this list: send the line unsubscribe linux-omap in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

RE: slap corruption

2009-05-19 Thread Guzman Lugo, Fernando 

Hi Hiroshi,

Yes, I can reproduce this running the ping.out sample, this are the 
steps:

# insmod dspbridge.ko
# insmod bridgedriver.ko
# ./cexec.out ddspbase_tiomap3430.dof64P
# ./ping.out 100
^C
# ./ping.out 10

Thanks and regards,
Fernando.

-Original Message-
From: Hiroshi DOYU [mailto:hiroshi.d...@nokia.com] 
Sent: Tuesday, May 19, 2009 6:27 AM
To: felipe.contre...@gmail.com
Cc: Guzman Lugo, Fernando; Kanigeri, Hari; Gupta, Ramesh; Menon, Nishanth; 
Ramirez Luna, Omar; Kargirwar, Pankaj; linux-omap@vger.kernel.org
Subject: Re: slap corruption

From: ext Felipe Contreras felipe.contre...@gmail.com
Subject: Re: slap corruption
Date: Tue, 19 May 2009 11:47:11 +0200

 On Tue, May 19, 2009 at 4:56 AM, Guzman Lugo, Fernando x0095...@ti.com 
 wrote:
 
 
 
  Hi everyone,
 
  We have been looking into the Slab Corruption error and we have found the 
  problem occurs after an application has been terminated abnormally, e.g. 
  Using CTL-C or Segmentation Fault or kill.
 
  The issue is observed not only for DMM test cases but for others test cases 
  that don't use DMM.
 
  The problem is reported by 'slab.c' when a slab object which is inactive, 
  it supposed to be filled with POISON_FREE (6b) and has some 00 within the 
  object, this seems to be corrupted.
 
  Slab corruption: size-64 start=c6bc71e0, len=64
  3Redzone: 0x9f911029d74e35b/0x9f911029d74e35b.
  Redzone: 0x9f911029d74e35b/0x9f911029d74e35b.
  3Last user: [bf063464]Last user: [bf063464](MEM_Free+0x9c/0xe0 
  [bridgedriv
  er])(MEM_Free+0x9c/0xe0 [bridgedriver])
 
  3010:010: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 
  6b 6b
   6b 00 00 00 00 00 00 00 00
 
 
  it seems it is not being freed correctly, because instead of being filled 
  only with 6b it has 00 maybe other process is overwriting it.
 
  Does anyone have any idea of what can be happening?
 
 Looping Hiroshi who is working on a similar issue... perhaps the same?

Thanks;) Maybe the same one, I guess.

 
 -- 
 Felipe Contreras

--
To unsubscribe from this list: send the line unsubscribe linux-omap in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html