2D test-video memory corruption

[GI Wan]

I found 2D test-video memory corruption failure when I was testing a Notebook computer UUT, this failure can be duplicated on both 1026 and 1028.

and I verified System module by Dell mpmemory and video memory by Nvdia test utility, no failures were found on both memory. can you help me with that?

PassMark BurnInTest Log file - http://www.passmark.com
================================================== ======
BurnInTest V5.3 Pro 1026
Logging detail level: Normal
******************
SYSTEM INFORMATION
******************
Computer Name: KFW10
Date: 09/18/08 17:44:17
Operating system: Windows XP Professional Service Pack 3 build 2600
Number of CPUs: 1 (2 Core(s)/CPU, 1 Logical(s)/Core)
CPU manufacturer: GenuineIntel
CPU type: Intel(R) Core(TM)2 Duo CPU T5850 @ 2.16GHz
CPU features: MMX SSE SSE2 SSE3 DEP PAE
CPU1 speed: 2165.9 MHz
CPU L2 Cache: 2 MB
RAM: 2046 MB
Video card: NVIDIA GeForce 8400M GS (Resolution: 1280x800x32)
Disk drive: Model ST9160310AS (Size: 149.0GB)
Optical drive: Slimtype DVDRW SSM-8515S

**************
RESULT SUMMARY
**************
Test Start time: Thu Sep 18 17:20:46 2008
Test Stop time: Thu Sep 18 17:40:50 2008
Test Duration: 000h 20m 04s
Test Name Cycles Operations Result Errors Last Error
CPU - Maths 499 11.847 Billion PASS 0 No errors
CPU - SIMD 192 7.525 Billion PASS 0 No errors
Memory (RAM) 0 1.352 Billion PASS 0 No errors
2D Graphics 0 959 FAIL 1 Video memory corruption
TEST RUN FAILED
******************
DETAILED EVENT LOG
******************
LOG NOTE: 2008-09-18 17:20:46, Status, Main Tests started
SERIOUS : 2008-09-18 17:32:53, 2D Graphics, Video memory corruption
LOG NOTE: 2008-09-18 17:32:53, 2D Graphics, 32 Bit ==> Offset 0xd5555555: Expected 0x55555555: Read 0xd5555555
LOG NOTE: 2008-09-18 17:40:50, Status, Test run stopped

[David (PassMark)]

Dell mpmemory only tests main RAM not video RAM. So you can't expect mpmemory to report on video RAM problems. I am not aware of any NVidia "test utility" that does write / read testing on video RAM. So I can't comment on that. (And would you trust NVidia tell you the full story about errors on NVidia hardware).

I think the error is a real hardware error, but one the user may never notice. See this past post on the topic.