AMD Processors
Decrease font size
Increase font size
Topic Title: Bug in 3500+ (Winchester) Stepping?
Topic Summary:
Created On: 08/23/2005 01:44 PM
Status: Read Only
Linear : Threading : Single : Branch
Search Topic Search Topic
Topic Tools Topic Tools
View similar topics View similar topics
View topic in raw text format. Print this topic.
 08/23/2005 01:44 PM
User is offline View Users Profile Print this message

Author Icon
nomenquis
Junior Member

Posts: 1
Joined: 08/23/2005

Hi,

I've quite a hard time with my AMD 3500+ Winchester Core (Cpu-z shows revision dh8-d0).

As soon as I use up the available memory bandwith by memsetting a 1 meg buffer in a loop my system will hang.

So far I've already changed everything (from memory to psu to motherboard) except the cpu.

I can easily hang my system by running a little program like this (running under linux 2.6.x in either 32 or 64 bit mode):
----
int main(int argc, char *argv[])
{
char buf[1024*1024];
int i;
for (i=0;i<1024*16*8*8;++i)
{
printf("%d\n",i);
memset(buf,0,1024*1024);
}
printf("Done\n");
return 0;
}
----

After about 1000 iterations in 64 bit mode and about 6-8000 in 32 bit mode my system will freeze.

This does not happen however if I add a slight delay into the loop or if I memcopy between two buffers.

So, after replacing almost all parts of my system I'd like to know if there possibly is a bug in my cpu version or what else I might try to change.

Currently I'm using an asus a8n-sli motherboard with 2 gigs of corsair memory.
The system is 100% stable during all normal operations, be it gaming in windows, compiling source for hours or whatever.

kind regards
Statistics
112018 users are registered to the AMD Processors forum.
There are currently 0 users logged in.

FuseTalk Hosting Executive Plan v3.2 - © 1999-2014 FuseTalk Inc. All rights reserved.



Contact AMD Terms and Conditions ©2007 Advanced Micro Devices, Inc. Privacy Trademark information