On Saturday 2008-04-26 03:37, David C. Rankin wrote:
Vahe Avedissian wrote:
Folks,
I don't know if anyone has seen this behavior with memtest86+ (v1.70). I downloaded and tried the one from memtest's web page after hearing about some issues with the SuSE one.
I have 6x1GB sticks in my machine and memtest is chugging along running tests at address 7537.3GB and of course failing. It clearly displays it found 6143MB of memory. The question is why is it testing memory at addresses above what it knows are valid?
I can't shed any light on the problem, but it certainly seems bizarre. It may be an issue for a bug report/email to the memtest86+ folks. The most bytes you could ever see with 6 Gig. is 6 * ( 2 ^30) = 6,442,450,944 bytes. Or, defining a MB as 1000K or 1000 * ( 2 ^ 10) = 1,024,000. Then 6 GB would be 6,144MB or 6,144,000,000B. Well below the weird 7537.3GB.
Interestingly enough, I though of something very different, namely that the PCI map hole was accidentally counted. But the difference (7537-6144)=1393 would be a bit high, well, unless you have a bunch of hottest graphic cards. /proc/iomem: 257 MB bd800000-cd9fffff : PCI Bus #01 33 MB cdb00000-cfcfffff : PCI Bus #01 127 MB d0000000-d7ffffff : 0000:00:00.0 so that's roughly 417 MB of mapped PCI space for me. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org