Hello, I have had similar problems with Debian kernels post 2.6.5. I am running these kernels on a IBM Thinkcenter (I think a M50). The computer runs fine for a while after rebooting, but as soon as its left idle for a while, the computer slows down to a crawl. After that every keyboard or mouse click take 10-15s to process. Rebooting was so slow that I had to do a hard reset. But kernels previous to 2.6.5 work fine, and the 2.4 series kernels run excellent.
On Thu, 24 Feb 2005 20:44:35 -0600, astrid jurgensen <[EMAIL PROTECTED]> wrote: > Memtest86 did not find any errors. > I ran the standard tests 37 times using the MemMap=e820-Std default setting > and 12 times using the MemMap=Probed setting. In both cases, no errors were > detected. I also ran the bit-fade test (Test #9) 5 times with > MemMap=e820-Std. Again there were no errors. > When I tried the standard tests using the MemMap=820-All option, I got a huge > number of consecutive addresses reported as bad. this suggests that Memtest86 > is confused about the size of memory, and hence testing non-existent memory, > rather than the presence of a genuine memory problem. > > Note: > If I choose option 9 (Adv. Options) in the configuration menu of Memtest86, I > get the on screen response: > Chipset Not supported ! > Are these advanced options only for ECC support, or does the above message > mean that Memtest86 cannot handle my chipset, despite the claim that it is > designed to work for all chipsets? > > I have a ATi Radeon 9100 IGP chipset in the machine. > > Cheers, > Astrid. > . > > ---------- Original Message ---------------------------------- > >While it is possible that it has something to do with the debian > >installation, e.g. maybe the kernel has an option built-in which makes > >the computer unhappy, IME, random lock-ups of this type are almost > >always a sign of bad memory. > > > >I would highly recommend making a memtest86 boot CD-ROM and running > >memtest to see what happens. > > > >If you have access to some memory dimm's trying swapping out memory. > > > >if you have more than 1 dimm, try using one at a time and see if > >the behavior changes. > > > > > >brian > > > > > > > > ________________________________________________________________ > Sent via the WebMail system at mail.src.wisc.edu > > -- > To UNSUBSCRIBE, email to [EMAIL PROTECTED] > with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED] > > -- Gokul -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
