I took one of my RAM chips out and moved the one from DIMM2 to DIMM3, so that I have DIMM1 & DIMM2 only filled now. The board supports dual channel memory, and I wonder if somehow it didn't like me putting different sizes in the same channel.
I just tried a kernel recompile (while still in X this time) and I had no issues (first time it got through a full compile with no hang). So it was either the chip or the configuration of using that chip with the others. The one I took out was an older pc2700 256mb chip. Thought I would post the result in case it helps anyone else. On 8/18/06, Andrew Robinson <[EMAIL PROTECTED]> wrote:
Just a note, that when the kernel build hangs, it seems to always happen at the same place. Seems very odd. Last 2 lines: SHIPPED drivers/scsi/aic7xxx/aic79xx_reg.h CC [M] drivers/scsi/aic7xxx/aic79xx_core.o 3 times it hung at the same line -Andrew On 8/18/06, Lennart Sorensen <[EMAIL PROTECTED]> wrote: > On Fri, Aug 18, 2006 at 02:02:10PM +1000, Hamish Moffatt wrote: > > On Thu, Aug 17, 2006 at 09:58:36PM -0600, Andrew Robinson wrote: > > > memtest86 test passed > > > > You can't have run it for long. It needs hours to be thorough > > (ie several passes). > > And even then many people have memtest86 pass and then after more > prodding and trying half the memory at a time in their system, they > discover that the problem only occours when one specific memory stick is > installed. > > memtest86 catches some errors, but far from all errors. A good kernel > compile or something else memory/cpu/disk intensive is usually the best > way to see if a system is actually stable or not. > > -- > Len Sorensen > > > -- > To UNSUBSCRIBE, email to [EMAIL PROTECTED] > with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED] > >
-- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]