Messages in this thread | | | Date | Sat, 11 Dec 1999 22:04:40 +0200 | From | Marius Aamodt Eriksen <> | Subject | Re: Kernel panics when using 64MB+ RAM |
| |
On Sat, Dec 11, 1999 at 09:07:05PM +0000, Alex Buell wrote: > On Sat, 11 Dec 1999, Marius Aamodt Eriksen wrote: > > > I've had a problem with my system, it goes like this: When I boot the > > system with my full amount of RAM (128 MB), it always crashes (at one > > point of another), a complete freezeup rather. It complaints about > > "Ther kernel is not able to handle null pointer at virtual address..." > > Sometimes, even MagicSysrq fails to save my day with this error. > > However, if I boot with 64MB RAM (linux mem=64M) the system works > > perfectly. First I thought it might have been some bad RAM, but > > somehow I doubt that would lock up the whole system, it just seems > > unlikely, especially with thought to how Linux is designed. > > Download memtest from http://www.qcc.sk.ca/~charlesc/software/memtester/ > > And make sure the memory *really* isn't bad. > > It saved my arse a few weeks ago from an expensive motherboard > replacement. > > Cheers, > Alex > -- > > Legalise cannabis today! Got GnuPG? Ask me for public key. > > http://www.tahallah.demon.co.uk - updated!
Hrm, how can I run memtest when I can't even allocate all the RAM without crashing, surely memtest can't allocate all 128 MB of it when it's not available from the kernel (hence malloc() won't do)...
But that really isn't the issue, because I _don't_ think (IMHO) that a little bad RAM will cause a kernel panic such as this.
Marius.
-- Marius Aamodt Eriksen linux.com - tuneup section ( http://tune.linux.com/ ) : marius@linux.com : mae@online.no : PGP Fingerprint = A8 75 CE 9A 9B 81 5A 4F 0D 60 39 71 CA 76 3A 17 : PGP Public Key @ http://marius.n3.net/key.asc [unhandled content-type:application/pgp-signature] | |