Messages in this thread | | | From | "Ulrich Windl" <> | Date | Mon, 22 Apr 1996 14:50:45 +0200 | Subject | Re: memtest86, built into kernel |
| |
On 22 Apr 96 at 9:35, Alan Cox wrote:
> > >2) A memory tester has no place in the kernel. > > Why not? It was a selling point for mainframes in days past. > > Big companies rely on their computers to give correct answers. > > Having a memory checker constanly running is one such way they > > helped guaranteed correctness. (it ran during the idle loop if > > I recall) > > No mainframes ran with ECC memory boards, so hardware faults got logged. > PC's _had_ parity to show you errors in RAM. Now they don't bother (I guess > because the people who make the PC's make so much bad memory they dont want > you to find out).
I think the problem with the PC parity detection was that it is practically useless, because you can't identify the location of the error. (I've seen a SunOS 4 message mapping out defective RAM, new HP hard- and software can also do this -- what about Linux; is it ready to do so (better hardware assumed)?)
> > Various things have run in idle loops - eg estimations of e. > > Alan ------------ Ulrich Windl Klinikum der Universitaet Regensburg Rechenzentrum DV-med Franz-Josef-Strauss-Allee 11 Tel: +49 941 944-5879 D-93053 Regensburg FAX: +49 941 944-5882 "The bad PC memory subsystems are really bad." (Linus Torvalds on 19.04.1996)
| |