Re: [Suggestion] Memory check

R.E.Wolff@BitWizard.nl, Rogier Wolff (wolff@dutepp0.et.tudelft.nl)
Mon, 17 Aug 1998 09:23:58 +0200 (MEST)


Carsten Pluntke wrote:
>
> Due to a really nasty surprise with a faulty block of memory which caused
> some programs (especially X) to malfunction or to crash, would it be an
> idea to implement a memory check routine early in the
> booting/configuration sequence, maybe settable using a kernel option?

Nope.

Being the maintainer of the "unexpected failures due to malfunctioning
hardware FAQ" (commonly known as the sig11 FAQ), I know that less than
10% of the malfunctioning memory is found by a memory test. I
implemented ALL of the algorithms in the book "Testing Semiconductor
Memories", and none of them find more than a small percentage of the
failing memory cases.

If your memory is VERY bad, the BIOS will find it. It has a braindead
memory test, but it will find simple "stuck-at" errors.

If your memory is a little less bad, the system will crash quickly
during boot, before starting "init". There is no saying if it would
get to testing the memory. But in any case it didn't make it to
starting "init".

Once "init" is started, you can do it in "userspace". No need to put
it in the kernel.

Roger.

-- 
| The secret of success is sincerity.  Once you can |  R.E.Wolff@BitWizard.nl 
| fake that, you've got it made.  -- Jean Giraudoux |       T: +31-15-2137555 
- Custom Linux device drivers for sale! Call for a quote. - F: +31-15-2138217

- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html