Re: Memory on boot via int 0x15

From: Alan Cox (
Date: Fri May 26 2000 - 07:27:02 EST

> I would suggest looking at the new memory detection patches from David Parson
> and <insert name of guy from AMD>. They use the 0xe820? function (which
> is what Windows uses) for extended memory detection, and only fall back to
> 0xe801 if 0xe820 is not supported. Apparently, the 0xe801 support has bugs
> in some BIOSes (as you have found out).

This is the first BIOS I've seen that over-reports memory with 0xE801. We do
have a similar report of E820 over-reporting on one series of laptops that
needs a fix to go in. Newer machines dont always have E801. If windows finds
E820 it doesnt use E801 so why put it in the BIOS (sigh..).

We also ought to look at int 15 ah=c7, as that tells you the top of cacheable
ram which on some pentiums would be useful some point post 2.4 when the NUMA
allocator is in (since uncached ram isnt that dissimilar to ram on the wrong
node of a numa box - its a performance hit fallback)

To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to
Please read the FAQ at

This archive was generated by hypermail 2b29 : Wed May 31 2000 - 21:00:15 EST