I've dropped Linus off the direct list. I think we should be able to debug
this one w/o his help. ;-)
Lines 594-606 of setup.c, which is the old section cause problems with
0-length regions, now properly handles them. This must be some new problem.
Could we have the e820 data out of /var/log/dmesg? head should work.
Nathan
-----Original Message-----
From: Jamie Lokier [mailto:lkd@tantalophile.demon.co.uk]
Sent: Wednesday, January 05, 2000 11:15 PM
To: Pavel Machek
Cc: torvalds@transmeta.com; David Ford; linux-kernel@vger.rutgers.edu;
Andre Hedrick; Zook, Nathan
Subject: Memory detection is still broken in 2.3.36
Pavel Machek wrote:
> > It's not the normal "I rebooted and have a few inconsistencies". fsck
> > complains about runs of contiguous inodes having mode 0177777, which
> > sometimes are actual files whose contents are lost. Sometimes fsck core
> > dumps (oh well).
> >
> > I imagine all-ones blocks are being written.
>
> That's your Toshiba, is it? Your BIOS tells ROM is writable. Patch
> your kernel to see it is wrong (patch attached, it will crash on
> boot). Then use something like mem=63M to fix it.
You're right. I tried your memory tester with 2.3.36 and it crashes if I
let the kernel detect memory. With mem=63M it's fine.
Do you know if this is the BIOS returning a bogus memory area (so we
have to test it and discard failing pages), or is it something that can
be fixed in the e820 code?
Your memory tester is a bit broken: the eaten amount is different to the
freed amount (with mem=63M):
Eating pages ....(61068K)
Freeing pages ....(61064K)
Nathan Zook wrote:
> Actually, the "famous" memory detection break IS fixed. If I had been
> paying better attention, it would have been fixed months ago instead of
> weeks.
As you can see, something is still broken with 2.3.36. It worked fine
with some older 2.3 kernel (probably before the e820 code).
Enjoy,
-- Jamie
-
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.tux.org/lkml/
This archive was generated by hypermail 2b29 : Fri Jan 07 2000 - 21:00:06 EST