From: Mel Gorman <mel@csn.ul.ie>
To: Andi Kleen <andi@firstfloor.org>
Cc: mingo@elte.hu, linux-mm@kvack.org, linux-kernel@vger.kernel.org,
apw@shadowen.org
Subject: Re: [PATCH 0/2] Relax restrictions on setting CONFIG_NUMA on x86
Date: Wed, 23 Jan 2008 10:57:57 +0000 [thread overview]
Message-ID: <20080123105757.GE21455@csn.ul.ie> (raw)
In-Reply-To: <200801231145.14915.andi@firstfloor.org>
On (23/01/08 11:45), Andi Kleen didst pronounce:
>
> > > i386 already has srat parsing code (just written in a horrible hackish
> > > way); but it exists arch/x86/kernel/srat_32.c
> >
> > Yes, I spotted that. Enabling it required a Kconfig change
>
> does it?
hmm, just a removal of (X86_SUMMIT || X86_GENERICARCH). When I was testing
this I was probably using a PC-compatible configuration as that is what I
was enabling at the time. It's a similar restriction for BOOT_IOREMAP.
> I was pretty sure that a !NUMAQ i386 CONFIG_NUMA build
> already used that.
Presumably a GENERICARCH configuration
> At least that was the case when I last looked. If that
> has changed it must have bitrotted recently.
>
I don't think it has bit-rotted. I was just not using a GENERICARCH
configuration.
> > or two and
> > enabling BOOT_IOREMAP. It then crashes early in boot on a call to strlen()
> > so I went with the stubs and SRAT disabled for the moment.
>
> Crashed on a Opteron box? That was always the case
>
Yeah, elm3b6 from test.kernel.org is an Opteron-based NUMA machine. Oddly
it's good to know that the nature of the crash has not changed.
> If it crashed on a (older) Summit then it likely bitrotted, because that
> worked at some point. The code was originally written by Pat G. for Summit1
> and I believe was at least used by some people (no distributions) for S2,
> possible 3 too.
>
Unfortunately, I don't have access to a Summit box so I cannot verify.
> > Ok, understood. When I next revisit this, I'll look at making ACPI_SRAT
> > and BOOT_IOREMAP work on normal machines and see what happens. Thanks.
>
> Again the problem shouldn't be normal machines, but non Summit NUMA systems.
>
Grand. I'll get back to this soon and see what can be done.
--
Mel Gorman
Part-time Phd Student Linux Technology Center
University of Limerick IBM Dublin Software Lab
next prev parent reply other threads:[~2008-01-23 10:58 UTC|newest]
Thread overview: 35+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-01-18 15:35 [PATCH 0/2] Relax restrictions on setting CONFIG_NUMA on x86 Mel Gorman
2008-01-18 15:35 ` [PATCH 1/2] Do not require CONFIG_HIGHMEM64G to set " Mel Gorman
2008-01-18 16:17 ` Ingo Molnar
2008-01-18 15:36 ` [PATCH 2/2] Allow any x86 sub-architecture type to set CONFIG_NUMA Mel Gorman
2008-01-18 16:17 ` Ingo Molnar
2008-01-19 6:35 ` [PATCH 0/2] Relax restrictions on setting CONFIG_NUMA on x86 Andi Kleen
2008-01-19 16:07 ` Mel Gorman
2008-01-22 12:14 ` Ingo Molnar
2008-01-22 12:29 ` Mel Gorman
2008-01-22 15:39 ` Ingo Molnar
2008-01-22 13:33 ` Andi Kleen
2008-01-23 10:28 ` Mel Gorman
2008-01-23 10:45 ` Andi Kleen
2008-01-23 10:57 ` Mel Gorman [this message]
2008-01-23 11:11 ` Andi Kleen
2008-01-23 11:15 ` [PATCH 0/2] Relax restrictions on setting CONFIG_NUMA on x86 II Andi Kleen
2008-01-23 11:24 ` Mel Gorman
2008-01-23 13:48 ` Andi Kleen
2008-01-23 14:15 ` Mel Gorman
2008-01-21 0:38 ` [PATCH 0/2] Relax restrictions on setting CONFIG_NUMA on x86 KOSAKI Motohiro
2008-01-21 14:35 ` Mel Gorman
2008-01-21 14:49 ` Ingo Molnar
2008-01-21 16:27 ` Mel Gorman
2008-01-23 2:04 ` KOSAKI Motohiro
2008-01-23 10:22 ` Mel Gorman
2008-01-24 3:19 ` KOSAKI Motohiro
2008-01-23 10:23 ` Mel Gorman
2008-01-26 14:10 ` KOSAKI Motohiro
2008-01-26 17:10 ` KOSAKI Motohiro
2008-01-26 17:18 ` Mel Gorman
2008-01-27 6:54 ` KOSAKI Motohiro
2008-01-28 15:02 ` Ingo Molnar
2008-01-29 2:57 ` KOSAKI Motohiro
2008-01-29 11:34 ` Mel Gorman
2008-02-03 9:32 ` KOSAKI Motohiro
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20080123105757.GE21455@csn.ul.ie \
--to=mel@csn.ul.ie \
--cc=andi@firstfloor.org \
--cc=apw@shadowen.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mingo@elte.hu \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).