From: Nishanth Aravamudan <nacc@us.ibm.com>
To: Christoph Lameter <clameter@sgi.com>
Cc: lee.schermerhorn@hp.com, Olaf Hering <olaf@aepfle.de>,
Linux MM <linux-mm@kvack.org>, Mel Gorman <mel@csn.ul.ie>,
linux-kernel@vger.kernel.org, linuxppc-dev@ozlabs.org,
Pekka Enberg <penberg@cs.helsinki.fi>,
"Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>,
akpm@linux-foundation.org,
KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Subject: Re: [PATCH] Fix boot problem in situations where the boot CPU is running on a memoryless node
Date: Wed, 23 Jan 2008 13:36:37 -0800 [thread overview]
Message-ID: <20080123213637.GE3848@us.ibm.com> (raw)
In-Reply-To: <Pine.LNX.4.64.0801231312580.15681@schroedinger.engr.sgi.com>
On 23.01.2008 [13:14:26 -0800], Christoph Lameter wrote:
> On Wed, 23 Jan 2008, Pekka Enberg wrote:
>
> > I think Mel said that their configuration did work with 2.6.23
> > although I also wonder how that's possible. AFAIK there has been some
> > changes in the page allocator that might explain this. That is, if
> > kmem_getpages() returned pages for memoryless node before, bootstrap
> > would have worked.
>
> Regular kmem_getpages is called with GFP_THISNODE set. There was some
> breakage in 2.6.22 and before with GFP_THISNODE returning pages from
> the wrong node if a node had no memory. So it may have worked
> accidentally and in an unsafe manner because the pages would have been
> associated with the wrong node which could trigger bug ons and locking
> troubles.
Right, so it might have functioned before, but the correctness was
wobbly at best... Certainly the memoryless patch series has tightened
that up, but we missed these SLAB issues.
I see that your patch fixed Olaf's machine, Pekka. Nice work on
everyone's part tracking this stuff down.
Thanks,
Nish
WARNING: multiple messages have this Message-ID (diff)
From: Nishanth Aravamudan <nacc@us.ibm.com>
To: Christoph Lameter <clameter@sgi.com>
Cc: Pekka Enberg <penberg@cs.helsinki.fi>, Mel Gorman <mel@csn.ul.ie>,
akpm@linux-foundation.org, linux-kernel@vger.kernel.org,
linuxppc-dev@ozlabs.org,
"Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>,
KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>,
lee.schermerhorn@hp.com, Linux MM <linux-mm@kvack.org>,
Olaf Hering <olaf@aepfle.de>
Subject: Re: [PATCH] Fix boot problem in situations where the boot CPU is running on a memoryless node
Date: Wed, 23 Jan 2008 13:36:37 -0800 [thread overview]
Message-ID: <20080123213637.GE3848@us.ibm.com> (raw)
In-Reply-To: <Pine.LNX.4.64.0801231312580.15681@schroedinger.engr.sgi.com>
On 23.01.2008 [13:14:26 -0800], Christoph Lameter wrote:
> On Wed, 23 Jan 2008, Pekka Enberg wrote:
>
> > I think Mel said that their configuration did work with 2.6.23
> > although I also wonder how that's possible. AFAIK there has been some
> > changes in the page allocator that might explain this. That is, if
> > kmem_getpages() returned pages for memoryless node before, bootstrap
> > would have worked.
>
> Regular kmem_getpages is called with GFP_THISNODE set. There was some
> breakage in 2.6.22 and before with GFP_THISNODE returning pages from
> the wrong node if a node had no memory. So it may have worked
> accidentally and in an unsafe manner because the pages would have been
> associated with the wrong node which could trigger bug ons and locking
> troubles.
Right, so it might have functioned before, but the correctness was
wobbly at best... Certainly the memoryless patch series has tightened
that up, but we missed these SLAB issues.
I see that your patch fixed Olaf's machine, Pekka. Nice work on
everyone's part tracking this stuff down.
Thanks,
Nish
WARNING: multiple messages have this Message-ID (diff)
From: Nishanth Aravamudan <nacc@us.ibm.com>
To: Christoph Lameter <clameter@sgi.com>
Cc: Pekka Enberg <penberg@cs.helsinki.fi>, Mel Gorman <mel@csn.ul.ie>,
akpm@linux-foundation.org, linux-kernel@vger.kernel.org,
linuxppc-dev@ozlabs.org,
"Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>,
KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>,
lee.schermerhorn@hp.com, Linux MM <linux-mm@kvack.org>,
Olaf Hering <olaf@aepfle.de>
Subject: Re: [PATCH] Fix boot problem in situations where the boot CPU is running on a memoryless node
Date: Wed, 23 Jan 2008 13:36:37 -0800 [thread overview]
Message-ID: <20080123213637.GE3848@us.ibm.com> (raw)
In-Reply-To: <Pine.LNX.4.64.0801231312580.15681@schroedinger.engr.sgi.com>
On 23.01.2008 [13:14:26 -0800], Christoph Lameter wrote:
> On Wed, 23 Jan 2008, Pekka Enberg wrote:
>
> > I think Mel said that their configuration did work with 2.6.23
> > although I also wonder how that's possible. AFAIK there has been some
> > changes in the page allocator that might explain this. That is, if
> > kmem_getpages() returned pages for memoryless node before, bootstrap
> > would have worked.
>
> Regular kmem_getpages is called with GFP_THISNODE set. There was some
> breakage in 2.6.22 and before with GFP_THISNODE returning pages from
> the wrong node if a node had no memory. So it may have worked
> accidentally and in an unsafe manner because the pages would have been
> associated with the wrong node which could trigger bug ons and locking
> troubles.
Right, so it might have functioned before, but the correctness was
wobbly at best... Certainly the memoryless patch series has tightened
that up, but we missed these SLAB issues.
I see that your patch fixed Olaf's machine, Pekka. Nice work on
everyone's part tracking this stuff down.
Thanks,
Nish
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
next prev parent reply other threads:[~2008-01-23 21:36 UTC|newest]
Thread overview: 177+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-01-15 15:09 crash in kmem_cache_init Olaf Hering
2008-01-15 15:58 ` Olaf Hering
2008-01-17 12:14 ` Pekka Enberg
2008-01-17 12:14 ` Pekka Enberg
2008-01-17 12:14 ` Pekka Enberg
2008-01-17 14:30 ` Christoph Lameter
2008-01-17 14:30 ` Christoph Lameter
2008-01-17 14:30 ` Christoph Lameter
2008-01-17 18:12 ` Olaf Hering
2008-01-17 18:12 ` Olaf Hering
2008-01-17 18:12 ` Olaf Hering
2008-01-17 18:58 ` Christoph Lameter
2008-01-17 18:58 ` Christoph Lameter
2008-01-17 18:58 ` Christoph Lameter
2008-01-17 19:54 ` Olaf Hering
2008-01-17 19:54 ` Olaf Hering
2008-01-17 20:20 ` Olaf Hering
2008-01-17 20:20 ` Olaf Hering
2008-01-17 20:20 ` Olaf Hering
2008-01-19 4:56 ` Christoph Lameter
2008-01-19 4:56 ` Christoph Lameter
2008-01-19 4:56 ` Christoph Lameter
2008-01-17 21:15 ` Olaf Hering
2008-01-17 21:15 ` Olaf Hering
2008-01-17 21:15 ` Olaf Hering
2008-01-18 6:56 ` Olaf Hering
2008-01-18 6:56 ` Olaf Hering
2008-01-18 6:56 ` Olaf Hering
2008-01-18 18:42 ` Christoph Lameter
2008-01-18 18:42 ` Christoph Lameter
2008-01-18 18:42 ` Christoph Lameter
2008-01-19 4:55 ` Christoph Lameter
2008-01-19 4:55 ` Christoph Lameter
2008-01-19 4:55 ` Christoph Lameter
2008-01-18 18:47 ` Christoph Lameter
2008-01-18 18:47 ` Christoph Lameter
2008-01-18 18:47 ` Christoph Lameter
2008-01-18 21:30 ` Mel Gorman
2008-01-18 21:30 ` Mel Gorman
2008-01-18 21:30 ` Mel Gorman
2008-01-18 21:43 ` Christoph Lameter
2008-01-18 21:43 ` Christoph Lameter
2008-01-18 21:43 ` Christoph Lameter
2008-01-18 22:16 ` Christoph Lameter
2008-01-18 22:16 ` Christoph Lameter
2008-01-18 22:16 ` Christoph Lameter
2008-01-18 22:19 ` Nish Aravamudan
2008-01-18 22:19 ` Nish Aravamudan
2008-01-18 22:19 ` Nish Aravamudan
2008-01-18 22:38 ` Christoph Lameter
2008-01-18 22:38 ` Christoph Lameter
2008-01-18 22:38 ` Christoph Lameter
2008-01-18 22:57 ` Olaf Hering
2008-01-18 22:57 ` Olaf Hering
2008-01-18 22:57 ` Olaf Hering
2008-01-22 19:54 ` Mel Gorman
2008-01-22 19:54 ` Mel Gorman
2008-01-22 19:54 ` Mel Gorman
2008-01-22 20:11 ` Christoph Lameter
2008-01-22 20:11 ` Christoph Lameter
2008-01-22 20:11 ` Christoph Lameter
2008-01-22 21:26 ` Mel Gorman
2008-01-22 21:26 ` Mel Gorman
2008-01-22 21:26 ` Mel Gorman
2008-01-22 21:34 ` Christoph Lameter
2008-01-22 21:34 ` Christoph Lameter
2008-01-22 21:34 ` Christoph Lameter
2008-01-22 22:50 ` Mel Gorman
2008-01-22 22:50 ` Mel Gorman
2008-01-22 22:57 ` Christoph Lameter
2008-01-22 22:57 ` Christoph Lameter
2008-01-22 22:57 ` Christoph Lameter
2008-01-22 23:10 ` Mel Gorman
2008-01-22 23:10 ` Mel Gorman
2008-01-22 23:10 ` Mel Gorman
2008-01-22 23:14 ` Christoph Lameter
2008-01-22 23:14 ` Christoph Lameter
2008-01-22 23:14 ` Christoph Lameter
2008-01-22 22:59 ` Pekka Enberg
2008-01-22 22:59 ` Pekka Enberg
2008-01-22 22:59 ` Pekka Enberg
2008-01-22 23:12 ` Christoph Lameter
2008-01-22 23:12 ` Christoph Lameter
2008-01-22 23:12 ` Christoph Lameter
2008-01-22 23:18 ` Christoph Lameter
2008-01-22 23:18 ` Christoph Lameter
2008-01-22 23:18 ` Christoph Lameter
2008-01-23 8:19 ` Pekka Enberg
2008-01-23 8:19 ` Pekka Enberg
2008-01-23 8:19 ` Pekka Enberg
2008-01-23 8:40 ` Olaf Hering
2008-01-23 8:40 ` Olaf Hering
2008-01-23 8:40 ` Olaf Hering
2008-01-22 21:45 ` Olaf Hering
2008-01-22 21:45 ` Olaf Hering
2008-01-22 21:45 ` Olaf Hering
2008-01-22 22:12 ` Nish Aravamudan
2008-01-22 22:12 ` Nish Aravamudan
2008-01-22 22:12 ` Nish Aravamudan
2008-01-22 22:23 ` Christoph Lameter
2008-01-22 22:23 ` Christoph Lameter
2008-01-22 22:23 ` Christoph Lameter
2008-01-23 7:58 ` Olaf Hering
2008-01-23 7:58 ` Olaf Hering
2008-01-23 7:58 ` Olaf Hering
2008-01-23 10:50 ` Mel Gorman
2008-01-23 10:50 ` Mel Gorman
2008-01-23 10:50 ` Mel Gorman
2008-01-23 12:14 ` Olaf Hering
2008-01-23 12:14 ` Olaf Hering
2008-01-23 12:14 ` Olaf Hering
2008-01-23 12:52 ` Olaf Hering
2008-01-23 12:52 ` Olaf Hering
2008-01-23 12:52 ` Olaf Hering
2008-01-23 13:55 ` [PATCH] Fix boot problem in situations where the boot CPU is running on a memoryless node Mel Gorman
2008-01-23 13:55 ` Mel Gorman
2008-01-23 13:55 ` Mel Gorman
2008-01-23 14:18 ` Pekka J Enberg
2008-01-23 14:18 ` Pekka J Enberg
2008-01-23 14:18 ` Pekka J Enberg
2008-01-23 14:32 ` Pekka J Enberg
2008-01-23 14:32 ` Pekka J Enberg
2008-01-23 14:32 ` Pekka J Enberg
2008-01-23 14:49 ` Pekka J Enberg
2008-01-23 14:49 ` Pekka J Enberg
2008-01-23 14:49 ` Pekka J Enberg
2008-01-23 15:56 ` Mel Gorman
2008-01-23 15:56 ` Mel Gorman
2008-01-23 15:56 ` Mel Gorman
2008-01-23 17:29 ` Pekka J Enberg
2008-01-23 17:29 ` Pekka J Enberg
2008-01-23 17:29 ` Pekka J Enberg
2008-01-23 17:42 ` Pekka J Enberg
2008-01-23 17:42 ` Pekka J Enberg
2008-01-23 17:42 ` Pekka J Enberg
2008-01-23 18:51 ` Christoph Lameter
2008-01-23 18:51 ` Christoph Lameter
2008-01-23 18:51 ` Christoph Lameter
2008-01-23 19:52 ` Nishanth Aravamudan
2008-01-23 19:52 ` Nishanth Aravamudan
2008-01-23 19:52 ` Nishanth Aravamudan
2008-01-23 21:02 ` Pekka Enberg
2008-01-23 21:02 ` Pekka Enberg
2008-01-23 21:02 ` Pekka Enberg
2008-01-23 21:14 ` Christoph Lameter
2008-01-23 21:14 ` Christoph Lameter
2008-01-23 21:14 ` Christoph Lameter
2008-01-23 21:36 ` Nishanth Aravamudan [this message]
2008-01-23 21:36 ` Nishanth Aravamudan
2008-01-23 21:36 ` Nishanth Aravamudan
2008-01-24 3:13 ` Christoph Lameter
2008-01-24 3:13 ` Christoph Lameter
2008-01-24 3:13 ` Christoph Lameter
2008-01-23 18:36 ` Christoph Lameter
2008-01-23 18:36 ` Christoph Lameter
2008-01-23 18:36 ` Christoph Lameter
2008-01-23 18:35 ` Christoph Lameter
2008-01-23 18:35 ` Christoph Lameter
2008-01-23 18:35 ` Christoph Lameter
2008-01-23 14:27 ` Olaf Hering
2008-01-23 14:27 ` Olaf Hering
2008-01-23 14:27 ` Olaf Hering
2008-01-23 14:42 ` Mel Gorman
2008-01-23 14:42 ` Mel Gorman
2008-01-23 14:42 ` Mel Gorman
2008-01-23 18:41 ` Christoph Lameter
2008-01-23 18:41 ` Christoph Lameter
2008-01-23 18:41 ` Christoph Lameter
2008-01-23 13:41 ` crash in kmem_cache_init Mel Gorman
2008-01-23 13:41 ` Mel Gorman
2008-01-23 13:41 ` Mel Gorman
2008-01-18 18:51 ` Christoph Lameter
2008-01-18 18:51 ` Christoph Lameter
2008-01-18 18:51 ` Christoph Lameter
2008-01-17 19:03 ` Christoph Lameter
2008-01-17 19:03 ` Christoph Lameter
2008-01-17 19:03 ` Christoph Lameter
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=20080123213637.GE3848@us.ibm.com \
--to=nacc@us.ibm.com \
--cc=akpm@linux-foundation.org \
--cc=aneesh.kumar@linux.vnet.ibm.com \
--cc=clameter@sgi.com \
--cc=kamezawa.hiroyu@jp.fujitsu.com \
--cc=lee.schermerhorn@hp.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=linuxppc-dev@ozlabs.org \
--cc=mel@csn.ul.ie \
--cc=olaf@aepfle.de \
--cc=penberg@cs.helsinki.fi \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.