From: Eric Dumazet <eric.dumazet@gmail.com>
To: Tejun Heo <tj@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>,
Brian Gerst <brgerst@gmail.com>,
x86@kernel.org, linux-kernel@vger.kernel.org,
torvalds@linux-foundation.org, mingo@elte.hu
Subject: Re: [PATCH] x86-32: Allocate irq stacks seperate from percpu area
Date: Wed, 27 Oct 2010 15:57:50 +0200 [thread overview]
Message-ID: <1288187870.2709.128.camel@edumazet-laptop> (raw)
In-Reply-To: <4CC82C2F.1020707@kernel.org>
Le mercredi 27 octobre 2010 à 15:42 +0200, Tejun Heo a écrit :
> Hello,
>
> On 10/27/2010 03:33 PM, Eric Dumazet wrote:
> > Le mercredi 27 octobre 2010 à 11:57 +0200, Peter Zijlstra a écrit :
> >> On Wed, 2010-10-27 at 08:07 +0200, Eric Dumazet wrote:
> >>>> - irqctx = &per_cpu(hardirq_stack, cpu);
> >>>> + irqctx = (union irq_ctx *)__get_free_pages(THREAD_FLAGS, THREAD_ORDER);
> >>>
> >>> Hmm, then we lose NUMA affinity for stacks.
> >>
> >> I guess we could use:
> >>
> >> alloc_pages_node(cpu_to_node(cpu), THREAD_FLAGS, THREAD_ORDER);
> >>
> >>
> >
> > Anyway, I just discovered per_cpu data on my machine (NUMA capable) all
> > sit on a single node, if 32bit kernel used.
> >
> > # cat /proc/buddyinfo
> > Node 0, zone DMA 0 1 0 1 2 1 1 0 1 1 3
> > Node 0, zone Normal 94 251 81 16 3 2 1 2 1 2 187
> > Node 0, zone HighMem 113 88 47 36 18 5 4 3 2 0 268
> > Node 1, zone HighMem 154 97 43 16 9 4 3 2 3 2 482
> ...
> >
> > I presume node 1 having only HighMem could be the reason ?
>
> What does cpu_to_node() on each cpu say? Also, do you know why
> num_possible_cpus() is 32, not 16?
>
I dont know, machine is HP ProLiant BL460c G6
[ 0.000000] SMP: Allowing 32 CPUs, 16 hotplug CPUs
for_each_possible_cpu(cpu) {
pr_err("cpu=%d node=%d\n", cpu, cpu_to_node(cpu));
}
cpu=0 node=1
cpu=1 node=0
cpu=2 node=1
cpu=3 node=0
cpu=4 node=1
cpu=5 node=0
cpu=6 node=1
cpu=7 node=0
cpu=8 node=1
cpu=9 node=0
cpu=10 node=1
cpu=11 node=0
cpu=12 node=1
cpu=13 node=0
cpu=14 node=1
cpu=15 node=0
cpu=16 node=0
cpu=17 node=0
cpu=18 node=0
cpu=19 node=0
cpu=20 node=0
cpu=21 node=0
cpu=22 node=0
cpu=23 node=0
cpu=24 node=0
cpu=25 node=0
cpu=26 node=0
cpu=27 node=0
cpu=28 node=0
cpu=29 node=0
cpu=30 node=0
cpu=31 node=0
next prev parent reply other threads:[~2010-10-27 13:57 UTC|newest]
Thread overview: 59+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-10-25 22:41 [PATCH] MN10300: Fix the PERCPU() alignment to allow for workqueues David Howells
2010-10-26 9:10 ` Tejun Heo
2010-10-26 10:22 ` David Howells
2010-10-26 12:14 ` Tejun Heo
2010-10-26 12:27 ` Tejun Heo
2010-10-26 12:45 ` [PATCH] x86, percpu: revert commit fe8e0c25 Tejun Heo
2010-10-26 13:25 ` Ingo Molnar
2010-10-26 13:34 ` Tejun Heo
2010-10-26 13:49 ` Brian Gerst
2010-10-26 15:08 ` Linus Torvalds
2010-10-27 5:43 ` [PATCH] x86-32: Allocate irq stacks seperate from percpu area Brian Gerst
2010-10-27 6:07 ` Eric Dumazet
2010-10-27 9:57 ` Peter Zijlstra
2010-10-27 13:33 ` Eric Dumazet
2010-10-27 13:42 ` Tejun Heo
2010-10-27 13:57 ` Eric Dumazet [this message]
2010-10-27 14:00 ` Tejun Heo
2010-10-27 14:24 ` Eric Dumazet
2010-10-27 14:39 ` Tejun Heo
2010-10-27 14:39 ` Eric Dumazet
2010-10-27 14:43 ` Tejun Heo
2010-10-27 15:21 ` Eric Dumazet
2010-10-27 15:35 ` Tejun Heo
2010-10-27 16:07 ` Eric Dumazet
2010-10-27 17:33 ` [PATCH] numa: fix slab_node(MPOL_BIND) Eric Dumazet
2010-10-28 15:59 ` Linus Torvalds
2010-10-28 16:27 ` Eric Dumazet
2010-10-28 16:45 ` Mel Gorman
2010-10-28 16:55 ` Christoph Lameter
2010-10-28 21:07 ` Andrew Morton
2010-10-29 14:55 ` Christoph Lameter
2010-10-27 20:55 ` [PATCH] x86-32: Allocate irq stacks seperate from percpu area Eric Dumazet
2010-10-28 12:01 ` Tejun Heo
2010-10-28 12:30 ` Eric Dumazet
2010-10-28 14:40 ` [PATCH] x86-32: NUMA irq stacks allocations Eric Dumazet
2010-10-29 6:43 ` [tip:x86/urgent] x86-32: Restore irq stacks NUMA-aware allocations tip-bot for Eric Dumazet
2010-10-29 18:32 ` Peter Zijlstra
2010-10-29 20:09 ` Cyrill Gorcunov
2010-10-29 20:28 ` Cyrill Gorcunov
2010-10-29 20:53 ` Eric Dumazet
2010-10-29 20:59 ` Cyrill Gorcunov
2010-10-29 20:58 ` Eric Dumazet
2010-10-29 21:21 ` Cyrill Gorcunov
2010-10-27 15:19 ` [PATCH] x86-32: Allocate irq stacks seperate from percpu area Linus Torvalds
2010-10-27 15:30 ` Ingo Molnar
2010-10-27 15:33 ` Ingo Molnar
2010-10-27 15:40 ` Tejun Heo
2010-10-27 15:43 ` Ingo Molnar
2010-10-27 16:03 ` [tip:x86/urgent] " tip-bot for Brian Gerst
2010-10-27 16:04 ` [tip:x86/urgent] percpu: Remove the multi-page alignment facility tip-bot for Ingo Molnar
2010-10-26 14:06 ` [RFC PATCH] percpu: always align percpu output section to PAGE_SIZE Tejun Heo
2011-03-24 6:46 ` [Uclinux-dist-devel] " Mike Frysinger
2011-03-24 8:25 ` Tejun Heo
2011-03-24 8:51 ` Tejun Heo
2011-03-24 13:46 ` Mike Frysinger
2011-03-24 17:51 ` Tejun Heo
2011-03-24 8:54 ` [PATCH UPDATED] " Tejun Heo
2010-10-26 14:50 ` [PATCH] MN10300: Fix the PERCPU() alignment to allow for workqueues David Howells
2010-10-26 14:56 ` Tejun Heo
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=1288187870.2709.128.camel@edumazet-laptop \
--to=eric.dumazet@gmail.com \
--cc=brgerst@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=peterz@infradead.org \
--cc=tj@kernel.org \
--cc=torvalds@linux-foundation.org \
--cc=x86@kernel.org \
/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
all inboxes | Powered by JetHome®