public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: ebiederm@xmission.com (Eric W. Biederman)
To: "Yinghai Lu" <yhlu.kernel@gmail.com>
Cc: "Ingo Molnar" <mingo@elte.hu>,
	"Thomas Gleixner" <tglx@linutronix.de>, hpa <hpa@zytor.com>,
	"Dhaval Giani" <dhaval@linux.vnet.ibm.com>,
	"Mike Travis" <travis@sgi.com>,
	"Andrew Morton" <akpm@linux-foundation.org>,
	linux-kernel@vger.kernel.org
Subject: Re: [PATCH 00/16] dyn_array and nr_irqs support v2
Date: Fri, 01 Aug 2008 15:45:23 -0700	[thread overview]
Message-ID: <m1abfwnym4.fsf@frodo.ebiederm.org> (raw)
In-Reply-To: <86802c440808011457g5b72d69bq768cf815046957ec@mail.gmail.com> (Yinghai Lu's message of "Fri, 1 Aug 2008 14:57:13 -0700")

"Yinghai Lu" <yhlu.kernel@gmail.com> writes:

> On Fri, Aug 1, 2008 at 2:30 PM, Yinghai Lu <yhlu.kernel@gmail.com> wrote:
>> On Fri, Aug 1, 2008 at 1:46 PM, Eric W. Biederman <ebiederm@xmission.com>
> wrote:
>
>>> http://lkml.org/lkml/2008/7/10/439
>
> you moved kstat_irqs to irqdesc, and it will not numa-aware. if
> irq_desc is not go with every cpu.

That part is a limitation of the per cpu allocator that the sgi guys
are in the process of fixing.  Which is one of the following goals
of folding the pda into a per cpu structure.

In practice it matters little as irqs only occur on one cpu at a time,
so we shouldn't have cache line contention.

I never got to the arch specific part of allocating irq_desc in a numa
aware fashion.  But I have always figured that if we move the work to
arch code it won't be too difficult, to do things appropriately.

Eric

  reply	other threads:[~2008-08-01 22:53 UTC|newest]

Thread overview: 39+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-08-01  9:37 [PATCH 00/16] dyn_array and nr_irqs support v2 Yinghai Lu
2008-08-01  9:37 ` [PATCH 01/16] x86: 64bit support more than 256 irq Yinghai Lu
2008-08-01  9:37   ` [PATCH 02/16] x86: introduce nr_irqs for 64bit v3 Yinghai Lu
2008-08-01  9:37     ` [PATCH 03/16] add dyn_array support Yinghai Lu
2008-08-01  9:37       ` [PATCH 04/16] make irq_timer_state to use dyn_array Yinghai Lu
2008-08-01  9:37         ` [PATCH 05/16] make irq2_iommu " Yinghai Lu
2008-08-01  9:37           ` [PATCH 06/16] make irq_desc " Yinghai Lu
2008-08-01  9:37             ` [PATCH 07/16] x86: make 64bit support dyn_array Yinghai Lu
2008-08-01  9:37               ` [PATCH 08/16] serial: change remove NR_IRQS in 8250.c v2 Yinghai Lu
2008-08-01  9:37                 ` [PATCH 09/16] add per_cpu_dyn_array support Yinghai Lu
2008-08-01  9:37                   ` [PATCH 10/16] irq: make irqs in kernel stat use per_cpu_dyn_array Yinghai Lu
2008-08-01  9:37                     ` [PATCH 11/16] x86 remove irq_vectors_limit.h Yinghai Lu
2008-08-01  9:37                       ` [PATCH 12/16] x86: make 32bit use dyn_array Yinghai Lu
2008-08-01  9:37                         ` [PATCH 13/16] add per_cpu_dyn_array for arch percpu support Yinghai Lu
2008-08-01  9:37                           ` [PATCH 14/16] x86: get mp_irqs from madt Yinghai Lu
2008-08-01  9:37                             ` [PATCH 15/16] x86: make 32bit more like with io_apic/dyn_array to 64 bit Yinghai Lu
2008-08-01  9:37                               ` [PATCH 16/16] x86: alloc dyn_array all alltogether Yinghai Lu
2008-08-01 20:46 ` [PATCH 00/16] dyn_array and nr_irqs support v2 Eric W. Biederman
2008-08-01 21:30   ` Yinghai Lu
2008-08-01 21:57     ` Yinghai Lu
2008-08-01 22:45       ` Eric W. Biederman [this message]
2008-08-01 22:10     ` Yinghai Lu
2008-08-01 22:38     ` Eric W. Biederman
2008-08-02  1:09       ` Yinghai Lu
2008-08-02  1:36         ` H. Peter Anvin
2008-08-02  1:41         ` Eric W. Biederman
2008-08-02  2:01           ` Yinghai Lu
2008-08-02  2:03             ` H. Peter Anvin
2008-08-02  2:39               ` Eric W. Biederman
2008-08-02  3:28                 ` H. Peter Anvin
2008-08-02  4:42                   ` Eric W. Biederman
2008-08-02 15:41                     ` H. Peter Anvin
2008-08-02 20:20                       ` Eric W. Biederman
2008-08-04 12:57           ` Mike Travis
2008-08-05  2:38             ` H. Peter Anvin
2008-08-05  3:40               ` Eric W. Biederman
2008-08-05  3:48                 ` H. Peter Anvin
2008-08-01 21:47   ` Mike Travis
2008-08-02  2:58   ` Yinghai Lu

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=m1abfwnym4.fsf@frodo.ebiederm.org \
    --to=ebiederm@xmission.com \
    --cc=akpm@linux-foundation.org \
    --cc=dhaval@linux.vnet.ibm.com \
    --cc=hpa@zytor.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=tglx@linutronix.de \
    --cc=travis@sgi.com \
    --cc=yhlu.kernel@gmail.com \
    /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