[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: IRQ Affinity

To: Scott Lovenberg <scott.lovenberg@xxxxxxxxx>
Subject: Re: IRQ Affinity
From: Elad Lahav <elad_lahav@xxxxxxxxxxxxxxxxxxxxx>
Date: Sat, 14 Jun 2008 07:43:28 -0400
Cc: kernelnewbies@xxxxxxxxxxxx
In-reply-to: <4853240D.9090501@gmail.com>
List-archive: <http://mail.nl.linux.org/kernelnewbies/>
List-help: <mailto:ecartis@nl.linux.org?Subject=help>
List-id: <kernelnewbies.nl.linux.org>
List-owner: <mailto:ecartis-owner@nl.linux.org>
List-post: <mailto:kernelnewbies@nl.linux.org>
List-software: Ecartis version 1.0.0
List-subscribe: <mailto:kernelnewbies-request@nl.linux.org?Subject=subscribe>
List-unsubscribe: <mailto:kernelnewbies-request@nl.linux.org?Subject=unsubscribe>
References: <48527A71.20100@users.sourceforge.net> <g2v25i$d7a$1@ger.gmane.org> <4853240D.9090501@gmail.com>
Sender: kernelnewbies-bounce@xxxxxxxxxxxx
User-agent: Thunderbird 2.0.0.14 (X11/20080522)

Is it possible that they're migrating due to the CPU being masked off for a critical section of code in another process/thread? I think everything gets booted off a CPU when it is being used for a mutex's critical section. The IRQs are masked and routed to other CPUs until the mutex is done and the caches are flushed again. Does this apply to your test environment at all? I can't remember exactly where the applicable code is, but I think it's around SMP.c or something to that effect. Concurrency.c perhaps?

I'll check that. Thanks.

Elad

--
To unsubscribe from this list: send an email with
"unsubscribe kernelnewbies" to ecartis@xxxxxxxxxxxx
Please read the FAQ at http://kernelnewbies.org/FAQ

References:
- IRQ Affinity
  - From: Elad Lahav
- Re: IRQ Affinity
  - From: Elad Lahav
- Re: IRQ Affinity
  - From: Scott Lovenberg

Prev by Date: Re: initrd???????????
Next by Date: Re: What is fixmaps?
Previous by thread: Re: IRQ Affinity
Next by thread: Modifying ext3
Index(es):
- Date
- Thread