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

Re: [RFC][PATCH] iowait statistics

To: Rik van Riel <riel@conectiva.com.br>,William Lee Irwin III <wli@holomorphy.com>
Subject: Re: [RFC][PATCH] iowait statistics
From: "Martin J. Bligh" <Martin.Bligh@us.ibm.com>
Date: Tue, 14 May 2002 11:19:26 -0700
cc: linux-kernel@vger.kernel.org, linux-mm@kvack.org
Fake-Sender: owner-linux-mm@kvack.org
In-Reply-To: <Pine.LNX.4.44L.0205141335080.9490-100000@duckman.distro.conectiva>
Original-Recipient: rfc822;linux-mm-archive@humbolt.geo.uu.nl
References: <Pine.LNX.4.44L.0205141335080.9490-100000@duckman.distro.conectiva>
Sender: Rik van Riel <riel@nl.linux.org>

>> This appears to be global across all cpu's. Maybe nr_iowait_tasks
>> should be accounted on a per-cpu basis, where
> 
> While your proposal should work, somehow I doubt it's worth
> the complexity. It's just a statistic to help sysadmins ;)

Depends how often you're going to end up bouncing that cacheline 
around ... do you do this for every IO?

M.

--
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/

References:
- Re: [RFC][PATCH] iowait statistics
  - From: Rik van Riel <riel@conectiva.com.br>

Prev by Date: Re: [RFC][PATCH] iowait statistics
Next by Date: Re: [RFC][PATCH] IO wait accounting
Prev by thread: Re: [RFC][PATCH] iowait statistics
Next by thread: Re: [RFC][PATCH] iowait statistics
Index(es):
- Date
- Thread