Date: Thu, 31 Jul 2008 14:38:27 -0500
From: Paul Jackson <pj@sgi.com>
To: "Paul Menage" <menage@google.com>
Cc: laijs@cn.fujitsu.com, akpm@linux-foundation.org,
       linux-kernel@vger.kernel.org
Subject: Re: [PATCH] cpuset: make ntasks to be a monotonic increasing value
Message-Id: <20080731143827.b8bf7bce.pj@sgi.com>
In-Reply-To: <6599ad830807311006r24e4d325ie505949ad760afd@mail.gmail.com>
References: <48912FDD.8060006@cn.fujitsu.com>
	<20080731072355.b582b2d6.pj@sgi.com>
	<4891B9E0.2090900@cn.fujitsu.com>
	<20080731083706.e6bd4acc.pj@sgi.com>
	<6599ad830807311006r24e4d325ie505949ad760afd@mail.gmail.com>
Organization: SGI
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1187
Lines: 30

Paul M wrote:
> That loop really could do with some updates though - currently it
> looks at the mm for every task in the cpuset, rather than filtering
> duplicate mms from threaded applications.

Interesting.

After a quick glance, I suppose that we'd still have:
 1) allocate an mmarray[] in that particular loop as we do now,
    sized large enough for all tasks,
 2) convert each task to it's mm, in the next code chunk, with:
	mm = get_task_mm(p);

but that then, before we call "mpol_rebind_mm()" for each such
mm, we could essentially do a "sort -u" (sort unique) on that
mmarray[], to remove duplicate mm's.  This would not change any
of the existing loops; rather just add one more code paragraph,
to remove the duplicate mm's.

Is that what you're thinking, Paul M?

-- 
                  I won't rest till it's the best ...
                  Programmer, Linux Scalability
                  Paul Jackson <pj@sgi.com> 1.940.382.4214
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/