[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <1203588090.6243.128.camel@lappy>
Date: Thu, 21 Feb 2008 11:01:30 +0100
From: Peter Zijlstra <a.p.zijlstra@...llo.nl>
To: balbir@...ux.vnet.ibm.com
Cc: Ingo Molnar <mingo@...e.hu>,
"Zhang, Yanmin" <yanmin_zhang@...ux.intel.com>,
Srivatsa Vaddagiri <vatsa@...ux.vnet.ibm.com>,
Dhaval Giani <dhaval@...ux.vnet.ibm.com>,
linux-kernel@...r.kernel.org
Subject: Re: Make yield_task_fair more efficient
On Thu, 2008-02-21 at 15:12 +0530, Balbir Singh wrote:
> Peter Zijlstra wrote:
> > On Thu, 2008-02-21 at 15:01 +0530, Balbir Singh wrote:
> >
> >> I have an alternate approach in mind (that I need to find time for),
> >> threaded-rbtrees. Walking the tree is really efficient, specially finding
> >> successor of a node.
> >
> > Threading the rbtrees would be even more expensive, it would require a
> > list_head in each node and a full list operation for every tree
> > operation.
> >
>
> Peter, when I say threaded, I don't mean threaded as in tasks. Please see
> http://www.nist.gov/dads/HTML/threadedtree.html and
> http://datastructures.itgo.com/trees/tbt.htm
I'm quite aware of the meaning of threaded in the context of data
structures. They usually require additional list data and operations to
implement.
Look at how the latter link you provided grows the rb_node with one
pointer to provide a single linked list.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists