Re: Make yield_task_fair more efficient

[Balbir Singh <balbir@linux.vnet.ibm.com>]

Peter Zijlstra wrote:
> On Thu, 2008-02-21 at 15:37 +0530, Balbir Singh wrote:
> 
>> You use the empty pointer (missing right child), so why do we need a list. May
>> be I am missing something.
> 
> A fully threaded tree also has back-pointer to traverse backwards
> through the ordered elements.
> 

The back pointer is implemented using missing left children. I am referring to
Don Knuth's volume 1: The art of computer programming, section 2.3.1, page 322.
Please see the threaded representation of the tree, it is compared with the
unthreaded representation.

> That said, overloading the right child pointer might not be the best
> thing for the linux kernel, as it will impact all the rb-tree lookups
> which are open-coded and often performance critical (this is the reason
> the colour isn't bit encoded in either of the child pointers either).
> 

We always look at the child, irrespective of whether it is to the right or left
to terminate our walk. Overloading them and setting a bit stating it is threaded
should not be that bad.

> But if you only want a uni directional thread, I guess we can stick it
> in the unsigned long we use for the node colour.
> 
> Still, perhaps it's worth it to grow rb_node to 4 words and do the fully
> threaded thing as there are also a lot of rb_prev() users in the kernel.
> Who knows..
> 

Why does the rb_node need to grow? We can encode the bits in the children

> Anyway, I agree that improving rb_next() is worth looking into for the
> scheduler.

Sure, will experiment and see if we can bump up performance.

-- 
	Warm Regards,
	Balbir Singh
	Linux Technology Center
	IBM, ISTL