Re: [PATCH v5 07/10] module: Optimize __module_address() using a latched RB-tree

[Peter Zijlstra <peterz@infradead.org>]

On Mon, Apr 13, 2015 at 06:49:49PM +0200, Ingo Molnar wrote:
> * Peter Zijlstra <peterz@infradead.org> wrote:

> > +static __always_inline int
> > +mod_tree_comp(void *key, struct latch_tree_node *n)
> > +{
> > +	unsigned long val = (unsigned long)key;
> > +	unsigned long start, end;
> > +
> > +	end = start = __mod_tree_val(n);
> > +	end += __mod_tree_size(n);
> > +
> > +	if (val < start)
> > +		return -1;
> > +
> > +	if (val >= end)
> > +		return 1;
> > +
> > +	return 0;
> 
> So since we are counting nanoseconds, I suspect this could be written 
> more optimally as:
> 
> {
> 	unsigned long val = (unsigned long)key;
> 	unsigned long start, end;
> 
> 	start = __mod_tree_val(n);
> 	if (val < start)
> 		return -1;
> 
> 	end = start + __mod_tree_size(n);
> 	if (val >= end)
> 		return 1;
> 
> 	return 0;
> }
> 
> right?

I was afraid it would rip apart the common bits of
__mod_tree_{val,size}(), iow. it would end up doing the whole
latch_tree_node -> mod_tree_node -> mod and mtn_init comparison dance
twice.

But GCC does the right thing, so yes.