Re: [PATCH 2/8] tracing: create automated trace defines

[Andi Kleen <andi@firstfloor.org>]

On Wed, Apr 22, 2009 at 02:24:17AM -0400, Steven Rostedt wrote:
> 
> On Wed, 22 Apr 2009, Andi Kleen wrote:
> 
> > > I think it was Ingo that let out the idea, and I'm starting to like it.
> > > 
> > > Perhaps we should fork off gcc and ship Linux with its own compiler. This 
> > > way we can optimize it for the kernel and not worry about any userland 
> > > optimizations.
> > > 
> > > I would like to do something like:
> > > 
> > > 	if (unlikely(err)) {
> > > 		__section__(".error_sect") {
> > 
> > 
> > gcc already supports that, you don't need to fork anything. It's called
> > hot/cold partitioning. Basically it splits functions into hot and cold
> > and unlikely parts and all the cold/unlikely parts go into a separate 
> > sections.
> > 
> > I think it's normally not enabled by default on x86 though, probably because
> > it doesn't help too much.
> > 
> > By default (unless you specify -fno-reorder-blocks) it does the same
> > without sections, just moving unlikely code out of line.
> 
> The unlikely code does not always get moved out that far. It still sits 
> inside a function, and looking at the tracepoint code it did not move it 
> far enough.

That's because you didn't enable the hot/cold partioning as I wrote.
These are separate options. By default it doesn't use partitions on x86,
but it can.

> If gcc can indeed move "unlikely" code completely out of the fast path, 
> and put it into its own sections, then I think we should go through the 
> kernel and start removing all "likely" and "unlikely"s that are not 99% 
> accurate. Then we can enable the separate section cold paths and perhaps 
> see a performance benefit.

iirc there wasn't much for using separate partitions with the usual
user space benchmarks (SpecCPU etc.) on x86. It helped a bit on POWER
apparently though.

-Andi

-- 
ak@linux.intel.com -- Speaking for myself only.