[PATCH] Prefer small negative constants in InstCombine

Chris Lattner clattner at apple.com
Wed Jul 3 15:53:52 PDT 2013


On Jul 3, 2013, at 8:12 AM, Tim Northover <t.p.northover at gmail.com> wrote:

>>> Just an update, it turned out to be a terrible idea. Linpack got 66% worse.
>> 
>> Have you analyzed why?
> 
> Turned out to be random variation runs (well, it'd better be since the
> two binaries were identical!).
> 
> Michael gave me some tips on improving consistency of the numbers,
> which may or may not have helped but certainly made me check what the
> test-suite says much more thoroughly. I've come to the conclusion that
> the timing differences are far too small to show up, so I resorted to
> code-size measurements.
> 
> The result was that only 3 binaries changed at all in the test-suite,
> all code-size regressions. An alternative patch at the DAGCombiner
> level had just 1 change, also a regression. I've not found a good
> combined diff/disassembler to really investage what happened there.
> 
> Rather a depressing day, all round.
> 
> I'll see if I can come up with another approach that actually makes
> things better.

Ok, thanks for digging in.  Sometimes you learn some interesting / surprising things if you look closely at artifacts from things like this.

-Chris



More information about the llvm-commits mailing list