I have compared 4.1.2 release (r121943) with three revisions of 4.2 on spec2k
on an 2GHz AMD Athlon64 box (in 64bit mode), detailed results are below.
In short, current 4.2 performs just as good as 4.1 on this target
with the exception of huge 80% win on 178.galgel. All other difference
lies almos
I checked out the gcc 4.2 branch because I want to start using the OpenMP
(gomp) capability.
When I tried to build it, I got an ICE somewhere in stage 3 and the message
said to submit a bug report.
I searched for a similar bug in bugzilla and decided to submit the bug
report, but when I had to ch
Here is the comparison of 4.1 and 4.2 branches (as of day before
yesterday) on SPEC2000 for ppc64.
In brief, gcc4.2 generates 3% faster code for SPECFP2000 and the same
code for SPECInt2000. In average the generated SPECInt2000 code size
is 0.5% smaller for gcc4.2. The SPECFp2000 code is 1.7% b
> Sure - I wonder if there is a reliable way of testing whether we face
> a non-base type in the middle-end. I suppose TREE_TYPE (type) != NULL
> won't work in all cases... (?)
That's the right way as far as I know.
> I agree. But appearantly fold does not care about base vs. non-base
> types a
On Fri, 23 Feb 2007, Richard Kenner wrote:
> > That said, this whole thing is a can of worms. Suppose the compiler wants
> > to
> > calculate t+1. Of course you do something like this:
> >
> > int_const_binop (PLUS_EXPR, t, build_int_cst (TREE_TYPE (t), 1), 0);
> >
> > But if 1 is not in the
You'll want to have a look at something like the SSE SIMD optimisation within
the back end for the x86.
I'd also download the GCC internals document for 4.3.x as it has some
information on CPU SIMD optimisations.
I've been reviewing this area heavily as I'm combining GCC and GPGPU (the same
id
