Re: Recognize vec_perm_expr in a constructor of bit_field_ref

[Marc Glisse <marc dot glisse at inria dot fr>]

On Tue, 11 Sep 2012, Richard Guenther wrote:

> On Tue, Sep 11, 2012 at 1:07 PM, Marc Glisse <marc.glisse@inria.fr> wrote:
> > Hello,
> >
> > here is a patch that turns {v[1],v[0]} into vec_perm_expr(v,v,{1,0}) if the
> > target is ok with it.
> >
> > I am attaching 2 versions of the patch. p-good is the one that passes
> > testing. p-bad, where I rely on fold_stmt to detect identity permutations,
> > ICEs towards the end of the pass while checking a bogus gimple stmt (one
> > that gimple_debug_stmt crashes on if I call it in gdb). From a performance
> > point of view, p-good makes sense, but I liked the simplicity of p-bad and I
> > am confused as to why it fails.
>
> Probably because you cannot simply increase num_ops ...

Ah... thanks, it makes sense now... For some reason I thought it was a 
fixed size structure and num_ops just told it how many of the fields were 
in use.

[...]
> Ok with that change.

Just to be sure, that means you prefer the version where I manually detect 
identity and don't call fold, right?

Thank you for all the quick reviews,

--
Marc Glisse