On Wed, Aug 17, 2016 at 4:17 PM, Jakub Jelinek wrote:
> On Tue, Aug 16, 2016 at 09:21:57PM +0200, Uros Bizjak wrote:
>> The idea is indeed good, but please leave full names in the *.def
>> file. We can change them later, if need arises.
>
> Ok, here it is.
> i386-builtin.def is basically moved the
On Tue, Aug 16, 2016 at 09:21:57PM +0200, Uros Bizjak wrote:
> The idea is indeed good, but please leave full names in the *.def
> file. We can change them later, if need arises.
Ok, here it is.
i386-builtin.def is basically moved the bdesc_* definitions, except that
{ ... },
is replaced with
BD
On Tue, Aug 16, 2016 at 09:21:57PM +0200, Uros Bizjak wrote:
> > Bootstrapped/regtested on x86_64-linux and i686-linux, ok for trunk?
>
> The patch is basically unreviwable, but we have many checks in the x86
> specific testsuite that would break left and right if something is
> wrong in the area
Hi!
I've noticed today that ix86_expand_builtin is very inefficient, in order to
expand each machine builtin (except for the ones handled in a switch), it
searches up to 2300+ bdesc_* array elements in order to find the one it is
looking for.
Following patch is an attempt to speed this up, by reo
