[Bug target/68261] GCC needs to use optimized version of memcpy

["manu at gcc dot gnu.org" <gcc-bugzilla at gcc dot gnu dot org>]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=68261

Manuel LÃpez-IbÃÃez <manu at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |manu at gcc dot gnu.org

--- Comment #3 from Manuel LÃpez-IbÃÃez <manu at gcc dot gnu.org> ---
(In reply to Geir Johansen from comment #0)
> The memcpy routine for GCC needs to be faster.  The following test case
> shows that the Intel compiler implementation of memcpy is over twice as fast
> as GCC.  I realize that memcpy is a part of GLIBC, but the GCC compiler
> should take advantage of the targetting information being provided and the
> context of the memcpy in order to provide more optimal code:

The right mailing list to discuss this is probably libc-alpha and the right
person to speak with is probably OndÅej BÃlka:

https://sourceware.org/ml/libc-alpha/2013-08/msg00161.html
https://sourceware.org/ml/libc-alpha/2015-05/msg00600.html
https://gcc.gnu.org/ml/gcc/2015-06/msg00057.html
https://gcc.gnu.org/ml/gcc/2015-06/msg00059.html

I think GCC still needs a person with the time and patience to serve as the
bridge between OndÅej (and GNU libc) and GCC on this issue, since it is obvious
that more collaboration is needed. If you are willing to be that person, it
would help to familiarize yourself with the latest discussion in libc-alpha and
gcc@gcc.gnu.org.