[PATCH] PowerPC: Optimized mpn functions for PowerPC64

[Will Schmidt]

On Mon, 2013-11-04 at 10:40 -0200, Adhemerval Zanella wrote:
> Hi all,
> 
> This patch add optimized __mpn_addmul, __mpn_addsub, __mpn_lshift, and
> __mpn_mul_1 implementations for PowerPC64. They are originally from GMP
> with adjustments for GLIBC.
> 
> With this patch I noticed improvements of about 20% for strtod with
> different kind of inputs (the speed up in the function itself are higher,
> due the fact strtod symbol itself takes about half of total strtod spent
> time).
> 
> Tested on PPC64 and PP64le.

Cosmetic comments about the copyright blurbs below.   I've read over
most of the assembly, no issues with the ode, LGTM.

Thanks, 
-Will

> 
> 2013-11-04  Adhemerval Zanella  <azanella@linux.vnet.ibm.com>
> 
> 	* sysdeps/powerpc/powerpc64/addmul_1.S: New file: optimized
> 	__mpn_addmul_1 for PowerPC64.
> 	* sysdeps/powerpc/powerpc64/submul_1.S: New file: optimized
> 	__mpn_submul_1 for PowerPC64.
> 	* sysdeps/powerpc/powerpc64/lshift.S: New file: optimized __mpn_lshift
> 	for PowerPC64.
> 	* sysdeps/powerpc/powerpc64/mul_1.S: New file: optimized __mpn_mul_1
> 	for PowerPC64.
> 



> diff --git a/sysdeps/powerpc/powerpc64/addmul_1.S b/sysdeps/powerpc/powerpc64/addmul_1.S
> new file mode 100644
> index 0000000..f256506
> --- /dev/null
> +++ b/sysdeps/powerpc/powerpc64/addmul_1.S
> @@ -0,0 +1,208 @@
> +/* PowerPC64 __mpn_addmul_1 -- Multiply a limb vector with a limb and add
> +   the result to a second limb vector.
> +   Copyright (C) 1999-2013 Free Software Foundation, Inc.


^ from a glance at other files, I think this should be just "Copyright
(C) 2013 ".  Similar comment for the other files. 


<snip..>

I've read over most of the .asm, nothing jumps out at me as an issue or
concern. 


> diff --git a/sysdeps/powerpc/powerpc64/submul_1.S b/sysdeps/powerpc/powerpc64/submul_1.S
> new file mode 100644
> index 0000000..145b1d4
> --- /dev/null
> +++ b/sysdeps/powerpc/powerpc64/submul_1.S
> @@ -0,0 +1,21 @@
> +/* PowerPC64 __mpn_addmul_1 -- Multiply a limb vector with a limb and subtract
> +   the result to a second limb vector.
> +   Copyright (C) 2013 Free Software Foundation, Inc.
> +   This file is part of the GNU C Library.
> +
> +   The GNU C Library is free software; you can redistribute it and/or
> +   modify it under the terms of the GNU Lesser General Public
> +   License as published by the Free Software Foundation; either
> +   version 2.1 of the License, or (at your option) any later version.
> +
> +   The GNU C Library is distributed in the hope that it will be useful,
> +   but WITHOUT ANY WARRANTY; without even the implied warranty of
> +   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
> +   Lesser General Public License for more details.
> +
> +   You should have received a copy of the GNU Lesser General Public
> +   License along with the GNU C Library; if not, see
> +   <http://www.gnu.org/licenses/>.  */
> +
> +#define USE_AS_SUBMUL
> +#include "addmul_1.S"

I see other files with just a #include often skip the Copyright blurb.
I'd be tempted to do the same here, but don't feel strongly about it,
and don't know if there is a coding style in place to address it.