mpir/x86_64 at c0e157e3b295ea153d1400c312532175f2c13ab1 - mpir

cheng/mpir

History

wbhart c0e157e3b2 Roughly speaking mpir should now build on an AMD64. At the present moment the config.guess doesn't distinguish a Core 2 from an AMD64 and so the same code is probably built on both. One must build yasm (included in the yasm directory) before building GMP, if building on an x86_64 machine. Note: make test and make tune do not currently build.		2008-05-26 22:11:40 +00:00
..
amd64	Roughly speaking mpir should now build on an AMD64. At the present moment the config.guess doesn't distinguish a Core 2 from an AMD64 and so the same code is probably built on both.	2008-05-26 22:11:40 +00:00
core2	Basic GMP files with a new core2 directory and amd_64 directory with Martin's and Gaudry's patches.	2008-04-17 21:03:07 +00:00
add_n.asm	Roughly speaking mpir should now build on an AMD64. At the present moment the config.guess doesn't distinguish a Core 2 from an AMD64 and so the same code is probably built on both.	2008-05-26 22:11:40 +00:00
addmul_1.asm	Basic GMP files with a new core2 directory and amd_64 directory with Martin's and Gaudry's patches.	2008-04-17 21:03:07 +00:00
copyd.asm	Roughly speaking mpir should now build on an AMD64. At the present moment the config.guess doesn't distinguish a Core 2 from an AMD64 and so the same code is probably built on both.	2008-05-26 22:11:40 +00:00
copyi.asm	Roughly speaking mpir should now build on an AMD64. At the present moment the config.guess doesn't distinguish a Core 2 from an AMD64 and so the same code is probably built on both.	2008-05-26 22:11:40 +00:00
dive_1.brg	Roughly speaking mpir should now build on an AMD64. At the present moment the config.guess doesn't distinguish a Core 2 from an AMD64 and so the same code is probably built on both.	2008-05-26 22:11:40 +00:00
gmp-mparam.h	Basic GMP files with a new core2 directory and amd_64 directory with Martin's and Gaudry's patches.	2008-04-17 21:03:07 +00:00
lshift.asm	Roughly speaking mpir should now build on an AMD64. At the present moment the config.guess doesn't distinguish a Core 2 from an AMD64 and so the same code is probably built on both.	2008-05-26 22:11:40 +00:00
mode1o.asm	Roughly speaking mpir should now build on an AMD64. At the present moment the config.guess doesn't distinguish a Core 2 from an AMD64 and so the same code is probably built on both.	2008-05-26 22:11:40 +00:00
mode1o.brg	Roughly speaking mpir should now build on an AMD64. At the present moment the config.guess doesn't distinguish a Core 2 from an AMD64 and so the same code is probably built on both.	2008-05-26 22:11:40 +00:00
mul_1.asm	Roughly speaking mpir should now build on an AMD64. At the present moment the config.guess doesn't distinguish a Core 2 from an AMD64 and so the same code is probably built on both.	2008-05-26 22:11:40 +00:00
README	Basic GMP files with a new core2 directory and amd_64 directory with Martin's and Gaudry's patches.	2008-04-17 21:03:07 +00:00
rshift.asm	Roughly speaking mpir should now build on an AMD64. At the present moment the config.guess doesn't distinguish a Core 2 from an AMD64 and so the same code is probably built on both.	2008-05-26 22:11:40 +00:00
sub_n.asm	Roughly speaking mpir should now build on an AMD64. At the present moment the config.guess doesn't distinguish a Core 2 from an AMD64 and so the same code is probably built on both.	2008-05-26 22:11:40 +00:00
submul_1.asm	Basic GMP files with a new core2 directory and amd_64 directory with Martin's and Gaudry's patches.	2008-04-17 21:03:07 +00:00
udiv.asm	Roughly speaking mpir should now build on an AMD64. At the present moment the config.guess doesn't distinguish a Core 2 from an AMD64 and so the same code is probably built on both.	2008-05-26 22:11:40 +00:00
umul.asm	Roughly speaking mpir should now build on an AMD64. At the present moment the config.guess doesn't distinguish a Core 2 from an AMD64 and so the same code is probably built on both.	2008-05-26 22:11:40 +00:00
x86_64-defs.m4	Basic GMP files with a new core2 directory and amd_64 directory with Martin's and Gaudry's patches.	2008-04-17 21:03:07 +00:00

README

Copyright 2003, 2004, 2006 Free Software Foundation, Inc.

This file is part of the GNU MP Library.

The GNU MP 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 MP 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 MP Library; see the file COPYING.LIB.  If not, write to
the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
02110-1301, USA.





			AMD64 MPN SUBROUTINES


This directory contains mpn functions for AMD64 chips.  It might also be
useful for 64-bit Pentiums, but that chip's poor carry handling makes it
unlikely.  We'll need completely separate code (in a subdirectory).


		     RELEVANT OPTIMIZATION ISSUES

The only AMD64 core as of this writing is the AMD Hammer, sold under the
names Opteron and Athlon64.  The Hammer can sustain up to 3 instructions per
cycle, but in practice that is only possible for integer instructions.  But
almost any three integer instructions can issue simultaneously, including
any 3 ALU operations, including shifts.  Up to two memory operations can
issue each cycle.

Scheduling typically requires that load-use instructions are split into
separate load and use instructions.  That requires more decode resources,
and it is rarely a win.  Hammer is a deep out-of-order core.


REFERENCES

"System V Application Binary Interface AMD64 Architecture Processor
Supplement", draft version 0.90, April 2003.