|
Packit |
5c3484 |
Copyright 1996, 2001 Free Software Foundation, Inc.
|
|
Packit |
5c3484 |
|
|
Packit |
5c3484 |
This file is part of the GNU MP Library.
|
|
Packit |
5c3484 |
|
|
Packit |
5c3484 |
The GNU MP Library is free software; you can redistribute it and/or modify
|
|
Packit |
5c3484 |
it under the terms of either:
|
|
Packit |
5c3484 |
|
|
Packit |
5c3484 |
* the GNU Lesser General Public License as published by the Free
|
|
Packit |
5c3484 |
Software Foundation; either version 3 of the License, or (at your
|
|
Packit |
5c3484 |
option) any later version.
|
|
Packit |
5c3484 |
|
|
Packit |
5c3484 |
or
|
|
Packit |
5c3484 |
|
|
Packit |
5c3484 |
* the GNU General Public License as published by the Free Software
|
|
Packit |
5c3484 |
Foundation; either version 2 of the License, or (at your option) any
|
|
Packit |
5c3484 |
later version.
|
|
Packit |
5c3484 |
|
|
Packit |
5c3484 |
or both in parallel, as here.
|
|
Packit |
5c3484 |
|
|
Packit |
5c3484 |
The GNU MP Library is distributed in the hope that it will be useful, but
|
|
Packit |
5c3484 |
WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
|
|
Packit |
5c3484 |
or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
|
|
Packit |
5c3484 |
for more details.
|
|
Packit |
5c3484 |
|
|
Packit |
5c3484 |
You should have received copies of the GNU General Public License and the
|
|
Packit |
5c3484 |
GNU Lesser General Public License along with the GNU MP Library. If not,
|
|
Packit |
5c3484 |
see https://www.gnu.org/licenses/.
|
|
Packit |
5c3484 |
|
|
Packit |
5c3484 |
|
|
Packit |
5c3484 |
|
|
Packit |
5c3484 |
|
|
Packit |
5c3484 |
|
|
Packit |
5c3484 |
This directory contains mpn functions for various SPARC chips. Code that
|
|
Packit |
5c3484 |
runs only on version 8 SPARC implementations, is in the v8 subdirectory.
|
|
Packit |
5c3484 |
|
|
Packit |
5c3484 |
RELEVANT OPTIMIZATION ISSUES
|
|
Packit |
5c3484 |
|
|
Packit |
5c3484 |
Load and Store timing
|
|
Packit |
5c3484 |
|
|
Packit |
5c3484 |
On most early SPARC implementations, the ST instructions takes multiple
|
|
Packit |
5c3484 |
cycles, while a STD takes just a single cycle more than an ST. For the CPUs
|
|
Packit |
5c3484 |
in SPARCstation I and II, the times are 3 and 4 cycles, respectively.
|
|
Packit |
5c3484 |
Therefore, combining two ST instructions into a STD when possible is a
|
|
Packit |
5c3484 |
significant optimization.
|
|
Packit |
5c3484 |
|
|
Packit |
5c3484 |
Later SPARC implementations have single cycle ST.
|
|
Packit |
5c3484 |
|
|
Packit |
5c3484 |
For SuperSPARC, we can perform just one memory instruction per cycle, even
|
|
Packit |
5c3484 |
if up to two integer instructions can be executed in its pipeline. For
|
|
Packit |
5c3484 |
programs that perform so many memory operations that there are not enough
|
|
Packit |
5c3484 |
non-memory operations to issue in parallel with all memory operations, using
|
|
Packit |
5c3484 |
LDD and STD when possible helps.
|
|
Packit |
5c3484 |
|
|
Packit |
5c3484 |
UltraSPARC-1/2 has very slow integer multiplication. In the v9 subdirectory,
|
|
Packit |
5c3484 |
we therefore use floating-point multiplication.
|
|
Packit |
5c3484 |
|
|
Packit |
5c3484 |
STATUS
|
|
Packit |
5c3484 |
|
|
Packit |
5c3484 |
1. On a SuperSPARC, mpn_lshift and mpn_rshift run at 3 cycles/limb, or 2.5
|
|
Packit |
5c3484 |
cycles/limb asymptotically. We could optimize speed for special counts
|
|
Packit |
5c3484 |
by using ADDXCC.
|
|
Packit |
5c3484 |
|
|
Packit |
5c3484 |
2. On a SuperSPARC, mpn_add_n and mpn_sub_n runs at 2.5 cycles/limb, or 2
|
|
Packit |
5c3484 |
cycles/limb asymptotically.
|
|
Packit |
5c3484 |
|
|
Packit |
5c3484 |
3. mpn_mul_1 runs at what is believed to be optimal speed.
|
|
Packit |
5c3484 |
|
|
Packit |
5c3484 |
4. On SuperSPARC, mpn_addmul_1 and mpn_submul_1 could both be improved by a
|
|
Packit |
5c3484 |
cycle by avoiding one of the add instructions. See a29k/addmul_1.
|
|
Packit |
5c3484 |
|
|
Packit |
5c3484 |
The speed of the code for other SPARC implementations is uncertain.
|