2
* Copyright (C) 1994, 1995, 1998, 1999,
3
* 2000, 2001, 2002 Free Software Foundation, Inc.
5
* This file is part of Libgcrypt.
7
* Libgcrypt is free software; you can redistribute it and/or modify
8
* it under the terms of the GNU Lesser General Public License as
9
* published by the Free Software Foundation; either version 2.1 of
10
* the License, or (at your option) any later version.
12
* Libgcrypt is distributed in the hope that it will be useful,
13
* but WITHOUT ANY WARRANTY; without even the implied warranty of
14
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
15
* GNU Lesser General Public License for more details.
17
* You should have received a copy of the GNU Lesser General Public
18
* License along with this program; if not, write to the Free Software
19
* Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA
21
* Note: This code is heavily based on the GNU MP Library.
22
* Actually it's the same code with only minor changes in the
23
* way the data is stored; this is to support the abstraction
24
* of an optional secure memory allocation which may be used
25
* to avoid revealing of sensitive data due to paging etc.
33
* _gcry_mpih_rshift( mpi_ptr_t wp, (r16)
35
* mpi_size_t usize, (r18)
38
* This code runs at 4.8 cycles/limb on the 21064. With infinite unrolling,
39
* it would take 4 cycles/limb. It should be possible to get down to 3
40
* cycles/limb since both ldq and stq can be paired with the other used
41
* instructions. But there are many restrictions in the 21064 pipeline that
42
* makes it hard, if not impossible, to get down to 3 cycles/limb:
44
* 1. ldq has a 3 cycle delay, srl and sll have a 2 cycle delay.
45
* 2. Only aligned instruction pairs can be paired.
46
* 3. The store buffer or silo might not be able to deal with the bandwidth.
53
.globl _gcry_mpih_rshift
54
.ent _gcry_mpih_rshift
58
ldq $4,0($17) # load first limb
62
and $18,4-1,$20 # number of limbs in first loop
63
sll $4,$7,$0 # compute function result
117
.end _gcry_mpih_rshift