1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
|
/* PowerPC-32 sub_n -- Subtract two limb vectors of the same length > 0
* and store difference in a third limb vector.
* Copyright (C) 1992, 1994, 1995, 1997, 1998 Free Software Foundation, Inc.
*
* This file is part of GnuPG.
*
* GnuPG is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 3 of the License, or
* (at your option) any later version.
*
* GnuPG 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 General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, see <http://www.gnu.org/licenses/>.
*/
#include "sysdep.h"
#include "asm-syntax.h"
#ifndef USE_PPC_PATCHES
/*******************
* mpi_limb_t
* mpihelp_sub_n( mpi_ptr_t res_ptr, (r3)
* mpi_ptr_t s1_ptr, (r4)
* mpi_ptr_t s2_ptr, (r5)
* mpi_size_t size) (r6)
*/
.toc
.extern mpihelp_sub_n[DS]
.extern .mpihelp_sub_n
.csect [PR]
.align 2
.globl mpihelp_sub_n
.globl .mpihelp_sub_n
.csect mpihelp_sub_n[DS]
mpihelp_sub_n:
.long .mpihelp_sub_n, TOC[tc0], 0
.csect [PR]
.mpihelp_sub_n:
mtctr 6 # copy size into CTR
lwz 8,0(4) # load least significant s1 limb
lwz 0,0(5) # load least significant s2 limb
addi 3,3,-4 # offset res_ptr, it is updated before used
subfc 7,0,8 # add least significant limbs, set cy
bdz Lend # If done, skip loop
Loop: lwzu 8,4(4) # load s1 limb and update s1_ptr
lwzu 0,4(5) # load s2 limb and update s2_ptr
stwu 7,4(3) # store previous limb in load latency slot
subfe 7,0,8 # add new limbs with cy, set cy
bdnz Loop # decrement CTR and loop back
Lend: stw 7,4(3) # store ultimate result limb
subfe 3,0,0 # load !cy into ...
subfic 3,3,0 # ... return value register
blr
#else
/* mp_limb_t mpn_sub_n (mp_ptr res_ptr, mp_srcptr s1_ptr, mp_srcptr s2_ptr,
mp_size_t size)
Calculate s1-s2 and put result in res_ptr; return borrow, 0 or 1. */
/* Note on optimisation: This code is optimal for the 601. Almost every other
possible 2-unrolled inner loop will not be. Also, watch out for the
alignment... */
EALIGN(mpihelp_sub_n,3,1)
/* Set up for loop below. */
mtcrf 0x01,%r6
srwi. %r7,%r6,1
mtctr %r7
bt 31,2f
/* Set the carry (clear the borrow). */
subfc %r0,%r0,%r0
/* Adjust pointers for loop. */
addi %r3,%r3,-4
addi %r4,%r4,-4
addi %r5,%r5,-4
b 0f
2: lwz %r7,0(%r5)
lwz %r6,0(%r4)
subfc %r6,%r7,%r6
stw %r6,0(%r3)
beq 1f
/* Align start of loop to an odd word boundary to guarantee that the
last two words can be fetched in one access (for 601). This turns
out to be important. */
0:
lwz %r9,4(%r4)
lwz %r8,4(%r5)
lwzu %r6,8(%r4)
lwzu %r7,8(%r5)
subfe %r8,%r8,%r9
stw %r8,4(%r3)
subfe %r6,%r7,%r6
stwu %r6,8(%r3)
bdnz 0b
/* Return the borrow. */
1: subfe %r3,%r3,%r3
neg %r3,%r3
blr
END(mpihelp_sub_n)
#endif
|
