1dnl  PowerPC-64 mpn_bdiv_q_1, mpn_pi1_bdiv_q_1 -- Hensel division by 1-limb
2dnl  divisor.
3
4dnl  Copyright 2006, 2010, 2017 Free Software Foundation, Inc.
5
6dnl  This file is part of the GNU MP Library.
7dnl
8dnl  The GNU MP Library is free software; you can redistribute it and/or modify
9dnl  it under the terms of either:
10dnl
11dnl    * the GNU Lesser General Public License as published by the Free
12dnl      Software Foundation; either version 3 of the License, or (at your
13dnl      option) any later version.
14dnl
15dnl  or
16dnl
17dnl    * the GNU General Public License as published by the Free Software
18dnl      Foundation; either version 2 of the License, or (at your option) any
19dnl      later version.
20dnl
21dnl  or both in parallel, as here.
22dnl
23dnl  The GNU MP Library is distributed in the hope that it will be useful, but
24dnl  WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
25dnl  or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
26dnl  for more details.
27dnl
28dnl  You should have received copies of the GNU General Public License and the
29dnl  GNU Lesser General Public License along with the GNU MP Library.  If not,
30dnl  see https://www.gnu.org/licenses/.
31
32include(`../config.m4')
33
34C                             cycles/limb
35C                             norm      unorm
36C POWER3/PPC630            13-19
37C POWER4/PPC970               16
38C POWER5            16        16
39C POWER6            37        46
40C POWER7            12        12
41C POWER8            12        12
42
43C INPUT PARAMETERS
44define(`rp', `r3')
45define(`up', `r4')
46define(`n',  `r5')
47define(`d',  `r6')
48define(`di', `r7')
49define(`cnt',`r8')
50
51define(`tnc',`r10')
52
53ASM_START()
54
55EXTERN(binvert_limb_table)
56
57PROLOGUE(mpn_bdiv_q_1,toc)
58          addi      r7, n, -1
59          cmpdi     cr1, n, 1
60          ld        r12, 0(up)
61          li        cnt, 0
62          neg       r0, d
63          and       r0, d, r0
64          cntlzd    r0, r0
65          subfic    cnt, r0, 63
66          srd       d, d, cnt
67L(7):
68          mtctr     r7
69          LEA(      r10, binvert_limb_table)
70          rldicl    r11, d, 63, 57
71          lbzx      r0, r10, r11
72          mulld     r9, r0, r0
73          sldi      r0, r0, 1
74          mulld     r9, d, r9
75          subf      r0, r9, r0
76          mulld     r10, r0, r0
77          sldi      r0, r0, 1
78          mulld     r10, d, r10
79          subf      r0, r10, r0
80          mulld     r9, r0, r0
81          sldi      r0, r0, 1
82          mulld     r9, d, r9
83          subf      di, r9, r0                    C di = 1/d mod 2^64
84ifdef(`AIX',
85`         C For AIX it is not clear how to jump into another function.
86          b         .mpn_pi1_bdiv_q_1
87',`
88          C For non-AIX, dispatch into the pi1 variant.
89          bne       cr0, L(norm)
90          b         L(unorm)
91')
92EPILOGUE()
93
94PROLOGUE(mpn_pi1_bdiv_q_1)
95          cmpdi     cr0, cnt, 0
96          ld        r12, 0(up)
97          addic     r0, n, -1           C set carry as side effect
98          cmpdi     cr1, n, 1
99          mtctr     r0
100          beq       cr0, L(norm)
101
102L(unorm):
103          subfic    tnc, cnt, 64                  C set carry as side effect
104          li        r5, 0
105          srd       r11, r12, cnt
106          beq       cr1, L(ed1)
107
108          ALIGN(16)
109L(tpu):   ld        r12, 8(up)
110          nop
111          addi      up, up, 8
112          sld       r0, r12, tnc
113          or        r11, r11, r0
114          subfe     r9, r5, r11
115          srd       r11, r12, cnt
116          mulld     r0, di, r9
117          mulhdu    r5, r0, d
118          std       r0, 0(rp)
119          addi      rp, rp, 8
120          bdnz      L(tpu)
121
122          subfe     r11, r5, r11
123L(ed1):   mulld     r0, di, r11
124          std       r0, 0(rp)
125          blr
126
127          ALIGN(16)
128L(norm):
129          mulld     r11, r12, di
130          mulhdu    r5, r11, d
131          std       r11, 0(rp)
132          beqlr     cr1
133
134          ALIGN(16)
135L(tpn):   ld        r9, 8(up)
136          addi      up, up, 8
137          subfe     r5, r5, r9
138          mulld     r11, di, r5
139          mulhdu    r5, r11, d          C result not used in last iteration
140          std       r11, 8(rp)
141          addi      rp, rp, 8
142          bdnz      L(tpn)
143
144          blr
145EPILOGUE()
146ASM_END()
147