1 | ; $Id: fma-asm.asm 96407 2022-08-22 17:43:14Z vboxsync $
|
---|
2 | ;; @file
|
---|
3 | ; IPRT - No-CRT fma alternatives - AMD64 & X86.
|
---|
4 | ;
|
---|
5 |
|
---|
6 | ;
|
---|
7 | ; Copyright (C) 2006-2022 Oracle and/or its affiliates.
|
---|
8 | ;
|
---|
9 | ; This file is part of VirtualBox base platform packages, as
|
---|
10 | ; available from https://www.virtualbox.org.
|
---|
11 | ;
|
---|
12 | ; This program is free software; you can redistribute it and/or
|
---|
13 | ; modify it under the terms of the GNU General Public License
|
---|
14 | ; as published by the Free Software Foundation, in version 3 of the
|
---|
15 | ; License.
|
---|
16 | ;
|
---|
17 | ; This program is distributed in the hope that it will be useful, but
|
---|
18 | ; WITHOUT ANY WARRANTY; without even the implied warranty of
|
---|
19 | ; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
|
---|
20 | ; General Public License for more details.
|
---|
21 | ;
|
---|
22 | ; You should have received a copy of the GNU General Public License
|
---|
23 | ; along with this program; if not, see <https://www.gnu.org/licenses>.
|
---|
24 | ;
|
---|
25 | ; The contents of this file may alternatively be used under the terms
|
---|
26 | ; of the Common Development and Distribution License Version 1.0
|
---|
27 | ; (CDDL), a copy of it is provided in the "COPYING.CDDL" file included
|
---|
28 | ; in the VirtualBox distribution, in which case the provisions of the
|
---|
29 | ; CDDL are applicable instead of those of the GPL.
|
---|
30 | ;
|
---|
31 | ; You may elect to license modified versions of this file under the
|
---|
32 | ; terms and conditions of either the GPL or the CDDL or both.
|
---|
33 | ;
|
---|
34 | ; SPDX-License-Identifier: GPL-3.0-only OR CDDL-1.0
|
---|
35 | ;
|
---|
36 |
|
---|
37 | %define RT_ASM_WITH_SEH64
|
---|
38 | %include "iprt/asmdefs.mac"
|
---|
39 |
|
---|
40 | BEGINCODE
|
---|
41 |
|
---|
42 | ;;
|
---|
43 | ; Fused multiplication and add, intel version.
|
---|
44 | ;
|
---|
45 | ; @returns st(0) / xmm0
|
---|
46 | ; @param rdFactor1 [rbp + 08h] / xmm0
|
---|
47 | ; @param rdFactor2 [rbp + 10h] / xmm1
|
---|
48 | ; @param rdAddend [rbp + 18h] / xmm2
|
---|
49 | BEGINPROC rtNoCrtMathFma3
|
---|
50 | push xBP
|
---|
51 | SEH64_PUSH_xBP
|
---|
52 | mov xBP, xSP
|
---|
53 | SEH64_SET_FRAME_xBP 0
|
---|
54 | SEH64_END_PROLOGUE
|
---|
55 |
|
---|
56 | %ifdef RT_ARCH_X86
|
---|
57 | movsd xmm0, qword [xBP + xCB*2 + 00h]
|
---|
58 | movsd xmm1, qword [xBP + xCB*2 + 08h]
|
---|
59 | movsd xmm2, qword [xBP + xCB*2 + 10h]
|
---|
60 | %endif
|
---|
61 |
|
---|
62 | vfmadd132sd xmm0, xmm2, xmm1 ; xmm0 = xmm0 * xmm1 + xmm2 (132 = multiply op1 with op3 and add op2)
|
---|
63 |
|
---|
64 | %ifdef RT_ARCH_X86
|
---|
65 | sub xSP, 10h
|
---|
66 | movsd [xSP], xmm0
|
---|
67 | fld qword [xSP]
|
---|
68 | %endif
|
---|
69 | leave
|
---|
70 | ret
|
---|
71 | ENDPROC rtNoCrtMathFma3
|
---|
72 |
|
---|
73 |
|
---|
74 | ;;
|
---|
75 | ; Fused multiplication and add, amd version.
|
---|
76 | ;
|
---|
77 | ; @returns st(0) / xmm0
|
---|
78 | ; @param rdFactor1 [rbp + 08h] / xmm0
|
---|
79 | ; @param rdFactor2 [rbp + 10h] / xmm1
|
---|
80 | ; @param rdAddend [rbp + 18h] / xmm2
|
---|
81 | BEGINPROC rtNoCrtMathFma4
|
---|
82 | push xBP
|
---|
83 | SEH64_PUSH_xBP
|
---|
84 | mov xBP, xSP
|
---|
85 | SEH64_SET_FRAME_xBP 0
|
---|
86 | SEH64_END_PROLOGUE
|
---|
87 |
|
---|
88 | %ifdef RT_ARCH_X86
|
---|
89 | movsd xmm0, qword [xBP + xCB*2 + 00h]
|
---|
90 | movsd xmm1, qword [xBP + xCB*2 + 08h]
|
---|
91 | movsd xmm2, qword [xBP + xCB*2 + 10h]
|
---|
92 | %endif
|
---|
93 |
|
---|
94 | vfmaddsd xmm0, xmm0, xmm1, xmm2 ; xmm0 = xmm0 * xmm1 + xmm2
|
---|
95 |
|
---|
96 | %ifdef RT_ARCH_X86
|
---|
97 | sub xSP, 10h
|
---|
98 | movsd [xSP], xmm0
|
---|
99 | fld qword [xSP]
|
---|
100 | %endif
|
---|
101 | leave
|
---|
102 | ret
|
---|
103 | ENDPROC rtNoCrtMathFma4
|
---|
104 |
|
---|