0
|
1 |
// Copyright (c) 1995-2009 Nokia Corporation and/or its subsidiary(-ies).
|
|
2 |
// All rights reserved.
|
|
3 |
// This component and the accompanying materials are made available
|
|
4 |
// under the terms of the License "Eclipse Public License v1.0"
|
|
5 |
// which accompanies this distribution, and is available
|
|
6 |
// at the URL "http://www.eclipse.org/legal/epl-v10.html".
|
|
7 |
//
|
|
8 |
// Initial Contributors:
|
|
9 |
// Nokia Corporation - initial contribution.
|
|
10 |
//
|
|
11 |
// Contributors:
|
|
12 |
//
|
|
13 |
// Description:
|
|
14 |
// e32\include\arm_vfp.h
|
|
15 |
//
|
|
16 |
// WARNING: This file contains some APIs which are internal and are subject
|
|
17 |
// to change without notice. Such APIs should therefore not be used
|
|
18 |
// outside the Kernel and Hardware Services package.
|
|
19 |
//
|
|
20 |
|
|
21 |
#ifndef __ARM_VFP_H__
|
|
22 |
#define __ARM_VFP_H__
|
|
23 |
#include <cpudefs.h>
|
|
24 |
#if defined(__CPU_ARM) && defined(__CPU_HAS_VFP)
|
|
25 |
|
|
26 |
// MRCcccc Ppppp, iii, Rdddd, Cnnnn, Cmmmm, ttt
|
|
27 |
// cccc 1110 iii1 nnnn dddd pppp ttt1 mmmm
|
|
28 |
// MCRcccc Ppppp, iii, Rdddd, Cnnnn, Cmmmm, ttt
|
|
29 |
// cccc 1110 iii0 nnnn dddd pppp ttt1 mmmm
|
|
30 |
// CDPcccc pppp, iiii, Cdddd, Cnnnn, Cmmmm, ttt
|
|
31 |
// cccc 1110 iiii nnnn dddd pppp ttt0 mmmm
|
|
32 |
// CDP2 pppp, iiii, Cdddd, Cnnnn, Cmmmm, ttt
|
|
33 |
// 1111 1110 iiii nnnn dddd pppp ttt0 mmmm
|
|
34 |
// LDCcccc
|
|
35 |
// cccc 110P UNW1 nnnn dddd pppp oooo oooo
|
|
36 |
// STCcccc
|
|
37 |
// cccc 110P UNW0 nnnn dddd pppp oooo oooo
|
|
38 |
#define _MRC(cc,p,i,r,c,c2,t) asm("mrc"#cc" p"#p", "#i", r"#r", c"#c", c"#c2", "#t )
|
|
39 |
#define _MCR(cc,p,i,r,c,c2,t) asm("mcr"#cc" p"#p", "#i", r"#r", c"#c", c"#c2", "#t )
|
|
40 |
#define _CDP(cc,p,i,d,n,m,t) asm("cdp"#cc" p"#p", "#i", c"#d", c"#n", c"#m", "#t )
|
|
41 |
|
|
42 |
#define _MRC(cc,p,i,r,c,c2,t) asm("mrc"#cc" p"#p", "#i", r"#r", c"#c", c"#c2", "#t )
|
|
43 |
#define _MCR(cc,p,i,r,c,c2,t) asm("mcr"#cc" p"#p", "#i", r"#r", c"#c", c"#c2", "#t )
|
|
44 |
#define _CDPS(cc,p,i,d,n,m,t) asm(".word %a0" : : "i" ((TInt) ((cc)<<28)|((p)<<8)|((i)<<20)|((t)<<5)| \
|
|
45 |
(((d)>>1)<<12)|(((n)>>1)<<16)|((m)>>1)| \
|
|
46 |
(((d)&1)<<22)|(((n)&1)<<7)|(((m)&1)<<5)| \
|
|
47 |
0x0e000000 ))
|
|
48 |
|
|
49 |
|
|
50 |
#define VFP_CPID_S 10 // coprocessor ID for single precision
|
|
51 |
#define VFP_CPID_D 11 // coprocessor ID for double precision
|
|
52 |
|
|
53 |
#define VFP_XREG_FPSID 0
|
|
54 |
#define VFP_XREG_FPSCR 1
|
|
55 |
#define VFP_XREG_MVFR0 7
|
|
56 |
#define VFP_XREG_FPEXC 8
|
|
57 |
#define VFP_XREG_FPINST 9
|
|
58 |
#define VFP_XREG_FPINST2 10
|
|
59 |
|
|
60 |
#define VFP_FPSID_IMP_SHIFT 24
|
|
61 |
#define VFP_FPSID_IMP_MASK (255u<<VFP_FPSID_IMP_SHIFT)
|
|
62 |
#define VFP_FPSID_SW 0x00800000 // software emulation
|
|
63 |
#define VFP_FPSID_FMT_SHIFT 21
|
|
64 |
#define VFP_FPSID_FMT_MASK (3<<VFP_FPSID_FMT_SHIFT)
|
|
65 |
#define VFP_FPSID_FMT1 0x00000000 // FLDMX/FSTMX format 1
|
|
66 |
#define VFP_FPSID_FMT2 0x00200000 // FLDMX/FSTMX format 2
|
|
67 |
#define VFP_FPSID_SNG 0x00100000 // single precision only
|
|
68 |
#define VFP_FPSID_ARCH_SHIFT 16
|
|
69 |
#define VFP_FPSID_ARCH_MASK (15<<VFP_FPSID_ARCH_SHIFT)
|
|
70 |
#define VFP_FPSID_PART_SHIFT 8
|
|
71 |
#define VFP_FPSID_PART_MASK (255<<VFP_FPSID_PART_SHIFT)
|
|
72 |
#define VFP_FPSID_VAR_SHIFT 4
|
|
73 |
#define VFP_FPSID_VAR_MASK (15<<VFP_FPSID_VAR_SHIFT)
|
|
74 |
#define VFP_FPSID_REV_MASK 15
|
|
75 |
|
|
76 |
|
|
77 |
#define VFP_FPSCR_N 0x80000000 // less than
|
|
78 |
#define VFP_FPSCR_Z 0x40000000 // equal
|
|
79 |
#define VFP_FPSCR_C 0x20000000 // equal greater or unordered
|
|
80 |
#define VFP_FPSCR_V 0x10000000 // unordered
|
|
81 |
#define VFP_FPSCR_DN 0x02000000 // enable default NAN mode
|
|
82 |
#define VFP_FPSCR_FZ 0x01000000 // enable flush to zero mode
|
|
83 |
#define VFP_FPSCR_RMODE_SHIFT 22
|
|
84 |
#define VFP_FPSCR_RMODE_MASK (3<<VFP_FPSCR_RMODE_SHIFT)
|
|
85 |
#define VFP_FPSCR_RMODE_NEAR 0x00000000 // round to nearest
|
|
86 |
#define VFP_FPSCR_RMODE_PLUS 0x00400000 // round up
|
|
87 |
#define VFP_FPSCR_RMODE_MINUS 0x00800000 // round down
|
|
88 |
#define VFP_FPSCR_RMODE_ZERO 0x00C00000 // round towards zero
|
|
89 |
|
|
90 |
#define VFP_FPSCR_STRIDE_SHIFT 20
|
|
91 |
#define VFP_FPSCR_STRIDE_MASK (3<<VFP_FPSCR_STRIDE_SHIFT)
|
|
92 |
#define VFP_FPSCR_LEN_SHIFT 16
|
|
93 |
#define VFP_FPSCR_LEN_MASK (7<<VFP_FPSCR_LEN_SHIFT)
|
|
94 |
#define VFP_FPSCR_IDE 0x00008000 // enable input subnormal exception
|
|
95 |
#define VFP_FPSCR_IXE 0x00001000 // enable inexact exception
|
|
96 |
#define VFP_FPSCR_UFE 0x00000800 // enable underflow exception
|
|
97 |
#define VFP_FPSCR_OFE 0x00000400 // enable overflow exception
|
|
98 |
#define VFP_FPSCR_DZE 0x00000200 // enable division by zero exception
|
|
99 |
#define VFP_FPSCR_IOE 0x00000100 // enable invalid operation exception
|
|
100 |
#define VFP_FPSCR_IDC 0x00000080 // input subnormal cumulative flag
|
|
101 |
#define VFP_FPSCR_IXC 0x00000010 // inexact cumulative flag
|
|
102 |
#define VFP_FPSCR_UFC 0x00000008 // underflow cumulative flag
|
|
103 |
#define VFP_FPSCR_OFC 0x00000004 // overflow cumulative flag
|
|
104 |
#define VFP_FPSCR_DZC 0x00000002 // division by zero cumulative flag
|
|
105 |
#define VFP_FPSCR_IOC 0x00000001 // invalid operation cumulative flag
|
|
106 |
|
|
107 |
#define VFP_FPSCR_RUNFAST (VFP_FPSCR_DN|VFP_FPSCR_FZ)
|
|
108 |
#define VFP_FPSCR_IEEE_NO_EXC 0
|
|
109 |
#define VFP_FPSCR_EXCEPTIONS (VFP_FPSCR_IDE|VFP_FPSCR_IXE|VFP_FPSCR_UFE|VFP_FPSCR_OFE|VFP_FPSCR_DZE|VFP_FPSCR_IOE)
|
|
110 |
#define VFP_FPSCR_MODE_MASK (VFP_FPSCR_EXCEPTIONS|VFP_FPSCR_RUNFAST|VFP_FPSCR_RMODE_MASK)
|
|
111 |
|
|
112 |
#define VFP_FPEXC_EX 0x80000000 // exceptional state
|
|
113 |
#define VFP_FPEXC_EN 0x40000000 // enable VFP
|
|
114 |
#define VFP_FPEXC_FP2V 0x10000000 // FPINST2 register valid
|
|
115 |
#define VFP_FPEXC_VECITR_SHIFT 8
|
|
116 |
#define VFP_FPEXC_VECITR_MASK (7<<VFP_FPEXC_VECITR_SHIFT) // (remaining iterations - 1) mod 7
|
|
117 |
#define VFP_FPEXC_INV 0x00000080 // input exception flag (subnormal or NaN)
|
|
118 |
#define VFP_FPEXC_UFC 0x00000008 // underflow cumulative flag
|
|
119 |
#define VFP_FPEXC_OFC 0x00000004 // overflow cumulative flag
|
|
120 |
#define VFP_FPEXC_IOC 0x00000001 // invalid operation cumulative flag
|
|
121 |
|
|
122 |
#define VFP_FPEXC_INIT (VFP_FPEXC_EN|VFP_FPEXC_VECITR_MASK)
|
|
123 |
#define VFP_FPEXC_THRD_INIT (VFP_FPEXC_VECITR_MASK)
|
|
124 |
|
|
125 |
#define VFP_MVFR0_ASIMD32 0x00000002 // Full 32 x 64-bit registers are supported for Advanced SIMD
|
|
126 |
#define VFP_CPACR_ASEDIS 0x80000000 // Access to the NEON unit is disabled
|
|
127 |
#define VFP_CPACR_D32DIS 0x40000000 // Access to the upper 16 64-bit registers is disabled
|
|
128 |
|
|
129 |
#define VFP_FMRX(cc,Rd,reg) _MRC(cc,10,7,Rd,reg,0,0)
|
|
130 |
#define VFP_FMXR(cc,reg,Rd) _MCR(cc,10,7,Rd,reg,0,0)
|
|
131 |
|
|
132 |
// VFPv3 adds D16-D31 extra double precision registers in previously UNDEF opcodes
|
|
133 |
#define _VFP_DN(Dn) (((Dn)>=16) ? ((Dn)-16) : (Dn))
|
|
134 |
#define _VFP_U(Dn) (((Dn)>=16) ? 1 : 0)
|
|
135 |
#define _VFP_D(Dn) (_VFP_U(Dn)<<22)
|
|
136 |
#define _VFP_N(Dn) (_VFP_U(Dn)<<7)
|
|
137 |
|
|
138 |
#define VFP_FLDMIAX(cc,Rn,Dd,N) asm(".word %a0" : : "i" ((TInt)( ((cc)<<28)|((Rn)<<16)|(_VFP_DN(Dd)<<12)|(2*(N)+1)|_VFP_D(Dd)|0x0c900b00 )) )
|
|
139 |
#define VFP_FSTMIAX(cc,Rn,Dd,N) asm(".word %a0" : : "i" ((TInt)( ((cc)<<28)|((Rn)<<16)|(_VFP_DN(Dd)<<12)|(2*(N)+1)|_VFP_D(Dd)|0x0c800b00 )) )
|
|
140 |
#define VFP_FLDMIAXW(cc,Rn,Dd,N) asm(".word %a0" : : "i" ((TInt)( ((cc)<<28)|((Rn)<<16)|(_VFP_DN(Dd)<<12)|(2*(N)+1)|_VFP_D(Dd)|0x0cb00b00 )) )
|
|
141 |
#define VFP_FSTMIAXW(cc,Rn,Dd,N) asm(".word %a0" : : "i" ((TInt)( ((cc)<<28)|((Rn)<<16)|(_VFP_DN(Dd)<<12)|(2*(N)+1)|_VFP_D(Dd)|0x0ca00b00 )) )
|
|
142 |
#define VFP_FLDMDBXW(cc,Rn,Dd,N) asm(".word %a0" : : "i" ((TInt)( ((cc)<<28)|((Rn)<<16)|(_VFP_DN(Dd)<<12)|(2*(N)+1)|_VFP_D(Dd)|0x0d300b00 )) )
|
|
143 |
#define VFP_FSTMDBXW(cc,Rn,Dd,N) asm(".word %a0" : : "i" ((TInt)( ((cc)<<28)|((Rn)<<16)|(_VFP_DN(Dd)<<12)|(2*(N)+1)|_VFP_D(Dd)|0x0d200b00 )) )
|
|
144 |
|
|
145 |
#define VFP_FMDLR(cc,Dn,Rd) asm(".word %a0" : : "i" ((TInt)( ((cc)<<28)|((Rd)<<12)|(_VFP_DN(Dn)<<16)|_VFP_N(Dn)|0x0e000b10 )) )
|
|
146 |
#define VFP_FMDHR(cc,Dn,Rd) asm(".word %a0" : : "i" ((TInt)( ((cc)<<28)|((Rd)<<12)|(_VFP_DN(Dn)<<16)|_VFP_N(Dn)|0x0e200b10 )) )
|
|
147 |
#define VFP_FMRDL(cc,Rd,Dn) asm(".word %a0" : : "i" ((TInt)( ((cc)<<28)|((Rd)<<12)|(_VFP_DN(Dn)<<16)|_VFP_N(Dn)|0x0e100b10 )) )
|
|
148 |
#define VFP_FMRDH(cc,Rd,Dn) asm(".word %a0" : : "i" ((TInt)( ((cc)<<28)|((Rd)<<12)|(_VFP_DN(Dn)<<16)|_VFP_N(Dn)|0x0e300b10 )) )
|
|
149 |
|
|
150 |
|
|
151 |
#ifdef __CPU_ARM_HAS_MCRR
|
|
152 |
#define VFP_FMRRD(cc,Rd,Rn,Dm) MRRCcc(cc,11,1,Rd,Rn,Dm)
|
|
153 |
#define VFP_FMDRR(cc,Dm,Rd,Rn) MCRRcc(cc,11,1,Rd,Rn,Dm)
|
|
154 |
#endif
|
|
155 |
|
|
156 |
// MRCcccc Ppppp, iii, Rdddd, Cnnnn, Cmmmm, ttt
|
|
157 |
// cccc 1110 iii1 nnnn dddd pppp ttt1 mmmm
|
|
158 |
// MCRcccc Ppppp, iii, Rdddd, Cnnnn, Cmmmm, ttt
|
|
159 |
// cccc 1110 iii0 nnnn dddd pppp ttt1 mmmm
|
|
160 |
// CDPcccc pppp, iiii, Cdddd, Cnnnn, Cmmmm, ttt
|
|
161 |
// cccc 1110 iiii nnnn dddd pppp ttt0 mmmm
|
|
162 |
#define VFP_FABSD(cc,Dd,Dm) _CDP(cc,11,11,Dd,0,Dm,6)
|
|
163 |
#define VFP_FADDD(cc,Dd,Dn,Dm) _CDP(cc,11,3,Dd,Dn,Dm,0)
|
|
164 |
#define VFP_FCMPD(cc,Dd,Dm) _CDP(cc,11,11,Dd,4,Dm,2)
|
|
165 |
#define VFP_FCMPED(cc,Dd,Dm) _CDP(cc,11,11,Dd,4,Dm,6)
|
|
166 |
#define VFP_FCMPEZD(cc,Dd) _CDP(cc,11,11,Dd,5,0,6)
|
|
167 |
#define VFP_FCMPZD(cc,Dd) _CDP(cc,11,11,Dd,5,0,2)
|
|
168 |
#define VFP_FCPYD(cc,Dd,Dm) _CDP(cc,11,11,Dd,0,Dm,2)
|
|
169 |
#define VFP_FDIVD(cc,Dd,Dn,Dm) _CDP(cc,11,8,Dd,Dn,Dm,0)
|
|
170 |
#define VFP_FMACD(cc,Dd,Dn,Dm) _CDP(cc,11,0,Dd,Dn,Dm,0)
|
|
171 |
#define VFP_FMSCD(cc,Dd,Dn,Dm) _CDP(cc,11,1,Dd,Dn,Dm,0)
|
|
172 |
#define VFP_FMULD(cc,Dd,Dn,Dm) _CDP(cc,11,2,Dd,Dn,Dm,0)
|
|
173 |
#define VFP_FNEGD(cc,Dd,Dm) _CDP(cc,11,11,Dd,1,Dm,2)
|
|
174 |
#define VFP_FNMACD(cc,Dd,Dn,Dm) _CDP(cc,11,0,Dd,Dn,Dm,2)
|
|
175 |
#define VFP_FNMSCD(cc,Dd,Dn,Dm) _CDP(cc,11,1,Dd,Dn,Dm,2)
|
|
176 |
#define VFP_FNMULD(cc,Dd,Dn,Dm) _CDP(cc,11,2,Dd,Dn,Dm,2)
|
|
177 |
#define VFP_FSQRTD(cc,Dd,Dm) _CDP(cc,11,11,Dd,1,Dm,6)
|
|
178 |
#define VFP_FSUBD(cc,Dd,Dn,Dm) _CDP(cc,11,3,Dd,Dn,Dm,2)
|
|
179 |
#define VFP_FMSTAT(cc) _MRC(cc,10,7,15,1,0,0)
|
|
180 |
|
|
181 |
#define _VFP_ADDR_U(off) ((off)>=0 ? 1 : 0)
|
|
182 |
#define _VFP_ADDR_O(off) ((off)>=0 ? (off) : -(off))
|
|
183 |
#define VFP_FLDD(cc,Dd,Rn,off) asm(".word %a0" : : "i" ((TInt)( ((cc)<<28)|(_VFP_D(Dd))|((Rn)<<16)|(_VFP_DN(Dd)<<12)|(_VFP_ADDR_U(off)<<23)|_VFP_ADDR_O(off)|0x0d100b00)))
|
|
184 |
#define VFP_FSTD(cc,Dd,Rn,off) asm(".word %a0" : : "i" ((TInt)( ((cc)<<28)|(_VFP_D(Dd))|((Rn)<<16)|(_VFP_DN(Dd)<<12)|(_VFP_ADDR_U(off)<<23)|_VFP_ADDR_O(off)|0x0d000b00)))
|
|
185 |
#define VFP_FLDMIAD(cc,Rn,Dd,N) asm(".word %a0" : : "i" ((TInt)( ((cc)<<28)|(_VFP_D(Dd))|((Rn)<<16)|(_VFP_DN(Dd)<<12)|(2*(N))|0x0c900b00 )) )
|
|
186 |
#define VFP_FSTMIAD(cc,Rn,Dd,N) asm(".word %a0" : : "i" ((TInt)( ((cc)<<28)|(_VFP_D(Dd))|((Rn)<<16)|(_VFP_DN(Dd)<<12)|(2*(N))|0x0c800b00 )) )
|
|
187 |
#define VFP_FLDMIADW(cc,Rn,Dd,N) asm(".word %a0" : : "i" ((TInt)( ((cc)<<28)|(_VFP_D(Dd))|((Rn)<<16)|(_VFP_DN(Dd)<<12)|(2*(N))|0x0cb00b00 )) )
|
|
188 |
#define VFP_FSTMIADW(cc,Rn,Dd,N) asm(".word %a0" : : "i" ((TInt)( ((cc)<<28)|(_VFP_D(Dd))|((Rn)<<16)|(_VFP_DN(Dd)<<12)|(2*(N))|0x0ca00b00 )) )
|
|
189 |
#define VFP_FLDMDBDW(cc,Rn,Dd,N) asm(".word %a0" : : "i" ((TInt)( ((cc)<<28)|(_VFP_D(Dd))|((Rn)<<16)|(_VFP_DN(Dd)<<12)|(2*(N))|0x0d300b00 )) )
|
|
190 |
#define VFP_FSTMDBDW(cc,Rn,Dd,N) asm(".word %a0" : : "i" ((TInt)( ((cc)<<28)|(_VFP_D(Dd))|((Rn)<<16)|(_VFP_DN(Dd)<<12)|(2*(N))|0x0d200b00 )) )
|
|
191 |
|
|
192 |
#define VFP_FABSS(cc,Sd,Sm) _CDPS(cc,10,11,Sd,0,Sm,6)
|
|
193 |
#define VFP_FADDS(cc,Sd,Sn,Sm) _CDPS(cc,10,3,Sd,Sn,Sm,0)
|
|
194 |
#define VFP_FCMPS(cc,Sd,Sm) _CDPS(cc,10,11,Sd,8,Sm,2)
|
|
195 |
#define VFP_FCMPES(cc,Sd,Sm) _CDPS(cc,10,11,Sd,8,Sm,6)
|
|
196 |
#define VFP_FCMPEZS(cc,Sd) _CDPS(cc,10,11,Sd,11,0,6)
|
|
197 |
#define VFP_FCMPZS(cc,Sd) _CDPS(cc,10,11,Sd,10,0,2)
|
|
198 |
#define VFP_FCPYS(cc,Sd,Sm) _CDPS(cc,10,11,Sd,0,Sm,2)
|
|
199 |
#define VFP_FDIVS(cc,Sd,Sn,Sm) _CDPS(cc,10,8,Sd,Sn,Sm,0)
|
|
200 |
#define VFP_FMACS(cc,Sd,Sn,Sm) _CDPS(cc,10,0,Sd,Sn,Sm,0)
|
|
201 |
#define VFP_FMSCS(cc,Sd,Sn,Sm) _CDPS(cc,10,1,Sd,Sn,Sm,0)
|
|
202 |
#define VFP_FMULS(cc,Sd,Sn,Sm) _CDPS(cc,10,2,Sd,Sn,Sm,0)
|
|
203 |
#define VFP_FNEGS(cc,Sd,Sm) _CDPS(cc,10,11,Sd,2,Sm,2)
|
|
204 |
#define VFP_FNMACS(cc,Sd,Sn,Sm) _CDPS(cc,10,0,Sd,Sn,Sm,2)
|
|
205 |
#define VFP_FNMSCS(cc,Sd,Sn,Sm) _CDPS(cc,10,1,Sd,Sn,Sm,2)
|
|
206 |
#define VFP_FNMULS(cc,Sd,Sn,Sm) _CDPS(cc,10,2,Sd,Sn,Sm,2)
|
|
207 |
#define VFP_FSQRTS(cc,Sd,Sm) _CDPS(cc,10,11,Sd,3,Sm,6)
|
|
208 |
#define VFP_FSUBS(cc,Sd,Sn,Sm) _CDPS(cc,10,3,Sd,Sn,Sm,2)
|
|
209 |
|
|
210 |
#define VFP_FLDS(cc,Sd,Rn,off) asm(".word %a0" : : "i" ((TInt) ((cc)<<28)|(((Sd)>>1)<<12)|(((Sd)&1)<<22)|((Rn)<<16)|(_VFP_ADDR_U(off)<<23)|_VFP_ADDR_O(off)|0x0d100a00))
|
|
211 |
#define VFP_FSTS(cc,Sd,Rn,off) asm(".word %a0" : : "i" ((TInt) ((cc)<<28)|(((Sd)>>1)<<12)|(((Sd)&1)<<22)|((Rn)<<16)|(_VFP_ADDR_U(off)<<23)|_VFP_ADDR_O(off)|0x0d000a00))
|
|
212 |
#define VFP_FLDMIAS(cc,Rn,Sd,N) asm(".word %a0" : : "i" ((TInt)( ((cc)<<28)|((Rn)<<16)|(((Sd)>>1)<<12)|(((Sd)&1)<<22)|(N)|0x0c900a00 )) )
|
|
213 |
#define VFP_FSTMIAS(cc,Rn,Sd,N) asm(".word %a0" : : "i" ((TInt)( ((cc)<<28)|((Rn)<<16)|(((Sd)>>1)<<12)|(((Sd)&1)<<22)|(N)|0x0c800a00 )) )
|
|
214 |
#define VFP_FLDMIASW(cc,Rn,Sd,N) asm(".word %a0" : : "i" ((TInt)( ((cc)<<28)|((Rn)<<16)|(((Sd)>>1)<<12)|(((Sd)&1)<<22)|(N)|0x0cb00a00 )) )
|
|
215 |
#define VFP_FSTMIASW(cc,Rn,Sd,N) asm(".word %a0" : : "i" ((TInt)( ((cc)<<28)|((Rn)<<16)|(((Sd)>>1)<<12)|(((Sd)&1)<<22)|(N)|0x0ca00a00 )) )
|
|
216 |
#define VFP_FLDMDBSW(cc,Rn,Sd,N) asm(".word %a0" : : "i" ((TInt)( ((cc)<<28)|((Rn)<<16)|(((Sd)>>1)<<12)|(((Sd)&1)<<22)|(N)|0x0d300a00 )) )
|
|
217 |
#define VFP_FSTMDBSW(cc,Rn,Sd,N) asm(".word %a0" : : "i" ((TInt)( ((cc)<<28)|((Rn)<<16)|(((Sd)>>1)<<12)|(((Sd)&1)<<22)|(N)|0x0d200a00 )) )
|
|
218 |
|
|
219 |
#define VFP_FMSR(cc,Sn,Rd) asm(".word %a0" : : "i" ((TInt) ((cc)<<28)|(((Sn)>>1)<<16)|(((Sn)&1)<<7)|((Rd)<<12)|0x0e000a10))
|
|
220 |
#define VFP_FMRS(cc,Rd,Sn) asm(".word %a0" : : "i" ((TInt) ((cc)<<28)|(((Sn)>>1)<<16)|(((Sn)&1)<<7)|((Rd)<<12)|0x0e100a10))
|
|
221 |
|
|
222 |
#define VFP_FCVTDS(cc,Dd,Sm) asm(".word %a0" : : "i" ((TInt) ((cc)<<28)|((Sm)>>1)|(((Sm)&1)<<5)|((Dd)<<12)|0x0eb70ac0))
|
|
223 |
#define VFP_FCVTSD(cc,Sd,Dm) asm(".word %a0" : : "i" ((TInt) ((cc)<<28)|(((Sd)>>1)<<12)|(((Sd)&1)<<22)|(Dm)|0x0eb70bc0))
|
|
224 |
|
|
225 |
#define VFP_FSITOD(cc,Dd,Sm) asm(".word %a0" : : "i" ((TInt) ((cc)<<28)|((Sm)>>1)|(((Sm)&1)<<5)|((Dd)<<12)|0x0eb80bc0))
|
|
226 |
#define VFP_FSITOS(cc,Sd,Sm) asm(".word %a0" : : "i" ((TInt) ((cc)<<28)|((Sm)>>1)|(((Sm)&1)<<5)|(((Sd)>>1)<<12)|(((Sd)&1)<<22)|0x0eb80ac0))
|
|
227 |
#define VFP_FTOSID(cc,Sd,Dm) asm(".word %a0" : : "i" ((TInt) ((cc)<<28)|(((Sd)>>1)<<12)|(((Sd)&1)<<22)|(Dm)|0x0ebd0b40))
|
|
228 |
#define VFP_FTOSIZD(cc,Sd,Dm) asm(".word %a0" : : "i" ((TInt) ((cc)<<28)|(((Sd)>>1)<<12)|(((Sd)&1)<<22)|(Dm)|0x0ebd0bc0))
|
|
229 |
#define VFP_FTOSIS(cc,Sd,Sm) asm(".word %a0" : : "i" ((TInt) ((cc)<<28)|((Sm)>>1)|(((Sm)&1)<<5)|(((Sd)>>1)<<12)|(((Sd)&1)<<22)|0x0ebd0a40))
|
|
230 |
#define VFP_FTOSIZS(cc,Sd,Sm) asm(".word %a0" : : "i" ((TInt) ((cc)<<28)|((Sm)>>1)|(((Sm)&1)<<5)|(((Sd)>>1)<<12)|(((Sd)&1)<<22)|0x0ebd0ac0))
|
|
231 |
#define VFP_FUITOD(cc,Dd,Sm) asm(".word %a0" : : "i" ((TInt) ((cc)<<28)|((Sm)>>1)|(((Sm)&1)<<5)|((Dd)<<12)|0x0eb80b40))
|
|
232 |
#define VFP_FUITOS(cc,Sd,Sm) asm(".word %a0" : : "i" ((TInt) ((cc)<<28)|((Sm)>>1)|(((Sm)&1)<<5)|(((Sd)>>1)<<12)|(((Sd)&1)<<22)|0x0eb80a40))
|
|
233 |
#define VFP_FTOUID(cc,Sd,Dm) asm(".word %a0" : : "i" ((TInt) ((cc)<<28)|(((Sd)>>1)<<12)|(((Sd)&1)<<22)|(Dm)|0x0ebc0b40))
|
|
234 |
#define VFP_FTOUIZD(cc,Sd,Dm) asm(".word %a0" : : "i" ((TInt) ((cc)<<28)|(((Sd)>>1)<<12)|(((Sd)&1)<<22)|(Dm)|0x0ebc0bc0))
|
|
235 |
#define VFP_FTOUIS(cc,Sd,Sm) asm(".word %a0" : : "i" ((TInt) ((cc)<<28)|((Sm)>>1)|(((Sm)&1)<<5)|(((Sd)>>1)<<12)|(((Sd)&1)<<22)|0x0ebc0a40))
|
|
236 |
#define VFP_FTOUIZS(cc,Sd,Sm) asm(".word %a0" : : "i" ((TInt) ((cc)<<28)|((Sm)>>1)|(((Sm)&1)<<5)|(((Sd)>>1)<<12)|(((Sd)&1)<<22)|0x0ebc0ac0))
|
|
237 |
|
|
238 |
// VFPv3
|
|
239 |
// conversion between floating point and fixed point
|
|
240 |
#define _VFP_VCVT_D(dp,d) ( (dp) ? ( (((d)>>4)<<22)|(((d)&0xf)<<12) ) : ( (((d)&1)<<22)|(((d)>>1)<<12) ) )
|
|
241 |
#define _VFP_I_IMM4(sx,fbits) (((sx)==0 ? 16 : 32) - (fbits))
|
|
242 |
#define _VFP_VCVT_FBITS(sx,fbits) ( ((_VFP_I_IMM4(sx,fbits)&1) <<5) | (_VFP_I_IMM4(sx,fbits)>>1) )
|
|
243 |
#define VFP_VCT(cc,op,sf,U,sx,d,fbits) asm(".word %a0" : : "i" ((TInt) ((cc)<<28)|((op)<<18)|((U)<<16)|((sf)<<8)|((sx)<<7)|_VFP_VCVT_D(sf,d)|_VFP_VCVT_FBITS(sx,fbits)|0x0eba0a40))
|
|
244 |
// from fixed to floating point
|
|
245 |
// S32=>F64
|
|
246 |
#define VFP_VCT_F64_S32(cc,Dd,fbits) VFP_VCT((cc),0,1,0,1,(Dd),(fbits))
|
|
247 |
// S32=>F32
|
|
248 |
#define VFP_VCT_F32_S32(cc,Sd,fbits) VFP_VCT((cc),0,0,0,1,(Sd),(fbits))
|
|
249 |
// from floating point to fix
|
|
250 |
// F64=>S32
|
|
251 |
#define VFP_VCT_S32_F64(cc,Dd,fbits) VFP_VCT((cc),1,1,0,1,(Dd),(fbits))
|
|
252 |
// F32=>S32
|
|
253 |
#define VFP_VCT_S32_F32(cc,Sd,fbits) VFP_VCT((cc),1,0,0,1,(Sd),(fbits))
|
|
254 |
|
|
255 |
|
|
256 |
// put immediate value to the register
|
|
257 |
// single_register=(sz==0)
|
|
258 |
// imm (abcdefgh)
|
|
259 |
/*
|
|
260 |
bcd 000 001 010 011 100 101 110 111
|
|
261 |
efgh
|
|
262 |
0000 2.0 4.0 8.0 16.0 0.125 0.25 0.5 1.0
|
|
263 |
0001 2.125 4.25 8.5 17.0 0.1328125 0.265625 0.53125 1.0625
|
|
264 |
0010 2.25 4.5 9.0 18.0 0.140625 0.28125 0.5625 1.125
|
|
265 |
0011 2.375 4.75 9.5 19.0 0.1484375 0.296875 0.59375 1.1875
|
|
266 |
0100 2.5 5.0 10.0 20.0 0.15625 0.3125 0.625 1.25
|
|
267 |
0101 2.625 5.25 10.5 21.0 0.1640625 0.328125 0.65625 1.3125
|
|
268 |
0110 2.75 5.5 11.0 22.0 0.171875 0.34375 0.6875 1.375
|
|
269 |
0111 2.875 5.75 11.5 23.0 0.1796875 0.359375 0.71875 1.4375
|
|
270 |
1000 3.0 6.0 12.0 24.0 0.1875 0.375 0.75 1.5
|
|
271 |
1001 3.125 6.25 12.5 25.0 0.1953125 0.390625 0.78125 1.5625
|
|
272 |
1010 3.25 6.5 13.0 26.0 0.203125 0.40625 0.8125 1.625
|
|
273 |
1011 3.375 6.75 13.5 27.0 0.2109375 0.421875 0.84375 1.6875
|
|
274 |
1100 3.5 7.0 14.0 28.0 0.21875 0.4375 0.875 1.75
|
|
275 |
1101 3.625 7.25 14.5 29.0 0.2265625 0.453125 0.90625 1.8125
|
|
276 |
1110 3.75 7.5 15.0 30.0 0.234375 0.46875 0.9375 1.875
|
|
277 |
1111 3.875 7.75 15.5 31.0 0.2421875 0.484375 0.96875 1.9375
|
|
278 |
*/
|
|
279 |
|
|
280 |
#define VFP_VMOV_IMM(cc,sz,d,imm) asm(".word %a0" : : "i" ((TInt) ((cc)<<28)|(((imm)>>4)<<16)|((imm)&0xf)|(sz)<<8|((sz)?((((d)>>4)<<22)|(((d)&0xf)<<12)):((((d)&1)<<22)|(((d)>>1)<<12)))|0x0eb00a00))
|
|
281 |
|
|
282 |
#endif
|
|
283 |
#endif
|