|
a/LibDBlasF77.def.m2pp |
|
b/LibDBlasF77.def.m2pp |
|
... |
|
... |
| 35 |
(* * ddot : dot product of two vectors *)
|
35 |
(* * ddot : dot product of two vectors *)
|
| 36 |
(* * idamax : index of vector element with largest absolute value *)
|
36 |
(* * idamax : index of vector element with largest absolute value *)
|
| 37 |
(* * idamin : index of vector element with smalles absolute value *)
|
37 |
(* * idamin : index of vector element with smalles absolute value *)
|
| 38 |
(* * dasum : sum of absolut values of real vector elements *)
|
38 |
(* * dasum : sum of absolut values of real vector elements *)
|
| 39 |
(* *)
|
39 |
(* *)
|
|
|
40 |
(* zswap : swap two vectors *)
|
|
|
41 |
(* zcopy : copy a vector to a vector *)
|
|
|
42 |
(* zdotc : dot product of two vectors *)
|
|
|
43 |
(* dznrm2 : euklidian norm of a vector *)
|
|
|
44 |
(* zscal : scale vector by a constant *)
|
|
|
45 |
(* * zaxpy : constant times a vector plus a vector *)
|
|
|
46 |
(* zdrot : plane rotation *)
|
|
|
47 |
(* *)
|
|
|
48 |
(* Level 2 *)
|
|
|
49 |
(* *)
|
|
|
50 |
(* dgemv : double precision matrix vector multiplication *)
|
|
|
51 |
(* zgemv : double complex matrix vector multiplication *)
|
|
|
52 |
(* *)
|
| 40 |
(* Level 3 *)
|
53 |
(* Level 3 *)
|
| 41 |
(* *)
|
54 |
(* *)
|
| 42 |
(* sgemm : single precision matrix matrix multiplication *)
|
55 |
(* sgemm : single precision matrix matrix multiplication *)
|
| 43 |
(* dgemm : double precision matrix matrix multiplication *)
|
56 |
(* dgemm : double precision matrix matrix multiplication *)
|
| 44 |
(* zgemm : double complex matrix matrix multiplication *)
|
57 |
(* zgemm : double complex matrix matrix multiplication *)
|
|
... |
|
... |
| 54 |
(* Changed ARRAY OF DOUBLEPRECISION to *)
|
67 |
(* Changed ARRAY OF DOUBLEPRECISION to *)
|
| 55 |
(* (* ARRAY OF *) DOUBLEPRECISION *)
|
68 |
(* (* ARRAY OF *) DOUBLEPRECISION *)
|
| 56 |
(* 29.10.17, MRi: Added dgemm *)
|
69 |
(* 29.10.17, MRi: Added dgemm *)
|
| 57 |
(* 01.11.17, MRi: Added idamin *)
|
70 |
(* 01.11.17, MRi: Added idamin *)
|
| 58 |
(* 23.06.18, MRi: Added zgemm *)
|
71 |
(* 23.06.18, MRi: Added zgemm *)
|
|
|
72 |
(* 11.09.18, MRi: Added zswap,zcopy,zdotc,dznrm2,zscal,zaxpy,zdrot *)
|
|
|
73 |
(* and dgemv, zgemv *)
|
| 59 |
(*------------------------------------------------------------------------*)
|
74 |
(*------------------------------------------------------------------------*)
|
| 60 |
(* Implementation : Michael Riedl *)
|
75 |
(* Implementation : Michael Riedl *)
|
| 61 |
(* Licence : GNU Lesser General Public License (LGPL) *)
|
76 |
(* Licence : GNU Lesser General Public License (LGPL) *)
|
| 62 |
(*------------------------------------------------------------------------*)
|
77 |
(*------------------------------------------------------------------------*)
|
| 63 |
|
78 |
|
| 64 |
(* $Id: LibDBlasF77.def.m2pp,v 1.1 2018/06/15 08:59:31 mriedl Exp mriedl $ *)
|
79 |
(* $Id: LibDBlasF77.def.m2pp,v 1.3 2018/09/12 13:20:49 mriedl Exp mriedl $ *)
|
| 65 |
|
80 |
|
| 66 |
IMPORT LibDBlasL1F77;
|
81 |
IMPORT LibDBlasL1F77;
|
|
|
82 |
IMPORT LibDBlasL2F77;
|
| 67 |
IMPORT LibDBlasL3F77;
|
83 |
IMPORT LibDBlasL3F77;
|
| 68 |
|
84 |
|
| 69 |
TYPE DOUBLEPRECISION = LibDBlasL1F77.DOUBLEPRECISION;
|
85 |
TYPE DOUBLEPRECISION = LibDBlasL1F77.DOUBLEPRECISION;
|
| 70 |
INTEGER4 = LibDBlasL1F77.INTEGER4;
|
86 |
INTEGER4 = LibDBlasL1F77.INTEGER4;
|
| 71 |
CHAR1 = LibDBlasL1F77.CHAR1;
|
87 |
CHAR1 = LibDBlasL1F77.CHAR1;
|
| 72 |
|
88 |
|
| 73 |
CONST dnrm2 = LibDBlasL1F77.dnrm2_;
|
|
|
| 74 |
|
|
|
| 75 |
(*
|
89 |
(*
|
| 76 |
* PROCEDURE dnrm2(VAR n : INTEGER4;
|
90 |
* PROCEDURE dnrm2(VAR n : INTEGER4;
|
| 77 |
* VAR X : (* ARRAY OF *) DOUBLEPRECISION;
|
91 |
* VAR X : (* ARRAY OF *) DOUBLEPRECISION;
|
| 78 |
* VAR IncX : INTEGER4): DOUBLEPRECISION;
|
92 |
* VAR IncX : INTEGER4): DOUBLEPRECISION;
|
| 79 |
*)
|
93 |
*)
|
|
|
94 |
|
|
|
95 |
CONST dnrm2 = LibDBlasL1F77.dnrm2_;
|
|
|
96 |
|
| 80 |
(*----------------------------------------------------------------*)
|
97 |
(*----------------------------------------------------------------*)
|
| 81 |
(* dnrm2 returns the euclidean norm of a vector via the function *)
|
98 |
(* dnrm2 returns the euclidean norm of a vector via the function *)
|
| 82 |
(* name, so that dnrm2 := sqrt( x^{tr}*x ) *)
|
99 |
(* name, so that dnrm2 := sqrt( x^{tr}*x ) *)
|
| 83 |
(*----------------------------------------------------------------*)
|
100 |
(*----------------------------------------------------------------*)
|
| 84 |
|
|
|
| 85 |
CONST dswap = LibDBlasL1F77.dswap_;
|
|
|
| 86 |
|
101 |
|
| 87 |
(*
|
102 |
(*
|
| 88 |
* PROCEDURE dswap(VAR N : INTEGER4;
|
103 |
* PROCEDURE dswap(VAR N : INTEGER4;
|
| 89 |
* VAR X : (* ARRAY OF *) DOUBLEPRECISION;
|
104 |
* VAR X : (* ARRAY OF *) DOUBLEPRECISION;
|
| 90 |
* VAR incx : INTEGER4;
|
105 |
* VAR incx : INTEGER4;
|
| 91 |
* VAR Y : (* ARRAY OF *) DOUBLEPRECISION;
|
106 |
* VAR Y : (* ARRAY OF *) DOUBLEPRECISION;
|
| 92 |
* VAR incy : INTEGER4);
|
107 |
* VAR incy : INTEGER4);
|
| 93 |
*)
|
108 |
*)
|
|
|
109 |
|
|
|
110 |
CONST dswap = LibDBlasL1F77.dswap_;
|
|
|
111 |
|
| 94 |
(*----------------------------------------------------------------*)
|
112 |
(*----------------------------------------------------------------*)
|
| 95 |
(* Interchanges two vectors using unrolled loops for increments *)
|
113 |
(* Interchanges two vectors using unrolled loops for increments *)
|
| 96 |
(* equal to 1. *)
|
114 |
(* equal to 1. *)
|
| 97 |
(*----------------------------------------------------------------*)
|
115 |
(*----------------------------------------------------------------*)
|
| 98 |
|
|
|
| 99 |
CONST dcopy = LibDBlasL1F77.dcopy_;
|
|
|
| 100 |
|
116 |
|
| 101 |
(*
|
117 |
(*
|
| 102 |
* PROCEDURE dcopy(VAR N : INTEGER4;
|
118 |
* PROCEDURE dcopy(VAR N : INTEGER4;
|
| 103 |
* VAR X : (* ARRAY OF *) DOUBLEPRECISION;
|
119 |
* VAR X : (* ARRAY OF *) DOUBLEPRECISION;
|
| 104 |
* VAR IncX : INTEGER4;
|
120 |
* VAR IncX : INTEGER4;
|
| 105 |
* VAR Y : (* ARRAY OF *) DOUBLEPRECISION;
|
121 |
* VAR Y : (* ARRAY OF *) DOUBLEPRECISION;
|
| 106 |
* VAR IncY : INTEGER);
|
122 |
* VAR IncY : INTEGER);
|
| 107 |
*)
|
123 |
*)
|
|
|
124 |
|
|
|
125 |
CONST dcopy = LibDBlasL1F77.dcopy_;
|
|
|
126 |
|
| 108 |
(*----------------------------------------------------------------*)
|
127 |
(*----------------------------------------------------------------*)
|
| 109 |
(* copies a vector, x, to a vector, y. *)
|
128 |
(* copies a vector, x, to a vector, y. *)
|
| 110 |
(* uses unrolled loops for increments equal to one. *)
|
129 |
(* uses unrolled loops for increments equal to one. *)
|
| 111 |
(* jack dongarra, linpack, 3/11/78. *)
|
130 |
(* jack dongarra, linpack, 3/11/78. *)
|
| 112 |
(*----------------------------------------------------------------*)
|
131 |
(*----------------------------------------------------------------*)
|
| 113 |
|
|
|
| 114 |
CONST drot = LibDBlasL1F77.drot_;
|
|
|
| 115 |
|
132 |
|
| 116 |
(*
|
133 |
(*
|
| 117 |
* PROCEDURE drot(VAR N : INTEGER4;
|
134 |
* PROCEDURE drot(VAR N : INTEGER4;
|
| 118 |
* VAR X : (* ARRAY OF *) DOUBLEPRECISION;
|
135 |
* VAR X : (* ARRAY OF *) DOUBLEPRECISION;
|
| 119 |
* VAR incX : INTEGER4;
|
136 |
* VAR incX : INTEGER4;
|
| 120 |
* VAR Y : (* ARRAY OF *) DOUBLEPRECISION;
|
137 |
* VAR Y : (* ARRAY OF *) DOUBLEPRECISION;
|
| 121 |
* VAR incY : INTEGER4;
|
138 |
* VAR incY : INTEGER4;
|
| 122 |
* VAR c,s : DOUBLEPRECISION);
|
139 |
* VAR c,s : DOUBLEPRECISION);
|
| 123 |
*)
|
140 |
*)
|
|
|
141 |
|
|
|
142 |
CONST drot = LibDBlasL1F77.drot_;
|
|
|
143 |
|
| 124 |
(*---------------------------------------------------------------*)
|
144 |
(*---------------------------------------------------------------*)
|
| 125 |
(* Applies a plane rotation. *)
|
145 |
(* Applies a plane rotation. *)
|
| 126 |
(* Jack Dongarra, linpack, 3/11/78. *)
|
146 |
(* Jack Dongarra, linpack, 3/11/78. *)
|
| 127 |
(*---------------------------------------------------------------*)
|
147 |
(*---------------------------------------------------------------*)
|
| 128 |
|
148 |
|
| 129 |
CONST drotg = LibDBlasL1F77.drotg_;
|
|
|
| 130 |
|
|
|
| 131 |
(*
|
149 |
(*
|
| 132 |
* PROCEDURE drotg(VAR da : DOUBLEPRECISION;
|
150 |
* PROCEDURE drotg(VAR da : DOUBLEPRECISION;
|
| 133 |
* VAR db : DOUBLEPRECISION;
|
151 |
* VAR db : DOUBLEPRECISION;
|
| 134 |
* VAR c : DOUBLEPRECISION;
|
152 |
* VAR c : DOUBLEPRECISION;
|
| 135 |
* VAR s : DOUBLEPRECISION);
|
153 |
* VAR s : DOUBLEPRECISION);
|
| 136 |
*)
|
154 |
*)
|
|
|
155 |
|
|
|
156 |
CONST drotg = LibDBlasL1F77.drotg_;
|
|
|
157 |
|
| 137 |
(*---------------------------------------------------------------*)
|
158 |
(*---------------------------------------------------------------*)
|
| 138 |
(* Construct a Givens plane rotation *)
|
159 |
(* Construct a Givens plane rotation *)
|
| 139 |
(* Jack Dongarra, linpack, 3/11/78. *)
|
160 |
(* Jack Dongarra, linpack, 3/11/78. *)
|
| 140 |
(*---------------------------------------------------------------*)
|
161 |
(*---------------------------------------------------------------*)
|
| 141 |
|
162 |
|
| 142 |
CONST dscal = LibDBlasL1F77.dscal_;
|
|
|
| 143 |
|
|
|
| 144 |
(*
|
163 |
(*
|
| 145 |
* PROCEDURE dscal(VAR n : INTEGER4;
|
164 |
* PROCEDURE dscal(VAR n : INTEGER4;
|
| 146 |
* VAR da : DOUBLEPRECISION;
|
165 |
* VAR da : DOUBLEPRECISION;
|
| 147 |
* VAR dx : (* ARRAY OF *) DOUBLEPRECISION;
|
166 |
* VAR dx : (* ARRAY OF *) DOUBLEPRECISION;
|
| 148 |
* VAR incx : INTEGER4);
|
167 |
* VAR incx : INTEGER4);
|
| 149 |
*)
|
168 |
*)
|
|
|
169 |
|
|
|
170 |
CONST dscal = LibDBlasL1F77.dscal_;
|
|
|
171 |
|
| 150 |
(*----------------------------------------------------------------*)
|
172 |
(*----------------------------------------------------------------*)
|
| 151 |
(* Scales a vector by a constant *)
|
173 |
(* Scales a vector by a constant *)
|
| 152 |
(*----------------------------------------------------------------*)
|
174 |
(*----------------------------------------------------------------*)
|
| 153 |
|
|
|
| 154 |
CONST daxpy = LibDBlasL1F77.daxpy_;
|
|
|
| 155 |
|
175 |
|
| 156 |
(*
|
176 |
(*
|
| 157 |
* PROCEDURE daxpy(VAR n : INTEGER4;
|
177 |
* PROCEDURE daxpy(VAR n : INTEGER4;
|
| 158 |
* VAR da : DOUBLEPRECISION;
|
178 |
* VAR da : DOUBLEPRECISION;
|
| 159 |
* VAR dx : (* ARRAY OF *) DOUBLEPRECISION;
|
179 |
* VAR dx : (* ARRAY OF *) DOUBLEPRECISION;
|
| 160 |
* VAR incx : INTEGER4;
|
180 |
* VAR incx : INTEGER4;
|
| 161 |
* VAR dy : (* ARRAY OF *) DOUBLEPRECISION;
|
181 |
* VAR dy : (* ARRAY OF *) DOUBLEPRECISION;
|
| 162 |
* VAR incy : INTEGER4);
|
182 |
* VAR incy : INTEGER4);
|
| 163 |
*)
|
183 |
*)
|
|
|
184 |
|
|
|
185 |
CONST daxpy = LibDBlasL1F77.daxpy_;
|
|
|
186 |
|
| 164 |
(*----------------------------------------------------------------*)
|
187 |
(*----------------------------------------------------------------*)
|
| 165 |
(* Constant times a vector plus a vector *)
|
188 |
(* Constant times a vector plus a vector *)
|
| 166 |
(*----------------------------------------------------------------*)
|
189 |
(*----------------------------------------------------------------*)
|
| 167 |
|
|
|
| 168 |
CONST ddot = LibDBlasL1F77.ddot_;
|
|
|
| 169 |
|
190 |
|
| 170 |
(*
|
191 |
(*
|
| 171 |
* PROCEDURE ddot(VAR N : INTEGER4;
|
192 |
* PROCEDURE ddot(VAR N : INTEGER4;
|
| 172 |
* VAR DX : (* ARRAY OF *) DOUBLEPRECISION;
|
193 |
* VAR DX : (* ARRAY OF *) DOUBLEPRECISION;
|
| 173 |
* VAR INCX : INTEGER4;
|
194 |
* VAR INCX : INTEGER4;
|
| 174 |
* VAR DY : (* ARRAY OF *) DOUBLEPRECISION;
|
195 |
* VAR DY : (* ARRAY OF *) DOUBLEPRECISION;
|
| 175 |
* VAR INCY : INTEGER4) : DOUBLEPRECISION;
|
196 |
* VAR INCY : INTEGER4) : DOUBLEPRECISION;
|
| 176 |
*)
|
197 |
*)
|
|
|
198 |
|
|
|
199 |
CONST ddot = LibDBlasL1F77.ddot_;
|
|
|
200 |
|
| 177 |
(*----------------------------------------------------------------*)
|
201 |
(*----------------------------------------------------------------*)
|
| 178 |
(* Forms the dot product of two vectors *)
|
202 |
(* Forms the dot product of two vectors *)
|
| 179 |
(*----------------------------------------------------------------*)
|
203 |
(*----------------------------------------------------------------*)
|
| 180 |
|
|
|
| 181 |
CONST idamax = LibDBlasL1F77.idamax_;
|
|
|
| 182 |
|
204 |
|
| 183 |
(*
|
205 |
(*
|
| 184 |
* PROCEDURE idamax(VAR n : INTEGER4;
|
206 |
* PROCEDURE idamax(VAR n : INTEGER4;
|
| 185 |
* VAR dx : (* ARRAY OF *) DOUBLEPRECISION;
|
207 |
* VAR dx : (* ARRAY OF *) DOUBLEPRECISION;
|
| 186 |
* VAR incx : INTEGER4) : INTEGER4;
|
208 |
* VAR incx : INTEGER4) : INTEGER4;
|
| 187 |
*)
|
209 |
*)
|
|
|
210 |
|
|
|
211 |
CONST idamax = LibDBlasL1F77.idamax_;
|
|
|
212 |
|
| 188 |
(*----------------------------------------------------------------*)
|
213 |
(*----------------------------------------------------------------*)
|
| 189 |
(* Finds the index of element having max. absolute value. *)
|
214 |
(* Finds the index of element having max. absolute value. *)
|
| 190 |
(*----------------------------------------------------------------*)
|
215 |
(*----------------------------------------------------------------*)
|
| 191 |
|
|
|
| 192 |
CONST idamin = LibDBlasL1F77.idamin_;
|
|
|
| 193 |
|
216 |
|
| 194 |
(*
|
217 |
(*
|
| 195 |
* PROCEDURE idamin(VAR n : INTEGER4;
|
218 |
* PROCEDURE idamin(VAR n : INTEGER4;
|
| 196 |
* VAR dx : (* ARRAY OF *) DOUBLEPRECISION;
|
219 |
* VAR dx : (* ARRAY OF *) DOUBLEPRECISION;
|
| 197 |
* VAR incx : INTEGER4) : INTEGER4;
|
220 |
* VAR incx : INTEGER4) : INTEGER4;
|
| 198 |
*)
|
221 |
*)
|
|
|
222 |
|
|
|
223 |
CONST idamin = LibDBlasL1F77.idamin_;
|
|
|
224 |
|
| 199 |
(*----------------------------------------------------------------*)
|
225 |
(*----------------------------------------------------------------*)
|
| 200 |
(* Finds the index of element having min. absolute value. *)
|
226 |
(* Finds the index of element having min. absolute value. *)
|
| 201 |
(*----------------------------------------------------------------*)
|
227 |
(*----------------------------------------------------------------*)
|
| 202 |
|
|
|
| 203 |
CONST dasum = LibDBlasL1F77.dasum_;
|
|
|
| 204 |
|
228 |
|
| 205 |
(*
|
229 |
(*
|
| 206 |
* PROCEDURE dasum(VAR dim : INTEGER4;
|
230 |
* PROCEDURE dasum(VAR dim : INTEGER4;
|
| 207 |
* VAR X : (* ARRAY OF *) DOUBLEPRECISION;
|
231 |
* VAR X : (* ARRAY OF *) DOUBLEPRECISION;
|
| 208 |
* VAR Inc : INTEGER4) : DOUBLEPRECISION;
|
232 |
* VAR Inc : INTEGER4) : DOUBLEPRECISION;
|
| 209 |
*)
|
233 |
*)
|
|
|
234 |
|
|
|
235 |
CONST dasum = LibDBlasL1F77.dasum_;
|
|
|
236 |
|
| 210 |
(*----------------------------------------------------------------*)
|
237 |
(*----------------------------------------------------------------*)
|
| 211 |
(* Berechnet die Summe der Absolutwerte der im Feld X gespeich- *)
|
238 |
(* Berechnet die Summe der Absolutwerte der im Feld X gespeich- *)
|
| 212 |
(* erten Zahlen. *)
|
239 |
(* erten Zahlen. *)
|
| 213 |
(*----------------------------------------------------------------*)
|
240 |
(*----------------------------------------------------------------*)
|
| 214 |
|
241 |
|
|
|
242 |
(*
|
|
|
243 |
* PROCEDURE dgemv( Trans : CHAR1;
|
|
|
244 |
* M,N : INTEGER4;
|
|
|
245 |
* Alpha : DOUBLEPRECISION;
|
|
|
246 |
* VAR A : ARRAY OF ARRAY OF DOUBLEPRECISION;
|
|
|
247 |
* LDA : INTEGER4;
|
|
|
248 |
* VAR X : ARRAY OF DOUBLEPRECISION;
|
|
|
249 |
* IncX : INTEGER4;
|
|
|
250 |
* Beta : DOUBLEPRECISION;
|
|
|
251 |
* VAR Y : ARRAY OF DOUBLEPRECISION;
|
|
|
252 |
* IncY : INTEGER4);
|
|
|
253 |
*)
|
|
|
254 |
|
| 215 |
CONST sgemm = LibDBlasL3F77.dgemm;
|
255 |
CONST dgemv = LibDBlasL2F77.dgemv;
|
|
|
256 |
|
|
|
257 |
(*----------------------------------------------------------------*)
|
|
|
258 |
(* Aufruf der Fortran Version von BLAS2 subroutine dgemv *)
|
|
|
259 |
(* ueber Schnittstelle die fuer eine korrekte Umsetzung der *)
|
|
|
260 |
(* Parameter sorgt. *)
|
|
|
261 |
(* Durch die Schnittstelle wird auch erreicht dass die Routine *)
|
|
|
262 |
(* "normal" aufgerufen werden kann wie von Modula-2 her gewoehnt *)
|
|
|
263 |
(*----------------------------------------------------------------*)
|
| 216 |
|
264 |
|
| 217 |
(*
|
265 |
(*
|
| 218 |
* PROCEDURE sgemm( TA : CHAR1;
|
266 |
* PROCEDURE sgemm( TA : CHAR1;
|
| 219 |
* TB : CHAR1;
|
267 |
* TB : CHAR1;
|
| 220 |
* M,N,K : INTEGER4;
|
268 |
* M,N,K : INTEGER4;
|
|
... |
|
... |
| 225 |
* ldb : INTEGER4;
|
273 |
* ldb : INTEGER4;
|
| 226 |
* Beta : REAL4;
|
274 |
* Beta : REAL4;
|
| 227 |
* VAR C : ARRAY OF ARRAY OF REAL4;
|
275 |
* VAR C : ARRAY OF ARRAY OF REAL4;
|
| 228 |
* ldc : INTEGER4);
|
276 |
* ldc : INTEGER4);
|
| 229 |
*)
|
277 |
*)
|
|
|
278 |
|
|
|
279 |
CONST sgemm = LibDBlasL3F77.dgemm;
|
|
|
280 |
|
| 230 |
(*----------------------------------------------------------------*)
|
281 |
(*----------------------------------------------------------------*)
|
| 231 |
(* Aufruf der Fortran Version von BLAS3 subroutine sgemm *)
|
282 |
(* Aufruf der Fortran Version von BLAS3 subroutine sgemm *)
|
| 232 |
(* ueber Schnittstelle die fuer eine korrekte Umsetzung der *)
|
283 |
(* ueber Schnittstelle die fuer eine korrekte Umsetzung der *)
|
| 233 |
(* Parameter sorgt. *)
|
284 |
(* Parameter sorgt. *)
|
| 234 |
(* Durch die Schnittstelle wird auch erreicht dass die Routine *)
|
285 |
(* Durch die Schnittstelle wird auch erreicht dass die Routine *)
|
| 235 |
(* "normal" aufgerufen werden kann wie von Modula-2 her gewoehnt *)
|
286 |
(* "normal" aufgerufen werden kann wie von Modula-2 her gewoehnt *)
|
| 236 |
(*----------------------------------------------------------------*)
|
287 |
(*----------------------------------------------------------------*)
|
| 237 |
|
|
|
| 238 |
<* IF (__OPENMP__) THEN *>
|
|
|
| 239 |
CONST dgemm = LibDBlasL3F77.dgemmOMP;
|
|
|
| 240 |
<* ELSE *>
|
|
|
| 241 |
CONST dgemm = LibDBlasL3F77.dgemm;
|
|
|
| 242 |
<* END *>
|
|
|
| 243 |
|
288 |
|
| 244 |
(*
|
289 |
(*
|
| 245 |
* PROCEDURE dgemm( TA : CHAR1;
|
290 |
* PROCEDURE dgemm( TA : CHAR1;
|
| 246 |
* TB : CHAR1;
|
291 |
* TB : CHAR1;
|
| 247 |
* M,N,K : INTEGER4;
|
292 |
* M,N,K : INTEGER4;
|
|
... |
|
... |
| 252 |
* ldb : INTEGER4;
|
297 |
* ldb : INTEGER4;
|
| 253 |
* Beta : DOUBLEPRECISION;
|
298 |
* Beta : DOUBLEPRECISION;
|
| 254 |
* VAR C : ARRAY OF ARRAY OF DOUBLEPRECISION;
|
299 |
* VAR C : ARRAY OF ARRAY OF DOUBLEPRECISION;
|
| 255 |
* ldc : INTEGER4);
|
300 |
* ldc : INTEGER4);
|
| 256 |
*)
|
301 |
*)
|
|
|
302 |
|
|
|
303 |
<* IF (__OPENMP__) THEN *>
|
|
|
304 |
CONST dgemm = LibDBlasL3F77.dgemmOMP;
|
|
|
305 |
<* ELSE *>
|
|
|
306 |
CONST dgemm = LibDBlasL3F77.dgemm;
|
|
|
307 |
<* END *>
|
|
|
308 |
|
| 257 |
(*----------------------------------------------------------------*)
|
309 |
(*----------------------------------------------------------------*)
|
| 258 |
(* Aufruf der Fortran Version von BLAS3 subroutine dgemm *)
|
310 |
(* Aufruf der Fortran Version von BLAS3 subroutine dgemm *)
|
| 259 |
(* ueber Schnittstelle die fuer eine korrekte Umsetzung der *)
|
311 |
(* ueber Schnittstelle die fuer eine korrekte Umsetzung der *)
|
| 260 |
(* Parameter sorgt. *)
|
312 |
(* Parameter sorgt. *)
|
| 261 |
(* Durch die Schnittstelle wird auch erreicht dass die Routine *)
|
313 |
(* Durch die Schnittstelle wird auch erreicht dass die Routine *)
|
|
... |
|
... |
| 280 |
(* C : Resulting matrix C *)
|
332 |
(* C : Resulting matrix C *)
|
| 281 |
(* ldb : The size of the first dimention of matrix C, if *)
|
333 |
(* ldb : The size of the first dimention of matrix C, if *)
|
| 282 |
(* passing a matrix C[m,n], the value should be m *)
|
334 |
(* passing a matrix C[m,n], the value should be m *)
|
| 283 |
(*----------------------------------------------------------------*)
|
335 |
(*----------------------------------------------------------------*)
|
| 284 |
|
336 |
|
|
|
337 |
(*=========================== Complex valued procedures ====================*)
|
|
|
338 |
|
|
|
339 |
(*
|
|
|
340 |
* PROCEDURE zswap(VAR N : INTEGER4;
|
|
|
341 |
* VAR X : (* ARRAY OF *) DOUBLECOMPLEX;
|
|
|
342 |
* VAR IncX : INTEGER4;
|
|
|
343 |
* VAR Y : (* ARRAY OF *) DOUBLECOMPLEX;
|
|
|
344 |
* VAR IncY : INTEGER4);
|
|
|
345 |
*)
|
|
|
346 |
|
|
|
347 |
CONST zswap = LibDBlasL1F77.zswap_;
|
|
|
348 |
|
|
|
349 |
(*----------------------------------------------------------------*)
|
|
|
350 |
(* Swap complex vectors X and Y *)
|
|
|
351 |
(*----------------------------------------------------------------*)
|
|
|
352 |
|
|
|
353 |
(*
|
|
|
354 |
* PROCEDURE zcopy(VAR N : INTEGER4;
|
|
|
355 |
* VAR X : (* ARRAY OF *) DOUBLECOMPLEX;
|
|
|
356 |
* VAR IncX : INTEGER4;
|
|
|
357 |
* VAR Y : (* ARRAY OF *) DOUBLECOMPLEX;
|
|
|
358 |
* VAR IncY : INTEGER4);
|
|
|
359 |
*)
|
|
|
360 |
|
|
|
361 |
CONST zcopy = LibDBlasL1F77.zcopy_;
|
|
|
362 |
|
|
|
363 |
(*----------------------------------------------------------------*)
|
|
|
364 |
(* copies a vector, x, to a vector, y. *)
|
|
|
365 |
(* uses unrolled loops for increments equal to one. *)
|
|
|
366 |
(* jack dongarra, linpack, 3/11/78. *)
|
|
|
367 |
(* MRi, Modula-2 10.04.16 | 09.09.18 (complex version) *)
|
|
|
368 |
(*----------------------------------------------------------------*)
|
|
|
369 |
|
|
|
370 |
(*
|
|
|
371 |
* PROCEDURE zdotc(VAR N : INTEGER4;
|
|
|
372 |
* VAR X : (* ARRAY OF *) DOUBLECOMPLEX;
|
|
|
373 |
* VAR IncX : INTEGER4;
|
|
|
374 |
* VAR Y : (* ARRAY OF *) DOUBLECOMPLEX;
|
|
|
375 |
* VAR IncY : INTEGER4) : DOUBLECOMPLEX;
|
|
|
376 |
*)
|
|
|
377 |
|
|
|
378 |
CONST zdotc = LibDBlasL1F77.zdotc_;
|
|
|
379 |
|
|
|
380 |
(*----------------------------------------------------------------*)
|
|
|
381 |
(* Forms the dot product of two vectors. Uses unrolled loops for *)
|
|
|
382 |
(* increments equal to one. *)
|
|
|
383 |
(*----------------------------------------------------------------*)
|
|
|
384 |
|
|
|
385 |
(*
|
|
|
386 |
* PROCEDURE dznrm2(VAR N : INTEGER4;
|
|
|
387 |
* VAR X : (* ARRAY OF *) DOUBLECOMPLEX;
|
|
|
388 |
* VAR IncX : INTEGER4) : DOUBLEPRECISION;
|
|
|
389 |
*)
|
|
|
390 |
|
|
|
391 |
CONST dznrm2 = LibDBlasL1F77.dznrm2_;
|
|
|
392 |
|
|
|
393 |
(*----------------------------------------------------------------*)
|
|
|
394 |
(* dznrm2 returns the euclidean norm of a vector so that *)
|
|
|
395 |
(* dznrm2 := sqrt( X**H*X ) *)
|
|
|
396 |
(*----------------------------------------------------------------*)
|
|
|
397 |
|
|
|
398 |
(*
|
|
|
399 |
* PROCEDURE zscal(VAR n : INTEGER4;
|
|
|
400 |
* VAR da : DOUBLECOMPLEX;
|
|
|
401 |
* VAR dx : (* ARRAY OF *) DOUBLECOMPLEX;
|
|
|
402 |
* VAR IncX : INTEGER4);
|
|
|
403 |
*)
|
|
|
404 |
|
|
|
405 |
CONST zscal = LibDBlasL1F77.zscal_;
|
|
|
406 |
|
|
|
407 |
(*----------------------------------------------------------------*)
|
|
|
408 |
(* Scales a vector by a constant (UNROLLED version) *)
|
|
|
409 |
(*----------------------------------------------------------------*)
|
|
|
410 |
|
|
|
411 |
(*
|
|
|
412 |
* PROCEDURE zaxpy(VAR n : INTEGER4;
|
|
|
413 |
* VAR da : DOUBLECOMPLEX;
|
|
|
414 |
* VAR X : (* ARRAY OF *) DOUBLECOMPLEX;
|
|
|
415 |
* VAR IncX : INTEGER4;
|
|
|
416 |
* VAR Y : (* ARRAY OF *) DOUBLECOMPLEX;
|
|
|
417 |
* VAR IncY : INTEGER4);
|
|
|
418 |
*)
|
|
|
419 |
|
|
|
420 |
CONST zaxpy = LibDBlasL1F77.zaxpy_;
|
|
|
421 |
|
|
|
422 |
|
|
|
423 |
(*----------------------------------------------------------------*)
|
|
|
424 |
(* constant times a vector plus a vector *)
|
|
|
425 |
(*----------------------------------------------------------------*)
|
|
|
426 |
|
|
|
427 |
(*
|
|
|
428 |
* PROCEDURE zdrot(VAR N : INTEGER4;
|
|
|
429 |
* VAR X : (* ARRAY OF *) DOUBLECOMPLEX;
|
|
|
430 |
* VAR IncX : INTEGER4;
|
|
|
431 |
* VAR Y : (* ARRAY OF *) DOUBLECOMPLEX;
|
|
|
432 |
* VAR IncY : INTEGER4;
|
|
|
433 |
* VAR c,s : REAL4);
|
|
|
434 |
*)
|
|
|
435 |
|
|
|
436 |
CONST zdrot = LibDBlasL1F77.zdrot_;
|
|
|
437 |
|
|
|
438 |
(*----------------------------------------------------------------*)
|
|
|
439 |
(* Applies a plane rotation. *)
|
|
|
440 |
(*----------------------------------------------------------------*)
|
|
|
441 |
|
|
|
442 |
(*
|
|
|
443 |
* PROCEDURE zgemv( Trans : CHAR;
|
|
|
444 |
* M,N : INTEGER;
|
|
|
445 |
* Alpha : LONGCOMPLEX;
|
|
|
446 |
* VAR A : ARRAY OF ARRAY OF LONGCOMPLEX;
|
|
|
447 |
* lda : INTEGER;
|
|
|
448 |
* VAR X : ARRAY OF LONGCOMPLEX;
|
|
|
449 |
* IncX : INTEGER;
|
|
|
450 |
* Beta : LONGCOMPLEX;
|
|
|
451 |
* VAR Y : ARRAY OF LONGCOMPLEX;
|
|
|
452 |
* IncY : INTEGER);
|
|
|
453 |
*)
|
|
|
454 |
|
| 285 |
CONST zgemm = LibDBlasL3F77.zgemm;
|
455 |
CONST zgemv = LibDBlasL2F77.zgemv;
|
|
|
456 |
|
|
|
457 |
(*----------------------------------------------------------------*)
|
|
|
458 |
(* Aufruf der Fortran Version von BLAS2 subroutine zgemv *)
|
|
|
459 |
(* ueber Schnittstelle die fuer eine korrekte Umsetzung der *)
|
|
|
460 |
(* Parameter sorgt. *)
|
|
|
461 |
(* Durch die Schnittstelle wird auch erreicht dass die Routine *)
|
|
|
462 |
(* "normal" aufgerufen werden kann wie von Modula-2 her gewoehnt *)
|
|
|
463 |
(* *)
|
|
|
464 |
(* Performs one of the matrix-vector operations *)
|
|
|
465 |
(* *)
|
|
|
466 |
(* y = alpha*a *x + beta*y, or *)
|
|
|
467 |
(* y = alpha*a'*x + beta*y, or; *)
|
|
|
468 |
(* y = alpha*conjg(a')*x + beta*y,; *)
|
|
|
469 |
(* *)
|
|
|
470 |
(* where Alpha and Beta are scalars, X and Y are vectors *)
|
|
|
471 |
(* and A is an M by N matrix. *)
|
|
|
472 |
(*----------------------------------------------------------------*)
|
| 286 |
|
473 |
|
| 287 |
(*
|
474 |
(*
|
| 288 |
* PROCEDURE zgemm( TA : CHAR1;
|
475 |
* PROCEDURE zgemm( TA : CHAR1;
|
| 289 |
* TB : CHAR1;
|
476 |
* TB : CHAR1;
|
| 290 |
* M,N,K : INTEGER4;
|
477 |
* M,N,K : INTEGER4;
|
|
... |
|
... |
| 295 |
* ldb : INTEGER4;
|
482 |
* ldb : INTEGER4;
|
| 296 |
* Beta : DOUBLECOMPLEX;
|
483 |
* Beta : DOUBLECOMPLEX;
|
| 297 |
* VAR C : ARRAY OF ARRAY OF DOUBLECOMPLEX;
|
484 |
* VAR C : ARRAY OF ARRAY OF DOUBLECOMPLEX;
|
| 298 |
* ldc : INTEGER4);
|
485 |
* ldc : INTEGER4);
|
| 299 |
*)
|
486 |
*)
|
|
|
487 |
|
|
|
488 |
CONST zgemm = LibDBlasL3F77.zgemm;
|
|
|
489 |
|
| 300 |
(*----------------------------------------------------------------*)
|
490 |
(*----------------------------------------------------------------*)
|
| 301 |
(* Aufruf der Fortran Version von BLAS3 subroutine zgemm *)
|
491 |
(* Aufruf der Fortran Version von BLAS3 subroutine zgemm *)
|
| 302 |
(* ueber Schnittstelle die fuer eine korrekte Umsetzung der *)
|
492 |
(* ueber Schnittstelle die fuer eine korrekte Umsetzung der *)
|
| 303 |
(* Parameter sorgt. *)
|
493 |
(* Parameter sorgt. *)
|
| 304 |
(* Durch die Schnittstelle wird auch erreicht dass die Routine *)
|
494 |
(* Durch die Schnittstelle wird auch erreicht dass die Routine *)
|