1 /*********************************************************************/
2 /* Copyright 2009, 2010 The University of Texas at Austin. */
3 /* All rights reserved. */
5 /* Redistribution and use in source and binary forms, with or */
6 /* without modification, are permitted provided that the following */
7 /* conditions are met: */
9 /* 1. Redistributions of source code must retain the above */
10 /* copyright notice, this list of conditions and the following */
13 /* 2. Redistributions in binary form must reproduce the above */
14 /* copyright notice, this list of conditions and the following */
15 /* disclaimer in the documentation and/or other materials */
16 /* provided with the distribution. */
18 /* THIS SOFTWARE IS PROVIDED BY THE UNIVERSITY OF TEXAS AT */
19 /* AUSTIN ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, */
20 /* INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF */
21 /* MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE */
22 /* DISCLAIMED. IN NO EVENT SHALL THE UNIVERSITY OF TEXAS AT */
23 /* AUSTIN OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, */
24 /* INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES */
25 /* (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE */
26 /* GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR */
27 /* BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF */
28 /* LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT */
29 /* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT */
30 /* OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE */
31 /* POSSIBILITY OF SUCH DAMAGE. */
33 /* The views and conclusions contained in the software and */
34 /* documentation are those of the authors and should not be */
35 /* interpreted as representing official policies, either expressed */
36 /* or implied, of The University of Texas at Austin. */
37 /*********************************************************************/
42 #ifdef FUNCTION_PROFILE
43 #include "functable.h"
47 #define ERROR_NAME "QTPMV "
49 #define ERROR_NAME "DTPMV "
51 #define ERROR_NAME "STPMV "
54 static int (*tpmv[])(BLASLONG, FLOAT *, FLOAT *, BLASLONG, void *) = {
56 qtpmv_NUU, qtpmv_NUN, qtpmv_NLU, qtpmv_NLN,
57 qtpmv_TUU, qtpmv_TUN, qtpmv_TLU, qtpmv_TLN,
59 dtpmv_NUU, dtpmv_NUN, dtpmv_NLU, dtpmv_NLN,
60 dtpmv_TUU, dtpmv_TUN, dtpmv_TLU, dtpmv_TLN,
62 stpmv_NUU, stpmv_NUN, stpmv_NLU, stpmv_NLN,
63 stpmv_TUU, stpmv_TUN, stpmv_TLU, stpmv_TLN,
68 static int (*tpmv_thread[])(BLASLONG, FLOAT *, FLOAT *, BLASLONG, FLOAT *, int) = {
70 qtpmv_thread_NUU, qtpmv_thread_NUN, qtpmv_thread_NLU, qtpmv_thread_NLN,
71 qtpmv_thread_TUU, qtpmv_thread_TUN, qtpmv_thread_TLU, qtpmv_thread_TLN,
73 dtpmv_thread_NUU, dtpmv_thread_NUN, dtpmv_thread_NLU, dtpmv_thread_NLN,
74 dtpmv_thread_TUU, dtpmv_thread_TUN, dtpmv_thread_TLU, dtpmv_thread_TLN,
76 stpmv_thread_NUU, stpmv_thread_NUN, stpmv_thread_NLU, stpmv_thread_NLN,
77 stpmv_thread_TUU, stpmv_thread_TUN, stpmv_thread_TLU, stpmv_thread_TLN,
84 void NAME(char *UPLO, char *TRANS, char *DIAG,
85 blasint *N, FLOAT *a, FLOAT *x, blasint *INCX){
87 char uplo_arg = *UPLO;
88 char trans_arg = *TRANS;
89 char diag_arg = *DIAG;
113 if (trans_arg == 'N') trans = 0;
114 if (trans_arg == 'T') trans = 1;
115 if (trans_arg == 'R') trans = 0;
116 if (trans_arg == 'C') trans = 1;
118 if (diag_arg == 'U') unit = 0;
119 if (diag_arg == 'N') unit = 1;
121 if (uplo_arg == 'U') uplo = 0;
122 if (uplo_arg == 'L') uplo = 1;
126 if (incx == 0) info = 7;
128 if (unit < 0) info = 3;
129 if (trans < 0) info = 2;
130 if (uplo < 0) info = 1;
133 BLASFUNC(xerbla)(ERROR_NAME, &info, sizeof(ERROR_NAME));
139 void CNAME(enum CBLAS_ORDER order, enum CBLAS_UPLO Uplo,
140 enum CBLAS_TRANSPOSE TransA, enum CBLAS_DIAG Diag,
141 blasint n, FLOAT *a, FLOAT *x, blasint incx) {
143 void CNAME(enum CBLAS_ORDER order, enum CBLAS_UPLO Uplo,
144 enum CBLAS_TRANSPOSE TransA, enum CBLAS_DIAG Diag,
145 blasint n, void *va, void *vx, blasint incx) {
146 FLOAT *a = (FLOAT*) va;
147 FLOAT *x = (FLOAT*) vx;
150 int trans, uplo, unit;
164 if (order == CblasColMajor) {
165 if (Uplo == CblasUpper) uplo = 0;
166 if (Uplo == CblasLower) uplo = 1;
168 if (TransA == CblasNoTrans) trans = 0;
169 if (TransA == CblasTrans) trans = 1;
170 if (TransA == CblasConjNoTrans) trans = 0;
171 if (TransA == CblasConjTrans) trans = 1;
173 if (Diag == CblasUnit) unit = 0;
174 if (Diag == CblasNonUnit) unit = 1;
178 if (incx == 0) info = 7;
180 if (unit < 0) info = 3;
181 if (trans < 0) info = 2;
182 if (uplo < 0) info = 1;
185 if (order == CblasRowMajor) {
186 if (Uplo == CblasUpper) uplo = 1;
187 if (Uplo == CblasLower) uplo = 0;
189 if (TransA == CblasNoTrans) trans = 1;
190 if (TransA == CblasTrans) trans = 0;
191 if (TransA == CblasConjNoTrans) trans = 1;
192 if (TransA == CblasConjTrans) trans = 0;
194 if (Diag == CblasUnit) unit = 0;
195 if (Diag == CblasNonUnit) unit = 1;
199 if (incx == 0) info = 7;
201 if (unit < 0) info = 3;
202 if (trans < 0) info = 2;
203 if (uplo < 0) info = 1;
207 BLASFUNC(xerbla)(ERROR_NAME, &info, sizeof(ERROR_NAME));
217 FUNCTION_PROFILE_START();
219 if (incx < 0 ) x -= (n - 1) * incx;
221 buffer = (FLOAT *)blas_memory_alloc(1);
224 nthreads = num_cpu_avail(2);
229 (tpmv[(trans<<2) | (uplo<<1) | unit])(n, a, x, incx, buffer);
234 (tpmv_thread[(trans<<2) | (uplo<<1) | unit])(n, a, x, incx, buffer, nthreads);
239 blas_memory_free(buffer);
241 FUNCTION_PROFILE_END(1, n * n / 2 + n, n * n);