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 "XTBMV "
49 #define ERROR_NAME "ZTBMV "
51 #define ERROR_NAME "CTBMV "
54 static int (*tbmv[])(BLASLONG, BLASLONG, FLOAT *, BLASLONG, FLOAT *, BLASLONG, void *) = {
56 xtbmv_NUU, xtbmv_NUN, xtbmv_NLU, xtbmv_NLN,
57 xtbmv_TUU, xtbmv_TUN, xtbmv_TLU, xtbmv_TLN,
58 xtbmv_RUU, xtbmv_RUN, xtbmv_RLU, xtbmv_RLN,
59 xtbmv_CUU, xtbmv_CUN, xtbmv_CLU, xtbmv_CLN,
61 ztbmv_NUU, ztbmv_NUN, ztbmv_NLU, ztbmv_NLN,
62 ztbmv_TUU, ztbmv_TUN, ztbmv_TLU, ztbmv_TLN,
63 ztbmv_RUU, ztbmv_RUN, ztbmv_RLU, ztbmv_RLN,
64 ztbmv_CUU, ztbmv_CUN, ztbmv_CLU, ztbmv_CLN,
66 ctbmv_NUU, ctbmv_NUN, ctbmv_NLU, ctbmv_NLN,
67 ctbmv_TUU, ctbmv_TUN, ctbmv_TLU, ctbmv_TLN,
68 ctbmv_RUU, ctbmv_RUN, ctbmv_RLU, ctbmv_RLN,
69 ctbmv_CUU, ctbmv_CUN, ctbmv_CLU, ctbmv_CLN,
74 static int (*tbmv_thread[])(BLASLONG, BLASLONG, FLOAT *, BLASLONG, FLOAT *, BLASLONG, FLOAT *, int) = {
76 xtbmv_thread_NUU, xtbmv_thread_NUN, xtbmv_thread_NLU, xtbmv_thread_NLN,
77 xtbmv_thread_TUU, xtbmv_thread_TUN, xtbmv_thread_TLU, xtbmv_thread_TLN,
78 xtbmv_thread_RUU, xtbmv_thread_RUN, xtbmv_thread_RLU, xtbmv_thread_RLN,
79 xtbmv_thread_CUU, xtbmv_thread_CUN, xtbmv_thread_CLU, xtbmv_thread_CLN,
81 ztbmv_thread_NUU, ztbmv_thread_NUN, ztbmv_thread_NLU, ztbmv_thread_NLN,
82 ztbmv_thread_TUU, ztbmv_thread_TUN, ztbmv_thread_TLU, ztbmv_thread_TLN,
83 ztbmv_thread_RUU, ztbmv_thread_RUN, ztbmv_thread_RLU, ztbmv_thread_RLN,
84 ztbmv_thread_CUU, ztbmv_thread_CUN, ztbmv_thread_CLU, ztbmv_thread_CLN,
86 ctbmv_thread_NUU, ctbmv_thread_NUN, ctbmv_thread_NLU, ctbmv_thread_NLN,
87 ctbmv_thread_TUU, ctbmv_thread_TUN, ctbmv_thread_TLU, ctbmv_thread_TLN,
88 ctbmv_thread_RUU, ctbmv_thread_RUN, ctbmv_thread_RLU, ctbmv_thread_RLN,
89 ctbmv_thread_CUU, ctbmv_thread_CUN, ctbmv_thread_CLU, ctbmv_thread_CLN,
96 void NAME(char *UPLO, char *TRANS, char *DIAG,
97 blasint *N, blasint *K,
98 FLOAT *a, blasint *LDA, FLOAT *x, blasint *INCX){
100 char uplo_arg = *UPLO;
101 char trans_arg = *TRANS;
102 char diag_arg = *DIAG;
107 blasint incx = *INCX;
128 if (trans_arg == 'N') trans = 0;
129 if (trans_arg == 'T') trans = 1;
130 if (trans_arg == 'R') trans = 2;
131 if (trans_arg == 'C') trans = 3;
133 if (diag_arg == 'U') unit = 0;
134 if (diag_arg == 'N') unit = 1;
136 if (uplo_arg == 'U') uplo = 0;
137 if (uplo_arg == 'L') uplo = 1;
141 if (incx == 0) info = 9;
142 if (lda < k + 1) info = 7;
145 if (unit < 0) info = 3;
146 if (trans < 0) info = 2;
147 if (uplo < 0) info = 1;
150 BLASFUNC(xerbla)(ERROR_NAME, &info, sizeof(ERROR_NAME));
156 void CNAME(enum CBLAS_ORDER order, enum CBLAS_UPLO Uplo,
157 enum CBLAS_TRANSPOSE TransA, enum CBLAS_DIAG Diag,
158 blasint n, blasint k, void *va, blasint lda, void *vx, blasint incx) {
160 FLOAT *a = (FLOAT*) va;
161 FLOAT *x = (FLOAT*) vx;
163 int trans, uplo, unit;
177 if (order == CblasColMajor) {
178 if (Uplo == CblasUpper) uplo = 0;
179 if (Uplo == CblasLower) uplo = 1;
181 if (TransA == CblasNoTrans) trans = 0;
182 if (TransA == CblasTrans) trans = 1;
183 if (TransA == CblasConjNoTrans) trans = 2;
184 if (TransA == CblasConjTrans) trans = 3;
186 if (Diag == CblasUnit) unit = 0;
187 if (Diag == CblasNonUnit) unit = 1;
191 if (incx == 0) info = 9;
192 if (lda < k + 1) info = 7;
195 if (unit < 0) info = 3;
196 if (trans < 0) info = 2;
197 if (uplo < 0) info = 1;
200 if (order == CblasRowMajor) {
201 if (Uplo == CblasUpper) uplo = 1;
202 if (Uplo == CblasLower) uplo = 0;
204 if (TransA == CblasNoTrans) trans = 1;
205 if (TransA == CblasTrans) trans = 0;
206 if (TransA == CblasConjNoTrans) trans = 3;
207 if (TransA == CblasConjTrans) trans = 2;
209 if (Diag == CblasUnit) unit = 0;
210 if (Diag == CblasNonUnit) unit = 1;
214 if (incx == 0) info = 9;
215 if (lda < k + 1) info = 7;
218 if (unit < 0) info = 3;
219 if (trans < 0) info = 2;
220 if (uplo < 0) info = 1;
224 BLASFUNC(xerbla)(ERROR_NAME, &info, sizeof(ERROR_NAME));
234 FUNCTION_PROFILE_START();
236 if (incx < 0 ) x -= (n - 1) * incx * 2;
238 buffer = (FLOAT *)blas_memory_alloc(1);
241 nthreads = num_cpu_avail(2);
246 (tbmv[(trans<<2) | (uplo<<1) | unit])(n, k, a, lda, x, incx, buffer);
251 (tbmv_thread[(trans<<2) | (uplo<<1) | unit])(n, k, a, lda, x, incx, buffer, nthreads);
256 blas_memory_free(buffer);
258 FUNCTION_PROFILE_END(4, n * k / 2 + n, n * k);