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 /*********************************************************************/
41 #ifdef FUNCTION_PROFILE
42 #include "functable.h"
52 #define ERROR_NAME "QGER "
54 #define ERROR_NAME "DGER "
56 #define ERROR_NAME "SGER "
62 #define GER_THREAD qger_thread
64 #define GER_THREAD dger_thread
66 #define GER_THREAD sger_thread
72 void NAME(blasint *M, blasint *N, FLOAT *Alpha,
73 FLOAT *x, blasint *INCX,
74 FLOAT *y, blasint *INCY,
75 FLOAT *a, blasint *LDA){
94 if (lda < MAX(1,m)) info = 9;
95 if (incy == 0) info = 7;
96 if (incx == 0) info = 5;
101 BLASFUNC(xerbla)(ERROR_NAME, &info, sizeof(ERROR_NAME));
107 void CNAME(enum CBLAS_ORDER order,
108 blasint m, blasint n,
110 FLOAT *x, blasint incx,
111 FLOAT *y, blasint incy,
112 FLOAT *a, blasint lda) {
124 if (order == CblasColMajor) {
127 if (lda < MAX(1,m)) info = 9;
128 if (incy == 0) info = 7;
129 if (incx == 0) info = 5;
134 if (order == CblasRowMajor) {
149 if (lda < MAX(1,m)) info = 9;
150 if (incy == 0) info = 7;
151 if (incx == 0) info = 5;
157 BLASFUNC(xerbla)(ERROR_NAME, &info, sizeof(ERROR_NAME));
163 /* Quick return if possible. */
164 if (m == 0 || n == 0) return;
165 if (alpha == 0.) return;
169 FUNCTION_PROFILE_START();
171 if (incy < 0) y -= (n - 1) * incy;
172 if (incx < 0) x -= (m - 1) * incx;
174 STACK_ALLOC(m, FLOAT, buffer);
177 // Threshold chosen so that speed-up is > 1 on a Xeon E5-2630
178 if(1L * m * n > 2048L * GEMM_MULTITHREAD_THRESHOLD)
179 nthreads = num_cpu_avail(2);
186 GER(m, n, 0, alpha, x, incx, y, incy, a, lda, buffer);
191 GER_THREAD(m, n, alpha, x, incx, y, incy, a, lda, buffer, nthreads);
197 FUNCTION_PROFILE_END(1, m * n + m + n, 2 * m * n);