/root/doris/contrib/openblas/interface/swap.c
Line | Count | Source |
1 | | /*********************************************************************/ |
2 | | /* Copyright 2009, 2010 The University of Texas at Austin. */ |
3 | | /* All rights reserved. */ |
4 | | /* */ |
5 | | /* Redistribution and use in source and binary forms, with or */ |
6 | | /* without modification, are permitted provided that the following */ |
7 | | /* conditions are met: */ |
8 | | /* */ |
9 | | /* 1. Redistributions of source code must retain the above */ |
10 | | /* copyright notice, this list of conditions and the following */ |
11 | | /* disclaimer. */ |
12 | | /* */ |
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. */ |
17 | | /* */ |
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. */ |
32 | | /* */ |
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 | | /*********************************************************************/ |
38 | | |
39 | | #include <stdio.h> |
40 | | #include "common.h" |
41 | | #ifdef FUNCTION_PROFILE |
42 | | #include "functable.h" |
43 | | #endif |
44 | | |
45 | | #if defined(THUNDERX2T99) || defined(VULCAN) || defined(ARMV8) || defined(THUNDERX3T110) |
46 | | // Multithreaded swap gives performance benefits in ThunderX2T99 |
47 | | #else |
48 | | // Disable multi-threading as it does not show any performance |
49 | | // benefits. Keep the multi-threading code for the record. |
50 | | #undef SMP |
51 | | #endif |
52 | | |
53 | | #ifndef CBLAS |
54 | | |
55 | 0 | void NAME(blasint *N, FLOAT *x, blasint *INCX, FLOAT *y, blasint *INCY){ |
56 | |
|
57 | 0 | blasint n = *N; |
58 | 0 | blasint incx = *INCX; |
59 | 0 | blasint incy = *INCY; |
60 | |
|
61 | | #else |
62 | | |
63 | | void CNAME(blasint n, FLOAT *x, blasint incx, FLOAT *y, blasint incy){ |
64 | | |
65 | | #endif |
66 | |
|
67 | | #ifdef SMP |
68 | | int mode, nthreads; |
69 | | FLOAT dummyalpha[2] = {ZERO, ZERO}; |
70 | | #endif |
71 | |
|
72 | 0 | #ifndef CBLAS |
73 | 0 | PRINT_DEBUG_NAME; |
74 | | #else |
75 | | PRINT_DEBUG_CNAME; |
76 | | #endif |
77 | |
|
78 | 0 | if (n <= 0) return; |
79 | | |
80 | 0 | IDEBUG_START; |
81 | |
|
82 | 0 | FUNCTION_PROFILE_START(); |
83 | |
|
84 | 0 | if (incx < 0) x -= (n - 1) * incx; |
85 | 0 | if (incy < 0) y -= (n - 1) * incy; |
86 | |
|
87 | | #ifdef SMP |
88 | | //disable multi-thread when incx==0 or incy==0 |
89 | | //In that case, the threads would be dependent. |
90 | | if (incx == 0 || incy == 0 || n < 2097152 * GEMM_MULTITHREAD_THRESHOLD / sizeof(FLOAT)) |
91 | | nthreads = 1; |
92 | | else |
93 | | nthreads = num_cpu_avail(1); |
94 | | |
95 | | if (nthreads == 1) { |
96 | | #endif |
97 | |
|
98 | 0 | SWAP_K(n, 0, 0, ZERO, x, incx, y, incy, NULL, 0); |
99 | |
|
100 | | #ifdef SMP |
101 | | } else { |
102 | | |
103 | | #ifdef XDOUBLE |
104 | | mode = BLAS_XDOUBLE | BLAS_REAL; |
105 | | #elif defined(DOUBLE) |
106 | | mode = BLAS_DOUBLE | BLAS_REAL; |
107 | | #else |
108 | | mode = BLAS_SINGLE | BLAS_REAL; |
109 | | #endif |
110 | | |
111 | | blas_level1_thread(mode, n, 0, 0, dummyalpha, |
112 | | x, incx, y, incy, NULL, 0, (void *)SWAP_K, nthreads); |
113 | | } |
114 | | |
115 | | #endif |
116 | |
|
117 | 0 | FUNCTION_PROFILE_END(1, 2 * n, 0); |
118 | |
|
119 | 0 | IDEBUG_END; |
120 | |
|
121 | 0 | return; |
122 | |
|
123 | 0 | } Unexecuted instantiation: sswap_ Unexecuted instantiation: dswap_ |