| /dports/math/libxsmm/libxsmm-1.16.3/samples/packed/getrf/ |
| H A D | getrf.c | 64 Atemp[i+j*m] = A[i*VLEN+j*lda*VLEN]; in dcopy_to_temp()
83 Atemp[i+j*m] = A[j*VLEN+i*lda*VLEN]; in dcopy_to_temp()
107 Atemp[i+j*m] = A[i*VLEN+j*lda*VLEN]; in scopy_to_temp()
117 Atemp[i+j*m] = A[j*VLEN+i*lda*VLEN]; in scopy_to_temp()
148 A[j*VLEN+i*lda*VLEN] = Atemp[i+j*m]; in dcopy_from_temp()
179 A[j*VLEN+i*lda*VLEN] = Atemp[i+j*m]; in scopy_from_temp()
308 address = (k1-1)*col*ld*VLEN + (j-1)*ld*VLEN + (i-1)*VLEN + (k-1); in dfill_matrix()
332 ia = h*ld*n*VLEN; in dfill_identity()
359 address = (k1-1)*col*ld*VLEN + (j-1)*ld*VLEN + (i-1)*VLEN + (k-1); in sfill_matrix()
394 address= (k1-1)*col*lda*VLEN + (j-1)*lda*VLEN + (i-1)*VLEN + (k-1); in residual_s()
[all …]
|
| /dports/math/libpgmath/flang-d07daf3/runtime/libpgmath/lib/common/sincos/ |
| H A D | gsincos.c | 44 #define VFLOAT CONCAT4(vr,PREC,VLEN,_t)
45 #define VINT CONCAT4(vi,PREC,VLEN,_t)
56 #define GENERICNAME CONCAT4(__g,PREC,_sincos_,VLEN)
57 #define GENERICNAMEMASK CONCAT5(__g,PREC,_sincos_,VLEN,m)
68 #define VEC_LOAD(_a) (VFLOAT)CONCAT4(_mm,_VLSIZE(PREC,VLEN),_load_p,PREC)((FLOAT *)_a)
88 FLOAT ts[VLEN] __attribute__((aligned((VLEN*sizeof(FLOAT))))); in GENERICNAME()
89 FLOAT tc[VLEN] __attribute__((aligned((VLEN*sizeof(FLOAT))))); in GENERICNAME()
91 for (i = 0 ; i < VLEN; i++) in GENERICNAME()
101 FLOAT ts[VLEN] __attribute__((aligned((VLEN*sizeof(FLOAT))))); in GENERICNAMEMASK()
102 FLOAT tc[VLEN] __attribute__((aligned((VLEN*sizeof(FLOAT))))); in GENERICNAMEMASK()
[all …]
|
| H A D | gsincos.cpp | 44 #define VFLOAT CONCAT4(vr,PREC,VLEN,_t)
45 #define VINT CONCAT4(vi,PREC,VLEN,_t)
56 #define GENERICNAME CONCAT4(__g,PREC,_sincos_,VLEN)
57 #define GENERICNAMEMASK CONCAT5(__g,PREC,_sincos_,VLEN,m)
68 #define VEC_LOAD(_a) (VFLOAT)CONCAT4(_mm,_VLSIZE(PREC,VLEN),_load_p,PREC)((FLOAT *)_a)
88 FLOAT ts[VLEN] __attribute__((__aligned__(16))); in GENERICNAME()
89 FLOAT tc[VLEN] __attribute__((__aligned__(16))); in GENERICNAME()
91 for (i = 0 ; i < VLEN; i++) in GENERICNAME()
101 FLOAT ts[VLEN] __attribute__((__aligned__(16))); in GENERICNAMEMASK()
102 FLOAT tc[VLEN] __attribute__((__aligned__(16))); in GENERICNAMEMASK()
[all …]
|
| /dports/math/libxsmm/libxsmm-1.16.3/samples/deeplearning/gxm/src/ |
| H A D | ConcatXSMM.cpp | 17 #define VLEN 16 macro
25 int nBOfm = gp->nOutput/VLEN; in forwardPropagate()
36 float (* __restrict output)[nBOfm][ofh][ofw][VLEN] = (float (*)[*][*][*][VLEN])outp; in forwardPropagate()
44 int nBIfm = gp->nInput[b]/VLEN; in forwardPropagate()
47 float (* __restrict input )[nBIfm][ifh][ifw][VLEN] = (float (*)[*][*][*][VLEN])inp; in forwardPropagate()
54 for(int v=0; v < VLEN; v++) { in forwardPropagate()
72 int nBOfm = gp->nOutput/VLEN; in backPropagate()
83 float (* __restrict del_output)[nBOfm][ofh][ofw][VLEN] = (float (*)[*][*][*][VLEN])deloutp; in backPropagate()
91 int nBIfm = gp->nInput[b]/VLEN; in backPropagate()
94 float (* __restrict del_input)[nBIfm][ifh][ifw][VLEN] = (float (*)[*][*][*][VLEN])delinp; in backPropagate()
[all …]
|
| /dports/math/libxsmm/libxsmm-1.16.3/samples/packed/trsm/ |
| H A D | trsm.c | 48 unsigned int VLEN ) in dcopy_to_temp() argument
64 Atemp[i+j*m] = A[i*VLEN+j*lda*VLEN]; in dcopy_to_temp()
83 Atemp[i+j*m] = A[j*VLEN+i*lda*VLEN]; in dcopy_to_temp()
107 Atemp[i+j*m] = A[i*VLEN+j*lda*VLEN]; in scopy_to_temp()
117 Atemp[i+j*m] = A[j*VLEN+i*lda*VLEN]; in scopy_to_temp()
148 A[j*VLEN+i*lda*VLEN] = Atemp[i+j*m]; in dcopy_from_temp()
179 A[j*VLEN+i*lda*VLEN] = Atemp[i+j*m]; in scopy_from_temp()
209 offsetb = (*ldb)*(*n)*(*VLEN); in compact_dtrsm_()
213 offsetb = (*ldb)*(*m)*(*VLEN); in compact_dtrsm_()
215 offseta = (*lda)*asize*(*VLEN); in compact_dtrsm_()
[all …]
|
| /dports/math/libxsmm/libxsmm-1.16.3/samples/packed/trmm/ |
| H A D | trmm.c | 48 unsigned int VLEN ) in dcopy_to_temp() argument
64 Atemp[i+j*m] = A[i*VLEN+j*lda*VLEN]; in dcopy_to_temp()
83 Atemp[i+j*m] = A[j*VLEN+i*lda*VLEN]; in dcopy_to_temp()
107 Atemp[i+j*m] = A[i*VLEN+j*lda*VLEN]; in scopy_to_temp()
117 Atemp[i+j*m] = A[j*VLEN+i*lda*VLEN]; in scopy_to_temp()
148 A[j*VLEN+i*lda*VLEN] = Atemp[i+j*m]; in dcopy_from_temp()
179 A[j*VLEN+i*lda*VLEN] = Atemp[i+j*m]; in scopy_from_temp()
209 offsetb = (*ldb)*(*n)*(*VLEN); in compact_dtrmm_()
213 offsetb = (*ldb)*(*m)*(*VLEN); in compact_dtrmm_()
215 offseta = (*lda)*asize*(*VLEN); in compact_dtrmm_()
[all …]
|
| /dports/math/libxsmm/libxsmm-1.16.3/samples/packed/gemm/ |
| H A D | gemm.c | 61 unsigned int VLEN ) in dcopy_to_temp() argument
77 Atemp[i+j*m] = A[i*VLEN+j*lda*VLEN]; in dcopy_to_temp()
96 Atemp[i+j*m] = A[j*VLEN+i*lda*VLEN]; in dcopy_to_temp()
104 unsigned int VLEN ) in scopy_to_temp() argument
120 Atemp[i+j*m] = A[i*VLEN+j*lda*VLEN]; in scopy_to_temp()
130 Atemp[i+j*m] = A[j*VLEN+i*lda*VLEN]; in scopy_to_temp()
161 A[j*VLEN+i*lda*VLEN] = Atemp[i+j*m]; in dcopy_from_temp()
192 A[j*VLEN+i*lda*VLEN] = Atemp[i+j*m]; in scopy_from_temp()
223 for ( j = 0; j < *VLEN; j++ ) in compact_dgemm_()
265 for ( j = 0; j < *VLEN; j++ ) in compact_sgemm_()
[all …]
|
| /dports/lang/gcc9/gcc-9.4.0/gcc/testsuite/gcc.c-torture/execute/ |
| H A D | 921202-1.c | 5 #define VLEN 2055 macro
7 #define VLEN ((STACK_SIZE/16) - 1) macro
11 long dx[VLEN+1]; in main()
12 long dy[VLEN+1]; in main()
13 long s1[VLEN]; in main()
20 size = VLEN; in main()
|
| /dports/lang/gcc10/gcc-10.3.0/gcc/testsuite/gcc.c-torture/execute/ |
| H A D | 921202-1.c | 5 #define VLEN 2055 macro
7 #define VLEN ((STACK_SIZE/16) - 1) macro
11 long dx[VLEN+1]; in main()
12 long dy[VLEN+1]; in main()
13 long s1[VLEN]; in main()
20 size = VLEN; in main()
|
| /dports/devel/arm-none-eabi-gcc492/gcc-4.9.2/gcc/testsuite/gcc.c-torture/execute/ |
| H A D | 921202-1.c | 2 #define VLEN 2055 macro
4 #define VLEN ((STACK_SIZE/16) - 1) macro
8 long dx[VLEN+1]; in main()
9 long dy[VLEN+1]; in main()
10 long s1[VLEN]; in main()
17 size = VLEN; in main()
|
| /dports/devel/avr-gcc/gcc-10.2.0/gcc/testsuite/gcc.c-torture/execute/ |
| H A D | 921202-1.c | 5 #define VLEN 2055 macro
7 #define VLEN ((STACK_SIZE/16) - 1) macro
11 long dx[VLEN+1]; in main()
12 long dy[VLEN+1]; in main()
13 long s1[VLEN]; in main()
20 size = VLEN; in main()
|
| /dports/lang/gcc48/gcc-4.8.5/gcc/testsuite/gcc.c-torture/execute/ |
| H A D | 921202-1.c | 2 #define VLEN 2055 macro
4 #define VLEN ((STACK_SIZE/16) - 1) macro
8 long dx[VLEN+1]; in main()
9 long dy[VLEN+1]; in main()
10 long s1[VLEN]; in main()
17 size = VLEN; in main()
|
| /dports/lang/gcc11/gcc-11.2.0/gcc/testsuite/gcc.c-torture/execute/ |
| H A D | 921202-1.c | 5 #define VLEN 2055 macro
7 #define VLEN ((STACK_SIZE/16) - 1) macro
11 long dx[VLEN+1]; in main()
12 long dy[VLEN+1]; in main()
13 long s1[VLEN]; in main()
20 size = VLEN; in main()
|
| /dports/devel/zpu-gcc/zpu-toolchain-1.0/toolchain/gcc/gcc/testsuite/gcc.c-torture/execute/ |
| H A D | 921202-1.c | 2 #define VLEN 2055 macro
4 #define VLEN ((STACK_SIZE/16) - 1) macro
8 long dx[VLEN]; in main()
9 long dy[VLEN]; in main()
10 long s1[VLEN]; in main()
17 size = VLEN; in main()
|
| /dports/devel/arm-none-eabi-gcc/gcc-8.4.0/gcc/testsuite/gcc.c-torture/execute/ |
| H A D | 921202-1.c | 5 #define VLEN 2055 macro
7 #define VLEN ((STACK_SIZE/16) - 1) macro
11 long dx[VLEN+1]; in main()
12 long dy[VLEN+1]; in main()
13 long s1[VLEN]; in main()
20 size = VLEN; in main()
|
| /dports/devel/riscv64-none-elf-gcc/gcc-8.4.0/gcc/testsuite/gcc.c-torture/execute/ |
| H A D | 921202-1.c | 5 #define VLEN 2055 macro
7 #define VLEN ((STACK_SIZE/16) - 1) macro
11 long dx[VLEN+1]; in main()
12 long dy[VLEN+1]; in main()
13 long s1[VLEN]; in main()
20 size = VLEN; in main()
|
| /dports/lang/gcc9-aux/gcc-9.1.0/gcc/testsuite/gcc.c-torture/execute/ |
| H A D | 921202-1.c | 5 #define VLEN 2055 macro
7 #define VLEN ((STACK_SIZE/16) - 1) macro
11 long dx[VLEN+1]; in main()
12 long dy[VLEN+1]; in main()
13 long s1[VLEN]; in main()
20 size = VLEN; in main()
|
| /dports/misc/cxx_atomics_pic/gcc-11.2.0/gcc/testsuite/gcc.c-torture/execute/ |
| H A D | 921202-1.c | 5 #define VLEN 2055 macro
7 #define VLEN ((STACK_SIZE/16) - 1) macro
11 long dx[VLEN+1]; in main()
12 long dy[VLEN+1]; in main()
13 long s1[VLEN]; in main()
20 size = VLEN; in main()
|
| /dports/devel/aarch64-none-elf-gcc/gcc-8.4.0/gcc/testsuite/gcc.c-torture/execute/ |
| H A D | 921202-1.c | 5 #define VLEN 2055 macro
7 #define VLEN ((STACK_SIZE/16) - 1) macro
11 long dx[VLEN+1]; in main()
12 long dy[VLEN+1]; in main()
13 long s1[VLEN]; in main()
20 size = VLEN; in main()
|
| /dports/lang/gcc9-devel/gcc-9-20211007/gcc/testsuite/gcc.c-torture/execute/ |
| H A D | 921202-1.c | 5 #define VLEN 2055 macro
7 #define VLEN ((STACK_SIZE/16) - 1) macro
11 long dx[VLEN+1]; in main()
12 long dy[VLEN+1]; in main()
13 long s1[VLEN]; in main()
20 size = VLEN; in main()
|
| /dports/devel/riscv32-unknown-elf-gcc/gcc-8.4.0/gcc/testsuite/gcc.c-torture/execute/ |
| H A D | 921202-1.c | 5 #define VLEN 2055 macro
7 #define VLEN ((STACK_SIZE/16) - 1) macro
11 long dx[VLEN+1]; in main()
12 long dy[VLEN+1]; in main()
13 long s1[VLEN]; in main()
20 size = VLEN; in main()
|
| /dports/lang/gcc11-devel/gcc-11-20211009/gcc/testsuite/gcc.c-torture/execute/ |
| H A D | 921202-1.c | 5 #define VLEN 2055 macro
7 #define VLEN ((STACK_SIZE/16) - 1) macro
11 long dx[VLEN+1]; in main()
12 long dy[VLEN+1]; in main()
13 long s1[VLEN]; in main()
20 size = VLEN; in main()
|
| /dports/lang/gcc8/gcc-8.5.0/gcc/testsuite/gcc.c-torture/execute/ |
| H A D | 921202-1.c | 5 #define VLEN 2055 macro
7 #define VLEN ((STACK_SIZE/16) - 1) macro
11 long dx[VLEN+1]; in main()
12 long dy[VLEN+1]; in main()
13 long s1[VLEN]; in main()
20 size = VLEN; in main()
|
| /dports/lang/gnat_util/gcc-6-20180516/gcc/testsuite/gcc.c-torture/execute/ |
| H A D | 921202-1.c | 3 #define VLEN 2055 macro
5 #define VLEN ((STACK_SIZE/16) - 1) macro
9 long dx[VLEN+1]; in main()
10 long dy[VLEN+1]; in main()
11 long s1[VLEN]; in main()
18 size = VLEN; in main()
|
| /dports/devel/riscv64-gcc/gcc-8.3.0/gcc/testsuite/gcc.c-torture/execute/ |
| H A D | 921202-1.c | 5 #define VLEN 2055 macro
7 #define VLEN ((STACK_SIZE/16) - 1) macro
11 long dx[VLEN+1]; in main()
12 long dy[VLEN+1]; in main()
13 long s1[VLEN]; in main()
20 size = VLEN; in main()
|