Add files via upload

oneMKL2/fftw/fftw3-mpi.h

@@ -0,0 +1,221 @@
+/*
+ * Copyright (c) 2003, 2007-14 Matteo Frigo
+ * Copyright (c) 2003, 2007-14 Massachusetts Institute of Technology
+ *
+ * The following statement of license applies *only* to this header file,
+ * and *not* to the other files distributed with FFTW or derived therefrom:
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ *
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS
+ * OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
+ * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY
+ * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE
+ * GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY,
+ * WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
+ * NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
+ * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+/***************************** NOTE TO USERS *********************************
+ *
+ *                 THIS IS A HEADER FILE, NOT A MANUAL
+ *
+ *    If you want to know how to use FFTW, please read the manual,
+ *    online at http://www.fftw.org/doc/ and also included with FFTW.
+ *    For a quick start, see the manual's tutorial section.
+ *
+ *   (Reading header files to learn how to use a library is a habit
+ *    stemming from code lacking a proper manual.  Arguably, it's a
+ *    *bad* habit in most cases, because header files can contain
+ *    interfaces that are not part of the public, stable API.)
+ *
+ ****************************************************************************/
+
+#ifndef FFTW3_MPI_H
+#define FFTW3_MPI_H
+
+#include "fftw3.h"
+#include <mpi.h>
+
+#ifdef __cplusplus
+extern "C"
+{
+#endif /* __cplusplus */
+
+struct fftw_mpi_ddim_do_not_use_me {
+     ptrdiff_t n;                     /* dimension size */
+     ptrdiff_t ib;                    /* input block */
+     ptrdiff_t ob;                    /* output block */
+};
+
+/*
+  huge second-order macro that defines prototypes for all API
+  functions.  We expand this macro for each supported precision
+
+  XM: name-mangling macro (MPI)
+  X: name-mangling macro (serial)
+  R: real data type
+  C: complex data type
+*/
+
+#define FFTW_MPI_DEFINE_API(XM, X, R, C)			\
+								\
+typedef struct fftw_mpi_ddim_do_not_use_me XM(ddim);		\
+								\
+FFTW_EXTERN void XM(init)(void);				\
+FFTW_EXTERN void XM(cleanup)(void);				\
+								\
+FFTW_EXTERN ptrdiff_t XM(local_size_many_transposed)		\
+     (int rnk, const ptrdiff_t *n, ptrdiff_t howmany,		\
+      ptrdiff_t block0, ptrdiff_t block1, MPI_Comm comm,	\
+      ptrdiff_t *local_n0, ptrdiff_t *local_0_start,		\
+      ptrdiff_t *local_n1, ptrdiff_t *local_1_start);		\
+FFTW_EXTERN ptrdiff_t XM(local_size_many)			\
+     (int rnk, const ptrdiff_t *n, ptrdiff_t howmany,		\
+      ptrdiff_t block0, MPI_Comm comm,				\
+      ptrdiff_t *local_n0, ptrdiff_t *local_0_start);		\
+FFTW_EXTERN ptrdiff_t XM(local_size_transposed)			\
+     (int rnk, const ptrdiff_t *n, MPI_Comm comm,		\
+      ptrdiff_t *local_n0, ptrdiff_t *local_0_start,		\
+      ptrdiff_t *local_n1, ptrdiff_t *local_1_start);		\
+FFTW_EXTERN ptrdiff_t XM(local_size)				\
+     (int rnk, const ptrdiff_t *n, MPI_Comm comm,		\
+      ptrdiff_t *local_n0, ptrdiff_t *local_0_start);		\
+FFTW_EXTERN ptrdiff_t XM(local_size_many_1d)(			\
+     ptrdiff_t n0, ptrdiff_t howmany,				\
+     MPI_Comm comm, int sign, unsigned flags,			\
+     ptrdiff_t *local_ni, ptrdiff_t *local_i_start,		\
+     ptrdiff_t *local_no, ptrdiff_t *local_o_start);		\
+FFTW_EXTERN ptrdiff_t XM(local_size_1d)(			\
+     ptrdiff_t n0, MPI_Comm comm, int sign, unsigned flags,	\
+     ptrdiff_t *local_ni, ptrdiff_t *local_i_start,		\
+     ptrdiff_t *local_no, ptrdiff_t *local_o_start);		\
+FFTW_EXTERN ptrdiff_t XM(local_size_2d)(			\
+     ptrdiff_t n0, ptrdiff_t n1, MPI_Comm comm,			\
+     ptrdiff_t *local_n0, ptrdiff_t *local_0_start);		\
+FFTW_EXTERN ptrdiff_t XM(local_size_2d_transposed)(		\
+     ptrdiff_t n0, ptrdiff_t n1, MPI_Comm comm,			\
+     ptrdiff_t *local_n0, ptrdiff_t *local_0_start,		\
+     ptrdiff_t *local_n1, ptrdiff_t *local_1_start);		\
+FFTW_EXTERN ptrdiff_t XM(local_size_3d)(			\
+     ptrdiff_t n0, ptrdiff_t n1, ptrdiff_t n2, MPI_Comm comm,	\
+     ptrdiff_t *local_n0, ptrdiff_t *local_0_start);		\
+FFTW_EXTERN ptrdiff_t XM(local_size_3d_transposed)(		\
+     ptrdiff_t n0, ptrdiff_t n1, ptrdiff_t n2, MPI_Comm comm,	\
+     ptrdiff_t *local_n0, ptrdiff_t *local_0_start,		\
+     ptrdiff_t *local_n1, ptrdiff_t *local_1_start);		\
+								\
+FFTW_EXTERN X(plan) XM(plan_many_transpose)			\
+     (ptrdiff_t n0, ptrdiff_t n1,				\
+      ptrdiff_t howmany, ptrdiff_t block0, ptrdiff_t block1,	\
+      R *in, R *out, MPI_Comm comm, unsigned flags);		\
+FFTW_EXTERN X(plan) XM(plan_transpose)				\
+     (ptrdiff_t n0, ptrdiff_t n1,				\
+      R *in, R *out, MPI_Comm comm, unsigned flags);		\
+								\
+FFTW_EXTERN X(plan) XM(plan_many_dft)				\
+     (int rnk, const ptrdiff_t *n, ptrdiff_t howmany,		\
+      ptrdiff_t block, ptrdiff_t tblock, C *in, C *out,		\
+      MPI_Comm comm, int sign, unsigned flags);			\
+FFTW_EXTERN X(plan) XM(plan_dft)				\
+     (int rnk, const ptrdiff_t *n, C *in, C *out,		\
+      MPI_Comm comm, int sign, unsigned flags);			\
+FFTW_EXTERN X(plan) XM(plan_dft_1d)				\
+     (ptrdiff_t n0, C *in, C *out,				\
+      MPI_Comm comm, int sign, unsigned flags);			\
+FFTW_EXTERN X(plan) XM(plan_dft_2d)				\
+     (ptrdiff_t n0, ptrdiff_t n1, C *in, C *out,		\
+      MPI_Comm comm, int sign, unsigned flags);			\
+FFTW_EXTERN X(plan) XM(plan_dft_3d)				\
+     (ptrdiff_t n0, ptrdiff_t n1, ptrdiff_t n2, C *in, C *out,	\
+      MPI_Comm comm, int sign, unsigned flags);			\
+								\
+FFTW_EXTERN X(plan) XM(plan_many_r2r)				\
+     (int rnk, const ptrdiff_t *n, ptrdiff_t howmany,		\
+      ptrdiff_t iblock, ptrdiff_t oblock, R *in, R *out,	\
+      MPI_Comm comm, const X(r2r_kind) *kind, unsigned flags);	\
+FFTW_EXTERN X(plan) XM(plan_r2r)				\
+     (int rnk, const ptrdiff_t *n, R *in, R *out,		\
+      MPI_Comm comm, const X(r2r_kind) *kind, unsigned flags);	\
+FFTW_EXTERN X(plan) XM(plan_r2r_2d)				\
+     (ptrdiff_t n0, ptrdiff_t n1, R *in, R *out, MPI_Comm comm,	\
+      X(r2r_kind) kind0, X(r2r_kind) kind1, unsigned flags);	\
+FFTW_EXTERN X(plan) XM(plan_r2r_3d)				\
+     (ptrdiff_t n0, ptrdiff_t n1, ptrdiff_t n2,			\
+      R *in, R *out, MPI_Comm comm, X(r2r_kind) kind0,		\
+      X(r2r_kind) kind1, X(r2r_kind) kind2, unsigned flags);	\
+								\
+FFTW_EXTERN X(plan) XM(plan_many_dft_r2c)			\
+     (int rnk, const ptrdiff_t *n, ptrdiff_t howmany,		\
+      ptrdiff_t iblock, ptrdiff_t oblock, R *in, C *out,	\
+      MPI_Comm comm, unsigned flags);				\
+FFTW_EXTERN X(plan) XM(plan_dft_r2c)				\
+     (int rnk, const ptrdiff_t *n, R *in, C *out,		\
+      MPI_Comm comm, unsigned flags);				\
+FFTW_EXTERN X(plan) XM(plan_dft_r2c_2d)				\
+     (ptrdiff_t n0, ptrdiff_t n1, R *in, C *out,		\
+      MPI_Comm comm, unsigned flags);				\
+FFTW_EXTERN X(plan) XM(plan_dft_r2c_3d)				\
+     (ptrdiff_t n0, ptrdiff_t n1, ptrdiff_t n2, R *in, C *out,	\
+      MPI_Comm comm, unsigned flags);				\
+								\
+FFTW_EXTERN X(plan) XM(plan_many_dft_c2r)			\
+     (int rnk, const ptrdiff_t *n, ptrdiff_t howmany,		\
+      ptrdiff_t iblock, ptrdiff_t oblock, C *in, R *out,	\
+      MPI_Comm comm, unsigned flags);				\
+FFTW_EXTERN X(plan) XM(plan_dft_c2r)				\
+     (int rnk, const ptrdiff_t *n, C *in, R *out,		\
+      MPI_Comm comm, unsigned flags);				\
+FFTW_EXTERN X(plan) XM(plan_dft_c2r_2d)				\
+     (ptrdiff_t n0, ptrdiff_t n1, C *in, R *out,		\
+      MPI_Comm comm, unsigned flags);				\
+FFTW_EXTERN X(plan) XM(plan_dft_c2r_3d)				\
+     (ptrdiff_t n0, ptrdiff_t n1, ptrdiff_t n2, C *in, R *out,	\
+      MPI_Comm comm, unsigned flags);				\
+								\
+FFTW_EXTERN void XM(gather_wisdom)(MPI_Comm comm_);		\
+FFTW_EXTERN void XM(broadcast_wisdom)(MPI_Comm comm_);          \
+								\
+FFTW_EXTERN void XM(execute_dft)(X(plan) p, C *in, C *out);	\
+FFTW_EXTERN void XM(execute_dft_r2c)(X(plan) p, R *in, C *out);	\
+FFTW_EXTERN void XM(execute_dft_c2r)(X(plan) p, C *in, R *out);	\
+FFTW_EXTERN void XM(execute_r2r)(X(plan) p, R *in, R *out);
+
+
+
+/* end of FFTW_MPI_DEFINE_API macro */
+
+#define FFTW_MPI_MANGLE_DOUBLE(name) FFTW_MANGLE_DOUBLE(FFTW_CONCAT(mpi_,name))
+#define FFTW_MPI_MANGLE_FLOAT(name) FFTW_MANGLE_FLOAT(FFTW_CONCAT(mpi_,name))
+#define FFTW_MPI_MANGLE_LONG_DOUBLE(name) FFTW_MANGLE_LONG_DOUBLE(FFTW_CONCAT(mpi_,name))
+
+FFTW_MPI_DEFINE_API(FFTW_MPI_MANGLE_DOUBLE, FFTW_MANGLE_DOUBLE, double, fftw_complex)
+FFTW_MPI_DEFINE_API(FFTW_MPI_MANGLE_FLOAT, FFTW_MANGLE_FLOAT, float, fftwf_complex)
+FFTW_MPI_DEFINE_API(FFTW_MPI_MANGLE_LONG_DOUBLE, FFTW_MANGLE_LONG_DOUBLE, long double, fftwl_complex)
+
+#define FFTW_MPI_DEFAULT_BLOCK (0)
+
+/* MPI-specific flags */
+#define FFTW_MPI_SCRAMBLED_IN (1U << 27)
+#define FFTW_MPI_SCRAMBLED_OUT (1U << 28)
+#define FFTW_MPI_TRANSPOSED_IN (1U << 29)
+#define FFTW_MPI_TRANSPOSED_OUT (1U << 30)
+
+#ifdef __cplusplus
+}  /* extern "C" */
+#endif /* __cplusplus */
+
+#endif /* FFTW3_MPI_H */

oneMKL2/fftw/fftw3-mpi_mkl.h

@@ -0,0 +1,53 @@
+/*******************************************************************************
+* Copyright 2010-2022 Intel Corporation.
+*
+* This software and the related documents are Intel copyrighted  materials,  and
+* your use of  them is  governed by the  express license  under which  they were
+* provided to you (License).  Unless the License provides otherwise, you may not
+* use, modify, copy, publish, distribute,  disclose or transmit this software or
+* the related documents without Intel's prior written permission.
+*
+* This software and the related documents  are provided as  is,  with no express
+* or implied  warranties,  other  than those  that are  expressly stated  in the
+* License.
+*******************************************************************************/
+
+/*
+ *
+ * Definitions for MPI FFTW3 wrappers to Intel(R) oneAPI Math Kernel Library (Intel(R) oneMKL).
+ *
+ ******************************************************************************
+ */
+
+#ifndef FFTW3_MPI_MKL_H
+#define FFTW3_MPI_MKL_H
+
+#include "fftw3-mpi.h"
+
+#if defined(MKL_SINGLE)
+typedef float real_t;
+typedef fftwf_complex complex_t;
+#define MPI_PREC MPI_FLOAT
+#define MKL_PREC DFTI_SINGLE
+#define FFTW_MPI_MANGLE(name) FFTW_MPI_MANGLE_FLOAT(name)
+#define FFTW_MANGLE(name) FFTW_MANGLE_FLOAT(name)
+#else
+typedef double real_t;
+typedef fftw_complex complex_t;
+#define MPI_PREC MPI_DOUBLE
+#define MKL_PREC DFTI_DOUBLE
+#define FFTW_MPI_MANGLE(name) FFTW_MPI_MANGLE_DOUBLE(name)
+#define FFTW_MANGLE(name) FFTW_MANGLE_DOUBLE(name)
+#endif
+
+#include "fftw3_mkl.h"
+#include "mkl_cdft.h"
+
+#define WANT_FAST_INPLACE_CLUSTER_FFT 1
+/* if WANT_FAST_INPLACE_CLUSTER_FFT set to 1, FFTW3 MPI wrappers internally
+ * allocate additional memory(workspace) needed for fast inplace Intel(R) oneMKL CDFT
+ * otherwise, no additional memory is used, though the perfomance would be
+ * worse, because of many MPI communications */
+
+#endif /* FFTW3_MPI_MKL_H */
+

oneMKL2/fftw/fftw3.f

@@ -0,0 +1,80 @@
+!*******************************************************************************
+!   Copyright (c) 2003, 2007-11 Matteo Frigo
+!   Copyright (c) 2003, 2007-11 Massachusetts Institute of Technology
+!
+!   This program is distributed with permission
+!
+!*******************************************************************************
+
+      INTEGER FFTW_R2HC
+      PARAMETER (FFTW_R2HC=0)
+      INTEGER FFTW_HC2R
+      PARAMETER (FFTW_HC2R=1)
+      INTEGER FFTW_DHT
+      PARAMETER (FFTW_DHT=2)
+      INTEGER FFTW_REDFT00
+      PARAMETER (FFTW_REDFT00=3)
+      INTEGER FFTW_REDFT01
+      PARAMETER (FFTW_REDFT01=4)
+      INTEGER FFTW_REDFT10
+      PARAMETER (FFTW_REDFT10=5)
+      INTEGER FFTW_REDFT11
+      PARAMETER (FFTW_REDFT11=6)
+      INTEGER FFTW_RODFT00
+      PARAMETER (FFTW_RODFT00=7)
+      INTEGER FFTW_RODFT01
+      PARAMETER (FFTW_RODFT01=8)
+      INTEGER FFTW_RODFT10
+      PARAMETER (FFTW_RODFT10=9)
+      INTEGER FFTW_RODFT11
+      PARAMETER (FFTW_RODFT11=10)
+      INTEGER FFTW_FORWARD
+      PARAMETER (FFTW_FORWARD=-1)
+      INTEGER FFTW_BACKWARD
+      PARAMETER (FFTW_BACKWARD=+1)
+      INTEGER FFTW_MEASURE
+      PARAMETER (FFTW_MEASURE=0)
+      INTEGER FFTW_DESTROY_INPUT
+      PARAMETER (FFTW_DESTROY_INPUT=1)
+      INTEGER FFTW_UNALIGNED
+      PARAMETER (FFTW_UNALIGNED=2)
+      INTEGER FFTW_CONSERVE_MEMORY
+      PARAMETER (FFTW_CONSERVE_MEMORY=4)
+      INTEGER FFTW_EXHAUSTIVE
+      PARAMETER (FFTW_EXHAUSTIVE=8)
+      INTEGER FFTW_PRESERVE_INPUT
+      PARAMETER (FFTW_PRESERVE_INPUT=16)
+      INTEGER FFTW_PATIENT
+      PARAMETER (FFTW_PATIENT=32)
+      INTEGER FFTW_ESTIMATE
+      PARAMETER (FFTW_ESTIMATE=64)
+      INTEGER FFTW_WISDOM_ONLY
+      PARAMETER (FFTW_WISDOM_ONLY=2097152)
+      INTEGER FFTW_ESTIMATE_PATIENT
+      PARAMETER (FFTW_ESTIMATE_PATIENT=128)
+      INTEGER FFTW_BELIEVE_PCOST
+      PARAMETER (FFTW_BELIEVE_PCOST=256)
+      INTEGER FFTW_NO_DFT_R2HC
+      PARAMETER (FFTW_NO_DFT_R2HC=512)
+      INTEGER FFTW_NO_NONTHREADED
+      PARAMETER (FFTW_NO_NONTHREADED=1024)
+      INTEGER FFTW_NO_BUFFERING
+      PARAMETER (FFTW_NO_BUFFERING=2048)
+      INTEGER FFTW_NO_INDIRECT_OP
+      PARAMETER (FFTW_NO_INDIRECT_OP=4096)
+      INTEGER FFTW_ALLOW_LARGE_GENERIC
+      PARAMETER (FFTW_ALLOW_LARGE_GENERIC=8192)
+      INTEGER FFTW_NO_RANK_SPLITS
+      PARAMETER (FFTW_NO_RANK_SPLITS=16384)
+      INTEGER FFTW_NO_VRANK_SPLITS
+      PARAMETER (FFTW_NO_VRANK_SPLITS=32768)
+      INTEGER FFTW_NO_VRECURSE
+      PARAMETER (FFTW_NO_VRECURSE=65536)
+      INTEGER FFTW_NO_SIMD
+      PARAMETER (FFTW_NO_SIMD=131072)
+      INTEGER FFTW_NO_SLOW
+      PARAMETER (FFTW_NO_SLOW=262144)
+      INTEGER FFTW_NO_FIXED_RADIX_LARGE_N
+      PARAMETER (FFTW_NO_FIXED_RADIX_LARGE_N=524288)
+      INTEGER FFTW_ALLOW_PRUNING
+      PARAMETER (FFTW_ALLOW_PRUNING=1048576)