view Lib/fftw-3.2.1/cell/spu/fftw-spu.h @ 1:e86e9c111b29

Updates stuff that potentially fixes the memory leak and also makes it work on Windows and Linux (Need to test). Still have to fix fftw include for linux in Jucer.

/*
 * Copyright (c) 2007 Massachusetts Institute of Technology
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License as published by
 * the Free Software Foundation; either version 2 of the License, or
 * (at your option) any later version.
 *
 * This program is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU General Public License for more details.
 *
 * You should have received a copy of the GNU General Public License
 * along with this program; if not, write to the Free Software
 * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
 *
 */

#include <spu_intrinsics.h>
#include <spu_mfcio.h>

#define FAKE_CRUDE_TIME		/* the spu does not have proper time headers */
#include <ifftw.h>

/* override the definition in <ifftw.h>, since this hack is not
   required for the spu */
#undef MAKE_VOLATILE_STRIDE
#define MAKE_VOLATILE_STRIDE(x) (void)0

#ifdef FFTW_SINGLE
#include "spu-single.h"
#else
#include "spu-double.h"
#endif

struct spu_plan {
     int r;   /* Radix.  R<0 means nontwiddle codelet and stop */
     int m;
     R *W;    /* twiddle factors */
};

typedef void (*n_codelet) (const R *ri, const R *ii, R *ro, R *io,
			   stride is, stride os, INT v, INT ivs, INT ovs);

typedef void (*t_codelet) (R *ri, R *ii, const R *W, 
			   stride rs, INT mb, INT me, INT ms);

extern const n_codelet X(n2fvtab)[33];
extern const t_codelet X(t1fvtab)[33];

void X(spu_execute_plan) (struct spu_plan * p, const R *in, R *out);

void X(spu_alloc_reset)(void);
void *X(spu_alloc)(size_t sz);
size_t X(spu_alloc_avail)(void);

void X(spu_dma1d)(void *spu_addr, unsigned long long ppu_addr, size_t sz,
		  unsigned int cmdl);

void X(spu_dma2d)(R *A, unsigned long long ppu_addr, 
		  int n, /* int spu_stride = 2 , */ int ppu_stride_bytes,
		  int v, /* int spu_vstride = 2 * n, */
		  int ppu_vstride_bytes,
		  unsigned int cmdl);

struct dft_context;
void X(spu_do_dft)(const struct dft_context *dft);

struct transpose_context;
void X(spu_do_transpose)(const struct transpose_context *t);

struct copy_context;
void X(spu_do_copy)(const struct copy_context *c);

/* DMA preferred alignment */
#define ALIGNMENT 128

/* add k to spu_addr, 0 < k <= ALIGNMENT, so that 
   (spu_addr + k) % ALIGNMENT == ppu_addr % ALIGNMENT */
#define ALIGN_LIKE(spu_addr, ppu_addr)					 \
(void *)(((char *)(spu_addr)) +						 \
  ((((unsigned)(ppu_addr)) - ((unsigned)(spu_addr))) & (ALIGNMENT - 1)))

/* list of codelets: */
void X(spu_n2fv_2) (const R *ri, const R *ii, R *ro, R *io, stride is,
		    stride os, INT v, INT ivs, INT ovs);
void X(spu_n2fv_3) (const R *ri, const R *ii, R *ro, R *io, stride is,
		    stride os, INT v, INT ivs, INT ovs);
void X(spu_n2fv_4) (const R *ri, const R *ii, R *ro, R *io, stride is,
		    stride os, INT v, INT ivs, INT ovs);
void X(spu_n2fv_5) (const R *ri, const R *ii, R *ro, R *io, stride is,
		    stride os, INT v, INT ivs, INT ovs);
void X(spu_n2fv_6) (const R *ri, const R *ii, R *ro, R *io, stride is,
		    stride os, INT v, INT ivs, INT ovs);
void X(spu_n2fv_7) (const R *ri, const R *ii, R *ro, R *io, stride is,
		    stride os, INT v, INT ivs, INT ovs);
void X(spu_n2fv_8) (const R *ri, const R *ii, R *ro, R *io, stride is,
		    stride os, INT v, INT ivs, INT ovs);
void X(spu_n2fv_9) (const R *ri, const R *ii, R *ro, R *io, stride is,
		    stride os, INT v, INT ivs, INT ovs);
void X(spu_n2fv_10) (const R *ri, const R *ii, R *ro, R *io, stride is,
		     stride os, INT v, INT ivs, INT ovs);
void X(spu_n2fv_11) (const R *ri, const R *ii, R *ro, R *io, stride is,
		    stride os, INT v, INT ivs, INT ovs);
void X(spu_n2fv_12) (const R *ri, const R *ii, R *ro, R *io, stride is,
		     stride os, INT v, INT ivs, INT ovs);
void X(spu_n2fv_13) (const R *ri, const R *ii, R *ro, R *io, stride is,
		    stride os, INT v, INT ivs, INT ovs);
void X(spu_n2fv_14) (const R *ri, const R *ii, R *ro, R *io, stride is,
		     stride os, INT v, INT ivs, INT ovs);
void X(spu_n2fv_15) (const R *ri, const R *ii, R *ro, R *io, stride is,
		    stride os, INT v, INT ivs, INT ovs);
void X(spu_n2fv_16) (const R *ri, const R *ii, R *ro, R *io, stride is,
		     stride os, INT v, INT ivs, INT ovs);
void X(spu_n2fv_32) (const R *ri, const R *ii, R *ro, R *io, stride is,
		     stride os, INT v, INT ivs, INT ovs);

void X(spu_t1fv_10) (R *ri, R *ii, const R *W, stride rs, 
		     INT mb, INT me, INT ms);
void X(spu_t1fv_12) (R *ri, R *ii, const R *W, stride rs,
		     INT mb, INT me, INT ms);
void X(spu_t1fv_15) (R *ri, R *ii, const R *W, stride rs,
		     INT mb, INT me, INT ms);
void X(spu_t1fv_16) (R *ri, R *ii, const R *W, stride rs,
		     INT mb, INT me, INT ms);
void X(spu_t1fv_2) (R *ri, R *ii, const R *W, stride rs,
		    INT mb, INT me, INT ms);
void X(spu_t1fv_32) (R *ri, R *ii, const R *W, stride rs,
		     INT mb, INT me, INT ms);
void X(spu_t1fv_3) (R *ri, R *ii, const R *W, stride rs,
		    INT mb, INT me, INT ms);
void X(spu_t1fv_4) (R *ri, R *ii, const R *W, stride rs,
		    INT mb, INT me, INT ms);
void X(spu_t1fv_5) (R *ri, R *ii, const R *W, stride rs,
		    INT mb, INT me, INT ms);
void X(spu_t1fv_6) (R *ri, R *ii, const R *W, stride rs,
		    INT mb, INT me, INT ms);
void X(spu_t1fv_7) (R *ri, R *ii, const R *W, stride rs,
		    INT mb, INT me, INT ms);
void X(spu_t1fv_8) (R *ri, R *ii, const R *W, stride rs,
		    INT mb, INT me, INT ms);
void X(spu_t1fv_9) (R *ri, R *ii, const R *W, stride rs,
		    INT mb, INT me, INT ms);