| /* Implementation of the SUM intrinsic | 
 |    Copyright 2002 Free Software Foundation, Inc. | 
 |    Contributed by Paul Brook <paul@nowt.org> | 
 |  | 
 | This file is part of the GNU Fortran 95 runtime library (libgfortran). | 
 |  | 
 | Libgfortran 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. | 
 |  | 
 | In addition to the permissions in the GNU General Public License, the | 
 | Free Software Foundation gives you unlimited permission to link the | 
 | compiled version of this file into combinations with other programs, | 
 | and to distribute those combinations without any restriction coming | 
 | from the use of this file.  (The General Public License restrictions | 
 | do apply in other respects; for example, they cover modification of | 
 | the file, and distribution when not linked into a combine | 
 | executable.) | 
 |  | 
 | Libgfortran 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 libgfortran; see the file COPYING.  If not, | 
 | write to the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, | 
 | Boston, MA 02110-1301, USA.  */ | 
 |  | 
 | #include "config.h" | 
 | #include <stdlib.h> | 
 | #include <assert.h> | 
 | #include "libgfortran.h" | 
 |  | 
 |  | 
 | #if defined (HAVE_GFC_INTEGER_1) && defined (HAVE_GFC_INTEGER_1) | 
 |  | 
 |  | 
 | extern void sum_i1 (gfc_array_i1 * const restrict,  | 
 | 	gfc_array_i1 * const restrict, const index_type * const restrict); | 
 | export_proto(sum_i1); | 
 |  | 
 | void | 
 | sum_i1 (gfc_array_i1 * const restrict retarray,  | 
 | 	gfc_array_i1 * const restrict array,  | 
 | 	const index_type * const restrict pdim) | 
 | { | 
 |   index_type count[GFC_MAX_DIMENSIONS]; | 
 |   index_type extent[GFC_MAX_DIMENSIONS]; | 
 |   index_type sstride[GFC_MAX_DIMENSIONS]; | 
 |   index_type dstride[GFC_MAX_DIMENSIONS]; | 
 |   const GFC_INTEGER_1 * restrict base; | 
 |   GFC_INTEGER_1 * restrict dest; | 
 |   index_type rank; | 
 |   index_type n; | 
 |   index_type len; | 
 |   index_type delta; | 
 |   index_type dim; | 
 |  | 
 |   /* Make dim zero based to avoid confusion.  */ | 
 |   dim = (*pdim) - 1; | 
 |   rank = GFC_DESCRIPTOR_RANK (array) - 1; | 
 |  | 
 |   len = array->dim[dim].ubound + 1 - array->dim[dim].lbound; | 
 |   delta = array->dim[dim].stride; | 
 |  | 
 |   for (n = 0; n < dim; n++) | 
 |     { | 
 |       sstride[n] = array->dim[n].stride; | 
 |       extent[n] = array->dim[n].ubound + 1 - array->dim[n].lbound; | 
 |  | 
 |       if (extent[n] < 0) | 
 | 	extent[n] = 0; | 
 |     } | 
 |   for (n = dim; n < rank; n++) | 
 |     { | 
 |       sstride[n] = array->dim[n + 1].stride; | 
 |       extent[n] = | 
 |         array->dim[n + 1].ubound + 1 - array->dim[n + 1].lbound; | 
 |  | 
 |       if (extent[n] < 0) | 
 | 	extent[n] = 0; | 
 |     } | 
 |  | 
 |   if (retarray->data == NULL) | 
 |     { | 
 |       size_t alloc_size; | 
 |  | 
 |       for (n = 0; n < rank; n++) | 
 |         { | 
 |           retarray->dim[n].lbound = 0; | 
 |           retarray->dim[n].ubound = extent[n]-1; | 
 |           if (n == 0) | 
 |             retarray->dim[n].stride = 1; | 
 |           else | 
 |             retarray->dim[n].stride = retarray->dim[n-1].stride * extent[n-1]; | 
 |         } | 
 |  | 
 |       retarray->offset = 0; | 
 |       retarray->dtype = (array->dtype & ~GFC_DTYPE_RANK_MASK) | rank; | 
 |  | 
 |       alloc_size = sizeof (GFC_INTEGER_1) * retarray->dim[rank-1].stride | 
 |     		   * extent[rank-1]; | 
 |  | 
 |       if (alloc_size == 0) | 
 | 	{ | 
 | 	  /* Make sure we have a zero-sized array.  */ | 
 | 	  retarray->dim[0].lbound = 0; | 
 | 	  retarray->dim[0].ubound = -1; | 
 | 	  return; | 
 | 	} | 
 |       else | 
 | 	retarray->data = internal_malloc_size (alloc_size); | 
 |     } | 
 |   else | 
 |     { | 
 |       if (rank != GFC_DESCRIPTOR_RANK (retarray)) | 
 | 	runtime_error ("rank of return array incorrect"); | 
 |     } | 
 |  | 
 |   for (n = 0; n < rank; n++) | 
 |     { | 
 |       count[n] = 0; | 
 |       dstride[n] = retarray->dim[n].stride; | 
 |       if (extent[n] <= 0) | 
 |         len = 0; | 
 |     } | 
 |  | 
 |   base = array->data; | 
 |   dest = retarray->data; | 
 |  | 
 |   while (base) | 
 |     { | 
 |       const GFC_INTEGER_1 * restrict src; | 
 |       GFC_INTEGER_1 result; | 
 |       src = base; | 
 |       { | 
 |  | 
 |   result = 0; | 
 |         if (len <= 0) | 
 | 	  *dest = 0; | 
 | 	else | 
 | 	  { | 
 | 	    for (n = 0; n < len; n++, src += delta) | 
 | 	      { | 
 |  | 
 |   result += *src; | 
 |           } | 
 | 	    *dest = result; | 
 | 	  } | 
 |       } | 
 |       /* Advance to the next element.  */ | 
 |       count[0]++; | 
 |       base += sstride[0]; | 
 |       dest += dstride[0]; | 
 |       n = 0; | 
 |       while (count[n] == extent[n]) | 
 |         { | 
 |           /* When we get to the end of a dimension, reset it and increment | 
 |              the next dimension.  */ | 
 |           count[n] = 0; | 
 |           /* We could precalculate these products, but this is a less | 
 |              frequently used path so probably not worth it.  */ | 
 |           base -= sstride[n] * extent[n]; | 
 |           dest -= dstride[n] * extent[n]; | 
 |           n++; | 
 |           if (n == rank) | 
 |             { | 
 |               /* Break out of the look.  */ | 
 |               base = NULL; | 
 |               break; | 
 |             } | 
 |           else | 
 |             { | 
 |               count[n]++; | 
 |               base += sstride[n]; | 
 |               dest += dstride[n]; | 
 |             } | 
 |         } | 
 |     } | 
 | } | 
 |  | 
 |  | 
 | extern void msum_i1 (gfc_array_i1 * const restrict,  | 
 | 	gfc_array_i1 * const restrict, const index_type * const restrict, | 
 | 	gfc_array_l4 * const restrict); | 
 | export_proto(msum_i1); | 
 |  | 
 | void | 
 | msum_i1 (gfc_array_i1 * const restrict retarray,  | 
 | 	gfc_array_i1 * const restrict array,  | 
 | 	const index_type * const restrict pdim,  | 
 | 	gfc_array_l4 * const restrict mask) | 
 | { | 
 |   index_type count[GFC_MAX_DIMENSIONS]; | 
 |   index_type extent[GFC_MAX_DIMENSIONS]; | 
 |   index_type sstride[GFC_MAX_DIMENSIONS]; | 
 |   index_type dstride[GFC_MAX_DIMENSIONS]; | 
 |   index_type mstride[GFC_MAX_DIMENSIONS]; | 
 |   GFC_INTEGER_1 * restrict dest; | 
 |   const GFC_INTEGER_1 * restrict base; | 
 |   const GFC_LOGICAL_4 * restrict mbase; | 
 |   int rank; | 
 |   int dim; | 
 |   index_type n; | 
 |   index_type len; | 
 |   index_type delta; | 
 |   index_type mdelta; | 
 |  | 
 |   dim = (*pdim) - 1; | 
 |   rank = GFC_DESCRIPTOR_RANK (array) - 1; | 
 |  | 
 |   len = array->dim[dim].ubound + 1 - array->dim[dim].lbound; | 
 |   if (len <= 0) | 
 |     return; | 
 |   delta = array->dim[dim].stride; | 
 |   mdelta = mask->dim[dim].stride; | 
 |  | 
 |   for (n = 0; n < dim; n++) | 
 |     { | 
 |       sstride[n] = array->dim[n].stride; | 
 |       mstride[n] = mask->dim[n].stride; | 
 |       extent[n] = array->dim[n].ubound + 1 - array->dim[n].lbound; | 
 |  | 
 |       if (extent[n] < 0) | 
 | 	extent[n] = 0; | 
 |  | 
 |     } | 
 |   for (n = dim; n < rank; n++) | 
 |     { | 
 |       sstride[n] = array->dim[n + 1].stride; | 
 |       mstride[n] = mask->dim[n + 1].stride; | 
 |       extent[n] = | 
 |         array->dim[n + 1].ubound + 1 - array->dim[n + 1].lbound; | 
 |  | 
 |       if (extent[n] < 0) | 
 | 	extent[n] = 0; | 
 |     } | 
 |  | 
 |   if (retarray->data == NULL) | 
 |     { | 
 |       size_t alloc_size; | 
 |  | 
 |       for (n = 0; n < rank; n++) | 
 |         { | 
 |           retarray->dim[n].lbound = 0; | 
 |           retarray->dim[n].ubound = extent[n]-1; | 
 |           if (n == 0) | 
 |             retarray->dim[n].stride = 1; | 
 |           else | 
 |             retarray->dim[n].stride = retarray->dim[n-1].stride * extent[n-1]; | 
 |         } | 
 |  | 
 |       alloc_size = sizeof (GFC_INTEGER_1) * retarray->dim[rank-1].stride | 
 |     		   * extent[rank-1]; | 
 |  | 
 |       retarray->offset = 0; | 
 |       retarray->dtype = (array->dtype & ~GFC_DTYPE_RANK_MASK) | rank; | 
 |  | 
 |       if (alloc_size == 0) | 
 | 	{ | 
 | 	  /* Make sure we have a zero-sized array.  */ | 
 | 	  retarray->dim[0].lbound = 0; | 
 | 	  retarray->dim[0].ubound = -1; | 
 | 	  return; | 
 | 	} | 
 |       else | 
 | 	retarray->data = internal_malloc_size (alloc_size); | 
 |  | 
 |     } | 
 |   else | 
 |     { | 
 |       if (rank != GFC_DESCRIPTOR_RANK (retarray)) | 
 | 	runtime_error ("rank of return array incorrect"); | 
 |     } | 
 |  | 
 |   for (n = 0; n < rank; n++) | 
 |     { | 
 |       count[n] = 0; | 
 |       dstride[n] = retarray->dim[n].stride; | 
 |       if (extent[n] <= 0) | 
 |         return; | 
 |     } | 
 |  | 
 |   dest = retarray->data; | 
 |   base = array->data; | 
 |   mbase = mask->data; | 
 |  | 
 |   if (GFC_DESCRIPTOR_SIZE (mask) != 4) | 
 |     { | 
 |       /* This allows the same loop to be used for all logical types.  */ | 
 |       assert (GFC_DESCRIPTOR_SIZE (mask) == 8); | 
 |       for (n = 0; n < rank; n++) | 
 |         mstride[n] <<= 1; | 
 |       mdelta <<= 1; | 
 |       mbase = (GFOR_POINTER_L8_TO_L4 (mbase)); | 
 |     } | 
 |  | 
 |   while (base) | 
 |     { | 
 |       const GFC_INTEGER_1 * restrict src; | 
 |       const GFC_LOGICAL_4 * restrict msrc; | 
 |       GFC_INTEGER_1 result; | 
 |       src = base; | 
 |       msrc = mbase; | 
 |       { | 
 |  | 
 |   result = 0; | 
 |         if (len <= 0) | 
 | 	  *dest = 0; | 
 | 	else | 
 | 	  { | 
 | 	    for (n = 0; n < len; n++, src += delta, msrc += mdelta) | 
 | 	      { | 
 |  | 
 |   if (*msrc) | 
 |     result += *src; | 
 |               } | 
 | 	    *dest = result; | 
 | 	  } | 
 |       } | 
 |       /* Advance to the next element.  */ | 
 |       count[0]++; | 
 |       base += sstride[0]; | 
 |       mbase += mstride[0]; | 
 |       dest += dstride[0]; | 
 |       n = 0; | 
 |       while (count[n] == extent[n]) | 
 |         { | 
 |           /* When we get to the end of a dimension, reset it and increment | 
 |              the next dimension.  */ | 
 |           count[n] = 0; | 
 |           /* We could precalculate these products, but this is a less | 
 |              frequently used path so probably not worth it.  */ | 
 |           base -= sstride[n] * extent[n]; | 
 |           mbase -= mstride[n] * extent[n]; | 
 |           dest -= dstride[n] * extent[n]; | 
 |           n++; | 
 |           if (n == rank) | 
 |             { | 
 |               /* Break out of the look.  */ | 
 |               base = NULL; | 
 |               break; | 
 |             } | 
 |           else | 
 |             { | 
 |               count[n]++; | 
 |               base += sstride[n]; | 
 |               mbase += mstride[n]; | 
 |               dest += dstride[n]; | 
 |             } | 
 |         } | 
 |     } | 
 | } | 
 |  | 
 |  | 
 | extern void ssum_i1 (gfc_array_i1 * const restrict,  | 
 | 	gfc_array_i1 * const restrict, const index_type * const restrict, | 
 | 	GFC_LOGICAL_4 *); | 
 | export_proto(ssum_i1); | 
 |  | 
 | void | 
 | ssum_i1 (gfc_array_i1 * const restrict retarray,  | 
 | 	gfc_array_i1 * const restrict array,  | 
 | 	const index_type * const restrict pdim,  | 
 | 	GFC_LOGICAL_4 * mask) | 
 | { | 
 |   index_type rank; | 
 |   index_type n; | 
 |   index_type dstride; | 
 |   GFC_INTEGER_1 *dest; | 
 |  | 
 |   if (*mask) | 
 |     { | 
 |       sum_i1 (retarray, array, pdim); | 
 |       return; | 
 |     } | 
 |     rank = GFC_DESCRIPTOR_RANK (array); | 
 |   if (rank <= 0) | 
 |     runtime_error ("Rank of array needs to be > 0"); | 
 |  | 
 |   if (retarray->data == NULL) | 
 |     { | 
 |       retarray->dim[0].lbound = 0; | 
 |       retarray->dim[0].ubound = rank-1; | 
 |       retarray->dim[0].stride = 1; | 
 |       retarray->dtype = (retarray->dtype & ~GFC_DTYPE_RANK_MASK) | 1; | 
 |       retarray->offset = 0; | 
 |       retarray->data = internal_malloc_size (sizeof (GFC_INTEGER_1) * rank); | 
 |     } | 
 |   else | 
 |     { | 
 |       if (GFC_DESCRIPTOR_RANK (retarray) != 1) | 
 | 	runtime_error ("rank of return array does not equal 1"); | 
 |  | 
 |       if (retarray->dim[0].ubound + 1 - retarray->dim[0].lbound != rank) | 
 |         runtime_error ("dimension of return array incorrect"); | 
 |     } | 
 |  | 
 |     dstride = retarray->dim[0].stride; | 
 |     dest = retarray->data; | 
 |  | 
 |     for (n = 0; n < rank; n++) | 
 |       dest[n * dstride] = 0 ; | 
 | } | 
 |  | 
 | #endif |