gcc/libgfortran/generated/matmul_l4.c
Janne Blomqvist 92e6f3a43e Introduce xmallocarray, an overflow checking variant of xmalloc.
2014-06-17  Janne Blomqvist  <jb@gcc.gnu.org>

	* libgfortran.h (xmallocarray): New prototype.
	* runtime/memory.c (xmallocarray): New function.
	(xcalloc): Check for nonzero separately instead of multiplying.
	* generated/*.c: Regenerated.
	* intrinsics/cshift0.c (cshift0): Call xmallocarray instead of
	xmalloc.
	* intrinsics/eoshift0.c (eoshift0): Likewise.
	* intrinsics/eoshift2.c (eoshift2): Likewise.
	* intrinsics/pack_generic.c (pack_internal): Likewise.
	(pack_s_internal): Likewise.
	* intrinsics/reshape_generic.c (reshape_internal): Likewise.
	* intrinsics/spread_generic.c (spread_internal): Likewise.
	(spread_internal_scalar): Likewise.
	* intrinsics/string_intrinsics_inc.c (string_trim): Likewise.
	(string_minmax): Likewise.
	* intrinsics/transpose_generic.c (transpose_internal): Likewise.
	* intrinsics/unpack_generic.c (unpack_internal): Likewise.
	* io/list_read.c (nml_touch_nodes): Don't cast xmalloc return value.
	* io/transfer.c (st_set_nml_var): Call xmallocarray instead of
	xmalloc.
	* io/unit.c (get_internal_unit): Likewise.
	(filename_from_unit): Don't cast xmalloc return value.
	* io/write.c (nml_write_obj): Likewise, formatting.
	* m4/bessel.m4 (bessel_jn_r'rtype_kind`): Call xmallocarray
	instead of xmalloc.
	(besse_yn_r'rtype_kind`): Likewise.
	* m4/cshift1.m4 (cshift1): Likewise.
	* m4/eoshift1.m4 (eoshift1): Likewise.
	* m4/eoshift3.m4 (eoshift3): Likewise.
	* m4/iforeach.m4: Likewise.
	* m4/ifunction.m4: Likewise.
	* m4/ifunction_logical.m4 (name`'rtype_qual`_'atype_code):
	Likewise.
	* m4/in_pack.m4 (internal_pack_'rtype_ccode`): Likewise.
	* m4/matmul.m4 (matmul_'rtype_code`): Likewise.
	* m4/matmull.m4 (matmul_'rtype_code`): Likewise.
	* m4/pack.m4 (pack_'rtype_code`): Likewise.
	* m4/reshape.m4 (reshape_'rtype_ccode`): Likewise.
	* m4/shape.m4 (shape_'rtype_kind`): Likewise.
	* m4/spread.m4 (spread_'rtype_code`): Likewise.
	(spread_scalar_'rtype_code`): Likewise.
	* m4/transpose.m4 (transpose_'rtype_code`): Likewise.
	* m4/unpack.m4 (unpack0_'rtype_code`): Likewise.
	(unpack1_'rtype_code`): Likewise.
	* runtime/convert_char.c (convert_char1_to_char4): Likewise.
	(convert_char4_to_char1): Simplify.
	* runtime/environ.c (init_unformatted): Call xmallocarray instead
	of xmalloc.
	* runtime/in_pack_generic.c (internal_pack): Likewise.

From-SVN: r211721
2014-06-17 06:50:34 +03:00

240 lines
6.8 KiB
C

/* Implementation of the MATMUL intrinsic
Copyright (C) 2002-2014 Free Software Foundation, Inc.
Contributed by Paul Brook <paul@nowt.org>
This file is part of the GNU Fortran 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 3 of the License, or (at your option) any later version.
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.
Under Section 7 of GPL version 3, you are granted additional
permissions described in the GCC Runtime Library Exception, version
3.1, as published by the Free Software Foundation.
You should have received a copy of the GNU General Public License and
a copy of the GCC Runtime Library Exception along with this program;
see the files COPYING3 and COPYING.RUNTIME respectively. If not, see
<http://www.gnu.org/licenses/>. */
#include "libgfortran.h"
#include <stdlib.h>
#include <assert.h>
#if defined (HAVE_GFC_LOGICAL_4)
/* Dimensions: retarray(x,y) a(x, count) b(count,y).
Either a or b can be rank 1. In this case x or y is 1. */
extern void matmul_l4 (gfc_array_l4 * const restrict,
gfc_array_l1 * const restrict, gfc_array_l1 * const restrict);
export_proto(matmul_l4);
void
matmul_l4 (gfc_array_l4 * const restrict retarray,
gfc_array_l1 * const restrict a, gfc_array_l1 * const restrict b)
{
const GFC_LOGICAL_1 * restrict abase;
const GFC_LOGICAL_1 * restrict bbase;
GFC_LOGICAL_4 * restrict dest;
index_type rxstride;
index_type rystride;
index_type xcount;
index_type ycount;
index_type xstride;
index_type ystride;
index_type x;
index_type y;
int a_kind;
int b_kind;
const GFC_LOGICAL_1 * restrict pa;
const GFC_LOGICAL_1 * restrict pb;
index_type astride;
index_type bstride;
index_type count;
index_type n;
assert (GFC_DESCRIPTOR_RANK (a) == 2
|| GFC_DESCRIPTOR_RANK (b) == 2);
if (retarray->base_addr == NULL)
{
if (GFC_DESCRIPTOR_RANK (a) == 1)
{
GFC_DIMENSION_SET(retarray->dim[0], 0,
GFC_DESCRIPTOR_EXTENT(b,1) - 1, 1);
}
else if (GFC_DESCRIPTOR_RANK (b) == 1)
{
GFC_DIMENSION_SET(retarray->dim[0], 0,
GFC_DESCRIPTOR_EXTENT(a,0) - 1, 1);
}
else
{
GFC_DIMENSION_SET(retarray->dim[0], 0,
GFC_DESCRIPTOR_EXTENT(a,0) - 1, 1);
GFC_DIMENSION_SET(retarray->dim[1], 0,
GFC_DESCRIPTOR_EXTENT(b,1) - 1,
GFC_DESCRIPTOR_EXTENT(retarray,0));
}
retarray->base_addr
= xmallocarray (size0 ((array_t *) retarray), sizeof (GFC_LOGICAL_4));
retarray->offset = 0;
}
else if (unlikely (compile_options.bounds_check))
{
index_type ret_extent, arg_extent;
if (GFC_DESCRIPTOR_RANK (a) == 1)
{
arg_extent = GFC_DESCRIPTOR_EXTENT(b,1);
ret_extent = GFC_DESCRIPTOR_EXTENT(retarray,0);
if (arg_extent != ret_extent)
runtime_error ("Incorrect extent in return array in"
" MATMUL intrinsic: is %ld, should be %ld",
(long int) ret_extent, (long int) arg_extent);
}
else if (GFC_DESCRIPTOR_RANK (b) == 1)
{
arg_extent = GFC_DESCRIPTOR_EXTENT(a,0);
ret_extent = GFC_DESCRIPTOR_EXTENT(retarray,0);
if (arg_extent != ret_extent)
runtime_error ("Incorrect extent in return array in"
" MATMUL intrinsic: is %ld, should be %ld",
(long int) ret_extent, (long int) arg_extent);
}
else
{
arg_extent = GFC_DESCRIPTOR_EXTENT(a,0);
ret_extent = GFC_DESCRIPTOR_EXTENT(retarray,0);
if (arg_extent != ret_extent)
runtime_error ("Incorrect extent in return array in"
" MATMUL intrinsic for dimension 1:"
" is %ld, should be %ld",
(long int) ret_extent, (long int) arg_extent);
arg_extent = GFC_DESCRIPTOR_EXTENT(b,1);
ret_extent = GFC_DESCRIPTOR_EXTENT(retarray,1);
if (arg_extent != ret_extent)
runtime_error ("Incorrect extent in return array in"
" MATMUL intrinsic for dimension 2:"
" is %ld, should be %ld",
(long int) ret_extent, (long int) arg_extent);
}
}
abase = a->base_addr;
a_kind = GFC_DESCRIPTOR_SIZE (a);
if (a_kind == 1 || a_kind == 2 || a_kind == 4 || a_kind == 8
#ifdef HAVE_GFC_LOGICAL_16
|| a_kind == 16
#endif
)
abase = GFOR_POINTER_TO_L1 (abase, a_kind);
else
internal_error (NULL, "Funny sized logical array");
bbase = b->base_addr;
b_kind = GFC_DESCRIPTOR_SIZE (b);
if (b_kind == 1 || b_kind == 2 || b_kind == 4 || b_kind == 8
#ifdef HAVE_GFC_LOGICAL_16
|| b_kind == 16
#endif
)
bbase = GFOR_POINTER_TO_L1 (bbase, b_kind);
else
internal_error (NULL, "Funny sized logical array");
dest = retarray->base_addr;
if (GFC_DESCRIPTOR_RANK (retarray) == 1)
{
rxstride = GFC_DESCRIPTOR_STRIDE(retarray,0);
rystride = rxstride;
}
else
{
rxstride = GFC_DESCRIPTOR_STRIDE(retarray,0);
rystride = GFC_DESCRIPTOR_STRIDE(retarray,1);
}
/* If we have rank 1 parameters, zero the absent stride, and set the size to
one. */
if (GFC_DESCRIPTOR_RANK (a) == 1)
{
astride = GFC_DESCRIPTOR_STRIDE_BYTES(a,0);
count = GFC_DESCRIPTOR_EXTENT(a,0);
xstride = 0;
rxstride = 0;
xcount = 1;
}
else
{
astride = GFC_DESCRIPTOR_STRIDE_BYTES(a,1);
count = GFC_DESCRIPTOR_EXTENT(a,1);
xstride = GFC_DESCRIPTOR_STRIDE_BYTES(a,0);
xcount = GFC_DESCRIPTOR_EXTENT(a,0);
}
if (GFC_DESCRIPTOR_RANK (b) == 1)
{
bstride = GFC_DESCRIPTOR_STRIDE_BYTES(b,0);
assert(count == GFC_DESCRIPTOR_EXTENT(b,0));
ystride = 0;
rystride = 0;
ycount = 1;
}
else
{
bstride = GFC_DESCRIPTOR_STRIDE_BYTES(b,0);
assert(count == GFC_DESCRIPTOR_EXTENT(b,0));
ystride = GFC_DESCRIPTOR_STRIDE_BYTES(b,1);
ycount = GFC_DESCRIPTOR_EXTENT(b,1);
}
for (y = 0; y < ycount; y++)
{
for (x = 0; x < xcount; x++)
{
/* Do the summation for this element. For real and integer types
this is the same as DOT_PRODUCT. For complex types we use do
a*b, not conjg(a)*b. */
pa = abase;
pb = bbase;
*dest = 0;
for (n = 0; n < count; n++)
{
if (*pa && *pb)
{
*dest = 1;
break;
}
pa += astride;
pb += bstride;
}
dest += rxstride;
abase += xstride;
}
abase -= xstride * xcount;
bbase += ystride;
dest += rystride - (rxstride * xcount);
}
}
#endif