OpenE2K/gcc
2
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

tree-loop-linear.c (try_interchange_loops): Compare memory access strides against cache sizes.

Browse Source 
* tree-loop-linear.c (try_interchange_loops): Compare memory access
	strides against cache sizes.

	* testsuite/gcc.dg/tree-ssa/ltrans-8.c: Increase the size of strides
	to make the interchange profitable.

From-SVN: r132765
This commit is contained in:
Sebastian Pop 2008-02-29 12:41:14 +00:00 committed by Sebastian Pop
parent d973ac5a9f
commit 0c96231b2f
4 changed files with 43 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
gcc/ChangeLog
View File

@ -1,3 +1,8 @@
2008-02-29 Sebastian Pop <sebastian.pop@amd.com>
* tree-loop-linear.c (try_interchange_loops): Compare memory access
strides against cache sizes.
2008-02-29 Kaz Kojima <kkojima@gcc.gnu.org>
* config/sh/sh.c (sh_secondary_reload): Handle loading a float

5
gcc/testsuite/ChangeLog
View File

@ -1,3 +1,8 @@
2008-02-29 Sebastian Pop <sebastian.pop@amd.com>
* testsuite/gcc.dg/tree-ssa/ltrans-8.c: Increase the size of strides
to make the interchange profitable.
2008-02-28 Daniel Franke <franke.daniel@gmail.com>
PR fortran/31463

6
gcc/testsuite/gcc.dg/tree-ssa/ltrans-8.c
View File

@ -4,9 +4,9 @@ double foo(double *a)
{
int i,j;
double r = 0.0;
for (i=0; i<8; ++i)
for (j=0; j<8; ++j)
r += a[j*8+i];
for (i=0; i<100; ++i)
for (j=0; j<1000; ++j)
r += a[j*100+i];
return r;
}

33
gcc/tree-loop-linear.c
View File

@ -179,10 +179,14 @@ try_interchange_loops (lambda_trans_matrix trans,
VEC (data_reference_p, heap) *datarefs,
struct loop *first_loop)
{
bool res;
struct loop *loop_i;
struct loop *loop_j;
unsigned int dependence_steps_i, dependence_steps_j;
double_int access_strides_i, access_strides_j;
double_int small, large, nb_iter;
double_int l1_cache_size, l2_cache_size;
int cmp;
unsigned int nb_deps_not_carried_by_i, nb_deps_not_carried_by_j;
struct data_dependence_relation *ddr;
@ -194,7 +198,10 @@ try_interchange_loops (lambda_trans_matrix trans,
ddr = VEC_index (ddr_p, dependence_relations, 0);
if (ddr == NULL || DDR_ARE_DEPENDENT (ddr) == chrec_dont_know)
return trans;
l1_cache_size = uhwi_to_double_int (L1_CACHE_SIZE * 1024);
l2_cache_size = uhwi_to_double_int (L2_CACHE_SIZE * 1024);
/* LOOP_I is always the outer loop. */
for (loop_j = first_loop->inner;
loop_j;
@ -216,18 +223,38 @@ try_interchange_loops (lambda_trans_matrix trans,
/* Heuristics for loop interchange profitability:
0. Don't transform if the smallest stride is larger than
the L2 cache, or if the largest stride multiplied by the
number of iterations is smaller than the L1 cache.
1. (spatial locality) Inner loops should have smallest
dependence steps.
2. (spatial locality) Inner loops should contain more
dependence relations not carried by the loop.
3. (temporal locality) Inner loops should have smallest
3. (temporal locality) Inner loops should have smallest
array access strides.
*/
cmp = double_int_ucmp (access_strides_i, access_strides_j);
small = cmp < 0 ? access_strides_i : access_strides_j;
large = cmp < 0 ? access_strides_j : access_strides_i;
if (double_int_ucmp (small, l2_cache_size) > 0)
continue;
res = cmp < 0 ?
estimated_loop_iterations (loop_j, false, &nb_iter):
estimated_loop_iterations (loop_i, false, &nb_iter);
large = double_int_mul (large, nb_iter);
if (res && double_int_ucmp (large, l1_cache_size) < 0)
continue;
if (dependence_steps_i < dependence_steps_j
|| nb_deps_not_carried_by_i > nb_deps_not_carried_by_j
|| double_int_ucmp (access_strides_i, access_strides_j) < 0)
|| cmp < 0)
{
lambda_matrix_row_exchange (LTM_MATRIX (trans),
loop_depth (loop_i) - loop_depth (first_loop),