diff options
| -rw-r--r-- | gcc/ChangeLog | 7 | ||||
| -rw-r--r-- | gcc/testsuite/ChangeLog | 5 | ||||
| -rw-r--r-- | gcc/testsuite/gcc.dg/vect/pr47001.c | 27 | ||||
| -rw-r--r-- | gcc/tree-vect-slp.c | 31 |
4 files changed, 69 insertions, 1 deletions
diff --git a/gcc/ChangeLog b/gcc/ChangeLog index bbf0545..bc44e7b 100644 --- a/gcc/ChangeLog +++ b/gcc/ChangeLog @@ -1,3 +1,10 @@ +2010-12-21 Ira Rosen <irar@il.ibm.com> + + PR tree-optimization/47001 + * tree-vect-slp.c (vect_supported_load_permutation_p): Check that + the loads in reduction are different and there are no gaps between + them. + 2010-12-21 Steven Bosscher <steven@gcc.gnu.org> PR middle-end/45310 diff --git a/gcc/testsuite/ChangeLog b/gcc/testsuite/ChangeLog index 007baf4..b03b602 100644 --- a/gcc/testsuite/ChangeLog +++ b/gcc/testsuite/ChangeLog @@ -1,3 +1,8 @@ +2010-12-21 Ira Rosen <irar@il.ibm.com>i + + PR tree-optimization/47001 + * gcc.dg/vect/pr47001.c: New. + 2010-12-21 Nicola Pero <nicola.pero@meta-innovation.com> PR libobjc/16110 diff --git a/gcc/testsuite/gcc.dg/vect/pr47001.c b/gcc/testsuite/gcc.dg/vect/pr47001.c new file mode 100644 index 0000000..9c5d08d --- /dev/null +++ b/gcc/testsuite/gcc.dg/vect/pr47001.c @@ -0,0 +1,27 @@ +/* { dg-do compile } */ + +#include <stdlib.h> + +#define N 128 + +int a[N]; + +int main1 (int res0, int res1) +{ + int i; + int sum0 = 0, sum1 = 0; + + for (i = 0; i < N/2; i++) { + sum1 += a[2*i]; + sum0 += a[2*i]; + } + + /* Check results: */ + if (sum0 != res0 + || sum1 != res1) + abort (); + + return 0; +} + +/* { dg-final { cleanup-tree-dump "vect" } } */ diff --git a/gcc/tree-vect-slp.c b/gcc/tree-vect-slp.c index fdaaff2..4f7c72c 100644 --- a/gcc/tree-vect-slp.c +++ b/gcc/tree-vect-slp.c @@ -1002,7 +1002,36 @@ vect_supported_load_permutation_p (slp_instance slp_instn, int group_size, if (!bad_permutation) { - /* This permutaion is valid for reduction. Since the order of the + /* Check that the loads in the first sequence are different and there + are no gaps between them. */ + load_index = sbitmap_alloc (group_size); + sbitmap_zero (load_index); + for (k = 0; k < group_size; k++) + { + first_group_load_index = VEC_index (int, load_permutation, k); + if (TEST_BIT (load_index, first_group_load_index)) + { + bad_permutation = true; + break; + } + + SET_BIT (load_index, first_group_load_index); + } + + if (!bad_permutation) + for (k = 0; k < group_size; k++) + if (!TEST_BIT (load_index, k)) + { + bad_permutation = true; + break; + } + + sbitmap_free (load_index); + } + + if (!bad_permutation) + { + /* This permutation is valid for reduction. Since the order of the statements in the nodes is not important unless they are memory accesses, we can rearrange the statements in all the nodes according to the order of the loads. */ |