| Message ID | BF230D13CA30DD48930C31D4099330003A4391D4@FMSMSX101.amr.corp.intel.com |
|---|---|
| State | New |
| Headers | show |
Index: gcc/testsuite/c-c++-common/cilk-plus/AN/sec_reduce_ind_same_value.c =================================================================== --- gcc/testsuite/c-c++-common/cilk-plus/AN/sec_reduce_ind_same_value.c (revision 0) +++ gcc/testsuite/c-c++-common/cilk-plus/AN/sec_reduce_ind_same_value.c (revision 0) @@ -0,0 +1,21 @@ +/* { dg-do run } */ +/* { dg-options "-fcilkplus" } */ + +int A[256]; + +int main () { + A[:] = 2; + int max_index = 0, min_index = 0; + + max_index = __sec_reduce_max_ind (A[:]); + + if (max_index != 255) + return 1; + + min_index = __sec_reduce_min_ind (A[:]); + if (min_index != 255) + return 2; + + return 0; +} +
Hello Everyone, I am adding a new execution test to the array notation suite. In array notation's __sec_reduce_max_ind and __sec_reduce_min_ind builtin functions, if there is a tie for max/min value, then the higher index should be returned. In this case, all the values are the same, so it should return the highest index of the array (i.e. size-1). I am going to check this in as obvious. I am willing to revert this patch (or fix it) if anyone has any objections. Here is the ChangeLog for it. 2013-06-17 Balaji V. Iyer <balaji.v.iyer@intel.com> * c-c++-common/cilk-plus/AN/sec_reduce_ind_same_value.c: New test. Here is the patch: Thanks, Balaji V. Iyer.