Commit 87f5a4b
committed
Rollup merge of #49231 - gnzlbg:fix_vec_fminmax, r=rkruppe
fix vector fmin/fmax non-fast/fast intrinsics NaN handling
This bugs shows up in release mode tests of `stdsimd`: rust-lang/stdarch#391 . The intrinsics are thoroughly tested there for roundoff errors, NaN, and overflow behavior.
The problem was that the non-fast intrinsics where specifying `NoNaNs == true`, which meant that they don't support NaNs. This is incorrect, the non-fast intrinsics should handle NaNs properly.
Also, the "fast" intrinsics where specifying `NoNaNs == false` which meant that they support NaNs and then fast-math, which probably disables this support. This was not intended either.
I've added a comment specifying what the boolean flags do.1 file changed
+4
-4
lines changed| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
1036 | 1036 | | |
1037 | 1037 | | |
1038 | 1038 | | |
1039 | | - | |
| 1039 | + | |
1040 | 1040 | | |
1041 | 1041 | | |
1042 | 1042 | | |
| |||
1046 | 1046 | | |
1047 | 1047 | | |
1048 | 1048 | | |
1049 | | - | |
| 1049 | + | |
1050 | 1050 | | |
1051 | 1051 | | |
1052 | 1052 | | |
| |||
1056 | 1056 | | |
1057 | 1057 | | |
1058 | 1058 | | |
1059 | | - | |
| 1059 | + | |
1060 | 1060 | | |
1061 | 1061 | | |
1062 | 1062 | | |
| |||
1067 | 1067 | | |
1068 | 1068 | | |
1069 | 1069 | | |
1070 | | - | |
| 1070 | + | |
1071 | 1071 | | |
1072 | 1072 | | |
1073 | 1073 | | |
| |||
0 commit comments