8000 DOC: Add release note about clang floating point error flag change · numpy/numpy@a303b33 · GitHub
[go: up one dir, main page]
Skip to content

Commit a303b33

Browse files
sebergseberg
committed
DOC: Add release note about clang floating point error flag change
1 parent 817684a commit a303b33

File tree

1 file changed

+6
-0
lines changed

1 file changed

+6
-0
lines changed

‎doc/release/upcoming_changes/19049.compatibility.rst

Lines changed: 6 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,6 @@
1+
Distutils forces strict floating point model on clang
2+
-----------------------------------------------------
3+
NumPy distutils will now always add the ``-ffp-exception-behavior=strict``
4+
compiler flag when compiling with clang. Clang defaults to a non-strict
5+
version, which allows the compiler to generate code that does not set
6+
floating point warnings/errors correctly.

0 commit comments

Comments
 (0)
0