diff options
author | Jonathan Wakely <jwakely@redhat.com> | 2025-02-25 18:06:46 +0000 |
---|---|---|
committer | Jonathan Wakely <redi@gcc.gnu.org> | 2025-02-25 22:34:16 +0000 |
commit | 2256e30874af2ef804bb19d2eba40f9c92953beb (patch) | |
tree | 29a3a68d0559588740cd99664964d2d42674365d /gcc/fortran/trans-expr.cc | |
parent | 892ee5ffba0760794a932e36771863a86ef2b271 (diff) | |
download | gcc-2256e30874af2ef804bb19d2eba40f9c92953beb.zip gcc-2256e30874af2ef804bb19d2eba40f9c92953beb.tar.gz gcc-2256e30874af2ef804bb19d2eba40f9c92953beb.tar.bz2 |
libstdc++: Fix typo in std::fill SFINAE constraint [PR93059]
The r15-4321-gd8ef4471cb9c9f change incorrectly used __value as the
member of the __memcpyable_integer trait, but it should have been
__width. That meant this overload was not being used for _Tp != _Up.
Also return after doing the loop for the consteval case. The missing
return wasn't causing incorrect behaviour because the consteval loop
increments the iterator until it equals the end of the range, so the
memset isn't done. But it's still better to return and not even try
to do the memset.
libstdc++-v3/ChangeLog:
PR libstdc++/93059
* include/bits/stl_algobase.h (__fill_a1): Fix typo in SFINAE
constraint.
Diffstat (limited to 'gcc/fortran/trans-expr.cc')
0 files changed, 0 insertions, 0 deletions