diff options
| author | Jason Merrill <jason@redhat.com> | 2025-01-19 05:15:01 -0500 |
|---|---|---|
| committer | Jason Merrill <jason@redhat.com> | 2025-04-17 13:20:39 -0400 |
| commit | 53d4e355db18fec21515d055924df8290ef5ce14 (patch) | |
| tree | 8befbd3e48fafb3597858e79821ba0ffe0c95248 /libjava/gnu/java | |
| parent | 20e31f507a2dd6cbd40cc65a7c06d07bfaa2a5e1 (diff) | |
| download | gcc-53d4e355db18fec21515d055924df8290ef5ce14.zip gcc-53d4e355db18fec21515d055924df8290ef5ce14.tar.gz gcc-53d4e355db18fec21515d055924df8290ef5ce14.tar.bz2 | |
c++: vec_safe_reserve usage tweaks
A couple of cleanups from noticing that the semantics of
std::vector<T>::reserve() (request the new minimum allocation) differ from
the GCC vec<...>::reserve() (request a minimum number of slots available).
In preserve_state, we were tripling the size of the vec when doubling it is
more than enough.
In get_tinfo_desc we were using vec_safe_reserve properly, but it's
simpler to use vec_safe_grow_cleared.
gcc/cp/ChangeLog:
* name-lookup.cc (name_lookup::preserve_state): Fix reserve call.
* rtti.cc (get_tinfo_desc): Use vec_safe_grow_cleared.
Diffstat (limited to 'libjava/gnu/java')
0 files changed, 0 insertions, 0 deletions