diff options
| author | Jakub Jelinek <jakub@redhat.com> | 2020-12-16 18:21:32 +0100 |
|---|---|---|
| committer | Jakub Jelinek <jakub@redhat.com> | 2020-12-16 18:21:32 +0100 |
| commit | 652702b0c93644a44ad5bdfd4ff260d0f69a084b (patch) | |
| tree | 6a766b7a48797950e8b09b0e68089015714311ba /gcc/config.gcc | |
| parent | 5098d35fb1997aa6fd6d7e37d0fd4501a5fe2f9d (diff) | |
| download | gcc-652702b0c93644a44ad5bdfd4ff260d0f69a084b.zip gcc-652702b0c93644a44ad5bdfd4ff260d0f69a084b.tar.gz gcc-652702b0c93644a44ad5bdfd4ff260d0f69a084b.tar.bz2 | |
libcody: fix --enable-checking=... follow-up [PR98311]
> The -enable-checking configure code in libcody didn't play well with
> us. This just uses libcpp's configurey for that piece.
This doesn't set is_release anywhere, which means when --enable-checking*
or --disable-checking isn't specified, it always treats it as
--enable-checking=yes, while the normal gcc behavior is treat only trunk
as --enable-checking=yes and treat release branches as
--enable-checking=release by default.
On the other side, nothing uses those ac_assert_checking and
ac_valgrind_checking variables, so it is a waste to compute those.
2020-12-16 Jakub Jelinek <jakub@redhat.com>
* configure.ac: Compute is_release.
(NMS_ENABLE_CHECKING): Simplify but not computing ac_assert_checking
and ac_valgrind_checking the code doesn't use.
* configure: Regenerated.
Diffstat (limited to 'gcc/config.gcc')
0 files changed, 0 insertions, 0 deletions