diff options
| author | Jörg Frings-Fürst <debian@jff.email> | 2024-10-20 15:21:58 +0200 |
|---|---|---|
| committer | Jörg Frings-Fürst <debian@jff.email> | 2024-10-20 15:21:58 +0200 |
| commit | 630f99f29bd31a76d8d24da2975a045452c763ef (patch) | |
| tree | 0c801f68561bfb0930a4ade80d7ca3a7940887ab /lib/signbitf.c | |
| parent | 84e26c587987e8484d55db4165f188b40c09e94b (diff) | |
| parent | 4682deeb62247d34de87f8e777f99e2d337fd377 (diff) | |
Update upstream source from tag 'upstream/1.3'
Update to upstream version '1.3'
with Debian dir 8a8e4828ddf646ece6c109b401e08d162be35936
Diffstat (limited to 'lib/signbitf.c')
| -rw-r--r-- | lib/signbitf.c | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/lib/signbitf.c b/lib/signbitf.c index f24caf4a..f4ad9567 100644 --- a/lib/signbitf.c +++ b/lib/signbitf.c @@ -32,7 +32,7 @@ gl_signbitf (float arg) { #if defined FLT_SIGNBIT_WORD && defined FLT_SIGNBIT_BIT /* The use of a union to extract the bits of the representation of a - 'long double' is safe in practice, despite of the "aliasing rules" of + 'float' is safe in practice, despite of the "aliasing rules" of C99, because the GCC docs say "Even with '-fstrict-aliasing', type-punning is allowed, provided the memory is accessed through the union type." |