diff options
| author | vlefevre <vlefevre@280ebfd0-de03-0410-8827-d642c229c3f4> | 2019-10-01 13:28:37 +0000 |
|---|---|---|
| committer | vlefevre <vlefevre@280ebfd0-de03-0410-8827-d642c229c3f4> | 2019-10-01 13:28:37 +0000 |
| commit | e2cb0fed0e2331e82a6861b19227ac3f754332ff (patch) | |
| tree | f56514585f311cf71e2b6eed8a3ad6458d36ba6c /acinclude.m4 | |
| parent | 5a33f50afa9f09897e5bb6c7d75cf443d7de3113 (diff) | |
| download | mpfr-e2cb0fed0e2331e82a6861b19227ac3f754332ff.tar.gz | |
[acinclude.m4] Updated comment about the tests for subnormal numbers.
git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@13625 280ebfd0-de03-0410-8827-d642c229c3f4
Diffstat (limited to 'acinclude.m4')
| -rw-r--r-- | acinclude.m4 | 4 |
1 files changed, 3 insertions, 1 deletions
diff --git a/acinclude.m4 b/acinclude.m4 index 05a31c1ba..755dd0604 100644 --- a/acinclude.m4 +++ b/acinclude.m4 @@ -304,7 +304,9 @@ dnl for the binary32 format, the smallest normal number is 2^(-126) dnl Note: One could double-check with the value of the macros dnl DBL_HAS_SUBNORM and FLT_HAS_SUBNORM, when defined (C2x), but dnl neither tests would be reliable on implementations with partial -dnl subnormal support. +dnl subnormal support. Anyway, this check is useful only for the +dnl tests. Thus in doubt, assume that subnormals are not supported, +dnl in order to disable the corresponding tests (which could fail). AC_CACHE_CHECK([for subnormal double-precision numbers], mpfr_cv_have_denorms, [ AC_RUN_IFELSE([AC_LANG_SOURCE([[ |