diff options
author | vlefevre <vlefevre@280ebfd0-de03-0410-8827-d642c229c3f4> | 2007-05-30 11:21:32 +0000 |
---|---|---|
committer | vlefevre <vlefevre@280ebfd0-de03-0410-8827-d642c229c3f4> | 2007-05-30 11:21:32 +0000 |
commit | 37b5763e2ac0aa84f7efdeb8799c93c4d9a7b2cf (patch) | |
tree | 84fd3979e92258e4f5b10f0bc1379bbdb3edf388 /TODO | |
parent | 42674a9e96dd0400a4de37051968e580aa655992 (diff) | |
download | mpfr-37b5763e2ac0aa84f7efdeb8799c93c4d9a7b2cf.tar.gz |
TODO: added an item about the behavior of mpfr_sgn on NaN.
git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@4510 280ebfd0-de03-0410-8827-d642c229c3f4
Diffstat (limited to 'TODO')
-rw-r--r-- | TODO | 7 |
1 files changed, 7 insertions, 0 deletions
@@ -41,6 +41,13 @@ Installation: Changes in existing functions: ############################################################################## +- TODO before the 2.3.0 release: decide whether the new behavior + of mpfr_sgn on NaN should be documented. This new behavior is + compatible with the old one since the old behavior on NaN was + undefined (with the same meaning as in the C standard). The + rationale is that it is better to avoid undefined behavior when + one can replace it by something that makes sense. + - many functions currently taking into account the precision of the *input* variable to set the initial working precison (acosh, asinh, cosh, ...). This is nonsense since the "average" working precision should only depend |