Current wording may cause people to mistakely believe "hey I use
-ftrapping-math so GCC should behave as-if it supported FENV_ACCESS and
I enabled it" and add more duplicates of PR 34678. Let's make it clear
such an expectation isn't working.
gcc/
PR middle-end/34678
* doc/invoke.texi (-fno-trapping-math): Note that the
-ftrapping-math effect is weaker than enabling FENV_ACCESS.
---
gcc/doc/invoke.texi | 8 +++++++-
1 file changed, 7 insertions(+), 1 deletion(-)
diff --git a/gcc/doc/invoke.texi b/gcc/doc/invoke.texi
index b1e6d703702..1f0988cfc6d 100644
--- a/gcc/doc/invoke.texi
+++ b/gcc/doc/invoke.texi
@@ -16129,7 +16129,13 @@ The default is @option{-ftrapping-math}.
Future versions of GCC may provide finer control of this setting
using C99's @code{FENV_ACCESS} pragma. This command-line option
will be used along with @option{-frounding-math} to specify the
-default state for @code{FENV_ACCESS}.
+default state for @code{FENV_ACCESS}. But note that as at now GCC
+does not support the @code{FENV_ACCESS} pragma yet, and the effect of
+@option{-ftrapping-math} is less strict than enabling the
+@code{FENV_ACCESS} pragma (on a compiler supporting it): GCC can still
+omit a floating-point exception if the result of the floating-point
+calculation can be determined at the compile time, even with
+@option{-ftrapping-math}.
@opindex frounding-math
@item -frounding-math
--
2.52.0
