diff options
| author | Brad Lucier <lucier@math.purdue.edu> | 2001-03-07 11:29:41 -0800 |
|---|---|---|
| committer | Richard Henderson <rth@gcc.gnu.org> | 2001-03-07 11:29:41 -0800 |
| commit | de6c59795844f185e1079034b9fa777fbc12ef5a (patch) | |
| tree | 996aa9dfa54be0d228dde6e7394f9adf66f7c76c /gcc/f | |
| parent | 6822468a694e59639c0f97975a2f48951501fd37 (diff) | |
| download | gcc-de6c59795844f185e1079034b9fa777fbc12ef5a.zip gcc-de6c59795844f185e1079034b9fa777fbc12ef5a.tar.gz gcc-de6c59795844f185e1079034b9fa777fbc12ef5a.tar.bz2 | |
Brad's -ffast-math breakup.
From-SVN: r40300
Diffstat (limited to 'gcc/f')
| -rw-r--r-- | gcc/f/ChangeLog | 11 | ||||
| -rw-r--r-- | gcc/f/g77.texi | 17 |
2 files changed, 25 insertions, 3 deletions
diff --git a/gcc/f/ChangeLog b/gcc/f/ChangeLog index 5da49e5..5dbe031 100644 --- a/gcc/f/ChangeLog +++ b/gcc/f/ChangeLog @@ -1,8 +1,13 @@ +2001-03-07 Brad Lucier <lucier@math.purdue.edu> + + * g77.texi: Document new options -funsafe-math-optimizations + and -fno-trapping-math. Revise documentation for -ffast-math. + 2001-03-01 Zack Weinberg <zackw@stanford.edu> - * f/proj.h: Delete 'bool' type. Don't include stddef.h here. - * f/com.c: Rename variables named 'true' and/or 'false'. - * f/intdoc.c: Delete 'bool' type. + * proj.h: Delete 'bool' type. Don't include stddef.h here. + * com.c: Rename variables named 'true' and/or 'false'. + * intdoc.c: Delete 'bool' type. 2001-03-01 Zack Weinberg <zackw@stanford.edu> diff --git a/gcc/f/g77.texi b/gcc/f/g77.texi index 93d1eab..5709f54 100644 --- a/gcc/f/g77.texi +++ b/gcc/f/g77.texi @@ -1446,6 +1446,7 @@ by type. Explanations are in the following sections. -malign-double -ffloat-store -fforce-mem -fforce-addr -fno-inline -ffast-math -fstrength-reduce -frerun-cse-after-loop +-funsafe-math-optimizations -fno-trapping-math -fexpensive-optimizations -fdelayed-branch -fschedule-insns -fschedule-insn2 -fcaller-saves -funroll-loops -funroll-all-loops @@ -2779,6 +2780,22 @@ Note that if you are not optimizing, no functions can be expanded inline. @cindex conformance, IEEE 754 Might allow some programs designed to not be too dependent on IEEE behavior for floating-point to run faster, or die trying. +Sets @samp{-funsafe-math-optimizations}, and +@samp{-fno-trapping-math}. + +@cindex -funsafe-math-optimizations option +@cindex options, -funsafe-math-optimizations +@item -funsafe-math-optimizations +Allow optimizations that may be give incorrect results +for certain IEEE inputs. + +@cindex -fno-trapping-math option +@cindex options, -fno-trapping-math +@item -fno-trapping-math +Allow the compiler to assume that floating-point arithmetic +will not generate traps on any inputs. This is useful, for +example, when running a program using IEEE "non-stop" +floating-point arithmetic. @cindex -fstrength-reduce option @cindex options, -fstrength-reduce |