diff options
author | Joseph Myers <jsm28@cam.ac.uk> | 2001-03-09 21:57:48 +0000 |
---|---|---|
committer | Joseph Myers <jsm28@gcc.gnu.org> | 2001-03-09 21:57:48 +0000 |
commit | 6ca3ad1bc5a51dce85970d6fa66167c98d7bfa47 (patch) | |
tree | 06926d75a9922ae7bb6ef902467b4f4ed2267b14 /gcc | |
parent | 11642c3a7cf7c6954afa4a174632535a46cf7157 (diff) | |
download | gcc-6ca3ad1bc5a51dce85970d6fa66167c98d7bfa47.zip gcc-6ca3ad1bc5a51dce85970d6fa66167c98d7bfa47.tar.gz gcc-6ca3ad1bc5a51dce85970d6fa66167c98d7bfa47.tar.bz2 |
invoke.texi: Fix typo.
* invoke.texi: Fix typo.
* gcc.1: Regenerate.
From-SVN: r40351
Diffstat (limited to 'gcc')
-rw-r--r-- | gcc/ChangeLog | 5 | ||||
-rw-r--r-- | gcc/gcc.1 | 57 | ||||
-rw-r--r-- | gcc/invoke.texi | 2 |
3 files changed, 51 insertions, 13 deletions
diff --git a/gcc/ChangeLog b/gcc/ChangeLog index 1528738..5abe6f6 100644 --- a/gcc/ChangeLog +++ b/gcc/ChangeLog @@ -1,3 +1,8 @@ +2001-03-09 Joseph S. Myers <jsm28@cam.ac.uk> + + * invoke.texi: Fix typo. + * gcc.1: Regenerate. + 2001-03-09 Zack Weinberg <zackw@stanford.edu> * configure.in: Prune nonexistent files from build_xm_file, @@ -1,5 +1,5 @@ .\" Automatically generated by Pod::Man version 1.1 -.\" Mon Feb 19 19:32:03 2001 +.\" Fri Mar 9 21:54:51 2001 .\" .\" Standard preamble: .\" ====================================================================== @@ -138,7 +138,7 @@ .\" ====================================================================== .\" .IX Title "GCC 1" -.TH GCC 1 "gcc-3.1" "2001-02-19" "GNU" +.TH GCC 1 "gcc-3.1" "2001-03-09" "GNU" .UC .SH "NAME" gcc \- \s-1GNU\s0 project C and \*(C+ compiler @@ -280,8 +280,9 @@ in the following sections. \&\-finline-functions \-finline-limit=\fR\fIn\fR \fB\-fkeep-inline-functions \&\-fkeep-static-consts \-fmove-all-movables \&\-fno-default-inline \-fno-defer-pop -\&\-fno-function-cse \-fno-guess-branch-probability +\&\-fno-function-cse \-fno-guess-branch-probability \&\-fno-inline \-fno-math-errno \-fno-peephole +\&\-funsafe-math-optimizations \-fno-trapping-math \&\-fomit-frame-pointer \-foptimize-register-move \&\-foptimize-sibling-calls \-freduce-all-givs \&\-fregmove \-frename-registers @@ -2819,11 +2820,12 @@ that alter the assembler output may be confused by the optimizations performed when this option is not used. .Ip "\fB\-ffast-math\fR" 4 .IX Item "-ffast-math" -This option allows \s-1GCC\s0 to violate some \s-1ISO\s0 or \s-1IEEE\s0 rules and/or -specifications in the interest of optimizing code for speed. For -example, it allows the compiler to assume arguments to the \f(CW\*(C`sqrt\*(C'\fR -function are non-negative numbers and that no floating-point values -are NaNs. +Sets \fB\-fno-math-errno\fR, \fB\-funsafe-math-optimizations\fR, +and \fB\-fno-trapping-math\fR. +.Sp +This option causes the preprocessor macro _\|_FAST_MATH_\|_ to be defined. +.Sp +This option causes the preprocessor macro _\|_FAST_MATH_\|_ to be defined. .Sp This option should never be turned on by any \fB\-O\fR option since it can result in incorrect output for programs which depend on @@ -2836,8 +2838,39 @@ with a single instruction, e.g., sqrt. A program that relies on \&\s-1IEEE\s0 exceptions for math error handling may want to use this flag for speed while maintaining \s-1IEEE\s0 arithmetic compatibility. .Sp +This option should never be turned on by any \fB\-O\fR option since +it can result in incorrect output for programs which depend on +an exact implementation of \s-1IEEE\s0 or \s-1ISO\s0 rules/specifications for +math functions. +.Sp The default is \fB\-fmath-errno\fR. The \fB\-ffast-math\fR option sets \fB\-fno-math-errno\fR. +.Ip "\fB\-funsafe-math-optimizations\fR" 4 +.IX Item "-funsafe-math-optimizations" +Allow optimizations for floating-point arithmetic that (a) assume +that arguments and results are valid and (b) may violate \s-1IEEE\s0 or +\&\s-1ANSI\s0 standards. +.Sp +This option should never be turned on by any \fB\-O\fR option since +it can result in incorrect output for programs which depend on +an exact implementation of \s-1IEEE\s0 or \s-1ISO\s0 rules/specifications for +math functions. +.Sp +The default is \fB\-fno-unsafe-math-optimizations\fR. The +\&\fB\-ffast-math\fR option sets \fB\-funsafe-math-optimizations\fR. +.Ip "\fB\-fno-trapping-math\fR" 4 +.IX Item "-fno-trapping-math" +Compile code assuming that floating-point operations cannot generate +user-visible traps. Setting this option may allow faster code +if one relies on ``non-stop'' \s-1IEEE\s0 arithmetic, for example. +.Sp +This option should never be turned on by any \fB\-O\fR option since +it can result in incorrect output for programs which depend on +an exact implementation of \s-1IEEE\s0 or \s-1ISO\s0 rules/specifications for +math functions. +.Sp +The default is \fB\-ftrapping-math\fR. The \fB\-ffast-math\fR +option sets \fB\-fno-trapping-math\fR. .PP The following options control specific optimizations. The \fB\-O2\fR option turns on all of these optimizations except \fB\-funroll-loops\fR @@ -5851,14 +5884,14 @@ These \fB\-m\fR options are defined for the i386 family of computers: Assume the defaults for the machine type \fIcpu type\fR when scheduling instructions. The choices for \fIcpu type\fR are \fBi386\fR, \&\fBi486\fR, \fBi586\fR, \fBi686\fR, \fBpentium\fR, -\&\fBpentiumpro\fR, \fBk6\fR, and \fBathlon\fR +\&\fBpentiumpro\fR, \fBpentium4\fR, \fBk6\fR, and \fBathlon\fR .Sp While picking a specific \fIcpu type\fR will schedule things appropriately for that particular chip, the compiler will not generate any code that does not run on the i386 without the \fB\-march=\fR\fIcpu type\fR option being used. \fBi586\fR is equivalent to \fBpentium\fR and \fBi686\fR -is equivalent to \fBpentiumpro\fR. \fBk6\fR is the \s-1AMD\s0 chip as -opposed to the Intel ones. +is equivalent to \fBpentiumpro\fR. \fBk6\fR and \fBathlon\fR are the +\&\s-1AMD\s0 chips as opposed to the Intel ones. .Ip "\fB\-march=\fR\fIcpu type\fR" 4 .IX Item "-march=cpu type" Generate instructions for the machine type \fIcpu type\fR. The choices @@ -5917,7 +5950,7 @@ Some 387 emulators do not support the \f(CW\*(C`sin\*(C'\fR, \f(CW\*(C`cos\*(C'\ \&\f(CW\*(C`sqrt\*(C'\fR instructions for the 387. Specify this option to avoid generating those instructions. This option is the default on FreeBSD. As of revision 2.6.1, these instructions are not generated unless you -also use the \fB\-ffast-math\fR switch. +also use the \fB\-funsafe-math-optimizations\fR switch. .Ip "\fB\-malign-double\fR" 4 .IX Item "-malign-double" .PD 0 diff --git a/gcc/invoke.texi b/gcc/invoke.texi index 0f3be5b..519dd92 100644 --- a/gcc/invoke.texi +++ b/gcc/invoke.texi @@ -6573,7 +6573,7 @@ While picking a specific @var{cpu type} will schedule things appropriately for that particular chip, the compiler will not generate any code that does not run on the i386 without the @samp{-march=@var{cpu type}} option being used. @samp{i586} is equivalent to @samp{pentium} and @samp{i686} -is equivalent to @samp{pentiumpro}. @samp{k6} and @samp{athlon} sre the +is equivalent to @samp{pentiumpro}. @samp{k6} and @samp{athlon} are the AMD chips as opposed to the Intel ones. @item -march=@var{cpu type} |