5626N/A--- gcc-3.4.3/gcc/doc/cppopts.texi.orig Sat Jan 30 17:09:08 2016

5626N/A+++ gcc-3.4.3/gcc/doc/cppopts.texi Sat Jan 30 17:09:28 2016

5626N/A@@ -671,7 +671,7 @@

5626N/A Enable special code to work around file systems which only permit very

5626N/A short file names, such as MS-DOS@.

5626N/A

5626N/A-@itemx --help

5626N/A+@item --help

5626N/A @itemx --target-help

5626N/A @opindex help

5626N/A @opindex target-help

5626N/A--- gcc-3.4.3/gcc/doc/invoke.texi.orig Sat Jan 30 17:13:18 2016

5626N/A+++ gcc-3.4.3/gcc/doc/invoke.texi Sat Jan 30 17:14:09 2016

5626N/A@@ -787,7 +787,7 @@

5626N/A generic, or subprogram renaming declaration). Such files are also

5626N/A called @dfn{specs}.

5626N/A

5626N/A-@itemx @var{file}.adb

5626N/A+@item @var{file}.adb

5626N/A Ada source code file containing a library unit body (a subprogram or

5626N/A package body). Such files are also called @dfn{bodies}.

5626N/A

5626N/A@@ -6658,7 +6658,7 @@

5626N/A @samp{arm1136js}, @samp{arm1136jfs} ,@samp{xscale}, @samp{iwmmxt},

5626N/A @samp{ep9312}.

5626N/A

5626N/A-@itemx -mtune=@var{name}

5626N/A+@item -mtune=@var{name}

5626N/A @opindex mtune

5626N/A This option is very similar to the @option{-mcpu=} option, except that

5626N/A instead of specifying the actual target processor type, and hence

5626N/A@@ -7968,7 +7968,7 @@

5626N/A Assume that the floating-point coprocessor only supports single-precision

5626N/A operations.

5626N/A

5626N/A-@itemx -mdouble-float

5626N/A+@item -mdouble-float

5626N/A @opindex mdouble-float

5626N/A Assume that the floating-point coprocessor supports double-precision

5626N/A operations. This is the default.

5626N/A--- gcc-3.4.3/gcc/doc/extend.texi.orig Sat Jan 30 17:14:48 2016

5626N/A+++ gcc-3.4.3/gcc/doc/extend.texi Sat Jan 30 17:20:30 2016

5626N/A@@ -3709,6 +3709,8 @@

5626N/A @option{-fstrict-aliasing}, which is on by default at @option{-O2} or

5626N/A above in recent GCC versions.

5626N/A

5626N/A+@end table

5626N/A+

5626N/A @subsection i386 Type Attributes

5626N/A

5626N/A Two attributes are currently defined for i386 configurations:

5626N/A--- gcc-3.4.3/gcc/doc/extend.texi.orig Sat Jan 30 17:24:18 2016

5626N/A+++ gcc-3.4.3/gcc/doc/extend.texi Sat Jan 30 17:25:08 2016

5626N/A@@ -3716,6 +3716,7 @@

5626N/A Two attributes are currently defined for i386 configurations:

5626N/A @code{ms_struct} and @code{gcc_struct}

5626N/A

5626N/A+@table @code

5626N/A @item ms_struct

5626N/A @itemx gcc_struct

5626N/A @cindex @code{ms_struct}

5626N/A--- gcc-3.4.3/gcc/doc/c-tree.texi.orig Sat Jan 30 17:29:36 2016

5626N/A+++ gcc-3.4.3/gcc/doc/c-tree.texi Sat Jan 30 17:29:47 2016

5626N/A@@ -2065,7 +2065,7 @@

5626N/A generate these expressions anyhow, if it can tell that strictness does

5626N/A not matter.

5626N/A

5626N/A-@itemx PLUS_EXPR

5626N/A+@item PLUS_EXPR

5626N/A @itemx MINUS_EXPR

5626N/A @itemx MULT_EXPR

5626N/A @itemx TRUNC_DIV_EXPR

5626N/A--- gcc-3.4.3/gcc/doc/md.texi.orig Sat Jan 30 17:30:12 2016

5626N/A+++ gcc-3.4.3/gcc/doc/md.texi Sat Jan 30 17:30:21 2016

5626N/A@@ -2539,7 +2539,7 @@

5626N/A Similar, for other arithmetic operations.

5626N/A @cindex @code{min@var{m}3} instruction pattern

5626N/A @cindex @code{max@var{m}3} instruction pattern

5626N/A-@itemx @samp{min@var{m}3}, @samp{max@var{m}3}

5626N/A+@item @samp{min@var{m}3}, @samp{max@var{m}3}

5626N/A Floating point min and max operations. If both operands are zeros,

5626N/A or if either operand is NaN, then it is unspecified which of the two

5626N/A operands is returned as the result.