The architecture used as an example for the MODE_CC Condition Codes in the doc
is the SPARC, but it turns out the quoted macros are either wrong or outdated:
e.g. SELECT_CC_MODE mentions CCFPEmode and CCFPmode, but REVERSIBLE_CC_MODE
says that the only mode for FP inequality comparisons is CCFPEmode.
Fixed thusly, applied on all active branches.
2014-11-13 Eric Botcazou <ebotca...@adacore.com>
* doc/tm.texi.in (SELECT_CC_MODE): Update example.
(REVERSIBLE_CC_MODE): Fix example.
(REVERSE_CONDITION): Fix typo.
* doc/tm.texi: Regenerate.
--
Eric BotcazouIndex: doc/tm.texi
===================================================================
--- doc/tm.texi (revision 217407)
+++ doc/tm.texi (working copy)
@@ -5941,10 +5941,11 @@ for comparisons whose argument is a @cod
@smallexample
#define SELECT_CC_MODE(OP,X,Y) \
- (GET_MODE_CLASS (GET_MODE (X)) == MODE_FLOAT \
- ? ((OP == EQ || OP == NE) ? CCFPmode : CCFPEmode) \
- : ((GET_CODE (X) == PLUS || GET_CODE (X) == MINUS \
- || GET_CODE (X) == NEG) \
+ (GET_MODE_CLASS (GET_MODE (X)) == MODE_FLOAT \
+ ? ((OP == LT || OP == LE || OP == GT || OP == GE) \
+ ? CCFPEmode : CCFPmode) \
+ : ((GET_CODE (X) == PLUS || GET_CODE (X) == MINUS \
+ || GET_CODE (X) == NEG || GET_CODE (x) == ASHIFT) \
? CC_NOOVmode : CCmode))
@end smallexample
@@ -5987,10 +5988,11 @@ then @code{REVERSIBLE_CC_MODE (@var{mode
You need not define this macro if it would always returns zero or if the
floating-point format is anything other than @code{IEEE_FLOAT_FORMAT}.
For example, here is the definition used on the SPARC, where floating-point
-inequality comparisons are always given @code{CCFPEmode}:
+inequality comparisons are given either @code{CCFPEmode} or @code{CCFPmode}:
@smallexample
-#define REVERSIBLE_CC_MODE(MODE) ((MODE) != CCFPEmode)
+#define REVERSIBLE_CC_MODE(MODE) \
+ ((MODE) != CCFPEmode && (MODE) != CCFPmode)
@end smallexample
@end defmac
@@ -6000,7 +6002,7 @@ comparison done in CC_MODE @var{mode}.
@code{REVERSIBLE_CC_MODE (@var{mode})} is nonzero. Define this macro in case
machine has some non-standard way how to reverse certain conditionals. For
instance in case all floating point conditions are non-trapping, compiler may
-freely convert unordered compares to ordered one. Then definition may look
+freely convert unordered compares to ordered ones. Then definition may look
like:
@smallexample
Index: doc/tm.texi.in
===================================================================
--- doc/tm.texi.in (revision 217407)
+++ doc/tm.texi.in (working copy)
@@ -4441,10 +4441,11 @@ for comparisons whose argument is a @cod
@smallexample
#define SELECT_CC_MODE(OP,X,Y) \
- (GET_MODE_CLASS (GET_MODE (X)) == MODE_FLOAT \
- ? ((OP == EQ || OP == NE) ? CCFPmode : CCFPEmode) \
- : ((GET_CODE (X) == PLUS || GET_CODE (X) == MINUS \
- || GET_CODE (X) == NEG) \
+ (GET_MODE_CLASS (GET_MODE (X)) == MODE_FLOAT \
+ ? ((OP == LT || OP == LE || OP == GT || OP == GE) \
+ ? CCFPEmode : CCFPmode) \
+ : ((GET_CODE (X) == PLUS || GET_CODE (X) == MINUS \
+ || GET_CODE (X) == NEG || GET_CODE (x) == ASHIFT) \
? CC_NOOVmode : CCmode))
@end smallexample
@@ -4467,10 +4468,11 @@ then @code{REVERSIBLE_CC_MODE (@var{mode
You need not define this macro if it would always returns zero or if the
floating-point format is anything other than @code{IEEE_FLOAT_FORMAT}.
For example, here is the definition used on the SPARC, where floating-point
-inequality comparisons are always given @code{CCFPEmode}:
+inequality comparisons are given either @code{CCFPEmode} or @code{CCFPmode}:
@smallexample
-#define REVERSIBLE_CC_MODE(MODE) ((MODE) != CCFPEmode)
+#define REVERSIBLE_CC_MODE(MODE) \
+ ((MODE) != CCFPEmode && (MODE) != CCFPmode)
@end smallexample
@end defmac
@@ -4480,7 +4482,7 @@ comparison done in CC_MODE @var{mode}.
@code{REVERSIBLE_CC_MODE (@var{mode})} is nonzero. Define this macro in case
machine has some non-standard way how to reverse certain conditionals. For
instance in case all floating point conditions are non-trapping, compiler may
-freely convert unordered compares to ordered one. Then definition may look
+freely convert unordered compares to ordered ones. Then definition may look
like:
@smallexample
