Re: c/5111: Misleading error message for "unsigned long double"

Re: c/5111: Misleading error message for "unsigned long double"

[bangerth]

Synopsis: Misleading error message for "unsigned long double"

State-Changed-From-To: open->closed
State-Changed-By: bangerth
State-Changed-When: Tue Nov  5 07:57:45 2002
State-Changed-Why:
    I think the warning is actually ok:
       unsigned short double fred;
    is recognized as "unsigned short" plus "double", while
       unsigned long double fred;
    is recognized as "unsigned" plus "long double", which makes
    all the difference in the messages. I think, the message
    as it is is actually sufficiently helpful to find this
    kind of error :-)

http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&database=gcc&pr=5111

c/5111: Misleading error message for "unsigned long double"

[brian]

>Number:         5111
>Category:       c
>Synopsis:       Misleading error message for "unsigned long double"
>Confidential:   no
>Severity:       non-critical
>Priority:       low
>Responsible:    unassigned
>State:          open
>Class:          doc-bug
>Submitter-Id:   net
>Arrival-Date:   Thu Dec 13 21:36:00 PST 2001
>Closed-Date:
>Last-Modified:
>Originator:     Brian Herlihy
>Release:        3.0.3 20011209 (Debian prerelease) (Debian testing/unstable)
>Organization:
Me
>Environment:
System: Linux btherl 2.4.16 #5 Mon Dec 3 10:34:12 EST 2001 i686 unknown
Architecture: i686

	
host: i386-pc-linux-gnu
build: i386-pc-linux-gnu
target: i386-pc-linux-gnu
configured with: ../src/configure -v --enable-languages=c,c++,java,f77,proto,objc --prefix=/usr --infodir=/share/info --mandir=/share/man --enable-shared --with-gnu-as --with-gnu-ld --with-system-zlib --enable-long-long --enable-nls --without-included-gettext --disable-checking --enable-threads=posix --enable-java-gc=boehm --with-cpp-install-dir=bin --enable-objc-gc i386-linux
>Description:
	Error message is misleading for illegal definition.  long doubles are
	allowed, although the error message implies that they are not.
>How-To-Repeat:
	cat > t.c
	unsigned long double fred;
	main () {}
	^D
	gcc-3.0 -o t t.c
>Fix:
	vi <file_with_error_message>  # :)
	The error message is:
	t.c:1: long, short, signed or unsigned invalid for `fred'
	which implies that all these qualifiers are invalid.  In fact,
	a long double is valid.  The error message for "unsigned short double"
	gives this information; it would be nice to have this message appear
	for "unsigned long double" as well.
	Believe it or not I found this because of code which had an unsigned
	long double in it that compiled in windows..
>Release-Note:
>Audit-Trail:
>Unformatted: