[Bug c++/11922] New: ICE on type_unification_real

[Bug c++/11922] New: ICE on type_unification_real

[gcc-bugzilla at x-infinity dot com]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11922

           Summary: ICE on type_unification_real
           Product: gcc
           Version: 3.3.1
            Status: UNCONFIRMED
          Severity: normal
          Priority: P2
         Component: c++
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: gcc-bugzilla at x-infinity dot com
                CC: gcc-bugs at gcc dot gnu dot org

The following output illustrates the ICE.
I'm not sure if this code should compile or is even syntactically correct, but 
ICE is still there.

xm@xm-debian:~/D/Projects/QSX/QSX/Source$ g++ -v -save-temps -pthread 
test.cpp -O3 -g2 -I ../Include -DNDEBUG ../Lib/release-
sse/QSX_Core.a ../Lib/release-sse/QSX_Math.a ../Lib/release-
sse/QSX_OS.a ../Lib/release-sse/QSX_Graphics.a ../Lib/release-sse/QSX_IO.a
Reading specs from /usr/lib/gcc-lib/i386-linux/3.3.1/specs
Configured with: ../src/configure -v --enable-
languages=c,c++,java,f77,pascal,objc,ada,treelang --prefix=/usr --
mandir=/usr/share/man --infodir=/usr/share/info --with-gxx-include-
dir=/usr/include/c++/3.3 --enable-shared --with-system-zlib --enable-nls --
without-included-gettext --enable-__cxa_atexit --enable-clocale=gnu --enable-
debug --enable-java-gc=boehm --enable-java-awt=xlib --enable-objc-gc i386-linux
Thread model: posix
gcc version 3.3.1 (Debian)
 /usr/lib/gcc-lib/i386-linux/3.3.1/cc1plus -E -D__GNUG__=3 -quiet -v -
I ../Include -D__GNUC__=3 -D__GNUC_MINOR__=3 -D__GNUC_PATCHLEVEL__=1 -
D_GNU_SOURCE -D_REENTRANT -DNDEBUG test.cpp -O3 test.ii
ignoring nonexistent directory "/usr/i386-linux/include"
#include "..." search starts here:
#include <...> search starts here:
 ../Include
 /usr/include/c++/3.3
 /usr/include/c++/3.3/i386-linux
 /usr/include/c++/3.3/backward
 /usr/local/include
 /usr/lib/gcc-lib/i386-linux/3.3.1/include
 /usr/include
End of search list.
 /usr/lib/gcc-lib/i386-linux/3.3.1/cc1plus -fpreprocessed test.ii -quiet -
dumpbase test.cpp -auxbase test -g2 -O3 -version -o test.s
GNU C++ version 3.3.1 (Debian) (i386-linux)
        compiled by GNU C version 3.3.1 (Debian).
GGC heuristics: --param ggc-min-expand=38 --param ggc-min-heapsize=15737
../Include/QSX_Graphics/ChannelFormatConverter.h: In static member function `
   static typename TraitsType::value_type
   QSX::Graphics::ChannelFormatConverter<TraitsType>::from(typename
   OtherTraitsType::value_type) [with OtherTraitsType =
   QSX::Graphics::ChannelTraits<FORMAT_BYTE>, TraitsType =
   QSX::Graphics::ChannelTraits<FORMAT_BYTE>]':
../Include/QSX_Graphics/pixel_converter_helper.h:51:   instantiated from `void 
QSX::Graphics::pixel_converter_helper<TargetPixelTraitsType, 
SourcePixelTraitsType, channel_offset>::operator()
(QSX::Graphics::Pixel<TargetPixelTraitsType>&, const 
QSX::Graphics::Pixel<SourcePixelTraitsType>&) [with TargetPixelTraitsType = 
QSX::Graphics::PixelTraits<QSX::Graphics::ChannelTraits<FORMAT_BYTE>, 
QSX::Graphics::PixelDescriptor_BGR>, SourcePixelTraitsType = 
QSX::Graphics::PixelTraits<QSX::Graphics::ChannelTraits<FORMAT_BYTE>, 
QSX::Graphics::PixelDescriptor_BGRA>, int channel_offset = 0]'
../Include/QSX_Graphics/PixelConverter.h:94:   instantiated from `static void 
QSX::Graphics::PixelConverter<TargetPixelTraitsType, SourcePixelTraitsType, 
metric>::fetch(QSX::Graphics::Pixel<TargetPixelTraitsType>&, const 
QSX::Graphics::Pixel<SourcePixelTraitsType>&) [with TargetPixelTraitsType = 
QSX::Graphics::PixelTraits<QSX::Graphics::ChannelTraits<FORMAT_BYTE>, 
QSX::Graphics::PixelDescriptor_BGR>, SourcePixelTraitsType = 
QSX::Graphics::PixelTraits<QSX::Graphics::ChannelTraits<FORMAT_BYTE>, 
QSX::Graphics::PixelDescriptor_BGRA>, int metric = 0]'
../Include/QSX_Graphics/Pixel.h:305:   instantiated from `void 
QSX::Graphics::Pixel<PixelTraitsType>::fetch(const 
QSX::Graphics::Pixel<SourcePixelTraitsType>&) [with int metric = 0, 
SourcePixelTraitsType = 
QSX::Graphics::PixelTraits<QSX::Graphics::ChannelTraits<FORMAT_BYTE>, 
QSX::Graphics::PixelDescriptor_BGRA>, PixelTraitsType = 
QSX::Graphics::PixelTraits<QSX::Graphics::ChannelTraits<FORMAT_BYTE>, 
QSX::Graphics::PixelDescriptor_BGR>]'
../Include/QSX_Graphics/Bitmap.h:407:   instantiated from `void 
QSX::Graphics::Bitmap<PixelType>::fetch(const 
QSX::Graphics::Bitmap<OtherPixelType>&) [with int metric = 0, OtherPixelType = 
QSX::Graphics::Pixel_ByteBGRA, PixelType = QSX::Graphics::Pixel_ByteBGR]'
../Include/QSX_Graphics/Bitmap.h:388:   instantiated from `void 
QSX::Graphics::Bitmap<PixelType>::fetch(const 
QSX::Graphics::Bitmap<OtherPixelType>&) [with OtherPixelType = 
QSX::Graphics::Pixel_ByteBGRA, PixelType = QSX::Graphics::Pixel_ByteBGR]'
../Include/QSX_Graphics/Bitmap.h:272:   instantiated from 
`QSX::Graphics::Bitmap<PixelType>::Bitmap(const 
QSX::Graphics::Bitmap<OtherPixelType>&) [with OtherPixelType = 
QSX::Graphics::Pixel_ByteBGRA, PixelType = QSX::Graphics::Pixel_ByteBGR]'
test.cpp:21:   instantiated from here
../Include/QSX_Graphics/ChannelFormatConverter.h:62: internal compiler error: 
in
   type_unification_real, at cp/pt.c:8275
Please submit a full bug report,
with preprocessed source if appropriate.
See <URL:http://gcc.gnu.org/bugs.html> for instructions.

[Bug c++/11922] ICE on type_unification_real

[gcc-bugzilla at x-infinity dot com]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11922



------- Additional Comments From gcc-bugzilla at x-infinity dot com  2003-08-14 14:27 -------
Created an attachment (id=4605)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=4605&action=view)
Preprocessed source file

[Bug c++/11922] ICE on type_unification_real

[gcc-bugzilla at x-infinity dot com]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11922



------- Additional Comments From gcc-bugzilla at x-infinity dot com  2003-08-14 14:28 -------
Created an attachment (id=4606)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=4606&action=view)
Preprocessed source file (.s)

[Bug c++/11922] ICE on type_unification_real

[bangerth at dealii dot org]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11922



------- Additional Comments From bangerth at dealii dot org  2003-08-14 22:01 -------
Just so nobody else tackles this: I'm about to reduce the testcase. Looks pretty ugly... 
W.

[Bug c++/11922] ICE on type_unification_real

[bangerth at dealii dot org]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11922



------- Additional Comments From bangerth at dealii dot org  2003-08-14 22:46 -------
Created an attachment (id=4607)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=4607&action=view)
Significantly reduced

I think I'm close but the girl is waiting :-)

[Bug c++/11922] ICE on type_unification_real

[pinskia at gcc dot gnu dot org]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11922


pinskia at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Attachment #4607|text/plain                  |application/x-gzip-
          mime type|                            |compressed

[Bug c++/11922] ICE on type_unification_real

[gcc-bugzilla at x-infinity dot com]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11922



------- Additional Comments From gcc-bugzilla at x-infinity dot com  2003-08-15 05:40 -------
The following replacement makes code compile (and run):

template <class TraitsType>
struct ChannelFormatConverter
{
//
//  ... code skipped ...
//
    template <class OtherTraitsType>
    inline static typename traits_type::value_type from(typename 
OtherTraitsType::value_type value)
    {
        return OtherTraitsType::template 
convert_format<traits_type::channel_format>()(value);
    }
//
//  ... code skipped ...
//
};

should be replaced with:

template <class TraitsType>
struct ChannelFormatConverter
{
//
//  ... code skipped ...
//
    template <class OtherTraitsType>
    inline static typename traits_type::value_type from(typename 
OtherTraitsType::value_type value)
    {
        typename OtherTraitsType::template 
convert_format<traits_type::channel_format> converter_type;
        return converter_type(value);
    }
//
//  ... code skipped ...
//
};


Imho, the first version just should not compile (no typename on a dependent 
type), it should report "undeclared type XXX" or something like that.

[Bug c++/11922] ICE on type_unification_real

[reichelt at gcc dot gnu dot org]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11922



------- Additional Comments From reichelt at gcc dot gnu dot org  2003-08-15 12:41 -------
Created an attachment (id=4611)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=4611&action=view)
Down to 100 lines

[Bug c++/11922] [3.3/3.4 regression] ICE on type_unification_real

[reichelt at gcc dot gnu dot org]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11922


reichelt at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |reichelt at gcc dot gnu dot
                   |                            |org
             Status|UNCONFIRMED                 |NEW
     Ever Confirmed|                            |1
           Keywords|                            |ice-on-invalid-code,
                   |                            |monitored
           Priority|P2                          |P1
   Last reconfirmed|0000-00-00 00:00:00         |2003-08-15 13:43:35
               date|                            |
            Summary|ICE on type_unification_real|[3.3/3.4 regression] ICE on
                   |                            |type_unification_real


------- Additional Comments From reichelt at gcc dot gnu dot org  2003-08-15 13:43 -------
Reduced testcase:

=======================================================
struct A
{
    template <bool> struct B;
};

template <> struct A::B<false> {};

template <typename T> void foo()
{
    T::template B<false>(); // missing typename
}

void bar()
{
    foo<A>();
}
=======================================================

With 3.3 branch I get:

test.cc: In function `void foo() [with T = A]':
test.cc:15:   instantiated from here
test.cc:10: internal compiler error: in type_unification_real, at cp/pt.c:8275
Please submit a full bug report, [etc.]

With mainline I get:

test.cc: In function `void foo() [with T = A]':
test.cc:15:   instantiated from here
test.cc:10: internal compiler error: in get_first_fn, at cp/tree.c:952
Please submit a full bug report, [etc.]

As Aristarkh already pointed out there's a typename missing.
Thus, we have an ICE on invalid code.

With gcc 3.2.3 I get a normal error message:

test.cc: In function `void foo() [with T = A]':
test.cc:15:   instantiated from here
test.cc:10: no method `A::B<false>'

Thus, we have a regression.

[Bug c++/11922] [3.3/3.4 regression] ICE on type_unification_real

[reichelt at gcc dot gnu dot org]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11922



------- Additional Comments From reichelt at gcc dot gnu dot org  2003-08-15 13:54 -------
Phil's regression tester says:
Search converges between 2002-12-24-trunk (#174) and 2002-12-25-trunk (#175).

[Bug c++/11922] [3.3/3.4 regression] ICE on type_unification_real

[bangerth at dealii dot org]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11922



------- Additional Comments From bangerth at dealii dot org  2003-08-15 14:00 -------
Volker, how mean of you! Can't one even go on a date without being deprived of the 
privilege to reduce a testcase??? :-) 
 
W.

[Bug c++/11922] [3.3/3.4 regression] ICE on type_unification_real

[pinskia at gcc dot gnu dot org]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11922



------- Additional Comments From pinskia at gcc dot gnu dot org  2003-08-15 14:03 -------
The dates which Volker quoted are right before the branch of 3.3.

[Bug c++/11922] [3.3/3.4 regression] ICE on type_unification_real

[reichelt at igpm dot rwth-aachen dot de]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11922



------- Additional Comments From reichelt at igpm dot rwth-aachen dot de  2003-08-15 14:15 -------
Subject: Re:  [3.3/3.4 regression] ICE on type_unification_real

On 15 Aug, bangerth at dealii dot org wrote:

> Volker, how mean of you! Can't one even go on a date without being deprived of the 
> privilege to reduce a testcase??? :-) 
>  
> W. 

No, you have to decide. A GCC maintainer is only allowed one kind of fun
per day. It's either dates or bugs. We don't wanna get greedy, do we?;-)

Regards,
Volker

[Bug c++/11922] [3.3/3.4 regression] ICE on type_unification_real

[bangerth at dealii dot org]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11922



------- Additional Comments From bangerth at dealii dot org  2003-08-15 15:23 -------
Aw. I should threaten to immediately step down as a bugmaster if this rule isn't lifted... 
W.

[Bug c++/11922] [3.3/3.4 regression] ICE on type_unification_real

[nathan at gcc dot gnu dot org]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11922


nathan at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         AssignedTo|unassigned at gcc dot gnu   |nathan at gcc dot gnu dot
                   |dot org                     |org
             Status|NEW                         |ASSIGNED

[Bug c++/11922] [3.3/3.4 regression] ICE on type_unification_real

[nathan at gcc dot gnu dot org]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11922


nathan at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |RESOLVED
         Resolution|                            |FIXED


------- Additional Comments From nathan at gcc dot gnu dot org  2003-09-05 08:37 -------
2003-09-05  Nathan Sidwell  <nathan@codesourcery.com>

	PR c++/11922
	* pt.c (tsubst_qualified_id): Make sure we get a non-type.
	(tsubst_expr, tsubst_copy_and_build): Pass false, not zero, as
	is_type_p to lookup_qualified_name.

	* semantics.c (finish_call_expr): Refactor some code.

[Bug c++/11922] [3.3/3.4 regression] ICE on type_unification_real

[cvs-commit at gcc dot gnu dot org]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11922



------- Additional Comments From cvs-commit at gcc dot gnu dot org  2003-09-05 08:38 -------
Subject: Bug 11922

CVSROOT:	/cvs/gcc
Module name:	gcc
Changes by:	nathan@gcc.gnu.org	2003-09-05 08:38:44

Modified files:
	gcc/cp         : ChangeLog pt.c semantics.c 
	gcc/testsuite  : ChangeLog 
Added files:
	gcc/testsuite/g++.dg/template: qualified-id1.C 

Log message:
	cp:
	PR c++/11922
	* pt.c (tsubst_qualified_id): Make sure we get a non-type.
	(tsubst_expr, tsubst_copy_and_build): Pass false, not zero, as
	is_type_p to lookup_qualified_name.
	
	* semantics.c (finish_call_expr): Refactor some code.
	testsuite:
	PR c++/11922
	* g++/dg/template/qualified-id1.C: New test.

Patches:
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/cp/ChangeLog.diff?cvsroot=gcc&r1=1.3647&r2=1.3648
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/cp/pt.c.diff?cvsroot=gcc&r1=1.773&r2=1.774
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/cp/semantics.c.diff?cvsroot=gcc&r1=1.355&r2=1.356
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/ChangeLog.diff?cvsroot=gcc&r1=1.3024&r2=1.3025
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/g++.dg/template/qualified-id1.C.diff?cvsroot=gcc&r1=NONE&r2=1.1