c++/8660: template overloading ICE in tsubst_expr, at cp/pt.c:7644

martin.dorey@bluearc.com martin.dorey@bluearc.com
Wed Nov 20 11:24:00 GMT 2002


>Number:         8660
>Category:       c++
>Synopsis:       template overloading ICE in tsubst_expr, at cp/pt.c:7644
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    unassigned
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Wed Nov 20 10:46:03 PST 2002
>Closed-Date:
>Last-Modified:
>Originator:     Martin Dorey
>Release:        GNU C++ version 3.2.1 20021111 (Debian prerelease) (i386-linux)
>Organization:
>Environment:
Linux trevithick 2.4.16 #2 Mon Dec 10 15:54:50 GMT 2001 i686 unknown unknown GNU/Linux
>Description:
trevithick:~/stingray/test/data_structures$ gcc-3.2 -v -c -W -Wall -pedantic --save-temps bug0.cpp 
Reading specs from /usr/lib/gcc-lib/i386-linux/3.2.1/specs
Configured with: /home/packages/gcc/3.2/gcc-3.2-3.2.1ds5/src/configure -v --enable-languages=c,c++,java,f77,proto,objc,ada --prefix=/usr --mandir=/usr/share/man --infodir=/usr/share/info --with-gxx-include-dir=/usr/include/c++/3.2 --enable-shared --with-system-zlib --enable-nls --without-included-gettext --enable-__cxa_atexit --enable-clocale=gnu --enable-java-gc=boehm --enable-objc-gc i386-linux
Thread model: posix
gcc version 3.2.1 20021111 (Debian prerelease)
 /usr/lib/gcc-lib/i386-linux/3.2.1/cpp0 -lang-c++ -D__GNUG__=3 -D__DEPRECATED -D__EXCEPTIONS -v -D__GNUC__=3 -D__GNUC_MINOR__=2 -D__GNUC_PATCHLEVEL__=1 -D__GXX_ABI_VERSION=102 -D__ELF__ -Dunix -D__gnu_linux__ -Dlinux -D__ELF__ -D__unix__ -D__gnu_linux__ -D__linux__ -D__unix -D__linux -Asystem=posix -D__NO_INLINE__ -D__STDC_HOSTED__=1 -D_GNU_SOURCE -Acpu=i386 -Amachine=i386 -Di386 -D__i386 -D__i386__ -D__tune_i386__ bug0.cpp -W -Wall -pedantic bug0.ii
GNU CPP version 3.2.1 20021111 (Debian prerelease) (cpplib) (i386 Linux/ELF)
ignoring nonexistent directory "/usr/i386-linux/include"
#include "..." search starts here:
#include <...> search starts here:
 /usr/include/c++/3.2
 /usr/include/c++/3.2/i386-linux
 /usr/include/c++/3.2/backward
 /usr/local/include
 /usr/lib/gcc-lib/i386-linux/3.2.1/include
 /usr/include
End of search list.
 /usr/lib/gcc-lib/i386-linux/3.2.1/cc1plus -fpreprocessed bug0.ii -quiet -dumpbase bug0.cpp -W -Wall -pedantic -version -o bug0.s
GNU CPP version 3.2.1 20021111 (Debian prerelease) (cpplib) (i386 Linux/ELF)
GNU C++ version 3.2.1 20021111 (Debian prerelease) (i386-linux)
        compiled by GNU C version 3.2.1 20021111 (Debian prerelease).
bug0.cpp: In member function `void BadgerBuf::ReadPod() [with B = int]':
bug0.cpp:10:   instantiated from here
bug0.cpp:10: Internal compiler error in tsubst_expr, at cp/pt.c:7644
Please submit a full bug report,
with preprocessed source if appropriate.
See <URL:http://www.gnu.org/software/gcc/bugs.html> for instructions.
>How-To-Repeat:
Source is:

struct BadgerBuf
{
  void ReadPod();
  template<class B>
  void ReadPod();
};

void BadgerBuf::ReadPod ()
  { ReadPod<int> (); }

Adding this-> to the call to ReadPod<int> doesn't change the behaviour.  Nor does "this->template ".  Putting brackets round the (ReadPod<int>) part causes a different error:

bug0.cpp: In member function `void BadgerBuf::ReadPod()':
bug0.cpp:10: internal error: Segmentation fault
Please submit a full bug report,
with preprocessed source if appropriate.
See <URL:http://www.gnu.org/software/gcc/bugs.html> for instructions.

Swapping the order of the declarations avoids the problem.  Unless you've left the brackets around (ReadPod<int>) in which case it still segfaults.

If you leave the declarations swapped but add another template between the class definition and the function definition, like so:

struct BadgerBuf
{
  template<class B>
  void ReadPod();
  void ReadPod();
};

template <class T>
struct Predicate 
  {};

void BadgerBuf::ReadPod ()
  { ReadPod<int> (); }

Then you get different behaviour again:

bug0.cpp: In member function `void BadgerBuf::ReadPod()':
bug0.cpp:14: parse error before `>' token

Perhaps I should have submitted these as separate bugs - but I can always do that later if they turn out to be unrelated.
>Fix:

>Release-Note:
>Audit-Trail:
>Unformatted:



More information about the Gcc-bugs mailing list