This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [patch, fortran] PR27954 ICE on garbage in DATA statement
- From: Jerry DeLisle <jvdelisle at verizon dot net>
- To: Jerry DeLisle <jvdelisle at verizon dot net>
- Cc: Paul Thomas <paulthomas2 at wanadoo dot fr>, Fortran List <fortran at gcc dot gnu dot org>, gcc-patches <gcc-patches at gcc dot gnu dot org>
- Date: Thu, 26 Oct 2006 21:42:42 -0700
- Subject: Re: [patch, fortran] PR27954 ICE on garbage in DATA statement
- References: <45395CD7.6040106@verizon.net> <4539A8C0.7080401@wanadoo.fr> <4539C89E.9030406@verizon.net>
Jerry DeLisle wrote:
Paul Thomas wrote:
Jerry
:REVIEWMAIL:
The three test cases provided gave three different internal errors on
i686-linux. One on a bad symbol, one on a bad type, and one segfault.
Putting the three tests in separate procedures would combine them into
one, would it not?
As long as the last one is the one that segfaults, and assuming the
other two don't segfault in a different setting. I suppose the test
case is not all that important for this particular patch.
I will give this a try. OK either way?
Jerry
PING See combined test case attached.
! { dg-do compile }
! PR27954 Internal compiler error on bad statements
! Derived from test case submitted in PR.
subroutine bad1
character*20 :: y, x 00 ! { dg-error "Syntax error" }
data y /'abcdef'/, x /'jbnhjk'/ pp ! { dg-error "Syntax error" }
end subroutine bad1
subroutine bad2
character*20 :: y, x 00 ! { dg-error "Syntax error" }
data y /'abcdef'/, x /'jbnhjk'/ pp ! { dg-error "Syntax error" }
print *, "basket case."
end subroutine bad2
subroutine bad3
implicit none
character*20 :: y, x 00 ! { dg-error "Syntax error" }
data y /'abcdef'/, x /'jbnhjk'/ pp ! { dg-error "Syntax error" }
print *, "basket case that segfaults without patch."
end subroutine bad3