This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug fortran/42736] [4.3/4.4/4.5 Regression] Wrong-code with allocatable compounds
- From: "burnus at gcc dot gnu dot org" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: 14 Jan 2010 11:10:05 -0000
- Subject: [Bug fortran/42736] [4.3/4.4/4.5 Regression] Wrong-code with allocatable compounds
- References: <bug-42736-13404@http.gcc.gnu.org/bugzilla/>
- Reply-to: gcc-bugzilla at gcc dot gnu dot org
------- Comment #5 from burnus at gcc dot gnu dot org 2010-01-14 11:10 -------
The crucial difference between 142153 and 142154 seems to be that the former
uses table->realdata.data directly, while the latter creates a "atmp"
descriptor, which does not fill .dim[1].stride:
table->realdata.dim[0].stride = 1;
table->realdata.dim[1].stride = 1;
and then:
D.947 = NON_LVALUE_EXPR <table->realdata.dim[1].stride> + D.936;
vs.
atmp.1.dim[0].stride = 1;
/* but not setting atmp.1.dim[1].{stride,lbound,ubound} */
D.959 = NON_LVALUE_EXPR <atmp.1.dim[1].stride> + D.957;
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=42736