This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug middle-end/37285] New: [4.4 Regression] ICE while building binutils on ppc
- From: "pinskia at gcc dot gnu dot org" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: 30 Aug 2008 01:28:46 -0000
- Subject: [Bug middle-end/37285] New: [4.4 Regression] ICE while building binutils on ppc
- Reply-to: gcc-bugzilla at gcc dot gnu dot org
Hi, it looks like some of the VRP enhancements for switch statements is causing
an ICE on some code in binutils.
Testcase:
_bfd_xcoff_canonicalize_dynamic_reloc (unsigned long long l_symndx)
{
if (l_symndx < 3)
{
switch (l_symndx)
{
case 0:
case 1:
break;
case 2:
_bfd_abort ("HI");
}
}
}
--- CUT ---
With VRP turned on we get:
switch (l_symndx) <case 0 ... 1: <L4>, case 2: <L3>>
But without it turned on we get:
switch (l_symndx) <default: <L4>, case 2: <L3>>
Most likely caused by the patch to fix PR 14495.
--
Summary: [4.4 Regression] ICE while building binutils on ppc
Product: gcc
Version: 4.4.0
Status: UNCONFIRMED
Keywords: ice-on-valid-code
Severity: normal
Priority: P3
Component: middle-end
AssignedTo: unassigned at gcc dot gnu dot org
ReportedBy: pinskia at gcc dot gnu dot org
GCC target triplet: powerpc-unknown-linux-gnu
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37285