From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 15408 invoked by alias); 29 May 2006 10:47:16 -0000 Received: (qmail 15386 invoked by uid 48); 29 May 2006 10:47:06 -0000 Date: Mon, 29 May 2006 10:47:00 -0000 Subject: [Bug middle-end/27793] New: num_ssa_names inconsistent or immediate use iterator wrong X-Bugzilla-Reason: CC Message-ID: Reply-To: gcc-bugzilla@gcc.gnu.org To: gcc-bugs@gcc.gnu.org From: "rguenth at gcc dot gnu dot org" Mailing-List: contact gcc-bugs-help@gcc.gnu.org; run by ezmlm Precedence: bulk List-Archive: List-Post: List-Help: Sender: gcc-bugs-owner@gcc.gnu.org X-SW-Source: 2006-05/txt/msg02921.txt.bz2 List-Id: The testcase I'll attach momentarily segfaults/ICEs at random places for a checking enabled 4.1.1 compiler. Valgrind has more useful information as ==31272== Invalid read of size 4 ==31272== at 0x82AC3C3: verify_ssa (tree-ssa.c:746) ==31272== by 0x8731ED8: execute_todo (passes.c:761) ... ==31272== Address 0x456FA58 is 8 bytes after a block of size 120 alloc'd ==31272== at 0x402064B: calloc (in /usr/lib/valgrind/x86-linux/vgpreload_memcheck.so) ==31272== by 0x88D0817: xcalloc (xmalloc.c:162) ==31272== by 0x82ABA7F: verify_ssa (tree-ssa.c:637) which means we access definition_block[SSA_NAME_VERSION (op)] with SSA_NAME_VERSION (op) >= num_ssa_names. Attaching with gdb at this point also shows op being corrupt (0x433230c), so it may be well the iterator being bogus. It's hard to reduce the testcase further as it is very sensitive to number of ssa names and allocation profile. -- Summary: num_ssa_names inconsistent or immediate use iterator wrong Product: gcc Version: 4.1.1 Status: UNCONFIRMED Keywords: ice-on-valid-code Severity: normal Priority: P3 Component: middle-end AssignedTo: unassigned at gcc dot gnu dot org ReportedBy: rguenth at gcc dot gnu dot org http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27793