From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 13921 invoked by alias); 2 Mar 2012 14:24:44 -0000 Received: (qmail 13908 invoked by uid 22791); 2 Mar 2012 14:24:41 -0000 X-SWARE-Spam-Status: No, hits=-2.8 required=5.0 tests=ALL_TRUSTED,AWL,BAYES_00 X-Spam-Check-By: sourceware.org Received: from localhost (HELO gcc.gnu.org) (127.0.0.1) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Fri, 02 Mar 2012 14:24:28 +0000 From: "bernds at gcc dot gnu.org" To: gcc-bugs@gcc.gnu.org Subject: [Bug target/52425] [4.6 Regression] ICE when compiling file from audacious on debian sparc Date: Fri, 02 Mar 2012 14:24:00 -0000 X-Bugzilla-Reason: CC X-Bugzilla-Type: changed X-Bugzilla-Watch-Reason: None X-Bugzilla-Product: gcc X-Bugzilla-Component: target X-Bugzilla-Keywords: X-Bugzilla-Severity: normal X-Bugzilla-Who: bernds at gcc dot gnu.org X-Bugzilla-Status: UNCONFIRMED X-Bugzilla-Priority: P3 X-Bugzilla-Assigned-To: unassigned at gcc dot gnu.org X-Bugzilla-Target-Milestone: 4.6.4 X-Bugzilla-Changed-Fields: Message-ID: In-Reply-To: References: X-Bugzilla-URL: http://gcc.gnu.org/bugzilla/ Auto-Submitted: auto-generated Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Mailing-List: contact gcc-bugs-help@gcc.gnu.org; run by ezmlm Precedence: bulk List-Id: List-Archive: List-Post: List-Help: Sender: gcc-bugs-owner@gcc.gnu.org X-SW-Source: 2012-03/txt/msg00221.txt.bz2 http://gcc.gnu.org/bugzilla/show_bug.cgi?id=3D52425 --- Comment #4 from Bernd Schmidt 2012-03-02 14= :22:54 UTC --- For some reason I can never reproduce sparc bugs with cross compilers. The original testcase doesn't fail here; the reduced one fails but with a diffe= rent error message, and it fails in instantiate_virtual_regs which is before any head-merging ever runs: sparcbug.i: In function =E2=80=98action_button_cb=E2=80=99: sparcbug.i:57:1: error: unrecognizable insn: (insn 7 6 8 3 (set (reg:SI 130) (lo_sum:SI (reg:SI 130) (unspec:SI [ (symbol_ref/f:DI ("*.LC0") [flags 0x2] ) ] 0))) sparcbug.i:49 -1 (nil)) Breakpoint 2, internal_error (gmsgid=3D0xd6093d "in %s, at %s:%d") at ../../trunk/gcc/diagnostic.c:839 839 { (gdb) bt #0 internal_error (gmsgid=3D0xd6093d "in %s, at %s:%d") at ../../trunk/gcc/diagnostic.c:839 #1 0x0000000000bcf7cc in fancy_abort (file=3D, line=3D2123, function=3D0xc99260 "extract_insn") at ../../trunk/gcc/diagnostic.c:899 #2 0x000000000081e8d3 in _fatal_insn (msgid=3D, insn=3D0x7ffff6bc5990,=20 file=3D0xc9885b "../../trunk/gcc/recog.c", line=3D2123, function=3D0xc9= 9260 "extract_insn") at ../../trunk/gcc/rtl-error.c:110 #3 0x000000000081e912 in _fatal_insn_not_found (insn=3D0xc99260, file=3D0x= c9886b "recog.c", line=3D2123,=20 function=3D0xa
) at ../../trunk/gcc/rtl-error.c:118 #4 0x00000000007ea2b4 in extract_insn (insn=3D0x7ffff6bc5990) at ../../trunk/gcc/recog.c:2123 #5 0x00000000006cfe1c in instantiate_virtual_regs_in_insn (insn=3D0x7ffff6bc5990) at ../../trunk/gcc/function.c:1555 #6 instantiate_virtual_regs () at ../../trunk/gcc/function.c:1911 Please post the actual command line that gets passed to cc1. Also, it would= be helpful if you could load cc1 into a debugger and do the following at the c= rash site in output_addr_const: (gdb) p debug_rtx (x) and at some higher level, also use debug_rtx to print out the insn it's try= ing to print.