public inbox for gcc-bugs@sourceware.org help / color / mirror / Atom feed
From: "steven at gcc dot gnu dot org" <gcc-bugzilla@gcc.gnu.org> To: gcc-bugs@gcc.gnu.org Subject: [Bug rtl-optimization/26449] [4.2 Regression] ICE with -march=pentium4 -ftree-vectorize in matmul_i4.c in loop invariant motion Date: Fri, 24 Feb 2006 20:49:00 -0000 [thread overview] Message-ID: <20060224204843.11504.qmail@sourceware.org> (raw) In-Reply-To: <bug-26449-11811@http.gcc.gnu.org/bugzilla/> ------- Comment #4 from steven at gcc dot gnu dot org 2006-02-24 20:48 ------- I'm not sure how to fix this. The debug trace here says it all: We try to force something into an operand using an optab with a mode for which we have no handler: Starting program: /home/steven/devel/build-trunk/gcc/cc1 -m32 -std=c99 -O1 -ftree-vectorize -march=pentium4 t.c matmul_i4 Analyzing compilation unitPerforming intraprocedural optimizations Assembling functions: matmul_i4 Breakpoint 3, move_invariant_reg (loop=0xd3f1d0, invno=0) at loop-invariant.c:1162 1162 op = force_operand (SET_SRC (set), reg); (gdb) p debug_rtx(inv->insn) (insn 82 81 143 10 (set (subreg:TI (reg:V4SI 98) 0) (lshiftrt:TI (subreg:TI (reg:V4SI 71 [ vect_cst_.47 ]) 0) (const_int 32 [0x20]))) -1 (nil) (nil)) $36 = void (gdb) p debug_rtx(set) (set (subreg:TI (reg:V4SI 98) 0) (lshiftrt:TI (subreg:TI (reg:V4SI 71 [ vect_cst_.47 ]) 0) (const_int 32 [0x20]))) $37 = void (gdb) step force_operand (value=0x2aaaab009cc0, target=0x2aaaab00e500) at expr.c:5793 5793 rtx subtarget = get_subtarget (target); (gdb) b 5888 Breakpoint 4 at 0x5ceb64: file expr.c, line 5888. (gdb) cont Continuing. Breakpoint 4, force_operand (value=0x2aaaab009cc0, target=0x2aaaab00e500) at expr.c:5888 5888 return expand_simple_binop (GET_MODE (value), code, op1, op2, (gdb) p debug_rtx(op1) (subreg:TI (reg:V4SI 71 [ vect_cst_.47 ]) 0) $38 = void (gdb) p debug_rtx(op2) (const_int 32 [0x20]) $39 = void (gdb) p code $40 = LSHIFTRT (gdb) step expand_simple_binop (mode=TImode, code=LSHIFTRT, op0=0x2aaaab009c80, op1=0x2aaaaadf5600, target=0x2aaaab00e500, unsignedp=1, methods=OPTAB_LIB_WIDEN) at optabs.c:1154 1154 optab binop = code_to_optab[(int) code]; (gdb) next 1157 return expand_binop (mode, binop, op0, op1, target, unsignedp, methods); (gdb) p *binop $41 = {code = LSHIFTRT, handlers = {{insn_code = CODE_FOR_nothing, libfunc = 0x0}, { insn_code = CODE_FOR_nothing, libfunc = 0x0}, {insn_code = CODE_FOR_nothing, libfunc = 0x0}, {insn_code = CODE_FOR_nothing, libfunc = 0x0}, { insn_code = CODE_FOR_nothing, libfunc = 0x0}, {insn_code = CODE_FOR_nothing, libfunc = 0x0}, {insn_code = CODE_FOR_nothing, libfunc = 0x0}, { insn_code = CODE_FOR_nothing, libfunc = 0x0}, {insn_code = CODE_FOR_nothing, libfunc = 0x0}, {insn_code = CODE_FOR_nothing, libfunc = 0x0}, { insn_code = CODE_FOR_lshrqi3, libfunc = 0x0}, {insn_code = CODE_FOR_lshrhi3, libfunc = 0x0}, {insn_code = CODE_FOR_lshrsi3, libfunc = 0x2aaaaaf88960}, { insn_code = CODE_FOR_lshrdi3, libfunc = 0x2aaaaaf889a0}, { insn_code = CODE_FOR_nothing, libfunc = 0x0} <repeats 23 times>, { insn_code = CODE_FOR_lshrv8hi3, libfunc = 0x0}, {insn_code = CODE_FOR_lshrv4si3, libfunc = 0x0}, {insn_code = CODE_FOR_lshrv2di3, libfunc = 0x0}, { insn_code = CODE_FOR_nothing, libfunc = 0x0}, {insn_code = CODE_FOR_nothing, libfunc = 0x0}, {insn_code = CODE_FOR_nothing, libfunc = 0x0}, { insn_code = CODE_FOR_nothing, libfunc = 0x0}, {insn_code = CODE_FOR_nothing, libfunc = 0x0}, {insn_code = CODE_FOR_nothing, libfunc = 0x0}, { insn_code = CODE_FOR_nothing, libfunc = 0x0}, {insn_code = CODE_FOR_nothing, libfunc = 0x0}, {insn_code = CODE_FOR_nothing, libfunc = 0x0}}} (gdb) -- steven at gcc dot gnu dot org changed: What |Removed |Added ---------------------------------------------------------------------------- Target Milestone|4.2.0 |--- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=26449
next prev parent reply other threads:[~2006-02-24 20:48 UTC|newest] Thread overview: 34+ messages / expand[flat|nested] mbox.gz Atom feed top 2006-02-23 18:31 [Bug c/26449] New: ICE with -march=pentium4 in matmul_i4.c aldot at gcc dot gnu dot org 2006-02-24 10:26 ` [Bug c/26449] ICE with -march=pentium4 -ftree-vectorize in matmul_i4.c in loop invaraint motion rguenth at gcc dot gnu dot org 2006-02-24 10:46 ` [Bug rtl-optimization/26449] " rguenth at gcc dot gnu dot org 2006-02-24 10:51 ` steven at gcc dot gnu dot org 2006-02-24 18:14 ` [Bug rtl-optimization/26449] [4.2 Regression] ICE with -march=pentium4 -ftree-vectorize in matmul_i4.c in loop invariant motion pinskia at gcc dot gnu dot org 2006-02-24 20:49 ` steven at gcc dot gnu dot org [this message] 2006-02-24 20:51 ` pinskia at gcc dot gnu dot org 2006-02-24 21:22 ` rakdver at atrey dot karlin dot mff dot cuni dot cz 2006-05-21 20:54 ` pinskia at gcc dot gnu dot org 2006-06-04 17:50 ` mmitchel at gcc dot gnu dot org 2006-06-04 18:00 ` steven at gcc dot gnu dot org 2006-06-06 10:11 ` rakdver at gcc dot gnu dot org 2006-06-06 18:03 ` rakdver at gcc dot gnu dot org 2006-06-08 8:20 ` rakdver at gcc dot gnu dot org 2006-06-08 15:36 ` pinskia at gcc dot gnu dot org 2007-06-01 6:39 ` [Bug rtl-optimization/26449] [4.2/4.3 Regression] ICE " ubizjak at gmail dot com 2007-06-01 6:42 ` ubizjak at gmail dot com 2007-06-01 6:42 ` ubizjak at gmail dot com 2007-06-01 7:30 ` ubizjak at gmail dot com 2007-09-08 11:33 ` uros at gcc dot gnu dot org 2007-09-08 11:35 ` ubizjak at gmail dot com 2007-09-08 11:48 ` uros at gcc dot gnu dot org 2007-09-08 11:51 ` ubizjak at gmail dot com 2007-09-14 5:42 ` ubizjak at gmail dot com 2007-09-14 5:49 ` ubizjak at gmail dot com 2007-09-14 15:57 ` rakdver at gcc dot gnu dot org 2007-09-17 15:39 ` rakdver at gcc dot gnu dot org 2007-09-17 16:44 ` [Bug rtl-optimization/26449] [4.2 " ubizjak at gmail dot com 2008-01-13 16:14 ` rguenth at gcc dot gnu dot org 2008-01-15 8:30 ` ubizjak at gmail dot com 2008-04-25 6:28 ` ubizjak at gmail dot com 2008-04-25 6:29 ` ubizjak at gmail dot com 2008-07-04 16:41 ` jsm28 at gcc dot gnu dot org 2009-03-30 15:50 ` jsm28 at gcc dot gnu dot org
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=20060224204843.11504.qmail@sourceware.org \ --to=gcc-bugzilla@gcc.gnu.org \ --cc=gcc-bugs@gcc.gnu.org \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).