[Bug middle-end/55046] New: [4.8 Regression] ICE in ira_reuse_stack_slot at ira-color.c:4065

[Bug middle-end/55046] New: [4.8 Regression] ICE in ira_reuse_stack_slot at ira-color.c:4065

[danglin at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=55046

             Bug #: 55046
           Summary: [4.8 Regression] ICE in ira_reuse_stack_slot at
                    ira-color.c:4065
    Classification: Unclassified
           Product: gcc
           Version: 4.8.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: middle-end
        AssignedTo: unassigned@gcc.gnu.org
        ReportedBy: danglin@gcc.gnu.org
              Host: hppa2.0w-hp-hpux11.11
            Target: hppa2.0w-hp-hpux11.11
             Build: hppa2.0w-hp-hpux11.11


Created attachment 28516
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=28516
Preprocessed source

The ICE breaks building the fortran library as follows:

/opt/gnu/bin/bash ./libtool  --tag=FC   --mode=compile
/test/gnu/gcc/objdir/./gc
c/gfortran -B/test/gnu/gcc/objdir/./gcc/
-B/opt/gnu/gcc/gcc-4.8/hppa2.0w-hp-hpux11.11/bin/
-B/opt/gnu/gcc/gcc-4.8/hppa2.0w-hp-hpux11.11/lib/ -isystem /opt/gnu/g
cc/gcc-4.8/hppa2.0w-hp-hpux11.11/include -isystem
/opt/gnu/gcc/gcc-4.8/hppa2.0w-
hp-hpux11.11/sys-include    -DHAVE_CONFIG_H -I. -I../../../gcc/libgfortran 
-iqu
ote../../../gcc/libgfortran/io -I../../../gcc/libgfortran/../gcc
-I../../../gcc/libgfortran/../gcc/config
-I../../../gcc/libgfortran/../libquadmath -I../.././gcc
-I../../../gcc/libgfortran/../libgcc -I../libgcc  -I . -Wall -Werror
-fimplicit-none -fno-repack-arrays -fno-underscoring -fallow-leading-underscore
-g -O2 -c -o _dim_r16.lo `test -f
'../../../gcc/libgfortran/generated/_dim_r16.F90' || ec
ho
'../../../gcc/libgfortran/'`../../../gcc/libgfortran/generated/_dim_r16.F90libtool:
compile:  /test/gnu/gcc/objdir/./gcc/gfortran -B/test/gnu/gcc/objdir/./gcc/
-B/opt/gnu/gcc/gcc-4.8/hppa2.0w-hp-hpux11.11/bin/
-B/opt/gnu/gcc/gcc-4.8/hppa2.0w-hp-hpux11.11/lib/ -isystem
/opt/gnu/gcc/gcc-4.8/hppa2.0w-hp-hpux11.11/include -isystem
/opt/gnu/gcc/gcc-4.8/hppa2.0w-hp-hpux11.11/sys-include -DHAVE_CONF
IG_H -I. -I../../../gcc/libgfortran -iquote../../../gcc/libgfortran/io
-I../../../gcc/libgfortran/../gcc -I../../../gcc/libgfortran/../gcc/config
-I../../../gcc/libgfortran/../libquadmath -I../.././gcc
-I../../../gcc/libgfortran/../libgcc -I../libgcc -I . -Wall -Werror
-fimplicit-none -fno-repack-arrays -fno-underscoring -fallow-leading-underscore
-g -O2 -c ../../../gcc/libgfortran/generated/_dim_
r16.F90  -fPIC -o .libs/_dim_r16.o
../../../gcc/libgfortran/generated/_dim_r16.F90: In function
'_gfortran_specific__dim_r16':
../../../gcc/libgfortran/generated/_dim_r16.F90:42:0: internal compiler error:
Segmentation fault
 end function
 ^

../../../gcc/libgfortran/generated/_dim_r16.F90:42:0: internal compiler error:
Aborted
gfortran: internal compiler error: Aborted (program f951)
./libtool: line 1156: 11697 ABORT instruction       (core dumped)
/test/gnu/gcc/objdir/./gcc/gfortran -B/test/gnu/gcc/objdir/./gcc/
-B/opt/gnu/gcc/gcc-4.8/hppa2.0w-hp-hpux11.11/bin/
-B/opt/gnu/gcc/gcc-4.8/hppa2.0w-hp-hpux11.11/lib/ -isystem
/opt/gnu/gcc/gcc-4.8/hppa2.0w-hp-hpux11.11/include -isystem
/opt/gnu/gcc/gcc-4.8/hppa2.0w-hp-hpux11.11/sys-include -DHAVE_CONFIG_H -I.
-I../../../gcc/libgfortran -iquote../../../gcc/libgfortran/io
-I../../../gcc/libgfortran/../gcc -I../../../gcc/libgfortran/../gcc/config
-I../../../gcc/libgfortran/../libquadmath -I../.././gcc
-I../../../gcc/libgfortran/../libgcc -I../libgcc -I . -Wall -Werror
-fimplicit-none -fno-repack-arrays -fno-underscoring -fallow-leading-underscore
-g -O2 -c ../../../gcc/libgfortran/generated/_dim_r16.F90 -fPIC -o
.libs/_dim_r16.o
make[3]: *** [_dim_r16.lo] Error 1

-bash-3.2$ ./xgcc -B./ -v
Reading specs from ./specs
COLLECT_GCC=./xgcc
COLLECT_LTO_WRAPPER=./lto-wrapper
Target: hppa2.0w-hp-hpux11.11
Configured with: ../gcc/configure --with-gnu-as --with-as=/opt/gnu/bin/as
--enable-shared --with-local-prefix=/opt/gnu --prefix=/opt/gnu/gcc/gcc-4.8
--with-gmp=/opt/gnu/gcc/gmp --enable-threads=posix --enable-debug=no
--disable-nls --without-cloog --without-ppl
--enable-languages=c,c++,objc,fortran,java,ada,obj-c++
Thread model: posix
gcc version 4.8.0 20121023 (experimental) [trunk revision 192710] (GCC) 

(gdb) r
Starting program: /test/gnu/gcc/objdir/gcc/f951
../../../gcc/libgfortran/generated/_dim_r16.F90 -cpp=_dim_r16.f90 -quiet -v -I
. -I ../../../gcc/libgfortran -I ../../../gcc/libgfortran/../gcc -I
../../../gcc/libgfortran/../gcc/config -I
../../../gcc/libgfortran/../libquadmath -I ../.././gcc -I
../../../gcc/libgfortran/../libgcc -I ../libgcc -I . -iprefix
/test/gnu/gcc/objdir/gcc/../lib/gcc/hppa2.0w-hp-hpux11.11/4.8.0/ -isystem
/test/gnu/gcc/objdir/./gcc/include -isystem
/test/gnu/gcc/objdir/./gcc/include-fixed -D HAVE_CONFIG_H -isystem
/opt/gnu/gcc/gcc-4.8/hppa2.0w-hp-hpux11.11/include -isystem
/opt/gnu/gcc/gcc-4.8/hppa2.0w-hp-hpux11.11/sys-include -iquote
../../../gcc/libgfortran/io ../../../gcc/libgfortran/generated/_dim_r16.F90
-quiet -dumpbase _dim_r16.F90 -auxbase-strip .libs/_dim_r16.o -g -O2 -Wall
-Werror -version -fimplicit-none -fno-repack-arrays -fno-underscoring
-fallow-leading-underscore -fPIC -fintrinsic-modules-path finclude -o
_dim_r16.s
warning: Private mapping of shared library text was not specified
by the executable; setting a breakpoint in a shared library which
is not privately mapped will not work.  See the HP-UX 11i v3 chatr
manpage for methods to privately map shared library text.
GNU Fortran (GCC) version 4.8.0 20121023 (experimental) [trunk revision 192710]
(hppa2.0w-hp-hpux11.11)
    compiled by GNU C version 4.8.0 20121023 (experimental) [trunk revision
192710], GMP version 5.0.2, MPFR version 3.1.0-p1, MPC version 1.0
warning: MPFR header version 3.1.0-p1 differs from library version 3.0.1-p4.
GGC heuristics: --param ggc-min-expand=30 --param ggc-min-heapsize=4096
ignoring nonexistent directory
"/test/gnu/gcc/objdir/gcc/../lib/gcc/hppa2.0w-hp-hpux11.11/4.8.0/include"
ignoring nonexistent directory
"/test/gnu/gcc/objdir/gcc/../lib/gcc/hppa2.0w-hp-hpux11.11/4.8.0/include-fixed"
ignoring nonexistent directory
"/test/gnu/gcc/objdir/gcc/../lib/gcc/hppa2.0w-hp-hpux11.11/4.8.0/../../../../hppa2.0w-hp-hpux11.11/include"
ignoring nonexistent directory
"/opt/gnu/gcc/gcc-4.8/lib/gcc/hppa2.0w-hp-hpux11.11/4.8.0/include"
ignoring nonexistent directory "/opt/gnu/gcc/gcc-4.8/include"
ignoring nonexistent directory
"/opt/gnu/gcc/gcc-4.8/lib/gcc/hppa2.0w-hp-hpux11.11/4.8.0/include-fixed"
ignoring nonexistent directory
"/opt/gnu/gcc/gcc-4.8/hppa2.0w-hp-hpux11.11/include"
ignoring duplicate directory "."
ignoring nonexistent directory
"/opt/gnu/gcc/gcc-4.8/hppa2.0w-hp-hpux11.11/include"
ignoring nonexistent directory
"/opt/gnu/gcc/gcc-4.8/hppa2.0w-hp-hpux11.11/sys-include"
#include "..." search starts here:
#include <...> search starts here:
 .
 ../../../gcc/libgfortran
 ../../../gcc/libgfortran/../gcc
 ../../../gcc/libgfortran/../gcc/config
 ../../../gcc/libgfortran/../libquadmath
 ../.././gcc
 ../../../gcc/libgfortran/../libgcc
 ../libgcc
 /test/gnu/gcc/objdir/./gcc/include
 /test/gnu/gcc/objdir/./gcc/include-fixed
 ../../../gcc/libgfortran/io
 /opt/gnu/include
 /usr/include
End of search list.
GNU Fortran (GCC) version 4.8.0 20121023 (experimental) [trunk revision 192710]
(hppa2.0w-hp-hpux11.11)
    compiled by GNU C version 4.8.0 20121023 (experimental) [trunk revision
192710], GMP version 5.0.2, MPFR version 3.1.0-p1, MPC version 1.0
warning: MPFR header version 3.1.0-p1 differs from library version 3.0.1-p4.
GGC heuristics: --param ggc-min-expand=30 --param ggc-min-heapsize=4096

Program received signal SIGSEGV, Segmentation fault.
0x004daff4 in _Z20ira_reuse_stack_slotijj (regno=102, inherent_size=4, 
    total_size=4) at ../../gcc/gcc/ira-color.c:4065
4065      ira_assert (inherent_size == PSEUDO_REGNO_BYTES (regno)
(gdb) bt
#0  0x004daff4 in _Z20ira_reuse_stack_slotijj (regno=102, inherent_size=4, 
    total_size=4) at ../../gcc/gcc/ira-color.c:4065
#1  0x005b1808 in _ZL9alter_regiib (i=102, from_reg=-1, dont_share_p=false)
    at ../../gcc/gcc/reload1.c:2205
#2  0x005b9b40 in _Z6reloadP7rtx_defi (
    first=0x40116b60 <global_options+1612>, global=1074854752)
    at ../../gcc/gcc/reload1.c:793
#3  0x004be268 in _ZL21rest_of_handle_reloadv () at ../../gcc/gcc/ira.c:4355
#4  0x0055c660 in _Z16execute_one_passP8opt_pass (
    pass=0x4005e428 <pass_reload>) at ../../gcc/gcc/passes.c:2320
#5  0x0055cad0 in _Z17execute_pass_listP8opt_pass (
    pass=0x4005e428 <pass_reload>) at ../../gcc/gcc/passes.c:2381
#6  0x0055cae8 in _Z17execute_pass_listP8opt_pass (
    pass=0x400687bc <pass_rest_of_compilation>) at ../../gcc/gcc/passes.c:2382
#7  0x002662e4 in _ZL15expand_functionP11cgraph_node (node=0x7ae78000)
    at ../../gcc/gcc/cgraphunit.c:1601
#8  0x00268600 in _Z7compilev () at ../../gcc/gcc/cgraphunit.c:1705
#9  0x00268e94 in _Z25finalize_compilation_unitv ()
    at ../../gcc/gcc/cgraphunit.c:2080
#10 0x004ec1a8 in _Z25write_global_declarationsv ()
    at ../../gcc/gcc/langhooks.c:323
#11 0x00111ac4 in _ZL29gfc_write_global_declarationsv ()
    at ../../gcc/gcc/fortran/f95-lang.c:265
---Type <return> to continue, or q <return> to quit---
#12 0x00633c88 in compile_file() () at ../../gcc/gcc/toplev.c:560
#13 0x00635d84 in _Z11toplev_mainiPPc (argc=0, argv=0x0)
    at ../../gcc/gcc/toplev.c:1866
#14 0x00921428 in main (argc=102, argv=0x4) at ../../gcc/gcc/main.c:36
(gdb) p regno
$1 = 102
(gdb) list
4060      ira_allocno_t another_allocno, allocno =
ira_regno_allocno_map[regno];
4061      rtx x;
4062      bitmap_iterator bi;
4063      struct ira_spilled_reg_stack_slot *slot = NULL;
4064    
4065      ira_assert (inherent_size == PSEUDO_REGNO_BYTES (regno)
4066              && inherent_size <= total_size
4067              && ALLOCNO_HARD_REGNO (allocno) < 0);
4068      if (! flag_ira_share_spill_slots)
4069        return NULL_RTX;
(gdb) p allocno
$2 = (ira_allocno_t) 0x0
(gdb) disass $pc-16,$pc+16
Dump of assembler code from 0x4dafe4 to 0x4db004:
   0x004dafe4 <_Z20ira_reuse_stack_slotijj+124>:    ldb 0(ret0),r14
   0x004dafe8 <_Z20ira_reuse_stack_slotijj+128>:    cmpb,<> r25,r14,0x4db594
<_Z20ira_reuse_stack_slotijj+1580>
   0x004dafec <_Z20ira_reuse_stack_slotijj+132>:    depw,z r26,29,30,r7
   0x004daff0 <_Z20ira_reuse_stack_slotijj+136>:    cmpb,<<,n r24,r14,0x4db594
<_Z20ira_reuse_stack_slotijj+1580>
=> 0x004daff4 <_Z20ira_reuse_stack_slotijj+140>:    ldh c(r4),r13
   0x004daff8 <_Z20ira_reuse_stack_slotijj+144>:    extrw,s r13,31,16,r13
   0x004daffc <_Z20ira_reuse_stack_slotijj+148>:    cmpib,<= 0,r13,0x4db594
<_Z20ira_reuse_stack_slotijj+1580>
   0x004db000 <_Z20ira_reuse_stack_slotijj+152>:    addil L%4d800,dp,r1
End of assembler dump.
(gdb) p/x $r4
$3 = 0x0
(gdb) info addr allocno
Symbol "allocno" is a variable in register r4.

[Bug middle-end/55046] [4.8 Regression] ICE in ira_reuse_stack_slot at ira-color.c:4065

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=55046

Richard Biener <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |4.8.0

[Bug middle-end/55046] [4.8 Regression] ICE in ira_reuse_stack_slot at ira-color.c:4065

[danglin at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=55046

John David Anglin <danglin at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |steven at gcc dot gnu.org

--- Comment #1 from John David Anglin <danglin at gcc dot gnu.org> 2012-10-27 20:42:30 UTC ---
This problem was introduced in revision 192440:

2012-10-14  Steven Bosscher  <steven@gcc.gnu.org>

        PR rtl-optimization/38711
        * df.h (df_get_live_out, df_get_live_in): Make static inline functions.
        * df-problems.c (df_get_live_out, df_get_live_in): Moved to df.h.
        * ira-lives.c (process_bb_node_lives): Use df_get_live_out instead of
        DF_LR_OUT.
        * ira-build.c (create_bb_allocnos): Likewise.
        (create_loop_allocnos): Likewise, and use df_get_live_in instead of
        DF_LR_IN.
        * ira-emit.c (generate_edge_moves): Likewise.
        (add_ranges_and_copies): Likewise.
        * ira-color.c (ira_loop_edge_freq): Use df_get_live_out instead of
        DF_LR_OUT, and df_get_live_in instead of DF_LR_IN.
        * ira.c (mark_elimination): Update DF_LR and DF_LIVE.
        (build_insn_chain): Use df_get_live_out instead of DF_LR_OUT.
        (do_reload): Remove the DF_LIVE problem for -O1.

[Bug middle-end/55046] [4.8 Regression] ICE in ira_reuse_stack_slot at ira-color.c:4065

[danglin at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=55046

--- Comment #2 from John David Anglin <danglin at gcc dot gnu.org> 2012-10-27 20:43:59 UTC ---
It doesn't help to return NULL_RTX when allocno is zero.

[Bug middle-end/55046] [4.8 Regression] ICE in ira_reuse_stack_slot at ira-color.c:4065

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=55046

Richard Biener <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Priority|P3                          |P1

[Bug middle-end/55046] [4.8 Regression] ICE in ira_reuse_stack_slot at ira-color.c:4065

[steven at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=55046

Steven Bosscher <steven at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |SUSPENDED
   Last reconfirmed|                            |2012-12-03
     Ever Confirmed|0                           |1

--- Comment #3 from Steven Bosscher <steven at gcc dot gnu.org> 2012-12-03 17:41:45 UTC ---
The patch to use DF_LIVE in IRA was already reverted.

[Bug middle-end/55046] ICE in ira_reuse_stack_slot at ira-color.c:4065

[danglin at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=55046

John David Anglin <danglin at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|SUSPENDED                   |RESOLVED
         Resolution|                            |FIXED

--- Comment #4 from John David Anglin <danglin at gcc dot gnu.org> 2012-12-03 23:24:58 UTC ---
Fixed.