------- Comment #5 from danglin at gcc dot gnu dot org 2010-07-28 15:00 -------
(In reply to comment #4)
> While the ICE nominally results from a hash lookup failure, replacing
> libiberty.a with the stage1 version makes no difference. Thus, it
> doesn't appear that the problem is with the hash code.
This is wrong. Hash related failure is different bug. The backtrace
for this ICE is:
Program received signal SIGSEGV, Segmentation fault.
commutative_operand_precedence (op=0x0) at ../../gcc/gcc/rtlanal.c:2914
2914 enum rtx_code code = GET_CODE (op);
(gdb) bt
#0 commutative_operand_precedence (op=0x0) at ../../gcc/gcc/rtlanal.c:2914
#1 0x003b135c in swap_commutative_operands_p (x=<value optimized out>, y=0x0)
at ../../gcc/gcc/rtlanal.c:2980
#2 0x003ee244 in simplify_const_relational_operation (code=NE, mode=SImode,
op0=0x400122b8, op1=0x0) at ../../gcc/gcc/simplify-rtx.c:4256
#3 0x003e4ee0 in simplify_relational_operation (code=NE, mode=SImode,
cmp_mode=SImode, op0=0x400122b8, op1=0x0)
at ../../gcc/gcc/simplify-rtx.c:3904
#4 0x003e565c in simplify_gen_relational (code=NE, mode=VOIDmode,
cmp_mode=<value optimized out>, op0=0x400122b8, op1=0x0)
at ../../gcc/gcc/simplify-rtx.c:346
#5 0x00317ec4 in iv_number_of_iterations (loop=0x7, desc=0x9b62d8)
at ../../gcc/gcc/loop-iv.c:2609
#6 check_simple_exit (loop=0x7, desc=0x9b62d8) at ../../gcc/gcc/loop-iv.c:2847
#7 find_simple_exit (loop=0x7, desc=0x9b62d8) at ../../gcc/gcc/loop-iv.c:2872
(gdb) p/x $pc
$2 = 0x3b11d8
(gdb) disass 0x3b11c8,0x3b11e8
Dump of assembler code from 0x3b11c8 to 0x3b11e8:
0x003b11c8 <regno_use_in+276>: ldo -4(r7),r7
0x003b11cc <regno_use_in+280>: b,l 0x3b10f8 <regno_use_in+68>,r0
0x003b11d0 <regno_use_in+284>: ldi 0,ret0
0x003b11d4 <commutative_operand_precedence+0>: stw rp,-14(sp)
=> 0x003b11d8 <commutative_operand_precedence+4>: ldh 0(r26),r19
0x003b11dc <commutative_operand_precedence+8>: ldo 40(sp),sp
0x003b11e0 <commutative_operand_precedence+12>: ldi 1e,r20
0x003b11e4 <commutative_operand_precedence+16>: cmpb,<>
r20,r19,0x3b11f8 <commutative_operand_precedence+36>
End of assembler dump.
(gdb) p/x $r26
$3 = 0x0
ICE also seen on hppa2.0w-hp-hpux11.11.
--
danglin at gcc dot gnu dot org changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |bernds at gcc dot gnu dot
| |org
GCC build triplet|hppa-unknown-linux-gnu |hppa*-*-* (32-bit)
GCC host triplet|hppa-unknown-linux-gnu |hppa*-*-* (32-bit)
GCC target triplet|hppa-unknown-linux-gnu |hppa*-*-* (32-bit)
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45063
