https://gcc.gnu.org/bugzilla/show_bug.cgi?id=69148
Bug ID: 69148
Summary: [5/6 Regression] ICE (floating point exception) on
s390x-linux-gnu
Product: gcc
Version: 5.3.1
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: target
Assignee: unassigned at gcc dot gnu.org
Reporter: doko at gcc dot gnu.org
Target Milestone: ---
seen building the mame package on s390x with gcc-5-branch 20160101
In file included from src/emu/cpu/sharc/sharc.c:227:0:
src/emu/cpu/sharc/sharcops.inc: In member function 'void
adsp21062_device::COMPUTE(UINT32)':
src/emu/cpu/sharc/sharcops.inc:1066:1: internal compiler error: Floating point
exception
}
^
Compiling src/emu/cpu/dsp32/dsp32.c...
0x802bb761 crash_signal
../../src/gcc/toplev.c:383
0x8067dfa6 validate_subreg(machine_mode, machine_mode, rtx_def const*, unsigned
int)
../../src/gcc/emit-rtl.c:904
0x8067e1d3 gen_rtx_SUBREG(machine_mode, rtx_def*, int)
../../src/gcc/emit-rtl.c:909
0x80229a8f emit_spill_move
../../src/gcc/lra-constraints.c:1054
0x8080a811 curr_insn_transform
../../src/gcc/lra-constraints.c:3561
0x808067ed lra_constraints(bool)
../../src/gcc/lra-constraints.c:4446
0x807ef89b lra(_IO_FILE*)
../../src/gcc/lra.c:2344
0x8079831d do_reload
../../src/gcc/ira.c:5421
0x8079831d execute
../../src/gcc/ira.c:5592
Please submit a full bug report,
with preprocessed source if appropriate.
reduced with a cross compiler
typedef int UINT32;
union SHARC_REG {
int r;
float f;
};
class adsp21062_device {
SHARC_REG m_r[];
double fabs(double);
UINT32 m_astat;
void compute_fclip(int, int, int);
};
void adsp21062_device::compute_fclip(int, int, int ry) {
SHARC_REG r_alu;
r_alu.f = fabs(m_r[ry].f);
m_astat = r_alu.r ? 4 : r_alu.r;
}
$ s390x-linux-gnu-g++ -c -O sharc.i
sharc.i: In member function 'void adsp21062_device::compute_fclip(int, int,
int)':
sharc.i:16:1: internal compiler error: Floating point exception
}
^
Please submit a full bug report,
with preprocessed source if appropriate.
