https://gcc.gnu.org/bugzilla/show_bug.cgi?id=109478
Richard Biener <rguenth at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Last reconfirmed| |2023-04-12
CC| |law at gcc dot gnu.org
Status|UNCONFIRMED |NEW
Ever confirmed|0 |1
--- Comment #1 from Richard Biener <rguenth at gcc dot gnu.org> ---
Confirmed.
t.ii:2:9: warning: width of 'a::b' exceeds its type
2 | short b : -1ULL;
| ^
Program received signal SIGSEGV, Segmentation fault.
0x00000000011d5bf4 in initialize_argument_information (num_actuals=1,
args=0x3398c30, args_size=0x7fffffffc750, n_named_args=1, exp=<call_expr
0x7ffff71c3930>, struct_value_addr_value=<tree 0x0>, fndecl=<function_decl
0x7ffff7329200 c>, fntype=<function_type 0x7ffff7205e70>, args_so_far=...,
reg_parm_stack_space=64, old_stack_level=0x7fffffffc6e0,
old_pending_adj=0x7fffffffc6d8, must_preallocate=0x7fffffffc708,
ecf_flags=0x7fffffffc700, may_tailcall=0x7fffffffc78f, call_from_thunk_p=false)
at /space/rguenther/src/gcc/gcc/calls.cc:1519
1519 && XEXP (XVECEXP (args[i].reg, 0, 0), 0) == 0)
Missing separate debuginfos, use: zypper install
libgmp10-debuginfo-6.1.2-4.9.1.x86_64 libmpc3-debuginfo-1.1.0-1.47.x86_64
libmpfr6-debuginfo-4.0.2-3.3.1.x86_64
libzstd1-debuginfo-1.5.0-150400.3.3.1.x86_64
(gdb) bt
#0 0x00000000011d5bf4 in initialize_argument_information (num_actuals=1,
args=0x3398c30, args_size=0x7fffffffc750, n_named_args=1,
exp=<call_expr 0x7ffff71c3930>, struct_value_addr_value=<tree 0x0>,
fndecl=<function_decl 0x7ffff7329200 c>,
fntype=<function_type 0x7ffff7205e70>, args_so_far=...,
reg_parm_stack_space=64, old_stack_level=0x7fffffffc6e0,
old_pending_adj=0x7fffffffc6d8, must_preallocate=0x7fffffffc708,
ecf_flags=0x7fffffffc700, may_tailcall=0x7fffffffc78f,
call_from_thunk_p=false) at /space/rguenther/src/gcc/gcc/calls.cc:1519
#1 0x00000000011da16d in expand_call (exp=<call_expr 0x7ffff71c3930>,
target=0x0, ignore=1) at /space/rguenther/src/gcc/gcc/calls.cc:2973
#2 0x00000000013af7cf in expand_expr_real_1 (exp=<call_expr 0x7ffff71c3930>,
target=0x0, tmode=E_VOIDmode, modifier=EXPAND_NORMAL, alt_rtl=0x0,
inner_reference_p=false) at /space/rguenther/src/gcc/gcc/expr.cc:11868
#3 0x00000000013a171e in expand_expr_real (exp=<call_expr 0x7ffff71c3930>,
target=0x7ffff71cb490, tmode=E_VOIDmode, modifier=EXPAND_NORMAL,
alt_rtl=0x0, inner_reference_p=false)
at /space/rguenther/src/gcc/gcc/expr.cc:9000
#4 0x00000000011cc17f in expand_expr (exp=<call_expr 0x7ffff71c3930>,
target=0x7ffff71cb490, mode=E_VOIDmode, modifier=EXPAND_NORMAL)
at /space/rguenther/src/gcc/gcc/expr.h:310
#5 0x00000000011f3928 in expand_call_stmt (stmt=0x7ffff73371b0)
at /space/rguenther/src/gcc/gcc/cfgexpand.cc:2831
(gdb) p debug_gimple_stmt (stmt)
# .MEM_3 = VDEF <.MEM_2>
c (D.2324);
(gdb) p debug_tree (exp)
<call_expr 0x7ffff71c3930
...
arg:0 <var_decl 0x7ffff7337120 D.2324
type <record_type 0x7ffff731bd20 a cxx-odr-p type_5 BLK
size <integer_cst 0x7ffff7338030 constant public overflow 0>
unit-size <integer_cst 0x7ffff732efa8 constant 2305843009213693952>
align:64 warn_if_not_align:0 symtab:0 alias-set 1 canonical-type
0x7ffff731bd20 fields <function_decl 0x7ffff7329c00 __dt > context
<translation_unit_decl 0x7ffff71d7000 t.ii>
full-name "struct a"
X() X(constX&) this=(X&) n_parents=0 use_template=0
interface-unknown
pointer_to_this <pointer_type 0x7ffff731b7e0> reference_to_this
<reference_type 0x7ffff731b690> chain <type_decl 0x7ffff7314688 a>>
used ignored BLK t.ii:4:16 size <integer_cst 0x7ffff7338030 overflow 0>
unit-size <integer_cst 0x7ffff732efa8 2305843009213693952>
align:64 warn_if_not_align:0 context <function_decl 0x7ffff7329200 c>
(mem/c:BLK (reg/f:DI 63 virtual-stack-vars) [1 D.2324+0
S2305843009213693952 A64])>
and the RTL for the argument is
(parallel:BLK [])
ick. pa_function_arg runs into
9786 arg_size = pa_function_arg_size (mode, type);
9800 if (arg_size > 1)
(gdb) p arg_size
$7 = 0
so isn't able to decipher things down to a "valid" argument spec. Note
above for the argument type we have TYPE_SIZE == 0 but a very
large TYPE_SIZE_UNIT.
One "obvious" mistake is to use 'int arg_size' for the HOST_WIDE_INT
pa_function_arg_size return value. Adjusting also downstream variable
types helps to some extent but then we ICE in
during RTL pass: dwarf2
t.ii: In function 'void c(...)':
t.ii:4:23: internal compiler error: in dwarf2out_frame_debug_expr, at
dwarf2cfi.cc:1960
4 | void c(...) { c(a()); }
| ^
0x12bd9d2 dwarf2out_frame_debug_expr
/space/rguenther/src/gcc/gcc/dwarf2cfi.cc:1960
0x12bea15 dwarf2out_frame_debug
/space/rguenther/src/gcc/gcc/dwarf2cfi.cc:2367
0x12bf81b scan_insn_after
/space/rguenther/src/gcc/gcc/dwarf2cfi.cc:2726
0x12bfe3c scan_trace
seeing
(set (reg:DI 1 %r1)
(plus:DI (reg/f:DI 30 %r30)
(const_int 4611686018427379840 [0x3fffffffffffe080])))
which probably means PA would need some alternate form when the
offset gets large. The following is what's required to get around
the first ICE:
diff --git a/gcc/config/pa/pa.cc b/gcc/config/pa/pa.cc
index 3f91ebce603..66371f7ced9 100644
--- a/gcc/config/pa/pa.cc
+++ b/gcc/config/pa/pa.cc
@@ -9775,7 +9775,7 @@ pa_function_arg (cumulative_args_t cum_v, const
function_arg_info &arg)
machine_mode mode = arg.mode;
int max_arg_words = (TARGET_64BIT ? 8 : 4);
int alignment = 0;
- int arg_size;
+ HOST_WIDE_INT arg_size;
int fpr_reg_base;
int gpr_reg_base;
rtx retval;
@@ -9840,7 +9840,8 @@ pa_function_arg (cumulative_args_t cum_v, const
function_arg_info &arg)
justified data on a big-endian target. */
rtx loc[8];
- int i, offset = 0, ub = arg_size;
+ int i;
+ HOST_WIDE_INT offset = 0, ub = arg_size;
/* Align the base register. */
gpr_reg_base -= alignment;
