https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114043
Richard Biener <rguenth at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Ever confirmed|0 |1
CC| |hubicka at gcc dot gnu.org
Last reconfirmed| |2024-02-22
Status|UNCONFIRMED |NEW
--- Comment #1 from Richard Biener <rguenth at gcc dot gnu.org> ---
Confirmed. Doesn't happen w/o -fstrict-enums (huh).
(gdb) p debug_tree (t)
<function_decl 0x7ffff71c0600 __ct_comp
type <method_type 0x7ffff71bdf18
type <void_type 0x7ffff702ff18 void VOID
align:8 warn_if_not_align:0 symtab:0 alias-set -1 canonical-type
0x7ffff702ff18
pointer_to_this <pointer_type 0x7ffff7036000>>
QI
size <integer_cst 0x7ffff70292b8 constant 8>
unit-size <integer_cst 0x7ffff70292d0 constant 1>
align:8 warn_if_not_align:0 symtab:0 alias-set -1 canonical-type
0x7ffff71bdf18 method basetype <record_type 0x7ffff71bdbd0 OverflowSafeInt>
arg-types <tree_list 0x7ffff71b7de8 value <pointer_type 0x7ffff71da9d8>
chain <tree_list 0x7ffff71b7dc0 value <integer_type 0x7ffff702f738
long int>
chain <tree_list 0x7ffff7024d98 value <void_type 0x7ffff702ff18
void>>>>
pointer_to_this <pointer_type 0x7ffff71da540>>
addressable used nothrow public static weak autoinline decl_5 QI t.ii:3:3
align:16 warn_if_not_align:0 context <translation_unit_decl 0x7ffff7018168
t.ii> initial <error_mark 0x7ffff70291b0> chain <type_decl 0x7ffff703ec78
OverflowSafeInt>>
that's the OverflowSafeInt(long) {} CTOR. We have cgraph_node::get
(t)->definition == 0.
It might be the --param disables some important cgraph initialization
(edges rebuild?). But not sure where -fstric-enums comes into play.
