https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85641
--- Comment #2 from Steve Kargl <sgk at troutmask dot apl.washington.edu> ---
On Thu, May 03, 2018 at 11:12:26PM +0000, kargl at gcc dot gnu.org wrote:
>
> #31 0x00000000008ad88d in gfc_code_walker (c=0x2ca231808,
> codefn=codefn@entry=0x8a90d0 <realloc_string_callback(gfc_code**, int*,
> void*)>,
> exprfn=exprfn@entry=0x8a6eb0 <dummy_expr_callback(gfc_expr**, int*,
> void*)>, data=data@entry=0x0) at ../../gcc/gcc/fortran/frontend-passes.c:4593
>
gfortran seems to be stuck in an infinite loop and slowly
swallowing memory.
(gdb) c
Continuing.
Breakpoint 1, gfc_code_walker (c=0x203dc5008,
codefn=codefn@entry=0x8a90d0 <realloc_string_callback(gfc_code**, int*,
void*)>,
exprfn=exprfn@entry=0x8a6eb0 <dummy_expr_callback(gfc_expr**, int*,
void*)>, data=data@entry=0x0) at ../../gcc/gcc/fortran/frontend-passes.c:4566
4566 for (; *c; c = &(*c)->next)
(gdb) c
Continuing.
Breakpoint 2, realloc_string_callback (c=0x203dc5008,
walk_subtrees=0x7fffffffd14c, data=0x0)
at ../../gcc/gcc/fortran/frontend-passes.c:248
248 if (co->op != EXEC_ASSIGN)
(gdb) list
4561
4562 int
4563 gfc_code_walker (gfc_code **c, walk_code_fn_t codefn, walk_expr_fn_t
exprfn,
4564 void *data)
4565 {
4566 for (; *c; c = &(*c)->next)
4567 {
4568 int walk_subtrees = 1;
4569 int result = codefn (c, &walk_subtrees, data);
4570 if (result)
:-(
