[Bug lto/99447] [11 Regression] ICE (segfault) in lookup_page_table_entry

["hubicka at ucw dot cz" <gcc-bugzilla@gcc.gnu.org>]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=99447

--- Comment #11 from Jan Hubicka <hubicka at ucw dot cz> ---
> https://gcc.gnu.org/bugzilla/show_bug.cgi?id=99447
> 
> --- Comment #10 from Richard Biener <rguenth at gcc dot gnu.org> ---
> So like this.
> 
> diff --git a/gcc/cgraph.c b/gcc/cgraph.c
> index 80140757d16..447d9a920f7 100644
> --- a/gcc/cgraph.c
> +++ b/gcc/cgraph.c
> @@ -1854,6 +1854,9 @@ cgraph_node::release_body (bool keep_arguments)
>       needed to emit debug info later.  */
>    if (!used_as_abstract_origin && DECL_INITIAL (decl))
>      DECL_INITIAL (decl) = error_mark_node;
> +  /* Zap references to call stmts of our body.  */
> +  for (cgraph_edge *e = callees; e; e = e->next_callee)
> +    e->call_stmt = NULL;

Looks good but will also need to warlk indirect calls.
However I wonder in what situations it makes sense to release body but
keep cgraph edges?

Honza
>    release_function_body (decl);
>    if (lto_file_data)
>      {
> 
> -- 
> You are receiving this mail because:
> You are on the CC list for the bug.