Re: [PATCH] c++: Fix up ICE in emit_support_tinfo_1 [PR109042]

[Jason Merrill <jason@redhat.com>]

On 3/7/23 04:07, Jakub Jelinek wrote:
> Hi!
> 
> In my recent rtti.cc change I assumed when emitting the support tinfos
> that the tinfos for the fundamental types haven't been created yet.
> Normally (in libsupc++.a (fundamental_type_info.o)) that is the case,
> but as can be seen on the testcase, one can violate it by using typeid
> etc. in the same TU and do it before ~__fundamental_type_info ()
> definition.
> 
> The following patch fixes that by popping from unemitted_tinfo_decls
> only in the normal case when it is there, and treating non-NULL
> DECL_INITIAL on a tinfo node as indication that emit_tinfo_decl has
> processed it already.
> 
> Bootstrapped/regtested on x86_64-linux and i686-linux, ok for trunk?
> 
> 2023-03-07  Jakub Jelinek  <jakub@redhat.com>
> 
> 	PR c++/109042
> 	* rtti.cc (emit_support_tinfo_1): Don't assert that last
> 	unemitted_tinfo_decls element is tinfo, instead pop from it only in
> 	that case.
> 	* decl2.cc (c_parse_final_cleanups): Don't call emit_tinfo_decl
> 	for unemitted_tinfO_decls which have already non-NULL DECL_INITIAL.
> 
> 	* g++.dg/rtti/pr109042.C: New test.
> 
> --- gcc/cp/rtti.cc.jj	2023-03-03 00:34:52.028567946 +0100
> +++ gcc/cp/rtti.cc	2023-03-06 19:06:27.433307136 +0100
> @@ -1581,10 +1581,10 @@ emit_support_tinfo_1 (tree bltn)
>         /* Emit it right away if not emitted already.  */
>         if (DECL_INITIAL (tinfo) == NULL_TREE)
>   	{
> -	  gcc_assert (unemitted_tinfo_decls->last () == tinfo);
>   	  bool ok = emit_tinfo_decl (tinfo);
>   	  gcc_assert (ok);
> -	  unemitted_tinfo_decls->pop ();
> +	  if (unemitted_tinfo_decls->last () == tinfo)
> +	    unemitted_tinfo_decls->pop ();

So if it's not last we'll leave it in the vec, even though it is no 
longer unemitted, and let c_parse_final_cleanups deal with removing it? 
  That could use a comment.  OK with that change.

>   	}
>       }
>   }
> --- gcc/cp/decl2.cc.jj	2023-01-18 16:11:47.053213397 +0100
> +++ gcc/cp/decl2.cc	2023-03-06 19:07:16.830582984 +0100
> @@ -4982,7 +4982,7 @@ c_parse_final_cleanups (void)
>   	 get emitted.  */
>         for (i = unemitted_tinfo_decls->length ();
>   	   unemitted_tinfo_decls->iterate (--i, &t);)
> -	if (emit_tinfo_decl (t))
> +	if (DECL_INITIAL (t) || emit_tinfo_decl (t))
>   	  {
>   	    reconsider = true;
>   	    unemitted_tinfo_decls->unordered_remove (i);
> --- gcc/testsuite/g++.dg/rtti/pr109042.C.jj	2023-03-06 19:11:06.995208812 +0100
> +++ gcc/testsuite/g++.dg/rtti/pr109042.C	2023-03-06 19:10:59.117324298 +0100
> @@ -0,0 +1,20 @@
> +// PR c++/109042
> +// { dg-do compile }
> +
> +namespace std { class type_info {}; }
> +
> +std::type_info
> +foo ()
> +{
> +  return typeid (void);
> +}
> +
> +namespace __cxxabiv1 {
> +  struct __fundamental_type_info {
> +    virtual ~__fundamental_type_info ();
> +  };
> +
> +  __fundamental_type_info::~__fundamental_type_info ()
> +  {
> +  }
> +}
> 
> 	Jakub
>