From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <gcc-bugzilla@gcc.gnu.org>
Received: by sourceware.org (Postfix, from userid 48)
	id E12C13857B85; Sat, 18 Nov 2023 00:40:32 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org E12C13857B85
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1700268032;
	bh=O34Ia8EZOKHcIAHhe+K+eacc26CjtPoOpuc5P6rTBLg=;
	h=From:To:Subject:Date:In-Reply-To:References:From;
	b=h92S4FRmH9SUL97nmJlUUJ5JG3Rkgrg266IRV4ZppVvp7AGaXsNIvCCDbN+E2E96K
	 wOO/tWAVU7rwvRR7jELcPRwPSnMvosYYd/E4rZJA3UeXHVTcH1Wv0QZ1r9bvvir/CQ
	 acLBixFwfJbmiEIAw5RtPDjneMU71QxOV8opZ36I=
From: "nathanieloshead at gmail dot com" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug c++/112588] ICE in make_decl_rtl when returning str literal
 when string header imported in module
Date: Sat, 18 Nov 2023 00:40:31 +0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: c++
X-Bugzilla-Version: 14.0
X-Bugzilla-Keywords: 
X-Bugzilla-Severity: normal
X-Bugzilla-Who: nathanieloshead at gmail dot com
X-Bugzilla-Status: UNCONFIRMED
X-Bugzilla-Resolution: 
X-Bugzilla-Priority: P3
X-Bugzilla-Assigned-To: unassigned at gcc dot gnu.org
X-Bugzilla-Target-Milestone: ---
X-Bugzilla-Flags: 
X-Bugzilla-Changed-Fields: cc
Message-ID: <bug-112588-4-ixiUfaDIaJ@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-112588-4@http.gcc.gnu.org/bugzilla/>
References: <bug-112588-4@http.gcc.gnu.org/bugzilla/>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
X-Bugzilla-URL: http://gcc.gnu.org/bugzilla/
Auto-Submitted: auto-generated
MIME-Version: 1.0
List-Id: <gcc-bugs.sourceware.org>

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=3D112588

Nathaniel Shead <nathanieloshead at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |nathanieloshead at gmail d=
ot com

--- Comment #1 from Nathaniel Shead <nathanieloshead at gmail dot com> ---
Minimised (the actual offender here is the with std::allocator<char>):


  // test.h
  void f(int*);

  template <typename T>
  struct S {
    void g(int n) { f(&n); }
  };

  template struct S<void>;


  // a.cpp
  module;
  #include "test.h"
  export module test;


  // b.cpp
  #include "test.h"
  import test;


So far it seems the issue is that the PARM_DECL in the expression tree of t=
he
body of the instantiation for `S<void>::g` is a different node from the act=
ual
PARM_DECL in g's DECL_ARGUMENTS; the latter gets RTL but the former does no=
t.
The issue is in the deduplication logic for instantiations somewhere.

The following patch fixes this issue but causes other issues in the testsui=
te,
and I don't think this is the correct approach anyway:


diff --git a/gcc/cp/module.cc b/gcc/cp/module.cc
index 4f5b6e2747a..f2d191fc408 100644
--- a/gcc/cp/module.cc
+++ b/gcc/cp/module.cc
@@ -8302,9 +8302,7 @@ trees_in::decl_value ()
       if (TREE_CODE (inner) =3D=3D FUNCTION_DECL)
        {
          tree e_inner =3D STRIP_TEMPLATE (existing);
-         for (auto parm =3D DECL_ARGUMENTS (inner);
-              parm; parm =3D DECL_CHAIN (parm))
-           DECL_CONTEXT (parm) =3D e_inner;
+         DECL_ARGUMENTS (inner) =3D DECL_ARGUMENTS (e_inner);
        }

       /* And our result is the existing node.  */


(I was originally working on this after attempting to reduce PR99999.)=