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 9C1C53858C00; Tue, 11 Jul 2023 18:13:20 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 9C1C53858C00
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1689099200;
	bh=oRCb6RRD6F5AYzYRW8cO0FfJYAWwoZH3JfF83oc02CM=;
	h=From:To:Subject:Date:In-Reply-To:References:From;
	b=WmetZxoUaHPLOMh+BYSbIK7HcwlYLnXjOzXq5A+7W5PHdH52FeuPLbied1ZwmVdXf
	 rMJ58zG0lfIC41br83o0QeeGfao3M79hc+yHCPihpxq0Q7FsJO3TFw3DqAeJ5X+DZs
	 W1gxz9YqrywOSEJO0ZgZoOj3RUb7n4GZVtFFD+p4=
From: "admin@tho-otto.de" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug modula2/110126] Variables are reported as unused when only
 referenced by ASM statements
Date: Tue, 11 Jul 2023 18:13:19 +0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: modula2
X-Bugzilla-Version: 13.1.0
X-Bugzilla-Keywords: 
X-Bugzilla-Severity: normal
X-Bugzilla-Who: admin@tho-otto.de
X-Bugzilla-Status: REOPENED
X-Bugzilla-Resolution: 
X-Bugzilla-Priority: P3
X-Bugzilla-Assigned-To: gaius at gcc dot gnu.org
X-Bugzilla-Target-Milestone: ---
X-Bugzilla-Flags: 
X-Bugzilla-Changed-Fields: bug_status resolution
Message-ID: <bug-110126-4-L51Khv4tOo@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-110126-4@http.gcc.gnu.org/bugzilla/>
References: <bug-110126-4@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=3D110126

Thorsten Otto <admin@tho-otto.de> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |REOPENED
         Resolution|FIXED                       |---

--- Comment #14 from Thorsten Otto <admin@tho-otto.de> ---
And another problem i just encountered. In the following example:

DEFINITION MODULE foo;

PROCEDURE ConWS(str: ARRAY OF CHAR);

END foo.


IMPLEMENTATION MODULE foo;

FROM SYSTEM     IMPORT ADR, CARDINAL16, CARDINAL32, INTEGER16, ADDRESS,
INTEGER32, WORD;

FROM StrIO IMPORT WriteString, WriteLn;

TYPE strtmp =3D ARRAY [0..255] OF CHAR;

PROCEDURE str0(VAR dst: strtmp; src: ARRAY OF CHAR);
VAR i, len: CARDINAL;
BEGIN
  len :=3D HIGH(dst);
  FOR i :=3D 0 TO len DO
    dst[i] :=3D src[i];
  END;
END str0;


PROCEDURE trap_1_wl(n: INTEGER16; a: ADDRESS): [ INTEGER32 ];
VAR
    retvalue: INTEGER32;
BEGIN
    ASM VOLATILE("move.l %2,-(%%sp); move.w %1,-(%%sp); trap #1; addq.l
#6,%%sp; move.l %%d0,%0"
       : "=3Dr"(retvalue)
       : "g"(n), "r"(a)
       : "memory"
    );
    RETURN retvalue
END trap_1_wl;


PROCEDURE ConWS(str: ARRAY OF CHAR);
VAR s: strtmp;
BEGIN
  str0(s, str);
  trap_1_wl(9, ADR(s));
END ConWS;


END foo.

The call to str0 (in reality supposed to ensure the string is 0-terminated;
just simplified for the example) is now completely eliminated. The assembler
statement is of course only for m68k, but it can even be triggered for other
targets when just producing assembler output.

It does not happen when replacing the call to trap_1_wl eg. by WriteString.=