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 E3B2F3858282; Wed, 27 Mar 2024 09:21:28 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org E3B2F3858282
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1711531288;
	bh=2/eHEMZiuP7hUaaVBrUUAFeG57ARDzH5WF6YBTNP7T0=;
	h=From:To:Subject:Date:In-Reply-To:References:From;
	b=TIKp7L4ukpH8ZRXVllgzCq1UMfhUMg1CcJWOseCMuHqtf9QHBGkMZu3eV/g0B2znj
	 6kwc0lCBD1sGfUduyEliNr5s2DTA8yaxuIhbV/MH1LD9rXbzUNzgKBFlk2rxa5p/E7
	 ebT68WjH9VWR36oYPdP6VB/X4ReTrmZeKGDsxcYQ=
From: "ubizjak at gmail dot com" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug target/114487] ICE when building libsdl2 on -mfpmath=sse x86
 with LTO
Date: Wed, 27 Mar 2024 09:21:28 +0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: target
X-Bugzilla-Version: 14.0
X-Bugzilla-Keywords: ice-on-valid-code, lto
X-Bugzilla-Severity: normal
X-Bugzilla-Who: ubizjak at gmail dot com
X-Bugzilla-Status: NEW
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: 
Message-ID: <bug-114487-4-nc1gRCz4Jc@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-114487-4@http.gcc.gnu.org/bugzilla/>
References: <bug-114487-4@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=3D114487

--- Comment #2 from Uro=C5=A1 Bizjak <ubizjak at gmail dot com> ---
(In reply to Richard Biener from comment #1)

> (insn 6 5 0 (set (reg/v:SF 99 [ gamma ])
>         (reg:SF 20 xmm0)) "testautomation-testautomation_pixels.i":15:17 =
-1
>      (nil))
>=20
> I'm not sure what's wrong - looks like a target issue to me.

We are working with SFmode, so -msse is enough to trigger the bug.

This is known issue. GCC assumes that at least moves of all hard registers =
are
working, which is not the case when LTO-compiling
testautomation-testautomation_pixels.i with SDL_test_fuzzer.o (that enables=
 and
uses XMM registers via -msse -mfpmath=3Dsse).

It looks to me that the compiler hits this part in function_value_32 when
LTO-compiling:

--cut here--
  /* Override FP return register with %xmm0 for local functions when
     SSE math is enabled or for functions with sseregparm attribute.  */
  if ((fn || fntype) && (mode =3D=3D SFmode || mode =3D=3D DFmode))
    {
      int sse_level =3D ix86_function_sseregparm (fntype, fn, false);
      if (sse_level =3D=3D -1)
        {
          error ("calling %qD with SSE calling convention without "
                 "SSE/SSE2 enabled", fn);
          sorry ("this is a GCC bug that can be worked around by adding "
                 "attribute used to function called");
        }
      else if ((sse_level >=3D 1 && mode =3D=3D SFmode)
               || (sse_level =3D=3D 2 && mode =3D=3D DFmode))
        regno =3D FIRST_SSE_REG;
    }
--cut here--

Adding -msse to the second compilation works OK, removing -mfpmath=3Dsse fr=
om the
first compilation also works OK.=