From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: by sourceware.org (Postfix, from userid 48) id C15C0384AB4E; Thu, 9 May 2024 13:55:55 +0000 (GMT) DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org C15C0384AB4E DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org; s=default; t=1715262955; bh=Eaha38+A646Fq722bgcb0VheguwmgNaXPu0esjxsKqQ=; h=From:To:Subject:Date:In-Reply-To:References:From; b=NdHDP0W/OsBCUXGsBViYgFJ7tc2R+djHrgGFY3Jy+UXS8R/W3WuPnmqQPs4GlJbdD 8ImNgcN9MZ5DrSWp8SCHOGJv62Z1Ixp3zIEgorkp6sP0QR/pcbBpQq5esOxGZ4xTbb SnnvCv2q6//wKxzTwkAtU/HpzdnsUOb6jqLAVnkA= From: "erhard_f at mailbox dot org" To: gcc-bugs@gcc.gnu.org Subject: [Bug target/113652] [14/15 regression] Failed bootstrap on ppc unrecognized opcode: `lfiwzx' with -mcpu=7450 Date: Thu, 09 May 2024 13:55:55 +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: assemble-failure, build X-Bugzilla-Severity: normal X-Bugzilla-Who: erhard_f at mailbox dot org X-Bugzilla-Status: ASSIGNED X-Bugzilla-Resolution: X-Bugzilla-Priority: P2 X-Bugzilla-Assigned-To: segher at gcc dot gnu.org X-Bugzilla-Target-Milestone: 14.2 X-Bugzilla-Flags: X-Bugzilla-Changed-Fields: Message-ID: In-Reply-To: References: 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: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=3D113652 --- Comment #27 from Erhard F. --- (In reply to Peter Bergner from comment #25) > I think this is the solution we want, meaning if the target we're building > supports VSX, then we'll build the IEEE128 bits, otherwise, we won't build > them. I think that is the only sane answer. >=20 > I also believe that if the user specifies a -mcpu=3D option (either impli= citly > or explicitly) that doesn't support VSX (eg, power4, or 7450, etc.) and t= hey > also explicitly use -mvsx, then we should emit an error message saying the > -mcpu=3D option doesn't support using -mvsx at the same time. Ditto for > -maltivec, -mmma, etc. We should not be silently enabling instruction > support over and above their -mcpu=3D selection just because its needed f= or > VSX/Altivec/MMA/etc. support. Currently we don't emit an error and inste= ad > silently enable generating instructions not supported by their -mcpu=3D o= ption. As user I would favor this approach. Because that way I get an error message I actually understand at the early stages of building a package. Not silently broken code and illegal opcode/unrecognized opcode messages later on which make me wonder where they come from when I specified -mcpu=3D7450 or -mcpu=3D970. Unrelated to libgcc but similar issues would be bug #97367 or my Gentoo Downstream bug report about nettle-3.9.1 (https://bugs.gentoo.org/920234).=