From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 4396 invoked by alias); 20 Jan 2012 11:08:06 -0000 Received: (qmail 4380 invoked by uid 22791); 20 Jan 2012 11:08:05 -0000 X-SWARE-Spam-Status: No, hits=-2.8 required=5.0 tests=ALL_TRUSTED,AWL,BAYES_00,TW_ZJ X-Spam-Check-By: sourceware.org Received: from localhost (HELO gcc.gnu.org) (127.0.0.1) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Fri, 20 Jan 2012 11:07:52 +0000 From: "ubizjak at gmail dot com" To: gcc-bugs@gcc.gnu.org Subject: [Bug target/49257] -mfpmath=sse generates x87 instructions on 32 bits OS Date: Fri, 20 Jan 2012 11:31:00 -0000 X-Bugzilla-Reason: CC X-Bugzilla-Type: changed X-Bugzilla-Watch-Reason: None X-Bugzilla-Product: gcc X-Bugzilla-Component: target X-Bugzilla-Keywords: wrong-code X-Bugzilla-Severity: normal X-Bugzilla-Who: ubizjak at gmail dot com X-Bugzilla-Status: RESOLVED X-Bugzilla-Priority: P3 X-Bugzilla-Assigned-To: ubizjak at gmail dot com X-Bugzilla-Target-Milestone: --- X-Bugzilla-Changed-Fields: Status Resolution Message-ID: In-Reply-To: References: X-Bugzilla-URL: http://gcc.gnu.org/bugzilla/ Auto-Submitted: auto-generated Content-Type: text/plain; charset="UTF-8" MIME-Version: 1.0 Mailing-List: contact gcc-bugs-help@gcc.gnu.org; run by ezmlm Precedence: bulk List-Id: List-Archive: List-Post: List-Help: Sender: gcc-bugs-owner@gcc.gnu.org X-SW-Source: 2012-01/txt/msg02297.txt.bz2 http://gcc.gnu.org/bugzilla/show_bug.cgi?id=49257 Uros Bizjak changed: What |Removed |Added ---------------------------------------------------------------------------- Status|ASSIGNED |RESOLVED Resolution| |WONTFIX --- Comment #18 from Uros Bizjak 2012-01-20 11:06:16 UTC --- The only way this could work is to put __builtin_ia32_emms into the loop, after __builtin_ia32_movntq. -mfpmath=sse does not mean that x87 is disabled, only that equivalent arithmetic instructions use SSE instructions. If there is no equivalent SSE insn, x87 insn is used. IIRC, even Intel's Instruction set reference suggests to group FP and MMX insn together and put emms after MMX block. Since a substantial effort would be needed to fix this questionable corner case, and the test is violating recommended practice, I'm marking this PR as WONTFIX.