From owner-svn-src-projects@freebsd.org Thu Feb 2 23:01:31 2017
Return-Path: <owner-svn-src-projects@freebsd.org>
Delivered-To: svn-src-projects@mailman.ysv.freebsd.org
Received: from mx1.freebsd.org (mx1.freebsd.org
[IPv6:2001:1900:2254:206a::19:1])
by mailman.ysv.freebsd.org (Postfix) with ESMTP id 2A094CCECEB
for <svn-src-projects@mailman.ysv.freebsd.org>;
Thu, 2 Feb 2017 23:01:31 +0000 (UTC) (envelope-from dim@FreeBSD.org)
Received: from repo.freebsd.org (repo.freebsd.org
[IPv6:2610:1c1:1:6068::e6a:0])
(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits))
(Client did not present a certificate)
by mx1.freebsd.org (Postfix) with ESMTPS id D5A6412EC;
Thu, 2 Feb 2017 23:01:30 +0000 (UTC) (envelope-from dim@FreeBSD.org)
Received: from repo.freebsd.org ([127.0.1.37])
by repo.freebsd.org (8.15.2/8.15.2) with ESMTP id v12N1Tov032358;
Thu, 2 Feb 2017 23:01:29 GMT (envelope-from dim@FreeBSD.org)
Received: (from dim@localhost)
by repo.freebsd.org (8.15.2/8.15.2/Submit) id v12N1T9k032357;
Thu, 2 Feb 2017 23:01:29 GMT (envelope-from dim@FreeBSD.org)
Message-Id: <201702022301.v12N1T9k032357@repo.freebsd.org>
X-Authentication-Warning: repo.freebsd.org: dim set sender to dim@FreeBSD.org
using -f
From: Dimitry Andric <dim@FreeBSD.org>
Date: Thu, 2 Feb 2017 23:01:29 +0000 (UTC)
To: src-committers@freebsd.org, svn-src-projects@freebsd.org
Subject: svn commit: r313110 -
projects/clang400-import/contrib/llvm/lib/Analysis
X-SVN-Group: projects
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
X-BeenThere: svn-src-projects@freebsd.org
X-Mailman-Version: 2.1.23
Precedence: list
List-Id: "SVN commit messages for the src " projects"
tree" <svn-src-projects.freebsd.org>
List-Unsubscribe: <https://lists.freebsd.org/mailman/options/svn-src-projects>,
<mailto:svn-src-projects-request@freebsd.org?subject=unsubscribe>
List-Archive: <http://lists.freebsd.org/pipermail/svn-src-projects/>
List-Post: <mailto:svn-src-projects@freebsd.org>
List-Help: <mailto:svn-src-projects-request@freebsd.org?subject=help>
List-Subscribe: <https://lists.freebsd.org/mailman/listinfo/svn-src-projects>,
<mailto:svn-src-projects-request@freebsd.org?subject=subscribe>
X-List-Received-Date: Thu, 02 Feb 2017 23:01:31 -0000
Author: dim
Date: Thu Feb 2 23:01:29 2017
New Revision: 313110
URL: https://svnweb.freebsd.org/changeset/base/313110
Log:
Pull in r293773 from upstream llvm trunk (by Sanjay Patel):
[ValueTracking] avoid crashing from bad assumptions (PR31809)
A program may contain llvm.assume info that disagrees with other
analysis. This may be caused by UB in the program, so we must not
crash because of that.
As noted in the code comments:
https://llvm.org/bugs/show_bug.cgi?id=31809
...we can do better, but this at least avoids the assert/crash in the
bug report.
Differential Revision: https://reviews.llvm.org/D29395
This fixes an assertion when building editors/emacs-devel.
PR: 216614
Modified:
projects/clang400-import/contrib/llvm/lib/Analysis/ValueTracking.cpp
Modified: projects/clang400-import/contrib/llvm/lib/Analysis/ValueTracking.cpp
==============================================================================
--- projects/clang400-import/contrib/llvm/lib/Analysis/ValueTracking.cpp Thu Feb 2 22:02:10 2017 (r313109)
+++ projects/clang400-import/contrib/llvm/lib/Analysis/ValueTracking.cpp Thu Feb 2 23:01:29 2017 (r313110)
@@ -781,6 +781,23 @@ static void computeKnownBitsFromAssume(c
APInt::getHighBitsSet(BitWidth, RHSKnownZero.countLeadingOnes());
}
}
+
+ // If assumptions conflict with each other or previous known bits, then we
+ // have a logical fallacy. This should only happen when a program has
+ // undefined behavior. We can't assert/crash, so clear out the known bits and
+ // hope for the best.
+
+ // FIXME: Publish a warning/remark that we have encountered UB or the compiler
+ // is broken.
+
+ // FIXME: Implement a stronger version of "I give up" by invalidating/clearing
+ // the assumption cache. This should indicate that the cache is corrupted so
+ // future callers will not waste time repopulating it with faulty assumptions.
+
+ if ((KnownZero & KnownOne) != 0) {
+ KnownZero.clearAllBits();
+ KnownOne.clearAllBits();
+ }
}
// Compute known bits from a shift operator, including those with a