From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756630Ab0CJNWJ (ORCPT ); Wed, 10 Mar 2010 08:22:09 -0500 Received: from hera.kernel.org ([140.211.167.34]:57558 "EHLO hera.kernel.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756607Ab0CJNWA (ORCPT ); Wed, 10 Mar 2010 08:22:00 -0500 Date: Wed, 10 Mar 2010 13:21:45 GMT From: tip-bot for Peter Zijlstra Cc: linux-kernel@vger.kernel.org, hpa@zytor.com, mingo@redhat.com, acme@infradead.org, a.p.zijlstra@chello.nl, tglx@linutronix.de, mingo@elte.hu Reply-To: mingo@redhat.com, hpa@zytor.com, linux-kernel@vger.kernel.org, a.p.zijlstra@chello.nl, acme@infradead.org, tglx@linutronix.de, mingo@elte.hu In-Reply-To: <20100305154129.042271287@chello.nl> References: <20100305154129.042271287@chello.nl> To: linux-tip-commits@vger.kernel.org Subject: [tip:perf/pebs] perf, x86: Robustify PEBS fixup Message-ID: Git-Commit-ID: a562b1871f7f7d2f3a835c3c1e07fa58d473cfb7 X-Mailer: tip-git-log-daemon MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Disposition: inline X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.2.3 (hera.kernel.org [127.0.0.1]); Wed, 10 Mar 2010 13:21:45 +0000 (UTC) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Commit-ID: a562b1871f7f7d2f3a835c3c1e07fa58d473cfb7 Gitweb: http://git.kernel.org/tip/a562b1871f7f7d2f3a835c3c1e07fa58d473cfb7 Author: Peter Zijlstra AuthorDate: Fri, 5 Mar 2010 16:29:14 +0100 Committer: Ingo Molnar CommitDate: Wed, 10 Mar 2010 13:23:35 +0100 perf, x86: Robustify PEBS fixup It turns out the LBR is massively unreliable on certain CPUs, so code the fixup a little more defensive to avoid crashing the kernel. Signed-off-by: Peter Zijlstra Cc: Arnaldo Carvalho de Melo Cc: paulus@samba.org Cc: eranian@google.com Cc: robert.richter@amd.com Cc: fweisbec@gmail.com LKML-Reference: <20100305154129.042271287@chello.nl> Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/perf_event_intel_ds.c | 21 +++++++++++++++++++-- 1 files changed, 19 insertions(+), 2 deletions(-) diff --git a/arch/x86/kernel/cpu/perf_event_intel_ds.c b/arch/x86/kernel/cpu/perf_event_intel_ds.c index a67fff1..e7ac517 100644 --- a/arch/x86/kernel/cpu/perf_event_intel_ds.c +++ b/arch/x86/kernel/cpu/perf_event_intel_ds.c @@ -399,10 +399,23 @@ static int intel_pmu_pebs_fixup_ip(struct pt_regs *regs) if (!x86_pmu.intel_cap.pebs_trap) return 1; + /* + * No LBR entry, no basic block, no rewinding + */ if (!cpuc->lbr_stack.nr || !from || !to) return 0; - if (ip < to) + /* + * Basic blocks should never cross user/kernel boundaries + */ + if (kernel_ip(ip) != kernel_ip(to)) + return 0; + + /* + * unsigned math, either ip is before the start (impossible) or + * the basic block is larger than 1 page (sanity) + */ + if ((ip - to) > PAGE_SIZE) return 0; /* @@ -420,7 +433,7 @@ static int intel_pmu_pebs_fixup_ip(struct pt_regs *regs) old_to = to; if (!kernel_ip(ip)) { - int bytes, size = min_t(int, MAX_INSN_SIZE, ip - to); + int bytes, size = MAX_INSN_SIZE; bytes = copy_from_user_nmi(buf, (void __user *)to, size); if (bytes != size) @@ -440,6 +453,10 @@ static int intel_pmu_pebs_fixup_ip(struct pt_regs *regs) return 1; } + /* + * Even though we decoded the basic block, the instruction stream + * never matched the given IP, either the TO or the IP got corrupted. + */ return 0; }