From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755545Ab3LCXPe (ORCPT ); Tue, 3 Dec 2013 18:15:34 -0500 Received: from relay3.sgi.com ([192.48.152.1]:50191 "EHLO relay.sgi.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1754950Ab3LCXPc (ORCPT ); Tue, 3 Dec 2013 18:15:32 -0500 To: linux-kernel@vger.kernel.org Subject: [PATCH] x86, UV: null pointer in uv_flush_tlb_others() Cc: mingo@elte.hu, x86@kernel.org Message-Id: From: cpw Date: Tue, 03 Dec 2013 17:15:30 -0600 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org The SGI UV tlb shootdown code panics the system with a null pointer deference if 'nobau' is specified on the boot commandline. uv_flush_tlb_other() gets called for every flush, whether the BAU is disabled or not. It should not be keeping the s_enters statistic while the BAU is disabled. The panic occurs because, during initialization, init_per_cpu_tunables() does not set the bcp->statp pointer if nobau was specified. The code is the same in the 3.12.2 tree. Cc: # 3.12.x Signed-off-by: Cliff Wickman --- arch/x86/platform/uv/tlb_uv.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) Index: linux/arch/x86/platform/uv/tlb_uv.c =================================================================== --- linux.orig/arch/x86/platform/uv/tlb_uv.c +++ linux/arch/x86/platform/uv/tlb_uv.c @@ -1070,12 +1070,13 @@ const struct cpumask *uv_flush_tlb_other unsigned long status; bcp = &per_cpu(bau_control, cpu); - stat = bcp->statp; - stat->s_enters++; if (bcp->nobau) return cpumask; + stat = bcp->statp; + stat->s_enters++; + if (bcp->busy) { descriptor_status = read_lmmr(UVH_LB_BAU_SB_ACTIVATION_STATUS_0);