From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752633AbdGRVlO (ORCPT ); Tue, 18 Jul 2017 17:41:14 -0400 Received: from mga11.intel.com ([192.55.52.93]:6637 "EHLO mga11.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752431AbdGRVlM (ORCPT ); Tue, 18 Jul 2017 17:41:12 -0400 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.40,378,1496127600"; d="scan'208";a="1152837438" From: Luis Felipe Sandoval Castro To: linux-mm@kvack.org, linux-kernel@vger.kernel.org Cc: Luis Felipe Sandoval Castro Subject: [PATCH v1] mm/mempolicy.c: Fix get_nodes() off-by-one error. Date: Tue, 18 Jul 2017 08:39:24 -0500 Message-Id: <1500385164-11062-1-git-send-email-luis.felipe.sandoval.castro@intel.com> X-Mailer: git-send-email 1.8.3.1 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org set_mempolicy() and mbind() take as argument a pointer to a bit mask (nodemask) and the number of bits in the mask the kernel will use (maxnode), among others. For instace on a system with 2 NUMA nodes valid masks are: 0b00, 0b01, 0b10 and 0b11 it's clear maxnode=2, however an off-by-one error in get_nodes() the function that copies the node mask from user space requires users to pass maxnode = 3 in this example and maxnode = actual_maxnode + 1 in the general case. This patch fixes such error. Signed-off-by: Luis Felipe Sandoval Castro --- mm/mempolicy.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/mm/mempolicy.c b/mm/mempolicy.c index d911fa5..5274e9d2 100644 --- a/mm/mempolicy.c +++ b/mm/mempolicy.c @@ -1208,11 +1208,10 @@ static int get_nodes(nodemask_t *nodes, const unsigned long __user *nmask, unsigned long nlongs; unsigned long endmask; - --maxnode; nodes_clear(*nodes); - if (maxnode == 0 || !nmask) + if (maxnode == 1 || !nmask) return 0; - if (maxnode > PAGE_SIZE*BITS_PER_BYTE) + if (maxnode - 1 > PAGE_SIZE * BITS_PER_BYTE) return -EINVAL; nlongs = BITS_TO_LONGS(maxnode); -- 1.8.3.1