Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1758853Ab2EII6h (ORCPT ); Wed, 9 May 2012 04:58:37 -0400 Received: from mx4-phx2.redhat.com ([209.132.183.25]:36686 "EHLO mx4-phx2.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757903Ab2EII6g (ORCPT ); Wed, 9 May 2012 04:58:36 -0400 Date: Wed, 09 May 2012 04:58:34 -0400 (EDT) From: Zhouping Liu To: linux-mm@kvack.org Cc: linux-kernel@vger.kernel.org Subject: mm: move_pages syscall can't return ENOENT when pages are not present Message-ID: <85e08d38-234a-4bc6-8c4f-6c92b50dc9b1@zmail13.collab.prod.int.phx2.redhat.com> In-Reply-To: <50e8b720-2459-4cf4-bfbd-fcc4cd408249@zmail13.collab.prod.int.phx2.redhat.com> Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit MIME-Version: 1.0 X-Originating-IP: [10.66.112.86] X-Mailer: Zimbra 7.1.2_GA_3268 (ZimbraWebClient - FF3.0 (Linux)/7.1.2_GA_3268) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2367 Lines: 93 hi, all Recently, I found an error in move_pages syscall: depending on move_pages(2), when page is not present, it should fail with ENOENT, in fact, it's ok without any errno. the following reproducer can easily reproduce the issue, suggest you get more details by strace. inside reproducer, I try to move a non-exist page from node 1 to node 0. I have tested it on the latest kernel 3.4-rc5 with 2 and 4 numa nodes. [zliu@ZhoupingLiu ~]$ gcc -o reproducer reproducer.c -lnuma [zliu@ZhoupingLiu ~]$ ./reproducer from_node is 1, to_node is 0 ERROR: move_pages expected FAIL. I'm not in mail list, please CC me. /* * Copyright (C) 2012 Red Hat, Inc. * * This work is licensed under the terms of the GNU GPL, version 2. See * the COPYING file in the top-level directory. * * Compiled: gcc -o reproducer reproducer.c -lnuma * Description: * it's designed to check move_pages syscall, when * page is not present, it should fail with ENOENT. */ #include #include #include #include #include #include #include #include #define TEST_PAGES 4 int main(int argc, char **argv) { void *pages[TEST_PAGES]; int onepage; int nodes[TEST_PAGES]; int status, ret; int i, from_node = 1, to_node = 0; onepage = getpagesize(); for (i = 0; i < TEST_PAGES - 1; i++) { pages[i] = numa_alloc_onnode(onepage, from_node); nodes[i] = to_node; } nodes[TEST_PAGES - 1] = to_node; /* * the follow page is not available, also not aligned, * depend on move_pages(2), it can't be moved, and should * return ENOENT errno. */ pages[TEST_PAGES - 1] = pages[TEST_PAGES - 2] - onepage * 4 + 1; printf("from_node is %u, to_node is %u\n", from_node, to_node); ret = move_pages(0, TEST_PAGES, pages, nodes, &status, MPOL_MF_MOVE); if (ret == -1) { if (errno != ENOENT) perror("move_pages expected ENOENT errno, but it's"); else printf("Succeed\n"); } else { printf("ERROR: move_pages expected FAIL.\n"); } for (i = 0; i < TEST_PAGES; i++) numa_free(pages[i], onepage); return 0; } -- Thanks, Zhouping -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/