Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S265091AbTGBQNb (ORCPT ); Wed, 2 Jul 2003 12:13:31 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S265101AbTGBQNb (ORCPT ); Wed, 2 Jul 2003 12:13:31 -0400 Received: from franka.aracnet.com ([216.99.193.44]:57564 "EHLO franka.aracnet.com") by vger.kernel.org with ESMTP id S265091AbTGBQN3 (ORCPT ); Wed, 2 Jul 2003 12:13:29 -0400 Date: Wed, 02 Jul 2003 09:27:42 -0700 From: "Martin J. Bligh" To: linux-kernel Subject: [Bug 863] New: cat /proc/buddyinfo + netstat -a kills machine Message-ID: <25670000.1057163262@[10.10.2.4]> X-Mailer: Mulberry/2.2.1 (Linux/x86) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2822 Lines: 79 Summary: cat /proc/buddyinfo + netstat -a kills machine Kernel Version: 2.5.73-bk10 Status: NEW Severity: blocking Owner: akpm@digeo.com Submitter: slpratt@us.ibm.com Distribution:SLES Hardware Environment:8way 900Mhz PIII Problem Description: on 2.5.73-bk10 as well as 2.5.73-mm3, issuing netsta -a followed by cat /proc/buddy infor causes segfaults and locks up the system. Repeated issuing of netstat -a prodices the following trap: Unable to handle kernel paging request at virtual address 00c1a000 printing eip: c01405ed *pde = 00000000 Oops: 0000 [#1] CPU: 1 EIP: 0060:[] Not tainted EFLAGS: 00010086 EIP is at kfree+0x3d/0x70 eax: 00000001 ebx: 00c1a000 ecx: f65b85c0 edx: c19f6009 esi: 00000100 edi: 00000206 ebp: f57c9720 esp: f5b13f4c ds: 007b es: 007b ss: 0068 Process netstat (pid: 2545, threadinfo=f5b12000 task=f58bb960) Stack: f68fd438 00000000 f68fd420 f65b85c0 f6caead0 c0173fd5 00000100 f65b85c0 f65b85c0 f7fdeb60 f6caead0 c0156c21 f6caead0 f65b85c0 f65b85c0 0805f038 f6aaf440 00000000 c01552a9 f65b85c0 f6aaf440 f65b85c0 0805f038 0805f038 Call Trace: [] seq_release_private+0x25/0x48 [] __fput+0xb1/0xc0 [] filp_close+0x99/0xd0 [] sys_close+0x5e/0x80 [] syscall_call+0x7/0xb Cating of /proc/buddy info produces: <1>Unable to handle kernel paging request at virtual address 08c19ec0 printing eip: c013d5e0 *pde = 00000000 Oops: 0000 [#2] CPU: 0 EIP: 0060:[] Not tainted EFLAGS: 00010006 EIP is at frag_show+0xd0/0x140 eax: 08c19ec0 ebx: c047e3f8 ecx: 08c19ec0 edx: c19f6009 esi: 000006b7 edi: c047e1f8 ebp: c047e3f8 esp: f60b7f0c ds: 007b es: 007b ss: 0068 Process cat (pid: 5355, threadinfo=f60b6000 task=f52aa0c0) Stack: f68fd520 c041caa6 00000000 c041ca07 00000078 00000000 00000206 c047e180 c047c980 f68fd520 c047c980 00000000 c01737a2 f68fd520 c047c980 00002000 f68fd538 00000000 c014a082 f6ba1dc0 00000000 00000000 00000000 c66d5a20 Call Trace: [] seq_read+0x102/0x300 [] do_brk+0x142/0x220 [] vfs_read+0xbe/0x130 [] sys_read+0x42/0x70 [] syscall_call+0x7/0xb Code: 8b 01 89 ca 46 89 c1 0f 18 00 90 39 da 75 f1 c7 44 24 04 a6 /autobench/scripts/getsysinfo: line 133: 5355 Segmentation fault cat /proc/buddyinfo > $2/proc/buddyinfo.$1$RUN_SUFFIX Steps to reproduce: run netstat -a a few times until it traps. then cat /proc/buddyinfo and hte system will be locked up. - 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/