Return-path: Received: from mga02.intel.com ([134.134.136.20]:20125 "EHLO mga02.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1759817AbYDCPZ4 convert rfc822-to-8bit (ORCPT ); Thu, 3 Apr 2008 11:25:56 -0400 MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Subject: RE: iwlwifi/rfkill cause iwlwifi problem and slab corruption Date: Thu, 3 Apr 2008 08:24:59 -0700 Message-ID: (sfid-20080403_162632_448650_D4C480C6) In-Reply-To: <1207145976.3957.7.camel@johannes.berg> References: <1207145976.3957.7.camel@johannes.berg> From: "Chatre, Reinette" To: "Johannes Berg" Cc: "linux-wireless" , "Ivo van Doorn" Sender: linux-wireless-owner@vger.kernel.org List-ID: On Wednesday, April 02, 2008 7:20 AM, Johannes Berg wrote: > Hi, > > When I tried to rmmod iwl4965, iwlcore and mac80211, I got: > > [ 176.440255] iwl4965: No space for Tx > [ 176.440262] iwl4965: Error sending REPLY_CARD_STATE_CMD: > enqueue_hcmd failed: -28 > > I cannot explain this. > > Also, later, this happened came out of the kernel: > > [ 173.589186] > =============================================================== > ============== [ 173.589209] BUG kmalloc-2048: Poison overwritten [ > 173.589216] > --------------------------------------------------------------- > -------------- > [ 173.589219] > [ 173.589231] INFO: 0xc000000118fc8b23-0xc000000118fc8b23. > First byte 0x6a instead of 0x6b > [ 173.589240] INFO: Allocated in .rfkill_allocate+0x40/0xf4 > [rfkill] age=2367858 cpu=2 pid=1125 > [ 173.589273] INFO: Freed in .rfkill_release+0x24/0x4c > [rfkill] age=509691 cpu=0 pid=6294 > [ 173.589290] INFO: Slab 0xc000000008c61b00 used=14 > fp=0xc000000118fc8848 flags=0x40c3 > [ 173.589297] INFO: Object 0xc000000118fc8848 @offset=2120 > fp=0x0000000000000000 [ 173.589300] > [ 173.589305] Bytes b4 0xc000000118fc8838: 00 00 00 00 ff fb > 93 eb 5a 5a 5a 5a 5a 5a 5a 5a ....XX.XZZZZZZZZ > [ 173.589433] Object 0xc000000118fc8848: 6b 6b 6b 6b 6b 6b > 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b kkkkkkkkkkkkkkkk > [ 173.589561] Object 0xc000000118fc8858: 6b 6b 6b 6b 6b 6b > 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b kkkkkkkkkkkkkkkk > [ 173.589689] Object 0xc000000118fc8868: 6b 6b 6b 6b 6b 6b > 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b kkkkkkkkkkkkkkkk > [ 173.589818] Object 0xc000000118fc8878: 6b 6b 6b 6b 6b 6b > 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b kkkkkkkkkkkkkkkk > [ 173.589949] Object 0xc000000118fc8888: 6b 6b 6b 6b 6b 6b > 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b kkkkkkkkkkkkkkkk > [ 173.590072] Object 0xc000000118fc8898: 6b 6b 6b 6b 6b 6b > 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b kkkkkkkkkkkkkkkk > [ 173.590133] Object 0xc000000118fc88a8: 6b 6b 6b 6b 6b 6b > 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b kkkkkkkkkkkkkkkk > [ 173.590133] Object 0xc000000118fc88b8: 6b 6b 6b 6b 6b 6b > 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b kkkkkkkkkkkkkkkk > [ 173.590133] Redzone 0xc000000118fc9048: bb bb bb bb bb bb > bb bb XXXXXXXX > [ 173.590133] Padding 0xc000000118fc9088: 5a 5a 5a 5a 5a 5a > 5a 5a ZZZZZZZZ > [ 173.590133] Call Trace: > [ 173.590133] [c000000115c3f290] [c00000000000fa0c] > .show_stack+0x78/0x1b0 (unreliable) > [ 173.590133] [c000000115c3f350] [c0000000003e154c] > .dump_stack+0x20/0x34 [ 173.590133] [c000000115c3f3d0] > [c0000000000d6aa0] .print_trailer+0x1a4/0x1d0 [ 173.590133] > [c000000115c3f470] [c0000000000d6bd8] > .check_bytes_and_report+0x10c/0x17c [ 173.590133] [c000000115c3f530] > [c0000000000d6da8] .check_object+0xfc/0x26c [ 173.590133] > [c000000115c3f5e0] [c0000000000d8ddc] .__slab_alloc+0x600/0x744 [ > 173.590133] [c000000115c3f6b0] [c0000000000d8fc0] > .kmem_cache_alloc+0xa0/0x12c [ 173.590133] [c000000115c3f760] > [c000000000348a10] .__scm_send+0xd8/0x330 [ 173.590133] > [c000000115c3f820] [c0000000003ca484] .unix_stream_sendmsg+0xd8/0x46c > [ 173.590133] [c000000115c3f950] [c00000000033baac] > .sock_sendmsg+0xc4/0xfc [ 173.590133] [c000000115c3fb50] > [c00000000033bce0] .sys_sendmsg+0x1fc/0x294 [ 173.590133] > [c000000115c3fd80] [c000000000361ddc] > .compat_sys_socketcall+0x1f4/0x238 [ 173.590133] [c000000115c3fe30] > [c000000000007754] syscall_exit+0x0/0x40 [ 173.590133] FIX > kmalloc-2048: Restoring 0xc000000118fc8b23-0xc000000118fc8b23=0x6b [ > 173.590133] [ 173.590133] FIX kmalloc-2048: Marking all objects used > > And, uh, don't ask me why my time seems to be going backwards. yeah ... we started seeing this here also. We are debugging it. May be related to us calling rfkill_unregister as well as rfkill_free, which we should not. Thanks Reinette