From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S261531AbVGTXTl (ORCPT ); Wed, 20 Jul 2005 19:19:41 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S261539AbVGTXTl (ORCPT ); Wed, 20 Jul 2005 19:19:41 -0400 Received: from torrent.CC.McGill.CA ([132.206.27.49]:13252 "EHLO torrent.cc.mcgill.ca") by vger.kernel.org with ESMTP id S261531AbVGTXTh (ORCPT ); Wed, 20 Jul 2005 19:19:37 -0400 Subject: kernel oops with x64_86 crossing 4Gb boundary From: Francois Pepin To: linux-kernel@vger.kernel.org Content-Type: text/plain Date: Wed, 20 Jul 2005 19:18:48 -0400 Message-Id: <1121901528.5121.21.camel@elm.mcb.mcgill.ca> Mime-Version: 1.0 X-Mailer: Evolution 2.0.4 (2.0.4-4) Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Hi, I am getting pseudo-random kernel oops on my Opteron box (Tyan Thunder K8W) with 4Gb RAM. I am running RedHad FC3 with kernel 2.6.11-1.35_FC3smp. It runs well with default BIOS settings, but only 3.5Gb RAM are visible. Using the "Software Memory Hole" option in the BIOS (version 3.02) remaps the last 2Gb to 4-6Gb. This causes kernel oops with various applications, generally killing them. I can reproduce it by loading lots of data into R (which caused the oops below). Various other applications have caused it too. Oops: 0000 [1] SMP CPU 1 Modules linked in: nfs lockd parport_pc lp parport autofs4 sunrpc pcmcia yenta_socket rsrc_nonstatic pcmcia_core dm_mod video button battery ac nvidia(U) md5 ipv6 ohci1394 ieee1394 ohci_hcd uhci_hcd ehci_hcd i2c_amd8111 i2c_core hw_random sata_sil libata scsi_mod snd_intel8x0 snd_ca0106 snd_ac97_codec snd_pcm_oss snd_mixer_oss snd_pcm snd_timer snd soundcore snd_page_alloc e1000 floppy ext3 jbd Pid: 5288, comm: R Tainted: P 2.6.11-1.35_FC3smp RIP: 0010:[] {pte_alloc_map+170} RSP: 0000:ffff8101553f5db8 EFLAGS: 00010293 RAX: ffffffff7fffffff RBX: 00002aaad0c00000 RCX: 0000000000000000 RDX: ffff810107c95000 RSI: ffff81000000f800 RDI: ffff81000000f000 RBP: ffff81012d20a430 R08: 0000000000000000 R09: 0000000000000000 R10: 000000000007f71c R11: 0000000000000000 R12: 0000000000000000 R13: ffff810173270040 R14: ffff8101732700b8 R15: ffff81012d20a430 FS: 00002aaaaae6b900(0000) GS:ffffffff804dd680(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000002730 CR3: 0000000156775000 CR4: 00000000000006e0 Process R (pid: 5288, threadinfo ffff8101553f4000, task ffff8101559f3810) Stack: ffff8101567752a8 00002aaad0c00000 ffff810159fe9a68 ffff810173270040 0000000000000001 ffffffff8016ed05 0000000000000000 ffff8101732700b8 ffff81015a033a80 00002aaaad975000 Call Trace:{handle_mm_fault+293} {do_page_fault+1172} {autoremove_wake_function+0} {dnotify_parent+46} {vfs_read+268} {error_exit+0} Code: 48 8b b1 30 27 00 00 76 0d b8 00 00 00 80 eb 10 66 66 90 66 RIP {pte_alloc_map+170} RSP CR2: 0000000000002730 Please tell me if more information is necessary. Francois