kernel BUG at rmap.c:398! invalid operand: 0000 sr_mod cramfs coda sg ide-cd cdrom nls_iso8859-1 nls_cp437 vfat fat nfsd netconsole autofs nfs lockd sunrpc tg3 eepro100 mii e1000 loop usb-ohci usbcore ext3 CPU: 0 EIP: 0010:[<c01436cf>] Not tainted EFLAGS: 00010246 EIP is at try_to_unmap_one [kernel] 0x13f (2.4.20-13.7smp) eax: 00000000 ebx: 00000000 ecx: fffed000 edx: 000ca239 esi: 00340000 edi: c1000030 ebp: fffedd00 esp: c54fbeb8 ds: 0018 es: 0018 ss: 0018 Process kswapd (pid: 5, stackpage=c54fb000) Stack: 00e5e500 00000000 c2d5cd60 c2d5cd60 c2d5cd60 00000006 c030e180 ffffffff c01437fa 00e5e500 c2d5cd60 c013e820 c03b8560 00000000 00000000 00000000 c2d5cd60 c030e180 c030bc20 c0139a34 00000001 000000dc c030e180 c030f338 Call Trace: [<c01437fa>] try_to_unmap [kernel] 0x6a (0xc54fbed8)) [<c013e820>] add_to_swap [kernel] 0x60 (0xc54fbee4)) [<c0139a34>] launder_page [kernel] 0x5a4 (0xc54fbf04)) [<c013be2a>] rebalance_dirty_zone [kernel] 0x9a (0xc54fbf1c)) [<c013c0ae>] rebalance_inactive_zone [kernel] 0x21e (0xc54fbf3c)) [<c013c21d>] rebalance_inactive [kernel] 0x3d (0xc54fbf6c)) [<c013c351>] do_try_to_free_pages_kswapd [kernel] 0x31 (0xc54fbf90)) [<c013c801>] kswapd [kernel] 0x141 (0xc54fbfd4)) [<c0105000>] stext [kernel] 0x0 (0xc54fbfe8)) [<c0107266>] arch_kernel_thread [kernel] 0x26 (0xc54fbff0)) [<c013c6c0>] kswapd [kernel] 0x0 (0xc54fbff8)) Code: 0f 0b 8e 01 a5 55 25 c0 31 c0 8b 54 24 04 86 42 2c 84 c0 7f
Seen the same oops on 2.4.20-18.7smp
We have quite a few systems that are experiencing this oops. If you need a place to test something let me know. Thanks Connie Sieh Fermi National Accelerator Laboratory
I'm also experiencing what appears to be the same failure on a uniprocessor (266MHz P4) system running 2.4.20-18.9: Jul 20 23:12:34 peanuts kernel: kernel BUG at rmap.c:470! Jul 20 23:12:34 peanuts kernel: invalid operand: 0000 Jul 20 23:12:34 peanuts kernel: cmaudio soundcore nvidia vpnmod autofs sis900 ipt_REJECT iptable_filter ip_tables sg sr_mod ide-scsi scsi_mod ide-cd cdrom ohci1394 ieee1394 nls_iso8859-1 ntf Jul 20 23:12:34 peanuts kernel: CPU: 0 Jul 20 23:12:34 peanuts kernel: EIP: 0060:[<c01444c0>] Tainted: P Jul 20 23:12:34 peanuts kernel: EFLAGS: 00013206 Jul 20 23:12:34 peanuts kernel: Jul 20 23:12:34 peanuts kernel: EIP is at try_to_unmap [kernel] 0x20 (2.4.20-18.9) Jul 20 23:12:34 peanuts kernel: eax: 01048149 ebx: c15dae68 ecx: d93b5728 edx: dfff3fa4 Jul 20 23:12:34 peanuts kernel: esi: c15dae84 edi: ffffffff ebp: c15dae68 esp: dfff3f78 Jul 20 23:12:34 peanuts kernel: ds: 0068 es: 0068 ss: 0068 Jul 20 23:12:34 peanuts kernel: Process kswapd (pid: 5, stackpage=dfff3000) Jul 20 23:12:34 peanuts kernel: Stack: 00000000 00000000 dfff3fa4 c15dae68 c15dae68 c15dae84 000001d0 c0310134 Jul 20 23:12:34 peanuts kernel: c013c0ed c15e7188 000001d0 00000001 00000033 c030ff00 0000000c c013cd0c Jul 20 23:12:34 peanuts kernel: c030ff00 000001d0 c15dae68 dfff2000 c030ff00 dfff2305 00000000 c013d32c Jul 20 23:12:34 peanuts kernel: Call Trace: [<c013c0ed>] launder_page [kernel] 0x54d (0xdfff3f98)) Jul 20 23:12:34 peanuts kernel: [<c013cd0c>] rebalance_dirty_zone [kernel] 0x5c (0xdfff3fb4)) Jul 20 23:12:34 peanuts kernel: [<c013d32c>] kswapd [kernel] 0x15c (0xdfff3fd4))Jul 20 23:12:34 peanuts kernel: [<c013d1d0>] kswapd [kernel] 0x0 (0xdfff3fe4)) Jul 20 23:12:34 peanuts kernel: [<c010742d>] kernel_thread_helper [kernel] 0x5 (0xdfff3ff0)) Jul 20 23:12:34 peanuts kernel: Jul 20 23:12:34 peanuts kernel: Jul 20 23:12:34 peanuts kernel: Code: 0f 0b d6 01 27 12 26 c0 8b 45 18 83 e0 01 75 08 0f 0b d8 01 According to rmap.c:470, The cause of the failure is trying to unmap a page that's reserved: int try_to_unmap(struct page * page) { struct pte_chain *pc, *next_pc, *start; int ret = SWAP_SUCCESS; int victim_i = -1; /* This page should not be on the pageout lists. */ if (PageReserved(page)) BUG();
Nigel Simpson: please read bug 73733
Bug 73733 concerns nvidia drivers. I clearly do not have nvidia drivers installed in my systems that are experiencing this bug as they do not even have monitors. These systems are very compute bound 2u physics analysis nodes. -Connie Sieh Fermi National Accelerator Laboratory
I also have seen this issue. The system seems to kernel panic about once very 2 weeks. Server: IBM EServer 345 (1 processor) Redhat 9.0 /proc/version: 20030222 (Red Hat Linux 3.2.2-5)) #1 SMP Thu Mar 13 17:45:54 EST 2003 .devel.redhat.com) (gcc version 3.2.2 20030222 (Red Hat Linux 3.2.2- 5)) #1 SMP Thu Mar 13 17:45:54 EST 2003 -Geoff Kuchera Siemens
Try 2.4.20-28 The last few erratas have contained numerous rmap stability fixes.
Thanks for the bug report. However, Red Hat no longer maintains this version of the product. Please upgrade to the latest version and open a new bug if the problem persists. The Fedora Legacy project (http://fedoralegacy.org/) maintains some older releases, and if you believe this bug is interesting to them, please report the problem in the bug tracker at: http://bugzilla.fedora.us/