(This was reported on the linux-cifs-client mailing list, posting here per Steve's suggestion.) Running dbench on a cifs mount as follows: ./dbench -c ./client_plain.txt 4 there was an oops as shown below - it seems to be in cifs_close. Please let me know if there's a fix for this. Thanks --suresh ------------------------------------------------------------------------------- ksymoops 2.4.9 on i686 2.6.8.1. Options used -v /usr/src/linux/vmlinux (specified) -K (specified) -l /proc/modules (default) -o /lib/modules/2.6.8.1/ (default) -m /usr/src/linux/System.map (default) No modules in ksyms, skipping objects No ksyms, skipping lsmod CPU 0 irqstacks, hard=c07ad000 soft=c078d000 CPU 1 irqstacks, hard=c07ae000 soft=c078e000 3c59x: Donald Becker and others. www.scyld.com/network/vortex.html 0000:00:09.0: 3Com PCI 3c905C Tornado at 0xec00. Vers LK1.1.19 Unable to handle kernel paging request at virtual address d1102f7c c026dcfe *pde = 00044067 Oops: 0002 [#1] CPU: 1 EIP: 0060:[<c026dcfe>] Not tainted Using defaults from ksymoops -t elf32-i386 -a i386 EFLAGS: 00010246 (2.6.8.1) Unable to handle kernel paging request at virtual address d1102f7c c026dcfe *pde = 00044067 eax: d1102f78 ebx: d240df78 ecx: d240df80 edx: d910ae00 esi: da10ff88 edi: da10ff5c ebp: d6640f54 esp: d6640f40 ds: 007b es: 007b ss: 0068 Stack: 001aeae4 00000000 da10ff5c dff7fe90 cfa83e9c d6640f74 c0162b36 cfa83e9c da10ff5c d1854f64 da10ff5c 00000000 d3844e54 d6640f98 c0161254 da10ff5c d3844e54 da10ff5c d3844e54 d3844e54 00000006 d3844e58 d6640fbc c01612fa Call Trace: [<c0108595>] show_stack+0x75/0x90 [<c01086f5>] show_registers+0x125/0x180 [<c0108870>] die+0x90/0x140 [<c0119890>] do_page_fault+0x290/0x5a4 [<c01081ed>] error_code+0x2d/0x40 [<c0162b36>] __fput+0xf6/0x110 [<c0161254>] filp_close+0x44/0x70 [<c01612fa>] sys_close+0x7a/0xf0 [<c0107149>] sysenter_past_esp+0x52/0x79 Code: 89 50 04 c7 43 04 00 02 20 00 f0 81 47 2c 00 00 00 01 8b 43 >>EIP; c026dcfe <cifs_close+de/1f0> <===== >>eax; d1102f78 <pg0+107f8f78/3f6f4000> >>ebx; d240df78 <pg0+11b03f78/3f6f4000> >>ecx; d240df80 <pg0+11b03f80/3f6f4000> >>edx; d910ae00 <pg0+18800e00/3f6f4000> >>esi; da10ff88 <pg0+19805f88/3f6f4000> >>edi; da10ff5c <pg0+19805f5c/3f6f4000> >>ebp; d6640f54 <pg0+15d36f54/3f6f4000> >>esp; d6640f40 <pg0+15d36f40/3f6f4000> Trace; c0108595 <show_stack+75/90> Trace; c01086f5 <show_registers+125/180> Trace; c0108870 <die+90/140> Trace; c0119890 <do_page_fault+290/5a4> Trace; c01081ed <error_code+2d/40> Trace; c0162b36 <__fput+f6/110> Trace; c0161254 <filp_close+44/70> Trace; c01612fa <sys_close+7a/f0> Trace; c0107149 <sysenter_past_esp+52/79> Code; c026dcfe <cifs_close+de/1f0> 00000000 <_EIP>: Code; c026dcfe <cifs_close+de/1f0> <===== 0: 89 50 04 mov %edx,0x4(%eax) <===== Code; c026dd01 <cifs_close+e1/1f0> 3: c7 43 04 00 02 20 00 movl $0x200200,0x4(%ebx) Code; c026dd08 <cifs_close+e8/1f0> a: f0 81 47 2c 00 00 00 lock addl $0x1000000,0x2c(%edi) Code; c026dd0f <cifs_close+ef/1f0> 11: 01 Code; c026dd10 <cifs_close+f0/1f0> 12: 8b 43 00 mov 0x0(%ebx),%eax <1>Oops: 0002 [#2] CPU: 0 EIP: 0060:[<c026dcfe>] Not tainted EFLAGS: 00010246 (2.6.8.1) eax: d1102f78 ebx: d16c0f78 ecx: d16c0f80 edx: d910ae00 esi: d0ef4f88 edi: d0ef4f5c ebp: d4ee2f54 esp: d4ee2f40 ds: 007b es: 007b ss: 0068 Stack: 001aeb5a 00000000 d0ef4f5c dff7fe90 d2149e9c d4ee2f74 c0162b36 d2149e9c d0ef4f5c cfdf0f64 d0ef4f5c 00000000 d665ae54 d4ee2f98 c0161254 d0ef4f5c d665ae54 d0ef4f5c d665ae54 d665ae54 00000005 d665ae58 d4ee2fbc c01612fa Call Trace: [<c0108595>] show_stack+0x75/0x90 [<c01086f5>] show_registers+0x125/0x180 [<c0108870>] die+0x90/0x140 [<c0119890>] do_page_fault+0x290/0x5a4 [<c01081ed>] error_code+0x2d/0x40 [<c0162b36>] __fput+0xf6/0x110 [<c0161254>] filp_close+0x44/0x70 [<c01612fa>] sys_close+0x7a/0xf0 [<c0107149>] sysenter_past_esp+0x52/0x79 Code: 89 50 04 c7 43 04 00 02 20 00 f0 81 47 2c 00 00 00 01 8b 43 >>EIP; c026dcfe <cifs_close+de/1f0> <===== >>eax; d1102f78 <pg0+107f8f78/3f6f4000> >>ebx; d16c0f78 <pg0+10db6f78/3f6f4000> >>ecx; d16c0f80 <pg0+10db6f80/3f6f4000> >>edx; d910ae00 <pg0+18800e00/3f6f4000> >>esi; d0ef4f88 <pg0+105eaf88/3f6f4000> >>edi; d0ef4f5c <pg0+105eaf5c/3f6f4000> >>ebp; d4ee2f54 <pg0+145d8f54/3f6f4000> >>esp; d4ee2f40 <pg0+145d8f40/3f6f4000> Trace; c0108595 <show_stack+75/90> Trace; c01086f5 <show_registers+125/180> Trace; c0108870 <die+90/140> Trace; c0119890 <do_page_fault+290/5a4> Trace; c01081ed <error_code+2d/40> Trace; c0162b36 <__fput+f6/110> Trace; c0161254 <filp_close+44/70> Trace; c01612fa <sys_close+7a/f0> Trace; c0107149 <sysenter_past_esp+52/79> Code; c026dcfe <cifs_close+de/1f0> 00000000 <_EIP>: Code; c026dcfe <cifs_close+de/1f0> <===== 0: 89 50 04 mov %edx,0x4(%eax) <===== Code; c026dd01 <cifs_close+e1/1f0> 3: c7 43 04 00 02 20 00 movl $0x200200,0x4(%ebx) Code; c026dd08 <cifs_close+e8/1f0> a: f0 81 47 2c 00 00 00 lock addl $0x1000000,0x2c(%edi) Code; c026dd0f <cifs_close+ef/1f0> 11: 01 Code; c026dd10 <cifs_close+f0/1f0> 12: 8b 43 00 mov 0x0(%ebx),%eax -------------------------------------------------------------------------------
I have got the same oops on a Debian/Unstable with a self compiled 2.6.8.1 oops(es): Feb 7 18:17:07 ramen kernel: c02180d4 Feb 7 18:17:07 ramen kernel: SMP Feb 7 18:17:07 ramen kernel: Modules linked in: loop ntfs appletalk ipx p8022 psnap llc usbcore Feb 7 18:17:07 ramen kernel: CPU: 1 Feb 7 18:17:07 ramen kernel: EIP: 0060:[cifs_readdir+2401/3445] Not tainted Feb 7 18:17:07 ramen kernel: EFLAGS: 00010246 (2.6.8.1) Feb 7 18:17:07 ramen kernel: EIP is at cifs_readdir+0x961/0xd75 Feb 7 18:17:07 ramen kernel: eax: c050c480 ebx: d3b79f4c ecx: f7db1340 edx: 91982a8e Feb 7 18:17:07 ramen kernel: esi: 00000000 edi: e9048bce ebp: d3b79f68 esp: d3b79ee0 Feb 7 18:17:07 ramen kernel: ds: 007b es: 007b ss: 0068 Feb 7 18:17:07 ramen kernel: Process smbd (pid: 30224, threadinfo=d3b78000 task=cbc4b290) Feb 7 18:17:07 ramen kernel: Stack: d3b79f4c e70d41e0 cc0e8480 c050c480 d3b79fa0 c050c480 d3b79f34 d3b79f38 Feb 7 18:17:07 ramen kernel: 00000001 000003e9 e70d4000 91982a8e c229ab00 dec6bb80 f7db1340 00000009 Feb 7 18:17:07 ramen kernel: 00004000 000c0451 00000000 4204f254 33390598 00000001 00000000 000a1800 Feb 7 18:17:07 ramen kernel: Call Trace: Feb 7 18:17:07 ramen kernel: [show_stack+128/150] show_stack+0x80/0x96 Feb 7 18:17:07 ramen kernel: [show_registers+351/430] show_registers+0x15f/0x1ae Feb 7 18:17:07 ramen kernel: [die+141/251] die+0x8d/0xfb Feb 7 18:17:07 ramen kernel: [do_page_fault+708/1390] do_page_fault+0x2c4/0x56e Feb 7 18:17:07 ramen kernel: [error_code+45/56] error_code+0x2d/0x38 Feb 7 18:17:07 ramen kernel: [vfs_readdir+150/177] vfs_readdir+0x96/0xb1 Feb 7 18:17:07 ramen kernel: [sys_getdents64+109/166] sys_getdents64+0x6d/0xa6 Feb 7 18:17:07 ramen kernel: [syscall_call+7/11] syscall_call+0x7/0xb Feb 7 18:17:07 ramen kernel: Code: 8b 42 3c d1 e8 89 44 24 08 89 d0 83 c0 40 89 44 24 04 89 04 Feb 7 18:17:27 ramen kernel: <3> CIFS VFS: Error 0xffffffec or on cifs_get_inode_info in lookup Feb 9 07:59:17 ramen -- MARK -- Feb 9 08:05:22 ramen kernel: c02180d4 Feb 9 08:05:22 ramen kernel: SMP Feb 9 08:05:22 ramen kernel: Modules linked in: loop ntfs appletalk ipx p8022 psnap llc usbcore Feb 9 08:05:22 ramen kernel: CPU: 1 Feb 9 08:05:22 ramen kernel: EIP: 0060:[cifs_readdir+2401/3445] Not tainted Feb 9 08:05:22 ramen kernel: EFLAGS: 00010246 (2.6.8.1) Feb 9 08:05:22 ramen kernel: EIP is at cifs_readdir+0x961/0xd75 Feb 9 08:05:22 ramen kernel: eax: c050c480 ebx: d1371f4c ecx: f7db1340 edx: 9542b9d7 Feb 9 08:05:22 ramen kernel: esi: 00000000 edi: d0fcf1c6 ebp: d1371f68 esp: d1371ee0 Feb 9 08:05:22 ramen kernel: ds: 007b es: 007b ss: 0068 Feb 9 08:05:22 ramen kernel: Process find (pid: 32008, threadinfo=d1370000 task=daf5c330) Feb 9 08:05:22 ramen kernel: Stack: d1371f4c e987cf28 c99ee080 c050c480 d1371fa0 c050c480 d1371f34 d1371f38 Feb 9 08:05:22 ramen kernel: 00000001 000003e9 e987c000 9542b9d7 d88c5100 d0fcfb80 f7db1340 0000002a Feb 9 08:05:22 ramen kernel: 00004000 000ffaf5 00000000 4204f243 03b9aca0 00000001 00000000 00291802 Feb 9 08:05:22 ramen kernel: Call Trace: Feb 9 08:05:22 ramen kernel: [show_stack+128/150] show_stack+0x80/0x96 Feb 9 08:05:22 ramen kernel: [show_registers+351/430] show_registers+0x15f/0x1ae Feb 9 08:05:22 ramen kernel: [die+141/251] die+0x8d/0xfb Feb 9 08:05:22 ramen kernel: [do_page_fault+708/1390] do_page_fault+0x2c4/0x56e Feb 9 08:05:22 ramen kernel: [error_code+45/56] error_code+0x2d/0x38 Feb 9 08:05:22 ramen kernel: [vfs_readdir+150/177] vfs_readdir+0x96/0xb1 Feb 9 08:05:22 ramen kernel: [sys_getdents64+109/166] sys_getdents64+0x6d/0xa6 Feb 9 08:05:22 ramen kernel: [syscall_call+7/11] syscall_call+0x7/0xb Feb 9 08:05:22 ramen kernel: Code: 8b 42 3c d1 e8 89 44 24 08 89 d0 83 c0 40 89 44 24 04 89 04
This is likely fixed in the big cifs merge that went into 2.6.10
This patch should fix the last hole in cifs_close http://cifs.bkbits.net:8080/linux-2.5cifs/gnupatch@424f86c1GG9cVtTEaLzyAGIQKVBbfQ It will be sent off for 2.6.12, and a patch (or at least replacement files for fs/cifs directory) will be made available in linux-cifs-client SVN for 2.6.8 kernels within a few days (it will be in version 1.32 of the cifs vfs).