[Xen-bugs] [Bug 372] Dom0 reboots instantly when accessing SCSI DLT tape drive

[bugzilla-daemon@xxxxxxxxxxxxxxxxxxx]

http://bugzilla.xensource.com/bugzilla/show_bug.cgi?id=372





------- Additional Comments From nospam@xxxxxxxxxxxxxx  2005-10-31 16:49 -------
I added 'noreboot', and here is a copy of as much of the error that fits on the 
50 line console.  I believe a significant amount of the panic dump scrolled 
past the readable screen:

 [<c02e5020>] scsi_done+0x0/0x30
 [<c02e4d59>] scsi_dispatch_cmd+0x189/0x300
 [<c02e5020>] scsi_done+0x0/0x30
 [<c02e8520>] scsi_times_out+0x0/0xc0
 [<c02ebf93>] scsi_request_fn+0x203/0x410
 [<c02984b5>] __elv_add_request+0xa5/0xe0
 [<c029bac8>] blk_insert_request+0xc8/0x120
 [<c02ea798>] scsi_insert_special_req+0x38/0x40
 [<c0317604>] st_do_scsi+0xf4/0x1c0
 [<c03174c0>] st_sleep_done+0x0/0x50
 [<c03195d7>] read_tape+0xf7/0x4e0
 [<c0318cab>] setup_buffering+0x14b/0x1c0
 [<c0319bf1>] st_read+0x231/0x3f0
 [<c016d0e1>] vfs_read+0xb1/0x130
 [<c016d3eb>] sys_read+0x4b/0x80
 [<c0109fb5>] syscall_call+0x7/0xb
Kernel panic - not syncing: BUG!
 Badness in smp_call_function at arch/xen/i386/kernel/smp.c:511
[<c0111821>] smp_call_function+0x131/0x140
[<c0286d74>] do_unblank_screen+0x14/0x140
[<c01118e7>] smp_send_stop+0x37/0xc0
[<c0111830>] stop_this_cpu+0x0/0x80
[<c0125381>] panic+0x61/0x160
[<c010ed72>] dma_map_single+0x102/0x160
[<c030d1ed>] ahc_linux_queue+0x75/0xe0
[<c02e4690>] scsi_done+0x0/0x30
[<c02e4d59>] scsi_dispatch_cmd+0x189/0x300
[<c02e5020>] scsi_done+0x0/0x30
[<c02e8520>] scsi_times_out+0x0/0xc0
[<c02ebf93>] scsi_request_fn+0x203/0x410
[<c02948b5>] __elv_add_request+0xa5/0xe0
[<c029bac8>] blk_insert_request+0xc8/0x120
[<c02ea798>] scsi_insert_special_req+0x38/0x40
[<c0317604>] st_do_scsi+0xf4/0x1c0
[<c03174c0>] st_sleep_done+0x0/0x50
[<c03195d7>] read_tape+0xf7/0x4e0
[<c0318cab>] setup_buffering+0x14b/0x1c0
[<c0319bf1>] st_read+0x231/0x3f0
[<c016d0e1>] vfs_read+0xb1/0x130
[<c016d3eb>] sys_read+0x4b/0x80
[<c0109fb5>] syscall_call+0x7/0xb


-------

No other physical devices are connected to the controller.  I can, however, 
issue commands to the media changer (this tape drive is two devices in one, two 
separate SCSI ids), and get the media changer to remove tapes from the drive, 
shuttle them to their bays, grab other tapes, and insert them into the drive.

I'm not sure if other people are successfully using AHA-2944UW's in their xen 
systems, but it's basically the standard Adaptec driver.  This card just talks 
to HVD drives instead of LVD/SE ones, which should be the same interface to the 
system as a 2940U2W - one of the most common SCSI cards.

Since I saw that the "Badness" was "in smp_call_function", I'll try compiling a 
single CPU kernel and see if I still get a panic.

-- 
Configure bugmail: 
http://bugzilla.xensource.com/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.

_______________________________________________
Xen-bugs mailing list
Xen-bugs@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-bugs