lkml.org 
[lkml]   [2003]   [Aug]   [13]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: Deep call trace from PCMCIA CF eject
On Wed, Aug 13, 2003 at 07:24:56AM -0700, Paul Dickson wrote:
> Besides the minor concern about the trace itself, a bigger concern on my
> part is how deep the call trace is (60+ levels). How's it doing for stack
> space?

Not all those functions listed were involved - never, ever believe an
x86 call trace as being the absolute truth. 8)

The real calltrace is much closer to:

> Aug 13 06:57:59 violet cardmgr[1559]: executing: './ide check hde'
> Aug 13 06:58:00 violet cardmgr[1559]: executing: './ide stop hde'
> Aug 13 06:58:01 violet kernel: hde: task_no_data_intr: status=0x51 { DriveReady SeekComplete Error }
> Aug 13 06:58:01 violet kernel: hde: task_no_data_intr: error=0x04 { DriveStatusError }
> Aug 13 06:58:01 violet kernel: hde: Write Cache FAILED Flushing!
> Aug 13 06:58:01 violet cardmgr[1559]: + /dev/hde1 umounted
> Aug 13 06:58:01 violet kernel: updfstab: numerical sysctl 1 23 is obsolete.
> Aug 13 06:58:02 violet kernel: hde: task_no_data_intr: status=0x51 { DriveReady SeekComplete Error }
> Aug 13 06:58:02 violet kernel: hde: task_no_data_intr: error=0x04 { DriveStatusError }
> Aug 13 06:58:02 violet kernel: hde: Write Cache FAILED Flushing!
> Aug 13 06:58:02 violet kernel: Debug: sleeping function called from invalid context at include/linux/rwsem.h:66
> Aug 13 06:58:02 violet kernel: Call Trace:
> Aug 13 06:58:02 violet kernel: [<c011bc81>] __might_sleep+0x61/0x80
> Aug 13 06:58:02 violet kernel: [<c02376dc>] unlink+0x3c/0xa0
> Aug 13 06:58:02 violet kernel: [<c0267a32>] kobj_unmap+0x62/0x110
> Aug 13 06:58:02 violet kernel: [<c026cf62>] blk_unregister_region+0x22/0x30
> Aug 13 06:58:02 violet kernel: [<c0287eb3>] ide_unregister+0x2f3/0x8d0
> Aug 13 06:58:02 violet kernel: [<d096cba4>] ide_release+0x94/0x130 [ide_cs]
> Aug 13 06:58:02 violet kernel: [<d096c1fc>] ide_detach+0xac/0xc0 [ide_cs]
> Aug 13 06:58:02 violet kernel: [<d09727e2>] unbind_request+0xd2/0xe0 [ds]
> Aug 13 06:58:02 violet kernel: [<d0972fa1>] ds_ioctl+0x441/0x770 [ds]
> Aug 13 06:58:02 violet kernel: [<c01652b0>] sys_ioctl+0x100/0x290
> Aug 13 06:58:02 violet kernel: [<c010b1f9>] sysenter_past_esp+0x52/0x71

The problem is IDE - ide_unregister() does the following:

spin_lock_irq(&ide_lock);
...
blk_unregister_region(MKDEV(hwif->major, 0), MAX_DRIVES<<PARTN_BITS);

Since blk_unregister_region is sleepy, and sleeping with a spinlock
held is a big NO NO, it's an IDE problem not a PCMCIA problem. It
should be calling blk_unregister_region with a spinlock held.

--
Russell King (rmk@arm.linux.org.uk) The developer of ARM Linux
http://www.arm.linux.org.uk/personal/aboutme.html

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2005-03-22 13:47    [W:0.045 / U:0.204 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site