cowloop driver

Denis Kirjanov <kirjanov@gmail.com kirjanov at gmail.com
Fri Sep 18 19:13:07 UTC 2009


On Thu, Sep 17, 2009 at 01:34 +0200, H.J. Thomassen wrote:
> Hello Greg - and cc. Denis Kirjanov
> 
> On Wed, Sep 16, 2009 at 03:50:47PM -0700, Greg KH wrote:
> > On Thu, Sep 17, 2009 at 01:16:29AM +0400, Denis Kirjanov <kirjanov at gmail.com wrote:
> > > Good time of day.
> > > I want to take on work related to the cowloop driver. Anybody else
> > > involved with this?
> > 
> > The original author is, you should also copy him.
> > 
> > Attached is his latest version, if you could turn this into a patch
> > against Linus's latest tree, that would be wonderful.  Unfortunaly, I
> > have to take a last-minute-business trip tomorrow and can't get to it
> > for a few weeks :(
> > 
> > > While loading the module I get an kernel crash. It seems that
> > > something wrong with cowloop kernel thread.
> > 
> > Handrik-Jan, have you heard of this?
> 
> Yes, Denis told me so. He indicated that it was inside a call
> to __blk_end_request_cur() and that's a call that you inserted
> to replace a call to end_request(). From the way you substituted
> these calls I assume that they are drop-in replacements?
> 
> Tomorrow I hope to have a VM running with the newest kernel version
> where I can start testing without porting the driver back to
> end_request() c.s.
> 
> > 
> > Denis, care to post a kernel oops trace?
> 

Here is the crash dump:
kernel BUG at block/blk-core.c:2043!
invalid opcode: 0000 [#1] SMP
last sysfs file: /sys/devices/virtual/block/cow0/removable
EIP is at blk_finish_request+0x1af/0x1e0

Call Trace:
add_disk_randomness
__blk_end_request
__blk_end_request_cur
cowlo_daemon
schedule_tail
ret_from_fork
autoremove_wake_function
cowlo_daemon
kernel_thread_helper



More information about the devel mailing list