Re: Unnecessary BKL contention in video1394

[Daniel Drake]

On Thu, 2006-10-19 at 16:36 +0200, Stefan Richter wrote:
> Thanks for the info. Daniel, do you want to resend a signed-off patch?
> And __video1394_ioctl and its wrapper video1394_ioctl can certainly be
> merged then.

Yep, I had already made that change locally. I will run it overnight on
several cameras just to be sure, and will send a patch tomorrow.

I also did some more investigation and straightened out my knowledge of
file_operations: release() can never be called while inside a read() or
ioctl(): This would imply that separate threads are in use, and the
driver's release() function is not called until *all* threads have
closed the fd.  In other words I'm now much more confident that this
patch is not removing any necessary locking.

Daniel


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