Comment # 15 on bug 1011250 from Johannes Weberhofer

(In reply to Neil Brown from comment #14)
> Created attachment 704742 [details]
> Revised patch
> 
> I looked at the code more closely and realized that the error scenario I
> described in the patch could not happen - I was misunderstanding some code.
> However I think I've found another error scenario that is credible.
> 
> Here is revised version of the patch which explains the details.  I've sent
> this upstream for comment/inclusion.
> No need to test this. The behavior isn't importantly different to the other
> patch, just the description.
> I will submit it to the SUSE kernels in a week or so unless I get negative
> feedback, either from upstream or your testing.

Thank you for your work on that issue. So far I haven't had any issue with the
testing kernel.