On Fri, Apr 08, 2016 at 12:18:15PM +0200, Mathieu Lirzin wrote: > Roel Janssen <r...@gnu.org> writes: > > > Strangely enough, I couldn't apply it either. I don't know what > > happened.. I regenerated the patch, and now I was able to apply it to > > my source tree. Hopefully it works for you too now. > > This problem has already happened to me last year. I suspect it is/was a > bug in Git.
I've attached the diffoscope report (HTML) between the first two patches in this thread. The diff header in the 'bad' patch says that the hunk applies to 7 lines. In the 'good' patch, the header instead says 6. The 'bad' patch can be applied when that parameter is changed to 6. I wonder if the bug can be reproduced?
