This is a meta-bug for all of the things causing a Windows bootstrap to fail. This should be closed when i686-pc-mingw32 successfully completes a full bootstrap.
Plus there is a new bootstrap failure also because of a change to use %I64 or what ever is the format is.
Can I close this now as the only one left is PR 17406?
No, let's wait until 17406 is fixed.
I am going to link PR 18459 to this one, even though it is not a bootstrap bug for mingw32, it does cause almost all the java testcases to fail.
18458 is failure to install in fixincludes.
Since the last bug was fixed so closing as fixed.