Re: [PATCH 0/3] Implement unionmount on top of unionfs

[Sergiu Ivanov]

Hello,

All the patches in this patch series have already been reviewed and
quite many changes have been introduced.  It's time for me to post the
corrected patches, which I'm doing right now.

Note, that PATCH 2/3 has changed essentially in the meantime: antrik
said it would be good to have the mountee started during
initialization of union mount.  As I have pointed out in several
E-mails, this is impossible in the simplest case due to a deadlock.
However, PATCH 2/3 moves the mountee startup code in a separate thread
(actually, reuses the update thread existing in unionfs), which
results in mountee starting immediately *after* the initialization of
unionfs.

Although this modification to PATCH 2/3 is an essential one, I'm still
keeping it in this patch series, because the meaning of the whole
series has not changed.

There also are some things about code formatting in PATCH 2/3, but I
hope I will be able to rule them out and post one more update.  (I
hope it won't be more than one.)

Regards,
scolobb