Re: [PATCH 4/7] ext4: fsync should wait for DIO writers

On Mon, Sep 10, 2012 at 11:51:35AM +0200, Jan Kara wrote:

On Sun 09-09-12 21:27:11, Dmitry Monakhov wrote:

quoted

fsync and punch_hole are the places where we have to wait for all
existing writers (writeback, aio, dio), but currently we simply
flush pended end_io request which is not sufficient.

  Why not? I guess you mean the fact that there can be DIO in flight for
which end_io() was not called so it is not queued in the queue? But that is
OK - we have not yet called aio_complete() for that IO so for userspace the
write has not happened yet. Thus there's no need to flush it to disk -
fsync() does not say anything about writes in progress while fsync is
called.

quoted

Even more i_mutex is not holded while punch_hole which obviously
result in dangerous data corruption due to write-after-free.

  Yes, that's a bug. I also noticed that but didn't get to fixing it (I'm
actually working on a more long term fix using range locking but that's
more of a research project so having somehow fixed at least the most
blatant locking problems is good).

Hi Jan,

Could you please share more detailed information about range locking
with me?  Actually, the goal of extent status tree is to implement a
range locking in ext4 [1], and I am working on it.  So I think that
you have some good ideas to share with me. :-)

1. http://www.spinics.net/lists/linux-ext4/msg32661.html

If you have some problems, please let me know.  Thanks!

Regards,
Zheng