rename() in rtems filesystems

[Joel Sherrill]

Dmitry Kargapolov wrote:
> 
> Hi All.
> I would like to know the your opinion about the rename() issue
> possibility and expedience. I wrote about this idea in my
> previous letter (subj. "FS: problems and patches").
> It's very important for the new filesystems implementation.

I wanted to take alonger look at this.  I know at the time that
the choice of filesystem primitive operations was made to be
as small as possible.  If something could be implemented as
a layer on top of those, then that was how we did it.

My initial tendency was to agree that it was a needed primitive
operation to allow for obvious optimizations.  Jennifer and I
just need to review the specifics of the patch. Don't take silence
as bad in this case. :)

My only concern is to verify that it can be implemented as a
single operation inside a single filesystem.  I am at home without
access to standards.  But the Linux man page says the source and
destination have to be in the same filesystem so it is reasonable
to assume this is allowed.

> Thank you in advance.
> 
> sincerely,
> Dmitry Kargapolov        void at dr.com

-- 
Joel Sherrill, Ph.D.             Director of Research & Development
joel at OARcorp.com                 On-Line Applications Research
Ask me about RTEMS: a free RTOS  Huntsville AL 35805
   Support Available             (256) 722-9985