On Tue, 6 Jun 2000, Hans Reiser wrote:
> James Sutherland wrote:
>
> > That wasn't my interpretation of it...
> >
> > * ReiserFS is still a moving target, unlike NWFS, HPFS etc.
>
> If it isn't a dead project it is a moving target, so yes. We are moving into
> synchronization with the standard linux kernel release methodology and
> schedule. I believe that methodology envisions movement....:-/
There is a difference between progress (better implementation) and changes
to the actual structure. If the version of ReiserFS built into the 2.4
series will not remain stable (backwards and forwards compatability, from
2.4.0 to 2.4.999, transparently), or it may not be properly maintained
throughout that period, then it should remain external until 2.5.
There are good and bad times to add code to the kernel: the very end of
the code freeze period is the worst possible time. This would also justify
deferring RFS until 2.5.
> > (Disclaimer: There's an FS I might be porting in a few months which would
> > benefit greatly from journalling support. It won't be competing with any
> > native Linux filing systems, though.)
>
> Go right ahead and use it. I'm not proposing stopping you. Just don't ask me
> to sit around and wait for the chance to use someone else's code when it is
> slower code than what I use now.
I don't have the source code to port yet...
If we decide between implementations based on benchmarks, when do we
abandon the Linux kernel and move to NT? Simply being faster (on paper)
does not justify the implementation.
Do not reinvent the wheel, even if your new wheel has better benchmark
results. Improve/replace the existing one, by all means - just don't go
adding new proprietary wheels to make your code a bit faster.
James.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Wed Jun 07 2000 - 21:00:24 EST