Hi all,

I just pushed a final few fixes to the io_transaction branch which makes
that nbd-server implementation pass all tests in the test suite, with
the sole exception of the integrityhuge one.

That's not to say I want to ignore that test, but I do think it means
I'm getting close to a working implementation on this alternate branch.
I would appreciate it if people could check it out. Don't try using it
on critical data yet, but for the most part it should be okay. It
*seems* to be slightly faster than the current implementation, too, but
I haven't yet done any serious benchmarks.

Note that beyond read and write, not much is supported yet (e.g.,
copy-on-write and multifile aren't supported yet; prerun and postrun
might work but I haven't tested; etc).

Especially interesting would be to figure out if the API as I'm
proposing it seems sane. The idea would be that people could write
custom backends for nbd-server; in theory this should allow OS-specific
sendfile() implementations, or things like gzipped storage, etc.

Note that there's a lot of dead code in the nbd-server.c file on that
branch currently. Yes, I do plan to throw that out, but I'm not quite
there yet.

Thanks,

-- 
It is easy to love a country that is famous for chocolate and beer

  -- Barack Obama, speaking in Brussels, Belgium, 2014-03-26


------------------------------------------------------------------------------
Want fast and easy access to all the code in your enterprise? Index and
search up to 200,000 lines of code with a free copy of Black Duck
Code Sight - the same software that powers the world's largest code
search on Ohloh, the Black Duck Open Hub! Try it now.
http://p.sf.net/sfu/bds
_______________________________________________
Nbd-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/nbd-general

Reply via email to