Re: Locking in disk(9) api

[Matthias Scheler <tron%zhadum.org.uk@localhost>]

On 29 Dec 2009, at 0:49, matthew green wrote:
> after thinking about this during the afternoon, i'm more convinced
> that the current method is OK.

I agree with that for two reasons:
1.) It seems correct for an architecture point of view. The disk(9) data
    is IMHO just another bit of driver instance data which should be
    protected by the driver.
2.) It makes the back-port of the "dm" driver to "netbsd-5" easier.

>  dm(4) just needs a little fix, and
> i'm aware that adam has already sent a test patch for tron.

The patch seems to do the job. My stress test (un-tarring a lot of NetBSD 
sources
to two LVM volumes at the same time) completed without problems with the
modified "dm" driver.

        Kind regards

-- 
Matthias Scheler                           http://zhadum.org.uk/