Re: [PATCH 2/3] relay: Fix race condition which occurs when reading across CPUs.

[Eduard - Gabriel Munteanu <eduard.munteanu@linux360.ro>]

On Mon, 16 Jun 2008 20:28:44 +0200
Jens Axboe <jens.axboe@oracle.com> wrote:
 
> Hmm dunno, that is what blktrace also did but primarily for
> performance reasons. It's tricky - Tom stated that he is working on a
> lib to abstract this from applications. While that is handy for
> telling you what to do, it also an annoyance that you HAVE to do it
> that way (it's supposed to just be a "normal" fs, not with funky
> restrictions).
> 
> So perhaps provide both versions in-kernel and let the kernel user
> device. For blktrace, we have one app and we know we can use the
> faster variant since readers are affine. For more debug style exports
> or where you don't know your consumer, use the safer variant (which
> should be the default action).

This sounds good. Though short debug info can be exported through
debugfs alone, there is another use to this patch: global channels,
which currently require kernel users to write their own locking
mechanism.

So, are you fine with me patching relay _and blktrace_ code to use
faster variants named relay_write_affine() and __relay_write_affine()?
This implies having relay_write() and __relay_write() be the slower,
safer paths. Do you agree with this names, provided the functions are
documented correctly?

kmemtrace will use the affine versions and set CPU affinity anyway, but
it would be nice to have a consistent behavior from relay's part.


	Cheers,
	Eduard