tech-kern archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: biodone() and splbio ?

On Tue, Jan 19, 2010 at 05:19:07PM -0500, Thor Lancelot Simon wrote:
> On Tue, Jan 19, 2010 at 11:16:41PM +0100, Manuel Bouyer wrote:
> [I wrote]
> > > I think this stuff should all be mutex-protected.  For what other
> > > reason do we need splbio/splvm in here any more?
> > 
> > Because it has not yet been made MP-safe ?
> > Sure, make these MP-safe but as a quick fix for netbsd-5, adding the
> > spl call is probably better.
> Well, let me rephrase: in what way is the handling of the queues not
> MP-safe that simply adding a mutex-per-queue would not remedy?  It might
> be about as easy to "MP-safe" this code as to sprinkle around the necessary
> splbio()/splvm() calls, using mutexes at IPL_VM or IPL_BIO as required.

Because I didn't make an in-depth analysis, the queue may not be the only
issue with this. splvm() is certainly safe, because it was before
biodone() was made MP-safe.

Manuel Bouyer <>
     NetBSD: 26 ans d'experience feront toujours la difference

Home | Main Index | Thread Index | Old Index