Re: [PATCH 0/21] MSI rework

[Greg KH <greg@kroah.com>]

On Thu, Mar 22, 2007 at 09:01:10AM -0600, Eric W. Biederman wrote:
> Michael Ellerman <michael@ellerman.id.au> writes:
> 
> > This is my series to rework the generic MSI code into something we can use
> > on powerpc[1].
> >
> > I've tried as much as possible not to change the semantics for other archs,
> > but there's a few little changes. I think they're all OK in their own right.
> >
> > I don't have any serious hardware to test on, but my little x86_64 box with
> > an e1000 using MSI still works with these changes. I've also got MSI working
> > on a powerpc blade with a tg3. I haven't tested MSI-X _at all_.
> >
> > I've also tested on the blade with a debug hack to make the MSI case
> > allocate/free 8 MSIs, but only use the last one, just to exercise the n > 1
> > case a little. All seems to work fine.
> 
> Generally I think this looks good.  However there is a lot here, and some
> of it potentially at least has some pretty subtle side effects.
> 
> So reviewing all of these patches at once is almost certain to cause
> something important to be missed :(
> 
> Can we slow this process down a few days by taking this one logical chunk
> at a time?
> 
> i.e.  First the simple bug fixes that should purely be restructure of
> msi.c with no affect on anything outside of it.
> 
> And then get into the architecture enhancements.

I agree, care to break these down into a smaller series of patches that
can go into -mm for testing?

thanks,

greg k-h