Re: Multiple MSI

[Michael Ellerman <michael@ellerman.id.au>]

[-- Attachment #1: Type: text/plain, Size: 1796 bytes --]

On Thu, 2008-07-03 at 21:41 +1000, Benjamin Herrenschmidt wrote:
> On Thu, 2008-07-03 at 05:31 -0600, Matthew Wilcox wrote:
> > On Thu, Jul 03, 2008 at 07:17:00PM +1000, Benjamin Herrenschmidt wrote:
> > > > Some years ago, we had discussions about getting rid of IRQ numbers
> > > > altogether, or at least the requirement to have device drivers know
> > > > about them. Does anyone remember what happened to that idea?
> > > 
> > > I think it's not totally dead. Last I heard, someone (jgarzik ?) was
> > > slowly, bit by bit, removing the dependencies on the irq argument on irq
> > > handlers which is one step in the direction.
> > 
> > I think that project's dead, Jim.  http://lkml.org/lkml/2008/4/22/578
> 
> Ouch, missed that one... sad, would have been a good idea in the long
> run. irq_desc array is a big PITA.

Well it looks like Linus' main objection is with changing the driver
API, which does make sense, _that_ would be a PITA.

> > You can't do that.  /proc/interrupts is so terribly useful for a
> > sysadmin that you can't remove information from it.
> 
> You can create a new one with informations about the new stuff..
> 
> Anyway, looks like it's not happening and we'll be stuck with the bloody
> array for the time being. Crap.

In most cases that I can see the conversion from irq number to irq_desc
is done in the genirq code, so I don't see why you couldn't just put a
remapping in there from irq numbers to descs that doesn't use an array. 
But maybe I'm missing something.

cheers

-- 
Michael Ellerman
OzLabs, IBM Australia Development Lab

wwweb: http://michael.ellerman.id.au
phone: +61 2 6212 1183 (tie line 70 21183)

We do not inherit the earth from our ancestors,
we borrow it from our children. - S.M.A.R.T Person

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]