On Mon, Jan 19, 2009 at 04:58:55PM +1100, Michael Ellerman wrote:
> I guess I'd rather we didn't do a "complete rewrite", but there's
> definitely work to be done before we can support lots of MSI-X irqs.
It turns out it's a bit less of a 'complete rewrite' than I thought. At
least, I tried branching off Linus' tree for it and was frustrated by
not having the changes I've made for multiple MSI.
> Are you seeing hardware with 2K MSIs? I've got _2_ on the SATA
> controller in my laptop.
Are you sure about that? Mine says:
Capabilities: [80] Message Signalled Interrupts: Mask- 64bit- Queue=2/2 Enable+
but that's because it's a non-fixed version of pciutils. Here's what
3.1.0 says:
Capabilities: [80] MSI: Mask- 64bit- Count=4/4 Enable+
so it's actually got 4. This laptop has an ICH8 chipset, fwiw.
As I'm sure you know (but others may not), MSI only supports up to 32
irqs; you need MSI-X to support up to 2048 irqs. I don't have access to
my machine with MSI-X devices right now, but it seems that manufacturers
have all kinds of good ideas about how to use hundreds if not yet
thousands of interrupts.
> Perhaps. Now that the other architecture implements arch_setup_msi_irqs
> that might be a bit more feasible, when I added it powerpc was the only
> user.
The architectures are going to have to implement _irqs anyway to support
multiple MSI.
> Sounds good. Have you got any code we can see?
>
> And if you can incorporate the other patch I sent today into the rewrite
> that'd be nice, we really need that functionality.
The rewrite feels like a longer-term project ... let's not hold anything
up for that. I'm going to rebase the multiple-msi tree on top of
Jesse's linux-next branch later today.
--
Matthew Wilcox Intel Open Source Technology Centre
"Bill, look, we understand that you're interested in selling us this
operating system, but compare it to ours. We can't possibly take such
a retrograde step."
--
To unsubscribe from this list: send the line "unsubscribe linux-pci" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
