Hello,
On Tue, 13 Sep 2016 10:34:02 -0500, Bjorn Helgaas wrote:
> > After looking at this myself, I'm not happy with this either. It feels
> > like there are bugs lurking here and we're just hiding one of them.
> >
> > Here are the callers of irq_domain_add_linear() for legacy INTx in
> > drivers/pci/host:
> >
> > advk_pcie_init_irq_domain LEGACY_IRQ_NUM (4)
> > dra7xx_pcie_init_irq_domain 4
> > ks_dw_pcie_host_init MAX_LEGACY_IRQS (4)
> > altera_pcie_init_irq_domain INTX_NUM + 1 (5)
> > nwl_pcie_init_irq_domain INTX_NUM + 1 (5)
> > xilinx_pcie_init_irq_domain 4
>
> The altera change corresponding to this was 99496bd2971f ("PCI: altera: Fix
> error when INTx is 4"). I should have noticed this inconsistency back
> then.
>
> Are aardvark, dra7xx, keystone, and xilinx (non-NWL) broken because they
> only request 4 IRQs and only INTA, INTB, and INTC work?
>
> > I think all of these use the of_irq_parse_and_map_pci() path you
> > mentioned, so if the problem is in the way that path works, I would
> > think these should *all* be requesting the same number of interrupts
> > in the domain.
> >
> > I agree with Marc that we should request 4 IRQs, because that's what
> > we need. If we can't do that for some reason, we ought to at least
> > make all these callers the same.
Thanks for Cc'ing about this issue. Indeed, the Aardvark driver
supports all of INT{A,B,C,D}, so the current situation doesn't work. As
suggested, the simplest solution is to just allocate an irq domain with
5 IRQs, like is done in the Altera driver.
However, my feeling is that a more correct solution would be to have a
translation between the PCI_INTERRUPT_PIN value (in the range 0x1 to
0x4) to the hwirq value (in the range 0x0 to 0x3).
Best regards,
Thomas
--
Thomas Petazzoni, CTO, Free Electrons
Embedded Linux and Kernel engineering
http://free-electrons.com
--
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
