On Tue, 2020-09-15 at 03:42 +0900, Keith Busch wrote:
> On Mon, Sep 14, 2020 at 01:32:55PM -0400, Jon Derrick wrote:
> > Client VMD platforms have a software-triggered MSI/X vector 0 that will
> > not forward hardware-remapped MSI/X vectors from the sub-device domain.
> > This causes an issue with VMD platforms that use AHCI behind VMD and
> > have a single MSI/X vector mapping to vector 0. Add an MSI/X vector
> > offset for these platforms.
> >
> > Signed-off-by: Jon Derrick <jonathan.derrick@xxxxxxxxx>
> > ---
> > drivers/pci/controller/vmd.c | 35 +++++++++++++++++++++++++----------
> > 1 file changed, 25 insertions(+), 10 deletions(-)
> >
> > diff --git a/drivers/pci/controller/vmd.c b/drivers/pci/controller/vmd.c
> > index f69ef8c89f72..f8195bad79d1 100644
> > --- a/drivers/pci/controller/vmd.c
> > +++ b/drivers/pci/controller/vmd.c
> > @@ -53,6 +53,12 @@ enum vmd_features {
> > * vendor-specific capability space
> > */
> > VMD_FEAT_HAS_MEMBAR_SHADOW_VSCAP = (1 << 2),
> > +
> > + /*
> > + * Device may use MSI/X vector 0 for software triggering and will not
> > + * be used for MSI/X remapping
> > + */
> > + VMD_FEAT_OFFSET_FIRST_VECTOR = (1 << 3),
> > };
> >
> > /*
> > @@ -104,6 +110,7 @@ struct vmd_dev {
> > struct irq_domain *irq_domain;
> > struct pci_bus *bus;
> > u8 busn_start;
> > + u8 first_vec;
> > };
> >
> > static inline struct vmd_dev *vmd_from_bus(struct pci_bus *bus)
> > @@ -199,25 +206,26 @@ static irq_hw_number_t vmd_get_hwirq(struct msi_domain_info *info,
> > */
> > static struct vmd_irq_list *vmd_next_irq(struct vmd_dev *vmd, struct msi_desc *desc)
> > {
> > - int i, best = 1;
> > unsigned long flags;
> > + int i, best;
> >
> > if (vmd->msix_count == 1)
>
> This condition needs account for the vector offset, right?
>
> if (vmd->msix_count == 1 + vmd->first_vec))
Looks like it, yes
Thank you
>
> > - return &vmd->irqs[0];
> > + return &vmd->irqs[vmd->first_vec];
