| Message ID | 20191121005607.274347-6-david@gibson.dropbear.id.au |
|---|---|
| State | New |
| Headers | show |
| Series | vfio/spapr: Handle changes of master irq chip for VFIO devices | expand |
On 21/11/2019 01:56, David Gibson wrote: > Traditional PCI INTx for vfio devices can only perform well if using > an in-kernel irqchip. Therefore, vfio_intx_update() issues a warning > if an in kernel irqchip is not available. > > We usually do have an in-kernel irqchip available for pseries machines > on POWER hosts. However, because the platform allows feature > negotiation of what interrupt controller model to use, we don't > currently initialize it until machine reset. vfio_intx_update() is > called (first) from vfio_realize() before that, so it can issue a > spurious warning, even if we will have an in kernel irqchip by the > time we need it. > > To workaround this, make a call to spapr_irq_update_active_intc() from > spapr_irq_init() which is called at machine realize time, before the > vfio realize. This call will be pretty much obsoleted by the later > call at reset time, but it serves to suppress the spurious warning > from VFIO. > > Cc: Alex Williamson <alex.williamson@redhat.com> > Cc: Alexey Kardashevskiy <aik@ozlabs.ru> > > Signed-off-by: David Gibson <david@gibson.dropbear.id.au> Reviewed-by: Cédric Le Goater <clg@kaod.org> > --- > hw/ppc/spapr_irq.c | 11 ++++++++++- > 1 file changed, 10 insertions(+), 1 deletion(-) > > diff --git a/hw/ppc/spapr_irq.c b/hw/ppc/spapr_irq.c > index 1d27034962..d6bb7fd2d6 100644 > --- a/hw/ppc/spapr_irq.c > +++ b/hw/ppc/spapr_irq.c > @@ -373,6 +373,14 @@ void spapr_irq_init(SpaprMachineState *spapr, Error **errp) > > spapr->qirqs = qemu_allocate_irqs(spapr_set_irq, spapr, > smc->nr_xirqs + SPAPR_XIRQ_BASE); > + > + /* > + * Mostly we don't actually need this until reset, except that not > + * having this set up can cause VFIO devices to issue a > + * false-positive warning during realize(), because they don't yet > + * have an in-kernel irq chip. > + */ > + spapr_irq_update_active_intc(spapr); > } > > int spapr_irq_claim(SpaprMachineState *spapr, int irq, bool lsi, Error **errp) > @@ -528,7 +536,8 @@ void spapr_irq_update_active_intc(SpaprMachineState *spapr) > * this. > */ > new_intc = SPAPR_INTC(spapr->xive); > - } else if (spapr_ovec_test(spapr->ov5_cas, OV5_XIVE_EXPLOIT)) { > + } else if (spapr->ov5_cas > + && spapr_ovec_test(spapr->ov5_cas, OV5_XIVE_EXPLOIT)) { > new_intc = SPAPR_INTC(spapr->xive); > } else { > new_intc = SPAPR_INTC(spapr->ics); >
diff --git a/hw/ppc/spapr_irq.c b/hw/ppc/spapr_irq.c index 1d27034962..d6bb7fd2d6 100644 --- a/hw/ppc/spapr_irq.c +++ b/hw/ppc/spapr_irq.c @@ -373,6 +373,14 @@ void spapr_irq_init(SpaprMachineState *spapr, Error **errp) spapr->qirqs = qemu_allocate_irqs(spapr_set_irq, spapr, smc->nr_xirqs + SPAPR_XIRQ_BASE); + + /* + * Mostly we don't actually need this until reset, except that not + * having this set up can cause VFIO devices to issue a + * false-positive warning during realize(), because they don't yet + * have an in-kernel irq chip. + */ + spapr_irq_update_active_intc(spapr); } int spapr_irq_claim(SpaprMachineState *spapr, int irq, bool lsi, Error **errp) @@ -528,7 +536,8 @@ void spapr_irq_update_active_intc(SpaprMachineState *spapr) * this. */ new_intc = SPAPR_INTC(spapr->xive); - } else if (spapr_ovec_test(spapr->ov5_cas, OV5_XIVE_EXPLOIT)) { + } else if (spapr->ov5_cas + && spapr_ovec_test(spapr->ov5_cas, OV5_XIVE_EXPLOIT)) { new_intc = SPAPR_INTC(spapr->xive); } else { new_intc = SPAPR_INTC(spapr->ics);
Traditional PCI INTx for vfio devices can only perform well if using an in-kernel irqchip. Therefore, vfio_intx_update() issues a warning if an in kernel irqchip is not available. We usually do have an in-kernel irqchip available for pseries machines on POWER hosts. However, because the platform allows feature negotiation of what interrupt controller model to use, we don't currently initialize it until machine reset. vfio_intx_update() is called (first) from vfio_realize() before that, so it can issue a spurious warning, even if we will have an in kernel irqchip by the time we need it. To workaround this, make a call to spapr_irq_update_active_intc() from spapr_irq_init() which is called at machine realize time, before the vfio realize. This call will be pretty much obsoleted by the later call at reset time, but it serves to suppress the spurious warning from VFIO. Cc: Alex Williamson <alex.williamson@redhat.com> Cc: Alexey Kardashevskiy <aik@ozlabs.ru> Signed-off-by: David Gibson <david@gibson.dropbear.id.au> --- hw/ppc/spapr_irq.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-)