6.13-stable review patch. If anyone has any objections, please let me know.
------------------
From: Sean Christopherson seanjc@google.com
commit 548183ea388c12b6d76d6982f3d72df3887af0da upstream.
Set the posted MSI irq_chip's irq_ack() hook to irq_move_irq() instead of a dummy/empty callback so that posted MSIs process pending changes to the IRQ's SMP affinity. Failure to honor a pending set-affinity results in userspace being unable to change the effective affinity of the IRQ, as IRQD_SETAFFINITY_PENDING is never cleared and so irq_set_affinity_locked() always defers moving the IRQ.
The issue is most easily reproducible by setting /proc/irq/xx/smp_affinity multiple times in quick succession, as only the first update is likely to be handled in process context.
Fixes: ed1e48ea4370 ("iommu/vt-d: Enable posted mode for device MSIs") Cc: Robert Lippert rlippert@google.com Cc: Thomas Gleixner tglx@linutronix.de Reported-by: Wentao Yang wentaoyang@google.com Cc: stable@vger.kernel.org Signed-off-by: Sean Christopherson seanjc@google.com Reviewed-by: Thomas Gleixner tglx@linutronix.de Link: https://lore.kernel.org/r/20250321194249.1217961-1-seanjc@google.com Signed-off-by: Lu Baolu baolu.lu@linux.intel.com Signed-off-by: Joerg Roedel jroedel@suse.de Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- drivers/iommu/intel/irq_remapping.c | 29 +++++++++++++++-------------- 1 file changed, 15 insertions(+), 14 deletions(-)
--- a/drivers/iommu/intel/irq_remapping.c +++ b/drivers/iommu/intel/irq_remapping.c @@ -1291,43 +1291,44 @@ static struct irq_chip intel_ir_chip = { };
/* - * With posted MSIs, all vectors are multiplexed into a single notification - * vector. Devices MSIs are then dispatched in a demux loop where - * EOIs can be coalesced as well. + * With posted MSIs, the MSI vectors are multiplexed into a single notification + * vector, and only the notification vector is sent to the APIC IRR. Device + * MSIs are then dispatched in a demux loop that harvests the MSIs from the + * CPU's Posted Interrupt Request bitmap. I.e. Posted MSIs never get sent to + * the APIC IRR, and thus do not need an EOI. The notification handler instead + * performs a single EOI after processing the PIR. * - * "INTEL-IR-POST" IRQ chip does not do EOI on ACK, thus the dummy irq_ack() - * function. Instead EOI is performed by the posted interrupt notification - * handler. + * Note! Pending SMP/CPU affinity changes, which are per MSI, must still be + * honored, only the APIC EOI is omitted. * * For the example below, 3 MSIs are coalesced into one CPU notification. Only - * one apic_eoi() is needed. + * one apic_eoi() is needed, but each MSI needs to process pending changes to + * its CPU affinity. * * __sysvec_posted_msi_notification() * irq_enter(); * handle_edge_irq() * irq_chip_ack_parent() - * dummy(); // No EOI + * irq_move_irq(); // No EOI * handle_irq_event() * driver_handler() * handle_edge_irq() * irq_chip_ack_parent() - * dummy(); // No EOI + * irq_move_irq(); // No EOI * handle_irq_event() * driver_handler() * handle_edge_irq() * irq_chip_ack_parent() - * dummy(); // No EOI + * irq_move_irq(); // No EOI * handle_irq_event() * driver_handler() * apic_eoi() * irq_exit() + * */ - -static void dummy_ack(struct irq_data *d) { } - static struct irq_chip intel_ir_chip_post_msi = { .name = "INTEL-IR-POST", - .irq_ack = dummy_ack, + .irq_ack = irq_move_irq, .irq_set_affinity = intel_ir_set_affinity, .irq_compose_msi_msg = intel_ir_compose_msi_msg, .irq_set_vcpu_affinity = intel_ir_set_vcpu_affinity,