iommu/hyper-v: Allow hyperv irq remapping without x2apic
Subsystems (4)
PATCH v2 1/1 iommu/hyper-v: Allow hyperv irq remapping without x2apic
If x2apic is not available, hyperv-iommu skips remapping irqs. This breaks root partition which always needs irqs remapped.
Fix this by allowing irq remapping regardless of x2apic, and change hyperv_enable_irq_remapping() to return IRQ_REMAP_XAPIC_MODE in case x2apic is missing.
Tested with root and non-root hyperv partitions.
--- arch/x86/kernel/cpu/mshyperv.c | 6 ++++++ drivers/iommu/Kconfig | 6 +++--- drivers/iommu/hyperv-iommu.c | 11 ++++++++--- 3 files changed, 17 insertions(+), 6 deletions(-)
arch/x86/kernel/cpu/mshyperv.c +6 -0
@@ -475,6 +475,12 @@ static bool __init ms_hyperv_x2apic_available(void)* (logically) generates MSIs directly to the system APIC irq domain. * There is no HPET, and PCI MSI/MSI-X interrupts are remapped by the * pci-hyperv host bridge. +* +* Note: for a Hyper-V root partition, this will always return false. +* The hypervisor doesn't expose these HYPERV_CPUID_VIRT_STACK_* cpuids by +* default, they are implemented as intercepts by the Windows Hyper-V stack. +* Even a nested root partition (L2 root) will not get them because the +* nested (L1) hypervisor filters them out. */ static bool __init ms_hyperv_msi_ext_dest_id (void ) {
drivers/iommu/Kconfig +3 -3
@@ -474,13 +474,13 @@ config QCOM_IOMMU Support for IOMMU on certain Qualcomm SoCs. config HYPERV_IOMMU - bool "Hyper-V x2APIC IRQ Handling" + bool "Hyper-V IRQ Handling" depends on HYPERV && X86 select IOMMU_API default HYPERV help - Stub IOMMU driver to handle IRQs as to allow Hyper-V Linux - guests to run with x2APIC mode enabled. + Stub IOMMU driver to handle IRQs to support Hyper-V Linux + guest and root partitions. config VIRTIO_IOMMU tristate "Virtio IOMMU driver"
drivers/iommu/hyperv-iommu.c +8 -3
@@ -122,9 +122,12 @@ static int __init hyperv_prepare_irq_remapping(void)const char *name;const struct irq_domain_ops *ops; +/* +* For a Hyper-V root partition, ms_hyperv_msi_ext_dest_id() +* will always return false. +*/ if (!hypervisor_is_type (X86_HYPER_MS_HYPERV ) || - x86_init.hyper.msi_ext_dest_id () || - !x2apic_supported ()) + x86_init.hyper.msi_ext_dest_id ())return -ENODEV ;if (hv_root_partition) {
@@ -170,7 +173,9 @@ static int __init hyperv_prepare_irq_remapping(void)static int __init hyperv_enable_irq_remapping (void ) { -return IRQ_REMAP_X2APIC_MODE ; +if (x2apic_supported ()) +return IRQ_REMAP_X2APIC_MODE ; +return IRQ_REMAP_XAPIC_MODE ; }struct irq_remap_ops hyperv_irq_remap_ops = {
Signature · 1 line
2.25.1
From: Nuno Das Neves <[email protected]> Sent: Thursday, November 17, 2022 12:12 PM
Quoted 87 lines
If x2apic is not available, hyperv-iommu skips remapping irqs. This breaks root partition which always needs irqs remapped.
Fix this by allowing irq remapping regardless of x2apic, and change
hyperv_enable_irq_remapping()to returnIRQ_REMAP_XAPIC_MODEin case x2apic is missing.Tested with root and non-root hyperv partitions.
Signed-off-by Nuno Das Neves <[email protected]>--- arch/x86/kernel/cpu/mshyperv.c | 6 ++++++ drivers/iommu/Kconfig | 6 +++--- drivers/iommu/hyperv-iommu.c | 11 ++++++++--- 3 files changed, 17 insertions(+), 6 deletions(-)arch/x86/kernel/cpu/mshyperv.c +6
-0@@ -475,6 +475,12 @@ static bool __init ms_hyperv_x2apic_available(void)* (logically) generates MSIs directly to the system APIC irq domain. * There is no HPET, and PCI MSI/MSI-X interrupts are remapped by the * pci-hyperv host bridge. +* +* Note: for a Hyper-V root partition, this will always return false. +* The hypervisor doesn't expose these HYPERV_CPUID_VIRT_STACK_* cpuids by +* default, they are implemented as intercepts by the Windows Hyper-V stack. +* Even a nested root partition (L2 root) will not get them because the +* nested (L1) hypervisor filters them out. */ static bool __init ms_hyperv_msi_ext_dest_id (void ) {drivers/iommu/Kconfig +3
-3@@ -474,13 +474,13 @@ config QCOM_IOMMU Support for IOMMU on certain Qualcomm SoCs. config HYPERV_IOMMU - bool "Hyper-V x2APIC IRQ Handling" + bool "Hyper-V IRQ Handling" depends on HYPERV && X86 select IOMMU_API default HYPERV help - Stub IOMMU driver to handle IRQs as to allow Hyper-V Linux - guests to run with x2APIC mode enabled. + Stub IOMMU driver to handle IRQs to support Hyper-V Linux + guest and root partitions. config VIRTIO_IOMMU tristate "Virtio IOMMU driver"drivers/iommu/hyperv-iommu.c +8
-4@@ -122,9 +122,12 @@ static int __init hyperv_prepare_irq_remapping(void)const char *name;const struct irq_domain_ops *ops; +/* +* For a Hyper-V root partition, ms_hyperv_msi_ext_dest_id() +* will always return false. +*/ if (!hypervisor_is_type (X86_HYPER_MS_HYPERV ) || - x86_init.hyper.msi_ext_dest_id () || - !x2apic_supported ()) + x86_init.hyper.msi_ext_dest_id ())return -ENODEV ;if (hv_root_partition) {@@ -170,7 +173,9 @@ static int __init hyperv_prepare_irq_remapping(void)static int __init hyperv_enable_irq_remapping (void ) { -return IRQ_REMAP_X2APIC_MODE ; +if (x2apic_supported ()) +return IRQ_REMAP_X2APIC_MODE ; +return IRQ_REMAP_XAPIC_MODE ; }struct irq_remap_ops hyperv_irq_remap_ops = { --
Quoted 1 line
2.25.1
Quoted 8 lines · Nuno Das Neves
If x2apic is not available, hyperv-iommu skips remapping irqs. This breaks root partition which always needs irqs remapped.
Fix this by allowing irq remapping regardless of x2apic, and change
hyperv_enable_irq_remapping()to returnIRQ_REMAP_XAPIC_MODEin case x2apic is missing.
Quoted 8 lines
Tested with root and non-root hyperv partitions.
Signed-off-by Nuno Das Neves<[email protected]>--- arch/x86/kernel/cpu/mshyperv.c | 6 ++++++ drivers/iommu/Kconfig | 6 +++--- drivers/iommu/hyperv-iommu.c | 11 ++++++++--- 3 files changed, 17 insertions(+), 6 deletions(-)
Quoted 11 lines · Nuno Das Neves
If x2apic is not available, hyperv-iommu skips remapping irqs. This breaks root partition which always needs irqs remapped.
Fix this by allowing irq remapping regardless of x2apic, and change
hyperv_enable_irq_remapping()to returnIRQ_REMAP_XAPIC_MODEin case x2apic is missing.Tested with root and non-root hyperv partitions.
Signed-off-by Nuno Das Neves <[email protected]>
Applied to hyperv-next. Thanks.