[PATCH v2] KVM: arm64: Move bundling vLPI and vSGI to vgic_supports_direct_msis()

[PATCH v2] KVM: arm64: Move bundling vLPI and vSGI to vgic_supports_direct_msis()

[Raghavendra Rao Ananta]

Commit <c652887a9288> ("KVM: arm64: vgic-v3: Allow userspace to write
GICD_TYPER2.nASSGIcap") bundles the vLPIs and vSGIs behind the
GICD_TYPER2.nASSGIcap field. While the vGIC v4 initialization and
teardown is handled correctly, it erroneously left out the cases when
KVM sets/unset vGIC v4 forwarding, which leads to a kernel panic of the
following nature:

  Unable to handle kernel NULL pointer dereference at virtual address 00000000000000a8
   Mem abort info:
     ESR = 0x0000000096000044
     EC = 0x25: DABT (current EL), IL = 32 bits
     SET = 0, FnV = 0
     EA = 0, S1PTW = 0
     FSC = 0x04: level 0 translation fault
   Data abort info:
     ISV = 0, ISS = 0x00000044, ISS2 = 0x00000000
     CM = 0, WnR = 1, TnD = 0, TagAccess = 0
     GCS = 0, Overlay = 0, DirtyBit = 0, Xs = 0
   user pgtable: 4k pages, 48-bit VAs, pgdp=00000073a453b000
   [00000000000000a8] pgd=0000000000000000, p4d=0000000000000000
   Internal error: Oops: 0000000096000044 [#1] SMP
   pstate: 23400009 (nzCv daif +PAN -UAO +TCO +DIT -SSBS BTYPE=--)
   pc : its_irq_set_vcpu_affinity+0x58c/0x95c
   lr : its_irq_set_vcpu_affinity+0x1e0/0x95c
   sp : ffff8001029bb9e0
   pmr_save: 00000060
   x29: ffff8001029bba20 x28: ffff0001ca5e28c0 x27: 0000000000000000
   x26: 0000000000000000 x25: ffff00019eee9f80 x24: ffff0001992b3f00
   x23: ffff8001029bbab8 x22: ffff00001159fb80 x21: 00000000000024a7
   x20: 00000000000024a7 x19: ffff00019eee9fb4 x18: 0000000000000494
   x17: 000000000000000e x16: 0000000000000494 x15: 0000000000000002
   x14: ffff0001a7f34600 x13: ffffccaad1203000 x12: 0000000000000018
   x11: ffff000011991000 x10: 0000000000000000 x9 : 00000000000000a2
   x8 : 00000000000020a8 x7 : 0000000000000000 x6 : 000000000000003f
   x5 : 0000000000000040 x4 : 0000000000000000 x3 : 0000000000000004
   x2 : 0000000000000000 x1 : ffff8001029bbab8 x0 : 00000000000000a8
   Call trace:
    its_irq_set_vcpu_affinity+0x58c/0x95c
    irq_set_vcpu_affinity+0x74/0xc8
    its_map_vlpi+0x4c/0x94
    kvm_vgic_v4_set_forwarding+0x134/0x298
    kvm_arch_irq_bypass_add_producer+0x28/0x34
    irq_bypass_register_producer+0xf8/0x1d8
    vfio_msi_set_vector_signal+0x2c8/0x308
    vfio_pci_set_msi_trigger+0x198/0x2d4
    vfio_pci_set_irqs_ioctl+0xf0/0x104
    vfio_pci_core_ioctl+0x6ac/0xc5c
    vfio_device_fops_unl_ioctl+0x128/0x370
    __arm64_sys_ioctl+0x98/0xd0
    el0_svc_common+0xd8/0x1d8
    do_el0_svc+0x28/0x34
    el0_svc+0x40/0xb8
    el0t_64_sync_handler+0x70/0xbc
    el0t_64_sync+0x1a8/0x1ac
   Code: 321f0129 f940094a 8b080148 d1400900 (39000009)
   ---[ end trace 0000000000000000 ]---

Fix this by moving the bundling logic to vgic_supports_direct_msis()
instead. This would still handle the vGIC v4 initialization
appropriately, while also helping with the vLPI set/unset forwarding.

Fixes: c652887a9288 ("KVM: arm64: vgic-v3: Allow userspace to write GICD_TYPER2.nASSGIcap")
Suggested-by: Oliver Upton <oliver.upton@linux.dev>
Signed-off-by: Raghavendra Rao Ananta <rananta@google.com>
---
 arch/arm64/kvm/vgic/vgic-mmio-v3.c |  8 ++++++++
 arch/arm64/kvm/vgic/vgic.h         | 10 +---------
 2 files changed, 9 insertions(+), 9 deletions(-)

diff --git a/arch/arm64/kvm/vgic/vgic-mmio-v3.c b/arch/arm64/kvm/vgic/vgic-mmio-v3.c
index a3ef185209e9..70d50c77e5dc 100644
--- a/arch/arm64/kvm/vgic/vgic-mmio-v3.c
+++ b/arch/arm64/kvm/vgic/vgic-mmio-v3.c
@@ -50,6 +50,14 @@ bool vgic_has_its(struct kvm *kvm)
 
 bool vgic_supports_direct_msis(struct kvm *kvm)
 {
+	/*
+	 * Deliberately conflate vLPI and vSGI support on GICv4.1 hardware,
+	 * indirectly allowing userspace to control whether or not vPEs are
+	 * allocated for the VM.
+	 */
+	if (system_supports_direct_sgis() && !vgic_supports_direct_sgis(kvm))
+		return false;
+
 	return kvm_vgic_global_state.has_gicv4 && vgic_has_its(kvm);
 }
 
diff --git a/arch/arm64/kvm/vgic/vgic.h b/arch/arm64/kvm/vgic/vgic.h
index 1384a04c0784..de1c1d3261c3 100644
--- a/arch/arm64/kvm/vgic/vgic.h
+++ b/arch/arm64/kvm/vgic/vgic.h
@@ -396,15 +396,7 @@ bool vgic_supports_direct_sgis(struct kvm *kvm);
 
 static inline bool vgic_supports_direct_irqs(struct kvm *kvm)
 {
-	/*
-	 * Deliberately conflate vLPI and vSGI support on GICv4.1 hardware,
-	 * indirectly allowing userspace to control whether or not vPEs are
-	 * allocated for the VM.
-	 */
-	if (system_supports_direct_sgis())
-		return vgic_supports_direct_sgis(kvm);
-
-	return vgic_supports_direct_msis(kvm);
+	return vgic_supports_direct_msis(kvm) || vgic_supports_direct_sgis(kvm);
 }
 
 int vgic_v4_init(struct kvm *kvm);

base-commit: 18ec25dd0e97653cdb576bb1750c31acf2513ea7
-- 
2.50.1.552.g942d659e1b-goog

Re: [PATCH v2] KVM: arm64: Move bundling vLPI and vSGI to vgic_supports_direct_msis()

[Oliver Upton]

On Tue, 29 Jul 2025 21:06:44 +0000, Raghavendra Rao Ananta wrote:
> Commit <c652887a9288> ("KVM: arm64: vgic-v3: Allow userspace to write
> GICD_TYPER2.nASSGIcap") bundles the vLPIs and vSGIs behind the
> GICD_TYPER2.nASSGIcap field. While the vGIC v4 initialization and
> teardown is handled correctly, it erroneously left out the cases when
> KVM sets/unset vGIC v4 forwarding, which leads to a kernel panic of the
> following nature:
> 
> [...]

Massaged the changelog a bit. Applied to fixes, thanks!

[1/1] KVM: arm64: Move bundling vLPI and vSGI to vgic_supports_direct_msis()
      https://git.kernel.org/kvmarm/kvmarm/c/7b8346bd9fce

--
Best,
Oliver