[PATCH v4 3/3] KVM: PPC: Add KVM_CAP_PPC_AIL_MODE_3

[Nicholas Piggin]

Add KVM_CAP_PPC_AIL_MODE_3 to advertise the capability to set the AIL
resource mode to 3 with the H_SET_MODE hypercall. This capability
differs between processor types and KVM types (PR, HV, Nested HV), and
affects guest-visible behaviour.

QEMU will implement a cap-ail-mode-3 to control this behaviour[1], and
use the KVM CAP if available to determine KVM support[2].

[1] https://lists.nongnu.org/archive/html/qemu-ppc/2022-02/msg00437.html
[2] https://lists.nongnu.org/archive/html/qemu-ppc/2022-02/msg00439.html

Reviewed-by: Fabiano Rosas <farosas at linux.ibm.com>
Signed-off-by: Nicholas Piggin <npiggin at gmail.com>
---
 Documentation/virt/kvm/api.rst         | 14 ++++++++++++++
 arch/powerpc/include/asm/setup.h       |  2 ++
 arch/powerpc/kvm/powerpc.c             | 17 +++++++++++++++++
 arch/powerpc/platforms/pseries/setup.c | 12 +++++++++++-
 include/uapi/linux/kvm.h               |  1 +
 tools/include/uapi/linux/kvm.h         |  1 +
 6 files changed, 46 insertions(+), 1 deletion(-)

diff --git a/Documentation/virt/kvm/api.rst b/Documentation/virt/kvm/api.rst
index a4267104db50..9954568c7eab 100644
--- a/Documentation/virt/kvm/api.rst
+++ b/Documentation/virt/kvm/api.rst
@@ -6997,6 +6997,20 @@ indicated by the fd to the VM this is called on.
 This is intended to support intra-host migration of VMs between userspace VMMs,
 upgrading the VMM process without interrupting the guest.
 
+7.30 KVM_CAP_PPC_AIL_MODE_3
+-------------------------------
+
+:Capability: KVM_CAP_PPC_AIL_MODE_3
+:Architectures: ppc
+:Type: vm
+
+This capability indicates that the kernel supports the mode 3 setting for the
+"Address Translation Mode on Interrupt" aka "Alternate Interrupt Location"
+resource that is controlled with the H_SET_MODE hypercall.
+
+This capability allows a guest kernel to use a better-performance mode for
+handling interrupts and system calls.
+
 8. Other capabilities.
 ======================
 
diff --git a/arch/powerpc/include/asm/setup.h b/arch/powerpc/include/asm/setup.h
index d0d3dd531c7f..a555fb77258a 100644
--- a/arch/powerpc/include/asm/setup.h
+++ b/arch/powerpc/include/asm/setup.h
@@ -28,11 +28,13 @@ void setup_panic(void);
 #define ARCH_PANIC_TIMEOUT 180
 
 #ifdef CONFIG_PPC_PSERIES
+extern bool pseries_reloc_on_exception(void);
 extern bool pseries_enable_reloc_on_exc(void);
 extern void pseries_disable_reloc_on_exc(void);
 extern void pseries_big_endian_exceptions(void);
 void __init pseries_little_endian_exceptions(void);
 #else
+static inline bool pseries_reloc_on_exception(void) { return false; }
 static inline bool pseries_enable_reloc_on_exc(void) { return false; }
 static inline void pseries_disable_reloc_on_exc(void) {}
 static inline void pseries_big_endian_exceptions(void) {}
diff --git a/arch/powerpc/kvm/powerpc.c b/arch/powerpc/kvm/powerpc.c
index 2ad0ccd202d5..56a6b66d16fe 100644
--- a/arch/powerpc/kvm/powerpc.c
+++ b/arch/powerpc/kvm/powerpc.c
@@ -678,6 +678,23 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext)
 		r = 1;
 		break;
 #endif
+	case KVM_CAP_PPC_AIL_MODE_3:
+		r = 0;
+		/*
+		 * KVM PR, POWER7, and some POWER9s don't support AIL=3 mode.
+		 * The POWER9s can support it if the guest runs in hash mode,
+		 * but QEMU doesn't necessarily query the capability in time.
+		 */
+		if (hv_enabled) {
+			if (kvmhv_on_pseries()) {
+				if (pseries_reloc_on_exception())
+					r = 1;
+			} else if (cpu_has_feature(CPU_FTR_ARCH_207S) &&
+				  !cpu_has_feature(CPU_FTR_P9_RADIX_PREFETCH_BUG)) {
+				r = 1;
+			}
+		}
+		break;
 	default:
 		r = 0;
 		break;
diff --git a/arch/powerpc/platforms/pseries/setup.c b/arch/powerpc/platforms/pseries/setup.c
index 83a04d967a59..182525c2abd5 100644
--- a/arch/powerpc/platforms/pseries/setup.c
+++ b/arch/powerpc/platforms/pseries/setup.c
@@ -353,6 +353,13 @@ static void pseries_lpar_idle(void)
 	pseries_idle_epilog();
 }
 
+static bool pseries_reloc_on_exception_enabled;
+
+bool pseries_reloc_on_exception(void)
+{
+	return pseries_reloc_on_exception_enabled;
+}
+
 /*
  * Enable relocation on during exceptions. This has partition wide scope and
  * may take a while to complete, if it takes longer than one second we will
@@ -377,6 +384,7 @@ bool pseries_enable_reloc_on_exc(void)
 					" on exceptions: %ld\n", rc);
 				return false;
 			}
+			pseries_reloc_on_exception_enabled = true;
 			return true;
 		}
 
@@ -404,7 +412,9 @@ void pseries_disable_reloc_on_exc(void)
 			break;
 		mdelay(get_longbusy_msecs(rc));
 	}
-	if (rc != H_SUCCESS)
+	if (rc == H_SUCCESS)
+		pseries_reloc_on_exception_enabled = false;
+	else
 		pr_warn("Warning: Failed to disable relocation on exceptions: %ld\n",
 			rc);
 }
diff --git a/include/uapi/linux/kvm.h b/include/uapi/linux/kvm.h
index 5191b57e1562..507ee1f2aa96 100644
--- a/include/uapi/linux/kvm.h
+++ b/include/uapi/linux/kvm.h
@@ -1134,6 +1134,7 @@ struct kvm_ppc_resize_hpt {
 #define KVM_CAP_VM_GPA_BITS 207
 #define KVM_CAP_XSAVE2 208
 #define KVM_CAP_SYS_ATTRIBUTES 209
+#define KVM_CAP_PPC_AIL_MODE_3 210
 
 #ifdef KVM_CAP_IRQ_ROUTING
 
diff --git a/tools/include/uapi/linux/kvm.h b/tools/include/uapi/linux/kvm.h
index 5191b57e1562..507ee1f2aa96 100644
--- a/tools/include/uapi/linux/kvm.h
+++ b/tools/include/uapi/linux/kvm.h
@@ -1134,6 +1134,7 @@ struct kvm_ppc_resize_hpt {
 #define KVM_CAP_VM_GPA_BITS 207
 #define KVM_CAP_XSAVE2 208
 #define KVM_CAP_SYS_ATTRIBUTES 209
+#define KVM_CAP_PPC_AIL_MODE_3 210
 
 #ifdef KVM_CAP_IRQ_ROUTING
 
-- 
2.23.0