KVM: s390: Implement KVM_PRE_FAULT_MEMORY

Implement and enable the KVM_PRE_FAULT_MEMORY ioctl for s390.

Faulted-in pages will be marked as accessed, unlike x86, otherwise they
will trigger a minor fault when accessed. Avoiding such faults is one of
the points of KVM_PRE_FAULT_MEMORY.

Reviewed-by: Steffen Eiden <seiden@linux.ibm.com>
Signed-off-by: Claudio Imbrenda <imbrenda@linux.ibm.com>
Message-ID: <20260527144358.186359-3-imbrenda@linux.ibm.com>

diff --git a/arch/s390/kvm/Kconfig b/arch/s390/kvm/Kconfig
index 5b835bc6a1941c758dcb116eb78c848ae5af3cbc..8d3ee17a1bcbcf208b2bb81c45c885fc3dbb03dd 100644 (file)
--- a/arch/s390/kvm/Kconfig
+++ b/arch/s390/kvm/Kconfig
@@ -30,6 +30,7 @@ config KVM
        select KVM_VFIO
        select VIRT_XFER_TO_GUEST_WORK
        select KVM_MMU_LOCKLESS_AGING
+       select KVM_GENERIC_PRE_FAULT_MEMORY
        help
          Support hosting paravirtualized guest machines using the SIE
          virtualization capability on the mainframe. This should work

diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c
index e09960c2e6ed1ad54808a2c330c9b145cb01a9e7..f6521f16532ae0a49da2cde891b01e3277025372 100644 (file)
--- a/arch/s390/kvm/kvm-s390.c
+++ b/arch/s390/kvm/kvm-s390.c
@@ -630,6 +630,7 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext)
        case KVM_CAP_S390_USER_OPEREXEC:
        case KVM_CAP_S390_KEYOP:
        case KVM_CAP_S390_VSIE_ESAMODE:
+       case KVM_CAP_PRE_FAULT_MEMORY:
                r = 1;
                break;
        case KVM_CAP_SET_GUEST_DEBUG2:
@@ -5736,6 +5737,50 @@ out:
        return;
 }
 
+/**
+ * kvm_arch_vcpu_pre_fault_memory() -- pre-fault and link gmap dat tables
+ * @vcpu: the vcpu that shall appear to have generated the fault-in.
+ * @range: the range that needs to be faulted in.
+ *
+ * The first page of the given range is faulted in and the corresponding gmap
+ * page tables are created, as if the given vCPU had performed a read
+ * operation.
+ * If the range starts outside any memslots, an error is returned. An error is
+ * also returned for UCONTROL VMs, which should instead use the
+ * KVM_S390_VCPU_FAULT ioctl.
+ *
+ * Return:
+ * * %-ENOENT if the range lies outside of a memslot.
+ * * %-EINVAL in case of invalid state (for example if the VM is UCONTROL).
+ * * %-EIO if errors happen while faulting-in the page (will trigger a warning
+ *   in the caller).
+ * * other error codes < 0 in case of other errors.
+ * * otherwise a number > 0 of bytes that have been faulted in successfully.
+ */
+long kvm_arch_vcpu_pre_fault_memory(struct kvm_vcpu *vcpu, struct kvm_pre_fault_memory *range)
+{
+       struct guest_fault f = { .gfn = gpa_to_gfn(range->gpa), };
+       gpa_t end;
+       int rc;
+
+       if (kvm_is_ucontrol(vcpu->kvm))
+               return -EINVAL;
+
+       rc = kvm_s390_faultin_gfn(vcpu, NULL, &f);
+       if (rc == PGM_ADDRESSING)
+               return -ENOENT;
+       if (rc > 0)
+               return -EIO;
+       if (rc < 0)
+               return rc;
+
+       if (f.ptep)
+               return PAGE_SIZE;
+
+       end = ALIGN(range->gpa + PAGE_SIZE, f.crste_region3 ? _REGION3_SIZE : HPAGE_SIZE);
+       return min(range->size, end - range->gpa);
+}
+
 /**
  * kvm_test_age_gfn() - test young
  * @kvm: the kvm instance