Re: [PATCH v7 20/22] KVM: s390: add KVM_S390_ZPCI_OP to manage guest zPCI devices

From: Christian Borntraeger
Date: Mon May 16 2022 - 12:37:49 EST


Am 16.05.22 um 17:35 schrieb Matthew Rosato:
On 5/16/22 5:52 AM, Thomas Huth wrote:
On 13/05/2022 21.15, Matthew Rosato wrote:
The KVM_S390_ZPCI_OP ioctl provides a mechanism for managing
hardware-assisted virtualization features for s390X zPCI passthrough.

s/s390X/s390x/

Add the first 2 operations, which can be used to enable/disable
the specified device for Adapter Event Notification interpretation.

Signed-off-by: Matthew Rosato <mjrosato@xxxxxxxxxxxxx>
---
  Documentation/virt/kvm/api.rst | 45 +++++++++++++++++++
  arch/s390/kvm/kvm-s390.c       | 23 ++++++++++
  arch/s390/kvm/pci.c            | 81 ++++++++++++++++++++++++++++++++++
  arch/s390/kvm/pci.h            |  2 +
  include/uapi/linux/kvm.h       | 31 +++++++++++++
  5 files changed, 182 insertions(+)

diff --git a/Documentation/virt/kvm/api.rst b/Documentation/virt/kvm/api.rst
index 4a900cdbc62e..a7cd5ebce031 100644
--- a/Documentation/virt/kvm/api.rst
+++ b/Documentation/virt/kvm/api.rst
@@ -5645,6 +5645,51 @@ enabled with ``arch_prctl()``, but this may change in the future.
  The offsets of the state save areas in struct kvm_xsave follow the contents
  of CPUID leaf 0xD on the host.
+4.135 KVM_S390_ZPCI_OP
+--------------------
+
+:Capability: KVM_CAP_S390_ZPCI_OP
+:Architectures: s390
+:Type: vcpu ioctl

vcpu? ... you're wiring it up in  kvm_arch_vm_ioctl() later, so I assume it's rather a VM ioctl?

Yup, VM ioctl, bad copy/paste job...

Can you maybe resend just the 1 or 2 patches with feedback? In the end this series might be "old" and
good enough to still be queued for the next merge window.
Would be good if Jason (Gunthorpe) would double check that his concerns are addressed and I will
have a look at the patches without RB/ACK.