Re: [PATCH v4 01/15] hw/pci: Add a pci_setup_iommu_ops() helper

[Cédric Le Goater]

On 10/6/23 19:59, Joao Martins wrote:
> On 06/10/2023 18:09, Cédric Le Goater wrote:
> > > > Getting acks from everyone will be difficultsince some PHBs are orphans.
> > >
> > > [...] This is what gets me a bit hesitant
> >
> > orphans shouldn't be an issue, nor the PPC emulated machines. We will see
> > what other maintainers have to say.
>
> How about this as a compromise: to have a separate patch at the end of the
> series that converts every other PHB? This way the rest can iterate, while we
> await maintainers feedback without potentially blocking everything else.

In patch [1], impacted files are :

* PCIIOMMUFunc -> PCIIOMMUOps change in models

  hw/ppc/spapr_pci.c       (David)  R-b
  hw/i386/intel_iommu.c    (Peter Xu) R-b
  hw/arm/smmu-common.c     (Eric)
  hw/virtio/virtio-iommu.c (Eric)
  hw/pci-host/pnv_phb3.c   (Cédric)
  hw/pci-host/pnv_phb4.c   (Cédric)
  hw/pci-host/designware.c (Peter M.)
  hw/pci-host/prep.c       (Hervé)
  hw/pci-host/sabre.c      (Mark)
  hw/s390x/s390-pci-bus.c  (Thomas)
  hw/alpha/typhoon.c       (Richard)
  hw/hppa/dino.c           (Richard)

* Common PCI

  hw/pci/pci.c             (Michael)
  include/hw/pci/pci.h     (Michael)
  include/hw/pci/pci_bus.h (Michael)

* Orphans
 
  hw/i386/amd_iommu.c      (Orphan)
  hw/ppc/ppc440_pcix.c     (Orphan)
  hw/pci-host/ppce500.c    (Orphan)

I will add my R-b on the PPC parts I maintain. The rest doesn't seem it
would raise issues and if so, it should be quick to have since the change
is simple.

[1] https://lore.kernel.org/all/20210302203827.437645-5-yi.l.liu@intel.com/

> Also, one other patch I'll add to this series at the end is this one:
>
> https://lore.kernel.org/qemu-devel/20230908120521.50903-1-joao.m.martins@oracle.com/
>
> This way the vIOMMU series is a complete thing for old and new guests, as
> opposed to just new.

Ok good. Let's have it.

Thanks,

C.