RE: [EXT] Re: [PATCH v5] iommu/virtio: Use page size bitmap supported by endpoint

From: Bharat Bhushan
Date: Thu May 07 2020 - 07:25:52 EST




> -----Original Message-----
> From: Michael S. Tsirkin <mst@xxxxxxxxxx>
> Sent: Wednesday, May 6, 2020 5:53 AM
> To: Bharat Bhushan <bbhushan2@xxxxxxxxxxx>
> Cc: jean-philippe@xxxxxxxxxx; joro@xxxxxxxxxx; jasowang@xxxxxxxxxx;
> virtualization@xxxxxxxxxxxxxxxxxxxxxxxxxx; iommu@xxxxxxxxxxxxxxxxxxxxxxxxxx;
> linux-kernel@xxxxxxxxxxxxxxx; eric.auger.pro@xxxxxxxxx; eric.auger@xxxxxxxxxx
> Subject: [EXT] Re: [PATCH v5] iommu/virtio: Use page size bitmap supported by
> endpoint
>
> External Email
>
> ----------------------------------------------------------------------
> On Tue, May 05, 2020 at 03:00:04PM +0530, Bharat Bhushan wrote:
> > Different endpoint can support different page size, probe endpoint if
> > it supports specific page size otherwise use global page sizes.
> >
> > Signed-off-by: Bharat Bhushan <bbhushan2@xxxxxxxxxxx>
> > ---
> > v4->v5:
> > - Rebase to Linux v5.7-rc4
> >
> > v3->v4:
> > - Fix whitespace error
> >
> > v2->v3:
> > - Fixed error return for incompatible endpoint
> > - __u64 changed to __le64 in header file
> >
> > drivers/iommu/virtio-iommu.c | 48 ++++++++++++++++++++++++++++---
> > include/uapi/linux/virtio_iommu.h | 7 +++++
> > 2 files changed, 51 insertions(+), 4 deletions(-)
> >
> > diff --git a/drivers/iommu/virtio-iommu.c
> > b/drivers/iommu/virtio-iommu.c index d5cac4f46ca5..9513d2ab819e 100644
> > --- a/drivers/iommu/virtio-iommu.c
> > +++ b/drivers/iommu/virtio-iommu.c
> > @@ -78,6 +78,7 @@ struct viommu_endpoint {
> > struct viommu_dev *viommu;
> > struct viommu_domain *vdomain;
> > struct list_head resv_regions;
> > + u64 pgsize_bitmap;
> > };
> >
> > struct viommu_request {
> > @@ -415,6 +416,19 @@ static int viommu_replay_mappings(struct
> viommu_domain *vdomain)
> > return ret;
> > }
> >
> > +static int viommu_set_pgsize_bitmap(struct viommu_endpoint *vdev,
> > + struct virtio_iommu_probe_pgsize_mask *mask,
> > + size_t len)
> > +{
> > + u64 pgsize_bitmap = le64_to_cpu(mask->pgsize_bitmap);
> > +
> > + if (len < sizeof(*mask))
>
> This is too late to validate length, you have dereferenced it already.
> do it before the read pls.

Yes, Will change here and other places as well

>
> > + return -EINVAL;
>
> OK but note that guest will then just proceed to ignore the property. Is that really
> OK? Wouldn't host want to know?


Guest need to be in sync with device, so yes seems like guest need to tell device which page-size-mask it is using.

Corresponding spec change patch (https://www.mail-archive.com/virtio-dev@xxxxxxxxxxxxxxxxxxxx/msg06214.html)

Would like Jean/Eric to comment here as well.

>
>
> > +
> > + vdev->pgsize_bitmap = pgsize_bitmap;
>
> what if bitmap is 0? Is that a valid size? I see a bunch of BUG_ON with that value ...

As per spec proposed device is supposed to set at-least one bit.
Will add a bug_on her.
Should we add bug_on or switch to global config page-size mask if this is zero (notify device which page-size-mask it is using).

>
> I also see a bunch of code like e.g. this:
>
> pg_size = 1UL << __ffs(pgsize_bitmap);
>
> which probably won't DTRT on a 32 bit guest if the bitmap has bits set in the high
> word.
>

My thought is that in that case viommu_domain_finalise() will fail, do not proceed.

>
>
> > + return 0;
> > +}
> > +
> > static int viommu_add_resv_mem(struct viommu_endpoint *vdev,
> > struct virtio_iommu_probe_resv_mem *mem,
> > size_t len)
> > @@ -499,6 +513,9 @@ static int viommu_probe_endpoint(struct viommu_dev
> *viommu, struct device *dev)
> > case VIRTIO_IOMMU_PROBE_T_RESV_MEM:
> > ret = viommu_add_resv_mem(vdev, (void *)prop, len);
> > break;
> > + case VIRTIO_IOMMU_PROBE_T_PAGE_SIZE_MASK:
> > + ret = viommu_set_pgsize_bitmap(vdev, (void *)prop, len);
> > + break;
> > default:
> > dev_err(dev, "unknown viommu prop 0x%x\n", type);
> > }
> > @@ -630,7 +647,7 @@ static int viommu_domain_finalise(struct
> > viommu_endpoint *vdev,
> >
> > vdomain->id = (unsigned int)ret;
> >
> > - domain->pgsize_bitmap = viommu->pgsize_bitmap;
> > + domain->pgsize_bitmap = vdev->pgsize_bitmap;
> > domain->geometry = viommu->geometry;
> >
> > vdomain->map_flags = viommu->map_flags;
> > @@ -654,6 +671,29 @@ static void viommu_domain_free(struct iommu_domain
> *domain)
> > kfree(vdomain);
> > }
> >
> > +/*
> > + * Check whether the endpoint's capabilities are compatible with
> > +other
> > + * endpoints in the domain. Report any inconsistency.
> > + */
> > +static bool viommu_endpoint_is_compatible(struct viommu_endpoint *vdev,
> > + struct viommu_domain *vdomain) {
> > + struct device *dev = vdev->dev;
> > +
> > + if (vdomain->viommu != vdev->viommu) {
> > + dev_err(dev, "cannot attach to foreign vIOMMU\n");
> > + return false;
> > + }
> > +
> > + if (vdomain->domain.pgsize_bitmap != vdev->pgsize_bitmap) {
> > + dev_err(dev, "incompatible domain bitmap 0x%lx != 0x%llx\n",
> > + vdomain->domain.pgsize_bitmap, vdev->pgsize_bitmap);
> > + return false;
> > + }
>
> I'm confused by this. So let's assume host supports pages sizes of 4k, 2M, 1G. It
> signals this in the properties. Nice.
> Now domain supports 4k, 2M and that's all. Why is that a problem?
> Just don't use 1G ...

Is not it too to change the existing domain properties, for devices already attached to domain? New devices must match to domain page-size.

>
>
> > +
> > + return true;
> > +}
> > +
> > static int viommu_attach_dev(struct iommu_domain *domain, struct
> > device *dev) {
> > int i;
> > @@ -670,9 +710,8 @@ static int viommu_attach_dev(struct iommu_domain
> *domain, struct device *dev)
> > * owns it.
> > */
> > ret = viommu_domain_finalise(vdev, domain);
> > - } else if (vdomain->viommu != vdev->viommu) {
> > - dev_err(dev, "cannot attach to foreign vIOMMU\n");
> > - ret = -EXDEV;
> > + } else if (!viommu_endpoint_is_compatible(vdev, vdomain)) {
> > + ret = -EINVAL;
> > }
> > mutex_unlock(&vdomain->mutex);
> >
> > @@ -886,6 +925,7 @@ static int viommu_add_device(struct device *dev)
> >
> > vdev->dev = dev;
> > vdev->viommu = viommu;
> > + vdev->pgsize_bitmap = viommu->pgsize_bitmap;
> > INIT_LIST_HEAD(&vdev->resv_regions);
> > dev_iommu_priv_set(dev, vdev);
> >
> > diff --git a/include/uapi/linux/virtio_iommu.h
> > b/include/uapi/linux/virtio_iommu.h
> > index 48e3c29223b5..2cced7accc99 100644
> > --- a/include/uapi/linux/virtio_iommu.h
> > +++ b/include/uapi/linux/virtio_iommu.h
>
> As any virtio UAPI change, you need to copy virtio TC at some point before this is
> merged ...

Jean already send patch for same
https://www.mail-archive.com/virtio-dev@xxxxxxxxxxxxxxxxxxxx/msg06214.html

Do we need to do anything additional?

>
> > @@ -111,6 +111,7 @@ struct virtio_iommu_req_unmap {
> >
> > #define VIRTIO_IOMMU_PROBE_T_NONE 0
> > #define VIRTIO_IOMMU_PROBE_T_RESV_MEM 1
> > +#define VIRTIO_IOMMU_PROBE_T_PAGE_SIZE_MASK 2
> >
> > #define VIRTIO_IOMMU_PROBE_T_MASK 0xfff
> >
>
> Does host need to know that guest will ignore the page size mask?
> Maybe we need a feature bit.
>
> > @@ -119,6 +120,12 @@ struct virtio_iommu_probe_property {
> > __le16 length;
> > };
> >
> > +struct virtio_iommu_probe_pgsize_mask {
> > + struct virtio_iommu_probe_property head;
> > + __u8 reserved[4];
> > + __le64 pgsize_bitmap;
> > +};
> > +
>
> This is UAPI. Document the format of pgsize_bitmap please.

Ok,

Thanks
-Bharat

>
>
> > #define VIRTIO_IOMMU_RESV_MEM_T_RESERVED 0
> > #define VIRTIO_IOMMU_RESV_MEM_T_MSI 1
> >
> > --
> > 2.17.1