| Message ID | 1443579237-9636-7-git-send-email-david@gibson.dropbear.id.au |
|---|---|
| State | New |
| Headers | show |
On 30/09/2015 04:13, David Gibson wrote: > At present the memory listener used by vfio to keep host IOMMU mappings > in sync with the guest memory image assumes that if a guest IOMMU > appears, then it has no existing mappings. > > This may not be true if a VFIO device is hotplugged onto a guest bus > which didn't previously include a VFIO device, and which has existing > guest IOMMU mappings. > > Therefore, use the memory_region_register_iommu_notifier_replay() > function in order to fix this case, replaying existing guest IOMMU > mappings, bringing the host IOMMU into sync with the guest IOMMU. > > Signed-off-by: David Gibson <david@gibson.dropbear.id.au> > --- > hw/vfio/common.c | 23 +++++++++-------------- > 1 file changed, 9 insertions(+), 14 deletions(-) > > diff --git a/hw/vfio/common.c b/hw/vfio/common.c > index f666de2..6797208 100644 > --- a/hw/vfio/common.c > +++ b/hw/vfio/common.c > @@ -312,6 +312,11 @@ out: > rcu_read_unlock(); > } > > +static hwaddr vfio_container_granularity(VFIOContainer *container) > +{ > + return (hwaddr)1 << ctz64(container->iova_pgsizes); > +} > + > static void vfio_listener_region_add(MemoryListener *listener, > MemoryRegionSection *section) > { > @@ -369,26 +374,16 @@ static void vfio_listener_region_add(MemoryListener *listener, > * would be the right place to wire that up (tell the KVM > * device emulation the VFIO iommu handles to use). > */ > - /* > - * This assumes that the guest IOMMU is empty of > - * mappings at this point. > - * > - * One way of doing this is: > - * 1. Avoid sharing IOMMUs between emulated devices or different > - * IOMMU groups. > - * 2. Implement VFIO_IOMMU_ENABLE in the host kernel to fail if > - * there are some mappings in IOMMU. > - * > - * VFIO on SPAPR does that. Other IOMMU models may do that different, > - * they must make sure there are no existing mappings or > - * loop through existing mappings to map them into VFIO. > - */ > giommu = g_malloc0(sizeof(*giommu)); > giommu->iommu = section->mr; > giommu->container = container; > giommu->n.notify = vfio_iommu_map_notify; > QLIST_INSERT_HEAD(&container->giommu_list, giommu, giommu_next); > + > memory_region_register_iommu_notifier(giommu->iommu, &giommu->n); > + memory_region_iommu_replay(giommu->iommu, &giommu->n, > + vfio_container_granularity(container), > + false); I'm wondering if it has any sense to provide the "is_write" information at this level of the API: I don't think we can have access to this information when we call this function (so it will be always used with false, or called twice once with false, once with true). I think it would be better to manage this internally. - > > return; > } >
On Wed, Sep 30, 2015 at 11:09:17AM +0200, Laurent Vivier wrote: > > > On 30/09/2015 04:13, David Gibson wrote: > > At present the memory listener used by vfio to keep host IOMMU mappings > > in sync with the guest memory image assumes that if a guest IOMMU > > appears, then it has no existing mappings. > > > > This may not be true if a VFIO device is hotplugged onto a guest bus > > which didn't previously include a VFIO device, and which has existing > > guest IOMMU mappings. > > > > Therefore, use the memory_region_register_iommu_notifier_replay() > > function in order to fix this case, replaying existing guest IOMMU > > mappings, bringing the host IOMMU into sync with the guest IOMMU. > > > > Signed-off-by: David Gibson <david@gibson.dropbear.id.au> > > --- > > hw/vfio/common.c | 23 +++++++++-------------- > > 1 file changed, 9 insertions(+), 14 deletions(-) > > > > diff --git a/hw/vfio/common.c b/hw/vfio/common.c > > index f666de2..6797208 100644 > > --- a/hw/vfio/common.c > > +++ b/hw/vfio/common.c > > @@ -312,6 +312,11 @@ out: > > rcu_read_unlock(); > > } > > > > +static hwaddr vfio_container_granularity(VFIOContainer *container) > > +{ > > + return (hwaddr)1 << ctz64(container->iova_pgsizes); > > +} > > + > > static void vfio_listener_region_add(MemoryListener *listener, > > MemoryRegionSection *section) > > { > > @@ -369,26 +374,16 @@ static void vfio_listener_region_add(MemoryListener *listener, > > * would be the right place to wire that up (tell the KVM > > * device emulation the VFIO iommu handles to use). > > */ > > - /* > > - * This assumes that the guest IOMMU is empty of > > - * mappings at this point. > > - * > > - * One way of doing this is: > > - * 1. Avoid sharing IOMMUs between emulated devices or different > > - * IOMMU groups. > > - * 2. Implement VFIO_IOMMU_ENABLE in the host kernel to fail if > > - * there are some mappings in IOMMU. > > - * > > - * VFIO on SPAPR does that. Other IOMMU models may do that different, > > - * they must make sure there are no existing mappings or > > - * loop through existing mappings to map them into VFIO. > > - */ > > giommu = g_malloc0(sizeof(*giommu)); > > giommu->iommu = section->mr; > > giommu->container = container; > > giommu->n.notify = vfio_iommu_map_notify; > > QLIST_INSERT_HEAD(&container->giommu_list, giommu, giommu_next); > > + > > memory_region_register_iommu_notifier(giommu->iommu, &giommu->n); > > + memory_region_iommu_replay(giommu->iommu, &giommu->n, > > + vfio_container_granularity(container), > > + false); > > I'm wondering if it has any sense to provide the "is_write" information > at this level of the API: I don't think we can have access to this > information when we call this function (so it will be always used with > false, or called twice once with false, once with true). I think it > would be better to manage this internally. I agree it's pretty ugly, but I'm not really sure how to handle it better. The translate function itself wants is_write; I'm pretty sure "false" is the right thing here, but I'm not sure it would be right for all potential replay cases.
diff --git a/hw/vfio/common.c b/hw/vfio/common.c index f666de2..6797208 100644 --- a/hw/vfio/common.c +++ b/hw/vfio/common.c @@ -312,6 +312,11 @@ out: rcu_read_unlock(); } +static hwaddr vfio_container_granularity(VFIOContainer *container) +{ + return (hwaddr)1 << ctz64(container->iova_pgsizes); +} + static void vfio_listener_region_add(MemoryListener *listener, MemoryRegionSection *section) { @@ -369,26 +374,16 @@ static void vfio_listener_region_add(MemoryListener *listener, * would be the right place to wire that up (tell the KVM * device emulation the VFIO iommu handles to use). */ - /* - * This assumes that the guest IOMMU is empty of - * mappings at this point. - * - * One way of doing this is: - * 1. Avoid sharing IOMMUs between emulated devices or different - * IOMMU groups. - * 2. Implement VFIO_IOMMU_ENABLE in the host kernel to fail if - * there are some mappings in IOMMU. - * - * VFIO on SPAPR does that. Other IOMMU models may do that different, - * they must make sure there are no existing mappings or - * loop through existing mappings to map them into VFIO. - */ giommu = g_malloc0(sizeof(*giommu)); giommu->iommu = section->mr; giommu->container = container; giommu->n.notify = vfio_iommu_map_notify; QLIST_INSERT_HEAD(&container->giommu_list, giommu, giommu_next); + memory_region_register_iommu_notifier(giommu->iommu, &giommu->n); + memory_region_iommu_replay(giommu->iommu, &giommu->n, + vfio_container_granularity(container), + false); return; }
At present the memory listener used by vfio to keep host IOMMU mappings in sync with the guest memory image assumes that if a guest IOMMU appears, then it has no existing mappings. This may not be true if a VFIO device is hotplugged onto a guest bus which didn't previously include a VFIO device, and which has existing guest IOMMU mappings. Therefore, use the memory_region_register_iommu_notifier_replay() function in order to fix this case, replaying existing guest IOMMU mappings, bringing the host IOMMU into sync with the guest IOMMU. Signed-off-by: David Gibson <david@gibson.dropbear.id.au> --- hw/vfio/common.c | 23 +++++++++-------------- 1 file changed, 9 insertions(+), 14 deletions(-)