Message ID | 1584880579-12178-9-git-send-email-yi.l.liu@intel.com (mailing list archive) |
---|---|
State | New, archived |
Headers | show |
Series | intel_iommu: expose Shared Virtual Addressing to VMs | expand |
> From: Peter Xu <peterx@redhat.com> > Sent: Tuesday, March 24, 2020 5:40 AM > To: Liu, Yi L <yi.l.liu@intel.com> > Subject: Re: [PATCH v1 08/22] vfio: init HostIOMMUContext per-container > > On Sun, Mar 22, 2020 at 05:36:05AM -0700, Liu Yi L wrote: > > After confirming dual stage DMA translation support with kernel by > > checking VFIO_TYPE1_NESTING_IOMMU, VFIO inits HostIOMMUContet instance > > and exposes it to PCI layer. Thus vIOMMU emualtors may make use of > > such capability by leveraging the methods provided by HostIOMMUContext. > > > > Cc: Kevin Tian <kevin.tian@intel.com> > > Cc: Jacob Pan <jacob.jun.pan@linux.intel.com> > > Cc: Peter Xu <peterx@redhat.com> > > Cc: Eric Auger <eric.auger@redhat.com> > > Cc: Yi Sun <yi.y.sun@linux.intel.com> > > Cc: David Gibson <david@gibson.dropbear.id.au> > > Cc: Alex Williamson <alex.williamson@redhat.com> > > Signed-off-by: Liu Yi L <yi.l.liu@intel.com> > > --- > > hw/vfio/common.c | 80 +++++++++++++++++++++++++++++++++++ > > hw/vfio/pci.c | 13 ++++++ > > include/hw/iommu/host_iommu_context.h | 3 ++ > > include/hw/vfio/vfio-common.h | 4 ++ > > 4 files changed, 100 insertions(+) > > > > diff --git a/hw/vfio/common.c b/hw/vfio/common.c > > index c276732..e4f5f10 100644 > > --- a/hw/vfio/common.c > > +++ b/hw/vfio/common.c > > @@ -1179,10 +1179,55 @@ static int vfio_get_iommu_type(VFIOContainer > *container, > > return -EINVAL; > > } > > > > +static int vfio_host_icx_pasid_alloc(HostIOMMUContext *host_icx, > > I'm not sure about Alex, but ... icx is confusing to me. Maybe "ctx" > as you always used? At first I used vfio_host_iommu_ctx_pasid_alloc(), found it is long, so I switched to "icx" which means iommu_context. Maybe the former one looks better as it gives more precise info. Regards, Yi Liu
On Tue, Mar 24, 2020 at 01:03:28PM +0000, Liu, Yi L wrote: > > From: Peter Xu <peterx@redhat.com> > > Sent: Tuesday, March 24, 2020 5:40 AM > > To: Liu, Yi L <yi.l.liu@intel.com> > > Subject: Re: [PATCH v1 08/22] vfio: init HostIOMMUContext per-container > > > > On Sun, Mar 22, 2020 at 05:36:05AM -0700, Liu Yi L wrote: > > > After confirming dual stage DMA translation support with kernel by > > > checking VFIO_TYPE1_NESTING_IOMMU, VFIO inits HostIOMMUContet instance > > > and exposes it to PCI layer. Thus vIOMMU emualtors may make use of > > > such capability by leveraging the methods provided by HostIOMMUContext. > > > > > > Cc: Kevin Tian <kevin.tian@intel.com> > > > Cc: Jacob Pan <jacob.jun.pan@linux.intel.com> > > > Cc: Peter Xu <peterx@redhat.com> > > > Cc: Eric Auger <eric.auger@redhat.com> > > > Cc: Yi Sun <yi.y.sun@linux.intel.com> > > > Cc: David Gibson <david@gibson.dropbear.id.au> > > > Cc: Alex Williamson <alex.williamson@redhat.com> > > > Signed-off-by: Liu Yi L <yi.l.liu@intel.com> > > > --- > > > hw/vfio/common.c | 80 +++++++++++++++++++++++++++++++++++ > > > hw/vfio/pci.c | 13 ++++++ > > > include/hw/iommu/host_iommu_context.h | 3 ++ > > > include/hw/vfio/vfio-common.h | 4 ++ > > > 4 files changed, 100 insertions(+) > > > > > > diff --git a/hw/vfio/common.c b/hw/vfio/common.c > > > index c276732..e4f5f10 100644 > > > --- a/hw/vfio/common.c > > > +++ b/hw/vfio/common.c > > > @@ -1179,10 +1179,55 @@ static int vfio_get_iommu_type(VFIOContainer > > *container, > > > return -EINVAL; > > > } > > > > > > +static int vfio_host_icx_pasid_alloc(HostIOMMUContext *host_icx, > > > > I'm not sure about Alex, but ... icx is confusing to me. Maybe "ctx" > > as you always used? > > At first I used vfio_host_iommu_ctx_pasid_alloc(), found it is long, so I > switched to "icx" which means iommu_context. Maybe the former one > looks better as it gives more precise info. vfio_host_iommu_ctx_pasid_alloc() isn't that bad imho. I'll omit the "ctx" if I want to make it even shorter, but "icx" might be ambiguous. Thanks,
> From: Peter Xu <peterx@redhat.com> > Sent: Tuesday, March 24, 2020 10:46 PM > To: Liu, Yi L <yi.l.liu@intel.com> > Subject: Re: [PATCH v1 08/22] vfio: init HostIOMMUContext per-container > > On Tue, Mar 24, 2020 at 01:03:28PM +0000, Liu, Yi L wrote: > > > From: Peter Xu <peterx@redhat.com> > > > Sent: Tuesday, March 24, 2020 5:40 AM > > > To: Liu, Yi L <yi.l.liu@intel.com> > > > Subject: Re: [PATCH v1 08/22] vfio: init HostIOMMUContext > > > per-container > > > > > > On Sun, Mar 22, 2020 at 05:36:05AM -0700, Liu Yi L wrote: > > > > After confirming dual stage DMA translation support with kernel by > > > > checking VFIO_TYPE1_NESTING_IOMMU, VFIO inits HostIOMMUContet > > > > instance and exposes it to PCI layer. Thus vIOMMU emualtors may > > > > make use of such capability by leveraging the methods provided by > HostIOMMUContext. > > > > > > > > Cc: Kevin Tian <kevin.tian@intel.com> > > > > Cc: Jacob Pan <jacob.jun.pan@linux.intel.com> > > > > Cc: Peter Xu <peterx@redhat.com> > > > > Cc: Eric Auger <eric.auger@redhat.com> > > > > Cc: Yi Sun <yi.y.sun@linux.intel.com> > > > > Cc: David Gibson <david@gibson.dropbear.id.au> > > > > Cc: Alex Williamson <alex.williamson@redhat.com> > > > > Signed-off-by: Liu Yi L <yi.l.liu@intel.com> > > > > --- > > > > hw/vfio/common.c | 80 > +++++++++++++++++++++++++++++++++++ > > > > hw/vfio/pci.c | 13 ++++++ > > > > include/hw/iommu/host_iommu_context.h | 3 ++ > > > > include/hw/vfio/vfio-common.h | 4 ++ > > > > 4 files changed, 100 insertions(+) > > > > > > > > diff --git a/hw/vfio/common.c b/hw/vfio/common.c index > > > > c276732..e4f5f10 100644 > > > > --- a/hw/vfio/common.c > > > > +++ b/hw/vfio/common.c > > > > @@ -1179,10 +1179,55 @@ static int > > > > vfio_get_iommu_type(VFIOContainer > > > *container, > > > > return -EINVAL; > > > > } > > > > > > > > +static int vfio_host_icx_pasid_alloc(HostIOMMUContext *host_icx, > > > > > > I'm not sure about Alex, but ... icx is confusing to me. Maybe "ctx" > > > as you always used? > > > > At first I used vfio_host_iommu_ctx_pasid_alloc(), found it is long, > > so I switched to "icx" which means iommu_context. Maybe the former one > > looks better as it gives more precise info. > > vfio_host_iommu_ctx_pasid_alloc() isn't that bad imho. I'll omit the "ctx" if I want > to make it even shorter, but "icx" might be ambiguous. Got it. let me modify the prefix. Regards, Yi Liu
diff --git a/hw/vfio/common.c b/hw/vfio/common.c index c276732..e4f5f10 100644 --- a/hw/vfio/common.c +++ b/hw/vfio/common.c @@ -1179,10 +1179,55 @@ static int vfio_get_iommu_type(VFIOContainer *container, return -EINVAL; } +static int vfio_host_icx_pasid_alloc(HostIOMMUContext *host_icx, + uint32_t min, uint32_t max, uint32_t *pasid) +{ + VFIOContainer *container = container_of(host_icx, VFIOContainer, host_icx); + struct vfio_iommu_type1_pasid_request req; + unsigned long argsz; + int ret; + + argsz = sizeof(req); + req.argsz = argsz; + req.flags = VFIO_IOMMU_PASID_ALLOC; + req.alloc_pasid.min = min; + req.alloc_pasid.max = max; + + if (ioctl(container->fd, VFIO_IOMMU_PASID_REQUEST, &req)) { + ret = -errno; + error_report("%s: %d, alloc failed", __func__, ret); + return ret; + } + *pasid = req.alloc_pasid.result; + return 0; +} + +static int vfio_host_icx_pasid_free(HostIOMMUContext *host_icx, + uint32_t pasid) +{ + VFIOContainer *container = container_of(host_icx, VFIOContainer, host_icx); + struct vfio_iommu_type1_pasid_request req; + unsigned long argsz; + int ret; + + argsz = sizeof(req); + req.argsz = argsz; + req.flags = VFIO_IOMMU_PASID_FREE; + req.free_pasid = pasid; + + if (ioctl(container->fd, VFIO_IOMMU_PASID_REQUEST, &req)) { + ret = -errno; + error_report("%s: %d, free failed", __func__, ret); + return ret; + } + return 0; +} + static int vfio_init_container(VFIOContainer *container, int group_fd, Error **errp) { int iommu_type, ret; + uint64_t flags = 0; iommu_type = vfio_get_iommu_type(container, errp); if (iommu_type < 0) { @@ -1210,6 +1255,18 @@ static int vfio_init_container(VFIOContainer *container, int group_fd, return -errno; } + if (iommu_type == VFIO_TYPE1_NESTING_IOMMU) { + /* + * TODO: config flags per host IOMMU nesting capability + * e.g. check if VFIO_TYPE1_NESTING_IOMMU supports PASID + * alloc/free + */ + host_iommu_ctx_init(&container->host_icx, + sizeof(container->host_icx), + TYPE_VFIO_HOST_IOMMU_CONTEXT, + flags); + } + container->iommu_type = iommu_type; return 0; } @@ -1456,6 +1513,7 @@ static void vfio_disconnect_container(VFIOGroup *group) } trace_vfio_disconnect_container(container->fd); + host_iommu_ctx_destroy(&container->host_icx); close(container->fd); g_free(container); @@ -1791,3 +1849,25 @@ int vfio_eeh_as_op(AddressSpace *as, uint32_t op) } return vfio_eeh_container_op(container, op); } + +static void vfio_host_iommu_context_class_init(ObjectClass *klass, + void *data) +{ + HostIOMMUContextClass *hicxc = HOST_IOMMU_CONTEXT_CLASS(klass); + + hicxc->pasid_alloc = vfio_host_icx_pasid_alloc; + hicxc->pasid_free = vfio_host_icx_pasid_free; +} + +static const TypeInfo vfio_host_iommu_context_info = { + .parent = TYPE_HOST_IOMMU_CONTEXT, + .name = TYPE_VFIO_HOST_IOMMU_CONTEXT, + .class_init = vfio_host_iommu_context_class_init, +}; + +static void vfio_register_types(void) +{ + type_register_static(&vfio_host_iommu_context_info); +} + +type_init(vfio_register_types) diff --git a/hw/vfio/pci.c b/hw/vfio/pci.c index 5e75a95..f099df3 100644 --- a/hw/vfio/pci.c +++ b/hw/vfio/pci.c @@ -2717,6 +2717,7 @@ static void vfio_realize(PCIDevice *pdev, Error **errp) VFIOPCIDevice *vdev = PCI_VFIO(pdev); VFIODevice *vbasedev_iter; VFIOGroup *group; + VFIOContainer *container; char *tmp, *subsys, group_path[PATH_MAX], *group_name; Error *err = NULL; ssize_t len; @@ -3028,6 +3029,11 @@ static void vfio_realize(PCIDevice *pdev, Error **errp) vfio_register_req_notifier(vdev); vfio_setup_resetfn_quirk(vdev); + container = vdev->vbasedev.group->container; + if (container->host_icx.initialized) { + pci_device_set_iommu_context(pdev, &container->host_icx); + } + return; out_deregister: @@ -3072,9 +3078,16 @@ static void vfio_instance_finalize(Object *obj) static void vfio_exitfn(PCIDevice *pdev) { VFIOPCIDevice *vdev = PCI_VFIO(pdev); + VFIOContainer *container; vfio_unregister_req_notifier(vdev); vfio_unregister_err_notifier(vdev); + + container = vdev->vbasedev.group->container; + if (container->host_icx.initialized) { + pci_device_unset_iommu_context(pdev); + } + pci_device_set_intx_routing_notifier(&vdev->pdev, NULL); if (vdev->irqchip_change_notifier.notify) { kvm_irqchip_remove_change_notifier(&vdev->irqchip_change_notifier); diff --git a/include/hw/iommu/host_iommu_context.h b/include/hw/iommu/host_iommu_context.h index cfbf5ac..5f11a4c 100644 --- a/include/hw/iommu/host_iommu_context.h +++ b/include/hw/iommu/host_iommu_context.h @@ -33,6 +33,9 @@ #define TYPE_HOST_IOMMU_CONTEXT "qemu:host-iommu-context" #define HOST_IOMMU_CONTEXT(obj) \ OBJECT_CHECK(HostIOMMUContext, (obj), TYPE_HOST_IOMMU_CONTEXT) +#define HOST_IOMMU_CONTEXT_CLASS(klass) \ + OBJECT_CLASS_CHECK(HostIOMMUContextClass, (klass), \ + TYPE_HOST_IOMMU_CONTEXT) #define HOST_IOMMU_CONTEXT_GET_CLASS(obj) \ OBJECT_GET_CLASS(HostIOMMUContextClass, (obj), \ TYPE_HOST_IOMMU_CONTEXT) diff --git a/include/hw/vfio/vfio-common.h b/include/hw/vfio/vfio-common.h index fd56420..532b78d 100644 --- a/include/hw/vfio/vfio-common.h +++ b/include/hw/vfio/vfio-common.h @@ -26,12 +26,15 @@ #include "qemu/notify.h" #include "ui/console.h" #include "hw/display/ramfb.h" +#include "hw/iommu/host_iommu_context.h" #ifdef CONFIG_LINUX #include <linux/vfio.h> #endif #define VFIO_MSG_PREFIX "vfio %s: " +#define TYPE_VFIO_HOST_IOMMU_CONTEXT "qemu:vfio-host-iommu-context" + enum { VFIO_DEVICE_TYPE_PCI = 0, VFIO_DEVICE_TYPE_PLATFORM = 1, @@ -71,6 +74,7 @@ typedef struct VFIOContainer { MemoryListener listener; MemoryListener prereg_listener; unsigned iommu_type; + HostIOMMUContext host_icx; Error *error; bool initialized; unsigned long pgsizes;
After confirming dual stage DMA translation support with kernel by checking VFIO_TYPE1_NESTING_IOMMU, VFIO inits HostIOMMUContet instance and exposes it to PCI layer. Thus vIOMMU emualtors may make use of such capability by leveraging the methods provided by HostIOMMUContext. Cc: Kevin Tian <kevin.tian@intel.com> Cc: Jacob Pan <jacob.jun.pan@linux.intel.com> Cc: Peter Xu <peterx@redhat.com> Cc: Eric Auger <eric.auger@redhat.com> Cc: Yi Sun <yi.y.sun@linux.intel.com> Cc: David Gibson <david@gibson.dropbear.id.au> Cc: Alex Williamson <alex.williamson@redhat.com> Signed-off-by: Liu Yi L <yi.l.liu@intel.com> --- hw/vfio/common.c | 80 +++++++++++++++++++++++++++++++++++ hw/vfio/pci.c | 13 ++++++ include/hw/iommu/host_iommu_context.h | 3 ++ include/hw/vfio/vfio-common.h | 4 ++ 4 files changed, 100 insertions(+)