vfio: Enable in-kernel acceleration via VFIO KVM device

TCE hypercalls (H_PUT_TCE, H_PUT_TCE_INDIRECT, H_STUFF_TCE) use a logical bus
number (LIOBN) to identify which TCE table the request is addressed to.
However VFIO kernel driver operates with IOMMU group IDs and has no idea
about which LIOBN corresponds to which group. If the host kernel supports
in-kernel acceleration for TCE calls, we have to provide the LIOBN to IOMMU
mapping information.

This makes use of a VFIO KVM device's
KVM_DEV_VFIO_GROUP_SET_SPAPR_TCE_LIOBN attribute to set the link
between LIOBN and IOMMU group.

The vfio_container_spapr_set_liobn() helper is implemented completely
in vfio.c because kvm_vfio_spapr_tce_liobn needs a group fd and
we do not want to share resources likes that outside vfio.c.

Signed-off-by: Alexey Kardashevskiy <[email protected]>

Author: aik

hw/misc/vfio.c

@@ -4560,3 +4560,43 @@ int vfio_container_ioctl(AddressSpace *as,
 
     return vfio_container_do_ioctl(as, req, param);
 }
+
+int vfio_container_spapr_set_liobn(AddressSpace *as,
+                                   uint64_t liobn,
+                                   uint64_t start_addr)
+{
+#ifdef CONFIG_KVM
+//    VFIOGroup *group;
+    int ret;
+    struct kvm_vfio_spapr_tce_liobn param = {
+        .argsz = sizeof(param),
+        .liobn = liobn,
+        .start_addr = start_addr
+    };
+    struct kvm_device_attr attr = {
+        .group = KVM_DEV_VFIO_GROUP,
+        .attr = KVM_DEV_VFIO_GROUP_SET_SPAPR_TCE_LIOBN,
+        .addr = (uint64_t)(unsigned long)&param,
+    };
+
+    if (vfio_kvm_device_fd < 0) {
+        return 0;
+    }
+
+//    group = vfio_get_group(groupid, as);
+//    if (!group) {
+//        return -1;
+//    }
+
+//    param.fd = group->fd;
+    ret = ioctl(vfio_kvm_device_fd, KVM_SET_DEVICE_ATTR, &attr);
+    if (ret) {
+        error_report("vfio: failed to setup liobn for a group: %s",
+                     strerror(errno));
+    }
+
+    return ret;
+#else
+    return 0;
+#endif
+}

hw/ppc/spapr_iommu.c

@@ -127,6 +127,7 @@ static int spapr_tce_table_realize(DeviceState *dev)
     if (!tcet->table) {
         size_t table_size = tcet->nb_table * sizeof(uint64_t);
         tcet->table = g_malloc0(table_size);
+        tcet->vfio_accel = false;
     }
 
     trace_spapr_iommu_new_table(tcet->liobn, tcet, tcet->table, tcet->fd);

hw/ppc/spapr_pci_vfio.c

@@ -22,6 +22,7 @@
 #include "hw/pci-host/spapr.h"
 #include "linux/vfio.h"
 #include "hw/misc/vfio.h"
+#include "qemu/error-report.h"
 
 #define DDW_PGSIZE_4K           0x01
 #define DDW_PGSIZE_64K          0x02
@@ -92,6 +93,16 @@ static void spapr_phb_vfio_reset(DeviceState *qdev)
     sphb->windows_num = 1;
 
     sphb->ddw_enabled = (info.windows_supported > 1);
+
+    if (!tcet->vfio_accel) {
+        return;
+    }
+    ret = vfio_container_spapr_set_liobn(&svphb->phb.iommu_as,
+                                         tcet->liobn,
+                                         tcet->bus_offset);
+    if (ret) {
+        error_report("spapr-vfio: failed to create link to IOMMU");
+    }
 }
 
 static int spapr_pci_vfio_ddw_query(sPAPRPHBState *sphb,
@@ -141,6 +152,16 @@ static int spapr_pci_vfio_ddw_create(sPAPRPHBState *sphb, uint32_t page_shift,
 
     ++sphb->windows_num;
 
+    if (!(*ptcet)->vfio_accel) {
+        return 0;
+    }
+    ret = vfio_container_spapr_set_liobn(&sphb->iommu_as,
+                                         liobn, (*ptcet)->bus_offset);
+    if (ret) {
+        error_report("spapr-vfio: failed to create link to IOMMU");
+        ret = 0;
+    }
+
     return ret;
 }
 

include/hw/misc/vfio.h

@@ -6,4 +6,8 @@
 extern int vfio_container_ioctl(AddressSpace *as,
                                 int req, void *param);
 
+extern int vfio_container_spapr_set_liobn(AddressSpace *as,
+                                          uint64_t liobn,
+                                          uint64_t start_addr);
+
 #endif

linux-headers/linux/kvm.h

@@ -973,13 +973,6 @@ struct kvm_vfio_spapr_tce_liobn {
 	__u64	start_addr;
 };
 
-struct kvm_vfio_spapr_tce_liobn {
-	__u32	argsz;
-	__s32	fd;
-	__u32	liobn;
-	__u64	start_addr;
-};
-
 /*
  * ioctls for VM fds
  */