powerpc/iommu: Add "borrowing" iommu_table_group_ops

PPC64 IOMMU API defines iommu_table_group_ops which handles DMA windows
for PEs: control the ownership, create/set/unset a table the hardware
for dynamic DMA windows (DDW). VFIO uses the API to implement support on
POWER.

So far only PowerNV IODA2 (POWER8 and newer machines) implemented this
and other cases (POWER7 or nested KVM) did not and instead reused
existing iommu_table structs. This means 1) no DDW 2) ownership transfer
is done directly in the VFIO SPAPR TCE driver.

Soon POWER is going to get its own iommu_ops and ownership control is
going to move there. This implements spapr_tce_table_group_ops which
borrows iommu_table tables. The upside is that VFIO needs to know less
about POWER.

The new ops returns the existing table from create_table() and only
checks if the same window is already set. This is only going to work if
the default DMA window starts table_group.tce32_start and as big as
pe->table_group.tce32_size (not the case for IODA2+ PowerNV).

This changes iommu_table_group_ops::take_ownership() to return an error
if borrowing a table failed.

This should not cause any visible change in behavior for PowerNV.
pSeries was not that well tested/supported anyway.

Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Timothy Pearson <tpearson@raptorengineering.com>
Acked-by: Alex Williamson <alex.williamson@redhat.com>
[mpe: Fix CONFIG_IOMMU_API=n build (skiroot_defconfig), & formatting]
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://msgid.link/525438831.16998517.1678123820075.JavaMail.zimbra@raptorengineeringinc.com

1 files changed, 16 insertions, 78 deletions

diff --git a/drivers/vfio/vfio_iommu_spapr_tce.c b/drivers/vfio/vfio_iommu_spapr_tce.c
index 60a50ce8701e..c3f8ae102ece 100644
--- a/drivers/vfio/vfio_iommu_spapr_tce.c
+++ b/drivers/vfio/vfio_iommu_spapr_tce.c
@@ -1190,52 +1190,6 @@ static long tce_iommu_ioctl(void *iommu_data,
 static void tce_iommu_release_ownership(struct tce_container *container,
 		struct iommu_table_group *table_group)
 {
-	int i;
-
-	for (i = 0; i < IOMMU_TABLE_GROUP_MAX_TABLES; ++i) {
-		struct iommu_table *tbl = container->tables[i];
-
-		if (!tbl)
-			continue;
-
-		tce_iommu_clear(container, tbl, tbl->it_offset, tbl->it_size);
-		if (tbl->it_map)
-			iommu_release_ownership(tbl);
-
-		container->tables[i] = NULL;
-	}
-}
-
-static int tce_iommu_take_ownership(struct tce_container *container,
-		struct iommu_table_group *table_group)
-{
-	int i, j, rc = 0;
-
-	for (i = 0; i < IOMMU_TABLE_GROUP_MAX_TABLES; ++i) {
-		struct iommu_table *tbl = table_group->tables[i];
-
-		if (!tbl || !tbl->it_map)
-			continue;
-
-		rc = iommu_take_ownership(tbl);
-		if (rc) {
-			for (j = 0; j < i; ++j)
-				iommu_release_ownership(
-						table_group->tables[j]);
-
-			return rc;
-		}
-	}
-
-	for (i = 0; i < IOMMU_TABLE_GROUP_MAX_TABLES; ++i)
-		container->tables[i] = table_group->tables[i];
-
-	return 0;
-}
-
-static void tce_iommu_release_ownership_ddw(struct tce_container *container,
-		struct iommu_table_group *table_group)
-{
 	long i;
 
 	if (!table_group->ops->unset_window) {
@@ -1250,18 +1204,14 @@ static void tce_iommu_release_ownership_ddw(struct tce_container *container,
 	table_group->ops->release_ownership(table_group);
 }
 
-static long tce_iommu_take_ownership_ddw(struct tce_container *container,
+static long tce_iommu_take_ownership(struct tce_container *container,
 		struct iommu_table_group *table_group)
 {
 	long i, ret = 0;
 
-	if (!table_group->ops->create_table || !table_group->ops->set_window ||
-			!table_group->ops->release_ownership) {
-		WARN_ON_ONCE(1);
-		return -EFAULT;
-	}
-
-	table_group->ops->take_ownership(table_group);
+	ret = table_group->ops->take_ownership(table_group);
+	if (ret)
+		return ret;
 
 	/* Set all windows to the new group */
 	for (i = 0; i < IOMMU_TABLE_GROUP_MAX_TABLES; ++i) {
@@ -1307,9 +1257,14 @@ static int tce_iommu_attach_group(void *iommu_data,
 		goto unlock_exit;
 	}
 
-	if (tce_groups_attached(container) && (!table_group->ops ||
-			!table_group->ops->take_ownership ||
-			!table_group->ops->release_ownership)) {
+	/* v2 requires full support of dynamic DMA windows */
+	if (container->v2 && table_group->max_dynamic_windows_supported == 0) {
+		ret = -EINVAL;
+		goto unlock_exit;
+	}
+
+	/* v1 reuses TCE tables and does not share them among PEs */
+	if (!container->v2 && tce_groups_attached(container)) {
 		ret = -EBUSY;
 		goto unlock_exit;
 	}
@@ -1344,29 +1299,15 @@ static int tce_iommu_attach_group(void *iommu_data,
 		goto unlock_exit;
 	}
 
-	if (!table_group->ops || !table_group->ops->take_ownership ||
-			!table_group->ops->release_ownership) {
-		if (container->v2) {
-			ret = -EPERM;
-			goto free_exit;
-		}
-		ret = tce_iommu_take_ownership(container, table_group);
-	} else {
-		if (!container->v2) {
-			ret = -EPERM;
-			goto free_exit;
-		}
-		ret = tce_iommu_take_ownership_ddw(container, table_group);
-		if (!tce_groups_attached(container) && !container->tables[0])
-			container->def_window_pending = true;
-	}
+	ret = tce_iommu_take_ownership(container, table_group);
+	if (!tce_groups_attached(container) && !container->tables[0])
+		container->def_window_pending = true;
 
 	if (!ret) {
 		tcegrp->grp = iommu_group;
 		list_add(&tcegrp->next, &container->group_list);
 	}
 
-free_exit:
 	if (ret && tcegrp)
 		kfree(tcegrp);
 
@@ -1405,10 +1346,7 @@ static void tce_iommu_detach_group(void *iommu_data,
 	table_group = iommu_group_get_iommudata(iommu_group);
 	BUG_ON(!table_group);
 
-	if (!table_group->ops || !table_group->ops->release_ownership)
-		tce_iommu_release_ownership(container, table_group);
-	else
-		tce_iommu_release_ownership_ddw(container, table_group);
+	tce_iommu_release_ownership(container, table_group);
 
 unlock_exit:
 	mutex_unlock(&container->lock);