diff mbox

[1/3] powerpc/vfio: Enable on POWERNV platform

Message ID 1369107191-28547-2-git-send-email-aik@ozlabs.ru (mailing list archive)
State Accepted, archived
Commit 4e13c1ac6baa1d6c2b650d66ca89e1e12727ec19
Delegated to: Benjamin Herrenschmidt
Headers show

Commit Message

Alexey Kardashevskiy May 21, 2013, 3:33 a.m. UTC
This initializes IOMMU groups based on the IOMMU configuration
discovered during the PCI scan on POWERNV (POWER non virtualized)
platform.  The IOMMU groups are to be used later by the VFIO driver,
which is used for PCI pass through.

It also implements an API for mapping/unmapping pages for
guest PCI drivers and providing DMA window properties.
This API is going to be used later by QEMU-VFIO to handle
h_put_tce hypercalls from the KVM guest.

The iommu_put_tce_user_mode() does only a single page mapping
as an API for adding many mappings at once is going to be
added later.

Although this driver has been tested only on the POWERNV
platform, it should work on any platform which supports
TCE tables.  As h_put_tce hypercall is received by the host
kernel and processed by the QEMU (what involves calling
the host kernel again), performance is not the best -
circa 220MB/s on 10Gb ethernet network.

To enable VFIO on POWER, enable SPAPR_TCE_IOMMU config
option and configure VFIO as required.

Cc: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Paul Mackerras <paulus@samba.org>

---

Changes:
* the PCI devices are added to groups from subsys_initcall_sync (used to
be done via bus_notifier callback registered for PCI bus)
* fixed parameters checking (the very last page in the address space was
not handled correctly)
---
 arch/powerpc/include/asm/iommu.h            |   26 +++
 arch/powerpc/kernel/iommu.c                 |  323 +++++++++++++++++++++++++++
 arch/powerpc/platforms/powernv/pci-ioda.c   |    1 +
 arch/powerpc/platforms/powernv/pci-p5ioc2.c |    5 +-
 arch/powerpc/platforms/powernv/pci.c        |    2 +
 drivers/iommu/Kconfig                       |    8 +
 6 files changed, 364 insertions(+), 1 deletion(-)

Comments

Scott Wood Dec. 12, 2013, 11:35 p.m. UTC | #1
On Tue, May 21, 2013 at 01:33:09PM +1000, Alexey Kardashevskiy wrote:
> +static int iommu_add_device(struct device *dev)
> +{
> +	struct iommu_table *tbl;
> +	int ret = 0;
> +
> +	if (WARN_ON(dev->iommu_group)) {
> +		pr_warn("iommu_tce: device %s is already in iommu group %d, skipping\n",
> +				dev_name(dev),
> +				iommu_group_id(dev->iommu_group));
> +		return -EBUSY;
> +	}
[snip]
> +static int __init tce_iommu_init(void)
> +{
> +	struct pci_dev *pdev = NULL;
> +
> +	BUILD_BUG_ON(PAGE_SIZE < IOMMU_PAGE_SIZE);
> +
> +	for_each_pci_dev(pdev)
> +		iommu_add_device(&pdev->dev);
> +
> +	bus_register_notifier(&pci_bus_type, &tce_iommu_bus_nb);
> +	return 0;
> +}
> +
> +subsys_initcall_sync(tce_iommu_init);

This is missing a check to see whether the appropriate hardware is
present.  This file should also be renamed to something less generic, and
depend on a kconfig symbol more specific than CONFIG_PPC64.

When this is combined with CONFIG_FSL_PAMU on hardware with a PAMU, I get
a bunch of those "WARN_ON(dev->iommu_group)" dumps because PAMU already
got to them.  Presumably without PAMU it silently (or with just pr_debug)
bails out at some other point.

-Scott
Alexey Kardashevskiy Dec. 13, 2013, 3:02 a.m. UTC | #2
On 12/13/2013 10:35 AM, Scott Wood wrote:
> On Tue, May 21, 2013 at 01:33:09PM +1000, Alexey Kardashevskiy wrote:
>> +static int iommu_add_device(struct device *dev)
>> +{
>> +	struct iommu_table *tbl;
>> +	int ret = 0;
>> +
>> +	if (WARN_ON(dev->iommu_group)) {
>> +		pr_warn("iommu_tce: device %s is already in iommu group %d, skipping\n",
>> +				dev_name(dev),
>> +				iommu_group_id(dev->iommu_group));
>> +		return -EBUSY;
>> +	}
> [snip]
>> +static int __init tce_iommu_init(void)
>> +{
>> +	struct pci_dev *pdev = NULL;
>> +
>> +	BUILD_BUG_ON(PAGE_SIZE < IOMMU_PAGE_SIZE);
>> +
>> +	for_each_pci_dev(pdev)
>> +		iommu_add_device(&pdev->dev);
>> +
>> +	bus_register_notifier(&pci_bus_type, &tce_iommu_bus_nb);
>> +	return 0;
>> +}
>> +
>> +subsys_initcall_sync(tce_iommu_init);
> 
> This is missing a check to see whether the appropriate hardware is
> present.  This file should also be renamed to something less generic, and
> depend on a kconfig symbol more specific than CONFIG_PPC64.
> 
> When this is combined with CONFIG_FSL_PAMU on hardware with a PAMU, I get
> a bunch of those "WARN_ON(dev->iommu_group)" dumps because PAMU already
> got to them.  Presumably without PAMU it silently (or with just pr_debug)
> bails out at some other point.


I posted (yet again) yesterday "[PATCH v11] PPC: POWERNV: move
iommu_add_device earlier" which should fix this. And Bharat asked many
times for this to get accepted :)
Scott Wood Dec. 13, 2013, 9:02 p.m. UTC | #3
On Fri, 2013-12-13 at 14:02 +1100, Alexey Kardashevskiy wrote:
> On 12/13/2013 10:35 AM, Scott Wood wrote:
> > On Tue, May 21, 2013 at 01:33:09PM +1000, Alexey Kardashevskiy wrote:
> >> +static int iommu_add_device(struct device *dev)
> >> +{
> >> +	struct iommu_table *tbl;
> >> +	int ret = 0;
> >> +
> >> +	if (WARN_ON(dev->iommu_group)) {
> >> +		pr_warn("iommu_tce: device %s is already in iommu group %d, skipping\n",
> >> +				dev_name(dev),
> >> +				iommu_group_id(dev->iommu_group));
> >> +		return -EBUSY;
> >> +	}
> > [snip]
> >> +static int __init tce_iommu_init(void)
> >> +{
> >> +	struct pci_dev *pdev = NULL;
> >> +
> >> +	BUILD_BUG_ON(PAGE_SIZE < IOMMU_PAGE_SIZE);
> >> +
> >> +	for_each_pci_dev(pdev)
> >> +		iommu_add_device(&pdev->dev);
> >> +
> >> +	bus_register_notifier(&pci_bus_type, &tce_iommu_bus_nb);
> >> +	return 0;
> >> +}
> >> +
> >> +subsys_initcall_sync(tce_iommu_init);
> > 
> > This is missing a check to see whether the appropriate hardware is
> > present.  This file should also be renamed to something less generic, and
> > depend on a kconfig symbol more specific than CONFIG_PPC64.
> > 
> > When this is combined with CONFIG_FSL_PAMU on hardware with a PAMU, I get
> > a bunch of those "WARN_ON(dev->iommu_group)" dumps because PAMU already
> > got to them.  Presumably without PAMU it silently (or with just pr_debug)
> > bails out at some other point.
> 
> 
> I posted (yet again) yesterday "[PATCH v11] PPC: POWERNV: move
> iommu_add_device earlier" which should fix this. And Bharat asked many
> times for this to get accepted :)

I still get the WARN_ONs even with that patch.  You're still registering
the bus notifier unconditionally.

-Scott
Bharat Bhushan Dec. 14, 2013, 4:56 a.m. UTC | #4
> -----Original Message-----
> From: Wood Scott-B07421
> Sent: Saturday, December 14, 2013 2:33 AM
> To: Alexey Kardashevskiy
> Cc: linuxppc-dev@lists.ozlabs.org; kvm@vger.kernel.org; linux-
> kernel@vger.kernel.org; Alex Williamson; Paul Mackerras; David Gibson; Sethi
> Varun-B16395; Bhushan Bharat-R65777
> Subject: Re: [1/3] powerpc/vfio: Enable on POWERNV platform
> 
> On Fri, 2013-12-13 at 14:02 +1100, Alexey Kardashevskiy wrote:
> > On 12/13/2013 10:35 AM, Scott Wood wrote:
> > > On Tue, May 21, 2013 at 01:33:09PM +1000, Alexey Kardashevskiy wrote:
> > >> +static int iommu_add_device(struct device *dev) {
> > >> +	struct iommu_table *tbl;
> > >> +	int ret = 0;
> > >> +
> > >> +	if (WARN_ON(dev->iommu_group)) {
> > >> +		pr_warn("iommu_tce: device %s is already in iommu group %d,
> skipping\n",
> > >> +				dev_name(dev),
> > >> +				iommu_group_id(dev->iommu_group));
> > >> +		return -EBUSY;
> > >> +	}
> > > [snip]
> > >> +static int __init tce_iommu_init(void) {
> > >> +	struct pci_dev *pdev = NULL;
> > >> +
> > >> +	BUILD_BUG_ON(PAGE_SIZE < IOMMU_PAGE_SIZE);
> > >> +
> > >> +	for_each_pci_dev(pdev)
> > >> +		iommu_add_device(&pdev->dev);
> > >> +
> > >> +	bus_register_notifier(&pci_bus_type, &tce_iommu_bus_nb);
> > >> +	return 0;
> > >> +}
> > >> +
> > >> +subsys_initcall_sync(tce_iommu_init);
> > >
> > > This is missing a check to see whether the appropriate hardware is
> > > present.  This file should also be renamed to something less
> > > generic, and depend on a kconfig symbol more specific than CONFIG_PPC64.
> > >
> > > When this is combined with CONFIG_FSL_PAMU on hardware with a PAMU,
> > > I get a bunch of those "WARN_ON(dev->iommu_group)" dumps because
> > > PAMU already got to them.  Presumably without PAMU it silently (or
> > > with just pr_debug) bails out at some other point.
> >
> >
> > I posted (yet again) yesterday "[PATCH v11] PPC: POWERNV: move
> > iommu_add_device earlier" which should fix this. And Bharat asked many
> > times for this to get accepted :)
> 
> I still get the WARN_ONs even with that patch.  You're still registering the bus
> notifier unconditionally.

I have not tried v11 but tested V9 version of that patch. And yes, in that version the bus notifier was not registered unconditionally in kernel/iommu.c .

Thanks
-Bharat

> 
> -Scott
>
diff mbox

Patch

diff --git a/arch/powerpc/include/asm/iommu.h b/arch/powerpc/include/asm/iommu.h
index cbfe678..98d1422 100644
--- a/arch/powerpc/include/asm/iommu.h
+++ b/arch/powerpc/include/asm/iommu.h
@@ -76,6 +76,9 @@  struct iommu_table {
 	struct iommu_pool large_pool;
 	struct iommu_pool pools[IOMMU_NR_POOLS];
 	unsigned long *it_map;       /* A simple allocation bitmap for now */
+#ifdef CONFIG_IOMMU_API
+	struct iommu_group *it_group;
+#endif
 };
 
 struct scatterlist;
@@ -98,6 +101,8 @@  extern void iommu_free_table(struct iommu_table *tbl, const char *node_name);
  */
 extern struct iommu_table *iommu_init_table(struct iommu_table * tbl,
 					    int nid);
+extern void iommu_register_group(struct iommu_table *tbl,
+				 int pci_domain_number, unsigned long pe_num);
 
 extern int iommu_map_sg(struct device *dev, struct iommu_table *tbl,
 			struct scatterlist *sglist, int nelems,
@@ -147,5 +152,26 @@  static inline void iommu_restore(void)
 }
 #endif
 
+/* The API to support IOMMU operations for VFIO */
+extern int iommu_tce_clear_param_check(struct iommu_table *tbl,
+		unsigned long ioba, unsigned long tce_value,
+		unsigned long npages);
+extern int iommu_tce_put_param_check(struct iommu_table *tbl,
+		unsigned long ioba, unsigned long tce);
+extern int iommu_tce_build(struct iommu_table *tbl, unsigned long entry,
+		unsigned long hwaddr, enum dma_data_direction direction);
+extern unsigned long iommu_clear_tce(struct iommu_table *tbl,
+		unsigned long entry);
+extern int iommu_clear_tces_and_put_pages(struct iommu_table *tbl,
+		unsigned long entry, unsigned long pages);
+extern int iommu_put_tce_user_mode(struct iommu_table *tbl,
+		unsigned long entry, unsigned long tce);
+
+extern void iommu_flush_tce(struct iommu_table *tbl);
+extern int iommu_take_ownership(struct iommu_table *tbl);
+extern void iommu_release_ownership(struct iommu_table *tbl);
+
+extern enum dma_data_direction iommu_tce_direction(unsigned long tce);
+
 #endif /* __KERNEL__ */
 #endif /* _ASM_IOMMU_H */
diff --git a/arch/powerpc/kernel/iommu.c b/arch/powerpc/kernel/iommu.c
index c0d0dbd..b20ff17 100644
--- a/arch/powerpc/kernel/iommu.c
+++ b/arch/powerpc/kernel/iommu.c
@@ -36,6 +36,8 @@ 
 #include <linux/hash.h>
 #include <linux/fault-inject.h>
 #include <linux/pci.h>
+#include <linux/iommu.h>
+#include <linux/sched.h>
 #include <asm/io.h>
 #include <asm/prom.h>
 #include <asm/iommu.h>
@@ -44,6 +46,7 @@ 
 #include <asm/kdump.h>
 #include <asm/fadump.h>
 #include <asm/vio.h>
+#include <asm/tce.h>
 
 #define DBG(...)
 
@@ -724,6 +727,13 @@  void iommu_free_table(struct iommu_table *tbl, const char *node_name)
 	if (tbl->it_offset == 0)
 		clear_bit(0, tbl->it_map);
 
+#ifdef CONFIG_IOMMU_API
+	if (tbl->it_group) {
+		iommu_group_put(tbl->it_group);
+		BUG_ON(tbl->it_group);
+	}
+#endif
+
 	/* verify that table contains no entries */
 	if (!bitmap_empty(tbl->it_map, tbl->it_size))
 		pr_warn("%s: Unexpected TCEs for %s\n", __func__, node_name);
@@ -860,3 +870,316 @@  void iommu_free_coherent(struct iommu_table *tbl, size_t size,
 		free_pages((unsigned long)vaddr, get_order(size));
 	}
 }
+
+#ifdef CONFIG_IOMMU_API
+/*
+ * SPAPR TCE API
+ */
+static void group_release(void *iommu_data)
+{
+	struct iommu_table *tbl = iommu_data;
+	tbl->it_group = NULL;
+}
+
+void iommu_register_group(struct iommu_table *tbl,
+		int pci_domain_number, unsigned long pe_num)
+{
+	struct iommu_group *grp;
+	char *name;
+
+	grp = iommu_group_alloc();
+	if (IS_ERR(grp)) {
+		pr_warn("powerpc iommu api: cannot create new group, err=%ld\n",
+				PTR_ERR(grp));
+		return;
+	}
+	tbl->it_group = grp;
+	iommu_group_set_iommudata(grp, tbl, group_release);
+	name = kasprintf(GFP_KERNEL, "domain%d-pe%lx",
+			pci_domain_number, pe_num);
+	if (!name)
+		return;
+	iommu_group_set_name(grp, name);
+	kfree(name);
+}
+
+enum dma_data_direction iommu_tce_direction(unsigned long tce)
+{
+	if ((tce & TCE_PCI_READ) && (tce & TCE_PCI_WRITE))
+		return DMA_BIDIRECTIONAL;
+	else if (tce & TCE_PCI_READ)
+		return DMA_TO_DEVICE;
+	else if (tce & TCE_PCI_WRITE)
+		return DMA_FROM_DEVICE;
+	else
+		return DMA_NONE;
+}
+EXPORT_SYMBOL_GPL(iommu_tce_direction);
+
+void iommu_flush_tce(struct iommu_table *tbl)
+{
+	/* Flush/invalidate TLB caches if necessary */
+	if (ppc_md.tce_flush)
+		ppc_md.tce_flush(tbl);
+
+	/* Make sure updates are seen by hardware */
+	mb();
+}
+EXPORT_SYMBOL_GPL(iommu_flush_tce);
+
+int iommu_tce_clear_param_check(struct iommu_table *tbl,
+		unsigned long ioba, unsigned long tce_value,
+		unsigned long npages)
+{
+	/* ppc_md.tce_free() does not support any value but 0 */
+	if (tce_value)
+		return -EINVAL;
+
+	if (ioba & ~IOMMU_PAGE_MASK)
+		return -EINVAL;
+
+	ioba >>= IOMMU_PAGE_SHIFT;
+	if (ioba < tbl->it_offset)
+		return -EINVAL;
+
+	if ((ioba + npages) > (tbl->it_offset + tbl->it_size))
+		return -EINVAL;
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(iommu_tce_clear_param_check);
+
+int iommu_tce_put_param_check(struct iommu_table *tbl,
+		unsigned long ioba, unsigned long tce)
+{
+	if (!(tce & (TCE_PCI_WRITE | TCE_PCI_READ)))
+		return -EINVAL;
+
+	if (tce & ~(IOMMU_PAGE_MASK | TCE_PCI_WRITE | TCE_PCI_READ))
+		return -EINVAL;
+
+	if (ioba & ~IOMMU_PAGE_MASK)
+		return -EINVAL;
+
+	ioba >>= IOMMU_PAGE_SHIFT;
+	if (ioba < tbl->it_offset)
+		return -EINVAL;
+
+	if ((ioba + 1) > (tbl->it_offset + tbl->it_size))
+		return -EINVAL;
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(iommu_tce_put_param_check);
+
+unsigned long iommu_clear_tce(struct iommu_table *tbl, unsigned long entry)
+{
+	unsigned long oldtce;
+	struct iommu_pool *pool = get_pool(tbl, entry);
+
+	spin_lock(&(pool->lock));
+
+	oldtce = ppc_md.tce_get(tbl, entry);
+	if (oldtce & (TCE_PCI_WRITE | TCE_PCI_READ))
+		ppc_md.tce_free(tbl, entry, 1);
+	else
+		oldtce = 0;
+
+	spin_unlock(&(pool->lock));
+
+	return oldtce;
+}
+EXPORT_SYMBOL_GPL(iommu_clear_tce);
+
+int iommu_clear_tces_and_put_pages(struct iommu_table *tbl,
+		unsigned long entry, unsigned long pages)
+{
+	unsigned long oldtce;
+	struct page *page;
+
+	for ( ; pages; --pages, ++entry) {
+		oldtce = iommu_clear_tce(tbl, entry);
+		if (!oldtce)
+			continue;
+
+		page = pfn_to_page(oldtce >> PAGE_SHIFT);
+		WARN_ON(!page);
+		if (page) {
+			if (oldtce & TCE_PCI_WRITE)
+				SetPageDirty(page);
+			put_page(page);
+		}
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(iommu_clear_tces_and_put_pages);
+
+/*
+ * hwaddr is a kernel virtual address here (0xc... bazillion),
+ * tce_build converts it to a physical address.
+ */
+int iommu_tce_build(struct iommu_table *tbl, unsigned long entry,
+		unsigned long hwaddr, enum dma_data_direction direction)
+{
+	int ret = -EBUSY;
+	unsigned long oldtce;
+	struct iommu_pool *pool = get_pool(tbl, entry);
+
+	spin_lock(&(pool->lock));
+
+	oldtce = ppc_md.tce_get(tbl, entry);
+	/* Add new entry if it is not busy */
+	if (!(oldtce & (TCE_PCI_WRITE | TCE_PCI_READ)))
+		ret = ppc_md.tce_build(tbl, entry, 1, hwaddr, direction, NULL);
+
+	spin_unlock(&(pool->lock));
+
+	/* if (unlikely(ret))
+		pr_err("iommu_tce: %s failed on hwaddr=%lx ioba=%lx kva=%lx ret=%d\n",
+				__func__, hwaddr, entry << IOMMU_PAGE_SHIFT,
+				hwaddr, ret); */
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(iommu_tce_build);
+
+int iommu_put_tce_user_mode(struct iommu_table *tbl, unsigned long entry,
+		unsigned long tce)
+{
+	int ret;
+	struct page *page = NULL;
+	unsigned long hwaddr, offset = tce & IOMMU_PAGE_MASK & ~PAGE_MASK;
+	enum dma_data_direction direction = iommu_tce_direction(tce);
+
+	ret = get_user_pages_fast(tce & PAGE_MASK, 1,
+			direction != DMA_TO_DEVICE, &page);
+	if (unlikely(ret != 1)) {
+		/* pr_err("iommu_tce: get_user_pages_fast failed tce=%lx ioba=%lx ret=%d\n",
+				tce, entry << IOMMU_PAGE_SHIFT, ret); */
+		return -EFAULT;
+	}
+	hwaddr = (unsigned long) page_address(page) + offset;
+
+	ret = iommu_tce_build(tbl, entry, hwaddr, direction);
+	if (ret)
+		put_page(page);
+
+	if (ret < 0)
+		pr_err("iommu_tce: %s failed ioba=%lx, tce=%lx, ret=%d\n",
+				__func__, entry << IOMMU_PAGE_SHIFT, tce, ret);
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(iommu_put_tce_user_mode);
+
+int iommu_take_ownership(struct iommu_table *tbl)
+{
+	unsigned long sz = (tbl->it_size + 7) >> 3;
+
+	if (tbl->it_offset == 0)
+		clear_bit(0, tbl->it_map);
+
+	if (!bitmap_empty(tbl->it_map, tbl->it_size)) {
+		pr_err("iommu_tce: it_map is not empty");
+		return -EBUSY;
+	}
+
+	memset(tbl->it_map, 0xff, sz);
+	iommu_clear_tces_and_put_pages(tbl, tbl->it_offset, tbl->it_size);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(iommu_take_ownership);
+
+void iommu_release_ownership(struct iommu_table *tbl)
+{
+	unsigned long sz = (tbl->it_size + 7) >> 3;
+
+	iommu_clear_tces_and_put_pages(tbl, tbl->it_offset, tbl->it_size);
+	memset(tbl->it_map, 0, sz);
+
+	/* Restore bit#0 set by iommu_init_table() */
+	if (tbl->it_offset == 0)
+		set_bit(0, tbl->it_map);
+}
+EXPORT_SYMBOL_GPL(iommu_release_ownership);
+
+static int iommu_add_device(struct device *dev)
+{
+	struct iommu_table *tbl;
+	int ret = 0;
+
+	if (WARN_ON(dev->iommu_group)) {
+		pr_warn("iommu_tce: device %s is already in iommu group %d, skipping\n",
+				dev_name(dev),
+				iommu_group_id(dev->iommu_group));
+		return -EBUSY;
+	}
+
+	tbl = get_iommu_table_base(dev);
+	if (!tbl || !tbl->it_group) {
+		pr_debug("iommu_tce: skipping device %s with no tbl\n",
+				dev_name(dev));
+		return 0;
+	}
+
+	pr_debug("iommu_tce: adding %s to iommu group %d\n",
+			dev_name(dev), iommu_group_id(tbl->it_group));
+
+	ret = iommu_group_add_device(tbl->it_group, dev);
+	if (ret < 0)
+		pr_err("iommu_tce: %s has not been added, ret=%d\n",
+				dev_name(dev), ret);
+
+	return ret;
+}
+
+static void iommu_del_device(struct device *dev)
+{
+	iommu_group_remove_device(dev);
+}
+
+static int iommu_bus_notifier(struct notifier_block *nb,
+			      unsigned long action, void *data)
+{
+	struct device *dev = data;
+
+	switch (action) {
+	case BUS_NOTIFY_ADD_DEVICE:
+		return iommu_add_device(dev);
+	case BUS_NOTIFY_DEL_DEVICE:
+		iommu_del_device(dev);
+		return 0;
+	default:
+		return 0;
+	}
+}
+
+static struct notifier_block tce_iommu_bus_nb = {
+	.notifier_call = iommu_bus_notifier,
+};
+
+static int __init tce_iommu_init(void)
+{
+	struct pci_dev *pdev = NULL;
+
+	BUILD_BUG_ON(PAGE_SIZE < IOMMU_PAGE_SIZE);
+
+	for_each_pci_dev(pdev)
+		iommu_add_device(&pdev->dev);
+
+	bus_register_notifier(&pci_bus_type, &tce_iommu_bus_nb);
+	return 0;
+}
+
+subsys_initcall_sync(tce_iommu_init);
+
+#else
+
+void iommu_register_group(struct iommu_table *tbl,
+		int pci_domain_number, unsigned long pe_num)
+{
+}
+
+#endif /* CONFIG_IOMMU_API */
diff --git a/arch/powerpc/platforms/powernv/pci-ioda.c b/arch/powerpc/platforms/powernv/pci-ioda.c
index 1da578b..cae2555 100644
--- a/arch/powerpc/platforms/powernv/pci-ioda.c
+++ b/arch/powerpc/platforms/powernv/pci-ioda.c
@@ -605,6 +605,7 @@  static void pnv_pci_ioda_setup_dma_pe(struct pnv_phb *phb,
 			       TCE_PCI_SWINV_PAIR;
 	}
 	iommu_init_table(tbl, phb->hose->node);
+	iommu_register_group(tbl, pci_domain_nr(pe->pbus), pe->pe_number);
 
 	return;
  fail:
diff --git a/arch/powerpc/platforms/powernv/pci-p5ioc2.c b/arch/powerpc/platforms/powernv/pci-p5ioc2.c
index 92b37a0..5d378f2 100644
--- a/arch/powerpc/platforms/powernv/pci-p5ioc2.c
+++ b/arch/powerpc/platforms/powernv/pci-p5ioc2.c
@@ -86,8 +86,11 @@  static void pnv_pci_init_p5ioc2_msis(struct pnv_phb *phb) { }
 static void pnv_pci_p5ioc2_dma_dev_setup(struct pnv_phb *phb,
 					 struct pci_dev *pdev)
 {
-	if (phb->p5ioc2.iommu_table.it_map == NULL)
+	if (phb->p5ioc2.iommu_table.it_map == NULL) {
 		iommu_init_table(&phb->p5ioc2.iommu_table, phb->hose->node);
+		iommu_register_group(&phb->p5ioc2.iommu_table,
+				pci_domain_nr(phb->hose->bus), phb->opal_id);
+	}
 
 	set_iommu_table_base(&pdev->dev, &phb->p5ioc2.iommu_table);
 }
diff --git a/arch/powerpc/platforms/powernv/pci.c b/arch/powerpc/platforms/powernv/pci.c
index 55dfca844..bd69285 100644
--- a/arch/powerpc/platforms/powernv/pci.c
+++ b/arch/powerpc/platforms/powernv/pci.c
@@ -20,6 +20,7 @@ 
 #include <linux/irq.h>
 #include <linux/io.h>
 #include <linux/msi.h>
+#include <linux/iommu.h>
 
 #include <asm/sections.h>
 #include <asm/io.h>
@@ -408,6 +409,7 @@  static struct iommu_table *pnv_pci_setup_bml_iommu(struct pci_controller *hose)
 	pnv_pci_setup_iommu_table(tbl, __va(be64_to_cpup(basep)),
 				  be32_to_cpup(sizep), 0);
 	iommu_init_table(tbl, hose->node);
+	iommu_register_group(tbl, pci_domain_nr(hose->bus), 0);
 
 	/* Deal with SW invalidated TCEs when needed (BML way) */
 	swinvp = of_get_property(hose->dn, "linux,tce-sw-invalidate-info",
diff --git a/drivers/iommu/Kconfig b/drivers/iommu/Kconfig
index c332fb9..3f3abde 100644
--- a/drivers/iommu/Kconfig
+++ b/drivers/iommu/Kconfig
@@ -261,4 +261,12 @@  config SHMOBILE_IOMMU_L1SIZE
 	default 256 if SHMOBILE_IOMMU_ADDRSIZE_64MB
 	default 128 if SHMOBILE_IOMMU_ADDRSIZE_32MB
 
+config SPAPR_TCE_IOMMU
+	bool "sPAPR TCE IOMMU Support"
+	depends on PPC_POWERNV
+	select IOMMU_API
+	help
+	  Enables bits of IOMMU API required by VFIO. The iommu_ops
+	  is not implemented as it is not necessary for VFIO.
+
 endif # IOMMU_SUPPORT