* [PATCH guest kernel] powerpc/pseries: Enable VFIO
@ 2017-03-24 6:37 Alexey Kardashevskiy
2017-03-26 22:41 ` David Gibson
2017-05-01 2:58 ` [guest,kernel] " Michael Ellerman
0 siblings, 2 replies; 3+ messages in thread
From: Alexey Kardashevskiy @ 2017-03-24 6:37 UTC (permalink / raw)
To: linuxppc-dev; +Cc: Alexey Kardashevskiy, David Gibson, Paul Mackerras
This enables VFIO on pseries host in order to allow VFIO in nested guest
under PR KVM or DPDK in a HV guest. This adds support of
the VFIO_SPAPR_TCE_IOMMU type.
This adds exchange() callback to allow TCE updates by the SPAPR TCE IOMMU
driver in VFIO.
This initializes DMA32 window parameters in iommu_table_group as
as this does not implement VFIO_SPAPR_TCE_v2_IOMMU and
VFIO_SPAPR_TCE_IOMMU just reuses the existing DMA32 window.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
---
arch/powerpc/platforms/pseries/iommu.c | 40 ++++++++++++++++++++++++++++++++--
1 file changed, 38 insertions(+), 2 deletions(-)
diff --git a/arch/powerpc/platforms/pseries/iommu.c b/arch/powerpc/platforms/pseries/iommu.c
index 4d757eaa46bf..7c8ed68d727e 100644
--- a/arch/powerpc/platforms/pseries/iommu.c
+++ b/arch/powerpc/platforms/pseries/iommu.c
@@ -550,6 +550,7 @@ static void iommu_table_setparms(struct pci_controller *phb,
static void iommu_table_setparms_lpar(struct pci_controller *phb,
struct device_node *dn,
struct iommu_table *tbl,
+ struct iommu_table_group *table_group,
const __be32 *dma_window)
{
unsigned long offset, size;
@@ -563,6 +564,9 @@ static void iommu_table_setparms_lpar(struct pci_controller *phb,
tbl->it_type = TCE_PCI;
tbl->it_offset = offset >> tbl->it_page_shift;
tbl->it_size = size >> tbl->it_page_shift;
+
+ table_group->tce32_start = offset;
+ table_group->tce32_size = size;
}
struct iommu_table_ops iommu_table_pseries_ops = {
@@ -651,8 +655,38 @@ static void pci_dma_bus_setup_pSeries(struct pci_bus *bus)
pr_debug("ISA/IDE, window size is 0x%llx\n", pci->phb->dma_window_size);
}
+#ifdef CONFIG_IOMMU_API
+static int tce_exchange_pSeries(struct iommu_table *tbl, long index,
+ unsigned long *tce, enum dma_data_direction *direction)
+{
+ long rc;
+ unsigned long ioba = (unsigned long) index << tbl->it_page_shift;
+ unsigned long flags, oldtce = 0;
+ u64 proto_tce = iommu_direction_to_tce_perm(*direction);
+ unsigned long newtce = *tce | proto_tce;
+
+ spin_lock_irqsave(&tbl->large_pool.lock, flags);
+
+ rc = plpar_tce_get((u64)tbl->it_index, ioba, &oldtce);
+ if (!rc)
+ rc = plpar_tce_put((u64)tbl->it_index, ioba, newtce);
+
+ if (!rc) {
+ *direction = iommu_tce_direction(oldtce);
+ *tce = oldtce & ~(TCE_PCI_READ | TCE_PCI_WRITE);
+ }
+
+ spin_unlock_irqrestore(&tbl->large_pool.lock, flags);
+
+ return rc;
+}
+#endif
+
struct iommu_table_ops iommu_table_lpar_multi_ops = {
.set = tce_buildmulti_pSeriesLP,
+#ifdef CONFIG_IOMMU_API
+ .exchange = tce_exchange_pSeries,
+#endif
.clear = tce_freemulti_pSeriesLP,
.get = tce_get_pSeriesLP
};
@@ -689,7 +723,8 @@ static void pci_dma_bus_setup_pSeriesLP(struct pci_bus *bus)
if (!ppci->table_group) {
ppci->table_group = iommu_pseries_alloc_group(ppci->phb->node);
tbl = ppci->table_group->tables[0];
- iommu_table_setparms_lpar(ppci->phb, pdn, tbl, dma_window);
+ iommu_table_setparms_lpar(ppci->phb, pdn, tbl,
+ ppci->table_group, dma_window);
tbl->it_ops = &iommu_table_lpar_multi_ops;
iommu_init_table(tbl, ppci->phb->node);
iommu_register_group(ppci->table_group,
@@ -1143,7 +1178,8 @@ static void pci_dma_dev_setup_pSeriesLP(struct pci_dev *dev)
if (!pci->table_group) {
pci->table_group = iommu_pseries_alloc_group(pci->phb->node);
tbl = pci->table_group->tables[0];
- iommu_table_setparms_lpar(pci->phb, pdn, tbl, dma_window);
+ iommu_table_setparms_lpar(pci->phb, pdn, tbl,
+ pci->table_group, dma_window);
tbl->it_ops = &iommu_table_lpar_multi_ops;
iommu_init_table(tbl, pci->phb->node);
iommu_register_group(pci->table_group,
--
2.11.0
^ permalink raw reply related [flat|nested] 3+ messages in thread* Re: [PATCH guest kernel] powerpc/pseries: Enable VFIO
2017-03-24 6:37 [PATCH guest kernel] powerpc/pseries: Enable VFIO Alexey Kardashevskiy
@ 2017-03-26 22:41 ` David Gibson
2017-05-01 2:58 ` [guest,kernel] " Michael Ellerman
1 sibling, 0 replies; 3+ messages in thread
From: David Gibson @ 2017-03-26 22:41 UTC (permalink / raw)
To: Alexey Kardashevskiy; +Cc: linuxppc-dev, Paul Mackerras
[-- Attachment #1: Type: text/plain, Size: 4266 bytes --]
On Fri, Mar 24, 2017 at 05:37:21PM +1100, Alexey Kardashevskiy wrote:
> This enables VFIO on pseries host in order to allow VFIO in nested guest
> under PR KVM or DPDK in a HV guest. This adds support of
> the VFIO_SPAPR_TCE_IOMMU type.
>
> This adds exchange() callback to allow TCE updates by the SPAPR TCE IOMMU
> driver in VFIO.
>
> This initializes DMA32 window parameters in iommu_table_group as
> as this does not implement VFIO_SPAPR_TCE_v2_IOMMU and
> VFIO_SPAPR_TCE_IOMMU just reuses the existing DMA32 window.
>
> Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
> ---
> arch/powerpc/platforms/pseries/iommu.c | 40 ++++++++++++++++++++++++++++++++--
> 1 file changed, 38 insertions(+), 2 deletions(-)
>
> diff --git a/arch/powerpc/platforms/pseries/iommu.c b/arch/powerpc/platforms/pseries/iommu.c
> index 4d757eaa46bf..7c8ed68d727e 100644
> --- a/arch/powerpc/platforms/pseries/iommu.c
> +++ b/arch/powerpc/platforms/pseries/iommu.c
> @@ -550,6 +550,7 @@ static void iommu_table_setparms(struct pci_controller *phb,
> static void iommu_table_setparms_lpar(struct pci_controller *phb,
> struct device_node *dn,
> struct iommu_table *tbl,
> + struct iommu_table_group *table_group,
> const __be32 *dma_window)
> {
> unsigned long offset, size;
> @@ -563,6 +564,9 @@ static void iommu_table_setparms_lpar(struct pci_controller *phb,
> tbl->it_type = TCE_PCI;
> tbl->it_offset = offset >> tbl->it_page_shift;
> tbl->it_size = size >> tbl->it_page_shift;
> +
> + table_group->tce32_start = offset;
> + table_group->tce32_size = size;
> }
>
> struct iommu_table_ops iommu_table_pseries_ops = {
> @@ -651,8 +655,38 @@ static void pci_dma_bus_setup_pSeries(struct pci_bus *bus)
> pr_debug("ISA/IDE, window size is 0x%llx\n", pci->phb->dma_window_size);
> }
>
> +#ifdef CONFIG_IOMMU_API
> +static int tce_exchange_pSeries(struct iommu_table *tbl, long index,
> + unsigned long *tce, enum dma_data_direction *direction)
> +{
> + long rc;
> + unsigned long ioba = (unsigned long) index << tbl->it_page_shift;
> + unsigned long flags, oldtce = 0;
> + u64 proto_tce = iommu_direction_to_tce_perm(*direction);
> + unsigned long newtce = *tce | proto_tce;
> +
> + spin_lock_irqsave(&tbl->large_pool.lock, flags);
> +
> + rc = plpar_tce_get((u64)tbl->it_index, ioba, &oldtce);
> + if (!rc)
> + rc = plpar_tce_put((u64)tbl->it_index, ioba, newtce);
> +
> + if (!rc) {
> + *direction = iommu_tce_direction(oldtce);
> + *tce = oldtce & ~(TCE_PCI_READ | TCE_PCI_WRITE);
> + }
> +
> + spin_unlock_irqrestore(&tbl->large_pool.lock, flags);
> +
> + return rc;
> +}
> +#endif
> +
> struct iommu_table_ops iommu_table_lpar_multi_ops = {
> .set = tce_buildmulti_pSeriesLP,
> +#ifdef CONFIG_IOMMU_API
> + .exchange = tce_exchange_pSeries,
> +#endif
> .clear = tce_freemulti_pSeriesLP,
> .get = tce_get_pSeriesLP
> };
> @@ -689,7 +723,8 @@ static void pci_dma_bus_setup_pSeriesLP(struct pci_bus *bus)
> if (!ppci->table_group) {
> ppci->table_group = iommu_pseries_alloc_group(ppci->phb->node);
> tbl = ppci->table_group->tables[0];
> - iommu_table_setparms_lpar(ppci->phb, pdn, tbl, dma_window);
> + iommu_table_setparms_lpar(ppci->phb, pdn, tbl,
> + ppci->table_group, dma_window);
> tbl->it_ops = &iommu_table_lpar_multi_ops;
> iommu_init_table(tbl, ppci->phb->node);
> iommu_register_group(ppci->table_group,
> @@ -1143,7 +1178,8 @@ static void pci_dma_dev_setup_pSeriesLP(struct pci_dev *dev)
> if (!pci->table_group) {
> pci->table_group = iommu_pseries_alloc_group(pci->phb->node);
> tbl = pci->table_group->tables[0];
> - iommu_table_setparms_lpar(pci->phb, pdn, tbl, dma_window);
> + iommu_table_setparms_lpar(pci->phb, pdn, tbl,
> + pci->table_group, dma_window);
> tbl->it_ops = &iommu_table_lpar_multi_ops;
> iommu_init_table(tbl, pci->phb->node);
> iommu_register_group(pci->table_group,
--
David Gibson | I'll have my music baroque, and my code
david AT gibson.dropbear.id.au | minimalist, thank you. NOT _the_ _other_
| _way_ _around_!
http://www.ozlabs.org/~dgibson
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 819 bytes --]
^ permalink raw reply [flat|nested] 3+ messages in thread* Re: [guest,kernel] powerpc/pseries: Enable VFIO
2017-03-24 6:37 [PATCH guest kernel] powerpc/pseries: Enable VFIO Alexey Kardashevskiy
2017-03-26 22:41 ` David Gibson
@ 2017-05-01 2:58 ` Michael Ellerman
1 sibling, 0 replies; 3+ messages in thread
From: Michael Ellerman @ 2017-05-01 2:58 UTC (permalink / raw)
To: Alexey Kardashevskiy, linuxppc-dev
Cc: Alexey Kardashevskiy, Paul Mackerras, David Gibson
On Fri, 2017-03-24 at 06:37:21 UTC, Alexey Kardashevskiy wrote:
> This enables VFIO on pseries host in order to allow VFIO in nested guest
> under PR KVM or DPDK in a HV guest. This adds support of
> the VFIO_SPAPR_TCE_IOMMU type.
>
> This adds exchange() callback to allow TCE updates by the SPAPR TCE IOMMU
> driver in VFIO.
>
> This initializes DMA32 window parameters in iommu_table_group as
> as this does not implement VFIO_SPAPR_TCE_v2_IOMMU and
> VFIO_SPAPR_TCE_IOMMU just reuses the existing DMA32 window.
>
> Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
> Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Applied to powerpc next, thanks.
https://git.kernel.org/powerpc/c/b6e1f6adce8068620f728f717f90f4
cheers
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2017-05-01 2:58 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2017-03-24 6:37 [PATCH guest kernel] powerpc/pseries: Enable VFIO Alexey Kardashevskiy
2017-03-26 22:41 ` David Gibson
2017-05-01 2:58 ` [guest,kernel] " Michael Ellerman
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).