powerpc/powernv: Add indirect levels to it_userspace
We want to support sparse memory and therefore huge chunks of DMA windows
do not need to be mapped. If a DMA window big enough to require 2 or more
indirect levels, and a DMA window is used to map all RAM (which is
a default case for 64bit window), we can actually save some memory by
not allocation TCE for regions which we are not going to map anyway.
The hardware tables alreary support indirect levels but we also keep
host-physical-to-userspace translation array which is allocated by
vmalloc() and is a flat array which might use quite some memory.
This converts it_userspace from vmalloc'ed array to a multi level table.
As the format becomes platform dependend, this replaces the direct access
to it_usespace with a iommu_table_ops::useraddrptr hook which returns
a pointer to the userspace copy of a TCE; future extension will return
NULL if the level was not allocated.
This should not change non-KVM handling of TCE tables and it_userspace
will not be allocated for non-KVM tables.
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
diff --git a/arch/powerpc/platforms/powernv/pci-ioda.c b/arch/powerpc/platforms/powernv/pci-ioda.c
index 4abf117..fc38f06 100644
--- a/arch/powerpc/platforms/powernv/pci-ioda.c
+++ b/arch/powerpc/platforms/powernv/pci-ioda.c
@@ -2036,6 +2036,7 @@
#ifdef CONFIG_IOMMU_API
.exchange = pnv_ioda1_tce_xchg,
.exchange_rm = pnv_ioda1_tce_xchg_rm,
+ .useraddrptr = pnv_tce_useraddrptr,
#endif
.clear = pnv_ioda1_tce_free,
.get = pnv_tce_get,
@@ -2200,6 +2201,7 @@
#ifdef CONFIG_IOMMU_API
.exchange = pnv_ioda2_tce_xchg,
.exchange_rm = pnv_ioda2_tce_xchg_rm,
+ .useraddrptr = pnv_tce_useraddrptr,
#endif
.clear = pnv_ioda2_tce_free,
.get = pnv_tce_get,
@@ -2455,7 +2457,7 @@
static long pnv_pci_ioda2_create_table(struct iommu_table_group *table_group,
int num, __u32 page_shift, __u64 window_size, __u32 levels,
- struct iommu_table **ptbl)
+ bool alloc_userspace_copy, struct iommu_table **ptbl)
{
struct pnv_ioda_pe *pe = container_of(table_group, struct pnv_ioda_pe,
table_group);
@@ -2472,7 +2474,7 @@
ret = pnv_pci_ioda2_table_alloc_pages(nid,
bus_offset, page_shift, window_size,
- levels, tbl);
+ levels, alloc_userspace_copy, tbl);
if (ret) {
iommu_tce_table_put(tbl);
return ret;
@@ -2505,7 +2507,7 @@
rc = pnv_pci_ioda2_create_table(&pe->table_group, 0,
IOMMU_PAGE_SHIFT_4K,
window_size,
- POWERNV_IOMMU_DEFAULT_LEVELS, &tbl);
+ POWERNV_IOMMU_DEFAULT_LEVELS, false, &tbl);
if (rc) {
pe_err(pe, "Failed to create 32-bit TCE table, err %ld",
rc);
@@ -2592,7 +2594,16 @@
tce_table_size, direct_table_size);
}
- return bytes;
+ return bytes + bytes; /* one for HW table, one for userspace copy */
+}
+
+static long pnv_pci_ioda2_create_table_userspace(
+ struct iommu_table_group *table_group,
+ int num, __u32 page_shift, __u64 window_size, __u32 levels,
+ struct iommu_table **ptbl)
+{
+ return pnv_pci_ioda2_create_table(table_group,
+ num, page_shift, window_size, levels, true, ptbl);
}
static void pnv_ioda2_take_ownership(struct iommu_table_group *table_group)
@@ -2621,7 +2632,7 @@
static struct iommu_table_group_ops pnv_pci_ioda2_ops = {
.get_table_size = pnv_pci_ioda2_get_table_size,
- .create_table = pnv_pci_ioda2_create_table,
+ .create_table = pnv_pci_ioda2_create_table_userspace,
.set_window = pnv_pci_ioda2_set_window,
.unset_window = pnv_pci_ioda2_unset_window,
.take_ownership = pnv_ioda2_take_ownership,
@@ -2726,7 +2737,7 @@
static struct iommu_table_group_ops pnv_pci_ioda2_npu_ops = {
.get_table_size = pnv_pci_ioda2_get_table_size,
- .create_table = pnv_pci_ioda2_create_table,
+ .create_table = pnv_pci_ioda2_create_table_userspace,
.set_window = pnv_pci_ioda2_npu_set_window,
.unset_window = pnv_pci_ioda2_npu_unset_window,
.take_ownership = pnv_ioda2_npu_take_ownership,