PAGE_SIZE is 64KB by default on PPC64 platform. When vfio
passthrough a pci device of which MMIO BARs are smaller than
64KB(PAGE_SIZE), guest would not handle the mmio accesses
to the BARs which leads to mmio emulations in host.
This is because vfio would not allow to passthrough one
BAR's mmio page which may be shared with other BARs.
To solve this performance issue, this patch enforces the
alignment of all MMIO BARs allocations to be at least PAGE_SIZE
on PPC64 platform because we have enough address space, so
that one BAR's mmio page would not be shared with other BARs.
Signed-off-by: Yongji Xie <xyj...@linux.vnet.ibm.com>
---
arch/powerpc/kernel/pci-common.c | 10 +++++++++-
1 file changed, 9 insertions(+), 1 deletion(-)
diff --git a/arch/powerpc/kernel/pci-common.c b/arch/powerpc/kernel/pci-common.c
index 0f7a60f..6989e0f 100644
--- a/arch/powerpc/kernel/pci-common.c
+++ b/arch/powerpc/kernel/pci-common.c
@@ -1074,6 +1074,11 @@ static int skip_isa_ioresource_align(struct pci_dev *dev)
* bits, so it's ok to allocate at, say, 0x2800-0x28ff,
* but we want to try to avoid allocating at 0x2900-0x2bff
* which might have be mirrored at 0x0100-0x03ff..
+ *
+ * And for PPC64, we enforce the alignment of all MMIO BARs
+ * allocations to be at least PAGE_SIZE(64KB). This would be
+ * helpful to improve performance when we passthrough
+ * a PCI device of which BARs are smaller than PAGE_SIZE
*/
resource_size_t pcibios_align_resource(void *data, const struct resource *res,
resource_size_t size, resource_size_t align)
@@ -1087,7 +1092,10 @@ resource_size_t pcibios_align_resource(void *data, const
struct resource *res,
if (start & 0x300)
start = (start + 0x3ff) & ~0x3ff;
}
-
+#ifdef CONFIG_PPC64
+ if (res->flags & IORESOURCE_MEM)
+ start = PAGE_ALIGN(start);
+#endif
return start;
}
EXPORT_SYMBOL(pcibios_align_resource);
--
1.7.9.5
--
To unsubscribe from this list: send the line "unsubscribe linux-api" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
