swiotlb: map_page fix for highmem systems

The current code calls virt_to_phys() on address that might
be in highmem, which is bad.  This wasn't needed, anyway, because
we already have the physical address we need.

Get rid of the now-unused virtual address as well.

Signed-off-by: Becky Bruce <[email protected]>
Acked-by: FUJITA Tomonori <[email protected]>
Signed-off-by: Kumar Gala <[email protected]>
Cc: [email protected]
Cc: [email protected]
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
This commit is contained in:
Becky Bruce 2009-04-08 09:09:17 -05:00 • committed by Ingo Molnar
commit dd6b02fe42

View file

@ -640,7 +640,6 @@ dma_addr_t swiotlb_map_page(struct device *dev, struct page *page,
struct dma_attrs *attrs)
{
phys_addr_t phys = page_to_phys(page) + offset;
void *ptr = page_address(page) + offset;
dma_addr_t dev_addr = swiotlb_phys_to_bus(dev, phys);
void *map;
@ -651,7 +650,7 @@ dma_addr_t swiotlb_map_page(struct device *dev, struct page *page,
* buffering it.
*/
if (!address_needs_mapping(dev, dev_addr, size) &&
!range_needs_mapping(virt_to_phys(ptr), size))
!range_needs_mapping(phys, size))
return dev_addr;
/*