MLK-15007-1 iommu: arm: pgtable: alloc pagetable in DMA area
Normally the iommu pagetable could be in 64bit address space,
but we have one patch to address PCIE driver, 'commit 9e03e5076269
("MLK-15064-2 ARM64: DMA: limit the dma mask to be 32bit")'
The patch restrict swiotlb and iommu dma to be in 32bit address.
So if we allocate pages in highmem, then dma_map_single will return
a 32bit address. Then, we will get "Cannot accommodate DMA
translation for IOMMU page tables", because `dma != virt_to_phys(pages)`.
So we strict the lpae iommu pgtable in DMA area to fix this issue.
Signed-off-by: Peng Fan <peng.fan@nxp.com>
This commit is contained in:
@ -14,6 +14,7 @@
|
||||
* along with this program. If not, see <http://www.gnu.org/licenses/>.
|
||||
*
|
||||
* Copyright (C) 2014 ARM Limited
|
||||
* Copyright 2017 NXP
|
||||
*
|
||||
* Author: Will Deacon <will.deacon@arm.com>
|
||||
*/
|
||||
@ -215,7 +216,7 @@ static void *__arm_lpae_alloc_pages(size_t size, gfp_t gfp,
|
||||
{
|
||||
struct device *dev = cfg->iommu_dev;
|
||||
dma_addr_t dma;
|
||||
void *pages = alloc_pages_exact(size, gfp | __GFP_ZERO);
|
||||
void *pages = alloc_pages_exact(size, gfp | __GFP_ZERO | __GFP_DMA);
|
||||
|
||||
if (!pages)
|
||||
return NULL;
|
||||
|
||||
Reference in New Issue
Block a user