gpu: nvgpu: use vidmem for page tables if available

Use the common gk20a_gmmu_alloc() that tries vidmem too.

Jira DNVGPU-20

Change-Id: I4ea02bc4962d299c6f71444048d4a2a22bd80f55
Signed-off-by: Konsta Holtta <kholtta@nvidia.com>
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: http://git-master/r/1206404
(cherry picked from commit 7297727cce8c5c7b26f82afe98cc5428135b4777)
Reviewed-on: http://git-master/r/1178831
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>

1 files changed, 2 insertions, 2 deletions

diff --git a/drivers/gpu/nvgpu/gk20a/mm_gk20a.c b/drivers/gpu/nvgpu/gk20a/mm_gk20a.c
index 1124a17d..553b3ac5 100644
--- a/drivers/gpu/nvgpu/gk20a/mm_gk20a.c
+++ b/drivers/gpu/nvgpu/gk20a/mm_gk20a.c
@@ -1173,9 +1173,9 @@ static int alloc_gmmu_pages(struct vm_gk20a *vm, u32 order,
          * default.
          */
         if (IS_ENABLED(CONFIG_ARM64))
-                err = gk20a_gmmu_alloc_sys(g, len, &entry->mem);
+                err = gk20a_gmmu_alloc(g, len, &entry->mem);
         else
-                err = gk20a_gmmu_alloc_attr_sys(g, DMA_ATTR_NO_KERNEL_MAPPING,
+                err = gk20a_gmmu_alloc_attr(g, DMA_ATTR_NO_KERNEL_MAPPING,
                                 len, &entry->mem);