gpu: nvgpu: Use common allocator for context

Reduce amount of duplicate code around memory allocation by using
common helpers, and common data structure for storing results of
allocations.

Bug 1605769

Change-Id: I10c226e2377aa867a5cf11be61d08a9d67206b1d
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/720507

1 files changed, 1 insertions, 1 deletions

diff --git a/drivers/gpu/nvgpu/gk20a/mm_gk20a.c b/drivers/gpu/nvgpu/gk20a/mm_gk20a.c
index e2e5fdd7..c86fee45 100644
--- a/drivers/gpu/nvgpu/gk20a/mm_gk20a.c
+++ b/drivers/gpu/nvgpu/gk20a/mm_gk20a.c
@@ -1581,7 +1581,7 @@ void gk20a_gmmu_free_attr(struct gk20a *g, enum dma_attr attr,
 {
         struct device *d = dev_from_gk20a(g);
 
-        if (mem->cpu_va) {
+        if (mem->cpu_va || mem->pages) {
                 if (attr) {
                         DEFINE_DMA_ATTRS(attrs);
                         dma_set_attr(attr, &attrs);