gpu: nvgpu: gp10x: support in-kernel vidmem mappings

Propagate the buffer aperture flag in gk20a_locked_gmmu_map up so that
buffers represented as a mem_desc and present in vidmem can be mapped to
gpu.

JIRA DNVGPU-18
JIRA DNVGPU-76

Change-Id: Icd675e83e3c28836f0ed8880425748697713bb0a
Signed-off-by: Konsta Holtta <kholtta@nvidia.com>
Reviewed-on: http://git-master/r/1169296
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>

1 files changed, 2 insertions, 1 deletions

diff --git a/drivers/gpu/nvgpu/gp10b/gr_gp10b.c b/drivers/gpu/nvgpu/gp10b/gr_gp10b.c
index 88893886..6f5016cb 100644
--- a/drivers/gpu/nvgpu/gp10b/gr_gp10b.c
+++ b/drivers/gpu/nvgpu/gp10b/gr_gp10b.c
@@ -836,7 +836,8 @@ int gr_gp10b_alloc_buffer(struct vm_gk20a *vm, size_t size,
                                 size,
                                 NVGPU_MAP_BUFFER_FLAGS_CACHEABLE_TRUE,
                                 gk20a_mem_flag_none,
-                                false);
+                                false,
+                                mem->aperture);
 
         if (!mem->gpu_va) {
                 err = -ENOMEM;