gpu: nvgpu: add vidmem manager

Use the nvgpu-internal buddy allocator for video memory allocations,
instead of nvmap. This allows better integration for copyengine, BAR1
mapping to userspace, etc.

Jira DNVGPU-38

Change-Id: I9fd67b76cd39721e4cd8e525ad0ed76f497e8b99
Signed-off-by: Konsta Holtta <kholtta@nvidia.com>
Reviewed-on: http://git-master/r/1181151
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>

1 files changed, 1 insertions, 1 deletions

diff --git a/drivers/gpu/nvgpu/gk20a/ce2_gk20a.c b/drivers/gpu/nvgpu/gk20a/ce2_gk20a.c
index e2f2d9e9..a69cce0c 100644
--- a/drivers/gpu/nvgpu/gk20a/ce2_gk20a.c
+++ b/drivers/gpu/nvgpu/gk20a/ce2_gk20a.c
@@ -224,7 +224,7 @@ static inline int gk20a_get_valid_launch_flags(struct gk20a *g, int launch_flags
 {
         /* there is no local memory available,
         don't allow local memory related CE flags */
-        if (!g->mm.vidmem_size) {
+        if (!g->mm.vidmem.size) {
                 launch_flags &= ~(NVGPU_CE_SRC_LOCATION_LOCAL_FB |
                         NVGPU_CE_DST_LOCATION_LOCAL_FB);
         }