gpu: nvgpu: use a separate big vm for cde

Allocate a separate VM for CDE channels instead of using the system
(PMU) vm, and make it much bigger than the PMU's to fit the maximum
number of CDE channels there.

Bug 1566740

Change-Id: I4f487c40c9ec79cc9ffb880b0ecd3f47eb450336
Signed-off-by: Konsta Holtta <kholtta@nvidia.com>
Reviewed-on: http://git-master/r/815149
Reviewed-by: Automatic_Commit_Validation_User

1 files changed, 2 insertions, 2 deletions

diff --git a/drivers/gpu/nvgpu/gk20a/cde_gk20a.c b/drivers/gpu/nvgpu/gk20a/cde_gk20a.c
index 6d8633a7..01fca058 100644
--- a/drivers/gpu/nvgpu/gk20a/cde_gk20a.c
+++ b/drivers/gpu/nvgpu/gk20a/cde_gk20a.c
@@ -1194,8 +1194,8 @@ static int gk20a_cde_load(struct gk20a_cde_ctx *cde_ctx)
         }
 
         /* bind the channel to the vm */
-        gk20a_vm_get(&g->mm.pmu.vm);
-        ch->vm = &g->mm.pmu.vm;
+        gk20a_vm_get(&g->mm.cde.vm);
+        ch->vm = &g->mm.cde.vm;
         err = channel_gk20a_commit_va(ch);
         if (err) {
                 gk20a_warn(&cde_ctx->pdev->dev, "cde: could not bind vm");