gpu: nvgpu: gm20b: fix issue with rail gating ref count

gpu rail gating reference count is going wrong because
"can_railgate" is set to false during probe(). For rail-gating
to work no gpu re-work is needed and by default rail-gating
is enabled with INT_MAX delay.

Bug 200044987

Change-Id: I9367275cd18c34cb19a51193353585789ba44c03
Signed-off-by: Seshendra Gadagottu <sgadagottu@nvidia.com>
Reviewed-on: http://git-master/r/556568
Reviewed-by: Mitch Luban <mluban@nvidia.com>

1 files changed, 1 insertions, 2 deletions

diff --git a/drivers/gpu/nvgpu/gk20a/platform_gk20a_tegra.c b/drivers/gpu/nvgpu/gk20a/platform_gk20a_tegra.c
index 59c83686..5c80294a 100644
--- a/drivers/gpu/nvgpu/gk20a/platform_gk20a_tegra.c
+++ b/drivers/gpu/nvgpu/gk20a/platform_gk20a_tegra.c
@@ -450,9 +450,8 @@ static int gk20a_tegra_probe(struct platform_device *dev)
         if (tegra_get_chipid() == TEGRA_CHIPID_TEGRA21) {
                 np = of_find_node_by_path("/gpu-dvfs-rework");
                 if (!(np && of_device_is_available(np))) {
-                        platform->can_railgate = false;
                         platform->devfreq_governor = "";
-                        dev_warn(&dev->dev, "board does not support scaling or powergating");
+                        dev_warn(&dev->dev, "board does not support scaling");
                 }
         }