gpu: nvgpu: Use gradual slowdown

Program clock slowdown to happen using gradual slowdown. It is
significantly faster than the default slowdown.

Change-Id: I9e5259889637fce2c0b083a424b54af12bb45c25
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/819698

1 files changed, 31 insertions, 0 deletions

diff --git a/drivers/gpu/nvgpu/gk20a/therm_gk20a.c b/drivers/gpu/nvgpu/gk20a/therm_gk20a.c
index b02113ad..fd4149ae 100644
--- a/drivers/gpu/nvgpu/gk20a/therm_gk20a.c
+++ b/drivers/gpu/nvgpu/gk20a/therm_gk20a.c
@@ -35,6 +35,8 @@ static int gk20a_init_therm_setup_sw(struct gk20a *g)
 
 static int gk20a_init_therm_setup_hw(struct gk20a *g)
 {
+        u32 v;
+
         /* program NV_THERM registers */
         gk20a_writel(g, therm_use_a_r(), NV_THERM_USE_A_INIT);
         gk20a_writel(g, therm_evt_ext_therm_0_r(),
@@ -44,6 +46,35 @@ static int gk20a_init_therm_setup_hw(struct gk20a *g)
         gk20a_writel(g, therm_evt_ext_therm_2_r(),
                 NV_THERM_EVT_EXT_THERM_2_INIT);
 
+        gk20a_writel(g, therm_grad_stepping_table_r(0),
+                therm_grad_stepping_table_slowdown_factor0_f(therm_grad_stepping_table_slowdown_factor0_fpdiv_by1p5_f()) |
+                therm_grad_stepping_table_slowdown_factor1_f(therm_grad_stepping_table_slowdown_factor0_fpdiv_by2_f()) |
+                therm_grad_stepping_table_slowdown_factor2_f(therm_grad_stepping_table_slowdown_factor0_fpdiv_by4_f()) |
+                therm_grad_stepping_table_slowdown_factor3_f(therm_grad_stepping_table_slowdown_factor0_fpdiv_by8_f()) |
+                therm_grad_stepping_table_slowdown_factor4_f(therm_grad_stepping_table_slowdown_factor0_fpdiv_by8_f()));
+        gk20a_writel(g, therm_grad_stepping_table_r(1),
+                therm_grad_stepping_table_slowdown_factor0_f(therm_grad_stepping_table_slowdown_factor0_fpdiv_by8_f()) |
+                therm_grad_stepping_table_slowdown_factor1_f(therm_grad_stepping_table_slowdown_factor0_fpdiv_by8_f()) |
+                therm_grad_stepping_table_slowdown_factor2_f(therm_grad_stepping_table_slowdown_factor0_fpdiv_by8_f()) |
+                therm_grad_stepping_table_slowdown_factor3_f(therm_grad_stepping_table_slowdown_factor0_fpdiv_by8_f()) |
+                therm_grad_stepping_table_slowdown_factor4_f(therm_grad_stepping_table_slowdown_factor0_fpdiv_by8_f()));
+
+        v = gk20a_readl(g, therm_clk_timing_r(0));
+        v |= therm_clk_timing_grad_slowdown_enabled_f();
+        gk20a_writel(g, therm_clk_timing_r(0), v);
+
+        v = gk20a_readl(g, therm_config2_r());
+        v |= therm_config2_grad_enable_f(1);
+        v |= therm_config2_slowdown_factor_extended_f(1);
+        gk20a_writel(g, therm_config2_r(), v);
+
+        gk20a_writel(g, therm_grad_stepping1_r(),
+                        therm_grad_stepping1_pdiv_duration_f(32));
+
+        v = gk20a_readl(g, therm_grad_stepping0_r());
+        v |= therm_grad_stepping0_feature_enable_f();
+        gk20a_writel(g, therm_grad_stepping0_r(), v);
+
         return 0;
 }