gpu: nvgpu: vgpu: add freq capping support

Currently callbacks from the PM_QOS framework (for
thermal events), result in a RPC call to set GPU frequency.

Since the governor will now be responsible for setting desired
rate, the max PM_QOS callback will now cap the possible
GPU frequency w/ a new RPC call to the server. The server
is responsible for setting the ultimate frequency
based on the cap & desired rates.

Jira VFND-3699

Change-Id: I806e309c40abc2f1381b6a23f2d898cfe26f9794
Signed-off-by: Sachit Kadle <skadle@nvidia.com>
Reviewed-on: http://git-master/r/1295543
(cherry picked from commit e81693c6e087f8f10a985be83715042fc590d6db)
Reviewed-on: http://git-master/r/1282467
(cherry picked from commit 7b4e0db647572e82a8d53e823c36b465781f4942)
Reviewed-on: http://git-master/r/1321836
(cherry picked from commit 57dafc08a57ea768eb302bf1adb901425ce2f835)
Reviewed-on: http://git-master/r/1313469
Tested-by: Aparna Das <aparnad@nvidia.com>
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Richard Zhao <rizhao@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>

1 files changed, 1 insertions, 8 deletions

diff --git a/drivers/gpu/nvgpu/vgpu/vgpu.c b/drivers/gpu/nvgpu/vgpu/vgpu.c
index e8a778f5..52f375f9 100644
--- a/drivers/gpu/nvgpu/vgpu/vgpu.c
+++ b/drivers/gpu/nvgpu/vgpu/vgpu.c
@@ -516,20 +516,13 @@ static int vgpu_qos_notify(struct notifier_block *nb,
                         container_of(nb, struct gk20a_scale_profile,
                         qos_notify_block);
         struct gk20a *g = get_gk20a(profile->dev);
-        struct tegra_vgpu_cmd_msg msg = {};
-        struct tegra_vgpu_gpu_clk_rate_params *p = &msg.params.gpu_clk_rate;
         u32 max_freq;
         int err;
 
         gk20a_dbg_fn("");
 
         max_freq = (u32)pm_qos_read_max_bound(PM_QOS_GPU_FREQ_BOUNDS);
-
-        msg.cmd = TEGRA_VGPU_CMD_SET_GPU_CLK_RATE;
-        msg.handle = vgpu_get_handle_from_dev(profile->dev);
-        p->rate = max_freq;
-        err = vgpu_comm_sendrecv(&msg, sizeof(msg), sizeof(msg));
-        err = err ? err : msg.ret;
+        err = vgpu_clk_cap_rate(profile->dev, max_freq);
         if (err)
                 nvgpu_err(g, "%s failed, err=%d", __func__, err);