gpu: nvgpu: Fix race in PMU state transitions.

PMU response(intr callback for messages) can run faster
than the kthread posting commands to PMU.

This causes the PMU message callback to skip important pmu
state change(which happens just after the PMU command is posted).

Solution:
State change should be triggered from only inside the intr callback.
Other places can only update the pmu_state variable.
This change also adds error check to print in case command post fails.

JIRA GPUT19X-20

Change-Id: Ib0a4275440455342a898c93ea9d86c5822e039a7
Signed-off-by: Deepak Goyal <dgoyal@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1583577
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>

drivers/gpu/nvgpu/common/pmu/pmu_ipc.c

@@ -640,7 +640,7 @@ int nvgpu_pmu_cmd_post(struct gk20a *g, struct pmu_cmd *cmd,
 
 	nvgpu_log_fn(g, "done");
 
-	return 0;
+	return err;
 
 clean_up:
 	nvgpu_log_fn(g, "fail");