drm/amd/display: set clocks to 0 on suspend on dce80
authorBhawanpreet Lakha <Bhawanpreet.Lakha@amd.com>
Tue, 5 Feb 2019 19:03:52 +0000 (14:03 -0500)
committerAlex Deucher <alexander.deucher@amd.com>
Wed, 20 Feb 2019 21:58:06 +0000 (16:58 -0500)
[Why]
When a dce80 asic was suspended, the clocks were not set to 0.
Upon resume, the new clock was compared to the existing clock,
they were found to be the same, and so the clock was not set.
This resulted in a blackscreen.

[How]
In atomic commit, check to see if there are any active pipes.
If no, set clocks to 0

Signed-off-by: Bhawanpreet Lakha <Bhawanpreet.Lakha@amd.com>
Reviewed-by: Nicholas Kazlauskas <Nicholas.Kazlauskas@amd.com>
Acked-by: Leo Li <sunpeng.li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c

index cdd1d6b7b9f2eb18cc1491bfe4b8ff9e5f470936..4e9ea50141bdaf0d10b701e4d5a47be55837fc13 100644 (file)
@@ -790,9 +790,22 @@ bool dce80_validate_bandwidth(
        struct dc *dc,
        struct dc_state *context)
 {
-       /* TODO implement when needed but for now hardcode max value*/
-       context->bw.dce.dispclk_khz = 681000;
-       context->bw.dce.yclk_khz = 250000 * MEMORY_TYPE_MULTIPLIER_CZ;
+       int i;
+       bool at_least_one_pipe = false;
+
+       for (i = 0; i < dc->res_pool->pipe_count; i++) {
+               if (context->res_ctx.pipe_ctx[i].stream)
+                       at_least_one_pipe = true;
+       }
+
+       if (at_least_one_pipe) {
+               /* TODO implement when needed but for now hardcode max value*/
+               context->bw.dce.dispclk_khz = 681000;
+               context->bw.dce.yclk_khz = 250000 * MEMORY_TYPE_MULTIPLIER_CZ;
+       } else {
+               context->bw.dce.dispclk_khz = 0;
+               context->bw.dce.yclk_khz = 0;
+       }
 
        return true;
 }