[PATCH 20/33] drm/amd/display: Optimize bandwidth on following fast update
Mikita Lipski
mikita.lipski at amd.com
Wed Sep 8 14:54:11 UTC 2021
From: Nicholas Kazlauskas <nicholas.kazlauskas at amd.com>
[Why]
The current call to optimize_bandwidth never occurs because flip is
always pending from the FULL and FAST updates.
[How]
Optimize on the following flip when it's a FAST update and we know we
aren't going to be modifying the clocks again.
Reviewed-by: Bhawanpreet Lakha <bhawanpreet.lakha at amd.com>
Acked-by: Mikita Lipski <mikita.lipski at amd.com>
Signed-off-by: Nicholas Kazlauskas <nicholas.kazlauskas at amd.com>
---
drivers/gpu/drm/amd/display/dc/core/dc.c | 17 ++++++++---------
1 file changed, 8 insertions(+), 9 deletions(-)
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index f58d3956f3e2..822239b59a78 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -1801,6 +1801,11 @@ void dc_post_update_surfaces_to_stream(struct dc *dc)
post_surface_trace(dc);
+ if (dc->ctx->dce_version >= DCE_VERSION_MAX)
+ TRACE_DCN_CLOCK_STATE(&context->bw_ctx.bw.dcn.clk);
+ else
+ TRACE_DCE_CLOCK_STATE(&context->bw_ctx.bw.dce);
+
if (is_flip_pending_in_pipes(dc, context))
return;
@@ -2986,6 +2991,9 @@ void dc_commit_updates_for_stream(struct dc *dc,
if (new_pipe->plane_state && new_pipe->plane_state != old_pipe->plane_state)
new_pipe->plane_state->force_full_update = true;
}
+ } else if (update_type == UPDATE_TYPE_FAST) {
+ /* Previous frame finished and HW is ready for optimization. */
+ dc_post_update_surfaces_to_stream(dc);
}
@@ -3042,15 +3050,6 @@ void dc_commit_updates_for_stream(struct dc *dc,
pipe_ctx->plane_state->force_full_update = false;
}
}
- /*let's use current_state to update watermark etc*/
- if (update_type >= UPDATE_TYPE_FULL) {
- dc_post_update_surfaces_to_stream(dc);
-
- if (dc_ctx->dce_version >= DCE_VERSION_MAX)
- TRACE_DCN_CLOCK_STATE(&context->bw_ctx.bw.dcn.clk);
- else
- TRACE_DCE_CLOCK_STATE(&context->bw_ctx.bw.dce);
- }
return;
--
2.25.1
More information about the amd-gfx
mailing list