[PATCH 08/14] drm/amd/display: Use correct helpers to compute timeslots

mikita.lipski at amd.com mikita.lipski at amd.com
Tue Oct 1 16:17:15 UTC 2019


From: David Francis <David.Francis at amd.com>

We were using drm helpers to convert a timing into its
bandwidth, its bandwidth into pbn, and its pbn into timeslots

These helpers
-Did not take DSC timings into account
-Used the link rate and lane count of the link's aux device,
which are not the same as the link's current cap
-Did not take FEC into account (FEC reduces the PBN per timeslot)

For converting timing into PBN, use the new function
drm_dp_calc_pbn_mode_dsc that handles the DSC case

For converting PBN into time slots, amdgpu doesn't use the
'correct' atomic method (drm_dp_atomic_find_vcpi_slots), so
don't add a new helper to cover our approach. Use the same
means of calculating pbn per time slot as the DSC code.

Cc: Jerry Zuo <Jerry.Zuo at amd.com>
Cc: Nicholas Kazlauskas <nicholas.kazlauskas at amd.com>
Signed-off-by: David Francis <David.Francis at amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c
index 5256abe32e92..fc537ae25eeb 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c
@@ -185,6 +185,7 @@ bool dm_helpers_dp_mst_write_payload_allocation_table(
 	struct drm_dp_mst_topology_mgr *mst_mgr;
 	struct drm_dp_mst_port *mst_port;
 	bool ret;
+	int pbn_per_timeslot = 0;
 
 	aconnector = (struct amdgpu_dm_connector *)stream->dm_stream_context;
 
@@ -200,6 +201,11 @@ bool dm_helpers_dp_mst_write_payload_allocation_table(
 
 	if (enable) {
 
+		/* Convert kilobits per second / 64 (for 64 timeslots) to pbn (54/64 megabytes per second) */
+		pbn_per_timeslot = dc_link_bandwidth_kbps(
+				stream->link, dc_link_get_link_cap(stream->link)) / (8 * 1000 * 54);
+		aconnector->vcpi_slots = DIV_ROUND_UP(aconnector->pbn, pbn_per_timeslot);
+
 		ret = drm_dp_mst_allocate_vcpi(mst_mgr, mst_port,
 					       aconnector->pbn,
 					       aconnector->vcpi_slots);
-- 
2.17.1



More information about the amd-gfx mailing list