drm/amd/display: Enforce minimum prefetch time for low memclk on DCN32

[WHY?]
Data return times when using lowest memclk can be <= 60us, which can cause
underflow on high bandwidth displays with a workload.

[HOW?]
Enforce a minimum prefetch time during validation for low memclk modes.
Reviewed-by: Jun Lei <Jun.Lei@amd.com>
Acked-by: Alan Liu <HaoPing.Liu@amd.com>
Signed-off-by: Dillon Varone <Dillon.Varone@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drivers/gpu/drm/amd/display/dc/dc.h

@@ -853,6 +853,7 @@ struct dc_debug_options {
 	bool enable_dp_dig_pixel_rate_div_policy;
 	enum lttpr_mode lttpr_mode_override;
 	unsigned int dsc_delay_factor_wa_x1000;
+	unsigned int min_prefetch_in_strobe_ns;
 };
 
 struct gpu_info_soc_bounding_box_v1_0;

drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c

@@ -724,6 +724,7 @@ static const struct dc_debug_options debug_defaults_drv = {
 	.enable_dp_dig_pixel_rate_div_policy = 1,
 	.allow_sw_cursor_fallback = false,
 	.alloc_extra_way_for_cursor = true,
+	.min_prefetch_in_strobe_ns = 60000, // 60us
 };
 
 static const struct dc_debug_options debug_defaults_diags = {

drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c

@@ -722,6 +722,7 @@ static const struct dc_debug_options debug_defaults_drv = {
 	.enable_dp_dig_pixel_rate_div_policy = 1,
 	.allow_sw_cursor_fallback = false,
 	.alloc_extra_way_for_cursor = true,
+	.min_prefetch_in_strobe_ns = 60000, // 60us
 };
 
 static const struct dc_debug_options debug_defaults_diags = {

drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c

@@ -2364,6 +2364,8 @@ void dcn32_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_pa
 	/* DML DSC delay factor workaround */
 	dcn3_2_ip.dsc_delay_factor_wa = dc->debug.dsc_delay_factor_wa_x1000 / 1000.0;
 
+	dcn3_2_ip.min_prefetch_in_strobe_us = dc->debug.min_prefetch_in_strobe_ns / 1000.0;
+
 	/* Override dispclk_dppclk_vco_speed_mhz from Clk Mgr */
 	dcn3_2_soc.dispclk_dppclk_vco_speed_mhz = dc->clk_mgr->dentist_vco_freq_khz / 1000.0;
 	dc->dml.soc.dispclk_dppclk_vco_speed_mhz = dc->clk_mgr->dentist_vco_freq_khz / 1000.0;

drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.c

@@ -786,6 +786,8 @@ static void DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndPerforman
 					v->SwathHeightY[k],
 					v->SwathHeightC[k],
 					TWait,
+					v->DRAMSpeedPerState[mode_lib->vba.VoltageLevel] <= MEM_STROBE_FREQ_MHZ ?
+							mode_lib->vba.ip.min_prefetch_in_strobe_us : 0,
 					/* Output */
 					&v->DSTXAfterScaler[k],
 					&v->DSTYAfterScaler[k],
@@ -3245,6 +3247,8 @@ void dml32_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l
 							v->swath_width_chroma_ub_this_state[k],
 							v->SwathHeightYThisState[k],
 							v->SwathHeightCThisState[k], v->TWait,
+							v->DRAMSpeedPerState[i] <= MEM_STROBE_FREQ_MHZ ?
+									mode_lib->vba.ip.min_prefetch_in_strobe_us : 0,
 
 							/* Output */
 							&v->dummy_vars.dml32_ModeSupportAndSystemConfigurationFull.DSTXAfterScaler[k],

drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.h

@@ -49,6 +49,9 @@
 #define BPP_INVALID 0
 #define BPP_BLENDED_PIPE 0xffffffff
 
+#define MEM_STROBE_FREQ_MHZ 1600
+#define MEM_STROBE_MAX_DELIVERY_TIME_US 60.0
+
 struct display_mode_lib;
 
 void dml32_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_lib);

drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c

@@ -3417,6 +3417,7 @@ bool dml32_CalculatePrefetchSchedule(
 		unsigned int SwathHeightY,
 		unsigned int SwathHeightC,
 		double TWait,
+		double TPreReq,
 		/* Output */
 		double   *DSTXAfterScaler,
 		double   *DSTYAfterScaler,
@@ -3474,6 +3475,7 @@ bool dml32_CalculatePrefetchSchedule(
 	double  min_Lsw;
 	double  Tsw_est1 = 0;
 	double  Tsw_est3 = 0;
+	double  TPreMargin = 0;
 
 	if (v->GPUVMEnable == true && v->HostVMEnable == true)
 		HostVMDynamicLevelsTrips = v->HostVMMaxNonCachedPageTableLevels;
@@ -3699,6 +3701,8 @@ bool dml32_CalculatePrefetchSchedule(
 
 	dst_y_prefetch_equ = dml_floor(4.0 * (dst_y_prefetch_equ + 0.125), 1) / 4.0;
 	Tpre_rounded = dst_y_prefetch_equ * LineTime;
+
+	TPreMargin = Tpre_rounded - TPreReq;
 #ifdef __DML_VBA_DEBUG__
 	dml_print("DML::%s: dst_y_prefetch_equ: %f (after round)\n", __func__, dst_y_prefetch_equ);
 	dml_print("DML::%s: LineTime: %f\n", __func__, LineTime);
@@ -3726,7 +3730,7 @@ bool dml32_CalculatePrefetchSchedule(
 	*VRatioPrefetchY = 0;
 	*VRatioPrefetchC = 0;
 	*RequiredPrefetchPixDataBWLuma = 0;
-	if (dst_y_prefetch_equ > 1) {
+	if (dst_y_prefetch_equ > 1 && TPreMargin > 0.0) {
 		double PrefetchBandwidth1;
 		double PrefetchBandwidth2;
 		double PrefetchBandwidth3;
@@ -3872,7 +3876,11 @@ bool dml32_CalculatePrefetchSchedule(
 		}
 
 		if (dst_y_prefetch_oto < dst_y_prefetch_equ) {
+			if (dst_y_prefetch_oto * LineTime < TPreReq) {
+				*DestinationLinesForPrefetch = dst_y_prefetch_equ;
+			} else {
 				*DestinationLinesForPrefetch = dst_y_prefetch_oto;
+			}
 			TimeForFetchingMetaPTE = Tvm_oto;
 			TimeForFetchingRowInVBlank = Tr0_oto;
 			*PrefetchBandwidth = prefetch_bw_oto;

drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.h

@@ -743,6 +743,7 @@ bool dml32_CalculatePrefetchSchedule(
 		unsigned int SwathHeightY,
 		unsigned int SwathHeightC,
 		double TWait,
+		double TPreReq,
 		/* Output */
 		double   *DSTXAfterScaler,
 		double   *DSTYAfterScaler,

drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c

@@ -544,6 +544,8 @@ void dcn321_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_p
 	/* DML DSC delay factor workaround */
 	dcn3_21_ip.dsc_delay_factor_wa = dc->debug.dsc_delay_factor_wa_x1000 / 1000.0;
 
+	dcn3_21_ip.min_prefetch_in_strobe_us = dc->debug.min_prefetch_in_strobe_ns / 1000.0;
+
 	/* Override dispclk_dppclk_vco_speed_mhz from Clk Mgr */
 	dcn3_21_soc.dispclk_dppclk_vco_speed_mhz = dc->clk_mgr->dentist_vco_freq_khz / 1000.0;
 	dc->dml.soc.dispclk_dppclk_vco_speed_mhz = dc->clk_mgr->dentist_vco_freq_khz / 1000.0;

drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h

@@ -367,6 +367,7 @@ struct _vcs_dpi_ip_params_st {
 
 	/* DM workarounds */
 	double dsc_delay_factor_wa; // TODO: Remove after implementing root cause fix
+	double min_prefetch_in_strobe_us;
 };
 
 struct _vcs_dpi_display_xfc_params_st {