diff options
Diffstat (limited to 'drivers/gpu/drm/amd/display/dc/dml')
25 files changed, 550 insertions, 117 deletions
diff --git a/drivers/gpu/drm/amd/display/dc/dml/Makefile b/drivers/gpu/drm/amd/display/dc/dml/Makefile index ca7d24000621..0ecea87cf48f 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dml/Makefile @@ -33,6 +33,10 @@ ifdef CONFIG_PPC64 dml_ccflags := -mhard-float -maltivec endif +ifdef CONFIG_ARM64 +dml_rcflags := -mgeneral-regs-only +endif + ifdef CONFIG_CC_IS_GCC ifneq ($(call gcc-min-version, 70100),y) IS_OLD_GCC = 1 @@ -55,8 +59,6 @@ frame_warn_flag := -Wframe-larger-than=2048 endif CFLAGS_$(AMDDALPATH)/dc/dml/display_mode_lib.o := $(dml_ccflags) - -ifdef CONFIG_DRM_AMD_DC_DCN CFLAGS_$(AMDDALPATH)/dc/dml/display_mode_vba.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/dcn10/dcn10_fpu.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/dcn20/dcn20_fpu.o := $(dml_ccflags) @@ -88,7 +90,6 @@ CFLAGS_$(AMDDALPATH)/dc/dml/calcs/dcn_calcs.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/calcs/dcn_calc_auto.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/calcs/dcn_calc_math.o := $(dml_ccflags) -Wno-tautological-compare CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/display_mode_vba.o := $(dml_rcflags) -CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn2x/dcn2x.o := $(dml_rcflags) CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn20/display_mode_vba_20.o := $(dml_rcflags) CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn20/display_rq_dlg_calc_20.o := $(dml_rcflags) CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn20/display_mode_vba_20v2.o := $(dml_rcflags) @@ -105,7 +106,18 @@ CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn32/display_mode_vba_util_32.o := $(dml_rcf CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn301/dcn301_fpu.o := $(dml_rcflags) CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/display_mode_lib.o := $(dml_rcflags) CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dsc/rc_calc_fpu.o := $(dml_rcflags) -endif +CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn10/dcn10_fpu.o := $(dml_rcflags) +CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn20/dcn20_fpu.o := $(dml_rcflags) +CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn314/display_mode_vba_314.o := $(dml_rcflags) +CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn314/display_rq_dlg_calc_314.o := $(dml_rcflags) +CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn314/dcn314_fpu.o := $(dml_rcflags) +CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn30/dcn30_fpu.o := $(dml_rcflags) +CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn32/dcn32_fpu.o := $(dml_rcflags) +CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn321/dcn321_fpu.o := $(dml_rcflags) +CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn31/dcn31_fpu.o := $(dml_rcflags) +CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn302/dcn302_fpu.o := $(dml_rcflags) +CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn303/dcn303_fpu.o := $(dml_rcflags) +CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/calcs/dcn_calc_math.o := $(dml_rcflags) CFLAGS_$(AMDDALPATH)/dc/dml/dml1_display_rq_dlg_calc.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/display_rq_dlg_helpers.o := $(dml_ccflags) CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dml1_display_rq_dlg_calc.o := $(dml_rcflags) diff --git a/drivers/gpu/drm/amd/display/dc/dml/dc_features.h b/drivers/gpu/drm/amd/display/dc/dml/dc_features.h index 74e86732e301..2cbdd75429ff 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dc_features.h +++ b/drivers/gpu/drm/amd/display/dc/dml/dc_features.h @@ -29,6 +29,13 @@ #define DC__PRESENT 1 #define DC__PRESENT__1 1 #define DC__NUM_DPP 4 + +/** + * @DC__VOLTAGE_STATES: + * + * Define the maximum amount of states supported by the ASIC. Every ASIC has a + * specific number of states; this macro defines the maximum number of states. + */ #define DC__VOLTAGE_STATES 20 #define DC__NUM_DPP__4 1 #define DC__NUM_DPP__0_PRESENT 1 diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn10/dcn10_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn10/dcn10_fpu.c index 99644d896222..c5e84190c17a 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn10/dcn10_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn10/dcn10_fpu.c @@ -27,6 +27,8 @@ #include "dcn10/dcn10_resource.h" #include "dcn10_fpu.h" +#include "resource.h" +#include "amdgpu_dm/dc_fpu.h" /** * DOC: DCN10 FPU manipulation Overview @@ -121,3 +123,37 @@ struct _vcs_dpi_soc_bounding_box_st dcn1_0_soc = { .writeback_dram_clock_change_latency_us = 23.0, .return_bus_width_bytes = 64, }; + +void dcn10_resource_construct_fp(struct dc *dc) +{ + dc_assert_fp_enabled(); + if (dc->ctx->dce_version == DCN_VERSION_1_01) { + struct dcn_soc_bounding_box *dcn_soc = dc->dcn_soc; + struct dcn_ip_params *dcn_ip = dc->dcn_ip; + struct display_mode_lib *dml = &dc->dml; + + dml->ip.max_num_dpp = 3; + /* TODO how to handle 23.84? */ + dcn_soc->dram_clock_change_latency = 23; + dcn_ip->max_num_dpp = 3; + } + if (ASICREV_IS_RV1_F0(dc->ctx->asic_id.hw_internal_rev)) { + dc->dcn_soc->urgent_latency = 3; + dc->debug.disable_dmcu = true; + dc->dcn_soc->fabric_and_dram_bandwidth_vmax0p9 = 41.60f; + } + + dc->dcn_soc->number_of_channels = dc->ctx->asic_id.vram_width / ddr4_dram_width; + ASSERT(dc->dcn_soc->number_of_channels < 3); + if (dc->dcn_soc->number_of_channels == 0)/*old sbios bug*/ + dc->dcn_soc->number_of_channels = 2; + + if (dc->dcn_soc->number_of_channels == 1) { + dc->dcn_soc->fabric_and_dram_bandwidth_vmax0p9 = 19.2f; + dc->dcn_soc->fabric_and_dram_bandwidth_vnom0p8 = 17.066f; + dc->dcn_soc->fabric_and_dram_bandwidth_vmid0p72 = 14.933f; + dc->dcn_soc->fabric_and_dram_bandwidth_vmin0p65 = 12.8f; + if (ASICREV_IS_RV1_F0(dc->ctx->asic_id.hw_internal_rev)) + dc->dcn_soc->fabric_and_dram_bandwidth_vmax0p9 = 20.80f; + } +} diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn10/dcn10_fpu.h b/drivers/gpu/drm/amd/display/dc/dml/dcn10/dcn10_fpu.h index e74ed4b4ce5b..63219ecd8478 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn10/dcn10_fpu.h +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn10/dcn10_fpu.h @@ -27,4 +27,6 @@ #ifndef __DCN10_FPU_H__ #define __DCN10_FPU_H__ +void dcn10_resource_construct_fp(struct dc *dc); + #endif /* __DCN20_FPU_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c index d680f1c5b69f..c26da3bb2892 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c @@ -565,7 +565,7 @@ struct _vcs_dpi_soc_bounding_box_st dcn2_1_soc = { .dppclk_mhz = 847.06, .phyclk_mhz = 810.0, .socclk_mhz = 953.0, - .dscclk_mhz = 489.0, + .dscclk_mhz = 300.0, .dram_speed_mts = 2400.0, }, { @@ -576,7 +576,7 @@ struct _vcs_dpi_soc_bounding_box_st dcn2_1_soc = { .dppclk_mhz = 960.00, .phyclk_mhz = 810.0, .socclk_mhz = 278.0, - .dscclk_mhz = 287.67, + .dscclk_mhz = 342.86, .dram_speed_mts = 2666.0, }, { @@ -587,7 +587,7 @@ struct _vcs_dpi_soc_bounding_box_st dcn2_1_soc = { .dppclk_mhz = 1028.57, .phyclk_mhz = 810.0, .socclk_mhz = 715.0, - .dscclk_mhz = 318.334, + .dscclk_mhz = 369.23, .dram_speed_mts = 3200.0, }, { @@ -949,6 +949,7 @@ static enum dcn_zstate_support_state decide_zstate_support(struct dc *dc, struc int plane_count; int i; unsigned int optimized_min_dst_y_next_start_us; + bool allow_z8 = context->bw_ctx.dml.vba.StutterPeriod > 1000.0; plane_count = 0; optimized_min_dst_y_next_start_us = 0; @@ -963,6 +964,8 @@ static enum dcn_zstate_support_state decide_zstate_support(struct dc *dc, struc * 2. single eDP, on link 0, 1 plane and stutter period > 5ms * Z10 only cases: * 1. single eDP, on link 0, 1 plane and stutter period >= 5ms + * Z8 cases: + * 1. stutter period sufficient * Zstate not allowed cases: * 1. Everything else */ @@ -989,12 +992,15 @@ static enum dcn_zstate_support_state decide_zstate_support(struct dc *dc, struc if (context->bw_ctx.dml.vba.StutterPeriod > 5000.0 || optimized_min_dst_y_next_start_us > 5000) return DCN_ZSTATE_SUPPORT_ALLOW; - else if (link->psr_settings.psr_version == DC_PSR_VERSION_1 && !dc->debug.disable_psr) - return DCN_ZSTATE_SUPPORT_ALLOW_Z10_ONLY; + else if (link->psr_settings.psr_version == DC_PSR_VERSION_1 && !link->panel_config.psr.disable_psr) + return allow_z8 ? DCN_ZSTATE_SUPPORT_ALLOW_Z8_Z10_ONLY : DCN_ZSTATE_SUPPORT_ALLOW_Z10_ONLY; else - return DCN_ZSTATE_SUPPORT_DISALLOW; - } else + return allow_z8 ? DCN_ZSTATE_SUPPORT_ALLOW_Z8_ONLY : DCN_ZSTATE_SUPPORT_DISALLOW; + } else if (allow_z8) { + return DCN_ZSTATE_SUPPORT_ALLOW_Z8_ONLY; + } else { return DCN_ZSTATE_SUPPORT_DISALLOW; + } } void dcn20_calculate_dlg_params( @@ -1228,6 +1234,7 @@ int dcn20_populate_dml_pipes_from_context( pipes[pipe_cnt].pipe.src.dcc = false; pipes[pipe_cnt].pipe.src.dcc_rate = 1; pipes[pipe_cnt].pipe.dest.synchronized_vblank_all_planes = synchronized_vblank; + pipes[pipe_cnt].pipe.dest.synchronize_timings = synchronized_vblank; pipes[pipe_cnt].pipe.dest.hblank_start = timing->h_total - timing->h_front_porch; pipes[pipe_cnt].pipe.dest.hblank_end = pipes[pipe_cnt].pipe.dest.hblank_start - timing->h_addressable @@ -1295,6 +1302,8 @@ int dcn20_populate_dml_pipes_from_context( case SIGNAL_TYPE_DISPLAY_PORT_MST: case SIGNAL_TYPE_DISPLAY_PORT: pipes[pipe_cnt].dout.output_type = dm_dp; + if (is_dp_128b_132b_signal(&res_ctx->pipe_ctx[i])) + pipes[pipe_cnt].dout.output_type = dm_dp2p0; break; case SIGNAL_TYPE_EDP: pipes[pipe_cnt].dout.output_type = dm_edp; diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn30/dcn30_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn30/dcn30_fpu.c index e1e92daba668..d4c0f9cdac8e 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn30/dcn30_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn30/dcn30_fpu.c @@ -520,9 +520,7 @@ void dcn30_fpu_calculate_wm_and_dlg( pipe_idx++; } - DC_FP_START(); dcn20_calculate_dlg_params(dc, context, pipes, pipe_cnt, vlevel); - DC_FP_END(); if (!pstate_en) /* Restore full p-state latency */ diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c b/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c index 479e2c1a1301..379729b02847 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c @@ -4851,7 +4851,7 @@ void dml30_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l v->SwathHeightYThisState[k], v->SwathHeightCThisState[k], v->HTotal[k] / v->PixelClock[k], - v->UrgentLatency, + v->UrgLatency[i], v->CursorBufferSize, v->CursorWidth[k][0], v->CursorBPP[k][0], diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.c index 7dd0845d1bd9..b37d14369a62 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.c @@ -483,7 +483,7 @@ void dcn31_calculate_wm_and_dlg_fp( int pipe_cnt, int vlevel) { - int i, pipe_idx, active_dpp_count = 0; + int i, pipe_idx, active_hubp_count = 0; double dcfclk = context->bw_ctx.dml.vba.DCFCLKState[vlevel][context->bw_ctx.dml.vba.maxMpcComb]; dc_assert_fp_enabled(); @@ -529,7 +529,7 @@ void dcn31_calculate_wm_and_dlg_fp( continue; if (context->res_ctx.pipe_ctx[i].plane_state) - active_dpp_count++; + active_hubp_count++; pipes[pipe_idx].clks_cfg.dispclk_mhz = get_dispclk_calculated(&context->bw_ctx.dml, pipes, pipe_cnt); pipes[pipe_idx].clks_cfg.dppclk_mhz = get_dppclk_calculated(&context->bw_ctx.dml, pipes, pipe_cnt, pipe_idx); @@ -547,9 +547,22 @@ void dcn31_calculate_wm_and_dlg_fp( } dcn20_calculate_dlg_params(dc, context, pipes, pipe_cnt, vlevel); - /* For 31x apu pstate change is only supported if possible in vactive or if there are no active dpps */ + /* For 31x apu pstate change is only supported if possible in vactive*/ context->bw_ctx.bw.dcn.clk.p_state_change_support = - context->bw_ctx.dml.vba.DRAMClockChangeSupport[vlevel][context->bw_ctx.dml.vba.maxMpcComb] == dm_dram_clock_change_vactive || !active_dpp_count; + context->bw_ctx.dml.vba.DRAMClockChangeSupport[vlevel][context->bw_ctx.dml.vba.maxMpcComb] == dm_dram_clock_change_vactive; + /* If DCN isn't making memory requests we can allow pstate change and lower clocks */ + if (!active_hubp_count) { + context->bw_ctx.bw.dcn.clk.socclk_khz = 0; + context->bw_ctx.bw.dcn.clk.dppclk_khz = 0; + context->bw_ctx.bw.dcn.clk.dcfclk_khz = 0; + context->bw_ctx.bw.dcn.clk.dcfclk_deep_sleep_khz = 0; + context->bw_ctx.bw.dcn.clk.dramclk_khz = 0; + context->bw_ctx.bw.dcn.clk.fclk_khz = 0; + context->bw_ctx.bw.dcn.clk.p_state_change_support = true; + for (i = 0; i < dc->res_pool->pipe_count; i++) + if (context->res_ctx.pipe_ctx[i].stream) + context->res_ctx.pipe_ctx[i].plane_res.bw.dppclk_khz = 0; + } } void dcn31_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params) @@ -797,3 +810,8 @@ void dcn316_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_param else dml_init_instance(&dc->dml, &dcn3_16_soc, &dcn3_16_ip, DML_PROJECT_DCN31_FPGA); } + +int dcn_get_max_non_odm_pix_rate_100hz(struct _vcs_dpi_soc_bounding_box_st *soc) +{ + return soc->clock_limits[0].dispclk_mhz * 10000.0 / (1.0 + soc->dcn_downspread_percent / 100.0); +} diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.h b/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.h index fd58b2561ec9..687d3522cc33 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.h +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.h @@ -46,5 +46,10 @@ void dcn31_calculate_wm_and_dlg_fp( void dcn31_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params); void dcn315_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params); void dcn316_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params); +int dcn_get_max_non_odm_pix_rate_100hz(struct _vcs_dpi_soc_bounding_box_st *soc); +int dcn31x_populate_dml_pipes_from_context(struct dc *dc, + struct dc_state *context, + display_e2e_pipe_params_st *pipes, + bool fast_validate); #endif /* __DCN31_FPU_H__*/ diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c b/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c index b612edb14417..ec351c8418cb 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c @@ -1052,17 +1052,16 @@ static bool CalculatePrefetchSchedule( else bytes_pp = myPipe->BytePerPixelY + myPipe->BytePerPixelC; /*rev 99*/ - prefetch_bw_pr = dml_min(1, bytes_pp * myPipe->PixelClock / (double) myPipe->DPPPerPlane); + prefetch_bw_pr = bytes_pp * myPipe->PixelClock / (double) myPipe->DPPPerPlane; + prefetch_bw_pr = dml_min(1, myPipe->VRatio) * prefetch_bw_pr; max_Tsw = dml_max(PrefetchSourceLinesY, PrefetchSourceLinesC) * LineTime; prefetch_sw_bytes = PrefetchSourceLinesY * swath_width_luma_ub * myPipe->BytePerPixelY + PrefetchSourceLinesC * swath_width_chroma_ub * myPipe->BytePerPixelC; - prefetch_bw_oto = dml_max(bytes_pp * myPipe->PixelClock / myPipe->DPPPerPlane, prefetch_sw_bytes / (dml_max(PrefetchSourceLinesY, PrefetchSourceLinesC) * LineTime)); prefetch_bw_oto = dml_max(prefetch_bw_pr, prefetch_sw_bytes / max_Tsw); min_Lsw = dml_max(1, dml_max(PrefetchSourceLinesY, PrefetchSourceLinesC) / max_vratio_pre); Lsw_oto = dml_ceil(4 * dml_max(prefetch_sw_bytes / prefetch_bw_oto / LineTime, min_Lsw), 1) / 4; Tsw_oto = Lsw_oto * LineTime; - prefetch_bw_oto = (PrefetchSourceLinesY * swath_width_luma_ub * myPipe->BytePerPixelY + PrefetchSourceLinesC * swath_width_chroma_ub * myPipe->BytePerPixelC) / Tsw_oto; #ifdef __DML_VBA_DEBUG__ dml_print("DML: HTotal: %d\n", myPipe->HTotal); @@ -5083,7 +5082,7 @@ void dml31_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l v->SwathHeightYThisState[k], v->SwathHeightCThisState[k], v->HTotal[k] / v->PixelClock[k], - v->UrgentLatency, + v->UrgLatency[i], v->CursorBufferSize, v->CursorWidth[k][0], v->CursorBPP[k][0], @@ -5361,6 +5360,58 @@ void dml31_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l v->ModeSupport[i][j] = true; } else { v->ModeSupport[i][j] = false; +#ifdef __DML_VBA_DEBUG__ + if (v->ScaleRatioAndTapsSupport == false) + dml_print("DML SUPPORT: ScaleRatioAndTapsSupport failed"); + if (v->SourceFormatPixelAndScanSupport == false) + dml_print("DML SUPPORT: SourceFormatPixelAndScanSupport failed"); + if (v->ViewportSizeSupport[i][j] == false) + dml_print("DML SUPPORT: ViewportSizeSupport failed"); + if (v->LinkCapacitySupport[i] == false) + dml_print("DML SUPPORT: LinkCapacitySupport failed"); + if (v->ODMCombine4To1SupportCheckOK[i] == false) + dml_print("DML SUPPORT: DSC422NativeNotSupported failed"); + if (v->NotEnoughDSCUnits[i] == true) + dml_print("DML SUPPORT: NotEnoughDSCUnits"); + if (v->DTBCLKRequiredMoreThanSupported[i] == true) + dml_print("DML SUPPORT: DTBCLKRequiredMoreThanSupported"); + if (v->ROBSupport[i][j] == false) + dml_print("DML SUPPORT: ROBSupport failed"); + if (v->DISPCLK_DPPCLK_Support[i][j] == false) + dml_print("DML SUPPORT: DISPCLK_DPPCLK_Support failed"); + if (v->TotalAvailablePipesSupport[i][j] == false) + dml_print("DML SUPPORT: DSC422NativeNotSupported failed"); + if (EnoughWritebackUnits == false) + dml_print("DML SUPPORT: DSC422NativeNotSupported failed"); + if (v->WritebackLatencySupport == false) + dml_print("DML SUPPORT: WritebackLatencySupport failed"); + if (v->WritebackScaleRatioAndTapsSupport == false) + dml_print("DML SUPPORT: DSC422NativeNotSupported "); + if (v->CursorSupport == false) + dml_print("DML SUPPORT: DSC422NativeNotSupported failed"); + if (v->PitchSupport == false) + dml_print("DML SUPPORT: PitchSupport failed"); + if (ViewportExceedsSurface == true) + dml_print("DML SUPPORT: ViewportExceedsSurface failed"); + if (v->PrefetchSupported[i][j] == false) + dml_print("DML SUPPORT: PrefetchSupported failed"); + if (v->DynamicMetadataSupported[i][j] == false) + dml_print("DML SUPPORT: DSC422NativeNotSupported failed"); + if (v->TotalVerticalActiveBandwidthSupport[i][j] == false) + dml_print("DML SUPPORT: TotalVerticalActiveBandwidthSupport failed"); + if (v->VRatioInPrefetchSupported[i][j] == false) + dml_print("DML SUPPORT: VRatioInPrefetchSupported failed"); + if (v->PTEBufferSizeNotExceeded[i][j] == false) + dml_print("DML SUPPORT: PTEBufferSizeNotExceeded failed"); + if (v->NonsupportedDSCInputBPC == true) + dml_print("DML SUPPORT: NonsupportedDSCInputBPC failed"); + if (!((v->HostVMEnable == false + && v->ImmediateFlipRequirement[0] != dm_immediate_flip_required) + || v->ImmediateFlipSupportedForState[i][j] == true)) + dml_print("DML SUPPORT: ImmediateFlipRequirement failed"); + if (FMTBufferExceeded == true) + dml_print("DML SUPPORT: FMTBufferExceeded failed"); +#endif } } } diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c index cf420ad2b8dc..6a1cf6adea77 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c @@ -29,6 +29,7 @@ #include "dcn31/dcn31_hubbub.h" #include "dcn314_fpu.h" #include "dml/dcn20/dcn20_fpu.h" +#include "dml/dcn31/dcn31_fpu.h" #include "dml/display_mode_vba.h" struct _vcs_dpi_ip_params_st dcn3_14_ip = { @@ -146,10 +147,10 @@ struct _vcs_dpi_soc_bounding_box_st dcn3_14_soc = { }, }, .num_states = 5, - .sr_exit_time_us = 9.0, - .sr_enter_plus_exit_time_us = 11.0, - .sr_exit_z8_time_us = 442.0, - .sr_enter_plus_exit_z8_time_us = 560.0, + .sr_exit_time_us = 16.5, + .sr_enter_plus_exit_time_us = 18.5, + .sr_exit_z8_time_us = 280.0, + .sr_enter_plus_exit_z8_time_us = 350.0, .writeback_latency_us = 12.0, .dram_channel_width_bytes = 4, .round_trip_ping_latency_dcfclk_cycles = 106, @@ -264,11 +265,8 @@ void dcn314_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_p dc->dml.soc.dispclk_dppclk_vco_speed_mhz = max_dispclk_mhz * 2; } - if ((int)(dcn3_14_soc.dram_clock_change_latency_us * 1000) - != dc->debug.dram_clock_change_latency_ns - && dc->debug.dram_clock_change_latency_ns) { - dcn3_14_soc.dram_clock_change_latency_us = dc->debug.dram_clock_change_latency_ns / 1000; - } + dcn20_patch_bounding_box(dc, &dcn3_14_soc); + if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) dml_init_instance(&dc->dml, &dcn3_14_soc, &dcn3_14_ip, DML_PROJECT_DCN314); else @@ -291,7 +289,7 @@ int dcn314_populate_dml_pipes_from_context_fpu(struct dc *dc, struct dc_state *c dc_assert_fp_enabled(); - dcn20_populate_dml_pipes_from_context(dc, context, pipes, fast_validate); + dcn31x_populate_dml_pipes_from_context(dc, context, pipes, fast_validate); for (i = 0, pipe_cnt = 0; i < dc->res_pool->pipe_count; i++) { struct dc_crtc_timing *timing; @@ -318,8 +316,6 @@ int dcn314_populate_dml_pipes_from_context_fpu(struct dc *dc, struct dc_state *c pipes[pipe_cnt].pipe.src.immediate_flip = true; pipes[pipe_cnt].pipe.src.unbounded_req_mode = false; - pipes[pipe_cnt].pipe.src.hostvm = dc->res_pool->hubbub->riommu_active; - pipes[pipe_cnt].pipe.src.gpuvm = true; pipes[pipe_cnt].pipe.src.dcc_fraction_of_zs_req_luma = 0; pipes[pipe_cnt].pipe.src.dcc_fraction_of_zs_req_chroma = 0; pipes[pipe_cnt].pipe.dest.vfront_porch = timing->v_front_porch; diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_mode_vba_314.c b/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_mode_vba_314.c index 0d12fd079cd6..950669f2c10d 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_mode_vba_314.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_mode_vba_314.c @@ -1074,17 +1074,16 @@ static bool CalculatePrefetchSchedule( else bytes_pp = myPipe->BytePerPixelY + myPipe->BytePerPixelC; /*rev 99*/ - prefetch_bw_pr = dml_min(1, bytes_pp * myPipe->PixelClock / (double) myPipe->DPPPerPlane); + prefetch_bw_pr = bytes_pp * myPipe->PixelClock / (double) myPipe->DPPPerPlane; + prefetch_bw_pr = dml_min(1, myPipe->VRatio) * prefetch_bw_pr; max_Tsw = dml_max(PrefetchSourceLinesY, PrefetchSourceLinesC) * LineTime; prefetch_sw_bytes = PrefetchSourceLinesY * swath_width_luma_ub * myPipe->BytePerPixelY + PrefetchSourceLinesC * swath_width_chroma_ub * myPipe->BytePerPixelC; - prefetch_bw_oto = dml_max(bytes_pp * myPipe->PixelClock / myPipe->DPPPerPlane, prefetch_sw_bytes / (dml_max(PrefetchSourceLinesY, PrefetchSourceLinesC) * LineTime)); prefetch_bw_oto = dml_max(prefetch_bw_pr, prefetch_sw_bytes / max_Tsw); min_Lsw = dml_max(1, dml_max(PrefetchSourceLinesY, PrefetchSourceLinesC) / max_vratio_pre); Lsw_oto = dml_ceil(4 * dml_max(prefetch_sw_bytes / prefetch_bw_oto / LineTime, min_Lsw), 1) / 4; Tsw_oto = Lsw_oto * LineTime; - prefetch_bw_oto = (PrefetchSourceLinesY * swath_width_luma_ub * myPipe->BytePerPixelY + PrefetchSourceLinesC * swath_width_chroma_ub * myPipe->BytePerPixelC) / Tsw_oto; #ifdef __DML_VBA_DEBUG__ dml_print("DML: HTotal: %d\n", myPipe->HTotal); @@ -5180,7 +5179,7 @@ void dml314_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_ v->SwathHeightYThisState[k], v->SwathHeightCThisState[k], v->HTotal[k] / v->PixelClock[k], - v->UrgentLatency, + v->UrgLatency[i], v->CursorBufferSize, v->CursorWidth[k][0], v->CursorBPP[k][0], diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c index 819de0f11012..f94abd124021 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c @@ -157,7 +157,7 @@ struct _vcs_dpi_soc_bounding_box_st dcn3_2_soc = { .dispclk_dppclk_vco_speed_mhz = 4300.0, .do_urgent_latency_adjustment = true, .urgent_latency_adjustment_fabric_clock_component_us = 1.0, - .urgent_latency_adjustment_fabric_clock_reference_mhz = 1000, + .urgent_latency_adjustment_fabric_clock_reference_mhz = 3000, }; void dcn32_build_wm_range_table_fpu(struct clk_mgr_internal *clk_mgr) @@ -211,7 +211,7 @@ void dcn32_build_wm_range_table_fpu(struct clk_mgr_internal *clk_mgr) /* 'DalDummyClockChangeLatencyNs' registry key option set to 0x7FFFFFFF can be used to disable Set C for dummy p-state */ if (clk_mgr->base.ctx->dc->bb_overrides.dummy_clock_change_latency_ns != 0x7FFFFFFF) { clk_mgr->base.bw_params->wm_table.nv_entries[WM_C].valid = true; - clk_mgr->base.bw_params->wm_table.nv_entries[WM_C].dml_input.pstate_latency_us = 38; + clk_mgr->base.bw_params->wm_table.nv_entries[WM_C].dml_input.pstate_latency_us = 50; clk_mgr->base.bw_params->wm_table.nv_entries[WM_C].dml_input.fclk_change_latency_us = fclk_change_latency_us; clk_mgr->base.bw_params->wm_table.nv_entries[WM_C].dml_input.sr_exit_time_us = sr_exit_time_us; clk_mgr->base.bw_params->wm_table.nv_entries[WM_C].dml_input.sr_enter_plus_exit_time_us = sr_enter_plus_exit_time_us; @@ -221,7 +221,7 @@ void dcn32_build_wm_range_table_fpu(struct clk_mgr_internal *clk_mgr) clk_mgr->base.bw_params->wm_table.nv_entries[WM_C].pmfw_breakdown.min_uclk = min_uclk_mhz; clk_mgr->base.bw_params->wm_table.nv_entries[WM_C].pmfw_breakdown.max_uclk = 0xFFFF; clk_mgr->base.bw_params->dummy_pstate_table[0].dram_speed_mts = clk_mgr->base.bw_params->clk_table.entries[0].memclk_mhz * 16; - clk_mgr->base.bw_params->dummy_pstate_table[0].dummy_pstate_latency_us = 38; + clk_mgr->base.bw_params->dummy_pstate_table[0].dummy_pstate_latency_us = 50; clk_mgr->base.bw_params->dummy_pstate_table[1].dram_speed_mts = clk_mgr->base.bw_params->clk_table.entries[1].memclk_mhz * 16; clk_mgr->base.bw_params->dummy_pstate_table[1].dummy_pstate_latency_us = 9; clk_mgr->base.bw_params->dummy_pstate_table[2].dram_speed_mts = clk_mgr->base.bw_params->clk_table.entries[2].memclk_mhz * 16; @@ -256,16 +256,24 @@ int dcn32_find_dummy_latency_index_for_fw_based_mclk_switch(struct dc *dc, int vlevel) { const int max_latency_table_entries = 4; - const struct vba_vars_st *vba = &context->bw_ctx.dml.vba; + struct vba_vars_st *vba = &context->bw_ctx.dml.vba; int dummy_latency_index = 0; + enum clock_change_support temp_clock_change_support = vba->DRAMClockChangeSupport[vlevel][context->bw_ctx.dml.vba.maxMpcComb]; dc_assert_fp_enabled(); while (dummy_latency_index < max_latency_table_entries) { + if (temp_clock_change_support != dm_dram_clock_change_unsupported) + vba->DRAMClockChangeSupport[vlevel][context->bw_ctx.dml.vba.maxMpcComb] = temp_clock_change_support; context->bw_ctx.dml.soc.dram_clock_change_latency_us = dc->clk_mgr->bw_params->dummy_pstate_table[dummy_latency_index].dummy_pstate_latency_us; dcn32_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, false); + /* for subvp + DRR case, if subvp pipes are still present we support pstate */ + if (vba->DRAMClockChangeSupport[vlevel][vba->maxMpcComb] == dm_dram_clock_change_unsupported && + dcn32_subvp_in_use(dc, context)) + vba->DRAMClockChangeSupport[vlevel][context->bw_ctx.dml.vba.maxMpcComb] = temp_clock_change_support; + if (vlevel < context->bw_ctx.dml.vba.soc.num_states && vba->DRAMClockChangeSupport[vlevel][vba->maxMpcComb] != dm_dram_clock_change_unsupported) break; @@ -531,9 +539,12 @@ void dcn32_set_phantom_stream_timing(struct dc *dc, unsigned int i, pipe_idx; struct pipe_ctx *pipe; uint32_t phantom_vactive, phantom_bp, pstate_width_fw_delay_lines; + unsigned int num_dpp; unsigned int vlevel = context->bw_ctx.dml.vba.VoltageLevel; unsigned int dcfclk = context->bw_ctx.dml.vba.DCFCLKState[vlevel][context->bw_ctx.dml.vba.maxMpcComb]; unsigned int socclk = context->bw_ctx.dml.vba.SOCCLKPerState[vlevel]; + struct vba_vars_st *vba = &context->bw_ctx.dml.vba; + struct dc_stream_state *main_stream = ref_pipe->stream; dc_assert_fp_enabled(); @@ -569,13 +580,26 @@ void dcn32_set_phantom_stream_timing(struct dc *dc, phantom_vactive = get_subviewport_lines_needed_in_mall(&context->bw_ctx.dml, pipes, pipe_cnt, pipe_idx) + pstate_width_fw_delay_lines + dc->caps.subvp_swath_height_margin_lines; + // W/A for DCC corruption with certain high resolution timings. + // Determing if pipesplit is used. If so, add meta_row_height to the phantom vactive. + num_dpp = vba->NoOfDPP[vba->VoltageLevel][vba->maxMpcComb][vba->pipe_plane[pipe_idx]]; + phantom_vactive += num_dpp > 1 ? vba->meta_row_height[vba->pipe_plane[pipe_idx]] : 0; + + /* dc->debug.subvp_extra_lines 0 by default*/ + phantom_vactive += dc->debug.subvp_extra_lines; + // For backporch of phantom pipe, use vstartup of the main pipe phantom_bp = get_vstartup(&context->bw_ctx.dml, pipes, pipe_cnt, pipe_idx); phantom_stream->dst.y = 0; phantom_stream->dst.height = phantom_vactive; + /* When scaling, DML provides the end to end required number of lines for MALL. + * dst.height is always correct for this case, but src.height is not which causes a + * delta between main and phantom pipe scaling outputs. Need to adjust src.height on + * phantom for this case. + */ phantom_stream->src.y = 0; - phantom_stream->src.height = phantom_vactive; + phantom_stream->src.height = (double)phantom_vactive * (double)main_stream->src.height / (double)main_stream->dst.height; phantom_stream->timing.v_addressable = phantom_vactive; phantom_stream->timing.v_front_porch = 1; @@ -1128,7 +1152,7 @@ static void dcn32_full_validate_bw_helper(struct dc *dc, context->bw_ctx.dml.soc.allow_for_pstate_or_stutter_in_vblank_final == dm_prefetch_support_uclk_fclk_and_stutter) { context->bw_ctx.dml.soc.allow_for_pstate_or_stutter_in_vblank_final = - dm_prefetch_support_stutter; + dm_prefetch_support_fclk_and_stutter; /* There are params (such as FabricClock) that need to be recalculated * after validation fails (otherwise it will be 0). Calculation for * phantom vactive requires call into DML, so we must ensure all the @@ -1179,7 +1203,7 @@ static void dcn32_full_validate_bw_helper(struct dc *dc, // If SubVP pipe config is unsupported (or cannot be used for UCLK switching) // remove phantom pipes and repopulate dml pipes if (!found_supported_config) { - dc->res_pool->funcs->remove_phantom_pipes(dc, context); + dc->res_pool->funcs->remove_phantom_pipes(dc, context, false); vba->DRAMClockChangeSupport[*vlevel][vba->maxMpcComb] = dm_dram_clock_change_unsupported; *pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, context, pipes, false); @@ -1191,9 +1215,7 @@ static void dcn32_full_validate_bw_helper(struct dc *dc, } } else { // Most populate phantom DLG params before programming hardware / timing for phantom pipe - DC_FP_START(); dcn32_helper_populate_phantom_dlg_params(dc, context, pipes, *pipe_cnt); - DC_FP_END(); /* Call validate_apply_pipe_split flags after calling DML getters for * phantom dlg params, or some of the VBA params indicating pipe split @@ -1230,7 +1252,7 @@ static void dcn32_calculate_dlg_params(struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, int pipe_cnt, int vlevel) { - int i, pipe_idx; + int i, pipe_idx, active_hubp_count = 0; bool usr_retraining_support = false; bool unbounded_req_enabled = false; @@ -1275,6 +1297,8 @@ static void dcn32_calculate_dlg_params(struct dc *dc, struct dc_state *context, for (i = 0, pipe_idx = 0; i < dc->res_pool->pipe_count; i++) { if (!context->res_ctx.pipe_ctx[i].stream) continue; + if (context->res_ctx.pipe_ctx[i].plane_state) + active_hubp_count++; pipes[pipe_idx].pipe.dest.vstartup_start = get_vstartup(&context->bw_ctx.dml, pipes, pipe_cnt, pipe_idx); pipes[pipe_idx].pipe.dest.vupdate_offset = get_vupdate_offset(&context->bw_ctx.dml, pipes, pipe_cnt, @@ -1296,10 +1320,23 @@ static void dcn32_calculate_dlg_params(struct dc *dc, struct dc_state *context, if (context->bw_ctx.bw.dcn.clk.dppclk_khz < pipes[pipe_idx].clks_cfg.dppclk_mhz * 1000) context->bw_ctx.bw.dcn.clk.dppclk_khz = pipes[pipe_idx].clks_cfg.dppclk_mhz * 1000; - context->res_ctx.pipe_ctx[i].plane_res.bw.dppclk_khz = pipes[pipe_idx].clks_cfg.dppclk_mhz * 1000; + if (context->res_ctx.pipe_ctx[i].plane_state) + context->res_ctx.pipe_ctx[i].plane_res.bw.dppclk_khz = pipes[pipe_idx].clks_cfg.dppclk_mhz * 1000; + else + context->res_ctx.pipe_ctx[i].plane_res.bw.dppclk_khz = 0; context->res_ctx.pipe_ctx[i].pipe_dlg_param = pipes[pipe_idx].pipe.dest; pipe_idx++; } + /* If DCN isn't making memory requests we can allow pstate change and lower clocks */ + if (!active_hubp_count) { + context->bw_ctx.bw.dcn.clk.socclk_khz = 0; + context->bw_ctx.bw.dcn.clk.dppclk_khz = 0; + context->bw_ctx.bw.dcn.clk.dcfclk_khz = 0; + context->bw_ctx.bw.dcn.clk.dcfclk_deep_sleep_khz = 0; + context->bw_ctx.bw.dcn.clk.dramclk_khz = 0; + context->bw_ctx.bw.dcn.clk.fclk_khz = 0; + context->bw_ctx.bw.dcn.clk.p_state_change_support = true; + } /*save a original dppclock copy*/ context->bw_ctx.bw.dcn.clk.bw_dppclk_khz = context->bw_ctx.bw.dcn.clk.dppclk_khz; context->bw_ctx.bw.dcn.clk.bw_dispclk_khz = context->bw_ctx.bw.dcn.clk.dispclk_khz; @@ -1481,7 +1518,7 @@ bool dcn32_internal_validate_bw(struct dc *dc, return false; // For each full update, remove all existing phantom pipes first - dc->res_pool->funcs->remove_phantom_pipes(dc, context); + dc->res_pool->funcs->remove_phantom_pipes(dc, context, fast_validate); dc->res_pool->funcs->update_soc_for_wm_a(dc, context); @@ -1494,11 +1531,8 @@ bool dcn32_internal_validate_bw(struct dc *dc, dml_log_pipe_params(&context->bw_ctx.dml, pipes, pipe_cnt); - if (!fast_validate) { - DC_FP_START(); + if (!fast_validate) dcn32_full_validate_bw_helper(dc, context, pipes, &vlevel, split, merge, &pipe_cnt); - DC_FP_END(); - } if (fast_validate || (dc->debug.dml_disallow_alternate_prefetch_modes && @@ -1734,6 +1768,10 @@ bool dcn32_internal_validate_bw(struct dc *dc, } if (repopulate_pipes) { + int flag_max_mpc_comb = vba->maxMpcComb; + int flag_vlevel = vlevel; + int i; + pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, context, pipes, fast_validate); /* repopulate_pipes = 1 means the pipes were either split or merged. In this case @@ -1741,10 +1779,28 @@ bool dcn32_internal_validate_bw(struct dc *dc, * ensure all the params are calculated correctly. We do not need to run the * pipe split check again after this call (pipes are already split / merged). * */ - if (!fast_validate) { - context->bw_ctx.dml.soc.allow_for_pstate_or_stutter_in_vblank_final = - dm_prefetch_support_uclk_fclk_and_stutter_if_possible; - vlevel = dml_get_voltage_level(&context->bw_ctx.dml, pipes, pipe_cnt); + context->bw_ctx.dml.soc.allow_for_pstate_or_stutter_in_vblank_final = + dm_prefetch_support_uclk_fclk_and_stutter_if_possible; + vlevel = dml_get_voltage_level(&context->bw_ctx.dml, pipes, pipe_cnt); + if (vlevel == context->bw_ctx.dml.soc.num_states) { + /* failed after DET size changes */ + goto validate_fail; + } else if (flag_max_mpc_comb == 0 && + flag_max_mpc_comb != context->bw_ctx.dml.vba.maxMpcComb) { + /* check the context constructed with pipe split flags is still valid*/ + bool flags_valid = false; + for (i = flag_vlevel; i < context->bw_ctx.dml.soc.num_states; i++) { + if (vba->ModeSupport[i][flag_max_mpc_comb]) { + vba->maxMpcComb = flag_max_mpc_comb; + vba->VoltageLevel = i; + vlevel = i; + flags_valid = true; + } + } + + /* this should never happen */ + if (!flags_valid) + goto validate_fail; } } *vlevel_out = vlevel; @@ -1775,14 +1831,38 @@ void dcn32_calculate_wm_and_dlg_fpu(struct dc *dc, struct dc_state *context, unsigned int dummy_latency_index = 0; int maxMpcComb = context->bw_ctx.dml.vba.maxMpcComb; unsigned int min_dram_speed_mts = context->bw_ctx.dml.vba.DRAMSpeed; + bool subvp_in_use = dcn32_subvp_in_use(dc, context); unsigned int min_dram_speed_mts_margin; + bool need_fclk_lat_as_dummy = false; + bool is_subvp_p_drr = true; dc_assert_fp_enabled(); - // Override DRAMClockChangeSupport for SubVP + DRR case where the DRR cannot switch without stretching it's VBLANK - if (!pstate_en && dcn32_subvp_in_use(dc, context)) { - context->bw_ctx.dml.vba.DRAMClockChangeSupport[vlevel][context->bw_ctx.dml.vba.maxMpcComb] = dm_dram_clock_change_vblank_w_mall_sub_vp; - pstate_en = true; + /* need to find dummy latency index for subvp */ + if (subvp_in_use) { + /* Override DRAMClockChangeSupport for SubVP + DRR case where the DRR cannot switch without stretching it's VBLANK */ + if (!pstate_en) { + context->bw_ctx.dml.vba.DRAMClockChangeSupport[vlevel][context->bw_ctx.dml.vba.maxMpcComb] = dm_dram_clock_change_vblank_w_mall_sub_vp; + pstate_en = true; + is_subvp_p_drr = true; + } + dummy_latency_index = dcn32_find_dummy_latency_index_for_fw_based_mclk_switch(dc, + context, pipes, pipe_cnt, vlevel); + + /* For DCN32/321 need to validate with fclk pstate change latency equal to dummy so prefetch is + * scheduled correctly to account for dummy pstate. + */ + if (context->bw_ctx.dml.soc.fclk_change_latency_us < dc->clk_mgr->bw_params->dummy_pstate_table[dummy_latency_index].dummy_pstate_latency_us) { + need_fclk_lat_as_dummy = true; + context->bw_ctx.dml.soc.fclk_change_latency_us = + dc->clk_mgr->bw_params->dummy_pstate_table[dummy_latency_index].dummy_pstate_latency_us; + } + context->bw_ctx.dml.soc.dram_clock_change_latency_us = + dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.pstate_latency_us; + dcn32_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, false); + if (is_subvp_p_drr) { + context->bw_ctx.dml.vba.DRAMClockChangeSupport[vlevel][context->bw_ctx.dml.vba.maxMpcComb] = dm_dram_clock_change_vblank_w_mall_sub_vp; + } } context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching = false; @@ -1803,6 +1883,14 @@ void dcn32_calculate_wm_and_dlg_fpu(struct dc *dc, struct dc_state *context, */ context->bw_ctx.dml.soc.dram_clock_change_latency_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.pstate_latency_us; + /* For DCN32/321 need to validate with fclk pstate change latency equal to dummy so + * prefetch is scheduled correctly to account for dummy pstate. + */ + if (context->bw_ctx.dml.soc.fclk_change_latency_us < dc->clk_mgr->bw_params->dummy_pstate_table[dummy_latency_index].dummy_pstate_latency_us) { + need_fclk_lat_as_dummy = true; + context->bw_ctx.dml.soc.fclk_change_latency_us = + dc->clk_mgr->bw_params->dummy_pstate_table[dummy_latency_index].dummy_pstate_latency_us; + } dcn32_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, false); maxMpcComb = context->bw_ctx.dml.vba.maxMpcComb; dcfclk_from_fw_based_mclk_switching = context->bw_ctx.dml.vba.DCFCLKState[vlevel][context->bw_ctx.dml.vba.maxMpcComb]; @@ -1904,13 +1992,13 @@ void dcn32_calculate_wm_and_dlg_fpu(struct dc *dc, struct dc_state *context, if (context->bw_ctx.dml.vba.DRAMClockChangeSupport[vlevel][maxMpcComb] == dm_dram_clock_change_unsupported) { - int min_dram_speed_mts_offset = dc->clk_mgr->bw_params->clk_table.num_entries - 1; + int min_dram_speed_mts_offset = dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_memclk_levels - 1; min_dram_speed_mts = dc->clk_mgr->bw_params->clk_table.entries[min_dram_speed_mts_offset].memclk_mhz * 16; } - if (!context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching) { + if (!context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching && !subvp_in_use) { /* find largest table entry that is lower than dram speed, * but lower than DPM0 still uses DPM0 */ @@ -1990,6 +2078,11 @@ void dcn32_calculate_wm_and_dlg_fpu(struct dc *dc, struct dc_state *context, context->perf_params.stutter_period_us = context->bw_ctx.dml.vba.StutterPeriod; + /* for proper prefetch calculations, if dummy lat > fclk lat, use fclk lat = dummy lat */ + if (need_fclk_lat_as_dummy) + context->bw_ctx.dml.soc.fclk_change_latency_us = + dc->clk_mgr->bw_params->dummy_pstate_table[dummy_latency_index].dummy_pstate_latency_us; + dcn32_calculate_dlg_params(dc, context, pipes, pipe_cnt, vlevel); if (!pstate_en) @@ -1997,8 +2090,14 @@ void dcn32_calculate_wm_and_dlg_fpu(struct dc *dc, struct dc_state *context, context->bw_ctx.dml.soc.dram_clock_change_latency_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.pstate_latency_us; - if (context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching) + if (context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching) { dcn30_setup_mclk_switch_using_fw_based_vblank_stretch(dc, context); + } + + /* revert fclk lat changes if required */ + if (need_fclk_lat_as_dummy) + context->bw_ctx.dml.soc.fclk_change_latency_us = + dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.fclk_change_latency_us; } static void dcn32_get_optimal_dcfclk_fclk_for_uclk(unsigned int uclk_mts, @@ -2145,9 +2244,7 @@ static int build_synthetic_soc_states(struct clk_bw_params *bw_params, entry.fabricclk_mhz = 0; entry.dram_speed_mts = 0; - DC_FP_START(); insert_entry_into_table_sorted(table, num_entries, &entry); - DC_FP_END(); } // Insert the max DCFCLK @@ -2155,9 +2252,7 @@ static int build_synthetic_soc_states(struct clk_bw_params *bw_params, entry.fabricclk_mhz = 0; entry.dram_speed_mts = 0; - DC_FP_START(); insert_entry_into_table_sorted(table, num_entries, &entry); - DC_FP_END(); // Insert the UCLK DPMS for (i = 0; i < num_uclk_dpms; i++) { @@ -2165,9 +2260,7 @@ static int build_synthetic_soc_states(struct clk_bw_params *bw_params, entry.fabricclk_mhz = 0; entry.dram_speed_mts = bw_params->clk_table.entries[i].memclk_mhz * 16; - DC_FP_START(); insert_entry_into_table_sorted(table, num_entries, &entry); - DC_FP_END(); } // If FCLK is coarse grained, insert individual DPMs. @@ -2177,9 +2270,7 @@ static int build_synthetic_soc_states(struct clk_bw_params *bw_params, entry.fabricclk_mhz = bw_params->clk_table.entries[i].fclk_mhz; entry.dram_speed_mts = 0; - DC_FP_START(); insert_entry_into_table_sorted(table, num_entries, &entry); - DC_FP_END(); } } // If FCLK fine grained, only insert max @@ -2188,9 +2279,7 @@ static int build_synthetic_soc_states(struct clk_bw_params *bw_params, entry.fabricclk_mhz = max_fclk_mhz; entry.dram_speed_mts = 0; - DC_FP_START(); insert_entry_into_table_sorted(table, num_entries, &entry); - DC_FP_END(); } // At this point, the table contains all "points of interest" based on @@ -2359,9 +2448,13 @@ void dcn32_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_pa if (dc->ctx->dc_bios->vram_info.dram_channel_width_bytes) dcn3_2_soc.dram_channel_width_bytes = dc->ctx->dc_bios->vram_info.dram_channel_width_bytes; - } + /* 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; @@ -2521,3 +2614,11 @@ void dcn32_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_pa } } +void dcn32_zero_pipe_dcc_fraction(display_e2e_pipe_params_st *pipes, + int pipe_cnt) +{ + dc_assert_fp_enabled(); + + pipes[pipe_cnt].pipe.src.dcc_fraction_of_zs_req_luma = 0; + pipes[pipe_cnt].pipe.src.dcc_fraction_of_zs_req_chroma = 0; +} diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.h b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.h index 3a3dc2ce4c73..ab010e7e840b 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.h +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.h @@ -73,4 +73,7 @@ int dcn32_find_dummy_latency_index_for_fw_based_mclk_switch(struct dc *dc, void dcn32_patch_dpm_table(struct clk_bw_params *bw_params); +void dcn32_zero_pipe_dcc_fraction(display_e2e_pipe_params_st *pipes, + int pipe_cnt); + #endif diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.c index 5b91660a6496..4b8f5fa0f0ad 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.c @@ -364,10 +364,11 @@ static void DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndPerforman for (k = 0; k < mode_lib->vba.NumberOfActiveSurfaces; ++k) { v->DSCDelay[k] = dml32_DSCDelayRequirement(mode_lib->vba.DSCEnabled[k], mode_lib->vba.ODMCombineEnabled[k], mode_lib->vba.DSCInputBitPerComponent[k], - mode_lib->vba.OutputBpp[k], mode_lib->vba.HActive[k], mode_lib->vba.HTotal[k], + mode_lib->vba.OutputBppPerState[mode_lib->vba.VoltageLevel][k], + mode_lib->vba.HActive[k], mode_lib->vba.HTotal[k], mode_lib->vba.NumberOfDSCSlices[k], mode_lib->vba.OutputFormat[k], mode_lib->vba.Output[k], mode_lib->vba.PixelClock[k], - mode_lib->vba.PixelClockBackEnd[k]); + mode_lib->vba.PixelClockBackEnd[k], mode_lib->vba.ip.dsc_delay_factor_wa); } for (k = 0; k < mode_lib->vba.NumberOfActiveSurfaces; ++k) @@ -669,6 +670,25 @@ static void DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndPerforman v->cursor_bw[k] = mode_lib->vba.NumberOfCursors[k] * mode_lib->vba.CursorWidth[k][0] * mode_lib->vba.CursorBPP[k][0] / 8 / (mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]) * mode_lib->vba.VRatio[k]; } + v->NotEnoughDETSwathFillLatencyHiding = dml32_CalculateDETSwathFillLatencyHiding( + mode_lib->vba.NumberOfActiveSurfaces, + mode_lib->vba.ReturnBW, + v->UrgentLatency, + mode_lib->vba.SwathHeightY, + mode_lib->vba.SwathHeightC, + v->swath_width_luma_ub, + v->swath_width_chroma_ub, + v->BytePerPixelDETY, + v->BytePerPixelDETC, + mode_lib->vba.DETBufferSizeY, + mode_lib->vba.DETBufferSizeC, + mode_lib->vba.DPPPerPlane, + mode_lib->vba.HTotal, + mode_lib->vba.PixelClock, + mode_lib->vba.VRatio, + mode_lib->vba.VRatioChroma, + mode_lib->vba.UsesMALLForPStateChange); + for (k = 0; k < mode_lib->vba.NumberOfActiveSurfaces; ++k) { v->MaxVStartupLines[k] = ((mode_lib->vba.Interlace[k] && !mode_lib->vba.ProgressiveToInterlaceUnitInOPP) ? @@ -717,6 +737,8 @@ static void DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndPerforman do { MaxTotalRDBandwidth = 0; + DestinationLineTimesForPrefetchLessThan2 = false; + VRatioPrefetchMoreThanMax = false; #ifdef __DML_VBA_DEBUG__ dml_print("DML::%s: Start loop: VStartup = %d\n", __func__, mode_lib->vba.VStartupLines); #endif @@ -785,6 +807,8 @@ 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], @@ -1627,7 +1651,7 @@ static void mode_support_configuration(struct vba_vars_st *v, && !mode_lib->vba.MSOOrODMSplitWithNonDPLink && !mode_lib->vba.NotEnoughLanesForMSO && mode_lib->vba.LinkCapacitySupport[i] == true && !mode_lib->vba.P2IWith420 - && !mode_lib->vba.DSCOnlyIfNecessaryWithBPP + //&& !mode_lib->vba.DSCOnlyIfNecessaryWithBPP && !mode_lib->vba.DSC422NativeNotSupported && !mode_lib->vba.MPCCombineMethodIncompatible && mode_lib->vba.ODMCombine2To1SupportCheckOK[i] == true @@ -1660,6 +1684,8 @@ static void mode_support_configuration(struct vba_vars_st *v, && mode_lib->vba.DCCMetaBufferSizeNotExceeded[i][j] == true && mode_lib->vba.NonsupportedDSCInputBPC == false && !mode_lib->vba.ExceededMALLSize + && (mode_lib->vba.NotEnoughDETSwathFillLatencyHidingPerState[i][j] == false + || i == v->soc.num_states - 1) && ((mode_lib->vba.HostVMEnable == false && !mode_lib->vba.ImmediateFlipRequiredFinal) || mode_lib->vba.ImmediateFlipSupportedForState[i][j]) @@ -2475,7 +2501,8 @@ void dml32_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l mode_lib->vba.OutputBppPerState[i][k], mode_lib->vba.HActive[k], mode_lib->vba.HTotal[k], mode_lib->vba.NumberOfDSCSlices[k], mode_lib->vba.OutputFormat[k], mode_lib->vba.Output[k], - mode_lib->vba.PixelClock[k], mode_lib->vba.PixelClockBackEnd[k]); + mode_lib->vba.PixelClock[k], mode_lib->vba.PixelClockBackEnd[k], + mode_lib->vba.ip.dsc_delay_factor_wa); } for (k = 0; k <= mode_lib->vba.NumberOfActiveSurfaces - 1; k++) { @@ -3152,6 +3179,25 @@ void dml32_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l mode_lib->vba.UrgentBurstFactorChroma, mode_lib->vba.UrgentBurstFactorCursor); + mode_lib->vba.NotEnoughDETSwathFillLatencyHidingPerState[i][j] = dml32_CalculateDETSwathFillLatencyHiding( + mode_lib->vba.NumberOfActiveSurfaces, + mode_lib->vba.ReturnBWPerState[i][j], + mode_lib->vba.UrgLatency[i], + mode_lib->vba.SwathHeightYThisState, + mode_lib->vba.SwathHeightCThisState, + mode_lib->vba.swath_width_luma_ub_this_state, + mode_lib->vba.swath_width_chroma_ub_this_state, + mode_lib->vba.BytePerPixelInDETY, + mode_lib->vba.BytePerPixelInDETC, + mode_lib->vba.DETBufferSizeYThisState, + mode_lib->vba.DETBufferSizeCThisState, + mode_lib->vba.NoOfDPPThisState, + mode_lib->vba.HTotal, + mode_lib->vba.PixelClock, + mode_lib->vba.VRatio, + mode_lib->vba.VRatioChroma, + mode_lib->vba.UsesMALLForPStateChange); + v->dummy_vars.dml32_ModeSupportAndSystemConfigurationFull.VMDataOnlyReturnBWPerState = dml32_get_return_bw_mbps_vm_only(&mode_lib->vba.soc, i, mode_lib->vba.DCFCLKState[i][j], mode_lib->vba.FabricClockPerState[i], mode_lib->vba.DRAMSpeedPerState[i]); @@ -3190,6 +3236,7 @@ void dml32_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l mode_lib->vba.FCLKChangeLatency, mode_lib->vba.UrgLatency[i], mode_lib->vba.SREnterPlusExitTime); + memset(&v->dummy_vars.dml32_ModeSupportAndSystemConfigurationFull, 0, sizeof(DmlPipe)); v->dummy_vars.dml32_ModeSupportAndSystemConfigurationFull.myPipe.Dppclk = mode_lib->vba.RequiredDPPCLK[i][j][k]; v->dummy_vars.dml32_ModeSupportAndSystemConfigurationFull.myPipe.Dispclk = mode_lib->vba.RequiredDISPCLK[i][j]; v->dummy_vars.dml32_ModeSupportAndSystemConfigurationFull.myPipe.PixelClock = mode_lib->vba.PixelClock[k]; @@ -3242,6 +3289,8 @@ 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], diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.h b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.h index c62e0991358b..c8b28c83ddf4 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.h +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.h @@ -46,9 +46,14 @@ // Prefetch schedule max vratio #define __DML_MAX_VRATIO_PRE__ 4.0 +#define __DML_VBA_MAX_DST_Y_PRE__ 63.75 + #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); diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c index ad66e241f9ae..5af601cff1a0 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c @@ -1726,7 +1726,8 @@ unsigned int dml32_DSCDelayRequirement(bool DSCEnabled, enum output_format_class OutputFormat, enum output_encoder_class Output, double PixelClock, - double PixelClockBackEnd) + double PixelClockBackEnd, + double dsc_delay_factor_wa) { unsigned int DSCDelayRequirement_val; @@ -1746,7 +1747,7 @@ unsigned int dml32_DSCDelayRequirement(bool DSCEnabled, } DSCDelayRequirement_val = DSCDelayRequirement_val + (HTotal - HActive) * - dml_ceil(DSCDelayRequirement_val / HActive, 1); + dml_ceil((double)DSCDelayRequirement_val / HActive, 1); DSCDelayRequirement_val = DSCDelayRequirement_val * PixelClock / PixelClockBackEnd; @@ -1764,7 +1765,7 @@ unsigned int dml32_DSCDelayRequirement(bool DSCEnabled, dml_print("DML::%s: DSCDelayRequirement_val = %d\n", __func__, DSCDelayRequirement_val); #endif - return DSCDelayRequirement_val; + return dml_ceil(DSCDelayRequirement_val * dsc_delay_factor_wa, 1); } void dml32_CalculateSurfaceSizeInMall( @@ -3416,6 +3417,7 @@ bool dml32_CalculatePrefetchSchedule( unsigned int SwathHeightY, unsigned int SwathHeightC, double TWait, + double TPreReq, /* Output */ double *DSTXAfterScaler, double *DSTYAfterScaler, @@ -3666,6 +3668,7 @@ bool dml32_CalculatePrefetchSchedule( dst_y_prefetch_equ = VStartup - (*TSetup + dml_max(TWait + TCalc, *Tdmdl)) / LineTime - (*DSTYAfterScaler + (double) *DSTXAfterScaler / (double) myPipe->HTotal); + dst_y_prefetch_equ = dml_min(dst_y_prefetch_equ, __DML_VBA_MAX_DST_Y_PRE__); #ifdef __DML_VBA_DEBUG__ dml_print("DML::%s: HTotal = %d\n", __func__, myPipe->HTotal); dml_print("DML::%s: min_Lsw = %f\n", __func__, min_Lsw); @@ -3725,7 +3728,8 @@ bool dml32_CalculatePrefetchSchedule( *VRatioPrefetchY = 0; *VRatioPrefetchC = 0; *RequiredPrefetchPixDataBWLuma = 0; - if (dst_y_prefetch_equ > 1) { + if (dst_y_prefetch_equ > 1 && + (Tpre_rounded >= TPreReq || dst_y_prefetch_equ == __DML_VBA_MAX_DST_Y_PRE__)) { double PrefetchBandwidth1; double PrefetchBandwidth2; double PrefetchBandwidth3; @@ -3871,7 +3875,11 @@ bool dml32_CalculatePrefetchSchedule( } if (dst_y_prefetch_oto < dst_y_prefetch_equ) { - *DestinationLinesForPrefetch = dst_y_prefetch_oto; + 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; @@ -4396,7 +4404,7 @@ void dml32_CalculateWatermarksMALLUseAndDRAMSpeedChangeSupport( if (v->NumberOfActiveSurfaces > 1) { ActiveClockChangeLatencyHidingY = ActiveClockChangeLatencyHidingY - - (1 - 1 / v->NumberOfActiveSurfaces) * SwathHeightY[k] * v->HTotal[k] + - (1.0 - 1.0 / v->NumberOfActiveSurfaces) * SwathHeightY[k] * v->HTotal[k] / v->PixelClock[k] / v->VRatio[k]; } @@ -6220,3 +6228,72 @@ void dml32_CalculateImmediateFlipBandwithSupport(unsigned int NumberOfActiveSurf *ImmediateFlipBandwidthSupport = (*TotalBandwidth <= ReturnBW); *FractionOfUrgentBandwidth = *TotalBandwidth / ReturnBW; } + +bool dml32_CalculateDETSwathFillLatencyHiding(unsigned int NumberOfActiveSurfaces, + double ReturnBW, + double UrgentLatency, + unsigned int SwathHeightY[], + unsigned int SwathHeightC[], + unsigned int SwathWidthY[], + unsigned int SwathWidthC[], + double BytePerPixelInDETY[], + double BytePerPixelInDETC[], + unsigned int DETBufferSizeY[], + unsigned int DETBufferSizeC[], + unsigned int NumOfDPP[], + unsigned int HTotal[], + double PixelClock[], + double VRatioY[], + double VRatioC[], + enum dm_use_mall_for_pstate_change_mode UsesMALLForPStateChange[DC__NUM_DPP__MAX]) +{ + int k; + double SwathSizeAllSurfaces = 0; + double SwathSizeAllSurfacesInFetchTimeUs; + double DETSwathLatencyHidingUs; + double DETSwathLatencyHidingYUs; + double DETSwathLatencyHidingCUs; + double SwathSizePerSurfaceY[DC__NUM_DPP__MAX]; + double SwathSizePerSurfaceC[DC__NUM_DPP__MAX]; + bool NotEnoughDETSwathFillLatencyHiding = false; + + /* calculate sum of single swath size for all pipes in bytes*/ + for (k = 0; k < NumberOfActiveSurfaces; k++) { + SwathSizePerSurfaceY[k] += SwathHeightY[k] * SwathWidthY[k] * BytePerPixelInDETY[k] * NumOfDPP[k]; + + if (SwathHeightC[k] != 0) + SwathSizePerSurfaceC[k] += SwathHeightC[k] * SwathWidthC[k] * BytePerPixelInDETC[k] * NumOfDPP[k]; + else + SwathSizePerSurfaceC[k] = 0; + + SwathSizeAllSurfaces += SwathSizePerSurfaceY[k] + SwathSizePerSurfaceC[k]; + } + + SwathSizeAllSurfacesInFetchTimeUs = SwathSizeAllSurfaces / ReturnBW + UrgentLatency; + + /* ensure all DET - 1 swath can hide a fetch for all surfaces */ + for (k = 0; k < NumberOfActiveSurfaces; k++) { + double LineTime = HTotal[k] / PixelClock[k]; + + /* only care if surface is not phantom */ + if (UsesMALLForPStateChange[k] != dm_use_mall_pstate_change_phantom_pipe) { + DETSwathLatencyHidingYUs = (dml_floor(DETBufferSizeY[k] / BytePerPixelInDETY[k] / SwathWidthY[k], 1.0) - SwathHeightY[k]) / VRatioY[k] * LineTime; + + if (SwathHeightC[k] != 0) { + DETSwathLatencyHidingCUs = (dml_floor(DETBufferSizeC[k] / BytePerPixelInDETC[k] / SwathWidthC[k], 1.0) - SwathHeightC[k]) / VRatioC[k] * LineTime; + + DETSwathLatencyHidingUs = dml_min(DETSwathLatencyHidingYUs, DETSwathLatencyHidingCUs); + } else { + DETSwathLatencyHidingUs = DETSwathLatencyHidingYUs; + } + + /* DET must be able to hide time to fetch 1 swath for each surface */ + if (DETSwathLatencyHidingUs < SwathSizeAllSurfacesInFetchTimeUs) { + NotEnoughDETSwathFillLatencyHiding = true; + break; + } + } + } + + return NotEnoughDETSwathFillLatencyHiding; +} diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.h b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.h index 55cead0d4237..779c6805f599 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.h +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.h @@ -30,7 +30,7 @@ #include "os_types.h" #include "../dc_features.h" #include "../display_mode_structs.h" -#include "dml/display_mode_vba.h" +#include "../display_mode_vba.h" unsigned int dml32_dscceComputeDelay( unsigned int bpc, @@ -327,7 +327,8 @@ unsigned int dml32_DSCDelayRequirement(bool DSCEnabled, enum output_format_class OutputFormat, enum output_encoder_class Output, double PixelClock, - double PixelClockBackEnd); + double PixelClockBackEnd, + double dsc_delay_factor_wa); void dml32_CalculateSurfaceSizeInMall( unsigned int NumberOfActiveSurfaces, @@ -742,6 +743,7 @@ bool dml32_CalculatePrefetchSchedule( unsigned int SwathHeightY, unsigned int SwathHeightC, double TWait, + double TPreReq, /* Output */ double *DSTXAfterScaler, double *DSTYAfterScaler, @@ -1139,4 +1141,22 @@ void dml32_CalculateImmediateFlipBandwithSupport(unsigned int NumberOfActiveSurf double *FractionOfUrgentBandwidth, bool *ImmediateFlipBandwidthSupport); +bool dml32_CalculateDETSwathFillLatencyHiding(unsigned int NumberOfActiveSurfaces, + double ReturnBW, + double UrgentLatency, + unsigned int SwathHeightY[], + unsigned int SwathHeightC[], + unsigned int SwathWidthY[], + unsigned int SwathWidthC[], + double BytePerPixelInDETY[], + double BytePerPixelInDETC[], + unsigned int DETBufferSizeY[], + unsigned int DETBufferSizeC[], + unsigned int NumOfDPP[], + unsigned int HTotal[], + double PixelClock[], + double VRatioY[], + double VRatioC[], + enum dm_use_mall_for_pstate_change_mode UsesMALLForPStateChange[DC__NUM_DPP__MAX]); + #endif diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_rq_dlg_calc_32.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_rq_dlg_calc_32.c index a1276f6b9581..395ae8761980 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_rq_dlg_calc_32.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_rq_dlg_calc_32.c @@ -291,8 +291,8 @@ void dml32_rq_dlg_get_dlg_reg(struct display_mode_lib *mode_lib, dml_print("DML_DLG: %s: vready_after_vcount0 = %d\n", __func__, dlg_regs->vready_after_vcount0); - dst_x_after_scaler = get_dst_x_after_scaler(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); - dst_y_after_scaler = get_dst_y_after_scaler(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); + dst_x_after_scaler = dml_ceil(get_dst_x_after_scaler(mode_lib, e2e_pipe_param, num_pipes, pipe_idx), 1); + dst_y_after_scaler = dml_ceil(get_dst_y_after_scaler(mode_lib, e2e_pipe_param, num_pipes, pipe_idx), 1); // do some adjustment on the dst_after scaler to account for odm combine mode dml_print("DML_DLG: %s: input dst_x_after_scaler = %d\n", __func__, dst_x_after_scaler); diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c index dd90f241e906..f4b176599be7 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c @@ -29,6 +29,7 @@ #include "dcn321_fpu.h" #include "dcn32/dcn32_resource.h" #include "dcn321/dcn321_resource.h" +#include "dml/dcn32/display_mode_vba_util_32.h" #define DCN3_2_DEFAULT_DET_SIZE 256 @@ -119,15 +120,15 @@ struct _vcs_dpi_soc_bounding_box_st dcn3_21_soc = { }, }, .num_states = 1, - .sr_exit_time_us = 12.36, - .sr_enter_plus_exit_time_us = 16.72, + .sr_exit_time_us = 19.95, + .sr_enter_plus_exit_time_us = 24.36, .sr_exit_z8_time_us = 285.0, .sr_enter_plus_exit_z8_time_us = 320, .writeback_latency_us = 12.0, .round_trip_ping_latency_dcfclk_cycles = 263, - .urgent_latency_pixel_data_only_us = 4.0, - .urgent_latency_pixel_mixed_with_vm_data_us = 4.0, - .urgent_latency_vm_data_only_us = 4.0, + .urgent_latency_pixel_data_only_us = 4, + .urgent_latency_pixel_mixed_with_vm_data_us = 4, + .urgent_latency_vm_data_only_us = 4, .fclk_change_latency_us = 20, .usr_retraining_latency_us = 2, .smn_latency_us = 2, @@ -155,7 +156,7 @@ struct _vcs_dpi_soc_bounding_box_st dcn3_21_soc = { .dispclk_dppclk_vco_speed_mhz = 4300.0, .do_urgent_latency_adjustment = true, .urgent_latency_adjustment_fabric_clock_component_us = 1.0, - .urgent_latency_adjustment_fabric_clock_reference_mhz = 1000, + .urgent_latency_adjustment_fabric_clock_reference_mhz = 3000, }; static void get_optimal_ntuple(struct _vcs_dpi_voltage_scaling_st *entry) @@ -538,9 +539,13 @@ void dcn321_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_p if (dc->ctx->dc_bios->vram_info.dram_channel_width_bytes) dcn3_21_soc.dram_channel_width_bytes = dc->ctx->dc_bios->vram_info.dram_channel_width_bytes; - } + /* 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; diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_enums.h b/drivers/gpu/drm/amd/display/dc/dml/display_mode_enums.h index f394b3f3922a..0bffae95f3a2 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_enums.h +++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_enums.h @@ -105,14 +105,39 @@ enum source_macro_tile_size { enum cursor_bpp { dm_cur_2bit = 0, dm_cur_32bit = 1, dm_cur_64bit = 2 }; + +/** + * @enum clock_change_support - It represents possible reasons to change the DRAM clock. + * + * DC may change the DRAM clock during its execution, and this enum tracks all + * the available methods. Note that every ASIC has their specific way to deal + * with these clock switch. + */ enum clock_change_support { + /** + * @dm_dram_clock_change_uninitialized: If you see this, we might have + * a code initialization issue + */ dm_dram_clock_change_uninitialized = 0, + + /** + * @dm_dram_clock_change_vactive: Support DRAM switch in VActive + */ dm_dram_clock_change_vactive, + + /** + * @dm_dram_clock_change_vblank: Support DRAM switch in VBlank + */ dm_dram_clock_change_vblank, + dm_dram_clock_change_vactive_w_mall_full_frame, dm_dram_clock_change_vactive_w_mall_sub_vp, dm_dram_clock_change_vblank_w_mall_full_frame, dm_dram_clock_change_vblank_w_mall_sub_vp, + + /** + * @dm_dram_clock_change_unsupported: Do not support DRAM switch + */ dm_dram_clock_change_unsupported }; diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h b/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h index f33a8879b05a..64d602e6412f 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h +++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h @@ -364,6 +364,10 @@ struct _vcs_dpi_ip_params_st { unsigned int max_num_dp2p0_outputs; unsigned int max_num_dp2p0_streams; unsigned int VBlankNomDefaultUS; + + /* 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 { diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c b/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c index 03924aed8d5c..8e6585dab20e 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c +++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c @@ -625,7 +625,7 @@ static void fetch_pipe_params(struct display_mode_lib *mode_lib) mode_lib->vba.skip_dio_check[mode_lib->vba.NumberOfActivePlanes] = dout->is_virtual; - if (!dout->dsc_enable) + if (dout->dsc_enable) mode_lib->vba.ForcedOutputLinkBPP[mode_lib->vba.NumberOfActivePlanes] = dout->output_bpp; else mode_lib->vba.ForcedOutputLinkBPP[mode_lib->vba.NumberOfActivePlanes] = 0.0; diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.h b/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.h index 630f3395e90a..81e53e67cd0b 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.h +++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.h @@ -419,6 +419,15 @@ struct vba_vars_st { double MinPixelChunkSizeBytes; unsigned int DCCMetaBufferSizeBytes; // Pipe/Plane Parameters + + /** @VoltageLevel: + * Every ASIC has a fixed number of DPM states, and some devices might + * have some particular voltage configuration that does not map + * directly to the DPM states. This field tells how many states the + * target device supports; even though this field combines the DPM and + * special SOC voltages, it mostly matches the total number of DPM + * states. + */ int VoltageLevel; double FabricClock; double DRAMSpeed; @@ -1041,6 +1050,7 @@ struct vba_vars_st { double MinFullDETBufferingTime; double AverageReadBandwidthGBytePerSecond; bool FirstMainPlane; + bool NotEnoughDETSwathFillLatencyHiding; unsigned int ViewportWidthChroma[DC__NUM_DPP__MAX]; unsigned int ViewportHeightChroma[DC__NUM_DPP__MAX]; @@ -1153,7 +1163,7 @@ struct vba_vars_st { double UrgBurstFactorLumaPre[DC__NUM_DPP__MAX]; double UrgBurstFactorChromaPre[DC__NUM_DPP__MAX]; bool NotUrgentLatencyHidingPre[DC__NUM_DPP__MAX]; - bool LinkCapacitySupport[DC__NUM_DPP__MAX]; + bool LinkCapacitySupport[DC__VOLTAGE_STATES]; bool VREADY_AT_OR_AFTER_VSYNC[DC__NUM_DPP__MAX]; unsigned int MIN_DST_Y_NEXT_START[DC__NUM_DPP__MAX]; unsigned int VFrontPorch[DC__NUM_DPP__MAX]; @@ -1224,6 +1234,7 @@ struct vba_vars_st { unsigned int BlockWidthC[DC__NUM_DPP__MAX]; unsigned int SubViewportLinesNeededInMALL[DC__NUM_DPP__MAX]; bool VActiveBandwithSupport[DC__VOLTAGE_STATES][2]; + bool NotEnoughDETSwathFillLatencyHidingPerState[DC__VOLTAGE_STATES][2]; struct dummy_vars dummy_vars; }; diff --git a/drivers/gpu/drm/amd/display/dc/dml/dsc/qp_tables.h b/drivers/gpu/drm/amd/display/dc/dml/dsc/qp_tables.h index e5fac9f4181d..dcff0dd2b6a1 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dsc/qp_tables.h +++ b/drivers/gpu/drm/amd/display/dc/dml/dsc/qp_tables.h @@ -25,7 +25,7 @@ */ -const qp_table qp_table_422_10bpc_min = { +static const qp_table qp_table_422_10bpc_min = { { 6, { 0, 4, 5, 6, 6, 6, 6, 7, 7, 8, 9, 9, 9, 12, 16} }, { 6.5, { 0, 4, 5, 6, 6, 6, 6, 7, 7, 8, 9, 9, 9, 12, 16} }, { 7, { 0, 4, 5, 6, 6, 6, 6, 7, 7, 7, 9, 9, 9, 11, 15} }, @@ -58,7 +58,7 @@ const qp_table qp_table_422_10bpc_min = { }; -const qp_table qp_table_444_8bpc_max = { +static const qp_table qp_table_444_8bpc_max = { { 6, { 4, 6, 8, 8, 9, 9, 9, 10, 11, 12, 12, 12, 12, 13, 15} }, { 6.5, { 4, 6, 7, 8, 8, 8, 9, 10, 11, 11, 12, 12, 12, 13, 15} }, { 7, { 4, 5, 7, 7, 8, 8, 8, 9, 10, 11, 11, 12, 12, 13, 14} }, @@ -99,7 +99,7 @@ const qp_table qp_table_444_8bpc_max = { }; -const qp_table qp_table_420_12bpc_max = { +static const qp_table qp_table_420_12bpc_max = { { 4, {11, 12, 13, 14, 15, 15, 15, 16, 17, 18, 18, 19, 19, 21, 22} }, { 4.5, {10, 11, 12, 13, 14, 15, 15, 16, 17, 18, 18, 19, 19, 20, 21} }, { 5, { 9, 11, 12, 13, 14, 15, 15, 16, 17, 17, 18, 18, 19, 20, 21} }, @@ -132,7 +132,7 @@ const qp_table qp_table_420_12bpc_max = { }; -const qp_table qp_table_444_10bpc_min = { +static const qp_table qp_table_444_10bpc_min = { { 6, { 0, 4, 7, 7, 9, 9, 9, 9, 9, 10, 10, 10, 10, 12, 18} }, { 6.5, { 0, 4, 6, 7, 8, 8, 9, 9, 9, 9, 10, 10, 10, 12, 18} }, { 7, { 0, 4, 6, 6, 8, 8, 8, 8, 8, 9, 9, 10, 10, 12, 17} }, @@ -185,7 +185,7 @@ const qp_table qp_table_444_10bpc_min = { }; -const qp_table qp_table_420_8bpc_max = { +static const qp_table qp_table_420_8bpc_max = { { 4, { 4, 4, 5, 6, 7, 7, 7, 8, 9, 10, 10, 11, 11, 13, 14} }, { 4.5, { 4, 4, 5, 6, 7, 7, 7, 8, 9, 10, 10, 11, 11, 12, 13} }, { 5, { 3, 4, 5, 6, 7, 7, 7, 8, 9, 9, 10, 10, 11, 12, 13} }, @@ -206,7 +206,7 @@ const qp_table qp_table_420_8bpc_max = { }; -const qp_table qp_table_444_8bpc_min = { +static const qp_table qp_table_444_8bpc_min = { { 6, { 0, 1, 3, 3, 5, 5, 5, 5, 5, 6, 6, 6, 6, 9, 14} }, { 6.5, { 0, 1, 2, 3, 4, 4, 5, 5, 5, 5, 6, 6, 6, 9, 14} }, { 7, { 0, 0, 2, 2, 4, 4, 4, 4, 4, 5, 5, 6, 6, 9, 13} }, @@ -247,7 +247,7 @@ const qp_table qp_table_444_8bpc_min = { }; -const qp_table qp_table_444_12bpc_min = { +static const qp_table qp_table_444_12bpc_min = { { 6, { 0, 5, 11, 11, 13, 13, 13, 13, 13, 14, 14, 14, 14, 17, 22} }, { 6.5, { 0, 5, 10, 11, 12, 12, 13, 13, 13, 13, 14, 14, 14, 17, 22} }, { 7, { 0, 5, 10, 10, 12, 12, 12, 12, 12, 13, 13, 14, 14, 17, 21} }, @@ -312,7 +312,7 @@ const qp_table qp_table_444_12bpc_min = { }; -const qp_table qp_table_420_12bpc_min = { +static const qp_table qp_table_420_12bpc_min = { { 4, { 0, 4, 9, 10, 11, 11, 11, 11, 11, 11, 13, 13, 13, 15, 21} }, { 4.5, { 0, 4, 8, 9, 10, 11, 11, 11, 11, 11, 13, 13, 13, 15, 20} }, { 5, { 0, 4, 8, 9, 10, 11, 11, 11, 11, 11, 13, 13, 13, 15, 20} }, @@ -345,7 +345,7 @@ const qp_table qp_table_420_12bpc_min = { }; -const qp_table qp_table_422_12bpc_min = { +static const qp_table qp_table_422_12bpc_min = { { 6, { 0, 4, 9, 10, 11, 11, 11, 11, 11, 11, 13, 13, 13, 16, 20} }, { 6.5, { 0, 4, 9, 10, 11, 11, 11, 11, 11, 11, 13, 13, 13, 16, 20} }, { 7, { 0, 4, 9, 10, 11, 11, 11, 11, 11, 11, 13, 13, 13, 15, 19} }, @@ -386,7 +386,7 @@ const qp_table qp_table_422_12bpc_min = { }; -const qp_table qp_table_422_12bpc_max = { +static const qp_table qp_table_422_12bpc_max = { { 6, {12, 12, 13, 14, 15, 15, 15, 16, 17, 18, 18, 19, 19, 20, 21} }, { 6.5, {12, 12, 13, 14, 15, 15, 15, 16, 17, 18, 18, 19, 19, 20, 21} }, { 7, {11, 12, 13, 14, 15, 15, 15, 16, 17, 17, 18, 18, 19, 19, 20} }, @@ -427,7 +427,7 @@ const qp_table qp_table_422_12bpc_max = { }; -const qp_table qp_table_444_12bpc_max = { +static const qp_table qp_table_444_12bpc_max = { { 6, {12, 14, 16, 16, 17, 17, 17, 18, 19, 20, 20, 20, 20, 21, 23} }, { 6.5, {12, 14, 15, 16, 16, 16, 17, 18, 19, 19, 20, 20, 20, 21, 23} }, { 7, {12, 13, 15, 15, 16, 16, 16, 17, 18, 19, 19, 20, 20, 21, 22} }, @@ -492,7 +492,7 @@ const qp_table qp_table_444_12bpc_max = { }; -const qp_table qp_table_420_8bpc_min = { +static const qp_table qp_table_420_8bpc_min = { { 4, { 0, 0, 1, 1, 3, 3, 3, 3, 3, 3, 5, 5, 5, 9, 13} }, { 4.5, { 0, 0, 1, 1, 3, 3, 3, 3, 3, 3, 5, 5, 5, 8, 12} }, { 5, { 0, 0, 1, 1, 3, 3, 3, 3, 3, 3, 5, 5, 5, 8, 12} }, @@ -513,7 +513,7 @@ const qp_table qp_table_420_8bpc_min = { }; -const qp_table qp_table_422_8bpc_min = { +static const qp_table qp_table_422_8bpc_min = { { 6, { 0, 0, 1, 2, 3, 3, 3, 3, 3, 3, 5, 5, 5, 8, 12} }, { 6.5, { 0, 0, 1, 2, 3, 3, 3, 3, 3, 3, 5, 5, 5, 8, 12} }, { 7, { 0, 0, 1, 2, 3, 3, 3, 3, 3, 3, 5, 5, 5, 7, 11} }, @@ -538,7 +538,7 @@ const qp_table qp_table_422_8bpc_min = { }; -const qp_table qp_table_422_10bpc_max = { +static const qp_table qp_table_422_10bpc_max = { { 6, { 8, 8, 9, 10, 11, 11, 11, 12, 13, 14, 14, 15, 15, 16, 17} }, { 6.5, { 8, 8, 9, 10, 11, 11, 11, 12, 13, 14, 14, 15, 15, 16, 17} }, { 7, { 7, 8, 9, 10, 11, 11, 11, 12, 13, 13, 14, 14, 15, 15, 16} }, @@ -571,7 +571,7 @@ const qp_table qp_table_422_10bpc_max = { }; -const qp_table qp_table_420_10bpc_max = { +static const qp_table qp_table_420_10bpc_max = { { 4, { 8, 8, 9, 10, 11, 11, 11, 12, 13, 14, 14, 15, 15, 17, 18} }, { 4.5, { 8, 8, 9, 10, 11, 11, 11, 12, 13, 14, 14, 15, 15, 16, 17} }, { 5, { 7, 8, 9, 10, 11, 11, 11, 12, 13, 13, 14, 14, 15, 16, 17} }, @@ -598,7 +598,7 @@ const qp_table qp_table_420_10bpc_max = { }; -const qp_table qp_table_420_10bpc_min = { +static const qp_table qp_table_420_10bpc_min = { { 4, { 0, 4, 4, 5, 7, 7, 7, 7, 7, 7, 9, 9, 9, 13, 17} }, { 4.5, { 0, 4, 4, 5, 7, 7, 7, 7, 7, 7, 9, 9, 9, 12, 16} }, { 5, { 0, 4, 4, 5, 7, 7, 7, 7, 7, 7, 9, 9, 9, 12, 16} }, @@ -625,7 +625,7 @@ const qp_table qp_table_420_10bpc_min = { }; -const qp_table qp_table_444_10bpc_max = { +static const qp_table qp_table_444_10bpc_max = { { 6, { 8, 10, 12, 12, 13, 13, 13, 14, 15, 16, 16, 16, 16, 17, 19} }, { 6.5, { 8, 10, 11, 12, 12, 12, 13, 14, 15, 15, 16, 16, 16, 17, 19} }, { 7, { 8, 9, 11, 11, 12, 12, 12, 13, 14, 15, 15, 16, 16, 17, 18} }, @@ -678,7 +678,7 @@ const qp_table qp_table_444_10bpc_max = { }; -const qp_table qp_table_422_8bpc_max = { +static const qp_table qp_table_422_8bpc_max = { { 6, { 4, 4, 5, 6, 7, 7, 7, 8, 9, 10, 10, 11, 11, 12, 13} }, { 6.5, { 4, 4, 5, 6, 7, 7, 7, 8, 9, 10, 10, 11, 11, 12, 13} }, { 7, { 3, 4, 5, 6, 7, 7, 7, 8, 9, 9, 10, 10, 11, 11, 12} }, |