diff options
Diffstat (limited to 'drivers/gpu/drm/amd/display')
40 files changed, 1118 insertions, 768 deletions
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c index a92cfb055c15..39b425d83bb1 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -769,7 +769,7 @@ static void dm_dmub_outbox1_low_irq(void *interrupt_params) do { dc_stat_get_dmub_notification(adev->dm.dc, ¬ify); - if (notify.type > ARRAY_SIZE(dm->dmub_thread_offload)) { + if (notify.type >= ARRAY_SIZE(dm->dmub_thread_offload)) { DRM_ERROR("DM: notify type %d invalid!", notify.type); continue; } @@ -2812,7 +2812,7 @@ static struct drm_mode_config_helper_funcs amdgpu_dm_mode_config_helperfuncs = { static void update_connector_ext_caps(struct amdgpu_dm_connector *aconnector) { - u32 max_cll, min_cll, max, min, q, r; + u32 max_avg, min_cll, max, min, q, r; struct amdgpu_dm_backlight_caps *caps; struct amdgpu_display_manager *dm; struct drm_connector *conn_base; @@ -2842,7 +2842,7 @@ static void update_connector_ext_caps(struct amdgpu_dm_connector *aconnector) caps = &dm->backlight_caps[i]; caps->ext_caps = &aconnector->dc_link->dpcd_sink_ext_caps; caps->aux_support = false; - max_cll = conn_base->hdr_sink_metadata.hdmi_type1.max_cll; + max_avg = conn_base->hdr_sink_metadata.hdmi_type1.max_fall; min_cll = conn_base->hdr_sink_metadata.hdmi_type1.min_cll; if (caps->ext_caps->bits.oled == 1 /*|| @@ -2870,8 +2870,8 @@ static void update_connector_ext_caps(struct amdgpu_dm_connector *aconnector) * The results of the above expressions can be verified at * pre_computed_values. */ - q = max_cll >> 5; - r = max_cll % 32; + q = max_avg >> 5; + r = max_avg % 32; max = (1 << q) * pre_computed_values[r]; // min luminance: maxLum * (CV/255)^2 / 100 @@ -5381,17 +5381,19 @@ fill_plane_buffer_attributes(struct amdgpu_device *adev, static void fill_blending_from_plane_state(const struct drm_plane_state *plane_state, - bool *per_pixel_alpha, bool *global_alpha, - int *global_alpha_value) + bool *per_pixel_alpha, bool *pre_multiplied_alpha, + bool *global_alpha, int *global_alpha_value) { *per_pixel_alpha = false; + *pre_multiplied_alpha = true; *global_alpha = false; *global_alpha_value = 0xff; if (plane_state->plane->type != DRM_PLANE_TYPE_OVERLAY) return; - if (plane_state->pixel_blend_mode == DRM_MODE_BLEND_PREMULTI) { + if (plane_state->pixel_blend_mode == DRM_MODE_BLEND_PREMULTI || + plane_state->pixel_blend_mode == DRM_MODE_BLEND_COVERAGE) { static const uint32_t alpha_formats[] = { DRM_FORMAT_ARGB8888, DRM_FORMAT_RGBA8888, @@ -5406,6 +5408,9 @@ fill_blending_from_plane_state(const struct drm_plane_state *plane_state, break; } } + + if (per_pixel_alpha && plane_state->pixel_blend_mode == DRM_MODE_BLEND_COVERAGE) + *pre_multiplied_alpha = false; } if (plane_state->alpha < 0xffff) { @@ -5568,7 +5573,7 @@ fill_dc_plane_info_and_addr(struct amdgpu_device *adev, return ret; fill_blending_from_plane_state( - plane_state, &plane_info->per_pixel_alpha, + plane_state, &plane_info->per_pixel_alpha, &plane_info->pre_multiplied_alpha, &plane_info->global_alpha, &plane_info->global_alpha_value); return 0; @@ -5615,6 +5620,7 @@ static int fill_dc_plane_attributes(struct amdgpu_device *adev, dc_plane_state->tiling_info = plane_info.tiling_info; dc_plane_state->visible = plane_info.visible; dc_plane_state->per_pixel_alpha = plane_info.per_pixel_alpha; + dc_plane_state->pre_multiplied_alpha = plane_info.pre_multiplied_alpha; dc_plane_state->global_alpha = plane_info.global_alpha; dc_plane_state->global_alpha_value = plane_info.global_alpha_value; dc_plane_state->dcc = plane_info.dcc; @@ -7911,7 +7917,8 @@ static int amdgpu_dm_plane_init(struct amdgpu_display_manager *dm, if (plane->type == DRM_PLANE_TYPE_OVERLAY && plane_cap && plane_cap->per_pixel_alpha) { unsigned int blend_caps = BIT(DRM_MODE_BLEND_PIXEL_NONE) | - BIT(DRM_MODE_BLEND_PREMULTI); + BIT(DRM_MODE_BLEND_PREMULTI) | + BIT(DRM_MODE_BLEND_COVERAGE); drm_plane_create_alpha_property(plane); drm_plane_create_blend_mode_property(plane, blend_caps); diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c index 02943ca65807..cf1b5f354ae9 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c @@ -122,7 +122,7 @@ static void rn_update_clocks_update_dpp_dto(struct clk_mgr_internal *clk_mgr, dpp_inst = clk_mgr->base.ctx->dc->res_pool->dpps[i]->inst; dppclk_khz = context->res_ctx.pipe_ctx[i].plane_res.bw.dppclk_khz; - prev_dppclk_khz = clk_mgr->dccg->pipe_dppclk_khz[i]; + prev_dppclk_khz = clk_mgr->dccg->pipe_dppclk_khz[dpp_inst]; if (safe_to_lower || prev_dppclk_khz < dppclk_khz) clk_mgr->dccg->funcs->update_dpp_dto( diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_clk_mgr.c index ceb34376decb..bca5f01da763 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_clk_mgr.c @@ -287,8 +287,11 @@ static void dcn31_enable_pme_wa(struct clk_mgr *clk_mgr_base) void dcn31_init_clocks(struct clk_mgr *clk_mgr) { + uint32_t ref_dtbclk = clk_mgr->clks.ref_dtbclk_khz; + memset(&(clk_mgr->clks), 0, sizeof(struct dc_clocks)); // Assumption is that boot state always supports pstate + clk_mgr->clks.ref_dtbclk_khz = ref_dtbclk; // restore ref_dtbclk clk_mgr->clks.p_state_change_support = true; clk_mgr->clks.prev_p_state_change_support = true; clk_mgr->clks.pwr_state = DCN_PWR_STATE_UNKNOWN; @@ -638,8 +641,14 @@ static void dcn31_set_low_power_state(struct clk_mgr *clk_mgr_base) } } +int dcn31_get_dtb_ref_freq_khz(struct clk_mgr *clk_mgr_base) +{ + return clk_mgr_base->clks.ref_dtbclk_khz; +} + static struct clk_mgr_funcs dcn31_funcs = { .get_dp_ref_clk_frequency = dce12_get_dp_ref_freq_khz, + .get_dtb_ref_clk_frequency = dcn31_get_dtb_ref_freq_khz, .update_clocks = dcn31_update_clocks, .init_clocks = dcn31_init_clocks, .enable_pme_wa = dcn31_enable_pme_wa, @@ -719,7 +728,7 @@ void dcn31_clk_mgr_construct( } clk_mgr->base.base.dprefclk_khz = 600000; - clk_mgr->base.dccg->ref_dtbclk_khz = 600000; + clk_mgr->base.base.clks.ref_dtbclk_khz = 600000; dce_clock_read_ss_info(&clk_mgr->base); /*if bios enabled SS, driver needs to adjust dtb clock, only enable with correct bios*/ //clk_mgr->base.dccg->ref_dtbclk_khz = dce_adjust_dp_ref_freq_for_ss(clk_mgr_internal, clk_mgr->base.base.dprefclk_khz); diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_clk_mgr.h b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_clk_mgr.h index 961b10a49486..be06fdbd0c22 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_clk_mgr.h +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_clk_mgr.h @@ -51,6 +51,8 @@ void dcn31_clk_mgr_construct(struct dc_context *ctx, struct pp_smu_funcs *pp_smu, struct dccg *dccg); +int dcn31_get_dtb_ref_freq_khz(struct clk_mgr *clk_mgr_base); + void dcn31_clk_mgr_destroy(struct clk_mgr_internal *clk_mgr_int); #endif //__DCN31_CLK_MGR_H__ diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_clk_mgr.c index 27501b735a9c..fb4ae800e919 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_clk_mgr.c @@ -41,9 +41,7 @@ #include "dc_dmub_srv.h" -#if defined (CONFIG_DRM_AMD_DC_DP2_0) #include "dc_link_dp.h" -#endif #define TO_CLK_MGR_DCN315(clk_mgr)\ container_of(clk_mgr, struct clk_mgr_dcn315, base) @@ -91,7 +89,8 @@ static void dcn315_disable_otg_wa(struct clk_mgr *clk_mgr_base, bool disable) if (pipe->top_pipe || pipe->prev_odm_pipe) continue; - if (pipe->stream && (pipe->stream->dpms_off || dc_is_virtual_signal(pipe->stream->signal))) { + if (pipe->stream && (pipe->stream->dpms_off || pipe->plane_state == NULL || + dc_is_virtual_signal(pipe->stream->signal))) { if (disable) pipe->stream_res.tg->funcs->immediate_disable_crtc(pipe->stream_res.tg); else @@ -579,6 +578,7 @@ static void dcn315_enable_pme_wa(struct clk_mgr *clk_mgr_base) static struct clk_mgr_funcs dcn315_funcs = { .get_dp_ref_clk_frequency = dce12_get_dp_ref_freq_khz, + .get_dtb_ref_clk_frequency = dcn31_get_dtb_ref_freq_khz, .update_clocks = dcn315_update_clocks, .init_clocks = dcn31_init_clocks, .enable_pme_wa = dcn315_enable_pme_wa, @@ -655,9 +655,9 @@ void dcn315_clk_mgr_construct( clk_mgr->base.base.dprefclk_khz = 600000; clk_mgr->base.base.dprefclk_khz = dcn315_smu_get_dpref_clk(&clk_mgr->base); - clk_mgr->base.dccg->ref_dtbclk_khz = clk_mgr->base.base.dprefclk_khz; + clk_mgr->base.base.clks.ref_dtbclk_khz = clk_mgr->base.base.dprefclk_khz; dce_clock_read_ss_info(&clk_mgr->base); - clk_mgr->base.dccg->ref_dtbclk_khz = dce_adjust_dp_ref_freq_for_ss(&clk_mgr->base, clk_mgr->base.base.dprefclk_khz); + clk_mgr->base.base.clks.ref_dtbclk_khz = dce_adjust_dp_ref_freq_for_ss(&clk_mgr->base, clk_mgr->base.base.dprefclk_khz); clk_mgr->base.base.bw_params = &dcn315_bw_params; diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn316/dcn316_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn316/dcn316_clk_mgr.c index 3121dd2d2a91..e4bb9c6193b5 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn316/dcn316_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn316/dcn316_clk_mgr.c @@ -122,7 +122,8 @@ static void dcn316_disable_otg_wa(struct clk_mgr *clk_mgr_base, bool disable) if (pipe->top_pipe || pipe->prev_odm_pipe) continue; - if (pipe->stream && (pipe->stream->dpms_off || dc_is_virtual_signal(pipe->stream->signal))) { + if (pipe->stream && (pipe->stream->dpms_off || pipe->plane_state == NULL || + dc_is_virtual_signal(pipe->stream->signal))) { if (disable) pipe->stream_res.tg->funcs->immediate_disable_crtc(pipe->stream_res.tg); else @@ -570,6 +571,7 @@ static void dcn316_clk_mgr_helper_populate_bw_params( static struct clk_mgr_funcs dcn316_funcs = { .enable_pme_wa = dcn316_enable_pme_wa, .get_dp_ref_clk_frequency = dce12_get_dp_ref_freq_khz, + .get_dtb_ref_clk_frequency = dcn31_get_dtb_ref_freq_khz, .update_clocks = dcn316_update_clocks, .init_clocks = dcn31_init_clocks, .are_clock_states_equal = dcn31_are_clock_states_equal, @@ -684,7 +686,7 @@ void dcn316_clk_mgr_construct( clk_mgr->base.base.dprefclk_khz = 600000; clk_mgr->base.base.dprefclk_khz = dcn316_smu_get_dpref_clk(&clk_mgr->base); - clk_mgr->base.dccg->ref_dtbclk_khz = clk_mgr->base.base.dprefclk_khz; + clk_mgr->base.base.clks.ref_dtbclk_khz = clk_mgr->base.base.dprefclk_khz; dce_clock_read_ss_info(&clk_mgr->base); /*clk_mgr->base.dccg->ref_dtbclk_khz = dce_adjust_dp_ref_freq_for_ss(&clk_mgr->base, clk_mgr->base.base.dprefclk_khz);*/ diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c index e41a48f596a3..f14449401188 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc.c @@ -2901,14 +2901,15 @@ static void commit_planes_for_stream(struct dc *dc, top_pipe_to_program->stream_res.tg); } - if (should_lock_all_pipes && dc->hwss.interdependent_update_lock) + if (should_lock_all_pipes && dc->hwss.interdependent_update_lock) { dc->hwss.interdependent_update_lock(dc, context, true); - else + } else { /* Lock the top pipe while updating plane addrs, since freesync requires * plane addr update event triggers to be synchronized. * top_pipe_to_program is expected to never be NULL */ dc->hwss.pipe_control_lock(dc, top_pipe_to_program, true); + } // Stream updates if (stream_update) @@ -2924,10 +2925,11 @@ static void commit_planes_for_stream(struct dc *dc, if (dc->hwss.program_front_end_for_ctx) dc->hwss.program_front_end_for_ctx(dc, context); - if (should_lock_all_pipes && dc->hwss.interdependent_update_lock) + if (should_lock_all_pipes && dc->hwss.interdependent_update_lock) { dc->hwss.interdependent_update_lock(dc, context, false); - else + } else { dc->hwss.pipe_control_lock(dc, top_pipe_to_program, false); + } dc->hwss.post_unlock_program_front_end(dc, context); return; } @@ -3052,10 +3054,11 @@ static void commit_planes_for_stream(struct dc *dc, } - if (should_lock_all_pipes && dc->hwss.interdependent_update_lock) + if (should_lock_all_pipes && dc->hwss.interdependent_update_lock) { dc->hwss.interdependent_update_lock(dc, context, false); - else + } else { dc->hwss.pipe_control_lock(dc, top_pipe_to_program, false); + } if ((update_type != UPDATE_TYPE_FAST) && stream->update_flags.bits.dsc_changed) if (top_pipe_to_program->stream_res.tg->funcs->lock_doublebuffer_enable) { diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link.c b/drivers/gpu/drm/amd/display/dc/core/dc_link.c index 67ef357e5798..a789ea8af27f 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link.c @@ -33,6 +33,7 @@ #include "gpio_service_interface.h" #include "core_status.h" #include "dc_link_dp.h" +#include "dc_link_dpia.h" #include "dc_link_ddc.h" #include "link_hwss.h" #include "opp.h" @@ -240,7 +241,7 @@ bool dc_link_detect_sink(struct dc_link *link, enum dc_connection_type *type) /* Link may not have physical HPD pin. */ if (link->ep_type != DISPLAY_ENDPOINT_PHY) { - if (link->is_hpd_pending || !link->hpd_status) + if (link->is_hpd_pending || !dc_link_dpia_query_hpd_status(link)) *type = dc_connection_none; else *type = dc_connection_single; @@ -1604,8 +1605,25 @@ static bool dc_link_construct_legacy(struct dc_link *link, if (link->hpd_gpio) { if (!link->dc->config.allow_edp_hotplug_detection) link->irq_source_hpd = DC_IRQ_SOURCE_INVALID; - link->irq_source_hpd_rx = - dal_irq_get_rx_source(link->hpd_gpio); + + switch (link->dc->config.allow_edp_hotplug_detection) { + case 1: // only the 1st eDP handles hotplug + if (link->link_index == 0) + link->irq_source_hpd_rx = + dal_irq_get_rx_source(link->hpd_gpio); + else + link->irq_source_hpd = DC_IRQ_SOURCE_INVALID; + break; + case 2: // only the 2nd eDP handles hotplug + if (link->link_index == 1) + link->irq_source_hpd_rx = + dal_irq_get_rx_source(link->hpd_gpio); + else + link->irq_source_hpd = DC_IRQ_SOURCE_INVALID; + break; + default: + break; + } } break; diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c index 340b5f90a82a..cbc47aecd00f 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c @@ -114,8 +114,8 @@ static const struct dc_link_settings fail_safe_link_settings = { static bool decide_fallback_link_setting( struct dc_link *link, - struct dc_link_settings initial_link_settings, - struct dc_link_settings *current_link_setting, + struct dc_link_settings *max, + struct dc_link_settings *cur, enum link_training_result training_result); static void maximize_lane_settings(const struct link_training_settings *lt_settings, struct dc_lane_settings lane_settings[LANE_COUNT_DP_MAX]); @@ -2783,31 +2783,37 @@ bool perform_link_training_with_retries( struct dc_link *link = stream->link; enum dp_panel_mode panel_mode = dp_get_panel_mode(link); enum link_training_result status = LINK_TRAINING_CR_FAIL_LANE0; - struct dc_link_settings current_setting = *link_setting; + struct dc_link_settings cur_link_settings = *link_setting; + struct dc_link_settings max_link_settings = *link_setting; const struct link_hwss *link_hwss = get_link_hwss(link, &pipe_ctx->link_res); int fail_count = 0; + bool is_link_bw_low = false; /* link bandwidth < stream bandwidth */ + bool is_link_bw_min = /* RBR x 1 */ + (cur_link_settings.link_rate <= LINK_RATE_LOW) && + (cur_link_settings.lane_count <= LANE_COUNT_ONE); dp_trace_commit_lt_init(link); - - if (dp_get_link_encoding_format(¤t_setting) == DP_8b_10b_ENCODING) + if (dp_get_link_encoding_format(&cur_link_settings) == DP_8b_10b_ENCODING) /* We need to do this before the link training to ensure the idle * pattern in SST mode will be sent right after the link training */ link_hwss->setup_stream_encoder(pipe_ctx); dp_trace_set_lt_start_timestamp(link, false); - for (j = 0; j < attempts; ++j) { + j = 0; + while (j < attempts && fail_count < (attempts * 10)) { - DC_LOG_HW_LINK_TRAINING("%s: Beginning link training attempt %u of %d\n", - __func__, (unsigned int)j + 1, attempts); + DC_LOG_HW_LINK_TRAINING("%s: Beginning link training attempt %u of %d @ rate(%d) x lane(%d)\n", + __func__, (unsigned int)j + 1, attempts, cur_link_settings.link_rate, + cur_link_settings.lane_count); dp_enable_link_phy( link, &pipe_ctx->link_res, signal, pipe_ctx->clock_source->id, - ¤t_setting); + &cur_link_settings); if (stream->sink_patches.dppowerup_delay > 0) { int delay_dp_power_up_in_ms = stream->sink_patches.dppowerup_delay; @@ -2832,30 +2838,30 @@ bool perform_link_training_with_retries( dp_set_panel_mode(link, panel_mode); if (link->aux_access_disabled) { - dc_link_dp_perform_link_training_skip_aux(link, &pipe_ctx->link_res, ¤t_setting); + dc_link_dp_perform_link_training_skip_aux(link, &pipe_ctx->link_res, &cur_link_settings); return true; } else { /** @todo Consolidate USB4 DP and DPx.x training. */ if (link->ep_type == DISPLAY_ENDPOINT_USB4_DPIA) { status = dc_link_dpia_perform_link_training(link, &pipe_ctx->link_res, - ¤t_setting, + &cur_link_settings, skip_video_pattern); /* Transmit idle pattern once training successful. */ - if (status == LINK_TRAINING_SUCCESS) + if (status == LINK_TRAINING_SUCCESS && !is_link_bw_low) dp_set_hw_test_pattern(link, &pipe_ctx->link_res, DP_TEST_PATTERN_VIDEO_MODE, NULL, 0); } else { status = dc_link_dp_perform_link_training(link, &pipe_ctx->link_res, - ¤t_setting, + &cur_link_settings, skip_video_pattern); } dp_trace_lt_total_count_increment(link, false); dp_trace_lt_result_update(link, status, false); dp_trace_set_lt_end_timestamp(link, false); - if (status == LINK_TRAINING_SUCCESS) + if (status == LINK_TRAINING_SUCCESS && !is_link_bw_low) return true; } @@ -2866,8 +2872,9 @@ bool perform_link_training_with_retries( if (j == (attempts - 1) && link->ep_type == DISPLAY_ENDPOINT_PHY) break; - DC_LOG_WARNING("%s: Link training attempt %u of %d failed\n", - __func__, (unsigned int)j + 1, attempts); + DC_LOG_WARNING("%s: Link training attempt %u of %d failed @ rate(%d) x lane(%d)\n", + __func__, (unsigned int)j + 1, attempts, cur_link_settings.link_rate, + cur_link_settings.lane_count); dp_disable_link_phy(link, &pipe_ctx->link_res, signal); @@ -2876,27 +2883,45 @@ bool perform_link_training_with_retries( enum dc_connection_type type = dc_connection_none; dc_link_detect_sink(link, &type); - if (type == dc_connection_none) + if (type == dc_connection_none) { + DC_LOG_HW_LINK_TRAINING("%s: Aborting training because sink unplugged\n", __func__); break; - } else if (do_fallback) { + } + } + + /* Try to train again at original settings if: + * - not falling back between training attempts; + * - aborted previous attempt due to reasons other than sink unplug; + * - successfully trained but at a link rate lower than that required by stream; + * - reached minimum link bandwidth. + */ + if (!do_fallback || (status == LINK_TRAINING_ABORT) || + (status == LINK_TRAINING_SUCCESS && is_link_bw_low) || + is_link_bw_min) { + j++; + cur_link_settings = *link_setting; + delay_between_attempts += LINK_TRAINING_RETRY_DELAY; + is_link_bw_low = false; + is_link_bw_min = (cur_link_settings.link_rate <= LINK_RATE_LOW) && + (cur_link_settings.lane_count <= LANE_COUNT_ONE); + + } else if (do_fallback) { /* Try training at lower link bandwidth if doing fallback. */ uint32_t req_bw; uint32_t link_bw; - decide_fallback_link_setting(link, *link_setting, ¤t_setting, status); + decide_fallback_link_setting(link, &max_link_settings, + &cur_link_settings, status); /* Fail link training if reduced link bandwidth no longer meets * stream requirements. */ req_bw = dc_bandwidth_in_kbps_from_timing(&stream->timing); - link_bw = dc_link_bandwidth_kbps(link, ¤t_setting); + link_bw = dc_link_bandwidth_kbps(link, &cur_link_settings); if (req_bw > link_bw) break; } msleep(delay_between_attempts); - - delay_between_attempts += LINK_TRAINING_RETRY_DELAY; } - return false; } @@ -3280,7 +3305,7 @@ static bool dp_verify_link_cap( int *fail_count) { struct dc_link_settings cur_link_settings = {0}; - struct dc_link_settings initial_link_settings = *known_limit_link_setting; + struct dc_link_settings max_link_settings = *known_limit_link_setting; bool success = false; bool skip_video_pattern; enum clock_source_id dp_cs_id = get_clock_source_id(link); @@ -3289,7 +3314,7 @@ static bool dp_verify_link_cap( struct link_resource link_res; memset(&irq_data, 0, sizeof(irq_data)); - cur_link_settings = initial_link_settings; + cur_link_settings = max_link_settings; /* Grant extended timeout request */ if ((link->lttpr_mode == LTTPR_MODE_NON_TRANSPARENT) && (link->dpcd_caps.lttpr_caps.max_ext_timeout > 0)) { @@ -3332,7 +3357,7 @@ static bool dp_verify_link_cap( dp_trace_lt_result_update(link, status, true); dp_disable_link_phy(link, &link_res, link->connector_signal); } while (!success && decide_fallback_link_setting(link, - initial_link_settings, &cur_link_settings, status)); + &max_link_settings, &cur_link_settings, status)); link->verified_link_cap = success ? cur_link_settings : fail_safe_link_settings; @@ -3567,16 +3592,19 @@ static bool decide_fallback_link_setting_max_bw_policy( */ static bool decide_fallback_link_setting( struct dc_link *link, - struct dc_link_settings initial_link_settings, - struct dc_link_settings *current_link_setting, + struct dc_link_settings *max, + struct dc_link_settings *cur, enum link_training_result training_result) { - if (!current_link_setting) + if (!cur) return false; - if (dp_get_link_encoding_format(&initial_link_settings) == DP_128b_132b_ENCODING || + if (!max) + return false; + + if (dp_get_link_encoding_format(max) == DP_128b_132b_ENCODING || link->dc->debug.force_dp2_lt_fallback_method) - return decide_fallback_link_setting_max_bw_policy(link, &initial_link_settings, - current_link_setting, training_result); + return decide_fallback_link_setting_max_bw_policy(link, max, cur, + training_result); switch (training_result) { case LINK_TRAINING_CR_FAIL_LANE0: @@ -3584,28 +3612,18 @@ static bool decide_fallback_link_setting( case LINK_TRAINING_CR_FAIL_LANE23: case LINK_TRAINING_LQA_FAIL: { - if (!reached_minimum_link_rate - (current_link_setting->link_rate)) { - current_link_setting->link_rate = - reduce_link_rate( - current_link_setting->link_rate); - } else if (!reached_minimum_lane_count - (current_link_setting->lane_count)) { - current_link_setting->link_rate = - initial_link_settings.link_rate; + if (!reached_minimum_link_rate(cur->link_rate)) { + cur->link_rate = reduce_link_rate(cur->link_rate); + } else if (!reached_minimum_lane_count(cur->lane_count)) { + cur->link_rate = max->link_rate; if (training_result == LINK_TRAINING_CR_FAIL_LANE0) return false; else if (training_result == LINK_TRAINING_CR_FAIL_LANE1) - current_link_setting->lane_count = - LANE_COUNT_ONE; - else if (training_result == - LINK_TRAINING_CR_FAIL_LANE23) - current_link_setting->lane_count = - LANE_COUNT_TWO; + cur->lane_count = LANE_COUNT_ONE; + else if (training_result == LINK_TRAINING_CR_FAIL_LANE23) + cur->lane_count = LANE_COUNT_TWO; else - current_link_setting->lane_count = - reduce_lane_count( - current_link_setting->lane_count); + cur->lane_count = reduce_lane_count(cur->lane_count); } else { return false; } @@ -3613,17 +3631,17 @@ static bool decide_fallback_link_setting( } case LINK_TRAINING_EQ_FAIL_EQ: { - if (!reached_minimum_lane_count - (current_link_setting->lane_count)) { - current_link_setting->lane_count = - reduce_lane_count( - current_link_setting->lane_count); - } else if (!reached_minimum_link_rate - (current_link_setting->link_rate)) { - current_link_setting->link_rate = - reduce_link_rate( - current_link_setting->link_rate); - current_link_setting->lane_count = initial_link_settings.lane_count; + if (!reached_minimum_lane_count(cur->lane_count)) { + cur->lane_count = reduce_lane_count(cur->lane_count); + } else if (!reached_minimum_link_rate(cur->link_rate)) { + cur->link_rate = reduce_link_rate(cur->link_rate); + /* Reduce max link rate to avoid potential infinite loop. + * Needed so that any subsequent CR_FAIL fallback can't + * re-set the link rate higher than the link rate from + * the latest EQ_FAIL fallback. + */ + max->link_rate = cur->link_rate; + cur->lane_count = max->lane_count; } else { return false; } @@ -3631,12 +3649,15 @@ static bool decide_fallback_link_setting( } case LINK_TRAINING_EQ_FAIL_CR: { - if (!reached_minimum_link_rate - (current_link_setting->link_rate)) { - current_link_setting->link_rate = - reduce_link_rate( - current_link_setting->link_rate); - current_link_setting->lane_count = initial_link_settings.lane_count; + if (!reached_minimum_link_rate(cur->link_rate)) { + cur->link_rate = reduce_link_rate(cur->link_rate); + /* Reduce max link rate to avoid potential infinite loop. + * Needed so that any subsequent CR_FAIL fallback can't + * re-set the link rate higher than the link rate from + * the latest EQ_FAIL fallback. + */ + max->link_rate = cur->link_rate; + cur->lane_count = max->lane_count; } else { return false; } @@ -5097,13 +5118,16 @@ static bool dpcd_read_sink_ext_caps(struct dc_link *link) return true; } -void dp_retrieve_lttpr_cap(struct dc_link *link) +bool dp_retrieve_lttpr_cap(struct dc_link *link) { + uint8_t lttpr_dpcd_data[8]; bool allow_lttpr_non_transparent_mode = 0; + bool vbios_lttpr_enable = link->dc->caps.vbios_lttpr_enable; bool vbios_lttpr_interop = link->dc->caps.vbios_lttpr_aware; enum dc_status status = DC_ERROR_UNEXPECTED; + bool is_lttpr_present = false; - memset(link->lttpr_dpcd_data, '\0', sizeof(link->lttpr_dpcd_data)); + memset(lttpr_dpcd_data, '\0', sizeof(lttpr_dpcd_data)); if ((link->dc->config.allow_lttpr_non_transparent_mode.bits.DP2_0 && link->dpcd_caps.channel_coding_cap.bits.DP_128b_132b_SUPPORTED)) { @@ -5113,116 +5137,82 @@ void dp_retrieve_lttpr_cap(struct dc_link *link) allow_lttpr_non_transparent_mode = 1; } - link->lttpr_mode = LTTPR_MODE_NON_LTTPR; - link->lttpr_support = LTTPR_UNSUPPORTED; - /* - * Logic to determine LTTPR support + * Logic to determine LTTPR mode */ - if (vbios_lttpr_interop) - link->lttpr_support = LTTPR_SUPPORTED; - else if (link->dc->config.allow_lttpr_non_transparent_mode.raw == 0 - || !link->dc->caps.extended_aux_timeout_support) - link->lttpr_support = LTTPR_UNSUPPORTED; - else - link->lttpr_support = LTTPR_CHECK_EXT_SUPPORT; + link->lttpr_mode = LTTPR_MODE_NON_LTTPR; + if (vbios_lttpr_enable && vbios_lttpr_interop) + link->lttpr_mode = LTTPR_MODE_NON_TRANSPARENT; + else if (!vbios_lttpr_enable && vbios_lttpr_interop) { + if (allow_lttpr_non_transparent_mode) + link->lttpr_mode = LTTPR_MODE_NON_TRANSPARENT; + else + link->lttpr_mode = LTTPR_MODE_TRANSPARENT; + } else if (!vbios_lttpr_enable && !vbios_lttpr_interop) { + if (!allow_lttpr_non_transparent_mode || !link->dc->caps.extended_aux_timeout_support) + link->lttpr_mode = LTTPR_MODE_NON_LTTPR; + else + link->lttpr_mode = LTTPR_MODE_NON_TRANSPARENT; + } +#if defined(CONFIG_DRM_AMD_DC_DCN) /* Check DP tunnel LTTPR mode debug option. */ if (link->ep_type == DISPLAY_ENDPOINT_USB4_DPIA && link->dc->debug.dpia_debug.bits.force_non_lttpr) - link->lttpr_support = LTTPR_UNSUPPORTED; + link->lttpr_mode = LTTPR_MODE_NON_LTTPR; +#endif - if (link->lttpr_support > LTTPR_UNSUPPORTED) { + if (link->lttpr_mode == LTTPR_MODE_NON_TRANSPARENT || link->lttpr_mode == LTTPR_MODE_TRANSPARENT) { /* By reading LTTPR capability, RX assumes that we will enable * LTTPR extended aux timeout if LTTPR is present. */ status = core_link_read_dpcd( link, DP_LT_TUNABLE_PHY_REPEATER_FIELD_DATA_STRUCTURE_REV, - link->lttpr_dpcd_data, - sizeof(link->lttpr_dpcd_data)); - } -} - -bool dp_parse_lttpr_mode(struct dc_link *link) -{ - bool dpcd_allow_lttpr_non_transparent_mode = false; - bool is_lttpr_present = false; - - bool vbios_lttpr_enable = link->dc->caps.vbios_lttpr_enable; - - if ((link->dc->config.allow_lttpr_non_transparent_mode.bits.DP2_0 && - link->dpcd_caps.channel_coding_cap.bits.DP_128b_132b_SUPPORTED)) { - dpcd_allow_lttpr_non_transparent_mode = true; - } else if (link->dc->config.allow_lttpr_non_transparent_mode.bits.DP1_4A && - !link->dpcd_caps.channel_coding_cap.bits.DP_128b_132b_SUPPORTED) { - dpcd_allow_lttpr_non_transparent_mode = true; + lttpr_dpcd_data, + sizeof(lttpr_dpcd_data)); + + link->dpcd_caps.lttpr_caps.revision.raw = + lttpr_dpcd_data[DP_LT_TUNABLE_PHY_REPEATER_FIELD_DATA_STRUCTURE_REV - + DP_LT_TUNABLE_PHY_REPEATER_FIELD_DATA_STRUCTURE_REV]; + + link->dpcd_caps.lttpr_caps.max_link_rate = + lttpr_dpcd_data[DP_MAX_LINK_RATE_PHY_REPEATER - + DP_LT_TUNABLE_PHY_REPEATER_FIELD_DATA_STRUCTURE_REV]; + + link->dpcd_caps.lttpr_caps.phy_repeater_cnt = + lttpr_dpcd_data[DP_PHY_REPEATER_CNT - + DP_LT_TUNABLE_PHY_REPEATER_FIELD_DATA_STRUCTURE_REV]; + + link->dpcd_caps.lttpr_caps.max_lane_count = + lttpr_dpcd_data[DP_MAX_LANE_COUNT_PHY_REPEATER - + DP_LT_TUNABLE_PHY_REPEATER_FIELD_DATA_STRUCTURE_REV]; + + link->dpcd_caps.lttpr_caps.mode = + lttpr_dpcd_data[DP_PHY_REPEATER_MODE - + DP_LT_TUNABLE_PHY_REPEATER_FIELD_DATA_STRUCTURE_REV]; + + link->dpcd_caps.lttpr_caps.max_ext_timeout = + lttpr_dpcd_data[DP_PHY_REPEATER_EXTENDED_WAIT_TIMEOUT - + DP_LT_TUNABLE_PHY_REPEATER_FIELD_DATA_STRUCTURE_REV]; + link->dpcd_caps.lttpr_caps.main_link_channel_coding.raw = + lttpr_dpcd_data[DP_MAIN_LINK_CHANNEL_CODING_PHY_REPEATER - + DP_LT_TUNABLE_PHY_REPEATER_FIELD_DATA_STRUCTURE_REV]; + + link->dpcd_caps.lttpr_caps.supported_128b_132b_rates.raw = + lttpr_dpcd_data[DP_PHY_REPEATER_128B132B_RATES - + DP_LT_TUNABLE_PHY_REPEATER_FIELD_DATA_STRUCTURE_REV]; + + /* Attempt to train in LTTPR transparent mode if repeater count exceeds 8. */ + is_lttpr_present = (link->dpcd_caps.lttpr_caps.max_lane_count > 0 && + link->dpcd_caps.lttpr_caps.max_lane_count <= 4 && + link->dpcd_caps.lttpr_caps.revision.raw >= 0x14); + if (is_lttpr_present) { + CONN_DATA_DETECT(link, lttpr_dpcd_data, sizeof(lttpr_dpcd_data), "LTTPR Caps: "); + configure_lttpr_mode_transparent(link); + } else + link->lttpr_mode = LTTPR_MODE_NON_LTTPR; } - - /* - * Logic to determine LTTPR mode - */ - if (link->lttpr_support == LTTPR_SUPPORTED) - if (vbios_lttpr_enable) - link->lttpr_mode = LTTPR_MODE_NON_TRANSPARENT; - else if (dpcd_allow_lttpr_non_transparent_mode) - link->lttpr_mode = LTTPR_MODE_NON_TRANSPARENT; - else - link->lttpr_mode = LTTPR_MODE_TRANSPARENT; - else // lttpr_support == LTTPR_CHECK_EXT_SUPPORT - if (dpcd_allow_lttpr_non_transparent_mode) { - link->lttpr_support = LTTPR_SUPPORTED; - link->lttpr_mode = LTTPR_MODE_NON_TRANSPARENT; - } else { - link->lttpr_support = LTTPR_UNSUPPORTED; - } - - if (link->lttpr_support == LTTPR_UNSUPPORTED) - return false; - - link->dpcd_caps.lttpr_caps.revision.raw = - link->lttpr_dpcd_data[DP_LT_TUNABLE_PHY_REPEATER_FIELD_DATA_STRUCTURE_REV - - DP_LT_TUNABLE_PHY_REPEATER_FIELD_DATA_STRUCTURE_REV]; - - link->dpcd_caps.lttpr_caps.max_link_rate = - link->lttpr_dpcd_data[DP_MAX_LINK_RATE_PHY_REPEATER - - DP_LT_TUNABLE_PHY_REPEATER_FIELD_DATA_STRUCTURE_REV]; - - link->dpcd_caps.lttpr_caps.phy_repeater_cnt = - link->lttpr_dpcd_data[DP_PHY_REPEATER_CNT - - DP_LT_TUNABLE_PHY_REPEATER_FIELD_DATA_STRUCTURE_REV]; - - link->dpcd_caps.lttpr_caps.max_lane_count = - link->lttpr_dpcd_data[DP_MAX_LANE_COUNT_PHY_REPEATER - - DP_LT_TUNABLE_PHY_REPEATER_FIELD_DATA_STRUCTURE_REV]; - - link->dpcd_caps.lttpr_caps.mode = - link->lttpr_dpcd_data[DP_PHY_REPEATER_MODE - - DP_LT_TUNABLE_PHY_REPEATER_FIELD_DATA_STRUCTURE_REV]; - - link->dpcd_caps.lttpr_caps.max_ext_timeout = - link->lttpr_dpcd_data[DP_PHY_REPEATER_EXTENDED_WAIT_TIMEOUT - - DP_LT_TUNABLE_PHY_REPEATER_FIELD_DATA_STRUCTURE_REV]; - - link->dpcd_caps.lttpr_caps.main_link_channel_coding.raw = - link->lttpr_dpcd_data[DP_MAIN_LINK_CHANNEL_CODING_PHY_REPEATER - - DP_LT_TUNABLE_PHY_REPEATER_FIELD_DATA_STRUCTURE_REV]; - - link->dpcd_caps.lttpr_caps.supported_128b_132b_rates.raw = - link->lttpr_dpcd_data[DP_PHY_REPEATER_128B132B_RATES - - DP_LT_TUNABLE_PHY_REPEATER_FIELD_DATA_STRUCTURE_REV]; - - - /* Attempt to train in LTTPR transparent mode if repeater count exceeds 8. */ - is_lttpr_present = (link->dpcd_caps.lttpr_caps.max_lane_count > 0 && - link->dpcd_caps.lttpr_caps.max_lane_count <= 4 && - link->dpcd_caps.lttpr_caps.revision.raw >= 0x14); - if (is_lttpr_present) { - CONN_DATA_DETECT(link, link->lttpr_dpcd_data, sizeof(link->lttpr_dpcd_data), "LTTPR Caps: "); - configure_lttpr_mode_transparent(link); - } else - link->lttpr_mode = LTTPR_MODE_NON_LTTPR; - return is_lttpr_present; } @@ -5374,8 +5364,7 @@ static bool retrieve_link_cap(struct dc_link *link) status = wa_try_to_wake_dprx(link, timeout_ms); } - dp_retrieve_lttpr_cap(link); - + is_lttpr_present = dp_retrieve_lttpr_cap(link); /* Read DP tunneling information. */ status = dpcd_get_tunneling_device_data(link); @@ -5411,9 +5400,6 @@ static bool retrieve_link_cap(struct dc_link *link) return false; } - if (link->lttpr_support > LTTPR_UNSUPPORTED) - is_lttpr_present = dp_parse_lttpr_mode(link); - if (!is_lttpr_present) dc_link_aux_try_to_configure_timeout(link->ddc, LINK_AUX_DEFAULT_TIMEOUT_PERIOD); diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_dpia.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_dpia.c index a5765f36d86f..1b7a8774b0c9 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link_dpia.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_dpia.c @@ -34,6 +34,7 @@ #include "dm_helpers.h" #include "dmub/inc/dmub_cmd.h" #include "inc/link_dpcd.h" +#include "dc_dmub_srv.h" #define DC_LOGGER \ link->ctx->logger @@ -69,6 +70,24 @@ enum dc_status dpcd_get_tunneling_device_data(struct dc_link *link) return status; } +bool dc_link_dpia_query_hpd_status(struct dc_link *link) +{ + union dmub_rb_cmd cmd = {0}; + struct dc_dmub_srv *dmub_srv = link->ctx->dmub_srv; + bool is_hpd_high = false; + + /* prepare QUERY_HPD command */ + cmd.query_hpd.header.type = DMUB_CMD__QUERY_HPD_STATE; + cmd.query_hpd.data.instance = link->link_id.enum_id - ENUM_ID_1; + cmd.query_hpd.data.ch_type = AUX_CHANNEL_DPIA; + + /* Return HPD status reported by DMUB if query successfully executed. */ + if (dc_dmub_srv_cmd_with_reply_data(dmub_srv, &cmd) && cmd.query_hpd.data.status == AUX_RET_SUCCESS) + is_hpd_high = cmd.query_hpd.data.result; + + return is_hpd_high; +} + /* Configure link as prescribed in link_setting; set LTTPR mode; and * Initialize link training settings. * Abort link training if sink unplug detected. diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_surface.c b/drivers/gpu/drm/amd/display/dc/core/dc_surface.c index e6b9c6a71841..5bc6ff2fa73e 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_surface.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_surface.c @@ -61,6 +61,8 @@ static void dc_plane_construct(struct dc_context *ctx, struct dc_plane_state *pl plane_state->blend_tf->type = TF_TYPE_BYPASS; } + plane_state->pre_multiplied_alpha = true; + } static void dc_plane_destruct(struct dc_plane_state *plane_state) diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h index 26c24db8f1da..817028d3c4a0 100644 --- a/drivers/gpu/drm/amd/display/dc/dc.h +++ b/drivers/gpu/drm/amd/display/dc/dc.h @@ -47,7 +47,7 @@ struct aux_payload; struct set_config_cmd_payload; struct dmub_notification; -#define DC_VER "3.2.185" +#define DC_VER "3.2.187" #define MAX_SURFACES 3 #define MAX_PLANES 6 @@ -329,7 +329,7 @@ struct dc_config { bool disable_dmcu; bool enable_4to1MPC; bool enable_windowed_mpo_odm; - bool allow_edp_hotplug_detection; + uint32_t allow_edp_hotplug_detection; bool clamp_min_dcfclk; uint64_t vblank_alignment_dto_params; uint8_t vblank_alignment_max_frame_time_diff; @@ -416,6 +416,7 @@ struct dc_clocks { bool p_state_change_support; enum dcn_zstate_support_state zstate_support; bool dtbclk_en; + int ref_dtbclk_khz; enum dcn_pwr_state pwr_state; /* * Elements below are not compared for the purposes of @@ -719,6 +720,8 @@ struct dc_debug_options { bool apply_vendor_specific_lttpr_wa; bool extended_blank_optimization; union aux_wake_wa_options aux_wake_wa; + /* uses value at boot and disables switch */ + bool disable_dtb_ref_clk_switch; uint8_t psr_power_use_phy_fsm; enum dml_hostvm_override_opts dml_hostvm_override; }; @@ -1011,6 +1014,7 @@ struct dc_plane_state { bool is_tiling_rotated; bool per_pixel_alpha; + bool pre_multiplied_alpha; bool global_alpha; int global_alpha_value; bool visible; @@ -1045,6 +1049,7 @@ struct dc_plane_info { bool horizontal_mirror; bool visible; bool per_pixel_alpha; + bool pre_multiplied_alpha; bool global_alpha; int global_alpha_value; bool input_csc_enabled; diff --git a/drivers/gpu/drm/amd/display/dc/dc_link.h b/drivers/gpu/drm/amd/display/dc/dc_link.h index 251f2bbc96b9..a3c37ee3f849 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_link.h +++ b/drivers/gpu/drm/amd/display/dc/dc_link.h @@ -129,8 +129,6 @@ struct dc_link { bool link_state_valid; bool aux_access_disabled; bool sync_lt_in_progress; - uint8_t lttpr_dpcd_data[8]; - enum lttpr_support lttpr_support; enum lttpr_mode lttpr_mode; bool is_internal_display; diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c b/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c index 29e20d92b0bb..9e39cd7b203e 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c @@ -87,7 +87,8 @@ static void release_engine( engine->ddc = NULL; - REG_UPDATE(AUX_ARB_CONTROL, AUX_SW_DONE_USING_AUX_REG, 1); + REG_UPDATE_2(AUX_ARB_CONTROL, AUX_SW_DONE_USING_AUX_REG, 1, + AUX_SW_USE_AUX_REG_REQ, 0); } #define SW_CAN_ACCESS_AUX 1 diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c b/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c index 5e6fea85a7b5..845aa8a1027d 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c @@ -1101,9 +1101,12 @@ static bool get_pixel_clk_frequency_100hz( * not be programmed equal to DPREFCLK */ modulo_hz = REG_READ(MODULO[inst]); - *pixel_clk_khz = div_u64((uint64_t)clock_hz* - clock_source->ctx->dc->clk_mgr->dprefclk_khz*10, - modulo_hz); + if (modulo_hz) + *pixel_clk_khz = div_u64((uint64_t)clock_hz* + clock_source->ctx->dc->clk_mgr->dprefclk_khz*10, + modulo_hz); + else + *pixel_clk_khz = 0; } else { /* NOTE: There is agreement with VBIOS here that MODULO is * programmed equal to DPREFCLK, in which case PHASE will be diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c index e02ac75afbf7..e3a62873c0e7 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c @@ -2550,12 +2550,21 @@ void dcn10_update_mpcc(struct dc *dc, struct pipe_ctx *pipe_ctx) blnd_cfg.overlap_only = false; blnd_cfg.global_gain = 0xff; - if (per_pixel_alpha && pipe_ctx->plane_state->global_alpha) { - blnd_cfg.alpha_mode = MPCC_ALPHA_BLEND_MODE_PER_PIXEL_ALPHA_COMBINED_GLOBAL_GAIN; - blnd_cfg.global_gain = pipe_ctx->plane_state->global_alpha_value; - } else if (per_pixel_alpha) { - blnd_cfg.alpha_mode = MPCC_ALPHA_BLEND_MODE_PER_PIXEL_ALPHA; + if (per_pixel_alpha) { + /* DCN1.0 has output CM before MPC which seems to screw with + * pre-multiplied alpha. + */ + blnd_cfg.pre_multiplied_alpha = (is_rgb_cspace( + pipe_ctx->stream->output_color_space) + && pipe_ctx->plane_state->pre_multiplied_alpha); + if (pipe_ctx->plane_state->global_alpha) { + blnd_cfg.alpha_mode = MPCC_ALPHA_BLEND_MODE_PER_PIXEL_ALPHA_COMBINED_GLOBAL_GAIN; + blnd_cfg.global_gain = pipe_ctx->plane_state->global_alpha_value; + } else { + blnd_cfg.alpha_mode = MPCC_ALPHA_BLEND_MODE_PER_PIXEL_ALPHA; + } } else { + blnd_cfg.pre_multiplied_alpha = false; blnd_cfg.alpha_mode = MPCC_ALPHA_BLEND_MODE_GLOBAL_ALPHA; } @@ -2564,14 +2573,6 @@ void dcn10_update_mpcc(struct dc *dc, struct pipe_ctx *pipe_ctx) else blnd_cfg.global_alpha = 0xff; - /* DCN1.0 has output CM before MPC which seems to screw with - * pre-multiplied alpha. - */ - blnd_cfg.pre_multiplied_alpha = is_rgb_cspace( - pipe_ctx->stream->output_color_space) - && per_pixel_alpha; - - /* * TODO: remove hack * Note: currently there is a bug in init_hw such that diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c index e1f87bd72e4a..ec6aa8d8b251 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c @@ -1773,7 +1773,6 @@ void dcn20_post_unlock_program_front_end( */ for (i = 0; i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; - if (pipe->plane_state && !pipe->top_pipe && pipe->update_flags.bits.enable) { struct hubp *hubp = pipe->plane_res.hubp; int j = 0; @@ -2346,12 +2345,16 @@ void dcn20_update_mpcc(struct dc *dc, struct pipe_ctx *pipe_ctx) blnd_cfg.overlap_only = false; blnd_cfg.global_gain = 0xff; - if (per_pixel_alpha && pipe_ctx->plane_state->global_alpha) { - blnd_cfg.alpha_mode = MPCC_ALPHA_BLEND_MODE_PER_PIXEL_ALPHA_COMBINED_GLOBAL_GAIN; - blnd_cfg.global_gain = pipe_ctx->plane_state->global_alpha_value; - } else if (per_pixel_alpha) { - blnd_cfg.alpha_mode = MPCC_ALPHA_BLEND_MODE_PER_PIXEL_ALPHA; + if (per_pixel_alpha) { + blnd_cfg.pre_multiplied_alpha = pipe_ctx->plane_state->pre_multiplied_alpha; + if (pipe_ctx->plane_state->global_alpha) { + blnd_cfg.alpha_mode = MPCC_ALPHA_BLEND_MODE_PER_PIXEL_ALPHA_COMBINED_GLOBAL_GAIN; + blnd_cfg.global_gain = pipe_ctx->plane_state->global_alpha_value; + } else { + blnd_cfg.alpha_mode = MPCC_ALPHA_BLEND_MODE_PER_PIXEL_ALPHA; + } } else { + blnd_cfg.pre_multiplied_alpha = false; blnd_cfg.alpha_mode = MPCC_ALPHA_BLEND_MODE_GLOBAL_ALPHA; } @@ -2365,7 +2368,7 @@ void dcn20_update_mpcc(struct dc *dc, struct pipe_ctx *pipe_ctx) blnd_cfg.top_gain = 0x1f000; blnd_cfg.bottom_inside_gain = 0x1f000; blnd_cfg.bottom_outside_gain = 0x1f000; - blnd_cfg.pre_multiplied_alpha = per_pixel_alpha; + if (pipe_ctx->plane_state->format == SURFACE_PIXEL_FORMAT_GRPH_RGBE_ALPHA) blnd_cfg.pre_multiplied_alpha = false; diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_optc.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_optc.c index f5e8916601d3..b604fb26f288 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_optc.c +++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_optc.c @@ -28,6 +28,8 @@ #include "dc.h" #include "dcn_calc_math.h" +#include "dml/dcn30/dcn30_fpu.h" + #define REG(reg)\ optc1->tg_regs->reg @@ -184,6 +186,14 @@ void optc3_set_dsc_config(struct timing_generator *optc, } +void optc3_set_vrr_m_const(struct timing_generator *optc, + double vtotal_avg) +{ + DC_FP_START(); + optc3_fpu_set_vrr_m_const(optc, vtotal_avg); + DC_FP_END(); +} + void optc3_set_odm_bypass(struct timing_generator *optc, const struct dc_crtc_timing *dc_crtc_timing) { diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c index 336b2ce6a636..1c1a67c4cec1 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c @@ -84,6 +84,7 @@ #include "dce/dce_aux.h" #include "dce/dce_i2c.h" +#include "dml/dcn30/dcn30_fpu.h" #include "dml/dcn30/display_mode_vba_30.h" #include "vm_helper.h" #include "dcn20/dcn20_vmid.h" @@ -91,137 +92,6 @@ #define DC_LOGGER_INIT(logger) -struct _vcs_dpi_ip_params_st dcn3_0_ip = { - .use_min_dcfclk = 0, - .clamp_min_dcfclk = 0, - .odm_capable = 1, - .gpuvm_enable = 0, - .hostvm_enable = 0, - .gpuvm_max_page_table_levels = 4, - .hostvm_max_page_table_levels = 4, - .hostvm_cached_page_table_levels = 0, - .pte_group_size_bytes = 2048, - .num_dsc = 6, - .rob_buffer_size_kbytes = 184, - .det_buffer_size_kbytes = 184, - .dpte_buffer_size_in_pte_reqs_luma = 84, - .pde_proc_buffer_size_64k_reqs = 48, - .dpp_output_buffer_pixels = 2560, - .opp_output_buffer_lines = 1, - .pixel_chunk_size_kbytes = 8, - .pte_enable = 1, - .max_page_table_levels = 2, - .pte_chunk_size_kbytes = 2, // ? - .meta_chunk_size_kbytes = 2, - .writeback_chunk_size_kbytes = 8, - .line_buffer_size_bits = 789504, - .is_line_buffer_bpp_fixed = 0, // ? - .line_buffer_fixed_bpp = 0, // ? - .dcc_supported = true, - .writeback_interface_buffer_size_kbytes = 90, - .writeback_line_buffer_buffer_size = 0, - .max_line_buffer_lines = 12, - .writeback_luma_buffer_size_kbytes = 12, // writeback_line_buffer_buffer_size = 656640 - .writeback_chroma_buffer_size_kbytes = 8, - .writeback_chroma_line_buffer_width_pixels = 4, - .writeback_max_hscl_ratio = 1, - .writeback_max_vscl_ratio = 1, - .writeback_min_hscl_ratio = 1, - .writeback_min_vscl_ratio = 1, - .writeback_max_hscl_taps = 1, - .writeback_max_vscl_taps = 1, - .writeback_line_buffer_luma_buffer_size = 0, - .writeback_line_buffer_chroma_buffer_size = 14643, - .cursor_buffer_size = 8, - .cursor_chunk_size = 2, - .max_num_otg = 6, - .max_num_dpp = 6, - .max_num_wb = 1, - .max_dchub_pscl_bw_pix_per_clk = 4, - .max_pscl_lb_bw_pix_per_clk = 2, - .max_lb_vscl_bw_pix_per_clk = 4, - .max_vscl_hscl_bw_pix_per_clk = 4, - .max_hscl_ratio = 6, - .max_vscl_ratio = 6, - .hscl_mults = 4, - .vscl_mults = 4, - .max_hscl_taps = 8, - .max_vscl_taps = 8, - .dispclk_ramp_margin_percent = 1, - .underscan_factor = 1.11, - .min_vblank_lines = 32, - .dppclk_delay_subtotal = 46, - .dynamic_metadata_vm_enabled = true, - .dppclk_delay_scl_lb_only = 16, - .dppclk_delay_scl = 50, - .dppclk_delay_cnvc_formatter = 27, - .dppclk_delay_cnvc_cursor = 6, - .dispclk_delay_subtotal = 119, - .dcfclk_cstate_latency = 5.2, // SRExitTime - .max_inter_dcn_tile_repeaters = 8, - .odm_combine_4to1_supported = true, - - .xfc_supported = false, - .xfc_fill_bw_overhead_percent = 10.0, - .xfc_fill_constant_bytes = 0, - .gfx7_compat_tiling_supported = 0, - .number_of_cursors = 1, -}; - -struct _vcs_dpi_soc_bounding_box_st dcn3_0_soc = { - .clock_limits = { - { - .state = 0, - .dispclk_mhz = 562.0, - .dppclk_mhz = 300.0, - .phyclk_mhz = 300.0, - .phyclk_d18_mhz = 667.0, - .dscclk_mhz = 405.6, - }, - }, - .min_dcfclk = 500.0, /* TODO: set this to actual min DCFCLK */ - .num_states = 1, - .sr_exit_time_us = 15.5, - .sr_enter_plus_exit_time_us = 20, - .urgent_latency_us = 4.0, - .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_out_of_order_return_per_channel_pixel_only_bytes = 4096, - .urgent_out_of_order_return_per_channel_pixel_and_vm_bytes = 4096, - .urgent_out_of_order_return_per_channel_vm_only_bytes = 4096, - .pct_ideal_dram_sdp_bw_after_urgent_pixel_only = 80.0, - .pct_ideal_dram_sdp_bw_after_urgent_pixel_and_vm = 60.0, - .pct_ideal_dram_sdp_bw_after_urgent_vm_only = 40.0, - .max_avg_sdp_bw_use_normal_percent = 60.0, - .max_avg_dram_bw_use_normal_percent = 40.0, - .writeback_latency_us = 12.0, - .max_request_size_bytes = 256, - .fabric_datapath_to_dcn_data_return_bytes = 64, - .dcn_downspread_percent = 0.5, - .downspread_percent = 0.38, - .dram_page_open_time_ns = 50.0, - .dram_rw_turnaround_time_ns = 17.5, - .dram_return_buffer_per_channel_bytes = 8192, - .round_trip_ping_latency_dcfclk_cycles = 191, - .urgent_out_of_order_return_per_channel_bytes = 4096, - .channel_interleave_bytes = 256, - .num_banks = 8, - .gpuvm_min_page_size_bytes = 4096, - .hostvm_min_page_size_bytes = 4096, - .dram_clock_change_latency_us = 404, - .dummy_pstate_latency_us = 5, - .writeback_dram_clock_change_latency_us = 23.0, - .return_bus_width_bytes = 64, - .dispclk_dppclk_vco_speed_mhz = 3650, - .xfc_bus_transport_time_us = 20, // ? - .xfc_xbuf_latency_tolerance_us = 4, // ? - .use_urgent_burst_bw = 1, // ? - .do_urgent_latency_adjustment = true, - .urgent_latency_adjustment_fabric_clock_component_us = 1.0, - .urgent_latency_adjustment_fabric_clock_reference_mhz = 1000, -}; - enum dcn30_clk_src_array_id { DCN30_CLK_SRC_PLL0, DCN30_CLK_SRC_PLL1, @@ -1480,90 +1350,9 @@ int dcn30_populate_dml_pipes_from_context( void dcn30_populate_dml_writeback_from_context( struct dc *dc, struct resource_context *res_ctx, display_e2e_pipe_params_st *pipes) { - int pipe_cnt, i, j; - double max_calc_writeback_dispclk; - double writeback_dispclk; - struct writeback_st dout_wb; - - for (i = 0, pipe_cnt = 0; i < dc->res_pool->pipe_count; i++) { - struct dc_stream_state *stream = res_ctx->pipe_ctx[i].stream; - - if (!stream) - continue; - max_calc_writeback_dispclk = 0; - - /* Set writeback information */ - pipes[pipe_cnt].dout.wb_enable = 0; - pipes[pipe_cnt].dout.num_active_wb = 0; - for (j = 0; j < stream->num_wb_info; j++) { - struct dc_writeback_info *wb_info = &stream->writeback_info[j]; - - if (wb_info->wb_enabled && wb_info->writeback_source_plane && - (wb_info->writeback_source_plane == res_ctx->pipe_ctx[i].plane_state)) { - pipes[pipe_cnt].dout.wb_enable = 1; - pipes[pipe_cnt].dout.num_active_wb++; - dout_wb.wb_src_height = wb_info->dwb_params.cnv_params.crop_en ? - wb_info->dwb_params.cnv_params.crop_height : - wb_info->dwb_params.cnv_params.src_height; - dout_wb.wb_src_width = wb_info->dwb_params.cnv_params.crop_en ? - wb_info->dwb_params.cnv_params.crop_width : - wb_info->dwb_params.cnv_params.src_width; - dout_wb.wb_dst_width = wb_info->dwb_params.dest_width; - dout_wb.wb_dst_height = wb_info->dwb_params.dest_height; - - /* For IP that doesn't support WB scaling, set h/v taps to 1 to avoid DML validation failure */ - if (dc->dml.ip.writeback_max_hscl_taps > 1) { - dout_wb.wb_htaps_luma = wb_info->dwb_params.scaler_taps.h_taps; - dout_wb.wb_vtaps_luma = wb_info->dwb_params.scaler_taps.v_taps; - } else { - dout_wb.wb_htaps_luma = 1; - dout_wb.wb_vtaps_luma = 1; - } - dout_wb.wb_htaps_chroma = 0; - dout_wb.wb_vtaps_chroma = 0; - dout_wb.wb_hratio = wb_info->dwb_params.cnv_params.crop_en ? - (double)wb_info->dwb_params.cnv_params.crop_width / - (double)wb_info->dwb_params.dest_width : - (double)wb_info->dwb_params.cnv_params.src_width / - (double)wb_info->dwb_params.dest_width; - dout_wb.wb_vratio = wb_info->dwb_params.cnv_params.crop_en ? - (double)wb_info->dwb_params.cnv_params.crop_height / - (double)wb_info->dwb_params.dest_height : - (double)wb_info->dwb_params.cnv_params.src_height / - (double)wb_info->dwb_params.dest_height; - if (wb_info->dwb_params.cnv_params.fc_out_format == DWB_OUT_FORMAT_64BPP_ARGB || - wb_info->dwb_params.cnv_params.fc_out_format == DWB_OUT_FORMAT_64BPP_RGBA) - dout_wb.wb_pixel_format = dm_444_64; - else - dout_wb.wb_pixel_format = dm_444_32; - - /* Workaround for cases where multiple writebacks are connected to same plane - * In which case, need to compute worst case and set the associated writeback parameters - * This workaround is necessary due to DML computation assuming only 1 set of writeback - * parameters per pipe - */ - writeback_dispclk = dml30_CalculateWriteBackDISPCLK( - dout_wb.wb_pixel_format, - pipes[pipe_cnt].pipe.dest.pixel_rate_mhz, - dout_wb.wb_hratio, - dout_wb.wb_vratio, - dout_wb.wb_htaps_luma, - dout_wb.wb_vtaps_luma, - dout_wb.wb_src_width, - dout_wb.wb_dst_width, - pipes[pipe_cnt].pipe.dest.htotal, - dc->current_state->bw_ctx.dml.ip.writeback_line_buffer_buffer_size); - - if (writeback_dispclk > max_calc_writeback_dispclk) { - max_calc_writeback_dispclk = writeback_dispclk; - pipes[pipe_cnt].dout.wb = dout_wb; - } - } - } - - pipe_cnt++; - } - + DC_FP_START(); + dcn30_fpu_populate_dml_writeback_from_context(dc, res_ctx, pipes); + DC_FP_END(); } unsigned int dcn30_calc_max_scaled_time( @@ -1598,7 +1387,7 @@ void dcn30_set_mcif_arb_params( enum mmhubbub_wbif_mode wbif_mode; struct display_mode_lib *dml = &context->bw_ctx.dml; struct mcif_arb_params *wb_arb_params; - int i, j, k, dwb_pipe; + int i, j, dwb_pipe; /* Writeback MCIF_WB arbitration parameters */ dwb_pipe = 0; @@ -1622,17 +1411,15 @@ void dcn30_set_mcif_arb_params( else wbif_mode = PACKED_444; - for (k = 0; k < sizeof(wb_arb_params->cli_watermark)/sizeof(wb_arb_params->cli_watermark[0]); k++) { - wb_arb_params->cli_watermark[k] = get_wm_writeback_urgent(dml, pipes, pipe_cnt) * 1000; - wb_arb_params->pstate_watermark[k] = get_wm_writeback_dram_clock_change(dml, pipes, pipe_cnt) * 1000; - } + DC_FP_START(); + dcn30_fpu_set_mcif_arb_params(wb_arb_params, dml, pipes, pipe_cnt, j); + DC_FP_END(); wb_arb_params->time_per_pixel = (1000000 << 6) / context->res_ctx.pipe_ctx[i].stream->phy_pix_clk; /* time_per_pixel should be in u6.6 format */ wb_arb_params->slice_lines = 32; wb_arb_params->arbitration_slice = 2; /* irrelevant since there is no YUV output */ wb_arb_params->max_scaled_time = dcn30_calc_max_scaled_time(wb_arb_params->time_per_pixel, wbif_mode, wb_arb_params->cli_watermark[0]); /* assume 4 watermark sets have the same value */ - wb_arb_params->dram_speed_change_duration = dml->vba.WritebackAllowDRAMClockChangeEndPosition[j] * pipes[0].clks_cfg.refclk_mhz; /* num_clock_cycles = us * MHz */ dwb_pipe++; @@ -2111,178 +1898,11 @@ validate_out: return out; } -/* - * This must be noinline to ensure anything that deals with FP registers - * is contained within this call; previously our compiling with hard-float - * would result in fp instructions being emitted outside of the boundaries - * of the DC_FP_START/END macros, which makes sense as the compiler has no - * idea about what is wrapped and what is not - * - * This is largely just a workaround to avoid breakage introduced with 5.6, - * ideally all fp-using code should be moved into its own file, only that - * should be compiled with hard-float, and all code exported from there - * should be strictly wrapped with DC_FP_START/END - */ -static noinline void dcn30_calculate_wm_and_dlg_fp( - struct dc *dc, struct dc_state *context, - display_e2e_pipe_params_st *pipes, - int pipe_cnt, - int vlevel) +void dcn30_update_soc_for_wm_a(struct dc *dc, struct dc_state *context) { - int maxMpcComb = context->bw_ctx.dml.vba.maxMpcComb; - int i, pipe_idx; - double dcfclk = context->bw_ctx.dml.vba.DCFCLKState[vlevel][maxMpcComb]; - bool pstate_en = context->bw_ctx.dml.vba.DRAMClockChangeSupport[vlevel][maxMpcComb] != dm_dram_clock_change_unsupported; - - if (context->bw_ctx.dml.soc.min_dcfclk > dcfclk) - dcfclk = context->bw_ctx.dml.soc.min_dcfclk; - - pipes[0].clks_cfg.voltage = vlevel; - pipes[0].clks_cfg.dcfclk_mhz = dcfclk; - pipes[0].clks_cfg.socclk_mhz = context->bw_ctx.dml.soc.clock_limits[vlevel].socclk_mhz; - - /* Set B: - * DCFCLK: 1GHz or min required above 1GHz - * FCLK/UCLK: Max - */ - if (dc->clk_mgr->bw_params->wm_table.nv_entries[WM_B].valid) { - if (vlevel == 0) { - pipes[0].clks_cfg.voltage = 1; - pipes[0].clks_cfg.dcfclk_mhz = context->bw_ctx.dml.soc.clock_limits[0].dcfclk_mhz; - } - context->bw_ctx.dml.soc.dram_clock_change_latency_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_B].dml_input.pstate_latency_us; - context->bw_ctx.dml.soc.sr_enter_plus_exit_time_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_B].dml_input.sr_enter_plus_exit_time_us; - context->bw_ctx.dml.soc.sr_exit_time_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_B].dml_input.sr_exit_time_us; - } - context->bw_ctx.bw.dcn.watermarks.b.urgent_ns = get_wm_urgent(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.b.cstate_pstate.cstate_enter_plus_exit_ns = get_wm_stutter_enter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.b.cstate_pstate.cstate_exit_ns = get_wm_stutter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.b.cstate_pstate.pstate_change_ns = get_wm_dram_clock_change(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.b.pte_meta_urgent_ns = get_wm_memory_trip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.b.frac_urg_bw_nom = get_fraction_of_urgent_bandwidth(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.b.frac_urg_bw_flip = get_fraction_of_urgent_bandwidth_imm_flip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.b.urgent_latency_ns = get_urgent_latency(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - - pipes[0].clks_cfg.voltage = vlevel; - pipes[0].clks_cfg.dcfclk_mhz = dcfclk; - - /* Set D: - * DCFCLK: Min Required - * FCLK(proportional to UCLK): 1GHz or Max - * MALL stutter, sr_enter_exit = 4, sr_exit = 2us - */ - /* - if (dc->clk_mgr->bw_params->wm_table.nv_entries[WM_D].valid) { - context->bw_ctx.dml.soc.dram_clock_change_latency_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_D].dml_input.pstate_latency_us; - context->bw_ctx.dml.soc.sr_enter_plus_exit_time_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_D].dml_input.sr_enter_plus_exit_time_us; - context->bw_ctx.dml.soc.sr_exit_time_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_D].dml_input.sr_exit_time_us; - } - context->bw_ctx.bw.dcn.watermarks.d.urgent_ns = get_wm_urgent(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.d.cstate_pstate.cstate_enter_plus_exit_ns = get_wm_stutter_enter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.d.cstate_pstate.cstate_exit_ns = get_wm_stutter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.d.cstate_pstate.pstate_change_ns = get_wm_dram_clock_change(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.d.pte_meta_urgent_ns = get_wm_memory_trip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.d.frac_urg_bw_nom = get_fraction_of_urgent_bandwidth(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.d.frac_urg_bw_flip = get_fraction_of_urgent_bandwidth_imm_flip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.d.urgent_latency_ns = get_urgent_latency(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - */ - - /* Set C: - * DCFCLK: Min Required - * FCLK(proportional to UCLK): 1GHz or Max - * pstate latency overridden to 5us - */ - if (dc->clk_mgr->bw_params->wm_table.nv_entries[WM_C].valid) { - unsigned int min_dram_speed_mts = context->bw_ctx.dml.vba.DRAMSpeed; - unsigned int min_dram_speed_mts_margin = 160; - - if (context->bw_ctx.dml.vba.DRAMClockChangeSupport[vlevel][context->bw_ctx.dml.vba.maxMpcComb] == dm_dram_clock_change_unsupported) - min_dram_speed_mts = dc->clk_mgr->bw_params->clk_table.entries[dc->clk_mgr->bw_params->clk_table.num_entries - 1].memclk_mhz * 16; - - /* find largest table entry that is lower than dram speed, but lower than DPM0 still uses DPM0 */ - for (i = 3; i > 0; i--) - if (min_dram_speed_mts + min_dram_speed_mts_margin > dc->clk_mgr->bw_params->dummy_pstate_table[i].dram_speed_mts) - break; - - context->bw_ctx.dml.soc.dram_clock_change_latency_us = dc->clk_mgr->bw_params->dummy_pstate_table[i].dummy_pstate_latency_us; - context->bw_ctx.dml.soc.sr_enter_plus_exit_time_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_C].dml_input.sr_enter_plus_exit_time_us; - context->bw_ctx.dml.soc.sr_exit_time_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_C].dml_input.sr_exit_time_us; - } - - context->bw_ctx.bw.dcn.watermarks.c.urgent_ns = get_wm_urgent(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.c.cstate_pstate.cstate_enter_plus_exit_ns = get_wm_stutter_enter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.c.cstate_pstate.cstate_exit_ns = get_wm_stutter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.c.cstate_pstate.pstate_change_ns = get_wm_dram_clock_change(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.c.pte_meta_urgent_ns = get_wm_memory_trip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.c.frac_urg_bw_nom = get_fraction_of_urgent_bandwidth(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.c.frac_urg_bw_flip = get_fraction_of_urgent_bandwidth_imm_flip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.c.urgent_latency_ns = get_urgent_latency(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - - if (!pstate_en) { - /* The only difference between A and C is p-state latency, if p-state is not supported we want to - * calculate DLG based on dummy p-state latency, and max out the set A p-state watermark - */ - context->bw_ctx.bw.dcn.watermarks.a = context->bw_ctx.bw.dcn.watermarks.c; - context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.pstate_change_ns = 0; - } else { - /* Set A: - * DCFCLK: Min Required - * FCLK(proportional to UCLK): 1GHz or Max - * - * Set A calculated last so that following calculations are based on Set A - */ - dc->res_pool->funcs->update_soc_for_wm_a(dc, context); - context->bw_ctx.bw.dcn.watermarks.a.urgent_ns = get_wm_urgent(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.cstate_enter_plus_exit_ns = get_wm_stutter_enter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.cstate_exit_ns = get_wm_stutter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.pstate_change_ns = get_wm_dram_clock_change(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.a.pte_meta_urgent_ns = get_wm_memory_trip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.a.frac_urg_bw_nom = get_fraction_of_urgent_bandwidth(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.a.frac_urg_bw_flip = get_fraction_of_urgent_bandwidth_imm_flip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.a.urgent_latency_ns = get_urgent_latency(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - } - - context->perf_params.stutter_period_us = context->bw_ctx.dml.vba.StutterPeriod; - - /* Make set D = set A until set D is enabled */ - context->bw_ctx.bw.dcn.watermarks.d = context->bw_ctx.bw.dcn.watermarks.a; - - for (i = 0, pipe_idx = 0; i < dc->res_pool->pipe_count; i++) { - if (!context->res_ctx.pipe_ctx[i].stream) - continue; - - 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); - - if (dc->config.forced_clocks) { - pipes[pipe_idx].clks_cfg.dispclk_mhz = context->bw_ctx.dml.soc.clock_limits[0].dispclk_mhz; - pipes[pipe_idx].clks_cfg.dppclk_mhz = context->bw_ctx.dml.soc.clock_limits[0].dppclk_mhz; - } - if (dc->debug.min_disp_clk_khz > pipes[pipe_idx].clks_cfg.dispclk_mhz * 1000) - pipes[pipe_idx].clks_cfg.dispclk_mhz = dc->debug.min_disp_clk_khz / 1000.0; - if (dc->debug.min_dpp_clk_khz > pipes[pipe_idx].clks_cfg.dppclk_mhz * 1000) - pipes[pipe_idx].clks_cfg.dppclk_mhz = dc->debug.min_dpp_clk_khz / 1000.0; - - pipe_idx++; - } - DC_FP_START(); - dcn20_calculate_dlg_params(dc, context, pipes, pipe_cnt, vlevel); + dcn30_fpu_update_soc_for_wm_a(dc, context); DC_FP_END(); - - if (!pstate_en) - /* Restore full p-state latency */ - 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; -} - -void dcn30_update_soc_for_wm_a(struct dc *dc, struct dc_state *context) -{ - if (dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].valid) { - 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; - context->bw_ctx.dml.soc.sr_enter_plus_exit_time_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.sr_enter_plus_exit_time_us; - context->bw_ctx.dml.soc.sr_exit_time_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.sr_exit_time_us; - } } void dcn30_calculate_wm_and_dlg( @@ -2292,7 +1912,7 @@ void dcn30_calculate_wm_and_dlg( int vlevel) { DC_FP_START(); - dcn30_calculate_wm_and_dlg_fp(dc, context, pipes, pipe_cnt, vlevel); + dcn30_fpu_calculate_wm_and_dlg(dc, context, pipes, pipe_cnt, vlevel); DC_FP_END(); } @@ -2351,40 +1971,6 @@ validate_out: return out; } -/* - * This must be noinline to ensure anything that deals with FP registers - * is contained within this call; previously our compiling with hard-float - * would result in fp instructions being emitted outside of the boundaries - * of the DC_FP_START/END macros, which makes sense as the compiler has no - * idea about what is wrapped and what is not - * - * This is largely just a workaround to avoid breakage introduced with 5.6, - * ideally all fp-using code should be moved into its own file, only that - * should be compiled with hard-float, and all code exported from there - * should be strictly wrapped with DC_FP_START/END - */ -static noinline void dcn30_get_optimal_dcfclk_fclk_for_uclk(unsigned int uclk_mts, - unsigned int *optimal_dcfclk, - unsigned int *optimal_fclk) -{ - double bw_from_dram, bw_from_dram1, bw_from_dram2; - - bw_from_dram1 = uclk_mts * dcn3_0_soc.num_chans * - dcn3_0_soc.dram_channel_width_bytes * (dcn3_0_soc.max_avg_dram_bw_use_normal_percent / 100); - bw_from_dram2 = uclk_mts * dcn3_0_soc.num_chans * - dcn3_0_soc.dram_channel_width_bytes * (dcn3_0_soc.max_avg_sdp_bw_use_normal_percent / 100); - - bw_from_dram = (bw_from_dram1 < bw_from_dram2) ? bw_from_dram1 : bw_from_dram2; - - if (optimal_fclk) - *optimal_fclk = bw_from_dram / - (dcn3_0_soc.fabric_datapath_to_dcn_data_return_bytes * (dcn3_0_soc.max_avg_sdp_bw_use_normal_percent / 100)); - - if (optimal_dcfclk) - *optimal_dcfclk = bw_from_dram / - (dcn3_0_soc.return_bus_width_bytes * (dcn3_0_soc.max_avg_sdp_bw_use_normal_percent / 100)); -} - void dcn30_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params) { unsigned int i, j; @@ -2399,47 +1985,43 @@ void dcn30_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params unsigned int num_dcfclk_sta_targets = 4; unsigned int num_uclk_states; + struct dc_bounding_box_max_clk dcn30_bb_max_clk; + + memset(&dcn30_bb_max_clk, 0, sizeof(dcn30_bb_max_clk)); + if (dc->ctx->dc_bios->vram_info.num_chans) dcn3_0_soc.num_chans = dc->ctx->dc_bios->vram_info.num_chans; - if (dc->ctx->dc_bios->vram_info.dram_channel_width_bytes) - dcn3_0_soc.dram_channel_width_bytes = dc->ctx->dc_bios->vram_info.dram_channel_width_bytes; - - dcn3_0_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; + DC_FP_START(); + dcn30_fpu_update_dram_channel_width_bytes(dc); + DC_FP_END(); if (bw_params->clk_table.entries[0].memclk_mhz) { - int max_dcfclk_mhz = 0, max_dispclk_mhz = 0, max_dppclk_mhz = 0, max_phyclk_mhz = 0; for (i = 0; i < MAX_NUM_DPM_LVL; i++) { - if (bw_params->clk_table.entries[i].dcfclk_mhz > max_dcfclk_mhz) - max_dcfclk_mhz = bw_params->clk_table.entries[i].dcfclk_mhz; - if (bw_params->clk_table.entries[i].dispclk_mhz > max_dispclk_mhz) - max_dispclk_mhz = bw_params->clk_table.entries[i].dispclk_mhz; - if (bw_params->clk_table.entries[i].dppclk_mhz > max_dppclk_mhz) - max_dppclk_mhz = bw_params->clk_table.entries[i].dppclk_mhz; - if (bw_params->clk_table.entries[i].phyclk_mhz > max_phyclk_mhz) - max_phyclk_mhz = bw_params->clk_table.entries[i].phyclk_mhz; + if (bw_params->clk_table.entries[i].dcfclk_mhz > dcn30_bb_max_clk.max_dcfclk_mhz) + dcn30_bb_max_clk.max_dcfclk_mhz = bw_params->clk_table.entries[i].dcfclk_mhz; + if (bw_params->clk_table.entries[i].dispclk_mhz > dcn30_bb_max_clk.max_dispclk_mhz) + dcn30_bb_max_clk.max_dispclk_mhz = bw_params->clk_table.entries[i].dispclk_mhz; + if (bw_params->clk_table.entries[i].dppclk_mhz > dcn30_bb_max_clk.max_dppclk_mhz) + dcn30_bb_max_clk.max_dppclk_mhz = bw_params->clk_table.entries[i].dppclk_mhz; + if (bw_params->clk_table.entries[i].phyclk_mhz > dcn30_bb_max_clk.max_phyclk_mhz) + dcn30_bb_max_clk.max_phyclk_mhz = bw_params->clk_table.entries[i].phyclk_mhz; } - if (!max_dcfclk_mhz) - max_dcfclk_mhz = dcn3_0_soc.clock_limits[0].dcfclk_mhz; - if (!max_dispclk_mhz) - max_dispclk_mhz = dcn3_0_soc.clock_limits[0].dispclk_mhz; - if (!max_dppclk_mhz) - max_dppclk_mhz = dcn3_0_soc.clock_limits[0].dppclk_mhz; - if (!max_phyclk_mhz) - max_phyclk_mhz = dcn3_0_soc.clock_limits[0].phyclk_mhz; + DC_FP_START(); + dcn30_fpu_update_max_clk(&dcn30_bb_max_clk); + DC_FP_END(); - if (max_dcfclk_mhz > dcfclk_sta_targets[num_dcfclk_sta_targets-1]) { + if (dcn30_bb_max_clk.max_dcfclk_mhz > dcfclk_sta_targets[num_dcfclk_sta_targets-1]) { // If max DCFCLK is greater than the max DCFCLK STA target, insert into the DCFCLK STA target array - dcfclk_sta_targets[num_dcfclk_sta_targets] = max_dcfclk_mhz; + dcfclk_sta_targets[num_dcfclk_sta_targets] = dcn30_bb_max_clk.max_dcfclk_mhz; num_dcfclk_sta_targets++; - } else if (max_dcfclk_mhz < dcfclk_sta_targets[num_dcfclk_sta_targets-1]) { + } else if (dcn30_bb_max_clk.max_dcfclk_mhz < dcfclk_sta_targets[num_dcfclk_sta_targets-1]) { // If max DCFCLK is less than the max DCFCLK STA target, cap values and remove duplicates for (i = 0; i < num_dcfclk_sta_targets; i++) { - if (dcfclk_sta_targets[i] > max_dcfclk_mhz) { - dcfclk_sta_targets[i] = max_dcfclk_mhz; + if (dcfclk_sta_targets[i] > dcn30_bb_max_clk.max_dcfclk_mhz) { + dcfclk_sta_targets[i] = dcn30_bb_max_clk.max_dcfclk_mhz; break; } } @@ -2452,7 +2034,7 @@ void dcn30_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params // Calculate optimal dcfclk for each uclk for (i = 0; i < num_uclk_states; i++) { DC_FP_START(); - dcn30_get_optimal_dcfclk_fclk_for_uclk(bw_params->clk_table.entries[i].memclk_mhz * 16, + dcn30_fpu_get_optimal_dcfclk_fclk_for_uclk(bw_params->clk_table.entries[i].memclk_mhz * 16, &optimal_dcfclk_for_uclk[i], NULL); DC_FP_END(); if (optimal_dcfclk_for_uclk[i] < bw_params->clk_table.entries[0].dcfclk_mhz) { @@ -2479,7 +2061,7 @@ void dcn30_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params dcfclk_mhz[num_states] = dcfclk_sta_targets[i]; dram_speed_mts[num_states++] = optimal_uclk_for_dcfclk_sta_targets[i++]; } else { - if (j < num_uclk_states && optimal_dcfclk_for_uclk[j] <= max_dcfclk_mhz) { + if (j < num_uclk_states && optimal_dcfclk_for_uclk[j] <= dcn30_bb_max_clk.max_dcfclk_mhz) { dcfclk_mhz[num_states] = optimal_dcfclk_for_uclk[j]; dram_speed_mts[num_states++] = bw_params->clk_table.entries[j++].memclk_mhz * 16; } else { @@ -2494,33 +2076,15 @@ void dcn30_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params } while (j < num_uclk_states && num_states < DC__VOLTAGE_STATES && - optimal_dcfclk_for_uclk[j] <= max_dcfclk_mhz) { + optimal_dcfclk_for_uclk[j] <= dcn30_bb_max_clk.max_dcfclk_mhz) { dcfclk_mhz[num_states] = optimal_dcfclk_for_uclk[j]; dram_speed_mts[num_states++] = bw_params->clk_table.entries[j++].memclk_mhz * 16; } dcn3_0_soc.num_states = num_states; - for (i = 0; i < dcn3_0_soc.num_states; i++) { - dcn3_0_soc.clock_limits[i].state = i; - dcn3_0_soc.clock_limits[i].dcfclk_mhz = dcfclk_mhz[i]; - dcn3_0_soc.clock_limits[i].fabricclk_mhz = dcfclk_mhz[i]; - dcn3_0_soc.clock_limits[i].dram_speed_mts = dram_speed_mts[i]; - - /* Fill all states with max values of all other clocks */ - dcn3_0_soc.clock_limits[i].dispclk_mhz = max_dispclk_mhz; - dcn3_0_soc.clock_limits[i].dppclk_mhz = max_dppclk_mhz; - dcn3_0_soc.clock_limits[i].phyclk_mhz = max_phyclk_mhz; - dcn3_0_soc.clock_limits[i].dtbclk_mhz = dcn3_0_soc.clock_limits[0].dtbclk_mhz; - /* These clocks cannot come from bw_params, always fill from dcn3_0_soc[1] */ - /* FCLK, PHYCLK_D18, SOCCLK, DSCCLK */ - dcn3_0_soc.clock_limits[i].phyclk_d18_mhz = dcn3_0_soc.clock_limits[0].phyclk_d18_mhz; - dcn3_0_soc.clock_limits[i].socclk_mhz = dcn3_0_soc.clock_limits[0].socclk_mhz; - dcn3_0_soc.clock_limits[i].dscclk_mhz = dcn3_0_soc.clock_limits[0].dscclk_mhz; - } - /* re-init DML with updated bb */ - dml_init_instance(&dc->dml, &dcn3_0_soc, &dcn3_0_ip, DML_PROJECT_DCN30); - if (dc->current_state) - dml_init_instance(&dc->current_state->bw_ctx.dml, &dcn3_0_soc, &dcn3_0_ip, DML_PROJECT_DCN30); + DC_FP_START(); + dcn30_fpu_update_bw_bounding_box(dc, bw_params, &dcn30_bb_max_clk, dcfclk_mhz, dram_speed_mts); + DC_FP_END(); } } diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.h b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.h index b92e4cc0232f..3330a1026fa5 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.h +++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.h @@ -35,6 +35,9 @@ struct dc; struct resource_pool; struct _vcs_dpi_display_pipe_params_st; +extern struct _vcs_dpi_ip_params_st dcn3_0_ip; +extern struct _vcs_dpi_soc_bounding_box_st dcn3_0_soc; + struct dcn30_resource_pool { struct resource_pool base; }; @@ -96,4 +99,6 @@ enum dc_status dcn30_add_stream_to_ctx( void dcn30_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params); +void dcn30_setup_mclk_switch_using_fw_based_vblank_stretch(struct dc *dc, struct dc_state *context); + #endif /* _DCN30_RESOURCE_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_resource.c b/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_resource.c index 4daf8931aa7c..a5df74110284 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_resource.c @@ -81,6 +81,8 @@ #include "dce/dce_aux.h" #include "dce/dce_i2c.h" +#include "dml/dcn30/dcn30_fpu.h" + #include "dml/dcn30/display_mode_vba_30.h" #include "dml/dcn301/dcn301_fpu.h" #include "vm_helper.h" diff --git a/drivers/gpu/drm/amd/display/dc/dcn302/dcn302_resource.c b/drivers/gpu/drm/amd/display/dc/dcn302/dcn302_resource.c index f0938653bb88..f537888f4fa6 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn302/dcn302_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn302/dcn302_resource.c @@ -43,6 +43,8 @@ #include "dcn20/dcn20_dsc.h" #include "dcn20/dcn20_resource.h" +#include "dml/dcn30/dcn30_fpu.h" + #include "dcn10/dcn10_resource.h" #include "dce/dce_abm.h" diff --git a/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c b/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c index 4fcbc0502808..76f863eb86ef 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c @@ -25,6 +25,8 @@ #include "dcn20/dcn20_dsc.h" #include "dcn20/dcn20_resource.h" +#include "dml/dcn30/dcn30_fpu.h" + #include "dcn10/dcn10_resource.h" #include "dc_link_ddc.h" diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dccg.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dccg.c index 287a1066b547..bbc58d167c63 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dccg.c +++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dccg.c @@ -513,12 +513,10 @@ void dccg31_set_physymclk( /* Controls the generation of pixel valid for OTG in (OTG -> HPO case) */ static void dccg31_set_dtbclk_dto( struct dccg *dccg, - int dtbclk_inst, - int req_dtbclk_khz, - int num_odm_segments, - const struct dc_crtc_timing *timing) + struct dtbclk_dto_params *params) { struct dcn_dccg *dccg_dcn = TO_DCN_DCCG(dccg); + int req_dtbclk_khz = params->pixclk_khz; uint32_t dtbdto_div; /* Mode DTBDTO Rate DTBCLK_DTO<x>_DIV Register @@ -529,57 +527,53 @@ static void dccg31_set_dtbclk_dto( * DSC native 4:2:2 pixel rate/2 4 * Other modes pixel rate 8 */ - if (num_odm_segments == 4) { + if (params->num_odm_segments == 4) { dtbdto_div = 2; - req_dtbclk_khz = req_dtbclk_khz / 4; - } else if ((num_odm_segments == 2) || - (timing->pixel_encoding == PIXEL_ENCODING_YCBCR420) || - (timing->flags.DSC && timing->pixel_encoding == PIXEL_ENCODING_YCBCR422 - && !timing->dsc_cfg.ycbcr422_simple)) { + req_dtbclk_khz = params->pixclk_khz / 4; + } else if ((params->num_odm_segments == 2) || + (params->timing->pixel_encoding == PIXEL_ENCODING_YCBCR420) || + (params->timing->flags.DSC && params->timing->pixel_encoding == PIXEL_ENCODING_YCBCR422 + && !params->timing->dsc_cfg.ycbcr422_simple)) { dtbdto_div = 4; - req_dtbclk_khz = req_dtbclk_khz / 2; + req_dtbclk_khz = params->pixclk_khz / 2; } else dtbdto_div = 8; - if (dccg->ref_dtbclk_khz && req_dtbclk_khz) { + if (params->ref_dtbclk_khz && req_dtbclk_khz) { uint32_t modulo, phase; // phase / modulo = dtbclk / dtbclk ref - modulo = dccg->ref_dtbclk_khz * 1000; - phase = div_u64((((unsigned long long)modulo * req_dtbclk_khz) + dccg->ref_dtbclk_khz - 1), - dccg->ref_dtbclk_khz); + modulo = params->ref_dtbclk_khz * 1000; + phase = div_u64((((unsigned long long)modulo * req_dtbclk_khz) + params->ref_dtbclk_khz - 1), + params->ref_dtbclk_khz); - REG_UPDATE(OTG_PIXEL_RATE_CNTL[dtbclk_inst], - DTBCLK_DTO_DIV[dtbclk_inst], dtbdto_div); + REG_UPDATE(OTG_PIXEL_RATE_CNTL[params->otg_inst], + DTBCLK_DTO_DIV[params->otg_inst], dtbdto_div); - REG_WRITE(DTBCLK_DTO_MODULO[dtbclk_inst], modulo); - REG_WRITE(DTBCLK_DTO_PHASE[dtbclk_inst], phase); + REG_WRITE(DTBCLK_DTO_MODULO[params->otg_inst], modulo); + REG_WRITE(DTBCLK_DTO_PHASE[params->otg_inst], phase); - REG_UPDATE(OTG_PIXEL_RATE_CNTL[dtbclk_inst], - DTBCLK_DTO_ENABLE[dtbclk_inst], 1); + REG_UPDATE(OTG_PIXEL_RATE_CNTL[params->otg_inst], + DTBCLK_DTO_ENABLE[params->otg_inst], 1); - REG_WAIT(OTG_PIXEL_RATE_CNTL[dtbclk_inst], - DTBCLKDTO_ENABLE_STATUS[dtbclk_inst], 1, + REG_WAIT(OTG_PIXEL_RATE_CNTL[params->otg_inst], + DTBCLKDTO_ENABLE_STATUS[params->otg_inst], 1, 1, 100); /* The recommended programming sequence to enable DTBCLK DTO to generate * valid pixel HPO DPSTREAM ENCODER, specifies that DTO source select should * be set only after DTO is enabled */ - REG_UPDATE(OTG_PIXEL_RATE_CNTL[dtbclk_inst], - PIPE_DTO_SRC_SEL[dtbclk_inst], 1); - - dccg->dtbclk_khz[dtbclk_inst] = req_dtbclk_khz; + REG_UPDATE(OTG_PIXEL_RATE_CNTL[params->otg_inst], + PIPE_DTO_SRC_SEL[params->otg_inst], 1); } else { - REG_UPDATE_3(OTG_PIXEL_RATE_CNTL[dtbclk_inst], - DTBCLK_DTO_ENABLE[dtbclk_inst], 0, - PIPE_DTO_SRC_SEL[dtbclk_inst], 0, - DTBCLK_DTO_DIV[dtbclk_inst], dtbdto_div); + REG_UPDATE_3(OTG_PIXEL_RATE_CNTL[params->otg_inst], + DTBCLK_DTO_ENABLE[params->otg_inst], 0, + PIPE_DTO_SRC_SEL[params->otg_inst], 0, + DTBCLK_DTO_DIV[params->otg_inst], dtbdto_div); - REG_WRITE(DTBCLK_DTO_MODULO[dtbclk_inst], 0); - REG_WRITE(DTBCLK_DTO_PHASE[dtbclk_inst], 0); - - dccg->dtbclk_khz[dtbclk_inst] = 0; + REG_WRITE(DTBCLK_DTO_MODULO[params->otg_inst], 0); + REG_WRITE(DTBCLK_DTO_PHASE[params->otg_inst], 0); } } @@ -606,16 +600,12 @@ void dccg31_set_audio_dtbclk_dto( REG_UPDATE(DCCG_AUDIO_DTO_SOURCE, DCCG_AUDIO_DTO_SEL, 4); // 04 - DCCG_AUDIO_DTO_SEL_AUDIO_DTO_DTBCLK - - dccg->audio_dtbclk_khz = req_audio_dtbclk_khz; } else { REG_WRITE(DCCG_AUDIO_DTBCLK_DTO_PHASE, 0); REG_WRITE(DCCG_AUDIO_DTBCLK_DTO_MODULO, 0); REG_UPDATE(DCCG_AUDIO_DTO_SOURCE, DCCG_AUDIO_DTO_SEL, 3); // 03 - DCCG_AUDIO_DTO_SEL_NO_AUDIO_DTO - - dccg->audio_dtbclk_khz = 0; } } diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dio_link_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dio_link_encoder.c index d94fd1010deb..8b12b4111c88 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dio_link_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dio_link_encoder.c @@ -230,9 +230,7 @@ static void enc31_hw_init(struct link_encoder *enc) AUX_RX_PHASE_DETECT_LEN, [21,20] = 0x3 default is 3 AUX_RX_DETECTION_THRESHOLD [30:28] = 1 */ - AUX_REG_WRITE(AUX_DPHY_RX_CONTROL0, 0x103d1110); - - AUX_REG_WRITE(AUX_DPHY_TX_CONTROL, 0x21c7a); + // dmub will read AUX_DPHY_RX_CONTROL0/AUX_DPHY_TX_CONTROL from vbios table in dp_aux_init //AUX_DPHY_TX_REF_CONTROL'AUX_TX_REF_DIV HW default is 0x32; // Set AUX_TX_REF_DIV Divider to generate 2 MHz reference from refclk diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.c index ccf1b71a8269..3d9f07d4770b 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.c @@ -36,6 +36,8 @@ #include "dcn20/dcn20_resource.h" #include "dcn30/dcn30_resource.h" +#include "dml/dcn30/dcn30_fpu.h" + #include "dcn10/dcn10_ipp.h" #include "dcn30/dcn30_hubbub.h" #include "dcn31/dcn31_hubbub.h" diff --git a/drivers/gpu/drm/amd/display/dc/dml/Makefile b/drivers/gpu/drm/amd/display/dc/dml/Makefile index ee911452c048..a64b88ca01a9 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dml/Makefile @@ -71,6 +71,7 @@ CFLAGS_$(AMDDALPATH)/dc/dml/dcn30/display_mode_vba_30.o := $(dml_ccflags) $(fram CFLAGS_$(AMDDALPATH)/dc/dml/dcn30/display_rq_dlg_calc_30.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/dcn31/display_mode_vba_31.o := $(dml_ccflags) $(frame_warn_flag) CFLAGS_$(AMDDALPATH)/dc/dml/dcn31/display_rq_dlg_calc_31.o := $(dml_ccflags) +CFLAGS_$(AMDDALPATH)/dc/dml/dcn30/dcn30_fpu.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/dcn31/dcn31_fpu.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/dcn301/dcn301_fpu.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/dcn302/dcn302_fpu.o := $(dml_ccflags) @@ -113,7 +114,7 @@ DML += dcn20/dcn20_fpu.o DML += display_mode_vba.o dcn20/display_rq_dlg_calc_20.o dcn20/display_mode_vba_20.o DML += dcn20/display_rq_dlg_calc_20v2.o dcn20/display_mode_vba_20v2.o DML += dcn21/display_rq_dlg_calc_21.o dcn21/display_mode_vba_21.o -DML += dcn30/display_mode_vba_30.o dcn30/display_rq_dlg_calc_30.o +DML += dcn30/dcn30_fpu.o dcn30/display_mode_vba_30.o dcn30/display_rq_dlg_calc_30.o DML += dcn31/display_mode_vba_31.o dcn31/display_rq_dlg_calc_31.o DML += dcn31/dcn31_fpu.o DML += dcn301/dcn301_fpu.o 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 new file mode 100644 index 000000000000..574676a0711a --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn30/dcn30_fpu.c @@ -0,0 +1,617 @@ +/* + * Copyright 2020-2021 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ +#include "resource.h" +#include "clk_mgr.h" +#include "reg_helper.h" +#include "dcn_calc_math.h" +#include "dcn20/dcn20_resource.h" +#include "dcn30/dcn30_resource.h" + + +#include "display_mode_vba_30.h" +#include "dcn30_fpu.h" + +#define REG(reg)\ + optc1->tg_regs->reg + +#define CTX \ + optc1->base.ctx + +#undef FN +#define FN(reg_name, field_name) \ + optc1->tg_shift->field_name, optc1->tg_mask->field_name + + +struct _vcs_dpi_ip_params_st dcn3_0_ip = { + .use_min_dcfclk = 0, + .clamp_min_dcfclk = 0, + .odm_capable = 1, + .gpuvm_enable = 0, + .hostvm_enable = 0, + .gpuvm_max_page_table_levels = 4, + .hostvm_max_page_table_levels = 4, + .hostvm_cached_page_table_levels = 0, + .pte_group_size_bytes = 2048, + .num_dsc = 6, + .rob_buffer_size_kbytes = 184, + .det_buffer_size_kbytes = 184, + .dpte_buffer_size_in_pte_reqs_luma = 84, + .pde_proc_buffer_size_64k_reqs = 48, + .dpp_output_buffer_pixels = 2560, + .opp_output_buffer_lines = 1, + .pixel_chunk_size_kbytes = 8, + .pte_enable = 1, + .max_page_table_levels = 2, + .pte_chunk_size_kbytes = 2, // ? + .meta_chunk_size_kbytes = 2, + .writeback_chunk_size_kbytes = 8, + .line_buffer_size_bits = 789504, + .is_line_buffer_bpp_fixed = 0, // ? + .line_buffer_fixed_bpp = 0, // ? + .dcc_supported = true, + .writeback_interface_buffer_size_kbytes = 90, + .writeback_line_buffer_buffer_size = 0, + .max_line_buffer_lines = 12, + .writeback_luma_buffer_size_kbytes = 12, // writeback_line_buffer_buffer_size = 656640 + .writeback_chroma_buffer_size_kbytes = 8, + .writeback_chroma_line_buffer_width_pixels = 4, + .writeback_max_hscl_ratio = 1, + .writeback_max_vscl_ratio = 1, + .writeback_min_hscl_ratio = 1, + .writeback_min_vscl_ratio = 1, + .writeback_max_hscl_taps = 1, + .writeback_max_vscl_taps = 1, + .writeback_line_buffer_luma_buffer_size = 0, + .writeback_line_buffer_chroma_buffer_size = 14643, + .cursor_buffer_size = 8, + .cursor_chunk_size = 2, + .max_num_otg = 6, + .max_num_dpp = 6, + .max_num_wb = 1, + .max_dchub_pscl_bw_pix_per_clk = 4, + .max_pscl_lb_bw_pix_per_clk = 2, + .max_lb_vscl_bw_pix_per_clk = 4, + .max_vscl_hscl_bw_pix_per_clk = 4, + .max_hscl_ratio = 6, + .max_vscl_ratio = 6, + .hscl_mults = 4, + .vscl_mults = 4, + .max_hscl_taps = 8, + .max_vscl_taps = 8, + .dispclk_ramp_margin_percent = 1, + .underscan_factor = 1.11, + .min_vblank_lines = 32, + .dppclk_delay_subtotal = 46, + .dynamic_metadata_vm_enabled = true, + .dppclk_delay_scl_lb_only = 16, + .dppclk_delay_scl = 50, + .dppclk_delay_cnvc_formatter = 27, + .dppclk_delay_cnvc_cursor = 6, + .dispclk_delay_subtotal = 119, + .dcfclk_cstate_latency = 5.2, // SRExitTime + .max_inter_dcn_tile_repeaters = 8, + .max_num_hdmi_frl_outputs = 1, + .odm_combine_4to1_supported = true, + + .xfc_supported = false, + .xfc_fill_bw_overhead_percent = 10.0, + .xfc_fill_constant_bytes = 0, + .gfx7_compat_tiling_supported = 0, + .number_of_cursors = 1, +}; + +struct _vcs_dpi_soc_bounding_box_st dcn3_0_soc = { + .clock_limits = { + { + .state = 0, + .dispclk_mhz = 562.0, + .dppclk_mhz = 300.0, + .phyclk_mhz = 300.0, + .phyclk_d18_mhz = 667.0, + .dscclk_mhz = 405.6, + }, + }, + + .min_dcfclk = 500.0, /* TODO: set this to actual min DCFCLK */ + .num_states = 1, + .sr_exit_time_us = 15.5, + .sr_enter_plus_exit_time_us = 20, + .urgent_latency_us = 4.0, + .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_out_of_order_return_per_channel_pixel_only_bytes = 4096, + .urgent_out_of_order_return_per_channel_pixel_and_vm_bytes = 4096, + .urgent_out_of_order_return_per_channel_vm_only_bytes = 4096, + .pct_ideal_dram_sdp_bw_after_urgent_pixel_only = 80.0, + .pct_ideal_dram_sdp_bw_after_urgent_pixel_and_vm = 60.0, + .pct_ideal_dram_sdp_bw_after_urgent_vm_only = 40.0, + .max_avg_sdp_bw_use_normal_percent = 60.0, + .max_avg_dram_bw_use_normal_percent = 40.0, + .writeback_latency_us = 12.0, + .max_request_size_bytes = 256, + .fabric_datapath_to_dcn_data_return_bytes = 64, + .dcn_downspread_percent = 0.5, + .downspread_percent = 0.38, + .dram_page_open_time_ns = 50.0, + .dram_rw_turnaround_time_ns = 17.5, + .dram_return_buffer_per_channel_bytes = 8192, + .round_trip_ping_latency_dcfclk_cycles = 191, + .urgent_out_of_order_return_per_channel_bytes = 4096, + .channel_interleave_bytes = 256, + .num_banks = 8, + .gpuvm_min_page_size_bytes = 4096, + .hostvm_min_page_size_bytes = 4096, + .dram_clock_change_latency_us = 404, + .dummy_pstate_latency_us = 5, + .writeback_dram_clock_change_latency_us = 23.0, + .return_bus_width_bytes = 64, + .dispclk_dppclk_vco_speed_mhz = 3650, + .xfc_bus_transport_time_us = 20, // ? + .xfc_xbuf_latency_tolerance_us = 4, // ? + .use_urgent_burst_bw = 1, // ? + .do_urgent_latency_adjustment = true, + .urgent_latency_adjustment_fabric_clock_component_us = 1.0, + .urgent_latency_adjustment_fabric_clock_reference_mhz = 1000, +}; + + +void optc3_fpu_set_vrr_m_const(struct timing_generator *optc, + double vtotal_avg) +{ +struct optc *optc1 = DCN10TG_FROM_TG(optc); + double vtotal_min, vtotal_max; + double ratio, modulo, phase; + uint32_t vblank_start; + uint32_t v_total_mask_value = 0; + + dc_assert_fp_enabled(); + + /* Compute VTOTAL_MIN and VTOTAL_MAX, so that + * VOTAL_MAX - VTOTAL_MIN = 1 + */ + v_total_mask_value = 16; + vtotal_min = dcn_bw_floor(vtotal_avg); + vtotal_max = dcn_bw_ceil(vtotal_avg); + + /* Check that bottom VBLANK is at least 2 lines tall when running with + * VTOTAL_MIN. Note that VTOTAL registers are defined as 'total number + * of lines in a frame - 1'. + */ + REG_GET(OTG_V_BLANK_START_END, OTG_V_BLANK_START, + &vblank_start); + ASSERT(vtotal_min >= vblank_start + 1); + + /* Special case where the average frame rate can be achieved + * without using the DTO + */ + if (vtotal_min == vtotal_max) { + REG_SET(OTG_V_TOTAL, 0, OTG_V_TOTAL, (uint32_t)vtotal_min); + + optc->funcs->set_vtotal_min_max(optc, 0, 0); + REG_SET(OTG_M_CONST_DTO0, 0, OTG_M_CONST_DTO_PHASE, 0); + REG_SET(OTG_M_CONST_DTO1, 0, OTG_M_CONST_DTO_MODULO, 0); + REG_UPDATE_3(OTG_V_TOTAL_CONTROL, + OTG_V_TOTAL_MIN_SEL, 0, + OTG_V_TOTAL_MAX_SEL, 0, + OTG_SET_V_TOTAL_MIN_MASK_EN, 0); + return; + } + + ratio = vtotal_max - vtotal_avg; + modulo = 65536.0 * 65536.0 - 1.0; /* 2^32 - 1 */ + phase = ratio * modulo; + + /* Special cases where the DTO phase gets rounded to 0 or + * to DTO modulo + */ + if (phase <= 0 || phase >= modulo) { + REG_SET(OTG_V_TOTAL, 0, OTG_V_TOTAL, + phase <= 0 ? + (uint32_t)vtotal_max : (uint32_t)vtotal_min); + REG_SET(OTG_V_TOTAL_MIN, 0, OTG_V_TOTAL_MIN, 0); + REG_SET(OTG_V_TOTAL_MAX, 0, OTG_V_TOTAL_MAX, 0); + REG_SET(OTG_M_CONST_DTO0, 0, OTG_M_CONST_DTO_PHASE, 0); + REG_SET(OTG_M_CONST_DTO1, 0, OTG_M_CONST_DTO_MODULO, 0); + REG_UPDATE_3(OTG_V_TOTAL_CONTROL, + OTG_V_TOTAL_MIN_SEL, 0, + OTG_V_TOTAL_MAX_SEL, 0, + OTG_SET_V_TOTAL_MIN_MASK_EN, 0); + return; + } + REG_UPDATE_6(OTG_V_TOTAL_CONTROL, + OTG_V_TOTAL_MIN_SEL, 1, + OTG_V_TOTAL_MAX_SEL, 1, + OTG_SET_V_TOTAL_MIN_MASK_EN, 1, + OTG_SET_V_TOTAL_MIN_MASK, v_total_mask_value, + OTG_VTOTAL_MID_REPLACING_MIN_EN, 0, + OTG_VTOTAL_MID_REPLACING_MAX_EN, 0); + REG_SET(OTG_V_TOTAL, 0, OTG_V_TOTAL, (uint32_t)vtotal_min); + optc->funcs->set_vtotal_min_max(optc, vtotal_min, vtotal_max); + REG_SET(OTG_M_CONST_DTO0, 0, OTG_M_CONST_DTO_PHASE, (uint32_t)phase); + REG_SET(OTG_M_CONST_DTO1, 0, OTG_M_CONST_DTO_MODULO, (uint32_t)modulo); +} + +void dcn30_fpu_populate_dml_writeback_from_context( + struct dc *dc, struct resource_context *res_ctx, display_e2e_pipe_params_st *pipes) +{ + int pipe_cnt, i, j; + double max_calc_writeback_dispclk; + double writeback_dispclk; + struct writeback_st dout_wb; + + dc_assert_fp_enabled(); + + for (i = 0, pipe_cnt = 0; i < dc->res_pool->pipe_count; i++) { + struct dc_stream_state *stream = res_ctx->pipe_ctx[i].stream; + + if (!stream) + continue; + max_calc_writeback_dispclk = 0; + + /* Set writeback information */ + pipes[pipe_cnt].dout.wb_enable = 0; + pipes[pipe_cnt].dout.num_active_wb = 0; + for (j = 0; j < stream->num_wb_info; j++) { + struct dc_writeback_info *wb_info = &stream->writeback_info[j]; + + if (wb_info->wb_enabled && wb_info->writeback_source_plane && + (wb_info->writeback_source_plane == res_ctx->pipe_ctx[i].plane_state)) { + pipes[pipe_cnt].dout.wb_enable = 1; + pipes[pipe_cnt].dout.num_active_wb++; + dout_wb.wb_src_height = wb_info->dwb_params.cnv_params.crop_en ? + wb_info->dwb_params.cnv_params.crop_height : + wb_info->dwb_params.cnv_params.src_height; + dout_wb.wb_src_width = wb_info->dwb_params.cnv_params.crop_en ? + wb_info->dwb_params.cnv_params.crop_width : + wb_info->dwb_params.cnv_params.src_width; + dout_wb.wb_dst_width = wb_info->dwb_params.dest_width; + dout_wb.wb_dst_height = wb_info->dwb_params.dest_height; + + /* For IP that doesn't support WB scaling, set h/v taps to 1 to avoid DML validation failure */ + if (dc->dml.ip.writeback_max_hscl_taps > 1) { + dout_wb.wb_htaps_luma = wb_info->dwb_params.scaler_taps.h_taps; + dout_wb.wb_vtaps_luma = wb_info->dwb_params.scaler_taps.v_taps; + } else { + dout_wb.wb_htaps_luma = 1; + dout_wb.wb_vtaps_luma = 1; + } + dout_wb.wb_htaps_chroma = 0; + dout_wb.wb_vtaps_chroma = 0; + dout_wb.wb_hratio = wb_info->dwb_params.cnv_params.crop_en ? + (double)wb_info->dwb_params.cnv_params.crop_width / + (double)wb_info->dwb_params.dest_width : + (double)wb_info->dwb_params.cnv_params.src_width / + (double)wb_info->dwb_params.dest_width; + dout_wb.wb_vratio = wb_info->dwb_params.cnv_params.crop_en ? + (double)wb_info->dwb_params.cnv_params.crop_height / + (double)wb_info->dwb_params.dest_height : + (double)wb_info->dwb_params.cnv_params.src_height / + (double)wb_info->dwb_params.dest_height; + if (wb_info->dwb_params.cnv_params.fc_out_format == DWB_OUT_FORMAT_64BPP_ARGB || + wb_info->dwb_params.cnv_params.fc_out_format == DWB_OUT_FORMAT_64BPP_RGBA) + dout_wb.wb_pixel_format = dm_444_64; + else + dout_wb.wb_pixel_format = dm_444_32; + + /* Workaround for cases where multiple writebacks are connected to same plane + * In which case, need to compute worst case and set the associated writeback parameters + * This workaround is necessary due to DML computation assuming only 1 set of writeback + * parameters per pipe + */ + writeback_dispclk = dml30_CalculateWriteBackDISPCLK( + dout_wb.wb_pixel_format, + pipes[pipe_cnt].pipe.dest.pixel_rate_mhz, + dout_wb.wb_hratio, + dout_wb.wb_vratio, + dout_wb.wb_htaps_luma, + dout_wb.wb_vtaps_luma, + dout_wb.wb_src_width, + dout_wb.wb_dst_width, + pipes[pipe_cnt].pipe.dest.htotal, + dc->current_state->bw_ctx.dml.ip.writeback_line_buffer_buffer_size); + + if (writeback_dispclk > max_calc_writeback_dispclk) { + max_calc_writeback_dispclk = writeback_dispclk; + pipes[pipe_cnt].dout.wb = dout_wb; + } + } + } + + pipe_cnt++; + } +} + +void dcn30_fpu_set_mcif_arb_params(struct mcif_arb_params *wb_arb_params, + struct display_mode_lib *dml, + display_e2e_pipe_params_st *pipes, + int pipe_cnt, + int cur_pipe) +{ + int i; + + dc_assert_fp_enabled(); + + for (i = 0; i < sizeof(wb_arb_params->cli_watermark)/sizeof(wb_arb_params->cli_watermark[0]); i++) { + wb_arb_params->cli_watermark[i] = get_wm_writeback_urgent(dml, pipes, pipe_cnt) * 1000; + wb_arb_params->pstate_watermark[i] = get_wm_writeback_dram_clock_change(dml, pipes, pipe_cnt) * 1000; + } + + wb_arb_params->dram_speed_change_duration = dml->vba.WritebackAllowDRAMClockChangeEndPosition[cur_pipe] * pipes[0].clks_cfg.refclk_mhz; /* num_clock_cycles = us * MHz */ +} + +void dcn30_fpu_update_soc_for_wm_a(struct dc *dc, struct dc_state *context) +{ + +dc_assert_fp_enabled(); + +if (dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].valid) { + 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; + context->bw_ctx.dml.soc.sr_enter_plus_exit_time_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.sr_enter_plus_exit_time_us; + context->bw_ctx.dml.soc.sr_exit_time_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.sr_exit_time_us; + } +} + +void dcn30_fpu_calculate_wm_and_dlg( + struct dc *dc, struct dc_state *context, + display_e2e_pipe_params_st *pipes, + int pipe_cnt, + int vlevel) +{ +int maxMpcComb = context->bw_ctx.dml.vba.maxMpcComb; + int i, pipe_idx; + double dcfclk = context->bw_ctx.dml.vba.DCFCLKState[vlevel][maxMpcComb]; + bool pstate_en = context->bw_ctx.dml.vba.DRAMClockChangeSupport[vlevel][maxMpcComb] != dm_dram_clock_change_unsupported; + +dc_assert_fp_enabled(); + + if (context->bw_ctx.dml.soc.min_dcfclk > dcfclk) + dcfclk = context->bw_ctx.dml.soc.min_dcfclk; + + pipes[0].clks_cfg.voltage = vlevel; + pipes[0].clks_cfg.dcfclk_mhz = dcfclk; + pipes[0].clks_cfg.socclk_mhz = context->bw_ctx.dml.soc.clock_limits[vlevel].socclk_mhz; + + /* Set B: + * DCFCLK: 1GHz or min required above 1GHz + * FCLK/UCLK: Max + */ + if (dc->clk_mgr->bw_params->wm_table.nv_entries[WM_B].valid) { + if (vlevel == 0) { + pipes[0].clks_cfg.voltage = 1; + pipes[0].clks_cfg.dcfclk_mhz = context->bw_ctx.dml.soc.clock_limits[0].dcfclk_mhz; + } + context->bw_ctx.dml.soc.dram_clock_change_latency_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_B].dml_input.pstate_latency_us; + context->bw_ctx.dml.soc.sr_enter_plus_exit_time_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_B].dml_input.sr_enter_plus_exit_time_us; + context->bw_ctx.dml.soc.sr_exit_time_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_B].dml_input.sr_exit_time_us; + } + context->bw_ctx.bw.dcn.watermarks.b.urgent_ns = get_wm_urgent(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.b.cstate_pstate.cstate_enter_plus_exit_ns = get_wm_stutter_enter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.b.cstate_pstate.cstate_exit_ns = get_wm_stutter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.b.cstate_pstate.pstate_change_ns = get_wm_dram_clock_change(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.b.pte_meta_urgent_ns = get_wm_memory_trip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.b.frac_urg_bw_nom = get_fraction_of_urgent_bandwidth(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.b.frac_urg_bw_flip = get_fraction_of_urgent_bandwidth_imm_flip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.b.urgent_latency_ns = get_urgent_latency(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + + pipes[0].clks_cfg.voltage = vlevel; + pipes[0].clks_cfg.dcfclk_mhz = dcfclk; + + /* Set D: + * DCFCLK: Min Required + * FCLK(proportional to UCLK): 1GHz or Max + * MALL stutter, sr_enter_exit = 4, sr_exit = 2us + */ + /* + if (dc->clk_mgr->bw_params->wm_table.nv_entries[WM_D].valid) { + context->bw_ctx.dml.soc.dram_clock_change_latency_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_D].dml_input.pstate_latency_us; + context->bw_ctx.dml.soc.sr_enter_plus_exit_time_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_D].dml_input.sr_enter_plus_exit_time_us; + context->bw_ctx.dml.soc.sr_exit_time_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_D].dml_input.sr_exit_time_us; + } + context->bw_ctx.bw.dcn.watermarks.d.urgent_ns = get_wm_urgent(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.d.cstate_pstate.cstate_enter_plus_exit_ns = get_wm_stutter_enter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.d.cstate_pstate.cstate_exit_ns = get_wm_stutter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.d.cstate_pstate.pstate_change_ns = get_wm_dram_clock_change(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.d.pte_meta_urgent_ns = get_wm_memory_trip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.d.frac_urg_bw_nom = get_fraction_of_urgent_bandwidth(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.d.frac_urg_bw_flip = get_fraction_of_urgent_bandwidth_imm_flip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.d.urgent_latency_ns = get_urgent_latency(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + */ + + /* Set C: + * DCFCLK: Min Required + * FCLK(proportional to UCLK): 1GHz or Max + * pstate latency overridden to 5us + */ + if (dc->clk_mgr->bw_params->wm_table.nv_entries[WM_C].valid) { + unsigned int min_dram_speed_mts = context->bw_ctx.dml.vba.DRAMSpeed; + unsigned int min_dram_speed_mts_margin = 160; + + if (context->bw_ctx.dml.vba.DRAMClockChangeSupport[vlevel][context->bw_ctx.dml.vba.maxMpcComb] == dm_dram_clock_change_unsupported) + min_dram_speed_mts = dc->clk_mgr->bw_params->clk_table.entries[dc->clk_mgr->bw_params->clk_table.num_entries - 1].memclk_mhz * 16; + + /* find largest table entry that is lower than dram speed, but lower than DPM0 still uses DPM0 */ + for (i = 3; i > 0; i--) + if (min_dram_speed_mts + min_dram_speed_mts_margin > dc->clk_mgr->bw_params->dummy_pstate_table[i].dram_speed_mts) + break; + + context->bw_ctx.dml.soc.dram_clock_change_latency_us = dc->clk_mgr->bw_params->dummy_pstate_table[i].dummy_pstate_latency_us; + + context->bw_ctx.dml.soc.sr_enter_plus_exit_time_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_C].dml_input.sr_enter_plus_exit_time_us; + context->bw_ctx.dml.soc.sr_exit_time_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_C].dml_input.sr_exit_time_us; + } + + context->bw_ctx.bw.dcn.watermarks.c.urgent_ns = get_wm_urgent(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.c.cstate_pstate.cstate_enter_plus_exit_ns = get_wm_stutter_enter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.c.cstate_pstate.cstate_exit_ns = get_wm_stutter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.c.cstate_pstate.pstate_change_ns = get_wm_dram_clock_change(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.c.pte_meta_urgent_ns = get_wm_memory_trip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.c.frac_urg_bw_nom = get_fraction_of_urgent_bandwidth(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.c.frac_urg_bw_flip = get_fraction_of_urgent_bandwidth_imm_flip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.c.urgent_latency_ns = get_urgent_latency(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + + if (!pstate_en) { + /* The only difference between A and C is p-state latency, if p-state is not supported we want to + * calculate DLG based on dummy p-state latency, and max out the set A p-state watermark + */ + context->bw_ctx.bw.dcn.watermarks.a = context->bw_ctx.bw.dcn.watermarks.c; + context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.pstate_change_ns = 0; + } else { + /* Set A: + * DCFCLK: Min Required + * FCLK(proportional to UCLK): 1GHz or Max + * + * Set A calculated last so that following calculations are based on Set A + */ + dc->res_pool->funcs->update_soc_for_wm_a(dc, context); + context->bw_ctx.bw.dcn.watermarks.a.urgent_ns = get_wm_urgent(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.cstate_enter_plus_exit_ns = get_wm_stutter_enter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.cstate_exit_ns = get_wm_stutter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.pstate_change_ns = get_wm_dram_clock_change(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.a.pte_meta_urgent_ns = get_wm_memory_trip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.a.frac_urg_bw_nom = get_fraction_of_urgent_bandwidth(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.a.frac_urg_bw_flip = get_fraction_of_urgent_bandwidth_imm_flip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.a.urgent_latency_ns = get_urgent_latency(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + } + + context->perf_params.stutter_period_us = context->bw_ctx.dml.vba.StutterPeriod; + + /* Make set D = set A until set D is enabled */ + context->bw_ctx.bw.dcn.watermarks.d = context->bw_ctx.bw.dcn.watermarks.a; + + for (i = 0, pipe_idx = 0; i < dc->res_pool->pipe_count; i++) { + if (!context->res_ctx.pipe_ctx[i].stream) + continue; + + 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); + + if (dc->config.forced_clocks) { + pipes[pipe_idx].clks_cfg.dispclk_mhz = context->bw_ctx.dml.soc.clock_limits[0].dispclk_mhz; + pipes[pipe_idx].clks_cfg.dppclk_mhz = context->bw_ctx.dml.soc.clock_limits[0].dppclk_mhz; + } + if (dc->debug.min_disp_clk_khz > pipes[pipe_idx].clks_cfg.dispclk_mhz * 1000) + pipes[pipe_idx].clks_cfg.dispclk_mhz = dc->debug.min_disp_clk_khz / 1000.0; + if (dc->debug.min_dpp_clk_khz > pipes[pipe_idx].clks_cfg.dppclk_mhz * 1000) + pipes[pipe_idx].clks_cfg.dppclk_mhz = dc->debug.min_dpp_clk_khz / 1000.0; + + 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 */ + 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; + +} + +void dcn30_fpu_update_dram_channel_width_bytes(struct dc *dc) +{ + dc_assert_fp_enabled(); + + if (dc->ctx->dc_bios->vram_info.dram_channel_width_bytes) + dcn3_0_soc.dram_channel_width_bytes = dc->ctx->dc_bios->vram_info.dram_channel_width_bytes; +} + +void dcn30_fpu_update_max_clk(struct dc_bounding_box_max_clk *dcn30_bb_max_clk) +{ + dc_assert_fp_enabled(); + + if (!dcn30_bb_max_clk->max_dcfclk_mhz) + dcn30_bb_max_clk->max_dcfclk_mhz = dcn3_0_soc.clock_limits[0].dcfclk_mhz; + if (!dcn30_bb_max_clk->max_dispclk_mhz) + dcn30_bb_max_clk->max_dispclk_mhz = dcn3_0_soc.clock_limits[0].dispclk_mhz; + if (!dcn30_bb_max_clk->max_dppclk_mhz) + dcn30_bb_max_clk->max_dppclk_mhz = dcn3_0_soc.clock_limits[0].dppclk_mhz; + if (!dcn30_bb_max_clk->max_phyclk_mhz) + dcn30_bb_max_clk->max_phyclk_mhz = dcn3_0_soc.clock_limits[0].phyclk_mhz; +} + +void dcn30_fpu_get_optimal_dcfclk_fclk_for_uclk(unsigned int uclk_mts, + unsigned int *optimal_dcfclk, + unsigned int *optimal_fclk) +{ + double bw_from_dram, bw_from_dram1, bw_from_dram2; + + dc_assert_fp_enabled(); + + bw_from_dram1 = uclk_mts * dcn3_0_soc.num_chans * + dcn3_0_soc.dram_channel_width_bytes * (dcn3_0_soc.max_avg_dram_bw_use_normal_percent / 100); + bw_from_dram2 = uclk_mts * dcn3_0_soc.num_chans * + dcn3_0_soc.dram_channel_width_bytes * (dcn3_0_soc.max_avg_sdp_bw_use_normal_percent / 100); + + bw_from_dram = (bw_from_dram1 < bw_from_dram2) ? bw_from_dram1 : bw_from_dram2; + + if (optimal_fclk) + *optimal_fclk = bw_from_dram / + (dcn3_0_soc.fabric_datapath_to_dcn_data_return_bytes * (dcn3_0_soc.max_avg_sdp_bw_use_normal_percent / 100)); + + if (optimal_dcfclk) + *optimal_dcfclk = bw_from_dram / + (dcn3_0_soc.return_bus_width_bytes * (dcn3_0_soc.max_avg_sdp_bw_use_normal_percent / 100)); +} + +void dcn30_fpu_update_bw_bounding_box(struct dc *dc, + struct clk_bw_params *bw_params, + struct dc_bounding_box_max_clk *dcn30_bb_max_clk, + unsigned int *dcfclk_mhz, + unsigned int *dram_speed_mts) +{ + unsigned int i; + + dc_assert_fp_enabled(); + + dcn3_0_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; + + for (i = 0; i < dcn3_0_soc.num_states; i++) { + dcn3_0_soc.clock_limits[i].state = i; + dcn3_0_soc.clock_limits[i].dcfclk_mhz = dcfclk_mhz[i]; + dcn3_0_soc.clock_limits[i].fabricclk_mhz = dcfclk_mhz[i]; + dcn3_0_soc.clock_limits[i].dram_speed_mts = dram_speed_mts[i]; + + /* Fill all states with max values of all other clocks */ + dcn3_0_soc.clock_limits[i].dispclk_mhz = dcn30_bb_max_clk->max_dispclk_mhz; + dcn3_0_soc.clock_limits[i].dppclk_mhz = dcn30_bb_max_clk->max_dppclk_mhz; + dcn3_0_soc.clock_limits[i].phyclk_mhz = dcn30_bb_max_clk->max_phyclk_mhz; + dcn3_0_soc.clock_limits[i].dtbclk_mhz = dcn3_0_soc.clock_limits[0].dtbclk_mhz; + /* These clocks cannot come from bw_params, always fill from dcn3_0_soc[1] */ + /* FCLK, PHYCLK_D18, SOCCLK, DSCCLK */ + dcn3_0_soc.clock_limits[i].phyclk_d18_mhz = dcn3_0_soc.clock_limits[0].phyclk_d18_mhz; + dcn3_0_soc.clock_limits[i].socclk_mhz = dcn3_0_soc.clock_limits[0].socclk_mhz; + dcn3_0_soc.clock_limits[i].dscclk_mhz = dcn3_0_soc.clock_limits[0].dscclk_mhz; + } + /* re-init DML with updated bb */ + dml_init_instance(&dc->dml, &dcn3_0_soc, &dcn3_0_ip, DML_PROJECT_DCN30); + if (dc->current_state) + dml_init_instance(&dc->current_state->bw_ctx.dml, &dcn3_0_soc, &dcn3_0_ip, DML_PROJECT_DCN30); + +} + + diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn30/dcn30_fpu.h b/drivers/gpu/drm/amd/display/dc/dml/dcn30/dcn30_fpu.h new file mode 100644 index 000000000000..dedfe7b5f173 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn30/dcn30_fpu.h @@ -0,0 +1,67 @@ +/* + * Copyright 2020-2021 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef __DCN30_FPU_H__ +#define __DCN30_FPU_H__ + +#include "core_types.h" +#include "dcn20/dcn20_optc.h" + +void optc3_fpu_set_vrr_m_const(struct timing_generator *optc, + double vtotal_avg); + +void dcn30_fpu_populate_dml_writeback_from_context( + struct dc *dc, struct resource_context *res_ctx, display_e2e_pipe_params_st *pipes); + +void dcn30_fpu_set_mcif_arb_params(struct mcif_arb_params *wb_arb_params, + struct display_mode_lib *dml, + display_e2e_pipe_params_st *pipes, + int pipe_cnt, + int cur_pipe); + +void dcn30_fpu_update_soc_for_wm_a(struct dc *dc, struct dc_state *context); + +void dcn30_fpu_calculate_wm_and_dlg( + struct dc *dc, struct dc_state *context, + display_e2e_pipe_params_st *pipes, + int pipe_cnt, + int vlevel); + +void dcn30_fpu_update_dram_channel_width_bytes(struct dc *dc); + +void dcn30_fpu_update_max_clk(struct dc_bounding_box_max_clk *dcn30_bb_max_clk); + +void dcn30_fpu_get_optimal_dcfclk_fclk_for_uclk(unsigned int uclk_mts, + unsigned int *optimal_dcfclk, + unsigned int *optimal_fclk); + +void dcn30_fpu_update_bw_bounding_box(struct dc *dc, + struct clk_bw_params *bw_params, + struct dc_bounding_box_max_clk *dcn30_bb_max_clk, + unsigned int *dcfclk_mhz, + unsigned int *dram_speed_mts); + + +#endif /* __DCN30_FPU_H__*/ diff --git a/drivers/gpu/drm/amd/display/dc/dml/dml_wrapper.c b/drivers/gpu/drm/amd/display/dc/dml/dml_wrapper.c index 789f7562cdc7..d2273674e872 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dml_wrapper.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dml_wrapper.c @@ -1284,10 +1284,8 @@ static bool is_dtbclk_required(struct dc *dc, struct dc_state *context) for (i = 0; i < dc->res_pool->pipe_count; i++) { if (!context->res_ctx.pipe_ctx[i].stream) continue; -#if defined (CONFIG_DRM_AMD_DC_DP2_0) if (is_dp_128b_132b_signal(&context->res_ctx.pipe_ctx[i])) return true; -#endif } return false; } diff --git a/drivers/gpu/drm/amd/display/dc/inc/core_types.h b/drivers/gpu/drm/amd/display/dc/inc/core_types.h index 26f3a55c35d7..555d4d9e1454 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/core_types.h +++ b/drivers/gpu/drm/amd/display/dc/inc/core_types.h @@ -486,4 +486,11 @@ struct dc_state { } perf_params; }; +struct dc_bounding_box_max_clk { + int max_dcfclk_mhz; + int max_dispclk_mhz; + int max_dppclk_mhz; + int max_phyclk_mhz; +}; + #endif /* _CORE_TYPES_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/inc/dc_link_dp.h b/drivers/gpu/drm/amd/display/dc/inc/dc_link_dp.h index 851b98299063..a3c1e9c56d8b 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/dc_link_dp.h +++ b/drivers/gpu/drm/amd/display/dc/inc/dc_link_dp.h @@ -217,8 +217,7 @@ void disable_dp_hpo_output(struct dc_link *link, void setup_dp_hpo_stream(struct pipe_ctx *pipe_ctx, bool enable); bool is_dp_128b_132b_signal(struct pipe_ctx *pipe_ctx); -void dp_retrieve_lttpr_cap(struct dc_link *link); -bool dp_apply_lttpr_mode(struct dc_link *link); +bool dp_retrieve_lttpr_cap(struct dc_link *link); void edp_panel_backlight_power_on(struct dc_link *link); void dp_receiver_power_ctrl(struct dc_link *link, bool on); void dp_source_sequence_trace(struct dc_link *link, uint8_t dp_test_mode); diff --git a/drivers/gpu/drm/amd/display/dc/inc/dc_link_dpia.h b/drivers/gpu/drm/amd/display/dc/inc/dc_link_dpia.h index 74dafd0f9d3d..39c1d1d07357 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/dc_link_dpia.h +++ b/drivers/gpu/drm/amd/display/dc/inc/dc_link_dpia.h @@ -87,6 +87,11 @@ union dpia_set_config_data { */ enum dc_status dpcd_get_tunneling_device_data(struct dc_link *link); +/* Query hot plug status of USB4 DP tunnel. + * Returns true if HPD high. + */ +bool dc_link_dpia_query_hpd_status(struct dc_link *link); + /* Train DP tunneling link for USB4 DPIA display endpoint. * DPIA equivalent of dc_link_dp_perfrorm_link_training. * Aborts link training upon detection of sink unplug. diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h b/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h index 46ce5a0ee4ec..b5570aa8e39d 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h @@ -237,6 +237,7 @@ struct clk_mgr_funcs { bool safe_to_lower); int (*get_dp_ref_clk_frequency)(struct clk_mgr *clk_mgr); + int (*get_dtb_ref_clk_frequency)(struct clk_mgr *clk_mgr); void (*set_low_power_state)(struct clk_mgr *clk_mgr); diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h index b2fa4de47734..c7021915bac8 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h @@ -60,8 +60,17 @@ struct dccg { const struct dccg_funcs *funcs; int pipe_dppclk_khz[MAX_PIPES]; int ref_dppclk; - int dtbclk_khz[MAX_PIPES]; - int audio_dtbclk_khz; + //int dtbclk_khz[MAX_PIPES];/* TODO needs to be removed */ + //int audio_dtbclk_khz;/* TODO needs to be removed */ + int ref_dtbclk_khz;/* TODO needs to be removed */ +}; + +struct dtbclk_dto_params { + const struct dc_crtc_timing *timing; + int otg_inst; + int pixclk_khz; + int req_audio_dtbclk_khz; + int num_odm_segments; int ref_dtbclk_khz; }; @@ -111,10 +120,7 @@ struct dccg_funcs { void (*set_dtbclk_dto)( struct dccg *dccg, - int dtbclk_inst, - int req_dtbclk_khz, - int num_odm_segments, - const struct dc_crtc_timing *timing); + struct dtbclk_dto_params *dto_params); void (*set_audio_dtbclk_dto)( struct dccg *dccg, diff --git a/drivers/gpu/drm/amd/display/dc/link/link_hwss_hpo_dp.c b/drivers/gpu/drm/amd/display/dc/link/link_hwss_hpo_dp.c index 87972dc8443d..ea6cf8bfce30 100644 --- a/drivers/gpu/drm/amd/display/dc/link/link_hwss_hpo_dp.c +++ b/drivers/gpu/drm/amd/display/dc/link/link_hwss_hpo_dp.c @@ -27,6 +27,7 @@ #include "core_types.h" #include "dccg.h" #include "dc_link_dp.h" +#include "clk_mgr.h" static enum phyd32clk_clock_source get_phyd32clk_src(struct dc_link *link) { @@ -106,14 +107,18 @@ static void setup_hpo_dp_stream_encoder(struct pipe_ctx *pipe_ctx) struct hpo_dp_link_encoder *link_enc = pipe_ctx->link_res.hpo_dp_link_enc; struct dccg *dccg = dc->res_pool->dccg; struct timing_generator *tg = pipe_ctx->stream_res.tg; - int odm_segment_count = get_odm_segment_count(pipe_ctx); + struct dtbclk_dto_params dto_params = {0}; enum phyd32clk_clock_source phyd32clk = get_phyd32clk_src(pipe_ctx->stream->link); + dto_params.otg_inst = tg->inst; + dto_params.pixclk_khz = pipe_ctx->stream->phy_pix_clk; + dto_params.num_odm_segments = get_odm_segment_count(pipe_ctx); + dto_params.timing = &pipe_ctx->stream->timing; + dto_params.ref_dtbclk_khz = dc->clk_mgr->funcs->get_dtb_ref_clk_frequency(dc->clk_mgr); + dccg->funcs->set_dpstreamclk(dccg, DTBCLK0, tg->inst); dccg->funcs->enable_symclk32_se(dccg, stream_enc->inst, phyd32clk); - dccg->funcs->set_dtbclk_dto(dccg, tg->inst, pipe_ctx->stream->phy_pix_clk, - odm_segment_count, - &pipe_ctx->stream->timing); + dccg->funcs->set_dtbclk_dto(dccg, &dto_params); stream_enc->funcs->enable_stream(stream_enc); stream_enc->funcs->map_stream_to_link(stream_enc, stream_enc->inst, link_enc->inst); } @@ -124,9 +129,13 @@ static void reset_hpo_dp_stream_encoder(struct pipe_ctx *pipe_ctx) struct hpo_dp_stream_encoder *stream_enc = pipe_ctx->stream_res.hpo_dp_stream_enc; struct dccg *dccg = dc->res_pool->dccg; struct timing_generator *tg = pipe_ctx->stream_res.tg; + struct dtbclk_dto_params dto_params = {0}; + + dto_params.otg_inst = tg->inst; + dto_params.timing = &pipe_ctx->stream->timing; stream_enc->funcs->disable(stream_enc); - dccg->funcs->set_dtbclk_dto(dccg, tg->inst, 0, 0, &pipe_ctx->stream->timing); + dccg->funcs->set_dtbclk_dto(dccg, &dto_params); dccg->funcs->disable_symclk32_se(dccg, stream_enc->inst); dccg->funcs->set_dpstreamclk(dccg, REFCLK, tg->inst); } diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c index 7c9330a61ac1..c7bd7e216710 100644 --- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c @@ -84,7 +84,7 @@ void dmub_dcn31_reset(struct dmub_srv *dmub) { union dmub_gpint_data_register cmd; const uint32_t timeout = 100; - uint32_t in_reset, scratch, i; + uint32_t in_reset, scratch, i, pwait_mode; REG_GET(DMCUB_CNTL2, DMCUB_SOFT_RESET, &in_reset); @@ -115,6 +115,13 @@ void dmub_dcn31_reset(struct dmub_srv *dmub) udelay(1); } + for (i = 0; i < timeout; ++i) { + REG_GET(DMCUB_CNTL, DMCUB_PWAIT_MODE_STATUS, &pwait_mode); + if (pwait_mode & (1 << 0)) + break; + + udelay(1); + } /* Force reset in case we timed out, DMCUB is likely hung. */ } @@ -125,6 +132,8 @@ void dmub_dcn31_reset(struct dmub_srv *dmub) REG_WRITE(DMCUB_INBOX1_WPTR, 0); REG_WRITE(DMCUB_OUTBOX1_RPTR, 0); REG_WRITE(DMCUB_OUTBOX1_WPTR, 0); + REG_WRITE(DMCUB_OUTBOX0_RPTR, 0); + REG_WRITE(DMCUB_OUTBOX0_WPTR, 0); REG_WRITE(DMCUB_SCRATCH0, 0); /* Clear the GPINT command manually so we don't send anything during boot. */ diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.h b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.h index 59ddc81b5a0e..f6db6f89d45d 100644 --- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.h +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.h @@ -151,7 +151,8 @@ struct dmub_srv; DMUB_SF(DCN_VM_FB_OFFSET, FB_OFFSET) \ DMUB_SF(DMCUB_INBOX0_WPTR, DMCUB_INBOX0_WPTR) \ DMUB_SF(DMCUB_INTERRUPT_ENABLE, DMCUB_GPINT_IH_INT_EN) \ - DMUB_SF(DMCUB_INTERRUPT_ACK, DMCUB_GPINT_IH_INT_ACK) + DMUB_SF(DMCUB_INTERRUPT_ACK, DMCUB_GPINT_IH_INT_ACK) \ + DMUB_SF(DMCUB_CNTL, DMCUB_PWAIT_MODE_STATUS) struct dmub_srv_dcn31_reg_offset { #define DMUB_SR(reg) uint32_t reg; diff --git a/drivers/gpu/drm/amd/display/include/ddc_service_types.h b/drivers/gpu/drm/amd/display/include/ddc_service_types.h index 73b9e0a87e54..20a3d4e23f66 100644 --- a/drivers/gpu/drm/amd/display/include/ddc_service_types.h +++ b/drivers/gpu/drm/amd/display/include/ddc_service_types.h @@ -127,6 +127,8 @@ struct av_sync_data { static const uint8_t DP_SINK_DEVICE_STR_ID_1[] = {7, 1, 8, 7, 3, 0}; static const uint8_t DP_SINK_DEVICE_STR_ID_2[] = {7, 1, 8, 7, 5, 0}; +static const u8 DP_SINK_BRANCH_DEV_NAME_7580[] = "7580\x80u"; + /*MST Dock*/ static const uint8_t SYNAPTICS_DEVICE_ID[] = "SYNA"; diff --git a/drivers/gpu/drm/amd/display/include/link_service_types.h b/drivers/gpu/drm/amd/display/include/link_service_types.h index 9f465b4d626e..447a56286dd0 100644 --- a/drivers/gpu/drm/amd/display/include/link_service_types.h +++ b/drivers/gpu/drm/amd/display/include/link_service_types.h @@ -80,12 +80,6 @@ enum link_training_result { DP_128b_132b_CDS_DONE_TIMEOUT, }; -enum lttpr_support { - LTTPR_UNSUPPORTED, - LTTPR_CHECK_EXT_SUPPORT, - LTTPR_SUPPORTED, -}; - enum lttpr_mode { LTTPR_MODE_NON_LTTPR, LTTPR_MODE_TRANSPARENT, |