Resguardo inicial

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/Makefile

@@ -0,0 +1,150 @@
+#
+# Copyright 2017 Advanced Micro Devices, Inc.
+# Copyright 2019 Raptor Engineering, LLC
+#
+# 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.
+#
+#
+# Makefile for the 'utils' sub-component of DAL.
+# It provides the general basic services required by other DAL
+# subcomponents.
+
+dml_ccflags := $(CC_FLAGS_FPU)
+dml_rcflags := $(CC_FLAGS_NO_FPU)
+
+ifneq ($(CONFIG_FRAME_WARN),0)
+    ifeq ($(filter y,$(CONFIG_KASAN)$(CONFIG_KCSAN)),y)
+        ifeq ($(CONFIG_CC_IS_CLANG)$(CONFIG_COMPILE_TEST),yy)
+            frame_warn_limit := 4096
+        else
+            frame_warn_limit := 3072
+        endif
+    else
+        frame_warn_limit := 2048
+    endif
+
+    ifeq ($(call test-lt, $(CONFIG_FRAME_WARN), $(frame_warn_limit)),y)
+        frame_warn_flag := -Wframe-larger-than=$(frame_warn_limit)
+    endif
+endif
+
+CFLAGS_$(AMDDALPATH)/dc/dml/display_mode_lib.o := $(dml_ccflags)
+CFLAGS_$(AMDDALPATH)/dc/dml/display_mode_vba.o := $(dml_ccflags)
+CFLAGS_$(AMDDALPATH)/dc/dml/dcn10/dcn10_fpu.o := $(dml_ccflags)
+CFLAGS_$(AMDDALPATH)/dc/dml/dcn20/dcn20_fpu.o := $(dml_ccflags)
+CFLAGS_$(AMDDALPATH)/dc/dml/dcn20/display_mode_vba_20.o := $(dml_ccflags) $(frame_warn_flag)
+CFLAGS_$(AMDDALPATH)/dc/dml/dcn20/display_rq_dlg_calc_20.o := $(dml_ccflags)
+CFLAGS_$(AMDDALPATH)/dc/dml/dcn20/display_mode_vba_20v2.o := $(dml_ccflags) $(frame_warn_flag)
+CFLAGS_$(AMDDALPATH)/dc/dml/dcn20/display_rq_dlg_calc_20v2.o := $(dml_ccflags)
+CFLAGS_$(AMDDALPATH)/dc/dml/dcn21/display_mode_vba_21.o := $(dml_ccflags) $(frame_warn_flag)
+CFLAGS_$(AMDDALPATH)/dc/dml/dcn21/display_rq_dlg_calc_21.o := $(dml_ccflags)
+CFLAGS_$(AMDDALPATH)/dc/dml/dcn30/display_mode_vba_30.o := $(dml_ccflags) $(frame_warn_flag)
+CFLAGS_$(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/dcn314/display_mode_vba_314.o := $(dml_ccflags) $(frame_warn_flag)
+CFLAGS_$(AMDDALPATH)/dc/dml/dcn314/display_rq_dlg_calc_314.o := $(dml_ccflags)
+CFLAGS_$(AMDDALPATH)/dc/dml/dcn314/dcn314_fpu.o := $(dml_ccflags)
+CFLAGS_$(AMDDALPATH)/dc/dml/dcn30/dcn30_fpu.o := $(dml_ccflags)
+CFLAGS_$(AMDDALPATH)/dc/dml/dcn32/dcn32_fpu.o := $(dml_ccflags)
+CFLAGS_$(AMDDALPATH)/dc/dml/dcn32/display_mode_vba_32.o := $(dml_ccflags) $(frame_warn_flag)
+CFLAGS_$(AMDDALPATH)/dc/dml/dcn32/display_rq_dlg_calc_32.o := $(dml_ccflags)
+CFLAGS_$(AMDDALPATH)/dc/dml/dcn32/display_mode_vba_util_32.o := $(dml_ccflags) $(frame_warn_flag)
+CFLAGS_$(AMDDALPATH)/dc/dml/dcn321/dcn321_fpu.o := $(dml_ccflags)
+CFLAGS_$(AMDDALPATH)/dc/dml/dcn35/dcn35_fpu.o := $(dml_ccflags)
+CFLAGS_$(AMDDALPATH)/dc/dml/dcn351/dcn351_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 := $(dml_ccflags)
+CFLAGS_$(AMDDALPATH)/dc/dml/dcn303/dcn303_fpu.o := $(dml_ccflags)
+CFLAGS_$(AMDDALPATH)/dc/dml/dsc/rc_calc_fpu.o := $(dml_ccflags)
+CFLAGS_$(AMDDALPATH)/dc/dml/calcs/dcn_calcs.o := $(dml_ccflags)
+CFLAGS_$(AMDDALPATH)/dc/dml/calcs/dcn_calc_auto.o := $(dml_ccflags)
+CFLAGS_$(AMDDALPATH)/dc/dml/calcs/dcn_calc_math.o := $(dml_ccflags) -Wno-tautological-compare
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/display_mode_vba.o := $(dml_rcflags)
+CFLAGS_$(AMDDALPATH)/dc/dml/dml1_frl_cap_chk.o := $(dml_ccflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn20/display_mode_vba_20.o := $(dml_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn20/display_rq_dlg_calc_20.o := $(dml_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn20/display_mode_vba_20v2.o := $(dml_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn20/display_rq_dlg_calc_20v2.o := $(dml_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn21/display_mode_vba_21.o := $(dml_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn21/display_rq_dlg_calc_21.o := $(dml_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn30/display_mode_vba_30.o := $(dml_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn30/display_rq_dlg_calc_30.o := $(dml_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn31/display_mode_vba_31.o := $(dml_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn31/display_rq_dlg_calc_31.o := $(dml_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn32/display_mode_vba_32.o := $(dml_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn32/display_rq_dlg_calc_32.o := $(dml_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn32/display_mode_vba_util_32.o := $(dml_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn314/display_mode_vba_314.o := $(dml_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn314/display_rq_dlg_calc_314.o := $(dml_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn301/dcn301_fpu.o := $(dml_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/display_mode_lib.o := $(dml_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dsc/rc_calc_fpu.o  := $(dml_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn10/dcn10_fpu.o := $(dml_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn20/dcn20_fpu.o := $(dml_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn314/display_mode_vba_314.o := $(dml_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn314/display_rq_dlg_calc_314.o := $(dml_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn314/dcn314_fpu.o := $(dml_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn30/dcn30_fpu.o := $(dml_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn32/dcn32_fpu.o := $(dml_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn321/dcn321_fpu.o := $(dml_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn35/dcn35_fpu.o := $(dml_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn351/dcn351_fpu.o := $(dml_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn31/dcn31_fpu.o := $(dml_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn302/dcn302_fpu.o := $(dml_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn303/dcn303_fpu.o := $(dml_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/calcs/dcn_calc_math.o := $(dml_rcflags)
+CFLAGS_$(AMDDALPATH)/dc/dml/dml1_display_rq_dlg_calc.o := $(dml_ccflags)
+CFLAGS_$(AMDDALPATH)/dc/dml/display_rq_dlg_helpers.o := $(dml_ccflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dml1_display_rq_dlg_calc.o := $(dml_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/display_rq_dlg_helpers.o := $(dml_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/calcs/dcn_calcs.o := $(dml_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/calcs/dcn_calc_auto.o := $(dml_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/calcs/dcn_calc_math.o := $(dml_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dml1_frl_cap_chk.o := $(dml_rcflags)
+
+ifdef CONFIG_DRM_AMD_DC_FP
+DML += display_mode_lib.o display_rq_dlg_helpers.o dml1_display_rq_dlg_calc.o
+DML += dml1_frl_cap_chk.o
+DML += dcn10/dcn10_fpu.o
+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/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 += dcn314/display_mode_vba_314.o dcn314/display_rq_dlg_calc_314.o
+DML += dcn32/display_mode_vba_32.o dcn32/display_rq_dlg_calc_32.o dcn32/display_mode_vba_util_32.o
+DML += dcn31/dcn31_fpu.o
+DML += dcn32/dcn32_fpu.o
+DML += dcn321/dcn321_fpu.o
+DML += dcn301/dcn301_fpu.o
+DML += dcn302/dcn302_fpu.o
+DML += dcn303/dcn303_fpu.o
+DML += dcn314/dcn314_fpu.o
+DML += dcn35/dcn35_fpu.o
+DML += dcn351/dcn351_fpu.o
+DML += dsc/rc_calc_fpu.o
+DML += calcs/dcn_calcs.o calcs/dcn_calc_math.o calcs/dcn_calc_auto.o
+endif
+
+AMD_DAL_DML = $(addprefix $(AMDDALPATH)/dc/dml/,$(DML))
+
+AMD_DISPLAY_FILES += $(AMD_DAL_DML)

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/calcs/dcn_calc_auto.h

@@ -0,0 +1,38 @@
+/*
+ * Copyright 2017 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 _DCN_CALC_AUTO_H_
+#define _DCN_CALC_AUTO_H_
+
+#include "dc.h"
+#include "dcn_calcs.h"
+
+void scaler_settings_calculation(struct dcn_bw_internal_vars *v);
+void mode_support_and_system_configuration(struct dcn_bw_internal_vars *v);
+void display_pipe_configuration(struct dcn_bw_internal_vars *v);
+void dispclkdppclkdcfclk_deep_sleep_prefetch_parameters_watermarks_and_performance_calculation(
+		struct dcn_bw_internal_vars *v);
+
+#endif /* _DCN_CALC_AUTO_H_ */

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/calcs/dcn_calc_math.c

@@ -0,0 +1,145 @@
+/*
+ * Copyright 2017 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 "os_types.h"
+#include "dcn_calc_math.h"
+
+#define isNaN(number) ((number) != (number))
+
+/*
+ * NOTE:
+ *   This file is gcc-parseable HW gospel, coming straight from HW engineers.
+ *
+ * It doesn't adhere to Linux kernel style and sometimes will do things in odd
+ * ways. Unless there is something clearly wrong with it the code should
+ * remain as-is as it provides us with a guarantee from HW that it is correct.
+ */
+
+float dcn_bw_mod(const float arg1, const float arg2)
+{
+	if (isNaN(arg1))
+		return arg2;
+	if (isNaN(arg2))
+		return arg1;
+	return arg1 - arg1 * ((int) (arg1 / arg2));
+}
+
+float dcn_bw_min2(const float arg1, const float arg2)
+{
+	if (isNaN(arg1))
+		return arg2;
+	if (isNaN(arg2))
+		return arg1;
+	return arg1 < arg2 ? arg1 : arg2;
+}
+
+unsigned int dcn_bw_max(const unsigned int arg1, const unsigned int arg2)
+{
+	return arg1 > arg2 ? arg1 : arg2;
+}
+float dcn_bw_max2(const float arg1, const float arg2)
+{
+	if (isNaN(arg1))
+		return arg2;
+	if (isNaN(arg2))
+		return arg1;
+	return arg1 > arg2 ? arg1 : arg2;
+}
+
+float dcn_bw_floor2(const float arg, const float significance)
+{
+	ASSERT(significance != 0);
+
+	return ((int) (arg / significance)) * significance;
+}
+float dcn_bw_floor(const float arg)
+{
+	return (float)((int)(arg));
+}
+
+float dcn_bw_ceil(const float arg)
+{
+	return (float)((int)(arg + 0.99999f));
+}
+
+float dcn_bw_ceil2(const float arg, const float significance)
+{
+	ASSERT(significance != 0);
+
+	return ((int) (arg / significance + 0.99999)) * significance;
+}
+
+float dcn_bw_max3(float v1, float v2, float v3)
+{
+	return v3 > dcn_bw_max2(v1, v2) ? v3 : dcn_bw_max2(v1, v2);
+}
+
+float dcn_bw_max5(float v1, float v2, float v3, float v4, float v5)
+{
+	return dcn_bw_max3(v1, v2, v3) > dcn_bw_max2(v4, v5) ? dcn_bw_max3(v1, v2, v3) : dcn_bw_max2(v4, v5);
+}
+
+float dcn_bw_pow(float a, float exp)
+{
+	float temp;
+	/*ASSERT(exp == (int)exp);*/
+	if ((int)exp == 0)
+		return 1;
+	temp = dcn_bw_pow(a, (float)((int)(exp / 2)));
+	if (((int)exp % 2) == 0) {
+		return temp * temp;
+	} else {
+		if ((int)exp > 0)
+			return a * temp * temp;
+		else
+			return (temp * temp) / a;
+	}
+}
+
+double dcn_bw_fabs(double a)
+{
+	if (a > 0)
+		return (a);
+	else
+		return (-a);
+}
+
+
+float dcn_bw_log(float a, float b)
+{
+	int * const exp_ptr = (int *)(&a);
+	int x = *exp_ptr;
+	const int log_2 = ((x >> 23) & 255) - 128;
+	x &= ~(255 << 23);
+	x += 127 << 23;
+	*exp_ptr = x;
+
+	a = ((-1.0f / 3) * a + 2) * a - 2.0f / 3;
+
+	if (b > 2.00001 || b < 1.99999)
+		return (a + log_2) / dcn_bw_log(b, 2);
+	else
+		return (a + log_2);
+}

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dc_features.h

@@ -0,0 +1,566 @@
+/*
+ * Copyright 2017 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 __DC_FEATURES_H__
+#define __DC_FEATURES_H__
+
+// local features
+#define DC__PRESENT 1
+#define DC__PRESENT__1 1
+#define DC__NUM_DPP 4
+
+/**
+ * @DC__VOLTAGE_STATES:
+ *
+ * Define the maximum amount of states supported by the ASIC. Every ASIC has a
+ * specific number of states; this macro defines the maximum number of states.
+ */
+#define DC__VOLTAGE_STATES 40
+#define DC__NUM_DPP__4 1
+#define DC__NUM_DPP__0_PRESENT 1
+#define DC__NUM_DPP__1_PRESENT 1
+#define DC__NUM_DPP__2_PRESENT 1
+#define DC__NUM_DPP__3_PRESENT 1
+#define DC__NUM_DPP__MAX 8
+#define DC__NUM_DPP__MAX__8 1
+#define DC__PIPE_10BIT 0
+#define DC__PIPE_10BIT__0 1
+#define DC__PIPE_10BIT__MAX 1
+#define DC__PIPE_10BIT__MAX__1 1
+#define DC__NUM_OPP 4
+#define DC__NUM_OPP__4 1
+#define DC__NUM_OPP__0_PRESENT 1
+#define DC__NUM_OPP__1_PRESENT 1
+#define DC__NUM_OPP__2_PRESENT 1
+#define DC__NUM_OPP__3_PRESENT 1
+#define DC__NUM_OPP__MAX 6
+#define DC__NUM_OPP__MAX__6 1
+#define DC__NUM_DSC 0
+#define DC__NUM_DSC__0 1
+#define DC__NUM_DSC__MAX 6
+#define DC__NUM_DSC__MAX__6 1
+#define DC__NUM_ABM 1
+#define DC__NUM_ABM__1 1
+#define DC__NUM_ABM__0_PRESENT 1
+#define DC__NUM_ABM__MAX 2
+#define DC__NUM_ABM__MAX__2 1
+#define DC__ODM_PRESENT 0
+#define DC__ODM_PRESENT__0 1
+#define DC__NUM_OTG 4
+#define DC__NUM_OTG__4 1
+#define DC__NUM_OTG__0_PRESENT 1
+#define DC__NUM_OTG__1_PRESENT 1
+#define DC__NUM_OTG__2_PRESENT 1
+#define DC__NUM_OTG__3_PRESENT 1
+#define DC__NUM_OTG__MAX 6
+#define DC__NUM_OTG__MAX__6 1
+#define DC__NUM_DWB 2
+#define DC__NUM_DWB__2 1
+#define DC__NUM_DWB__0_PRESENT 1
+#define DC__NUM_DWB__1_PRESENT 1
+#define DC__NUM_DWB__MAX 2
+#define DC__NUM_DWB__MAX__2 1
+#define DC__NUM_DIG 4
+#define DC__NUM_DIG__4 1
+#define DC__NUM_DIG__0_PRESENT 1
+#define DC__NUM_DIG__1_PRESENT 1
+#define DC__NUM_DIG__2_PRESENT 1
+#define DC__NUM_DIG__3_PRESENT 1
+#define DC__NUM_DIG__MAX 6
+#define DC__NUM_DIG__MAX__6 1
+#define DC__NUM_AUX 4
+#define DC__NUM_AUX__4 1
+#define DC__NUM_AUX__0_PRESENT 1
+#define DC__NUM_AUX__1_PRESENT 1
+#define DC__NUM_AUX__2_PRESENT 1
+#define DC__NUM_AUX__3_PRESENT 1
+#define DC__NUM_AUX__MAX 6
+#define DC__NUM_AUX__MAX__6 1
+#define DC__NUM_AUDIO_STREAMS 4
+#define DC__NUM_AUDIO_STREAMS__4 1
+#define DC__NUM_AUDIO_STREAMS__0_PRESENT 1
+#define DC__NUM_AUDIO_STREAMS__1_PRESENT 1
+#define DC__NUM_AUDIO_STREAMS__2_PRESENT 1
+#define DC__NUM_AUDIO_STREAMS__3_PRESENT 1
+#define DC__NUM_AUDIO_STREAMS__MAX 8
+#define DC__NUM_AUDIO_STREAMS__MAX__8 1
+#define DC__NUM_AUDIO_ENDPOINTS 6
+#define DC__NUM_AUDIO_ENDPOINTS__6 1
+#define DC__NUM_AUDIO_ENDPOINTS__0_PRESENT 1
+#define DC__NUM_AUDIO_ENDPOINTS__1_PRESENT 1
+#define DC__NUM_AUDIO_ENDPOINTS__2_PRESENT 1
+#define DC__NUM_AUDIO_ENDPOINTS__3_PRESENT 1
+#define DC__NUM_AUDIO_ENDPOINTS__4_PRESENT 1
+#define DC__NUM_AUDIO_ENDPOINTS__5_PRESENT 1
+#define DC__NUM_AUDIO_ENDPOINTS__MAX 8
+#define DC__NUM_AUDIO_ENDPOINTS__MAX__8 1
+#define DC__NUM_AUDIO_INPUT_STREAMS 0
+#define DC__NUM_AUDIO_INPUT_STREAMS__0 1
+#define DC__NUM_AUDIO_INPUT_STREAMS__MAX 8
+#define DC__NUM_AUDIO_INPUT_STREAMS__MAX__8 1
+#define DC__NUM_AUDIO_INPUT_ENDPOINTS 0
+#define DC__NUM_AUDIO_INPUT_ENDPOINTS__0 1
+#define DC__NUM_AUDIO_INPUT_ENDPOINTS__MAX 8
+#define DC__NUM_AUDIO_INPUT_ENDPOINTS__MAX__8 1
+#define DC__NUM_CURSOR 1
+#define DC__NUM_CURSOR__1 1
+#define DC__NUM_CURSOR__0_PRESENT 1
+#define DC__NUM_CURSOR__MAX 2
+#define DC__NUM_CURSOR__MAX__2 1
+#define DC__DIGITAL_BYPASS_PRESENT 0
+#define DC__DIGITAL_BYPASS_PRESENT__0 1
+#define DC__HCID_HWMAJVER 1
+#define DC__HCID_HWMAJVER__1 1
+#define DC__HCID_HWMINVER 0
+#define DC__HCID_HWMINVER__0 1
+#define DC__HCID_HWREV 0
+#define DC__HCID_HWREV__0 1
+#define DC__ROMSTRAP_PRESENT 0
+#define DC__ROMSTRAP_PRESENT__0 1
+#define DC__NUM_RBBMIF_DECODES 30
+#define DC__NUM_RBBMIF_DECODES__30 1
+#define DC__NUM_DBG_REGS 36
+#define DC__NUM_DBG_REGS__36 1
+#define DC__NUM_PIPES_UNDERLAY 0
+#define DC__NUM_PIPES_UNDERLAY__0 1
+#define DC__NUM_PIPES_UNDERLAY__MAX 2
+#define DC__NUM_PIPES_UNDERLAY__MAX__2 1
+#define DC__NUM_VCE_ENGINE 1
+#define DC__NUM_VCE_ENGINE__1 1
+#define DC__NUM_VCE_ENGINE__0_PRESENT 1
+#define DC__NUM_VCE_ENGINE__MAX 2
+#define DC__NUM_VCE_ENGINE__MAX__2 1
+#define DC__OTG_EXTERNAL_SYNC_PRESENT 0
+#define DC__OTG_EXTERNAL_SYNC_PRESENT__0 1
+#define DC__OTG_CRC_PRESENT 1
+#define DC__OTG_CRC_PRESENT__1 1
+#define DC__VIP_PRESENT 0
+#define DC__VIP_PRESENT__0 1
+#define DC__DTMTEST_PRESENT 0
+#define DC__DTMTEST_PRESENT__0 1
+#define DC__POWER_GATE_PRESENT 1
+#define DC__POWER_GATE_PRESENT__1 1
+#define DC__MEM_PG 1
+#define DC__MEM_PG__1 1
+#define DC__FMT_SRC_SEL_PRESENT 0
+#define DC__FMT_SRC_SEL_PRESENT__0 1
+#define DC__DIG_FEATURES__HDMI_PRESENT 1
+#define DC__DIG_FEATURES__HDMI_PRESENT__1 1
+#define DC__DIG_FEATURES__DP_PRESENT 1
+#define DC__DIG_FEATURES__DP_PRESENT__1 1
+#define DC__DIG_FEATURES__DP_MST_PRESENT 1
+#define DC__DIG_FEATURES__DP_MST_PRESENT__1 1
+#define DC__DIG_LP_FEATURES__HDMI_PRESENT 0
+#define DC__DIG_LP_FEATURES__HDMI_PRESENT__0 1
+#define DC__DIG_LP_FEATURES__DP_PRESENT 1
+#define DC__DIG_LP_FEATURES__DP_PRESENT__1 1
+#define DC__DIG_LP_FEATURES__DP_MST_PRESENT 0
+#define DC__DIG_LP_FEATURES__DP_MST_PRESENT__0 1
+#define DC__DIG_RESYNC_FIFO_SIZE 14
+#define DC__DIG_RESYNC_FIFO_SIZE__14 1
+#define DC__DIG_RESYNC_FIFO_SIZE__0_PRESENT 1
+#define DC__DIG_RESYNC_FIFO_SIZE__1_PRESENT 1
+#define DC__DIG_RESYNC_FIFO_SIZE__2_PRESENT 1
+#define DC__DIG_RESYNC_FIFO_SIZE__3_PRESENT 1
+#define DC__DIG_RESYNC_FIFO_SIZE__4_PRESENT 1
+#define DC__DIG_RESYNC_FIFO_SIZE__5_PRESENT 1
+#define DC__DIG_RESYNC_FIFO_SIZE__6_PRESENT 1
+#define DC__DIG_RESYNC_FIFO_SIZE__7_PRESENT 1
+#define DC__DIG_RESYNC_FIFO_SIZE__8_PRESENT 1
+#define DC__DIG_RESYNC_FIFO_SIZE__9_PRESENT 1
+#define DC__DIG_RESYNC_FIFO_SIZE__10_PRESENT 1
+#define DC__DIG_RESYNC_FIFO_SIZE__11_PRESENT 1
+#define DC__DIG_RESYNC_FIFO_SIZE__12_PRESENT 1
+#define DC__DIG_RESYNC_FIFO_SIZE__13_PRESENT 1
+#define DC__DIG_RESYNC_FIFO_SIZE__MAX 16
+#define DC__DIG_RESYNC_FIFO_SIZE__MAX__16 1
+#define DC__DAC_RESYNC_FIFO_SIZE 12
+#define DC__DAC_RESYNC_FIFO_SIZE__12 1
+#define DC__DAC_RESYNC_FIFO_SIZE__0_PRESENT 1
+#define DC__DAC_RESYNC_FIFO_SIZE__1_PRESENT 1
+#define DC__DAC_RESYNC_FIFO_SIZE__2_PRESENT 1
+#define DC__DAC_RESYNC_FIFO_SIZE__3_PRESENT 1
+#define DC__DAC_RESYNC_FIFO_SIZE__4_PRESENT 1
+#define DC__DAC_RESYNC_FIFO_SIZE__5_PRESENT 1
+#define DC__DAC_RESYNC_FIFO_SIZE__6_PRESENT 1
+#define DC__DAC_RESYNC_FIFO_SIZE__7_PRESENT 1
+#define DC__DAC_RESYNC_FIFO_SIZE__8_PRESENT 1
+#define DC__DAC_RESYNC_FIFO_SIZE__9_PRESENT 1
+#define DC__DAC_RESYNC_FIFO_SIZE__10_PRESENT 1
+#define DC__DAC_RESYNC_FIFO_SIZE__11_PRESENT 1
+#define DC__DAC_RESYNC_FIFO_SIZE__MAX 16
+#define DC__DAC_RESYNC_FIFO_SIZE__MAX__16 1
+#define DC__DVO_RESYNC_FIFO_SIZE 12
+#define DC__DVO_RESYNC_FIFO_SIZE__12 1
+#define DC__DVO_RESYNC_FIFO_SIZE__0_PRESENT 1
+#define DC__DVO_RESYNC_FIFO_SIZE__1_PRESENT 1
+#define DC__DVO_RESYNC_FIFO_SIZE__2_PRESENT 1
+#define DC__DVO_RESYNC_FIFO_SIZE__3_PRESENT 1
+#define DC__DVO_RESYNC_FIFO_SIZE__4_PRESENT 1
+#define DC__DVO_RESYNC_FIFO_SIZE__5_PRESENT 1
+#define DC__DVO_RESYNC_FIFO_SIZE__6_PRESENT 1
+#define DC__DVO_RESYNC_FIFO_SIZE__7_PRESENT 1
+#define DC__DVO_RESYNC_FIFO_SIZE__8_PRESENT 1
+#define DC__DVO_RESYNC_FIFO_SIZE__9_PRESENT 1
+#define DC__DVO_RESYNC_FIFO_SIZE__10_PRESENT 1
+#define DC__DVO_RESYNC_FIFO_SIZE__11_PRESENT 1
+#define DC__DVO_RESYNC_FIFO_SIZE__MAX 16
+#define DC__DVO_RESYNC_FIFO_SIZE__MAX__16 1
+#define DC__MEM_CDC_PRESENT 1
+#define DC__MEM_CDC_PRESENT__1 1
+#define DC__NUM_HPD 4
+#define DC__NUM_HPD__4 1
+#define DC__NUM_HPD__0_PRESENT 1
+#define DC__NUM_HPD__1_PRESENT 1
+#define DC__NUM_HPD__2_PRESENT 1
+#define DC__NUM_HPD__3_PRESENT 1
+#define DC__NUM_HPD__MAX 6
+#define DC__NUM_HPD__MAX__6 1
+#define DC__NUM_DDC_PAIRS 4
+#define DC__NUM_DDC_PAIRS__4 1
+#define DC__NUM_DDC_PAIRS__0_PRESENT 1
+#define DC__NUM_DDC_PAIRS__1_PRESENT 1
+#define DC__NUM_DDC_PAIRS__2_PRESENT 1
+#define DC__NUM_DDC_PAIRS__3_PRESENT 1
+#define DC__NUM_DDC_PAIRS__MAX 6
+#define DC__NUM_DDC_PAIRS__MAX__6 1
+#define DC__NUM_AUDIO_PLL 0
+#define DC__NUM_AUDIO_PLL__0 1
+#define DC__NUM_AUDIO_PLL__MAX 2
+#define DC__NUM_AUDIO_PLL__MAX__2 1
+#define DC__NUM_PIXEL_PLL 1
+#define DC__NUM_PIXEL_PLL__1 1
+#define DC__NUM_PIXEL_PLL__0_PRESENT 1
+#define DC__NUM_PIXEL_PLL__MAX 4
+#define DC__NUM_PIXEL_PLL__MAX__4 1
+#define DC__NUM_CASCADED_PLL 0
+#define DC__NUM_CASCADED_PLL__0 1
+#define DC__NUM_CASCADED_PLL__MAX 3
+#define DC__NUM_CASCADED_PLL__MAX__3 1
+#define DC__PIXCLK_FROM_PHYPLL 1
+#define DC__PIXCLK_FROM_PHYPLL__1 1
+#define DC__NB_STUTTER_MODE_PRESENT 0
+#define DC__NB_STUTTER_MODE_PRESENT__0 1
+#define DC__I2S0_AND_SPDIF0_PRESENT 0
+#define DC__I2S0_AND_SPDIF0_PRESENT__0 1
+#define DC__I2S1_PRESENT 0
+#define DC__I2S1_PRESENT__0 1
+#define DC__SPDIF1_PRESENT 0
+#define DC__SPDIF1_PRESENT__0 1
+#define DC__DSI_PRESENT 0
+#define DC__DSI_PRESENT__0 1
+#define DC__DACA_PRESENT 0
+#define DC__DACA_PRESENT__0 1
+#define DC__DACB_PRESENT 0
+#define DC__DACB_PRESENT__0 1
+#define DC__NUM_PIPES 4
+#define DC__NUM_PIPES__4 1
+#define DC__NUM_PIPES__0_PRESENT 1
+#define DC__NUM_PIPES__1_PRESENT 1
+#define DC__NUM_PIPES__2_PRESENT 1
+#define DC__NUM_PIPES__3_PRESENT 1
+#define DC__NUM_PIPES__MAX 6
+#define DC__NUM_PIPES__MAX__6 1
+#define DC__NUM_DIG_LP 0
+#define DC__NUM_DIG_LP__0 1
+#define DC__NUM_DIG_LP__MAX 2
+#define DC__NUM_DIG_LP__MAX__2 1
+#define DC__DPDEBUG_PRESENT 0
+#define DC__DPDEBUG_PRESENT__0 1
+#define DC__DISPLAY_WB_PRESENT 1
+#define DC__DISPLAY_WB_PRESENT__1 1
+#define DC__NUM_CWB 0
+#define DC__NUM_CWB__0 1
+#define DC__NUM_CWB__MAX 2
+#define DC__NUM_CWB__MAX__2 1
+#define DC__MVP_PRESENT 0
+#define DC__MVP_PRESENT__0 1
+#define DC__DVO_PRESENT 0
+#define DC__DVO_PRESENT__0 1
+#define DC__ABM_PRESENT 0
+#define DC__ABM_PRESENT__0 1
+#define DC__BPHYC_PLL_PRESENT 0
+#define DC__BPHYC_PLL_PRESENT__0 1
+#define DC__BPHYC_UNIPHY_PRESENT 0
+#define DC__BPHYC_UNIPHY_PRESENT__0 1
+#define DC__PHY_BROADCAST_PRESENT 0
+#define DC__PHY_BROADCAST_PRESENT__0 1
+#define DC__NUM_OF_DCRX_SD 0
+#define DC__NUM_OF_DCRX_SD__0 1
+#define DC__DVO_17BIT_MAPPING 0
+#define DC__DVO_17BIT_MAPPING__0 1
+#define DC__AVSYNC_PRESENT 0
+#define DC__AVSYNC_PRESENT__0 1
+#define DC__NUM_OF_DCRX_PORTS 0
+#define DC__NUM_OF_DCRX_PORTS__0 1
+#define DC__NUM_OF_DCRX_PORTS__MAX 1
+#define DC__NUM_OF_DCRX_PORTS__MAX__1 1
+#define DC__NUM_PHY 4
+#define DC__NUM_PHY__4 1
+#define DC__NUM_PHY__0_PRESENT 1
+#define DC__NUM_PHY__1_PRESENT 1
+#define DC__NUM_PHY__2_PRESENT 1
+#define DC__NUM_PHY__3_PRESENT 1
+#define DC__NUM_PHY__MAX 7
+#define DC__NUM_PHY__MAX__7 1
+#define DC__NUM_PHY_LP 0
+#define DC__NUM_PHY_LP__0 1
+#define DC__NUM_PHY_LP__MAX 2
+#define DC__NUM_PHY_LP__MAX__2 1
+#define DC__SYNC_CELL vid_sync_gf14lpp
+#define DC__SYNC_CELL__VID_SYNC_GF14LPP 1
+#define DC__USE_NEW_VSS 1
+#define DC__USE_NEW_VSS__1 1
+#define DC__SYNC_CELL_DISPCLK_NUM_LATCHES 6
+#define DC__SYNC_CELL_DISPCLK_NUM_LATCHES__6 1
+#define DC__SYNC_CELL_DVOCLK_NUM_LATCHES 6
+#define DC__SYNC_CELL_DVOCLK_NUM_LATCHES__6 1
+#define DC__SYNC_CELL_PIXCLK_NUM_LATCHES 6
+#define DC__SYNC_CELL_PIXCLK_NUM_LATCHES__6 1
+#define DC__SYNC_CELL_SYMCLK_NUM_LATCHES 6
+#define DC__SYNC_CELL_SYMCLK_NUM_LATCHES__6 1
+#define DC__SYNC_CELL_DPPCLK_NUM_LATCHES 6
+#define DC__SYNC_CELL_DPPCLK_NUM_LATCHES__6 1
+#define DC__SYNC_CELL_DPREFCLK_NUM_LATCHES 6
+#define DC__SYNC_CELL_DPREFCLK_NUM_LATCHES__6 1
+#define DC__SYNC_CELL_REFCLK_NUM_LATCHES 6
+#define DC__SYNC_CELL_REFCLK_NUM_LATCHES__6 1
+#define DC__SYNC_CELL_PCIE_REFCLK_NUM_LATCHES 6
+#define DC__SYNC_CELL_PCIE_REFCLK_NUM_LATCHES__6 1
+#define DC__SYNC_CELL_MVPCLK_NUM_LATCHES 6
+#define DC__SYNC_CELL_MVPCLK_NUM_LATCHES__6 1
+#define DC__SYNC_CELL_SCLK_NUM_LATCHES 6
+#define DC__SYNC_CELL_SCLK_NUM_LATCHES__6 1
+#define DC__SYNC_CELL_DCEFCLK_NUM_LATCHES 6
+#define DC__SYNC_CELL_DCEFCLK_NUM_LATCHES__6 1
+#define DC__SYNC_CELL_AMCLK_NUM_LATCHES 6
+#define DC__SYNC_CELL_AMCLK_NUM_LATCHES__6 1
+#define DC__SYNC_CELL_DSICLK_NUM_LATCHES 6
+#define DC__SYNC_CELL_DSICLK_NUM_LATCHES__6 1
+#define DC__SYNC_CELL_BYTECLK_NUM_LATCHES 6
+#define DC__SYNC_CELL_BYTECLK_NUM_LATCHES__6 1
+#define DC__SYNC_CELL_ESCCLK_NUM_LATCHES 6
+#define DC__SYNC_CELL_ESCCLK_NUM_LATCHES__6 1
+#define DC__SYNC_CELL_DB_CLK_NUM_LATCHES 6
+#define DC__SYNC_CELL_DB_CLK_NUM_LATCHES__6 1
+#define UNIPHYA_PRESENT 1
+#define UNIPHYA_PRESENT__1 1
+#define DC__UNIPHYA_PRESENT 1
+#define DC__UNIPHYA_PRESENT__1 1
+#define UNIPHYB_PRESENT 1
+#define UNIPHYB_PRESENT__1 1
+#define DC__UNIPHYB_PRESENT 1
+#define DC__UNIPHYB_PRESENT__1 1
+#define UNIPHYC_PRESENT 1
+#define UNIPHYC_PRESENT__1 1
+#define DC__UNIPHYC_PRESENT 1
+#define DC__UNIPHYC_PRESENT__1 1
+#define UNIPHYD_PRESENT 1
+#define UNIPHYD_PRESENT__1 1
+#define DC__UNIPHYD_PRESENT 1
+#define DC__UNIPHYD_PRESENT__1 1
+#define UNIPHYE_PRESENT 0
+#define UNIPHYE_PRESENT__0 1
+#define DC__UNIPHYE_PRESENT 0
+#define DC__UNIPHYE_PRESENT__0 1
+#define UNIPHYF_PRESENT 0
+#define UNIPHYF_PRESENT__0 1
+#define DC__UNIPHYF_PRESENT 0
+#define DC__UNIPHYF_PRESENT__0 1
+#define UNIPHYG_PRESENT 0
+#define UNIPHYG_PRESENT__0 1
+#define DC__UNIPHYG_PRESENT 0
+#define DC__UNIPHYG_PRESENT__0 1
+#define DC__TMDS_LINK tmds_link_dual
+#define DC__TMDS_LINK__TMDS_LINK_DUAL 1
+#define DC__WBSCL_PIXBW 8
+#define DC__WBSCL_PIXBW__8 1
+#define DC__DWB_CSC_PRESENT 0
+#define DC__DWB_CSC_PRESENT__0 1
+#define DC__DWB_LUMA_SCL_PRESENT 0
+#define DC__DWB_LUMA_SCL_PRESENT__0 1
+#define DC__DENTIST_INTERFACE_PRESENT 1
+#define DC__DENTIST_INTERFACE_PRESENT__1 1
+#define DC__GENERICA_PRESENT 1
+#define DC__GENERICA_PRESENT__1 1
+#define DC__GENERICB_PRESENT 1
+#define DC__GENERICB_PRESENT__1 1
+#define DC__GENERICC_PRESENT 0
+#define DC__GENERICC_PRESENT__0 1
+#define DC__GENERICD_PRESENT 0
+#define DC__GENERICD_PRESENT__0 1
+#define DC__GENERICE_PRESENT 0
+#define DC__GENERICE_PRESENT__0 1
+#define DC__GENERICF_PRESENT 0
+#define DC__GENERICF_PRESENT__0 1
+#define DC__GENERICG_PRESENT 0
+#define DC__GENERICG_PRESENT__0 1
+#define DC__UNIPHY_VOLTAGE_MODE 1
+#define DC__UNIPHY_VOLTAGE_MODE__1 1
+#define DC__BLON_TYPE dedicated
+#define DC__BLON_TYPE__DEDICATED 1
+#define DC__UNIPHY_STAGGER_CH_PRESENT 1
+#define DC__UNIPHY_STAGGER_CH_PRESENT__1 1
+#define DC__XDMA_PRESENT 0
+#define DC__XDMA_PRESENT__0 1
+#define XDMA__PRESENT 0
+#define XDMA__PRESENT__0 1
+#define DC__DP_MEM_PG 0
+#define DC__DP_MEM_PG__0 1
+#define DP__MEM_PG 0
+#define DP__MEM_PG__0 1
+#define DC__AFMT_MEM_PG 0
+#define DC__AFMT_MEM_PG__0 1
+#define AFMT__MEM_PG 0
+#define AFMT__MEM_PG__0 1
+#define DC__HDMI_MEM_PG 0
+#define DC__HDMI_MEM_PG__0 1
+#define HDMI__MEM_PG 0
+#define HDMI__MEM_PG__0 1
+#define DC__I2C_MEM_PG 0
+#define DC__I2C_MEM_PG__0 1
+#define I2C__MEM_PG 0
+#define I2C__MEM_PG__0 1
+#define DC__DSCL_MEM_PG 0
+#define DC__DSCL_MEM_PG__0 1
+#define DSCL__MEM_PG 0
+#define DSCL__MEM_PG__0 1
+#define DC__CM_MEM_PG 0
+#define DC__CM_MEM_PG__0 1
+#define CM__MEM_PG 0
+#define CM__MEM_PG__0 1
+#define DC__OBUF_MEM_PG 0
+#define DC__OBUF_MEM_PG__0 1
+#define OBUF__MEM_PG 0
+#define OBUF__MEM_PG__0 1
+#define DC__WBIF_MEM_PG 1
+#define DC__WBIF_MEM_PG__1 1
+#define WBIF__MEM_PG 1
+#define WBIF__MEM_PG__1 1
+#define DC__VGA_MEM_PG 0
+#define DC__VGA_MEM_PG__0 1
+#define VGA__MEM_PG 0
+#define VGA__MEM_PG__0 1
+#define DC__FMT_MEM_PG 0
+#define DC__FMT_MEM_PG__0 1
+#define FMT__MEM_PG 0
+#define FMT__MEM_PG__0 1
+#define DC__ODM_MEM_PG 0
+#define DC__ODM_MEM_PG__0 1
+#define ODM__MEM_PG 0
+#define ODM__MEM_PG__0 1
+#define DC__DSI_MEM_PG 0
+#define DC__DSI_MEM_PG__0 1
+#define DSI__MEM_PG 0
+#define DSI__MEM_PG__0 1
+#define DC__AZ_MEM_PG 1
+#define DC__AZ_MEM_PG__1 1
+#define AZ__MEM_PG 1
+#define AZ__MEM_PG__1 1
+#define DC__WBSCL_MEM1P1024X64QS_MEM_PG 1
+#define DC__WBSCL_MEM1P1024X64QS_MEM_PG__1 1
+#define WBSCL_MEM1P1024X64QS__MEM_PG 1
+#define WBSCL_MEM1P1024X64QS__MEM_PG__1 1
+#define DC__WBSCL_MEM1P528X64QS_MEM_PG 1
+#define DC__WBSCL_MEM1P528X64QS_MEM_PG__1 1
+#define WBSCL_MEM1P528X64QS__MEM_PG 1
+#define WBSCL_MEM1P528X64QS__MEM_PG__1 1
+#define DC__DMCU_MEM1P1024X32BQS_MEM_PG 1
+#define DC__DMCU_MEM1P1024X32BQS_MEM_PG__1 1
+#define DMCU_MEM1P1024X32BQS__MEM_PG 1
+#define DMCU_MEM1P1024X32BQS__MEM_PG__1 1
+#define DC__HUBBUB_SDP_TAG_INT_MEM_PG 0
+#define DC__HUBBUB_SDP_TAG_INT_MEM_PG__0 1
+#define HUBBUB_SDP_TAG_INT__MEM_PG 0
+#define HUBBUB_SDP_TAG_INT__MEM_PG__0 1
+#define DC__HUBBUB_SDP_TAG_EXT_MEM_PG 0
+#define DC__HUBBUB_SDP_TAG_EXT_MEM_PG__0 1
+#define HUBBUB_SDP_TAG_EXT__MEM_PG 0
+#define HUBBUB_SDP_TAG_EXT__MEM_PG__0 1
+#define DC__HUBBUB_RET_ZERO_MEM_PG 0
+#define DC__HUBBUB_RET_ZERO_MEM_PG__0 1
+#define HUBBUB_RET_ZERO__MEM_PG 0
+#define HUBBUB_RET_ZERO__MEM_PG__0 1
+#define DC__HUBBUB_RET_ROB_MEM_PG 0
+#define DC__HUBBUB_RET_ROB_MEM_PG__0 1
+#define HUBBUB_RET_ROB__MEM_PG 0
+#define HUBBUB_RET_ROB__MEM_PG__0 1
+#define DC__HUBPRET_CUR_ROB_MEM_PG 0
+#define DC__HUBPRET_CUR_ROB_MEM_PG__0 1
+#define HUBPRET_CUR_ROB__MEM_PG 0
+#define HUBPRET_CUR_ROB__MEM_PG__0 1
+#define DC__HUBPRET_CUR_CDC_MEM_PG 0
+#define DC__HUBPRET_CUR_CDC_MEM_PG__0 1
+#define HUBPRET_CUR_CDC__MEM_PG 0
+#define HUBPRET_CUR_CDC__MEM_PG__0 1
+#define DC__HUBPREQ_MPTE_MEM_PG 0
+#define DC__HUBPREQ_MPTE_MEM_PG__0 1
+#define HUBPREQ_MPTE__MEM_PG 0
+#define HUBPREQ_MPTE__MEM_PG__0 1
+#define DC__HUBPREQ_META_MEM_PG 0
+#define DC__HUBPREQ_META_MEM_PG__0 1
+#define HUBPREQ_META__MEM_PG 0
+#define HUBPREQ_META__MEM_PG__0 1
+#define DC__HUBPREQ_DPTE_MEM_PG 0
+#define DC__HUBPREQ_DPTE_MEM_PG__0 1
+#define HUBPREQ_DPTE__MEM_PG 0
+#define HUBPREQ_DPTE__MEM_PG__0 1
+#define DC__HUBPRET_DET_MEM_PG 0
+#define DC__HUBPRET_DET_MEM_PG__0 1
+#define HUBPRET_DET__MEM_PG 0
+#define HUBPRET_DET__MEM_PG__0 1
+#define DC__HUBPRET_PIX_CDC_MEM_PG 0
+#define DC__HUBPRET_PIX_CDC_MEM_PG__0 1
+#define HUBPRET_PIX_CDC__MEM_PG 0
+#define HUBPRET_PIX_CDC__MEM_PG__0 1
+#define DC__TOP_BLKS__DCCG 1
+#define DC__TOP_BLKS__DCHUBBUB 1
+#define DC__TOP_BLKS__DCHUBP 1
+#define DC__TOP_BLKS__HDA 1
+#define DC__TOP_BLKS__DIO 1
+#define DC__TOP_BLKS__DCIO 1
+#define DC__TOP_BLKS__DMU 1
+#define DC__TOP_BLKS__DPP 1
+#define DC__TOP_BLKS__MPC 1
+#define DC__TOP_BLKS__OPP 1
+#define DC__TOP_BLKS__OPTC 1
+#define DC__TOP_BLKS__MMHUBBUB 1
+#define DC__TOP_BLKS__WB 1
+#define DC__TOP_BLKS__MAX 13
+#define DC__TOP_BLKS__MAX__13 1
+#define DC__DCHUBP_DPP_SF_PIXEL_CREDITS 9
+#define DC__DCHUBP_DPP_SF_PIXEL_CREDITS__9 1
+#define DC__DPP_MPC_SF_PIXEL_CREDITS 9
+#define DC__DPP_MPC_SF_PIXEL_CREDITS__9 1
+#define DC__MPC_OPP_SF_PIXEL_CREDITS 8
+#define DC__MPC_OPP_SF_PIXEL_CREDITS__8 1
+#define DC__OPP_OPTC_SF_PIXEL_CREDITS 8
+#define DC__OPP_OPTC_SF_PIXEL_CREDITS__8 1
+#define DC__SFR_SFT_ROUND_TRIP_DELAY 5
+#define DC__SFR_SFT_ROUND_TRIP_DELAY__5 1
+#define DC__REPEATER_PROJECT_MAX 8
+#define DC__REPEATER_PROJECT_MAX__8 1
+#define DC__SURFACE_422_CAPABLE 0
+#define DC__SURFACE_422_CAPABLE__0 1
+#endif

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dcn10/dcn10_fpu.c

@@ -0,0 +1,159 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright 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 "dcn10/dcn10_resource.h"
+
+#include "dcn10_fpu.h"
+#include "resource.h"
+#include "amdgpu_dm/dc_fpu.h"
+
+/**
+ * DOC: DCN10 FPU manipulation Overview
+ *
+ * The DCN architecture relies on FPU operations, which require special
+ * compilation flags and the use of kernel_fpu_begin/end functions; ideally, we
+ * want to avoid spreading FPU access across multiple files. With this idea in
+ * mind, this file aims to centralize DCN10 functions that require FPU access
+ * in a single place. Code in this file follows the following code pattern:
+ *
+ * 1. Functions that use FPU operations should be isolated in static functions.
+ * 2. The FPU functions should have the noinline attribute to ensure anything
+ *    that deals with FP register is contained within this call.
+ * 3. All function that needs to be accessed outside this file requires a
+ *    public interface that not uses any FPU reference.
+ * 4. Developers **must not** use DC_FP_START/END in this file, but they need
+ *    to ensure that the caller invokes it before access any function available
+ *    in this file. For this reason, public functions in this file must invoke
+ *    dc_assert_fp_enabled();
+ *
+ * Let's expand a little bit more the idea in the code pattern. To fully
+ * isolate FPU operations in a single place, we must avoid situations where
+ * compilers spill FP values to registers due to FP enable in a specific C
+ * file. Note that even if we isolate all FPU functions in a single file and
+ * call its interface from other files, the compiler might enable the use of
+ * FPU before we call DC_FP_START. Nevertheless, it is the programmer's
+ * responsibility to invoke DC_FP_START/END in the correct place. To highlight
+ * situations where developers forgot to use the FP protection before calling
+ * the DC FPU interface functions, we introduce a helper that checks if the
+ * function is invoked under FP protection. If not, it will trigger a kernel
+ * warning.
+ */
+
+struct _vcs_dpi_ip_params_st dcn1_0_ip = {
+	.rob_buffer_size_kbytes = 64,
+	.det_buffer_size_kbytes = 164,
+	.dpte_buffer_size_in_pte_reqs_luma = 42,
+	.dpp_output_buffer_pixels = 2560,
+	.opp_output_buffer_lines = 1,
+	.pixel_chunk_size_kbytes = 8,
+	.pte_enable = 1,
+	.pte_chunk_size_kbytes = 2,
+	.meta_chunk_size_kbytes = 2,
+	.writeback_chunk_size_kbytes = 2,
+	.line_buffer_size_bits = 589824,
+	.max_line_buffer_lines = 12,
+	.IsLineBufferBppFixed = 0,
+	.LineBufferFixedBpp = (unsigned int)-1,
+	.writeback_luma_buffer_size_kbytes = 12,
+	.writeback_chroma_buffer_size_kbytes = 8,
+	.max_num_dpp = 4,
+	.max_num_wb = 2,
+	.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 = 4,
+	.max_vscl_ratio = 4,
+	.hscl_mults = 4,
+	.vscl_mults = 4,
+	.max_hscl_taps = 8,
+	.max_vscl_taps = 8,
+	.dispclk_ramp_margin_percent = 1,
+	.underscan_factor = 1.10,
+	.min_vblank_lines = 14,
+	.dppclk_delay_subtotal = 90,
+	.dispclk_delay_subtotal = 42,
+	.dcfclk_cstate_latency = 10,
+	.max_inter_dcn_tile_repeaters = 8,
+	.can_vstartup_lines_exceed_vsync_plus_back_porch_lines_minus_one = 0,
+	.bug_forcing_LC_req_same_size_fixed = 0,
+};
+
+struct _vcs_dpi_soc_bounding_box_st dcn1_0_soc = {
+	.sr_exit_time_us = 9.0,
+	.sr_enter_plus_exit_time_us = 11.0,
+	.urgent_latency_us = 4.0,
+	.writeback_latency_us = 12.0,
+	.ideal_dram_bw_after_urgent_percent = 80.0,
+	.max_request_size_bytes = 256,
+	.downspread_percent = 0.5,
+	.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 = 128,
+	.urgent_out_of_order_return_per_channel_bytes = 256,
+	.channel_interleave_bytes = 256,
+	.num_banks = 8,
+	.num_chans = 2,
+	.vmm_page_size_bytes = 4096,
+	.dram_clock_change_latency_us = 17.0,
+	.writeback_dram_clock_change_latency_us = 23.0,
+	.return_bus_width_bytes = 64,
+};
+
+void dcn10_resource_construct_fp(struct dc *dc)
+{
+	dc_assert_fp_enabled();
+	if (dc->ctx->dce_version == DCN_VERSION_1_01) {
+		struct dcn_soc_bounding_box *dcn_soc = dc->dcn_soc;
+		struct dcn_ip_params *dcn_ip = dc->dcn_ip;
+		struct display_mode_lib *dml = &dc->dml;
+
+		dml->ip.max_num_dpp = 3;
+		/* TODO how to handle 23.84? */
+		dcn_soc->dram_clock_change_latency = 23;
+		dcn_ip->max_num_dpp = 3;
+	}
+	if (ASICREV_IS_RV1_F0(dc->ctx->asic_id.hw_internal_rev)) {
+		dc->dcn_soc->urgent_latency = 3;
+		dc->debug.disable_dmcu = true;
+		dc->dcn_soc->fabric_and_dram_bandwidth_vmax0p9 = 41.60f;
+	}
+
+	dc->dcn_soc->number_of_channels = dc->ctx->asic_id.vram_width / ddr4_dram_width;
+	ASSERT(dc->dcn_soc->number_of_channels < 3);
+	if (dc->dcn_soc->number_of_channels == 0)/*old sbios bug*/
+		dc->dcn_soc->number_of_channels = 2;
+
+	if (dc->dcn_soc->number_of_channels == 1) {
+		dc->dcn_soc->fabric_and_dram_bandwidth_vmax0p9 = 19.2f;
+		dc->dcn_soc->fabric_and_dram_bandwidth_vnom0p8 = 17.066f;
+		dc->dcn_soc->fabric_and_dram_bandwidth_vmid0p72 = 14.933f;
+		dc->dcn_soc->fabric_and_dram_bandwidth_vmin0p65 = 12.8f;
+		if (ASICREV_IS_RV1_F0(dc->ctx->asic_id.hw_internal_rev))
+			dc->dcn_soc->fabric_and_dram_bandwidth_vmax0p9 = 20.80f;
+	}
+}

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dcn10/dcn10_fpu.h

@@ -0,0 +1,32 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright 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 __DCN10_FPU_H__
+#define __DCN10_FPU_H__
+
+void dcn10_resource_construct_fp(struct dc *dc);
+
+#endif /* __DCN10_FPU_H__ */

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.h

@@ -0,0 +1,89 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright 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 "core_types.h"
+
+#ifndef __DCN20_FPU_H__
+#define __DCN20_FPU_H__
+
+void dcn20_populate_dml_writeback_from_context(struct dc *dc,
+					       struct resource_context *res_ctx,
+					       display_e2e_pipe_params_st *pipes);
+
+void dcn20_fpu_set_wb_arb_params(struct mcif_arb_params *wb_arb_params,
+				 struct dc_state *context,
+				 display_e2e_pipe_params_st *pipes,
+				 int pipe_cnt, int i);
+void dcn20_calculate_dlg_params(struct dc *dc,
+				struct dc_state *context,
+				display_e2e_pipe_params_st *pipes,
+				int pipe_cnt,
+				int vlevel);
+int dcn20_populate_dml_pipes_from_context(struct dc *dc,
+					  struct dc_state *context,
+					  display_e2e_pipe_params_st *pipes,
+					  enum dc_validate_mode validate_mode);
+void dcn20_calculate_wm(struct dc *dc,
+			struct dc_state *context,
+			display_e2e_pipe_params_st *pipes,
+			int *out_pipe_cnt,
+			int *pipe_split_from,
+			int vlevel,
+			enum dc_validate_mode validate_mode);
+void dcn20_cap_soc_clocks(struct _vcs_dpi_soc_bounding_box_st *bb,
+			  struct pp_smu_nv_clock_table max_clocks);
+void dcn20_update_bounding_box(struct dc *dc,
+			       struct _vcs_dpi_soc_bounding_box_st *bb,
+			       struct pp_smu_nv_clock_table *max_clocks,
+			       unsigned int *uclk_states,
+			       unsigned int num_states);
+void dcn20_patch_bounding_box(struct dc *dc,
+			      struct _vcs_dpi_soc_bounding_box_st *bb);
+bool dcn20_validate_bandwidth_fp(struct dc *dc, struct dc_state *context,
+				 enum dc_validate_mode validate_mode, display_e2e_pipe_params_st *pipes);
+void dcn20_fpu_set_wm_ranges(int i,
+			     struct pp_smu_wm_range_sets *ranges,
+			     struct _vcs_dpi_soc_bounding_box_st *loaded_bb);
+void dcn20_fpu_adjust_dppclk(struct vba_vars_st *v,
+			     int vlevel,
+			     int max_mpc_comb,
+			     int pipe_idx,
+			     bool is_validating_bw);
+
+int dcn21_populate_dml_pipes_from_context(struct dc *dc,
+					  struct dc_state *context,
+					  display_e2e_pipe_params_st *pipes,
+					  enum dc_validate_mode validate_mode);
+bool dcn21_validate_bandwidth_fp(struct dc *dc, struct dc_state *context, enum
+				 dc_validate_mode, display_e2e_pipe_params_st *pipes);
+void dcn21_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_params);
+
+void dcn21_clk_mgr_set_bw_params_wm_table(struct clk_bw_params *bw_params);
+
+void dcn201_populate_dml_writeback_from_context_fpu(struct dc *dc,
+						struct resource_context *res_ctx,
+						display_e2e_pipe_params_st *pipes);
+
+#endif /* __DCN20_FPU_H__ */

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_mode_vba_20.h

@@ -0,0 +1,32 @@
+/*
+ * Copyright 2018 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 _DCN20_DISPLAY_MODE_VBA_H_
+#define _DCN20_DISPLAY_MODE_VBA_H_
+
+void dml20_recalculate(struct display_mode_lib *mode_lib);
+void dml20_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_lib);
+
+#endif

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_mode_vba_20v2.h

@@ -0,0 +1,32 @@
+/*
+ * Copyright 2018 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 _DCN20V2_DISPLAY_MODE_VBA_H_
+#define _DCN20V2_DISPLAY_MODE_VBA_H_
+
+void dml20v2_recalculate(struct display_mode_lib *mode_lib);
+void dml20v2_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_lib);
+
+#endif

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20.h

@@ -0,0 +1,73 @@
+/*
+ * Copyright 2018 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 __DML20_DISPLAY_RQ_DLG_CALC_H__
+#define __DML20_DISPLAY_RQ_DLG_CALC_H__
+
+#include "../display_rq_dlg_helpers.h"
+
+struct display_mode_lib;
+
+
+// Function: dml_rq_dlg_get_rq_reg
+//  Main entry point for test to get the register values out of this DML class.
+//  This function calls <get_rq_param> and <extract_rq_regs> functions to calculate
+//  and then populate the rq_regs struct
+// Input:
+//  pipe_src_param - pipe source configuration (e.g. vp, pitch, etc.)
+// Output:
+//  rq_regs - struct that holds all the RQ registers field value.
+//            See also: <display_rq_regs_st>
+void dml20_rq_dlg_get_rq_reg(
+		struct display_mode_lib *mode_lib,
+		display_rq_regs_st *rq_regs,
+		const display_pipe_params_st *pipe_param);
+
+
+// Function: dml_rq_dlg_get_dlg_reg
+//   Calculate and return DLG and TTU register struct given the system setting
+// Output:
+//  dlg_regs - output DLG register struct
+//  ttu_regs - output DLG TTU register struct
+// Input:
+//  e2e_pipe_param - "compacted" array of e2e pipe param struct
+//  num_pipes - num of active "pipe" or "route"
+//  pipe_idx - index that identifies the e2e_pipe_param that corresponding to this dlg
+//  cstate - 0: when calculate min_ttu_vblank it is assumed cstate is not required. 1: Normal mode, cstate is considered.
+//           Added for legacy or unrealistic timing tests.
+void dml20_rq_dlg_get_dlg_reg(
+		struct display_mode_lib *mode_lib,
+		display_dlg_regs_st *dlg_regs,
+		display_ttu_regs_st *ttu_regs,
+		const display_e2e_pipe_params_st *e2e_pipe_param,
+		const unsigned int num_pipes,
+		const unsigned int pipe_idx,
+		const bool cstate_en,
+		const bool pstate_en,
+		const bool vm_en,
+		const bool ignore_viewport_pos,
+		const bool immediate_flip_support);
+
+#endif

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20v2.h

@@ -0,0 +1,73 @@
+/*
+ * Copyright 2018 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 __DML20V2_DISPLAY_RQ_DLG_CALC_H__
+#define __DML20V2_DISPLAY_RQ_DLG_CALC_H__
+
+#include "../display_rq_dlg_helpers.h"
+
+struct display_mode_lib;
+
+
+// Function: dml_rq_dlg_get_rq_reg
+//  Main entry point for test to get the register values out of this DML class.
+//  This function calls <get_rq_param> and <extract_rq_regs> functions to calculate
+//  and then populate the rq_regs struct
+// Input:
+//  pipe_src_param - pipe source configuration (e.g. vp, pitch, etc.)
+// Output:
+//  rq_regs - struct that holds all the RQ registers field value.
+//            See also: <display_rq_regs_st>
+void dml20v2_rq_dlg_get_rq_reg(
+		struct display_mode_lib *mode_lib,
+		display_rq_regs_st *rq_regs,
+		const display_pipe_params_st *pipe_param);
+
+
+// Function: dml_rq_dlg_get_dlg_reg
+//   Calculate and return DLG and TTU register struct given the system setting
+// Output:
+//  dlg_regs - output DLG register struct
+//  ttu_regs - output DLG TTU register struct
+// Input:
+//  e2e_pipe_param - "compacted" array of e2e pipe param struct
+//  num_pipes - num of active "pipe" or "route"
+//  pipe_idx - index that identifies the e2e_pipe_param that corresponding to this dlg
+//  cstate - 0: when calculate min_ttu_vblank it is assumed cstate is not required. 1: Normal mode, cstate is considered.
+//           Added for legacy or unrealistic timing tests.
+void dml20v2_rq_dlg_get_dlg_reg(
+		struct display_mode_lib *mode_lib,
+		display_dlg_regs_st *dlg_regs,
+		display_ttu_regs_st *ttu_regs,
+		const display_e2e_pipe_params_st *e2e_pipe_param,
+		const unsigned int num_pipes,
+		const unsigned int pipe_idx,
+		const bool cstate_en,
+		const bool pstate_en,
+		const bool vm_en,
+		const bool ignore_viewport_pos,
+		const bool immediate_flip_support);
+
+#endif

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_mode_vba_21.h

@@ -0,0 +1,32 @@
+/*
+ * Copyright 2017 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 __DML21_DISPLAY_MODE_VBA_H__
+#define __DML21_DISPLAY_MODE_VBA_H__
+
+void dml21_recalculate(struct display_mode_lib *mode_lib);
+void dml21_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_lib);
+
+#endif /* _DML21_DISPLAY_MODE_VBA_H_ */

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_rq_dlg_calc_21.h

@@ -0,0 +1,73 @@
+/*
+ * Copyright 2017 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 __DML21_DISPLAY_RQ_DLG_CALC_H__
+#define __DML21_DISPLAY_RQ_DLG_CALC_H__
+
+#include "dm_services.h"
+#include "../display_rq_dlg_helpers.h"
+
+struct display_mode_lib;
+
+
+// Function: dml_rq_dlg_get_rq_reg
+//  Main entry point for test to get the register values out of this DML class.
+//  This function calls <get_rq_param> and <extract_rq_regs> functions to calculate
+//  and then populate the rq_regs struct
+// Input:
+//  pipe_src_param - pipe source configuration (e.g. vp, pitch, etc.)
+// Output:
+//  rq_regs - struct that holds all the RQ registers field value.
+//            See also: <display_rq_regs_st>
+void dml21_rq_dlg_get_rq_reg(
+		struct display_mode_lib *mode_lib,
+		display_rq_regs_st *rq_regs,
+		const display_pipe_params_st *pipe_param);
+
+// Function: dml_rq_dlg_get_dlg_reg
+//   Calculate and return DLG and TTU register struct given the system setting
+// Output:
+//  dlg_regs - output DLG register struct
+//  ttu_regs - output DLG TTU register struct
+// Input:
+//  e2e_pipe_param - "compacted" array of e2e pipe param struct
+//  num_pipes - num of active "pipe" or "route"
+//  pipe_idx - index that identifies the e2e_pipe_param that corresponding to this dlg
+//  cstate - 0: when calculate min_ttu_vblank it is assumed cstate is not required. 1: Normal mode, cstate is considered.
+//           Added for legacy or unrealistic timing tests.
+void dml21_rq_dlg_get_dlg_reg(
+		struct display_mode_lib *mode_lib,
+		display_dlg_regs_st *dlg_regs,
+		display_ttu_regs_st *ttu_regs,
+		const display_e2e_pipe_params_st *e2e_pipe_param,
+		const unsigned int num_pipes,
+		const unsigned int pipe_idx,
+		const bool cstate_en,
+		const bool pstate_en,
+		const bool vm_en,
+		const bool ignore_viewport_pos,
+		const bool immediate_flip_support);
+
+#endif

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dcn30/dcn30_fpu.c

@@ -0,0 +1,870 @@
+/*
+ * 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 "clk_mgr/dcn30/dcn30_smu11_driver_if.h"
+#include "display_mode_vba_30.h"
+#include "dcn30_fpu.h"
+
+#include "../dml1_frl_cap_chk.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,
+				.dtbclk_mhz = 1217.0,
+			},
+		},
+
+	.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 dcn30_fpu_populate_dml_writeback_from_context(
+		struct dc *dc, struct resource_context *res_ctx, display_e2e_pipe_params_st *pipes)
+{
+	int pipe_cnt;
+	unsigned int i, j;
+	double max_calc_writeback_dispclk;
+	double writeback_dispclk;
+	struct writeback_st dout_wb = {0};
+
+	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 < ARRAY_SIZE(wb_arb_params->cli_watermark); i++) {
+		wb_arb_params->cli_watermark[i] = (unsigned int)(get_wm_writeback_urgent(dml, pipes, pipe_cnt) * 1000);
+		wb_arb_params->pstate_watermark[i] = (unsigned int)(get_wm_writeback_dram_clock_change(dml, pipes, pipe_cnt) * 1000);
+	}
+
+	wb_arb_params->dram_speed_change_duration = (unsigned int)(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) {
+		if (!context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching ||
+				context->bw_ctx.dml.soc.dram_clock_change_latency_us == 0)
+			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;
+	unsigned int pipe_i, state_i;
+	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;
+	unsigned int dummy_latency_index = 0;
+	struct dc_stream_status *stream_status = NULL;
+
+	dc_assert_fp_enabled();
+
+	context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching = false;
+	for (i = 0; i < context->stream_count; i++) {
+		stream_status = NULL;
+		if (context->streams[i])
+			stream_status = dc_state_get_stream_status(context, context->streams[i]);
+		if (stream_status)
+			stream_status->fpo_in_use = false;
+	}
+
+	if (!pstate_en) {
+		/* only when the mclk switch can not be natural, is the fw based vblank stretch attempted */
+		context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching =
+			dcn30_can_support_mclk_switch_using_fw_based_vblank_stretch(dc, context);
+
+		if (context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching) {
+			dummy_latency_index = dcn30_find_dummy_latency_index_for_fw_based_mclk_switch(dc,
+				context, pipes, pipe_cnt, vlevel);
+
+			/* After calling dcn30_find_dummy_latency_index_for_fw_based_mclk_switch
+			 * we reinstate the original dram_clock_change_latency_us on the context
+			 * and all variables that may have changed up to this point, except the
+			 * newly found dummy_latency_index
+			 */
+			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;
+			dcn30_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel,
+				DC_VALIDATE_MODE_AND_PROGRAMMING, true);
+			maxMpcComb = context->bw_ctx.dml.vba.maxMpcComb;
+			dcfclk = context->bw_ctx.dml.vba.DCFCLKState[vlevel][context->bw_ctx.dml.vba.maxMpcComb];
+			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 = (uint32_t)(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 = (uint32_t)(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 = (uint32_t)(get_wm_stutter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000);
+	context->bw_ctx.bw.dcn.watermarks.b.cstate_pstate.pstate_change_ns = (uint32_t)(get_wm_dram_clock_change(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000);
+	context->bw_ctx.bw.dcn.watermarks.b.pte_meta_urgent_ns = (uint32_t)(get_wm_memory_trip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000);
+	context->bw_ctx.bw.dcn.watermarks.b.frac_urg_bw_nom = (uint32_t)(get_fraction_of_urgent_bandwidth(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000);
+	context->bw_ctx.bw.dcn.watermarks.b.frac_urg_bw_flip = (uint32_t)(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 = (uint32_t)(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 = (unsigned int)context->bw_ctx.dml.vba.DRAMSpeed;
+		unsigned int min_dram_speed_mts_margin = 160;
+
+		context->bw_ctx.dml.soc.dram_clock_change_latency_us =
+			dc->clk_mgr->bw_params->dummy_pstate_table[0].dummy_pstate_latency_us;
+
+		if (context->bw_ctx.dml.vba.DRAMClockChangeSupport[vlevel][maxMpcComb] ==
+			dm_dram_clock_change_unsupported) {
+			int min_dram_speed_mts_offset = dc->clk_mgr->bw_params->clk_table.num_entries - 1;
+
+			min_dram_speed_mts =
+				dc->clk_mgr->bw_params->clk_table.entries[min_dram_speed_mts_offset].memclk_mhz * 16;
+		}
+
+		if (!context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching) {
+			/* find largest table entry that is lower than dram speed,
+			 * but lower than DPM0 still uses DPM0
+			 */
+			for (dummy_latency_index = 3; dummy_latency_index > 0; dummy_latency_index--)
+				if (min_dram_speed_mts + min_dram_speed_mts_margin >
+					dc->clk_mgr->bw_params->dummy_pstate_table[dummy_latency_index].dram_speed_mts)
+					break;
+		}
+
+		context->bw_ctx.dml.soc.dram_clock_change_latency_us =
+			dc->clk_mgr->bw_params->dummy_pstate_table[dummy_latency_index].dummy_pstate_latency_us;
+
+		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 = (uint32_t)(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 = (uint32_t)(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 = (uint32_t)(get_wm_stutter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000);
+	context->bw_ctx.bw.dcn.watermarks.c.cstate_pstate.pstate_change_ns = (uint32_t)(get_wm_dram_clock_change(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000);
+	context->bw_ctx.bw.dcn.watermarks.c.pte_meta_urgent_ns = (uint32_t)(get_wm_memory_trip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000);
+	context->bw_ctx.bw.dcn.watermarks.c.frac_urg_bw_nom = (uint32_t)(get_fraction_of_urgent_bandwidth(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000);
+	context->bw_ctx.bw.dcn.watermarks.c.frac_urg_bw_flip = (uint32_t)(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 = (uint32_t)(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 = (uint32_t)(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 = (uint32_t)(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 = (uint32_t)(get_wm_stutter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000);
+		context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.pstate_change_ns = (uint32_t)(get_wm_dram_clock_change(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000);
+		context->bw_ctx.bw.dcn.watermarks.a.pte_meta_urgent_ns = (uint32_t)(get_wm_memory_trip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000);
+		context->bw_ctx.bw.dcn.watermarks.a.frac_urg_bw_nom = (uint32_t)(get_fraction_of_urgent_bandwidth(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000);
+		context->bw_ctx.bw.dcn.watermarks.a.frac_urg_bw_flip = (uint32_t)(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 = (uint32_t)(get_urgent_latency(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000);
+	}
+
+	context->perf_params.stutter_period_us = (unsigned int)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 (pipe_i = 0, pipe_idx = 0; pipe_i < dc->res_pool->pipe_count; pipe_i++) {
+		if (!context->res_ctx.pipe_ctx[pipe_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++;
+	}
+
+	// WA: restrict FPO to use first non-strobe mode (NV24 BW issue)
+	if (context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching &&
+			dc->dml.soc.num_chans <= 4 &&
+			context->bw_ctx.dml.vba.DRAMSpeed <= 1700 &&
+			context->bw_ctx.dml.vba.DRAMSpeed >= 1500) {
+
+		for (state_i = 0; state_i < dc->dml.soc.num_states; state_i++) {
+			if (dc->dml.soc.clock_limits[state_i].dram_speed_mts > 1700) {
+				context->bw_ctx.dml.vba.DRAMSpeed = dc->dml.soc.clock_limits[state_i].dram_speed_mts;
+				break;
+			}
+		}
+	}
+
+	dcn20_calculate_dlg_params(dc, context, pipes, pipe_cnt, vlevel);
+
+	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;
+
+	if (context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching)
+		dcn30_setup_mclk_switch_using_fw_based_vblank_stretch(dc, context);
+}
+
+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 = (int)dcn3_0_soc.clock_limits[0].dcfclk_mhz;
+		if (!dcn30_bb_max_clk->max_dispclk_mhz)
+			dcn30_bb_max_clk->max_dispclk_mhz = (int)dcn3_0_soc.clock_limits[0].dispclk_mhz;
+		if (!dcn30_bb_max_clk->max_dppclk_mhz)
+			dcn30_bb_max_clk->max_dppclk_mhz = (int)dcn3_0_soc.clock_limits[0].dppclk_mhz;
+		if (!dcn30_bb_max_clk->max_phyclk_mhz)
+			dcn30_bb_max_clk->max_phyclk_mhz = (int)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 = (unsigned int)(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 = (unsigned int)(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)
+{
+	(void)bw_params;
+	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);
+
+}
+
+/**
+ * dcn30_find_dummy_latency_index_for_fw_based_mclk_switch() - Finds
+ * dummy_latency_index when MCLK switching using firmware based vblank stretch
+ * is enabled. This function will iterate through the table of dummy pstate
+ * latencies until the lowest value that allows
+ * dm_allow_self_refresh_and_mclk_switch to happen is found
+ *
+ * @dc: Current DC state
+ * @context: new dc state
+ * @pipes: DML pipe params
+ * @pipe_cnt: number of DML pipes
+ * @vlevel: Voltage level calculated by DML
+ *
+ * Return: lowest dummy_latency_index value
+ */
+int dcn30_find_dummy_latency_index_for_fw_based_mclk_switch(struct dc *dc,
+							    struct dc_state *context,
+							    display_e2e_pipe_params_st *pipes,
+							    int pipe_cnt,
+							    int vlevel)
+{
+	const int max_latency_table_entries = 4;
+	int dummy_latency_index = 0;
+
+	dc_assert_fp_enabled();
+
+	while (dummy_latency_index < max_latency_table_entries) {
+		context->bw_ctx.dml.soc.dram_clock_change_latency_us =
+				dc->clk_mgr->bw_params->dummy_pstate_table[dummy_latency_index].dummy_pstate_latency_us;
+		dcn30_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel,
+			DC_VALIDATE_MODE_AND_PROGRAMMING, true);
+
+		if (context->bw_ctx.dml.soc.allow_dram_self_refresh_or_dram_clock_change_in_vblank ==
+			dm_allow_self_refresh_and_mclk_switch)
+			break;
+
+		dummy_latency_index++;
+	}
+
+	if (dummy_latency_index == max_latency_table_entries) {
+		ASSERT(dummy_latency_index != max_latency_table_entries);
+		/* If the execution gets here, it means dummy p_states are
+		 * not possible. This should never happen and would mean
+		 * something is severely wrong.
+		 * Here we reset dummy_latency_index to 3, because it is
+		 * better to have underflows than system crashes.
+		 */
+		dummy_latency_index = 3;
+	}
+
+	return dummy_latency_index;
+}
+
+void dcn3_fpu_build_wm_range_table(struct clk_mgr *base)
+{
+	/* defaults */
+	double pstate_latency_us = base->ctx->dc->dml.soc.dram_clock_change_latency_us;
+	double sr_exit_time_us = base->ctx->dc->dml.soc.sr_exit_time_us;
+	double sr_enter_plus_exit_time_us = base->ctx->dc->dml.soc.sr_enter_plus_exit_time_us;
+	uint16_t min_uclk_mhz = (uint16_t)base->bw_params->clk_table.entries[0].memclk_mhz;
+
+	dc_assert_fp_enabled();
+
+	/* Set A - Normal - default values*/
+	base->bw_params->wm_table.nv_entries[WM_A].valid = true;
+	base->bw_params->wm_table.nv_entries[WM_A].dml_input.pstate_latency_us = pstate_latency_us;
+	base->bw_params->wm_table.nv_entries[WM_A].dml_input.sr_exit_time_us = sr_exit_time_us;
+	base->bw_params->wm_table.nv_entries[WM_A].dml_input.sr_enter_plus_exit_time_us = sr_enter_plus_exit_time_us;
+	base->bw_params->wm_table.nv_entries[WM_A].pmfw_breakdown.wm_type = WATERMARKS_CLOCK_RANGE;
+	base->bw_params->wm_table.nv_entries[WM_A].pmfw_breakdown.min_dcfclk = 0;
+	base->bw_params->wm_table.nv_entries[WM_A].pmfw_breakdown.max_dcfclk = 0xFFFF;
+	base->bw_params->wm_table.nv_entries[WM_A].pmfw_breakdown.min_uclk = min_uclk_mhz;
+	base->bw_params->wm_table.nv_entries[WM_A].pmfw_breakdown.max_uclk = 0xFFFF;
+
+	/* Set B - Performance - higher minimum clocks */
+//	base->bw_params->wm_table.nv_entries[WM_B].valid = true;
+//	base->bw_params->wm_table.nv_entries[WM_B].dml_input.pstate_latency_us = pstate_latency_us;
+//	base->bw_params->wm_table.nv_entries[WM_B].dml_input.sr_exit_time_us = sr_exit_time_us;
+//	base->bw_params->wm_table.nv_entries[WM_B].dml_input.sr_enter_plus_exit_time_us = sr_enter_plus_exit_time_us;
+//	base->bw_params->wm_table.nv_entries[WM_B].pmfw_breakdown.wm_type = WATERMARKS_CLOCK_RANGE;
+//	base->bw_params->wm_table.nv_entries[WM_B].pmfw_breakdown.min_dcfclk = TUNED VALUE;
+//	base->bw_params->wm_table.nv_entries[WM_B].pmfw_breakdown.max_dcfclk = 0xFFFF;
+//	base->bw_params->wm_table.nv_entries[WM_B].pmfw_breakdown.min_uclk = TUNED VALUE;
+//	base->bw_params->wm_table.nv_entries[WM_B].pmfw_breakdown.max_uclk = 0xFFFF;
+
+	/* Set C - Dummy P-State - P-State latency set to "dummy p-state" value */
+	base->bw_params->wm_table.nv_entries[WM_C].valid = true;
+	base->bw_params->wm_table.nv_entries[WM_C].dml_input.pstate_latency_us = 0;
+	base->bw_params->wm_table.nv_entries[WM_C].dml_input.sr_exit_time_us = sr_exit_time_us;
+	base->bw_params->wm_table.nv_entries[WM_C].dml_input.sr_enter_plus_exit_time_us = sr_enter_plus_exit_time_us;
+	base->bw_params->wm_table.nv_entries[WM_C].pmfw_breakdown.wm_type = WATERMARKS_DUMMY_PSTATE;
+	base->bw_params->wm_table.nv_entries[WM_C].pmfw_breakdown.min_dcfclk = 0;
+	base->bw_params->wm_table.nv_entries[WM_C].pmfw_breakdown.max_dcfclk = 0xFFFF;
+	base->bw_params->wm_table.nv_entries[WM_C].pmfw_breakdown.min_uclk = min_uclk_mhz;
+	base->bw_params->wm_table.nv_entries[WM_C].pmfw_breakdown.max_uclk = 0xFFFF;
+	base->bw_params->dummy_pstate_table[0].dram_speed_mts = 1600;
+	base->bw_params->dummy_pstate_table[0].dummy_pstate_latency_us = 38;
+	base->bw_params->dummy_pstate_table[1].dram_speed_mts = 8000;
+	base->bw_params->dummy_pstate_table[1].dummy_pstate_latency_us = 9;
+	base->bw_params->dummy_pstate_table[2].dram_speed_mts = 10000;
+	base->bw_params->dummy_pstate_table[2].dummy_pstate_latency_us = 8;
+	base->bw_params->dummy_pstate_table[3].dram_speed_mts = 16000;
+	base->bw_params->dummy_pstate_table[3].dummy_pstate_latency_us = 5;
+
+	/* Set D - MALL - SR enter and exit times adjusted for MALL */
+	base->bw_params->wm_table.nv_entries[WM_D].valid = true;
+	base->bw_params->wm_table.nv_entries[WM_D].dml_input.pstate_latency_us = pstate_latency_us;
+	base->bw_params->wm_table.nv_entries[WM_D].dml_input.sr_exit_time_us = 2;
+	base->bw_params->wm_table.nv_entries[WM_D].dml_input.sr_enter_plus_exit_time_us = 4;
+	base->bw_params->wm_table.nv_entries[WM_D].pmfw_breakdown.wm_type = WATERMARKS_MALL;
+	base->bw_params->wm_table.nv_entries[WM_D].pmfw_breakdown.min_dcfclk = 0;
+	base->bw_params->wm_table.nv_entries[WM_D].pmfw_breakdown.max_dcfclk = 0xFFFF;
+	base->bw_params->wm_table.nv_entries[WM_D].pmfw_breakdown.min_uclk = min_uclk_mhz;
+	base->bw_params->wm_table.nv_entries[WM_D].pmfw_breakdown.max_uclk = 0xFFFF;
+}
+
+void patch_dcn30_soc_bounding_box(struct dc *dc, struct _vcs_dpi_soc_bounding_box_st *soc_bb)
+{
+	(void)soc_bb;
+	dc_assert_fp_enabled();
+
+	if (dc->ctx->dc_bios->funcs->get_soc_bb_info) {
+		struct bp_soc_bb_info bb_info = {0};
+
+		if (dc->ctx->dc_bios->funcs->get_soc_bb_info(dc->ctx->dc_bios, &bb_info) == BP_RESULT_OK) {
+			if (bb_info.dram_clock_change_latency_100ns > 0)
+				dcn3_0_soc.dram_clock_change_latency_us = bb_info.dram_clock_change_latency_100ns * 10;
+
+			if (bb_info.dram_sr_enter_exit_latency_100ns > 0)
+				dcn3_0_soc.sr_enter_plus_exit_time_us = bb_info.dram_sr_enter_exit_latency_100ns * 10;
+
+			if (bb_info.dram_sr_exit_latency_100ns > 0)
+				dcn3_0_soc.sr_exit_time_us = bb_info.dram_sr_exit_latency_100ns * 10;
+		}
+	}
+}
+
+#undef DC_LOGGER
+#define DC_LOGGER \
+		enc3->base.ctx->logger
+
+#define hdmi_frl_print(str, ...) {DC_LOG_HDMI_FRL(str, ##__VA_ARGS__); }
+
+#define DEBUG_FRL_CAP_CHK 1
+
+void hpo_fpu_enc3_validate_hdmi_frl_output_link(struct hpo_frl_stream_encoder *enc,
+						struct dc_hdmi_frl_link_settings *frl_link_settings,
+						struct frl_cap_chk_params *frl_params,
+						const struct dc_crtc_timing *timing,
+						unsigned int dsc_max_rate)
+{
+	(void)enc;
+	dc_assert_fp_enabled();
+
+	switch (frl_link_settings->frl_link_rate) {
+	case HDMI_FRL_LINK_RATE_3GBPS:
+		frl_params->r_bit_nominal = 3.0e9;
+		break;
+	case HDMI_FRL_LINK_RATE_6GBPS:
+	case HDMI_FRL_LINK_RATE_6GBPS_4LANE:
+		frl_params->r_bit_nominal = 6.0e9;
+		break;
+	case HDMI_FRL_LINK_RATE_8GBPS:
+		frl_params->r_bit_nominal = 8.0e9;
+		break;
+	case HDMI_FRL_LINK_RATE_10GBPS:
+	default:
+		frl_params->r_bit_nominal = 10.0e9;
+		break;
+	case HDMI_FRL_LINK_RATE_12GBPS:
+		frl_params->r_bit_nominal = 12.0e9;
+		break;
+	}
+
+	if (timing->flags.DSC &&
+			(unsigned int)frl_link_settings->frl_link_rate > dsc_max_rate) {
+		if (dsc_max_rate < HDMI_FRL_LINK_RATE_6GBPS_4LANE) {
+			frl_params->lanes = 3;
+		} else {
+			frl_params->lanes = 4;
+		}
+
+		switch (dsc_max_rate) {
+		case HDMI_FRL_LINK_RATE_3GBPS:
+			frl_params->r_bit_nominal = 3.0e9;
+			break;
+		case HDMI_FRL_LINK_RATE_6GBPS:
+		case HDMI_FRL_LINK_RATE_6GBPS_4LANE:
+			frl_params->r_bit_nominal = 6.0e9;
+			break;
+		case HDMI_FRL_LINK_RATE_8GBPS:
+			frl_params->r_bit_nominal = 8.0e9;
+			break;
+		case HDMI_FRL_LINK_RATE_10GBPS:
+		default:
+			frl_params->r_bit_nominal = 10.0e9;
+			break;
+		case HDMI_FRL_LINK_RATE_12GBPS:
+			frl_params->r_bit_nominal = 12.0e9;
+			break;
+		}
+	}
+
+	if (timing->flags.DSC && timing->rid > 0)
+		frl_params->is_ovt = true;
+
+	frl_params->f_pixel_clock_nominal = (double)timing->pix_clk_100hz * 100;
+	frl_params->h_active = timing->h_addressable + timing->h_border_left + timing->h_border_right;
+	frl_params->h_blank = timing->h_total - frl_params->h_active;
+	frl_params->vic = timing->vic;
+}
+
+void hpo_fpu_enc3_validate_hdmi_frl_output_timing(
+		const struct dc_crtc_timing *timing,
+		const struct audio_check *audio,
+		struct frl_cap_chk_params *frl_params)
+{
+	dc_assert_fp_enabled();
+
+	if (timing->flags.DSC) {
+		frl_params->compressed = true;
+		frl_params->slices = timing->dsc_cfg.num_slices_h;
+		frl_params->slice_width = (frl_params->h_active + frl_params->slices - 1) / frl_params->slices;
+		// If YCBCR420 slice width must be even
+		if (timing->pixel_encoding == PIXEL_ENCODING_YCBCR420 && frl_params->slice_width % 2 != 0)
+			frl_params->slice_width++;
+		frl_params->bpp_target = timing->dsc_cfg.bits_per_pixel / 16.0;
+	} else {
+		frl_params->compressed = false;
+	}
+
+	frl_params->audio_packet_type = audio->audio_packet_type;
+	frl_params->f_audio = audio->max_audiosample_rate;
+	frl_params->acat = audio->acat;
+}
+
+enum frl_cap_chk_result frl_fpu_cap_chk_common(struct hpo_frl_stream_encoder *enc,
+					       struct frl_cap_chk_intermediates *inter,
+					       struct frl_cap_chk_params *params)
+{
+	(void)enc;
+	return dml1_frl_cap_chk_common(inter, params);
+}
+
+
+enum frl_cap_chk_result frl_fpu_cap_chk_uncompressed(struct hpo_frl_stream_encoder *enc,
+						     struct frl_cap_chk_params *params,
+						     struct frl_cap_chk_intermediates *inter)
+{
+	(void)enc;
+	return dml1_frl_cap_chk_uncompressed(params, inter);
+}
+
+enum frl_cap_chk_result frl_fpu_cap_chk_compressed(struct hpo_frl_stream_encoder *enc,
+						   struct frl_cap_chk_params *params,
+						   struct frl_cap_chk_intermediates *inter)
+{
+	(void)enc;
+	return dml1_frl_cap_chk_compressed(params, inter);
+}

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dcn30/dcn30_fpu.h

@@ -0,0 +1,95 @@
+/*
+ * 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 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);
+
+int dcn30_find_dummy_latency_index_for_fw_based_mclk_switch(struct dc *dc,
+							    struct dc_state *context,
+							    display_e2e_pipe_params_st *pipes,
+							    int pipe_cnt,
+							    int vlevel);
+
+void dcn3_fpu_build_wm_range_table(struct clk_mgr *base);
+
+void patch_dcn30_soc_bounding_box(struct dc *dc, struct _vcs_dpi_soc_bounding_box_st *dcn3_0_ip);
+
+void hpo_fpu_enc3_validate_hdmi_frl_output_link(struct hpo_frl_stream_encoder *enc,
+						struct dc_hdmi_frl_link_settings *frl_link_settings,
+						struct frl_cap_chk_params *frl_params,
+						const struct dc_crtc_timing *timing,
+						unsigned int dsc_max_rate);
+
+void hpo_fpu_enc3_validate_hdmi_frl_output_timing(const struct dc_crtc_timing *timing,
+						  const struct audio_check *audio,
+						  struct frl_cap_chk_params *frl_params);
+
+enum frl_cap_chk_result frl_fpu_cap_chk_common(struct hpo_frl_stream_encoder *enc,
+					       struct frl_cap_chk_intermediates *inter,
+					       struct frl_cap_chk_params *params);
+
+enum frl_cap_chk_result frl_fpu_cap_chk_uncompressed(struct hpo_frl_stream_encoder *enc,
+						     struct frl_cap_chk_params *params,
+						     struct frl_cap_chk_intermediates *inter);
+
+enum frl_cap_chk_result frl_fpu_cap_chk_compressed(struct hpo_frl_stream_encoder *enc,
+						   struct frl_cap_chk_params *params,
+						   struct frl_cap_chk_intermediates *inter);
+
+#endif /* __DCN30_FPU_H__*/

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.h

@@ -0,0 +1,54 @@
+/*
+ * Copyright 2020 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 __DML30_DISPLAY_MODE_VBA_H__
+#define __DML30_DISPLAY_MODE_VBA_H__
+
+void dml30_recalculate(struct display_mode_lib *mode_lib);
+void dml30_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_lib);
+double dml30_CalculateWriteBackDISPCLK(
+		enum source_format_class WritebackPixelFormat,
+		double PixelClock,
+		double WritebackHRatio,
+		double WritebackVRatio,
+		unsigned int WritebackHTaps,
+		unsigned int WritebackVTaps,
+		long   WritebackSourceWidth,
+		long   WritebackDestinationWidth,
+		unsigned int HTotal,
+		unsigned int WritebackLineBufferSize);
+void dml30_CalculateBytePerPixelAnd256BBlockSizes(
+		enum source_format_class SourcePixelFormat,
+		enum dm_swizzle_mode SurfaceTiling,
+		unsigned int *BytePerPixelY,
+		unsigned int *BytePerPixelC,
+		double       *BytePerPixelDETY,
+		double       *BytePerPixelDETC,
+		unsigned int *BlockHeight256BytesY,
+		unsigned int *BlockHeight256BytesC,
+		unsigned int *BlockWidth256BytesY,
+		unsigned int *BlockWidth256BytesC);
+
+#endif /* __DML30_DISPLAY_MODE_VBA_H__ */

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_rq_dlg_calc_30.h

@@ -0,0 +1,69 @@
+/*
+ * Copyright 2020 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 __DML30_DISPLAY_RQ_DLG_CALC_H__
+#define __DML30_DISPLAY_RQ_DLG_CALC_H__
+
+#include "../display_rq_dlg_helpers.h"
+
+struct display_mode_lib;
+
+// Function: dml_rq_dlg_get_rq_reg
+//  Main entry point for test to get the register values out of this DML class.
+//  This function calls <get_rq_param> and <extract_rq_regs> functions to calculate
+//  and then populate the rq_regs struct
+// Input:
+//  pipe_param - pipe source configuration (e.g. vp, pitch, scaling, dest, etc.)
+// Output:
+//  rq_regs - struct that holds all the RQ registers field value.
+//            See also: <display_rq_regs_st>
+void dml30_rq_dlg_get_rq_reg(struct display_mode_lib *mode_lib,
+		display_rq_regs_st *rq_regs,
+		const display_pipe_params_st *pipe_param);
+
+// Function: dml_rq_dlg_get_dlg_reg
+//   Calculate and return DLG and TTU register struct given the system setting
+// Output:
+//  dlg_regs - output DLG register struct
+//  ttu_regs - output DLG TTU register struct
+// Input:
+//  e2e_pipe_param - "compacted" array of e2e pipe param struct
+//  num_pipes - num of active "pipe" or "route"
+//  pipe_idx - index that identifies the e2e_pipe_param that corresponding to this dlg
+//  cstate - 0: when calculate min_ttu_vblank it is assumed cstate is not required. 1: Normal mode, cstate is considered.
+//           Added for legacy or unrealistic timing tests.
+void dml30_rq_dlg_get_dlg_reg(struct display_mode_lib             *mode_lib,
+		display_dlg_regs_st          *dlg_regs,
+		display_ttu_regs_st          *ttu_regs,
+		const display_e2e_pipe_params_st   *e2e_pipe_param,
+		const unsigned int            num_pipes,
+		const unsigned int            pipe_idx,
+		const bool                    cstate_en,
+		const bool                    pstate_en,
+		const bool                    vm_en,
+		const bool                    ignore_viewport_pos,
+		const bool                    immediate_flip_support);
+
+#endif

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dcn301/dcn301_fpu.c

@@ -0,0 +1,487 @@
+/*
+ * Copyright 2019-2021 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 "dcn20/dcn20_resource.h"
+#include "dcn301/dcn301_resource.h"
+#include "clk_mgr/dcn301/vg_clk_mgr.h"
+
+#include "dml/dcn20/dcn20_fpu.h"
+#include "dcn301_fpu.h"
+
+#define TO_DCN301_RES_POOL(pool)\
+	container_of(pool, struct dcn301_resource_pool, base)
+
+/* Based on: //vidip/dc/dcn3/doc/architecture/DCN3x_Display_Mode.xlsm#83 */
+struct _vcs_dpi_ip_params_st dcn3_01_ip = {
+	.odm_capable = 1,
+	.gpuvm_enable = 1,
+	.hostvm_enable = 1,
+	.gpuvm_max_page_table_levels = 1,
+	.hostvm_max_page_table_levels = 2,
+	.hostvm_cached_page_table_levels = 0,
+	.pte_group_size_bytes = 2048,
+	.num_dsc = 3,
+	.rob_buffer_size_kbytes = 184,
+	.det_buffer_size_kbytes = 184,
+	.dpte_buffer_size_in_pte_reqs_luma = 64,
+	.dpte_buffer_size_in_pte_reqs_chroma = 32,
+	.pde_proc_buffer_size_64k_reqs = 48,
+	.dpp_output_buffer_pixels = 2560,
+	.opp_output_buffer_lines = 1,
+	.pixel_chunk_size_kbytes = 8,
+	.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 = 48,     // ?
+	.dcc_supported = true,
+	.writeback_interface_buffer_size_kbytes = 90,
+	.writeback_line_buffer_buffer_size = 656640,
+	.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 = 4,
+	.max_num_dpp = 4,
+	.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 = 0,
+	.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_01_soc = {
+	.clock_limits = {
+		{
+			.state = 0,
+			.dram_speed_mts = 2400.0,
+			.fabricclk_mhz = 600,
+			.socclk_mhz = 278.0,
+			.dcfclk_mhz = 400.0,
+			.dscclk_mhz = 206.0,
+			.dppclk_mhz = 1015.0,
+			.dispclk_mhz = 1015.0,
+			.phyclk_mhz = 600.0,
+		},
+
+		{
+			.state = 1,
+			.dram_speed_mts = 2400.0,
+			.fabricclk_mhz = 688,
+			.socclk_mhz = 278.0,
+			.dcfclk_mhz = 400.0,
+			.dscclk_mhz = 206.0,
+			.dppclk_mhz = 1015.0,
+			.dispclk_mhz = 1015.0,
+			.phyclk_mhz = 600.0,
+		},
+
+		{
+			.state = 2,
+			.dram_speed_mts = 4267.0,
+			.fabricclk_mhz = 1067,
+			.socclk_mhz = 278.0,
+			.dcfclk_mhz = 608.0,
+			.dscclk_mhz = 296.0,
+			.dppclk_mhz = 1015.0,
+			.dispclk_mhz = 1015.0,
+			.phyclk_mhz = 810.0,
+		},
+
+		{
+			.state = 3,
+			.dram_speed_mts = 4267.0,
+			.fabricclk_mhz = 1067,
+			.socclk_mhz = 715.0,
+			.dcfclk_mhz = 676.0,
+			.dscclk_mhz = 338.0,
+			.dppclk_mhz = 1015.0,
+			.dispclk_mhz = 1015.0,
+			.phyclk_mhz = 810.0,
+		},
+
+		{
+			.state = 4,
+			.dram_speed_mts = 4267.0,
+			.fabricclk_mhz = 1067,
+			.socclk_mhz = 953.0,
+			.dcfclk_mhz = 810.0,
+			.dscclk_mhz = 338.0,
+			.dppclk_mhz = 1015.0,
+			.dispclk_mhz = 1015.0,
+			.phyclk_mhz = 810.0,
+		},
+	},
+
+	.sr_exit_time_us = 9.0,
+	.sr_enter_plus_exit_time_us = 11.0,
+	.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 = 75.0,
+	.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 = 60.0,
+	.writeback_latency_us = 12.0,
+	.max_request_size_bytes = 256,
+	.dram_channel_width_bytes = 4,
+	.fabric_datapath_to_dcn_data_return_bytes = 32,
+	.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,
+	.num_chans = 4,
+	.gpuvm_min_page_size_bytes = 4096,
+	.hostvm_min_page_size_bytes = 4096,
+	.dram_clock_change_latency_us = 23.84,
+	.writeback_dram_clock_change_latency_us = 23.0,
+	.return_bus_width_bytes = 64,
+	.dispclk_dppclk_vco_speed_mhz = 3550,
+	.xfc_bus_transport_time_us = 20,      // ?
+	.xfc_xbuf_latency_tolerance_us = 4,  // ?
+	.use_urgent_burst_bw = 1,            // ?
+	.num_states = 5,
+	.do_urgent_latency_adjustment = false,
+	.urgent_latency_adjustment_fabric_clock_component_us = 0,
+	.urgent_latency_adjustment_fabric_clock_reference_mhz = 0,
+};
+
+struct wm_table ddr4_wm_table = {
+	.entries = {
+		{
+			.wm_inst = WM_A,
+			.wm_type = WM_TYPE_PSTATE_CHG,
+			.pstate_latency_us = 11.72,
+			.sr_exit_time_us = 6.09,
+			.sr_enter_plus_exit_time_us = 7.14,
+			.valid = true,
+		},
+		{
+			.wm_inst = WM_B,
+			.wm_type = WM_TYPE_PSTATE_CHG,
+			.pstate_latency_us = 11.72,
+			.sr_exit_time_us = 10.12,
+			.sr_enter_plus_exit_time_us = 11.48,
+			.valid = true,
+		},
+		{
+			.wm_inst = WM_C,
+			.wm_type = WM_TYPE_PSTATE_CHG,
+			.pstate_latency_us = 11.72,
+			.sr_exit_time_us = 10.12,
+			.sr_enter_plus_exit_time_us = 11.48,
+			.valid = true,
+		},
+		{
+			.wm_inst = WM_D,
+			.wm_type = WM_TYPE_PSTATE_CHG,
+			.pstate_latency_us = 11.72,
+			.sr_exit_time_us = 10.12,
+			.sr_enter_plus_exit_time_us = 11.48,
+			.valid = true,
+		},
+	}
+};
+
+struct wm_table lpddr5_wm_table = {
+	.entries = {
+		{
+			.wm_inst = WM_A,
+			.wm_type = WM_TYPE_PSTATE_CHG,
+			.pstate_latency_us = 11.65333,
+			.sr_exit_time_us = 13.5,
+			.sr_enter_plus_exit_time_us = 16.5,
+			.valid = true,
+		},
+		{
+			.wm_inst = WM_B,
+			.wm_type = WM_TYPE_PSTATE_CHG,
+			.pstate_latency_us = 11.65333,
+			.sr_exit_time_us = 13.5,
+			.sr_enter_plus_exit_time_us = 16.5,
+			.valid = true,
+		},
+		{
+			.wm_inst = WM_C,
+			.wm_type = WM_TYPE_PSTATE_CHG,
+			.pstate_latency_us = 11.65333,
+			.sr_exit_time_us = 13.5,
+			.sr_enter_plus_exit_time_us = 16.5,
+			.valid = true,
+		},
+		{
+			.wm_inst = WM_D,
+			.wm_type = WM_TYPE_PSTATE_CHG,
+			.pstate_latency_us = 11.65333,
+			.sr_exit_time_us = 13.5,
+			.sr_enter_plus_exit_time_us = 16.5,
+			.valid = true,
+		},
+	}
+};
+
+static void calculate_wm_set_for_vlevel(int vlevel,
+		struct wm_range_table_entry *table_entry,
+		struct dcn_watermarks *wm_set,
+		struct display_mode_lib *dml,
+		display_e2e_pipe_params_st *pipes,
+		int pipe_cnt)
+{
+	double dram_clock_change_latency_cached = dml->soc.dram_clock_change_latency_us;
+
+	ASSERT(vlevel < (int)dml->soc.num_states);
+	/* only pipe 0 is read for voltage and dcf/soc clocks */
+	pipes[0].clks_cfg.voltage = vlevel;
+	pipes[0].clks_cfg.dcfclk_mhz = dml->soc.clock_limits[vlevel].dcfclk_mhz;
+	pipes[0].clks_cfg.socclk_mhz = dml->soc.clock_limits[vlevel].socclk_mhz;
+
+	dml->soc.dram_clock_change_latency_us = table_entry->pstate_latency_us;
+	dml->soc.sr_exit_time_us = table_entry->sr_exit_time_us;
+	dml->soc.sr_enter_plus_exit_time_us = table_entry->sr_enter_plus_exit_time_us;
+
+	wm_set->urgent_ns = (uint32_t)(get_wm_urgent(dml, pipes, pipe_cnt) * 1000);
+	wm_set->cstate_pstate.cstate_enter_plus_exit_ns = (uint32_t)(get_wm_stutter_enter_exit(dml, pipes, pipe_cnt) * 1000);
+	wm_set->cstate_pstate.cstate_exit_ns = (uint32_t)(get_wm_stutter_exit(dml, pipes, pipe_cnt) * 1000);
+	wm_set->cstate_pstate.pstate_change_ns = (uint32_t)(get_wm_dram_clock_change(dml, pipes, pipe_cnt) * 1000);
+	wm_set->pte_meta_urgent_ns = (uint32_t)(get_wm_memory_trip(dml, pipes, pipe_cnt) * 1000);
+	wm_set->frac_urg_bw_nom = (uint32_t)(get_fraction_of_urgent_bandwidth(dml, pipes, pipe_cnt) * 1000);
+	wm_set->frac_urg_bw_flip = (uint32_t)(get_fraction_of_urgent_bandwidth_imm_flip(dml, pipes, pipe_cnt) * 1000);
+	wm_set->urgent_latency_ns = (uint32_t)(get_urgent_latency(dml, pipes, pipe_cnt) * 1000);
+	dml->soc.dram_clock_change_latency_us = dram_clock_change_latency_cached;
+
+}
+
+void dcn301_fpu_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params)
+{
+	struct _vcs_dpi_voltage_scaling_st *s = dc->scratch.update_bw_bounding_box.clock_limits;
+	struct dcn301_resource_pool *pool = TO_DCN301_RES_POOL(dc->res_pool);
+	struct clk_limit_table *clk_table = &bw_params->clk_table;
+	unsigned int i, closest_clk_lvl;
+	int j = 0;
+	unsigned int max_dispclk_mhz = 0, max_dppclk_mhz = 0;
+
+	dc_assert_fp_enabled();
+
+	memcpy(s, dcn3_01_soc.clock_limits, sizeof(dcn3_01_soc.clock_limits));
+
+	/* Default clock levels are used for diags, which may lead to overclocking. */
+	dcn3_01_ip.max_num_otg = pool->base.res_cap->num_timing_generator;
+	dcn3_01_ip.max_num_dpp = pool->base.pipe_count;
+	dcn3_01_soc.num_chans = bw_params->num_channels;
+
+	ASSERT(clk_table->num_entries);
+
+	/* Prepass to find max clocks independent of voltage level. */
+	for (i = 0; i < clk_table->num_entries; ++i) {
+		if (clk_table->entries[i].dispclk_mhz > max_dispclk_mhz)
+			max_dispclk_mhz = clk_table->entries[i].dispclk_mhz;
+		if (clk_table->entries[i].dppclk_mhz > max_dppclk_mhz)
+			max_dppclk_mhz = clk_table->entries[i].dppclk_mhz;
+	}
+
+	for (i = 0; i < clk_table->num_entries; i++) {
+		/* loop backwards*/
+		for (closest_clk_lvl = 0, j = dcn3_01_soc.num_states - 1; j >= 0; j--) {
+			if ((unsigned int) dcn3_01_soc.clock_limits[j].dcfclk_mhz <= clk_table->entries[i].dcfclk_mhz) {
+				closest_clk_lvl = j;
+				break;
+			}
+		}
+
+		s[i].state = i;
+		s[i].dcfclk_mhz = clk_table->entries[i].dcfclk_mhz;
+		s[i].fabricclk_mhz = clk_table->entries[i].fclk_mhz;
+		s[i].socclk_mhz = clk_table->entries[i].socclk_mhz;
+		s[i].dram_speed_mts = clk_table->entries[i].memclk_mhz * 2;
+
+		/* Clocks independent of voltage level. */
+		s[i].dispclk_mhz = max_dispclk_mhz ? max_dispclk_mhz :
+			dcn3_01_soc.clock_limits[closest_clk_lvl].dispclk_mhz;
+
+		s[i].dppclk_mhz = max_dppclk_mhz ? max_dppclk_mhz :
+			dcn3_01_soc.clock_limits[closest_clk_lvl].dppclk_mhz;
+
+		s[i].dram_bw_per_chan_gbps =
+			dcn3_01_soc.clock_limits[closest_clk_lvl].dram_bw_per_chan_gbps;
+		s[i].dscclk_mhz = dcn3_01_soc.clock_limits[closest_clk_lvl].dscclk_mhz;
+		s[i].dtbclk_mhz = dcn3_01_soc.clock_limits[closest_clk_lvl].dtbclk_mhz;
+		s[i].phyclk_d18_mhz =
+			dcn3_01_soc.clock_limits[closest_clk_lvl].phyclk_d18_mhz;
+		s[i].phyclk_mhz = dcn3_01_soc.clock_limits[closest_clk_lvl].phyclk_mhz;
+	}
+
+	if (clk_table->num_entries) {
+		dcn3_01_soc.num_states = clk_table->num_entries;
+		/* duplicate last level */
+		s[dcn3_01_soc.num_states] =
+			dcn3_01_soc.clock_limits[dcn3_01_soc.num_states - 1];
+		s[dcn3_01_soc.num_states].state = dcn3_01_soc.num_states;
+	}
+
+	memcpy(dcn3_01_soc.clock_limits, s, sizeof(dcn3_01_soc.clock_limits));
+
+	dcn3_01_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;
+
+	if ((int)(dcn3_01_soc.dram_clock_change_latency_us * 1000)
+				!= dc->debug.dram_clock_change_latency_ns
+			&& dc->debug.dram_clock_change_latency_ns) {
+		dcn3_01_soc.dram_clock_change_latency_us = dc->debug.dram_clock_change_latency_ns / 1000.0;
+	}
+	dml_init_instance(&dc->dml, &dcn3_01_soc, &dcn3_01_ip, DML_PROJECT_DCN30);
+}
+
+void dcn301_fpu_set_wm_ranges(int i,
+	struct pp_smu_wm_range_sets *ranges,
+	struct _vcs_dpi_soc_bounding_box_st *loaded_bb)
+{
+	dc_assert_fp_enabled();
+
+	ranges->reader_wm_sets[i].min_fill_clk_mhz = (uint16_t)((i > 0) ? (loaded_bb->clock_limits[i - 1].dram_speed_mts / 16) + 1 : 0);
+	ranges->reader_wm_sets[i].max_fill_clk_mhz = (uint16_t)(loaded_bb->clock_limits[i].dram_speed_mts / 16);
+}
+
+void dcn301_fpu_init_soc_bounding_box(struct bp_soc_bb_info bb_info)
+{
+	dc_assert_fp_enabled();
+
+	if (bb_info.dram_clock_change_latency_100ns > 0)
+		dcn3_01_soc.dram_clock_change_latency_us = bb_info.dram_clock_change_latency_100ns * 10;
+
+	if (bb_info.dram_sr_enter_exit_latency_100ns > 0)
+		dcn3_01_soc.sr_enter_plus_exit_time_us = bb_info.dram_sr_enter_exit_latency_100ns * 10;
+
+	if (bb_info.dram_sr_exit_latency_100ns > 0)
+		dcn3_01_soc.sr_exit_time_us = bb_info.dram_sr_exit_latency_100ns * 10;
+}
+
+void dcn301_fpu_calculate_wm_and_dlg(struct dc *dc,
+		struct dc_state *context,
+		display_e2e_pipe_params_st *pipes,
+		int pipe_cnt,
+		int vlevel_req)
+{
+	unsigned int i, pipe_idx;
+	int vlevel, vlevel_max;
+	struct wm_range_table_entry *table_entry;
+	struct clk_bw_params *bw_params = dc->clk_mgr->bw_params;
+
+	ASSERT(bw_params);
+	dc_assert_fp_enabled();
+
+	vlevel_max = bw_params->clk_table.num_entries - 1;
+
+	/* WM Set D */
+	table_entry = &bw_params->wm_table.entries[WM_D];
+	if (table_entry->wm_type == WM_TYPE_RETRAINING)
+		vlevel = 0;
+	else
+		vlevel = vlevel_max;
+	calculate_wm_set_for_vlevel(vlevel, table_entry, &context->bw_ctx.bw.dcn.watermarks.d,
+						&context->bw_ctx.dml, pipes, pipe_cnt);
+	/* WM Set C */
+	table_entry = &bw_params->wm_table.entries[WM_C];
+	vlevel = clamp(vlevel_req, 2, vlevel_max);
+	calculate_wm_set_for_vlevel(vlevel, table_entry, &context->bw_ctx.bw.dcn.watermarks.c,
+						&context->bw_ctx.dml, pipes, pipe_cnt);
+	/* WM Set B */
+	table_entry = &bw_params->wm_table.entries[WM_B];
+	vlevel = clamp(vlevel_req, 1, vlevel_max);
+	calculate_wm_set_for_vlevel(vlevel, table_entry, &context->bw_ctx.bw.dcn.watermarks.b,
+						&context->bw_ctx.dml, pipes, pipe_cnt);
+
+	/* WM Set A */
+	table_entry = &bw_params->wm_table.entries[WM_A];
+	vlevel = min(vlevel_req, vlevel_max);
+	calculate_wm_set_for_vlevel(vlevel, table_entry, &context->bw_ctx.bw.dcn.watermarks.a,
+						&context->bw_ctx.dml, pipes, pipe_cnt);
+
+	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++;
+	}
+
+	dcn20_calculate_dlg_params(dc, context, pipes, pipe_cnt, vlevel);
+}

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dcn301/dcn301_fpu.h

@@ -0,0 +1,41 @@
+/*
+ * Copyright 2019-2021 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 __DCN301_FPU_H__
+#define __DCN301_FPU_H__
+
+void dcn301_fpu_init_soc_bounding_box(struct bp_soc_bb_info bb_info);
+void dcn301_fpu_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params);
+
+void dcn301_fpu_set_wm_ranges(int i,
+	struct pp_smu_wm_range_sets *ranges,
+	struct _vcs_dpi_soc_bounding_box_st *loaded_bb);
+
+void dcn301_fpu_calculate_wm_and_dlg(struct dc *dc,
+		struct dc_state *context,
+		display_e2e_pipe_params_st *pipes,
+		int pipe_cnt,
+		int vlevel_req);
+#endif /* __DCN301_FPU_H__*/

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dcn302/dcn302_fpu.c

@@ -0,0 +1,368 @@
+/*
+ * Copyright 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 "dcn20/dcn20_resource.h"
+#include "dcn302/dcn302_resource.h"
+
+#include "dml/dcn20/dcn20_fpu.h"
+#include "dcn302_fpu.h"
+
+struct _vcs_dpi_ip_params_st dcn3_02_ip = {
+		.use_min_dcfclk = 0,
+		.clamp_min_dcfclk = 0,
+		.odm_capable = 1,
+		.gpuvm_enable = 1,
+		.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 = 5,
+		.rob_buffer_size_kbytes = 184,
+		.det_buffer_size_kbytes = 184,
+		.dpte_buffer_size_in_pte_reqs_luma = 64,
+		.dpte_buffer_size_in_pte_reqs_chroma = 34,
+		.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 = 5,
+		.max_num_dpp = 5,
+		.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_02_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,
+						.dtbclk_mhz = 1217.0,
+				},
+		},
+
+		.min_dcfclk = 500.0, /* TODO: set this to actual min DCFCLK */
+		.num_states = 1,
+		.sr_exit_time_us = 26.5,
+		.sr_enter_plus_exit_time_us = 31,
+		.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 = 156,
+		.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,
+};
+
+static void dcn302_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_02_soc.num_chans *
+				dcn3_02_soc.dram_channel_width_bytes *
+				(dcn3_02_soc.max_avg_dram_bw_use_normal_percent / 100);
+			bw_from_dram2 = uclk_mts * dcn3_02_soc.num_chans *
+				dcn3_02_soc.dram_channel_width_bytes *
+				(dcn3_02_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 = (unsigned int)(bw_from_dram /
+				(dcn3_02_soc.fabric_datapath_to_dcn_data_return_bytes *
+				 (dcn3_02_soc.max_avg_sdp_bw_use_normal_percent / 100)));
+
+			if (optimal_dcfclk)
+				*optimal_dcfclk = (unsigned int)(bw_from_dram /
+				(dcn3_02_soc.return_bus_width_bytes *
+				 (dcn3_02_soc.max_avg_sdp_bw_use_normal_percent / 100)));
+}
+
+void dcn302_fpu_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params)
+{
+	unsigned int i, j;
+	unsigned int num_states = 0;
+
+	unsigned int dcfclk_mhz[DC__VOLTAGE_STATES] = {0};
+	unsigned int dram_speed_mts[DC__VOLTAGE_STATES] = {0};
+	unsigned int optimal_uclk_for_dcfclk_sta_targets[DC__VOLTAGE_STATES] = {0};
+	unsigned int optimal_dcfclk_for_uclk[DC__VOLTAGE_STATES] = {0};
+
+	unsigned int dcfclk_sta_targets[DC__VOLTAGE_STATES] = {694, 875, 1000, 1200};
+	unsigned int num_dcfclk_sta_targets = 4;
+	unsigned int num_uclk_states;
+
+	dc_assert_fp_enabled();
+
+	if (dc->ctx->dc_bios->vram_info.num_chans)
+		dcn3_02_soc.num_chans = dc->ctx->dc_bios->vram_info.num_chans;
+
+	if (dc->ctx->dc_bios->vram_info.dram_channel_width_bytes)
+		dcn3_02_soc.dram_channel_width_bytes = dc->ctx->dc_bios->vram_info.dram_channel_width_bytes;
+
+	dcn3_02_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;
+
+	if (bw_params->clk_table.entries[0].memclk_mhz) {
+		unsigned 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 (!max_dcfclk_mhz)
+			max_dcfclk_mhz = (int)dcn3_02_soc.clock_limits[0].dcfclk_mhz;
+		if (!max_dispclk_mhz)
+			max_dispclk_mhz = (int)dcn3_02_soc.clock_limits[0].dispclk_mhz;
+		if (!max_dppclk_mhz)
+			max_dppclk_mhz = (int)dcn3_02_soc.clock_limits[0].dppclk_mhz;
+		if (!max_phyclk_mhz)
+			max_phyclk_mhz = (int)dcn3_02_soc.clock_limits[0].phyclk_mhz;
+
+		if (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;
+			num_dcfclk_sta_targets++;
+		} else if (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;
+					break;
+				}
+			}
+			/* Update size of array since we "removed" duplicates */
+			num_dcfclk_sta_targets = i + 1;
+		}
+
+		num_uclk_states = bw_params->clk_table.num_entries;
+
+		/* Calculate optimal dcfclk for each uclk */
+		for (i = 0; i < num_uclk_states; i++) {
+			dcn302_get_optimal_dcfclk_fclk_for_uclk(bw_params->clk_table.entries[i].memclk_mhz * 16,
+					&optimal_dcfclk_for_uclk[i], NULL);
+			if (optimal_dcfclk_for_uclk[i] < bw_params->clk_table.entries[0].dcfclk_mhz)
+				optimal_dcfclk_for_uclk[i] = bw_params->clk_table.entries[0].dcfclk_mhz;
+		}
+
+		/* Calculate optimal uclk for each dcfclk sta target */
+		for (i = 0; i < num_dcfclk_sta_targets; i++) {
+			for (j = 0; j < num_uclk_states; j++) {
+				if (dcfclk_sta_targets[i] < optimal_dcfclk_for_uclk[j]) {
+					optimal_uclk_for_dcfclk_sta_targets[i] =
+							bw_params->clk_table.entries[j].memclk_mhz * 16;
+					break;
+				}
+			}
+		}
+
+		i = 0;
+		j = 0;
+		/* create the final dcfclk and uclk table */
+		while (i < num_dcfclk_sta_targets && j < num_uclk_states && num_states < DC__VOLTAGE_STATES) {
+			if (dcfclk_sta_targets[i] < optimal_dcfclk_for_uclk[j]) {
+				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) {
+					dcfclk_mhz[num_states] = optimal_dcfclk_for_uclk[j];
+					dram_speed_mts[num_states++] = bw_params->clk_table.entries[j++].memclk_mhz * 16;
+				} else {
+					j = num_uclk_states;
+				}
+			}
+		}
+
+		while (i < num_dcfclk_sta_targets && num_states < DC__VOLTAGE_STATES) {
+			dcfclk_mhz[num_states] = dcfclk_sta_targets[i];
+			dram_speed_mts[num_states++] = optimal_uclk_for_dcfclk_sta_targets[i++];
+		}
+
+		while (j < num_uclk_states && num_states < DC__VOLTAGE_STATES &&
+				optimal_dcfclk_for_uclk[j] <= 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;
+		}
+
+		/* bw_params->clk_table.entries[MAX_NUM_DPM_LVL].
+		 * MAX_NUM_DPM_LVL is 8.
+		 * dcn3_02_soc.clock_limits[DC__VOLTAGE_STATES].
+		 * DC__VOLTAGE_STATES is 40.
+		 */
+		if (num_states > MAX_NUM_DPM_LVL) {
+			ASSERT(0);
+			return;
+		}
+
+		dcn3_02_soc.num_states = num_states;
+		for (i = 0; i < dcn3_02_soc.num_states; i++) {
+			dcn3_02_soc.clock_limits[i].state = i;
+			dcn3_02_soc.clock_limits[i].dcfclk_mhz = dcfclk_mhz[i];
+			dcn3_02_soc.clock_limits[i].fabricclk_mhz = dcfclk_mhz[i];
+			dcn3_02_soc.clock_limits[i].dram_speed_mts = dram_speed_mts[i];
+
+			/* Fill all states with max values of all other clocks */
+			dcn3_02_soc.clock_limits[i].dispclk_mhz = max_dispclk_mhz;
+			dcn3_02_soc.clock_limits[i].dppclk_mhz  = max_dppclk_mhz;
+			dcn3_02_soc.clock_limits[i].phyclk_mhz  = max_phyclk_mhz;
+			/* Populate from bw_params for DTBCLK, SOCCLK */
+			if (!bw_params->clk_table.entries[i].dtbclk_mhz && i > 0)
+				dcn3_02_soc.clock_limits[i].dtbclk_mhz  = dcn3_02_soc.clock_limits[i-1].dtbclk_mhz;
+			else
+				dcn3_02_soc.clock_limits[i].dtbclk_mhz  = bw_params->clk_table.entries[i].dtbclk_mhz;
+			if (!bw_params->clk_table.entries[i].socclk_mhz && i > 0)
+				dcn3_02_soc.clock_limits[i].socclk_mhz = dcn3_02_soc.clock_limits[i-1].socclk_mhz;
+			else
+				dcn3_02_soc.clock_limits[i].socclk_mhz = bw_params->clk_table.entries[i].socclk_mhz;
+			/* These clocks cannot come from bw_params, always fill from dcn3_02_soc[1] */
+			/* FCLK, PHYCLK_D18, DSCCLK */
+			dcn3_02_soc.clock_limits[i].phyclk_d18_mhz = dcn3_02_soc.clock_limits[0].phyclk_d18_mhz;
+			dcn3_02_soc.clock_limits[i].dscclk_mhz = dcn3_02_soc.clock_limits[0].dscclk_mhz;
+		}
+		/* re-init DML with updated bb */
+		dml_init_instance(&dc->dml, &dcn3_02_soc, &dcn3_02_ip, DML_PROJECT_DCN30);
+		if (dc->current_state)
+			dml_init_instance(&dc->current_state->bw_ctx.dml, &dcn3_02_soc, &dcn3_02_ip, DML_PROJECT_DCN30);
+	}
+}
+
+void dcn302_fpu_init_soc_bounding_box(struct bp_soc_bb_info bb_info)
+{
+
+			dc_assert_fp_enabled();
+
+			if (bb_info.dram_clock_change_latency_100ns > 0)
+				dcn3_02_soc.dram_clock_change_latency_us =
+					bb_info.dram_clock_change_latency_100ns * 10;
+
+			if (bb_info.dram_sr_enter_exit_latency_100ns > 0)
+				dcn3_02_soc.sr_enter_plus_exit_time_us =
+					bb_info.dram_sr_enter_exit_latency_100ns * 10;
+
+			if (bb_info.dram_sr_exit_latency_100ns > 0)
+				dcn3_02_soc.sr_exit_time_us =
+					bb_info.dram_sr_exit_latency_100ns * 10;
+}
+
+

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dcn302/dcn302_fpu.h

@@ -0,0 +1,32 @@
+/*
+ * Copyright 2019-2021 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 __DCN302_FPU_H__
+#define __DCN302_FPU_H__
+
+void dcn302_fpu_init_soc_bounding_box(struct bp_soc_bb_info bb_info);
+void dcn302_fpu_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params);
+
+#endif /* __DCN302_FPU_H__*/

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dcn303/dcn303_fpu.c

@@ -0,0 +1,381 @@
+/*
+ * Copyright 2019-2021 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 "dcn20/dcn20_resource.h"
+#include "dcn303/dcn303_resource.h"
+
+#include "dml/dcn20/dcn20_fpu.h"
+#include "dcn303_fpu.h"
+
+struct _vcs_dpi_ip_params_st dcn3_03_ip = {
+		.use_min_dcfclk = 0,
+		.clamp_min_dcfclk = 0,
+		.odm_capable = 1,
+		.gpuvm_enable = 1,
+		.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 = 2,
+		.rob_buffer_size_kbytes = 184,
+		.det_buffer_size_kbytes = 184,
+		.dpte_buffer_size_in_pte_reqs_luma = 64,
+		.dpte_buffer_size_in_pte_reqs_chroma = 34,
+		.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 = 2,
+		.max_num_dpp = 2,
+		.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 = false,
+
+		.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_03_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,
+						.dtbclk_mhz = 1217.0,
+				},
+		},
+
+		.min_dcfclk = 500.0, /* TODO: set this to actual min DCFCLK */
+		.num_states = 1,
+		.sr_exit_time_us = 35.5,
+		.sr_enter_plus_exit_time_us = 40,
+		.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 = 156,
+		.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,
+};
+
+static void dcn303_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_03_soc.num_chans *
+		dcn3_03_soc.dram_channel_width_bytes * (dcn3_03_soc.max_avg_dram_bw_use_normal_percent / 100);
+	bw_from_dram2 = uclk_mts * dcn3_03_soc.num_chans *
+		dcn3_03_soc.dram_channel_width_bytes * (dcn3_03_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 = (unsigned int)(bw_from_dram /
+		(dcn3_03_soc.fabric_datapath_to_dcn_data_return_bytes *
+				(dcn3_03_soc.max_avg_sdp_bw_use_normal_percent / 100)));
+
+	if (optimal_dcfclk)
+		*optimal_dcfclk = (unsigned int)(bw_from_dram /
+		(dcn3_03_soc.return_bus_width_bytes * (dcn3_03_soc.max_avg_sdp_bw_use_normal_percent / 100)));
+}
+
+
+void dcn303_fpu_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params)
+{
+	unsigned int i, j;
+	unsigned int num_states = 0;
+
+	unsigned int dcfclk_mhz[DC__VOLTAGE_STATES] = {0};
+	unsigned int dram_speed_mts[DC__VOLTAGE_STATES] = {0};
+	unsigned int optimal_uclk_for_dcfclk_sta_targets[DC__VOLTAGE_STATES] = {0};
+	unsigned int optimal_dcfclk_for_uclk[DC__VOLTAGE_STATES] = {0};
+
+	unsigned int dcfclk_sta_targets[DC__VOLTAGE_STATES] = {694, 875, 1000, 1200};
+	unsigned int num_dcfclk_sta_targets = 4;
+	unsigned int num_uclk_states;
+
+	dc_assert_fp_enabled();
+
+	if (dc->ctx->dc_bios->vram_info.num_chans)
+		dcn3_03_soc.num_chans = dc->ctx->dc_bios->vram_info.num_chans;
+
+	if (dc->ctx->dc_bios->vram_info.dram_channel_width_bytes)
+		dcn3_03_soc.dram_channel_width_bytes = dc->ctx->dc_bios->vram_info.dram_channel_width_bytes;
+
+	dcn3_03_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;
+
+	if (bw_params->clk_table.entries[0].memclk_mhz) {
+		unsigned 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 (!max_dcfclk_mhz)
+			max_dcfclk_mhz = (int)dcn3_03_soc.clock_limits[0].dcfclk_mhz;
+		if (!max_dispclk_mhz)
+			max_dispclk_mhz = (int)dcn3_03_soc.clock_limits[0].dispclk_mhz;
+		if (!max_dppclk_mhz)
+			max_dppclk_mhz = (int)dcn3_03_soc.clock_limits[0].dppclk_mhz;
+		if (!max_phyclk_mhz)
+			max_phyclk_mhz = (int)dcn3_03_soc.clock_limits[0].phyclk_mhz;
+
+		if (max_dcfclk_mhz > dcfclk_sta_targets[num_dcfclk_sta_targets-1]) {
+			dcfclk_sta_targets[num_dcfclk_sta_targets] = max_dcfclk_mhz;
+			num_dcfclk_sta_targets++;
+		} else if (max_dcfclk_mhz < dcfclk_sta_targets[num_dcfclk_sta_targets-1]) {
+			for (i = 0; i < num_dcfclk_sta_targets; i++) {
+				if (dcfclk_sta_targets[i] > max_dcfclk_mhz) {
+					dcfclk_sta_targets[i] = max_dcfclk_mhz;
+					break;
+				}
+			}
+			/* Update size of array since we "removed" duplicates */
+			num_dcfclk_sta_targets = i + 1;
+		}
+
+		num_uclk_states = bw_params->clk_table.num_entries;
+
+		/* Calculate optimal dcfclk for each uclk */
+		for (i = 0; i < num_uclk_states; i++) {
+			dcn303_get_optimal_dcfclk_fclk_for_uclk(bw_params->clk_table.entries[i].memclk_mhz * 16,
+					&optimal_dcfclk_for_uclk[i], NULL);
+			if (optimal_dcfclk_for_uclk[i] < bw_params->clk_table.entries[0].dcfclk_mhz)
+				optimal_dcfclk_for_uclk[i] = bw_params->clk_table.entries[0].dcfclk_mhz;
+		}
+
+		/* Calculate optimal uclk for each dcfclk sta target */
+		for (i = 0; i < num_dcfclk_sta_targets; i++) {
+			for (j = 0; j < num_uclk_states; j++) {
+				if (dcfclk_sta_targets[i] < optimal_dcfclk_for_uclk[j]) {
+					optimal_uclk_for_dcfclk_sta_targets[i] =
+							bw_params->clk_table.entries[j].memclk_mhz * 16;
+					break;
+				} else {
+					/* condition where (dcfclk_sta_targets[i] >= optimal_dcfclk_for_uclk[j]):
+					 * This is required for dcn303 because it just so happens that the memory
+					 * bandwidth is low enough such that all the optimal DCFCLK for each UCLK
+					 * is lower than the smallest DCFCLK STA target. In this case we need to
+					 * populate the optimal UCLK for each DCFCLK STA target to be the max UCLK.
+					 */
+					if (j == num_uclk_states - 1) {
+						optimal_uclk_for_dcfclk_sta_targets[i] =
+								bw_params->clk_table.entries[j].memclk_mhz * 16;
+					}
+				}
+			}
+		}
+
+		i = 0;
+		j = 0;
+		/* create the final dcfclk and uclk table */
+		while (i < num_dcfclk_sta_targets && j < num_uclk_states && num_states < DC__VOLTAGE_STATES) {
+			if (dcfclk_sta_targets[i] < optimal_dcfclk_for_uclk[j]) {
+				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) {
+					dcfclk_mhz[num_states] = optimal_dcfclk_for_uclk[j];
+					dram_speed_mts[num_states++] =
+							bw_params->clk_table.entries[j++].memclk_mhz * 16;
+				} else {
+					j = num_uclk_states;
+				}
+			}
+		}
+
+		while (i < num_dcfclk_sta_targets && num_states < DC__VOLTAGE_STATES) {
+			dcfclk_mhz[num_states] = dcfclk_sta_targets[i];
+			dram_speed_mts[num_states++] = optimal_uclk_for_dcfclk_sta_targets[i++];
+		}
+
+		while (j < num_uclk_states && num_states < DC__VOLTAGE_STATES &&
+				optimal_dcfclk_for_uclk[j] <= 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;
+		}
+
+		/* bw_params->clk_table.entries[MAX_NUM_DPM_LVL].
+		 * MAX_NUM_DPM_LVL is 8.
+		 * dcn3_02_soc.clock_limits[DC__VOLTAGE_STATES].
+		 * DC__VOLTAGE_STATES is 40.
+		 */
+		if (num_states > MAX_NUM_DPM_LVL) {
+			ASSERT(0);
+			return;
+		}
+
+		dcn3_03_soc.num_states = num_states;
+		for (i = 0; i < dcn3_03_soc.num_states; i++) {
+			dcn3_03_soc.clock_limits[i].state = i;
+			dcn3_03_soc.clock_limits[i].dcfclk_mhz = dcfclk_mhz[i];
+			dcn3_03_soc.clock_limits[i].fabricclk_mhz = dcfclk_mhz[i];
+			dcn3_03_soc.clock_limits[i].dram_speed_mts = dram_speed_mts[i];
+
+			/* Fill all states with max values of all other clocks */
+			dcn3_03_soc.clock_limits[i].dispclk_mhz = max_dispclk_mhz;
+			dcn3_03_soc.clock_limits[i].dppclk_mhz  = max_dppclk_mhz;
+			dcn3_03_soc.clock_limits[i].phyclk_mhz  = max_phyclk_mhz;
+			/* Populate from bw_params for DTBCLK, SOCCLK */
+			if (!bw_params->clk_table.entries[i].dtbclk_mhz && i > 0)
+				dcn3_03_soc.clock_limits[i].dtbclk_mhz = dcn3_03_soc.clock_limits[i-1].dtbclk_mhz;
+			else
+				dcn3_03_soc.clock_limits[i].dtbclk_mhz = bw_params->clk_table.entries[i].dtbclk_mhz;
+			if (!bw_params->clk_table.entries[i].socclk_mhz && i > 0)
+				dcn3_03_soc.clock_limits[i].socclk_mhz = dcn3_03_soc.clock_limits[i-1].socclk_mhz;
+			else
+				dcn3_03_soc.clock_limits[i].socclk_mhz = bw_params->clk_table.entries[i].socclk_mhz;
+			/* These clocks cannot come from bw_params, always fill from dcn3_03_soc[1] */
+			/* FCLK, PHYCLK_D18, DSCCLK */
+			dcn3_03_soc.clock_limits[i].phyclk_d18_mhz = dcn3_03_soc.clock_limits[0].phyclk_d18_mhz;
+			dcn3_03_soc.clock_limits[i].dscclk_mhz = dcn3_03_soc.clock_limits[0].dscclk_mhz;
+		}
+
+		if (dcn3_03_soc.num_chans <= 4) {
+			for (i = 0; i < dcn3_03_soc.num_states; i++) {
+				if (dcn3_03_soc.clock_limits[i].dram_speed_mts > 1700)
+					break;
+
+				if (dcn3_03_soc.clock_limits[i].dram_speed_mts >= 1500) {
+					dcn3_03_soc.clock_limits[i].dcfclk_mhz = 100;
+					dcn3_03_soc.clock_limits[i].fabricclk_mhz = 100;
+				}
+			}
+		}
+
+		/* re-init DML with updated bb */
+		dml_init_instance(&dc->dml, &dcn3_03_soc, &dcn3_03_ip, DML_PROJECT_DCN30);
+		if (dc->current_state)
+			dml_init_instance(&dc->current_state->bw_ctx.dml, &dcn3_03_soc, &dcn3_03_ip, DML_PROJECT_DCN30);
+	}
+}
+
+void dcn303_fpu_init_soc_bounding_box(struct bp_soc_bb_info bb_info)
+{
+	dc_assert_fp_enabled();
+
+	if (bb_info.dram_clock_change_latency_100ns > 0)
+		dcn3_03_soc.dram_clock_change_latency_us = bb_info.dram_clock_change_latency_100ns * 10;
+
+	if (bb_info.dram_sr_enter_exit_latency_100ns > 0)
+		dcn3_03_soc.sr_enter_plus_exit_time_us = bb_info.dram_sr_enter_exit_latency_100ns * 10;
+
+	if (bb_info.dram_sr_exit_latency_100ns > 0)
+		dcn3_03_soc.sr_exit_time_us = bb_info.dram_sr_exit_latency_100ns * 10;
+}

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dcn303/dcn303_fpu.h

@@ -0,0 +1,32 @@
+/*
+ * Copyright 2019-2021 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 __DCN303_FPU_H__
+#define __DCN303_FPU_H__
+
+void dcn303_fpu_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params);
+void dcn303_fpu_init_soc_bounding_box(struct bp_soc_bb_info bb_info);
+
+#endif /* __DCN303_FPU_H__*/

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.c

@@ -0,0 +1,830 @@
+/*
+ * Copyright 2019-2021 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 "dcn31/dcn31_resource.h"
+#include "dcn315/dcn315_resource.h"
+#include "dcn316/dcn316_resource.h"
+
+#include "dml/dcn20/dcn20_fpu.h"
+#include "dcn31_fpu.h"
+
+/**
+ * DOC: DCN31x FPU manipulation Overview
+ *
+ * The DCN architecture relies on FPU operations, which require special
+ * compilation flags and the use of kernel_fpu_begin/end functions; ideally, we
+ * want to avoid spreading FPU access across multiple files. With this idea in
+ * mind, this file aims to centralize all DCN3.1.x functions that require FPU
+ * access in a single place. Code in this file follows the following code
+ * pattern:
+ *
+ * 1. Functions that use FPU operations should be isolated in static functions.
+ * 2. The FPU functions should have the noinline attribute to ensure anything
+ *    that deals with FP register is contained within this call.
+ * 3. All function that needs to be accessed outside this file requires a
+ *    public interface that not uses any FPU reference.
+ * 4. Developers **must not** use DC_FP_START/END in this file, but they need
+ *    to ensure that the caller invokes it before access any function available
+ *    in this file. For this reason, public functions in this file must invoke
+ *    dc_assert_fp_enabled();
+ */
+
+struct _vcs_dpi_ip_params_st dcn3_1_ip = {
+	.gpuvm_enable = 1,
+	.gpuvm_max_page_table_levels = 1,
+	.hostvm_enable = 1,
+	.hostvm_max_page_table_levels = 2,
+	.rob_buffer_size_kbytes = 64,
+	.det_buffer_size_kbytes = DCN3_1_DEFAULT_DET_SIZE,
+	.config_return_buffer_size_in_kbytes = 1792,
+	.compressed_buffer_segment_size_in_kbytes = 64,
+	.meta_fifo_size_in_kentries = 32,
+	.zero_size_buffer_entries = 512,
+	.compbuf_reserved_space_64b = 256,
+	.compbuf_reserved_space_zs = 64,
+	.dpp_output_buffer_pixels = 2560,
+	.opp_output_buffer_lines = 1,
+	.pixel_chunk_size_kbytes = 8,
+	.meta_chunk_size_kbytes = 2,
+	.min_meta_chunk_size_bytes = 256,
+	.writeback_chunk_size_kbytes = 8,
+	.ptoi_supported = false,
+	.num_dsc = 3,
+	.maximum_dsc_bits_per_component = 10,
+	.dsc422_native_support = false,
+	.is_line_buffer_bpp_fixed = true,
+	.line_buffer_fixed_bpp = 48,
+	.line_buffer_size_bits = 789504,
+	.max_line_buffer_lines = 12,
+	.writeback_interface_buffer_size_kbytes = 90,
+	.max_num_dpp = 4,
+	.max_num_otg = 4,
+	.max_num_hdmi_frl_outputs = 1,
+	.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,
+	.max_hscl_taps = 8,
+	.max_vscl_taps = 8,
+	.dpte_buffer_size_in_pte_reqs_luma = 64,
+	.dpte_buffer_size_in_pte_reqs_chroma = 34,
+	.dispclk_ramp_margin_percent = 1,
+	.max_inter_dcn_tile_repeaters = 8,
+	.cursor_buffer_size = 16,
+	.cursor_chunk_size = 2,
+	.writeback_line_buffer_buffer_size = 0,
+	.writeback_min_hscl_ratio = 1,
+	.writeback_min_vscl_ratio = 1,
+	.writeback_max_hscl_ratio = 1,
+	.writeback_max_vscl_ratio = 1,
+	.writeback_max_hscl_taps = 1,
+	.writeback_max_vscl_taps = 1,
+	.dppclk_delay_subtotal = 46,
+	.dppclk_delay_scl = 50,
+	.dppclk_delay_scl_lb_only = 16,
+	.dppclk_delay_cnvc_formatter = 27,
+	.dppclk_delay_cnvc_cursor = 6,
+	.dispclk_delay_subtotal = 119,
+	.dynamic_metadata_vm_enabled = false,
+	.odm_combine_4to1_supported = false,
+	.dcc_supported = true,
+};
+
+static struct _vcs_dpi_soc_bounding_box_st dcn3_1_soc = {
+		/*TODO: correct dispclk/dppclk voltage level determination*/
+	.clock_limits = {
+		{
+			.state = 0,
+			.dispclk_mhz = 1200.0,
+			.dppclk_mhz = 1200.0,
+			.phyclk_mhz = 600.0,
+			.phyclk_d18_mhz = 667.0,
+			.dscclk_mhz = 186.0,
+			.dtbclk_mhz = 625.0,
+		},
+		{
+			.state = 1,
+			.dispclk_mhz = 1200.0,
+			.dppclk_mhz = 1200.0,
+			.phyclk_mhz = 810.0,
+			.phyclk_d18_mhz = 667.0,
+			.dscclk_mhz = 209.0,
+			.dtbclk_mhz = 625.0,
+		},
+		{
+			.state = 2,
+			.dispclk_mhz = 1200.0,
+			.dppclk_mhz = 1200.0,
+			.phyclk_mhz = 810.0,
+			.phyclk_d18_mhz = 667.0,
+			.dscclk_mhz = 209.0,
+			.dtbclk_mhz = 625.0,
+		},
+		{
+			.state = 3,
+			.dispclk_mhz = 1200.0,
+			.dppclk_mhz = 1200.0,
+			.phyclk_mhz = 810.0,
+			.phyclk_d18_mhz = 667.0,
+			.dscclk_mhz = 371.0,
+			.dtbclk_mhz = 625.0,
+		},
+		{
+			.state = 4,
+			.dispclk_mhz = 1200.0,
+			.dppclk_mhz = 1200.0,
+			.phyclk_mhz = 810.0,
+			.phyclk_d18_mhz = 667.0,
+			.dscclk_mhz = 417.0,
+			.dtbclk_mhz = 625.0,
+		},
+	},
+	.num_states = 5,
+	.sr_exit_time_us = 9.0,
+	.sr_enter_plus_exit_time_us = 11.0,
+	.sr_exit_z8_time_us = 442.0,
+	.sr_enter_plus_exit_z8_time_us = 560.0,
+	.writeback_latency_us = 12.0,
+	.dram_channel_width_bytes = 4,
+	.round_trip_ping_latency_dcfclk_cycles = 106,
+	.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_sdp_bw_after_urgent = 80.0,
+	.pct_ideal_dram_sdp_bw_after_urgent_pixel_only = 65.0,
+	.pct_ideal_dram_sdp_bw_after_urgent_pixel_and_vm = 60.0,
+	.pct_ideal_dram_sdp_bw_after_urgent_vm_only = 30.0,
+	.max_avg_sdp_bw_use_normal_percent = 60.0,
+	.max_avg_dram_bw_use_normal_percent = 60.0,
+	.fabric_datapath_to_dcn_data_return_bytes = 32,
+	.return_bus_width_bytes = 64,
+	.downspread_percent = 0.38,
+	.dcn_downspread_percent = 0.5,
+	.gpuvm_min_page_size_bytes = 4096,
+	.hostvm_min_page_size_bytes = 4096,
+	.do_urgent_latency_adjustment = false,
+	.urgent_latency_adjustment_fabric_clock_component_us = 0,
+	.urgent_latency_adjustment_fabric_clock_reference_mhz = 0,
+};
+
+struct _vcs_dpi_ip_params_st dcn3_15_ip = {
+	.gpuvm_enable = 1,
+	.gpuvm_max_page_table_levels = 1,
+	.hostvm_enable = 1,
+	.hostvm_max_page_table_levels = 2,
+	.rob_buffer_size_kbytes = 64,
+	.det_buffer_size_kbytes = DCN3_15_DEFAULT_DET_SIZE,
+	.min_comp_buffer_size_kbytes = 64,
+	.config_return_buffer_size_in_kbytes = 1024,
+	.compressed_buffer_segment_size_in_kbytes = 64,
+	.meta_fifo_size_in_kentries = 32,
+	.zero_size_buffer_entries = 512,
+	.compbuf_reserved_space_64b = 256,
+	.compbuf_reserved_space_zs = 64,
+	.dpp_output_buffer_pixels = 2560,
+	.opp_output_buffer_lines = 1,
+	.pixel_chunk_size_kbytes = 8,
+	.meta_chunk_size_kbytes = 2,
+	.min_meta_chunk_size_bytes = 256,
+	.writeback_chunk_size_kbytes = 8,
+	.ptoi_supported = false,
+	.num_dsc = 3,
+	.maximum_dsc_bits_per_component = 10,
+	.dsc422_native_support = false,
+	.is_line_buffer_bpp_fixed = true,
+	.line_buffer_fixed_bpp = 48,
+	.line_buffer_size_bits = 789504,
+	.max_line_buffer_lines = 12,
+	.writeback_interface_buffer_size_kbytes = 90,
+	.max_num_dpp = 4,
+	.max_num_otg = 4,
+	.max_num_hdmi_frl_outputs = 1,
+	.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,
+	.max_hscl_taps = 8,
+	.max_vscl_taps = 8,
+	.dpte_buffer_size_in_pte_reqs_luma = 64,
+	.dpte_buffer_size_in_pte_reqs_chroma = 34,
+	.dispclk_ramp_margin_percent = 1,
+	.max_inter_dcn_tile_repeaters = 9,
+	.cursor_buffer_size = 16,
+	.cursor_chunk_size = 2,
+	.writeback_line_buffer_buffer_size = 0,
+	.writeback_min_hscl_ratio = 1,
+	.writeback_min_vscl_ratio = 1,
+	.writeback_max_hscl_ratio = 1,
+	.writeback_max_vscl_ratio = 1,
+	.writeback_max_hscl_taps = 1,
+	.writeback_max_vscl_taps = 1,
+	.dppclk_delay_subtotal = 46,
+	.dppclk_delay_scl = 50,
+	.dppclk_delay_scl_lb_only = 16,
+	.dppclk_delay_cnvc_formatter = 27,
+	.dppclk_delay_cnvc_cursor = 6,
+	.dispclk_delay_subtotal = 119,
+	.dynamic_metadata_vm_enabled = false,
+	.odm_combine_4to1_supported = false,
+	.dcc_supported = true,
+};
+
+static struct _vcs_dpi_soc_bounding_box_st dcn3_15_soc = {
+	.sr_exit_time_us = 9.0,
+	.sr_enter_plus_exit_time_us = 11.0,
+	.sr_exit_z8_time_us = 50.0,
+	.sr_enter_plus_exit_z8_time_us = 50.0,
+	.writeback_latency_us = 12.0,
+	.dram_channel_width_bytes = 4,
+	.round_trip_ping_latency_dcfclk_cycles = 106,
+	.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_sdp_bw_after_urgent = 80.0,
+	.pct_ideal_dram_sdp_bw_after_urgent_pixel_only = 65.0,
+	.pct_ideal_dram_sdp_bw_after_urgent_pixel_and_vm = 60.0,
+	.pct_ideal_dram_sdp_bw_after_urgent_vm_only = 30.0,
+	.max_avg_sdp_bw_use_normal_percent = 60.0,
+	.max_avg_dram_bw_use_normal_percent = 60.0,
+	.fabric_datapath_to_dcn_data_return_bytes = 32,
+	.return_bus_width_bytes = 64,
+	.downspread_percent = 0.38,
+	.dcn_downspread_percent = 0.38,
+	.gpuvm_min_page_size_bytes = 4096,
+	.hostvm_min_page_size_bytes = 4096,
+	.do_urgent_latency_adjustment = false,
+	.urgent_latency_adjustment_fabric_clock_component_us = 0,
+	.urgent_latency_adjustment_fabric_clock_reference_mhz = 0,
+	.dispclk_dppclk_vco_speed_mhz = 2400.0,
+	.num_chans = 4,
+	.dummy_pstate_latency_us = 10.0
+};
+
+struct _vcs_dpi_ip_params_st dcn3_16_ip = {
+	.gpuvm_enable = 1,
+	.gpuvm_max_page_table_levels = 1,
+	.hostvm_enable = 1,
+	.hostvm_max_page_table_levels = 2,
+	.rob_buffer_size_kbytes = 64,
+	.det_buffer_size_kbytes = DCN3_16_DEFAULT_DET_SIZE,
+	.min_comp_buffer_size_kbytes = 64,
+	.config_return_buffer_size_in_kbytes = 1024,
+	.compressed_buffer_segment_size_in_kbytes = 64,
+	.meta_fifo_size_in_kentries = 32,
+	.zero_size_buffer_entries = 512,
+	.compbuf_reserved_space_64b = 256,
+	.compbuf_reserved_space_zs = 64,
+	.dpp_output_buffer_pixels = 2560,
+	.opp_output_buffer_lines = 1,
+	.pixel_chunk_size_kbytes = 8,
+	.meta_chunk_size_kbytes = 2,
+	.min_meta_chunk_size_bytes = 256,
+	.writeback_chunk_size_kbytes = 8,
+	.ptoi_supported = false,
+	.num_dsc = 3,
+	.maximum_dsc_bits_per_component = 10,
+	.dsc422_native_support = false,
+	.is_line_buffer_bpp_fixed = true,
+	.line_buffer_fixed_bpp = 48,
+	.line_buffer_size_bits = 789504,
+	.max_line_buffer_lines = 12,
+	.writeback_interface_buffer_size_kbytes = 90,
+	.max_num_dpp = 4,
+	.max_num_otg = 4,
+	.max_num_hdmi_frl_outputs = 1,
+	.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,
+	.max_hscl_taps = 8,
+	.max_vscl_taps = 8,
+	.dpte_buffer_size_in_pte_reqs_luma = 64,
+	.dpte_buffer_size_in_pte_reqs_chroma = 34,
+	.dispclk_ramp_margin_percent = 1,
+	.max_inter_dcn_tile_repeaters = 8,
+	.cursor_buffer_size = 16,
+	.cursor_chunk_size = 2,
+	.writeback_line_buffer_buffer_size = 0,
+	.writeback_min_hscl_ratio = 1,
+	.writeback_min_vscl_ratio = 1,
+	.writeback_max_hscl_ratio = 1,
+	.writeback_max_vscl_ratio = 1,
+	.writeback_max_hscl_taps = 1,
+	.writeback_max_vscl_taps = 1,
+	.dppclk_delay_subtotal = 46,
+	.dppclk_delay_scl = 50,
+	.dppclk_delay_scl_lb_only = 16,
+	.dppclk_delay_cnvc_formatter = 27,
+	.dppclk_delay_cnvc_cursor = 6,
+	.dispclk_delay_subtotal = 119,
+	.dynamic_metadata_vm_enabled = false,
+	.odm_combine_4to1_supported = false,
+	.dcc_supported = true,
+};
+
+static struct _vcs_dpi_soc_bounding_box_st dcn3_16_soc = {
+		/*TODO: correct dispclk/dppclk voltage level determination*/
+	.clock_limits = {
+		{
+			.state = 0,
+			.dispclk_mhz = 556.0,
+			.dppclk_mhz = 556.0,
+			.phyclk_mhz = 600.0,
+			.phyclk_d18_mhz = 445.0,
+			.dscclk_mhz = 186.0,
+			.dtbclk_mhz = 625.0,
+		},
+		{
+			.state = 1,
+			.dispclk_mhz = 625.0,
+			.dppclk_mhz = 625.0,
+			.phyclk_mhz = 810.0,
+			.phyclk_d18_mhz = 667.0,
+			.dscclk_mhz = 209.0,
+			.dtbclk_mhz = 625.0,
+		},
+		{
+			.state = 2,
+			.dispclk_mhz = 625.0,
+			.dppclk_mhz = 625.0,
+			.phyclk_mhz = 810.0,
+			.phyclk_d18_mhz = 667.0,
+			.dscclk_mhz = 209.0,
+			.dtbclk_mhz = 625.0,
+		},
+		{
+			.state = 3,
+			.dispclk_mhz = 1112.0,
+			.dppclk_mhz = 1112.0,
+			.phyclk_mhz = 810.0,
+			.phyclk_d18_mhz = 667.0,
+			.dscclk_mhz = 371.0,
+			.dtbclk_mhz = 625.0,
+		},
+		{
+			.state = 4,
+			.dispclk_mhz = 1250.0,
+			.dppclk_mhz = 1250.0,
+			.phyclk_mhz = 810.0,
+			.phyclk_d18_mhz = 667.0,
+			.dscclk_mhz = 417.0,
+			.dtbclk_mhz = 625.0,
+		},
+	},
+	.num_states = 5,
+	.sr_exit_time_us = 9.0,
+	.sr_enter_plus_exit_time_us = 11.0,
+	.sr_exit_z8_time_us = 442.0,
+	.sr_enter_plus_exit_z8_time_us = 560.0,
+	.writeback_latency_us = 12.0,
+	.dram_channel_width_bytes = 4,
+	.round_trip_ping_latency_dcfclk_cycles = 106,
+	.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_sdp_bw_after_urgent = 80.0,
+	.pct_ideal_dram_sdp_bw_after_urgent_pixel_only = 65.0,
+	.pct_ideal_dram_sdp_bw_after_urgent_pixel_and_vm = 60.0,
+	.pct_ideal_dram_sdp_bw_after_urgent_vm_only = 30.0,
+	.max_avg_sdp_bw_use_normal_percent = 60.0,
+	.max_avg_dram_bw_use_normal_percent = 60.0,
+	.fabric_datapath_to_dcn_data_return_bytes = 32,
+	.return_bus_width_bytes = 64,
+	.downspread_percent = 0.38,
+	.dcn_downspread_percent = 0.5,
+	.gpuvm_min_page_size_bytes = 4096,
+	.hostvm_min_page_size_bytes = 4096,
+	.do_urgent_latency_adjustment = false,
+	.urgent_latency_adjustment_fabric_clock_component_us = 0,
+	.urgent_latency_adjustment_fabric_clock_reference_mhz = 0,
+	.dispclk_dppclk_vco_speed_mhz = 2500.0,
+};
+
+void dcn31_zero_pipe_dcc_fraction(display_e2e_pipe_params_st *pipes,
+				  int pipe_cnt)
+{
+	dc_assert_fp_enabled();
+
+	pipes[pipe_cnt].pipe.src.dcc_fraction_of_zs_req_luma = 0;
+	pipes[pipe_cnt].pipe.src.dcc_fraction_of_zs_req_chroma = 0;
+}
+
+void dcn31_update_soc_for_wm_a(struct dc *dc, struct dc_state *context)
+{
+	dc_assert_fp_enabled();
+
+	if (dc->clk_mgr->bw_params->wm_table.entries[WM_A].valid) {
+		context->bw_ctx.dml.soc.dram_clock_change_latency_us = dc->clk_mgr->bw_params->wm_table.entries[WM_A].pstate_latency_us;
+		context->bw_ctx.dml.soc.sr_enter_plus_exit_time_us = dc->clk_mgr->bw_params->wm_table.entries[WM_A].sr_enter_plus_exit_time_us;
+		context->bw_ctx.dml.soc.sr_exit_time_us = dc->clk_mgr->bw_params->wm_table.entries[WM_A].sr_exit_time_us;
+	}
+}
+
+void dcn315_update_soc_for_wm_a(struct dc *dc, struct dc_state *context)
+{
+	dc_assert_fp_enabled();
+
+	if (dc->clk_mgr->bw_params->wm_table.entries[WM_A].valid) {
+		/* For 315 pstate change is only supported if possible in vactive */
+		if (context->bw_ctx.dml.vba.DRAMClockChangeSupport[context->bw_ctx.dml.vba.VoltageLevel][context->bw_ctx.dml.vba.maxMpcComb] != dm_dram_clock_change_vactive)
+			context->bw_ctx.dml.soc.dram_clock_change_latency_us = context->bw_ctx.dml.soc.dummy_pstate_latency_us;
+		else
+			context->bw_ctx.dml.soc.dram_clock_change_latency_us = dc->clk_mgr->bw_params->wm_table.entries[WM_A].pstate_latency_us;
+		context->bw_ctx.dml.soc.sr_enter_plus_exit_time_us =
+				dc->clk_mgr->bw_params->wm_table.entries[WM_A].sr_enter_plus_exit_time_us;
+		context->bw_ctx.dml.soc.sr_exit_time_us =
+				dc->clk_mgr->bw_params->wm_table.entries[WM_A].sr_exit_time_us;
+	}
+}
+
+void dcn31_calculate_wm_and_dlg_fp(
+		struct dc *dc, struct dc_state *context,
+		display_e2e_pipe_params_st *pipes,
+		int pipe_cnt,
+		int vlevel)
+{
+	int total_det = 0, active_hubp_count = 0;
+	unsigned int i, pipe_idx;
+	double dcfclk = context->bw_ctx.dml.vba.DCFCLKState[vlevel][context->bw_ctx.dml.vba.maxMpcComb];
+	uint32_t cstate_enter_plus_exit_z8_ns;
+	uint32_t minimum_z8_residency_time_ns =
+			(uint32_t)dc->debug.minimum_z8_residency_time * 1000U;
+
+	dc_assert_fp_enabled();
+
+	if (context->bw_ctx.dml.soc.min_dcfclk > dcfclk)
+		dcfclk = context->bw_ctx.dml.soc.min_dcfclk;
+
+	/* We don't recalculate clocks for 0 pipe configs, which can block
+	 * S0i3 as high clocks will block low power states
+	 * Override any clocks that can block S0i3 to min here
+	 */
+	if (pipe_cnt == 0) {
+		context->bw_ctx.bw.dcn.clk.dcfclk_khz = (int)dcfclk; // always should be vlevel 0
+		return;
+	}
+
+	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;
+
+	cstate_enter_plus_exit_z8_ns =
+		(uint32_t)(get_wm_z8_stutter_enter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000);
+
+	if (get_stutter_period(&context->bw_ctx.dml, pipes, pipe_cnt) < dc->debug.minimum_z8_residency_time &&
+				cstate_enter_plus_exit_z8_ns < minimum_z8_residency_time_ns)
+		cstate_enter_plus_exit_z8_ns = minimum_z8_residency_time_ns;
+
+	/* Set A:
+	 * All clocks min required
+	 *
+	 * 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 = (uint32_t)(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 = (uint32_t)(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 = (uint32_t)(get_wm_stutter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000);
+	context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.pstate_change_ns = (uint32_t)(get_wm_dram_clock_change(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000);
+	context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.cstate_enter_plus_exit_z8_ns = cstate_enter_plus_exit_z8_ns;
+	context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.cstate_exit_z8_ns = (uint32_t)(get_wm_z8_stutter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000);
+	context->bw_ctx.bw.dcn.watermarks.a.pte_meta_urgent_ns = (uint32_t)(get_wm_memory_trip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000);
+	context->bw_ctx.bw.dcn.watermarks.a.frac_urg_bw_nom = (uint32_t)(get_fraction_of_urgent_bandwidth(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000);
+	context->bw_ctx.bw.dcn.watermarks.a.frac_urg_bw_flip = (uint32_t)(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 = (uint32_t)(get_urgent_latency(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000);
+	context->bw_ctx.bw.dcn.watermarks.b = context->bw_ctx.bw.dcn.watermarks.a;
+	context->bw_ctx.bw.dcn.watermarks.c = context->bw_ctx.bw.dcn.watermarks.a;
+	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;
+
+		if (context->res_ctx.pipe_ctx[i].plane_state)
+			active_hubp_count++;
+
+		pipes[pipe_idx].clks_cfg.dispclk_mhz = get_dispclk_calculated(&context->bw_ctx.dml, pipes, pipe_cnt);
+		pipes[pipe_idx].clks_cfg.dppclk_mhz = get_dppclk_calculated(&context->bw_ctx.dml, pipes, pipe_cnt, pipe_idx);
+
+		if (dc->config.forced_clocks || dc->debug.max_disp_clk) {
+			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++;
+	}
+
+	dcn20_calculate_dlg_params(dc, context, pipes, pipe_cnt, vlevel);
+	/* For 31x apu pstate change is only supported if possible in vactive*/
+	context->bw_ctx.bw.dcn.clk.p_state_change_support =
+			context->bw_ctx.dml.vba.DRAMClockChangeSupport[vlevel][context->bw_ctx.dml.vba.maxMpcComb] == dm_dram_clock_change_vactive;
+	/* If DCN isn't making memory requests we can allow pstate change and lower clocks */
+	if (!active_hubp_count) {
+		context->bw_ctx.bw.dcn.clk.socclk_khz = 0;
+		context->bw_ctx.bw.dcn.clk.dppclk_khz = 0;
+		context->bw_ctx.bw.dcn.clk.dcfclk_khz = 0;
+		context->bw_ctx.bw.dcn.clk.dcfclk_deep_sleep_khz = 0;
+		context->bw_ctx.bw.dcn.clk.dramclk_khz = 0;
+		context->bw_ctx.bw.dcn.clk.fclk_khz = 0;
+		context->bw_ctx.bw.dcn.clk.p_state_change_support = true;
+		for (i = 0; i < dc->res_pool->pipe_count; i++)
+			if (context->res_ctx.pipe_ctx[i].stream)
+				context->res_ctx.pipe_ctx[i].plane_res.bw.dppclk_khz = 0;
+	}
+	for (i = 0, pipe_idx = 0; i < dc->res_pool->pipe_count; i++) {
+		if (!context->res_ctx.pipe_ctx[i].stream)
+			continue;
+
+		context->res_ctx.pipe_ctx[i].det_buffer_size_kb =
+				(int)get_det_buffer_size_kbytes(&context->bw_ctx.dml, pipes, pipe_cnt, pipe_idx);
+		if (context->res_ctx.pipe_ctx[i].det_buffer_size_kb > 384)
+			context->res_ctx.pipe_ctx[i].det_buffer_size_kb /= 2;
+		total_det += context->res_ctx.pipe_ctx[i].det_buffer_size_kb;
+		pipe_idx++;
+	}
+	context->bw_ctx.bw.dcn.compbuf_size_kb = context->bw_ctx.dml.ip.config_return_buffer_size_in_kbytes - total_det;
+}
+
+void dcn31_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_params)
+{
+	struct _vcs_dpi_voltage_scaling_st *s = dc->scratch.update_bw_bounding_box.clock_limits;
+	struct clk_limit_table *clk_table = &bw_params->clk_table;
+	unsigned int i, closest_clk_lvl;
+	unsigned int max_dispclk_mhz = 0, max_dppclk_mhz = 0;
+	int j;
+
+	dc_assert_fp_enabled();
+
+	memcpy(s, dcn3_1_soc.clock_limits, sizeof(dcn3_1_soc.clock_limits));
+
+	// Default clock levels are used for diags, which may lead to overclocking.
+	dcn3_1_ip.max_num_otg = dc->res_pool->res_cap->num_timing_generator;
+	dcn3_1_ip.max_num_dpp = dc->res_pool->pipe_count;
+	dcn3_1_soc.num_chans = bw_params->num_channels;
+
+	ASSERT(clk_table->num_entries);
+
+	/* Prepass to find max clocks independent of voltage level. */
+	for (i = 0; i < clk_table->num_entries; ++i) {
+		if (clk_table->entries[i].dispclk_mhz > max_dispclk_mhz)
+			max_dispclk_mhz = clk_table->entries[i].dispclk_mhz;
+		if (clk_table->entries[i].dppclk_mhz > max_dppclk_mhz)
+			max_dppclk_mhz = clk_table->entries[i].dppclk_mhz;
+	}
+
+	for (i = 0; i < clk_table->num_entries; i++) {
+		/* loop backwards*/
+		for (closest_clk_lvl = 0, j = dcn3_1_soc.num_states - 1; j >= 0; j--) {
+			if ((unsigned int) dcn3_1_soc.clock_limits[j].dcfclk_mhz <= clk_table->entries[i].dcfclk_mhz) {
+				closest_clk_lvl = j;
+				break;
+			}
+		}
+
+		s[i].state = i;
+
+		/* Clocks dependent on voltage level. */
+		s[i].dcfclk_mhz = clk_table->entries[i].dcfclk_mhz;
+		s[i].fabricclk_mhz = clk_table->entries[i].fclk_mhz;
+		s[i].socclk_mhz = clk_table->entries[i].socclk_mhz;
+		s[i].dram_speed_mts = clk_table->entries[i].memclk_mhz *
+			2 * clk_table->entries[i].wck_ratio;
+
+		/* Clocks independent of voltage level. */
+		s[i].dispclk_mhz = max_dispclk_mhz ? max_dispclk_mhz :
+			dcn3_1_soc.clock_limits[closest_clk_lvl].dispclk_mhz;
+
+		s[i].dppclk_mhz = max_dppclk_mhz ? max_dppclk_mhz :
+			dcn3_1_soc.clock_limits[closest_clk_lvl].dppclk_mhz;
+
+		s[i].dram_bw_per_chan_gbps =
+			dcn3_1_soc.clock_limits[closest_clk_lvl].dram_bw_per_chan_gbps;
+		s[i].dscclk_mhz = dcn3_1_soc.clock_limits[closest_clk_lvl].dscclk_mhz;
+		s[i].dtbclk_mhz = dcn3_1_soc.clock_limits[closest_clk_lvl].dtbclk_mhz;
+		s[i].phyclk_d18_mhz =
+			dcn3_1_soc.clock_limits[closest_clk_lvl].phyclk_d18_mhz;
+		s[i].phyclk_mhz = dcn3_1_soc.clock_limits[closest_clk_lvl].phyclk_mhz;
+	}
+
+	if (clk_table->num_entries)
+		dcn3_1_soc.num_states = clk_table->num_entries;
+
+	memcpy(dcn3_1_soc.clock_limits, s, sizeof(dcn3_1_soc.clock_limits));
+
+	dcn3_1_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;
+
+	if ((int)(dcn3_1_soc.dram_clock_change_latency_us * 1000)
+				!= dc->debug.dram_clock_change_latency_ns
+			&& dc->debug.dram_clock_change_latency_ns) {
+		dcn3_1_soc.dram_clock_change_latency_us = dc->debug.dram_clock_change_latency_ns / 1000;
+	}
+
+	dml_init_instance(&dc->dml, &dcn3_1_soc, &dcn3_1_ip, DML_PROJECT_DCN31);
+}
+
+void dcn315_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_params)
+{
+	struct clk_limit_table *clk_table = &bw_params->clk_table;
+	unsigned int i, max_dispclk_mhz = 0, max_dppclk_mhz = 0;
+
+	dc_assert_fp_enabled();
+
+	dcn3_15_ip.max_num_otg = dc->res_pool->res_cap->num_timing_generator;
+	dcn3_15_ip.max_num_dpp = dc->res_pool->pipe_count;
+
+	if (bw_params->num_channels > 0)
+		dcn3_15_soc.num_chans = bw_params->num_channels;
+	if (bw_params->dram_channel_width_bytes > 0)
+		dcn3_15_soc.dram_channel_width_bytes = bw_params->dram_channel_width_bytes;
+
+	ASSERT(clk_table->num_entries);
+
+	/* Setup soc to always use max dispclk/dppclk to avoid odm-to-lower-voltage */
+	for (i = 0; i < clk_table->num_entries; ++i) {
+		if (clk_table->entries[i].dispclk_mhz > max_dispclk_mhz)
+			max_dispclk_mhz = clk_table->entries[i].dispclk_mhz;
+		if (clk_table->entries[i].dppclk_mhz > max_dppclk_mhz)
+			max_dppclk_mhz = clk_table->entries[i].dppclk_mhz;
+	}
+
+	for (i = 0; i < clk_table->num_entries; i++) {
+		dcn3_15_soc.clock_limits[i].state = i;
+
+		/* Clocks dependent on voltage level. */
+		dcn3_15_soc.clock_limits[i].dcfclk_mhz = clk_table->entries[i].dcfclk_mhz;
+		dcn3_15_soc.clock_limits[i].fabricclk_mhz = clk_table->entries[i].fclk_mhz;
+		dcn3_15_soc.clock_limits[i].socclk_mhz = clk_table->entries[i].socclk_mhz;
+		dcn3_15_soc.clock_limits[i].dram_speed_mts = clk_table->entries[i].memclk_mhz * 2 * clk_table->entries[i].wck_ratio;
+
+		/* These aren't actually read from smu, but rather set in clk_mgr defaults */
+		dcn3_15_soc.clock_limits[i].dtbclk_mhz = clk_table->entries[i].dtbclk_mhz;
+		dcn3_15_soc.clock_limits[i].phyclk_d18_mhz = clk_table->entries[i].phyclk_d18_mhz;
+		dcn3_15_soc.clock_limits[i].phyclk_mhz = clk_table->entries[i].phyclk_mhz;
+
+		/* Clocks independent of voltage level. */
+		dcn3_15_soc.clock_limits[i].dispclk_mhz = max_dispclk_mhz;
+		dcn3_15_soc.clock_limits[i].dppclk_mhz = max_dppclk_mhz;
+		dcn3_15_soc.clock_limits[i].dscclk_mhz = max_dispclk_mhz / 3.0;
+	}
+	dcn3_15_soc.num_states = clk_table->num_entries;
+
+
+	/* Set vco to max_dispclk * 2 to make sure the highest dispclk is always available for dml calcs,
+	 * no impact outside of dml validation
+	 */
+	dcn3_15_soc.dispclk_dppclk_vco_speed_mhz = max_dispclk_mhz * 2;
+
+	if ((int)(dcn3_15_soc.dram_clock_change_latency_us * 1000)
+				!= dc->debug.dram_clock_change_latency_ns
+			&& dc->debug.dram_clock_change_latency_ns) {
+		dcn3_15_soc.dram_clock_change_latency_us = dc->debug.dram_clock_change_latency_ns / 1000;
+	}
+
+	dml_init_instance(&dc->dml, &dcn3_15_soc, &dcn3_15_ip, DML_PROJECT_DCN315);
+}
+
+void dcn316_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_params)
+{
+	struct _vcs_dpi_voltage_scaling_st *s = dc->scratch.update_bw_bounding_box.clock_limits;
+	struct clk_limit_table *clk_table = &bw_params->clk_table;
+	unsigned int i, closest_clk_lvl;
+	unsigned int max_dispclk_mhz = 0, max_dppclk_mhz = 0;
+	int j;
+
+	dc_assert_fp_enabled();
+
+	memcpy(s, dcn3_16_soc.clock_limits, sizeof(dcn3_16_soc.clock_limits));
+
+	// Default clock levels are used for diags, which may lead to overclocking.
+	dcn3_16_ip.max_num_otg = dc->res_pool->res_cap->num_timing_generator;
+	dcn3_16_ip.max_num_dpp = dc->res_pool->pipe_count;
+	dcn3_16_soc.num_chans = bw_params->num_channels;
+
+	ASSERT(clk_table->num_entries);
+
+	/* Prepass to find max clocks independent of voltage level. */
+	for (i = 0; i < clk_table->num_entries; ++i) {
+		if (clk_table->entries[i].dispclk_mhz > max_dispclk_mhz)
+			max_dispclk_mhz = clk_table->entries[i].dispclk_mhz;
+		if (clk_table->entries[i].dppclk_mhz > max_dppclk_mhz)
+			max_dppclk_mhz = clk_table->entries[i].dppclk_mhz;
+	}
+
+	for (i = 0; i < clk_table->num_entries; i++) {
+		/* loop backwards*/
+		for (closest_clk_lvl = 0, j = dcn3_16_soc.num_states - 1; j >= 0; j--) {
+			if ((unsigned int) dcn3_16_soc.clock_limits[j].dcfclk_mhz <=
+			    clk_table->entries[i].dcfclk_mhz) {
+				closest_clk_lvl = j;
+				break;
+			}
+		}
+
+		s[i].state = i;
+
+		/* Clocks dependent on voltage level. */
+		s[i].dcfclk_mhz = clk_table->entries[i].dcfclk_mhz;
+		s[i].fabricclk_mhz = clk_table->entries[i].fclk_mhz;
+		s[i].socclk_mhz = clk_table->entries[i].socclk_mhz;
+		s[i].dram_speed_mts = clk_table->entries[i].memclk_mhz *
+			2 * clk_table->entries[i].wck_ratio;
+
+		/* Clocks independent of voltage level. */
+		s[i].dispclk_mhz = max_dispclk_mhz ? max_dispclk_mhz :
+			dcn3_16_soc.clock_limits[closest_clk_lvl].dispclk_mhz;
+
+		s[i].dppclk_mhz = max_dppclk_mhz ? max_dppclk_mhz :
+			dcn3_16_soc.clock_limits[closest_clk_lvl].dppclk_mhz;
+
+		s[i].dram_bw_per_chan_gbps =
+			dcn3_16_soc.clock_limits[closest_clk_lvl].dram_bw_per_chan_gbps;
+		s[i].dscclk_mhz = dcn3_16_soc.clock_limits[closest_clk_lvl].dscclk_mhz;
+		s[i].dtbclk_mhz = dcn3_16_soc.clock_limits[closest_clk_lvl].dtbclk_mhz;
+		s[i].phyclk_d18_mhz =
+			dcn3_16_soc.clock_limits[closest_clk_lvl].phyclk_d18_mhz;
+		s[i].phyclk_mhz = dcn3_16_soc.clock_limits[closest_clk_lvl].phyclk_mhz;
+	}
+
+	if (clk_table->num_entries)
+		dcn3_16_soc.num_states = clk_table->num_entries;
+
+	memcpy(dcn3_16_soc.clock_limits, s, sizeof(dcn3_16_soc.clock_limits));
+
+	if (max_dispclk_mhz) {
+		dcn3_16_soc.dispclk_dppclk_vco_speed_mhz = max_dispclk_mhz * 2;
+		dc->dml.soc.dispclk_dppclk_vco_speed_mhz = max_dispclk_mhz * 2;
+	}
+	if ((int)(dcn3_16_soc.dram_clock_change_latency_us * 1000)
+				!= dc->debug.dram_clock_change_latency_ns
+			&& dc->debug.dram_clock_change_latency_ns) {
+		dcn3_16_soc.dram_clock_change_latency_us = dc->debug.dram_clock_change_latency_ns / 1000;
+	}
+
+	dml_init_instance(&dc->dml, &dcn3_16_soc, &dcn3_16_ip, DML_PROJECT_DCN31);
+}
+
+int dcn_get_max_non_odm_pix_rate_100hz(struct _vcs_dpi_soc_bounding_box_st *soc)
+{
+	dc_assert_fp_enabled();
+
+	return (int)(soc->clock_limits[0].dispclk_mhz * 10000.0 / (1.0 + soc->dcn_downspread_percent / 100.0));
+}
+
+int dcn_get_approx_det_segs_required_for_pstate(
+		struct _vcs_dpi_soc_bounding_box_st *soc,
+		int pix_clk_100hz, int bpp, int seg_size_kb)
+{
+	dc_assert_fp_enabled();
+
+	/* Roughly calculate required crb to hide latency. In practice there is slightly
+	 * more buffer available for latency hiding
+	 */
+	return (int)(soc->dram_clock_change_latency_us * pix_clk_100hz * bpp
+					/ 10240000 + seg_size_kb - 1) /	seg_size_kb;
+}

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.h

@@ -0,0 +1,59 @@
+/*
+ * Copyright 2019-2021 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 __DCN31_FPU_H__
+#define __DCN31_FPU_H__
+
+#define DCN3_1_DEFAULT_DET_SIZE 384
+#define DCN3_15_DEFAULT_DET_SIZE 192
+#define DCN3_15_MIN_COMPBUF_SIZE_KB 128
+#define DCN3_16_DEFAULT_DET_SIZE 192
+#define DCN3_16_MIN_COMPBUF_SIZE_KB 128
+
+void dcn31_zero_pipe_dcc_fraction(display_e2e_pipe_params_st *pipes,
+				  int pipe_cnt);
+
+void dcn31_update_soc_for_wm_a(struct dc *dc, struct dc_state *context);
+void dcn315_update_soc_for_wm_a(struct dc *dc, struct dc_state *context);
+
+void dcn31_calculate_wm_and_dlg_fp(
+		struct dc *dc, struct dc_state *context,
+		display_e2e_pipe_params_st *pipes,
+		int pipe_cnt,
+		int vlevel);
+
+void dcn31_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_params);
+void dcn315_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_params);
+void dcn316_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_params);
+int dcn_get_max_non_odm_pix_rate_100hz(struct _vcs_dpi_soc_bounding_box_st *soc);
+int dcn_get_approx_det_segs_required_for_pstate(
+		struct _vcs_dpi_soc_bounding_box_st *soc,
+		int pix_clk_100hz, int bpp, int seg_size_kb);
+
+int dcn31x_populate_dml_pipes_from_context(struct dc *dc,
+					  struct dc_state *context,
+					  display_e2e_pipe_params_st *pipes,
+					  enum dc_validate_mode validate_mode);
+#endif /* __DCN31_FPU_H__*/

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.h

@@ -0,0 +1,43 @@
+/*
+ * Copyright 2017 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 __DML31_DISPLAY_MODE_VBA_H__
+#define __DML31_DISPLAY_MODE_VBA_H__
+
+void dml31_recalculate(struct display_mode_lib *mode_lib);
+void dml31_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_lib);
+double dml31_CalculateWriteBackDISPCLK(
+		enum source_format_class WritebackPixelFormat,
+		double PixelClock,
+		double WritebackHRatio,
+		double WritebackVRatio,
+		unsigned int WritebackHTaps,
+		unsigned int WritebackVTaps,
+		long   WritebackSourceWidth,
+		long   WritebackDestinationWidth,
+		unsigned int HTotal,
+		unsigned int WritebackLineBufferSize);
+
+#endif /* __DML31_DISPLAY_MODE_VBA_H__ */

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_rq_dlg_calc_31.h

@@ -0,0 +1,69 @@
+/*
+ * Copyright 2017 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 __DML31_DISPLAY_RQ_DLG_CALC_H__
+#define __DML31_DISPLAY_RQ_DLG_CALC_H__
+
+#include "../display_rq_dlg_helpers.h"
+
+struct display_mode_lib;
+
+// Function: dml_rq_dlg_get_rq_reg
+//  Main entry point for test to get the register values out of this DML class.
+//  This function calls <get_rq_param> and <extract_rq_regs> fucntions to calculate
+//  and then populate the rq_regs struct
+// Input:
+//  pipe_param - pipe source configuration (e.g. vp, pitch, scaling, dest, etc.)
+// Output:
+//  rq_regs - struct that holds all the RQ registers field value.
+//            See also: <display_rq_regs_st>
+void dml31_rq_dlg_get_rq_reg(struct display_mode_lib *mode_lib,
+		display_rq_regs_st *rq_regs,
+		const display_pipe_params_st *pipe_param);
+
+// Function: dml_rq_dlg_get_dlg_reg
+//   Calculate and return DLG and TTU register struct given the system setting
+// Output:
+//  dlg_regs - output DLG register struct
+//  ttu_regs - output DLG TTU register struct
+// Input:
+//  e2e_pipe_param - "compacted" array of e2e pipe param struct
+//  num_pipes - num of active "pipe" or "route"
+//  pipe_idx - index that identifies the e2e_pipe_param that corresponding to this dlg
+//  cstate - 0: when calculate min_ttu_vblank it is assumed cstate is not required. 1: Normal mode, cstate is considered.
+//           Added for legacy or unrealistic timing tests.
+void dml31_rq_dlg_get_dlg_reg(struct display_mode_lib             *mode_lib,
+		display_dlg_regs_st          *dlg_regs,
+		display_ttu_regs_st          *ttu_regs,
+		const display_e2e_pipe_params_st *e2e_pipe_param,
+		const unsigned int            num_pipes,
+		const unsigned int            pipe_idx,
+		const bool                    cstate_en,
+		const bool                    pstate_en,
+		const bool                    vm_en,
+		const bool                    ignore_viewport_pos,
+		const bool                    immediate_flip_support);
+
+#endif

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c

@@ -0,0 +1,429 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright 2022 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 "clk_mgr.h"
+#include "resource.h"
+#include "dcn31/dcn31_hubbub.h"
+#include "dcn314_fpu.h"
+#include "dml/dcn20/dcn20_fpu.h"
+#include "dml/dcn31/dcn31_fpu.h"
+#include "dml/display_mode_vba.h"
+#include "dml/dml_inline_defs.h"
+
+struct _vcs_dpi_ip_params_st dcn3_14_ip = {
+	.VBlankNomDefaultUS = 668,
+	.gpuvm_enable = 1,
+	.gpuvm_max_page_table_levels = 1,
+	.hostvm_enable = 1,
+	.hostvm_max_page_table_levels = 2,
+	.rob_buffer_size_kbytes = 64,
+	.det_buffer_size_kbytes = DCN3_14_DEFAULT_DET_SIZE,
+	.config_return_buffer_size_in_kbytes = 1792,
+	.compressed_buffer_segment_size_in_kbytes = 64,
+	.meta_fifo_size_in_kentries = 32,
+	.zero_size_buffer_entries = 512,
+	.compbuf_reserved_space_64b = 256,
+	.compbuf_reserved_space_zs = 64,
+	.dpp_output_buffer_pixels = 2560,
+	.opp_output_buffer_lines = 1,
+	.pixel_chunk_size_kbytes = 8,
+	.meta_chunk_size_kbytes = 2,
+	.min_meta_chunk_size_bytes = 256,
+	.writeback_chunk_size_kbytes = 8,
+	.ptoi_supported = false,
+	.num_dsc = 4,
+	.maximum_dsc_bits_per_component = 10,
+	.dsc422_native_support = false,
+	.is_line_buffer_bpp_fixed = true,
+	.line_buffer_fixed_bpp = 48,
+	.line_buffer_size_bits = 789504,
+	.max_line_buffer_lines = 12,
+	.writeback_interface_buffer_size_kbytes = 90,
+	.max_num_dpp = 4,
+	.max_num_otg = 4,
+	.max_num_hdmi_frl_outputs = 1,
+	.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,
+	.max_hscl_taps = 8,
+	.max_vscl_taps = 8,
+	.dpte_buffer_size_in_pte_reqs_luma = 64,
+	.dpte_buffer_size_in_pte_reqs_chroma = 34,
+	.dispclk_ramp_margin_percent = 1,
+	.max_inter_dcn_tile_repeaters = 8,
+	.cursor_buffer_size = 16,
+	.cursor_chunk_size = 2,
+	.writeback_line_buffer_buffer_size = 0,
+	.writeback_min_hscl_ratio = 1,
+	.writeback_min_vscl_ratio = 1,
+	.writeback_max_hscl_ratio = 1,
+	.writeback_max_vscl_ratio = 1,
+	.writeback_max_hscl_taps = 1,
+	.writeback_max_vscl_taps = 1,
+	.dppclk_delay_subtotal = 46,
+	.dppclk_delay_scl = 50,
+	.dppclk_delay_scl_lb_only = 16,
+	.dppclk_delay_cnvc_formatter = 27,
+	.dppclk_delay_cnvc_cursor = 6,
+	.dispclk_delay_subtotal = 119,
+	.dynamic_metadata_vm_enabled = false,
+	.odm_combine_4to1_supported = false,
+	.dcc_supported = true,
+};
+
+static struct _vcs_dpi_soc_bounding_box_st dcn3_14_soc = {
+		/*TODO: correct dispclk/dppclk voltage level determination*/
+	.clock_limits = {
+		{
+			.state = 0,
+			.dispclk_mhz = 1200.0,
+			.dppclk_mhz = 1200.0,
+			.phyclk_mhz = 600.0,
+			.phyclk_d18_mhz = 667.0,
+			.dscclk_mhz = 186.0,
+			.dtbclk_mhz = 600.0,
+		},
+		{
+			.state = 1,
+			.dispclk_mhz = 1200.0,
+			.dppclk_mhz = 1200.0,
+			.phyclk_mhz = 810.0,
+			.phyclk_d18_mhz = 667.0,
+			.dscclk_mhz = 209.0,
+			.dtbclk_mhz = 600.0,
+		},
+		{
+			.state = 2,
+			.dispclk_mhz = 1200.0,
+			.dppclk_mhz = 1200.0,
+			.phyclk_mhz = 810.0,
+			.phyclk_d18_mhz = 667.0,
+			.dscclk_mhz = 209.0,
+			.dtbclk_mhz = 600.0,
+		},
+		{
+			.state = 3,
+			.dispclk_mhz = 1200.0,
+			.dppclk_mhz = 1200.0,
+			.phyclk_mhz = 810.0,
+			.phyclk_d18_mhz = 667.0,
+			.dscclk_mhz = 371.0,
+			.dtbclk_mhz = 600.0,
+		},
+		{
+			.state = 4,
+			.dispclk_mhz = 1200.0,
+			.dppclk_mhz = 1200.0,
+			.phyclk_mhz = 810.0,
+			.phyclk_d18_mhz = 667.0,
+			.dscclk_mhz = 417.0,
+			.dtbclk_mhz = 600.0,
+		},
+	},
+	.num_states = 5,
+	.sr_exit_time_us = 16.5,
+	.sr_enter_plus_exit_time_us = 18.5,
+	.sr_exit_z8_time_us = 268.0,
+	.sr_enter_plus_exit_z8_time_us = 393.0,
+	.writeback_latency_us = 12.0,
+	.dram_channel_width_bytes = 4,
+	.round_trip_ping_latency_dcfclk_cycles = 106,
+	.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_sdp_bw_after_urgent = 80.0,
+	.pct_ideal_dram_sdp_bw_after_urgent_pixel_only = 65.0,
+	.pct_ideal_dram_sdp_bw_after_urgent_pixel_and_vm = 60.0,
+	.pct_ideal_dram_sdp_bw_after_urgent_vm_only = 30.0,
+	.max_avg_sdp_bw_use_normal_percent = 60.0,
+	.max_avg_dram_bw_use_normal_percent = 60.0,
+	.fabric_datapath_to_dcn_data_return_bytes = 32,
+	.return_bus_width_bytes = 64,
+	.downspread_percent = 0.38,
+	.dcn_downspread_percent = 0.5,
+	.gpuvm_min_page_size_bytes = 4096,
+	.hostvm_min_page_size_bytes = 4096,
+	.do_urgent_latency_adjustment = false,
+	.urgent_latency_adjustment_fabric_clock_component_us = 0,
+	.urgent_latency_adjustment_fabric_clock_reference_mhz = 0,
+};
+
+
+void dcn314_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_params)
+{
+	struct clk_limit_table *clk_table = &bw_params->clk_table;
+	struct _vcs_dpi_voltage_scaling_st *clock_limits =
+		dcn3_14_soc.clock_limits;
+	unsigned int i, closest_clk_lvl;
+	unsigned int max_dispclk_mhz = 0, max_dppclk_mhz = 0;
+	int j;
+
+	dc_assert_fp_enabled();
+
+	// Default clock levels are used for diags, which may lead to overclocking.
+	if (dc->config.use_default_clock_table == false) {
+		dcn3_14_ip.max_num_otg = dc->res_pool->res_cap->num_timing_generator;
+		dcn3_14_ip.max_num_dpp = dc->res_pool->pipe_count;
+
+		if (bw_params->dram_channel_width_bytes > 0)
+			dcn3_14_soc.dram_channel_width_bytes = bw_params->dram_channel_width_bytes;
+
+		if (bw_params->num_channels > 0)
+			dcn3_14_soc.num_chans = bw_params->num_channels;
+
+		ASSERT(dcn3_14_soc.num_chans);
+		ASSERT(clk_table->num_entries);
+
+		/* Prepass to find max clocks independent of voltage level. */
+		for (i = 0; i < clk_table->num_entries; ++i) {
+			if (clk_table->entries[i].dispclk_mhz > max_dispclk_mhz)
+				max_dispclk_mhz = clk_table->entries[i].dispclk_mhz;
+			if (clk_table->entries[i].dppclk_mhz > max_dppclk_mhz)
+				max_dppclk_mhz = clk_table->entries[i].dppclk_mhz;
+		}
+
+		for (i = 0; i < clk_table->num_entries; i++) {
+			/* loop backwards*/
+			for (closest_clk_lvl = 0, j = dcn3_14_soc.num_states - 1; j >= 0; j--) {
+				if ((unsigned int) dcn3_14_soc.clock_limits[j].dcfclk_mhz <= clk_table->entries[i].dcfclk_mhz) {
+					closest_clk_lvl = j;
+					break;
+				}
+			}
+			if (clk_table->num_entries == 1) {
+				/*smu gives one DPM level, let's take the highest one*/
+				closest_clk_lvl = dcn3_14_soc.num_states - 1;
+			}
+
+			clock_limits[i].state = i;
+
+			/* Clocks dependent on voltage level. */
+			clock_limits[i].dcfclk_mhz = clk_table->entries[i].dcfclk_mhz;
+			if (clk_table->num_entries == 1 &&
+				clock_limits[i].dcfclk_mhz < dcn3_14_soc.clock_limits[closest_clk_lvl].dcfclk_mhz) {
+				/*SMU fix not released yet*/
+				clock_limits[i].dcfclk_mhz = dcn3_14_soc.clock_limits[closest_clk_lvl].dcfclk_mhz;
+			}
+			clock_limits[i].fabricclk_mhz = clk_table->entries[i].fclk_mhz;
+			clock_limits[i].socclk_mhz = clk_table->entries[i].socclk_mhz;
+
+			if (clk_table->entries[i].memclk_mhz && clk_table->entries[i].wck_ratio)
+				clock_limits[i].dram_speed_mts = clk_table->entries[i].memclk_mhz * 2 * clk_table->entries[i].wck_ratio;
+
+			/* Clocks independent of voltage level. */
+			clock_limits[i].dispclk_mhz = max_dispclk_mhz ? max_dispclk_mhz :
+				dcn3_14_soc.clock_limits[closest_clk_lvl].dispclk_mhz;
+
+			clock_limits[i].dppclk_mhz = max_dppclk_mhz ? max_dppclk_mhz :
+				dcn3_14_soc.clock_limits[closest_clk_lvl].dppclk_mhz;
+
+			clock_limits[i].dram_bw_per_chan_gbps = dcn3_14_soc.clock_limits[closest_clk_lvl].dram_bw_per_chan_gbps;
+			clock_limits[i].dscclk_mhz = dcn3_14_soc.clock_limits[closest_clk_lvl].dscclk_mhz;
+			clock_limits[i].dtbclk_mhz = dcn3_14_soc.clock_limits[closest_clk_lvl].dtbclk_mhz;
+			clock_limits[i].phyclk_d18_mhz = dcn3_14_soc.clock_limits[closest_clk_lvl].phyclk_d18_mhz;
+			clock_limits[i].phyclk_mhz = dcn3_14_soc.clock_limits[closest_clk_lvl].phyclk_mhz;
+		}
+		for (i = 0; i < clk_table->num_entries; i++)
+			dcn3_14_soc.clock_limits[i] = clock_limits[i];
+		if (clk_table->num_entries) {
+			dcn3_14_soc.num_states = clk_table->num_entries;
+		}
+	}
+
+	if (max_dispclk_mhz) {
+		dcn3_14_soc.dispclk_dppclk_vco_speed_mhz = max_dispclk_mhz * 2;
+		dc->dml.soc.dispclk_dppclk_vco_speed_mhz = max_dispclk_mhz * 2;
+	}
+
+	dcn20_patch_bounding_box(dc, &dcn3_14_soc);
+	dml_init_instance(&dc->dml, &dcn3_14_soc, &dcn3_14_ip, DML_PROJECT_DCN314);
+}
+
+static bool is_dual_plane(enum surface_pixel_format format)
+{
+	return format >= SURFACE_PIXEL_FORMAT_VIDEO_BEGIN || format == SURFACE_PIXEL_FORMAT_GRPH_RGBE_ALPHA;
+}
+
+/*
+ * micro_sec_to_vert_lines () - converts time to number of vertical lines for a given timing
+ *
+ * @param: num_us: number of microseconds
+ * @return: number of vertical lines. If exact number of vertical lines is not found then
+ *          it will round up to next number of lines to guarantee num_us
+ */
+static unsigned int micro_sec_to_vert_lines(unsigned int num_us, struct dc_crtc_timing *timing)
+{
+	unsigned int num_lines = 0;
+	double lines_time_in_ns = 1000.0 *
+			(((double)timing->h_total * 1000.0) /
+			 ((double)timing->pix_clk_100hz / 10.0));
+
+	num_lines = (unsigned int)dml_ceil(1000.0 * num_us / lines_time_in_ns, 1.0);
+
+	return num_lines;
+}
+
+static unsigned int get_vertical_back_porch(struct dc_crtc_timing *timing)
+{
+	unsigned int v_active = 0, v_blank = 0, v_back_porch = 0;
+
+	v_active = timing->v_border_top + timing->v_addressable + timing->v_border_bottom;
+	v_blank = timing->v_total - v_active;
+	v_back_porch = v_blank - timing->v_front_porch - timing->v_sync_width;
+
+	return v_back_porch;
+}
+
+int dcn314_populate_dml_pipes_from_context_fpu(struct dc *dc, struct dc_state *context,
+					       display_e2e_pipe_params_st *pipes,
+					       enum dc_validate_mode validate_mode)
+{
+	int pipe_cnt;
+	unsigned int i;
+	struct resource_context *res_ctx = &context->res_ctx;
+	struct pipe_ctx *pipe = 0;
+	bool upscaled = false;
+	const unsigned int max_allowed_vblank_nom = 1023;
+
+	dc_assert_fp_enabled();
+
+	dcn31x_populate_dml_pipes_from_context(dc, context, pipes, validate_mode);
+
+	for (i = 0, pipe_cnt = 0; i < dc->res_pool->pipe_count; i++) {
+		struct dc_crtc_timing *timing;
+		unsigned int num_lines = 0;
+		unsigned int v_back_porch = 0;
+
+		if (!res_ctx->pipe_ctx[i].stream)
+			continue;
+		pipe = &res_ctx->pipe_ctx[i];
+		timing = &pipe->stream->timing;
+
+		num_lines = micro_sec_to_vert_lines(dcn3_14_ip.VBlankNomDefaultUS, timing);
+
+		if (pipe->stream->adjust.v_total_min != 0)
+			pipes[pipe_cnt].pipe.dest.vtotal = pipe->stream->adjust.v_total_min;
+		else
+			pipes[pipe_cnt].pipe.dest.vtotal = timing->v_total;
+
+		v_back_porch  = get_vertical_back_porch(timing);
+
+		pipes[pipe_cnt].pipe.dest.vblank_nom = timing->v_total - pipes[pipe_cnt].pipe.dest.vactive;
+		pipes[pipe_cnt].pipe.dest.vblank_nom = min(pipes[pipe_cnt].pipe.dest.vblank_nom, num_lines);
+		// vblank_nom should not smaller than (VSync (timing->v_sync_width + v_back_porch) + 2)
+		// + 2 is because
+		// 1 -> VStartup_start should be 1 line before VSync
+		// 1 -> always reserve 1 line between start of vblank to vstartup signal
+		pipes[pipe_cnt].pipe.dest.vblank_nom =
+			max(pipes[pipe_cnt].pipe.dest.vblank_nom, timing->v_sync_width + v_back_porch + 2);
+		pipes[pipe_cnt].pipe.dest.vblank_nom = min(pipes[pipe_cnt].pipe.dest.vblank_nom, max_allowed_vblank_nom);
+
+		if (pipe->plane_state &&
+				(pipe->plane_state->src_rect.height < pipe->plane_state->dst_rect.height ||
+				pipe->plane_state->src_rect.width < pipe->plane_state->dst_rect.width))
+			upscaled = true;
+
+		/* Apply HostVM policy - either based on hypervisor globally enabled, or rIOMMU active */
+		if (dc->debug.dml_hostvm_override == DML_HOSTVM_NO_OVERRIDE)
+			pipes[i].pipe.src.hostvm = dc->vm_pa_config.is_hvm_enabled || dc->res_pool->hubbub->riommu_active;
+
+		/*
+		 * Immediate flip can be set dynamically after enabling the plane.
+		 * We need to require support for immediate flip or underflow can be
+		 * intermittently experienced depending on peak b/w requirements.
+		 */
+		pipes[pipe_cnt].pipe.src.immediate_flip = true;
+
+		pipes[pipe_cnt].pipe.src.unbounded_req_mode = false;
+		pipes[pipe_cnt].pipe.src.dcc_fraction_of_zs_req_luma = 0;
+		pipes[pipe_cnt].pipe.src.dcc_fraction_of_zs_req_chroma = 0;
+		pipes[pipe_cnt].pipe.dest.vfront_porch = timing->v_front_porch;
+		pipes[pipe_cnt].pipe.src.dcc_rate = 3;
+		pipes[pipe_cnt].dout.dsc_input_bpc = 0;
+
+		if (pipes[pipe_cnt].dout.dsc_enable) {
+			switch (timing->display_color_depth) {
+			case COLOR_DEPTH_888:
+				pipes[pipe_cnt].dout.dsc_input_bpc = 8;
+				break;
+			case COLOR_DEPTH_101010:
+				pipes[pipe_cnt].dout.dsc_input_bpc = 10;
+				break;
+			case COLOR_DEPTH_121212:
+				pipes[pipe_cnt].dout.dsc_input_bpc = 12;
+				break;
+			default:
+				ASSERT(0);
+				break;
+			}
+		}
+
+		pipe_cnt++;
+	}
+	context->bw_ctx.dml.ip.det_buffer_size_kbytes = DCN3_14_DEFAULT_DET_SIZE;
+
+	if (pipe_cnt == 1 && pipe->plane_state
+		&& pipe->plane_state->rotation == ROTATION_ANGLE_0 && !dc->debug.disable_z9_mpc) {
+		if (!is_dual_plane(pipe->plane_state->format) && pipe->plane_state->src_rect.width <= 5120) {
+			/* Limit to 5k max to avoid forced pipe split when there is not enough detile for swath */
+			context->bw_ctx.dml.ip.det_buffer_size_kbytes = 192;
+			pipes[0].pipe.src.unbounded_req_mode = true;
+		}
+	} else if (context->stream_count >= dc->debug.crb_alloc_policy_min_disp_count
+			&& dc->debug.crb_alloc_policy > DET_SIZE_DEFAULT) {
+		context->bw_ctx.dml.ip.det_buffer_size_kbytes = dc->debug.crb_alloc_policy * 64;
+	} else if (context->stream_count >= 3 && upscaled) {
+		context->bw_ctx.dml.ip.det_buffer_size_kbytes = 192;
+	}
+
+	if (dc->debug.force_odm_combine_4to1)
+		context->bw_ctx.dml.ip.odm_combine_4to1_supported = true;
+
+	for (i = 0; i < dc->res_pool->pipe_count; i++) {
+		struct pipe_ctx *cur_pipe = &context->res_ctx.pipe_ctx[i];
+
+		if (!cur_pipe->stream)
+			continue;
+
+		if (cur_pipe->stream->signal == SIGNAL_TYPE_EDP && dc->debug.seamless_boot_odm_combine &&
+				cur_pipe->stream->apply_seamless_boot_optimization) {
+
+			if (cur_pipe->stream->apply_boot_odm_mode == dm_odm_combine_policy_2to1) {
+				context->bw_ctx.dml.vba.ODMCombinePolicy = dm_odm_combine_policy_2to1;
+				break;
+			}
+		}
+	}
+
+	return pipe_cnt;
+}

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.h

@@ -0,0 +1,40 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright 2022 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 __DCN314_FPU_H__
+#define __DCN314_FPU_H__
+
+#define DCN3_14_DEFAULT_DET_SIZE 384
+#define DCN3_14_MAX_DET_SIZE 384
+#define DCN3_14_MIN_COMPBUF_SIZE_KB 128
+#define DCN3_14_CRB_SEGMENT_SIZE_KB 64
+
+void dcn314_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_params);
+int dcn314_populate_dml_pipes_from_context_fpu(struct dc *dc, struct dc_state *context,
+					       display_e2e_pipe_params_st *pipes,
+					       enum dc_validate_mode validate_mode);
+
+#endif

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_mode_vba_314.h

@@ -0,0 +1,44 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright 2022 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 __DML314_DISPLAY_MODE_VBA_H__
+#define __DML314_DISPLAY_MODE_VBA_H__
+
+void dml314_recalculate(struct display_mode_lib *mode_lib);
+void dml314_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_lib);
+double dml314_CalculateWriteBackDISPCLK(
+		enum source_format_class WritebackPixelFormat,
+		double PixelClock,
+		double WritebackHRatio,
+		double WritebackVRatio,
+		unsigned int WritebackHTaps,
+		unsigned int WritebackVTaps,
+		long   WritebackSourceWidth,
+		long   WritebackDestinationWidth,
+		unsigned int HTotal,
+		unsigned int WritebackLineBufferSize);
+
+#endif /* __DML314_DISPLAY_MODE_VBA_H__ */

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_rq_dlg_calc_314.h

@@ -0,0 +1,70 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright 2022 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 __DML314_DISPLAY_RQ_DLG_CALC_H__
+#define __DML314_DISPLAY_RQ_DLG_CALC_H__
+
+#include "../display_rq_dlg_helpers.h"
+
+struct display_mode_lib;
+
+// Function: dml_rq_dlg_get_rq_reg
+//  Main entry point for test to get the register values out of this DML class.
+//  This function calls <get_rq_param> and <extract_rq_regs> fucntions to calculate
+//  and then populate the rq_regs struct
+// Input:
+//  pipe_param - pipe source configuration (e.g. vp, pitch, scaling, dest, etc.)
+// Output:
+//  rq_regs - struct that holds all the RQ registers field value.
+//            See also: <display_rq_regs_st>
+void dml314_rq_dlg_get_rq_reg(struct display_mode_lib *mode_lib,
+		display_rq_regs_st *rq_regs,
+		const display_pipe_params_st *pipe_param);
+
+// Function: dml_rq_dlg_get_dlg_reg
+//   Calculate and return DLG and TTU register struct given the system setting
+// Output:
+//  dlg_regs - output DLG register struct
+//  ttu_regs - output DLG TTU register struct
+// Input:
+//  e2e_pipe_param - "compacted" array of e2e pipe param struct
+//  num_pipes - num of active "pipe" or "route"
+//  pipe_idx - index that identifies the e2e_pipe_param that corresponding to this dlg
+//  cstate - 0: when calculate min_ttu_vblank it is assumed cstate is not required. 1: Normal mode, cstate is considered.
+//           Added for legacy or unrealistic timing tests.
+void dml314_rq_dlg_get_dlg_reg(struct display_mode_lib             *mode_lib,
+		display_dlg_regs_st          *dlg_regs,
+		display_ttu_regs_st          *ttu_regs,
+		const display_e2e_pipe_params_st *e2e_pipe_param,
+		const unsigned int            num_pipes,
+		const unsigned int            pipe_idx,
+		const bool                    cstate_en,
+		const bool                    pstate_en,
+		const bool                    vm_en,
+		const bool                    ignore_viewport_pos,
+		const bool                    immediate_flip_support);
+
+#endif

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.h

@@ -0,0 +1,82 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright 2022 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 __DCN32_FPU_H__
+#define __DCN32_FPU_H__
+
+#include "clk_mgr_internal.h"
+
+void dcn32_build_wm_range_table_fpu(struct clk_mgr_internal *clk_mgr);
+
+void dcn32_helper_populate_phantom_dlg_params(struct dc *dc,
+					      struct dc_state *context,
+					      display_e2e_pipe_params_st *pipes,
+					      int pipe_cnt);
+
+void dcn32_set_phantom_stream_timing(struct dc *dc,
+				     struct dc_state *context,
+				     struct pipe_ctx *ref_pipe,
+				     struct dc_stream_state *phantom_stream,
+				     display_e2e_pipe_params_st *pipes,
+				     unsigned int pipe_cnt,
+				     unsigned int dc_pipe_idx);
+
+bool dcn32_internal_validate_bw(struct dc *dc,
+				struct dc_state *context,
+				display_e2e_pipe_params_st *pipes,
+				int *pipe_cnt_out,
+				int *vlevel_out,
+				enum dc_validate_mode validate_mode);
+
+void dcn32_calculate_wm_and_dlg_fpu(struct dc *dc, struct dc_state *context,
+				display_e2e_pipe_params_st *pipes,
+				int pipe_cnt,
+				int vlevel);
+
+void dcn32_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_params);
+
+int dcn32_find_dummy_latency_index_for_fw_based_mclk_switch(struct dc *dc,
+							    struct dc_state *context,
+							    display_e2e_pipe_params_st *pipes,
+							    int pipe_cnt,
+							    int vlevel);
+
+void dcn32_patch_dpm_table(struct clk_bw_params *bw_params);
+
+void dcn32_zero_pipe_dcc_fraction(display_e2e_pipe_params_st *pipes,
+				  int pipe_cnt);
+
+void dcn32_assign_fpo_vactive_candidate(struct dc *dc, const struct dc_state *context, struct dc_stream_state **fpo_candidate_stream);
+
+bool dcn32_find_vactive_pipe(struct dc *dc, const struct dc_state *context, struct dc_stream_state *fpo_candidate_stream, uint32_t vactive_margin_req);
+
+void dcn32_override_min_req_memclk(struct dc *dc, struct dc_state *context);
+
+void dcn32_set_clock_limits(const struct _vcs_dpi_soc_bounding_box_st *soc_bb);
+
+unsigned int dcn32_get_max_dispclk_mhz(struct dc *dc, struct dc_state *context);
+
+#endif

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.h

@@ -0,0 +1,64 @@
+/*
+ * Copyright 2022 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 __DML32_DISPLAY_MODE_VBA_H__
+#define __DML32_DISPLAY_MODE_VBA_H__
+
+#include "../display_mode_enums.h"
+
+// To enable a lot of debug msg
+//#define __DML_VBA_DEBUG__
+// For DML-C changes that hasn't been propagated to VBA yet
+//#define __DML_VBA_ALLOW_DELTA__
+
+// Move these to ip parameters/constant
+// At which vstartup the DML start to try if the mode can be supported
+#define __DML_VBA_MIN_VSTARTUP__    9
+
+// Delay in DCFCLK from ARB to DET (1st num is ARB to SDPIF, 2nd number is SDPIF to DET)
+#define __DML_ARB_TO_RET_DELAY__    7 + 95
+
+// fudge factor for min dcfclk calclation
+#define __DML_MIN_DCFCLK_FACTOR__   1.15
+
+// Prefetch schedule max vratio
+#define __DML_MAX_VRATIO_PRE__ 7.9
+#define __DML_MAX_BW_RATIO_PRE__ 4.0
+
+#define __DML_VBA_MAX_DST_Y_PRE__    63.75
+
+#define BPP_INVALID 0
+#define BPP_BLENDED_PIPE 0xffffffff
+
+#define MEM_STROBE_FREQ_MHZ 1600
+#define DCFCLK_FREQ_EXTRA_PREFETCH_REQ_MHZ 300
+#define MEM_STROBE_MAX_DELIVERY_TIME_US 60.0
+
+struct display_mode_lib;
+
+void dml32_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_lib);
+void dml32_recalculate(struct display_mode_lib *mode_lib);
+
+#endif

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_rq_dlg_calc_32.c

@@ -0,0 +1,614 @@
+/*
+ * Copyright 2022 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 "../display_mode_lib.h"
+#include "../display_mode_vba.h"
+#include "../dml_inline_defs.h"
+#include "display_rq_dlg_calc_32.h"
+
+static bool is_dual_plane(enum source_format_class source_format)
+{
+	bool ret_val = 0;
+
+	if ((source_format == dm_420_12) || (source_format == dm_420_8) || (source_format == dm_420_10)
+		|| (source_format == dm_rgbe_alpha))
+		ret_val = 1;
+
+	return ret_val;
+}
+
+void dml32_rq_dlg_get_rq_reg(display_rq_regs_st *rq_regs,
+		struct display_mode_lib *mode_lib,
+		const display_e2e_pipe_params_st *e2e_pipe_param,
+		const unsigned int num_pipes,
+		const unsigned int pipe_idx)
+{
+	const display_pipe_source_params_st *src = &e2e_pipe_param[pipe_idx].pipe.src;
+	bool dual_plane = is_dual_plane((enum source_format_class) (src->source_format));
+	double stored_swath_l_bytes;
+	double stored_swath_c_bytes;
+	bool is_phantom_pipe;
+	uint32_t pixel_chunk_bytes = 0;
+	uint32_t min_pixel_chunk_bytes = 0;
+	uint32_t meta_chunk_bytes = 0;
+	uint32_t min_meta_chunk_bytes = 0;
+	uint32_t dpte_group_bytes = 0;
+	uint32_t mpte_group_bytes = 0;
+
+	uint32_t p1_pixel_chunk_bytes = 0;
+	uint32_t p1_min_pixel_chunk_bytes = 0;
+	uint32_t p1_meta_chunk_bytes = 0;
+	uint32_t p1_min_meta_chunk_bytes = 0;
+	uint32_t p1_dpte_group_bytes = 0;
+	uint32_t p1_mpte_group_bytes = 0;
+
+	unsigned int detile_buf_size_in_bytes;
+	unsigned int detile_buf_plane1_addr;
+	unsigned int pte_row_height_linear;
+
+	memset(rq_regs, 0, sizeof(*rq_regs));
+
+	dml_print("DML_DLG::%s: Calculation for pipe[%d] start, num_pipes=%d\n", __func__, pipe_idx, num_pipes);
+
+	pixel_chunk_bytes = (uint32_t)(get_pixel_chunk_size_in_kbyte(mode_lib, e2e_pipe_param, num_pipes) * 1024); // From VBA
+	min_pixel_chunk_bytes = (uint32_t)get_min_pixel_chunk_size_in_byte(mode_lib, e2e_pipe_param, num_pipes); // From VBA
+
+	if (pixel_chunk_bytes == 64 * 1024)
+		min_pixel_chunk_bytes = 0;
+
+	meta_chunk_bytes = (uint32_t)(get_meta_chunk_size_in_kbyte(mode_lib, e2e_pipe_param, num_pipes) * 1024); // From VBA
+	min_meta_chunk_bytes = (uint32_t)get_min_meta_chunk_size_in_byte(mode_lib, e2e_pipe_param, num_pipes); // From VBA
+
+	dpte_group_bytes = (uint32_t)get_dpte_group_size_in_bytes(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); // From VBA
+	mpte_group_bytes = (uint32_t)get_vm_group_size_in_bytes(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); // From VBA
+
+	p1_pixel_chunk_bytes = pixel_chunk_bytes;
+	p1_min_pixel_chunk_bytes = min_pixel_chunk_bytes;
+	p1_meta_chunk_bytes = meta_chunk_bytes;
+	p1_min_meta_chunk_bytes = min_meta_chunk_bytes;
+	p1_dpte_group_bytes = dpte_group_bytes;
+	p1_mpte_group_bytes = mpte_group_bytes;
+
+	if ((enum source_format_class) src->source_format == dm_rgbe_alpha)
+		p1_pixel_chunk_bytes = (uint32_t)(get_alpha_pixel_chunk_size_in_kbyte(mode_lib, e2e_pipe_param, num_pipes) * 1024);
+
+	rq_regs->rq_regs_l.chunk_size = (uint32_t)(dml_log2(pixel_chunk_bytes) - 10);
+	rq_regs->rq_regs_c.chunk_size = (uint32_t)(dml_log2(p1_pixel_chunk_bytes) - 10);
+
+	if (min_pixel_chunk_bytes == 0)
+		rq_regs->rq_regs_l.min_chunk_size = 0;
+	else
+		rq_regs->rq_regs_l.min_chunk_size = dml_log2(min_pixel_chunk_bytes) - 8 + 1;
+
+	if (p1_min_pixel_chunk_bytes == 0)
+		rq_regs->rq_regs_c.min_chunk_size = 0;
+	else
+		rq_regs->rq_regs_c.min_chunk_size = dml_log2(p1_min_pixel_chunk_bytes) - 8 + 1;
+
+	rq_regs->rq_regs_l.meta_chunk_size = (uint32_t)(dml_log2(meta_chunk_bytes) - 10);
+	rq_regs->rq_regs_c.meta_chunk_size = (uint32_t)(dml_log2(p1_meta_chunk_bytes) - 10);
+
+	if (min_meta_chunk_bytes == 0)
+		rq_regs->rq_regs_l.min_meta_chunk_size = 0;
+	else
+		rq_regs->rq_regs_l.min_meta_chunk_size = dml_log2(min_meta_chunk_bytes) - 6 + 1;
+
+	if (p1_min_meta_chunk_bytes == 0)
+		rq_regs->rq_regs_c.min_meta_chunk_size = 0;
+	else
+		rq_regs->rq_regs_c.min_meta_chunk_size = dml_log2(p1_min_meta_chunk_bytes) - 6 + 1;
+
+	rq_regs->rq_regs_l.dpte_group_size = (uint32_t)(dml_log2(dpte_group_bytes) - 6);
+	rq_regs->rq_regs_l.mpte_group_size = dml_log2(mpte_group_bytes) - 6;
+	rq_regs->rq_regs_c.dpte_group_size = dml_log2(p1_dpte_group_bytes) - 6;
+	rq_regs->rq_regs_c.mpte_group_size = dml_log2(p1_mpte_group_bytes) - 6;
+
+	detile_buf_size_in_bytes = (unsigned int)(get_det_buffer_size_kbytes(mode_lib, e2e_pipe_param, num_pipes, pipe_idx) * 1024);
+	detile_buf_plane1_addr = 0;
+	pte_row_height_linear = (unsigned int)get_dpte_row_height_linear_l(mode_lib, e2e_pipe_param, num_pipes,
+			pipe_idx);
+
+	if (src->sw_mode == dm_sw_linear)
+		ASSERT(pte_row_height_linear >= 8);
+
+	rq_regs->rq_regs_l.pte_row_height_linear = (unsigned int)(dml_floor(dml_log2(pte_row_height_linear), 1) - 3);
+
+	if (dual_plane) {
+		unsigned int p1_pte_row_height_linear = (unsigned int)get_dpte_row_height_linear_c(mode_lib, e2e_pipe_param,
+				num_pipes, pipe_idx);
+		if (src->sw_mode == dm_sw_linear)
+			ASSERT(p1_pte_row_height_linear >= 8);
+
+		rq_regs->rq_regs_c.pte_row_height_linear = (unsigned int)(dml_floor(dml_log2(p1_pte_row_height_linear), 1) - 3);
+	}
+
+	rq_regs->rq_regs_l.swath_height = (unsigned int)dml_log2(get_swath_height_l(mode_lib, e2e_pipe_param, num_pipes, pipe_idx));
+	rq_regs->rq_regs_c.swath_height = (unsigned int)dml_log2(get_swath_height_c(mode_lib, e2e_pipe_param, num_pipes, pipe_idx));
+
+	// FIXME: take the max between luma, chroma chunk size?
+	// okay for now, as we are setting pixel_chunk_bytes to 8kb anyways
+	if (pixel_chunk_bytes >= 32 * 1024 || (dual_plane && p1_pixel_chunk_bytes >= 32 * 1024)) { //32kb
+		rq_regs->drq_expansion_mode = 0;
+	} else {
+		rq_regs->drq_expansion_mode = 2;
+	}
+	rq_regs->prq_expansion_mode = 1;
+	rq_regs->mrq_expansion_mode = 1;
+	rq_regs->crq_expansion_mode = 1;
+
+	stored_swath_l_bytes = get_det_stored_buffer_size_l_bytes(mode_lib, e2e_pipe_param, num_pipes,
+			pipe_idx);
+	stored_swath_c_bytes = get_det_stored_buffer_size_c_bytes(mode_lib, e2e_pipe_param, num_pipes,
+			pipe_idx);
+	is_phantom_pipe = get_is_phantom_pipe(mode_lib, e2e_pipe_param, num_pipes, pipe_idx);
+
+	// Note: detile_buf_plane1_addr is in unit of 1KB
+	if (dual_plane) {
+		if (is_phantom_pipe) {
+			detile_buf_plane1_addr = (unsigned int)((1024.0 * 1024.0) / 2.0 / 1024.0); // half to chroma
+		} else {
+			if (stored_swath_l_bytes / stored_swath_c_bytes <= 1.5) {
+				detile_buf_plane1_addr = (unsigned int)(detile_buf_size_in_bytes / 2.0 / 1024.0); // half to chroma
+#ifdef __DML_RQ_DLG_CALC_DEBUG__
+				dml_print("DML_DLG: %s: detile_buf_plane1_addr = %d (1/2 to chroma)\n",
+						__func__, detile_buf_plane1_addr);
+#endif
+			} else {
+				detile_buf_plane1_addr = (unsigned int)(
+						dml_round_to_multiple(
+								(unsigned int) ((2.0 * detile_buf_size_in_bytes) / 3.0),
+								1024, 0) / 1024.0); // 2/3 to luma
+#ifdef __DML_RQ_DLG_CALC_DEBUG__
+				dml_print("DML_DLG: %s: detile_buf_plane1_addr = %d (1/3 chroma)\n",
+						__func__, detile_buf_plane1_addr);
+#endif
+			}
+		}
+	}
+	rq_regs->plane1_base_address = detile_buf_plane1_addr;
+
+#ifdef __DML_RQ_DLG_CALC_DEBUG__
+	dml_print("DML_DLG: %s: is_phantom_pipe = %d\n", __func__, is_phantom_pipe);
+	dml_print("DML_DLG: %s: stored_swath_l_bytes = %f\n", __func__, stored_swath_l_bytes);
+	dml_print("DML_DLG: %s: stored_swath_c_bytes = %f\n", __func__, stored_swath_c_bytes);
+	dml_print("DML_DLG: %s: detile_buf_size_in_bytes = %d\n", __func__, detile_buf_size_in_bytes);
+	dml_print("DML_DLG: %s: detile_buf_plane1_addr = %d\n", __func__, detile_buf_plane1_addr);
+	dml_print("DML_DLG: %s: plane1_base_address = %d\n", __func__, rq_regs->plane1_base_address);
+#endif
+	print__rq_regs_st(mode_lib, rq_regs);
+	dml_print("DML_DLG::%s: Calculation for pipe[%d] done, num_pipes=%d\n", __func__, pipe_idx, num_pipes);
+}
+
+void dml32_rq_dlg_get_dlg_reg(struct display_mode_lib *mode_lib,
+		display_dlg_regs_st *dlg_regs,
+		display_ttu_regs_st *ttu_regs,
+		display_e2e_pipe_params_st *e2e_pipe_param,
+		const unsigned int num_pipes,
+		const unsigned int pipe_idx)
+{
+	const display_pipe_source_params_st *src = &e2e_pipe_param[pipe_idx].pipe.src;
+	const display_pipe_dest_params_st *dst = &e2e_pipe_param[pipe_idx].pipe.dest;
+	const display_clocks_and_cfg_st *clks = &e2e_pipe_param[pipe_idx].clks_cfg;
+	double refcyc_per_req_delivery_pre_cur0 = 0.;
+	double refcyc_per_req_delivery_cur0 = 0.;
+	double refcyc_per_req_delivery_pre_c = 0.;
+	double refcyc_per_req_delivery_c = 0.;
+	double refcyc_per_req_delivery_pre_l;
+	double refcyc_per_req_delivery_l;
+	double refcyc_per_line_delivery_pre_c = 0.;
+	double refcyc_per_line_delivery_c = 0.;
+	double refcyc_per_line_delivery_pre_l;
+	double refcyc_per_line_delivery_l;
+	double min_ttu_vblank;
+	double vratio_pre_l;
+	double vratio_pre_c;
+	unsigned int min_dst_y_next_start;
+	unsigned int htotal = dst->htotal;
+	unsigned int hblank_end = dst->hblank_end;
+	unsigned int vblank_end = dst->vblank_end;
+	bool interlaced = dst->interlaced;
+	double pclk_freq_in_mhz = dst->pixel_rate_mhz;
+	unsigned int vready_after_vcount0;
+	double refclk_freq_in_mhz = clks->refclk_mhz;
+	double ref_freq_to_pix_freq = refclk_freq_in_mhz / pclk_freq_in_mhz;
+	bool dual_plane = 0;
+	unsigned int pipe_index_in_combine[DC__NUM_PIPES__MAX];
+	unsigned int dst_x_after_scaler;
+	unsigned int dst_y_after_scaler;
+	double dst_y_prefetch;
+	double dst_y_per_vm_vblank;
+	double dst_y_per_row_vblank;
+	double dst_y_per_vm_flip;
+	double dst_y_per_row_flip;
+	double max_dst_y_per_vm_vblank = 32.0;
+	double max_dst_y_per_row_vblank = 16.0;
+	double dst_y_per_pte_row_nom_l;
+	double dst_y_per_pte_row_nom_c;
+	double dst_y_per_meta_row_nom_l;
+	double dst_y_per_meta_row_nom_c;
+	double refcyc_per_pte_group_nom_l;
+	double refcyc_per_pte_group_nom_c;
+	double refcyc_per_pte_group_vblank_l;
+	double refcyc_per_pte_group_vblank_c;
+	double refcyc_per_pte_group_flip_l;
+	double refcyc_per_pte_group_flip_c;
+	double refcyc_per_meta_chunk_nom_l;
+	double refcyc_per_meta_chunk_nom_c;
+	double refcyc_per_meta_chunk_vblank_l;
+	double refcyc_per_meta_chunk_vblank_c;
+	double refcyc_per_meta_chunk_flip_l;
+	double refcyc_per_meta_chunk_flip_c;
+
+	memset(dlg_regs, 0, sizeof(*dlg_regs));
+	memset(ttu_regs, 0, sizeof(*ttu_regs));
+	dml_print("DML_DLG::%s: Calculation for pipe[%d] starts, num_pipes=%d\n", __func__, pipe_idx, num_pipes);
+	dml_print("DML_DLG: %s: refclk_freq_in_mhz     = %3.2f\n", __func__, refclk_freq_in_mhz);
+	dml_print("DML_DLG: %s: pclk_freq_in_mhz = %3.2f\n", __func__, pclk_freq_in_mhz);
+	dml_print("DML_DLG: %s: ref_freq_to_pix_freq   = %3.2f\n", __func__, ref_freq_to_pix_freq);
+	dml_print("DML_DLG: %s: interlaced = %d\n", __func__, interlaced);
+	ASSERT(ref_freq_to_pix_freq < 4.0);
+
+	dlg_regs->ref_freq_to_pix_freq = (unsigned int) (ref_freq_to_pix_freq * dml_pow(2, 19));
+	dlg_regs->refcyc_per_htotal = (unsigned int) (ref_freq_to_pix_freq * (double) htotal * dml_pow(2, 8));
+	dlg_regs->dlg_vblank_end = interlaced ? (vblank_end / 2) : vblank_end; // 15 bits
+
+	min_ttu_vblank = get_min_ttu_vblank_in_us(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); // From VBA
+	min_dst_y_next_start = (unsigned int)get_min_dst_y_next_start(mode_lib, e2e_pipe_param, num_pipes, pipe_idx);
+
+	dml_print("DML_DLG: %s: min_ttu_vblank (us)    = %3.2f\n", __func__, min_ttu_vblank);
+	dml_print("DML_DLG: %s: min_dst_y_next_start = %d\n", __func__, min_dst_y_next_start);
+	dml_print("DML_DLG: %s: ref_freq_to_pix_freq   = %3.2f\n", __func__, ref_freq_to_pix_freq);
+
+	dual_plane = is_dual_plane((enum source_format_class) (src->source_format));
+
+	vready_after_vcount0 = (unsigned int)get_vready_at_or_after_vsync(mode_lib, e2e_pipe_param, num_pipes,
+			pipe_idx); // From VBA
+	dlg_regs->vready_after_vcount0 = vready_after_vcount0;
+
+	dml_print("DML_DLG: %s: vready_after_vcount0 = %d\n", __func__, dlg_regs->vready_after_vcount0);
+
+	dst_x_after_scaler = (unsigned int)dml_ceil(get_dst_x_after_scaler(mode_lib, e2e_pipe_param, num_pipes, pipe_idx), 1);
+	dst_y_after_scaler = (unsigned int)dml_ceil(get_dst_y_after_scaler(mode_lib, e2e_pipe_param, num_pipes, pipe_idx), 1);
+
+	// do some adjustment on the dst_after scaler to account for odm combine mode
+	dml_print("DML_DLG: %s: input dst_x_after_scaler   = %d\n", __func__, dst_x_after_scaler);
+	dml_print("DML_DLG: %s: input dst_y_after_scaler   = %d\n", __func__, dst_y_after_scaler);
+
+	// need to figure out which side of odm combine we're in
+	if (dst->odm_combine == dm_odm_combine_mode_2to1 || dst->odm_combine == dm_odm_combine_mode_4to1) {
+		// figure out which pipes go together
+		bool visited[DC__NUM_PIPES__MAX];
+		unsigned int i, j, k;
+
+		for (k = 0; k < num_pipes; ++k) {
+			visited[k] = false;
+			pipe_index_in_combine[k] = 0;
+		}
+
+		for (i = 0; i < num_pipes; i++) {
+			if (e2e_pipe_param[i].pipe.src.is_hsplit && !visited[i]) {
+
+				unsigned int grp = e2e_pipe_param[i].pipe.src.hsplit_grp;
+				unsigned int grp_idx = 0;
+
+				for (j = i; j < num_pipes; j++) {
+					if (e2e_pipe_param[j].pipe.src.hsplit_grp == grp
+						&& e2e_pipe_param[j].pipe.src.is_hsplit && !visited[j]) {
+						pipe_index_in_combine[j] = grp_idx;
+						dml_print("DML_DLG: %s: pipe[%d] is in grp %d idx %d\n",
+								__func__, j, grp, grp_idx);
+						grp_idx++;
+						visited[j] = true;
+					}
+				}
+			}
+		}
+	}
+
+	if (dst->odm_combine == dm_odm_combine_mode_disabled) {
+		// FIXME how about ODM split??
+		dlg_regs->refcyc_h_blank_end = (unsigned int) ((double) hblank_end * ref_freq_to_pix_freq);
+	} else {
+		if (dst->odm_combine == dm_odm_combine_mode_2to1 || dst->odm_combine == dm_odm_combine_mode_4to1) {
+			// TODO: We should really check that 4to1 is supported before setting it to 4
+			unsigned int odm_combine_factor = (dst->odm_combine == dm_odm_combine_mode_2to1 ? 2 : 4);
+			unsigned int odm_pipe_index = pipe_index_in_combine[pipe_idx];
+
+			dlg_regs->refcyc_h_blank_end = (unsigned int) (((double) hblank_end
+				+ odm_pipe_index * (double) dst->hactive / odm_combine_factor) * ref_freq_to_pix_freq);
+		}
+	}
+	ASSERT(dlg_regs->refcyc_h_blank_end < (unsigned int)dml_pow(2, 13));
+
+	dml_print("DML_DLG: %s: htotal= %d\n", __func__, htotal);
+	dml_print("DML_DLG: %s: dst_x_after_scaler[%d]= %d\n", __func__, pipe_idx, dst_x_after_scaler);
+	dml_print("DML_DLG: %s: dst_y_after_scaler[%d] = %d\n", __func__, pipe_idx, dst_y_after_scaler);
+
+	dst_y_prefetch = get_dst_y_prefetch(mode_lib, e2e_pipe_param, num_pipes, pipe_idx);        // From VBA
+	// From VBA
+	dst_y_per_vm_vblank = get_dst_y_per_vm_vblank(mode_lib, e2e_pipe_param, num_pipes, pipe_idx);
+	// From VBA
+	dst_y_per_row_vblank = get_dst_y_per_row_vblank(mode_lib, e2e_pipe_param, num_pipes, pipe_idx);
+	dst_y_per_vm_flip = get_dst_y_per_vm_flip(mode_lib, e2e_pipe_param, num_pipes, pipe_idx);    // From VBA
+	dst_y_per_row_flip = get_dst_y_per_row_flip(mode_lib, e2e_pipe_param, num_pipes, pipe_idx);  // From VBA
+
+	// magic!
+	if (htotal <= 75) {
+		max_dst_y_per_vm_vblank = 100.0;
+		max_dst_y_per_row_vblank = 100.0;
+	}
+
+	dml_print("DML_DLG: %s: dst_y_prefetch (after rnd) = %3.2f\n", __func__, dst_y_prefetch);
+	dml_print("DML_DLG: %s: dst_y_per_vm_flip    = %3.2f\n", __func__, dst_y_per_vm_flip);
+	dml_print("DML_DLG: %s: dst_y_per_row_flip   = %3.2f\n", __func__, dst_y_per_row_flip);
+	dml_print("DML_DLG: %s: dst_y_per_vm_vblank  = %3.2f\n", __func__, dst_y_per_vm_vblank);
+	dml_print("DML_DLG: %s: dst_y_per_row_vblank = %3.2f\n", __func__, dst_y_per_row_vblank);
+
+	ASSERT(dst_y_per_vm_vblank < max_dst_y_per_vm_vblank);
+	ASSERT(dst_y_per_row_vblank < max_dst_y_per_row_vblank);
+	ASSERT(dst_y_prefetch > (dst_y_per_vm_vblank + dst_y_per_row_vblank));
+
+	vratio_pre_l = get_vratio_prefetch_l(mode_lib, e2e_pipe_param, num_pipes, pipe_idx);    // From VBA
+	vratio_pre_c = get_vratio_prefetch_c(mode_lib, e2e_pipe_param, num_pipes, pipe_idx);    // From VBA
+
+	dml_print("DML_DLG: %s: vratio_pre_l = %3.2f\n", __func__, vratio_pre_l);
+	dml_print("DML_DLG: %s: vratio_pre_c = %3.2f\n", __func__, vratio_pre_c);
+
+	// Active
+	refcyc_per_line_delivery_pre_l = get_refcyc_per_line_delivery_pre_l_in_us(mode_lib, e2e_pipe_param, num_pipes, pipe_idx) * refclk_freq_in_mhz;   // From VBA
+	refcyc_per_line_delivery_l = get_refcyc_per_line_delivery_l_in_us(mode_lib, e2e_pipe_param, num_pipes,
+			pipe_idx) * refclk_freq_in_mhz;       // From VBA
+
+	dml_print("DML_DLG: %s: refcyc_per_line_delivery_pre_l = %3.2f\n", __func__, refcyc_per_line_delivery_pre_l);
+	dml_print("DML_DLG: %s: refcyc_per_line_delivery_l     = %3.2f\n", __func__, refcyc_per_line_delivery_l);
+
+	if (dual_plane) {
+		refcyc_per_line_delivery_pre_c = get_refcyc_per_line_delivery_pre_c_in_us(mode_lib, e2e_pipe_param,
+				num_pipes, pipe_idx) * refclk_freq_in_mhz;     // From VBA
+		refcyc_per_line_delivery_c = get_refcyc_per_line_delivery_c_in_us(mode_lib, e2e_pipe_param, num_pipes,
+				pipe_idx) * refclk_freq_in_mhz; // From VBA
+
+		dml_print("DML_DLG: %s: refcyc_per_line_delivery_pre_c = %3.2f\n",
+				__func__, refcyc_per_line_delivery_pre_c);
+		dml_print("DML_DLG: %s: refcyc_per_line_delivery_c     = %3.2f\n",
+				__func__, refcyc_per_line_delivery_c);
+	}
+
+	if (src->dynamic_metadata_enable && src->gpuvm)
+		dlg_regs->refcyc_per_vm_dmdata = (unsigned int)(get_refcyc_per_vm_dmdata_in_us(mode_lib, e2e_pipe_param, num_pipes,
+				pipe_idx) * refclk_freq_in_mhz); // From VBA
+
+	dlg_regs->dmdata_dl_delta = (unsigned int)(get_dmdata_dl_delta_in_us(mode_lib, e2e_pipe_param, num_pipes, pipe_idx)
+		* refclk_freq_in_mhz); // From VBA
+
+	refcyc_per_req_delivery_pre_l = (unsigned int)(get_refcyc_per_req_delivery_pre_l_in_us(mode_lib, e2e_pipe_param, num_pipes, pipe_idx) * refclk_freq_in_mhz); // From VBA
+	refcyc_per_req_delivery_l = (unsigned int)(get_refcyc_per_req_delivery_l_in_us(mode_lib, e2e_pipe_param, num_pipes,
+			pipe_idx) * refclk_freq_in_mhz);     // From VBA
+
+	dml_print("DML_DLG: %s: refcyc_per_req_delivery_pre_l = %3.2f\n", __func__, refcyc_per_req_delivery_pre_l);
+	dml_print("DML_DLG: %s: refcyc_per_req_delivery_l     = %3.2f\n", __func__, refcyc_per_req_delivery_l);
+
+	if (dual_plane) {
+		refcyc_per_req_delivery_pre_c = get_refcyc_per_req_delivery_pre_c_in_us(mode_lib, e2e_pipe_param,
+				num_pipes, pipe_idx) * refclk_freq_in_mhz;  // From VBA
+		refcyc_per_req_delivery_c = get_refcyc_per_req_delivery_c_in_us(mode_lib, e2e_pipe_param, num_pipes,
+				pipe_idx) * refclk_freq_in_mhz;      // From VBA
+
+		dml_print("DML_DLG: %s: refcyc_per_req_delivery_pre_c = %3.2f\n",
+				__func__, refcyc_per_req_delivery_pre_c);
+		dml_print("DML_DLG: %s: refcyc_per_req_delivery_c     = %3.2f\n", __func__, refcyc_per_req_delivery_c);
+	}
+
+	// TTU - Cursor
+	ASSERT(src->num_cursors <= 1);
+	if (src->num_cursors > 0) {
+		refcyc_per_req_delivery_pre_cur0 = get_refcyc_per_cursor_req_delivery_pre_in_us(mode_lib,
+				e2e_pipe_param, num_pipes, pipe_idx) * refclk_freq_in_mhz;  // From VBA
+		refcyc_per_req_delivery_cur0 = get_refcyc_per_cursor_req_delivery_in_us(mode_lib, e2e_pipe_param,
+				num_pipes, pipe_idx) * refclk_freq_in_mhz;      // From VBA
+
+		dml_print("DML_DLG: %s: refcyc_per_req_delivery_pre_cur0 = %3.2f\n",
+				__func__, refcyc_per_req_delivery_pre_cur0);
+		dml_print("DML_DLG: %s: refcyc_per_req_delivery_cur0     = %3.2f\n",
+				__func__, refcyc_per_req_delivery_cur0);
+	}
+
+	// Assign to register structures
+	dlg_regs->min_dst_y_next_start = (unsigned int)(min_dst_y_next_start * dml_pow(2, 2));
+	ASSERT(dlg_regs->min_dst_y_next_start < (unsigned int)dml_pow(2, 18));
+
+	dlg_regs->dst_y_after_scaler = dst_y_after_scaler; // in terms of line
+	dlg_regs->refcyc_x_after_scaler = (unsigned int)(dst_x_after_scaler * ref_freq_to_pix_freq); // in terms of refclk
+	dlg_regs->dst_y_prefetch = (unsigned int) (dst_y_prefetch * dml_pow(2, 2));
+	dlg_regs->dst_y_per_vm_vblank = (unsigned int) (dst_y_per_vm_vblank * dml_pow(2, 2));
+	dlg_regs->dst_y_per_row_vblank = (unsigned int) (dst_y_per_row_vblank * dml_pow(2, 2));
+	dlg_regs->dst_y_per_vm_flip = (unsigned int) (dst_y_per_vm_flip * dml_pow(2, 2));
+	dlg_regs->dst_y_per_row_flip = (unsigned int) (dst_y_per_row_flip * dml_pow(2, 2));
+
+	dlg_regs->vratio_prefetch = (unsigned int) (vratio_pre_l * dml_pow(2, 19));
+	dlg_regs->vratio_prefetch_c = (unsigned int) (vratio_pre_c * dml_pow(2, 19));
+
+	dml_print("DML_DLG: %s: dlg_regs->dst_y_per_vm_vblank  = 0x%x\n", __func__, dlg_regs->dst_y_per_vm_vblank);
+	dml_print("DML_DLG: %s: dlg_regs->dst_y_per_row_vblank = 0x%x\n", __func__, dlg_regs->dst_y_per_row_vblank);
+	dml_print("DML_DLG: %s: dlg_regs->dst_y_per_vm_flip    = 0x%x\n", __func__, dlg_regs->dst_y_per_vm_flip);
+	dml_print("DML_DLG: %s: dlg_regs->dst_y_per_row_flip   = 0x%x\n", __func__, dlg_regs->dst_y_per_row_flip);
+
+	dlg_regs->refcyc_per_vm_group_vblank = (unsigned int)(get_refcyc_per_vm_group_vblank_in_us(mode_lib, e2e_pipe_param,
+			num_pipes, pipe_idx) * refclk_freq_in_mhz);               // From VBA
+	dlg_regs->refcyc_per_vm_group_flip = (unsigned int)(get_refcyc_per_vm_group_flip_in_us(mode_lib, e2e_pipe_param, num_pipes,
+			pipe_idx) * refclk_freq_in_mhz);                 // From VBA
+	dlg_regs->refcyc_per_vm_req_vblank = (unsigned int)(get_refcyc_per_vm_req_vblank_in_us(mode_lib, e2e_pipe_param, num_pipes,
+			pipe_idx) * refclk_freq_in_mhz * dml_pow(2, 10));                 // From VBA
+	dlg_regs->refcyc_per_vm_req_flip = (unsigned int)(get_refcyc_per_vm_req_flip_in_us(mode_lib, e2e_pipe_param, num_pipes,
+			pipe_idx) * refclk_freq_in_mhz * dml_pow(2, 10));  // From VBA
+
+	// From VBA
+	dst_y_per_pte_row_nom_l = get_dst_y_per_pte_row_nom_l(mode_lib, e2e_pipe_param, num_pipes, pipe_idx);
+	// From VBA
+	dst_y_per_pte_row_nom_c = get_dst_y_per_pte_row_nom_c(mode_lib, e2e_pipe_param, num_pipes, pipe_idx);
+	// From VBA
+	dst_y_per_meta_row_nom_l = get_dst_y_per_meta_row_nom_l(mode_lib, e2e_pipe_param, num_pipes, pipe_idx);
+	// From VBA
+	dst_y_per_meta_row_nom_c = get_dst_y_per_meta_row_nom_c(mode_lib, e2e_pipe_param, num_pipes, pipe_idx);
+
+	refcyc_per_pte_group_nom_l = get_refcyc_per_pte_group_nom_l_in_us(mode_lib, e2e_pipe_param, num_pipes,
+			pipe_idx) * refclk_freq_in_mhz;         // From VBA
+	refcyc_per_pte_group_nom_c = get_refcyc_per_pte_group_nom_c_in_us(mode_lib, e2e_pipe_param, num_pipes,
+			pipe_idx) * refclk_freq_in_mhz;         // From VBA
+	refcyc_per_pte_group_vblank_l = get_refcyc_per_pte_group_vblank_l_in_us(mode_lib, e2e_pipe_param,
+			num_pipes, pipe_idx) * refclk_freq_in_mhz;      // From VBA
+	refcyc_per_pte_group_vblank_c = get_refcyc_per_pte_group_vblank_c_in_us(mode_lib, e2e_pipe_param,
+			num_pipes, pipe_idx) * refclk_freq_in_mhz;      // From VBA
+	refcyc_per_pte_group_flip_l = get_refcyc_per_pte_group_flip_l_in_us(mode_lib, e2e_pipe_param, num_pipes,
+			pipe_idx) * refclk_freq_in_mhz;        // From VBA
+	refcyc_per_pte_group_flip_c = get_refcyc_per_pte_group_flip_c_in_us(mode_lib, e2e_pipe_param, num_pipes,
+			pipe_idx) * refclk_freq_in_mhz;        // From VBA
+
+	refcyc_per_meta_chunk_nom_l = get_refcyc_per_meta_chunk_nom_l_in_us(mode_lib, e2e_pipe_param, num_pipes,
+			pipe_idx) * refclk_freq_in_mhz;        // From VBA
+	refcyc_per_meta_chunk_nom_c = get_refcyc_per_meta_chunk_nom_c_in_us(mode_lib, e2e_pipe_param, num_pipes,
+			pipe_idx) * refclk_freq_in_mhz;        // From VBA
+	refcyc_per_meta_chunk_vblank_l = get_refcyc_per_meta_chunk_vblank_l_in_us(mode_lib, e2e_pipe_param,
+			num_pipes, pipe_idx) * refclk_freq_in_mhz;     // From VBA
+	refcyc_per_meta_chunk_vblank_c = get_refcyc_per_meta_chunk_vblank_c_in_us(mode_lib, e2e_pipe_param,
+			num_pipes, pipe_idx) * refclk_freq_in_mhz;     // From VBA
+	refcyc_per_meta_chunk_flip_l = get_refcyc_per_meta_chunk_flip_l_in_us(mode_lib, e2e_pipe_param,
+			num_pipes, pipe_idx) * refclk_freq_in_mhz;       // From VBA
+	refcyc_per_meta_chunk_flip_c = get_refcyc_per_meta_chunk_flip_c_in_us(mode_lib, e2e_pipe_param,
+			num_pipes, pipe_idx) * refclk_freq_in_mhz;       // From VBA
+
+	dlg_regs->dst_y_per_pte_row_nom_l = (unsigned int)(dst_y_per_pte_row_nom_l * dml_pow(2, 2));
+	dlg_regs->dst_y_per_pte_row_nom_c = (unsigned int)(dst_y_per_pte_row_nom_c * dml_pow(2, 2));
+	dlg_regs->dst_y_per_meta_row_nom_l = (unsigned int)(dst_y_per_meta_row_nom_l * dml_pow(2, 2));
+	dlg_regs->dst_y_per_meta_row_nom_c = (unsigned int)(dst_y_per_meta_row_nom_c * dml_pow(2, 2));
+	dlg_regs->refcyc_per_pte_group_nom_l = (unsigned int)refcyc_per_pte_group_nom_l;
+	dlg_regs->refcyc_per_pte_group_nom_c = (unsigned int)refcyc_per_pte_group_nom_c;
+	dlg_regs->refcyc_per_pte_group_vblank_l = (unsigned int)refcyc_per_pte_group_vblank_l;
+	dlg_regs->refcyc_per_pte_group_vblank_c = (unsigned int)refcyc_per_pte_group_vblank_c;
+	dlg_regs->refcyc_per_pte_group_flip_l = (unsigned int)refcyc_per_pte_group_flip_l;
+	dlg_regs->refcyc_per_pte_group_flip_c = (unsigned int)refcyc_per_pte_group_flip_c;
+	dlg_regs->refcyc_per_meta_chunk_nom_l = (unsigned int)refcyc_per_meta_chunk_nom_l;
+	dlg_regs->refcyc_per_meta_chunk_nom_c = (unsigned int)refcyc_per_meta_chunk_nom_c;
+	dlg_regs->refcyc_per_meta_chunk_vblank_l = (unsigned int)refcyc_per_meta_chunk_vblank_l;
+	dlg_regs->refcyc_per_meta_chunk_vblank_c = (unsigned int)refcyc_per_meta_chunk_vblank_c;
+	dlg_regs->refcyc_per_meta_chunk_flip_l = (unsigned int)refcyc_per_meta_chunk_flip_l;
+	dlg_regs->refcyc_per_meta_chunk_flip_c = (unsigned int)refcyc_per_meta_chunk_flip_c;
+	dlg_regs->refcyc_per_line_delivery_pre_l = (unsigned int) dml_floor(refcyc_per_line_delivery_pre_l, 1);
+	dlg_regs->refcyc_per_line_delivery_l = (unsigned int) dml_floor(refcyc_per_line_delivery_l, 1);
+	dlg_regs->refcyc_per_line_delivery_pre_c = (unsigned int) dml_floor(refcyc_per_line_delivery_pre_c, 1);
+	dlg_regs->refcyc_per_line_delivery_c = (unsigned int) dml_floor(refcyc_per_line_delivery_c, 1);
+
+	dlg_regs->chunk_hdl_adjust_cur0 = 3;
+	dlg_regs->dst_y_offset_cur0 = 0;
+	dlg_regs->chunk_hdl_adjust_cur1 = 3;
+	dlg_regs->dst_y_offset_cur1 = 0;
+
+	dlg_regs->dst_y_delta_drq_limit = 0x7fff; // off
+
+	ttu_regs->refcyc_per_req_delivery_pre_l = (unsigned int) (refcyc_per_req_delivery_pre_l * dml_pow(2, 10));
+	ttu_regs->refcyc_per_req_delivery_l = (unsigned int) (refcyc_per_req_delivery_l * dml_pow(2, 10));
+	ttu_regs->refcyc_per_req_delivery_pre_c = (unsigned int) (refcyc_per_req_delivery_pre_c * dml_pow(2, 10));
+	ttu_regs->refcyc_per_req_delivery_c = (unsigned int) (refcyc_per_req_delivery_c * dml_pow(2, 10));
+	ttu_regs->refcyc_per_req_delivery_pre_cur0 =
+			(unsigned int) (refcyc_per_req_delivery_pre_cur0 * dml_pow(2, 10));
+	ttu_regs->refcyc_per_req_delivery_cur0 = (unsigned int) (refcyc_per_req_delivery_cur0 * dml_pow(2, 10));
+	ttu_regs->refcyc_per_req_delivery_pre_cur1 = 0;
+	ttu_regs->refcyc_per_req_delivery_cur1 = 0;
+	ttu_regs->qos_level_low_wm = 0;
+
+	ttu_regs->qos_level_high_wm = (unsigned int) (4.0 * (double) htotal * ref_freq_to_pix_freq);
+
+	ttu_regs->qos_level_flip = 14;
+	ttu_regs->qos_level_fixed_l = 8;
+	ttu_regs->qos_level_fixed_c = 8;
+	ttu_regs->qos_level_fixed_cur0 = 8;
+	ttu_regs->qos_ramp_disable_l = 0;
+	ttu_regs->qos_ramp_disable_c = 0;
+	ttu_regs->qos_ramp_disable_cur0 = 0;
+	ttu_regs->min_ttu_vblank = (unsigned int)(min_ttu_vblank * refclk_freq_in_mhz);
+
+	// CHECK for HW registers' range, assert or clamp
+	ASSERT(refcyc_per_req_delivery_pre_l < dml_pow(2, 13));
+	ASSERT(refcyc_per_req_delivery_l < dml_pow(2, 13));
+	ASSERT(refcyc_per_req_delivery_pre_c < dml_pow(2, 13));
+	ASSERT(refcyc_per_req_delivery_c < dml_pow(2, 13));
+	if (dlg_regs->refcyc_per_vm_group_vblank >= (unsigned int) dml_pow(2, 23))
+		dlg_regs->refcyc_per_vm_group_vblank = (unsigned int)(dml_pow(2, 23) - 1);
+
+	if (dlg_regs->refcyc_per_vm_group_flip >= (unsigned int) dml_pow(2, 23))
+		dlg_regs->refcyc_per_vm_group_flip = (unsigned int)(dml_pow(2, 23) - 1);
+
+	if (dlg_regs->refcyc_per_vm_req_vblank >= (unsigned int) dml_pow(2, 23))
+		dlg_regs->refcyc_per_vm_req_vblank = (unsigned int)(dml_pow(2, 23) - 1);
+
+	if (dlg_regs->refcyc_per_vm_req_flip >= (unsigned int) dml_pow(2, 23))
+		dlg_regs->refcyc_per_vm_req_flip = (unsigned int)(dml_pow(2, 23) - 1);
+
+	ASSERT(dlg_regs->dst_y_after_scaler < (unsigned int) 8);
+	ASSERT(dlg_regs->refcyc_x_after_scaler < (unsigned int)dml_pow(2, 13));
+	ASSERT(dlg_regs->dst_y_per_pte_row_nom_l < (unsigned int)dml_pow(2, 17));
+	if (dual_plane) {
+		if (dlg_regs->dst_y_per_pte_row_nom_c >= (unsigned int) dml_pow(2, 17)) {
+			// FIXME what so special about chroma, can we just assert?
+			dml_print("DML_DLG: %s: Warning dst_y_per_pte_row_nom_c %u > register max U15.2 %u\n",
+					__func__, dlg_regs->dst_y_per_pte_row_nom_c, (unsigned int)dml_pow(2, 17) - 1);
+		}
+	}
+	ASSERT(dlg_regs->dst_y_per_meta_row_nom_l < (unsigned int)dml_pow(2, 17));
+	ASSERT(dlg_regs->dst_y_per_meta_row_nom_c < (unsigned int)dml_pow(2, 17));
+
+	if (dlg_regs->refcyc_per_pte_group_nom_l >= (unsigned int) dml_pow(2, 23))
+		dlg_regs->refcyc_per_pte_group_nom_l = (unsigned int)(dml_pow(2, 23) - 1);
+	if (dual_plane) {
+		if (dlg_regs->refcyc_per_pte_group_nom_c >= (unsigned int) dml_pow(2, 23))
+			dlg_regs->refcyc_per_pte_group_nom_c = (unsigned int)(dml_pow(2, 23) - 1);
+	}
+	ASSERT(dlg_regs->refcyc_per_pte_group_vblank_l < (unsigned int)dml_pow(2, 13));
+	if (dual_plane) {
+		ASSERT(dlg_regs->refcyc_per_pte_group_vblank_c < (unsigned int)dml_pow(2, 13));
+	}
+
+	if (dlg_regs->refcyc_per_meta_chunk_nom_l >= (unsigned int) dml_pow(2, 23))
+		dlg_regs->refcyc_per_meta_chunk_nom_l = (unsigned int)(dml_pow(2, 23) - 1);
+	if (dual_plane) {
+		if (dlg_regs->refcyc_per_meta_chunk_nom_c >= (unsigned int) dml_pow(2, 23))
+			dlg_regs->refcyc_per_meta_chunk_nom_c = (unsigned int)(dml_pow(2, 23) - 1);
+	}
+	ASSERT(dlg_regs->refcyc_per_meta_chunk_vblank_l < (unsigned int)dml_pow(2, 13));
+	ASSERT(dlg_regs->refcyc_per_meta_chunk_vblank_c < (unsigned int)dml_pow(2, 13));
+	ASSERT(dlg_regs->refcyc_per_line_delivery_pre_l < (unsigned int)dml_pow(2, 13));
+	ASSERT(dlg_regs->refcyc_per_line_delivery_l < (unsigned int)dml_pow(2, 13));
+	ASSERT(dlg_regs->refcyc_per_line_delivery_pre_c < (unsigned int)dml_pow(2, 13));
+	ASSERT(dlg_regs->refcyc_per_line_delivery_c < (unsigned int)dml_pow(2, 13));
+	ASSERT(ttu_regs->qos_level_low_wm < dml_pow(2, 14));
+	ASSERT(ttu_regs->qos_level_high_wm < dml_pow(2, 14));
+	ASSERT(ttu_regs->min_ttu_vblank < dml_pow(2, 24));
+
+	print__ttu_regs_st(mode_lib, ttu_regs);
+	print__dlg_regs_st(mode_lib, dlg_regs);
+	dml_print("DML_DLG::%s: Calculation for pipe[%d] done, num_pipes=%d\n", __func__, pipe_idx, num_pipes);
+}
+

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_rq_dlg_calc_32.h

@@ -0,0 +1,70 @@
+/*
+ * Copyright 2022 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 __DML32_DISPLAY_RQ_DLG_CALC_H__
+#define __DML32_DISPLAY_RQ_DLG_CALC_H__
+
+#include "../display_rq_dlg_helpers.h"
+
+struct display_mode_lib;
+
+/*
+* Function: dml_rq_dlg_get_rq_reg
+*  Main entry point for test to get the register values out of this DML class.
+*  This function calls <get_rq_param> and <extract_rq_regs> functions to calculate
+*  and then populate the rq_regs struct
+* Input:
+*  pipe_param - pipe source configuration (e.g. vp, pitch, scaling, dest, etc.)
+* Output:
+*  rq_regs - struct that holds all the RQ registers field value.
+*            See also: <display_rq_regs_st>
+*/
+void dml32_rq_dlg_get_rq_reg(display_rq_regs_st *rq_regs,
+		struct display_mode_lib *mode_lib,
+		const display_e2e_pipe_params_st *e2e_pipe_param,
+		const unsigned int num_pipes,
+		const unsigned int pipe_idx);
+
+/*
+* Function: dml_rq_dlg_get_dlg_reg
+*   Calculate and return DLG and TTU register struct given the system setting
+* Output:
+*  dlg_regs - output DLG register struct
+*  ttu_regs - output DLG TTU register struct
+* Input:
+*  e2e_pipe_param - "compacted" array of e2e pipe param struct
+*  num_pipes - num of active "pipe" or "route"
+*  pipe_idx - index that identifies the e2e_pipe_param that corresponding to this dlg
+*  cstate - 0: when calculate min_ttu_vblank it is assumed cstate is not required. 1: Normal mode, cstate is considered.
+*           Added for legacy or unrealistic timing tests.
+*/
+void dml32_rq_dlg_get_dlg_reg(struct display_mode_lib *mode_lib,
+		display_dlg_regs_st *dlg_regs,
+		display_ttu_regs_st *ttu_regs,
+		display_e2e_pipe_params_st *e2e_pipe_param,
+		const unsigned int num_pipes,
+		const unsigned int pipe_idx);
+
+#endif

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c

@@ -0,0 +1,934 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright 2022 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 "clk_mgr.h"
+#include "resource.h"
+#include "dcn321_fpu.h"
+#include "dcn32/dcn32_resource.h"
+#include "dcn321/dcn321_resource.h"
+#include "dml/dcn32/display_mode_vba_util_32.h"
+
+#define DCN3_2_DEFAULT_DET_SIZE 256
+
+struct _vcs_dpi_ip_params_st dcn3_21_ip = {
+	.gpuvm_enable = 0,
+	.gpuvm_max_page_table_levels = 4,
+	.hostvm_enable = 0,
+	.rob_buffer_size_kbytes = 128,
+	.det_buffer_size_kbytes = DCN3_2_DEFAULT_DET_SIZE,
+	.config_return_buffer_size_in_kbytes = 1280,
+	.compressed_buffer_segment_size_in_kbytes = 64,
+	.meta_fifo_size_in_kentries = 22,
+	.zero_size_buffer_entries = 512,
+	.compbuf_reserved_space_64b = 256,
+	.compbuf_reserved_space_zs = 64,
+	.dpp_output_buffer_pixels = 2560,
+	.opp_output_buffer_lines = 1,
+	.pixel_chunk_size_kbytes = 8,
+	.alpha_pixel_chunk_size_kbytes = 4,
+	.min_pixel_chunk_size_bytes = 1024,
+	.dcc_meta_buffer_size_bytes = 6272,
+	.meta_chunk_size_kbytes = 2,
+	.min_meta_chunk_size_bytes = 256,
+	.writeback_chunk_size_kbytes = 8,
+	.ptoi_supported = false,
+	.num_dsc = 4,
+	.maximum_dsc_bits_per_component = 12,
+	.maximum_pixels_per_line_per_dsc_unit = 6016,
+	.dsc422_native_support = true,
+	.is_line_buffer_bpp_fixed = true,
+	.line_buffer_fixed_bpp = 57,
+	.line_buffer_size_bits = 1171920,
+	.max_line_buffer_lines = 32,
+	.writeback_interface_buffer_size_kbytes = 90,
+	.max_num_dpp = 4,
+	.max_num_otg = 4,
+	.max_num_hdmi_frl_outputs = 1,
+	.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,
+	.max_hscl_taps = 8,
+	.max_vscl_taps = 8,
+	.dpte_buffer_size_in_pte_reqs_luma = 64,
+	.dpte_buffer_size_in_pte_reqs_chroma = 34,
+	.dispclk_ramp_margin_percent = 1,
+	.max_inter_dcn_tile_repeaters = 8,
+	.cursor_buffer_size = 16,
+	.cursor_chunk_size = 2,
+	.writeback_line_buffer_buffer_size = 0,
+	.writeback_min_hscl_ratio = 1,
+	.writeback_min_vscl_ratio = 1,
+	.writeback_max_hscl_ratio = 1,
+	.writeback_max_vscl_ratio = 1,
+	.writeback_max_hscl_taps = 1,
+	.writeback_max_vscl_taps = 1,
+	.dppclk_delay_subtotal = 47,
+	.dppclk_delay_scl = 50,
+	.dppclk_delay_scl_lb_only = 16,
+	.dppclk_delay_cnvc_formatter = 28,
+	.dppclk_delay_cnvc_cursor = 6,
+	.dispclk_delay_subtotal = 125,
+	.dynamic_metadata_vm_enabled = false,
+	.odm_combine_4to1_supported = false,
+	.dcc_supported = true,
+	.max_num_dp2p0_outputs = 2,
+	.max_num_dp2p0_streams = 4,
+};
+
+struct _vcs_dpi_soc_bounding_box_st dcn3_21_soc = {
+	.clock_limits = {
+		{
+			.state = 0,
+			.dcfclk_mhz = 1434.0,
+			.fabricclk_mhz = 2250.0,
+			.dispclk_mhz = 1720.0,
+			.dppclk_mhz = 1720.0,
+			.phyclk_mhz = 810.0,
+			.phyclk_d18_mhz = 667.0,
+			.phyclk_d32_mhz = 313.0,
+			.socclk_mhz = 1200.0,
+			.dscclk_mhz = 573.333,
+			.dram_speed_mts = 16000.0,
+			.dtbclk_mhz = 1564.0,
+		},
+	},
+	.num_states = 1,
+	.sr_exit_time_us = 19.95,
+	.sr_enter_plus_exit_time_us = 24.36,
+	.sr_exit_z8_time_us = 285.0,
+	.sr_enter_plus_exit_z8_time_us = 320,
+	.writeback_latency_us = 12.0,
+	.round_trip_ping_latency_dcfclk_cycles = 207,
+	.urgent_latency_pixel_data_only_us = 4,
+	.urgent_latency_pixel_mixed_with_vm_data_us = 4,
+	.urgent_latency_vm_data_only_us = 4,
+	.fclk_change_latency_us = 7,
+	.usr_retraining_latency_us = 0,
+	.smn_latency_us = 0,
+	.mall_allocated_for_dcn_mbytes = 32,
+	.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_sdp_bw_after_urgent = 90.0,
+	.pct_ideal_fabric_bw_after_urgent = 67.0,
+	.pct_ideal_dram_sdp_bw_after_urgent_pixel_only = 20.0,
+	.pct_ideal_dram_sdp_bw_after_urgent_pixel_and_vm = 60.0,
+	.pct_ideal_dram_sdp_bw_after_urgent_vm_only = 30.0,
+	.pct_ideal_dram_bw_after_urgent_strobe = 67.0,
+	.max_avg_sdp_bw_use_normal_percent = 80.0,
+	.max_avg_fabric_bw_use_normal_percent = 60.0,
+	.max_avg_dram_bw_use_normal_strobe_percent = 50.0,
+	.max_avg_dram_bw_use_normal_percent = 15.0,
+	.num_chans = 8,
+	.dram_channel_width_bytes = 2,
+	.fabric_datapath_to_dcn_data_return_bytes = 64,
+	.return_bus_width_bytes = 64,
+	.downspread_percent = 0.38,
+	.dcn_downspread_percent = 0.5,
+	.dram_clock_change_latency_us = 400,
+	.dispclk_dppclk_vco_speed_mhz = 4300.0,
+	.do_urgent_latency_adjustment = true,
+	.urgent_latency_adjustment_fabric_clock_component_us = 1.0,
+	.urgent_latency_adjustment_fabric_clock_reference_mhz = 3000,
+};
+
+static void get_optimal_ntuple(struct _vcs_dpi_voltage_scaling_st *entry)
+{
+	if (entry->dcfclk_mhz > 0) {
+		float bw_on_sdp = (float)(entry->dcfclk_mhz * dcn3_21_soc.return_bus_width_bytes * ((float)dcn3_21_soc.pct_ideal_sdp_bw_after_urgent / 100));
+
+		entry->fabricclk_mhz = bw_on_sdp / (dcn3_21_soc.return_bus_width_bytes * ((float)dcn3_21_soc.pct_ideal_fabric_bw_after_urgent / 100));
+		entry->dram_speed_mts = bw_on_sdp / (dcn3_21_soc.num_chans *
+				dcn3_21_soc.dram_channel_width_bytes * ((float)dcn3_21_soc.pct_ideal_dram_sdp_bw_after_urgent_pixel_only / 100));
+	} else if (entry->fabricclk_mhz > 0) {
+		float bw_on_fabric = (float)(entry->fabricclk_mhz * dcn3_21_soc.return_bus_width_bytes * ((float)dcn3_21_soc.pct_ideal_fabric_bw_after_urgent / 100));
+
+		entry->dcfclk_mhz = bw_on_fabric / (dcn3_21_soc.return_bus_width_bytes * ((float)dcn3_21_soc.pct_ideal_sdp_bw_after_urgent / 100));
+		entry->dram_speed_mts = bw_on_fabric / (dcn3_21_soc.num_chans *
+				dcn3_21_soc.dram_channel_width_bytes * ((float)dcn3_21_soc.pct_ideal_dram_sdp_bw_after_urgent_pixel_only / 100));
+	} else if (entry->dram_speed_mts > 0) {
+		float bw_on_dram = (float)(entry->dram_speed_mts * dcn3_21_soc.num_chans *
+				dcn3_21_soc.dram_channel_width_bytes * ((float)dcn3_21_soc.pct_ideal_dram_sdp_bw_after_urgent_pixel_only / 100));
+
+		entry->fabricclk_mhz = bw_on_dram / (dcn3_21_soc.return_bus_width_bytes * ((float)dcn3_21_soc.pct_ideal_fabric_bw_after_urgent / 100));
+		entry->dcfclk_mhz = bw_on_dram / (dcn3_21_soc.return_bus_width_bytes * ((float)dcn3_21_soc.pct_ideal_sdp_bw_after_urgent / 100));
+	}
+}
+
+static float calculate_net_bw_in_kbytes_sec(struct _vcs_dpi_voltage_scaling_st *entry)
+{
+	float memory_bw_kbytes_sec;
+	float fabric_bw_kbytes_sec;
+	float sdp_bw_kbytes_sec;
+	float limiting_bw_kbytes_sec;
+
+	memory_bw_kbytes_sec = (float)(entry->dram_speed_mts * dcn3_21_soc.num_chans *
+			dcn3_21_soc.dram_channel_width_bytes * ((float)dcn3_21_soc.pct_ideal_dram_sdp_bw_after_urgent_pixel_only / 100));
+
+	fabric_bw_kbytes_sec = (float)(entry->fabricclk_mhz * dcn3_21_soc.return_bus_width_bytes * ((float)dcn3_21_soc.pct_ideal_fabric_bw_after_urgent / 100));
+
+	sdp_bw_kbytes_sec = (float)(entry->dcfclk_mhz * dcn3_21_soc.return_bus_width_bytes * ((float)dcn3_21_soc.pct_ideal_sdp_bw_after_urgent / 100));
+
+	limiting_bw_kbytes_sec = memory_bw_kbytes_sec;
+
+	if (fabric_bw_kbytes_sec < limiting_bw_kbytes_sec)
+		limiting_bw_kbytes_sec = fabric_bw_kbytes_sec;
+
+	if (sdp_bw_kbytes_sec < limiting_bw_kbytes_sec)
+		limiting_bw_kbytes_sec = sdp_bw_kbytes_sec;
+
+	return limiting_bw_kbytes_sec;
+}
+
+static void dcn321_insert_entry_into_table_sorted(struct _vcs_dpi_voltage_scaling_st *table,
+					   unsigned int *num_entries,
+					   struct _vcs_dpi_voltage_scaling_st *entry)
+{
+	unsigned int i = 0;
+	unsigned int index = 0;
+
+	dc_assert_fp_enabled();
+
+	if (*num_entries == 0) {
+		table[0] = *entry;
+		(*num_entries)++;
+	} else {
+		while (entry->net_bw_in_kbytes_sec > table[index].net_bw_in_kbytes_sec) {
+			index++;
+			if (index >= *num_entries)
+				break;
+		}
+
+		for (i = *num_entries; i > index; i--)
+			table[i] = table[i - 1];
+
+		table[index] = *entry;
+		(*num_entries)++;
+	}
+}
+
+static void remove_entry_from_table_at_index(struct _vcs_dpi_voltage_scaling_st *table, unsigned int *num_entries,
+		unsigned int index)
+{
+	unsigned int i;
+
+	if (*num_entries == 0)
+		return;
+
+	for (i = index; i < *num_entries - 1; i++) {
+		table[i] = table[i + 1];
+	}
+	memset(&table[--(*num_entries)], 0, sizeof(struct _vcs_dpi_voltage_scaling_st));
+}
+
+static void swap_table_entries(struct _vcs_dpi_voltage_scaling_st *first_entry,
+		struct _vcs_dpi_voltage_scaling_st *second_entry)
+{
+	struct _vcs_dpi_voltage_scaling_st temp_entry = *first_entry;
+	*first_entry = *second_entry;
+	*second_entry = temp_entry;
+}
+
+/*
+ * sort_entries_with_same_bw - Sort entries sharing the same bandwidth by DCFCLK
+ */
+static void sort_entries_with_same_bw(struct _vcs_dpi_voltage_scaling_st *table, unsigned int *num_entries)
+{
+	unsigned int start_index = 0;
+	unsigned int end_index = 0;
+	unsigned int current_bw = 0;
+
+	for (unsigned int i = 0; i + 1 < *num_entries; i++) {
+		if (table[i].net_bw_in_kbytes_sec == table[i+1].net_bw_in_kbytes_sec) {
+			current_bw = (unsigned int)table[i].net_bw_in_kbytes_sec;
+			start_index = i;
+			end_index = ++i;
+
+			while ((i + 1 < *num_entries) && (table[i+1].net_bw_in_kbytes_sec == current_bw))
+				end_index = ++i;
+		}
+
+		if (start_index != end_index) {
+			for (unsigned int j = start_index; j < end_index; j++) {
+				for (unsigned int k = start_index; k < end_index; k++) {
+					if (table[k].dcfclk_mhz > table[k+1].dcfclk_mhz)
+						swap_table_entries(&table[k], &table[k+1]);
+				}
+			}
+		}
+
+		start_index = 0;
+		end_index = 0;
+
+	}
+}
+
+/*
+ * remove_inconsistent_entries - Ensure entries with the same bandwidth have MEMCLK and FCLK monotonically increasing
+ *                               and remove entries that do not follow this order
+ */
+static void remove_inconsistent_entries(struct _vcs_dpi_voltage_scaling_st *table, unsigned int *num_entries)
+{
+	for (unsigned int i = 0; i + 1 < *num_entries; i++) {
+		if (table[i].net_bw_in_kbytes_sec == table[i+1].net_bw_in_kbytes_sec) {
+			if ((table[i].dram_speed_mts > table[i+1].dram_speed_mts) ||
+				(table[i].fabricclk_mhz > table[i+1].fabricclk_mhz))
+				remove_entry_from_table_at_index(table, num_entries, i);
+		}
+	}
+}
+
+/*
+ * override_max_clk_values - Overwrite the max clock frequencies with the max DC mode timings
+ * Input:
+ *	max_clk_limit - struct containing the desired clock timings
+ * Output:
+ *	curr_clk_limit  - struct containing the timings that need to be overwritten
+ * Return: 0 upon success, non-zero for failure
+ */
+static int override_max_clk_values(struct clk_limit_table_entry *max_clk_limit,
+		struct clk_limit_table_entry *curr_clk_limit)
+{
+	if (NULL == max_clk_limit || NULL == curr_clk_limit)
+		return -1; //invalid parameters
+
+	//only overwrite if desired max clock frequency is initialized
+	if (max_clk_limit->dcfclk_mhz != 0)
+		curr_clk_limit->dcfclk_mhz = max_clk_limit->dcfclk_mhz;
+
+	if (max_clk_limit->fclk_mhz != 0)
+		curr_clk_limit->fclk_mhz = max_clk_limit->fclk_mhz;
+
+	if (max_clk_limit->memclk_mhz != 0)
+		curr_clk_limit->memclk_mhz = max_clk_limit->memclk_mhz;
+
+	if (max_clk_limit->socclk_mhz != 0)
+		curr_clk_limit->socclk_mhz = max_clk_limit->socclk_mhz;
+
+	if (max_clk_limit->dtbclk_mhz != 0)
+		curr_clk_limit->dtbclk_mhz = max_clk_limit->dtbclk_mhz;
+
+	if (max_clk_limit->dispclk_mhz != 0)
+		curr_clk_limit->dispclk_mhz = max_clk_limit->dispclk_mhz;
+
+	return 0;
+}
+
+static int build_synthetic_soc_states(bool disable_dc_mode_overwrite, struct clk_bw_params *bw_params,
+		struct _vcs_dpi_voltage_scaling_st *table, unsigned int *num_entries)
+{
+	int i;
+	unsigned int j, k;
+	struct _vcs_dpi_voltage_scaling_st entry = {0};
+	struct clk_limit_table_entry max_clk_data = {0};
+
+	unsigned int min_dcfclk_mhz = 199, min_fclk_mhz = 299;
+
+	static const unsigned int num_dcfclk_stas = 5;
+	unsigned int dcfclk_sta_targets[DC__VOLTAGE_STATES] = {199, 615, 906, 1324, 1564};
+
+	unsigned int num_uclk_dpms = 0;
+	unsigned int num_fclk_dpms = 0;
+	unsigned int num_dcfclk_dpms = 0;
+
+	unsigned int num_dc_uclk_dpms = 0;
+	unsigned int num_dc_fclk_dpms = 0;
+	unsigned int num_dc_dcfclk_dpms = 0;
+
+	for (i = 0; i < MAX_NUM_DPM_LVL; i++) {
+		if (bw_params->clk_table.entries[i].dcfclk_mhz > max_clk_data.dcfclk_mhz)
+			max_clk_data.dcfclk_mhz = bw_params->clk_table.entries[i].dcfclk_mhz;
+		if (bw_params->clk_table.entries[i].fclk_mhz > max_clk_data.fclk_mhz)
+			max_clk_data.fclk_mhz = bw_params->clk_table.entries[i].fclk_mhz;
+		if (bw_params->clk_table.entries[i].memclk_mhz > max_clk_data.memclk_mhz)
+			max_clk_data.memclk_mhz = bw_params->clk_table.entries[i].memclk_mhz;
+		if (bw_params->clk_table.entries[i].dispclk_mhz > max_clk_data.dispclk_mhz)
+			max_clk_data.dispclk_mhz = bw_params->clk_table.entries[i].dispclk_mhz;
+		if (bw_params->clk_table.entries[i].dppclk_mhz > max_clk_data.dppclk_mhz)
+			max_clk_data.dppclk_mhz = bw_params->clk_table.entries[i].dppclk_mhz;
+		if (bw_params->clk_table.entries[i].phyclk_mhz > max_clk_data.phyclk_mhz)
+			max_clk_data.phyclk_mhz = bw_params->clk_table.entries[i].phyclk_mhz;
+		if (bw_params->clk_table.entries[i].dtbclk_mhz > max_clk_data.dtbclk_mhz)
+			max_clk_data.dtbclk_mhz = bw_params->clk_table.entries[i].dtbclk_mhz;
+
+		if (bw_params->clk_table.entries[i].memclk_mhz > 0) {
+			num_uclk_dpms++;
+			if (bw_params->clk_table.entries[i].memclk_mhz <= bw_params->dc_mode_limit.memclk_mhz)
+				num_dc_uclk_dpms++;
+		}
+		if (bw_params->clk_table.entries[i].fclk_mhz > 0) {
+			num_fclk_dpms++;
+			if (bw_params->clk_table.entries[i].fclk_mhz <= bw_params->dc_mode_limit.fclk_mhz)
+				num_dc_fclk_dpms++;
+		}
+		if (bw_params->clk_table.entries[i].dcfclk_mhz > 0) {
+			num_dcfclk_dpms++;
+			if (bw_params->clk_table.entries[i].dcfclk_mhz <= bw_params->dc_mode_limit.dcfclk_mhz)
+				num_dc_dcfclk_dpms++;
+		}
+	}
+
+	if (!disable_dc_mode_overwrite) {
+		//Overwrite max frequencies with max DC mode frequencies for DC mode systems
+		override_max_clk_values(&bw_params->dc_mode_limit, &max_clk_data);
+		num_uclk_dpms = num_dc_uclk_dpms;
+		num_fclk_dpms = num_dc_fclk_dpms;
+		num_dcfclk_dpms = num_dc_dcfclk_dpms;
+		bw_params->clk_table.num_entries_per_clk.num_memclk_levels = num_uclk_dpms;
+		bw_params->clk_table.num_entries_per_clk.num_fclk_levels = num_fclk_dpms;
+	}
+
+	if (num_dcfclk_dpms > 0 && bw_params->clk_table.entries[0].fclk_mhz > min_fclk_mhz)
+		min_fclk_mhz = bw_params->clk_table.entries[0].fclk_mhz;
+
+	if (!max_clk_data.dcfclk_mhz || !max_clk_data.dispclk_mhz || !max_clk_data.dtbclk_mhz)
+		return -1;
+
+	if (max_clk_data.dppclk_mhz == 0)
+		max_clk_data.dppclk_mhz = max_clk_data.dispclk_mhz;
+
+	if (max_clk_data.fclk_mhz == 0)
+		max_clk_data.fclk_mhz = (unsigned int)(max_clk_data.dcfclk_mhz *
+				dcn3_21_soc.pct_ideal_sdp_bw_after_urgent /
+				dcn3_21_soc.pct_ideal_fabric_bw_after_urgent);
+
+	if (max_clk_data.phyclk_mhz == 0)
+		max_clk_data.phyclk_mhz = (unsigned int)dcn3_21_soc.clock_limits[0].phyclk_mhz;
+
+	*num_entries = 0;
+	entry.dispclk_mhz = max_clk_data.dispclk_mhz;
+	entry.dscclk_mhz = max_clk_data.dispclk_mhz / 3;
+	entry.dppclk_mhz = max_clk_data.dppclk_mhz;
+	entry.dtbclk_mhz = max_clk_data.dtbclk_mhz;
+	entry.phyclk_mhz = max_clk_data.phyclk_mhz;
+	entry.phyclk_d18_mhz = dcn3_21_soc.clock_limits[0].phyclk_d18_mhz;
+	entry.phyclk_d32_mhz = dcn3_21_soc.clock_limits[0].phyclk_d32_mhz;
+
+	// Insert all the DCFCLK STAs
+	for (k = 0; k < num_dcfclk_stas; k++) {
+		entry.dcfclk_mhz = dcfclk_sta_targets[k];
+		entry.fabricclk_mhz = 0;
+		entry.dram_speed_mts = 0;
+
+		get_optimal_ntuple(&entry);
+		entry.net_bw_in_kbytes_sec = calculate_net_bw_in_kbytes_sec(&entry);
+		dcn321_insert_entry_into_table_sorted(table, num_entries, &entry);
+	}
+
+	// Insert the max DCFCLK
+	entry.dcfclk_mhz = max_clk_data.dcfclk_mhz;
+	entry.fabricclk_mhz = 0;
+	entry.dram_speed_mts = 0;
+
+	get_optimal_ntuple(&entry);
+	entry.net_bw_in_kbytes_sec = calculate_net_bw_in_kbytes_sec(&entry);
+	dcn321_insert_entry_into_table_sorted(table, num_entries, &entry);
+
+	// Insert the UCLK DPMS
+	for (k = 0; k < num_uclk_dpms; k++) {
+		entry.dcfclk_mhz = 0;
+		entry.fabricclk_mhz = 0;
+		entry.dram_speed_mts = bw_params->clk_table.entries[k].memclk_mhz * 16;
+
+		get_optimal_ntuple(&entry);
+		entry.net_bw_in_kbytes_sec = calculate_net_bw_in_kbytes_sec(&entry);
+		dcn321_insert_entry_into_table_sorted(table, num_entries, &entry);
+	}
+
+	// If FCLK is coarse grained, insert individual DPMs.
+	if (num_fclk_dpms > 2) {
+		for (k = 0; k < num_fclk_dpms; k++) {
+			entry.dcfclk_mhz = 0;
+			entry.fabricclk_mhz = bw_params->clk_table.entries[k].fclk_mhz;
+			entry.dram_speed_mts = 0;
+
+			get_optimal_ntuple(&entry);
+			entry.net_bw_in_kbytes_sec = calculate_net_bw_in_kbytes_sec(&entry);
+			dcn321_insert_entry_into_table_sorted(table, num_entries, &entry);
+		}
+	}
+	// If FCLK fine grained, only insert max
+	else {
+		entry.dcfclk_mhz = 0;
+		entry.fabricclk_mhz = max_clk_data.fclk_mhz;
+		entry.dram_speed_mts = 0;
+
+		get_optimal_ntuple(&entry);
+		entry.net_bw_in_kbytes_sec = calculate_net_bw_in_kbytes_sec(&entry);
+		dcn321_insert_entry_into_table_sorted(table, num_entries, &entry);
+	}
+
+	// At this point, the table contains all "points of interest" based on
+	// DPMs from PMFW, and STAs.  Table is sorted by BW, and all clock
+	// ratios (by derate, are exact).
+
+	// Remove states that require higher clocks than are supported
+	for (i = (int)*num_entries - 1; i >= 0 ; i--) {
+		if (table[i].dcfclk_mhz > max_clk_data.dcfclk_mhz ||
+				table[i].fabricclk_mhz > max_clk_data.fclk_mhz ||
+				table[i].dram_speed_mts > max_clk_data.memclk_mhz * 16)
+			remove_entry_from_table_at_index(table, num_entries, i);
+	}
+
+	// Insert entry with all max dc limits without bandwitch matching
+	if (!disable_dc_mode_overwrite) {
+		struct _vcs_dpi_voltage_scaling_st max_dc_limits_entry = entry;
+
+		max_dc_limits_entry.dcfclk_mhz = max_clk_data.dcfclk_mhz;
+		max_dc_limits_entry.fabricclk_mhz = max_clk_data.fclk_mhz;
+		max_dc_limits_entry.dram_speed_mts = max_clk_data.memclk_mhz * 16;
+
+		max_dc_limits_entry.net_bw_in_kbytes_sec = calculate_net_bw_in_kbytes_sec(&max_dc_limits_entry);
+		dcn321_insert_entry_into_table_sorted(table, num_entries, &max_dc_limits_entry);
+
+		sort_entries_with_same_bw(table, num_entries);
+		remove_inconsistent_entries(table, num_entries);
+	}
+
+
+
+	// At this point, the table only contains supported points of interest
+	// it could be used as is, but some states may be redundant due to
+	// coarse grained nature of some clocks, so we want to round up to
+	// coarse grained DPMs and remove duplicates.
+
+	// Round up UCLKs
+	for (i = (int)*num_entries - 1; i >= 0 ; i--) {
+		for (j = 0; j < num_uclk_dpms; j++) {
+			if (bw_params->clk_table.entries[j].memclk_mhz * 16 >= table[i].dram_speed_mts) {
+				table[i].dram_speed_mts = bw_params->clk_table.entries[j].memclk_mhz * 16;
+				break;
+			}
+		}
+	}
+
+	// If FCLK is coarse grained, round up to next DPMs
+	if (num_fclk_dpms > 2) {
+		for (i = (int)*num_entries - 1; i >= 0 ; i--) {
+			for (j = 0; j < num_fclk_dpms; j++) {
+				if (bw_params->clk_table.entries[j].fclk_mhz >= table[i].fabricclk_mhz) {
+					table[i].fabricclk_mhz = bw_params->clk_table.entries[j].fclk_mhz;
+					break;
+				}
+			}
+		}
+	}
+	// Otherwise, round up to minimum.
+	else {
+		for (i = (int)*num_entries - 1; i >= 0 ; i--) {
+			if (table[i].fabricclk_mhz < min_fclk_mhz) {
+				table[i].fabricclk_mhz = min_fclk_mhz;
+			}
+		}
+	}
+
+	// Round DCFCLKs up to minimum
+	for (i = (int)*num_entries - 1; i >= 0 ; i--) {
+		if (table[i].dcfclk_mhz < min_dcfclk_mhz) {
+			table[i].dcfclk_mhz = min_dcfclk_mhz;
+		}
+	}
+
+	// Remove duplicate states, note duplicate states are always neighbouring since table is sorted.
+	i = 0;
+	while (i < (int)*num_entries - 1) {
+		if (table[i].dcfclk_mhz == table[i + 1].dcfclk_mhz &&
+				table[i].fabricclk_mhz == table[i + 1].fabricclk_mhz &&
+				table[i].dram_speed_mts == table[i + 1].dram_speed_mts)
+			remove_entry_from_table_at_index(table, num_entries, i + 1);
+		else
+			i++;
+	}
+
+	// Fix up the state indicies
+	for (i = (int)*num_entries - 1; i >= 0 ; i--)
+		table[i].state = i;
+
+	return 0;
+}
+
+static void dcn321_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_21_soc.num_chans *
+		dcn3_21_soc.dram_channel_width_bytes * (dcn3_21_soc.max_avg_dram_bw_use_normal_percent / 100);
+	bw_from_dram2 = uclk_mts * dcn3_21_soc.num_chans *
+		dcn3_21_soc.dram_channel_width_bytes * (dcn3_21_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 = (unsigned int)(bw_from_dram /
+		(dcn3_21_soc.fabric_datapath_to_dcn_data_return_bytes * (dcn3_21_soc.max_avg_sdp_bw_use_normal_percent / 100)));
+
+	if (optimal_dcfclk)
+		*optimal_dcfclk =  (unsigned int)(bw_from_dram /
+		(dcn3_21_soc.return_bus_width_bytes * (dcn3_21_soc.max_avg_sdp_bw_use_normal_percent / 100)));
+}
+
+/** dcn321_update_bw_bounding_box
+ * This would override some dcn3_2 ip_or_soc initial parameters hardcoded from spreadsheet
+ * with actual values as per dGPU SKU:
+ * -with passed few options from dc->config
+ * -with dentist_vco_frequency from Clk Mgr (currently hardcoded, but might need to get it from PM FW)
+ * -with passed latency values (passed in ns units) in dc-> bb override for debugging purposes
+ * -with passed latencies from VBIOS (in 100_ns units) if available for certain dGPU SKU
+ * -with number of DRAM channels from VBIOS (which differ for certain dGPU SKU of the same ASIC)
+ * -clocks levels with passed clk_table entries from Clk Mgr as reported by PM FW for different
+ *  clocks (which might differ for certain dGPU SKU of the same ASIC)
+ */
+void dcn321_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_params)
+{
+	dc_assert_fp_enabled();
+	/* Overrides from dc->config options */
+	dcn3_21_ip.clamp_min_dcfclk = dc->config.clamp_min_dcfclk;
+
+	/* Override from passed dc->bb_overrides if available*/
+	if ((int)(dcn3_21_soc.sr_exit_time_us * 1000) != dc->bb_overrides.sr_exit_time_ns
+			&& dc->bb_overrides.sr_exit_time_ns) {
+		dc->dml2_options.bbox_overrides.sr_exit_latency_us =
+		dcn3_21_soc.sr_exit_time_us = dc->bb_overrides.sr_exit_time_ns / 1000.0;
+	}
+
+	if ((int)(dcn3_21_soc.sr_enter_plus_exit_time_us * 1000)
+			!= dc->bb_overrides.sr_enter_plus_exit_time_ns
+			&& dc->bb_overrides.sr_enter_plus_exit_time_ns) {
+		dc->dml2_options.bbox_overrides.sr_enter_plus_exit_latency_us =
+		dcn3_21_soc.sr_enter_plus_exit_time_us =
+			dc->bb_overrides.sr_enter_plus_exit_time_ns / 1000.0;
+	}
+
+	if ((int)(dcn3_21_soc.urgent_latency_us * 1000) != dc->bb_overrides.urgent_latency_ns
+		&& dc->bb_overrides.urgent_latency_ns) {
+		dcn3_21_soc.urgent_latency_us = dc->bb_overrides.urgent_latency_ns / 1000.0;
+		dc->dml2_options.bbox_overrides.urgent_latency_us =
+		dcn3_21_soc.urgent_latency_pixel_data_only_us = dc->bb_overrides.urgent_latency_ns / 1000.0;
+	}
+
+	if ((int)(dcn3_21_soc.dram_clock_change_latency_us * 1000)
+			!= dc->bb_overrides.dram_clock_change_latency_ns
+			&& dc->bb_overrides.dram_clock_change_latency_ns) {
+		dc->dml2_options.bbox_overrides.dram_clock_change_latency_us =
+		dcn3_21_soc.dram_clock_change_latency_us =
+			dc->bb_overrides.dram_clock_change_latency_ns / 1000.0;
+	}
+
+	if ((int)(dcn3_21_soc.fclk_change_latency_us * 1000)
+			!= dc->bb_overrides.fclk_clock_change_latency_ns
+			&& dc->bb_overrides.fclk_clock_change_latency_ns) {
+		dc->dml2_options.bbox_overrides.fclk_change_latency_us =
+		dcn3_21_soc.fclk_change_latency_us =
+			dc->bb_overrides.fclk_clock_change_latency_ns / 1000;
+	}
+
+	if ((int)(dcn3_21_soc.dummy_pstate_latency_us * 1000)
+			!= dc->bb_overrides.dummy_clock_change_latency_ns
+			&& dc->bb_overrides.dummy_clock_change_latency_ns) {
+		dcn3_21_soc.dummy_pstate_latency_us =
+			dc->bb_overrides.dummy_clock_change_latency_ns / 1000.0;
+	}
+
+	/* Override from VBIOS if VBIOS bb_info available */
+	if (dc->ctx->dc_bios->funcs->get_soc_bb_info) {
+		struct bp_soc_bb_info bb_info = {0};
+
+		if (dc->ctx->dc_bios->funcs->get_soc_bb_info(dc->ctx->dc_bios, &bb_info) == BP_RESULT_OK) {
+			if (bb_info.dram_clock_change_latency_100ns > 0)
+				dc->dml2_options.bbox_overrides.dram_clock_change_latency_us =
+				dcn3_21_soc.dram_clock_change_latency_us =
+					bb_info.dram_clock_change_latency_100ns * 10;
+
+			if (bb_info.dram_sr_enter_exit_latency_100ns > 0)
+				dc->dml2_options.bbox_overrides.sr_enter_plus_exit_latency_us =
+				dcn3_21_soc.sr_enter_plus_exit_time_us =
+					bb_info.dram_sr_enter_exit_latency_100ns * 10;
+
+			if (bb_info.dram_sr_exit_latency_100ns > 0)
+				dc->dml2_options.bbox_overrides.sr_exit_latency_us =
+				dcn3_21_soc.sr_exit_time_us =
+					bb_info.dram_sr_exit_latency_100ns * 10;
+		}
+	}
+
+	/* Override from VBIOS for num_chan */
+	if (dc->ctx->dc_bios->vram_info.num_chans) {
+		dc->dml2_options.bbox_overrides.dram_num_chan =
+		dcn3_21_soc.num_chans = dc->ctx->dc_bios->vram_info.num_chans;
+		dcn3_21_soc.mall_allocated_for_dcn_mbytes = (double)(dcn32_calc_num_avail_chans_for_mall(dc,
+			dc->ctx->dc_bios->vram_info.num_chans) * dc->caps.mall_size_per_mem_channel);
+	}
+
+	if (dc->ctx->dc_bios->vram_info.dram_channel_width_bytes) {
+		unsigned int dram_channel_width_bytes = (unsigned int)dc->ctx->dc_bios->vram_info.dram_channel_width_bytes;
+
+		dc->dml2_options.bbox_overrides.dram_chanel_width_bytes = dram_channel_width_bytes;
+		dcn3_21_soc.dram_channel_width_bytes = dram_channel_width_bytes;
+	}
+
+	/* DML DSC delay factor workaround */
+	dcn3_21_ip.dsc_delay_factor_wa = dc->debug.dsc_delay_factor_wa_x1000 / 1000.0;
+
+	dcn3_21_ip.min_prefetch_in_strobe_us = dc->debug.min_prefetch_in_strobe_ns / 1000.0;
+
+	/* Override dispclk_dppclk_vco_speed_mhz from Clk Mgr */
+	dcn3_21_soc.dispclk_dppclk_vco_speed_mhz = dc->clk_mgr->dentist_vco_freq_khz / 1000.0;
+	dc->dml.soc.dispclk_dppclk_vco_speed_mhz = dc->clk_mgr->dentist_vco_freq_khz / 1000.0;
+	dc->dml2_options.bbox_overrides.disp_pll_vco_speed_mhz = dc->clk_mgr->dentist_vco_freq_khz / 1000.0;
+	dc->dml2_options.bbox_overrides.xtalclk_mhz = dc->ctx->dc_bios->fw_info.pll_info.crystal_frequency / 1000.0;
+	dc->dml2_options.bbox_overrides.dchub_refclk_mhz = dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000.0;
+	dc->dml2_options.bbox_overrides.dprefclk_mhz = dc->clk_mgr->dprefclk_khz / 1000.0;
+
+	/* Overrides Clock levelsfrom CLK Mgr table entries as reported by PM FW */
+	if (dc->debug.use_legacy_soc_bb_mechanism) {
+		unsigned int i = 0, j = 0, num_states = 0;
+
+		unsigned int dcfclk_mhz[DC__VOLTAGE_STATES] = {0};
+		unsigned int dram_speed_mts[DC__VOLTAGE_STATES] = {0};
+		unsigned int optimal_uclk_for_dcfclk_sta_targets[DC__VOLTAGE_STATES] = {0};
+		unsigned int optimal_dcfclk_for_uclk[DC__VOLTAGE_STATES] = {0};
+
+		unsigned int dcfclk_sta_targets[DC__VOLTAGE_STATES] = {615, 906, 1324, 1564};
+		unsigned int num_dcfclk_sta_targets = 4, num_uclk_states = 0;
+		unsigned 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 (!max_dcfclk_mhz)
+			max_dcfclk_mhz = (unsigned int)dcn3_21_soc.clock_limits[0].dcfclk_mhz;
+		if (!max_dispclk_mhz)
+			max_dispclk_mhz = (unsigned int)dcn3_21_soc.clock_limits[0].dispclk_mhz;
+		if (!max_dppclk_mhz)
+			max_dppclk_mhz = (unsigned int)dcn3_21_soc.clock_limits[0].dppclk_mhz;
+		if (!max_phyclk_mhz)
+			max_phyclk_mhz = (unsigned int)dcn3_21_soc.clock_limits[0].phyclk_mhz;
+
+		if (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;
+			num_dcfclk_sta_targets++;
+		} else if (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;
+					break;
+				}
+			}
+			// Update size of array since we "removed" duplicates
+			num_dcfclk_sta_targets = i + 1;
+		}
+
+		num_uclk_states = bw_params->clk_table.num_entries;
+
+		// Calculate optimal dcfclk for each uclk
+		for (i = 0; i < num_uclk_states; i++) {
+			dcn321_get_optimal_dcfclk_fclk_for_uclk(bw_params->clk_table.entries[i].memclk_mhz * 16,
+					&optimal_dcfclk_for_uclk[i], NULL);
+			if (optimal_dcfclk_for_uclk[i] < bw_params->clk_table.entries[0].dcfclk_mhz) {
+				optimal_dcfclk_for_uclk[i] = bw_params->clk_table.entries[0].dcfclk_mhz;
+			}
+		}
+
+		// Calculate optimal uclk for each dcfclk sta target
+		for (i = 0; i < num_dcfclk_sta_targets; i++) {
+			for (j = 0; j < num_uclk_states; j++) {
+				if (dcfclk_sta_targets[i] < optimal_dcfclk_for_uclk[j]) {
+					optimal_uclk_for_dcfclk_sta_targets[i] =
+							bw_params->clk_table.entries[j].memclk_mhz * 16;
+					break;
+				}
+			}
+		}
+
+		i = 0;
+		j = 0;
+		// create the final dcfclk and uclk table
+		while (i < num_dcfclk_sta_targets && j < num_uclk_states && num_states < DC__VOLTAGE_STATES) {
+			if (dcfclk_sta_targets[i] < optimal_dcfclk_for_uclk[j]) {
+				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) {
+					dcfclk_mhz[num_states] = optimal_dcfclk_for_uclk[j];
+					dram_speed_mts[num_states++] = bw_params->clk_table.entries[j++].memclk_mhz * 16;
+				} else {
+					j = num_uclk_states;
+				}
+			}
+		}
+
+		while (i < num_dcfclk_sta_targets && num_states < DC__VOLTAGE_STATES) {
+			dcfclk_mhz[num_states] = dcfclk_sta_targets[i];
+			dram_speed_mts[num_states++] = optimal_uclk_for_dcfclk_sta_targets[i++];
+		}
+
+		while (j < num_uclk_states && num_states < DC__VOLTAGE_STATES &&
+				optimal_dcfclk_for_uclk[j] <= 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;
+		}
+
+		/* bw_params->clk_table.entries[MAX_NUM_DPM_LVL].
+		 * MAX_NUM_DPM_LVL is 8.
+		 * dcn3_02_soc.clock_limits[DC__VOLTAGE_STATES].
+		 * DC__VOLTAGE_STATES is 40.
+		 */
+		if (num_states > MAX_NUM_DPM_LVL) {
+			ASSERT(0);
+			return;
+		}
+
+		dcn3_21_soc.num_states = num_states;
+		for (i = 0; i < dcn3_21_soc.num_states; i++) {
+			dcn3_21_soc.clock_limits[i].state = i;
+			dcn3_21_soc.clock_limits[i].dcfclk_mhz = dcfclk_mhz[i];
+			dcn3_21_soc.clock_limits[i].fabricclk_mhz = dcfclk_mhz[i];
+
+			/* Fill all states with max values of all these clocks */
+			dcn3_21_soc.clock_limits[i].dispclk_mhz = max_dispclk_mhz;
+			dcn3_21_soc.clock_limits[i].dppclk_mhz  = max_dppclk_mhz;
+			dcn3_21_soc.clock_limits[i].phyclk_mhz  = max_phyclk_mhz;
+			dcn3_21_soc.clock_limits[i].dscclk_mhz  = max_dispclk_mhz / 3;
+
+			/* Populate from bw_params for DTBCLK, SOCCLK */
+			if (i > 0) {
+				if (!bw_params->clk_table.entries[i].dtbclk_mhz) {
+					dcn3_21_soc.clock_limits[i].dtbclk_mhz  = dcn3_21_soc.clock_limits[i-1].dtbclk_mhz;
+				} else {
+					dcn3_21_soc.clock_limits[i].dtbclk_mhz  = bw_params->clk_table.entries[i].dtbclk_mhz;
+				}
+			} else if (bw_params->clk_table.entries[i].dtbclk_mhz) {
+				dcn3_21_soc.clock_limits[i].dtbclk_mhz  = bw_params->clk_table.entries[i].dtbclk_mhz;
+			}
+
+			if (!bw_params->clk_table.entries[i].socclk_mhz && i > 0)
+				dcn3_21_soc.clock_limits[i].socclk_mhz = dcn3_21_soc.clock_limits[i-1].socclk_mhz;
+			else
+				dcn3_21_soc.clock_limits[i].socclk_mhz = bw_params->clk_table.entries[i].socclk_mhz;
+
+			if (!dram_speed_mts[i] && i > 0)
+				dcn3_21_soc.clock_limits[i].dram_speed_mts = dcn3_21_soc.clock_limits[i-1].dram_speed_mts;
+			else
+				dcn3_21_soc.clock_limits[i].dram_speed_mts = dram_speed_mts[i];
+
+			/* These clocks cannot come from bw_params, always fill from dcn3_21_soc[0] */
+			/* PHYCLK_D18, PHYCLK_D32 */
+			dcn3_21_soc.clock_limits[i].phyclk_d18_mhz = dcn3_21_soc.clock_limits[0].phyclk_d18_mhz;
+			dcn3_21_soc.clock_limits[i].phyclk_d32_mhz = dcn3_21_soc.clock_limits[0].phyclk_d32_mhz;
+		}
+	} else {
+		build_synthetic_soc_states(dc->debug.disable_dc_mode_overwrite, bw_params,
+			dcn3_21_soc.clock_limits, &dcn3_21_soc.num_states);
+	}
+
+	/* Re-init DML with updated bb */
+	dml_init_instance(&dc->dml, &dcn3_21_soc, &dcn3_21_ip, DML_PROJECT_DCN32);
+	if (dc->current_state)
+		dml_init_instance(&dc->current_state->bw_ctx.dml, &dcn3_21_soc, &dcn3_21_ip, DML_PROJECT_DCN32);
+
+	if (dc->clk_mgr->bw_params->clk_table.num_entries > 1) {
+		unsigned int i = 0;
+
+		dc->dml2_options.bbox_overrides.clks_table.num_states = dc->clk_mgr->bw_params->clk_table.num_entries;
+
+		dc->dml2_options.bbox_overrides.clks_table.num_entries_per_clk.num_dcfclk_levels =
+			dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_dcfclk_levels;
+
+		dc->dml2_options.bbox_overrides.clks_table.num_entries_per_clk.num_fclk_levels =
+			dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_fclk_levels;
+
+		dc->dml2_options.bbox_overrides.clks_table.num_entries_per_clk.num_memclk_levels =
+			dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_memclk_levels;
+
+		dc->dml2_options.bbox_overrides.clks_table.num_entries_per_clk.num_socclk_levels =
+			dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_socclk_levels;
+
+		dc->dml2_options.bbox_overrides.clks_table.num_entries_per_clk.num_dtbclk_levels =
+			dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_dtbclk_levels;
+
+		dc->dml2_options.bbox_overrides.clks_table.num_entries_per_clk.num_dispclk_levels =
+			dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_dispclk_levels;
+
+		dc->dml2_options.bbox_overrides.clks_table.num_entries_per_clk.num_dppclk_levels =
+			dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_dppclk_levels;
+
+
+		for (i = 0; i < dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_dcfclk_levels; i++) {
+			if (dc->clk_mgr->bw_params->clk_table.entries[i].dcfclk_mhz)
+				dc->dml2_options.bbox_overrides.clks_table.clk_entries[i].dcfclk_mhz =
+					dc->clk_mgr->bw_params->clk_table.entries[i].dcfclk_mhz;
+		}
+
+		for (i = 0; i < dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_fclk_levels; i++) {
+			if (dc->clk_mgr->bw_params->clk_table.entries[i].fclk_mhz)
+				dc->dml2_options.bbox_overrides.clks_table.clk_entries[i].fclk_mhz =
+					dc->clk_mgr->bw_params->clk_table.entries[i].fclk_mhz;
+		}
+
+		for (i = 0; i < dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_memclk_levels; i++) {
+			if (dc->clk_mgr->bw_params->clk_table.entries[i].memclk_mhz)
+				dc->dml2_options.bbox_overrides.clks_table.clk_entries[i].memclk_mhz =
+					dc->clk_mgr->bw_params->clk_table.entries[i].memclk_mhz;
+		}
+
+		for (i = 0; i < dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_socclk_levels; i++) {
+			if (dc->clk_mgr->bw_params->clk_table.entries[i].socclk_mhz)
+				dc->dml2_options.bbox_overrides.clks_table.clk_entries[i].socclk_mhz =
+					dc->clk_mgr->bw_params->clk_table.entries[i].socclk_mhz;
+		}
+
+		for (i = 0; i < dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_dtbclk_levels; i++) {
+			if (dc->clk_mgr->bw_params->clk_table.entries[i].dtbclk_mhz)
+				dc->dml2_options.bbox_overrides.clks_table.clk_entries[i].dtbclk_mhz =
+					dc->clk_mgr->bw_params->clk_table.entries[i].dtbclk_mhz;
+		}
+
+		for (i = 0; i < dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_dispclk_levels; i++) {
+			if (dc->clk_mgr->bw_params->clk_table.entries[i].dispclk_mhz) {
+				dc->dml2_options.bbox_overrides.clks_table.clk_entries[i].dispclk_mhz =
+					dc->clk_mgr->bw_params->clk_table.entries[i].dispclk_mhz;
+				dc->dml2_options.bbox_overrides.clks_table.clk_entries[i].dppclk_mhz =
+					dc->clk_mgr->bw_params->clk_table.entries[i].dispclk_mhz;
+			}
+		}
+	}
+}
+

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.h

@@ -0,0 +1,34 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright 2022 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 __DCN32_FPU_H__
+#define __DCN32_FPU_H__
+
+#include "dml/display_mode_vba.h"
+
+void dcn321_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_params);
+
+#endif

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.c

@@ -0,0 +1,617 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright 2023 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 "dcn35_fpu.h"
+#include "dcn31/dcn31_resource.h"
+#include "dcn32/dcn32_resource.h"
+#include "dcn35/dcn35_resource.h"
+#include "dml/dcn31/dcn31_fpu.h"
+#include "dml/dml_inline_defs.h"
+
+#include "link_service.h"
+
+#define DC_LOGGER_INIT(logger)
+
+struct _vcs_dpi_ip_params_st dcn3_5_ip = {
+	.VBlankNomDefaultUS = 668,
+	.gpuvm_enable = 1,
+	.gpuvm_max_page_table_levels = 1,
+	.hostvm_enable = 1,
+	.hostvm_max_page_table_levels = 2,
+	.rob_buffer_size_kbytes = 64,
+	.det_buffer_size_kbytes = 1536,
+	.config_return_buffer_size_in_kbytes = 1792,
+	.compressed_buffer_segment_size_in_kbytes = 64,
+	.meta_fifo_size_in_kentries = 32,
+	.zero_size_buffer_entries = 512,
+	.compbuf_reserved_space_64b = 256,
+	.compbuf_reserved_space_zs = 64,
+	.dpp_output_buffer_pixels = 2560,/*not used*/
+	.opp_output_buffer_lines = 1,/*not used*/
+	.pixel_chunk_size_kbytes = 8,
+	//.alpha_pixel_chunk_size_kbytes = 4;/*new*/
+	//.min_pixel_chunk_size_bytes = 1024;/*new*/
+	.meta_chunk_size_kbytes = 2,
+	.min_meta_chunk_size_bytes = 256,
+	.writeback_chunk_size_kbytes = 8,
+	.ptoi_supported = false,
+	.num_dsc = 4,
+	.maximum_dsc_bits_per_component = 12,/*delta from 10*/
+	.dsc422_native_support = true,/*delta from false*/
+	.is_line_buffer_bpp_fixed = true,/*new*/
+	.line_buffer_fixed_bpp = 32,/*delta from 48*/
+	.line_buffer_size_bits = 986880,/*delta from 789504*/
+	.max_line_buffer_lines = 32,/*delta from 12*/
+	.writeback_interface_buffer_size_kbytes = 90,
+	.max_num_dpp = 4,
+	.max_num_otg = 4,
+	.max_num_hdmi_frl_outputs = 1,
+	.max_num_wb = 1,
+	/*.max_num_hdmi_frl_outputs = 1; new in dml2*/
+	/*.max_num_dp2p0_outputs = 2; new in dml2*/
+	/*.max_num_dp2p0_streams = 4; new in dml2*/
+	.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,
+	.max_hscl_taps = 8,
+	.max_vscl_taps = 8,
+	.dpte_buffer_size_in_pte_reqs_luma = 68,/*changed from 64,*/
+	.dpte_buffer_size_in_pte_reqs_chroma = 36,/*changed from 34*/
+	/*.dcc_meta_buffer_size_bytes = 6272; new to dml2*/
+	.dispclk_ramp_margin_percent = 1.11,/*delta from 1*/
+	/*.dppclk_delay_subtotal = 47;
+	.dppclk_delay_scl = 50;
+	.dppclk_delay_scl_lb_only = 16;
+	.dppclk_delay_cnvc_formatter = 28;
+	.dppclk_delay_cnvc_cursor = 6;
+	.dispclk_delay_subtotal = 125;*/ /*new to dml2*/
+	.max_inter_dcn_tile_repeaters = 8,
+	.cursor_buffer_size = 16,
+	.cursor_chunk_size = 2,
+	.writeback_line_buffer_buffer_size = 0,
+	.writeback_min_hscl_ratio = 1,
+	.writeback_min_vscl_ratio = 1,
+	.writeback_max_hscl_ratio = 1,
+	.writeback_max_vscl_ratio = 1,
+	.writeback_max_hscl_taps = 1,
+	.writeback_max_vscl_taps = 1,
+	.dppclk_delay_subtotal = 47, /* changed from 46,*/
+	.dppclk_delay_scl = 50,
+	.dppclk_delay_scl_lb_only = 16,
+	.dppclk_delay_cnvc_formatter = 28,/*changed from 27,*/
+	.dppclk_delay_cnvc_cursor = 6,
+	.dispclk_delay_subtotal = 125, /*changed from 119,*/
+	.dynamic_metadata_vm_enabled = false,
+	.odm_combine_4to1_supported = false,
+	.dcc_supported = true,
+//	.config_return_buffer_segment_size_in_kbytes = 64;/*required, hard coded in dml2_translate_ip_params*/
+
+};
+
+struct _vcs_dpi_soc_bounding_box_st dcn3_5_soc = {
+		/*TODO: correct dispclk/dppclk voltage level determination*/
+	.clock_limits = {
+		{
+			.state = 0,
+			.dispclk_mhz = 1200.0,
+			.dppclk_mhz = 1200.0,
+			.phyclk_mhz = 600.0,
+			.phyclk_d18_mhz = 667.0,
+			.dscclk_mhz = 186.0,
+			.dtbclk_mhz = 600.0,
+		},
+		{
+			.state = 1,
+			.dispclk_mhz = 1200.0,
+			.dppclk_mhz = 1200.0,
+			.phyclk_mhz = 810.0,
+			.phyclk_d18_mhz = 667.0,
+			.dscclk_mhz = 209.0,
+			.dtbclk_mhz = 600.0,
+		},
+		{
+			.state = 2,
+			.dispclk_mhz = 1200.0,
+			.dppclk_mhz = 1200.0,
+			.phyclk_mhz = 810.0,
+			.phyclk_d18_mhz = 667.0,
+			.dscclk_mhz = 209.0,
+			.dtbclk_mhz = 600.0,
+		},
+		{
+			.state = 3,
+			.dispclk_mhz = 1200.0,
+			.dppclk_mhz = 1200.0,
+			.phyclk_mhz = 810.0,
+			.phyclk_d18_mhz = 667.0,
+			.dscclk_mhz = 371.0,
+			.dtbclk_mhz = 600.0,
+		},
+		{
+			.state = 4,
+			.dispclk_mhz = 1200.0,
+			.dppclk_mhz = 1200.0,
+			.phyclk_mhz = 810.0,
+			.phyclk_d18_mhz = 667.0,
+			.dscclk_mhz = 400.0,
+			.dtbclk_mhz = 600.0,
+		},
+	},
+	.num_states = 5,
+	.sr_exit_time_us = 31.0,
+	.sr_enter_plus_exit_time_us = 33.0,
+	.sr_exit_z8_time_us = 250.0,
+	.sr_enter_plus_exit_z8_time_us = 350.0,
+	.fclk_change_latency_us = 24.0,
+	.usr_retraining_latency_us = 2,
+	.writeback_latency_us = 12.0,
+
+	.dram_channel_width_bytes = 4,/*not exist in dml2*/
+	.round_trip_ping_latency_dcfclk_cycles = 106,/*not exist in dml2*/
+	.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,
+	.dram_clock_change_latency_us = 34.0,
+	.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_sdp_bw_after_urgent = 80.0,
+	.pct_ideal_fabric_bw_after_urgent = 80.0, /*new to dml2*/
+	.pct_ideal_dram_sdp_bw_after_urgent_pixel_only = 65.0,
+	.pct_ideal_dram_sdp_bw_after_urgent_pixel_and_vm = 60.0,
+	.pct_ideal_dram_sdp_bw_after_urgent_vm_only = 30.0,
+	.max_avg_sdp_bw_use_normal_percent = 60.0,
+	.max_avg_dram_bw_use_normal_percent = 60.0,
+	.fabric_datapath_to_dcn_data_return_bytes = 32,
+	.return_bus_width_bytes = 64,
+	.downspread_percent = 0.38,
+	.dcn_downspread_percent = 0.5,
+	.gpuvm_min_page_size_bytes = 4096,
+	.hostvm_min_page_size_bytes = 4096,
+	.do_urgent_latency_adjustment = 0,
+	.urgent_latency_adjustment_fabric_clock_component_us = 0,
+	.urgent_latency_adjustment_fabric_clock_reference_mhz = 0,
+};
+
+void dcn35_build_wm_range_table_fpu(struct clk_mgr *clk_mgr)
+{
+	(void)clk_mgr;
+	//TODO
+}
+
+
+/*
+ * dcn35_update_bw_bounding_box
+ *
+ * This would override some dcn3_5 ip_or_soc initial parameters hardcoded from
+ * spreadsheet with actual values as per dGPU SKU:
+ * - with passed few options from dc->config
+ * - with dentist_vco_frequency from Clk Mgr (currently hardcoded, but might
+ *   need to get it from PM FW)
+ * - with passed latency values (passed in ns units) in dc-> bb override for
+ *   debugging purposes
+ * - with passed latencies from VBIOS (in 100_ns units) if available for
+ *   certain dGPU SKU
+ * - with number of DRAM channels from VBIOS (which differ for certain dGPU SKU
+ *   of the same ASIC)
+ * - clocks levels with passed clk_table entries from Clk Mgr as reported by PM
+ *   FW for different clocks (which might differ for certain dGPU SKU of the
+ *   same ASIC)
+ */
+void dcn35_update_bw_bounding_box_fpu(struct dc *dc,
+				      struct clk_bw_params *bw_params)
+{
+	unsigned int i, closest_clk_lvl;
+	int j;
+	struct clk_limit_table *clk_table = &bw_params->clk_table;
+	struct _vcs_dpi_voltage_scaling_st *clock_limits =
+		dc->scratch.update_bw_bounding_box.clock_limits;
+	unsigned int max_dispclk_mhz = 0, max_dppclk_mhz = 0;
+
+	dc_assert_fp_enabled();
+
+	dcn3_5_ip.max_num_otg =
+		dc->res_pool->res_cap->num_timing_generator;
+	dcn3_5_ip.max_num_dpp = dc->res_pool->pipe_count;
+	dcn3_5_soc.num_chans = bw_params->num_channels;
+
+	ASSERT(clk_table->num_entries);
+
+	/* Prepass to find max clocks independent of voltage level. */
+	for (i = 0; i < clk_table->num_entries; ++i) {
+		if (clk_table->entries[i].dispclk_mhz > max_dispclk_mhz)
+			max_dispclk_mhz = clk_table->entries[i].dispclk_mhz;
+		if (clk_table->entries[i].dppclk_mhz > max_dppclk_mhz)
+			max_dppclk_mhz = clk_table->entries[i].dppclk_mhz;
+	}
+
+	for (i = 0; i < clk_table->num_entries; i++) {
+		/* loop backwards*/
+		for (closest_clk_lvl = 0, j = dcn3_5_soc.num_states - 1;
+			j >= 0; j--) {
+			if (dcn3_5_soc.clock_limits[j].dcfclk_mhz <=
+				clk_table->entries[i].dcfclk_mhz) {
+				closest_clk_lvl = j;
+				break;
+			}
+		}
+		if (clk_table->num_entries == 1) {
+			/*smu gives one DPM level, let's take the highest one*/
+			closest_clk_lvl = dcn3_5_soc.num_states - 1;
+		}
+
+		clock_limits[i].state = i;
+
+		/* Clocks dependent on voltage level. */
+		clock_limits[i].dcfclk_mhz = clk_table->entries[i].dcfclk_mhz;
+		if (clk_table->num_entries == 1 &&
+			clock_limits[i].dcfclk_mhz <
+			dcn3_5_soc.clock_limits[closest_clk_lvl].dcfclk_mhz) {
+			/*SMU fix not released yet*/
+			clock_limits[i].dcfclk_mhz =
+				dcn3_5_soc.clock_limits[closest_clk_lvl].dcfclk_mhz;
+		}
+
+		clock_limits[i].fabricclk_mhz =
+			clk_table->entries[i].fclk_mhz;
+		clock_limits[i].socclk_mhz =
+			clk_table->entries[i].socclk_mhz;
+
+		if (clk_table->entries[i].memclk_mhz &&
+			clk_table->entries[i].wck_ratio)
+			clock_limits[i].dram_speed_mts =
+				clk_table->entries[i].memclk_mhz * 2 *
+				clk_table->entries[i].wck_ratio;
+
+		/* Clocks independent of voltage level. */
+		clock_limits[i].dispclk_mhz = max_dispclk_mhz ?
+			max_dispclk_mhz :
+			dcn3_5_soc.clock_limits[closest_clk_lvl].dispclk_mhz;
+
+		clock_limits[i].dppclk_mhz = max_dppclk_mhz ?
+			max_dppclk_mhz :
+			dcn3_5_soc.clock_limits[closest_clk_lvl].dppclk_mhz;
+
+		clock_limits[i].dram_bw_per_chan_gbps =
+			dcn3_5_soc.clock_limits[closest_clk_lvl].dram_bw_per_chan_gbps;
+		clock_limits[i].dscclk_mhz =
+			dcn3_5_soc.clock_limits[closest_clk_lvl].dscclk_mhz;
+		clock_limits[i].dtbclk_mhz =
+			dcn3_5_soc.clock_limits[closest_clk_lvl].dtbclk_mhz;
+		clock_limits[i].phyclk_d18_mhz =
+			dcn3_5_soc.clock_limits[closest_clk_lvl].phyclk_d18_mhz;
+		clock_limits[i].phyclk_mhz =
+			dcn3_5_soc.clock_limits[closest_clk_lvl].phyclk_mhz;
+	}
+
+	memcpy(dcn3_5_soc.clock_limits, clock_limits,
+		sizeof(dcn3_5_soc.clock_limits));
+
+	if (clk_table->num_entries)
+		dcn3_5_soc.num_states = clk_table->num_entries;
+
+	if (max_dispclk_mhz) {
+		dcn3_5_soc.dispclk_dppclk_vco_speed_mhz = max_dispclk_mhz * 2;
+		dc->dml.soc.dispclk_dppclk_vco_speed_mhz = max_dispclk_mhz * 2;
+	}
+	if ((int)(dcn3_5_soc.dram_clock_change_latency_us * 1000)
+				!= dc->debug.dram_clock_change_latency_ns
+			&& dc->debug.dram_clock_change_latency_ns) {
+		dcn3_5_soc.dram_clock_change_latency_us =
+			dc->debug.dram_clock_change_latency_ns / 1000.0;
+	}
+
+	if (dc->bb_overrides.dram_clock_change_latency_ns > 0)
+		dcn3_5_soc.dram_clock_change_latency_us =
+			dc->bb_overrides.dram_clock_change_latency_ns / 1000.0;
+
+	if (dc->bb_overrides.sr_exit_time_ns > 0)
+		dcn3_5_soc.sr_exit_time_us = dc->bb_overrides.sr_exit_time_ns / 1000.0;
+
+	if (dc->bb_overrides.sr_enter_plus_exit_time_ns > 0)
+		dcn3_5_soc.sr_enter_plus_exit_time_us =
+			dc->bb_overrides.sr_enter_plus_exit_time_ns / 1000.0;
+
+	if (dc->bb_overrides.sr_exit_z8_time_ns > 0)
+		dcn3_5_soc.sr_exit_z8_time_us = dc->bb_overrides.sr_exit_z8_time_ns / 1000.0;
+
+	if (dc->bb_overrides.sr_enter_plus_exit_z8_time_ns > 0)
+		dcn3_5_soc.sr_enter_plus_exit_z8_time_us =
+			dc->bb_overrides.sr_enter_plus_exit_z8_time_ns / 1000.0;
+
+	/*temp till dml2 fully work without dml1*/
+	dml_init_instance(&dc->dml, &dcn3_5_soc, &dcn3_5_ip,
+				DML_PROJECT_DCN31);
+
+	/*copy to dml2, before dml2_create*/
+	if (clk_table->num_entries > 2) {
+
+		for (i = 0; i < clk_table->num_entries; i++) {
+			dc->dml2_options.bbox_overrides.clks_table.num_states =
+				clk_table->num_entries;
+			dc->dml2_options.bbox_overrides.clks_table.clk_entries[i].dcfclk_mhz =
+				(unsigned int)clock_limits[i].dcfclk_mhz;
+			dc->dml2_options.bbox_overrides.clks_table.clk_entries[i].fclk_mhz =
+				(unsigned int)clock_limits[i].fabricclk_mhz;
+			dc->dml2_options.bbox_overrides.clks_table.clk_entries[i].dispclk_mhz =
+				(unsigned int)clock_limits[i].dispclk_mhz;
+			dc->dml2_options.bbox_overrides.clks_table.clk_entries[i].dppclk_mhz =
+				(unsigned int)clock_limits[i].dppclk_mhz;
+			dc->dml2_options.bbox_overrides.clks_table.clk_entries[i].socclk_mhz =
+				(unsigned int)clock_limits[i].socclk_mhz;
+			dc->dml2_options.bbox_overrides.clks_table.clk_entries[i].memclk_mhz =
+				clk_table->entries[i].memclk_mhz * clk_table->entries[i].wck_ratio;
+
+			dc->dml2_options.bbox_overrides.clks_table.clk_entries[i].dram_speed_mts = (unsigned int)clock_limits[i].dram_speed_mts;
+			dc->dml2_options.bbox_overrides.clks_table.clk_entries[i].dtbclk_mhz =
+				(unsigned int)clock_limits[i].dtbclk_mhz;
+			dc->dml2_options.bbox_overrides.clks_table.num_entries_per_clk.num_dcfclk_levels =
+				clk_table->num_entries;
+			dc->dml2_options.bbox_overrides.clks_table.num_entries_per_clk.num_fclk_levels =
+				clk_table->num_entries;
+			dc->dml2_options.bbox_overrides.clks_table.num_entries_per_clk.num_dispclk_levels =
+				clk_table->num_entries;
+			dc->dml2_options.bbox_overrides.clks_table.num_entries_per_clk.num_dppclk_levels =
+				clk_table->num_entries;
+			dc->dml2_options.bbox_overrides.clks_table.num_entries_per_clk.num_socclk_levels =
+				clk_table->num_entries;
+			dc->dml2_options.bbox_overrides.clks_table.num_entries_per_clk.num_memclk_levels =
+				clk_table->num_entries;
+			dc->dml2_options.bbox_overrides.clks_table.num_entries_per_clk.num_dtbclk_levels =
+				clk_table->num_entries;
+		}
+	}
+
+	/* Update latency values */
+	dc->dml2_options.bbox_overrides.dram_clock_change_latency_us = dcn3_5_soc.dram_clock_change_latency_us;
+
+	dc->dml2_options.bbox_overrides.sr_exit_latency_us = dcn3_5_soc.sr_exit_time_us;
+	dc->dml2_options.bbox_overrides.sr_enter_plus_exit_latency_us = dcn3_5_soc.sr_enter_plus_exit_time_us;
+
+	dc->dml2_options.bbox_overrides.sr_exit_z8_time_us = dcn3_5_soc.sr_exit_z8_time_us;
+	dc->dml2_options.bbox_overrides.sr_enter_plus_exit_z8_time_us = dcn3_5_soc.sr_enter_plus_exit_z8_time_us;
+}
+
+static bool is_dual_plane(enum surface_pixel_format format)
+{
+	return format >= SURFACE_PIXEL_FORMAT_VIDEO_BEGIN ||
+		format == SURFACE_PIXEL_FORMAT_GRPH_RGBE_ALPHA;
+}
+
+/*
+ * micro_sec_to_vert_lines () - converts time to number of vertical lines for a given timing
+ *
+ * @param: num_us: number of microseconds
+ * @return: number of vertical lines. If exact number of vertical lines is not found then
+ *          it will round up to next number of lines to guarantee num_us
+ */
+static unsigned int micro_sec_to_vert_lines(unsigned int num_us, struct dc_crtc_timing *timing)
+{
+	unsigned int num_lines = 0;
+	double lines_time_in_ns = 1000.0 *
+			(((double)timing->h_total * 1000.0) /
+			 ((double)timing->pix_clk_100hz / 10.0));
+
+	num_lines = (unsigned int)dml_ceil(1000.0 * num_us / lines_time_in_ns, 1.0);
+
+	return num_lines;
+}
+
+static unsigned int get_vertical_back_porch(struct dc_crtc_timing *timing)
+{
+	unsigned int v_active = 0, v_blank = 0, v_back_porch = 0;
+
+	v_active = timing->v_border_top + timing->v_addressable + timing->v_border_bottom;
+	v_blank = timing->v_total - v_active;
+	v_back_porch = v_blank - timing->v_front_porch - timing->v_sync_width;
+
+	return v_back_porch;
+}
+
+int dcn35_populate_dml_pipes_from_context_fpu(struct dc *dc,
+					      struct dc_state *context,
+					      display_e2e_pipe_params_st *pipes,
+					      enum dc_validate_mode validate_mode)
+{
+	unsigned int i;
+	int pipe_cnt;
+	struct resource_context *res_ctx = &context->res_ctx;
+	struct pipe_ctx *pipe = 0;
+	bool upscaled = false;
+	const unsigned int max_allowed_vblank_nom = 1023;
+
+	dc_assert_fp_enabled();
+
+	dcn31_populate_dml_pipes_from_context(dc, context, pipes,
+					      validate_mode);
+
+	for (i = 0, pipe_cnt = 0; i < dc->res_pool->pipe_count; i++) {
+		struct dc_crtc_timing *timing;
+		unsigned int num_lines = 0;
+		unsigned int v_back_porch = 0;
+
+		if (!res_ctx->pipe_ctx[i].stream)
+			continue;
+
+		pipe = &res_ctx->pipe_ctx[i];
+		timing = &pipe->stream->timing;
+
+		num_lines = micro_sec_to_vert_lines(dcn3_5_ip.VBlankNomDefaultUS, timing);
+		v_back_porch  = get_vertical_back_porch(timing);
+
+		if (pipe->stream->adjust.v_total_max ==
+		    pipe->stream->adjust.v_total_min &&
+		    pipe->stream->adjust.v_total_min > timing->v_total) {
+			pipes[pipe_cnt].pipe.dest.vtotal =
+				pipe->stream->adjust.v_total_min;
+			pipes[pipe_cnt].pipe.dest.vblank_nom = timing->v_total -
+				pipes[pipe_cnt].pipe.dest.vactive;
+		}
+
+		pipes[pipe_cnt].pipe.dest.vblank_nom = timing->v_total - pipes[pipe_cnt].pipe.dest.vactive;
+		pipes[pipe_cnt].pipe.dest.vblank_nom = min(pipes[pipe_cnt].pipe.dest.vblank_nom, num_lines);
+		// vblank_nom should not smaller than (VSync (timing->v_sync_width + v_back_porch) + 2)
+		// + 2 is because
+		// 1 -> VStartup_start should be 1 line before VSync
+		// 1 -> always reserve 1 line between start of vblank to vstartup signal
+		pipes[pipe_cnt].pipe.dest.vblank_nom =
+			max(pipes[pipe_cnt].pipe.dest.vblank_nom, timing->v_sync_width + v_back_porch + 2);
+		pipes[pipe_cnt].pipe.dest.vblank_nom = min(pipes[pipe_cnt].pipe.dest.vblank_nom, max_allowed_vblank_nom);
+
+		if (pipe->plane_state &&
+		    (pipe->plane_state->src_rect.height <
+		     pipe->plane_state->dst_rect.height ||
+		     pipe->plane_state->src_rect.width <
+		     pipe->plane_state->dst_rect.width))
+			upscaled = true;
+
+		/*
+		 * Immediate flip can be set dynamically after enabling the
+		 * plane. We need to require support for immediate flip or
+		 * underflow can be intermittently experienced depending on peak
+		 * b/w requirements.
+		 */
+		pipes[pipe_cnt].pipe.src.immediate_flip = true;
+
+		pipes[pipe_cnt].pipe.src.unbounded_req_mode = false;
+
+		dcn31_zero_pipe_dcc_fraction(pipes, pipe_cnt);
+
+		pipes[pipe_cnt].pipe.dest.vfront_porch = timing->v_front_porch;
+		pipes[pipe_cnt].pipe.src.dcc_rate = 3;
+		pipes[pipe_cnt].dout.dsc_input_bpc = 0;
+		pipes[pipe_cnt].pipe.src.gpuvm_min_page_size_kbytes = 256;
+
+		if (pipes[pipe_cnt].dout.dsc_enable) {
+			switch (timing->display_color_depth) {
+			case COLOR_DEPTH_888:
+				pipes[pipe_cnt].dout.dsc_input_bpc = 8;
+				break;
+			case COLOR_DEPTH_101010:
+				pipes[pipe_cnt].dout.dsc_input_bpc = 10;
+				break;
+			case COLOR_DEPTH_121212:
+				pipes[pipe_cnt].dout.dsc_input_bpc = 12;
+				break;
+			default:
+				ASSERT(0);
+				break;
+			}
+		}
+
+		pipe_cnt++;
+	}
+
+	context->bw_ctx.dml.ip.det_buffer_size_kbytes = 384;/*per guide*/
+
+	if (pipe_cnt == 1 && pipe->plane_state && !dc->debug.disable_z9_mpc) {
+		if (!is_dual_plane(pipe->plane_state->format) &&
+			   pipe->plane_state->src_rect.width <= 5120) {
+			/*
+			 * Limit to 5k max to avoid forced pipe split when there
+			 * is not enough detile for swath
+			 */
+			context->bw_ctx.dml.ip.det_buffer_size_kbytes = 192;
+			pipes[0].pipe.src.unbounded_req_mode = true;
+		}
+	} else if (context->stream_count >=
+		   dc->debug.crb_alloc_policy_min_disp_count &&
+		   dc->debug.crb_alloc_policy > DET_SIZE_DEFAULT) {
+		context->bw_ctx.dml.ip.det_buffer_size_kbytes =
+			dc->debug.crb_alloc_policy * 64;
+	} else if (context->stream_count >= 3 && upscaled) {
+		context->bw_ctx.dml.ip.det_buffer_size_kbytes = 192;
+	}
+
+	for (i = 0; i < dc->res_pool->pipe_count; i++) {
+		struct pipe_ctx *cur_pipe = &context->res_ctx.pipe_ctx[i];
+
+		if (!cur_pipe->stream)
+			continue;
+
+		if (cur_pipe->stream->signal == SIGNAL_TYPE_EDP &&
+		    dc->debug.seamless_boot_odm_combine &&
+		    cur_pipe->stream->apply_seamless_boot_optimization) {
+
+			if (cur_pipe->stream->apply_boot_odm_mode ==
+			    dm_odm_combine_policy_2to1) {
+				context->bw_ctx.dml.vba.ODMCombinePolicy =
+					dm_odm_combine_policy_2to1;
+				break;
+			}
+		}
+	}
+
+	return pipe_cnt;
+}
+
+void dcn35_decide_zstate_support(struct dc *dc, struct dc_state *context)
+{
+	enum dcn_zstate_support_state support = DCN_ZSTATE_SUPPORT_DISALLOW;
+	unsigned int i, plane_count = 0;
+	DC_LOGGER_INIT(dc->ctx->logger);
+
+	dc_assert_fp_enabled();
+
+	for (i = 0; i < dc->res_pool->pipe_count; i++) {
+		if (context->res_ctx.pipe_ctx[i].plane_state)
+			plane_count++;
+	}
+
+	if (context->stream_count == 0 || plane_count == 0) {
+		support = DCN_ZSTATE_SUPPORT_ALLOW;
+	} else if (context->stream_count == 1 && context->streams[0]->signal == SIGNAL_TYPE_EDP) {
+		struct dc_link *link = context->streams[0]->sink->link;
+		bool is_pwrseq0 = link && link->link_index == 0;
+		bool is_psr = (link && (link->psr_settings.psr_version == DC_PSR_VERSION_1 ||
+								link->psr_settings.psr_version == DC_PSR_VERSION_SU_1) && !link->panel_config.psr.disable_psr);
+		bool is_replay = link && link->replay_settings.replay_feature_enabled;
+		int minmum_z8_residency =
+			dc->debug.minimum_z8_residency_time > 0 ? dc->debug.minimum_z8_residency_time : 1000;
+		bool allow_z8 = context->bw_ctx.dml.vba.StutterPeriod > (double)minmum_z8_residency;
+		int minmum_z10_residency =
+			dc->debug.minimum_z10_residency_time > 0 ? dc->debug.minimum_z10_residency_time : 5000;
+		bool allow_z10 = context->bw_ctx.dml.vba.StutterPeriod > (double)minmum_z10_residency;
+
+		/*for psr1/psr-su, we allow z8 and z10 based on latency, for replay with IPS enabled, it will enter ips2*/
+		if (is_pwrseq0 && allow_z10)
+			support = DCN_ZSTATE_SUPPORT_ALLOW;
+		else if (is_pwrseq0 && (is_psr || is_replay))
+			support = DCN_ZSTATE_SUPPORT_ALLOW_Z8_Z10_ONLY;
+		else if (allow_z8)
+			support = DCN_ZSTATE_SUPPORT_ALLOW_Z8_ONLY;
+
+	}
+
+	DC_LOG_SMU("zstate_support: %d, StutterPeriod: %d\n", support,
+		   (int)context->bw_ctx.dml.vba.StutterPeriod);
+
+	context->bw_ctx.bw.dcn.clk.zstate_support = support;
+}

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.h

@@ -0,0 +1,44 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright 2023 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 __DCN35_FPU_H__
+#define __DCN35_FPU_H__
+
+#include "clk_mgr.h"
+
+void dcn35_build_wm_range_table_fpu(struct clk_mgr *clk_mgr);
+
+void dcn35_update_bw_bounding_box_fpu(struct dc *dc,
+				      struct clk_bw_params *bw_params);
+
+int dcn35_populate_dml_pipes_from_context_fpu(struct dc *dc,
+					      struct dc_state *context,
+					      display_e2e_pipe_params_st *pipes,
+					      enum dc_validate_mode validate_mode);
+
+void dcn35_decide_zstate_support(struct dc *dc, struct dc_state *context);
+
+#endif

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dcn351/dcn351_fpu.c

@@ -0,0 +1,635 @@
+/* SPDX-License-Identifier: MIT */
+/* Copyright 2024 Advanced Micro Devices, Inc. */
+#include "resource.h"
+#include "dcn351_fpu.h"
+#include "dcn31/dcn31_resource.h"
+#include "dcn32/dcn32_resource.h"
+#include "dcn35/dcn35_resource.h"
+#include "dcn351/dcn351_resource.h"
+#include "dml/dcn31/dcn31_fpu.h"
+#include "dml/dcn35/dcn35_fpu.h"
+#include "dml/dml_inline_defs.h"
+
+#include "link_service.h"
+
+#define DC_LOGGER_INIT(logger)
+
+struct _vcs_dpi_ip_params_st dcn3_51_ip = {
+	.VBlankNomDefaultUS = 668,
+	.gpuvm_enable = 1,
+	.gpuvm_max_page_table_levels = 1,
+	.hostvm_enable = 1,
+	.hostvm_max_page_table_levels = 2,
+	.rob_buffer_size_kbytes = 64,
+	.det_buffer_size_kbytes = 1536,
+	.config_return_buffer_size_in_kbytes = 1792,
+	.compressed_buffer_segment_size_in_kbytes = 64,
+	.meta_fifo_size_in_kentries = 32,
+	.zero_size_buffer_entries = 512,
+	.compbuf_reserved_space_64b = 256,
+	.compbuf_reserved_space_zs = 64,
+	.dpp_output_buffer_pixels = 2560,/*not used*/
+	.opp_output_buffer_lines = 1,/*not used*/
+	.pixel_chunk_size_kbytes = 8,
+	//.alpha_pixel_chunk_size_kbytes = 4;/*new*/
+	//.min_pixel_chunk_size_bytes = 1024;/*new*/
+	.meta_chunk_size_kbytes = 2,
+	.min_meta_chunk_size_bytes = 256,
+	.writeback_chunk_size_kbytes = 8,
+	.ptoi_supported = false,
+	.num_dsc = 4,
+	.maximum_dsc_bits_per_component = 12,/*delta from 10*/
+	.dsc422_native_support = true,/*delta from false*/
+	.is_line_buffer_bpp_fixed = true,/*new*/
+	.line_buffer_fixed_bpp = 32,/*delta from 48*/
+	.line_buffer_size_bits = 986880,/*delta from 789504*/
+	.max_line_buffer_lines = 32,/*delta from 12*/
+	.writeback_interface_buffer_size_kbytes = 90,
+	.max_num_dpp = 4,
+	.max_num_otg = 4,
+	.max_num_hdmi_frl_outputs = 1,
+	.max_num_wb = 1,
+	/*.max_num_hdmi_frl_outputs = 1; new in dml2*/
+	/*.max_num_dp2p0_outputs = 2; new in dml2*/
+	/*.max_num_dp2p0_streams = 4; new in dml2*/
+	.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,
+	.max_hscl_taps = 8,
+	.max_vscl_taps = 8,
+	.dpte_buffer_size_in_pte_reqs_luma = 68,/*changed from 64,*/
+	.dpte_buffer_size_in_pte_reqs_chroma = 36,/*changed from 34*/
+	/*.dcc_meta_buffer_size_bytes = 6272; new to dml2*/
+	.dispclk_ramp_margin_percent = 1.11,/*delta from 1*/
+	/*.dppclk_delay_subtotal = 47;
+	.dppclk_delay_scl = 50;
+	.dppclk_delay_scl_lb_only = 16;
+	.dppclk_delay_cnvc_formatter = 28;
+	.dppclk_delay_cnvc_cursor = 6;
+	.dispclk_delay_subtotal = 125;*/ /*new to dml2*/
+	.max_inter_dcn_tile_repeaters = 8,
+	.cursor_buffer_size = 16,
+	.cursor_chunk_size = 2,
+	.writeback_line_buffer_buffer_size = 0,
+	.writeback_min_hscl_ratio = 1,
+	.writeback_min_vscl_ratio = 1,
+	.writeback_max_hscl_ratio = 1,
+	.writeback_max_vscl_ratio = 1,
+	.writeback_max_hscl_taps = 1,
+	.writeback_max_vscl_taps = 1,
+	.dppclk_delay_subtotal = 47, /* changed from 46,*/
+	.dppclk_delay_scl = 50,
+	.dppclk_delay_scl_lb_only = 16,
+	.dppclk_delay_cnvc_formatter = 28,/*changed from 27,*/
+	.dppclk_delay_cnvc_cursor = 6,
+	.dispclk_delay_subtotal = 125, /*changed from 119,*/
+	.dynamic_metadata_vm_enabled = false,
+	.odm_combine_4to1_supported = false,
+	.dcc_supported = true,
+//	.config_return_buffer_segment_size_in_kbytes = 64;/*required, hard coded in dml2_translate_ip_params*/
+
+};
+
+struct _vcs_dpi_soc_bounding_box_st dcn3_51_soc = {
+		/*TODO: correct dispclk/dppclk voltage level determination*/
+	.clock_limits = {
+		{
+			.state = 0,
+			.dcfclk_mhz = 400.0,
+			.fabricclk_mhz = 400.0,
+			.socclk_mhz = 600.0,
+			.dram_speed_mts = 3200.0,
+			.dispclk_mhz = 600.0,
+			.dppclk_mhz = 600.0,
+			.phyclk_mhz = 600.0,
+			.phyclk_d18_mhz = 667.0,
+			.dscclk_mhz = 200.0,
+			.dtbclk_mhz = 600.0,
+		},
+		{
+			.state = 1,
+			.dcfclk_mhz = 600.0,
+			.fabricclk_mhz = 1000.0,
+			.socclk_mhz = 733.0,
+			.dram_speed_mts = 6400.0,
+			.dispclk_mhz = 800.0,
+			.dppclk_mhz = 800.0,
+			.phyclk_mhz = 810.0,
+			.phyclk_d18_mhz = 667.0,
+			.dscclk_mhz = 266.7,
+			.dtbclk_mhz = 600.0,
+		},
+		{
+			.state = 2,
+			.dcfclk_mhz = 738.0,
+			.fabricclk_mhz = 1200.0,
+			.socclk_mhz = 880.0,
+			.dram_speed_mts = 7500.0,
+			.dispclk_mhz = 800.0,
+			.dppclk_mhz = 800.0,
+			.phyclk_mhz = 810.0,
+			.phyclk_d18_mhz = 667.0,
+			.dscclk_mhz = 266.7,
+			.dtbclk_mhz = 600.0,
+		},
+		{
+			.state = 3,
+			.dcfclk_mhz = 800.0,
+			.fabricclk_mhz = 1400.0,
+			.socclk_mhz = 978.0,
+			.dram_speed_mts = 7500.0,
+			.dispclk_mhz = 960.0,
+			.dppclk_mhz = 960.0,
+			.phyclk_mhz = 810.0,
+			.phyclk_d18_mhz = 667.0,
+			.dscclk_mhz = 320.0,
+			.dtbclk_mhz = 600.0,
+		},
+		{
+			.state = 4,
+			.dcfclk_mhz = 873.0,
+			.fabricclk_mhz = 1600.0,
+			.socclk_mhz = 1100.0,
+			.dram_speed_mts = 8533.0,
+			.dispclk_mhz = 1066.7,
+			.dppclk_mhz = 1066.7,
+			.phyclk_mhz = 810.0,
+			.phyclk_d18_mhz = 667.0,
+			.dscclk_mhz = 355.6,
+			.dtbclk_mhz = 600.0,
+		},
+		{
+			.state = 5,
+			.dcfclk_mhz = 960.0,
+			.fabricclk_mhz = 1700.0,
+			.socclk_mhz = 1257.0,
+			.dram_speed_mts = 8533.0,
+			.dispclk_mhz = 1200.0,
+			.dppclk_mhz = 1200.0,
+			.phyclk_mhz = 810.0,
+			.phyclk_d18_mhz = 667.0,
+			.dscclk_mhz = 400.0,
+			.dtbclk_mhz = 600.0,
+		},
+		{
+			.state = 6,
+			.dcfclk_mhz = 1067.0,
+			.fabricclk_mhz = 1850.0,
+			.socclk_mhz = 1257.0,
+			.dram_speed_mts = 8533.0,
+			.dispclk_mhz = 1371.4,
+			.dppclk_mhz = 1371.4,
+			.phyclk_mhz = 810.0,
+			.phyclk_d18_mhz = 667.0,
+			.dscclk_mhz = 457.1,
+			.dtbclk_mhz = 600.0,
+		},
+		{
+			.state = 7,
+			.dcfclk_mhz = 1200.0,
+			.fabricclk_mhz = 2000.0,
+			.socclk_mhz = 1467.0,
+			.dram_speed_mts = 8533.0,
+			.dispclk_mhz = 1600.0,
+			.dppclk_mhz = 1600.0,
+			.phyclk_mhz = 810.0,
+			.phyclk_d18_mhz = 667.0,
+			.dscclk_mhz = 533.3,
+			.dtbclk_mhz = 600.0,
+		},
+	},
+	.num_states = 8,
+	.sr_exit_time_us = 28.0,
+	.sr_enter_plus_exit_time_us = 30.0,
+	.sr_exit_z8_time_us = 263.0,
+	.sr_enter_plus_exit_z8_time_us = 363.0,
+	.fclk_change_latency_us = 24.0,
+	.usr_retraining_latency_us = 2,
+	.writeback_latency_us = 12.0,
+
+	.dram_channel_width_bytes = 4,/*not exist in dml2*/
+	.round_trip_ping_latency_dcfclk_cycles = 106,/*not exist in dml2*/
+	.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,
+	.dram_clock_change_latency_us = 34,
+	.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_sdp_bw_after_urgent = 80.0,
+	.pct_ideal_fabric_bw_after_urgent = 80.0, /*new to dml2*/
+	.pct_ideal_dram_sdp_bw_after_urgent_pixel_only = 65.0,
+	.pct_ideal_dram_sdp_bw_after_urgent_pixel_and_vm = 60.0,
+	.pct_ideal_dram_sdp_bw_after_urgent_vm_only = 30.0,
+	.max_avg_sdp_bw_use_normal_percent = 60.0,
+	.max_avg_dram_bw_use_normal_percent = 60.0,
+	.fabric_datapath_to_dcn_data_return_bytes = 32,
+	.return_bus_width_bytes = 64,
+	.downspread_percent = 0.38,
+	.dcn_downspread_percent = 0.5,
+	.gpuvm_min_page_size_bytes = 4096,
+	.hostvm_min_page_size_bytes = 4096,
+	.do_urgent_latency_adjustment = 0,
+	.urgent_latency_adjustment_fabric_clock_component_us = 0,
+	.urgent_latency_adjustment_fabric_clock_reference_mhz = 0,
+	.num_chans = 4,
+	.dispclk_dppclk_vco_speed_mhz = 2400.0,
+};
+
+/*
+ * dcn351_update_bw_bounding_box
+ *
+ * This would override some dcn3_51 ip_or_soc initial parameters hardcoded from
+ * spreadsheet with actual values as per dGPU SKU:
+ * - with passed few options from dc->config
+ * - with dentist_vco_frequency from Clk Mgr (currently hardcoded, but might
+ *   need to get it from PM FW)
+ * - with passed latency values (passed in ns units) in dc-> bb override for
+ *   debugging purposes
+ * - with passed latencies from VBIOS (in 100_ns units) if available for
+ *   certain dGPU SKU
+ * - with number of DRAM channels from VBIOS (which differ for certain dGPU SKU
+ *   of the same ASIC)
+ * - clocks levels with passed clk_table entries from Clk Mgr as reported by PM
+ *   FW for different clocks (which might differ for certain dGPU SKU of the
+ *   same ASIC)
+ */
+void dcn351_update_bw_bounding_box_fpu(struct dc *dc,
+				      struct clk_bw_params *bw_params)
+{
+	unsigned int i, closest_clk_lvl;
+	int j;
+	struct clk_limit_table *clk_table = &bw_params->clk_table;
+	struct _vcs_dpi_voltage_scaling_st *clock_limits =
+		dc->scratch.update_bw_bounding_box.clock_limits;
+	unsigned int max_dispclk_mhz = 0, max_dppclk_mhz = 0;
+
+	dc_assert_fp_enabled();
+
+	dcn3_51_ip.max_num_otg =
+		dc->res_pool->res_cap->num_timing_generator;
+	dcn3_51_ip.max_num_dpp = dc->res_pool->pipe_count;
+	dcn3_51_soc.num_chans = bw_params->num_channels;
+
+	ASSERT(clk_table->num_entries);
+
+	/* Prepass to find max clocks independent of voltage level. */
+	for (i = 0; i < clk_table->num_entries; ++i) {
+		if (clk_table->entries[i].dispclk_mhz > max_dispclk_mhz)
+			max_dispclk_mhz = clk_table->entries[i].dispclk_mhz;
+		if (clk_table->entries[i].dppclk_mhz > max_dppclk_mhz)
+			max_dppclk_mhz = clk_table->entries[i].dppclk_mhz;
+	}
+
+	for (i = 0; i < clk_table->num_entries; i++) {
+		/* loop backwards*/
+		for (closest_clk_lvl = 0, j = dcn3_51_soc.num_states - 1;
+			j >= 0; j--) {
+			if (dcn3_51_soc.clock_limits[j].dcfclk_mhz <=
+				clk_table->entries[i].dcfclk_mhz) {
+				closest_clk_lvl = j;
+				break;
+			}
+		}
+		if (clk_table->num_entries == 1) {
+			/*smu gives one DPM level, let's take the highest one*/
+			closest_clk_lvl = dcn3_51_soc.num_states - 1;
+		}
+
+		clock_limits[i].state = i;
+
+		/* Clocks dependent on voltage level. */
+		clock_limits[i].dcfclk_mhz = clk_table->entries[i].dcfclk_mhz;
+		if (clk_table->num_entries == 1 &&
+			clock_limits[i].dcfclk_mhz <
+			dcn3_51_soc.clock_limits[closest_clk_lvl].dcfclk_mhz) {
+			/*SMU fix not released yet*/
+			clock_limits[i].dcfclk_mhz =
+				dcn3_51_soc.clock_limits[closest_clk_lvl].dcfclk_mhz;
+		}
+
+		clock_limits[i].fabricclk_mhz =
+			clk_table->entries[i].fclk_mhz;
+		clock_limits[i].socclk_mhz =
+			clk_table->entries[i].socclk_mhz;
+
+		if (clk_table->entries[i].memclk_mhz &&
+			clk_table->entries[i].wck_ratio)
+			clock_limits[i].dram_speed_mts =
+				clk_table->entries[i].memclk_mhz * 2 *
+				clk_table->entries[i].wck_ratio;
+
+		/* Clocks independent of voltage level. */
+		clock_limits[i].dispclk_mhz = max_dispclk_mhz ?
+			max_dispclk_mhz :
+			dcn3_51_soc.clock_limits[closest_clk_lvl].dispclk_mhz;
+
+		clock_limits[i].dppclk_mhz = max_dppclk_mhz ?
+			max_dppclk_mhz :
+			dcn3_51_soc.clock_limits[closest_clk_lvl].dppclk_mhz;
+
+		clock_limits[i].dram_bw_per_chan_gbps =
+			dcn3_51_soc.clock_limits[closest_clk_lvl].dram_bw_per_chan_gbps;
+		clock_limits[i].dscclk_mhz =
+			dcn3_51_soc.clock_limits[closest_clk_lvl].dscclk_mhz;
+		clock_limits[i].dtbclk_mhz =
+			dcn3_51_soc.clock_limits[closest_clk_lvl].dtbclk_mhz;
+		clock_limits[i].phyclk_d18_mhz =
+			dcn3_51_soc.clock_limits[closest_clk_lvl].phyclk_d18_mhz;
+		clock_limits[i].phyclk_mhz =
+			dcn3_51_soc.clock_limits[closest_clk_lvl].phyclk_mhz;
+	}
+
+	memcpy(dcn3_51_soc.clock_limits, clock_limits,
+		sizeof(dcn3_51_soc.clock_limits));
+
+	if (clk_table->num_entries)
+		dcn3_51_soc.num_states = clk_table->num_entries;
+
+	if (max_dispclk_mhz) {
+		dcn3_51_soc.dispclk_dppclk_vco_speed_mhz = max_dispclk_mhz * 2;
+		dc->dml.soc.dispclk_dppclk_vco_speed_mhz = max_dispclk_mhz * 2;
+	}
+	if ((int)(dcn3_51_soc.dram_clock_change_latency_us * 1000)
+				!= dc->debug.dram_clock_change_latency_ns
+			&& dc->debug.dram_clock_change_latency_ns) {
+		dcn3_51_soc.dram_clock_change_latency_us =
+			dc->debug.dram_clock_change_latency_ns / 1000.0;
+	}
+
+	if (dc->bb_overrides.dram_clock_change_latency_ns > 0)
+		dcn3_51_soc.dram_clock_change_latency_us =
+			dc->bb_overrides.dram_clock_change_latency_ns / 1000.0;
+
+	if (dc->bb_overrides.sr_exit_time_ns > 0)
+		dcn3_51_soc.sr_exit_time_us = dc->bb_overrides.sr_exit_time_ns / 1000.0;
+
+	if (dc->bb_overrides.sr_enter_plus_exit_time_ns > 0)
+		dcn3_51_soc.sr_enter_plus_exit_time_us =
+			dc->bb_overrides.sr_enter_plus_exit_time_ns / 1000.0;
+
+	if (dc->bb_overrides.sr_exit_z8_time_ns > 0)
+		dcn3_51_soc.sr_exit_z8_time_us = dc->bb_overrides.sr_exit_z8_time_ns / 1000.0;
+
+	if (dc->bb_overrides.sr_enter_plus_exit_z8_time_ns > 0)
+		dcn3_51_soc.sr_enter_plus_exit_z8_time_us =
+			dc->bb_overrides.sr_enter_plus_exit_z8_time_ns / 1000.0;
+
+	/*temp till dml2 fully work without dml1*/
+	dml_init_instance(&dc->dml, &dcn3_51_soc, &dcn3_51_ip,
+			  DML_PROJECT_DCN31);
+
+	/*copy to dml2, before dml2_create*/
+	if (clk_table->num_entries > 2) {
+
+		for (i = 0; i < clk_table->num_entries; i++) {
+			dc->dml2_options.bbox_overrides.clks_table.num_states =
+				clk_table->num_entries;
+			dc->dml2_options.bbox_overrides.clks_table.clk_entries[i].dcfclk_mhz =
+				(unsigned int)clock_limits[i].dcfclk_mhz;
+			dc->dml2_options.bbox_overrides.clks_table.clk_entries[i].fclk_mhz =
+				(unsigned int)clock_limits[i].fabricclk_mhz;
+			dc->dml2_options.bbox_overrides.clks_table.clk_entries[i].dispclk_mhz =
+				(unsigned int)clock_limits[i].dispclk_mhz;
+			dc->dml2_options.bbox_overrides.clks_table.clk_entries[i].dppclk_mhz =
+				(unsigned int)clock_limits[i].dppclk_mhz;
+			dc->dml2_options.bbox_overrides.clks_table.clk_entries[i].socclk_mhz =
+				(unsigned int)clock_limits[i].socclk_mhz;
+			dc->dml2_options.bbox_overrides.clks_table.clk_entries[i].memclk_mhz =
+				clk_table->entries[i].memclk_mhz * clk_table->entries[i].wck_ratio;
+			dc->dml2_options.bbox_overrides.clks_table.clk_entries[i].dram_speed_mts = (unsigned int)clock_limits[i].dram_speed_mts;
+			dc->dml2_options.bbox_overrides.clks_table.clk_entries[i].dtbclk_mhz =
+				(unsigned int)clock_limits[i].dtbclk_mhz;
+			dc->dml2_options.bbox_overrides.clks_table.num_entries_per_clk.num_dcfclk_levels =
+				clk_table->num_entries;
+			dc->dml2_options.bbox_overrides.clks_table.num_entries_per_clk.num_fclk_levels =
+				clk_table->num_entries;
+			dc->dml2_options.bbox_overrides.clks_table.num_entries_per_clk.num_dispclk_levels =
+				clk_table->num_entries;
+			dc->dml2_options.bbox_overrides.clks_table.num_entries_per_clk.num_dppclk_levels =
+				clk_table->num_entries;
+			dc->dml2_options.bbox_overrides.clks_table.num_entries_per_clk.num_socclk_levels =
+				clk_table->num_entries;
+			dc->dml2_options.bbox_overrides.clks_table.num_entries_per_clk.num_memclk_levels =
+				clk_table->num_entries;
+			dc->dml2_options.bbox_overrides.clks_table.num_entries_per_clk.num_dtbclk_levels =
+				clk_table->num_entries;
+		}
+	}
+
+	/* Update latency values */
+	dc->dml2_options.bbox_overrides.dram_clock_change_latency_us = dcn3_51_soc.dram_clock_change_latency_us;
+
+	dc->dml2_options.bbox_overrides.sr_exit_latency_us = dcn3_51_soc.sr_exit_time_us;
+	dc->dml2_options.bbox_overrides.sr_enter_plus_exit_latency_us = dcn3_51_soc.sr_enter_plus_exit_time_us;
+
+	dc->dml2_options.bbox_overrides.sr_exit_z8_time_us = dcn3_51_soc.sr_exit_z8_time_us;
+	dc->dml2_options.bbox_overrides.sr_enter_plus_exit_z8_time_us = dcn3_51_soc.sr_enter_plus_exit_z8_time_us;
+}
+
+static bool is_dual_plane(enum surface_pixel_format format)
+{
+	return format >= SURFACE_PIXEL_FORMAT_VIDEO_BEGIN ||
+		format == SURFACE_PIXEL_FORMAT_GRPH_RGBE_ALPHA;
+}
+
+/*
+ * micro_sec_to_vert_lines () - converts time to number of vertical lines for a given timing
+ *
+ * @param: num_us: number of microseconds
+ * @return: number of vertical lines. If exact number of vertical lines is not found then
+ *          it will round up to next number of lines to guarantee num_us
+ */
+static unsigned int micro_sec_to_vert_lines(unsigned int num_us, struct dc_crtc_timing *timing)
+{
+	unsigned int num_lines = 0;
+	double lines_time_in_ns = 1000.0 *
+			(((double)timing->h_total * 1000.0) /
+			 ((double)timing->pix_clk_100hz / 10.0));
+
+	num_lines = (unsigned int)dml_ceil(1000.0 * num_us / lines_time_in_ns, 1.0);
+
+	return num_lines;
+}
+
+static unsigned int get_vertical_back_porch(struct dc_crtc_timing *timing)
+{
+	unsigned int v_active = 0, v_blank = 0, v_back_porch = 0;
+
+	v_active = timing->v_border_top + timing->v_addressable + timing->v_border_bottom;
+	v_blank = timing->v_total - v_active;
+	v_back_porch = v_blank - timing->v_front_porch - timing->v_sync_width;
+
+	return v_back_porch;
+}
+
+int dcn351_populate_dml_pipes_from_context_fpu(struct dc *dc,
+					      struct dc_state *context,
+					      display_e2e_pipe_params_st *pipes,
+					      enum dc_validate_mode validate_mode)
+{
+	int pipe_cnt;
+	unsigned int i;
+	struct resource_context *res_ctx = &context->res_ctx;
+	struct pipe_ctx *pipe = 0;
+	bool upscaled = false;
+	const unsigned int max_allowed_vblank_nom = 1023;
+
+	dc_assert_fp_enabled();
+
+	dcn31_populate_dml_pipes_from_context(dc, context, pipes,
+					      validate_mode);
+
+	for (i = 0, pipe_cnt = 0; i < dc->res_pool->pipe_count; i++) {
+		struct dc_crtc_timing *timing;
+		unsigned int num_lines = 0;
+		unsigned int v_back_porch = 0;
+
+		if (!res_ctx->pipe_ctx[i].stream)
+			continue;
+
+		pipe = &res_ctx->pipe_ctx[i];
+		timing = &pipe->stream->timing;
+
+		num_lines = micro_sec_to_vert_lines(dcn3_51_ip.VBlankNomDefaultUS, timing);
+		v_back_porch  = get_vertical_back_porch(timing);
+
+		if (pipe->stream->adjust.v_total_max ==
+		    pipe->stream->adjust.v_total_min &&
+		    pipe->stream->adjust.v_total_min > timing->v_total) {
+			pipes[pipe_cnt].pipe.dest.vtotal =
+				pipe->stream->adjust.v_total_min;
+			pipes[pipe_cnt].pipe.dest.vblank_nom = timing->v_total -
+				pipes[pipe_cnt].pipe.dest.vactive;
+		}
+
+		pipes[pipe_cnt].pipe.dest.vblank_nom = timing->v_total - pipes[pipe_cnt].pipe.dest.vactive;
+		pipes[pipe_cnt].pipe.dest.vblank_nom = min(pipes[pipe_cnt].pipe.dest.vblank_nom, num_lines);
+		// vblank_nom should not smaller than (VSync (timing->v_sync_width + v_back_porch) + 2)
+		// + 2 is because
+		// 1 -> VStartup_start should be 1 line before VSync
+		// 1 -> always reserve 1 line between start of vblank to vstartup signal
+		pipes[pipe_cnt].pipe.dest.vblank_nom =
+			max(pipes[pipe_cnt].pipe.dest.vblank_nom, timing->v_sync_width + v_back_porch + 2);
+		pipes[pipe_cnt].pipe.dest.vblank_nom = min(pipes[pipe_cnt].pipe.dest.vblank_nom, max_allowed_vblank_nom);
+
+		if (pipe->plane_state &&
+		    (pipe->plane_state->src_rect.height <
+		     pipe->plane_state->dst_rect.height ||
+		     pipe->plane_state->src_rect.width <
+		     pipe->plane_state->dst_rect.width))
+			upscaled = true;
+
+		/*
+		 * Immediate flip can be set dynamically after enabling the
+		 * plane. We need to require support for immediate flip or
+		 * underflow can be intermittently experienced depending on peak
+		 * b/w requirements.
+		 */
+		pipes[pipe_cnt].pipe.src.immediate_flip = true;
+
+		pipes[pipe_cnt].pipe.src.unbounded_req_mode = false;
+
+		dcn31_zero_pipe_dcc_fraction(pipes, pipe_cnt);
+
+		pipes[pipe_cnt].pipe.dest.vfront_porch = timing->v_front_porch;
+		pipes[pipe_cnt].pipe.src.dcc_rate = 3;
+		pipes[pipe_cnt].dout.dsc_input_bpc = 0;
+		pipes[pipe_cnt].pipe.src.gpuvm_min_page_size_kbytes = 256;
+
+		if (pipes[pipe_cnt].dout.dsc_enable) {
+			switch (timing->display_color_depth) {
+			case COLOR_DEPTH_888:
+				pipes[pipe_cnt].dout.dsc_input_bpc = 8;
+				break;
+			case COLOR_DEPTH_101010:
+				pipes[pipe_cnt].dout.dsc_input_bpc = 10;
+				break;
+			case COLOR_DEPTH_121212:
+				pipes[pipe_cnt].dout.dsc_input_bpc = 12;
+				break;
+			default:
+				ASSERT(0);
+				break;
+			}
+		}
+
+		pipe_cnt++;
+	}
+
+	context->bw_ctx.dml.ip.det_buffer_size_kbytes = 384;/*per guide*/
+
+	if (pipe_cnt == 1 && pipe->plane_state && !dc->debug.disable_z9_mpc) {
+		if (!is_dual_plane(pipe->plane_state->format) &&
+			   pipe->plane_state->src_rect.width <= 5120) {
+			/*
+			 * Limit to 5k max to avoid forced pipe split when there
+			 * is not enough detile for swath
+			 */
+			context->bw_ctx.dml.ip.det_buffer_size_kbytes = 192;
+			pipes[0].pipe.src.unbounded_req_mode = true;
+		}
+	} else if (context->stream_count >=
+		   dc->debug.crb_alloc_policy_min_disp_count &&
+		   dc->debug.crb_alloc_policy > DET_SIZE_DEFAULT) {
+		context->bw_ctx.dml.ip.det_buffer_size_kbytes =
+			dc->debug.crb_alloc_policy * 64;
+	} else if (context->stream_count >= 3 && upscaled) {
+		context->bw_ctx.dml.ip.det_buffer_size_kbytes = 192;
+	}
+
+	for (i = 0; i < dc->res_pool->pipe_count; i++) {
+		struct pipe_ctx *cur_pipe = &context->res_ctx.pipe_ctx[i];
+
+		if (!cur_pipe->stream)
+			continue;
+
+		if (cur_pipe->stream->signal == SIGNAL_TYPE_EDP &&
+		    dc->debug.seamless_boot_odm_combine &&
+		    cur_pipe->stream->apply_seamless_boot_optimization) {
+
+			if (cur_pipe->stream->apply_boot_odm_mode ==
+			    dm_odm_combine_policy_2to1) {
+				context->bw_ctx.dml.vba.ODMCombinePolicy =
+					dm_odm_combine_policy_2to1;
+				break;
+			}
+		}
+	}
+
+	return pipe_cnt;
+}
+
+void dcn351_decide_zstate_support(struct dc *dc, struct dc_state *context)
+{
+	enum dcn_zstate_support_state support = DCN_ZSTATE_SUPPORT_DISALLOW;
+	unsigned int i, plane_count = 0;
+
+	for (i = 0; i < dc->res_pool->pipe_count; i++) {
+		if (context->res_ctx.pipe_ctx[i].plane_state)
+			plane_count++;
+	}
+
+	/*dcn351 does not support z9/z10*/
+	if (context->stream_count == 0 || plane_count == 0) {
+		support = DCN_ZSTATE_SUPPORT_ALLOW_Z8_ONLY;
+	} else if (context->stream_count == 1 && context->streams[0]->signal == SIGNAL_TYPE_EDP) {
+		struct dc_link *link = context->streams[0]->sink->link;
+		bool is_pwrseq0 = link && link->link_index == 0;
+		bool is_psr = (link && (link->psr_settings.psr_version == DC_PSR_VERSION_1 ||
+								link->psr_settings.psr_version == DC_PSR_VERSION_SU_1) && !link->panel_config.psr.disable_psr);
+		bool is_replay = link && link->replay_settings.replay_feature_enabled;
+		int minmum_z8_residency =
+			dc->debug.minimum_z8_residency_time > 0 ? dc->debug.minimum_z8_residency_time : 1000;
+		bool allow_z8 = context->bw_ctx.dml.vba.StutterPeriod > (double)minmum_z8_residency;
+
+		/*for psr1/psr-su, we allow z8 and z10 based on latency, for replay with IPS enabled, it will enter ips2*/
+		if (is_pwrseq0 && (is_psr || is_replay))
+			support = allow_z8 ? allow_z8 : DCN_ZSTATE_SUPPORT_DISALLOW;
+	}
+	context->bw_ctx.bw.dcn.clk.zstate_support = support;
+}

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dcn351/dcn351_fpu.h

@@ -0,0 +1,19 @@
+/* SPDX-License-Identifier: MIT */
+/* Copyright 2024 Advanced Micro Devices, Inc. */
+
+#ifndef __DCN351_FPU_H__
+#define __DCN351_FPU_H__
+
+#include "clk_mgr.h"
+
+void dcn351_update_bw_bounding_box_fpu(struct dc *dc,
+				      struct clk_bw_params *bw_params);
+
+int dcn351_populate_dml_pipes_from_context_fpu(struct dc *dc,
+					      struct dc_state *context,
+					      display_e2e_pipe_params_st *pipes,
+					      enum dc_validate_mode validate_mode);
+
+void dcn351_decide_zstate_support(struct dc *dc, struct dc_state *context);
+
+#endif

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/display_mode_enums.h

@@ -0,0 +1,312 @@
+/*
+ * Copyright 2017 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 __DISPLAY_MODE_ENUMS_H__
+#define __DISPLAY_MODE_ENUMS_H__
+
+enum output_encoder_class {
+	dm_dp = 0,
+	dm_hdmi = 1,
+	dm_wb = 2,
+	dm_edp = 3,
+	dm_hdmifrl = 4,
+	dm_dp2p0 = 5,
+};
+enum output_format_class {
+	dm_444 = 0, dm_420 = 1, dm_n422, dm_s422
+};
+enum source_format_class {
+	dm_444_16 = 0,
+	dm_444_32 = 1,
+	dm_444_64 = 2,
+	dm_420_8 = 3,
+	dm_420_10 = 4,
+	dm_420_12 = 5,
+	dm_422_8 = 6,
+	dm_422_10 = 7,
+	dm_444_8 = 8,
+	dm_mono_8 = dm_444_8,
+	dm_mono_16 = dm_444_16,
+	dm_rgbe = 9,
+	dm_rgbe_alpha = 10,
+};
+enum output_bpc_class {
+	dm_out_6 = 0, dm_out_8 = 1, dm_out_10 = 2, dm_out_12 = 3, dm_out_16 = 4
+};
+enum scan_direction_class {
+	dm_horz = 0, dm_vert = 1
+};
+enum dm_swizzle_mode {
+	dm_sw_linear = 0,
+	dm_sw_256b_s = 1,
+	dm_sw_256b_d = 2,
+	dm_sw_SPARE_0 = 3,
+	dm_sw_SPARE_1 = 4,
+	dm_sw_4kb_s = 5,
+	dm_sw_4kb_d = 6,
+	dm_sw_SPARE_2 = 7,
+	dm_sw_SPARE_3 = 8,
+	dm_sw_64kb_s = 9,
+	dm_sw_64kb_d = 10,
+	dm_sw_SPARE_4 = 11,
+	dm_sw_SPARE_5 = 12,
+	dm_sw_var_s = 13,
+	dm_sw_var_d = 14,
+	dm_sw_SPARE_6 = 15,
+	dm_sw_SPARE_7 = 16,
+	dm_sw_64kb_s_t = 17,
+	dm_sw_64kb_d_t = 18,
+	dm_sw_SPARE_10 = 19,
+	dm_sw_SPARE_11 = 20,
+	dm_sw_4kb_s_x = 21,
+	dm_sw_4kb_d_x = 22,
+	dm_sw_SPARE_12 = 23,
+	dm_sw_SPARE_13 = 24,
+	dm_sw_64kb_s_x = 25,
+	dm_sw_64kb_d_x = 26,
+	dm_sw_64kb_r_x = 27,
+	dm_sw_SPARE_15 = 28,
+	dm_sw_var_s_x = 29,
+	dm_sw_var_d_x = 30,
+	dm_sw_var_r_x = 31,
+	dm_sw_gfx7_2d_thin_l_vp,
+	dm_sw_gfx7_2d_thin_gl,
+};
+enum lb_depth {
+	dm_lb_10 = 0, dm_lb_8 = 1, dm_lb_6 = 2, dm_lb_12 = 3, dm_lb_16 = 4,
+	dm_lb_19 = 5
+};
+enum voltage_state {
+	dm_vmin = 0, dm_vmid = 1, dm_vnom = 2, dm_vmax = 3
+};
+enum source_macro_tile_size {
+	dm_4k_tile = 0, dm_64k_tile = 1, dm_256k_tile = 2
+};
+enum cursor_bpp {
+	dm_cur_2bit = 0, dm_cur_32bit = 1, dm_cur_64bit = 2
+};
+
+/**
+ * @enum clock_change_support - It represents possible reasons to change the DRAM clock.
+ *
+ * DC may change the DRAM clock during its execution, and this enum tracks all
+ * the available methods. Note that every ASIC has their specific way to deal
+ * with these clock switch.
+ */
+enum clock_change_support {
+	/**
+	 * @dm_dram_clock_change_uninitialized: If you see this, we might have
+	 * a code initialization issue
+	 */
+	dm_dram_clock_change_uninitialized = 0,
+
+	/**
+	 * @dm_dram_clock_change_vactive: Support DRAM switch in VActive
+	 */
+	dm_dram_clock_change_vactive,
+
+	/**
+	 * @dm_dram_clock_change_vblank: Support DRAM switch in VBlank
+	 */
+	dm_dram_clock_change_vblank,
+
+	dm_dram_clock_change_vactive_w_mall_full_frame,
+	dm_dram_clock_change_vactive_w_mall_sub_vp,
+	dm_dram_clock_change_vblank_w_mall_full_frame,
+	dm_dram_clock_change_vblank_w_mall_sub_vp,
+
+	/**
+	 * @dm_dram_clock_change_unsupported: Do not support DRAM switch
+	 */
+	dm_dram_clock_change_unsupported
+};
+
+enum output_standard {
+	dm_std_uninitialized = 0,
+	dm_std_cvtr2,
+	dm_std_cvt
+};
+
+enum mpc_combine_affinity {
+	dm_mpc_always_when_possible,
+	dm_mpc_reduce_voltage,
+	dm_mpc_reduce_voltage_and_clocks,
+	dm_mpc_never
+};
+
+enum RequestType {
+	REQ_256Bytes, REQ_128BytesNonContiguous, REQ_128BytesContiguous, REQ_NA
+};
+
+enum self_refresh_affinity {
+	dm_try_to_allow_self_refresh_and_mclk_switch,
+	dm_allow_self_refresh_and_mclk_switch,
+	dm_allow_self_refresh,
+	dm_neither_self_refresh_nor_mclk_switch
+};
+
+enum dm_validation_status {
+	DML_VALIDATION_OK,
+	DML_FAIL_SCALE_RATIO_TAP,
+	DML_FAIL_SOURCE_PIXEL_FORMAT,
+	DML_FAIL_VIEWPORT_SIZE,
+	DML_FAIL_TOTAL_V_ACTIVE_BW,
+	DML_FAIL_DIO_SUPPORT,
+	DML_FAIL_NOT_ENOUGH_DSC,
+	DML_FAIL_DSC_CLK_REQUIRED,
+	DML_FAIL_DSC_VALIDATION_FAILURE,
+	DML_FAIL_URGENT_LATENCY,
+	DML_FAIL_REORDERING_BUFFER,
+	DML_FAIL_DISPCLK_DPPCLK,
+	DML_FAIL_TOTAL_AVAILABLE_PIPES,
+	DML_FAIL_NUM_OTG,
+	DML_FAIL_WRITEBACK_MODE,
+	DML_FAIL_WRITEBACK_LATENCY,
+	DML_FAIL_WRITEBACK_SCALE_RATIO_TAP,
+	DML_FAIL_CURSOR_SUPPORT,
+	DML_FAIL_PITCH_SUPPORT,
+	DML_FAIL_PTE_BUFFER_SIZE,
+	DML_FAIL_HOST_VM_IMMEDIATE_FLIP,
+	DML_FAIL_DSC_INPUT_BPC,
+	DML_FAIL_PREFETCH_SUPPORT,
+	DML_FAIL_V_RATIO_PREFETCH,
+	DML_FAIL_P2I_WITH_420,
+	DML_FAIL_DSC_ONLY_IF_NECESSARY_WITH_BPP,
+	DML_FAIL_NOT_DSC422_NATIVE,
+	DML_FAIL_ODM_COMBINE4TO1,
+	DML_FAIL_ENOUGH_WRITEBACK_UNITS,
+	DML_FAIL_VIEWPORT_EXCEEDS_SURFACE,
+	DML_FAIL_DYNAMIC_METADATA,
+	DML_FAIL_FMT_BUFFER_EXCEEDED,
+};
+
+enum writeback_config {
+	dm_normal,
+	dm_whole_buffer_for_single_stream_no_interleave,
+	dm_whole_buffer_for_single_stream_interleave,
+};
+
+enum odm_combine_mode {
+	dm_odm_combine_mode_disabled,
+	dm_odm_combine_mode_2to1,
+	dm_odm_combine_mode_4to1,
+	dm_odm_split_mode_1to2,
+	dm_odm_mode_mso_1to2,
+	dm_odm_mode_mso_1to4
+};
+
+enum odm_combine_policy {
+	dm_odm_combine_policy_dal,
+	dm_odm_combine_policy_none,
+	dm_odm_combine_policy_2to1,
+	dm_odm_combine_policy_4to1,
+	dm_odm_split_policy_1to2,
+	dm_odm_mso_policy_1to2,
+	dm_odm_mso_policy_1to4,
+};
+
+enum immediate_flip_requirement {
+	dm_immediate_flip_not_required,
+	dm_immediate_flip_required,
+	dm_immediate_flip_opportunistic,
+};
+
+enum unbounded_requesting_policy {
+	dm_unbounded_requesting,
+	dm_unbounded_requesting_edp_only,
+	dm_unbounded_requesting_disable
+};
+
+enum dm_rotation_angle {
+	dm_rotation_0,
+	dm_rotation_90,
+	dm_rotation_180,
+	dm_rotation_270,
+	dm_rotation_0m,
+	dm_rotation_90m,
+	dm_rotation_180m,
+	dm_rotation_270m,
+};
+
+enum dm_use_mall_for_pstate_change_mode {
+	dm_use_mall_pstate_change_disable,
+	dm_use_mall_pstate_change_full_frame,
+	dm_use_mall_pstate_change_sub_viewport,
+	dm_use_mall_pstate_change_phantom_pipe
+};
+
+enum dm_use_mall_for_static_screen_mode {
+	dm_use_mall_static_screen_disable,
+	dm_use_mall_static_screen_optimize,
+	dm_use_mall_static_screen_enable,
+};
+
+enum dm_output_link_dp_rate {
+	dm_dp_rate_na,
+	dm_dp_rate_hbr,
+	dm_dp_rate_hbr2,
+	dm_dp_rate_hbr3,
+	dm_dp_rate_uhbr10,
+	dm_dp_rate_uhbr13p5,
+	dm_dp_rate_uhbr20,
+};
+
+enum dm_fclock_change_support {
+	dm_fclock_change_vactive,
+	dm_fclock_change_vblank,
+	dm_fclock_change_unsupported,
+};
+
+enum dm_prefetch_modes {
+	dm_prefetch_support_uclk_fclk_and_stutter_if_possible,
+	dm_prefetch_support_uclk_fclk_and_stutter,
+	dm_prefetch_support_fclk_and_stutter,
+	dm_prefetch_support_stutter,
+	dm_prefetch_support_none,
+};
+enum dm_output_type {
+	dm_output_type_unknown,
+	dm_output_type_dp,
+	dm_output_type_edp,
+	dm_output_type_dp2p0,
+	dm_output_type_hdmi,
+	dm_output_type_hdmifrl,
+};
+
+enum dm_output_rate {
+	dm_output_rate_unknown,
+	dm_output_rate_dp_rate_hbr,
+	dm_output_rate_dp_rate_hbr2,
+	dm_output_rate_dp_rate_hbr3,
+	dm_output_rate_dp_rate_uhbr10,
+	dm_output_rate_dp_rate_uhbr13p5,
+	dm_output_rate_dp_rate_uhbr20,
+	dm_output_rate_hdmi_rate_3x3,
+	dm_output_rate_hdmi_rate_6x3,
+	dm_output_rate_hdmi_rate_6x4,
+	dm_output_rate_hdmi_rate_8x4,
+	dm_output_rate_hdmi_rate_10x4,
+	dm_output_rate_hdmi_rate_12x4,
+};
+#endif

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.c

@@ -0,0 +1,321 @@
+/*
+ * Copyright 2017 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 "display_mode_lib.h"
+#include "dc_features.h"
+#include "dcn20/display_mode_vba_20.h"
+#include "dcn20/display_rq_dlg_calc_20.h"
+#include "dcn20/display_mode_vba_20v2.h"
+#include "dcn20/display_rq_dlg_calc_20v2.h"
+#include "dcn21/display_mode_vba_21.h"
+#include "dcn21/display_rq_dlg_calc_21.h"
+#include "dcn30/display_mode_vba_30.h"
+#include "dcn30/display_rq_dlg_calc_30.h"
+#include "dcn31/display_mode_vba_31.h"
+#include "dcn31/display_rq_dlg_calc_31.h"
+#include "dcn314/display_mode_vba_314.h"
+#include "dcn314/display_rq_dlg_calc_314.h"
+#include "dcn32/display_mode_vba_32.h"
+#include "dcn32/display_rq_dlg_calc_32.h"
+#include "dml_logger.h"
+
+static const struct dml_funcs dml20_funcs = {
+	.validate = dml20_ModeSupportAndSystemConfigurationFull,
+	.recalculate = dml20_recalculate,
+	.rq_dlg_get_dlg_reg = dml20_rq_dlg_get_dlg_reg,
+	.rq_dlg_get_rq_reg = dml20_rq_dlg_get_rq_reg
+};
+
+static const struct dml_funcs dml20v2_funcs = {
+	.validate = dml20v2_ModeSupportAndSystemConfigurationFull,
+	.recalculate = dml20v2_recalculate,
+	.rq_dlg_get_dlg_reg = dml20v2_rq_dlg_get_dlg_reg,
+	.rq_dlg_get_rq_reg = dml20v2_rq_dlg_get_rq_reg
+};
+
+static const struct dml_funcs dml21_funcs = {
+	.validate = dml21_ModeSupportAndSystemConfigurationFull,
+	.recalculate = dml21_recalculate,
+	.rq_dlg_get_dlg_reg = dml21_rq_dlg_get_dlg_reg,
+	.rq_dlg_get_rq_reg = dml21_rq_dlg_get_rq_reg
+};
+
+static const struct dml_funcs dml30_funcs = {
+	.validate = dml30_ModeSupportAndSystemConfigurationFull,
+	.recalculate = dml30_recalculate,
+	.rq_dlg_get_dlg_reg = dml30_rq_dlg_get_dlg_reg,
+	.rq_dlg_get_rq_reg = dml30_rq_dlg_get_rq_reg
+};
+
+static const struct dml_funcs dml31_funcs = {
+	.validate = dml31_ModeSupportAndSystemConfigurationFull,
+	.recalculate = dml31_recalculate,
+	.rq_dlg_get_dlg_reg = dml31_rq_dlg_get_dlg_reg,
+	.rq_dlg_get_rq_reg = dml31_rq_dlg_get_rq_reg
+};
+
+static const struct dml_funcs dml314_funcs = {
+	.validate = dml314_ModeSupportAndSystemConfigurationFull,
+	.recalculate = dml314_recalculate,
+	.rq_dlg_get_dlg_reg = dml314_rq_dlg_get_dlg_reg,
+	.rq_dlg_get_rq_reg = dml314_rq_dlg_get_rq_reg
+};
+
+static const struct dml_funcs dml32_funcs = {
+	.validate = dml32_ModeSupportAndSystemConfigurationFull,
+	.recalculate = dml32_recalculate,
+	.rq_dlg_get_dlg_reg_v2 = dml32_rq_dlg_get_dlg_reg,
+	.rq_dlg_get_rq_reg_v2 = dml32_rq_dlg_get_rq_reg
+};
+
+void dml_init_instance(struct display_mode_lib *lib,
+		const struct _vcs_dpi_soc_bounding_box_st *soc_bb,
+		const struct _vcs_dpi_ip_params_st *ip_params,
+		enum dml_project project)
+{
+	lib->soc = *soc_bb;
+	lib->ip = *ip_params;
+	lib->project = project;
+	switch (project) {
+	case DML_PROJECT_NAVI10:
+	case DML_PROJECT_DCN201:
+		lib->funcs = dml20_funcs;
+		break;
+	case DML_PROJECT_NAVI10v2:
+		lib->funcs = dml20v2_funcs;
+		break;
+        case DML_PROJECT_DCN21:
+                lib->funcs = dml21_funcs;
+                break;
+	case DML_PROJECT_DCN30:
+		lib->funcs = dml30_funcs;
+		break;
+	case DML_PROJECT_DCN31:
+	case DML_PROJECT_DCN315:
+		lib->funcs = dml31_funcs;
+		break;
+	case DML_PROJECT_DCN314:
+		lib->funcs = dml314_funcs;
+		break;
+	case DML_PROJECT_DCN32:
+		lib->funcs = dml32_funcs;
+		break;
+
+	default:
+		break;
+	}
+}
+
+const char *dml_get_status_message(enum dm_validation_status status)
+{
+	switch (status) {
+	case DML_VALIDATION_OK:                   return "Validation OK";
+	case DML_FAIL_SCALE_RATIO_TAP:            return "Scale ratio/tap";
+	case DML_FAIL_SOURCE_PIXEL_FORMAT:        return "Source pixel format";
+	case DML_FAIL_VIEWPORT_SIZE:              return "Viewport size";
+	case DML_FAIL_TOTAL_V_ACTIVE_BW:          return "Total vertical active bandwidth";
+	case DML_FAIL_DIO_SUPPORT:                return "DIO support";
+	case DML_FAIL_NOT_ENOUGH_DSC:             return "Not enough DSC Units";
+	case DML_FAIL_DSC_CLK_REQUIRED:           return "DSC clock required";
+	case DML_FAIL_URGENT_LATENCY:             return "Urgent latency";
+	case DML_FAIL_REORDERING_BUFFER:          return "Re-ordering buffer";
+	case DML_FAIL_DISPCLK_DPPCLK:             return "Dispclk and Dppclk";
+	case DML_FAIL_TOTAL_AVAILABLE_PIPES:      return "Total available pipes";
+	case DML_FAIL_NUM_OTG:                    return "Number of OTG";
+	case DML_FAIL_WRITEBACK_MODE:             return "Writeback mode";
+	case DML_FAIL_WRITEBACK_LATENCY:          return "Writeback latency";
+	case DML_FAIL_WRITEBACK_SCALE_RATIO_TAP:  return "Writeback scale ratio/tap";
+	case DML_FAIL_CURSOR_SUPPORT:             return "Cursor support";
+	case DML_FAIL_PITCH_SUPPORT:              return "Pitch support";
+	case DML_FAIL_PTE_BUFFER_SIZE:            return "PTE buffer size";
+	case DML_FAIL_DSC_INPUT_BPC:              return "DSC input bpc";
+	case DML_FAIL_PREFETCH_SUPPORT:           return "Prefetch support";
+	case DML_FAIL_V_RATIO_PREFETCH:           return "Vertical ratio prefetch";
+	default:                                  return "Unknown Status";
+	}
+}
+
+void dml_log_pipe_params(
+		struct display_mode_lib *mode_lib,
+		display_e2e_pipe_params_st *pipes,
+		int pipe_cnt)
+{
+	(void)mode_lib;
+	display_pipe_source_params_st *pipe_src;
+	display_pipe_dest_params_st   *pipe_dest;
+	scaler_ratio_depth_st         *scale_ratio_depth;
+	scaler_taps_st                *scale_taps;
+	display_output_params_st      *dout;
+	display_clocks_and_cfg_st     *clks_cfg;
+	int i;
+
+	for (i = 0; i < pipe_cnt; i++) {
+		pipe_src = &(pipes[i].pipe.src);
+		pipe_dest = &(pipes[i].pipe.dest);
+		scale_ratio_depth = &(pipes[i].pipe.scale_ratio_depth);
+		scale_taps = &(pipes[i].pipe.scale_taps);
+		dout = &(pipes[i].dout);
+		clks_cfg = &(pipes[i].clks_cfg);
+
+		dml_print("DML PARAMS: =====================================\n");
+		dml_print("DML PARAMS: PIPE [%d] SOURCE PARAMS:\n", i);
+		dml_print("DML PARAMS:     source_format              = %d\n", pipe_src->source_format);
+		dml_print("DML PARAMS:     dcc                        = %d\n", pipe_src->dcc);
+		dml_print("DML PARAMS:     dcc_rate                   = %d\n", pipe_src->dcc_rate);
+		dml_print("DML PARAMS:     dcc_use_global             = %d\n", pipe_src->dcc_use_global);
+		dml_print("DML PARAMS:     vm                         = %d\n", pipe_src->vm);
+		dml_print("DML PARAMS:     gpuvm                      = %d\n", pipe_src->gpuvm);
+		dml_print("DML PARAMS:     hostvm                     = %d\n", pipe_src->hostvm);
+		dml_print("DML PARAMS:     gpuvm_levels_force_en      = %d\n", pipe_src->gpuvm_levels_force_en);
+		dml_print("DML PARAMS:     gpuvm_levels_force         = %d\n", pipe_src->gpuvm_levels_force);
+		dml_print("DML PARAMS:     source_scan                = %d\n", pipe_src->source_scan);
+		dml_print("DML PARAMS:     sw_mode                    = %d\n", pipe_src->sw_mode);
+		dml_print("DML PARAMS:     macro_tile_size            = %d\n", pipe_src->macro_tile_size);
+		dml_print("DML PARAMS:     viewport_width             = %d\n", pipe_src->viewport_width);
+		dml_print("DML PARAMS:     viewport_height            = %d\n", pipe_src->viewport_height);
+		dml_print("DML PARAMS:     viewport_y_y               = %d\n", pipe_src->viewport_y_y);
+		dml_print("DML PARAMS:     viewport_y_c               = %d\n", pipe_src->viewport_y_c);
+		dml_print("DML PARAMS:     viewport_width_c           = %d\n", pipe_src->viewport_width_c);
+		dml_print("DML PARAMS:     viewport_height_c          = %d\n", pipe_src->viewport_height_c);
+		dml_print("DML PARAMS:     data_pitch                 = %d\n", pipe_src->data_pitch);
+		dml_print("DML PARAMS:     data_pitch_c               = %d\n", pipe_src->data_pitch_c);
+		dml_print("DML PARAMS:     meta_pitch                 = %d\n", pipe_src->meta_pitch);
+		dml_print("DML PARAMS:     meta_pitch_c               = %d\n", pipe_src->meta_pitch_c);
+		dml_print("DML PARAMS:     cur0_src_width             = %d\n", pipe_src->cur0_src_width);
+		dml_print("DML PARAMS:     cur0_bpp                   = %d\n", pipe_src->cur0_bpp);
+		dml_print("DML PARAMS:     cur1_src_width             = %d\n", pipe_src->cur1_src_width);
+		dml_print("DML PARAMS:     cur1_bpp                   = %d\n", pipe_src->cur1_bpp);
+		dml_print("DML PARAMS:     num_cursors                = %d\n", pipe_src->num_cursors);
+		dml_print("DML PARAMS:     is_hsplit                  = %d\n", pipe_src->is_hsplit);
+		dml_print("DML PARAMS:     hsplit_grp                 = %d\n", pipe_src->hsplit_grp);
+		dml_print("DML PARAMS:     dynamic_metadata_enable    = %d\n", pipe_src->dynamic_metadata_enable);
+		dml_print("DML PARAMS:     dmdata_lines_before_active = %d\n", pipe_src->dynamic_metadata_lines_before_active);
+		dml_print("DML PARAMS:     dmdata_xmit_bytes          = %d\n", pipe_src->dynamic_metadata_xmit_bytes);
+		dml_print("DML PARAMS:     immediate_flip             = %d\n", pipe_src->immediate_flip);
+		dml_print("DML PARAMS:     v_total_min                = %d\n", pipe_src->v_total_min);
+		dml_print("DML PARAMS:     v_total_max                = %d\n", pipe_src->v_total_max);
+		dml_print("DML PARAMS: =====================================\n");
+
+		dml_print("DML PARAMS: PIPE [%d] DESTINATION PARAMS:\n", i);
+		dml_print("DML PARAMS:     recout_width               = %d\n", pipe_dest->recout_width);
+		dml_print("DML PARAMS:     recout_height              = %d\n", pipe_dest->recout_height);
+		dml_print("DML PARAMS:     full_recout_width          = %d\n", pipe_dest->full_recout_width);
+		dml_print("DML PARAMS:     full_recout_height         = %d\n", pipe_dest->full_recout_height);
+		dml_print("DML PARAMS:     hblank_start               = %d\n", pipe_dest->hblank_start);
+		dml_print("DML PARAMS:     hblank_end                 = %d\n", pipe_dest->hblank_end);
+		dml_print("DML PARAMS:     vblank_start               = %d\n", pipe_dest->vblank_start);
+		dml_print("DML PARAMS:     vblank_end                 = %d\n", pipe_dest->vblank_end);
+		dml_print("DML PARAMS:     htotal                     = %d\n", pipe_dest->htotal);
+		dml_print("DML PARAMS:     vtotal                     = %d\n", pipe_dest->vtotal);
+		dml_print("DML PARAMS:     vactive                    = %d\n", pipe_dest->vactive);
+		dml_print("DML PARAMS:     hactive                    = %d\n", pipe_dest->hactive);
+		dml_print("DML PARAMS:     vstartup_start             = %d\n", pipe_dest->vstartup_start);
+		dml_print("DML PARAMS:     vupdate_offset             = %d\n", pipe_dest->vupdate_offset);
+		dml_print("DML PARAMS:     vupdate_width              = %d\n", pipe_dest->vupdate_width);
+		dml_print("DML PARAMS:     vready_offset              = %d\n", pipe_dest->vready_offset);
+		dml_print("DML PARAMS:     interlaced                 = %d\n", pipe_dest->interlaced);
+		dml_print("DML PARAMS:     pixel_rate_mhz             = %3.2f\n", pipe_dest->pixel_rate_mhz);
+		dml_print("DML PARAMS:     sync_vblank_all_planes     = %d\n", pipe_dest->synchronized_vblank_all_planes);
+		dml_print("DML PARAMS:     otg_inst                   = %d\n", pipe_dest->otg_inst);
+		dml_print("DML PARAMS:     odm_combine                = %d\n", pipe_dest->odm_combine);
+		dml_print("DML PARAMS:     use_maximum_vstartup       = %d\n", pipe_dest->use_maximum_vstartup);
+		dml_print("DML PARAMS:     vtotal_max                 = %d\n", pipe_dest->vtotal_max);
+		dml_print("DML PARAMS:     vtotal_min                 = %d\n", pipe_dest->vtotal_min);
+		dml_print("DML PARAMS: =====================================\n");
+
+		dml_print("DML PARAMS: PIPE [%d] SCALER PARAMS:\n", i);
+		dml_print("DML PARAMS:     hscl_ratio                 = %3.4f\n", scale_ratio_depth->hscl_ratio);
+		dml_print("DML PARAMS:     vscl_ratio                 = %3.4f\n", scale_ratio_depth->vscl_ratio);
+		dml_print("DML PARAMS:     hscl_ratio_c               = %3.4f\n", scale_ratio_depth->hscl_ratio_c);
+		dml_print("DML PARAMS:     vscl_ratio_c               = %3.4f\n", scale_ratio_depth->vscl_ratio_c);
+		dml_print("DML PARAMS:     vinit                      = %3.4f\n", scale_ratio_depth->vinit);
+		dml_print("DML PARAMS:     vinit_c                    = %3.4f\n", scale_ratio_depth->vinit_c);
+		dml_print("DML PARAMS:     vinit_bot                  = %3.4f\n", scale_ratio_depth->vinit_bot);
+		dml_print("DML PARAMS:     vinit_bot_c                = %3.4f\n", scale_ratio_depth->vinit_bot_c);
+		dml_print("DML PARAMS:     lb_depth                   = %d\n", scale_ratio_depth->lb_depth);
+		dml_print("DML PARAMS:     scl_enable                 = %d\n", scale_ratio_depth->scl_enable);
+		dml_print("DML PARAMS:     htaps                      = %d\n", scale_taps->htaps);
+		dml_print("DML PARAMS:     vtaps                      = %d\n", scale_taps->vtaps);
+		dml_print("DML PARAMS:     htaps_c                    = %d\n", scale_taps->htaps_c);
+		dml_print("DML PARAMS:     vtaps_c                    = %d\n", scale_taps->vtaps_c);
+		dml_print("DML PARAMS: =====================================\n");
+
+		dml_print("DML PARAMS: PIPE [%d] DISPLAY OUTPUT PARAMS:\n", i);
+		dml_print("DML PARAMS:     output_type                = %d\n", dout->output_type);
+		dml_print("DML PARAMS:     output_format              = %d\n", dout->output_format);
+		dml_print("DML PARAMS:     dsc_input_bpc              = %d\n", dout->dsc_input_bpc);
+		dml_print("DML PARAMS:     output_bpp                 = %3.4f\n", dout->output_bpp);
+		dml_print("DML PARAMS:     dp_lanes                   = %d\n", dout->dp_lanes);
+		dml_print("DML PARAMS:     dsc_enable                 = %d\n", dout->dsc_enable);
+		dml_print("DML PARAMS:     dsc_slices                 = %d\n", dout->dsc_slices);
+		dml_print("DML PARAMS:     wb_enable                  = %d\n", dout->wb_enable);
+		dml_print("DML PARAMS:     num_active_wb              = %d\n", dout->num_active_wb);
+		dml_print("DML PARAMS: =====================================\n");
+
+		dml_print("DML PARAMS: PIPE [%d] CLOCK CONFIG PARAMS:\n", i);
+		dml_print("DML PARAMS:     voltage                    = %d\n", clks_cfg->voltage);
+		dml_print("DML PARAMS:     dppclk_mhz                 = %3.2f\n", clks_cfg->dppclk_mhz);
+		dml_print("DML PARAMS:     refclk_mhz                 = %3.2f\n", clks_cfg->refclk_mhz);
+		dml_print("DML PARAMS:     dispclk_mhz                = %3.2f\n", clks_cfg->dispclk_mhz);
+		dml_print("DML PARAMS:     dcfclk_mhz                 = %3.2f\n", clks_cfg->dcfclk_mhz);
+		dml_print("DML PARAMS:     socclk_mhz                 = %3.2f\n", clks_cfg->socclk_mhz);
+		dml_print("DML PARAMS: =====================================\n");
+	}
+}
+
+void dml_log_mode_support_params(struct display_mode_lib *mode_lib)
+{
+	int i;
+
+	for (i = mode_lib->vba.soc.num_states; i >= 0; i--) {
+		dml_print("DML SUPPORT: ===============================================\n");
+		dml_print("DML SUPPORT: Voltage State %d\n", i);
+		dml_print("DML SUPPORT:     Mode Supported              : %s\n", mode_lib->vba.ModeSupport[i][0] ? "Supported" : "NOT Supported");
+		dml_print("DML SUPPORT:     Mode Supported (pipe split) : %s\n", mode_lib->vba.ModeSupport[i][1] ? "Supported" : "NOT Supported");
+		dml_print("DML SUPPORT:     Scale Ratio And Taps                : %s\n", mode_lib->vba.ScaleRatioAndTapsSupport ? "Supported" : "NOT Supported");
+		dml_print("DML SUPPORT:     Source Format Pixel And Scan        : %s\n", mode_lib->vba.SourceFormatPixelAndScanSupport ? "Supported" : "NOT Supported");
+		dml_print("DML SUPPORT:     Viewport Size                       : [%s, %s]\n", mode_lib->vba.ViewportSizeSupport[i][0] ? "Supported" : "NOT Supported", mode_lib->vba.ViewportSizeSupport[i][1] ? "Supported" : "NOT Supported");
+		dml_print("DML SUPPORT:     DIO Support                         : %s\n", mode_lib->vba.DIOSupport[i] ? "Supported" : "NOT Supported");
+		dml_print("DML SUPPORT:     ODM Combine 4To1 Support Check      : %s\n", mode_lib->vba.ODMCombine4To1SupportCheckOK[i] ? "Supported" : "NOT Supported");
+		dml_print("DML SUPPORT:     DSC Units                           : %s\n", mode_lib->vba.NotEnoughDSCUnits[i] ? "Not Supported" : "Supported");
+		dml_print("DML SUPPORT:     DSCCLK Required                     : %s\n", mode_lib->vba.DSCCLKRequiredMoreThanSupported[i] ? "Not Supported" : "Supported");
+		dml_print("DML SUPPORT:     DTBCLK Required                     : %s\n", mode_lib->vba.DTBCLKRequiredMoreThanSupported[i] ? "Not Supported" : "Supported");
+		dml_print("DML SUPPORT:     Re-ordering Buffer                  : [%s, %s]\n", mode_lib->vba.ROBSupport[i][0] ? "Supported" : "NOT Supported", mode_lib->vba.ROBSupport[i][1] ? "Supported" : "NOT Supported");
+		dml_print("DML SUPPORT:     DISPCLK and DPPCLK                  : [%s, %s]\n", mode_lib->vba.DISPCLK_DPPCLK_Support[i][0] ? "Supported" : "NOT Supported", mode_lib->vba.DISPCLK_DPPCLK_Support[i][1] ? "Supported" : "NOT Supported");
+		dml_print("DML SUPPORT:     Total Available Pipes               : [%s, %s]\n", mode_lib->vba.TotalAvailablePipesSupport[i][0] ? "Supported" : "NOT Supported", mode_lib->vba.TotalAvailablePipesSupport[i][1] ? "Supported" : "NOT Supported");
+		dml_print("DML SUPPORT:     Writeback Latency                   : %s\n", mode_lib->vba.WritebackLatencySupport ? "Supported" : "NOT Supported");
+		dml_print("DML SUPPORT:     Writeback Scale Ratio And Taps      : %s\n", mode_lib->vba.WritebackScaleRatioAndTapsSupport ? "Supported" : "NOT Supported");
+		dml_print("DML SUPPORT:     Cursor                              : %s\n", mode_lib->vba.CursorSupport ? "Supported" : "NOT Supported");
+		dml_print("DML SUPPORT:     Pitch                               : %s\n", mode_lib->vba.PitchSupport ? "Supported" : "NOT Supported");
+		dml_print("DML SUPPORT:     Prefetch                            : [%s, %s]\n", mode_lib->vba.PrefetchSupported[i][0] ? "Supported" : "NOT Supported", mode_lib->vba.PrefetchSupported[i][1] ? "Supported" : "NOT Supported");
+		dml_print("DML SUPPORT:     Dynamic Metadata                    : [%s, %s]\n", mode_lib->vba.DynamicMetadataSupported[i][0] ? "Supported" : "NOT Supported", mode_lib->vba.DynamicMetadataSupported[i][1] ? "Supported" : "NOT Supported");
+		dml_print("DML SUPPORT:     Total Vertical Active Bandwidth     : [%s, %s]\n", mode_lib->vba.TotalVerticalActiveBandwidthSupport[i][0] ? "Supported" : "NOT Supported", mode_lib->vba.TotalVerticalActiveBandwidthSupport[i][1] ? "Supported" : "NOT Supported");
+		dml_print("DML SUPPORT:     VRatio In Prefetch                  : [%s, %s]\n", mode_lib->vba.VRatioInPrefetchSupported[i][0] ? "Supported" : "NOT Supported", mode_lib->vba.VRatioInPrefetchSupported[i][1] ? "Supported" : "NOT Supported");
+		dml_print("DML SUPPORT:     PTE Buffer Size Not Exceeded        : [%s, %s]\n", mode_lib->vba.PTEBufferSizeNotExceeded[i][0] ? "Supported" : "NOT Supported", mode_lib->vba.PTEBufferSizeNotExceeded[i][1] ? "Supported" : "NOT Supported");
+		dml_print("DML SUPPORT:     DSC Input BPC                       : %s\n", mode_lib->vba.NonsupportedDSCInputBPC ? "Not Supported" : "Supported");
+		dml_print("DML SUPPORT:     HostVMEnable                        : %d\n", mode_lib->vba.HostVMEnable);
+		dml_print("DML SUPPORT:     ImmediateFlipSupportedForState      : [%d, %d]\n", mode_lib->vba.ImmediateFlipSupportedForState[i][0], mode_lib->vba.ImmediateFlipSupportedForState[i][1]);
+	}
+}

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.h

@@ -0,0 +1,109 @@
+/*
+ * Copyright 2017 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 __DISPLAY_MODE_LIB_H__
+#define __DISPLAY_MODE_LIB_H__
+
+#include "dm_services.h"
+#include "dc_features.h"
+#include "display_mode_structs.h"
+#include "display_mode_enums.h"
+#include "display_mode_vba.h"
+
+enum dml_project {
+	DML_PROJECT_UNDEFINED,
+	DML_PROJECT_RAVEN1,
+	DML_PROJECT_NAVI10,
+	DML_PROJECT_NAVI10v2,
+	DML_PROJECT_DCN201,
+	DML_PROJECT_DCN21,
+	DML_PROJECT_DCN30,
+	DML_PROJECT_DCN31,
+	DML_PROJECT_DCN315,
+	DML_PROJECT_DCN314,
+	DML_PROJECT_DCN32,
+};
+
+struct display_mode_lib;
+
+struct dml_funcs {
+	void (*rq_dlg_get_dlg_reg)(
+			struct display_mode_lib *mode_lib,
+			display_dlg_regs_st *dlg_regs,
+			display_ttu_regs_st *ttu_regs,
+			const display_e2e_pipe_params_st *e2e_pipe_param,
+			const unsigned int num_pipes,
+			const unsigned int pipe_idx,
+			const bool cstate_en,
+			const bool pstate_en,
+			const bool vm_en,
+			const bool ignore_viewport_pos,
+			const bool immediate_flip_support);
+	void (*rq_dlg_get_rq_reg)(
+		struct display_mode_lib *mode_lib,
+		display_rq_regs_st *rq_regs,
+		const display_pipe_params_st *pipe_param);
+	// DLG interfaces have different function parameters in DCN32.
+	// Create new function pointers to address the changes
+	void (*rq_dlg_get_dlg_reg_v2)(
+			struct display_mode_lib *mode_lib,
+			display_dlg_regs_st *dlg_regs,
+			display_ttu_regs_st *ttu_regs,
+			display_e2e_pipe_params_st *e2e_pipe_param,
+			const unsigned int num_pipes,
+			const unsigned int pipe_idx);
+	void (*rq_dlg_get_rq_reg_v2)(display_rq_regs_st *rq_regs,
+			struct display_mode_lib *mode_lib,
+			const display_e2e_pipe_params_st *e2e_pipe_param,
+			const unsigned int num_pipes,
+			const unsigned int pipe_idx);
+	void (*recalculate)(struct display_mode_lib *mode_lib);
+	void (*validate)(struct display_mode_lib *mode_lib);
+};
+
+struct display_mode_lib {
+	struct _vcs_dpi_ip_params_st ip;
+	struct _vcs_dpi_soc_bounding_box_st soc;
+	enum dml_project project;
+	struct vba_vars_st vba;
+	struct dal_logger *logger;
+	struct dml_funcs funcs;
+	struct _vcs_dpi_display_e2e_pipe_params_st dml_pipe_state[6];
+	bool validate_max_state;
+};
+
+void dml_init_instance(struct display_mode_lib *lib,
+		const struct _vcs_dpi_soc_bounding_box_st *soc_bb,
+		const struct _vcs_dpi_ip_params_st *ip_params,
+		enum dml_project project);
+
+const char *dml_get_status_message(enum dm_validation_status status);
+
+void dml_log_pipe_params(
+		struct display_mode_lib *mode_lib,
+		display_e2e_pipe_params_st *pipes,
+		int pipe_cnt);
+
+void dml_log_mode_support_params(struct display_mode_lib *mode_lib);
+#endif

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h

@@ -0,0 +1,741 @@
+/*
+ * Copyright 2017 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 "dc_features.h"
+#include "display_mode_enums.h"
+
+/**
+ * DOC: overview
+ *
+ * Most of the DML code is automatically generated and tested via hardware
+ * description language. Usually, we use the reference _vcs_dpi in the code
+ * where VCS means "Verilog Compiled Simulator" and DPI stands for "Direct
+ * Programmer Interface". In other words, those structs can be used to
+ * interface with Verilog with other languages such as C.
+ */
+
+#ifndef __DISPLAY_MODE_STRUCTS_H__
+#define __DISPLAY_MODE_STRUCTS_H__
+
+typedef struct _vcs_dpi_voltage_scaling_st voltage_scaling_st;
+typedef struct _vcs_dpi_soc_bounding_box_st soc_bounding_box_st;
+typedef struct _vcs_dpi_ip_params_st ip_params_st;
+typedef struct _vcs_dpi_display_pipe_source_params_st display_pipe_source_params_st;
+typedef struct _vcs_dpi_display_output_params_st display_output_params_st;
+typedef struct _vcs_dpi_scaler_ratio_depth_st scaler_ratio_depth_st;
+typedef struct _vcs_dpi_scaler_taps_st scaler_taps_st;
+typedef struct _vcs_dpi_display_pipe_dest_params_st display_pipe_dest_params_st;
+typedef struct _vcs_dpi_display_pipe_params_st display_pipe_params_st;
+typedef struct _vcs_dpi_display_clocks_and_cfg_st display_clocks_and_cfg_st;
+typedef struct _vcs_dpi_display_e2e_pipe_params_st display_e2e_pipe_params_st;
+typedef struct _vcs_dpi_display_data_rq_misc_params_st display_data_rq_misc_params_st;
+typedef struct _vcs_dpi_display_data_rq_sizing_params_st display_data_rq_sizing_params_st;
+typedef struct _vcs_dpi_display_data_rq_dlg_params_st display_data_rq_dlg_params_st;
+typedef struct _vcs_dpi_display_rq_dlg_params_st display_rq_dlg_params_st;
+typedef struct _vcs_dpi_display_rq_sizing_params_st display_rq_sizing_params_st;
+typedef struct _vcs_dpi_display_rq_misc_params_st display_rq_misc_params_st;
+typedef struct _vcs_dpi_display_rq_params_st display_rq_params_st;
+typedef struct _vcs_dpi_display_dlg_regs_st display_dlg_regs_st;
+typedef struct _vcs_dpi_display_ttu_regs_st display_ttu_regs_st;
+typedef struct _vcs_dpi_display_data_rq_regs_st display_data_rq_regs_st;
+typedef struct _vcs_dpi_display_rq_regs_st display_rq_regs_st;
+typedef struct _vcs_dpi_display_dlg_sys_params_st display_dlg_sys_params_st;
+typedef struct _vcs_dpi_display_arb_params_st display_arb_params_st;
+
+typedef struct {
+	double UrgentWatermark;
+	double WritebackUrgentWatermark;
+	double DRAMClockChangeWatermark;
+	double FCLKChangeWatermark;
+	double WritebackDRAMClockChangeWatermark;
+	double WritebackFCLKChangeWatermark;
+	double StutterExitWatermark;
+	double StutterEnterPlusExitWatermark;
+	double Z8StutterExitWatermark;
+	double Z8StutterEnterPlusExitWatermark;
+	double USRRetrainingWatermark;
+} Watermarks;
+
+typedef struct {
+	double UrgentLatency;
+	double ExtraLatency;
+	double WritebackLatency;
+	double DRAMClockChangeLatency;
+	double FCLKChangeLatency;
+	double SRExitTime;
+	double SREnterPlusExitTime;
+	double SRExitZ8Time;
+	double SREnterPlusExitZ8Time;
+	double USRRetrainingLatencyPlusSMNLatency;
+} Latencies;
+
+typedef struct {
+	double Dppclk;
+	double Dispclk;
+	double PixelClock;
+	double DCFClkDeepSleep;
+	unsigned int DPPPerSurface;
+	bool ScalerEnabled;
+	enum dm_rotation_angle SourceRotation;
+	unsigned int ViewportHeight;
+	unsigned int ViewportHeightChroma;
+	unsigned int BlockWidth256BytesY;
+	unsigned int BlockHeight256BytesY;
+	unsigned int BlockWidth256BytesC;
+	unsigned int BlockHeight256BytesC;
+	unsigned int BlockWidthY;
+	unsigned int BlockHeightY;
+	unsigned int BlockWidthC;
+	unsigned int BlockHeightC;
+	unsigned int InterlaceEnable;
+	unsigned int NumberOfCursors;
+	unsigned int VBlank;
+	unsigned int HTotal;
+	unsigned int HActive;
+	bool DCCEnable;
+	enum odm_combine_mode ODMMode;
+	enum source_format_class SourcePixelFormat;
+	enum dm_swizzle_mode SurfaceTiling;
+	unsigned int BytePerPixelY;
+	unsigned int BytePerPixelC;
+	bool ProgressiveToInterlaceUnitInOPP;
+	double VRatio;
+	double VRatioChroma;
+	unsigned int VTaps;
+	unsigned int VTapsChroma;
+	unsigned int PitchY;
+	unsigned int DCCMetaPitchY;
+	unsigned int PitchC;
+	unsigned int DCCMetaPitchC;
+	bool ViewportStationary;
+	unsigned int ViewportXStart;
+	unsigned int ViewportYStart;
+	unsigned int ViewportXStartC;
+	unsigned int ViewportYStartC;
+	bool FORCE_ONE_ROW_FOR_FRAME;
+	unsigned int SwathHeightY;
+	unsigned int SwathHeightC;
+} DmlPipe;
+
+typedef struct {
+	double UrgentLatency;
+	double ExtraLatency;
+	double WritebackLatency;
+	double DRAMClockChangeLatency;
+	double FCLKChangeLatency;
+	double SRExitTime;
+	double SREnterPlusExitTime;
+	double SRExitZ8Time;
+	double SREnterPlusExitZ8Time;
+	double USRRetrainingLatency;
+	double SMNLatency;
+} SOCParametersList;
+
+struct _vcs_dpi_voltage_scaling_st {
+	int state;
+	double dscclk_mhz;
+	double dcfclk_mhz;
+	double socclk_mhz;
+	double phyclk_d18_mhz;
+	double phyclk_d32_mhz;
+	double dram_speed_mts;
+	double fabricclk_mhz;
+	double dispclk_mhz;
+	double dram_bw_per_chan_gbps;
+	double phyclk_mhz;
+	double dppclk_mhz;
+	double dtbclk_mhz;
+	float net_bw_in_kbytes_sec;
+};
+
+/**
+ * _vcs_dpi_soc_bounding_box_st: SOC definitions
+ *
+ * This struct maintains the SOC Bounding Box information for the ASIC; it
+ * defines things such as clock, voltage, performance, etc. Usually, we load
+ * these values from VBIOS; if something goes wrong, we use some hard-coded
+ * values, which will enable the ASIC to light up with limitations.
+ */
+struct _vcs_dpi_soc_bounding_box_st {
+	struct _vcs_dpi_voltage_scaling_st clock_limits[DC__VOLTAGE_STATES];
+	/**
+	 * @num_states: It represents the total of Display Power Management
+	 * (DPM) supported by the specific ASIC.
+	 */
+	unsigned int num_states;
+	double sr_exit_time_us;
+	double sr_enter_plus_exit_time_us;
+	double sr_exit_z8_time_us;
+	double sr_enter_plus_exit_z8_time_us;
+	double urgent_latency_us;
+	double urgent_latency_pixel_data_only_us;
+	double urgent_latency_pixel_mixed_with_vm_data_us;
+	double urgent_latency_vm_data_only_us;
+	double usr_retraining_latency_us;
+	double smn_latency_us;
+	double fclk_change_latency_us;
+	double mall_allocated_for_dcn_mbytes;
+	double pct_ideal_fabric_bw_after_urgent;
+	double pct_ideal_dram_bw_after_urgent_strobe;
+	double max_avg_fabric_bw_use_normal_percent;
+	double max_avg_dram_bw_use_normal_strobe_percent;
+	enum dm_prefetch_modes allow_for_pstate_or_stutter_in_vblank_final;
+	bool dram_clock_change_requirement_final;
+	double writeback_latency_us;
+	double ideal_dram_bw_after_urgent_percent;
+	double pct_ideal_dram_sdp_bw_after_urgent_pixel_only; // PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyPixelDataOnly
+	double pct_ideal_dram_sdp_bw_after_urgent_pixel_and_vm;
+	double pct_ideal_dram_sdp_bw_after_urgent_vm_only;
+	double pct_ideal_sdp_bw_after_urgent;
+	double max_avg_sdp_bw_use_normal_percent;
+	double max_avg_dram_bw_use_normal_percent;
+	unsigned int max_request_size_bytes;
+	double downspread_percent;
+	double dram_page_open_time_ns;
+	double dram_rw_turnaround_time_ns;
+	double dram_return_buffer_per_channel_bytes;
+	double dram_channel_width_bytes;
+	double fabric_datapath_to_dcn_data_return_bytes;
+	double dcn_downspread_percent;
+	double dispclk_dppclk_vco_speed_mhz;
+	double dfs_vco_period_ps;
+	unsigned int urgent_out_of_order_return_per_channel_pixel_only_bytes;
+	unsigned int urgent_out_of_order_return_per_channel_pixel_and_vm_bytes;
+	unsigned int urgent_out_of_order_return_per_channel_vm_only_bytes;
+	unsigned int round_trip_ping_latency_dcfclk_cycles;
+	unsigned int urgent_out_of_order_return_per_channel_bytes;
+	unsigned int channel_interleave_bytes;
+	unsigned int num_banks;
+	unsigned int num_chans;
+	unsigned int vmm_page_size_bytes;
+	unsigned int hostvm_min_page_size_bytes;
+	unsigned int gpuvm_min_page_size_bytes;
+	double dram_clock_change_latency_us;
+	double dummy_pstate_latency_us;
+	double writeback_dram_clock_change_latency_us;
+	unsigned int return_bus_width_bytes;
+	unsigned int voltage_override;
+	double xfc_bus_transport_time_us;
+	double xfc_xbuf_latency_tolerance_us;
+	int use_urgent_burst_bw;
+	double min_dcfclk;
+	bool do_urgent_latency_adjustment;
+	double urgent_latency_adjustment_fabric_clock_component_us;
+	double urgent_latency_adjustment_fabric_clock_reference_mhz;
+	bool disable_dram_clock_change_vactive_support;
+	bool allow_dram_clock_one_display_vactive;
+	enum self_refresh_affinity allow_dram_self_refresh_or_dram_clock_change_in_vblank;
+	double max_vratio_pre;
+};
+
+/**
+ * @_vcs_dpi_ip_params_st: IP configuraion for DCN blocks
+ *
+ * In this struct you can find the DCN configuration associated to the specific
+ * ASIC. For example, here we can save how many DPPs the ASIC is using and it
+ * is available.
+ *
+ */
+struct _vcs_dpi_ip_params_st {
+	bool use_min_dcfclk;
+	bool clamp_min_dcfclk;
+	bool gpuvm_enable;
+	bool hostvm_enable;
+	bool dsc422_native_support;
+	unsigned int gpuvm_max_page_table_levels;
+	unsigned int hostvm_max_page_table_levels;
+	unsigned int hostvm_cached_page_table_levels;
+	unsigned int pte_group_size_bytes;
+	unsigned int max_inter_dcn_tile_repeaters;
+	unsigned int num_dsc;
+	unsigned int odm_capable;
+	unsigned int rob_buffer_size_kbytes;
+	unsigned int det_buffer_size_kbytes;
+	unsigned int min_comp_buffer_size_kbytes;
+	unsigned int dpte_buffer_size_in_pte_reqs_luma;
+	unsigned int dpte_buffer_size_in_pte_reqs_chroma;
+	unsigned int pde_proc_buffer_size_64k_reqs;
+	unsigned int dpp_output_buffer_pixels;
+	unsigned int opp_output_buffer_lines;
+	unsigned int pixel_chunk_size_kbytes;
+	unsigned int alpha_pixel_chunk_size_kbytes;
+	unsigned int min_pixel_chunk_size_bytes;
+	unsigned int dcc_meta_buffer_size_bytes;
+	unsigned char pte_enable;
+	unsigned int pte_chunk_size_kbytes;
+	unsigned int meta_chunk_size_kbytes;
+	unsigned int min_meta_chunk_size_bytes;
+	unsigned int writeback_chunk_size_kbytes;
+	unsigned int line_buffer_size_bits;
+	unsigned int max_line_buffer_lines;
+	unsigned int writeback_luma_buffer_size_kbytes;
+	unsigned int writeback_chroma_buffer_size_kbytes;
+	unsigned int writeback_chroma_line_buffer_width_pixels;
+
+	unsigned int writeback_interface_buffer_size_kbytes;
+	unsigned int writeback_line_buffer_buffer_size;
+
+	unsigned int writeback_10bpc420_supported;
+	double writeback_max_hscl_ratio;
+	double writeback_max_vscl_ratio;
+	double writeback_min_hscl_ratio;
+	double writeback_min_vscl_ratio;
+	unsigned int maximum_dsc_bits_per_component;
+	unsigned int maximum_pixels_per_line_per_dsc_unit;
+	unsigned int writeback_max_hscl_taps;
+	unsigned int writeback_max_vscl_taps;
+	unsigned int writeback_line_buffer_luma_buffer_size;
+	unsigned int writeback_line_buffer_chroma_buffer_size;
+
+	unsigned int max_page_table_levels;
+	/**
+	 * @max_num_dpp: Maximum number of DPP supported in the target ASIC.
+	 */
+	unsigned int max_num_dpp;
+	unsigned int max_num_otg;
+	unsigned int cursor_chunk_size;
+	unsigned int cursor_buffer_size;
+	unsigned int max_num_wb;
+	unsigned int max_dchub_pscl_bw_pix_per_clk;
+	unsigned int max_pscl_lb_bw_pix_per_clk;
+	unsigned int max_lb_vscl_bw_pix_per_clk;
+	unsigned int max_vscl_hscl_bw_pix_per_clk;
+	double max_hscl_ratio;
+	double max_vscl_ratio;
+	unsigned int hscl_mults;
+	unsigned int vscl_mults;
+	unsigned int max_hscl_taps;
+	unsigned int max_vscl_taps;
+	unsigned int xfc_supported;
+	unsigned int ptoi_supported;
+	unsigned int gfx7_compat_tiling_supported;
+
+	bool odm_combine_4to1_supported;
+	bool dynamic_metadata_vm_enabled;
+	unsigned int max_num_hdmi_frl_outputs;
+
+	unsigned int xfc_fill_constant_bytes;
+	double dispclk_ramp_margin_percent;
+	double xfc_fill_bw_overhead_percent;
+	double underscan_factor;
+	unsigned int min_vblank_lines;
+	unsigned int dppclk_delay_subtotal;
+	unsigned int dispclk_delay_subtotal;
+	double dcfclk_cstate_latency;
+	unsigned int dppclk_delay_scl;
+	unsigned int dppclk_delay_scl_lb_only;
+	unsigned int dppclk_delay_cnvc_formatter;
+	unsigned int dppclk_delay_cnvc_cursor;
+	unsigned int is_line_buffer_bpp_fixed;
+	unsigned int line_buffer_fixed_bpp;
+	unsigned int dcc_supported;
+	unsigned int config_return_buffer_size_in_kbytes;
+	unsigned int compressed_buffer_segment_size_in_kbytes;
+	unsigned int meta_fifo_size_in_kentries;
+	unsigned int zero_size_buffer_entries;
+	unsigned int compbuf_reserved_space_64b;
+	unsigned int compbuf_reserved_space_zs;
+
+	unsigned int IsLineBufferBppFixed;
+	unsigned int LineBufferFixedBpp;
+	unsigned int can_vstartup_lines_exceed_vsync_plus_back_porch_lines_minus_one;
+	unsigned int bug_forcing_LC_req_same_size_fixed;
+	unsigned int number_of_cursors;
+	unsigned int max_num_dp2p0_outputs;
+	unsigned int max_num_dp2p0_streams;
+	unsigned int VBlankNomDefaultUS;
+
+	/* DM workarounds */
+	double dsc_delay_factor_wa; // TODO: Remove after implementing root cause fix
+	double min_prefetch_in_strobe_us;
+};
+
+struct _vcs_dpi_display_xfc_params_st {
+	double xfc_tslv_vready_offset_us;
+	double xfc_tslv_vupdate_width_us;
+	double xfc_tslv_vupdate_offset_us;
+	int xfc_slv_chunk_size_bytes;
+};
+
+struct _vcs_dpi_display_pipe_source_params_st {
+	int source_format;
+	double dcc_fraction_of_zs_req_luma;
+	double dcc_fraction_of_zs_req_chroma;
+	unsigned char dcc;
+	unsigned int dcc_rate;
+	unsigned int dcc_rate_chroma;
+	unsigned char dcc_use_global;
+	unsigned char vm;
+	bool unbounded_req_mode;
+	bool gpuvm;    // gpuvm enabled
+	bool hostvm;    // hostvm enabled
+	bool gpuvm_levels_force_en;
+	unsigned int gpuvm_levels_force;
+	bool hostvm_levels_force_en;
+	unsigned int hostvm_levels_force;
+	int source_scan;
+	int source_rotation; // new in dml32
+	unsigned int det_size_override; // use to populate DETSizeOverride in vba struct
+	int sw_mode;
+	int macro_tile_size;
+	unsigned int surface_width_y;
+	unsigned int surface_height_y;
+	unsigned int surface_width_c;
+	unsigned int surface_height_c;
+	unsigned int viewport_width;
+	unsigned int viewport_height;
+	unsigned int viewport_y_y;
+	unsigned int viewport_y_c;
+	unsigned int viewport_width_c;
+	unsigned int viewport_height_c;
+	unsigned int viewport_width_max;
+	unsigned int viewport_height_max;
+	unsigned int viewport_x_y;
+	unsigned int viewport_x_c;
+	bool viewport_stationary;
+	unsigned int dcc_rate_luma;
+	unsigned int gpuvm_min_page_size_kbytes;
+	unsigned int use_mall_for_pstate_change;
+	unsigned int use_mall_for_static_screen;
+	bool force_one_row_for_frame;
+	bool pte_buffer_mode;
+	unsigned int data_pitch;
+	unsigned int data_pitch_c;
+	unsigned int meta_pitch;
+	unsigned int meta_pitch_c;
+	unsigned int cur0_src_width;
+	int cur0_bpp;
+	unsigned int cur1_src_width;
+	int cur1_bpp;
+	int num_cursors;
+	unsigned char is_hsplit;
+	unsigned char dynamic_metadata_enable;
+	unsigned int dynamic_metadata_lines_before_active;
+	unsigned int dynamic_metadata_xmit_bytes;
+	unsigned int hsplit_grp;
+	unsigned char xfc_enable;
+	unsigned char xfc_slave;
+	unsigned char immediate_flip;
+	struct _vcs_dpi_display_xfc_params_st xfc_params;
+	//for vstartuplines calculation freesync
+	unsigned char v_total_min;
+	unsigned char v_total_max;
+};
+struct writeback_st {
+	int wb_src_height;
+	int wb_src_width;
+	int wb_dst_width;
+	int wb_dst_height;
+	int wb_pixel_format;
+	int wb_htaps_luma;
+	int wb_vtaps_luma;
+	int wb_htaps_chroma;
+	int wb_vtaps_chroma;
+	unsigned int wb_htaps;
+	unsigned int wb_vtaps;
+	double wb_hratio;
+	double wb_vratio;
+};
+
+struct display_audio_params_st {
+	unsigned int   audio_sample_rate_khz;
+	int    		   audio_sample_layout;
+};
+
+struct _vcs_dpi_display_output_params_st {
+	int dp_lanes;
+	double output_bpp;
+	unsigned int dsc_input_bpc;
+	int dsc_enable;
+	int wb_enable;
+	int num_active_wb;
+	int output_type;
+	int is_virtual;
+	int output_format;
+	int dsc_slices;
+	int max_audio_sample_rate;
+	struct writeback_st wb;
+	struct display_audio_params_st audio;
+	unsigned int output_bpc;
+	int dp_rate;
+	unsigned int dp_multistream_id;
+	bool dp_multistream_en;
+};
+
+struct _vcs_dpi_scaler_ratio_depth_st {
+	double hscl_ratio;
+	double vscl_ratio;
+	double hscl_ratio_c;
+	double vscl_ratio_c;
+	double vinit;
+	double vinit_c;
+	double vinit_bot;
+	double vinit_bot_c;
+	int lb_depth;
+	int scl_enable;
+};
+
+struct _vcs_dpi_scaler_taps_st {
+	unsigned int htaps;
+	unsigned int vtaps;
+	unsigned int htaps_c;
+	unsigned int vtaps_c;
+};
+
+struct _vcs_dpi_display_pipe_dest_params_st {
+	unsigned int recout_width;
+	unsigned int recout_height;
+	unsigned int full_recout_width;
+	unsigned int full_recout_height;
+	unsigned int hblank_start;
+	unsigned int hblank_end;
+	unsigned int vblank_start;
+	unsigned int vblank_end;
+	unsigned int htotal;
+	unsigned int vtotal;
+	unsigned int vfront_porch;
+	unsigned int vblank_nom;
+	unsigned int vactive;
+	unsigned int hactive;
+	unsigned int vstartup_start;
+	unsigned int vupdate_offset;
+	unsigned int vupdate_width;
+	unsigned int vready_offset;
+	unsigned int pstate_keepout;
+	unsigned char interlaced;
+	double pixel_rate_mhz;
+	unsigned char synchronized_vblank_all_planes;
+	unsigned char otg_inst;
+	unsigned int odm_combine;
+	unsigned char use_maximum_vstartup;
+	unsigned int vtotal_max;
+	unsigned int vtotal_min;
+	unsigned int refresh_rate;
+	bool synchronize_timings;
+	unsigned int odm_combine_policy;
+	bool drr_display;
+};
+
+struct _vcs_dpi_display_pipe_params_st {
+	display_pipe_source_params_st src;
+	display_pipe_dest_params_st dest;
+	scaler_ratio_depth_st scale_ratio_depth;
+	scaler_taps_st scale_taps;
+};
+
+struct _vcs_dpi_display_clocks_and_cfg_st {
+	int voltage;
+	double dppclk_mhz;
+	double refclk_mhz;
+	double dispclk_mhz;
+	double dcfclk_mhz;
+	double socclk_mhz;
+};
+
+struct _vcs_dpi_display_e2e_pipe_params_st {
+	display_pipe_params_st pipe;
+	display_output_params_st dout;
+	display_clocks_and_cfg_st clks_cfg;
+};
+
+struct _vcs_dpi_display_data_rq_misc_params_st {
+	unsigned int full_swath_bytes;
+	unsigned int stored_swath_bytes;
+	unsigned int blk256_height;
+	unsigned int blk256_width;
+	unsigned int req_height;
+	unsigned int req_width;
+};
+
+struct _vcs_dpi_display_data_rq_sizing_params_st {
+	unsigned int chunk_bytes;
+	unsigned int min_chunk_bytes;
+	unsigned int meta_chunk_bytes;
+	unsigned int min_meta_chunk_bytes;
+	unsigned int mpte_group_bytes;
+	unsigned int dpte_group_bytes;
+};
+
+struct _vcs_dpi_display_data_rq_dlg_params_st {
+	unsigned int swath_width_ub;
+	unsigned int swath_height;
+	unsigned int req_per_swath_ub;
+	unsigned int meta_pte_bytes_per_frame_ub;
+	unsigned int dpte_req_per_row_ub;
+	unsigned int dpte_groups_per_row_ub;
+	unsigned int dpte_row_height;
+	unsigned int dpte_bytes_per_row_ub;
+	unsigned int meta_chunks_per_row_ub;
+	unsigned int meta_req_per_row_ub;
+	unsigned int meta_row_height;
+	unsigned int meta_bytes_per_row_ub;
+};
+
+struct _vcs_dpi_display_rq_dlg_params_st {
+	display_data_rq_dlg_params_st rq_l;
+	display_data_rq_dlg_params_st rq_c;
+};
+
+struct _vcs_dpi_display_rq_sizing_params_st {
+	display_data_rq_sizing_params_st rq_l;
+	display_data_rq_sizing_params_st rq_c;
+};
+
+struct _vcs_dpi_display_rq_misc_params_st {
+	display_data_rq_misc_params_st rq_l;
+	display_data_rq_misc_params_st rq_c;
+};
+
+struct _vcs_dpi_display_rq_params_st {
+	unsigned char yuv420;
+	unsigned char yuv420_10bpc;
+	unsigned char rgbe_alpha;
+	display_rq_misc_params_st misc;
+	display_rq_sizing_params_st sizing;
+	display_rq_dlg_params_st dlg;
+};
+
+struct _vcs_dpi_display_dlg_regs_st {
+	unsigned int refcyc_h_blank_end;
+	unsigned int dlg_vblank_end;
+	unsigned int min_dst_y_next_start;
+	unsigned int min_dst_y_next_start_us;
+	unsigned int refcyc_per_htotal;
+	unsigned int refcyc_x_after_scaler;
+	unsigned int dst_y_after_scaler;
+	unsigned int dst_y_prefetch;
+	unsigned int dst_y_per_vm_vblank;
+	unsigned int dst_y_per_row_vblank;
+	unsigned int dst_y_per_vm_flip;
+	unsigned int dst_y_per_row_flip;
+	unsigned int ref_freq_to_pix_freq;
+	unsigned int vratio_prefetch;
+	unsigned int vratio_prefetch_c;
+	unsigned int refcyc_per_tdlut_group;
+	unsigned int refcyc_per_pte_group_vblank_l;
+	unsigned int refcyc_per_pte_group_vblank_c;
+	unsigned int refcyc_per_meta_chunk_vblank_l;
+	unsigned int refcyc_per_meta_chunk_vblank_c;
+	unsigned int refcyc_per_pte_group_flip_l;
+	unsigned int refcyc_per_pte_group_flip_c;
+	unsigned int refcyc_per_meta_chunk_flip_l;
+	unsigned int refcyc_per_meta_chunk_flip_c;
+	unsigned int dst_y_per_pte_row_nom_l;
+	unsigned int dst_y_per_pte_row_nom_c;
+	unsigned int refcyc_per_pte_group_nom_l;
+	unsigned int refcyc_per_pte_group_nom_c;
+	unsigned int dst_y_per_meta_row_nom_l;
+	unsigned int dst_y_per_meta_row_nom_c;
+	unsigned int refcyc_per_meta_chunk_nom_l;
+	unsigned int refcyc_per_meta_chunk_nom_c;
+	unsigned int refcyc_per_line_delivery_pre_l;
+	unsigned int refcyc_per_line_delivery_pre_c;
+	unsigned int refcyc_per_line_delivery_l;
+	unsigned int refcyc_per_line_delivery_c;
+	unsigned int chunk_hdl_adjust_cur0;
+	unsigned int chunk_hdl_adjust_cur1;
+	unsigned int vready_after_vcount0;
+	unsigned int dst_y_offset_cur0;
+	unsigned int dst_y_offset_cur1;
+	unsigned int xfc_reg_transfer_delay;
+	unsigned int xfc_reg_precharge_delay;
+	unsigned int xfc_reg_remote_surface_flip_latency;
+	unsigned int xfc_reg_prefetch_margin;
+	unsigned int dst_y_delta_drq_limit;
+	unsigned int refcyc_per_vm_group_vblank;
+	unsigned int refcyc_per_vm_group_flip;
+	unsigned int refcyc_per_vm_req_vblank;
+	unsigned int refcyc_per_vm_req_flip;
+	unsigned int refcyc_per_vm_dmdata;
+	unsigned int dmdata_dl_delta;
+};
+
+struct _vcs_dpi_display_ttu_regs_st {
+	unsigned int qos_level_low_wm;
+	unsigned int qos_level_high_wm;
+	unsigned int min_ttu_vblank;
+	unsigned int qos_level_flip;
+	unsigned int refcyc_per_req_delivery_l;
+	unsigned int refcyc_per_req_delivery_c;
+	unsigned int refcyc_per_req_delivery_cur0;
+	unsigned int refcyc_per_req_delivery_cur1;
+	unsigned int refcyc_per_req_delivery_pre_l;
+	unsigned int refcyc_per_req_delivery_pre_c;
+	unsigned int refcyc_per_req_delivery_pre_cur0;
+	unsigned int refcyc_per_req_delivery_pre_cur1;
+	unsigned int qos_level_fixed_l;
+	unsigned int qos_level_fixed_c;
+	unsigned int qos_level_fixed_cur0;
+	unsigned int qos_level_fixed_cur1;
+	unsigned int qos_ramp_disable_l;
+	unsigned int qos_ramp_disable_c;
+	unsigned int qos_ramp_disable_cur0;
+	unsigned int qos_ramp_disable_cur1;
+};
+
+struct _vcs_dpi_display_data_rq_regs_st {
+	unsigned int chunk_size;
+	unsigned int min_chunk_size;
+	unsigned int meta_chunk_size;
+	unsigned int min_meta_chunk_size;
+	unsigned int dpte_group_size;
+	unsigned int mpte_group_size;
+	unsigned int swath_height;
+	unsigned int pte_row_height_linear;
+};
+
+struct _vcs_dpi_display_rq_regs_st {
+	display_data_rq_regs_st rq_regs_l;
+	display_data_rq_regs_st rq_regs_c;
+	unsigned int drq_expansion_mode;
+	unsigned int prq_expansion_mode;
+	unsigned int mrq_expansion_mode;
+	unsigned int crq_expansion_mode;
+	unsigned int plane1_base_address;
+	unsigned int aperture_low_addr;   // bits [47:18]
+	unsigned int aperture_high_addr;  // bits [47:18]
+};
+
+struct _vcs_dpi_display_dlg_sys_params_st {
+	double t_mclk_wm_us;
+	double t_urg_wm_us;
+	double t_sr_wm_us;
+	double t_extra_us;
+	double mem_trip_us;
+	double deepsleep_dcfclk_mhz;
+	double total_flip_bw;
+	unsigned int total_flip_bytes;
+};
+
+struct _vcs_dpi_display_arb_params_st {
+	int max_req_outstanding;
+	int min_req_outstanding;
+	int sat_level_us;
+	int hvm_min_req_outstand_commit_threshold;
+	int hvm_max_qos_commit_threshold;
+	int compbuf_reserved_space_kbytes;
+};
+
+#endif /*__DISPLAY_MODE_STRUCTS_H__*/

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/display_rq_dlg_helpers.c

@@ -0,0 +1,396 @@
+/*
+ * Copyright 2017 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 "display_rq_dlg_helpers.h"
+#include "dml_logger.h"
+
+void print__rq_params_st(struct display_mode_lib *mode_lib, const struct _vcs_dpi_display_rq_params_st *rq_param)
+{
+	dml_print("DML_RQ_DLG_CALC: ***************************\n");
+	dml_print("DML_RQ_DLG_CALC: DISPLAY_RQ_PARAM_ST\n");
+	dml_print("DML_RQ_DLG_CALC:  <LUMA>\n");
+	print__data_rq_sizing_params_st(mode_lib, &rq_param->sizing.rq_l);
+	dml_print("DML_RQ_DLG_CALC:  <CHROMA> ===\n");
+	print__data_rq_sizing_params_st(mode_lib, &rq_param->sizing.rq_c);
+
+	dml_print("DML_RQ_DLG_CALC: <LUMA>\n");
+	print__data_rq_dlg_params_st(mode_lib, &rq_param->dlg.rq_l);
+	dml_print("DML_RQ_DLG_CALC: <CHROMA>\n");
+	print__data_rq_dlg_params_st(mode_lib, &rq_param->dlg.rq_c);
+
+	dml_print("DML_RQ_DLG_CALC: <LUMA>\n");
+	print__data_rq_misc_params_st(mode_lib, &rq_param->misc.rq_l);
+	dml_print("DML_RQ_DLG_CALC: <CHROMA>\n");
+	print__data_rq_misc_params_st(mode_lib, &rq_param->misc.rq_c);
+	dml_print("DML_RQ_DLG_CALC: ***************************\n");
+}
+
+void print__data_rq_sizing_params_st(struct display_mode_lib *mode_lib, const struct _vcs_dpi_display_data_rq_sizing_params_st *rq_sizing)
+{
+	(void)mode_lib;
+	(void)rq_sizing;
+	dml_print("DML_RQ_DLG_CALC: =====================================\n");
+	dml_print("DML_RQ_DLG_CALC: DISPLAY_DATA_RQ_SIZING_PARAM_ST\n");
+	dml_print("DML_RQ_DLG_CALC:    chunk_bytes           = %0d\n", rq_sizing->chunk_bytes);
+	dml_print("DML_RQ_DLG_CALC:    min_chunk_bytes       = %0d\n", rq_sizing->min_chunk_bytes);
+	dml_print("DML_RQ_DLG_CALC:    meta_chunk_bytes      = %0d\n", rq_sizing->meta_chunk_bytes);
+	dml_print(
+			"DML_RQ_DLG_CALC:    min_meta_chunk_bytes  = %0d\n",
+			rq_sizing->min_meta_chunk_bytes);
+	dml_print("DML_RQ_DLG_CALC:    mpte_group_bytes      = %0d\n", rq_sizing->mpte_group_bytes);
+	dml_print("DML_RQ_DLG_CALC:    dpte_group_bytes      = %0d\n", rq_sizing->dpte_group_bytes);
+	dml_print("DML_RQ_DLG_CALC: =====================================\n");
+}
+
+void print__data_rq_dlg_params_st(struct display_mode_lib *mode_lib, const struct _vcs_dpi_display_data_rq_dlg_params_st *rq_dlg_param)
+{
+	(void)mode_lib;
+	(void)rq_dlg_param;
+	dml_print("DML_RQ_DLG_CALC: =====================================\n");
+	dml_print("DML_RQ_DLG_CALC: DISPLAY_DATA_RQ_DLG_PARAM_ST\n");
+	dml_print(
+			"DML_RQ_DLG_CALC:    swath_width_ub              = %0d\n",
+			rq_dlg_param->swath_width_ub);
+	dml_print(
+			"DML_RQ_DLG_CALC:    swath_height                = %0d\n",
+			rq_dlg_param->swath_height);
+	dml_print(
+			"DML_RQ_DLG_CALC:    req_per_swath_ub            = %0d\n",
+			rq_dlg_param->req_per_swath_ub);
+	dml_print(
+			"DML_RQ_DLG_CALC:    meta_pte_bytes_per_frame_ub = %0d\n",
+			rq_dlg_param->meta_pte_bytes_per_frame_ub);
+	dml_print(
+			"DML_RQ_DLG_CALC:    dpte_req_per_row_ub         = %0d\n",
+			rq_dlg_param->dpte_req_per_row_ub);
+	dml_print(
+			"DML_RQ_DLG_CALC:    dpte_groups_per_row_ub      = %0d\n",
+			rq_dlg_param->dpte_groups_per_row_ub);
+	dml_print(
+			"DML_RQ_DLG_CALC:    dpte_row_height             = %0d\n",
+			rq_dlg_param->dpte_row_height);
+	dml_print(
+			"DML_RQ_DLG_CALC:    dpte_bytes_per_row_ub       = %0d\n",
+			rq_dlg_param->dpte_bytes_per_row_ub);
+	dml_print(
+			"DML_RQ_DLG_CALC:    meta_chunks_per_row_ub      = %0d\n",
+			rq_dlg_param->meta_chunks_per_row_ub);
+	dml_print(
+			"DML_RQ_DLG_CALC:    meta_req_per_row_ub         = %0d\n",
+			rq_dlg_param->meta_req_per_row_ub);
+	dml_print(
+			"DML_RQ_DLG_CALC:    meta_row_height             = %0d\n",
+			rq_dlg_param->meta_row_height);
+	dml_print(
+			"DML_RQ_DLG_CALC:    meta_bytes_per_row_ub       = %0d\n",
+			rq_dlg_param->meta_bytes_per_row_ub);
+	dml_print("DML_RQ_DLG_CALC: =====================================\n");
+}
+
+void print__data_rq_misc_params_st(struct display_mode_lib *mode_lib, const struct _vcs_dpi_display_data_rq_misc_params_st *rq_misc_param)
+{
+	(void)mode_lib;
+	(void)rq_misc_param;
+	dml_print("DML_RQ_DLG_CALC: =====================================\n");
+	dml_print("DML_RQ_DLG_CALC: DISPLAY_DATA_RQ_MISC_PARAM_ST\n");
+	dml_print(
+			"DML_RQ_DLG_CALC:     full_swath_bytes   = %0d\n",
+			rq_misc_param->full_swath_bytes);
+	dml_print(
+			"DML_RQ_DLG_CALC:     stored_swath_bytes = %0d\n",
+			rq_misc_param->stored_swath_bytes);
+	dml_print("DML_RQ_DLG_CALC:     blk256_width       = %0d\n", rq_misc_param->blk256_width);
+	dml_print("DML_RQ_DLG_CALC:     blk256_height      = %0d\n", rq_misc_param->blk256_height);
+	dml_print("DML_RQ_DLG_CALC:     req_width          = %0d\n", rq_misc_param->req_width);
+	dml_print("DML_RQ_DLG_CALC:     req_height         = %0d\n", rq_misc_param->req_height);
+	dml_print("DML_RQ_DLG_CALC: =====================================\n");
+}
+
+void print__dlg_sys_params_st(struct display_mode_lib *mode_lib, const struct _vcs_dpi_display_dlg_sys_params_st *dlg_sys_param)
+{
+	(void)dlg_sys_param;
+	(void)mode_lib;
+	dml_print("DML_RQ_DLG_CALC: =====================================\n");
+	dml_print("DML_RQ_DLG_CALC: DISPLAY_RQ_DLG_PARAM_ST\n");
+	dml_print("DML_RQ_DLG_CALC:    t_mclk_wm_us         = %3.2f\n", dlg_sys_param->t_mclk_wm_us);
+	dml_print("DML_RQ_DLG_CALC:    t_urg_wm_us          = %3.2f\n", dlg_sys_param->t_urg_wm_us);
+	dml_print("DML_RQ_DLG_CALC:    t_sr_wm_us           = %3.2f\n", dlg_sys_param->t_sr_wm_us);
+	dml_print("DML_RQ_DLG_CALC:    t_extra_us           = %3.2f\n", dlg_sys_param->t_extra_us);
+	dml_print(
+			"DML_RQ_DLG_CALC:    deepsleep_dcfclk_mhz = %3.2f\n",
+			dlg_sys_param->deepsleep_dcfclk_mhz);
+	dml_print(
+			"DML_RQ_DLG_CALC:    total_flip_bw        = %3.2f\n",
+			dlg_sys_param->total_flip_bw);
+	dml_print(
+			"DML_RQ_DLG_CALC:    total_flip_bytes     = %i\n",
+			dlg_sys_param->total_flip_bytes);
+	dml_print("DML_RQ_DLG_CALC: =====================================\n");
+}
+
+void print__data_rq_regs_st(struct display_mode_lib *mode_lib, const struct _vcs_dpi_display_data_rq_regs_st *rq_regs)
+{
+	(void)mode_lib;
+	(void)rq_regs;
+	dml_print("DML_RQ_DLG_CALC: =====================================\n");
+	dml_print("DML_RQ_DLG_CALC: DISPLAY_DATA_RQ_REGS_ST\n");
+	dml_print("DML_RQ_DLG_CALC:    chunk_size              = 0x%0x\n", rq_regs->chunk_size);
+	dml_print("DML_RQ_DLG_CALC:    min_chunk_size          = 0x%0x\n", rq_regs->min_chunk_size);
+	dml_print("DML_RQ_DLG_CALC:    meta_chunk_size         = 0x%0x\n", rq_regs->meta_chunk_size);
+	dml_print(
+			"DML_RQ_DLG_CALC:    min_meta_chunk_size     = 0x%0x\n",
+			rq_regs->min_meta_chunk_size);
+	dml_print("DML_RQ_DLG_CALC:    dpte_group_size         = 0x%0x\n", rq_regs->dpte_group_size);
+	dml_print("DML_RQ_DLG_CALC:    mpte_group_size         = 0x%0x\n", rq_regs->mpte_group_size);
+	dml_print("DML_RQ_DLG_CALC:    swath_height            = 0x%0x\n", rq_regs->swath_height);
+	dml_print(
+			"DML_RQ_DLG_CALC:    pte_row_height_linear   = 0x%0x\n",
+			rq_regs->pte_row_height_linear);
+	dml_print("DML_RQ_DLG_CALC: =====================================\n");
+}
+
+void print__rq_regs_st(struct display_mode_lib *mode_lib, const struct _vcs_dpi_display_rq_regs_st *rq_regs)
+{
+	dml_print("DML_RQ_DLG_CALC: =====================================\n");
+	dml_print("DML_RQ_DLG_CALC: DISPLAY_RQ_REGS_ST\n");
+	dml_print("DML_RQ_DLG_CALC:  <LUMA>\n");
+	print__data_rq_regs_st(mode_lib, &rq_regs->rq_regs_l);
+	dml_print("DML_RQ_DLG_CALC:  <CHROMA>\n");
+	print__data_rq_regs_st(mode_lib, &rq_regs->rq_regs_c);
+	dml_print("DML_RQ_DLG_CALC:    drq_expansion_mode  = 0x%0x\n", rq_regs->drq_expansion_mode);
+	dml_print("DML_RQ_DLG_CALC:    prq_expansion_mode  = 0x%0x\n", rq_regs->prq_expansion_mode);
+	dml_print("DML_RQ_DLG_CALC:    mrq_expansion_mode  = 0x%0x\n", rq_regs->mrq_expansion_mode);
+	dml_print("DML_RQ_DLG_CALC:    crq_expansion_mode  = 0x%0x\n", rq_regs->crq_expansion_mode);
+	dml_print("DML_RQ_DLG_CALC:    plane1_base_address = 0x%0x\n", rq_regs->plane1_base_address);
+	dml_print("DML_RQ_DLG_CALC: =====================================\n");
+}
+
+void print__dlg_regs_st(struct display_mode_lib *mode_lib, const struct _vcs_dpi_display_dlg_regs_st *dlg_regs)
+{
+	(void)dlg_regs;
+	(void)mode_lib;
+	dml_print("DML_RQ_DLG_CALC: =====================================\n");
+	dml_print("DML_RQ_DLG_CALC: DISPLAY_DLG_REGS_ST\n");
+	dml_print(
+			"DML_RQ_DLG_CALC:    refcyc_h_blank_end              = 0x%0x\n",
+			dlg_regs->refcyc_h_blank_end);
+	dml_print(
+			"DML_RQ_DLG_CALC:    dlg_vblank_end                  = 0x%0x\n",
+			dlg_regs->dlg_vblank_end);
+	dml_print(
+			"DML_RQ_DLG_CALC:    min_dst_y_next_start            = 0x%0x\n",
+			dlg_regs->min_dst_y_next_start);
+	dml_print(
+			"DML_RQ_DLG_CALC:    refcyc_per_htotal               = 0x%0x\n",
+			dlg_regs->refcyc_per_htotal);
+	dml_print(
+			"DML_RQ_DLG_CALC:    refcyc_x_after_scaler           = 0x%0x\n",
+			dlg_regs->refcyc_x_after_scaler);
+	dml_print(
+			"DML_RQ_DLG_CALC:    dst_y_after_scaler              = 0x%0x\n",
+			dlg_regs->dst_y_after_scaler);
+	dml_print(
+			"DML_RQ_DLG_CALC:    dst_y_prefetch                  = 0x%0x\n",
+			dlg_regs->dst_y_prefetch);
+	dml_print(
+			"DML_RQ_DLG_CALC:    dst_y_per_vm_vblank             = 0x%0x\n",
+			dlg_regs->dst_y_per_vm_vblank);
+	dml_print(
+			"DML_RQ_DLG_CALC:    dst_y_per_row_vblank            = 0x%0x\n",
+			dlg_regs->dst_y_per_row_vblank);
+	dml_print(
+			"DML_RQ_DLG_CALC:    dst_y_per_vm_flip               = 0x%0x\n",
+			dlg_regs->dst_y_per_vm_flip);
+	dml_print(
+			"DML_RQ_DLG_CALC:    dst_y_per_row_flip              = 0x%0x\n",
+			dlg_regs->dst_y_per_row_flip);
+	dml_print(
+			"DML_RQ_DLG_CALC:    ref_freq_to_pix_freq            = 0x%0x\n",
+			dlg_regs->ref_freq_to_pix_freq);
+	dml_print(
+			"DML_RQ_DLG_CALC:    vratio_prefetch                 = 0x%0x\n",
+			dlg_regs->vratio_prefetch);
+	dml_print(
+			"DML_RQ_DLG_CALC:    vratio_prefetch_c               = 0x%0x\n",
+			dlg_regs->vratio_prefetch_c);
+	dml_print(
+			"DML_RQ_DLG_CALC:    refcyc_per_pte_group_vblank_l   = 0x%0x\n",
+			dlg_regs->refcyc_per_pte_group_vblank_l);
+	dml_print(
+			"DML_RQ_DLG_CALC:    refcyc_per_pte_group_vblank_c   = 0x%0x\n",
+			dlg_regs->refcyc_per_pte_group_vblank_c);
+	dml_print(
+			"DML_RQ_DLG_CALC:    refcyc_per_meta_chunk_vblank_l  = 0x%0x\n",
+			dlg_regs->refcyc_per_meta_chunk_vblank_l);
+	dml_print(
+			"DML_RQ_DLG_CALC:    refcyc_per_meta_chunk_vblank_c  = 0x%0x\n",
+			dlg_regs->refcyc_per_meta_chunk_vblank_c);
+	dml_print(
+			"DML_RQ_DLG_CALC:    refcyc_per_pte_group_flip_l     = 0x%0x\n",
+			dlg_regs->refcyc_per_pte_group_flip_l);
+	dml_print(
+			"DML_RQ_DLG_CALC:    refcyc_per_pte_group_flip_c     = 0x%0x\n",
+			dlg_regs->refcyc_per_pte_group_flip_c);
+	dml_print(
+			"DML_RQ_DLG_CALC:    refcyc_per_meta_chunk_flip_l    = 0x%0x\n",
+			dlg_regs->refcyc_per_meta_chunk_flip_l);
+	dml_print(
+			"DML_RQ_DLG_CALC:    refcyc_per_meta_chunk_flip_c    = 0x%0x\n",
+			dlg_regs->refcyc_per_meta_chunk_flip_c);
+	dml_print(
+			"DML_RQ_DLG_CALC:    dst_y_per_pte_row_nom_l         = 0x%0x\n",
+			dlg_regs->dst_y_per_pte_row_nom_l);
+	dml_print(
+			"DML_RQ_DLG_CALC:    dst_y_per_pte_row_nom_c         = 0x%0x\n",
+			dlg_regs->dst_y_per_pte_row_nom_c);
+	dml_print(
+			"DML_RQ_DLG_CALC:    refcyc_per_pte_group_nom_l      = 0x%0x\n",
+			dlg_regs->refcyc_per_pte_group_nom_l);
+	dml_print(
+			"DML_RQ_DLG_CALC:    refcyc_per_pte_group_nom_c      = 0x%0x\n",
+			dlg_regs->refcyc_per_pte_group_nom_c);
+	dml_print(
+			"DML_RQ_DLG_CALC:    dst_y_per_meta_row_nom_l        = 0x%0x\n",
+			dlg_regs->dst_y_per_meta_row_nom_l);
+	dml_print(
+			"DML_RQ_DLG_CALC:    dst_y_per_meta_row_nom_c        = 0x%0x\n",
+			dlg_regs->dst_y_per_meta_row_nom_c);
+	dml_print(
+			"DML_RQ_DLG_CALC:    refcyc_per_meta_chunk_nom_l     = 0x%0x\n",
+			dlg_regs->refcyc_per_meta_chunk_nom_l);
+	dml_print(
+			"DML_RQ_DLG_CALC:    refcyc_per_meta_chunk_nom_c     = 0x%0x\n",
+			dlg_regs->refcyc_per_meta_chunk_nom_c);
+	dml_print(
+			"DML_RQ_DLG_CALC:    refcyc_per_line_delivery_pre_l  = 0x%0x\n",
+			dlg_regs->refcyc_per_line_delivery_pre_l);
+	dml_print(
+			"DML_RQ_DLG_CALC:    refcyc_per_line_delivery_pre_c  = 0x%0x\n",
+			dlg_regs->refcyc_per_line_delivery_pre_c);
+	dml_print(
+			"DML_RQ_DLG_CALC:    refcyc_per_line_delivery_l      = 0x%0x\n",
+			dlg_regs->refcyc_per_line_delivery_l);
+	dml_print(
+			"DML_RQ_DLG_CALC:    refcyc_per_line_delivery_c      = 0x%0x\n",
+			dlg_regs->refcyc_per_line_delivery_c);
+	dml_print(
+			"DML_RQ_DLG_CALC:    chunk_hdl_adjust_cur0           = 0x%0x\n",
+			dlg_regs->chunk_hdl_adjust_cur0);
+	dml_print(
+			"DML_RQ_DLG_CALC:    dst_y_offset_cur1               = 0x%0x\n",
+			dlg_regs->dst_y_offset_cur1);
+	dml_print(
+			"DML_RQ_DLG_CALC:    chunk_hdl_adjust_cur1           = 0x%0x\n",
+			dlg_regs->chunk_hdl_adjust_cur1);
+	dml_print(
+			"DML_RQ_DLG_CALC:    vready_after_vcount0            = 0x%0x\n",
+			dlg_regs->vready_after_vcount0);
+	dml_print(
+			"DML_RQ_DLG_CALC:    dst_y_delta_drq_limit           = 0x%0x\n",
+			dlg_regs->dst_y_delta_drq_limit);
+	dml_print(
+			"DML_RQ_DLG_CALC:    xfc_reg_transfer_delay          = 0x%0x\n",
+			dlg_regs->xfc_reg_transfer_delay);
+	dml_print(
+			"DML_RQ_DLG_CALC:    xfc_reg_precharge_delay         = 0x%0x\n",
+			dlg_regs->xfc_reg_precharge_delay);
+	dml_print(
+			"DML_RQ_DLG_CALC:    xfc_reg_remote_surface_flip_latency = 0x%0x\n",
+			dlg_regs->xfc_reg_remote_surface_flip_latency);
+	dml_print(
+			"DML_RQ_DLG_CALC:    refcyc_per_vm_dmdata            = 0x%0x\n",
+			dlg_regs->refcyc_per_vm_dmdata);
+
+	dml_print("DML_RQ_DLG_CALC: =====================================\n");
+}
+
+void print__ttu_regs_st(struct display_mode_lib *mode_lib, const struct _vcs_dpi_display_ttu_regs_st *ttu_regs)
+{
+	(void)mode_lib;
+	(void)ttu_regs;
+	dml_print("DML_RQ_DLG_CALC: =====================================\n");
+	dml_print("DML_RQ_DLG_CALC: DISPLAY_TTU_REGS_ST\n");
+	dml_print(
+			"DML_RQ_DLG_CALC:    qos_level_low_wm                  = 0x%0x\n",
+			ttu_regs->qos_level_low_wm);
+	dml_print(
+			"DML_RQ_DLG_CALC:    qos_level_high_wm                 = 0x%0x\n",
+			ttu_regs->qos_level_high_wm);
+	dml_print(
+			"DML_RQ_DLG_CALC:    min_ttu_vblank                    = 0x%0x\n",
+			ttu_regs->min_ttu_vblank);
+	dml_print(
+			"DML_RQ_DLG_CALC:    qos_level_flip                    = 0x%0x\n",
+			ttu_regs->qos_level_flip);
+	dml_print(
+			"DML_RQ_DLG_CALC:    refcyc_per_req_delivery_pre_l     = 0x%0x\n",
+			ttu_regs->refcyc_per_req_delivery_pre_l);
+	dml_print(
+			"DML_RQ_DLG_CALC:    refcyc_per_req_delivery_l         = 0x%0x\n",
+			ttu_regs->refcyc_per_req_delivery_l);
+	dml_print(
+			"DML_RQ_DLG_CALC:    refcyc_per_req_delivery_pre_c     = 0x%0x\n",
+			ttu_regs->refcyc_per_req_delivery_pre_c);
+	dml_print(
+			"DML_RQ_DLG_CALC:    refcyc_per_req_delivery_c         = 0x%0x\n",
+			ttu_regs->refcyc_per_req_delivery_c);
+	dml_print(
+			"DML_RQ_DLG_CALC:    refcyc_per_req_delivery_cur0      = 0x%0x\n",
+			ttu_regs->refcyc_per_req_delivery_cur0);
+	dml_print(
+			"DML_RQ_DLG_CALC:    refcyc_per_req_delivery_pre_cur0  = 0x%0x\n",
+			ttu_regs->refcyc_per_req_delivery_pre_cur0);
+	dml_print(
+			"DML_RQ_DLG_CALC:    refcyc_per_req_delivery_cur1      = 0x%0x\n",
+			ttu_regs->refcyc_per_req_delivery_cur1);
+	dml_print(
+			"DML_RQ_DLG_CALC:    refcyc_per_req_delivery_pre_cur1  = 0x%0x\n",
+			ttu_regs->refcyc_per_req_delivery_pre_cur1);
+	dml_print(
+			"DML_RQ_DLG_CALC:    qos_level_fixed_l                 = 0x%0x\n",
+			ttu_regs->qos_level_fixed_l);
+	dml_print(
+			"DML_RQ_DLG_CALC:    qos_ramp_disable_l                = 0x%0x\n",
+			ttu_regs->qos_ramp_disable_l);
+	dml_print(
+			"DML_RQ_DLG_CALC:    qos_level_fixed_c                 = 0x%0x\n",
+			ttu_regs->qos_level_fixed_c);
+	dml_print(
+			"DML_RQ_DLG_CALC:    qos_ramp_disable_c                = 0x%0x\n",
+			ttu_regs->qos_ramp_disable_c);
+	dml_print(
+			"DML_RQ_DLG_CALC:    qos_level_fixed_cur0              = 0x%0x\n",
+			ttu_regs->qos_level_fixed_cur0);
+	dml_print(
+			"DML_RQ_DLG_CALC:    qos_ramp_disable_cur0             = 0x%0x\n",
+			ttu_regs->qos_ramp_disable_cur0);
+	dml_print(
+			"DML_RQ_DLG_CALC:    qos_level_fixed_cur1              = 0x%0x\n",
+			ttu_regs->qos_level_fixed_cur1);
+	dml_print(
+			"DML_RQ_DLG_CALC:    qos_ramp_disable_cur1             = 0x%0x\n",
+			ttu_regs->qos_ramp_disable_cur1);
+	dml_print("DML_RQ_DLG_CALC: =====================================\n");
+}

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/display_rq_dlg_helpers.h

@@ -0,0 +1,45 @@
+/*
+ * Copyright 2017 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 __DISPLAY_RQ_DLG_HELPERS_H__
+#define __DISPLAY_RQ_DLG_HELPERS_H__
+
+#include "display_mode_lib.h"
+
+/* Function: Printer functions
+ *  Print various struct
+ */
+void print__rq_params_st(struct display_mode_lib *mode_lib, const struct _vcs_dpi_display_rq_params_st *rq_param);
+void print__data_rq_sizing_params_st(struct display_mode_lib *mode_lib, const struct _vcs_dpi_display_data_rq_sizing_params_st *rq_sizing);
+void print__data_rq_dlg_params_st(struct display_mode_lib *mode_lib, const struct _vcs_dpi_display_data_rq_dlg_params_st *rq_dlg_param);
+void print__data_rq_misc_params_st(struct display_mode_lib *mode_lib, const struct _vcs_dpi_display_data_rq_misc_params_st *rq_misc_param);
+void print__dlg_sys_params_st(struct display_mode_lib *mode_lib, const struct _vcs_dpi_display_dlg_sys_params_st *dlg_sys_param);
+
+void print__data_rq_regs_st(struct display_mode_lib *mode_lib, const struct _vcs_dpi_display_data_rq_regs_st *rq_regs);
+void print__rq_regs_st(struct display_mode_lib *mode_lib, const struct _vcs_dpi_display_rq_regs_st *rq_regs);
+void print__dlg_regs_st(struct display_mode_lib *mode_lib, const struct _vcs_dpi_display_dlg_regs_st *dlg_regs);
+void print__ttu_regs_st(struct display_mode_lib *mode_lib, const struct _vcs_dpi_display_ttu_regs_st *ttu_regs);
+
+#endif

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dml1_display_rq_dlg_calc.h

@@ -0,0 +1,66 @@
+/*
+ * Copyright 2017 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 __DISPLAY_RQ_DLG_CALC_H__
+#define __DISPLAY_RQ_DLG_CALC_H__
+
+struct display_mode_lib;
+
+#include "display_rq_dlg_helpers.h"
+
+void dml1_extract_rq_regs(
+		struct display_mode_lib *mode_lib,
+		struct _vcs_dpi_display_rq_regs_st *rq_regs,
+		const struct _vcs_dpi_display_rq_params_st *rq_param);
+/* Function: dml_rq_dlg_get_rq_params
+ *  Calculate requestor related parameters that register definition agnostic
+ *  (i.e. this layer does try to separate real values from register definition)
+ * Input:
+ *  pipe_src_param - pipe source configuration (e.g. vp, pitch, etc.)
+ * Output:
+ *  rq_param - values that can be used to setup RQ (e.g. swath_height, plane1_addr, etc.)
+ */
+void dml1_rq_dlg_get_rq_params(
+		struct display_mode_lib *mode_lib,
+		struct _vcs_dpi_display_rq_params_st *rq_param,
+		const struct _vcs_dpi_display_pipe_source_params_st *pipe_src_param);
+
+
+/* Function: dml_rq_dlg_get_dlg_params
+ *  Calculate deadline related parameters
+ */
+void dml1_rq_dlg_get_dlg_params(
+		struct display_mode_lib *mode_lib,
+		struct _vcs_dpi_display_dlg_regs_st *dlg_regs,
+		struct _vcs_dpi_display_ttu_regs_st *ttu_regs,
+		const struct _vcs_dpi_display_rq_dlg_params_st *rq_dlg_param,
+		const struct _vcs_dpi_display_dlg_sys_params_st *dlg_sys_param,
+		const struct _vcs_dpi_display_e2e_pipe_params_st *e2e_pipe_param,
+		const bool cstate_en,
+		const bool pstate_en,
+		const bool vm_en,
+		const bool iflip_en);
+
+#endif

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dml1_frl_cap_chk.c

@@ -0,0 +1,790 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright 2023 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 "dml_logger.h"
+#include "dml1_frl_cap_chk.h"
+#include "dml_inline_defs.h"
+
+static const double __maybe_unused EPSILON			= 0.01;
+static const double __maybe_unused DBL_EPSILON			= 2.2204460492503131e-16;
+static const double __maybe_unused OVERHEAD_M			= 0.003;  /* %   */
+static const double __maybe_unused TOLERANCE_PIXEL_CLOCK	= 0.005;  /* %   */
+static const double __maybe_unused DML_TOLERANCE_AUDIO_CLOCK	= 1000;   /* ppm */
+
+#define frl_dump_var(fmt, var) {}
+#define frl_print(fmt, ...) {}
+
+const struct frl_primary_format prim_format_444[] = {
+/* VIC/Rate/Lanes/HCactive/HCBlank */
+	{64,  3, 3, 960,  360}, /* 1920x1080 @ 100 */
+	{77,  3, 3, 960,  360}, /* 1920x1080 @ 100 */
+	{63,  3, 3, 960,  140}, /* 1920x1080 @ 120 */
+	{78,  3, 3, 960,  140}, /* 1920x1080 @ 120 */
+	{93,  3, 3, 1920, 828}, /* 3840x2160 @ 24 */
+	{103, 3, 3, 1920, 828}, /* 3840x2160 @ 24 */
+	{94,  3, 3, 1920, 720}, /* 3840x2160 @ 25 */
+	{104, 3, 3, 1920, 720}, /* 3840x2160 @ 25 */
+	{95,  3, 3, 1920, 280}, /* 3840x2160 @ 30 */
+	{105, 3, 3, 1920, 280}, /* 3840x2160 @ 30 */
+	{114, 3, 3, 1920, 828}, /* 3840x2160 @ 48 */
+	{116, 3, 3, 1920, 828}, /* 3840x2160 @ 48 */
+	{96,  3, 3, 1920, 720}, /* 3840x2160 @ 50 */
+	{106, 3, 3, 1920, 720}, /* 3840x2160 @ 50 */
+	{97,  3, 3, 1920, 280}, /* 3840x2160 @ 60 */
+	{107, 3, 3, 1920, 280}, /* 3840x2160 @ 60 */
+	{117, 6, 3, 1920, 720}, /* 3840x2160 @ 100 */
+	{119, 6, 3, 1920, 720}, /* 3840x2160 @ 100 */
+	{118, 6, 3, 1920, 280}, /* 3840x2160 @ 120 */
+	{120, 6, 3, 1920, 280}, /* 3840x2160 @ 120 */
+	{98,  3, 3, 2048, 700}, /* 4096x2160 @ 24 */
+	{99,  3, 3, 2048, 592}, /* 4096x2160 @ 25 */
+	{100, 3, 3, 2048, 152}, /* 4096x2160 @ 30 */
+	{115, 3, 3, 2048, 700}, /* 4096x2160 @ 48 */
+	{101, 3, 3, 2048, 592}, /* 4096x2160 @ 50 */
+	{102, 3, 3, 2048, 152}, /* 4096x2160 @ 60 */
+	{218, 6, 3, 2048, 592}, /* 4096x2160 @ 100 */
+	{219, 6, 3, 2048, 152}, /* 4096x2160 @ 120 */
+	{121, 3, 3, 2560, 1188}, /* 5120x2160 @ 24 */
+	{122, 3, 3, 2560, 1040}, /* 5120x2160 @ 25 */
+	{123, 3, 3, 2560, 440}, /* 5120x2160 @ 30 */
+	{124, 3, 3, 2560, 256}, /* 5120x2160 @ 48 */
+	{125, 3, 3, 2560, 484}, /* 5120x2160 @ 50 */
+	{126, 3, 3, 2307, 144}, /* 5120x2160 @ 60 */
+	{127, 6, 3, 2560, 484}, /* 5120x2160 @ 100 */
+	{193, 6, 3, 2334, 104}, /* 5120x2160 @ 120 */
+	{194, 6, 3, 3840, 1660}, /* 7680x2160 @ 24 */
+	{202, 6, 3, 3840, 1660}, /* 7680x2160 @ 24 */
+	{195, 6, 3, 3840, 1560}, /* 7680x2160 @ 25 */
+	{203, 6, 3, 3840, 1560}, /* 7680x2160 @ 25 */
+	{196, 6, 3, 3840, 660}, /* 7680x2160 @ 30 */
+	{204, 6, 3, 3840, 660}, /* 7680x2160 @ 30 */
+	{197, 6, 4, 3142, 1292}, /* 7680x2160 @ 48 */
+	{205, 6, 4, 3142, 1292}, /* 7680x2160 @ 48 */
+	{198, 6, 4, 3142, 1180}, /* 7680x2160 @ 50 */
+	{206, 6, 4, 3142, 1180}, /* 7680x2160 @ 50 */
+	{199, 6, 4, 3182, 140}, /* 7680x2160 @ 60 */
+	{207, 6, 4, 3182, 140}, /* 7680x2160 @ 60 */
+	{200, 10, 4, 2680, 784}, /* 7680x2160 @ 100 */
+	{208, 10, 4, 2680, 784}, /* 7680x2160 @ 100 */
+	{201, 10, 4, 2600, 100}, /* 7680x2160 @ 120 */
+	{209, 10, 4, 2600, 100}, /* 7680x2160 @ 120 */
+	{210, 6, 3, 4854, 912}, /* 10240x4320 @ 24 */
+	{211, 6, 3, 4827, 1536}, /* 10240x4320 @ 25 */
+	{212, 6, 3, 4720, 128}, /* 10240x4320 @ 30 */
+	{213, 8, 4, 4347, 756}, /* 10240x4320 @ 48 */
+	{214, 8, 4, 4320, 1376}, /* 10240x4320 @ 50 */
+	{215, 8, 4, 4187, 124}, /* 10240x4320 @ 60 */
+};
+
+const struct frl_primary_format prim_format_422[] = {
+/* VIC/Rate/Lanes/HCactive/HCBlank */
+	{64,  3, 3, 960,  360}, /* 1920x1080 @ 100 */
+	{77,  3, 3, 960,  360}, /* 1920x1080 @ 100 */
+	{63,  3, 3, 960,  140}, /* 1920x1080 @ 120 */
+	{78,  3, 3, 960,  140}, /* 1920x1080 @ 120 */
+	{93,  3, 3, 1920, 828}, /* 3840x2160 @ 24 */
+	{103, 3, 3, 1920, 828}, /* 3840x2160 @ 24 */
+	{94,  3, 3, 1920, 720}, /* 3840x2160 @ 25 */
+	{104, 3, 3, 1920, 720}, /* 3840x2160 @ 25 */
+	{95,  3, 3, 1920, 280}, /* 3840x2160 @ 30 */
+	{105, 3, 3, 1920, 280}, /* 3840x2160 @ 30 */
+	{114, 3, 3, 1920, 828}, /* 3840x2160 @ 48 */
+	{116, 3, 3, 1920, 828}, /* 3840x2160 @ 48 */
+	{96,  3, 3, 1920, 720}, /* 3840x2160 @ 50 */
+	{106, 3, 3, 1920, 720}, /* 3840x2160 @ 50 */
+	{97,  3, 3, 1920, 280}, /* 3840x2160 @ 60 */
+	{107, 3, 3, 1920, 280}, /* 3840x2160 @ 60 */
+	{117, 3, 3, 1370, 104}, /* 3840x2160 @ 100 */
+	{119, 3, 3, 1370, 104}, /* 3840x2160 @ 100 */
+	{118, 3, 3, 1130, 104}, /* 3840x2160 @ 120 */
+	{120, 3, 3, 1130, 104}, /* 3840x2160 @ 120 */
+	{98,  3, 3, 2048, 700}, /* 4096x2160 @ 24 */
+	{99,  3, 3, 2048, 592}, /* 4096x2160 @ 25 */
+	{100, 3, 3, 2048, 152}, /* 4096x2160 @ 30 */
+	{115, 3, 3, 2048, 700}, /* 4096x2160 @ 48 */
+	{101, 3, 3, 2048, 592}, /* 4096x2160 @ 50 */
+	{102, 3, 3, 2048, 152}, /* 4096x2160 @ 60 */
+	{218, 6, 3, 2048, 592}, /* 4096x2160 @ 100 */
+	{219, 6, 3, 2048, 152}, /* 4096x2160 @ 120 */
+	{121, 3, 3, 2560, 1188}, /* 5120x2160 @ 24 */
+	{122, 3, 3, 2560, 1040}, /* 5120x2160 @ 25 */
+	{123, 3, 3, 2560, 440}, /* 5120x2160 @ 30 */
+	{124, 3, 3, 2560, 256}, /* 5120x2160 @ 48 */
+	{125, 3, 3, 2560, 484}, /* 5120x2160 @ 50 */
+	{126, 3, 3, 2307, 144}, /* 5120x2160 @ 60 */
+	{127, 6, 3, 2560, 484}, /* 5120x2160 @ 100 */
+	{193, 6, 3, 2334, 104}, /* 5120x2160 @ 120 */
+	{194, 3, 3, 2460, 816}, /* 7680x2160 @ 24 */
+	{202, 3, 3, 2460, 816}, /* 7680x2160 @ 24 */
+	{195, 3, 3, 2460, 732}, /* 7680x2160 @ 25 */
+	{203, 3, 3, 2460, 732}, /* 7680x2160 @ 25 */
+	{196, 3, 3, 2360, 144}, /* 7680x2160 @ 30 */
+	{204, 3, 3, 2360, 144}, /* 7680x2160 @ 30 */
+	{197, 6, 3, 2460, 816}, /* 7680x2160 @ 48 */
+	{205, 6, 3, 2460, 816}, /* 7680x2160 @ 48 */
+	{198, 6, 3, 2460, 732}, /* 7680x2160 @ 50 */
+	{206, 6, 3, 2460, 732}, /* 7680x2160 @ 50 */
+	{199, 6, 3, 2380, 116}, /* 7680x2160 @ 60 */
+	{207, 6, 3, 2380, 116}, /* 7680x2160 @ 60 */
+	{200, 10, 4, 2680, 784}, /* 7680x2160 @ 100 */
+	{208, 10, 4, 2680, 784}, /* 7680x2160 @ 100 */
+	{201, 10, 4, 2600, 100}, /* 7680x2160 @ 120 */
+	{209, 10, 4, 2600, 100}, /* 7680x2160 @ 120 */
+	{210, 6, 3, 4854, 912}, /* 10240x4320 @ 24 */
+	{211, 6, 3, 4827, 1536}, /* 10240x4320 @ 25 */
+	{212, 6, 3, 4720, 128}, /* 10240x4320 @ 30 */
+	{213, 6, 4, 3360, 420}, /* 10240x4320 @ 48 */
+	{214, 6, 4, 3334, 892}, /* 10240x4320 @ 50 */
+	{215, 6, 4, 3120, 124}, /* 10240x4320 @ 60 */
+	{216, 12, 4, 3334, 764}, /* 10240x4320 @ 100 */
+	{217, 12, 4, 3120, 124}, /* 10240x4320 @ 120 */
+};
+
+const struct frl_primary_format prim_format_420[] = {
+/* VIC/Rate/Lanes/HCactive/HCBlank */
+	{114, 3, 3, 1920, 828}, /* 3840x2160 @ 48 */
+	{116, 3, 3, 1920, 828}, /* 3840x2160 @ 48 */
+	{96,  3, 3, 1920, 720}, /* 3840x2160 @ 50 */
+	{106, 3, 3, 1920, 720}, /* 3840x2160 @ 50 */
+	{97,  3, 3, 1920, 280}, /* 3840x2160 @ 60 */
+	{107, 3, 3, 1920, 280}, /* 3840x2160 @ 60 */
+	{117, 3, 3, 1370, 104}, /* 3840x2160 @ 100 */
+	{119, 3, 3, 1370, 104}, /* 3840x2160 @ 100 */
+	{118, 3, 3, 1130, 104}, /* 3840x2160 @ 120 */
+	{120, 3, 3, 1130, 104}, /* 3840x2160 @ 120 */
+	{115, 3, 3, 2048, 700}, /* 4096x2160 @ 48 */
+	{101, 3, 3, 2048, 592}, /* 4096x2160 @ 50 */
+	{102, 3, 3, 2048, 152}, /* 4096x2160 @ 60 */
+	{218, 3, 3, 1376, 96}, /* 4096x2160 @ 100 */
+	{219, 3, 3, 1131, 84}, /* 4096x2160 @ 120 */
+	{124, 3, 3, 2560, 256}, /* 5120x2160 @ 48 */
+	{125, 3, 3, 2560, 484}, /* 5120x2160 @ 50 */
+	{126, 3, 3, 2307, 144}, /* 5120x2160 @ 60 */
+	{127, 6, 3, 2560, 484}, /* 5120x2160 @ 100 */
+	{193, 6, 3, 2334, 104}, /* 5120x2160 @ 120 */
+	{194, 3, 3, 2460, 816}, /* 7680x2160 @ 24 */
+	{202, 3, 3, 2460, 816}, /* 7680x2160 @ 24 */
+	{195, 3, 3, 2460, 732}, /* 7680x2160 @ 25 */
+	{203, 3, 3, 2460, 732}, /* 7680x2160 @ 25 */
+	{196, 3, 3, 2360, 144}, /* 7680x2160 @ 30 */
+	{204, 3, 3, 2360, 144}, /* 7680x2160 @ 30 */
+	{197, 6, 3, 2460, 816}, /* 7680x2160 @ 48 */
+	{205, 6, 3, 2460, 816}, /* 7680x2160 @ 48 */
+	{198, 6, 3, 2460, 732}, /* 7680x2160 @ 50 */
+	{206, 6, 3, 2460, 732}, /* 7680x2160 @ 50 */
+	{199, 6, 3, 2380, 116}, /* 7680x2160 @ 60 */
+	{207, 6, 3, 2380, 116}, /* 7680x2160 @ 60 */
+	{200, 8, 4, 2240, 480}, /* 7680x2160 @ 100 */
+	{208, 8, 4, 2240, 480}, /* 7680x2160 @ 100 */
+	{201, 8, 4, 2062, 108}, /* 7680x2160 @ 120 */
+	{209, 8, 4, 2062, 108}, /* 7680x2160 @ 120 */
+	{210, 3, 3, 2614, 172}, /* 10240x4320 @ 24 */
+	{211, 3, 3, 2614, 500}, /* 10240x4320 @ 25 */
+	{212, 6, 3, 4720, 128}, /* 10240x4320 @ 30 */
+	{213, 6, 3, 2614, 172}, /* 10240x4320 @ 48 */
+	{214, 6, 4, 3334, 892}, /* 10240x4320 @ 50 */
+	{215, 6, 4, 3120, 124}, /* 10240x4320 @ 60 */
+	{216, 10, 4, 2854, 520}, /* 10240x4320 @ 100 */
+	{217, 10, 4, 2587, 120}, /* 10240x4320 @ 120 */
+};
+
+enum frl_cap_chk_result dml1_frl_cap_chk_common(struct frl_cap_chk_intermediates *inter,
+						struct frl_cap_chk_params *params)
+{
+	double audio_bw_reserve = (params->compressed ? 192000.0 : 0.0);
+
+	dc_assert_fp_enabled();
+
+#ifdef DEBUG_FRL_CAP_CHK
+	{
+		printf("frl_cap_chk inputs:\n");
+		printf("-------------------\n");
+		frl_dump_var("%i",  params->lanes);
+		frl_dump_var("%le", params->f_pixel_clock_nominal);
+		frl_dump_var("%le", params->r_bit_nominal);
+		frl_dump_var("%i",  params->audio_packet_type);
+		frl_dump_var("%le", params->f_audio);
+		frl_dump_var("%i",  params->h_active);
+		frl_dump_var("%i",  params->h_blank);
+		frl_dump_var("%i",  params->bpc);
+		frl_dump_var("%i",  params->pixel_encoding);
+		frl_dump_var("%i",  params->compressed);
+		frl_dump_var("%i",  params->slices);
+		frl_dump_var("%i",  params->slice_width);
+		frl_dump_var("%le", params->bpp_target);
+		frl_dump_var("%i",  params->layout);
+		frl_dump_var("%i",  params->acat);
+		printf("frl_cap_chk outputs:\n");
+		printf("---------------------\n");
+	}
+#endif
+
+	inter->c_frl_sb          = 4 * C_FRL_CB + params->lanes;
+	inter->overhead_sb       = (double)params->lanes / inter->c_frl_sb;
+	inter->overhead_rs       = 8.0 * 4.0 / inter->c_frl_sb;
+	inter->overhead_map      = 2.5 / inter->c_frl_sb;
+	inter->overhead_min      = inter->overhead_sb + inter->overhead_rs + inter->overhead_map;
+	inter->overhead_max      = inter->overhead_min + OVERHEAD_M;
+	inter->f_pixel_clock_max = params->f_pixel_clock_nominal * (1.0 + TOLERANCE_PIXEL_CLOCK);
+	inter->t_line            = (params->h_active + params->h_blank) / inter->f_pixel_clock_max;
+	inter->r_bit_min         = params->r_bit_nominal * (1.0 - TOLERANCE_FRL_BIT / 1000000.0);
+	inter->r_frl_char_min    = inter->r_bit_min / 18.0;
+	inter->c_frl_line        = dml_floor(inter->t_line * inter->r_frl_char_min * params->lanes, 1);
+
+#ifdef DEBUG_FRL_CAP_CHK
+	{
+		frl_dump_var("%i",  inter->c_frl_sb);
+		frl_dump_var("%le", inter->overhead_sb);
+		frl_dump_var("%le", inter->overhead_rs);
+		frl_dump_var("%le", inter->overhead_map);
+		frl_dump_var("%le", inter->overhead_min);
+		frl_dump_var("%le", inter->overhead_max);
+		frl_dump_var("%le", inter->f_pixel_clock_max);
+		frl_dump_var("%le", inter->t_line);
+		frl_dump_var("%le", inter->r_bit_min);
+		frl_dump_var("%le", inter->r_frl_char_min);
+		frl_dump_var("%le", inter->c_frl_line);
+	}
+#endif
+
+	switch (params->audio_packet_type) {
+	case 0x02:
+		if (params->layout == 0)
+			inter->ap = 0.25;
+		else if (params->layout == 1)
+			inter->ap = 1.0;
+		break;
+	case 0x08:
+		inter->ap = 0.25;
+		break;
+	case 0x09:
+		inter->ap = 1.0;
+		break;
+	case 0x07:
+	case 0x0e:
+	case 0x0f:
+	case 0x0b:
+	case 0x0c:
+		/* Unsupported audio format */
+		return FRL_CAP_CHK_ERROR_UNSUPPORTED_AUDIO;
+	default:
+		inter->ap = 0.0;
+	}
+
+	inter->r_ap                   = (dml_max(audio_bw_reserve, params->f_audio * inter->ap) + 2 * ACR_RATE_MAX) * (1 + DML_TOLERANCE_AUDIO_CLOCK / 1000000.0);
+	inter->avg_audio_packets_line = inter->r_ap * inter->t_line;
+	inter->audio_packets_line     = (int)dml_ceil(inter->avg_audio_packets_line, 1);
+	inter->blank_audio_min        = 32 + 32 * inter->audio_packets_line; // h_blank_audio_min or hc_blank_audio_min
+
+	params->borrow_params.audio_packets_line = inter->audio_packets_line;
+
+#ifdef DEBUG_FRL_CAP_CHK
+	{
+		frl_dump_var("%le", inter->ap);
+		frl_dump_var("%le", inter->r_ap);
+		frl_dump_var("%le", inter->avg_audio_packets_line);
+		frl_dump_var("%i",  inter->audio_packets_line);
+		frl_dump_var("%i",  inter->blank_audio_min);
+	}
+#endif
+
+	return FRL_CAP_CHK_OK;
+}
+
+enum frl_cap_chk_result dml1_frl_cap_chk_uncompressed(struct frl_cap_chk_params *params,
+						      struct frl_cap_chk_intermediates *inter)
+{
+	enum frl_cap_chk_result res;
+	int      k_420;
+	double   k_cd;
+	int      c_frl_free;
+	int      c_frl_rc_margin;
+	int      c_frl_rc_savings;
+	int      bpp;
+	double   bytes_line;
+	int      tb_active;
+	int      tb_blank;
+	double   f_tb_average;
+	double   t_active_ref;
+	double   t_blank_ref;
+	double   t_active_min;
+	double   t_blank_min;
+	double   t_borrowed;
+	double   tb_borrowed;
+	int      c_frl_actual_payload;
+	double   utilization;
+	double   margin;
+
+	dc_assert_fp_enabled();
+
+	res = dml1_frl_cap_chk_common(inter, params);
+	if (res != FRL_CAP_CHK_OK)
+		return res;
+
+	k_420            = params->pixel_encoding == HDMI_FRL_PIXEL_ENCODING_420 ? 2 : 1;
+	k_cd             = params->pixel_encoding == HDMI_FRL_PIXEL_ENCODING_422 ? 1.0 : params->bpc / 8.0;
+	c_frl_free       = (int)dml_max(params->h_blank * k_cd / k_420 - 32 * (1 + inter->audio_packets_line) - 7, 0);
+	c_frl_rc_margin  = 4;
+	c_frl_rc_savings = (int)dml_floor(dml_max(((7.0 / 8.0) * c_frl_free) - c_frl_rc_margin, 0.0), 1);
+	bpp              = (int)(24 * k_cd / k_420);
+	bytes_line       = bpp * params->h_active / 8.0;
+	tb_active        = (int)dml_ceil(bytes_line / 3, 1);
+	tb_blank         = (int)dml_ceil(params->h_blank * k_cd / k_420, 1);
+
+#ifdef DEBUG_FRL_CAP_CHK
+	{
+		frl_dump_var("%i", k_420);
+		frl_dump_var("%le", k_cd);
+		frl_dump_var("%i", c_frl_free);
+		frl_dump_var("%i", c_frl_rc_margin);
+		frl_dump_var("%i", c_frl_rc_savings);
+		frl_dump_var("%i", bpp);
+		frl_dump_var("%le", bytes_line);
+		frl_dump_var("%i", tb_active);
+		frl_dump_var("%i", tb_blank);
+	}
+#endif
+
+	if (!(inter->blank_audio_min <= tb_blank)) {
+		frl_dump_var("%i", inter->blank_audio_min);
+		frl_dump_var("%i", tb_blank);
+		return FRL_CAP_CHK_ERROR_AUDIO_BW;
+	}
+
+	f_tb_average = (inter->f_pixel_clock_max / (params->h_active + params->h_blank)) * (tb_active + tb_blank);
+	t_active_ref = inter->t_line * ((double)params->h_active / (params->h_active + params->h_blank));
+	t_blank_ref  = inter->t_line * ((double)params->h_blank / (params->h_active + params->h_blank));
+	t_active_min = (3.0 / 2.0) * tb_active / (params->lanes * inter->r_frl_char_min * (1.0 - inter->overhead_max));
+	t_blank_min  = tb_blank / (params->lanes * inter->r_frl_char_min * (1.0 - inter->overhead_max));
+
+#ifdef DEBUG_FRL_CAP_CHK
+	{
+		frl_dump_var("%le", f_tb_average);
+		frl_dump_var("%le", t_active_ref);
+		frl_dump_var("%le", t_blank_ref);
+		frl_dump_var("%le", t_active_min);
+		frl_dump_var("%le", t_blank_min);
+	}
+#endif
+
+	if (t_active_ref >= t_active_min && t_blank_ref >= t_blank_min) {
+		t_borrowed = 0;
+		params->borrow_params.borrow_mode = FRL_BORROW_MODE_NONE;
+	} else if ((t_active_ref < t_active_min) && (t_blank_ref >= t_blank_min)) {
+		t_borrowed = t_active_min - t_active_ref;
+		params->borrow_params.borrow_mode = FRL_BORROW_MODE_FROM_BLANK;
+	} else {
+		return FRL_CAP_CHK_ERROR_BORROW;
+	}
+
+	tb_borrowed = dml_ceil(t_borrowed * f_tb_average, 1);
+
+#ifdef DEBUG_FRL_CAP_CHK
+	{
+		frl_dump_var("%le", tb_borrowed);
+		frl_dump_var("%i", params->borrow_params.borrow_mode);
+	}
+#endif
+
+	if (!(tb_borrowed <= TB_BORROWED_MAX))
+		return FRL_CAP_CHK_ERROR_MAX_BORROW;
+
+	c_frl_actual_payload = (int)(dml_ceil((3.0 / 2.0) * tb_active, 1) + tb_blank - c_frl_rc_savings);
+	utilization          = c_frl_actual_payload / inter->c_frl_line;
+	margin               = 1.0 - (utilization + inter->overhead_max);
+
+#ifdef DEBUG_FRL_CAP_CHK
+	{
+		frl_dump_var("%i",  c_frl_actual_payload);
+		frl_dump_var("%le", utilization);
+		frl_dump_var("%le", margin);
+	}
+#endif
+
+	if (margin < 0 && dcn_bw_fabs(margin) > EPSILON)
+		return FRL_CAP_CHK_ERROR_MARGIN;
+
+	return FRL_CAP_CHK_OK;
+}
+
+#if defined (CONFIG_DRM_AMD_DC_FP)
+enum frl_cap_chk_result dml1_frl_cap_chk_compressed(struct frl_cap_chk_params *params,
+						    struct frl_cap_chk_intermediates *inter)
+{
+	enum frl_cap_chk_result res;
+	int      c_frl_available;
+#if defined(DEBUG_FRL_CAP_CHK)
+	int      c_frl_active_available;
+	int      c_frl_blank_available;
+#endif
+	int      bytes_target = 0;
+	int      hc_active_target;
+	int      hc_blank_target_est1;
+	int      hc_blank_target_est2;
+	int      hc_blank_target = 0;
+	int      c_frl_actual_target_payload;
+	double   utilization_targeted;
+	double   margin_target;
+	double   f_tb_average;
+	double   t_active_ref;
+	double   t_blank_ref;
+	double   t_active_target;
+	double   t_blank_target;
+	double   tb_borrowed;
+#ifdef DEBUG_FRL_CAP_CHK
+	double   tb_delta;
+	double   tb_delta_limit;
+	int      tb_worst;
+#endif
+	int table_size_444 = ARRAY_SIZE(prim_format_444);
+	int table_size_422 = ARRAY_SIZE(prim_format_422);
+	int table_size_420 = ARRAY_SIZE(prim_format_420);
+	int i;
+	bool hc_active_blank_predefined = false;
+
+	dc_assert_fp_enabled();
+
+	res = dml1_frl_cap_chk_common(inter, params);
+
+	if (res != FRL_CAP_CHK_OK)
+		return res;
+
+	c_frl_available        = (int)dml_floor((1 - inter->overhead_max) * inter->c_frl_line, 1);
+#if defined(DEBUG_FRL_CAP_CHK)
+	c_frl_active_available = dml_floor(c_frl_available * ((double)params->h_active / (params->h_active + params->h_blank)), 1);
+	c_frl_blank_available  = dml_floor(c_frl_available * ((double)params->h_blank / (params->h_active + params->h_blank)), 1);
+#endif
+	bytes_target           = (int)(params->slices * dml_ceil(params->bpp_target * params->slice_width / 8.0, 1));
+
+	if (!params->bypass_hc_target_calc)
+			hc_active_target = (int)dml_ceil(bytes_target / 3.0, 1);
+	else
+		hc_active_target = params->borrow_params.hc_active_target;
+
+	if (!params->allow_all_bpp && params->vic != 0) {
+		if (params->pixel_encoding == HDMI_FRL_PIXEL_ENCODING_444) {
+			for (i = 0; i < table_size_444 ; i++) {
+				if (prim_format_444[i].vic == params->vic) {
+					params->borrow_params.hc_active_target = prim_format_444[i].hc_active;
+					params->borrow_params.hc_blank_target  = prim_format_444[i].hc_blank;
+					hc_active_blank_predefined = true;
+					break;
+				}
+			}
+		} else if (params->pixel_encoding == HDMI_FRL_PIXEL_ENCODING_422) {
+			for (i = 0; i < table_size_422 ; i++) {
+				if (prim_format_422[i].vic == params->vic) {
+					params->borrow_params.hc_active_target = prim_format_422[i].hc_active;
+					params->borrow_params.hc_blank_target  = prim_format_422[i].hc_blank;
+					hc_active_blank_predefined = true;
+					break;
+				}
+			}
+		} else if (params->pixel_encoding == HDMI_FRL_PIXEL_ENCODING_420) {
+			for (i = 0; i < table_size_420 ; i++) {
+				if (prim_format_420[i].vic == params->vic) {
+					params->borrow_params.hc_active_target = prim_format_420[i].hc_active;
+					params->borrow_params.hc_blank_target  = prim_format_420[i].hc_blank;
+					hc_active_blank_predefined = true;
+					break;
+				}
+			}
+		}
+
+		if (hc_active_blank_predefined) {
+			hc_active_target = params->borrow_params.hc_active_target;
+			hc_blank_target = params->borrow_params.hc_blank_target;
+		}
+	}
+
+	hc_blank_target_est1 = (int)dml_ceil(hc_active_target * ((double)params->h_blank / params->h_active), 1);
+	hc_blank_target_est2 = (int)dml_max(hc_blank_target_est1, inter->blank_audio_min);
+
+	if (!hc_active_blank_predefined) {
+		if (!params->bypass_hc_target_calc) {
+			hc_blank_target = (int)(4 * dml_floor(dml_min(hc_blank_target_est2, c_frl_available - 3.0 / 2.0 * hc_active_target) / 4.0, 1));
+
+			params->borrow_params.hc_active_target = hc_active_target;
+			params->borrow_params.hc_blank_target  = hc_blank_target;
+		} else {
+			hc_blank_target  = params->borrow_params.hc_blank_target;
+		}
+	}
+
+#ifdef DEBUG_FRL_CAP_CHK
+	{
+		frl_dump_var("%i", c_frl_available);
+		frl_dump_var("%i", c_frl_active_available);
+		frl_dump_var("%i", c_frl_blank_available);
+		frl_dump_var("%i", bytes_target);
+		frl_dump_var("%i", hc_active_target);
+		frl_dump_var("%i", hc_blank_target_est1);
+		frl_dump_var("%i", hc_blank_target_est2);
+		frl_dump_var("%i", hc_blank_target);
+	}
+#endif
+
+	if (!(inter->blank_audio_min <= hc_blank_target)) {
+		frl_dump_var("%i", inter->blank_audio_min);
+		frl_dump_var("%i", hc_blank_target);
+		return FRL_CAP_CHK_ERROR_AUDIO_BW;
+	}
+
+	f_tb_average    = inter->f_pixel_clock_max / (params->h_active + params->h_blank) * (hc_active_target + hc_blank_target);
+	t_active_ref    = inter->t_line * ((double)params->h_active / (params->h_active + params->h_blank));
+	t_blank_ref     = inter->t_line - t_active_ref; // * ((double) params->h_blank / (params->h_active + params->h_blank));
+	t_active_target = dml_max((hc_active_target / f_tb_average),
+				  (3.0 / 2.0 * hc_active_target) /
+				  (params->lanes * inter->r_frl_char_min * (1.0 - inter->overhead_max)));
+	t_blank_target  = inter->t_line - t_active_target;
+
+	tb_borrowed     = t_active_target * f_tb_average - hc_active_target;
+#ifdef DEBUG_FRL_CAP_CHK
+	tb_delta        = dcn_bw_fabs(t_active_target - t_active_ref) * (hc_active_target + hc_blank_target_est1) / inter->t_line;
+
+	{
+		frl_dump_var("%le", f_tb_average);
+		frl_dump_var("%le", t_active_ref);
+		frl_dump_var("%le", t_blank_ref);
+		frl_dump_var("%le", t_active_target);
+		frl_dump_var("%le", t_blank_target);
+		frl_dump_var("%le", tb_delta);
+	}
+#endif
+
+	if (t_blank_target - t_blank_ref > DBL_EPSILON) {
+#ifdef DEBUG_FRL_CAP_CHK
+		tb_delta_limit = (t_active_ref - hc_active_target / f_tb_average) * (hc_active_target + hc_blank_target_est1) / inter->t_line;
+#endif
+		params->borrow_params.borrow_mode = FRL_BORROW_MODE_FROM_ACTIVE;
+	} else if (t_active_target - t_active_ref > DBL_EPSILON) {
+#ifdef DEBUG_FRL_CAP_CHK
+		tb_delta_limit = tb_delta;
+#endif
+		params->borrow_params.borrow_mode = FRL_BORROW_MODE_FROM_BLANK;
+	} else {
+#ifdef DEBUG_FRL_CAP_CHK
+		tb_delta_limit = 0;
+#endif
+		params->borrow_params.borrow_mode = FRL_BORROW_MODE_NONE;
+	}
+
+#ifdef DEBUG_FRL_CAP_CHK
+	tb_worst = dml_ceil(dml_max(tb_borrowed, tb_delta_limit), 1);
+
+	{
+		frl_dump_var("%le", tb_delta_limit);
+		frl_dump_var("%le", tb_borrowed);
+		frl_dump_var("%i", params->borrow_params.borrow_mode);
+		frl_dump_var("%i", tb_worst);
+	}
+#endif
+
+	if (!(tb_borrowed <= TB_BORROWED_MAX))
+		return FRL_CAP_CHK_ERROR_MAX_BORROW;
+
+	c_frl_actual_target_payload = (int)(dml_ceil(3.0 / 2.0 * hc_active_target, 1) + hc_blank_target);
+	utilization_targeted        = c_frl_actual_target_payload / inter->c_frl_line;
+	margin_target               = 1.0 - (utilization_targeted + inter->overhead_max);
+
+#ifdef DEBUG_FRL_CAP_CHK
+	{
+		frl_dump_var("%i",  c_frl_actual_target_payload);
+		frl_dump_var("%le", utilization_targeted);
+		frl_dump_var("%le", margin_target);
+	}
+#endif
+
+	// oversubscribed bandwidth relative to margin
+	if (margin_target < 0 && dcn_bw_fabs(margin_target) > EPSILON)
+		return FRL_CAP_CHK_ERROR_MARGIN;
+
+	return FRL_CAP_CHK_OK;
+}
+#endif
+
+enum frl_cap_chk_result dml1_frl_cap_chk(struct frl_cap_chk_params *params)
+{
+	struct frl_cap_chk_intermediates inter;
+
+#if defined (CONFIG_DRM_AMD_DC_FP)
+	if (params->compressed)
+		return dml1_frl_cap_chk_compressed(params, &inter);
+#endif
+
+	return dml1_frl_cap_chk_inter(params, &inter);
+}
+
+enum frl_cap_chk_result dml1_frl_cap_chk_inter(struct frl_cap_chk_params *params,
+					       struct frl_cap_chk_intermediates *inter)
+{
+	return dml1_frl_cap_chk_uncompressed(params, inter);
+}
+
+static double calculate_compressed_active_time(uint32_t h_active,
+	const uint32_t h_blank,
+	const int hc_active,
+	const int hc_blank,
+	const uint32_t frl_num_lanes,
+	const double pix_clk,
+	const int frl_link_rate)
+{
+	double f_tb_average;
+	double r_bit_nominal;
+	double r_bit_min;
+	double r_frl_char_min;
+	double t_active_est_1;
+	double t_active_est_2;
+	double t_active_target;
+	int c_frl_sb = 510;
+	int frl_bit_tolerance = 300;
+	double overhead_m = 0.003;
+	double overhead_sb;
+	double overhead_rs;
+	double overhead_map;
+	double overhead_min;
+	double overhead_max;
+
+	switch (frl_link_rate) {
+	case FRL_LINK_RATE_3GBPS:
+		r_bit_nominal = 3.0e9;
+		break;
+	case FRL_LINK_RATE_6GBPS:
+	case FRL_LINK_RATE_6GBPS_4LANE:
+		r_bit_nominal = 6.0e9;
+		break;
+	case FRL_LINK_RATE_8GBPS:
+		r_bit_nominal = 8.0e9;
+		break;
+	case FRL_LINK_RATE_10GBPS:
+	default:
+		r_bit_nominal = 10.0e9;
+		break;
+	case FRL_LINK_RATE_12GBPS:
+		r_bit_nominal = 12.0e9;
+		break;
+	}
+
+	f_tb_average = pix_clk / (h_active + h_blank)
+					* (hc_active + hc_blank);
+
+	c_frl_sb = 4 * c_frl_sb + frl_num_lanes;
+	overhead_sb = (double)frl_num_lanes / c_frl_sb;
+	overhead_rs = 8.0 * 4.0 / c_frl_sb;
+	overhead_map = 2.5 / c_frl_sb;
+	overhead_min = overhead_sb + overhead_rs + overhead_map;
+	overhead_max = overhead_min + overhead_m;
+
+	r_bit_min = r_bit_nominal * (1.0 - frl_bit_tolerance / 1000000.0);
+	r_frl_char_min = r_bit_min / 18.0;
+	t_active_est_1 = hc_active / f_tb_average;
+	t_active_est_2 = (3.0 / 2.0 * hc_active) /
+				  (frl_num_lanes * r_frl_char_min * (1.0 - overhead_max));
+
+	if (t_active_est_1 > t_active_est_2) {
+		t_active_target = t_active_est_1;
+	} else {
+		t_active_target = t_active_est_2;
+	}
+
+	return t_active_target;
+}
+
+void frl_modified_pix_clock_for_dsc_padding(const int hc_active_target,
+	const int hc_blank_target,
+	const uint8_t frl_num_lanes,
+	const uint32_t pix_clk_100hz,
+	const int frl_link_rate,
+	const uint32_t h_addressable,
+	const uint32_t h_border_left,
+	const uint32_t h_border_right,
+	const uint32_t h_total,
+	const uint32_t h_addressable_otg,
+	uint32_t *pix_clk_100hz_otg,
+	uint32_t *h_total_otg)
+{
+	double pix_clk;
+	int h_active;
+	int h_blank;
+	double t_active_target;
+	double hw_pix_clk;
+	double h_total_otg_temp;
+
+	pix_clk = (double)pix_clk_100hz * 100;
+
+	h_active = h_addressable + h_border_left + h_border_right;
+	h_blank = h_total - h_active;
+
+	t_active_target = calculate_compressed_active_time(h_active, h_blank, hc_active_target, hc_blank_target, frl_num_lanes, pix_clk, frl_link_rate);
+
+	h_total_otg_temp = ((double)h_addressable_otg * (double)h_total) / ((double)pix_clk_100hz * 100.0 * t_active_target);
+	/* Htotal must be a multiple of 4, also take the ceiling */
+	*h_total_otg = (uint32_t)dml_ceil(h_total_otg_temp, 4.0);
+
+	hw_pix_clk = (double)(pix_clk_100hz * 100.0 * (double)*h_total_otg) / (double)h_total;
+	*pix_clk_100hz_otg = (uint32_t)(hw_pix_clk / 100.0);
+}
+
+int frl_modify_borrow_mode_for_dsc_padding(const uint32_t pix_clk_100hz,
+	const uint32_t h_active,
+	const uint32_t h_active_padded,
+	const uint32_t h_blank,
+	const uint32_t h_blank_padded,
+	const int hc_active,
+	const int hc_blank,
+	const uint8_t frl_num_lanes,
+	const int frl_link_rate)
+{
+	double f_pixel_clock_max;
+	double t_line;
+	double t_active;
+	double t_blank;
+	double t_active_target;
+	double t_blank_target;
+	double pix_clk_tolerance = 0.005;
+
+	enum frl_borrow_mode borrow_mode;
+
+	f_pixel_clock_max = (double)pix_clk_100hz * (1.0 + pix_clk_tolerance);
+	t_line = (double)(h_active + h_blank) / f_pixel_clock_max;
+
+	t_active_target = calculate_compressed_active_time(h_active, h_blank, hc_active, hc_blank, frl_num_lanes, f_pixel_clock_max, frl_link_rate);
+
+	t_active = t_line * ((double)h_active_padded / (h_active_padded + h_blank_padded));
+	t_blank = t_line - t_active;
+
+	t_blank_target = t_line - t_active_target;
+
+	if (t_blank_target - t_blank > DBL_EPSILON) {
+		borrow_mode = FRL_BORROW_MODE_FROM_ACTIVE;
+	} else if (t_active_target - t_active > DBL_EPSILON) {
+		borrow_mode = FRL_BORROW_MODE_FROM_BLANK;
+	} else {
+		borrow_mode = FRL_BORROW_MODE_NONE;
+	}
+
+	return borrow_mode;
+}

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dml1_frl_cap_chk.h

@@ -0,0 +1,173 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright 2022 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 __DML1_FRL_CAP_CHK_H__
+#define __DML1_FRL_CAP_CHK_H__
+
+#include "os_types.h"
+
+#define TB_BORROWED_MAX	  400
+#define C_FRL_CB	  510
+#define TOLERANCE_FRL_BIT 300    /* ppm */
+#define ACR_RATE_MAX	  1500
+
+enum hdmi_frl_pixel_encoding {
+	HDMI_FRL_PIXEL_ENCODING_444,
+	HDMI_FRL_PIXEL_ENCODING_422,
+	HDMI_FRL_PIXEL_ENCODING_420
+};
+
+enum frl_cap_chk_result {
+	FRL_CAP_CHK_OK = 0,
+
+	FRL_CAP_CHK_ERROR_AUDIO_BW   = -1,
+	FRL_CAP_CHK_ERROR_BORROW     = -2,
+	FRL_CAP_CHK_ERROR_MAX_BORROW = -3,
+	FRL_CAP_CHK_ERROR_MARGIN     = -4,
+
+	FRL_CAP_CHK_ERROR_UNSUPPORTED_AUDIO = -1000
+};
+
+enum frl_borrow_mode {
+	FRL_BORROW_MODE_NONE,
+	FRL_BORROW_MODE_FROM_ACTIVE,
+	FRL_BORROW_MODE_FROM_BLANK
+};
+
+enum frl_link_rate {
+	FRL_LINK_RATE_DISABLE = 0,
+	FRL_LINK_RATE_3GBPS,
+	FRL_LINK_RATE_6GBPS,
+	FRL_LINK_RATE_6GBPS_4LANE,
+	FRL_LINK_RATE_8GBPS,
+	FRL_LINK_RATE_10GBPS,
+	FRL_LINK_RATE_12GBPS,
+	FRL_LINK_RATE_16GBPS,
+	FRL_LINK_RATE_20GBPS,
+	FRL_LINK_RATE_24GBPS
+};
+
+struct frl_dml_borrow_params {
+	int audio_packets_line;
+	int hc_active_target;
+	int hc_blank_target;
+	enum frl_borrow_mode borrow_mode;
+};
+
+struct frl_primary_format {
+	uint32_t vic;
+	uint32_t frl_rate;
+	uint32_t frl_lanes;
+	uint32_t hc_active;
+	uint32_t hc_blank;
+};
+
+struct frl_cap_chk_intermediates {
+	int      c_frl_sb;
+	double   overhead_sb;
+	double   overhead_rs;
+	double   overhead_map;
+	double   overhead_min;
+	double   overhead_max;
+	double   f_pixel_clock_max;
+	double   t_line;
+	double   r_bit_min;
+	double   r_frl_char_min;
+	double   c_frl_line;
+	double   ap;
+	double   r_ap;
+	double   avg_audio_packets_line;
+	int      audio_packets_line;
+	int      blank_audio_min;
+};
+
+struct frl_cap_chk_params {
+	int      lanes;
+	double   f_pixel_clock_nominal;   /* Pixel Clock rate (Hz)  */
+	double   r_bit_nominal;           /* FRL bitrate (bps) */
+	int      audio_packet_type;
+	double   f_audio;                 /* Audio rate (Hz) */
+	int      h_active;                /* Active pixels per line */
+	int      h_blank;                 /* Blanking pixels per line */
+	int      bpc;                     /* Bits per component */
+	int      vic;                     /* Video Identification Code */
+
+	enum hdmi_frl_pixel_encoding    pixel_encoding;
+
+	bool     compressed;              /* set to true if DSC is enabled */
+	bool     bypass_hc_target_calc;   /* debug only */
+	bool     allow_all_bpp;           /* dsc_all_bpp */
+
+	/* DSC parameters */
+	int      slices;
+	int      slice_width;
+	double   bpp_target;
+	bool     is_ovt;
+	int      layout;
+	int      acat;    /* not supported */
+
+	/* outputs */
+	struct frl_dml_borrow_params borrow_params;
+	int      average_tribyte_rate;
+};
+
+enum frl_cap_chk_result dml1_frl_cap_chk(struct frl_cap_chk_params *params);
+
+enum frl_cap_chk_result dml1_frl_cap_chk_inter(struct frl_cap_chk_params *params,
+					       struct frl_cap_chk_intermediates *inter);
+
+enum frl_cap_chk_result dml1_frl_cap_chk_common(struct frl_cap_chk_intermediates *inter,
+						struct frl_cap_chk_params *params);
+
+enum frl_cap_chk_result dml1_frl_cap_chk_uncompressed(struct frl_cap_chk_params *params,
+						      struct frl_cap_chk_intermediates *inter);
+
+enum frl_cap_chk_result dml1_frl_cap_chk_compressed(struct frl_cap_chk_params *params,
+						    struct frl_cap_chk_intermediates *inter);
+#endif
+
+void frl_modified_pix_clock_for_dsc_padding(const int hc_active_target,
+	const int hc_blank_target,
+	const uint8_t frl_num_lanes,
+	const uint32_t pix_clk_100hz,
+	const int frl_link_rate,
+	const uint32_t h_addressable,
+	const uint32_t h_border_left,
+	const uint32_t h_border_right,
+	const uint32_t h_total,
+	const uint32_t h_addressable_otg,
+	uint32_t *pix_clk_100hz_otg,
+	uint32_t *h_total_otg);
+
+int frl_modify_borrow_mode_for_dsc_padding(const uint32_t pix_clk_100hz,
+	const uint32_t h_active,
+	const uint32_t h_active_padded,
+	const uint32_t h_blank,
+	const uint32_t h_blank_padded,
+	const int hc_active,
+	const int hc_blank,
+	const uint8_t frl_num_lanes,
+	const int frl_link_rate);

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dml_inline_defs.h

@@ -0,0 +1,160 @@
+/*
+ * Copyright 2017 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 __DML_INLINE_DEFS_H__
+#define __DML_INLINE_DEFS_H__
+
+#include "dcn_calc_math.h"
+#include "dml_logger.h"
+
+static inline double dml_min(double a, double b)
+{
+	return (double)dcn_bw_min2((float)a, (float)b);
+}
+
+static inline double dml_min3(double a, double b, double c)
+{
+	return dml_min(dml_min(a, b), c);
+}
+
+static inline double dml_min4(double a, double b, double c, double d)
+{
+	return dml_min(dml_min(a, b), dml_min(c, d));
+}
+
+static inline double dml_max(double a, double b)
+{
+	return (double)dcn_bw_max2((float)a, (float)b);
+}
+
+static inline double dml_max3(double a, double b, double c)
+{
+	return dml_max(dml_max(a, b), c);
+}
+
+static inline double dml_max4(double a, double b, double c, double d)
+{
+	return dml_max(dml_max(a, b), dml_max(c, d));
+}
+
+static inline double dml_max5(double a, double b, double c, double d, double e)
+{
+	return dml_max(dml_max4(a, b, c, d), e);
+}
+
+static inline double dml_ceil(double a, double granularity)
+{
+	if (granularity == 0)
+		return 0;
+	return (double)dcn_bw_ceil2((float)a, (float)granularity);
+}
+
+static inline double dml_floor(double a, double granularity)
+{
+	if (granularity == 0)
+		return 0;
+	return (double)dcn_bw_floor2((float)a, (float)granularity);
+}
+
+static inline double dml_round(double a)
+{
+	const double round_pt = 0.5;
+
+	return dml_floor(a + round_pt, 1);
+}
+
+/* float
+static inline int dml_log2(float x)
+{
+	unsigned int ix = *((unsigned int *)&x);
+
+	return (int)((ix >> 23) & 0xff) - 127;
+}*/
+
+/* double */
+static inline int dml_log2(double x)
+{
+	unsigned long long ix = *((unsigned long long *)&x);
+
+	return (int)((ix >> 52) & 0x7ff) - 1023;
+}
+
+static inline double dml_pow(double a, int exp)
+{
+	return (double)dcn_bw_pow((float)a, (float)exp);
+}
+
+static inline double dml_fmod(double f, int val)
+{
+	return (double)dcn_bw_mod((float)f, (float)val);
+}
+
+static inline double dml_ceil_2(double f)
+{
+	return (double)dcn_bw_ceil2((float)f, 2.0f);
+}
+
+static inline double dml_ceil_ex(double x, double granularity)
+{
+	if (granularity == 0)
+		return 0;
+	return (double)dcn_bw_ceil2((float)x, (float)granularity);
+}
+
+static inline double dml_floor_ex(double x, double granularity)
+{
+	if (granularity == 0)
+		return 0;
+	return (double)dcn_bw_floor2((float)x, (float)granularity);
+}
+
+static inline unsigned int dml_round_to_multiple(unsigned int num,
+						 unsigned int multiple,
+						 unsigned char up)
+{
+	unsigned int remainder;
+
+	if (multiple == 0)
+		return num;
+
+	remainder = num % multiple;
+
+	if (remainder == 0)
+		return num;
+
+	if (up)
+		return (num + multiple - remainder);
+	else
+		return (num - remainder);
+}
+static inline double dml_abs(double a)
+{
+	if (a > 0)
+		return a;
+	else
+		return (a*(-1));
+}
+
+#endif

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dml_logger.h

@@ -0,0 +1,38 @@
+/*
+ * Copyright 2018 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 __DML_LOGGER_H_
+#define __DML_LOGGER_H_
+
+#define DC_LOGGER \
+	mode_lib->logger
+
+#define dml_print(str, ...) {DC_LOG_DML(str, ##__VA_ARGS__); }
+#define DTRACE(str, ...) {DC_LOG_DML(str, ##__VA_ARGS__); }
+
+#endif
+
+

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dsc/qp_tables.h

@@ -0,0 +1,704 @@
+
+/*
+ * Copyright 2017 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
+ *
+ */
+
+
+static const qp_table   qp_table_422_10bpc_min = {
+	{   6, { 0, 4, 5, 6, 6, 6, 6, 7, 7, 8, 9, 9, 9, 12, 16} },
+	{ 6.5, { 0, 4, 5, 6, 6, 6, 6, 7, 7, 8, 9, 9, 9, 12, 16} },
+	{   7, { 0, 4, 5, 6, 6, 6, 6, 7, 7, 7, 9, 9, 9, 11, 15} },
+	{ 7.5, { 0, 2, 4, 6, 6, 6, 6, 7, 7, 7, 8, 9, 9, 11, 15} },
+	{   8, { 0, 2, 3, 5, 5, 6, 6, 7, 7, 7, 8, 8, 9, 11, 14} },
+	{ 8.5, { 0, 2, 3, 4, 5, 5, 5, 6, 6, 7, 8, 8, 9, 11, 14} },
+	{   9, { 0, 2, 3, 4, 5, 5, 5, 6, 6, 7, 8, 8, 9, 11, 13} },
+	{ 9.5, { 0, 2, 3, 4, 4, 5, 5, 6, 6, 7, 8, 8, 9, 11, 13} },
+	{  10, { 0, 2, 2, 3, 4, 5, 5, 6, 6, 7, 8, 8, 9, 11, 12} },
+	{10.5, { 0, 2, 2, 3, 4, 4, 5, 6, 6, 7, 8, 8, 9, 11, 12} },
+	{  11, { 0, 2, 2, 3, 4, 4, 5, 6, 6, 7, 8, 8, 9, 10, 11} },
+	{11.5, { 0, 2, 2, 3, 4, 4, 5, 5, 6, 7, 7, 8, 8, 10, 11} },
+	{  12, { 0, 2, 2, 3, 4, 4, 5, 5, 6, 6, 7, 7, 8, 9, 10} },
+	{12.5, { 0, 1, 2, 2, 4, 4, 4, 5, 5, 6, 6, 7, 8, 9, 10} },
+	{  13, { 0, 1, 2, 2, 4, 4, 4, 5, 5, 6, 6, 6, 8, 8, 9} },
+	{13.5, { 0, 1, 2, 2, 3, 4, 4, 4, 5, 6, 6, 6, 7, 8, 9} },
+	{  14, { 0, 1, 2, 2, 3, 4, 4, 4, 4, 5, 5, 6, 7, 7, 8} },
+	{14.5, { 0, 1, 1, 2, 3, 3, 4, 4, 4, 5, 5, 6, 6, 7, 8} },
+	{  15, { 0, 1, 1, 2, 2, 3, 3, 4, 4, 4, 5, 6, 6, 6, 8} },
+	{15.5, { 0, 0, 1, 1, 2, 3, 3, 3, 4, 4, 5, 5, 5, 6, 7} },
+	{  16, { 0, 0, 1, 1, 2, 3, 3, 3, 4, 4, 5, 5, 5, 5, 7} },
+	{16.5, { 0, 0, 0, 1, 1, 2, 2, 3, 3, 3, 3, 4, 4, 5, 6} },
+	{  17, { 0, 0, 0, 0, 1, 1, 2, 2, 2, 3, 3, 3, 3, 4, 6} },
+	{17.5, { 0, 0, 0, 0, 1, 1, 1, 2, 2, 2, 2, 3, 3, 4, 5} },
+	{  18, { 0, 0, 0, 0, 0, 0, 1, 1, 1, 2, 2, 2, 3, 3, 5} },
+	{18.5, { 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 2, 2, 2, 3, 5} },
+	{  19, { 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 2, 2, 4} },
+	{19.5, { 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 2, 4} },
+	{  20, { 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 3} }
+};
+
+
+static const qp_table   qp_table_444_8bpc_max = {
+	{   6, { 4, 6, 8, 8, 9, 9, 9, 10, 11, 12, 12, 12, 12, 13, 15} },
+	{ 6.5, { 4, 6, 7, 8, 8, 8, 9, 10, 11, 11, 12, 12, 12, 13, 15} },
+	{   7, { 4, 5, 7, 7, 8, 8, 8, 9, 10, 11, 11, 12, 12, 13, 14} },
+	{ 7.5, { 4, 5, 6, 7, 7, 8, 8, 9, 10, 10, 11, 11, 12, 13, 14} },
+	{   8, { 4, 4, 5, 6, 7, 7, 7, 8, 9, 10, 10, 11, 11, 12, 13} },
+	{ 8.5, { 4, 4, 5, 6, 7, 7, 7, 8, 9, 10, 10, 11, 11, 12, 13} },
+	{   9, { 3, 4, 5, 6, 7, 7, 7, 8, 9, 9, 10, 10, 11, 11, 13} },
+	{ 9.5, { 3, 4, 5, 6, 7, 7, 7, 8, 9, 9, 10, 10, 11, 11, 13} },
+	{  10, { 3, 4, 5, 6, 7, 7, 7, 8, 9, 9, 10, 10, 11, 11, 12} },
+	{10.5, { 3, 4, 5, 6, 7, 7, 7, 8, 9, 9, 10, 10, 10, 11, 12} },
+	{  11, { 2, 4, 5, 6, 7, 7, 7, 8, 8, 9, 9, 9, 10, 10, 11} },
+	{11.5, { 2, 4, 5, 6, 7, 7, 7, 8, 8, 9, 9, 9, 9, 10, 11} },
+	{  12, { 2, 3, 4, 5, 6, 6, 7, 8, 8, 9, 9, 9, 9, 10, 11} },
+	{12.5, { 2, 3, 4, 5, 6, 6, 6, 7, 7, 8, 8, 9, 9, 10, 11} },
+	{  13, { 1, 3, 4, 5, 5, 6, 6, 7, 7, 8, 8, 8, 8, 9, 10} },
+	{13.5, { 1, 2, 3, 4, 4, 5, 5, 6, 6, 7, 7, 8, 8, 9, 10} },
+	{  14, { 1, 2, 2, 3, 4, 4, 4, 5, 6, 6, 7, 8, 8, 8, 10} },
+	{14.5, { 0, 1, 2, 3, 3, 4, 4, 5, 5, 6, 6, 7, 7, 8, 9} },
+	{  15, { 0, 1, 1, 2, 2, 3, 3, 4, 5, 5, 6, 7, 7, 8, 9} },
+	{15.5, { 0, 1, 1, 2, 2, 3, 3, 4, 5, 5, 6, 7, 7, 8, 9} },
+	{  16, { 0, 1, 1, 2, 2, 3, 3, 4, 5, 5, 5, 6, 6, 7, 8} },
+	{16.5, { 0, 1, 1, 2, 2, 3, 3, 4, 5, 5, 5, 6, 6, 7, 8} },
+	{  17, { 0, 1, 1, 2, 2, 3, 3, 3, 4, 4, 4, 5, 5, 6, 8} },
+	{17.5, { 0, 1, 1, 2, 2, 3, 3, 3, 4, 4, 4, 5, 5, 6, 8} },
+	{  18, { 0, 1, 1, 2, 2, 3, 3, 3, 4, 4, 4, 5, 5, 6, 7} },
+	{18.5, { 0, 1, 1, 2, 2, 3, 3, 3, 4, 4, 4, 5, 5, 6, 7} },
+	{  19, { 0, 1, 1, 1, 2, 2, 2, 2, 3, 3, 3, 4, 4, 5, 6} },
+	{19.5, { 0, 1, 1, 1, 2, 2, 2, 2, 3, 3, 3, 4, 4, 5, 6} },
+	{  20, { 0, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 3, 3, 4, 6} },
+	{20.5, { 0, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 3, 3, 4, 6} },
+	{  21, { 0, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 3, 3, 4, 5} },
+	{21.5, { 0, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 3, 3, 4, 5} },
+	{  22, { 0, 0, 0, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 3, 5} },
+	{22.5, { 0, 0, 0, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 3, 4} },
+	{  23, { 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 4} },
+	{23.5, { 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 4} },
+	{  24, { 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 2, 4} }
+};
+
+
+static const qp_table   qp_table_420_12bpc_max = {
+	{   4, {11, 12, 13, 14, 15, 15, 15, 16, 17, 18, 18, 19, 19, 21, 22} },
+	{ 4.5, {10, 11, 12, 13, 14, 15, 15, 16, 17, 18, 18, 19, 19, 20, 21} },
+	{   5, { 9, 11, 12, 13, 14, 15, 15, 16, 17, 17, 18, 18, 19, 20, 21} },
+	{ 5.5, { 8, 10, 11, 12, 13, 14, 15, 16, 16, 17, 17, 18, 18, 19, 20} },
+	{   6, { 6, 9, 11, 12, 13, 14, 15, 16, 16, 17, 17, 17, 17, 18, 19} },
+	{ 6.5, { 6, 8, 10, 11, 11, 13, 14, 15, 15, 16, 16, 17, 17, 18, 19} },
+	{   7, { 5, 7, 9, 10, 10, 12, 13, 14, 14, 15, 16, 16, 17, 17, 18} },
+	{ 7.5, { 5, 7, 8, 9, 9, 11, 12, 13, 14, 14, 15, 15, 16, 16, 17} },
+	{   8, { 4, 6, 7, 8, 8, 10, 11, 12, 13, 13, 14, 15, 15, 16, 17} },
+	{ 8.5, { 3, 6, 6, 7, 7, 10, 11, 12, 13, 13, 14, 14, 15, 15, 16} },
+	{   9, { 3, 5, 6, 7, 7, 10, 11, 12, 12, 13, 13, 14, 14, 14, 15} },
+	{ 9.5, { 2, 5, 6, 6, 7, 9, 10, 11, 12, 12, 13, 13, 13, 14, 15} },
+	{  10, { 2, 4, 5, 6, 7, 8, 9, 10, 11, 12, 12, 13, 13, 13, 15} },
+	{10.5, { 2, 3, 5, 5, 6, 7, 8, 9, 11, 11, 12, 12, 12, 12, 14} },
+	{  11, { 1, 3, 4, 5, 6, 6, 7, 9, 10, 11, 11, 11, 12, 12, 13} },
+	{11.5, { 1, 2, 3, 4, 5, 6, 6, 8, 9, 10, 10, 11, 11, 11, 13} },
+	{  12, { 1, 1, 3, 3, 4, 5, 6, 7, 8, 9, 9, 10, 10, 10, 12} },
+	{12.5, { 1, 1, 2, 3, 4, 5, 6, 7, 8, 8, 9, 9, 9, 10, 11} },
+	{  13, { 1, 1, 1, 2, 4, 4, 6, 6, 7, 8, 8, 9, 9, 9, 11} },
+	{13.5, { 1, 1, 1, 2, 3, 4, 5, 5, 6, 7, 8, 8, 8, 9, 11} },
+	{  14, { 1, 1, 1, 2, 3, 4, 4, 5, 6, 6, 7, 8, 8, 8, 10} },
+	{14.5, { 0, 1, 1, 1, 2, 3, 4, 4, 5, 5, 6, 7, 7, 7, 9} },
+	{  15, { 0, 1, 1, 1, 1, 2, 3, 3, 5, 5, 5, 6, 6, 7, 9} },
+	{15.5, { 0, 0, 0, 1, 1, 2, 2, 3, 3, 4, 4, 5, 5, 6, 8} },
+	{  16, { 0, 0, 0, 0, 1, 1, 2, 2, 3, 3, 4, 4, 5, 5, 7} },
+	{16.5, { 0, 0, 0, 0, 0, 1, 2, 2, 2, 3, 3, 4, 4, 5, 7} },
+	{  17, { 0, 0, 0, 0, 0, 0, 1, 2, 2, 2, 3, 3, 3, 4, 6} },
+	{17.5, { 0, 0, 0, 0, 0, 0, 1, 1, 1, 2, 2, 3, 3, 4, 6} },
+	{  18, { 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 2, 2, 2, 3, 5} }
+};
+
+
+static const qp_table   qp_table_444_10bpc_min = {
+	{   6, { 0, 4, 7, 7, 9, 9, 9, 9, 9, 10, 10, 10, 10, 12, 18} },
+	{ 6.5, { 0, 4, 6, 7, 8, 8, 9, 9, 9, 9, 10, 10, 10, 12, 18} },
+	{   7, { 0, 4, 6, 6, 8, 8, 8, 8, 8, 9, 9, 10, 10, 12, 17} },
+	{ 7.5, { 0, 4, 6, 6, 7, 8, 8, 8, 8, 8, 9, 9, 10, 12, 17} },
+	{   8, { 0, 4, 5, 5, 7, 7, 7, 7, 7, 8, 9, 9, 9, 12, 16} },
+	{ 8.5, { 0, 4, 5, 5, 7, 7, 7, 7, 7, 8, 9, 9, 9, 12, 16} },
+	{   9, { 0, 4, 5, 5, 7, 7, 7, 7, 7, 7, 9, 9, 9, 12, 16} },
+	{ 9.5, { 0, 4, 5, 5, 7, 7, 7, 7, 7, 7, 9, 9, 9, 12, 16} },
+	{  10, { 0, 4, 5, 5, 7, 7, 7, 7, 7, 7, 9, 9, 9, 12, 15} },
+	{10.5, { 0, 4, 5, 5, 7, 7, 7, 7, 7, 7, 9, 9, 9, 12, 15} },
+	{  11, { 0, 3, 5, 5, 7, 7, 7, 7, 7, 7, 9, 9, 9, 11, 14} },
+	{11.5, { 0, 3, 5, 5, 7, 7, 7, 7, 7, 7, 9, 9, 9, 11, 14} },
+	{  12, { 0, 2, 4, 4, 6, 6, 7, 7, 7, 7, 9, 9, 9, 11, 14} },
+	{12.5, { 0, 2, 4, 4, 6, 6, 7, 7, 7, 7, 8, 9, 9, 11, 14} },
+	{  13, { 0, 2, 4, 4, 5, 6, 7, 7, 7, 7, 8, 9, 9, 11, 13} },
+	{13.5, { 0, 2, 3, 4, 5, 6, 6, 7, 7, 7, 8, 9, 9, 11, 13} },
+	{  14, { 0, 2, 3, 4, 5, 5, 6, 6, 7, 7, 8, 9, 9, 11, 13} },
+	{14.5, { 0, 2, 3, 4, 5, 5, 6, 6, 6, 7, 7, 8, 9, 11, 12} },
+	{  15, { 0, 2, 3, 4, 4, 5, 5, 6, 6, 6, 7, 8, 9, 11, 12} },
+	{15.5, { 0, 2, 3, 4, 4, 5, 5, 6, 6, 6, 7, 8, 9, 11, 12} },
+	{  16, { 0, 2, 3, 4, 4, 5, 5, 6, 6, 6, 7, 8, 8, 10, 11} },
+	{16.5, { 0, 1, 2, 3, 4, 5, 5, 6, 6, 6, 7, 8, 8, 10, 11} },
+	{  17, { 0, 1, 2, 3, 3, 4, 5, 5, 5, 6, 6, 7, 8, 9, 11} },
+	{17.5, { 0, 1, 2, 3, 3, 4, 4, 5, 5, 6, 6, 7, 8, 9, 11} },
+	{  18, { 0, 1, 2, 3, 3, 4, 4, 5, 5, 6, 6, 7, 8, 9, 10} },
+	{18.5, { 0, 1, 2, 3, 3, 4, 4, 5, 5, 6, 6, 7, 8, 9, 10} },
+	{  19, { 0, 1, 1, 2, 3, 3, 3, 4, 5, 6, 6, 7, 7, 8, 9} },
+	{19.5, { 0, 1, 1, 2, 3, 3, 3, 4, 5, 6, 6, 7, 7, 8, 9} },
+	{  20, { 0, 1, 1, 2, 2, 3, 3, 4, 5, 5, 5, 6, 6, 7, 9} },
+	{20.5, { 0, 1, 1, 2, 2, 3, 3, 4, 4, 5, 5, 6, 6, 7, 9} },
+	{  21, { 0, 1, 1, 2, 2, 3, 3, 4, 4, 4, 5, 6, 6, 7, 9} },
+	{21.5, { 0, 1, 1, 2, 2, 2, 3, 4, 4, 4, 5, 6, 6, 7, 8} },
+	{  22, { 0, 0, 1, 2, 2, 2, 3, 3, 4, 4, 5, 5, 5, 6, 8} },
+	{22.5, { 0, 0, 1, 2, 2, 2, 3, 3, 4, 4, 5, 5, 5, 6, 7} },
+	{  23, { 0, 0, 1, 2, 2, 2, 3, 3, 3, 3, 5, 5, 5, 5, 7} },
+	{23.5, { 0, 0, 0, 2, 2, 2, 3, 3, 3, 3, 5, 5, 5, 5, 7} },
+	{  24, { 0, 0, 0, 1, 1, 2, 3, 3, 3, 3, 4, 4, 4, 5, 7} },
+	{24.5, { 0, 0, 0, 1, 1, 2, 2, 3, 3, 3, 4, 4, 4, 5, 7} },
+	{  25, { 0, 0, 0, 0, 1, 2, 2, 2, 3, 3, 4, 4, 4, 4, 6} },
+	{25.5, { 0, 0, 0, 0, 1, 2, 2, 2, 3, 3, 3, 4, 4, 4, 6} },
+	{  26, { 0, 0, 0, 0, 0, 1, 1, 2, 2, 3, 3, 3, 3, 3, 5} },
+	{26.5, { 0, 0, 0, 0, 0, 1, 1, 2, 2, 2, 3, 3, 3, 3, 5} },
+	{  27, { 0, 0, 0, 0, 0, 1, 1, 1, 2, 2, 2, 3, 3, 3, 5} },
+	{27.5, { 0, 0, 0, 0, 0, 0, 1, 1, 1, 2, 2, 3, 3, 3, 5} },
+	{  28, { 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 2, 2, 2, 2, 4} },
+	{28.5, { 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 2, 2, 2, 4} },
+	{  29, { 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 2, 2, 3} },
+	{29.5, { 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 2, 2, 3} },
+	{  30, { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 3} }
+};
+
+
+static const qp_table   qp_table_420_8bpc_max = {
+	{   4, { 4, 4, 5, 6, 7, 7, 7, 8, 9, 10, 10, 11, 11, 13, 14} },
+	{ 4.5, { 4, 4, 5, 6, 7, 7, 7, 8, 9, 10, 10, 11, 11, 12, 13} },
+	{   5, { 3, 4, 5, 6, 7, 7, 7, 8, 9, 9, 10, 10, 11, 12, 13} },
+	{ 5.5, { 3, 4, 5, 6, 7, 7, 7, 8, 8, 9, 9, 10, 10, 11, 12} },
+	{   6, { 2, 4, 5, 6, 7, 7, 7, 8, 8, 9, 9, 9, 9, 10, 11} },
+	{ 6.5, { 2, 3, 4, 5, 5, 6, 6, 7, 7, 8, 8, 9, 9, 10, 11} },
+	{   7, { 1, 2, 3, 4, 4, 5, 5, 6, 6, 7, 8, 8, 9, 9, 10} },
+	{ 7.5, { 1, 2, 2, 3, 3, 4, 4, 5, 6, 6, 7, 7, 8, 8, 9} },
+	{   8, { 0, 1, 1, 2, 2, 3, 3, 4, 5, 5, 6, 7, 7, 8, 9} },
+	{ 8.5, { 0, 1, 1, 2, 2, 3, 3, 4, 5, 5, 6, 6, 7, 7, 8} },
+	{   9, { 0, 1, 1, 2, 2, 3, 3, 4, 4, 5, 5, 6, 6, 6, 7} },
+	{ 9.5, { 0, 1, 1, 1, 2, 2, 3, 3, 4, 4, 5, 5, 5, 6, 7} },
+	{  10, { 0, 0, 1, 1, 2, 2, 2, 3, 3, 4, 4, 5, 5, 5, 6} },
+	{10.5, { 0, 0, 1, 1, 1, 2, 2, 2, 3, 3, 4, 4, 4, 4, 6} },
+	{  11, { 0, 0, 0, 1, 1, 1, 2, 2, 2, 3, 3, 3, 4, 4, 5} },
+	{11.5, { 0, 0, 0, 0, 1, 1, 1, 2, 2, 2, 2, 3, 3, 4, 5} },
+	{  12, { 0, 0, 0, 0, 0, 0, 1, 1, 1, 2, 2, 2, 2, 3, 4} }
+};
+
+
+static const qp_table   qp_table_444_8bpc_min = {
+	{   6, { 0, 1, 3, 3, 5, 5, 5, 5, 5, 6, 6, 6, 6, 9, 14} },
+	{ 6.5, { 0, 1, 2, 3, 4, 4, 5, 5, 5, 5, 6, 6, 6, 9, 14} },
+	{   7, { 0, 0, 2, 2, 4, 4, 4, 4, 4, 5, 5, 6, 6, 9, 13} },
+	{ 7.5, { 0, 0, 2, 2, 3, 4, 4, 4, 4, 4, 5, 5, 6, 9, 13} },
+	{   8, { 0, 0, 1, 1, 3, 3, 3, 3, 3, 4, 5, 5, 5, 8, 12} },
+	{ 8.5, { 0, 0, 1, 1, 3, 3, 3, 3, 3, 4, 5, 5, 5, 8, 12} },
+	{   9, { 0, 0, 1, 1, 3, 3, 3, 3, 3, 3, 5, 5, 5, 7, 12} },
+	{ 9.5, { 0, 0, 1, 1, 3, 3, 3, 3, 3, 3, 5, 5, 5, 7, 12} },
+	{  10, { 0, 0, 1, 1, 3, 3, 3, 3, 3, 3, 5, 5, 5, 7, 11} },
+	{10.5, { 0, 0, 1, 1, 3, 3, 3, 3, 3, 3, 5, 5, 5, 7, 11} },
+	{  11, { 0, 0, 1, 1, 3, 3, 3, 3, 3, 3, 5, 5, 5, 7, 10} },
+	{11.5, { 0, 0, 1, 1, 3, 3, 3, 3, 3, 3, 5, 5, 5, 7, 10} },
+	{  12, { 0, 0, 1, 1, 3, 3, 3, 3, 3, 3, 5, 5, 5, 7, 10} },
+	{12.5, { 0, 0, 1, 1, 3, 3, 3, 3, 3, 3, 4, 5, 5, 7, 10} },
+	{  13, { 0, 0, 1, 1, 2, 3, 3, 3, 3, 3, 4, 5, 5, 7, 9} },
+	{13.5, { 0, 0, 0, 1, 1, 2, 2, 2, 2, 3, 4, 5, 5, 7, 9} },
+	{  14, { 0, 0, 0, 0, 1, 1, 2, 2, 2, 3, 4, 5, 5, 7, 9} },
+	{14.5, { 0, 0, 0, 0, 1, 1, 2, 2, 2, 3, 4, 4, 5, 7, 8} },
+	{  15, { 0, 0, 0, 0, 1, 1, 1, 2, 2, 3, 4, 4, 5, 7, 8} },
+	{15.5, { 0, 0, 0, 0, 1, 1, 1, 2, 2, 3, 4, 4, 5, 7, 8} },
+	{  16, { 0, 0, 0, 0, 1, 1, 1, 2, 2, 3, 4, 4, 5, 6, 7} },
+	{16.5, { 0, 0, 0, 0, 1, 1, 1, 2, 2, 3, 4, 4, 5, 6, 7} },
+	{  17, { 0, 0, 0, 0, 1, 1, 1, 2, 2, 3, 3, 4, 4, 5, 7} },
+	{17.5, { 0, 0, 0, 0, 1, 1, 1, 2, 2, 3, 3, 4, 4, 5, 7} },
+	{  18, { 0, 0, 0, 0, 1, 1, 1, 2, 2, 3, 3, 4, 4, 5, 6} },
+	{18.5, { 0, 0, 0, 0, 1, 1, 1, 2, 2, 3, 3, 4, 4, 5, 6} },
+	{  19, { 0, 0, 0, 0, 1, 1, 1, 1, 2, 2, 2, 3, 3, 4, 5} },
+	{19.5, { 0, 0, 0, 0, 1, 1, 1, 1, 2, 2, 2, 3, 3, 4, 5} },
+	{  20, { 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 2, 2, 3, 5} },
+	{20.5, { 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 2, 2, 3, 5} },
+	{  21, { 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 2, 2, 3, 4} },
+	{21.5, { 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 2, 2, 3, 4} },
+	{  22, { 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 2, 4} },
+	{22.5, { 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 2, 3} },
+	{  23, { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 3} },
+	{23.5, { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 3} },
+	{  24, { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 3} }
+};
+
+
+static const qp_table   qp_table_444_12bpc_min = {
+	{   6, { 0, 5, 11, 11, 13, 13, 13, 13, 13, 14, 14, 14, 14, 17, 22} },
+	{ 6.5, { 0, 5, 10, 11, 12, 12, 13, 13, 13, 13, 14, 14, 14, 17, 22} },
+	{   7, { 0, 5, 10, 10, 12, 12, 12, 12, 12, 13, 13, 14, 14, 17, 21} },
+	{ 7.5, { 0, 5, 9, 10, 11, 12, 12, 12, 12, 12, 13, 13, 14, 17, 21} },
+	{   8, { 0, 4, 8, 9, 11, 11, 11, 11, 11, 12, 13, 13, 13, 16, 20} },
+	{ 8.5, { 0, 4, 8, 9, 11, 11, 11, 11, 11, 12, 13, 13, 13, 16, 20} },
+	{   9, { 0, 4, 8, 9, 11, 11, 11, 11, 11, 11, 13, 13, 13, 15, 20} },
+	{ 9.5, { 0, 4, 8, 9, 11, 11, 11, 11, 11, 11, 13, 13, 13, 15, 20} },
+	{  10, { 0, 4, 8, 9, 11, 11, 11, 11, 11, 11, 13, 13, 13, 15, 19} },
+	{10.5, { 0, 4, 8, 9, 11, 11, 11, 11, 11, 11, 13, 13, 13, 15, 19} },
+	{  11, { 0, 4, 8, 9, 11, 11, 11, 11, 11, 11, 13, 13, 13, 15, 18} },
+	{11.5, { 0, 4, 8, 9, 11, 11, 11, 11, 11, 11, 13, 13, 13, 15, 18} },
+	{  12, { 0, 4, 7, 8, 10, 11, 11, 11, 11, 11, 13, 13, 13, 15, 18} },
+	{12.5, { 0, 4, 7, 8, 10, 11, 11, 11, 11, 11, 13, 13, 13, 15, 18} },
+	{  13, { 0, 4, 7, 8, 9, 11, 11, 11, 11, 11, 13, 13, 13, 15, 17} },
+	{13.5, { 0, 3, 6, 7, 9, 10, 10, 11, 11, 11, 12, 13, 13, 15, 17} },
+	{  14, { 0, 3, 5, 6, 9, 9, 9, 10, 11, 11, 12, 13, 13, 15, 17} },
+	{14.5, { 0, 2, 5, 6, 8, 9, 9, 10, 11, 11, 12, 13, 13, 15, 16} },
+	{  15, { 0, 2, 4, 6, 7, 8, 9, 10, 11, 11, 12, 13, 13, 15, 16} },
+	{15.5, { 0, 2, 4, 6, 7, 8, 9, 10, 11, 11, 12, 13, 13, 15, 16} },
+	{  16, { 0, 2, 4, 6, 7, 8, 9, 10, 11, 11, 11, 12, 12, 14, 15} },
+	{16.5, { 0, 2, 3, 5, 7, 8, 9, 10, 11, 11, 11, 12, 12, 14, 15} },
+	{  17, { 0, 2, 3, 5, 5, 6, 9, 9, 10, 10, 11, 11, 12, 13, 15} },
+	{17.5, { 0, 2, 3, 5, 5, 6, 8, 9, 10, 10, 11, 11, 12, 13, 15} },
+	{  18, { 0, 2, 3, 5, 5, 6, 8, 9, 10, 10, 11, 11, 12, 13, 14} },
+	{18.5, { 0, 2, 3, 5, 5, 6, 8, 9, 10, 10, 11, 11, 12, 13, 14} },
+	{  19, { 0, 1, 2, 4, 5, 5, 7, 8, 9, 9, 10, 11, 11, 12, 13} },
+	{19.5, { 0, 1, 2, 4, 5, 5, 7, 8, 9, 9, 10, 11, 11, 12, 13} },
+	{  20, { 0, 1, 2, 3, 4, 5, 7, 8, 8, 8, 9, 10, 10, 11, 13} },
+	{20.5, { 0, 1, 2, 3, 4, 5, 6, 7, 7, 8, 9, 10, 10, 11, 13} },
+	{  21, { 0, 1, 2, 3, 4, 5, 5, 7, 7, 8, 9, 10, 10, 11, 13} },
+	{21.5, { 0, 1, 2, 3, 3, 4, 5, 7, 7, 8, 9, 10, 10, 11, 12} },
+	{  22, { 0, 0, 1, 3, 3, 4, 5, 6, 7, 8, 9, 9, 9, 10, 12} },
+	{22.5, { 0, 0, 1, 3, 3, 4, 5, 6, 7, 8, 9, 9, 9, 10, 11} },
+	{  23, { 0, 0, 1, 3, 3, 4, 5, 6, 6, 7, 9, 9, 9, 9, 11} },
+	{23.5, { 0, 0, 1, 3, 3, 4, 5, 6, 6, 7, 9, 9, 9, 9, 11} },
+	{  24, { 0, 0, 1, 2, 3, 4, 5, 6, 6, 7, 8, 8, 8, 9, 11} },
+	{24.5, { 0, 0, 1, 2, 3, 4, 4, 6, 6, 7, 8, 8, 8, 9, 11} },
+	{  25, { 0, 0, 1, 2, 3, 4, 4, 5, 6, 7, 8, 8, 8, 8, 10} },
+	{25.5, { 0, 0, 1, 2, 3, 4, 4, 5, 6, 6, 7, 8, 8, 8, 10} },
+	{  26, { 0, 0, 1, 2, 2, 3, 4, 5, 5, 6, 7, 7, 7, 7, 9} },
+	{26.5, { 0, 0, 1, 2, 2, 3, 4, 5, 5, 5, 7, 7, 7, 7, 9} },
+	{  27, { 0, 0, 1, 2, 2, 3, 4, 4, 5, 5, 6, 7, 7, 7, 9} },
+	{27.5, { 0, 0, 1, 1, 2, 2, 4, 4, 4, 5, 6, 7, 7, 7, 9} },
+	{  28, { 0, 0, 0, 1, 1, 2, 3, 4, 4, 4, 6, 6, 6, 7, 9} },
+	{28.5, { 0, 0, 0, 1, 1, 2, 3, 3, 4, 4, 5, 6, 6, 6, 8} },
+	{  29, { 0, 0, 0, 1, 1, 2, 3, 3, 4, 4, 5, 5, 6, 6, 8} },
+	{29.5, { 0, 0, 0, 1, 1, 1, 2, 2, 3, 4, 5, 5, 6, 6, 7} },
+	{  30, { 0, 0, 0, 1, 1, 1, 2, 2, 3, 3, 5, 5, 5, 5, 7} },
+	{30.5, { 0, 0, 0, 0, 0, 1, 1, 2, 2, 3, 4, 4, 4, 5, 7} },
+	{  31, { 0, 0, 0, 0, 0, 1, 1, 2, 2, 3, 4, 4, 4, 5, 7} },
+	{31.5, { 0, 0, 0, 0, 0, 1, 1, 2, 2, 3, 3, 4, 4, 5, 7} },
+	{  32, { 0, 0, 0, 0, 0, 1, 1, 2, 2, 3, 3, 3, 3, 4, 6} },
+	{32.5, { 0, 0, 0, 0, 0, 0, 1, 1, 1, 2, 3, 3, 3, 4, 6} },
+	{  33, { 0, 0, 0, 0, 0, 0, 1, 1, 1, 2, 2, 3, 3, 3, 5} },
+	{33.5, { 0, 0, 0, 0, 0, 0, 1, 1, 1, 2, 2, 3, 3, 3, 5} },
+	{  34, { 0, 0, 0, 0, 0, 0, 0, 1, 1, 2, 2, 2, 2, 3, 5} },
+	{34.5, { 0, 0, 0, 0, 0, 0, 0, 0, 1, 2, 2, 2, 2, 3, 5} },
+	{  35, { 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 2, 4} },
+	{35.5, { 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 2, 4} },
+	{  36, { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 3} }
+};
+
+
+static const qp_table   qp_table_420_12bpc_min = {
+	{   4, { 0, 4, 9, 10, 11, 11, 11, 11, 11, 11, 13, 13, 13, 15, 21} },
+	{ 4.5, { 0, 4, 8, 9, 10, 11, 11, 11, 11, 11, 13, 13, 13, 15, 20} },
+	{   5, { 0, 4, 8, 9, 10, 11, 11, 11, 11, 11, 13, 13, 13, 15, 20} },
+	{ 5.5, { 0, 4, 7, 8, 10, 11, 11, 11, 11, 11, 13, 13, 13, 15, 19} },
+	{   6, { 0, 4, 7, 8, 10, 11, 11, 11, 11, 11, 13, 13, 13, 15, 18} },
+	{ 6.5, { 0, 4, 6, 8, 9, 10, 11, 11, 11, 11, 13, 13, 13, 15, 18} },
+	{   7, { 0, 3, 5, 7, 9, 10, 10, 11, 11, 11, 13, 13, 13, 15, 17} },
+	{ 7.5, { 0, 3, 5, 7, 8, 9, 10, 10, 11, 11, 12, 13, 13, 15, 16} },
+	{   8, { 0, 2, 4, 6, 7, 9, 9, 10, 11, 11, 12, 13, 13, 15, 16} },
+	{ 8.5, { 0, 2, 4, 6, 6, 9, 9, 10, 11, 11, 12, 12, 13, 14, 15} },
+	{   9, { 0, 2, 4, 6, 6, 9, 9, 10, 10, 11, 11, 12, 13, 13, 14} },
+	{ 9.5, { 0, 2, 4, 5, 6, 8, 8, 9, 10, 10, 11, 12, 12, 13, 14} },
+	{  10, { 0, 2, 3, 5, 6, 7, 8, 8, 9, 10, 10, 12, 12, 12, 14} },
+	{10.5, { 0, 2, 3, 4, 5, 6, 7, 8, 9, 9, 10, 11, 11, 11, 13} },
+	{  11, { 0, 2, 3, 4, 5, 5, 6, 8, 8, 9, 9, 10, 11, 11, 12} },
+	{11.5, { 0, 1, 2, 3, 4, 5, 5, 7, 8, 8, 9, 10, 10, 10, 12} },
+	{  12, { 0, 0, 2, 2, 3, 4, 5, 6, 7, 8, 8, 9, 9, 9, 11} },
+	{12.5, { 0, 0, 1, 2, 3, 4, 5, 6, 7, 7, 8, 8, 8, 9, 10} },
+	{  13, { 0, 0, 0, 1, 3, 3, 5, 5, 6, 7, 7, 8, 8, 8, 10} },
+	{13.5, { 0, 0, 0, 1, 2, 3, 4, 4, 5, 6, 7, 7, 7, 8, 10} },
+	{  14, { 0, 0, 0, 1, 2, 3, 3, 4, 5, 5, 6, 7, 7, 7, 9} },
+	{14.5, { 0, 0, 0, 0, 1, 2, 3, 3, 4, 4, 5, 6, 6, 6, 8} },
+	{  15, { 0, 0, 0, 0, 0, 1, 2, 2, 4, 4, 4, 5, 5, 6, 8} },
+	{15.5, { 0, 0, 0, 0, 0, 1, 1, 2, 2, 3, 3, 4, 4, 5, 7} },
+	{  16, { 0, 0, 0, 0, 0, 0, 1, 1, 2, 2, 3, 3, 4, 4, 6} },
+	{16.5, { 0, 0, 0, 0, 0, 0, 1, 1, 1, 2, 2, 3, 3, 4, 6} },
+	{  17, { 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 2, 2, 2, 3, 5} },
+	{17.5, { 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 2, 2, 3, 5} },
+	{  18, { 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 2, 4} }
+};
+
+
+static const qp_table   qp_table_422_12bpc_min = {
+	{   6, { 0, 4, 9, 10, 11, 11, 11, 11, 11, 11, 13, 13, 13, 16, 20} },
+	{ 6.5, { 0, 4, 9, 10, 11, 11, 11, 11, 11, 11, 13, 13, 13, 16, 20} },
+	{   7, { 0, 4, 9, 10, 11, 11, 11, 11, 11, 11, 13, 13, 13, 15, 19} },
+	{ 7.5, { 0, 4, 8, 10, 11, 11, 11, 11, 11, 11, 13, 13, 13, 15, 19} },
+	{   8, { 0, 4, 7, 8, 10, 11, 11, 11, 11, 11, 13, 13, 13, 15, 18} },
+	{ 8.5, { 0, 3, 6, 8, 9, 10, 10, 11, 11, 11, 12, 13, 13, 15, 18} },
+	{   9, { 0, 3, 5, 8, 9, 10, 10, 10, 11, 11, 12, 13, 13, 15, 17} },
+	{ 9.5, { 0, 3, 5, 7, 8, 9, 10, 10, 11, 11, 12, 13, 13, 15, 17} },
+	{  10, { 0, 2, 4, 6, 7, 9, 9, 10, 11, 11, 12, 13, 13, 15, 16} },
+	{10.5, { 0, 2, 4, 6, 7, 8, 9, 10, 11, 11, 12, 13, 13, 15, 16} },
+	{  11, { 0, 2, 4, 6, 7, 8, 9, 10, 11, 11, 12, 12, 13, 14, 15} },
+	{11.5, { 0, 2, 4, 6, 7, 7, 9, 9, 10, 11, 11, 12, 12, 14, 15} },
+	{  12, { 0, 2, 4, 6, 6, 6, 8, 8, 9, 9, 11, 11, 12, 13, 14} },
+	{12.5, { 0, 1, 4, 5, 6, 6, 7, 8, 8, 9, 10, 11, 11, 13, 14} },
+	{  13, { 0, 1, 3, 4, 5, 5, 7, 8, 8, 9, 10, 10, 11, 12, 13} },
+	{13.5, { 0, 1, 3, 3, 4, 5, 7, 7, 8, 8, 10, 10, 10, 12, 13} },
+	{  14, { 0, 0, 2, 3, 4, 5, 6, 6, 7, 7, 9, 10, 10, 11, 12} },
+	{14.5, { 0, 0, 1, 3, 4, 4, 6, 6, 6, 7, 9, 9, 9, 11, 12} },
+	{  15, { 0, 0, 1, 3, 3, 4, 5, 6, 6, 6, 8, 9, 9, 10, 12} },
+	{15.5, { 0, 0, 1, 2, 3, 4, 5, 5, 6, 6, 8, 8, 8, 10, 11} },
+	{  16, { 0, 0, 1, 2, 3, 4, 5, 5, 6, 6, 8, 8, 8, 9, 11} },
+	{16.5, { 0, 0, 0, 2, 2, 3, 4, 5, 5, 5, 6, 7, 7, 9, 10} },
+	{  17, { 0, 0, 0, 1, 2, 2, 4, 4, 4, 5, 6, 6, 6, 8, 10} },
+	{17.5, { 0, 0, 0, 1, 2, 2, 3, 4, 4, 4, 5, 6, 6, 8, 9} },
+	{  18, { 0, 0, 0, 1, 2, 2, 3, 3, 3, 4, 5, 5, 6, 7, 9} },
+	{18.5, { 0, 0, 0, 1, 2, 2, 3, 3, 3, 3, 5, 5, 5, 7, 9} },
+	{  19, { 0, 0, 0, 1, 2, 2, 2, 3, 3, 3, 4, 4, 5, 6, 8} },
+	{19.5, { 0, 0, 0, 1, 1, 1, 2, 3, 3, 3, 4, 4, 4, 6, 8} },
+	{  20, { 0, 0, 0, 1, 1, 1, 2, 3, 3, 3, 4, 4, 4, 5, 7} },
+	{20.5, { 0, 0, 0, 0, 1, 1, 2, 2, 3, 3, 4, 4, 4, 5, 7} },
+	{  21, { 0, 0, 0, 0, 0, 1, 2, 2, 3, 3, 3, 4, 4, 4, 6} },
+	{21.5, { 0, 0, 0, 0, 0, 0, 1, 2, 2, 2, 3, 3, 3, 4, 6} },
+	{  22, { 0, 0, 0, 0, 0, 0, 1, 1, 2, 2, 2, 3, 3, 4, 6} },
+	{22.5, { 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 2, 2, 2, 3, 5} },
+	{  23, { 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 2, 2, 3, 5} },
+	{23.5, { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 2, 4} },
+	{  24, { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 2, 4} }
+};
+
+
+static const qp_table   qp_table_422_12bpc_max = {
+	{   6, {12, 12, 13, 14, 15, 15, 15, 16, 17, 18, 18, 19, 19, 20, 21} },
+	{ 6.5, {12, 12, 13, 14, 15, 15, 15, 16, 17, 18, 18, 19, 19, 20, 21} },
+	{   7, {11, 12, 13, 14, 15, 15, 15, 16, 17, 17, 18, 18, 19, 19, 20} },
+	{ 7.5, { 9, 10, 12, 14, 15, 15, 15, 16, 16, 17, 17, 18, 18, 19, 20} },
+	{   8, { 6, 9, 10, 12, 14, 15, 15, 16, 16, 17, 17, 17, 17, 18, 19} },
+	{ 8.5, { 6, 8, 9, 11, 13, 14, 14, 15, 15, 16, 16, 17, 17, 18, 19} },
+	{   9, { 5, 7, 8, 10, 12, 13, 13, 14, 14, 15, 15, 16, 16, 17, 18} },
+	{ 9.5, { 5, 7, 7, 9, 10, 12, 12, 13, 14, 14, 15, 15, 16, 17, 18} },
+	{  10, { 4, 6, 6, 8, 9, 11, 11, 12, 13, 13, 14, 15, 15, 16, 17} },
+	{10.5, { 4, 6, 6, 8, 9, 10, 11, 12, 13, 13, 14, 15, 15, 16, 17} },
+	{  11, { 4, 5, 6, 8, 9, 10, 11, 12, 13, 13, 14, 14, 15, 15, 16} },
+	{11.5, { 3, 5, 6, 8, 9, 9, 11, 11, 12, 13, 13, 14, 14, 15, 16} },
+	{  12, { 3, 5, 6, 8, 8, 8, 10, 10, 11, 11, 13, 13, 14, 14, 15} },
+	{12.5, { 3, 4, 6, 7, 8, 8, 9, 10, 10, 11, 12, 13, 13, 14, 15} },
+	{  13, { 2, 4, 5, 6, 7, 7, 9, 10, 10, 11, 12, 12, 13, 13, 14} },
+	{13.5, { 2, 4, 5, 5, 6, 7, 9, 9, 10, 10, 12, 12, 12, 13, 14} },
+	{  14, { 2, 3, 4, 5, 6, 7, 8, 8, 9, 9, 11, 12, 12, 12, 13} },
+	{14.5, { 2, 3, 3, 5, 6, 6, 8, 8, 8, 9, 11, 11, 11, 12, 13} },
+	{  15, { 2, 3, 3, 5, 5, 6, 7, 8, 8, 8, 10, 11, 11, 11, 13} },
+	{15.5, { 2, 2, 3, 4, 5, 6, 7, 7, 8, 8, 10, 10, 10, 11, 12} },
+	{  16, { 2, 2, 3, 4, 5, 6, 7, 7, 8, 8, 10, 10, 10, 10, 12} },
+	{16.5, { 1, 2, 2, 4, 4, 5, 6, 7, 7, 7, 8, 9, 9, 10, 11} },
+	{  17, { 1, 1, 2, 3, 4, 4, 6, 6, 6, 7, 8, 8, 8, 9, 11} },
+	{17.5, { 1, 1, 2, 3, 4, 4, 5, 6, 6, 6, 7, 8, 8, 9, 10} },
+	{  18, { 1, 1, 1, 2, 3, 3, 5, 5, 5, 6, 7, 7, 8, 8, 10} },
+	{18.5, { 1, 1, 1, 2, 3, 3, 5, 5, 5, 5, 7, 7, 7, 8, 10} },
+	{  19, { 1, 1, 1, 2, 3, 3, 4, 5, 5, 5, 6, 6, 7, 7, 9} },
+	{19.5, { 1, 1, 1, 2, 2, 2, 4, 5, 5, 5, 6, 6, 6, 7, 9} },
+	{  20, { 1, 1, 1, 2, 2, 2, 4, 5, 5, 5, 6, 6, 6, 6, 8} },
+	{20.5, { 0, 0, 0, 1, 2, 2, 3, 3, 4, 4, 5, 5, 5, 6, 8} },
+	{  21, { 0, 0, 0, 1, 1, 2, 3, 3, 4, 4, 4, 5, 5, 5, 7} },
+	{21.5, { 0, 0, 0, 0, 1, 1, 2, 3, 3, 3, 4, 4, 4, 5, 7} },
+	{  22, { 0, 0, 0, 0, 1, 1, 2, 2, 3, 3, 3, 4, 4, 5, 7} },
+	{22.5, { 0, 0, 0, 0, 1, 1, 1, 2, 2, 2, 3, 3, 3, 4, 6} },
+	{  23, { 0, 0, 0, 0, 0, 1, 1, 1, 1, 2, 2, 3, 3, 4, 6} },
+	{23.5, { 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 2, 2, 2, 3, 5} },
+	{  24, { 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 2, 2, 2, 3, 5} }
+};
+
+
+static const qp_table   qp_table_444_12bpc_max = {
+	{   6, {12, 14, 16, 16, 17, 17, 17, 18, 19, 20, 20, 20, 20, 21, 23} },
+	{ 6.5, {12, 14, 15, 16, 16, 16, 17, 18, 19, 19, 20, 20, 20, 21, 23} },
+	{   7, {12, 13, 15, 15, 16, 16, 16, 17, 18, 19, 19, 20, 20, 21, 22} },
+	{ 7.5, {12, 13, 14, 15, 15, 16, 16, 17, 18, 18, 19, 19, 20, 21, 22} },
+	{   8, {12, 12, 13, 14, 15, 15, 15, 16, 17, 18, 18, 19, 19, 20, 21} },
+	{ 8.5, {12, 12, 13, 14, 15, 15, 15, 16, 17, 18, 18, 19, 19, 20, 21} },
+	{   9, {11, 12, 13, 14, 15, 15, 15, 16, 17, 17, 18, 18, 19, 19, 21} },
+	{ 9.5, {11, 12, 13, 14, 15, 15, 15, 16, 17, 17, 18, 18, 19, 19, 21} },
+	{  10, {11, 12, 13, 14, 15, 15, 15, 16, 17, 17, 18, 18, 19, 19, 20} },
+	{10.5, {10, 12, 13, 14, 15, 15, 15, 16, 17, 17, 18, 18, 18, 19, 20} },
+	{  11, { 9, 11, 13, 14, 15, 15, 15, 16, 16, 17, 17, 17, 18, 18, 19} },
+	{11.5, { 9, 11, 13, 14, 15, 15, 15, 16, 16, 17, 17, 17, 17, 18, 19} },
+	{  12, { 6, 9, 12, 13, 14, 14, 15, 16, 16, 17, 17, 17, 17, 18, 19} },
+	{12.5, { 6, 9, 12, 13, 14, 14, 14, 15, 15, 16, 16, 17, 17, 18, 19} },
+	{  13, { 5, 9, 12, 13, 13, 14, 14, 15, 15, 16, 16, 16, 16, 17, 18} },
+	{13.5, { 5, 8, 11, 12, 12, 13, 13, 14, 14, 15, 15, 16, 16, 17, 18} },
+	{  14, { 5, 8, 10, 11, 12, 12, 12, 13, 14, 14, 15, 16, 16, 16, 18} },
+	{14.5, { 4, 7, 10, 11, 11, 12, 12, 13, 13, 14, 14, 15, 15, 16, 17} },
+	{  15, { 4, 7, 9, 10, 10, 11, 11, 12, 13, 13, 14, 15, 15, 16, 17} },
+	{15.5, { 4, 7, 9, 10, 10, 11, 11, 12, 13, 13, 14, 15, 15, 16, 17} },
+	{  16, { 4, 7, 9, 10, 10, 11, 11, 12, 13, 13, 13, 14, 14, 15, 16} },
+	{16.5, { 4, 5, 7, 8, 10, 11, 11, 12, 13, 13, 13, 14, 14, 15, 16} },
+	{  17, { 4, 5, 7, 8, 8, 9, 11, 11, 12, 12, 12, 13, 13, 14, 16} },
+	{17.5, { 3, 5, 7, 8, 8, 9, 10, 11, 12, 12, 12, 13, 13, 14, 16} },
+	{  18, { 3, 5, 7, 8, 8, 9, 10, 11, 12, 12, 12, 13, 13, 14, 15} },
+	{18.5, { 3, 5, 7, 8, 8, 9, 10, 11, 12, 12, 12, 13, 13, 14, 15} },
+	{  19, { 3, 4, 6, 7, 8, 8, 9, 10, 11, 11, 11, 12, 12, 13, 14} },
+	{19.5, { 3, 4, 6, 7, 8, 8, 9, 10, 11, 11, 11, 12, 12, 13, 14} },
+	{  20, { 2, 4, 5, 6, 7, 8, 9, 10, 10, 10, 10, 11, 11, 12, 14} },
+	{20.5, { 2, 3, 5, 5, 7, 8, 8, 8, 9, 10, 10, 11, 11, 12, 14} },
+	{  21, { 2, 3, 5, 5, 7, 7, 7, 8, 8, 9, 10, 11, 11, 12, 14} },
+	{21.5, { 2, 3, 5, 5, 6, 6, 7, 8, 8, 9, 10, 11, 11, 12, 13} },
+	{  22, { 2, 2, 4, 5, 6, 6, 7, 7, 8, 9, 10, 10, 10, 11, 13} },
+	{22.5, { 2, 2, 4, 5, 5, 6, 7, 7, 8, 9, 10, 10, 10, 11, 12} },
+	{  23, { 2, 2, 4, 5, 5, 6, 7, 7, 7, 8, 10, 10, 10, 10, 12} },
+	{23.5, { 2, 2, 3, 5, 5, 6, 7, 7, 7, 8, 10, 10, 10, 10, 12} },
+	{  24, { 2, 2, 3, 4, 4, 5, 7, 7, 7, 8, 9, 9, 9, 10, 12} },
+	{24.5, { 1, 2, 3, 4, 4, 5, 6, 7, 7, 8, 9, 9, 9, 10, 12} },
+	{  25, { 1, 2, 3, 3, 4, 5, 6, 6, 7, 8, 9, 9, 9, 9, 11} },
+	{25.5, { 1, 1, 3, 3, 4, 5, 6, 6, 7, 7, 8, 9, 9, 9, 11} },
+	{  26, { 1, 1, 3, 3, 3, 4, 5, 6, 6, 7, 8, 8, 8, 8, 10} },
+	{26.5, { 1, 1, 2, 3, 3, 4, 5, 6, 6, 6, 8, 8, 8, 8, 10} },
+	{  27, { 1, 1, 2, 3, 3, 4, 5, 5, 6, 6, 7, 8, 8, 8, 10} },
+	{27.5, { 1, 1, 2, 2, 3, 3, 5, 5, 5, 6, 7, 8, 8, 8, 10} },
+	{  28, { 0, 1, 1, 2, 2, 3, 4, 5, 5, 5, 7, 7, 7, 8, 10} },
+	{28.5, { 0, 1, 1, 2, 2, 3, 4, 4, 5, 5, 6, 7, 7, 7, 9} },
+	{  29, { 0, 1, 1, 2, 2, 3, 4, 4, 5, 5, 6, 6, 7, 7, 9} },
+	{29.5, { 0, 1, 1, 2, 2, 2, 3, 3, 4, 5, 6, 6, 7, 7, 8} },
+	{  30, { 0, 1, 1, 2, 2, 2, 3, 3, 4, 4, 6, 6, 6, 6, 8} },
+	{30.5, { 0, 0, 0, 1, 1, 2, 2, 3, 3, 4, 5, 5, 5, 6, 8} },
+	{  31, { 0, 0, 0, 1, 1, 2, 2, 3, 3, 4, 5, 5, 5, 6, 8} },
+	{31.5, { 0, 0, 0, 0, 1, 2, 2, 3, 3, 4, 4, 5, 5, 6, 8} },
+	{  32, { 0, 0, 0, 0, 1, 2, 2, 3, 3, 4, 4, 4, 4, 5, 7} },
+	{32.5, { 0, 0, 0, 0, 1, 1, 2, 2, 2, 3, 4, 4, 4, 5, 7} },
+	{  33, { 0, 0, 0, 0, 1, 1, 2, 2, 2, 3, 3, 4, 4, 4, 6} },
+	{33.5, { 0, 0, 0, 0, 1, 1, 2, 2, 2, 3, 3, 4, 4, 4, 6} },
+	{  34, { 0, 0, 0, 0, 1, 1, 1, 2, 2, 3, 3, 3, 3, 4, 6} },
+	{34.5, { 0, 0, 0, 0, 1, 1, 1, 1, 2, 3, 3, 3, 3, 4, 6} },
+	{  35, { 0, 0, 0, 0, 1, 1, 1, 1, 2, 2, 2, 2, 2, 3, 5} },
+	{35.5, { 0, 0, 0, 0, 0, 1, 1, 1, 1, 2, 2, 2, 2, 3, 5} },
+	{  36, { 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 2, 2, 2, 2, 4} }
+};
+
+
+static const qp_table   qp_table_420_8bpc_min = {
+	{   4, { 0, 0, 1, 1, 3, 3, 3, 3, 3, 3, 5, 5, 5, 9, 13} },
+	{ 4.5, { 0, 0, 1, 1, 3, 3, 3, 3, 3, 3, 5, 5, 5, 8, 12} },
+	{   5, { 0, 0, 1, 1, 3, 3, 3, 3, 3, 3, 5, 5, 5, 8, 12} },
+	{ 5.5, { 0, 0, 1, 1, 3, 3, 3, 3, 3, 3, 5, 5, 5, 7, 11} },
+	{   6, { 0, 0, 1, 1, 3, 3, 3, 3, 3, 3, 5, 5, 5, 7, 10} },
+	{ 6.5, { 0, 0, 1, 1, 2, 2, 3, 3, 3, 3, 4, 5, 5, 7, 10} },
+	{   7, { 0, 0, 1, 1, 1, 2, 2, 2, 2, 3, 4, 5, 5, 7, 9} },
+	{ 7.5, { 0, 0, 0, 1, 1, 1, 2, 2, 2, 3, 4, 4, 5, 7, 8} },
+	{   8, { 0, 0, 0, 0, 1, 1, 1, 2, 2, 3, 4, 4, 5, 7, 8} },
+	{ 8.5, { 0, 0, 0, 0, 1, 1, 1, 2, 2, 3, 4, 4, 5, 6, 7} },
+	{   9, { 0, 0, 0, 0, 1, 1, 1, 2, 2, 3, 3, 4, 5, 5, 6} },
+	{ 9.5, { 0, 0, 0, 0, 1, 1, 1, 2, 2, 3, 3, 4, 4, 5, 6} },
+	{  10, { 0, 0, 0, 0, 1, 1, 1, 2, 2, 3, 3, 4, 4, 4, 5} },
+	{10.5, { 0, 0, 0, 0, 0, 1, 1, 1, 2, 2, 3, 3, 3, 3, 5} },
+	{  11, { 0, 0, 0, 0, 0, 0, 1, 1, 1, 2, 2, 2, 3, 3, 4} },
+	{11.5, { 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 2, 2, 3, 4} },
+	{  12, { 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 2, 3} }
+};
+
+
+static const qp_table   qp_table_422_8bpc_min = {
+	{   6, { 0, 0, 1, 2, 3, 3, 3, 3, 3, 3, 5, 5, 5, 8, 12} },
+	{ 6.5, { 0, 0, 1, 2, 3, 3, 3, 3, 3, 3, 5, 5, 5, 8, 12} },
+	{   7, { 0, 0, 1, 2, 3, 3, 3, 3, 3, 3, 5, 5, 5, 7, 11} },
+	{ 7.5, { 0, 0, 1, 2, 3, 3, 3, 3, 3, 3, 5, 5, 5, 7, 11} },
+	{   8, { 0, 0, 1, 2, 3, 3, 3, 3, 3, 3, 5, 5, 5, 7, 10} },
+	{ 8.5, { 0, 0, 1, 2, 2, 2, 2, 3, 3, 3, 4, 5, 5, 7, 10} },
+	{   9, { 0, 0, 0, 1, 2, 2, 2, 2, 2, 3, 4, 5, 5, 7, 9} },
+	{ 9.5, { 0, 0, 0, 1, 1, 1, 1, 2, 2, 3, 4, 4, 5, 7, 9} },
+	{  10, { 0, 0, 0, 0, 1, 1, 1, 2, 2, 3, 4, 4, 5, 7, 8} },
+	{10.5, { 0, 0, 0, 0, 1, 1, 1, 2, 2, 3, 4, 4, 5, 7, 8} },
+	{  11, { 0, 0, 0, 0, 1, 1, 1, 2, 2, 3, 4, 4, 5, 6, 7} },
+	{11.5, { 0, 0, 0, 0, 1, 1, 1, 2, 2, 3, 4, 4, 5, 6, 7} },
+	{  12, { 0, 0, 0, 0, 1, 1, 1, 2, 2, 3, 4, 4, 5, 5, 6} },
+	{12.5, { 0, 0, 0, 0, 1, 1, 1, 2, 2, 3, 3, 4, 4, 5, 6} },
+	{  13, { 0, 0, 0, 0, 1, 1, 1, 2, 2, 3, 3, 3, 4, 4, 5} },
+	{13.5, { 0, 0, 0, 0, 0, 1, 1, 1, 2, 2, 3, 3, 3, 4, 5} },
+	{  14, { 0, 0, 0, 0, 0, 1, 1, 1, 2, 2, 2, 3, 3, 3, 4} },
+	{14.5, { 0, 0, 0, 0, 0, 0, 1, 1, 1, 2, 2, 2, 2, 3, 4} },
+	{  15, { 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 2, 2, 2, 4} },
+	{15.5, { 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 2, 3} },
+	{  16, { 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 2, 3} }
+};
+
+
+static const qp_table   qp_table_422_10bpc_max = {
+	{   6, { 8, 8, 9, 10, 11, 11, 11, 12, 13, 14, 14, 15, 15, 16, 17} },
+	{ 6.5, { 8, 8, 9, 10, 11, 11, 11, 12, 13, 14, 14, 15, 15, 16, 17} },
+	{   7, { 7, 8, 9, 10, 11, 11, 11, 12, 13, 13, 14, 14, 15, 15, 16} },
+	{ 7.5, { 5, 6, 8, 10, 11, 11, 11, 12, 12, 13, 13, 14, 14, 15, 16} },
+	{   8, { 4, 6, 7, 9, 10, 11, 11, 12, 12, 13, 13, 13, 13, 14, 15} },
+	{ 8.5, { 4, 5, 6, 8, 9, 10, 10, 11, 11, 12, 12, 13, 13, 14, 15} },
+	{   9, { 3, 4, 5, 7, 8, 9, 9, 10, 10, 11, 11, 12, 12, 13, 14} },
+	{ 9.5, { 3, 4, 4, 6, 6, 8, 8, 9, 10, 10, 11, 11, 12, 13, 14} },
+	{  10, { 2, 3, 3, 5, 5, 7, 7, 8, 9, 9, 10, 11, 11, 12, 13} },
+	{10.5, { 2, 3, 3, 5, 5, 6, 7, 8, 9, 9, 10, 11, 11, 12, 13} },
+	{  11, { 2, 3, 3, 5, 5, 6, 7, 8, 9, 9, 10, 10, 11, 11, 12} },
+	{11.5, { 2, 3, 3, 5, 5, 5, 7, 7, 8, 9, 9, 10, 10, 11, 12} },
+	{  12, { 2, 3, 3, 5, 5, 5, 7, 7, 8, 8, 9, 9, 10, 10, 11} },
+	{12.5, { 2, 2, 3, 4, 5, 5, 6, 7, 7, 8, 8, 9, 9, 10, 11} },
+	{  13, { 1, 2, 3, 4, 5, 5, 6, 7, 7, 8, 8, 8, 9, 9, 10} },
+	{13.5, { 1, 2, 3, 3, 4, 5, 6, 6, 7, 7, 8, 8, 8, 9, 10} },
+	{  14, { 1, 2, 3, 3, 4, 5, 5, 5, 6, 6, 7, 8, 8, 8, 9} },
+	{14.5, { 1, 2, 2, 3, 4, 4, 5, 5, 5, 6, 7, 7, 7, 8, 9} },
+	{  15, { 1, 2, 2, 3, 3, 4, 4, 5, 5, 5, 6, 7, 7, 7, 9} },
+	{15.5, { 1, 1, 2, 2, 3, 4, 4, 4, 5, 5, 6, 6, 6, 7, 8} },
+	{  16, { 1, 1, 2, 2, 3, 4, 4, 4, 5, 5, 6, 6, 6, 6, 8} },
+	{16.5, { 0, 1, 1, 2, 2, 3, 3, 4, 4, 4, 4, 5, 5, 6, 7} },
+	{  17, { 0, 0, 1, 1, 2, 2, 3, 3, 3, 4, 4, 4, 4, 5, 7} },
+	{17.5, { 0, 0, 1, 1, 2, 2, 2, 3, 3, 3, 3, 4, 4, 5, 6} },
+	{  18, { 0, 0, 0, 1, 1, 1, 2, 2, 2, 3, 3, 3, 4, 4, 6} },
+	{18.5, { 0, 0, 0, 1, 1, 1, 2, 2, 2, 2, 3, 3, 3, 4, 6} },
+	{  19, { 0, 0, 0, 1, 1, 1, 1, 2, 2, 2, 2, 2, 3, 3, 5} },
+	{19.5, { 0, 0, 0, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 3, 5} },
+	{  20, { 0, 0, 0, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 4} }
+};
+
+
+static const qp_table qp_table_420_10bpc_max = {
+	{   4, { 8, 8, 9, 10, 11, 11, 11, 12, 13, 14, 14, 15, 15, 17, 18} },
+	{ 4.5, { 8, 8, 9, 10, 11, 11, 11, 12, 13, 14, 14, 15, 15, 16, 17} },
+	{   5, { 7, 8, 9, 10, 11, 11, 11, 12, 13, 13, 14, 14, 15, 16, 17} },
+	{ 5.5, { 6, 7, 8, 9, 10, 10, 11, 12, 12, 13, 13, 14, 14, 15, 16} },
+	{   6, { 4, 6, 8, 9, 10, 10, 11, 12, 12, 13, 13, 13, 13, 14, 15} },
+	{ 6.5, { 4, 5, 7, 8, 8, 9, 10, 11, 11, 12, 12, 13, 13, 14, 15} },
+	{   7, { 3, 4, 6, 7, 7, 8, 9, 10, 10, 11, 12, 12, 13, 13, 14} },
+	{ 7.5, { 3, 4, 5, 6, 6, 7, 8, 9, 10, 10, 11, 11, 12, 12, 13} },
+	{   8, { 2, 3, 4, 5, 5, 6, 7, 8, 9, 9, 10, 11, 11, 12, 13} },
+	{ 8.5, { 1, 3, 3, 4, 4, 6, 7, 8, 9, 9, 10, 10, 11, 11, 12} },
+	{   9, { 1, 3, 3, 4, 4, 6, 7, 8, 8, 9, 9, 10, 10, 10, 11} },
+	{ 9.5, { 1, 3, 3, 3, 4, 5, 6, 7, 8, 8, 9, 9, 9, 10, 11} },
+	{  10, { 1, 2, 3, 3, 4, 4, 5, 6, 7, 8, 8, 9, 9, 9, 11} },
+	{10.5, { 1, 1, 3, 3, 3, 4, 5, 5, 7, 7, 8, 8, 8, 8, 10} },
+	{  11, { 0, 1, 2, 3, 3, 3, 4, 5, 6, 7, 7, 7, 8, 8, 9} },
+	{11.5, { 0, 1, 1, 2, 3, 3, 3, 4, 5, 6, 6, 7, 7, 7, 9} },
+	{  12, { 0, 0, 1, 1, 2, 2, 3, 3, 4, 5, 5, 6, 6, 6, 8} },
+	{12.5, { 0, 0, 1, 1, 2, 2, 3, 3, 4, 4, 5, 5, 5, 6, 7} },
+	{  13, { 0, 0, 0, 1, 2, 2, 3, 3, 3, 4, 4, 5, 5, 5, 7} },
+	{13.5, { 0, 0, 0, 1, 1, 2, 2, 2, 3, 3, 4, 4, 4, 4, 6} },
+	{  14, { 0, 0, 0, 1, 1, 2, 2, 2, 3, 3, 3, 4, 4, 4, 6} },
+	{14.5, { 0, 0, 0, 0, 1, 1, 2, 2, 2, 2, 3, 3, 3, 3, 5} },
+	{  15, { 0, 0, 0, 0, 0, 1, 1, 1, 2, 2, 2, 2, 2, 3, 5} }
+};
+
+
+static const qp_table   qp_table_420_10bpc_min = {
+	{   4, { 0, 4, 4, 5, 7, 7, 7, 7, 7, 7, 9, 9, 9, 13, 17} },
+	{ 4.5, { 0, 4, 4, 5, 7, 7, 7, 7, 7, 7, 9, 9, 9, 12, 16} },
+	{   5, { 0, 4, 4, 5, 7, 7, 7, 7, 7, 7, 9, 9, 9, 12, 16} },
+	{ 5.5, { 0, 3, 3, 4, 6, 7, 7, 7, 7, 7, 9, 9, 9, 11, 15} },
+	{   6, { 0, 2, 3, 4, 6, 7, 7, 7, 7, 7, 9, 9, 9, 11, 14} },
+	{ 6.5, { 0, 2, 3, 4, 5, 6, 6, 7, 7, 7, 8, 9, 9, 11, 14} },
+	{   7, { 0, 2, 3, 4, 5, 5, 6, 6, 7, 7, 8, 8, 9, 11, 13} },
+	{ 7.5, { 0, 2, 3, 4, 4, 5, 5, 6, 7, 7, 8, 8, 9, 11, 12} },
+	{   8, { 0, 2, 3, 4, 4, 5, 5, 6, 6, 7, 8, 8, 9, 11, 12} },
+	{ 8.5, { 0, 2, 2, 3, 3, 5, 5, 6, 6, 7, 8, 8, 9, 10, 11} },
+	{   9, { 0, 2, 2, 3, 3, 5, 5, 6, 6, 7, 7, 8, 9, 9, 10} },
+	{ 9.5, { 0, 2, 2, 2, 3, 4, 4, 5, 6, 6, 7, 8, 8, 9, 10} },
+	{  10, { 0, 1, 2, 2, 3, 3, 4, 4, 5, 6, 6, 8, 8, 8, 10} },
+	{10.5, { 0, 0, 2, 2, 2, 3, 4, 4, 5, 5, 6, 7, 7, 7, 9} },
+	{  11, { 0, 0, 1, 2, 2, 2, 3, 4, 4, 5, 5, 6, 7, 7, 8} },
+	{11.5, { 0, 0, 0, 1, 2, 2, 2, 3, 4, 4, 5, 6, 6, 6, 8} },
+	{  12, { 0, 0, 0, 0, 1, 1, 2, 2, 3, 4, 4, 5, 5, 5, 7} },
+	{12.5, { 0, 0, 0, 0, 1, 1, 2, 2, 3, 3, 4, 4, 4, 5, 6} },
+	{  13, { 0, 0, 0, 0, 1, 1, 2, 2, 2, 3, 3, 4, 4, 4, 6} },
+	{13.5, { 0, 0, 0, 0, 0, 1, 1, 1, 2, 2, 3, 3, 3, 3, 5} },
+	{  14, { 0, 0, 0, 0, 0, 1, 1, 1, 2, 2, 2, 3, 3, 3, 5} },
+	{14.5, { 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 2, 2, 2, 2, 4} },
+	{  15, { 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 2, 4} }
+};
+
+
+static const qp_table   qp_table_444_10bpc_max = {
+	{   6, { 8, 10, 12, 12, 13, 13, 13, 14, 15, 16, 16, 16, 16, 17, 19} },
+	{ 6.5, { 8, 10, 11, 12, 12, 12, 13, 14, 15, 15, 16, 16, 16, 17, 19} },
+	{   7, { 8, 9, 11, 11, 12, 12, 12, 13, 14, 15, 15, 16, 16, 17, 18} },
+	{ 7.5, { 8, 9, 10, 11, 11, 12, 12, 13, 14, 14, 15, 15, 16, 17, 18} },
+	{   8, { 8, 8, 9, 10, 11, 11, 11, 12, 13, 14, 14, 15, 15, 16, 17} },
+	{ 8.5, { 8, 8, 9, 10, 11, 11, 11, 12, 13, 14, 14, 15, 15, 16, 17} },
+	{   9, { 7, 8, 9, 10, 11, 11, 11, 12, 13, 13, 14, 14, 15, 15, 17} },
+	{ 9.5, { 7, 8, 9, 10, 11, 11, 11, 12, 13, 13, 14, 14, 15, 15, 17} },
+	{  10, { 7, 8, 9, 10, 11, 11, 11, 12, 13, 13, 14, 14, 15, 15, 16} },
+	{10.5, { 6, 8, 9, 10, 11, 11, 11, 12, 13, 13, 14, 14, 14, 15, 16} },
+	{  11, { 5, 7, 9, 10, 11, 11, 11, 12, 12, 13, 13, 13, 14, 14, 15} },
+	{11.5, { 5, 7, 9, 10, 11, 11, 11, 12, 12, 13, 13, 13, 13, 14, 15} },
+	{  12, { 4, 6, 8, 9, 10, 10, 11, 12, 12, 13, 13, 13, 13, 14, 15} },
+	{12.5, { 4, 6, 8, 9, 10, 10, 10, 11, 11, 12, 12, 13, 13, 14, 15} },
+	{  13, { 3, 6, 8, 9, 9, 10, 10, 11, 11, 12, 12, 12, 12, 13, 14} },
+	{13.5, { 3, 5, 7, 8, 8, 9, 9, 10, 10, 11, 11, 12, 12, 13, 14} },
+	{  14, { 3, 5, 6, 7, 8, 8, 8, 9, 10, 10, 11, 12, 12, 12, 14} },
+	{14.5, { 2, 4, 6, 7, 7, 8, 8, 9, 9, 10, 10, 11, 11, 12, 13} },
+	{  15, { 2, 4, 5, 6, 6, 7, 7, 8, 9, 9, 10, 11, 11, 12, 13} },
+	{15.5, { 2, 4, 5, 6, 6, 7, 7, 8, 9, 9, 10, 11, 11, 12, 13} },
+	{  16, { 2, 4, 5, 6, 6, 7, 7, 8, 9, 9, 9, 10, 10, 11, 12} },
+	{16.5, { 2, 3, 4, 5, 6, 7, 7, 8, 9, 9, 9, 10, 10, 11, 12} },
+	{  17, { 2, 3, 4, 5, 5, 6, 7, 7, 8, 8, 8, 9, 9, 10, 12} },
+	{17.5, { 1, 3, 4, 5, 5, 6, 6, 7, 8, 8, 8, 9, 9, 10, 12} },
+	{  18, { 1, 3, 4, 5, 5, 6, 6, 7, 8, 8, 8, 9, 9, 10, 11} },
+	{18.5, { 1, 3, 4, 5, 5, 6, 6, 7, 8, 8, 8, 9, 9, 10, 11} },
+	{  19, { 1, 2, 3, 4, 5, 5, 5, 6, 7, 7, 7, 8, 8, 9, 10} },
+	{19.5, { 1, 2, 3, 4, 5, 5, 5, 6, 7, 7, 7, 8, 8, 9, 10} },
+	{  20, { 1, 2, 3, 3, 4, 5, 5, 6, 6, 6, 6, 7, 7, 8, 10} },
+	{20.5, { 1, 2, 3, 3, 4, 5, 5, 5, 5, 6, 6, 7, 7, 8, 10} },
+	{  21, { 1, 2, 3, 3, 4, 4, 4, 5, 5, 5, 6, 7, 7, 8, 10} },
+	{21.5, { 1, 2, 3, 3, 3, 3, 4, 5, 5, 5, 6, 7, 7, 8, 9} },
+	{  22, { 1, 1, 2, 3, 3, 3, 4, 4, 5, 5, 6, 6, 6, 7, 9} },
+	{22.5, { 1, 1, 2, 3, 3, 3, 4, 4, 5, 5, 6, 6, 6, 7, 8} },
+	{  23, { 1, 1, 2, 3, 3, 3, 4, 4, 4, 4, 6, 6, 6, 6, 8} },
+	{23.5, { 1, 1, 1, 3, 3, 3, 4, 4, 4, 4, 6, 6, 6, 6, 8} },
+	{  24, { 1, 1, 1, 2, 2, 3, 4, 4, 4, 4, 5, 5, 5, 6, 8} },
+	{24.5, { 0, 1, 1, 2, 2, 3, 3, 4, 4, 4, 5, 5, 5, 6, 8} },
+	{  25, { 0, 1, 1, 1, 2, 3, 3, 3, 4, 4, 5, 5, 5, 5, 7} },
+	{25.5, { 0, 0, 1, 1, 2, 3, 3, 3, 4, 4, 4, 5, 5, 5, 7} },
+	{  26, { 0, 0, 1, 1, 1, 2, 2, 3, 3, 4, 4, 4, 4, 4, 6} },
+	{26.5, { 0, 0, 0, 1, 1, 2, 2, 3, 3, 3, 4, 4, 4, 4, 6} },
+	{  27, { 0, 0, 0, 1, 1, 2, 2, 2, 3, 3, 3, 4, 4, 4, 6} },
+	{27.5, { 0, 0, 0, 0, 1, 1, 2, 2, 2, 3, 3, 4, 4, 4, 6} },
+	{  28, { 0, 0, 0, 0, 0, 1, 1, 2, 2, 2, 3, 3, 3, 3, 5} },
+	{28.5, { 0, 0, 0, 0, 0, 1, 1, 1, 2, 2, 2, 3, 3, 3, 5} },
+	{  29, { 0, 0, 0, 0, 0, 1, 1, 1, 2, 2, 2, 2, 3, 3, 4} },
+	{29.5, { 0, 0, 0, 0, 0, 0, 1, 1, 1, 2, 2, 2, 3, 3, 4} },
+	{  30, { 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 2, 2, 2, 2, 4} }
+};
+
+
+static const qp_table   qp_table_422_8bpc_max = {
+	{   6, { 4, 4, 5, 6, 7, 7, 7, 8, 9, 10, 10, 11, 11, 12, 13} },
+	{ 6.5, { 4, 4, 5, 6, 7, 7, 7, 8, 9, 10, 10, 11, 11, 12, 13} },
+	{   7, { 3, 4, 5, 6, 7, 7, 7, 8, 9, 9, 10, 10, 11, 11, 12} },
+	{ 7.5, { 3, 4, 5, 6, 7, 7, 7, 8, 8, 9, 9, 10, 10, 11, 12} },
+	{   8, { 2, 4, 5, 6, 7, 7, 7, 8, 8, 9, 9, 9, 9, 10, 11} },
+	{ 8.5, { 2, 3, 4, 5, 6, 6, 6, 7, 7, 8, 8, 9, 9, 10, 11} },
+	{   9, { 1, 2, 3, 4, 5, 5, 5, 6, 6, 7, 7, 8, 8, 9, 10} },
+	{ 9.5, { 1, 2, 2, 3, 3, 4, 4, 5, 6, 6, 7, 7, 8, 9, 10} },
+	{  10, { 0, 1, 1, 2, 2, 3, 3, 4, 5, 5, 6, 7, 7, 8, 9} },
+	{10.5, { 0, 1, 1, 2, 2, 3, 3, 4, 5, 5, 6, 7, 7, 8, 9} },
+	{  11, { 0, 1, 1, 2, 2, 3, 3, 4, 5, 5, 6, 6, 7, 7, 8} },
+	{11.5, { 0, 1, 1, 2, 2, 2, 3, 3, 4, 5, 5, 6, 6, 7, 8} },
+	{  12, { 0, 1, 1, 2, 2, 2, 3, 3, 4, 4, 5, 5, 6, 6, 7} },
+	{12.5, { 0, 0, 1, 1, 2, 2, 2, 3, 3, 4, 4, 5, 5, 6, 7} },
+	{  13, { 0, 0, 1, 1, 2, 2, 2, 3, 3, 4, 4, 4, 5, 5, 6} },
+	{13.5, { 0, 0, 1, 1, 1, 2, 2, 2, 3, 3, 4, 4, 4, 5, 6} },
+	{  14, { 0, 0, 1, 1, 1, 2, 2, 2, 3, 3, 3, 4, 4, 4, 5} },
+	{14.5, { 0, 0, 0, 1, 1, 1, 2, 2, 2, 3, 3, 3, 3, 4, 5} },
+	{  15, { 0, 0, 0, 1, 1, 1, 1, 2, 2, 2, 2, 3, 3, 3, 5} },
+	{15.5, { 0, 0, 0, 0, 1, 1, 1, 1, 2, 2, 2, 2, 2, 3, 4} },
+	{  16, { 0, 0, 0, 0, 1, 1, 1, 1, 2, 2, 2, 2, 2, 3, 4} }
+};
+

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dsc/rc_calc_fpu.c

@@ -0,0 +1,260 @@
+/*
+ * Copyright 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 "rc_calc_fpu.h"
+
+#include "qp_tables.h"
+#include "amdgpu_dm/dc_fpu.h"
+
+#define table_hash(mode, bpc, max_min) ((mode << 16) | (bpc << 8) | max_min)
+
+#define MODE_SELECT(val444, val422, val420) \
+	(cm == CM_444 || cm == CM_RGB) ? (val444) : (cm == CM_422 ? (val422) : (val420))
+
+
+#define TABLE_CASE(mode, bpc, max)   case (table_hash(mode, BPC_##bpc, max)): \
+	table = qp_table_##mode##_##bpc##bpc_##max; \
+	table_size = sizeof(qp_table_##mode##_##bpc##bpc_##max)/sizeof(*qp_table_##mode##_##bpc##bpc_##max); \
+	break
+
+static int median3(int a, int b, int c)
+{
+	if (a > b)
+		swap(a, b);
+	if (b > c)
+		swap(b, c);
+	if (a > b)
+		swap(b, c);
+
+	return b;
+}
+
+static double dsc_roundf(double num)
+{
+	if (num < 0.0)
+		num = num - 0.5;
+	else
+		num = num + 0.5;
+
+	return (int)(num);
+}
+
+static void get_qp_set(qp_set qps, enum colour_mode cm, enum bits_per_comp bpc,
+		       enum max_min max_min, float bpp)
+{
+	int mode = MODE_SELECT(444, 422, 420);
+	int sel = table_hash(mode, bpc, max_min);
+	int table_size = 0;
+	int index;
+	const struct qp_entry *table = NULL;
+
+	// alias enum
+	enum { min = DAL_MM_MIN, max = DAL_MM_MAX };
+	switch (sel) {
+		TABLE_CASE(444,  8, max);
+		TABLE_CASE(444,  8, min);
+		TABLE_CASE(444, 10, max);
+		TABLE_CASE(444, 10, min);
+		TABLE_CASE(444, 12, max);
+		TABLE_CASE(444, 12, min);
+		TABLE_CASE(422,  8, max);
+		TABLE_CASE(422,  8, min);
+		TABLE_CASE(422, 10, max);
+		TABLE_CASE(422, 10, min);
+		TABLE_CASE(422, 12, max);
+		TABLE_CASE(422, 12, min);
+		TABLE_CASE(420,  8, max);
+		TABLE_CASE(420,  8, min);
+		TABLE_CASE(420, 10, max);
+		TABLE_CASE(420, 10, min);
+		TABLE_CASE(420, 12, max);
+		TABLE_CASE(420, 12, min);
+	}
+
+	if (!table)
+		return;
+
+	index = (int)((bpp - table[0].bpp) * 2);
+
+	/* requested size is bigger than the table */
+	if (index >= table_size) {
+		dm_error("ERROR: Requested rc_calc to find a bpp entry that exceeds the table size\n");
+		return;
+	}
+
+	memcpy(qps, table[index].qps, sizeof(qp_set));
+}
+
+static void get_ofs_set(qp_set ofs, enum colour_mode mode, float bpp)
+{
+	int   *p = ofs;
+
+	if (mode == CM_444 || mode == CM_RGB) {
+		*p++ = (int)((bpp <=  6) ? (0) : ((((bpp >=  8) && (bpp <= 12))) ? (2) : ((bpp >= 15) ? (10) : ((((bpp > 6) && (bpp < 8))) ? (0 + dsc_roundf((bpp -  6) * (2 / 2.0))) : (2 + dsc_roundf((bpp - 12) * (8 / 3.0)))))));
+		*p++ = (int)((bpp <=  6) ? (-2) : ((((bpp >=  8) && (bpp <= 12))) ? (0) : ((bpp >= 15) ? (8) : ((((bpp > 6) && (bpp < 8))) ? (-2 + dsc_roundf((bpp -  6) * (2 / 2.0))) : (0 + dsc_roundf((bpp - 12) * (8 / 3.0)))))));
+		*p++ = (int)((bpp <=  6) ? (-2) : ((((bpp >=  8) && (bpp <= 12))) ? (0) : ((bpp >= 15) ? (6) : ((((bpp > 6) && (bpp < 8))) ? (-2 + dsc_roundf((bpp -  6) * (2 / 2.0))) : (0 + dsc_roundf((bpp - 12) * (6 / 3.0)))))));
+		*p++ = (int)((bpp <=  6) ? (-4) : ((((bpp >=  8) && (bpp <= 12))) ? (-2) : ((bpp >= 15) ? (4) : ((((bpp > 6) && (bpp < 8))) ? (-4 + dsc_roundf((bpp -  6) * (2 / 2.0))) : (-2 + dsc_roundf((bpp - 12) * (6 / 3.0)))))));
+		*p++ = (int)((bpp <=  6) ? (-6) : ((((bpp >=  8) && (bpp <= 12))) ? (-4) : ((bpp >= 15) ? (2) : ((((bpp > 6) && (bpp < 8))) ? (-6 + dsc_roundf((bpp -  6) * (2 / 2.0))) : (-4 + dsc_roundf((bpp - 12) * (6 / 3.0)))))));
+		*p++ = (int)((bpp <= 12) ? (-6) : ((bpp >= 15) ? (0) : (-6 + dsc_roundf((bpp - 12) * (6 / 3.0)))));
+		*p++ = (int)((bpp <= 12) ? (-8) : ((bpp >= 15) ? (-2) : (-8 + dsc_roundf((bpp - 12) * (6 / 3.0)))));
+		*p++ = (int)((bpp <= 12) ? (-8) : ((bpp >= 15) ? (-4) : (-8 + dsc_roundf((bpp - 12) * (4 / 3.0)))));
+		*p++ = (int)((bpp <= 12) ? (-8) : ((bpp >= 15) ? (-6) : (-8 + dsc_roundf((bpp - 12) * (2 / 3.0)))));
+		*p++ = (int)((bpp <= 12) ? (-10) : ((bpp >= 15) ? (-8) : (-10 + dsc_roundf((bpp - 12) * (2 / 3.0)))));
+		*p++ = -10;
+		*p++ = (int)((bpp <=  6) ? (-12) : ((bpp >=  8) ? (-10) : (-12 + dsc_roundf((bpp -  6) * (2 / 2.0)))));
+		*p++ = -12;
+		*p++ = -12;
+		*p++ = -12;
+	} else if (mode == CM_422) {
+		*p++ = (int)((bpp <=  8) ? (2) : ((bpp >= 10) ? (10) : (2 + dsc_roundf((bpp -  8) * (8 / 2.0)))));
+		*p++ = (int)((bpp <=  8) ? (0) : ((bpp >= 10) ? (8) : (0 + dsc_roundf((bpp -  8) * (8 / 2.0)))));
+		*p++ = (int)((bpp <=  8) ? (0) : ((bpp >= 10) ? (6) : (0 + dsc_roundf((bpp -  8) * (6 / 2.0)))));
+		*p++ = (int)((bpp <=  8) ? (-2) : ((bpp >= 10) ? (4) : (-2 + dsc_roundf((bpp -  8) * (6 / 2.0)))));
+		*p++ = (int)((bpp <=  8) ? (-4) : ((bpp >= 10) ? (2) : (-4 + dsc_roundf((bpp -  8) * (6 / 2.0)))));
+		*p++ = (int)((bpp <=  8) ? (-6) : ((bpp >= 10) ? (0) : (-6 + dsc_roundf((bpp -  8) * (6 / 2.0)))));
+		*p++ = (int)((bpp <=  8) ? (-8) : ((bpp >= 10) ? (-2) : (-8 + dsc_roundf((bpp -  8) * (6 / 2.0)))));
+		*p++ = (int)((bpp <=  8) ? (-8) : ((bpp >= 10) ? (-4) : (-8 + dsc_roundf((bpp -  8) * (4 / 2.0)))));
+		*p++ = (int)((bpp <=  8) ? (-8) : ((bpp >= 10) ? (-6) : (-8 + dsc_roundf((bpp -  8) * (2 / 2.0)))));
+		*p++ = (int)((bpp <=  8) ? (-10) : ((bpp >= 10) ? (-8) : (-10 + dsc_roundf((bpp -  8) * (2 / 2.0)))));
+		*p++ = -10;
+		*p++ = (int)((bpp <=  6) ? (-12) : ((bpp >= 7) ? (-10) : (-12 + dsc_roundf((bpp -  6) * (2.0 / 1)))));
+		*p++ = -12;
+		*p++ = -12;
+		*p++ = -12;
+	} else {
+		*p++ = (int)((bpp <=  6) ? (2) : ((bpp >=  8) ? (10) : (2 + dsc_roundf((bpp -  6) * (8 / 2.0)))));
+		*p++ = (int)((bpp <=  6) ? (0) : ((bpp >=  8) ? (8) : (0 + dsc_roundf((bpp -  6) * (8 / 2.0)))));
+		*p++ = (int)((bpp <=  6) ? (0) : ((bpp >=  8) ? (6) : (0 + dsc_roundf((bpp -  6) * (6 / 2.0)))));
+		*p++ = (int)((bpp <=  6) ? (-2) : ((bpp >=  8) ? (4) : (-2 + dsc_roundf((bpp -  6) * (6 / 2.0)))));
+		*p++ = (int)((bpp <=  6) ? (-4) : ((bpp >=  8) ? (2) : (-4 + dsc_roundf((bpp -  6) * (6 / 2.0)))));
+		*p++ = (int)((bpp <=  6) ? (-6) : ((bpp >=  8) ? (0) : (-6 + dsc_roundf((bpp -  6) * (6 / 2.0)))));
+		*p++ = (int)((bpp <=  6) ? (-8) : ((bpp >=  8) ? (-2) : (-8 + dsc_roundf((bpp -  6) * (6 / 2.0)))));
+		*p++ = (int)((bpp <=  6) ? (-8) : ((bpp >=  8) ? (-4) : (-8 + dsc_roundf((bpp -  6) * (4 / 2.0)))));
+		*p++ = (int)((bpp <=  6) ? (-8) : ((bpp >=  8) ? (-6) : (-8 + dsc_roundf((bpp -  6) * (2 / 2.0)))));
+		*p++ = (int)((bpp <=  6) ? (-10) : ((bpp >=  8) ? (-8) : (-10 + dsc_roundf((bpp -  6) * (2 / 2.0)))));
+		*p++ = -10;
+		*p++ = (int)((bpp <=  4) ? (-12) : ((bpp >=  5) ? (-10) : (-12 + dsc_roundf((bpp -  4) * (2 / 1.0)))));
+		*p++ = -12;
+		*p++ = -12;
+		*p++ = -12;
+	}
+}
+
+void _do_calc_rc_params(struct rc_params *rc,
+		enum colour_mode cm,
+		enum bits_per_comp bpc,
+		u16 drm_bpp,
+		bool is_navite_422_or_420,
+		int slice_width,
+		int slice_height,
+		int minor_version)
+{
+	float bpp;
+	float bpp_group;
+	float initial_xmit_delay_factor;
+	int padding_pixels;
+	int i;
+
+	dc_assert_fp_enabled();
+
+	bpp = (float)((float)drm_bpp / 16.0);
+	/* in native_422 or native_420 modes, the bits_per_pixel is double the
+	 * target bpp (the latter is what calc_rc_params expects)
+	 */
+	if (is_navite_422_or_420)
+		bpp /= 2.0;
+
+	rc->rc_quant_incr_limit0 = ((bpc == BPC_8) ? 11 : (bpc == BPC_10 ? 15 : 19)) - ((minor_version == 1 && cm == CM_444) ? 1 : 0);
+	rc->rc_quant_incr_limit1 = ((bpc == BPC_8) ? 11 : (bpc == BPC_10 ? 15 : 19)) - ((minor_version == 1 && cm == CM_444) ? 1 : 0);
+
+	bpp_group = MODE_SELECT(bpp, bpp * 2.0f, bpp * 2.0f);
+
+	switch (cm) {
+	case CM_420:
+		rc->initial_fullness_offset = (int)((bpp >=  6) ? (2048) : ((bpp <=  4) ? (6144) : ((((bpp >  4) && (bpp <=  5))) ? (6144 - dsc_roundf((bpp - 4) * (512))) : (5632 - dsc_roundf((bpp -  5) * (3584))))));
+		rc->first_line_bpg_offset   = median3(0, (12 + (int) (0.09 *  min(34, slice_height - 8))), (int)((3 * bpc * 3) - (3 * bpp_group)));
+		rc->second_line_bpg_offset  = median3(0, 12, (int)((3 * bpc * 3) - (3 * bpp_group)));
+		break;
+	case CM_422:
+		rc->initial_fullness_offset = (int)((bpp >=  8) ? (2048) : ((bpp <=  7) ? (5632) : (5632 - dsc_roundf((bpp - 7) * (3584)))));
+		rc->first_line_bpg_offset   = median3(0, (12 + (int) (0.09 *  min(34, slice_height - 8))), (int)((3 * bpc * 4) - (3 * bpp_group)));
+		rc->second_line_bpg_offset  = 0;
+		break;
+	case CM_444:
+	case CM_RGB:
+		rc->initial_fullness_offset = (int)((bpp >= 12) ? (2048) : ((bpp <=  8) ? (6144) : ((((bpp >  8) && (bpp <= 10))) ? (6144 - dsc_roundf((bpp - 8) * (512 / 2))) : (5632 - dsc_roundf((bpp - 10) * (3584 / 2))))));
+		rc->first_line_bpg_offset   = median3(0, (12 + (int) (0.09 *  min(34, slice_height - 8))), (int)(((3 * bpc + (cm == CM_444 ? 0 : 2)) * 3) - (3 * bpp_group)));
+		rc->second_line_bpg_offset  = 0;
+		break;
+	}
+
+	initial_xmit_delay_factor = (float)((cm == CM_444 || cm == CM_RGB) ? 1.0 : 2.0);
+	rc->initial_xmit_delay = (int)dsc_roundf(8192.0/2.0/bpp/initial_xmit_delay_factor);
+
+	if (cm == CM_422 || cm == CM_420)
+		slice_width /= 2;
+
+	padding_pixels = ((slice_width % 3) != 0) ? (3 - (slice_width % 3)) * (rc->initial_xmit_delay / slice_width) : 0;
+	if (3 * bpp_group >= (((rc->initial_xmit_delay + 2) / 3) * (3 + (cm == CM_422)))) {
+		if ((rc->initial_xmit_delay + padding_pixels) % 3 == 1)
+			rc->initial_xmit_delay++;
+	}
+
+	rc->flatness_min_qp     = ((bpc == BPC_8) ?  (3) : ((bpc == BPC_10) ? (7)  : (11))) - ((minor_version == 1 && cm == CM_444) ? 1 : 0);
+	rc->flatness_max_qp     = ((bpc == BPC_8) ? (12) : ((bpc == BPC_10) ? (16) : (20))) - ((minor_version == 1 && cm == CM_444) ? 1 : 0);
+	rc->flatness_det_thresh = 2 << (bpc - 8);
+
+	get_qp_set(rc->qp_min, cm, bpc, DAL_MM_MIN, bpp);
+	get_qp_set(rc->qp_max, cm, bpc, DAL_MM_MAX, bpp);
+	if (cm == CM_444 && minor_version == 1) {
+		for (i = 0; i < QP_SET_SIZE; ++i) {
+			rc->qp_min[i] = rc->qp_min[i] > 0 ? rc->qp_min[i] - 1 : 0;
+			rc->qp_max[i] = rc->qp_max[i] > 0 ? rc->qp_max[i] - 1 : 0;
+		}
+	}
+	get_ofs_set(rc->ofs, cm, bpp);
+
+	/* fixed parameters */
+	rc->rc_model_size    = 8192;
+	rc->rc_edge_factor   = 6;
+	rc->rc_tgt_offset_hi = 3;
+	rc->rc_tgt_offset_lo = 3;
+
+	rc->rc_buf_thresh[0] = 896;
+	rc->rc_buf_thresh[1] = 1792;
+	rc->rc_buf_thresh[2] = 2688;
+	rc->rc_buf_thresh[3] = 3584;
+	rc->rc_buf_thresh[4] = 4480;
+	rc->rc_buf_thresh[5] = 5376;
+	rc->rc_buf_thresh[6] = 6272;
+	rc->rc_buf_thresh[7] = 6720;
+	rc->rc_buf_thresh[8] = 7168;
+	rc->rc_buf_thresh[9] = 7616;
+	rc->rc_buf_thresh[10] = 7744;
+	rc->rc_buf_thresh[11] = 7872;
+	rc->rc_buf_thresh[12] = 8000;
+	rc->rc_buf_thresh[13] = 8064;
+}
+

linux-7.2.3/drivers/gpu/drm/amd/display/dc/dml/dsc/rc_calc_fpu.h

@@ -0,0 +1,90 @@
+/*
+ * Copyright 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 __RC_CALC_FPU_H__
+#define __RC_CALC_FPU_H__
+
+#include "os_types.h"
+#include <drm/display/drm_dsc.h>
+
+#define QP_SET_SIZE 15
+
+typedef int qp_set[QP_SET_SIZE];
+
+struct rc_params {
+	int      rc_quant_incr_limit0;
+	int      rc_quant_incr_limit1;
+	int      initial_fullness_offset;
+	int      initial_xmit_delay;
+	int      first_line_bpg_offset;
+	int      second_line_bpg_offset;
+	int      flatness_min_qp;
+	int      flatness_max_qp;
+	int      flatness_det_thresh;
+	qp_set   qp_min;
+	qp_set   qp_max;
+	qp_set   ofs;
+	int      rc_model_size;
+	int      rc_edge_factor;
+	int      rc_tgt_offset_hi;
+	int      rc_tgt_offset_lo;
+	int      rc_buf_thresh[QP_SET_SIZE - 1];
+};
+
+enum colour_mode {
+	CM_RGB,   /* 444 RGB */
+	CM_444,   /* 444 YUV or simple 422 */
+	CM_422,   /* native 422 */
+	CM_420    /* native 420 */
+};
+
+enum bits_per_comp {
+	BPC_8  =  8,
+	BPC_10 = 10,
+	BPC_12 = 12
+};
+
+enum max_min {
+	DAL_MM_MIN = 0,
+	DAL_MM_MAX = 1
+};
+
+struct qp_entry {
+	float         bpp;
+	const qp_set  qps;
+};
+
+typedef struct qp_entry qp_table[];
+
+void _do_calc_rc_params(struct rc_params *rc,
+		enum colour_mode cm,
+		enum bits_per_comp bpc,
+		u16 drm_bpp,
+		bool is_navite_422_or_420,
+		int slice_width,
+		int slice_height,
+		int minor_version);
+
+#endif