/*
* 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
*
*/
#include "dm_services.h"
#include "core_types.h"
#include "reg_helper.h"
#include "dcn30/dcn30_dpp.h"
#include "basics/conversion.h"
#include "dcn30/dcn30_cm_common.h"
#define REG(reg)\
dpp->tf_regs->reg
#define CTX \
dpp->base.ctx
#undef FN
#define FN(reg_name, field_name) \
dpp->tf_shift->field_name, dpp->tf_mask->field_name
void dpp30_read_state(struct dpp *dpp_base, struct dcn_dpp_state *s)
{
struct dcn3_dpp *dpp = TO_DCN30_DPP(dpp_base);
uint32_t gamcor_lut_mode, rgam_lut_mode;
REG_GET(DPP_CONTROL,
DPP_CLOCK_ENABLE, &s->is_enabled);
// Pre-degamma (ROM)
REG_GET_2(PRE_DEGAM,
PRE_DEGAM_MODE, &s->pre_dgam_mode,
PRE_DEGAM_SELECT, &s->pre_dgam_select);
// Gamma Correction (RAM)
REG_GET(CM_GAMCOR_CONTROL,
CM_GAMCOR_MODE_CURRENT, &s->gamcor_mode);
if (s->gamcor_mode) {
REG_GET(CM_GAMCOR_CONTROL, CM_GAMCOR_SELECT_CURRENT, &gamcor_lut_mode);
if (!gamcor_lut_mode)
s->gamcor_mode = LUT_RAM_A; // Otherwise, LUT_RAM_B
}
// Shaper LUT (RAM), 3D LUT (mode, bit-depth, size)
if (REG(CM_SHAPER_CONTROL))
REG_GET(CM_SHAPER_CONTROL, CM_SHAPER_LUT_MODE, &s->shaper_lut_mode);
if (REG(CM_3DLUT_MODE))
REG_GET(CM_3DLUT_MODE, CM_3DLUT_MODE_CURRENT, &s->lut3d_mode);
if (REG(CM_3DLUT_READ_WRITE_CONTROL))
REG_GET(CM_3DLUT_READ_WRITE_CONTROL, CM_3DLUT_30BIT_EN, &s->lut3d_bit_depth);
if (REG(CM_3DLUT_MODE))
REG_GET(CM_3DLUT_MODE, CM_3DLUT_SIZE, &s->lut3d_size);
// Blend/Out Gamma (RAM)
if (REG(CM_BLNDGAM_CONTROL)) {
REG_GET(CM_BLNDGAM_CONTROL, CM_BLNDGAM_MODE_CURRENT, &s->rgam_lut_mode);
if (s->rgam_lut_mode) {
REG_GET(CM_BLNDGAM_CONTROL, CM_BLNDGAM_SELECT_CURRENT, &rgam_lut_mode);
if (!rgam_lut_mode)
s->rgam_lut_mode = LUT_RAM_A; // Otherwise, LUT_RAM_B
}
}
}
/*program post scaler scs block in dpp CM*/
void dpp3_program_post_csc(
struct dpp *dpp_base,
enum dc_color_space color_space,
enum dcn10_input_csc_select input_select,
const struct out_csc_color_matrix *tbl_entry)
{
struct dcn3_dpp *dpp = TO_DCN30_DPP(dpp_base);
int i;
int arr_size = sizeof(dpp_input_csc_matrix)/sizeof(struct dpp_input_csc_matrix);
const uint16_t *regval = NULL;
uint32_t cur_select = 0;
enum dcn10_input_csc_select select;
struct color_matrices_reg gam_regs;
if (input_select == INPUT_CSC_SELECT_BYPASS) {
REG_SET(CM_POST_CSC_CONTROL, 0, CM_POST_CSC_MODE, 0);
return;
}
if (tbl_entry == NULL) {
for (i = 0; i < arr_size; i++)
if (dpp_input_csc_matrix[i].color_space == color_space) {
regval = dpp_input_csc_matrix[i].regval;
break;
}
if (regval == NULL) {
BREAK_TO_DEBUGGER();
return;
}
} else {
regval = tbl_entry->regval;
}
/* determine which CSC matrix (icsc or coma) we are using
* currently. select the alternate set to double buffer
* the CSC update so CSC is updated on frame boundary
*/
REG_GET(CM_POST_CSC_CONTROL,
CM_POST_CSC_MODE_CURRENT, &cur_select);
if (cur_select != INPUT_CSC_SELECT_ICSC)
select = INPUT_CSC_SELECT_ICSC;
else
select = INPUT_CSC_SELECT_COMA;
gam_regs.shifts.csc_c11 = dpp->tf_shift->CM_POST_CSC_C11;
gam_regs.masks.csc_c11 = dpp->tf_mask->CM_POST_CSC_C11;
gam_regs.shifts.csc_c12 = dpp->tf_shift->CM_POST_CSC_C12;
gam_regs.masks.csc_c12 = dpp->tf_mask->CM_POST_CSC_C12;
if (select == INPUT_CSC_SELECT_ICSC) {
gam_regs.csc_c11_c12 = REG(CM_POST_CSC_C11_C12);
gam_regs.csc_c33_c34 = REG(CM_POST_CSC_C33_C34);
} else {
gam_regs.csc_c11_c12 = REG(CM_POST_CSC_B_C11_C12);
gam_regs.csc_c33_c34 = REG(CM_POST_CSC_B_C33_C34);
}
cm_helper_program_color_matrices(
dpp->base.ctx,
regval,
&gam_regs);
REG_SET(CM_POST_CSC_CONTROL, 0,
CM_POST_CSC_MODE, select);
}
/*CNVC degam unit has read only LUTs*/
void dpp3_set_pre_degam(struct dpp *dpp_base, enum dc_transfer_func_predefined tr)
{
struct dcn3_dpp *dpp = TO_DCN30_DPP(dpp_base);
int pre_degam_en = 1;
int degamma_lut_selection = 0;
switch (tr) {
case TRANSFER_FUNCTION_LINEAR:
case TRANSFER_FUNCTION_UNITY:
pre_degam_en = 0; //bypass
break;
case TRANSFER_FUNCTION_SRGB:
degamma_lut_selection = 0;
break;
case TRANSFER_FUNCTION_BT709:
degamma_lut_selection = 4;
break;
case TRANSFER_FUNCTION_PQ:
degamma_lut_selection = 5;
break;
case TRANSFER_FUNCTION_HLG:
degamma_lut_selection = 6;
break;
case TRANSFER_FUNCTION_GAMMA22:
degamma_lut_selection = 1;
break;
case TRANSFER_FUNCTION_GAMMA24:
degamma_lut_selection = 2;
break;
case TRANSFER_FUNCTION_GAMMA26:
degamma_lut_selection = 3;
break;
default:
pre_degam_en = 0;
break;
}
REG_SET_2(PRE_DEGAM, 0,
PRE_DEGAM_MODE, pre_degam_en,
PRE_DEGAM_SELECT, degamma_lut_selection);
}
void dpp3_cnv_setup (
struct dpp *dpp_base,
enum surface_pixel_format format,
enum expansion_mode mode,
struct dc_csc_transform input_csc_color_matrix,
enum dc_color_space input_color_space,
struct cnv_alpha_2bit_lut *alpha_2bit_lut)
{
struct dcn3_dpp *dpp = TO_DCN30_DPP(dpp_base);
uint32_t pixel_format = 0;
uint32_t alpha_en = 1;
enum dc_color_space color_space = COLOR_SPACE_SRGB;
enum dcn10_input_csc_select select = INPUT_CSC_SELECT_BYPASS;
bool force_disable_cursor = false;
uint32_t is_2bit = 0;
uint32_t alpha_plane_enable = 0;
uint32_t dealpha_en = 0, dealpha_ablnd_en = 0;
uint32_t realpha_en = 0, realpha_ablnd_en = 0;
struct out_csc_color_matrix tbl_entry;
int i;
REG_SET_2(FORMAT_CONTROL, 0,
CNVC_BYPASS, 0,
FORMAT_EXPANSION_MODE, mode);
REG_UPDATE(FORMAT_CONTROL, FORMAT_CNV16, 0);
REG_UPDATE(FORMAT_CONTROL, CNVC_BYPASS_MSB_ALIGN, 0);
REG_UPDATE(FORMAT_CONTROL, CLAMP_POSITIVE, 0);
REG_UPDATE(FORMAT_CONTROL, CLAMP_POSITIVE_C, 0);
REG_UPDATE(FORMAT_CONTROL, FORMAT_CROSSBAR_R, 0);
REG_UPDATE(FORMAT_CONTROL, FORMAT_CROSSBAR_G, 1);
REG_UPDATE(FORMAT_CONTROL, FORMAT_CROSSBAR_B, 2);
switch (format) {
case SURFACE_PIXEL_FORMAT_GRPH_ARGB1555:
pixel_format = 1;
break;
case SURFACE_PIXEL_FORMAT_GRPH_RGB565:
pixel_format = 3;
alpha_en = 0;
break;
case SURFACE_PIXEL_FORMAT_GRPH_ARGB8888:
case SURFACE_PIXEL_FORMAT_GRPH_ABGR8888:
pixel_format = 8;
break;
case SURFACE_PIXEL_FORMAT_GRPH_ARGB2101010:
case SURFACE_PIXEL_FORMAT_GRPH_ABGR2101010:
pixel_format = 10;
is_2bit = 1;
break;
case SURFACE_PIXEL_FORMAT_VIDEO_420_YCbCr:
force_disable_cursor = false;
pixel_format = 65;
color_space = COLOR_SPACE_YCBCR709;
select = INPUT_CSC_SELECT_ICSC;
break;
case SURFACE_PIXEL_FORMAT_VIDEO_420_YCrCb:
force_disable_cursor = true;
pixel_format = 64;
color_space = COLOR_SPACE_YCBCR709;
select = INPUT_CSC_SELECT_ICSC;
break;
case SURFACE_PIXEL_FORMAT_VIDEO_420_10bpc_YCbCr:
force_disable_cursor = true;
pixel_format = 67;
color_space = COLOR_SPACE_YCBCR709;
select = INPUT_CSC_SELECT_ICSC;
break;
case SURFACE_PIXEL_FORMAT_VIDEO_420_10bpc_YCrCb:
force_disable_cursor = true;
pixel_format = 66;
color_space = COLOR_SPACE_YCBCR709;
select = INPUT_CSC_SELECT_ICSC;
break;
case SURFACE_PIXEL_FORMAT_GRPH_ARGB16161616:
case SURFACE_PIXEL_FORMAT_GRPH_ABGR16161616:
pixel_format = 26; /* ARGB16161616_UNORM */
break;
case SURFACE_PIXEL_FORMAT_GRPH_ARGB16161616F:
pixel_format = 24;
break;
case SURFACE_PIXEL_FORMAT_GRPH_ABGR16161616F:
pixel_format = 25;
break;
case SURFACE_PIXEL_FORMAT_VIDEO_AYCrCb8888:
pixel_format = 12;
color_space = COLOR_SPACE_YCBCR709;
select = INPUT_CSC_SELECT_ICSC;
break;
case SURFACE_PIXEL_FORMAT_GRPH_RGB111110_FIX:
pixel_format = 112;
alpha_en = 0;
break;
case SURFACE_PIXEL_FORMAT_GRPH_BGR101111_FIX:
pixel_format = 113;
alpha_en = 0;
break;
case SURFACE_PIXEL_FORMAT_VIDEO_ACrYCb2101010:
pixel_format = 114;
color_space = COLOR_SPACE_YCBCR709;
select = INPUT_CSC_SELECT_ICSC;
is_2bit = 1;
break;
case SURFACE_PIXEL_FORMAT_VIDEO_CrYCbA1010102:
pixel_format = 115;
color_space = COLOR_SPACE_YCBCR709;
select = INPUT_CSC_SELECT_ICSC;
is_2bit = 1;
break;
case SURFACE_PIXEL_FORMAT_GRPH_RGBE:
pixel_format = 116;
alpha_plane_enable = 0;
break;
case SURFACE_PIXEL_FORMAT_GRPH_RGBE_ALPHA:
pixel_format = 116;
alpha_plane_enable = 1;
break;
case SURFACE_PIXEL_FORMAT_GRPH_RGB111110_FLOAT:
pixel_format = 118;
alpha_en = 0;
break;
case SURFACE_PIXEL_FORMAT_GRPH_BGR101111_FLOAT:
pixel_format = 119;
alpha_en = 0;
break;
default:
break;
}
/* Set default color space based on format if none is given. */
color_space = input_color_space ? input_color_space : color_space;
if (is_2bit == 1 && alpha_2bit_lut != NULL) {
REG_UPDATE(ALPHA_2BIT_LUT, ALPHA_2BIT_LUT0, alpha_2bit_lut->lut0);
REG_UPDATE(ALPHA_2BIT_LUT, ALPHA_2BIT_LUT1, alpha_2bit_lut->lut1);
REG_UPDATE(ALPHA_2BIT_LUT, ALPHA_2BIT_LUT2, alpha_2bit_lut->lut2);
REG_UPDATE(ALPHA_2BIT_LUT, ALPHA_2BIT_LUT3, alpha_2bit_lut->lut3);
}
REG_SET_2(CNVC_SURFACE_PIXEL_FORMAT, 0,
CNVC_SURFACE_PIXEL_FORMAT, pixel_format,
CNVC_ALPHA_PLANE_ENABLE, alpha_plane_enable);
REG_UPDATE(FORMAT_CONTROL, FORMAT_CONTROL__ALPHA_EN, alpha_en);
REG_SET_2(PRE_DEALPHA, 0,
PRE_DEALPHA_EN, dealpha_en,
PRE_DEALPHA_ABLND_EN, dealpha_ablnd_en);
REG_SET_2(PRE_REALPHA, 0,
PRE_REALPHA_EN, realpha_en,
PRE_REALPHA_ABLND_EN, realpha_ablnd_en);
/* If input adjustment exists, program the ICSC with those values. */
if (input_csc_color_matrix.enable_adjustment == true) {
for (i = 0; i < 12; i++)
tbl_entry.regval[i] = input_csc_color_matrix.matrix[i];
tbl_entry.color_space = input_color_space;
if (color_space >= COLOR_SPACE_YCBCR601)
select = INPUT_CSC_SELECT_ICSC;
else
select = INPUT_CSC_SELECT_BYPASS;
dpp3_program_post_csc(dpp_base, color_space, select,
&tbl_entry);
} else {
dpp3_program_post_csc(dpp_base, color_space, select, NULL);
}
if (force_disable_cursor) {
REG_UPDATE(CURSOR_CONTROL,
CURSOR_ENABLE, 0);
REG_UPDATE(CURSOR0_CONTROL,
CUR0_ENABLE, 0);
}
}
#define IDENTITY_RATIO(ratio) (dc_fixpt_u3d19(ratio) == (1 << 19))
void dpp3_set_cursor_attributes(
struct dpp *dpp_base,
struct dc_cursor_attributes *cursor_attributes)
{
enum dc_cursor_color_format color_format = cursor_attributes->color_format;
struct dcn3_dpp *dpp = TO_DCN30_DPP(dpp_base);
int cur_rom_en = 0;
if (color_format == CURSOR_MODE_COLOR_PRE_MULTIPLIED_ALPHA ||
color_format == CURSOR_MODE_COLOR_UN_PRE_MULTIPLIED_ALPHA) {
if (cursor_attributes->attribute_flags.bits.ENABLE_CURSOR_DEGAMMA) {
cur_rom_en = 1;
}
}
REG_UPDATE_3(CURSOR0_CONTROL,
CUR0_MODE, color_format,
CUR0_EXPANSION_MODE, 0,
CUR0_ROM_EN, cur_rom_en);
if (color_format == CURSOR_MODE_MONO) {
/* todo: clarify what to program these to */
REG_UPDATE(CURSOR0_COLOR0,
CUR0_COLOR0, 0x00000000);
REG_UPDATE(CURSOR0_COLOR1,
CUR0_COLOR1, 0xFFFFFFFF);
}
dpp_base->att.cur0_ctl.bits.expansion_mode = 0;
dpp_base->att.cur0_ctl.bits.cur0_rom_en = cur_rom_en;
dpp_base->att.cur0_ctl.bits.mode = color_format;
}
bool dpp3_get_optimal_number_of_taps(
struct dpp *dpp,
struct scaler_data *scl_data,
const struct scaling_taps *in_taps)
{
int num_part_y, num_part_c;
int max_taps_y, max_taps_c;
int min_taps_y, min_taps_c;
enum lb_memory_config lb_config;
if (scl_data->viewport.width > scl_data->h_active &&
dpp->ctx->dc->debug.max_downscale_src_width != 0 &&
scl_data->viewport.width > dpp->ctx->dc->debug.max_downscale_src_width)
return false;
/*
* Set default taps if none are provided
* From programming guide: taps = min{ ceil(2*H_RATIO,1), 8} for downscaling
* taps = 4 for upscaling
*/
if (in_taps->h_taps == 0) {
if (dc_fixpt_ceil(scl_data->ratios.horz) > 1)
scl_data->taps.h_taps = min(2 * dc_fixpt_ceil(scl_data->ratios.horz), 8);
else
scl_data->taps.h_taps = 4;
} else
scl_data->taps.h_taps = in_taps->h_taps;
if (in_taps->v_taps == 0) {
if (dc_fixpt_ceil(scl_data->ratios.vert) > 1)
scl_data->taps.v_taps = min(dc_fixpt_ceil(dc_fixpt_mul_int(scl_data->ratios.vert, 2)), 8);
else
scl_data->taps.v_taps = 4;
} else
scl_data->taps.v_taps = in_taps->v_taps;
if (in_taps->v_taps_c == 0) {
if (dc_fixpt_ceil(scl_data->ratios.vert_c) > 1)
scl_data->taps.v_taps_c = min(dc_fixpt_ceil(dc_fixpt_mul_int(scl_data->ratios.vert_c, 2)), 8);
else
scl_data->taps.v_taps_c = 4;
} else
scl_data->taps.v_taps_c = in_taps->v_taps_c;
if (in_taps->h_taps_c == 0) {
if (dc_fixpt_ceil(scl_data->ratios.horz_c) > 1)
scl_data->taps.h_taps_c = min(2 * dc_fixpt_ceil(scl_data->ratios.horz_c), 8);
else
scl_data->taps.h_taps_c = 4;
} else if ((in_taps->h_taps_c % 2) != 0 && in_taps->h_taps_c != 1)
/* Only 1 and even h_taps_c are supported by hw */
scl_data->taps.h_taps_c = in_taps->h_taps_c - 1;
else
scl_data->taps.h_taps_c = in_taps->h_taps_c;
/*Ensure we can support the requested number of vtaps*/
min_taps_y = dc_fixpt_ceil(scl_data->ratios.vert);
min_taps_c = dc_fixpt_ceil(scl_data->ratios.vert_c);
/* Use LB_MEMORY_CONFIG_3 for 4:2:0 */
if ((scl_data->format == PIXEL_FORMAT_420BPP8) || (scl_data->format == PIXEL_FORMAT_420BPP10))
lb_config = LB_MEMORY_CONFIG_3;
else
lb_config = LB_MEMORY_CONFIG_0;
dpp->caps->dscl_calc_lb_num_partitions(
scl_data, lb_config, &num_part_y, &num_part_c);
/* MAX_V_TAPS = MIN (NUM_LINES - MAX(CEILING(V_RATIO,1)-2, 0), 8) */
if (dc_fixpt_ceil(scl_data->ratios.vert) > 2)
max_taps_y = num_part_y - (dc_fixpt_ceil(scl_data->ratios.vert) - 2);
else
max_taps_y = num_part_y;
if (dc_fixpt_ceil(scl_data->ratios.vert_c) > 2)
max_taps_c = num_part_c - (dc_fixpt_ceil(scl_data->ratios.vert_c) - 2);
else
max_taps_c = num_part_c;
if (max_taps_y < min_taps_y)
return false;
else if (max_taps_c < min_taps_c)
return false;
if (scl_data->taps.v_taps > max_taps_y)
scl_data->taps.v_taps = max_taps_y;
if (scl_data->taps.v_taps_c > max_taps_c)
scl_data->taps.v_taps_c = max_taps_c;
if (!dpp->ctx->dc->debug.always_scale) {
if (IDENTITY_RATIO(scl_data->ratios.horz))
scl_data->taps.h_taps = 1;
if (IDENTITY_RATIO(scl_data->ratios.vert))
scl_data->taps.v_taps = 1;
if (IDENTITY_RATIO(scl_data->ratios.horz_c))
scl_data->taps.h_taps_c = 1;
if (IDENTITY_RATIO(scl_data->ratios.vert_c))
scl_data->taps.v_taps_c = 1;
}
return true;
}
static void dpp3_deferred_update(struct dpp *dpp_base)
{
int bypass_state;
struct dcn3_dpp *dpp = TO_DCN30_DPP(dpp_base);
if (dpp_base->deferred_reg_writes.bits.disable_dscl) {
REG_UPDATE(DSCL_MEM_PWR_CTRL, LUT_MEM_PWR_FORCE, 3);
dpp_base->deferred_reg_writes.bits.disable_dscl = false;
}
if (dpp_base->deferred_reg_writes.bits.disable_gamcor) {
REG_GET(CM_GAMCOR_CONTROL, CM_GAMCOR_MODE_CURRENT, &bypass_state);
if (bypass_state == 0) { // only program if bypass was latched
REG_UPDATE(CM_MEM_PWR_CTRL, GAMCOR_MEM_PWR_FORCE, 3);
} else
ASSERT(0); // LUT select was updated again before vupdate
dpp_base->deferred_reg_writes.bits.disable_gamcor = false;
}
if (dpp_base->deferred_reg_writes.bits.disable_blnd_lut) {
REG_GET(CM_BLNDGAM_CONTROL, CM_BLNDGAM_MODE_CURRENT, &bypass_state);
if (bypass_state == 0) { // only program if bypass was latched
REG_UPDATE(CM_MEM_PWR_CTRL, BLNDGAM_MEM_PWR_FORCE, 3);
} else
ASSERT(0); // LUT select was updated again before vupdate
dpp_base->deferred_reg_writes.bits.disable_blnd_lut = false;
}
if (dpp_base->deferred_reg_writes.bits.disable_3dlut) {
REG_GET(CM_3DLUT_MODE, CM_3DLUT_MODE_CURRENT, &bypass_state);
if (bypass_state == 0) { // only program if bypass was latched
REG_UPDATE(CM_MEM_PWR_CTRL2, HDR3DLUT_MEM_PWR_FORCE, 3);
} else
ASSERT(0); // LUT select was updated again before vupdate
dpp_base->deferred_reg_writes.bits.disable_3dlut = false;
}
if (dpp_base->deferred_reg_writes.bits.disable_shaper) {
REG_GET(CM_SHAPER_CONTROL, CM_SHAPER_MODE_CURRENT, &bypass_state);
if (bypass_state == 0) { // only program if bypass was latched
REG_UPDATE(CM_MEM_PWR_CTRL2, SHAPER_MEM_PWR_FORCE, 3);
} else
ASSERT(0); // LUT select was updated again before vupdate
dpp_base->deferred_reg_writes.bits.disable_shaper = false;
}
}
static void dpp3_power_on_blnd_lut(
struct dpp *dpp_base,
bool power_on)
{
struct dcn3_dpp *dpp = TO_DCN30_DPP(dpp_base);
if (dpp_base->ctx->dc->debug.enable_mem_low_power.bits.cm) {
if (power_on) {
REG_UPDATE(CM_MEM_PWR_CTRL, BLNDGAM_MEM_PWR_FORCE, 0);
REG_WAIT(CM_MEM_PWR_STATUS, BLNDGAM_MEM_PWR_STATE, 0, 1, 5);
} else {
dpp_base->ctx->dc->optimized_required = true;
dpp_base->deferred_reg_writes.bits.disable_blnd_lut = true;
}
} else {
REG_SET(CM_MEM_PWR_CTRL, 0,
BLNDGAM_MEM_PWR_FORCE, power_on == true ? 0 : 1);
}
}
static void dpp3_power_on_hdr3dlut(
struct dpp *dpp_base,
bool power_on)
{
struct dcn3_dpp *dpp = TO_DCN30_DPP(dpp_base);
if (dpp_base->ctx->dc->debug.enable_mem_low_power.bits.cm) {
if (power_on) {
REG_UPDATE(CM_MEM_PWR_CTRL2, HDR3DLUT_MEM_PWR_FORCE, 0);
REG_WAIT(CM_MEM_PWR_STATUS2, HDR3DLUT_MEM_PWR_STATE, 0, 1, 5);
} else {
dpp_base->ctx->dc->optimized_required = true;
dpp_base->deferred_reg_writes.bits.disable_3dlut = true;
}
}
}
static void dpp3_power_on_shaper(
struct dpp *dpp_base,
bool power_on)
{
struct dcn3_dpp *dpp = TO_DCN30_DPP(dpp_base);
if (dpp_base->ctx->dc->debug.enable_mem_low_power.bits.cm) {
if (power_on) {
REG_UPDATE(CM_MEM_PWR_CTRL2, SHAPER_MEM_PWR_FORCE, 0);
REG_WAIT(CM_MEM_PWR_STATUS2, SHAPER_MEM_PWR_STATE, 0, 1, 5);
} else {
dpp_base->ctx->dc->optimized_required = true;
dpp_base->deferred_reg_writes.bits.disable_shaper = true;
}
}
}
static void dpp3_configure_blnd_lut(
struct dpp *dpp_base,
bool is_ram_a)
{
struct dcn3_dpp *dpp = TO_DCN30_DPP(dpp_base);
REG_UPDATE_2(CM_BLNDGAM_LUT_CONTROL,
CM_BLNDGAM_LUT_WRITE_COLOR_MASK, 7,
CM_BLNDGAM_LUT_HOST_SEL, is_ram_a == true ? 0 : 1);
REG_SET(CM_BLNDGAM_LUT_INDEX, 0, CM_BLNDGAM_LUT_INDEX, 0);
}
static void dpp3_program_blnd_pwl(
struct dpp *dpp_base,
const struct pwl_result_data *rgb,
uint32_t num)
{
uint32_t i;
struct dcn3_dpp *dpp = TO_DCN30_DPP(dpp_base);
uint32_t last_base_value_red = rgb[num-1].red_reg + rgb[num-1].delta_red_reg;
uint32_t last_base_value_green = rgb[num-1].green_reg + rgb[num-1].delta_green_reg;
uint32_t last_base_value_blue = rgb[num-1].blue_reg + rgb[num-1].delta_blue_reg;
if (is_rgb_equal(rgb, num)) {
for (i = 0 ; i < num; i++)
REG_SET(CM_BLNDGAM_LUT_DATA, 0, CM_BLNDGAM_LUT_DATA, rgb[i].red_reg);
REG_SET(CM_BLNDGAM_LUT_DATA, 0, CM_BLNDGAM_LUT_DATA, last_base_value_red);
} else {
REG_SET(CM_BLNDGAM_LUT_INDEX, 0, CM_BLNDGAM_LUT_INDEX, 0);
REG_UPDATE(CM_BLNDGAM_LUT_CONTROL, CM_BLNDGAM_LUT_WRITE_COLOR_MASK, 4);
for (i = 0 ; i < num; i++)
REG_SET(CM_BLNDGAM_LUT_DATA, 0, CM_BLNDGAM_LUT_DATA, rgb[i].red_reg);
REG_SET(CM_BLNDGAM_LUT_DATA, 0, CM_BLNDGAM_LUT_DATA, last_base_value_red);
REG_SET(CM_BLNDGAM_LUT_INDEX, 0, CM_BLNDGAM_LUT_INDEX, 0);
REG_UPDATE(CM_BLNDGAM_LUT_CONTROL, CM_BLNDGAM_LUT_WRITE_COLOR_MASK, 2);
for (i = 0 ; i < num; i++)
REG_SET(CM_BLNDGAM_LUT_DATA, 0, CM_BLNDGAM_LUT_DATA, rgb[i].green_reg);
REG_SET(CM_BLNDGAM_LUT_DATA, 0, CM_BLNDGAM_LUT_DATA, last_base_value_green);
REG_SET(CM_BLNDGAM_LUT_INDEX, 0, CM_BLNDGAM_LUT_INDEX, 0);
REG_UPDATE(CM_BLNDGAM_LUT_CONTROL, CM_BLNDGAM_LUT_WRITE_COLOR_MASK, 1);
for (i = 0 ; i < num; i++)
REG_SET(CM_BLNDGAM_LUT_DATA, 0, CM_BLNDGAM_LUT_DATA, rgb[i].blue_reg);
REG_SET(CM_BLNDGAM_LUT_DATA, 0, CM_BLNDGAM_LUT_DATA, last_base_value_blue);
}
}
static void dcn3_dpp_cm_get_reg_field(
struct dcn3_dpp *dpp,
struct dcn3_xfer_func_reg *reg)
{
reg->shifts.exp_region0_lut_offset = dpp->tf_shift->CM_BLNDGAM_RAMA_EXP_REGION0_LUT_OFFSET;
reg->masks.exp_region0_lut_offset = dpp->tf_mask->CM_BLNDGAM_RAMA_EXP_REGION0_LUT_OFFSET;
reg->shifts.exp_region0_num_segments = dpp->tf_shift->CM_BLNDGAM_RAMA_EXP_REGION0_NUM_SEGMENTS;
reg->masks.exp_region0_num_segments = dpp->tf_mask->CM_BLNDGAM_RAMA_EXP_REGION0_NUM_SEGMENTS;
reg->shifts.exp_region1_lut_offset = dpp->tf_shift->CM_BLNDGAM_RAMA_EXP_REGION1_LUT_OFFSET;
reg->masks.exp_region1_lut_offset = dpp->tf_mask->CM_BLNDGAM_RAMA_EXP_REGION1_LUT_OFFSET;
reg->shifts.exp_region1_num_segments = dpp->tf_shift->CM_BLNDGAM_RAMA_EXP_REGION1_NUM_SEGMENTS;
reg->masks.exp_region1_num_segments = dpp->tf_mask->CM_BLNDGAM_RAMA_EXP_REGION1_NUM_SEGMENTS;
reg->shifts.field_region_end = dpp->tf_shift->CM_BLNDGAM_RAMA_EXP_REGION_END_B;
reg->masks.field_region_end = dpp->tf_mask->CM_BLNDGAM_RAMA_EXP_REGION_END_B;
reg->shifts.field_region_end_slope = dpp->tf_shift->CM_BLNDGAM_RAMA_EXP_REGION_END_SLOPE_B;
reg->masks.field_region_end_slope = dpp->tf_mask->CM_BLNDGAM_RAMA_EXP_REGION_END_SLOPE_B;
reg->shifts.field_region_end_base = dpp->tf_shift->CM_BLNDGAM_RAMA_EXP_REGION_END_BASE_B;
reg->masks.field_region_end_base = dpp->tf_mask->CM_BLNDGAM_RAMA_EXP_REGION_END_BASE_B;
reg->shifts.field_region_linear_slope = dpp->tf_shift->CM_BLNDGAM_RAMA_EXP_REGION_START_SLOPE_B;
reg->masks.field_region_linear_slope = dpp->tf_mask->CM_BLNDGAM_RAMA_EXP_REGION_START_SLOPE_B;
reg->shifts.exp_region_start = dpp->tf_shift->CM_BLNDGAM_RAMA_EXP_REGION_START_B;
reg->masks.exp_region_start = dpp->tf_mask->CM_BLNDGAM_RAMA_EXP_REGION_START_B;
reg->shifts.exp_resion_start_segment = dpp->tf_shift->CM_BLNDGAM_RAMA_EXP_REGION_START_SEGMENT_B;
reg->masks.exp_resion_start_segment = dpp->tf_mask->CM_BLNDGAM_RAMA_EXP_REGION_START_SEGMENT_B;
}
/*program blnd lut RAM A*/
static void dpp3_program_blnd_luta_settings(
struct dpp *dpp_base,
const struct pwl_params *params)
{
struct dcn3_dpp *dpp = TO_DCN30_DPP(dpp_base);
struct dcn3_xfer_func_reg gam_regs;
dcn3_dpp_cm_get_reg_field(dpp, &gam_regs);
gam_regs.start_cntl_b = REG(CM_BLNDGAM_RAMA_START_CNTL_B);
gam_regs.start_cntl_g = REG(CM_BLNDGAM_RAMA_START_CNTL_G);
gam_regs.start_cntl_r = REG(CM_BLNDGAM_RAMA_START_CNTL_R);
gam_regs.start_slope_cntl_b = REG(CM_BLNDGAM_RAMA_START_SLOPE_CNTL_B);
gam_regs.start_slope_cntl_g = REG(CM_BLNDGAM_RAMA_START_SLOPE_CNTL_G);
gam_regs.start_slope_cntl_r = REG(CM_BLNDGAM_RAMA_START_SLOPE_CNTL_R);
gam_regs.start_end_cntl1_b = REG(CM_BLNDGAM_RAMA_END_CNTL1_B);
gam_regs.start_end_cntl2_b = REG(CM_BLNDGAM_RAMA_END_CNTL2_B);
gam_regs.start_end_cntl1_g = REG(CM_BLNDGAM_RAMA_END_CNTL1_G);
gam_regs.start_end_cntl2_g = REG(CM_BLNDGAM_RAMA_END_CNTL2_G);
gam_regs.start_end_cntl1_r = REG(CM_BLNDGAM_RAMA_END_CNTL1_R);
gam_regs.start_end_cntl2_r = REG(CM_BLNDGAM_RAMA_END_CNTL2_R);
gam_regs.region_start = REG(CM_BLNDGAM_RAMA_REGION_0_1);
gam_regs.region_end = REG(CM_BLNDGAM_RAMA_REGION_32_33);
cm_helper_program_gamcor_xfer_func(dpp->base.ctx, params, &gam_regs);
}
/*program blnd lut RAM B*/
static void dpp3_program_blnd_lutb_settings(
struct dpp *dpp_base,
const struct pwl_params *params)
{
struct dcn3_dpp *dpp = TO_DCN30_DPP(dpp_base);
struct dcn3_xfer_func_reg gam_regs;
dcn3_dpp_cm_get_reg_field(dpp, &gam_regs);
gam_regs.start_cntl_b = REG(CM_BLNDGAM_RAMB_START_CNTL_B);
gam_regs.start_cntl_g = REG(CM_BLNDGAM_RAMB_START_CNTL_G);
gam_regs.start_cntl_r = REG(CM_BLNDGAM_RAMB_START_CNTL_R);
gam_regs.start_slope_cntl_b = REG(CM_BLNDGAM_RAMB_START_SLOPE_CNTL_B);
gam_regs.start_slope_cntl_g = REG(CM_BLNDGAM_RAMB_START_SLOPE_CNTL_G);
gam_regs.start_slope_cntl_r = REG(CM_BLNDGAM_RAMB_START_SLOPE_CNTL_R);
gam_regs.start_end_cntl1_b = REG(CM_BLNDGAM_RAMB_END_CNTL1_B);
gam_regs.start_end_cntl2_b = REG(CM_BLNDGAM_RAMB_END_CNTL2_B);
gam_regs.start_end_cntl1_g = REG(CM_BLNDGAM_RAMB_END_CNTL1_G);
gam_regs.start_end_cntl2_g = REG(CM_BLNDGAM_RAMB_END_CNTL2_G);
gam_regs.start_end_cntl1_r = REG(CM_BLNDGAM_RAMB_END_CNTL1_R);
gam_regs.start_end_cntl2_r = REG(CM_BLNDGAM_RAMB_END_CNTL2_R);
gam_regs.region_start = REG(CM_BLNDGAM_RAMB_REGION_0_1);
gam_regs.region_end = REG(CM_BLNDGAM_RAMB_REGION_32_33);
cm_helper_program_gamcor_xfer_func(dpp->base.ctx, params, &gam_regs);
}
static enum dc_lut_mode dpp3_get_blndgam_current(struct dpp *dpp_base)
{
enum dc_lut_mode mode;
uint32_t mode_current = 0;
uint32_t in_use = 0;
struct dcn3_dpp *dpp = TO_DCN30_DPP(dpp_base);
REG_GET(CM_BLNDGAM_CONTROL, CM_BLNDGAM_MODE_CURRENT, &mode_current);
REG_GET(CM_BLNDGAM_CONTROL, CM_BLNDGAM_SELECT_CURRENT, &in_use);
switch (mode_current) {
case 0:
case 1:
mode = LUT_BYPASS;
break;
case 2:
if (in_use == 0)
mode = LUT_RAM_A;
else
mode = LUT_RAM_B;
break;
default:
mode = LUT_BYPASS;
break;
}
return mode;
}
static bool dpp3_program_blnd_lut(struct dpp *dpp_base,
const struct pwl_params *params)
{
enum dc_lut_mode current_mode;
enum dc_lut_mode next_mode;
struct dcn3_dpp *dpp = TO_DCN30_DPP(dpp_base);
if (params == NULL) {
REG_SET(CM_BLNDGAM_CONTROL, 0, CM_BLNDGAM_MODE, 0);
if (dpp_base->ctx->dc->debug.enable_mem_low_power.bits.cm)
dpp3_power_on_blnd_lut(dpp_base, false);
return false;
}
current_mode = dpp3_get_blndgam_current(dpp_base);
if (current_mode == LUT_BYPASS || current_mode == LUT_RAM_B)
next_mode = LUT_RAM_A;
else
next_mode = LUT_RAM_B;
dpp3_power_on_blnd_lut(dpp_base, true);
dpp3_configure_blnd_lut(dpp_base, next_mode == LUT_RAM_A);
if (next_mode == LUT_RAM_A)
dpp3_program_blnd_luta_settings(dpp_base, params);
else
dpp3_program_blnd_lutb_settings(dpp_base, params);
dpp3_program_blnd_pwl(
dpp_base, params->rgb_resulted, params->hw_points_num);
REG_UPDATE_2(CM_BLNDGAM_CONTROL,
CM_BLNDGAM_MODE, 2,
CM_BLNDGAM_SELECT, next_mode == LUT_RAM_A ? 0 : 1);
return true;
}
static void dpp3_program_shaper_lut(
struct dpp *dpp_base,
const struct pwl_result_data *rgb,
uint32_t num)
{
uint32_t i, red, green, blue;
uint32_t red_delta, green_delta, blue_delta;
uint32_t red_value, green_value, blue_value;
struct dcn3_dpp *dpp = TO_DCN30_DPP(dpp_base);
for (i = 0 ; i < num; i++) {
red = rgb[i].red_reg;
green = rgb[i].green_reg;
blue = rgb[i].blue_reg;
red_delta = rgb[i].delta_red_reg;
green_delta = rgb[i].delta_green_reg;
blue_delta = rgb[i].delta_blue_reg;
red_value = ((red_delta & 0x3ff) << 14) | (red & 0x3fff);
green_value = ((green_delta & 0x3ff) << 14) | (green & 0x3fff);
blue_value = ((blue_delta & 0x3ff) << 14) | (blue & 0x3fff);
REG_SET(CM_SHAPER_LUT_DATA, 0, CM_SHAPER_LUT_DATA, red_value);
REG_SET(CM_SHAPER_LUT_DATA, 0, CM_SHAPER_LUT_DATA, green_value);
REG_SET(CM_SHAPER_LUT_DATA, 0, CM_SHAPER_LUT_DATA, blue_value);
}
}
static enum dc_lut_mode dpp3_get_shaper_current(struct dpp *dpp_base)
{
enum dc_lut_mode mode;
uint32_t state_mode;
struct dcn3_dpp *dpp = TO_DCN30_DPP(dpp_base);
REG_GET(CM_SHAPER_CONTROL, CM_SHAPER_MODE_CURRENT, &state_mode);
switch (state_mode) {
case 0:
mode = LUT_BYPASS;
break;
case 1:
mode = LUT_RAM_A;
break;
case 2:
mode = LUT_RAM_B;
break;
default:
mode = LUT_BYPASS;
break;
}
return mode;
}
static void dpp3_configure_shaper_lut(
struct dpp *dpp_base,
bool is_ram_a)
{
struct dcn3_dpp *dpp = TO_DCN30_DPP(dpp_base);
REG_UPDATE(CM_SHAPER_LUT_WRITE_EN_MASK,
CM_SHAPER_LUT_WRITE_EN_MASK, 7);
REG_UPDATE(CM_SHAPER_LUT_WRITE_EN_MASK,
CM_SHAPER_LUT_WRITE_SEL, is_ram_a == true ? 0:1);
REG_SET(CM_SHAPER_LUT_INDEX, 0, CM_SHAPER_LUT_INDEX, 0);
}
/*program shaper RAM A*/
static void dpp3_program_shaper_luta_settings(
struct dpp *dpp_base,
const struct pwl_params *params)
{
const struct gamma_curve *curve;
struct dcn3_dpp *dpp = TO_DCN30_DPP(dpp_base);
REG_SET_2(CM_SHAPER_RAMA_START_CNTL_B, 0,
CM_SHAPER_RAMA_EXP_REGION_START_B, params->corner_points[0].blue.custom_float_x,
CM_SHAPER_RAMA_EXP_REGION_START_SEGMENT_B, 0);
REG_SET_2(CM_SHAPER_RAMA_START_CNTL_G, 0,
CM_SHAPER_RAMA_EXP_REGION_START_G, params->corner_points[0].green.custom_float_x,
CM_SHAPER_RAMA_EXP_REGION_START_SEGMENT_G, 0);
REG_SET_2(CM_SHAPER_RAMA_START_CNTL_R, 0,
CM_SHAPER_RAMA_EXP_REGION_START_R, params->corner_points[0].red.custom_float_x,
CM_SHAPER_RAMA_EXP_REGION_START_SEGMENT_R, 0);
REG_SET_2(CM_SHAPER_RAMA_END_CNTL_B, 0,
CM_SHAPER_RAMA_EXP_REGION_END_B, params->corner_points[1].blue.custom_float_x,
CM_SHAPER_RAMA_EXP_REGION_END_BASE_B, params->corner_points[1].blue.custom_float_y);
REG_SET_2(CM_SHAPER_RAMA_END_CNTL_G, 0,
CM_SHAPER_RAMA_EXP_REGION_END_G, params->corner_points[1].green.custom_float_x,
CM_SHAPER_RAMA_EXP_REGION_END_BASE_G, params->corner_points[1].green.custom_float_y);
REG_SET_2(CM_SHAPER_RAMA_END_CNTL_R, 0,
CM_SHAPER_RAMA_EXP_REGION_END_R, params->corner_points[1].red.custom_float_x,
CM_SHAPER_RAMA_EXP_REGION_END_BASE_R, params->corner_points[1].red.custom_float_y);
curve = params->arr_curve_points;
REG_SET_4(CM_SHAPER_RAMA_REGION_0_1, 0,
CM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
CM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
CM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
CM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
curve += 2;
REG_SET_4(CM_SHAPER_RAMA_REGION_2_3, 0,
CM_SHAPER_RAMA_EXP_REGION2_LUT_OFFSET, curve[0].offset,
CM_SHAPER_RAMA_EXP_REGION2_NUM_SEGMENTS, curve[0].segments_num,
CM_SHAPER_RAMA_EXP_REGION3_LUT_OFFSET, curve[1].offset,
CM_SHAPER_RAMA_EXP_REGION3_NUM_SEGMENTS, curve[1].segments_num);
curve += 2;
REG_SET_4(CM_SHAPER_RAMA_REGION_4_5, 0,
CM_SHAPER_RAMA_EXP_REGION4_LUT_OFFSET, curve[0].offset,
CM_SHAPER_RAMA_EXP_REGION4_NUM_SEGMENTS, curve[0].segments_num,
CM_SHAPER_RAMA_EXP_REGION5_LUT_OFFSET, curve[1].offset,
CM_SHAPER_RAMA_EXP_REGION5_NUM_SEGMENTS, curve[1].segments_num);
curve += 2;
REG_SET_4(CM_SHAPER_RAMA_REGION_6_7, 0,
CM_SHAPER_RAMA_EXP_REGION6_LUT_OFFSET, curve[0].offset,
CM_SHAPER_RAMA_EXP_REGION6_NUM_SEGMENTS, curve[0].segments_num,
CM_SHAPER_RAMA_EXP_REGION7_LUT_OFFSET, curve[1].offset,
CM_SHAPER_RAMA_EXP_REGION7_NUM_SEGMENTS, curve[1].segments_num);
curve += 2;
REG_SET_4(CM_SHAPER_RAMA_REGION_8_9, 0,
CM_SHAPER_RAMA_EXP_REGION8_LUT_OFFSET, curve[0].offset,
CM_SHAPER_RAMA_EXP_REGION8_NUM_SEGMENTS, curve[0].segments_num,
CM_SHAPER_RAMA_EXP_REGION9_LUT_OFFSET, curve[1].offset,
CM_SHAPER_RAMA_EXP_REGION9_NUM_SEGMENTS, curve[1].segments_num);
curve += 2;
REG_SET_4(CM_SHAPER_RAMA_REGION_10_11, 0,
CM_SHAPER_RAMA_EXP_REGION10_LUT_OFFSET, curve[0].offset,
CM_SHAPER_RAMA_EXP_REGION10_NUM_SEGMENTS, curve[0].segments_num,
CM_SHAPER_RAMA_EXP_REGION11_LUT_OFFSET, curve[1].offset,
CM_SHAPER_RAMA_EXP_REGION11_NUM_SEGMENTS, curve[1].segments_num);
curve += 2;
REG_SET_4(CM_SHAPER_RAMA_REGION_12_13, 0,
CM_SHAPER_RAMA_EXP_REGION12_LUT_OFFSET, curve[0].offset,
CM_SHAPER_RAMA_EXP_REGION12_NUM_SEGMENTS, curve[0].segments_num,
CM_SHAPER_RAMA_EXP_REGION13_LUT_OFFSET, curve[1].offset,
CM_SHAPER_RAMA_EXP_REGION13_NUM_SEGMENTS, curve[1].segments_num);
curve += 2;
REG_SET_4(CM_SHAPER_RAMA_REGION_14_15, 0,
CM_SHAPER_RAMA_EXP_REGION14_LUT_OFFSET, curve[0].offset,
CM_SHAPER_RAMA_EXP_REGION14_NUM_SEGMENTS, curve[0].segments_num,
CM_SHAPER_RAMA_EXP_REGION15_LUT_OFFSET, curve[1].offset,
CM_SHAPER_RAMA_EXP_REGION15_NUM_SEGMENTS, curve[1].segments_num);
curve += 2;
REG_SET_4(CM_SHAPER_RAMA_REGION_16_17, 0,
CM_SHAPER_RAMA_EXP_REGION16_LUT_OFFSET, curve[0].offset,
CM_SHAPER_RAMA_EXP_REGION16_NUM_SEGMENTS, curve[0].segments_num,
CM_SHAPER_RAMA_EXP_REGION17_LUT_OFFSET, curve[1].offset,
CM_SHAPER_RAMA_EXP_REGION17_NUM_SEGMENTS, curve[1].segments_num);
curve += 2;
REG_SET_4(CM_SHAPER_RAMA_REGION_18_19, 0,
CM_SHAPER_RAMA_EXP_REGION18_LUT_OFFSET, curve[0].offset,
CM_SHAPER_RAMA_EXP_REGION18_NUM_SEGMENTS, curve[0].segments_num,
CM_SHAPER_RAMA_EXP_REGION19_LUT_OFFSET, curve[1].offset,
CM_SHAPER_RAMA_EXP_REGION19_NUM_SEGMENTS, curve[1].segments_num);
curve += 2;
REG_SET_4(CM_SHAPER_RAMA_REGION_20_21, 0,
CM_SHAPER_RAMA_EXP_REGION20_LUT_OFFSET, curve[0].offset,
CM_SHAPER_RAMA_EXP_REGION20_NUM_SEGMENTS, curve[0].segments_num,
CM_SHAPER_RAMA_EXP_REGION21_LUT_OFFSET, curve[1].offset,
CM_SHAPER_RAMA_EXP_REGION21_NUM_SEGMENTS, curve[1].segments_num);
curve += 2;
REG_SET_4(CM_SHAPER_RAMA_REGION_22_23, 0,
CM_SHAPER_RAMA_EXP_REGION22_LUT_OFFSET, curve[0].offset,
CM_SHAPER_RAMA_EXP_REGION22_NUM_SEGMENTS, curve[0].segments_num,
CM_SHAPER_RAMA_EXP_REGION23_LUT_OFFSET, curve[1].offset,
CM_SHAPER_RAMA_EXP_REGION23_NUM_SEGMENTS, curve[1].segments_num);
curve += 2;
REG_SET_4(CM_SHAPER_RAMA_REGION_24_25, 0,
CM_SHAPER_RAMA_EXP_REGION24_LUT_OFFSET, curve[0].offset,
CM_SHAPER_RAMA_EXP_REGION24_NUM_SEGMENTS, curve[0].segments_num,
CM_SHAPER_RAMA_EXP_REGION25_LUT_OFFSET, curve[1].offset,
CM_SHAPER_RAMA_EXP_REGION25_NUM_SEGMENTS, curve[1].segments_num);
curve += 2;
REG_SET_4(CM_SHAPER_RAMA_REGION_26_27, 0,
CM_SHAPER_RAMA_EXP_REGION26_LUT_OFFSET, curve[0].offset,
CM_SHAPER_RAMA_EXP_REGION26_NUM_SEGMENTS, curve[0].segments_num,
CM_SHAPER_RAMA_EXP_REGION27_LUT_OFFSET, curve[1].offset,
CM_SHAPER_RAMA_EXP_REGION27_NUM_SEGMENTS, curve[1].segments_num);
curve += 2;
REG_SET_4(CM_SHAPER_RAMA_REGION_28_29, 0,
CM_SHAPER_RAMA_EXP_REGION28_LUT_OFFSET, curve[0].offset,
CM_SHAPER_RAMA_EXP_REGION28_NUM_SEGMENTS, curve[0].segments_num,
CM_SHAPER_RAMA_EXP_REGION29_LUT_OFFSET, curve[1].offset,
CM_SHAPER_RAMA_EXP_REGION29_NUM_SEGMENTS, curve[1].segments_num);
curve += 2;
REG_SET_4(CM_SHAPER_RAMA_REGION_30_31, 0,
CM_SHAPER_RAMA_EXP_REGION30_LUT_OFFSET, curve[0].offset,
CM_SHAPER_RAMA_EXP_REGION30_NUM_SEGMENTS, curve[0].segments_num,
CM_SHAPER_RAMA_EXP_REGION31_LUT_OFFSET, curve[1].offset,
CM_SHAPER_RAMA_EXP_REGION31_NUM_SEGMENTS, curve[1].segments_num);
curve += 2;
REG_SET_4(CM_SHAPER_RAMA_REGION_32_33, 0,
CM_SHAPER_RAMA_EXP_REGION32_LUT_OFFSET, curve[0].offset,
CM_SHAPER_RAMA_EXP_REGION32_NUM_SEGMENTS, curve[0].segments_num,
CM_SHAPER_RAMA_EXP_REGION33_LUT_OFFSET, curve[1].offset,
CM_SHAPER_RAMA_EXP_REGION33_NUM_SEGMENTS, curve[1].segments_num);
}
/*program shaper RAM B*/
static void dpp3_program_shaper_lutb_settings(
struct dpp *dpp_base,
const struct pwl_params *params)
{
const struct gamma_curve *curve;
struct dcn3_dpp *dpp = TO_DCN30_DPP(dpp_base);
REG_SET_2(CM_SHAPER_RAMB_START_CNTL_B, 0,
CM_SHAPER_RAMB_EXP_REGION_START_B, params->corner_points[0].blue.custom_float_x,
CM_SHAPER_RAMB_EXP_REGION_START_SEGMENT_B, 0);
REG_SET_2(CM_SHAPER_RAMB_START_CNTL_G, 0,
CM_SHAPER_RAMB_EXP_REGION_START_G, params->corner_points[0].green.custom_float_x,
CM_SHAPER_RAMB_EXP_REGION_START_SEGMENT_G, 0);
REG_SET_2(CM_SHAPER_RAMB_START_CNTL_R, 0,
CM_SHAPER_RAMB_EXP_REGION_START_R, params->corner_points[0].red.custom_float_x,
CM_SHAPER_RAMB_EXP_REGION_START_SEGMENT_R, 0);
REG_SET_2(CM_SHAPER_RAMB_END_CNTL_B, 0,
CM_SHAPER_RAMB_EXP_REGION_END_B, params->corner_points[1].blue.custom_float_x,
CM_SHAPER_RAMB_EXP_REGION_END_BASE_B, params->corner_points[1].blue.custom_float_y);
REG_SET_2(CM_SHAPER_RAMB_END_CNTL_G, 0,
CM_SHAPER_RAMB_EXP_REGION_END_G, params->corner_points[1].green.custom_float_x,
CM_SHAPER_RAMB_EXP_REGION_END_BASE_G, params->corner_points[1].green.custom_float_y);
REG_SET_2(CM_SHAPER_RAMB_END_CNTL_R, 0,
CM_SHAPER_RAMB_EXP_REGION_END_R, params->corner_points[1].red.custom_float_x,
CM_SHAPER_RAMB_EXP_REGION_END_BASE_R, params->corner_points[1].red.custom_float_y);
curve = params->arr_curve_points;
REG_SET_4(CM_SHAPER_RAMB_REGION_0_1, 0,
CM_SHAPER_RAMB_EXP_REGION0_LUT_OFFSET, curve[0].offset,
CM_SHAPER_RAMB_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
CM_SHAPER_RAMB_EXP_REGION1_LUT_OFFSET, curve[1].offset,
CM_SHAPER_RAMB_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
curve += 2;
REG_SET_4(CM_SHAPER_RAMB_REGION_2_3, 0,
CM_SHAPER_RAMB_EXP_REGION2_LUT_OFFSET, curve[0].offset,
CM_SHAPER_RAMB_EXP_REGION2_NUM_SEGMENTS, curve[0].segments_num,
CM_SHAPER_RAMB_EXP_REGION3_LUT_OFFSET, curve[1].offset,
CM_SHAPER_RAMB_EXP_REGION3_NUM_SEGMENTS, curve[1].segments_num);
curve += 2;
REG_SET_4(CM_SHAPER_RAMB_REGION_4_5, 0,
CM_SHAPER_RAMB_EXP_REGION4_LUT_OFFSET, curve[0].offset,
CM_SHAPER_RAMB_EXP_REGION4_NUM_SEGMENTS, curve[0].segments_num,
CM_SHAPER_RAMB_EXP_REGION5_LUT_OFFSET, curve[1].offset,
CM_SHAPER_RAMB_EXP_REGION5_NUM_SEGMENTS, curve[1].segments_num);
curve += 2;
REG_SET_4(CM_SHAPER_RAMB_REGION_6_7, 0,
CM_SHAPER_RAMB_EXP_REGION6_LUT_OFFSET, curve[0].offset,
CM_SHAPER_RAMB_EXP_REGION6_NUM_SEGMENTS, curve[0].segments_num,
CM_SHAPER_RAMB_EXP_REGION7_LUT_OFFSET, curve[1].offset,
CM_SHAPER_RAMB_EXP_REGION7_NUM_SEGMENTS, curve[1].segments_num);
curve += 2;
REG_SET_4(CM_SHAPER_RAMB_REGION_8_9, 0,
CM_SHAPER_RAMB_EXP_REGION8_LUT_OFFSET, curve[0].offset,
CM_SHAPER_RAMB_EXP_REGION8_NUM_SEGMENTS, curve[0].segments_num,
CM_SHAPER_RAMB_EXP_REGION9_LUT_OFFSET, curve[1].offset,
CM_SHAPER_RAMB_EXP_REGION9_NUM_SEGMENTS, curve[1].segments_num);
curve += 2;
REG_SET_4(CM_SHAPER_RAMB_REGION_10_11, 0,
CM_SHAPER_RAMB_EXP_REGION10_LUT_OFFSET, curve[0].offset,
CM_SHAPER_RAMB_EXP_REGION10_NUM_SEGMENTS, curve[0].segments_num,
CM_SHAPER_RAMB_EXP_REGION11_LUT_OFFSET, curve[1].offset,
CM_SHAPER_RAMB_EXP_REGION11_NUM_SEGMENTS, curve[1].segments_num);
curve += 2;
REG_SET_4(CM_SHAPER_RAMB_REGION_12_13, 0,
CM_SHAPER_RAMB_EXP_REGION12_LUT_OFFSET, curve[0].offset,
CM_SHAPER_RAMB_EXP_REGION12_NUM_SEGMENTS, curve[0].segments_num,
CM_SHAPER_RAMB_EXP_REGION13_LUT_OFFSET, curve[1].offset,
CM_SHAPER_RAMB_EXP_REGION13_NUM_SEGMENTS, curve[1].segments_num);
curve += 2;
REG_SET_4(CM_SHAPER_RAMB_REGION_14_15, 0,
CM_SHAPER_RAMB_EXP_REGION14_LUT_OFFSET, curve[0].offset,
CM_SHAPER_RAMB_EXP_REGION14_NUM_SEGMENTS, curve[0].segments_num,
CM_SHAPER_RAMB_EXP_REGION15_LUT_OFFSET, curve[1].offset,
CM_SHAPER_RAMB_EXP_REGION15_NUM_SEGMENTS, curve[1].segments_num);
curve += 2;
REG_SET_4(CM_SHAPER_RAMB_REGION_16_17, 0,
CM_SHAPER_RAMB_EXP_REGION16_LUT_OFFSET, curve[0].offset,
CM_SHAPER_RAMB_EXP_REGION16_NUM_SEGMENTS, curve[0].segments_num,
CM_SHAPER_RAMB_EXP_REGION17_LUT_OFFSET, curve[1].offset,
CM_SHAPER_RAMB_EXP_REGION17_NUM_SEGMENTS, curve[1].segments_num);
curve += 2;
REG_SET_4(CM_SHAPER_RAMB_REGION_18_19, 0,
CM_SHAPER_RAMB_EXP_REGION18_LUT_OFFSET, curve[0].offset,
CM_SHAPER_RAMB_EXP_REGION18_NUM_SEGMENTS, curve[0].segments_num,
CM_SHAPER_RAMB_EXP_REGION19_LUT_OFFSET, curve[1].offset,
CM_SHAPER_RAMB_EXP_REGION19_NUM_SEGMENTS, curve[1].segments_num);
curve += 2;
REG_SET_4(CM_SHAPER_RAMB_REGION_20_21, 0,
CM_SHAPER_RAMB_EXP_REGION20_LUT_OFFSET, curve[0].offset,
CM_SHAPER_RAMB_EXP_REGION20_NUM_SEGMENTS, curve[0].segments_num,
CM_SHAPER_RAMB_EXP_REGION21_LUT_OFFSET, curve[1].offset,
CM_SHAPER_RAMB_EXP_REGION21_NUM_SEGMENTS, curve[1].segments_num);
curve += 2;
REG_SET_4(CM_SHAPER_RAMB_REGION_22_23, 0,
CM_SHAPER_RAMB_EXP_REGION22_LUT_OFFSET, curve[0].offset,
CM_SHAPER_RAMB_EXP_REGION22_NUM_SEGMENTS, curve[0].segments_num,
CM_SHAPER_RAMB_EXP_REGION23_LUT_OFFSET, curve[1].offset,
CM_SHAPER_RAMB_EXP_REGION23_NUM_SEGMENTS, curve[1].segments_num);
curve += 2;
REG_SET_4(CM_SHAPER_RAMB_REGION_24_25, 0,
CM_SHAPER_RAMB_EXP_REGION24_LUT_OFFSET, curve[0].offset,
CM_SHAPER_RAMB_EXP_REGION24_NUM_SEGMENTS, curve[0].segments_num,
CM_SHAPER_RAMB_EXP_REGION25_LUT_OFFSET, curve[1].offset,
CM_SHAPER_RAMB_EXP_REGION25_NUM_SEGMENTS, curve[1].segments_num);
curve += 2;
REG_SET_4(CM_SHAPER_RAMB_REGION_26_27, 0,
CM_SHAPER_RAMB_EXP_REGION26_LUT_OFFSET, curve[0].offset,
CM_SHAPER_RAMB_EXP_REGION26_NUM_SEGMENTS, curve[0].segments_num,
CM_SHAPER_RAMB_EXP_REGION27_LUT_OFFSET, curve[1].offset,
CM_SHAPER_RAMB_EXP_REGION27_NUM_SEGMENTS, curve[1].segments_num);
curve += 2;
REG_SET_4(CM_SHAPER_RAMB_REGION_28_29, 0,
CM_SHAPER_RAMB_EXP_REGION28_LUT_OFFSET, curve[0].offset,
CM_SHAPER_RAMB_EXP_REGION28_NUM_SEGMENTS, curve[0].segments_num,
CM_SHAPER_RAMB_EXP_REGION29_LUT_OFFSET, curve[1].offset,
CM_SHAPER_RAMB_EXP_REGION29_NUM_SEGMENTS, curve[1].segments_num);
curve += 2;
REG_SET_4(CM_SHAPER_RAMB_REGION_30_31, 0,
CM_SHAPER_RAMB_EXP_REGION30_LUT_OFFSET, curve[0].offset,
CM_SHAPER_RAMB_EXP_REGION30_NUM_SEGMENTS, curve[0].segments_num,
CM_SHAPER_RAMB_EXP_REGION31_LUT_OFFSET, curve[1].offset,
CM_SHAPER_RAMB_EXP_REGION31_NUM_SEGMENTS, curve[1].segments_num);
curve += 2;
REG_SET_4(CM_SHAPER_RAMB_REGION_32_33, 0,
CM_SHAPER_RAMB_EXP_REGION32_LUT_OFFSET, curve[0].offset,
CM_SHAPER_RAMB_EXP_REGION32_NUM_SEGMENTS, curve[0].segments_num,
CM_SHAPER_RAMB_EXP_REGION33_LUT_OFFSET, curve[1].offset,
CM_SHAPER_RAMB_EXP_REGION33_NUM_SEGMENTS, curve[1].segments_num);
}
static bool dpp3_program_shaper(struct dpp *dpp_base,
const struct pwl_params *params)
{
enum dc_lut_mode current_mode;
enum dc_lut_mode next_mode;
struct dcn3_dpp *dpp = TO_DCN30_DPP(dpp_base);
if (params == NULL) {
REG_SET(CM_SHAPER_CONTROL, 0, CM_SHAPER_LUT_MODE, 0);
if (dpp_base->ctx->dc->debug.enable_mem_low_power.bits.cm)
dpp3_power_on_shaper(dpp_base, false);
return false;
}
if (dpp_base->ctx->dc->debug.enable_mem_low_power.bits.cm)
dpp3_power_on_shaper(dpp_base, true);
current_mode = dpp3_get_shaper_current(dpp_base);
if (current_mode == LUT_BYPASS || current_mode == LUT_RAM_A)
next_mode = LUT_RAM_B;
else
next_mode = LUT_RAM_A;
dpp3_configure_shaper_lut(dpp_base, next_mode == LUT_RAM_A);
if (next_mode == LUT_RAM_A)
dpp3_program_shaper_luta_settings(dpp_base, params);
else
dpp3_program_shaper_lutb_settings(dpp_base, params);
dpp3_program_shaper_lut(
dpp_base, params->rgb_resulted, params->hw_points_num);
REG_SET(CM_SHAPER_CONTROL, 0, CM_SHAPER_LUT_MODE, next_mode == LUT_RAM_A ? 1:2);
return true;
}
static enum dc_lut_mode get3dlut_config(
struct dpp *dpp_base,
bool *is_17x17x17,
bool *is_12bits_color_channel)
{
uint32_t i_mode, i_enable_10bits, lut_size;
enum dc_lut_mode mode;
struct dcn3_dpp *dpp = TO_DCN30_DPP(dpp_base);
REG_GET(CM_3DLUT_READ_WRITE_CONTROL,
CM_3DLUT_30BIT_EN, &i_enable_10bits);
REG_GET(CM_3DLUT_MODE,
CM_3DLUT_MODE_CURRENT, &i_mode);
switch (i_mode) {
case 0:
mode = LUT_BYPASS;
break;
case 1:
mode = LUT_RAM_A;
break;
case 2:
mode = LUT_RAM_B;
break;
default:
mode = LUT_BYPASS;
break;
}
if (i_enable_10bits > 0)
*is_12bits_color_channel = false;
else
*is_12bits_color_channel = true;
REG_GET(CM_3DLUT_MODE, CM_3DLUT_SIZE, &lut_size);
if (lut_size == 0)
*is_17x17x17 = true;
else
*is_17x17x17 = false;
return mode;
}
/*
* select ramA or ramB, or bypass
* select color channel size 10 or 12 bits
* select 3dlut size 17x17x17 or 9x9x9
*/
static void dpp3_set_3dlut_mode(
struct dpp *dpp_base,
enum dc_lut_mode mode,
bool is_color_channel_12bits,
bool is_lut_size17x17x17)
{
uint32_t lut_mode;
struct dcn3_dpp *dpp = TO_DCN30_DPP(dpp_base);
if (mode == LUT_BYPASS)
lut_mode = 0;
else if (mode == LUT_RAM_A)
lut_mode = 1;
else
lut_mode = 2;
REG_UPDATE_2(CM_3DLUT_MODE,
CM_3DLUT_MODE, lut_mode,
CM_3DLUT_SIZE, is_lut_size17x17x17 == true ? 0 : 1);
}
static void dpp3_select_3dlut_ram(
struct dpp *dpp_base,
enum dc_lut_mode mode,
bool is_color_channel_12bits)
{
struct dcn3_dpp *dpp = TO_DCN30_DPP(dpp_base);
REG_UPDATE_2(CM_3DLUT_READ_WRITE_CONTROL,
CM_3DLUT_RAM_SEL, mode == LUT_RAM_A ? 0 : 1,
CM_3DLUT_30BIT_EN,
is_color_channel_12bits == true ? 0:1);
}
static void dpp3_set3dlut_ram12(
struct dpp *dpp_base,
const struct dc_rgb *lut,
uint32_t entries)
{
uint32_t i, red, green, blue, red1, green1, blue1;
struct dcn3_dpp *dpp = TO_DCN30_DPP(dpp_base);
for (i = 0 ; i < entries; i += 2) {
red = lut[i].red<<4;
green = lut[i].green<<4;
blue = lut[i].blue<<4;
red1 = lut[i+1].red<<4;
green1 = lut[i+1].green<<4;
blue1 = lut[i+1].blue<<4;
REG_SET_2(CM_3DLUT_DATA, 0,
CM_3DLUT_DATA0, red,
CM_3DLUT_DATA1, red1);
REG_SET_2(CM_3DLUT_DATA, 0,
CM_3DLUT_DATA0, green,
CM_3DLUT_DATA1, green1);
REG_SET_2(CM_3DLUT_DATA, 0,
CM_3DLUT_DATA0, blue,
CM_3DLUT_DATA1, blue1);
}
}
/*
* load selected lut with 10 bits color channels
*/
static void dpp3_set3dlut_ram10(
struct dpp *dpp_base,
const struct dc_rgb *lut,
uint32_t entries)
{
uint32_t i, red, green, blue, value;
struct dcn3_dpp *dpp = TO_DCN30_DPP(dpp_base);
for (i = 0; i < entries; i++) {
red = lut[i].red;
green = lut[i].green;
blue = lut[i].blue;
value = (red<<20) | (green<<10) | blue;
REG_SET(CM_3DLUT_DATA_30BIT, 0, CM_3DLUT_DATA_30BIT, value);
}
}
static void dpp3_select_3dlut_ram_mask(
struct dpp *dpp_base,
uint32_t ram_selection_mask)
{
struct dcn3_dpp *dpp = TO_DCN30_DPP(dpp_base);
REG_UPDATE(CM_3DLUT_READ_WRITE_CONTROL, CM_3DLUT_WRITE_EN_MASK,
ram_selection_mask);
REG_SET(CM_3DLUT_INDEX, 0, CM_3DLUT_INDEX, 0);
}
static bool dpp3_program_3dlut(struct dpp *dpp_base,
const struct tetrahedral_params *params)
{
enum dc_lut_mode mode;
bool is_17x17x17;
bool is_12bits_color_channel;
const struct dc_rgb *lut0;
const struct dc_rgb *lut1;
const struct dc_rgb *lut2;
const struct dc_rgb *lut3;
int lut_size0;
int lut_size;
if (params == NULL) {
dpp3_set_3dlut_mode(dpp_base, LUT_BYPASS, false, false);
if (dpp_base->ctx->dc->debug.enable_mem_low_power.bits.cm)
dpp3_power_on_hdr3dlut(dpp_base, false);
return false;
}
if (dpp_base->ctx->dc->debug.enable_mem_low_power.bits.cm)
dpp3_power_on_hdr3dlut(dpp_base, true);
mode = get3dlut_config(dpp_base, &is_17x17x17, &is_12bits_color_channel);
if (mode == LUT_BYPASS || mode == LUT_RAM_B)
mode = LUT_RAM_A;
else
mode = LUT_RAM_B;
is_17x17x17 = !params->use_tetrahedral_9;
is_12bits_color_channel = params->use_12bits;
if (is_17x17x17) {
lut0 = params->tetrahedral_17.lut0;
lut1 = params->tetrahedral_17.lut1;
lut2 = params->tetrahedral_17.lut2;
lut3 = params->tetrahedral_17.lut3;
lut_size0 = sizeof(params->tetrahedral_17.lut0)/
sizeof(params->tetrahedral_17.lut0[0]);
lut_size = sizeof(params->tetrahedral_17.lut1)/
sizeof(params->tetrahedral_17.lut1[0]);
} else {
lut0 = params->tetrahedral_9.lut0;
lut1 = params->tetrahedral_9.lut1;
lut2 = params->tetrahedral_9.lut2;
lut3 = params->tetrahedral_9.lut3;
lut_size0 = sizeof(params->tetrahedral_9.lut0)/
sizeof(params->tetrahedral_9.lut0[0]);
lut_size = sizeof(params->tetrahedral_9.lut1)/
sizeof(params->tetrahedral_9.lut1[0]);
}
dpp3_select_3dlut_ram(dpp_base, mode,
is_12bits_color_channel);
dpp3_select_3dlut_ram_mask(dpp_base, 0x1);
if (is_12bits_color_channel)
dpp3_set3dlut_ram12(dpp_base, lut0, lut_size0);
else
dpp3_set3dlut_ram10(dpp_base, lut0, lut_size0);
dpp3_select_3dlut_ram_mask(dpp_base, 0x2);
if (is_12bits_color_channel)
dpp3_set3dlut_ram12(dpp_base, lut1, lut_size);
else
dpp3_set3dlut_ram10(dpp_base, lut1, lut_size);
dpp3_select_3dlut_ram_mask(dpp_base, 0x4);
if (is_12bits_color_channel)
dpp3_set3dlut_ram12(dpp_base, lut2, lut_size);
else
dpp3_set3dlut_ram10(dpp_base, lut2, lut_size);
dpp3_select_3dlut_ram_mask(dpp_base, 0x8);
if (is_12bits_color_channel)
dpp3_set3dlut_ram12(dpp_base, lut3, lut_size);
else
dpp3_set3dlut_ram10(dpp_base, lut3, lut_size);
dpp3_set_3dlut_mode(dpp_base, mode, is_12bits_color_channel,
is_17x17x17);
return true;
}
static struct dpp_funcs dcn30_dpp_funcs = {
.dpp_program_gamcor_lut = dpp3_program_gamcor_lut,
.dpp_read_state = dpp30_read_state,
.dpp_reset = dpp_reset,
.dpp_set_scaler = dpp1_dscl_set_scaler_manual_scale,
.dpp_get_optimal_number_of_taps = dpp3_get_optimal_number_of_taps,
.dpp_set_gamut_remap = dpp3_cm_set_gamut_remap,
.dpp_set_csc_adjustment = NULL,
.dpp_set_csc_default = NULL,
.dpp_program_regamma_pwl = NULL,
.dpp_set_pre_degam = dpp3_set_pre_degam,
.dpp_program_input_lut = NULL,
.dpp_full_bypass = dpp1_full_bypass,
.dpp_setup = dpp3_cnv_setup,
.dpp_program_degamma_pwl = NULL,
.dpp_program_cm_dealpha = dpp3_program_cm_dealpha,
.dpp_program_cm_bias = dpp3_program_cm_bias,
.dpp_program_blnd_lut = dpp3_program_blnd_lut,
.dpp_program_shaper_lut = dpp3_program_shaper,
.dpp_program_3dlut = dpp3_program_3dlut,
.dpp_deferred_update = dpp3_deferred_update,
.dpp_program_bias_and_scale = NULL,
.dpp_cnv_set_alpha_keyer = dpp2_cnv_set_alpha_keyer,
.set_cursor_attributes = dpp3_set_cursor_attributes,
.set_cursor_position = dpp1_set_cursor_position,
.set_optional_cursor_attributes = dpp1_cnv_set_optional_cursor_attributes,
.dpp_dppclk_control = dpp1_dppclk_control,
.dpp_set_hdr_multiplier = dpp3_set_hdr_multiplier,
.dpp_get_gamut_remap = dpp3_cm_get_gamut_remap,
};
static struct dpp_caps dcn30_dpp_cap = {
.dscl_data_proc_format = DSCL_DATA_PRCESSING_FLOAT_FORMAT,
.dscl_calc_lb_num_partitions = dscl2_calc_lb_num_partitions,
};
bool dpp3_construct(
struct dcn3_dpp *dpp,
struct dc_context *ctx,
uint32_t inst,
const struct dcn3_dpp_registers *tf_regs,
const struct dcn3_dpp_shift *tf_shift,
const struct dcn3_dpp_mask *tf_mask)
{
dpp->base.ctx = ctx;
dpp->base.inst = inst;
dpp->base.funcs = &dcn30_dpp_funcs;
dpp->base.caps = &dcn30_dpp_cap;
dpp->tf_regs = tf_regs;
dpp->tf_shift = tf_shift;
dpp->tf_mask = tf_mask;
return true;
}