From 31393a296b4773645cb926ff60c67571903f9f4b Mon Sep 17 00:00:00 2001 From: Vitaly Prosyak Date: Tue, 7 Feb 2017 10:41:37 -0600 Subject: [PATCH 0208/4131] drm/amd/display: Enable regamma 25 segments and use double buffer. Moved custom floating point calculation to the shared place between dce's. Signed-off-by: Vitaly Prosyak Reviewed-by: Tony Cheng Acked-by: Harry Wentland Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/display/dc/calcs/Makefile | 2 +- .../gpu/drm/amd/display/dc/calcs/custom_float.c | 197 +++++++++++++++++++++ .../amd/display/dc/dce110/dce110_hw_sequencer.c | 171 +----------------- .../amd/display/dc/dce110/dce110_hw_sequencer.h | 3 - drivers/gpu/drm/amd/display/dc/inc/custom_float.h | 40 +++++ 5 files changed, 240 insertions(+), 173 deletions(-) create mode 100644 drivers/gpu/drm/amd/display/dc/calcs/custom_float.c create mode 100644 drivers/gpu/drm/amd/display/dc/inc/custom_float.h diff --git a/drivers/gpu/drm/amd/display/dc/calcs/Makefile b/drivers/gpu/drm/amd/display/dc/calcs/Makefile index 4bb08ae..2f4c8e7 100644 --- a/drivers/gpu/drm/amd/display/dc/calcs/Makefile +++ b/drivers/gpu/drm/amd/display/dc/calcs/Makefile @@ -3,7 +3,7 @@ # It calculates Bandwidth and Watermarks values for HW programming # -BW_CALCS = bandwidth_calcs.o bw_fixed.o +BW_CALCS = bandwidth_calcs.o bw_fixed.o custom_float.o AMD_DAL_BW_CALCS = $(addprefix $(AMDDALPATH)/dc/calcs/,$(BW_CALCS)) diff --git a/drivers/gpu/drm/amd/display/dc/calcs/custom_float.c b/drivers/gpu/drm/amd/display/dc/calcs/custom_float.c new file mode 100644 index 0000000..7243c37 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/calcs/custom_float.c @@ -0,0 +1,197 @@ +/* + * 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 "dm_services.h" +#include "custom_float.h" + + +static bool build_custom_float( + struct fixed31_32 value, + const struct custom_float_format *format, + bool *negative, + uint32_t *mantissa, + uint32_t *exponenta) +{ + uint32_t exp_offset = (1 << (format->exponenta_bits - 1)) - 1; + + const struct fixed31_32 mantissa_constant_plus_max_fraction = + dal_fixed31_32_from_fraction( + (1LL << (format->mantissa_bits + 1)) - 1, + 1LL << format->mantissa_bits); + + struct fixed31_32 mantiss; + + if (dal_fixed31_32_eq( + value, + dal_fixed31_32_zero)) { + *negative = false; + *mantissa = 0; + *exponenta = 0; + return true; + } + + if (dal_fixed31_32_lt( + value, + dal_fixed31_32_zero)) { + *negative = format->sign; + value = dal_fixed31_32_neg(value); + } else { + *negative = false; + } + + if (dal_fixed31_32_lt( + value, + dal_fixed31_32_one)) { + uint32_t i = 1; + + do { + value = dal_fixed31_32_shl(value, 1); + ++i; + } while (dal_fixed31_32_lt( + value, + dal_fixed31_32_one)); + + --i; + + if (exp_offset <= i) { + *mantissa = 0; + *exponenta = 0; + return true; + } + + *exponenta = exp_offset - i; + } else if (dal_fixed31_32_le( + mantissa_constant_plus_max_fraction, + value)) { + uint32_t i = 1; + + do { + value = dal_fixed31_32_shr(value, 1); + ++i; + } while (dal_fixed31_32_lt( + mantissa_constant_plus_max_fraction, + value)); + + *exponenta = exp_offset + i - 1; + } else { + *exponenta = exp_offset; + } + + mantiss = dal_fixed31_32_sub( + value, + dal_fixed31_32_one); + + if (dal_fixed31_32_lt( + mantiss, + dal_fixed31_32_zero) || + dal_fixed31_32_lt( + dal_fixed31_32_one, + mantiss)) + mantiss = dal_fixed31_32_zero; + else + mantiss = dal_fixed31_32_shl( + mantiss, + format->mantissa_bits); + + *mantissa = dal_fixed31_32_floor(mantiss); + + return true; +} + +static bool setup_custom_float( + const struct custom_float_format *format, + bool negative, + uint32_t mantissa, + uint32_t exponenta, + uint32_t *result) +{ + uint32_t i = 0; + uint32_t j = 0; + + uint32_t value = 0; + + /* verification code: + * once calculation is ok we can remove it + */ + + const uint32_t mantissa_mask = + (1 << (format->mantissa_bits + 1)) - 1; + + const uint32_t exponenta_mask = + (1 << (format->exponenta_bits + 1)) - 1; + + if (mantissa & ~mantissa_mask) { + BREAK_TO_DEBUGGER(); + mantissa = mantissa_mask; + } + + if (exponenta & ~exponenta_mask) { + BREAK_TO_DEBUGGER(); + exponenta = exponenta_mask; + } + + /* end of verification code */ + + while (i < format->mantissa_bits) { + uint32_t mask = 1 << i; + + if (mantissa & mask) + value |= mask; + + ++i; + } + + while (j < format->exponenta_bits) { + uint32_t mask = 1 << j; + + if (exponenta & mask) + value |= mask << i; + + ++j; + } + + if (negative && format->sign) + value |= 1 << (i + j); + + *result = value; + + return true; +} + +bool convert_to_custom_float_format( + struct fixed31_32 value, + const struct custom_float_format *format, + uint32_t *result) +{ + uint32_t mantissa; + uint32_t exponenta; + bool negative; + + return build_custom_float( + value, format, &negative, &mantissa, &exponenta) && + setup_custom_float( + format, negative, mantissa, exponenta, result); +} + + diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c index 6a5cec0..88a2fa9 100644 --- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c +++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c @@ -49,6 +49,7 @@ /* include DCE11 register header files */ #include "dce/dce_11_0_d.h" #include "dce/dce_11_0_sh_mask.h" +#include "custom_float.h" struct dce110_hw_seq_reg_offsets { uint32_t crtc; @@ -286,174 +287,6 @@ static bool dce110_set_input_transfer_func( return result; } -static bool build_custom_float( - struct fixed31_32 value, - const struct custom_float_format *format, - bool *negative, - uint32_t *mantissa, - uint32_t *exponenta) -{ - uint32_t exp_offset = (1 << (format->exponenta_bits - 1)) - 1; - - const struct fixed31_32 mantissa_constant_plus_max_fraction = - dal_fixed31_32_from_fraction( - (1LL << (format->mantissa_bits + 1)) - 1, - 1LL << format->mantissa_bits); - - struct fixed31_32 mantiss; - - if (dal_fixed31_32_eq( - value, - dal_fixed31_32_zero)) { - *negative = false; - *mantissa = 0; - *exponenta = 0; - return true; - } - - if (dal_fixed31_32_lt( - value, - dal_fixed31_32_zero)) { - *negative = format->sign; - value = dal_fixed31_32_neg(value); - } else { - *negative = false; - } - - if (dal_fixed31_32_lt( - value, - dal_fixed31_32_one)) { - uint32_t i = 1; - - do { - value = dal_fixed31_32_shl(value, 1); - ++i; - } while (dal_fixed31_32_lt( - value, - dal_fixed31_32_one)); - - --i; - - if (exp_offset <= i) { - *mantissa = 0; - *exponenta = 0; - return true; - } - - *exponenta = exp_offset - i; - } else if (dal_fixed31_32_le( - mantissa_constant_plus_max_fraction, - value)) { - uint32_t i = 1; - - do { - value = dal_fixed31_32_shr(value, 1); - ++i; - } while (dal_fixed31_32_lt( - mantissa_constant_plus_max_fraction, - value)); - - *exponenta = exp_offset + i - 1; - } else { - *exponenta = exp_offset; - } - - mantiss = dal_fixed31_32_sub( - value, - dal_fixed31_32_one); - - if (dal_fixed31_32_lt( - mantiss, - dal_fixed31_32_zero) || - dal_fixed31_32_lt( - dal_fixed31_32_one, - mantiss)) - mantiss = dal_fixed31_32_zero; - else - mantiss = dal_fixed31_32_shl( - mantiss, - format->mantissa_bits); - - *mantissa = dal_fixed31_32_floor(mantiss); - - return true; -} - -static bool setup_custom_float( - const struct custom_float_format *format, - bool negative, - uint32_t mantissa, - uint32_t exponenta, - uint32_t *result) -{ - uint32_t i = 0; - uint32_t j = 0; - - uint32_t value = 0; - - /* verification code: - * once calculation is ok we can remove it - */ - - const uint32_t mantissa_mask = - (1 << (format->mantissa_bits + 1)) - 1; - - const uint32_t exponenta_mask = - (1 << (format->exponenta_bits + 1)) - 1; - - if (mantissa & ~mantissa_mask) { - BREAK_TO_DEBUGGER(); - mantissa = mantissa_mask; - } - - if (exponenta & ~exponenta_mask) { - BREAK_TO_DEBUGGER(); - exponenta = exponenta_mask; - } - - /* end of verification code */ - - while (i < format->mantissa_bits) { - uint32_t mask = 1 << i; - - if (mantissa & mask) - value |= mask; - - ++i; - } - - while (j < format->exponenta_bits) { - uint32_t mask = 1 << j; - - if (exponenta & mask) - value |= mask << i; - - ++j; - } - - if (negative && format->sign) - value |= 1 << (i + j); - - *result = value; - - return true; -} - -static bool convert_to_custom_float_format( - struct fixed31_32 value, - const struct custom_float_format *format, - uint32_t *result) -{ - uint32_t mantissa; - uint32_t exponenta; - bool negative; - - return build_custom_float( - value, format, &negative, &mantissa, &exponenta) && - setup_custom_float( - format, negative, mantissa, exponenta, result); -} - static bool convert_to_custom_float( struct pwl_result_data *rgb_resulted, struct curve_points *arr_points, @@ -579,7 +412,7 @@ static bool convert_to_custom_float( return true; } -bool dce110_translate_regamma_to_hw_format(const struct dc_transfer_func +static bool dce110_translate_regamma_to_hw_format(const struct dc_transfer_func *output_tf, struct pwl_params *regamma_params) { struct curve_points *arr_points; diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.h index 68632dd..a6b4d0d 100644 --- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.h +++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.h @@ -58,8 +58,5 @@ void dce110_power_down(struct core_dc *dc); void dce110_update_pending_status(struct pipe_ctx *pipe_ctx); -bool dce110_translate_regamma_to_hw_format(const struct dc_transfer_func - *output_tf, struct pwl_params *regamma_params); - #endif /* __DC_HWSS_DCE110_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/inc/custom_float.h b/drivers/gpu/drm/amd/display/dc/inc/custom_float.h new file mode 100644 index 0000000..f572396 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/inc/custom_float.h @@ -0,0 +1,40 @@ +/* + * 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 CUSTOM_FLOAT_H_ +#define CUSTOM_FLOAT_H_ + +#include "bw_fixed.h" +#include "hw_shared.h" +#include "opp.h" + + +bool convert_to_custom_float_format( + struct fixed31_32 value, + const struct custom_float_format *format, + uint32_t *result); + + +#endif //CUSTOM_FLOAT_H_ -- 2.7.4