1
0
mirror of https://github.com/TomHarte/CLK.git synced 2024-11-26 23:52:26 +00:00

Merge pull request #149 from TomHarte/BetterVectors

Corrects some long-standing unnecessarily convoluted ways to handle semi-dynamic arrays
This commit is contained in:
Thomas Harte 2017-07-16 15:02:52 -04:00 committed by GitHub
commit 561373e793
8 changed files with 32 additions and 37 deletions

View File

@ -69,8 +69,8 @@ unsigned int Machine::perform_bus_operation(CPU::MOS6502::BusOperation operation
return 1;
}
void Machine::set_rom(const uint8_t *rom) {
memcpy(rom_, rom, sizeof(rom_));
void Machine::set_rom(const std::vector<uint8_t> &rom) {
memcpy(rom_, rom.data(), std::min(sizeof(rom_), rom.size()));
}
void Machine::set_disk(std::shared_ptr<Storage::Disk::Disk> disk) {

View File

@ -131,7 +131,7 @@ class Machine:
/*!
Sets the ROM image to use for this drive; it is assumed that the buffer provided will be at least 16 kb in size.
*/
void set_rom(const uint8_t *rom);
void set_rom(const std::vector<uint8_t> &rom);
/*!
Sets the serial bus to which this drive should attach itself.

View File

@ -246,8 +246,8 @@ void Machine::set_rom(ROMSlot slot, size_t length, const uint8_t *data) {
case Characters: target = character_rom_; max_length = 0x1000; break;
case BASIC: target = basic_rom_; break;
case Drive:
drive_rom_.reset(new uint8_t[length]);
memcpy(drive_rom_.get(), data, length);
drive_rom_.resize(length);
memcpy(drive_rom_.data(), data, length);
install_disk_rom();
return;
}
@ -313,10 +313,10 @@ void Machine::tape_did_change_input(Storage::Tape::BinaryTapePlayer *tape) {
#pragma mark - Disc
void Machine::install_disk_rom() {
if(drive_rom_ && c1540_) {
c1540_->set_rom(drive_rom_.get());
if(!drive_rom_.empty() && c1540_) {
c1540_->set_rom(drive_rom_);
c1540_->run_for_cycles(2000000);
drive_rom_.reset();
drive_rom_.clear();
}
}

View File

@ -199,7 +199,7 @@ class Machine:
uint8_t user_basic_memory_[0x0400];
uint8_t screen_memory_[0x1000];
uint8_t colour_memory_[0x0400];
std::unique_ptr<uint8_t> drive_rom_;
std::vector<uint8_t> drive_rom_;
uint8_t *processor_read_memory_map_[64];
uint8_t *processor_write_memory_map_[64];

View File

@ -21,8 +21,8 @@
struct SpeakerDelegate: public Outputs::Speaker::Delegate {
__weak CSMachine *machine;
void speaker_did_complete_samples(Outputs::Speaker *speaker, const int16_t *buffer, int buffer_size) {
[machine speaker:speaker didCompleteSamples:buffer length:buffer_size];
void speaker_did_complete_samples(Outputs::Speaker *speaker, const std::vector<int16_t> &buffer) {
[machine speaker:speaker didCompleteSamples:buffer.data() length:(int)buffer.size()];
}
};

View File

@ -15,6 +15,7 @@
#include <memory>
#include <list>
#include <vector>
#include "../SignalProcessing/Stepper.hpp"
#include "../SignalProcessing/FIRFilter.hpp"
@ -34,7 +35,7 @@ class Speaker {
public:
class Delegate {
public:
virtual void speaker_did_complete_samples(Speaker *speaker, const int16_t *buffer, int buffer_size) = 0;
virtual void speaker_did_complete_samples(Speaker *speaker, const std::vector<int16_t> &buffer) = 0;
};
float get_ideal_clock_rate_in_range(float minimum, float maximum) {
@ -53,10 +54,7 @@ class Speaker {
void set_output_rate(float cycles_per_second, int buffer_size) {
output_cycles_per_second_ = cycles_per_second;
if(buffer_size_ != buffer_size) {
buffer_in_progress_.reset(new int16_t[buffer_size]);
buffer_size_ = buffer_size;
}
buffer_in_progress_.resize((size_t)buffer_size);
set_needs_updated_filter_coefficients();
}
@ -104,9 +102,8 @@ class Speaker {
}
std::shared_ptr<std::list<std::function<void(void)>>> queued_functions_;
std::unique_ptr<int16_t> buffer_in_progress_;
std::vector<int16_t> buffer_in_progress_;
float high_frequency_cut_off_;
int buffer_size_;
int buffer_in_progress_pointer_;
int number_of_taps_, requested_number_of_taps_;
bool coefficients_are_dirty_;
@ -151,17 +148,17 @@ template <class T> class Filter: public Speaker {
// if input and output rates exactly match, just accumulate results and pass on
if(input_cycles_per_second_ == output_cycles_per_second_ && high_frequency_cut_off_ < 0.0) {
while(cycles_remaining) {
unsigned int cycles_to_read = (unsigned int)(buffer_size_ - buffer_in_progress_pointer_);
unsigned int cycles_to_read = (unsigned int)(buffer_in_progress_.size() - (size_t)buffer_in_progress_pointer_);
if(cycles_to_read > cycles_remaining) cycles_to_read = cycles_remaining;
static_cast<T *>(this)->get_samples(cycles_to_read, &buffer_in_progress_.get()[buffer_in_progress_pointer_]);
static_cast<T *>(this)->get_samples(cycles_to_read, &buffer_in_progress_[(size_t)buffer_in_progress_pointer_]);
buffer_in_progress_pointer_ += cycles_to_read;
// announce to delegate if full
if(buffer_in_progress_pointer_ == buffer_size_) {
if(buffer_in_progress_pointer_ == buffer_in_progress_.size()) {
buffer_in_progress_pointer_ = 0;
if(delegate_) {
delegate_->speaker_did_complete_samples(this, buffer_in_progress_.get(), buffer_size_);
delegate_->speaker_did_complete_samples(this, buffer_in_progress_);
}
}
@ -175,19 +172,19 @@ template <class T> class Filter: public Speaker {
if(input_cycles_per_second_ > output_cycles_per_second_ || (input_cycles_per_second_ == output_cycles_per_second_ && high_frequency_cut_off_ >= 0.0)) {
while(cycles_remaining) {
unsigned int cycles_to_read = (unsigned int)std::min((int)cycles_remaining, number_of_taps_ - input_buffer_depth_);
static_cast<T *>(this)->get_samples(cycles_to_read, &input_buffer_.get()[input_buffer_depth_]);
static_cast<T *>(this)->get_samples(cycles_to_read, &input_buffer_[(size_t)input_buffer_depth_]);
cycles_remaining -= cycles_to_read;
input_buffer_depth_ += cycles_to_read;
if(input_buffer_depth_ == number_of_taps_) {
buffer_in_progress_.get()[buffer_in_progress_pointer_] = filter_->apply(input_buffer_.get());
buffer_in_progress_[(size_t)buffer_in_progress_pointer_] = filter_->apply(input_buffer_.data());
buffer_in_progress_pointer_++;
// announce to delegate if full
if(buffer_in_progress_pointer_ == buffer_size_) {
if(buffer_in_progress_pointer_ == buffer_in_progress_.size()) {
buffer_in_progress_pointer_ = 0;
if(delegate_) {
delegate_->speaker_did_complete_samples(this, buffer_in_progress_.get(), buffer_size_);
delegate_->speaker_did_complete_samples(this, buffer_in_progress_);
}
}
@ -196,7 +193,7 @@ template <class T> class Filter: public Speaker {
// anything. Otherwise skip as required to get to the next sample batch and don't expect to reuse.
uint64_t steps = stepper_->step();
if(steps < number_of_taps_) {
int16_t *input_buffer = input_buffer_.get();
int16_t *input_buffer = input_buffer_.data();
memmove(input_buffer, &input_buffer[steps], sizeof(int16_t) * ((size_t)number_of_taps_ - (size_t)steps));
input_buffer_depth_ -= steps;
} else {
@ -218,7 +215,7 @@ template <class T> class Filter: public Speaker {
std::unique_ptr<SignalProcessing::Stepper> stepper_;
std::unique_ptr<SignalProcessing::FIRFilter> filter_;
std::unique_ptr<int16_t> input_buffer_;
std::vector<int16_t> input_buffer_;
int input_buffer_depth_;
void update_filter_coefficients() {
@ -244,7 +241,7 @@ template <class T> class Filter: public Speaker {
}
filter_.reset(new SignalProcessing::FIRFilter((unsigned int)number_of_taps_, (float)input_cycles_per_second_, 0.0, high_pass_frequency, SignalProcessing::FIRFilter::DefaultAttenuation));
input_buffer_.reset(new int16_t[number_of_taps_]);
input_buffer_.resize((size_t)number_of_taps_);
input_buffer_depth_ = 0;
}
};

View File

@ -17,9 +17,8 @@ DigitalPhaseLockedLoop::DigitalPhaseLockedLoop(int clocks_per_bit, int tolerance
tolerance_(tolerance),
phase_(0),
window_length_(clocks_per_bit),
phase_error_pointer_(0) {
phase_error_history_.reset(new std::vector<int>(length_of_history, 0));
}
phase_error_pointer_(0),
phase_error_history_(length_of_history, 0) {}
void DigitalPhaseLockedLoop::run_for_cycles(int number_of_cycles) {
phase_ += number_of_cycles;
@ -51,15 +50,14 @@ void DigitalPhaseLockedLoop::post_phase_error(int error) {
phase_ -= (error + 1) >> 1;
// use the average of the last few errors to affect frequency
std::vector<int> *phase_error_history = phase_error_history_.get();
size_t phase_error_history_size = phase_error_history->size();
size_t phase_error_history_size = phase_error_history_.size();
(*phase_error_history)[phase_error_pointer_] = error;
phase_error_history_[phase_error_pointer_] = error;
phase_error_pointer_ = (phase_error_pointer_ + 1)%phase_error_history_size;
int total_error = 0;
for(size_t c = 0; c < phase_error_history_size; c++) {
total_error += (*phase_error_history)[c];
total_error += phase_error_history_[c];
}
int denominator = (int)(phase_error_history_size * 4);
window_length_ += (total_error + (denominator >> 1)) / denominator;

View File

@ -52,7 +52,7 @@ class DigitalPhaseLockedLoop {
Delegate *delegate_;
void post_phase_error(int error);
std::unique_ptr<std::vector<int>> phase_error_history_;
std::vector<int> phase_error_history_;
size_t phase_error_pointer_;
int phase_;