circuitpython/ports/atmel-samd/common-hal/audiobusio/PDMIn.c

462 lines
17 KiB
C
Raw Normal View History

/*
* This file is part of the MicroPython project, http://micropython.org/
*
* The MIT License (MIT)
*
* Copyright (c) 2017 Scott Shawcroft for Adafruit Industries
*
* 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
* AUTHORS OR COPYRIGHT HOLDERS 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.
*/
#include <stdint.h>
#include <string.h>
2018-01-02 21:25:41 -05:00
#include <math.h>
#include "py/gc.h"
#include "py/mperrno.h"
#include "py/runtime.h"
#include "common-hal/analogio/AnalogOut.h"
#include "common-hal/audiobusio/PDMIn.h"
#include "shared-bindings/analogio/AnalogOut.h"
#include "shared-bindings/audiobusio/PDMIn.h"
#include "shared-bindings/microcontroller/Pin.h"
#include "atmel_start_pins.h"
#include "hal/include/hal_gpio.h"
#include "hal/utils/include/utils.h"
#include "peripherals/clocks.h"
#include "peripherals/events.h"
#include "peripherals/i2s.h"
#include "peripherals/pins.h"
#include "peripherals/dma.h"
#include "audio_dma.h"
#include "tick.h"
2018-01-02 21:25:41 -05:00
#define OVERSAMPLING 64
#define SAMPLES_PER_BUFFER 32
2018-01-02 21:25:41 -05:00
// MEMS microphones must be clocked at at least 1MHz.
#define MIN_MIC_CLOCK 1000000
#ifdef SAMD21
#define SERCTRL(name) I2S_SERCTRL_ ## name
#endif
#ifdef SAMD51
#define SERCTRL(name) I2S_RXCTRL_ ## name
#endif
void pdmin_reset(void) {
while (I2S->SYNCBUSY.reg & I2S_SYNCBUSY_ENABLE) {}
I2S->INTENCLR.reg = I2S_INTENCLR_MASK;
I2S->INTFLAG.reg = I2S_INTFLAG_MASK;
I2S->CTRLA.reg &= ~I2S_SYNCBUSY_ENABLE;
while (I2S->SYNCBUSY.reg & I2S_SYNCBUSY_ENABLE) {}
I2S->CTRLA.reg = I2S_CTRLA_SWRST;
}
void common_hal_audiobusio_pdmin_construct(audiobusio_pdmin_obj_t* self,
const mcu_pin_obj_t* clock_pin,
const mcu_pin_obj_t* data_pin,
uint32_t sample_rate,
uint8_t bit_depth,
bool mono,
uint8_t oversample) {
self->clock_pin = clock_pin; // PA10, PA20 -> SCK0, PB11 -> SCK1
#ifdef SAMD21
if (clock_pin == &pin_PA10
#ifdef PIN_PA20
|| clock_pin == &pin_PA20
#endif
) {
self->clock_unit = 0;
#ifdef PIN_PB11
} else if (clock_pin == &pin_PB11) {
self->clock_unit = 1;
#endif
#endif
#ifdef SAMD51
if (clock_pin == &pin_PA10 || clock_pin == &pin_PB16) {
self->clock_unit = 0;
} else if (clock_pin == &pin_PB12
#ifdef PIN_PB28
|| data_pin == &pin_PB28) {
#else
) {
#endif
self->clock_unit = 1;
#endif
} else {
mp_raise_ValueError("Invalid clock pin");
}
self->data_pin = data_pin; // PA07, PA19 -> SD0, PA08, PB16 -> SD1
#ifdef SAMD21
if (data_pin == &pin_PA07 || data_pin == &pin_PA19) {
self->serializer = 0;
} else if (data_pin == &pin_PA08
#ifdef PIN_PB16
|| data_pin == &pin_PB16) {
#else
) {
#endif
self->serializer = 1;
#endif
#ifdef SAMD51
if (data_pin == &pin_PB10 || data_pin == &pin_PA22) {
self->serializer = 1;
#endif
} else {
mp_raise_ValueError("Invalid data pin");
}
2018-01-02 21:25:41 -05:00
if (!(bit_depth == 16 || bit_depth == 8) || !mono || oversample != OVERSAMPLING) {
mp_raise_NotImplementedError("Only 8 or 16 bit mono with " MP_STRINGIFY(OVERSAMPLING) "x oversampling is supported.");
}
turn_on_i2s();
if (I2S->CTRLA.bit.ENABLE == 0) {
I2S->CTRLA.bit.SWRST = 1;
while (I2S->CTRLA.bit.SWRST == 1) {}
} else {
#ifdef SAMD21
if ((I2S->CTRLA.vec.SEREN & (1 << self->serializer)) != 0) {
mp_raise_RuntimeError("Serializer in use");
}
#endif
#ifdef SAMD51
if (I2S->CTRLA.bit.RXEN == 1) {
mp_raise_RuntimeError("Serializer in use");
}
#endif
}
#ifdef SAMD51
#define GPIO_I2S_FUNCTION GPIO_PIN_FUNCTION_J
#endif
#ifdef SAMD21
#define GPIO_I2S_FUNCTION GPIO_PIN_FUNCTION_G
#endif
assert_pin_free(clock_pin);
assert_pin_free(data_pin);
uint32_t clock_divisor = (uint32_t) roundf( 48000000.0f / sample_rate / oversample);
float mic_clock_freq = 48000000.0f / clock_divisor;
self->sample_rate = mic_clock_freq / oversample;
if (mic_clock_freq < MIN_MIC_CLOCK || clock_divisor == 0) {
mp_raise_ValueError("sampling rate out of range");
}
// Find a free GCLK to generate the MCLK signal.
uint8_t gclk = find_free_gclk(clock_divisor);
if (gclk > GCLK_GEN_NUM) {
mp_raise_RuntimeError("Unable to find free GCLK");
}
self->gclk = gclk;
enable_clock_generator(self->gclk, CLOCK_48MHZ, clock_divisor);
connect_gclk_to_peripheral(self->gclk, I2S_GCLK_ID_0 + self->clock_unit);
// Clock unit configuration
uint32_t clkctrl = I2S_CLKCTRL_MCKSEL_GCLK |
I2S_CLKCTRL_NBSLOTS(2) |
I2S_CLKCTRL_FSWIDTH_SLOT |
I2S_CLKCTRL_SLOTSIZE_16;
// Serializer configuration
#ifdef SAMD21
uint32_t serctrl = (self->clock_unit << I2S_SERCTRL_CLKSEL_Pos) | SERCTRL(SERMODE_PDM2) | SERCTRL(DATASIZE_32);
#endif
#ifdef SAMD51
uint32_t serctrl = (self->clock_unit << I2S_RXCTRL_CLKSEL_Pos) | SERCTRL(SERMODE_PDM2) | SERCTRL(DATASIZE_32);
#endif
// Configure the I2S peripheral
i2s_set_enable(false);
I2S->CLKCTRL[self->clock_unit].reg = clkctrl;
#ifdef SAMD21
I2S->SERCTRL[self->serializer].reg = serctrl;
#endif
#ifdef SAMD51
I2S->RXCTRL.reg = serctrl;
#endif
i2s_set_enable(true);
2018-01-02 21:25:41 -05:00
// Run the serializer all the time. This eliminates startup delay for the microphone.
i2s_set_clock_unit_enable(self->clock_unit, true);
i2s_set_serializer_enable(self->serializer, true);
claim_pin(clock_pin);
claim_pin(data_pin);
gpio_set_pin_function(self->clock_pin->pin, GPIO_I2S_FUNCTION);
gpio_set_pin_function(self->data_pin->pin, GPIO_I2S_FUNCTION);
2018-01-02 21:25:41 -05:00
self->bytes_per_sample = oversample >> 3;
self->bit_depth = bit_depth;
}
bool common_hal_audiobusio_pdmin_deinited(audiobusio_pdmin_obj_t* self) {
return self->clock_pin == mp_const_none;
}
void common_hal_audiobusio_pdmin_deinit(audiobusio_pdmin_obj_t* self) {
if (common_hal_audiobusio_pdmin_deinited(self)) {
return;
}
i2s_set_serializer_enable(self->serializer, false);
i2s_set_clock_unit_enable(self->clock_unit, false);
i2s_set_enable(false);
disconnect_gclk_from_peripheral(self->gclk, I2S_GCLK_ID_0 + self->clock_unit);
disable_clock_generator(self->gclk);
reset_pin(self->clock_pin->pin);
reset_pin(self->data_pin->pin);
self->clock_pin = mp_const_none;
self->data_pin = mp_const_none;
}
uint8_t common_hal_audiobusio_pdmin_get_bit_depth(audiobusio_pdmin_obj_t* self) {
return self->bit_depth;
}
uint32_t common_hal_audiobusio_pdmin_get_sample_rate(audiobusio_pdmin_obj_t* self) {
return self->sample_rate;
}
static void setup_dma(audiobusio_pdmin_obj_t* self, uint32_t length,
DmacDescriptor* descriptor,
DmacDescriptor* second_descriptor,
uint32_t words_per_buffer, uint8_t words_per_sample,
uint32_t* first_buffer, uint32_t* second_buffer) {
descriptor->BTCTRL.reg = DMAC_BTCTRL_VALID |
DMAC_BTCTRL_BLOCKACT_NOACT |
DMAC_BTCTRL_EVOSEL_BLOCK |
DMAC_BTCTRL_DSTINC |
DMAC_BTCTRL_BEATSIZE_WORD;
// Block transfer count is the number of beats per block (aka descriptor).
// In this case there are two bytes per beat so divide the length by two.
uint16_t block_transfer_count = words_per_buffer;
if (length * words_per_sample < words_per_buffer) {
block_transfer_count = length * words_per_sample;
}
descriptor->BTCNT.reg = block_transfer_count;
descriptor->DSTADDR.reg = ((uint32_t) first_buffer + sizeof(uint32_t) * block_transfer_count);
descriptor->DESCADDR.reg = 0;
if (length * words_per_sample > words_per_buffer) {
descriptor->DESCADDR.reg = ((uint32_t)second_descriptor);
}
#ifdef SAMD21
descriptor->SRCADDR.reg = (uint32_t)&I2S->DATA[self->serializer];
#endif
#ifdef SAMD51
descriptor->SRCADDR.reg = (uint32_t)&I2S->RXDATA;
#endif
2018-01-02 21:25:41 -05:00
// Do we need more values than will fit in the first buffer?
// If so, set up a second buffer chained to be filled after the first buffer.
if (length * words_per_sample > words_per_buffer) {
block_transfer_count = words_per_buffer;
second_descriptor->DESCADDR.reg = ((uint32_t)descriptor);
if (length * words_per_sample < 2 * words_per_buffer) {
2018-01-02 21:25:41 -05:00
// Length needed is more than one buffer but less than two.
// Subtract off the size of the first buffer, and what remains is the count we need.
block_transfer_count = length * words_per_sample - words_per_buffer;
second_descriptor->DESCADDR.reg = 0;
}
second_descriptor->DSTADDR.reg = ((uint32_t) second_buffer + sizeof(uint32_t) * block_transfer_count);
second_descriptor->BTCNT.reg = block_transfer_count;
#ifdef SAMD21
second_descriptor->SRCADDR.reg = (uint32_t)&I2S->DATA[self->serializer];
#endif
#ifdef SAMD51
second_descriptor->SRCADDR.reg = (uint32_t)&I2S->RXDATA;
#endif
second_descriptor->BTCTRL.reg = DMAC_BTCTRL_VALID |
DMAC_BTCTRL_BLOCKACT_NOACT |
DMAC_BTCTRL_EVOSEL_BLOCK |
DMAC_BTCTRL_DSTINC |
DMAC_BTCTRL_BEATSIZE_WORD;
}
}
2018-01-02 21:25:41 -05:00
// a windowed sinc filter for 44 khz, 64 samples
//
// This filter is good enough to use for lower sample rates as
// well. It does not increase the noise enough to be a problem.
//
// In the long run we could use a fast filter like this to do the
// decimation and initial filtering in real time, filtering to a
// higher sample rate than specified. Then after the audio is
// recorded, a more expensive filter non-real-time filter could be
// used to down-sample and low-pass.
uint16_t sinc_filter [OVERSAMPLING] = {
0, 2, 9, 21, 39, 63, 94, 132,
179, 236, 302, 379, 467, 565, 674, 792,
920, 1055, 1196, 1341, 1487, 1633, 1776, 1913,
2042, 2159, 2263, 2352, 2422, 2474, 2506, 2516,
2506, 2474, 2422, 2352, 2263, 2159, 2042, 1913,
1776, 1633, 1487, 1341, 1196, 1055, 920, 792,
674, 565, 467, 379, 302, 236, 179, 132,
94, 63, 39, 21, 9, 2, 0, 0
};
2018-01-02 21:25:41 -05:00
#define REPEAT_16_TIMES(X) X X X X X X X X X X X X X X X X
static uint16_t filter_sample(uint32_t pdm_samples[4]) {
2018-01-02 21:25:41 -05:00
uint16_t running_sum = 0;
const uint16_t *filter_ptr = sinc_filter;
for (uint8_t i = 0; i < OVERSAMPLING/16; i++) {
// The sample is 16-bits right channel in the upper two bytes and 16-bits left channel
// in the lower two bytes.
// We just ignore the upper bits
uint32_t pdm_sample = pdm_samples[i];
REPEAT_16_TIMES( {
if (pdm_sample & 0x8000) {
running_sum += *filter_ptr;
}
filter_ptr++;
pdm_sample <<= 1;
}
2018-01-02 21:25:41 -05:00
)
}
2018-01-02 21:25:41 -05:00
return running_sum;
}
2018-01-02 21:25:41 -05:00
// output_buffer may be a byte buffer or a halfword buffer.
// output_buffer_length is the number of slots, not the number of bytes.
uint32_t common_hal_audiobusio_pdmin_record_to_buffer(audiobusio_pdmin_obj_t* self,
2018-01-02 21:25:41 -05:00
uint16_t* output_buffer, uint32_t output_buffer_length) {
uint8_t dma_channel = find_free_audio_dma_channel();
uint8_t event_channel = find_sync_event_channel();
2018-01-02 21:25:41 -05:00
// We allocate two buffers on the stack to use for double buffering.
const uint8_t samples_per_buffer = SAMPLES_PER_BUFFER;
// For every word we record, we throw away 2 bytes of a phantom second channel.
uint8_t words_per_sample = self->bytes_per_sample / 2;
uint32_t words_per_buffer = samples_per_buffer * words_per_sample;
uint32_t first_buffer[words_per_buffer];
uint32_t second_buffer[words_per_buffer];
turn_on_event_system();
COMPILER_ALIGNED(16) DmacDescriptor second_descriptor;
setup_dma(self, output_buffer_length, dma_descriptor(dma_channel), &second_descriptor,
words_per_buffer, words_per_sample, first_buffer, second_buffer);
uint8_t trigger_source = I2S_DMAC_ID_RX_0;
#ifdef SAMD21
trigger_source += self->serializer;
#endif
dma_configure(dma_channel, trigger_source, true);
init_event_channel_interrupt(event_channel, CORE_GCLK, EVSYS_ID_GEN_DMAC_CH_0 + dma_channel);
dma_enable_channel(dma_channel);
// Record
uint32_t buffers_processed = 0;
2018-01-02 21:25:41 -05:00
uint32_t values_output = 0;
2018-01-02 21:25:41 -05:00
uint32_t remaining_samples_needed = output_buffer_length;
while (values_output < output_buffer_length) {
if (event_interrupt_overflow(event_channel)) {
// Looks like we aren't keeping up. We shouldn't skip a buffer so stop early.
break;
}
// Wait for the next buffer to fill
while (!event_interrupt_active(event_channel)) {
#ifdef MICROPY_VM_HOOK_LOOP
MICROPY_VM_HOOK_LOOP
#endif
}
2018-01-02 21:25:41 -05:00
// The mic is running all the time, so we don't need to wait the usual 10msec or 100msec
// for it to start up.
// Flip back and forth between processing the first and second buffers.
uint32_t *buffer = first_buffer;
DmacDescriptor* descriptor = dma_descriptor(dma_channel);
if (buffers_processed % 2 == 1) {
buffer = second_buffer;
descriptor = &second_descriptor;
}
2018-01-02 21:25:41 -05:00
// Decimate and filter the buffer that was just filled.
uint32_t samples_gathered = descriptor->BTCNT.reg / words_per_sample;
// Don't run off the end of output buffer. Process only as many as needed.
uint32_t samples_to_process = min(remaining_samples_needed, samples_gathered);
for (uint32_t i = 0; i < samples_to_process; i++) {
// Call filter_sample just one place so it can be inlined.
uint16_t value = filter_sample(buffer + i * words_per_sample);
if (self->bit_depth == 8) {
2018-01-02 21:25:41 -05:00
// Truncate to 8 bits.
((uint8_t*) output_buffer)[values_output] = value >> 8;
} else {
output_buffer[values_output] = value;
}
2018-01-02 21:25:41 -05:00
values_output++;
}
2018-01-02 21:25:41 -05:00
buffers_processed++;
2018-01-02 21:25:41 -05:00
// Compute how many more samples we need, and if the last buffer is the last
// set of samples needed, adjust the DMA count to only fetch as necessary.
remaining_samples_needed = output_buffer_length - values_output;
if (remaining_samples_needed <= samples_per_buffer*2 &&
remaining_samples_needed > samples_per_buffer) {
// Adjust the DMA settings for the current buffer, which will be processed
// after the other buffer, which is now receiving samples via DMA.
// We don't adjust the DMA in progress, but the one after that.
// Timeline:
// 1. current buffer (already processed)
// 2. alternate buffer (DMA in progress)
// 3. current buffer (last set of samples needed)
// Set up to receive the last set of samples (don't include the alternate buffer, now in use).
uint32_t samples_needed_for_last_buffer = remaining_samples_needed - samples_per_buffer;
descriptor->BTCNT.reg = samples_needed_for_last_buffer * words_per_sample;
descriptor->DSTADDR.reg = ((uint32_t) buffer)
+ samples_needed_for_last_buffer * words_per_sample * sizeof(buffer[0]);
// Break chain to alternate buffer.
descriptor->DESCADDR.reg = 0;
}
}
disable_event_channel(event_channel);
dma_disable_channel(dma_channel);
2018-01-02 21:25:41 -05:00
return values_output;
}
void common_hal_audiobusio_pdmin_record_to_file(audiobusio_pdmin_obj_t* self, uint8_t* buffer, uint32_t length) {
}