Matthias Blankertz
7f8282315e
All checks were successful
Build RPi Pico firmware image / Build-Firmware (push) Successful in 5m29s
Check code formatting / Check-C-Format (push) Successful in 9s
Check code formatting / Check-Python-Flake8 (push) Successful in 9s
Check code formatting / Check-Bash-Shellcheck (push) Successful in 4s
Run unit tests on host / Run-Unit-Tests (push) Successful in 8s
The python and C modules that are supposed to be built into the firmware image (i.e. those that are in manifest.py or in USER_C_MODULES) have been moved to the software/modules directory. The software/src directory should now only contain python scripts and other files that should be installed to the Picos flash filesystem. The idea is that these should be those scripts that implement the application behaviour, as these are the ones that a user who does not want to build the whole firmware themself wants to modify.
95 lines
3.8 KiB
C
95 lines
3.8 KiB
C
// SPDX-License-Identifier: MIT
|
|
// Copyright (c) 2024-2025 Matthias Blankertz <matthias@blankertz.org>
|
|
|
|
#pragma once
|
|
|
|
#include <hardware/sync.h>
|
|
|
|
#include <stdint.h>
|
|
#include <string.h>
|
|
|
|
/* Access rules
|
|
* audiocore processing runs on core 1 and, unless stated otherwise, all
|
|
* variables may only be accessed from core 1. Take care of interrupt safety
|
|
* where needed. The micropython interface lives in module.c and is invoked from
|
|
* the micropython runtime on core 0. Micropython objects may only be handled in
|
|
* that context The audiocore_shared_context struct defined below is used for
|
|
* communication between the cores.
|
|
*/
|
|
|
|
#define MP3_BUFFER_ALIGN (sizeof(uintptr_t))
|
|
#define MP3_BUFFER_SIZE 4096
|
|
#define MP3_BUFFER_PREAREA 1048
|
|
|
|
_Static_assert(MP3_BUFFER_PREAREA % MP3_BUFFER_ALIGN == 0,
|
|
"Prearea must be a multiple of machine word size for alignment");
|
|
|
|
// Context shared between the micropython runtime on core0 and the audio task on
|
|
// core1 All access must hold "lock" unless otherwise noted
|
|
struct audiocore_shared_context {
|
|
spin_lock_t *lock;
|
|
|
|
// Set by module.c before core1 is launched and then never changed, can be read without lock
|
|
int out_pin, sideset_base, samplerate;
|
|
|
|
// Must hold lock. The indices 0..MP3_BUFFER_PREAREA-1 may only be read and written on core1 (no
|
|
// lock needed) The buffer is aligned to, and MP3_BUFFER_PREAREA is a multiple of, the machine
|
|
// word size to prevent these accesses from crossing word boundaries.
|
|
char mp3_buffer[MP3_BUFFER_PREAREA + MP3_BUFFER_SIZE] __attribute__((aligned(MP3_BUFFER_ALIGN)));
|
|
int mp3_buffer_write, mp3_buffer_read;
|
|
int underruns;
|
|
};
|
|
|
|
extern struct audiocore_shared_context shared_context;
|
|
|
|
/* Must hold audiocore_shared_context.lock */
|
|
static inline unsigned audiocore_get_buffer_space(void)
|
|
{
|
|
if (shared_context.mp3_buffer_write >= shared_context.mp3_buffer_read)
|
|
return MP3_BUFFER_SIZE - 1 - (shared_context.mp3_buffer_write - shared_context.mp3_buffer_read);
|
|
else
|
|
return shared_context.mp3_buffer_read - shared_context.mp3_buffer_write - 1;
|
|
}
|
|
|
|
/* Must hold audiocore_shared_context.lock */
|
|
static inline unsigned audiocore_get_buffer_avail(void)
|
|
{
|
|
if (shared_context.mp3_buffer_write >= shared_context.mp3_buffer_read)
|
|
return shared_context.mp3_buffer_write - shared_context.mp3_buffer_read;
|
|
else
|
|
return MP3_BUFFER_SIZE - (shared_context.mp3_buffer_read - shared_context.mp3_buffer_write);
|
|
}
|
|
|
|
/* Must hold audiocore_shared_context.lock */
|
|
static inline void audiocore_buffer_put(const char *restrict src, const size_t len)
|
|
{
|
|
const unsigned end_bytes = MP3_BUFFER_SIZE - shared_context.mp3_buffer_write;
|
|
memcpy(shared_context.mp3_buffer + MP3_BUFFER_PREAREA + shared_context.mp3_buffer_write, src,
|
|
((end_bytes >= len) ? len : end_bytes));
|
|
if (end_bytes < len) {
|
|
memcpy(shared_context.mp3_buffer + MP3_BUFFER_PREAREA, src + end_bytes, len - end_bytes);
|
|
shared_context.mp3_buffer_write = len - end_bytes;
|
|
} else {
|
|
shared_context.mp3_buffer_write += len;
|
|
}
|
|
shared_context.mp3_buffer_write %= MP3_BUFFER_SIZE;
|
|
}
|
|
|
|
void __time_critical_func(volume_adjust)(int16_t *buf, size_t samples, uint16_t scalef);
|
|
|
|
void core1_main(void);
|
|
|
|
// For data sent from core1 to core0, signals that this is a return value from some function call
|
|
// Otherwise it it just a trigger to wake core0 and the data can be discarded.
|
|
#define AUDIOCORE_FIFO_DATA_FLAG 0x80000000
|
|
|
|
// SHUTDOWN - no arguments - return 0
|
|
#define AUDIOCORE_CMD_SHUTDOWN 0xdeadc0de
|
|
|
|
// FLUSH - signals end of file and stop decoding when buffer empty - no arguments - return 0 when decoding is finished
|
|
#define AUDIOCORE_CMD_FLUSH 0xdeadc0dd
|
|
|
|
#define AUDIOCORE_MAX_VOLUME 0x8000u
|
|
// SET VOLUME - one argument: uint32_t range 0 to AUDIOCORE_MAX_VOLUME - return 0 if volume changed, 1 on error
|
|
#define AUDIOCORE_CMD_SET_VOLUME 0xdead0001
|