mirror of
https://github.com/wpilibsuite/allwpilib
synced 2026-06-19 00:41:43 +00:00
587 lines
19 KiB
C++
587 lines
19 KiB
C++
/*----------------------------------------------------------------------------*/
|
|
/* Copyright (c) FIRST 2016. All Rights Reserved. */
|
|
/* Open Source Software - may be modified and shared by FRC teams. The code */
|
|
/* must be accompanied by the FIRST BSD license file in the root directory of */
|
|
/* the project. */
|
|
/*----------------------------------------------------------------------------*/
|
|
|
|
#include "HAL/SPI.h"
|
|
|
|
#include <atomic>
|
|
|
|
#include "DigitalInternal.h"
|
|
#include "HAL/DIO.h"
|
|
#include "HAL/HAL.h"
|
|
#include "HAL/cpp/priority_mutex.h"
|
|
#include "spilib/spi-lib.h"
|
|
|
|
static_assert(sizeof(uint32_t) <= sizeof(void*),
|
|
"This file shoves uint32_ts into pointers.");
|
|
|
|
using namespace hal;
|
|
|
|
static int32_t m_spiCS0Handle = 0;
|
|
static int32_t m_spiCS1Handle = 0;
|
|
static int32_t m_spiCS2Handle = 0;
|
|
static int32_t m_spiCS3Handle = 0;
|
|
static int32_t m_spiMXPHandle = 0;
|
|
static priority_recursive_mutex spiOnboardMutex;
|
|
static priority_recursive_mutex spiMXPMutex;
|
|
static tSPI* spiSystem;
|
|
|
|
static HAL_DigitalHandle spiMXPDigitalHandle1 = HAL_kInvalidHandle;
|
|
static HAL_DigitalHandle spiMXPDigitalHandle2 = HAL_kInvalidHandle;
|
|
static HAL_DigitalHandle spiMXPDigitalHandle3 = HAL_kInvalidHandle;
|
|
static HAL_DigitalHandle spiMXPDigitalHandle4 = HAL_kInvalidHandle;
|
|
|
|
/**
|
|
* Get the semaphore for a SPI port
|
|
*
|
|
* @param port The number of the port to use. 0-3 for Onboard CS0-CS2, 4 for MXP
|
|
* @return The semaphore for the SPI port.
|
|
*/
|
|
static priority_recursive_mutex& spiGetMutex(int32_t port) {
|
|
if (port < 4)
|
|
return spiOnboardMutex;
|
|
else
|
|
return spiMXPMutex;
|
|
}
|
|
|
|
extern "C" {
|
|
|
|
struct SPIAccumulator {
|
|
std::atomic<HAL_NotifierHandle> notifier{0};
|
|
uint64_t triggerTime;
|
|
int32_t period;
|
|
|
|
int64_t value = 0;
|
|
uint32_t count = 0;
|
|
int32_t last_value = 0;
|
|
|
|
int32_t center = 0;
|
|
int32_t deadband = 0;
|
|
|
|
uint8_t cmd[4]; // command to send (up to 4 bytes)
|
|
int32_t valid_mask;
|
|
int32_t valid_value;
|
|
int32_t data_max; // one more than max data value
|
|
int32_t data_msb_mask; // data field MSB mask (for signed)
|
|
uint8_t data_shift; // data field shift right amount, in bits
|
|
uint8_t xfer_size; // SPI transfer size, in bytes (up to 4)
|
|
uint8_t port;
|
|
bool is_signed; // is data field signed?
|
|
bool big_endian; // is response big endian?
|
|
};
|
|
SPIAccumulator* spiAccumulators[5] = {nullptr, nullptr, nullptr, nullptr,
|
|
nullptr};
|
|
|
|
/*
|
|
* Initialize the spi port. Opens the port if necessary and saves the handle.
|
|
* If opening the MXP port, also sets up the pin functions appropriately
|
|
* @param port The number of the port to use. 0-3 for Onboard CS0-CS2, 4 for MXP
|
|
*/
|
|
void HAL_InitializeSPI(int32_t port, int32_t* status) {
|
|
if (spiSystem == nullptr) spiSystem = tSPI::create(status);
|
|
if (HAL_GetSPIHandle(port) != 0) return;
|
|
switch (port) {
|
|
case 0:
|
|
HAL_SetSPIHandle(0, spilib_open("/dev/spidev0.0"));
|
|
break;
|
|
case 1:
|
|
HAL_SetSPIHandle(1, spilib_open("/dev/spidev0.1"));
|
|
break;
|
|
case 2:
|
|
HAL_SetSPIHandle(2, spilib_open("/dev/spidev0.2"));
|
|
break;
|
|
case 3:
|
|
HAL_SetSPIHandle(3, spilib_open("/dev/spidev0.3"));
|
|
break;
|
|
case 4:
|
|
initializeDigital(status);
|
|
if (*status != 0) return;
|
|
if ((spiMXPDigitalHandle1 = HAL_InitializeDIOPort(
|
|
HAL_GetPort(14), false, status)) == HAL_kInvalidHandle) {
|
|
printf("Failed to allocate DIO 14\n");
|
|
return;
|
|
}
|
|
if ((spiMXPDigitalHandle2 = HAL_InitializeDIOPort(
|
|
HAL_GetPort(15), false, status)) == HAL_kInvalidHandle) {
|
|
printf("Failed to allocate DIO 15\n");
|
|
HAL_FreeDIOPort(spiMXPDigitalHandle1); // free the first port allocated
|
|
return;
|
|
}
|
|
if ((spiMXPDigitalHandle3 = HAL_InitializeDIOPort(
|
|
HAL_GetPort(16), false, status)) == HAL_kInvalidHandle) {
|
|
printf("Failed to allocate DIO 16\n");
|
|
HAL_FreeDIOPort(spiMXPDigitalHandle1); // free the first port allocated
|
|
HAL_FreeDIOPort(
|
|
spiMXPDigitalHandle2); // free the second port allocated
|
|
return;
|
|
}
|
|
if ((spiMXPDigitalHandle4 = HAL_InitializeDIOPort(
|
|
HAL_GetPort(17), false, status)) == HAL_kInvalidHandle) {
|
|
printf("Failed to allocate DIO 17\n");
|
|
HAL_FreeDIOPort(spiMXPDigitalHandle1); // free the first port allocated
|
|
HAL_FreeDIOPort(
|
|
spiMXPDigitalHandle2); // free the second port allocated
|
|
HAL_FreeDIOPort(spiMXPDigitalHandle3); // free the third port allocated
|
|
return;
|
|
}
|
|
digitalSystem->writeEnableMXPSpecialFunction(
|
|
digitalSystem->readEnableMXPSpecialFunction(status) | 0x00F0, status);
|
|
HAL_SetSPIHandle(4, spilib_open("/dev/spidev1.0"));
|
|
break;
|
|
default:
|
|
*status = PARAMETER_OUT_OF_RANGE;
|
|
break;
|
|
}
|
|
return;
|
|
}
|
|
|
|
/**
|
|
* Generic transaction.
|
|
*
|
|
* This is a lower-level interface to the spi hardware giving you more control
|
|
* over each transaction.
|
|
*
|
|
* @param port The number of the port to use. 0-3 for Onboard CS0-CS2, 4 for MXP
|
|
* @param dataToSend Buffer of data to send as part of the transaction.
|
|
* @param dataReceived Buffer to read data into.
|
|
* @param size Number of bytes to transfer. [0..7]
|
|
* @return Number of bytes transferred, -1 for error
|
|
*/
|
|
int32_t HAL_TransactionSPI(int32_t port, uint8_t* dataToSend,
|
|
uint8_t* dataReceived, int32_t size) {
|
|
std::lock_guard<priority_recursive_mutex> sync(spiGetMutex(port));
|
|
return spilib_writeread(
|
|
HAL_GetSPIHandle(port), reinterpret_cast<const char*>(dataToSend),
|
|
reinterpret_cast<char*>(dataReceived), static_cast<int32_t>(size));
|
|
}
|
|
|
|
/**
|
|
* Execute a write transaction with the device.
|
|
*
|
|
* Write to a device and wait until the transaction is complete.
|
|
*
|
|
* @param port The number of the port to use. 0-3 for Onboard CS0-CS2, 4 for MXP
|
|
* @param datToSend The data to write to the register on the device.
|
|
* @param sendSize The number of bytes to be written
|
|
* @return The number of bytes written. -1 for an error
|
|
*/
|
|
int32_t HAL_WriteSPI(int32_t port, uint8_t* dataToSend, int32_t sendSize) {
|
|
std::lock_guard<priority_recursive_mutex> sync(spiGetMutex(port));
|
|
return spilib_write(HAL_GetSPIHandle(port),
|
|
reinterpret_cast<const char*>(dataToSend),
|
|
static_cast<int32_t>(sendSize));
|
|
}
|
|
|
|
/**
|
|
* Execute a read from the device.
|
|
*
|
|
* This method does not write any data out to the device
|
|
* Most spi devices will require a register address to be written before
|
|
* they begin returning data
|
|
*
|
|
* @param port The number of the port to use. 0-3 for Onboard CS0-CS2, 4 for MXP
|
|
* @param buffer A pointer to the array of bytes to store the data read from the
|
|
* device.
|
|
* @param count The number of bytes to read in the transaction. [1..7]
|
|
* @return Number of bytes read. -1 for error.
|
|
*/
|
|
int32_t HAL_ReadSPI(int32_t port, uint8_t* buffer, int32_t count) {
|
|
std::lock_guard<priority_recursive_mutex> sync(spiGetMutex(port));
|
|
return spilib_read(HAL_GetSPIHandle(port), reinterpret_cast<char*>(buffer),
|
|
static_cast<int32_t>(count));
|
|
}
|
|
|
|
/**
|
|
* Close the SPI port
|
|
*
|
|
* @param port The number of the port to use. 0-3 for Onboard CS0-CS2, 4 for MXP
|
|
*/
|
|
void HAL_CloseSPI(int32_t port) {
|
|
std::lock_guard<priority_recursive_mutex> sync(spiGetMutex(port));
|
|
if (spiAccumulators[port]) {
|
|
int32_t status = 0;
|
|
HAL_FreeSPIAccumulator(port, &status);
|
|
}
|
|
spilib_close(HAL_GetSPIHandle(port));
|
|
HAL_SetSPIHandle(port, 0);
|
|
if (port == 4) {
|
|
HAL_FreeDIOPort(spiMXPDigitalHandle1);
|
|
HAL_FreeDIOPort(spiMXPDigitalHandle2);
|
|
HAL_FreeDIOPort(spiMXPDigitalHandle3);
|
|
HAL_FreeDIOPort(spiMXPDigitalHandle4);
|
|
}
|
|
return;
|
|
}
|
|
|
|
/**
|
|
* Set the clock speed for the SPI bus.
|
|
*
|
|
* @param port The number of the port to use. 0-3 for Onboard CS0-CS2, 4 for MXP
|
|
* @param speed The speed in Hz (0-1MHz)
|
|
*/
|
|
void HAL_SetSPISpeed(int32_t port, int32_t speed) {
|
|
std::lock_guard<priority_recursive_mutex> sync(spiGetMutex(port));
|
|
spilib_setspeed(HAL_GetSPIHandle(port), speed);
|
|
}
|
|
|
|
/**
|
|
* Set the SPI options
|
|
*
|
|
* @param port The number of the port to use. 0-3 for Onboard CS0-CS2, 4 for MXP
|
|
* @param msb_first True to write the MSB first, False for LSB first
|
|
* @param sample_on_trailing True to sample on the trailing edge, False to
|
|
* sample on the leading edge
|
|
* @param clk_idle_high True to set the clock to active low, False to set the
|
|
* clock active high
|
|
*/
|
|
void HAL_SetSPIOpts(int32_t port, HAL_Bool msb_first,
|
|
HAL_Bool sample_on_trailing, HAL_Bool clk_idle_high) {
|
|
std::lock_guard<priority_recursive_mutex> sync(spiGetMutex(port));
|
|
spilib_setopts(HAL_GetSPIHandle(port), msb_first, sample_on_trailing,
|
|
clk_idle_high);
|
|
}
|
|
|
|
/**
|
|
* Set the CS Active high for a SPI port
|
|
*
|
|
* @param port The number of the port to use. 0-3 for Onboard CS0-CS2, 4 for MXP
|
|
*/
|
|
void HAL_SetSPIChipSelectActiveHigh(int32_t port, int32_t* status) {
|
|
std::lock_guard<priority_recursive_mutex> sync(spiGetMutex(port));
|
|
if (port < 4) {
|
|
spiSystem->writeChipSelectActiveHigh_Hdr(
|
|
spiSystem->readChipSelectActiveHigh_Hdr(status) | (1 << port), status);
|
|
} else {
|
|
spiSystem->writeChipSelectActiveHigh_MXP(1, status);
|
|
}
|
|
}
|
|
|
|
/**
|
|
* Set the CS Active low for a SPI port
|
|
*
|
|
* @param port The number of the port to use. 0-3 for Onboard CS0-CS2, 4 for MXP
|
|
*/
|
|
void HAL_SetSPIChipSelectActiveLow(int32_t port, int32_t* status) {
|
|
std::lock_guard<priority_recursive_mutex> sync(spiGetMutex(port));
|
|
if (port < 4) {
|
|
spiSystem->writeChipSelectActiveHigh_Hdr(
|
|
spiSystem->readChipSelectActiveHigh_Hdr(status) & ~(1 << port), status);
|
|
} else {
|
|
spiSystem->writeChipSelectActiveHigh_MXP(0, status);
|
|
}
|
|
}
|
|
|
|
/**
|
|
* Get the stored handle for a SPI port
|
|
*
|
|
* @param port The number of the port to use. 0-3 for Onboard CS0-CS2, 4 for MXP
|
|
* @return The stored handle for the SPI port. 0 represents no stored handle.
|
|
*/
|
|
int32_t HAL_GetSPIHandle(int32_t port) {
|
|
std::lock_guard<priority_recursive_mutex> sync(spiGetMutex(port));
|
|
switch (port) {
|
|
case 0:
|
|
return m_spiCS0Handle;
|
|
case 1:
|
|
return m_spiCS1Handle;
|
|
case 2:
|
|
return m_spiCS2Handle;
|
|
case 3:
|
|
return m_spiCS3Handle;
|
|
case 4:
|
|
return m_spiMXPHandle;
|
|
default:
|
|
return 0;
|
|
}
|
|
}
|
|
|
|
/**
|
|
* Set the stored handle for a SPI port
|
|
*
|
|
* @param port The number of the port to use. 0-3 for Onboard CS0-CS2, 4 for
|
|
* MXP.
|
|
* @param handle The value of the handle for the port.
|
|
*/
|
|
void HAL_SetSPIHandle(int32_t port, int32_t handle) {
|
|
std::lock_guard<priority_recursive_mutex> sync(spiGetMutex(port));
|
|
switch (port) {
|
|
case 0:
|
|
m_spiCS0Handle = handle;
|
|
break;
|
|
case 1:
|
|
m_spiCS1Handle = handle;
|
|
break;
|
|
case 2:
|
|
m_spiCS2Handle = handle;
|
|
break;
|
|
case 3:
|
|
m_spiCS3Handle = handle;
|
|
break;
|
|
case 4:
|
|
m_spiMXPHandle = handle;
|
|
break;
|
|
default:
|
|
break;
|
|
}
|
|
}
|
|
|
|
static void spiAccumulatorProcess(uint64_t currentTime, void* param) {
|
|
SPIAccumulator* accum = static_cast<SPIAccumulator*>(param);
|
|
|
|
// perform SPI transaction
|
|
uint8_t resp_b[4];
|
|
std::lock_guard<priority_recursive_mutex> sync(spiGetMutex(accum->port));
|
|
spilib_writeread(
|
|
HAL_GetSPIHandle(accum->port), reinterpret_cast<const char*>(accum->cmd),
|
|
reinterpret_cast<char*>(resp_b), static_cast<int32_t>(accum->xfer_size));
|
|
|
|
// convert from bytes
|
|
uint32_t resp = 0;
|
|
if (accum->big_endian) {
|
|
for (int i = 0; i < accum->xfer_size; ++i) {
|
|
resp <<= 8;
|
|
resp |= resp_b[i] & 0xff;
|
|
}
|
|
} else {
|
|
for (int i = accum->xfer_size - 1; i >= 0; --i) {
|
|
resp <<= 8;
|
|
resp |= resp_b[i] & 0xff;
|
|
}
|
|
}
|
|
|
|
// process response
|
|
if ((resp & accum->valid_mask) == static_cast<uint32_t>(accum->valid_value)) {
|
|
// valid sensor data; extract data field
|
|
int32_t data = static_cast<int32_t>(resp >> accum->data_shift);
|
|
data &= accum->data_max - 1;
|
|
// 2s complement conversion if signed MSB is set
|
|
if (accum->is_signed && (data & accum->data_msb_mask) != 0)
|
|
data -= accum->data_max;
|
|
// center offset
|
|
data -= accum->center;
|
|
// only accumulate if outside deadband
|
|
if (data < -accum->deadband || data > accum->deadband) accum->value += data;
|
|
++accum->count;
|
|
accum->last_value = data;
|
|
} else {
|
|
// no data from the sensor; just clear the last value
|
|
accum->last_value = 0;
|
|
}
|
|
|
|
// reschedule timer
|
|
accum->triggerTime += accum->period;
|
|
// handle timer slip
|
|
if (accum->triggerTime < currentTime)
|
|
accum->triggerTime = currentTime + accum->period;
|
|
int32_t status = 0;
|
|
HAL_UpdateNotifierAlarm(accum->notifier, accum->triggerTime, &status);
|
|
}
|
|
|
|
/**
|
|
* Initialize a SPI accumulator.
|
|
*
|
|
* @param port SPI port
|
|
* @param period Time between reads, in us
|
|
* @param cmd SPI command to send to request data
|
|
* @param xfer_size SPI transfer size, in bytes
|
|
* @param valid_mask Mask to apply to received data for validity checking
|
|
* @param valid_data After valid_mask is applied, required matching value for
|
|
* validity checking
|
|
* @param data_shift Bit shift to apply to received data to get actual data
|
|
* value
|
|
* @param data_size Size (in bits) of data field
|
|
* @param is_signed Is data field signed?
|
|
* @param big_endian Is device big endian?
|
|
*/
|
|
void HAL_InitSPIAccumulator(int32_t port, int32_t period, int32_t cmd,
|
|
int32_t xfer_size, int32_t valid_mask,
|
|
int32_t valid_value, int32_t data_shift,
|
|
int32_t data_size, HAL_Bool is_signed,
|
|
HAL_Bool big_endian, int32_t* status) {
|
|
std::lock_guard<priority_recursive_mutex> sync(spiGetMutex(port));
|
|
if (port > 4) return;
|
|
if (!spiAccumulators[port]) spiAccumulators[port] = new SPIAccumulator();
|
|
SPIAccumulator* accum = spiAccumulators[port];
|
|
if (big_endian) {
|
|
for (int i = xfer_size - 1; i >= 0; --i) {
|
|
accum->cmd[i] = cmd & 0xff;
|
|
cmd >>= 8;
|
|
}
|
|
} else {
|
|
accum->cmd[0] = cmd & 0xff;
|
|
cmd >>= 8;
|
|
accum->cmd[1] = cmd & 0xff;
|
|
cmd >>= 8;
|
|
accum->cmd[2] = cmd & 0xff;
|
|
cmd >>= 8;
|
|
accum->cmd[3] = cmd & 0xff;
|
|
}
|
|
accum->period = period;
|
|
accum->xfer_size = xfer_size;
|
|
accum->valid_mask = valid_mask;
|
|
accum->valid_value = valid_value;
|
|
accum->data_shift = data_shift;
|
|
accum->data_max = (1 << data_size);
|
|
accum->data_msb_mask = (1 << (data_size - 1));
|
|
accum->is_signed = is_signed;
|
|
accum->big_endian = big_endian;
|
|
if (!accum->notifier) {
|
|
accum->notifier =
|
|
HAL_InitializeNotifier(spiAccumulatorProcess, accum, status);
|
|
accum->triggerTime = HAL_GetFPGATime(status) + period;
|
|
if (*status != 0) return;
|
|
HAL_UpdateNotifierAlarm(accum->notifier, accum->triggerTime, status);
|
|
}
|
|
}
|
|
|
|
/**
|
|
* Frees a SPI accumulator.
|
|
*/
|
|
void HAL_FreeSPIAccumulator(int32_t port, int32_t* status) {
|
|
std::lock_guard<priority_recursive_mutex> sync(spiGetMutex(port));
|
|
SPIAccumulator* accum = spiAccumulators[port];
|
|
if (!accum) {
|
|
*status = NULL_PARAMETER;
|
|
return;
|
|
}
|
|
HAL_NotifierHandle handle = accum->notifier.exchange(0);
|
|
HAL_CleanNotifier(handle, status);
|
|
delete accum;
|
|
spiAccumulators[port] = nullptr;
|
|
}
|
|
|
|
/**
|
|
* Resets the accumulator to zero.
|
|
*/
|
|
void HAL_ResetSPIAccumulator(int32_t port, int32_t* status) {
|
|
std::lock_guard<priority_recursive_mutex> sync(spiGetMutex(port));
|
|
SPIAccumulator* accum = spiAccumulators[port];
|
|
if (!accum) {
|
|
*status = NULL_PARAMETER;
|
|
return;
|
|
}
|
|
accum->value = 0;
|
|
accum->count = 0;
|
|
accum->last_value = 0;
|
|
}
|
|
|
|
/**
|
|
* Set the center value of the accumulator.
|
|
*
|
|
* The center value is subtracted from each value before it is added to the
|
|
* accumulator. This
|
|
* is used for the center value of devices like gyros and accelerometers to make
|
|
* integration work
|
|
* and to take the device offset into account when integrating.
|
|
*/
|
|
void HAL_SetSPIAccumulatorCenter(int32_t port, int32_t center,
|
|
int32_t* status) {
|
|
std::lock_guard<priority_recursive_mutex> sync(spiGetMutex(port));
|
|
SPIAccumulator* accum = spiAccumulators[port];
|
|
if (!accum) {
|
|
*status = NULL_PARAMETER;
|
|
return;
|
|
}
|
|
accum->center = center;
|
|
}
|
|
|
|
/**
|
|
* Set the accumulator's deadband.
|
|
*/
|
|
void HAL_SetSPIAccumulatorDeadband(int32_t port, int32_t deadband,
|
|
int32_t* status) {
|
|
std::lock_guard<priority_recursive_mutex> sync(spiGetMutex(port));
|
|
SPIAccumulator* accum = spiAccumulators[port];
|
|
if (!accum) {
|
|
*status = NULL_PARAMETER;
|
|
return;
|
|
}
|
|
accum->deadband = deadband;
|
|
}
|
|
|
|
/**
|
|
* Read the last value read by the accumulator engine.
|
|
*/
|
|
int32_t HAL_GetSPIAccumulatorLastValue(int32_t port, int32_t* status) {
|
|
std::lock_guard<priority_recursive_mutex> sync(spiGetMutex(port));
|
|
SPIAccumulator* accum = spiAccumulators[port];
|
|
if (!accum) {
|
|
*status = NULL_PARAMETER;
|
|
return 0;
|
|
}
|
|
return accum->last_value;
|
|
}
|
|
|
|
/**
|
|
* Read the accumulated value.
|
|
*
|
|
* @return The 64-bit value accumulated since the last Reset().
|
|
*/
|
|
int64_t HAL_GetSPIAccumulatorValue(int32_t port, int32_t* status) {
|
|
std::lock_guard<priority_recursive_mutex> sync(spiGetMutex(port));
|
|
SPIAccumulator* accum = spiAccumulators[port];
|
|
if (!accum) {
|
|
*status = NULL_PARAMETER;
|
|
return 0;
|
|
}
|
|
return accum->value;
|
|
}
|
|
|
|
/**
|
|
* Read the number of accumulated values.
|
|
*
|
|
* Read the count of the accumulated values since the accumulator was last
|
|
* Reset().
|
|
*
|
|
* @return The number of times samples from the channel were accumulated.
|
|
*/
|
|
int64_t HAL_GetSPIAccumulatorCount(int32_t port, int32_t* status) {
|
|
std::lock_guard<priority_recursive_mutex> sync(spiGetMutex(port));
|
|
SPIAccumulator* accum = spiAccumulators[port];
|
|
if (!accum) {
|
|
*status = NULL_PARAMETER;
|
|
return 0;
|
|
}
|
|
return accum->count;
|
|
}
|
|
|
|
/**
|
|
* Read the average of the accumulated value.
|
|
*
|
|
* @return The accumulated average value (value / count).
|
|
*/
|
|
double HAL_GetSPIAccumulatorAverage(int32_t port, int32_t* status) {
|
|
int64_t value;
|
|
int64_t count;
|
|
HAL_GetAccumulatorOutput(port, &value, &count, status);
|
|
if (count == 0) return 0.0;
|
|
return static_cast<double>(value) / count;
|
|
}
|
|
|
|
/**
|
|
* Read the accumulated value and the number of accumulated values atomically.
|
|
*
|
|
* This function reads the value and count atomically.
|
|
* This can be used for averaging.
|
|
*
|
|
* @param value Pointer to the 64-bit accumulated output.
|
|
* @param count Pointer to the number of accumulation cycles.
|
|
*/
|
|
void HAL_GetSPIAccumulatorOutput(int32_t port, int64_t* value, int64_t* count,
|
|
int32_t* status) {
|
|
std::lock_guard<priority_recursive_mutex> sync(spiGetMutex(port));
|
|
SPIAccumulator* accum = spiAccumulators[port];
|
|
if (!accum) {
|
|
*status = NULL_PARAMETER;
|
|
*value = 0;
|
|
*count = 0;
|
|
return;
|
|
}
|
|
*value = accum->value;
|
|
*count = accum->count;
|
|
}
|
|
}
|