58 lines
1.9 KiB
C
58 lines
1.9 KiB
C
/*
|
|
* This file is subject to the terms of the GFX License. If a copy of
|
|
* the license was not distributed with this file, you can obtain one at:
|
|
*
|
|
* http://ugfx.org/license.html
|
|
*/
|
|
|
|
/**
|
|
* @file drivers/gadc/AT91SAM7/gadc_lld_config.h
|
|
* @brief GADC Driver config file.
|
|
*
|
|
* @addtogroup GADC
|
|
* @{
|
|
*/
|
|
|
|
#ifndef GADC_LLD_CONFIG_H
|
|
#define GADC_LLD_CONFIG_H
|
|
|
|
#if GFX_USE_GADC
|
|
|
|
/*===========================================================================*/
|
|
/* Driver hardware support. */
|
|
/*===========================================================================*/
|
|
|
|
/**
|
|
* @brief ChibiOS has a nasty bug in its _adc_isr_full_code() routine (defined in adc.h as a macro).
|
|
* Do we have the version of ChibiOS with this bug.
|
|
* @detail Set to TRUE if ChibiOS has this bug.
|
|
* @note Fixed in ChibiOS 2.4.4stable and 2.5.2unstable (and the repository from 18th Feb 2013)
|
|
* @note This bug prevents us re-calling adcStartConversionI() from with the ISR even though
|
|
* it is clearly designed to handle it. For some reason (on this micro) the high speed timer
|
|
* is not affected only the single sample low speed timer. In that situation we wait until
|
|
* we get back to thread land. This is terrible for the accuracy of the high speed timer
|
|
* but what can we do (other than fix the bug).
|
|
* @note For the AT91SAM7 ADC driver, it post-dates the finding of the bug so we safely
|
|
* say that the bug doesn't exist for this driver.
|
|
*/
|
|
#define CHIBIOS_ADC_ISR_FULL_CODE_BUG FALSE
|
|
|
|
/**
|
|
* @brief The maximum sample frequency supported by this CPU
|
|
*/
|
|
#define GADC_MAX_SAMPLE_FREQUENCY 132000
|
|
|
|
/**
|
|
* @brief The sample format
|
|
*/
|
|
#define GADC_SAMPLE_FORMAT ARRAY_DATA_10BITUNSIGNED
|
|
|
|
/* Include the user supplied board definitions */
|
|
#include "gadc_lld_board.h"
|
|
|
|
#endif /* GFX_USE_GADC */
|
|
|
|
#endif /* GADC_LLD_CONFIG_H */
|
|
/** @} */
|
|
|