Part Number: AM2634
Tool/software:
Hi TI experts,
I tested "fsi_loopback_polling_am263x-cc_r5fss0-0_nortos_ti-arm-clang" examples.
(MCU+SDK version : 10.1.0.31)
-
1. MPU settings in syscfg
It seems that used memory ranges are not for AM263x.
This is not an important problem.
-
2. fixed valid DATA WORD size
In the example, it said
/* Configuring Frame - can be between 1-16U */
#define FSI_APP_FRAME_DATA_WORD_SIZE (16U)
But when I lowered the value, mismatches between TX and RX data occurred.
It was due to buffer pointer error of RX module.
Before or After receiving RX data, this was needed.
status = FSI_setRxBufferPtr(rxBaseAddr, 0U);
-
3. duplicated resets in initialization of TX and RX
In a function "FSI_appTxConfig",
/* TX init and reset */
status = FSI_performTxInitialization(txBaseAddr, FSI_APP_TX_PRESCALER_VAL);
status += FSI_resetTxModule(txBaseAddr, FSI_TX_MAIN_CORE_RESET);
FSI_clearTxModuleReset(txBaseAddr, FSI_TX_MAIN_CORE_RESET);
When I checked the code inside of "FSI_performTxInitialization", the same reset process (second and third sentences) was found.
I wonder why this initialization operate a reset again.
It was the same in a function "Fsi_appRxConfig",
/* RX init and reset */
status = FSI_performRxInitialization(rxBaseAddr);
status += FSI_resetRxModule(rxBaseAddr, FSI_RX_MAIN_CORE_RESET);
FSI_clearRxModuleReset(rxBaseAddr, FSI_RX_MAIN_CORE_RESET);
"FSI_performRXInitialization" included the same reset process.
Is there a reason to do the reset again?
-
Regards,
Jeeuk
