与USB主机并行传输后，Harmony I2C SDA为低电平

你好！我看到，在USB棒（1024字节块）上连续写操作期间，如果I2C以“并行”方式运行，那么SDA线路突然保持低位，并且不再看到I2C通信……我使用与USB主机（USB Host）一些其他SPI外围设备和具有动态中断驱动程序的I2C-2总线的和谐1.09。他I2C连接的RealtimeCox或我的DSP/DAC似乎被一个I2C操作弄糊涂了。USB写操作只需要37sI，只需要重置硬件或执行以下功能，就可以“重启”它们，执行10次I2C停止“bitban.”，bool I2C_MasterStop_Manual(bool EnableI2C){uint32_t starttime;int i；/*Wait 1BRG*/starttime(ReadCoreTimer();ReadCoreTimer()；.(ReadCoreTimer()-启动时间<_BRG_1_BRG_1_TIME_2)；/**禁用I2C，如果启用*/PLIB_IIB_II2C_I2C_I2C禁用(I2C_ID 2)；//手动执行10x I2C STOP 10STOP STOP STOP(SCL高，然后SDALOLOLO->；HIGH)PLIB PLIB_POIB_PORTS_PORTS_PORTS_PORTS_ID_ID_ID_ID_0，PORTS_ID_ID_ID_ID_0，PORT_CHANNEL_PORT_CHANNEL_S_PinSet（PORTS_ID_0，PORT_CHANNEL_APIN_SCL2；PLIB_PORTS_OpenDrainEn可(PORTS_ID_0，PORT_CHANNEL_F，（1<；<；PIN_SDA2），（PIN_SCL2）；；PLIB_PORTS_PORTS_PORTS_PORTS_PORTS_ID ID_0，PORT_PORT_信道_CHANNEL_F，（1<；（1<；<；<；PIN_SDA2(PIN_SCL2)；PLIB_POIB_PORTS_PORTS_PORTS_PORTS_PORTS_PORTS_PI端口_Pin方向针针针方向输出输出输出集(PORTS_ID 0，PORTS_ID_ID(i=0;i<10;i++){/*等待2BRG*/启动时间ReadCoreTimer()；.(ReadCoreTimer()-starttime

以上来自于百度翻译

      以下为原文

    Hello !

I saw that during continuous writing actions on a USB Stick (1024 byte blocks)
if I2C is runnig in "parallel" suddenly the SDA line stays low and no I2C
communication was seen any more...

I am using hARMony 1.09 with a USB Host some other SPI peripheral and
the I2C-2 bus with dynamic Interrupt driver.

The I2C connected  realtimeclock or my DSP/DAC seems to be "confused" by an I2C
action. The realtime is requeste every 1s. The USB write action needs about 37s

I only can bring them "back to live" with hardware reset or a function like below,
which executes 10 times I2C STOPs "bit banged".

bool I2C_MasterStop_Manual(bool EnableI2C)
{   
    uint32_t starttime;
    int i;
   
    /* Wait 1 BRG */
    starttime = ReadCoreTimer();
    while(ReadCoreTimer()-starttime < BRG_1_TIME_2);

    /* Disable I2C if enabled*/
    PLIB_I2C_Disable(I2C_ID_2);

    // Manually exectute 10 x I2C STOP (SCL High,then SDA LOW -> HIGH)
    PLIB_PORTS_PinSet(PORTS_ID_0, PORT_CHANNEL_A, PIN_SDA2);
    PLIB_PORTS_PinSet(PORTS_ID_0, PORT_CHANNEL_A, PIN_SCL2);
    PLIB_PORTS_OpenDrainEnable(PORTS_ID_0, PORT_CHANNEL_F, (1<     PLIB_PORTS_PinDirectionOutputSet(PORTS_ID_0, PORT_CHANNEL_A, PIN_SDA2);
    PLIB_PORTS_PinDirectionOutputSet(PORTS_ID_0, PORT_CHANNEL_A, PIN_SCL2);
   
    for (i=0; i<10; i++)
    {
        /* Wait 2 BRG */
            starttime = ReadCoreTimer();
            while(ReadCoreTimer()- starttime < BRG_1_TIME_2);
        PLIB_PORTS_PinClear(PORTS_ID_0, PORT_CHANNEL_A, PIN_SDA2);

            starttime = ReadCoreTimer();
            while(ReadCoreTimer()- starttime < BRG_1_TIME_2);
        PLIB_PORTS_PinClear(PORTS_ID_0, PORT_CHANNEL_A, PIN_SCL2);

            starttime = ReadCoreTimer();
            while(ReadCoreTimer()- starttime < BRG_1_TIME_2);
        PLIB_PORTS_PinSet(PORTS_ID_0, PORT_CHANNEL_A, PIN_SCL2);

            starttime = ReadCoreTimer();
            while(ReadCoreTimer()- starttime < BRG_1_TIME_2);
        PLIB_PORTS_PinSet(PORTS_ID_0, PORT_CHANNEL_A, PIN_SDA2);

    }
    PLIB_PORTS_PinDirectionInputSet(PORTS_ID_0, PORT_CHANNEL_A, PIN_SDA2);
    PLIB_PORTS_PinDirectionInputSet(PORTS_ID_0, PORT_CHANNEL_A, PIN_SCL2);
   
    // reenable I2C
    if (EnableI2C)
      PLIB_I2C_Enable(I2C_ID_2);
   
    return true;
}


I currently have no real idea why this happens, but I think it's a timing
issue because if USB is writing in the main task, the timing is not so
stable than without. It seems to happen always on the last byte of the
7 bytes read from the Realtime clock (if it happens).

There is NO error in the I2C driver's callback function produced !

void RTC_I2CBufferEventFunction( DRV_I2C_BUFFER_EVENT event,
                                 DRV_I2C_BUFFER_HANDLE handle,
                                 uintptr_t context)
{
....
}

The software does not really hang, but I never get a "ready" from the USB
driver and also no error...

Any idea ?

0