Interfacing RS485 transceiver with AM335x

Hi,

Awaiting for your reply at an earliest.

Regards,

Mitesh

If you are interested, we put together a driver patch for driving an RS-485 transceiver for the AM-3359 to support the MityARM-3359 processor SOM.

A description of the application code, traces of the performance and links to kernel patches can be found on this link.

-Mike

Hi Mike,

Thanks for your reply. Hope this will resolve our problem.

Regards,

Mitesh

Hi Biser,

Thanks for reply.

Regards,

Mitesh

Hi,

Thank you for your patch and I've adapted your patch for the Kernel 3.8 for the BeagleBone Black .

But now, I've a problem, when I'm sending data with the call of the function write, I can see that the data is sent in two blocks !

http://i.stack.imgur.com/N8beT.jpg

most of the time, it works, but i don't why, when i'm running a special sequence, the driver choose to cut in half the frame ......

Maybe it's my configuration of the serial port that is not correct :

	handler = open( hPort->acAddress, O_RDWR | O_NOCTTY | O_NDELAY );
	if(!handler){
		iStatus=MICKA_STATUS_ERROR_OPEN;
		return MICKA_STATUS_OK;
	}else{
		rs485conf.flags = SER_RS485_ENABLED;
		rs485conf.flags |= SER_RS485_USE_GPIO;
	}

	rs485conf.gpio_pin = 32;	
	rs485conf.delay_rts_before_send = 0;
	rs485conf.delay_rts_after_send = 0;
	

	if(rs485conf.delay_rts_after_send<0 || iStatus!=MICKA_STATUS_OK){
		iStatus=MICKA_STATUS_ERROR_PARAMETER;
		return MICKA_STATUS_OK;
	}else{
		rv = ioctl (handler, TIOCSRS485, &rs485conf);
		if (rv) {
			iStatus=MICKA_STATUS_ERROR_SYSTEM;
			return MICKA_STATUS_OK;
		}
	}

	if ( tcgetattr ( handler , & mode ) >= 0 && iStatus==MICKA_STATUS_OK) {
		mode . c_cflag &= ( ~ CBAUD )  ;
		//mode . c_cflag |= atoi(token) ;

		mode.c_cflag |=B38400;
		

		// 8N1 :
		mode . c_cflag &= ( ~ CSIZE )  ; mode . c_cflag |= CS8 ;
		mode . c_cflag &= ( ~ PARENB ) ;


		mode . c_cflag &= ( ~ CRTSCTS ) ;
		/*
		  raw same as -ignbrk -brkint -ignpar -parmrk -inpck -istrip -inlcr -igncr -icrnl -ixon -ixoff -iuclc -ixany -imaxbel -opost -isig -icanon -xcase min 1 $
		 */
		mode . c_iflag |=     IGNBRK   ;
		mode . c_iflag |=     IGNPAR   ;
		mode . c_iflag &= ( ~ INLCR  ) ;
		mode . c_iflag &= ( ~ ICRNL  ) ;
		mode . c_iflag &= ( ~ IXON   ) ;
		mode . c_iflag &= ( ~ IXOFF  ) ;
		mode . c_oflag &= ( ~ ONLCR  ) ;
		mode . c_oflag &= ( ~ OCRNL  ) ;
		mode . c_lflag &= ( ~ ISIG   ) ;
		mode . c_lflag &= ( ~ ICANON ) ;
		mode . c_lflag &= ( ~ ECHO   ) ;
		rv=tcsetattr ( handler , 0 /* no special action */ , & mode );
		if(rv){
			return MICKA_STATUS_OK;
		}
		ioctl ( handler , FIONBIO , & ser1half ) ;
	}else {
		iStatus=MICKA_STATUS_ERROR_SYSTEM;
		return MICKA_STATUS_OK;
	}

How long is your transmit sequence?  I am wondering if you are exceeding the driver FIFO/buffer depth....

-Mike

no more than 23 bytes.

I'm repeating this sequence :

02 00 10 03 00 00 02 00 bb aa 55 00 08 00 01 09 00 00 01 00 00 6e 24

and I'm increasing the byte 15, and changing of course the CRc :

02 00 10 03 00 00 02 00 bb aa 55 00 08 00 01 00 00 00 01 00 00 26 c6

....

02 00 10 03 00 00 02 00 bb aa 55 00 08 00 01 09 00 00 01 00 00 6e 24

It doesn't make sense for me ....

In the beginning, it works, but when I'm going to the 9 frame, I got the problems....

Problem Solved !

Hi all,

As per above discussion i could under stand that , after transmit holding register empty , the driver will wait for a fixed time and reassert direction control pin. does that work for wide variety of baud rates from 300bps to 115200bps ? For different baud rates,time taken for transmitting 1 byte will be different . After Transmit holding register , there is 16byte FIFO and shift register .THRE interrupt only indicates that transmit holding register empty   and not whole frame physically transmitted .Am i correct ? So what will be the final solution for the same ? 

Thanks and regards 

Gokul CG[gokul.g@kalkitech.in]