[SOLVED] Verify this bi-di (Verilog)

[dpaul]

Verify this IIC SDA/SCL connection (Verilog)

Hello,

I want to connect the o/p of an IIC master to external SDA and SCL. The o/p of the master IIC block has ports sda_i, sda_o, sda_t_i, and scl_i, scl_o, scl_t_i.

Given below is the connection sketch as I have visualised the connections.



I have written the tri-state logic in Verilog and is given below.

module axi_iic_bi_di 
  (
   // IIC master interface signals
   // data
   input  wire sda_i,   // IIC data in
   output wire sda_o,   // IIC data out
   input  wire sda_t_i, // Data control
   //clock
   input  wire scl_i,   // IIC clk input
   output wire scl_o,   // IIC clk output
   input wire  scl_t_i, // clock control
   // IIC slave interface signals
   inout wire Sda_io,
   inout wire Scl_io
  );
  
  assign sda_o = Sda_io;
  
  assign Sda_io = (sda_t_i == 1'b1) ? sda_i : 1'bz;
  
  //pullup(Sda_io); // pullup done in TB
  
  
  assign scl_o = Scl_io;
  
  assign Scl_io = (scl_t_i == 1'b1) ? scl_i : 1'bz;
  
  //pullup(Scl_io); // pullup done in TB
  
endmodule

Can anyone confirm if my Verilog code will deliver the functionality required to connect the IIC master to the inout SDA and SCL?

Thanks in advance.

- - - Updated - - -

If the above is not the proper way to implent the SDA and SCL then what should be done?
Use a MUX, with sda_t_i being the output enable (the IIC master spec says that Sda_T is the "serial data o/p enable to 3-state buffer"; same for the Scl)?

 

[ads-ee]

if scl_t_i is an active low tri-state control then it's coded correctly, otherwise if it's tri-state when high then the mux inputs should be swapped.

 

[dpaul]

HI,

The scl_t_i (this is an i/p port in my tri-state module) is connected to the output pin SCL_T of the IIC master.

I am giving the following info directly from the IIC master module spec...
The SCL_T is an output port with should have initial state 0x0 and is "IIC Serial Clock Output Enable to 3-state buffer".
But there isn't any info on it being active high or low. I have interpreted it to be active high and have coded accordingly.
What should I do? Is it correct then?

 

[ads-ee]

Given the name SCL_T it looks like a active high tri-state (i.e. if the SCL_T is logic 1 the buffer is to be tri-stated)

Most naming conventions would use something like SCL_E, SCL_ENA, SCL_OE etc to mean enable the output (i.e. drive the buffer)

 

[dpaul]

Given the name SCL_T it looks like a active high tri-state (i.e. if the SCL_T is logic 1 the buffer is to be tri-stated)

You are right ads-ee! I just changed my logic in the 3-state RTL code and can clearly see the SCL and SDA transitions in my simulation. Now I feel relieved!

I think one could have also used the Xilinx EDK XPS to generate a 3-state buffer that could be instantiated (verilog example from Bi-Directional IOBUF template in Vivado) with the AXI IIC IP:

IOBUF #(
.DRIVE(12), // Specify the output drive strength
.IBUF_LOW_PWR("TRUE"), // Low Power - "TRUE", High Performance = "FALSE"
.IOSTANDARD("DEFAULT"), // Specify the I/O standard
.SLEW("SLOW") // Specify the output slew rate
) IOBUF_inst (
.O(O), // Buffer output
.IO(IO), // Buffer inout port (connect directly to top-level port)
.I(I), // Buffer input
.T(T) // 3-state enable input, high=input, low=output
);