coreali 0.1.0

Convenient Register Access Library

coreali - Convenient Register Access Library

Generates Python code from a register model. This python code can the be used to access the registers of an FPGA, a MCU or another device which has a register map.

Installation

Coreali can be installed with pip by executing the following command

pip install coreali

How to use

The typical use case for coreali is:

1. Make a register description of your system in systemRDL or any other supported format (ipxact) of the systemrdl-compiler
2. Compile the register description with the systemrdl-compiler
3. Generate the hierarchical register model class and objects with the coreali.RegisterModel
4. Access the registers of your FPGA device with coreali.registerio

Quickstart - Logger example

1. Create a register description of your system or module

mem log_mem {
    mementries = 64;
    memwidth = 8;
};

addrmap logger { 
	reg  {	
		desc = "Control register";
		field {sw = rw; hw = r;} ENABLE;
		field {sw = rw; hw = r;} TRIGGERED;
	} Ctrl;
	reg  {	
		desc = "Trigger configuration for channel 0 and 1";
		field {sw = rw; hw = r;} SW_TRIG;
		field {sw = rw; hw = r;} LEVEL[8];
	} Trig[2];
  	external log_mem LogMem[2];
};

1. Import dependencies and compile register model with systemrdl-compiler

import sys
from systemrdl import RDLCompiler
import coreali 
import numpy as np
rdlc = RDLCompiler()
rdlc.compile_file("../systemrdl/logger.rdl")
root = rdlc.elaborate()

2. Create the register model object

rio = coreali.registerio.RegIoNoHW(np.zeros([256], np.uint8()))
logger = coreali.RegisterModel(root, rio)

In this step the rio register input/output obkect is created. This object handles the low level access. The RegIoNoHW class allows to access virtual registers without having a hardware at hand. The XXXXX example shows how to create your own RegIo class to access your own hardware.TODO

3. Use the generated register model

logger.Ctrl.read()
logger.LogMem.write(0,[1,2,3])
logger.LogMem.read()
logger.LogMem[1].write(0,[11,12,13])
print(logger)

The registers can now be read and written by hierarchically accessing them. With code completion of your python IDE the registers can be accessed conveniently.

When using the print function on a register, register node or address map all registers and field will be readout and printed as a string.

The full source code of this example can be found in examples/quickstart/