Configuring parity error control

While it is normal in modern systems with error-corrected links to set serial ports up with no parity, there will be occasions when enabling detection of parity errors is both useful and desirable.

There are a number of different ways that parity errors can be handled:

• ignore parity errors and discard any bytes with bad parity

• replace bytes received with bad parity with the iParityErrorChar character

• immediately complete any read operation with a KErrCommsParity error

Implementing any of the available options is simply a matter of using the TCommConfig package to change the value of iParityError. The value of iParityErrorChar can be changed in the same way.

Note that iParityError can be masked with KConfigXonXoffDebug. This has no effect on parity error control at all. It is an internal developmental debugging control which forces the high water mark to software flow control level of 5 bytes below 50% level instead the usual 75% level, irrespective of whether software flow control is enabled or not.