ECC exception handling

AN4276

14/21

DocID024408 Rev 2

Figure 7. Pseudo Code for known address jump

An advantage of this approach is that ECC error is immediately found and possible read

accesses stopped. This can save some time and exception load when reading long buffers

without immediate error check.

The disadvantage is that there must be only one function that can cause ECC error. It

means that in whole application there is just one place where data Flash is read. It is

strongly dependent on software implementation.

4.4

Clearing error flags

ECC error flags in the ECSM module should be cleared for recoverable scenarios in order to

enable proper exception handling for other causes, like register protection etc. Other reason

for clearing ECSM status bits is the property of the ECSM behavior where new ECC error

overwrites the old information, because there is only one set of registers storing extended

information. The recommendation is to clear these flags always, even in the case of non-

recoverable situations.

Application function FCE1()

Value = ReadDataFlashWord(addr)

Buffer[i] = StoreValue

EndOfFunction

Z0 core stores SRR0/SRR1

Start of handler execution

Reason found

Move ECC_error_check symbol to

SRR0 register

Clear the ECC error flag

Finish handler (rfi)

Z0 core restores MSR (SRR1) and

load PC counter with value of SRR0

IVOR2 exception (ECC on data flash access)

ECC_error_check: