duane> [ describing a method to do cross target stuff with ]
duane> [ a 'quaint' little C program ]
oyvind> How would you handle the C code for a particular flash chip?
oyvind> Would that algorithm have to go into the single "giant" app
oyvind> that is downloaded?
The general idea is this:
We could easily create a fairly generic "run_helper" code.
That would be quite re-usable across platforms.
Today - we have complex solutions in each chip.
(ie: Run inner,and run outer, etc).
It would generally be come:
load_this_standard_helper( &task_specific_helper );
while( !done ){
address = get_standard_buffer_address();
len = get_standard_buffer_length();
download_to_buffer( address, length, mydata );
success= run_standard_helper( opcode_to_perfom );
if( !success ){
issue_error_messages();
}
}
The point is, today - the above are *complex* for every target,
The above approach would help 'standardize' the method of doing this.
Nothing says it must be *one*giant* app in fact that might be bad.
But maybe there is a "chip specific helper" a specific target might use.
-Duane.
_______________________________________________
Openocd-development mailing list
Openocd-development@lists.berlios.de
https://lists.berlios.de/mailman/listinfo/openocd-development
