Hi,
Personally I wouldn't call it issues.... It's just the way it works and there should be NO corrupted registers if you DON'T use PBP statements in your ISR's - have you experienced that?
I have no idea of how other compilers are working, if they are using similar scheme with system variables like PBP, and/or their performance when it comes to latency etc. As far as I understand the way to work around it in PBP IS to do exactly what DT-INTS does. However my point is that if MELABS changed the way PBP handles interrupts and incorporated DT-INTS then perhaps we wouldn't have the luxuary to select the best method for the task at hand. DT-INTS (with PBP handlers) are GREAT for ease of use but not so great when you need high speed.
However, I'm all for any improvements, in any area so if something good comes up I'd be happy too.
/Henrik.
Bookmarks