Instant Interrupts - Revisited

[sayzer]

Hi sayzer,

2 years and 3 months since you joined.
What took you so long?

WOW! Saved 3100 bytes. That's pretty impressive.

Can I ask what the difference was? or maybe where the savings came from?
<br>

The interesting part is that I did not change anything other the "int" part.

That is why it was surprising and "Really very interesting for me".

[Ioannis]

You mean that comparing to On-Interrupt of PBP you did a reduction of 3100?

Ioannis

[sayzer]

You mean that comparing to On-Interrupt of PBP you did a reduction of 3100?

Ioannis

Hi Ioannis,

Actually I tried different codes after this experiment and came up with these simple results.

1. In a relatively large code, DT's 18F int routine saves a lot of code space compared to On-Interrupt of PBP as you said.
Ex: I saved 3100 bytes for the same code (code size : approx. 18,000 bytes).

2. In a small code, DT's 18F routine becomes code hungry.
Ex: PBP On-Interrupt produces 318 bytes for a sample code I just tried, but
DT's 18F int routine produces 762 bytes. More then twice larger!


As the code gets longer, DT's int routine gets handier I suppose.

[Ioannis]

Since On Interrupts of PBP puts a check every assembly instructions is obvious why is that code explosion!

Sure Darrels routines are fixed in length, so either your code is short or long, these routines are known in length. And in my humbble opinion, comparing Darrels interrupts with PBP interrupts is like comparing apples and oranges. Two different things.

Ioannis