OS_Claim (0x1F)

Is there any specific reason why you really don’t want to use the proper API and would prefer to do things behind RISC OS’s back?

Possibly because Adrian is someone used to other OS’s that don’t expose the bottom end of the system anywhere near as much/well.
Basically, when you use an OS produced for a generic platform you find you can improve on certain aspects when you delve under the bonnet. Here in RISC OS land we are already under the bonnet the moment we start programming, many people don’t realise this until they develop a familiarity.

An additional benefit – unless Adrian’s code is extremely specific to the Pi, it will work on other supported RISC OS platforms when using the API. Does the Beagle or Panda offer at least two available high resolution timers? One for RISC OS and one for us? If the answer is Yes then this will “just work”.

We talk to the HAL and the HAL talks to the hardware. That is how it should be – for HAL means exactly that → Hardware Abstraction Layer. As much as possible software should use the standard API and not try to do clever tricks, for history is littered with examples of software (primarily copy protection schemes) ultimately failing due to being too clever for its own good or doomed to work with only one specific piece of hardware.

An additional benefit – unless Adrian’s code is extremely specific to the Pi, it will work on other supported RISC OS platforms when using the API.

Well it doesn’t crash on an emulated RPC when the emulation is using RO5…

As much as possible software should use the standard API and not try to do clever tricks, for history is littered with examples of software (primarily copy protection schemes)

I have to say that my normal reaction to such things was always to see how I could “unprotect” the item in question. No intention of piracy, I just didn’t like the lock down.

I fixed a bug or two in a database my mother and colleague used at work, but had to unprotect it to get at the software to fix it. Easy fix, bit of a pain using a disc editor on a clone disc to access the software.

Is there any way of determining whether a timer is already in use?
If I were writing this I’d have something like:

For timer = 0 to (qty of timers – 1)
If timer is available Then hooray!
Error “No timer available”

I don’t believe there’s a way to check if a timer is in use, there was some discussion about adding this functionality, but I’m not sure what the final outcome was.

A layer above the HAL that virtualises them, to allow as many timers as a user requests would be useful.