> Your original Question..... "there is some startup code that calls
That's not a question - it's a statement. Please refer to my original message at the beginning of this thread. It contained only one question and that question was "How to determine the location of the first instruction that will get executed in the code 1 resource?".
> and my answer was correct. so it *was* your question....
Nope. Your answer was wrong (in the sense that the name you gave is compiler-dependent) and it was not the answer of my question anyway, because my question was different. :-)
> No, it would not be, the *name* could be different, the position
> could not possibly be different.
Sure it can be - although the fault for the misunderstanding is mine here; I didn't express correctly what I meant. I meant that the location of the startup code (the one that calls PilotMain) is compiler-dependent and could be *anywhere* in the code 1 resource. It's not at a fixed address. There must be, however, a standard way the OS uses to locate it. The other poster cleared that issue - the OS obviously transfers control to address 0 of the code 1 segment; the code there transfers control to wherever the beginning of the startup module happens to have been put by the compiler/linker - and that startup code eventually calls PilotMain.
> If it was you're effectively saying an M515 would start execution
> at a different place to an M505, which would cause a nightmare
> for the compiler.
No. I was saying that one and the same source compiled with different compilers could result in the startup code being in different places in the code 1 resource. But PalmOS has to have a standard way of locating that place. The different devices have nothing to do with that and the compiler isn't device-aware anyway. It's the OS that has to be able to locate the first executable instruction of an executable application on any device.
> I digress, I'm just in need of a smoke....
Smoking kills you slowly. But, I guess, you're not in a hurry. :-)