AFAIK The difference in behavior is caused by the difference in the
instruction set. JIT-compiler attempts to compile the code but hits size
limitations when generating short jumps, which due to the nature of x86
instruction set cannot exceed 64K. I believe it to be a JIT-compiler
shortcoming, which unfortunately you have to live with.
OTOH consider the fact that if form designer code has exceeded 64 K, you
have a form which has too many controls. I cannot envision a form which
would run over 64 K in intialization and yet consist of one page - you must
be using tab control with a large number of tabs. You may want to redesign
the form to split it into several subforms. The upside is that form
initialization time will become tolerable. In fact it is likely that if you
add several more controls to your form, you will end up over the limit on
ARM cpu as well...
As for feeling like being back in 1988 - for some reason people tend to
forget, that we are still dealing with embedded devices. Small memory
footprint, limited resources etc.
"khorad" <(E-Mail Removed)> wrote in message
news:(E-Mail Removed)...
> i get a system.notsupportedexception on InitializeComponent()
>
> from previous threads it appears there is a 64KB method size limit or
> something when using the emulator
>
> it does work on the real device
>
> his solution was to split initialize into two methods, but i think that
> would mess up the form designer
>
> so is there anyway to get this to work on the emulator while still using
the
> form designer?
>
> i feel like i'm stuck back in 1988 worrying about 64k limitations
>
>
|
Similar Threads
