> From: John Marshall [mailto:[EMAIL PROTECTED]]
> we won't have to have this [linked to] thread ever again:
Famous last words?
> You just have to find some other way to disambiguate your constructors.
Heck, I fixed the code _before_ I posted the complaint. :-) And writing
the email took longer than fixing the code.
> You'll be happy to hear that I broke the IntlGlueGetRoutineAddress
> part of your slides too! That function wasn't previously documented
Hmm. I'm sure I wouldn't have mentioned that function in the presentation
if it were officially undocumented. Maybe this was back before the glue was
documented in the main PDF files, so the headers were the only real docs.
Or I screwed up.
Actually though, all this trap fiddling is definitely not going to work when
running in 68K emulation on the ARM, right? In David Fedor's ARM talk at
PalmSource, he said trap patching (including Hackmaster) won't be supported,
and I'm assuming that just fetching trap addresses won't either. That makes
sense since there's no 68K function there to take the address of. So this
makes the whole SpeedTrap technique less safe and less useful. When running
68K code on the ARM it just won't work, and when running native ARM code
it'll be so fast it won't be needed. (If this technique is even available
at all in native ARM code.)
-slj-
--
For information on using the Palm Developer Forums, or to unsubscribe, please see
http://www.palmos.com/dev/tech/support/forums/
