I'm working on an experimental extension API, which is going
well, but I'm pondering how the GC should cooperate with
native calls.
Currently I have some code like:
# t.pasm
loadlib P0, "libpqt.so"
print "Loaded\n"
callnative P1, P0, "PQt_init"
end
/* libpqt.c */
int PQt_init(Interp * interp) {
At 02:17 PM 6/1/2002 -0400, Melvin Smith wrote:
># t.pasm
>loadlib P0, "libpqt.so"
>print "Loaded\n"
>callnative P1, P0, "PQt_init"
>end
The correct code is actually:
loadlib P0, "libpqt.so"
print "Loaded\n"
# Save an argument for native method
save "Testing..."
# callnative calls a native metho
>Now there are a dozen ways to handle this. Using flags, keeping args on the
>stack until return, yada yada yada...
One option is we don't pass args to native on the normal stack, rather
create a scratchpad Array PMC and pass it instead, so the routine
can access it randomly. However, I'm unsure
At 2:17 PM -0400 6/1/02, Melvin Smith wrote:
>Now there are a dozen ways to handle this. Using flags, keeping args on the
>stack until return, yada yada yada...
>
>I'd like to hear suggestions, or maybe reference to past discussions I might
>have missed on how we shall handle this. Actually I want
At 02:50 PM 6/1/2002 -0400, Dan Sugalski wrote:
>At 2:17 PM -0400 6/1/02, Melvin Smith wrote:
>>Now there are a dozen ways to handle this. Using flags, keeping args on the
>>stack until return, yada yada yada...
>>Since extensions are supposed to be isolated from the interpreter
>>internals, I'm
