Jens Rieks <[EMAIL PROTECTED]> wrote:
> Hi,
> this time a funny bug :-)
> 2. create a file hello.imc with the following content:
> .emit
> .pcc_sub @LOAD _onload:
> print "foo\n"
> end
^^^
It's called as a PCC subroutine, so you have to return from it via
"invoke P1". I wrote that "end" is equivalent, but that's wrong. While
"end" terminates the run loop, it does *not* restore the context,
specifically, it does not restore the byte code segment (with the
constant table inside)
The code runs than on an arbitrary code position which happens to be a
print statement with constant tables swapped (print_sc has an index into
the constants) - or something like this, I didn't look further.
> jens
leo
