At Thu, 3 Nov 2011 14:52:52 -0700, John Clements wrote: > If I have a place that's blocked on an mzrt-semaphore > (not a regular one, but a full-on-block-an-OS-thread semaphore), will > (place-kill ...) successfully kill the place?
[Sorry for the long delay!] No --- `place-kill' is cooperative at that level. To allow a place to be killed, you have to use scheme_block_until(), in general. Unfortunaetly, scheme_block_until() doesn't interact nicely with semaphores, except on Windows. An alternative for non-Windows platforms is to use an OS pipe instead of a semaphore, since file-descriptor waiting can cooperate with scheme_block_until(). _________________________________________________ For list-related administrative tasks: http://lists.racket-lang.org/listinfo/users
