In the edge-trigger mode before this change, if a socket
receives 2 connections before one can be handled, then only a single
connection is handle and 1 connection will never be handled in some
cases. Reverting to level-trigger mode makes this impossible.
Allows for builting with cosmopolitan, both with meson
and Makefile. Use:
CC=comsocc meson setup -Dipv6=false -Ddynamic_modules=false
-Dshared=false -Dos_name=cosmopolitan
to configure for cosmopolitan build.
A stream may have a fiber attached for memory management purposes, but
not actually be waiting on anything. Be more seletive with poll, which
is not edge-triggered, to not poll for readiness on these streams.
Big issue with IOCP vs. poll variants is that the overlapped
structures have a longer lifetime than intermediate state needed
for epoll. One cannot free overlapped structures after closing a
handle/socket, like one can do with any intermediate state when using
readiness-based IO.
This results in fewer system calls and presumably more effcient code. It
also brings the epoll (and kqueue) code more in line with how the
windows IOCP code works, incidentally.
The sandboxing API is meant to make janet a bit more attractive
for certain application embedding use cases. The sandboxing API
puts limits on what system resources the interpreter can access.
The issue is that there was no synchronization on writes.
The stability of the test relied on the fact that the server
would read in an entire message in one call to ev/read, which
would _almost_ always happen since the messages are so small.
We were casting a pointer to the wrong type, which caused all sorts of
wonderful chaos, but only on windows and only when the garbage collector
ran after setting up a server in a specific configuration. We were
casting a closure pointer to an abstract type during the mark phase,
which resulted in memory corruption.
