Implementation of true asynchronous signaling between processes

Communication between Erlang processes has conceptually always been
performed through asynchronous signaling. The runtime system
implementation has however previously preformed most operation
synchronously. In a system with only one true thread of execution, this
is not problematic (often the opposite). In a system with multiple threads
of execution (as current runtime system implementation with SMP support)
it becomes problematic. This since it often involves locking of structures
when updating them which in turn cause resource contention. Utilizing
true asynchronous communication often avoids these resource contention
issues.

The case that triggered this change was contention on the link lock due
to frequent updates of the monitor trees during communication with a
frequently used server. The signal order delivery guarantees of the
language makes it hard to change the implementation of only some signals
to use true asynchronous signaling. Therefore the implementations
of (almost) all signals have been changed.

Currently the following signals have been implemented as true
asynchronous signals:
- Message signals
- Exit signals
- Monitor signals
- Demonitor signals
- Monitor triggered signals (DOWN, CHANGE, etc)
- Link signals
- Unlink signals
- Group leader signals

All of the above already defined as asynchronous signals in the
language. The implementation of messages signals was quite
asynchronous to begin with, but had quite strict delivery constraints
due to the ordering guarantees of signals between a pair of processes.

The previously used message queue partitioned into two halves has been
replaced by a more general signal queue partitioned into three parts
that service all kinds of signals. More details regarding the signal
queue can be found in comments in the erl_proc_sig_queue.h file.

The monitor and link implementations have also been completely replaced
in order to fit the new asynchronous signaling implementation as good
as possible. More details regarding the new monitor and link
implementations can be found in the erl_monitor_link.h file.

1 files changed, 35 insertions, 0 deletions

diff --git a/lib/hipe/doc/src/hipe_app.xml b/lib/hipe/doc/src/hipe_app.xml
index aaeb06193d..fc42ecd97d 100644
--- a/lib/hipe/doc/src/hipe_app.xml
+++ b/lib/hipe/doc/src/hipe_app.xml
@@ -99,6 +99,41 @@
       each mode.</p>
       </item>
 
+      <tag>Optimization for <c>receive</c> with unique references</tag>
+      <item>
+	<p>
+	  The BEAM compiler can do an optimization when a receive
+	  statement is only waiting for messages containing a reference
+	  created before the receive. All messages that existed in the
+	  queue when the reference was created will be bypassed, as they
+	  cannot possibly contain the reference. HiPE currently has an
+	  optimization similar this, but it is not guaranteed to
+	  bypass all messages. In the worst case scenario it, cannot
+	  bypass any messages at all.
+	</p>
+	<p>
+	  An example of this is when <c>gen_server:call()</c> waits for
+	  the reply message.
+	</p>
+      </item>
+
+    </taglist>
+  </section>
+  <section>
+    <title>Stability Issues</title>
+    <taglist>
+      <tag>Not yielding in <c>receive</c> statements</tag>
+      <item>
+	<p>HiPE will not yield in <c>receive</c> statements where
+	appropriate. If a process have lots of signals in its signal
+	queue and execute a HiPE compiled <c>receive</c> statement,
+	the scheduler thread performing the execution may be stuck
+	in the <c>receive</c> statement for a very long time. This
+	can in turn cause various severe issues such as for example
+	prevent the runtime system from being able to release
+	memory.
+	</p>
+      </item>
     </taglist>
   </section>
   <section>