This test was provided by Lewis Muir, it fails with seq-2.1.5.

The connection handler now releases all semaphores waiting for an
event on this PV. Furthermore, the pvGet/pvPut[Complete] built-in
functions check the connected status whenever they successfully waited
for such a semaphore.

This patch also fixes a few problems with error handling in these
procedures; for instance, the SYNC versions released semaphores even
in case of timeout, which is definitely wrong.

This is done by setting all event flags that have anonymous PVs sync'ed to
them, once right after all named (real) PVs have connected.

Each event flag now knows the channels which are synced to it.
This is a singly linked list of channel pointers. Since each channel
can be synced to at most one event flag, we can embed the list nodes
into the channel structure, so no runtime allocation is needed.
The pvSync builtin function must now update these lists, while
ss_read_buffer_selective traverses them.

The fix is to declare efTest and efTestAndClear to be synchronization
points in safe mode. Both now check whether any of the variables sync'ed
with the event flag have been changed (have their dirty flag set) and
update the state set local copy of these variables. This happens
regardless of whether the call is part of a when() condition or inside
an action block.

This was a subtle concurrency bug discovered and analyzed by Lewis Muir:
if pvPutComplete or pvGetComplete are used inside a when-clause, waking
up the state set before signalling completion could result in a deadlock.

Funny, I had an assertion in the code that should have caught the bug
but unfortunately the assertion itself had an off-by-one error and so
didn't fire. So I had to debug this the hard way. The compiler will now
give a warning message saying that it discards the excess PV names.

When the lexer encountered a new line marker, it freed the old source
file string, leading to bad memory references when the string is used
later in the parser (for errors and warnings), which led to wrong
reported source file names (and potential compiler crashes).

Also removed builtin constants from lexer. Constants are now looked up
in analysis when lookup for a declared variable fails.

This fixes the long-standing problem that pvStatOK are not known
to snc and thus must be declared as foreign (or you get a warning).

This is now configurable in test/validate/Makefile.

A sync'ed event flags gets set on any kind of event:
monitor, as well as put and get completion.