Now that VCLs can be referenced, there are two different situations
stalling the cooldown. The "cooling" one is kept for VCLs that have
received a COLD event, and the "busy" one is introduced to wait for
ongoing transactions.

Basically, if there are transactions, the VCL is still active even
though it's not *the* active VCL and should therefore be considered
warm. This guarantees that WARM and COLD events are balanced, and
also that a "busy" VCL doesn't need to warm up if it is set to warm
again.

The new VCL temperature engine looks like this:

  .----------W>---------------.
  |              .--W>---.    v
init ---> cold --|       |-- warm --.
          ^  ^   '---<C--'    |     |
          |  |                |     |
          |  '--- cooling <C--'     |
          |         ^               |
          |         C               |
          |         |    .---<---.  |
          '--<C-- busy --|       |--'
                         '--->---'

The transitions marked with a 'W' or a 'C' are the one dispatching
WARM and COLD events respectively.

Since the "busy" state is considered warm, VMODs should operate as
such and backend creation remains possible.

Broken implementations do it contrary to what the RFC mandates.

Fixes #1826.

This works around a Linux kernel bug where the epoll_wait will return
EINTR when the process is subjected to a ptrace or the OS wakes from
suspend.

Fixes: #1763

Update built-in usage text to reflect that -T is optional.

Forgotten in commit 8139164b.

* Adhere to current style guide.
* Sort argument list.
* Add missing reference to related man pages.

There is a Squid server listening in today's office...

Also for 4.1

And stop using it in the test suite.

Also for 4.1

A VCL may show up as auto/cooling in the CLI, breaking the columns align
in the output.

Also for 4.1

Amends 83aa5402 to handle FreeBSD and OpenBSD (at least).

Correct typo that checked for pthread_set_name_np() instead of
pthread_setname_np().

Thread names should now be visible in /proc/CHILDPID/tasks/*/comm on
Linux systems.

have to walk the full list all the time.

This is a minimal fix for backporting to 4.1.x

Fixes:	#1798
Fixes:	#1788

Fixes:	#1821

context.

Cherry-picked and modified from patch (pull#71) by:	lochiiconnectivity

(which gets the busyobj) as a miss.

Testcase by:	daghf
Fixes:	1818

`orderto` -> `order to` for docs at https://www.varnish-cache.org/docs/trunk/reference/varnishd.html