A time limit can be set on PHP script execution via `set_time_limit` (or .ini file).
When the time limit is reached, the OS will notify PHP and `timed_out` and `vm_interrupt`
flags are set. While these flags are regularly checked when executing PHP code, once the
end of the script is reached, they are not checked while invoking shutdown functions
(registered via `register_shutdown_function`).
Of course, if the shutdown functions are implemented *in* PHP, then the interrupt flag
will be checked while the VM is running PHP bytecode and the timeout will take effect.
But if the shutdown functions are built-in (implemented in C), it will not.
Since the shutdown functions are invoked through `zend_call_function`, add a check of the
`vm_interrupt` flag there. Then, the script time limit will be respected when *entering*
each shutdown function. The fact still remains that if a shutdown function is built-in and
runs for a long time, script execution will not time out until it finishes and the
interpreter tries to invoke the next one.
Still, the behavior of scripts with execution time limits will be more consistent after
this patch. To make the execution time-out feature work even more precisely, it would
be necessary to scrutinize all the built-in functions and add checks of the `vm_interrupt`
flag in any which can run for a long time. That might not be worth the effort, though.
It should be mentioned that this patch does not solely affect shutdown functions, neither
does it solely allow for interruption of running code due to script execution timeout.
Anything else which causes `vm_interrupt` to be set, such as the PHP interpreter receiving
a signal, will take effect when exiting from an internal function. And not just internal
functions which are called because they were registered to run at shutdown; there are
other cases where a series of internal functions might run in the midst of a script. In
all such cases, it will be possible to interrupt the interpreter now.
Closes GH-5543.
After taking a more detailed look at our commonly failing timeout
tests... turns out that most of them are useless as written and
don't test what they're supposed to.
This PR has a couple of changes:
* Tests for timeout in while/for/foreach should just have the loop
as an infinite loop. Calling into something like busy_wait means
that we just end up always testing whatever busy_wait does.
* Tests for timeouts in calls need to be based on something like
sleep, otherwise we'd have to introduce a loop, and we'd end up
testing timeout of the looping structure instead. Using sleep only
works on Windows, because that's the only system where sleep counts
towards the timeout. As such, many of those tests are now Windows only.
* Removed some tests where I don't see a good way to test what they're
supposed to test. E.g. how can we test a timeout in eval() specifically?
The shutdown function tests are marked as XFAIL, as we are currently
missing a timeout check in call_user_function. I believe that's a
legitimate issue.
Closes GH-4969.
