This replaces the GUI element used for execution timeout handling
on Windows. Instead a timer queue technique is used, which is indeed
a thread pool. A timer queue timer is a lightweight object handled
but that thread pool and the timer thread spends most of the time
sleeping and waiting for an alert.
Please note also that this introduces neither binary nor source
breach. The custom timeout thread functions are deleted, however
they was not exported throug DLL, so couldn't be used by any
external code. As well they couldn't be used anywhere in the core
except in executor api, because those custom timeout thread
functions they used to operate on static variables which would
be overwritten (and that would blow).
So instead a relatively modern technique is used for the timeout
handling. It's still not perfect because the executor still has to
check EX(timed_out). This can be a topic for an improvement in
master. But brobably can be tricky as currently it seems to be not
possible to signal an individual thread. Also note another issue
that static variables aren't thread safe, but the current timer
implementation is.
Revert unintentional docblock change
It looks like commit dd8e59da8f
introduced an unintended docbloc change. I have reverted this
change in this commit.
* pull-request/932:
add initial install
switch to C travis project instead of PHP
use the generic TRAVIS environment var to check for travis (see http://docs.travis-ci.com/user/ci-environment/)
With that, also fixed bug #68297 (Application Popup provides too few
information) as a better error message is provided to the event log.
In the second case, the condition for display_startup_error was removed. The
condition was added as a fix for bug #30760 which was preventing the
MessageBox to appear. When display_startup_error=on, the error will be seen
on the console. But when running under a webserver it'll probably get lost,
so we need to log it other way into the event log (would be done automatically
when message box was used, but that would eventually cause issues of blocked
execution).
Generally speaking - any MessageBox or other graphical element is a potential
issue and that was repeatedly reported. Graphical elements shouldn't be used
in the core. Even being a rare one it can cause a bad situation where the
server is blocked. Yet some places have to cleaned up.
* PHP-5.4:
make sure that we don't truncate the stack trace and cause false test failures when the test is executed in a directory with long path
Conflicts:
run-tests.php
Add port range verification of listening port with -S option for the php cli.
This fixes inconsistent listening port due to unverified cast from long to short
with htons(3).
Remi Collet