Add a function timerfd_tracker::enter_critical_section_cancelable,
which is like enter_critical_section but honors a cancel event. Call
this when a timer expires while the timerfd thread is in its inner
loop. This avoids a deadlock if timerfd_tracker::dtor has entered its
critical section and is trying to cancel the thread. See
https://cygwin.com/ml/cygwin/2019-06/msg00096.html.
timerfd_tracker and timerfd_shared classes:
- Just because handles are shared, we don't have to store them in
shared memory. Move share handles into timerfd_tracker class.
- Drop shared instance counter since it's not required anymore.
timerfd_shared only stores the actual timer data.
- Drop timerfd_shared::create, just set clock id.
- Drop timerfd_shared::dtor, it's not required anymore.
- Drop timerfd_tracker::close, just call dtor where required.
- Rename timerfd_tracker::increment_instances to timerfd_tracker::dup.
It's the only reason it exists...
- timerfd_tracker::dtor now checks the non-shared pointers for NULL
before attempting to close them.
- timerfd_tracker::dtor handles decrementing the local instance count
by itself.
- Add a method timerfd_tracker::init_fixup_after_fork_exec to set
non-shared pointers to NULL. Together with the dtor patches it
fixes a problem with close_on_exec timerfd descriptors.
- Fix a bug in handling the thread synchronization event. It's
actually nice to create it before using it...
- Drop using sec_none{_nih} in InitializeObjectAttributes. It's
an unnecessary roundabout route just to get a NULL pointer.
- Slightly rework timechange window handling.
- Add more comments to explain what happens.
fhandler_timerfd:
- Drop cnew macro, it just hides what happens.
- fhandler_timerfd::fixup_after_exec now calls
timerfd_tracker::init_fixup_after_fork_exec first, so a subsequent
call to timerfd_tracker::dtor only works on valid handles.
- fhandler_timerfd::close directly calls timerfd_tracker::dtor now.
- Drop dtor call in fhandler_timerfd destructor.
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
The value returned by reading from a timerfd is not an overrun
count in the same sense as for posix timers, it's an expiry counter.
Reflect that in the name.
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
Using posix timers "timer_tracker" as base class for timerfd was flawed.
Posix timers are not inherited by child processes and don't survive
execve. The method used by posix timers didn't allow to share timers
between processes. The timers were still per-process timers and worked
entirely separate from each other. Reading from these timers via
different descriptors was only synchronized within the same process.
This does not reflect the timerfd semantics in Linux: The per-file
timers can be dup'ed and survive fork and execve. They are still just
descriptors pointing to the same timer object originally created by
timerfd_create. Synchronization is performed between all descriptor
instances of the same timer, system-wide.
Thus, reimplement timerfd using a timer instance in shared memory,
a kernel timer, and a handful of sync objects.
Every process maintains a per-process timerfd struct on the cygheap
maintaining a per-process thread. Every process sharing the same
timerfd will run this thread checking the state of the timer, similar
to the posix timer thread, just working on the shared objects and
synchronizing its job with each other thread.
Drop the timerfd implementation in the posix timer code and move the
public API to fhandler_timerfd.c. Keep the ttstart timer_tracker
anchor out of "NO_COPY" since the fixup_after_fork code should run to
avoid memory leakage.
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
Ken Brown
Corinna Vinschen