| Commit message (Collapse) | Author | Age |
|---|
| |
|
|
|
| |
* Expose cmake variable to set number of DNS resolver threads: NNG_RESOLV_CONCURRENCY
* Expose cmake variable to set number of taskq threads: NNG_NUM_TASKQ_THREADS
|
| | |
|
| |
|
|
|
|
| |
This should work on both Windows and the most common POSIX
variants. We will create at least two threads for running
completions, but there are numerous other threads in the code.
|
| |
|
|
|
|
|
|
|
|
| |
Essentially, if we're destroying an aio, and we are doing so from the
thread that is running the callback, then we should defer the destruction
of the task until it returns.
Note that calling nni_aio_wait() or anything else that calls it from
the callback is still verboten and will result in a single party
deadlock.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This changes nni_task_fini to always run synchronously, waiting
for the task to finish before cleaning up. Much simpler code.
Additionally, we've refactored the resolver code to avoid the
use of taskqs, which added complexity and inefficiency. The
approach of just allocating its own threads and a work queue
to process them turns out to be vastly simpler, and actually
reduces extra allocations and context switches.
wip
POSIX resolv threads.
(Taskqs are just overhead and complexity here.)
Windows resolver changes.
Task cleanup.
fix up windows mutex.
|
| |
|
|
|
|
|
|
| |
This changes the array of flags, which was confusing, brittle, and
racy, into a much simpler reference (busy) count on the task structures.
This allows us to support certain kinds of "reentrant" dispatching,
where either a synchronous or asynchronous task can reschedule / dispatch
itself. The new code also helps reduce certain lock pressure, as a bonus.
|
| |
|
|
|
|
| |
While here I also improved the taskq.h comments (and removed a
stale prototype for nni_task_cancel), and addressed leaks in
the reqstress and multistress test programs.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
| |
fixes #429 async websocket reap leads to crash
This tightens up the code for shutdown, ensuring that transport
callbacks are completely stopped before advancing to the next step
of teardown of transport pipes or endpoints.
It also fixes a problem where task_wait would sometimes get "stuck"
as tasks transitioned between asynch and synchronous completions.
Finally, it saves a few cycles by only calling a cancellation callback
once during cancellation of an aio.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
fixes #326 consider nni_taskq_exec_synch()
fixes #410 kqueue implementation could be smarter
fixes #411 epoll_implementation could be smarter
fixes #426 synchronous completion can lead to panic
fixes #421 pipe close race condition/duplicate destroy
This is a major refactoring of two significant parts of the code base,
which are closely interrelated.
First the aio and taskq framework have undergone a number of simplifications,
and improvements. We have ditched a few parts of the internal API (for
example tasks no longer support cancellation) that weren't terribly useful
but added a lot of complexity, and we've made aio_schedule something that
now checks for cancellation or other "premature" completions. The
aio framework now uses the tasks more tightly, so that aio wait can
devolve into just nni_task_wait(). We did have to add a "task_prep()"
step to prevent race conditions.
Second, the entire POSIX poller framework has been simplified, and made
more robust, and more scalable. There were some fairly inherent race
conditions around the shutdown/close code, where we *thought* we were
synchronizing against the other thread, but weren't doing so adequately.
With a cleaner design, we've been able to tighten up the implementation
to remove these race conditions, while substantially reducing the chance
for lock contention, thereby improving scalability. The illumos poller
also got a performance boost by polling for multiple events.
In highly "busy" systems, we expect to see vast reductions in lock
contention, and therefore greater scalability, in addition to overall
improved reliability.
One area where we currently can do better is that there is still only
a single poller thread run. Scaling this out is a task that has to be done
differently for each poller, and carefuly to ensure that close conditions
are safe on all pollers, and that no chance for deadlock/livelock waiting
for pfd finalizers can occur.
|
| | |
|
| |
|
|
|
|
|
|
|
|
| |
While here we added a test for the aio stuff, and cleaned up some dead
code for the old fd notifications. There were a few improvements to
shorten & clean code elsewhere, such as short-circuiting task wait
when the task has no callback.
The legacy sendmsg() and recvmsg() APIs are still in the socket core
until we convert the device code to use the aios.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
If the underlying platform fails (FreeBSD is the only one I'm aware
of that does this!), we use a global lock or condition variable instead.
This means that our lock initializers never ever fail.
Probably we could eliminate most of this for Linux and Darwin, since
on those platforms, mutex and condvar initialization reasonably never
fails. Initial benchmarks show little difference either way -- so we
can revisit (optimize) later.
This removes a lot of otherwise untested code in error cases and so forth,
improving coverage and resilience in the face of allocation failures.
Platforms other than POSIX should follow a similar pattern if they need
this. (VxWorks, I'm thinking of you.) Most sane platforms won't have
an issue here, since normally these initializations do not need to allocate
memory. (Reportedly, even FreeBSD has plans to "fix" this in libthr2.)
While here, some bugs were fixed in initialization & teardown.
The fallback code is properly tested with dedicated test cases.
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
| |
A little benchmarking showed that we were encountering far too many
wakeups, leading to severe performance degradation; we had a bunch
of threads all sleeping on the same condition variable (taskqs)
and this woke them all up, resulting in heavy mutex contention.
Since we only need one of the threads to wake, and we don't care which
one, let's just wake only one. This reduced RTT latency from about
240 us down to about 30 s. (1/8 of the former cost.)
There's still a bunch of tuning to do; performance remains worse than
we would like.
|
| | |
|
| |
|
|
|
|
|
|
|
| |
This passes valgrind 100% clean for both helgrind and deep leak
checks. This represents a complete rethink of how the AIOs work,
and much simpler synchronization; the provider API is a bit simpler
to boot, as a number of failure modes have been simply eliminated.
While here a few other minor bugs were squashed.
|
| | |
|
| |
|
|
| |
block for any AIO completion.
|
| |
|
|
|
|
|
| |
The queue is bound at initialization time of the task, and we call
entries just tasks, so we don't have to pass around a taskq pointer
across all the calls. Further, nni_task_dispatch is now guaranteed
to succeed.
|
| |
|
|
|
|
|
|
|
| |
We need to remember that protocol stops can run synchronously, and
therefore we need to wait for the aio to complete. Further, we need
to break apart shutting down aio activity from deallocation, as we need
to shut down *all* async activity before deallocating *anything*.
Noticed that we had a pipe race in the surveyor pattern too.
|
| |
|
|
|
|
|
|
| |
We have seen some yet another weird situation where we had an orphaned
pipe, which was caused by not completing the callback. If we are going
to run nni_aio_fini, we should still run the callback (albeit with a
return value of NNG_ECANCELED or somesuch) to be sure that we can't
orphan stuff.
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
|
