| Commit message (Collapse) | Author | Age |
|---|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
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.
|
| |
|
|
|
|
|
| |
We enabled verbose compiler warnings, and found a lot of issues.
Some of these were even real bugs. As a bonus, we actually save
some initialization steps in the compat layer, and avoid passing
some variables we don't need.
|
| | |
|
| |
|
|
|
| |
This addresses a number of problems that were found on Windows,
including one bug that actually turned up in testing on POSIX.
|
| |
|
|
|
|
| |
There is now a public nng_duration type. We have also updated the
zerotier work to work with the signed int64_t's that the latst ZeroTier
dev branch is using.
|
| |
|
|
|
|
|
| |
This implements the basic UDP functionality for Windows (required
for ZeroTier for example). We have also introduced a UDP test suite
to validate that this actually works. While here a few Windows
compilation warnings / nits were fixed.
|
| |
|
|
|
|
|
|
|
|
| |
We only compile files that are appropriate for the platform. (We
still have guards in place, to allow for a future single .C file
to be built from all the sources.) We also remove the subsystem defines;
if a new platform needs to deviate from POSIX in ways beyond what we
intended here, then that platform should just copy those parts into
a new platform directory, rather than cross including portions from
POSIX.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
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.
|
| |
|
|
|
|
|
| |
This fixes one major problem, which was that if nni_fini() was called
once on Windows, it would not be further possible to call nni_init().
While here fixed a few compilation issues.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
| |
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 is only lightly tested, and I expect that there remain
some race conditions. Endpoint logic in particular needs
work.
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
| |
Modern Windows (Vista and later) have light weight Slim Read/Write locks
which only occupy 64 bits, and don't require any memory allocation to
create.
While here clean up a few more unreferenced variables found with the
Microsoft compilers.
|
| | |
|
| |
|
|
|
|
|
|
|
|
| |
This compiles correctly, but doesn't actually deliver events yet.
As part of this, I've made most of the initializables in nng
safe to tear-down if uninitialized (or set to zero e.g. via calloc).
This makes it loads easier to write the teardown on error code, since
I can deinit everything, without worrying about which things have been
initialized and which have not.
|
| | |
|
| | |
|
| |
|
|
|
| |
Test code needs to use the static libraries so that they can get access
to the entire set of symbols, including private ones that are not exported.
|
| |
|
|
|
|
|
|
|
|
|
|
| |
Since we use the tick counter to sleep, we should use the same clock
for validation. The problem is that the high performance tick counter
on the CPU may be slightly out of agreement with the windows clock.
Furthermore, the tick counter is probably lots faster to retrieve since
it is already updated, and needn't be recalculated each time.
(We should consider just switching to millisecond clock resolution
internally as well. It turns out that I don't think that timers that
are shorter than 1ms are very useful.)
|
| |
|
|
|
|
|
| |
There are lots of changes here, mostly stuff we did in support of
Windows TCP. However, there are some bugs that were fixed, and we
added some new error codes, and generalized the handling of some failures
during accept. Windows IPC (NamedPipes) is still missing.
|
| |
|
|
|
|
| |
Windows is getting there. Needs a couple of more more hours to enable
everything, especially IPC, and most of the work at this point is probably
some combination of debug and tweaking things like error handling.
|
| |
|
