aboutsummaryrefslogtreecommitdiff
path: root/src/core
Commit message (Collapse)AuthorAge
* Add endpoint tuning of maxrcv size. Fix cmsg API.Garrett D'Amore2017-01-24
| | | | | | | | | | | | | | The CMSG handling was completely borked. This is fixed now, and we stash the SP header size (ugh) in the CMSG contents to match what nanomsg does. We now pass the cmsg validation test. We also fixed handling of certain endpoint-related options, so that endpoints can get options from the socket at initialization time. This required a minor change to the transport API for endpoints. Finally, we fixed a critical fault in the REP handling of RAW sockets, which caused them to always return NNG_ESTATE in all cases. It should now honor the actual socket option.
* Add nni_ep_hold, nni_ep_hold_close, nni_ep_rele, nng_endpoint_close.Garrett D'Amore2017-01-24
|
* Implement reconnect timer including backoff.Garrett D'Amore2017-01-24
| | | | This allows us to enable the last test case for compat_reqrep.
* Added a bunch more compatibility stuff.Garrett D'Amore2017-01-23
| | | | | | | | I implemented the reqrep compatibility test, which uncovered a few semantic issues I had in the REQ/REP protocol, which I've fixed. There are still missing things. and at least one portion of the req/rep test suite cannot be enabled until I add tuning of the reconnect timeout, which is currently way too long (1 sec) for the test suite to work.
* Initial start of compat layer with bind, connect, etc. Untested.Garrett D'Amore2017-01-22
|
* Event notification via pollable FDs verified working.Garrett D'Amore2017-01-22
|
* Adds NNG_OPT_SENDFD and NNG_OPT_RECVFD socket options (untested).Garrett D'Amore2017-01-21
|
* Uncrustify last couple sets of changes.Garrett D'Amore2017-01-21
|
* Added protocol flags (which ones can send, and which can receive).Garrett D'Amore2017-01-21
|
* Initial swag at notification pipes (not used yet).Garrett D'Amore2017-01-21
|
* Fix compilation warnings found on Windows.Garrett D'Amore2017-01-21
|
* Fix leaks in bus, socket leaks, tighten up close-side refcnting.Garrett D'Amore2017-01-21
| | | | | | | | | | | | | | This does a few things. First it closes some preexisting leaks. Second it tightens the overall close logic so that we automatically discard idhash resources (while keeping numeric values for next id etc. around) when the last socket is closed. This then eliminates the need for applications to ever explicitly terminate resources. It turns out platform-specific resources established at nni_init() time might still be leaked, but it's also the case that we now no longer dynamically allocate anything at platform initialization time. (This presumes that the platform doesn't do so under the hood when creating critical sections or mutexes for example.)
* Resolve deadlock during close.Garrett D'Amore2017-01-21
|
* fixes #18 Sockets should be uint32_t's (handles) not pointers.Garrett D'Amore2017-01-20
|
* Thread fini shouldl be idempotent.Garrett D'Amore2017-01-19
|
* Fix synchronization problem in msgqueue with multiple consumers.Garrett D'Amore2017-01-19
|
* fixes #12 SURVEY hang in TravisGarrett D'Amore2017-01-18
|
* Reduce lock contention and simplify logic for endpoint creation.Garrett D'Amore2017-01-18
|
* Reaper exits too soon, leaking bits.Garrett D'Amore2017-01-18
|
* Fix leaking notify structs.Garrett D'Amore2017-01-18
|
* Windows compilation fixes.Garrett D'Amore2017-01-18
|
* Public pipe and endpoint APIs use IDs instead of pointers.Garrett D'Amore2017-01-17
|
* Pipe IDs are now tracked on global ID hashes.Garrett D'Amore2017-01-17
|
* Pipe IDs are now tracked by hash table for performance.Garrett D'Amore2017-01-17
| | | | | This gives a better idea of pipe ID uniqueness, and is a step towards conversion of the API to use IDs instead of pointers.
* Added dynamic ID generation & management for idhash tables.Garrett D'Amore2017-01-17
| | | | | | This will allow us to use idhash tables to manage id handles a bit more flexibly. For example, sockets, pipe IDs, etc. can all be generated, and we can use hash tables to ensure that values do not collide.
* External event API for send/recv implemented.Garrett D'Amore2017-01-16
| | | | | | This was the main blocker, I think, for the nanomsg legacy compat shim. Now that we have this, it should be relatively straight-forward to implement the legacy nanomsg API, including the SENDFD, RECVFD thing.
* Recv/Send event plumbing implemented (msgqueue and up).Garrett D'Amore2017-01-16
| | | | | | | | This change provides for a private callback in the message queues, which can be used to notify the socket, and which than arranges for the appropriate event thread to run. Upper layer hooks to access this still need to be written.
* Start of event framework.Garrett D'Amore2017-01-16
| | | | | | | | | | 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.
* Various complaints found in AppVeyor build.Garrett D'Amore2017-01-16
|
* Cleanup winsock somewhat.Garrett D'Amore2017-01-15
| | | | | | It turns out that I didn't quite understand overlapped I/O. We can and should always do the GetOverlappedResult(), regardless of how the routine returns.
* Windows IPC works now.Garrett D'Amore2017-01-15
|
* Windows TCP now working.Garrett D'Amore2017-01-14
| | | | | | | 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.
* Many fixes for Windows. It compiles, and some tests work.Garrett D'Amore2017-01-13
| | | | | | 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.
* Initial swag at Win32. Much to do still.Garrett D'Amore2017-01-12
|
* Pull out the posix clock stuff for plat_clock.Garrett D'Amore2017-01-12
|
* Add IPC (UNIX domain sockets) for POSIX, and test suite.Garrett D'Amore2017-01-12
|
* Initial (untested) bus implementation.Garrett D'Amore2017-01-10
|
* Initialize some mq vars.Garrett D'Amore2017-01-09
|
* Add survey test (and fix survey pattern).Garrett D'Amore2017-01-09
| | | | | | | As part of this, we've added a way to unblock callers in a message queue with an error, even without a signal channel. This was necessary to interrupt blockers upon survey timeout. They will get NNG_ETIMEDOUT, but afterwards callers get NNG_ESTATE.
* New ISAAC pRNG. This replaces other local hacks for random data.Garrett D'Amore2017-01-08
| | | | | | Platforms must seed the pRNGs by offering an nni_plat_seed_prng() routine. Implementations for POSIX using various options (including the /dev/urandom device) are supplied.
* Add surveyor protocol (no tests yet).Garrett D'Amore2017-01-08
| | | | | | | This adds the surveyor protocol, and updates the respondent somewhat. I've switched to using generic names for per-pipe and per-socket protocol data. Hopefully this will make 'cut-n-paste' from other protocol implementations easier.
* Move to generic socket & pipe workers, and up to 4 each.Garrett D'Amore2017-01-08
| | | | | This should eliminate all need for protocols to do their own thread management tasks.
* Add respondent protocol. (Half of survey pattern.)Garrett D'Amore2017-01-08
|
* More close race fixing.Garrett D'Amore2017-01-07
| | | | | Don't drop the lock in sock_close while holding the pipe reference. I'm pretty sure this is responsible for the use-after-free race.
* Clean up sender/recver threads.Garrett D'Amore2017-01-07
|
* Pipe bail shouldn't mess with lists. Doh!Garrett D'Amore2017-01-07
|
* Uncrustify fix for last.Garrett D'Amore2017-01-07
|
* Race condition fix attempt.Garrett D'Amore2017-01-07
|
* Simplify locking for protocols.Garrett D'Amore2017-01-07
| | | | | | | | | In an attempt to simplify the protocol implementation, and hopefully track down a close related race, we've made it so that most protocols need not worry about locks, and can access the socket lock if they do need a lock. They also let the socket manage their workers, for the most part. (The req protocol is special, since it needs a top level work distributor, *and* a resender.)
* Initial swag at pipeline (PUSH/PULL).Garrett D'Amore2017-01-06
| | | | | | | | | | | | | | | PUSH attempts to do a round-robin based distribution. However, I noticed that there is a bug in REQ, because REQ sockets will continue to pull down work until the first one no longer has room. This can in theory lead to scheduliung imbalances when the load is very light. (Under heavy load, the backpressure dominates.) Also, I note that mangos suffers the same problem. It does not make any attempt to deliver work equally, basically each pipe winds up pulling messages until its own buffers are full. This is bad. We can borrow the logic here for both REQ and mangos. None of this is tested yet.
generated by cgit v1.2.3-70-g09d2 (git 2.46.0) at 2026-01-20 07:57:40 +0000