| Commit message (Collapse) | Author | Age | ||
|---|---|---|---|---|
| ... | ||||
| * | Minor cleanups. | Garrett D'Amore | 2017-01-10 | |
| | | ||||
| * | Cleanup debug. | Garrett D'Amore | 2017-01-09 | |
| | | ||||
| * | Probably this fixes the hang -- something was clearing the error condition. | Garrett D'Amore | 2017-01-09 | |
| | | ||||
| * | (Briefly) debug output for hang in survey test. | Garrett D'Amore | 2017-01-09 | |
| | | ||||
| * | Initialize some mq vars. | Garrett D'Amore | 2017-01-09 | |
| | | ||||
| * | Add survey test (and fix survey pattern). | Garrett D'Amore | 2017-01-09 | |
| | | | | | | | | 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. | |||
| * | POSIX /dev/urandom compilation bug. | Garrett D'Amore | 2017-01-08 | |
| | | ||||
| * | New ISAAC pRNG. This replaces other local hacks for random data. | Garrett D'Amore | 2017-01-08 | |
| | | | | | | | 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. | |||
| * | Prevent uncrustify "error" checks from trigger travis errors. | Garrett D'Amore | 2017-01-08 | |
| | | | | | | | | There are multiple different versions of uncrustify, and they do not always generate the same output. (Arguably this is due to defects in uncrustify.) So for now we punt and don't throw an error, but we do still generate the output. Pay attention to this going forward. | |||
| * | Add RAW option support (no-op) for PAIR. | Garrett D'Amore | 2017-01-08 | |
| | | ||||
| * | Add surveyor protocol (no tests yet). | Garrett D'Amore | 2017-01-08 | |
| | | | | | | | | 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'Amore | 2017-01-08 | |
| | | | | | | This should eliminate all need for protocols to do their own thread management tasks. | |||
| * | Add respondent protocol. (Half of survey pattern.) | Garrett D'Amore | 2017-01-08 | |
| | | ||||
| * | Fix small message leak in pipeline test. | Garrett D'Amore | 2017-01-07 | |
| | | ||||
| * | More close race fixing. | Garrett D'Amore | 2017-01-07 | |
| | | | | | | 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'Amore | 2017-01-07 | |
| | | ||||
| * | Pipe bail shouldn't mess with lists. Doh! | Garrett D'Amore | 2017-01-07 | |
| | | ||||
| * | Uncrustify fix for last. | Garrett D'Amore | 2017-01-07 | |
| | | ||||
| * | Race condition fix attempt. | Garrett D'Amore | 2017-01-07 | |
| | | ||||
| * | Simplify locking for protocols. | Garrett D'Amore | 2017-01-07 | |
| | | | | | | | | | | 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.) | |||
| * | Add another dial to ensure that accept loop has made progress. | Garrett D'Amore | 2017-01-07 | |
| | | ||||
| * | Pipeline fixes. | Garrett D'Amore | 2017-01-07 | |
| | | ||||
| * | Pipeline test. | Garrett D'Amore | 2017-01-07 | |
| | | ||||
| * | Fix startup race in the tests. | Garrett D'Amore | 2017-01-06 | |
| | | | | | | Pub pipes might not be connected yet. Do the dial from the pub side synchronously, and we can be sure no data will be lost. | |||
| * | Bump the timeouts even more for Travis. | Garrett D'Amore | 2017-01-06 | |
| | | ||||
| * | Bump timeout for crummy Travis -- 50 ms isn't enough! | Garrett D'Amore | 2017-01-06 | |
| | | ||||
| * | Shut up uncrustify. | Garrett D'Amore | 2017-01-06 | |
| | | ||||
| * | Fix for valgrind leak. | Garrett D'Amore | 2017-01-06 | |
| | | ||||
| * | Initial swag at pipeline (PUSH/PULL). | Garrett D'Amore | 2017-01-06 | |
| | | | | | | | | | | | | | | | | 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. | |||
| * | Bunch of copyright fixes. | Garrett D'Amore | 2017-01-06 | |
| | | ||||
| * | Fixes for PUB/SUB. | Garrett D'Amore | 2017-01-06 | |
| | | | | | | This fixes several issues, and brings PUB/SUB to operational correctness. Included is test code to verify that. | |||
| * | Message API was awkward. | Garrett D'Amore | 2017-01-06 | |
| | | | | | | | | The use of a single function to get both size and length actually turned out to be awkward to use; better to have separate functions to get each. While here, disable some of the initialization/fork checks, because it turns out they aren't needed. | |||
| * | Start of PUB/SUB testing. Fixed a subscribe bug. | Garrett D'Amore | 2017-01-06 | |
| | | ||||
| * | Add inproc performance tests. | Garrett D'Amore | 2017-01-05 | |
| | | ||||
| * | Added test for cancellation, fixed retry bug. | Garrett D'Amore | 2017-01-05 | |
| | | | | | | | On retry we were pushing back to the queue. The problem with this is that we could wind up pushing back many copies of the message if no reader was present. The new code ensures at most one retry is outstanding. | |||
| * | Oops forgot to git commit pub.c | Garrett D'Amore | 2017-01-05 | |
| | | ||||
| * | Change a bunch of copyrights to 2017 for work done since the 1st. | Garrett D'Amore | 2017-01-05 | |
| | | ||||
| * | PUB protocol (untested). | Garrett D'Amore | 2017-01-05 | |
| | | ||||
| * | Add nng_shutdown() for sockets to help avoid close race. | Garrett D'Amore | 2017-01-05 | |
| | | | | | Also we added a two phase shutdown for threads. | |||
| * | Fix early termination, and buffer for throughput. | Garrett D'Amore | 2017-01-05 | |
| | | ||||
| * | Naive attempt to avoid allocator was misguided. This helps performance. | Garrett D'Amore | 2017-01-05 | |
| | | ||||
| * | Implement throughput performance tests. | Garrett D'Amore | 2017-01-05 | |
| | | | | | | | | The throughput performance tests "try" to avoid hitting the allocator, but I think this actually causes other cache related performance, and the receive thread still has to perform a message allocation, leading to really rotten performance. Its probably time to think about a message pool. | |||
| * | Add initial performanced tests. | Garrett D'Amore | 2017-01-05 | |
| | | ||||
| * | Oops -- forgot to test compile. | Garrett D'Amore | 2017-01-05 | |
| | | ||||
| * | Release subscriptions on socket close. | Garrett D'Amore | 2017-01-05 | |
| | | ||||
| * | SUB protocol implemented (uses sorted linked list for topics). | Garrett D'Amore | 2017-01-05 | |
| | | ||||
| * | Fixes for TCP transport. Working now. | Garrett D'Amore | 2017-01-04 | |
| | | ||||
| * | Recv returns 0 on EOF. | Garrett D'Amore | 2017-01-04 | |
| | | ||||
| * | Fix close related races (POSIX close is a PITA). | Garrett D'Amore | 2017-01-04 | |
| | | ||||
| * | Compilation fixes for Linux. | Garrett D'Amore | 2017-01-04 | |
| | | ||||
 |
index : nng | |
| A mirror of https://github.com/nanomsg/nng |
| aboutsummaryrefslogtreecommitdiff |