aboutsummaryrefslogtreecommitdiff
path: root/src/core
Commit message (Expand)AuthorAge
* Windows fixes; especially idempotent init/fini.Garrett D'Amore2017-08-11
* Leaking poll fds.Garrett D'Amore2017-08-11
* Unify the msg API.Garrett D'Amore2017-08-10
* Add new PAIR_V1 protocol.Garrett D'Amore2017-08-10
* Thundering herd kills performance.Garrett D'Amore2017-08-10
* fixes #47 compat_reqttls fails sometimesGarrett D'Amore2017-08-09
* Fix problems found in Windows build.Garrett D'Amore2017-08-09
* fixes #44 open protocol by "name" (symbol) instead numberGarrett D'Amore2017-08-09
* fixes #37 Make transports pluggableGarrett D'Amore2017-08-08
* Added nn_compat code for option handling, fixed other bugs.Garrett D'Amore2017-08-08
* Don't clear the AIO provider data in finish.Garrett D'Amore2017-08-07
* We use NNG_ETIMEDOUT from msgq now.Garrett D'Amore2017-08-07
* Remove unused idhash_reclaim.Garrett D'Amore2017-08-07
* Remove some dead code; msg queue depths are always unsigned.Garrett D'Amore2017-08-07
* Subsystem initialize is idempotent; simplify cleanup.Garrett D'Amore2017-08-07
* Don't shut down taskq too soon.Garrett D'Amore2017-08-05
* Minor improvements to reap start/stop logic.Garrett D'Amore2017-08-05
* Use dedicate reap thread for endpoints too.Garrett D'Amore2017-08-05
* Use a dedicated reap thread instead of taskq.Garrett D'Amore2017-08-05
* Refactor AIO logic to close numerous races and reduce complexity.Garrett D'Amore2017-08-04
* More reliable taskq fini; avoids deadlock during shutdown.Garrett D'Amore2017-08-02
* Initial swag at UDP (POSIX only) low level handling.Garrett D'Amore2017-07-25
* Eliminate the separate AIO wake callback, making nni_aio_waitGarrett D'Amore2017-07-21
* Simpler taskq API.Garrett D'Amore2017-07-21
* Lets hold the lock a little while longer.Garrett D'Amore2017-07-20
* Yet more race condition fixes.Garrett D'Amore2017-07-20
* Always run the AIO completion logic.Garrett D'Amore2017-07-19
* fixes #27 Double free found in stress testingGarrett D'Amore2017-07-19
* Possible division by zero error (unset backoff start time).Garrett D'Amore2017-07-18
* Fixes most of the raaces in posix; but at least one remains outstanding.Garrett D'Amore2017-07-18
* Sometimes providers don't clear the prov data details. (Backoff).Garrett D'Amore2017-07-18
* Fix close-related leak of pipes.Garrett D'Amore2017-07-18
* Fix unreferenced variable warnings and errors reported by MVSC.Garrett D'Amore2017-07-17
* Scalability test fixes.Garrett D'Amore2017-07-17
* Fix hang on double-close of socket.Garrett D'Amore2017-07-17
* Ditch unused nni_sock_hold() call.Garrett D'Amore2017-07-17
* Clean up pipes on fini. EP close sync with pipes.Garrett D'Amore2017-07-16
* Close negotiation race.Garrett D'Amore2017-07-16
* Fix EAGAIN (timeout thread can run before we finish scheduling!)Garrett D'Amore2017-07-16
* Bind the pipe to the ep properly, and wake any closers needed.Garrett D'Amore2017-07-16
* Delete old #ifdef 0 pipe_create logic.Garrett D'Amore2017-07-16
* Fix locking errors in endpoints, and simplify some logic.Garrett D'Amore2017-07-16
* Reconnect automatically, but do backoff on failures. (Accept too!)Garrett D'Amore2017-07-16
* AIO timeouts work correctly now, using their own timer logic.Garrett D'Amore2017-07-16
* Fix incorrect attempt to proceed inproc.Garrett D'Amore2017-07-15
* Race conditions removed... TCP tests work well know.Garrett D'Amore2017-07-15
* Some initial progress on *connect* async.Garrett D'Amore2017-07-15
* Implemented asynchronous (fully) accept.Garrett D'Amore2017-07-14
* Close a race during pipe creation.Garrett D'Amore2017-07-13
* Use the same pipe teardown in all circumstances.Garrett D'Amore2017-07-13
generated by cgit v1.2.3-70-g09d2 (git 2.46.0) at 2026-01-18 14:17:12 +0000