aboutsummaryrefslogtreecommitdiff
path: root/src
Commit message (Collapse)AuthorAge
...
* Use static initialization for lists and mutexes.Garrett D'Amore2021-12-05
| | | | | | This eliminates some run-time initialization, moving it to compile time. Additional follow up work will expand on this to simplify initialization and reduce the need for certain locks.
* Provide atomic pointer support.Garrett D'Amore2021-12-05
| | | | | This is initially used for TLS to make loading the engine pointer faster, eliminating a much more expensive lock operation.
* fixes #1541 bug in nni_chunk_insertGarrett D'Amore2021-12-02
|
* Fix incorrect elides of deprecated code.Garrett D'Amore2021-11-29
| | | | | | (Note that this includes some incorrect formatting due to an apparent bug in clang-format 13. Hopefully they'll fix it later, but for now I'm preserving the bad whitespace.)
* fixes #1536 nng_sendmsg and nng_recvmsg could be fasterGarrett D'Amore2021-11-28
| | | | fixes #1535 Desire nng_ctx_sendmsg and nng_ctx_recvmsg
* Don't use deprecated function in test.Garrett D'Amore2021-11-28
|
* fixes #1526 NNG_USE_CLOCKID Bug?Garrett D'Amore2021-11-27
| | | | | | | This makes CLOCK_MONOTONIC the default (as it should have been) for platforms that have it defined, except for Apple platforms which lack support for using anything other than the real time clock with condition variables. (And unfortunately silently ignore attempts to do otherwise.)
* fix building with PAIR and PUBSUB disabled (#1530)Edward Rudd2021-11-26
| | | | | When building with PAIR and PUBSUB protocols disabled, the perf tools fail to compile. This makes some minor tweaks to correct that
* fixes #1346 windows ipc winsec fails frequently in CI/CDGarrett D'Amore2021-11-02
|
* Use env shebangs everywhere (#1515)Leonard Pollak2021-10-26
| | | | Change all shebangs to use '#!/usr/bin/env bash'. This increases portability to platforms which do not cohere to the FHS.
* fixes #1518 Disconnect during negotiation breaks listenerGarrett D'Amore2021-10-16
|
* Remove unused eq_len member.Garrett D'Amore2021-10-11
|
* Minor cleanups.Garrett D'Amore2021-09-06
|
* Remove redundant check.Garrett D'Amore2021-09-06
|
* fixes #1347 windows IPC accept cancellation not wired upGarrett D'Amore2021-09-06
|
* Eliminate the pipe mutex and use atomic for pipe closed.Garrett D'Amore2021-09-06
| | | | | | | | | | This eliminates several mutex operations done each time a pipe is created or destroyed. For large scale systems this should reduce overall pressure on the memory subsystem, and scale better as many threads are coming and going. This also reduces the overall size of nni_pipe -- on Linux by 36 bytes typically.
* SP initialization cannot fail.Garrett D'Amore2021-09-06
|
* fixes #1498 Endpoint close/shutdown could be synchronous (#1499)Garrett D'Amore2021-09-04
|
* Minor cleanups.Garrett D'Amore2021-08-21
|
* Remove unused prototypes.Garrett D'Amore2021-08-21
|
* Move protocol.c into SP tree (it is SP specific.)Garrett D'Amore2021-08-21
| | | | Also, remove an extraneous initialization call.
* fixes #1488 aio expiration list performance work neededGarrett D'Amore2021-08-09
| | | | | | There were several problems with the array implementation, both from performance and from correctness. This corrects those errors (hopefully) and restores the expiration lists as linked lists.
* fixes #1478 mbedTLS 3.0 is not API compatible with 2.xGarrett D'Amore2021-08-09
|
* * FIX #1486 by waking up latest aio each time. (#1487)JaylinYu2021-08-09
|
* Remove separate protocol initialization step.Garrett D'Amore2021-08-07
| | | | | Nothing is using this, but it adds complexity and also requires additional lock activity each time a socket is opened.
* Fix premature transport registration. Mark it deprecated.Garrett D'Amore2021-08-01
| | | | | | | Originally the idea was to better support having the transports be separate loadable modules. This isn't needed for the builtin transports, so we make the explicit initialization of them deprecated, and document it as such.
* Simplify the SP transport initialization process.Garrett D'Amore2021-07-31
|
* Minor format and spelling in comments.Garrett D'Amore2021-07-31
|
* Remove extra wrapper for platform clock support.Garrett D'Amore2021-07-22
|
* Fix some unused variables.Garrett D'Amore2021-07-22
|
* Fix header guards.Garrett D'Amore2021-07-22
|
* fixes #1471 Use of strdup instead of nni_strdup in mbedtls/tls.cGarrett D'Amore2021-07-22
|
* fixes #1475 nni_aio_begin should not dispatch task on stopped aioGarrett D'Amore2021-07-22
|
* Fix valgrind error epoll_ctl points to uninitialised bytes (#1469)Jesper Wramberg2021-07-21
| | | | While not strictly an error, valgrind reports it as an error. This commit zero initializes the epoll_event before parsing it to the system call.
* Fix the wrong ratio when expire queue shrink. (#1470)wangha2021-07-16
|
* Minor clang-tidy fixes.Garrett D'Amore2021-07-11
|
* fixes #1409 reader/writer lock desiredGarrett D'Amore2021-07-11
| | | | | | This provides the initial implementation, and converts the transport lookup routines to use it. This is probably of limited performance benefit, but rwlock's may be useful in further future work.
* More symbol renamingGarrett D'Amore2021-07-11
|
* Remove unused nni_notifyfdGarrett D'Amore2021-07-10
|
* More work on moving SP stuff out of common. Remove unused defs.Garrett D'Amore2021-07-10
|
* Several minor cleanups. Fix socket id stat for listener.v1.5.0Garrett D'Amore2021-07-09
|
* Move transport.[ch] to SPGarrett D'Amore2021-07-09
|
* fixes #1465 NNG_OPT_IPC_PERMISSIONS stopped workingGarrett D'Amore2021-07-09
| | | | | This backs out the changes for #1326, because fchmod on sockets doesn't actually work on Linux, even though it returns success.
* Test fixes.Garrett D'Amore2021-07-07
|
* Add test cases for nng_msg_reserve, etc. Convert message tests to NUTS.Garrett D'Amore2021-07-07
|
* Fix leaking pipes after close.Garrett D'Amore2021-07-07
|
* Fix LeakAnalyzer complaints due to deferred reap.Garrett D'Amore2021-07-07
|
* Disable clock-dependent checks on macOS in CI/CD.Garrett D'Amore2021-07-06
| | | | | | GitHub's darwin server farm appears possibly overloaded, and the timing specific checks in that environment appear to be busted. Local instances of macOS don't seem to have problems though.
* Fix for UB (NULL pointer plus zero length). (#1459)Garrett D'Amore2021-07-06
|
* Add & document msg_capacity, msg_reserve (#1458)Evan Balster2021-07-06
| | | | | | | * Add & document msg_capacity, msg_reserve * reserve/capacity code style * Documentation references to reserve/capacity
generated by cgit v1.2.3-70-g09d2 (git 2.46.0) at 2026-01-19 19:23:21 +0000