aboutsummaryrefslogtreecommitdiff
path: root/src/platform
Commit message (Collapse)AuthorAge
* Working towards TCP support.Garrett D'Amore2017-01-03
|
* Fixes to enable REQ/REP to operate.Garrett D'Amore2017-01-02
| | | | | | | This uncovered a few problems - inproc was not moving the headers to the body on transmit, and the message chunk allocator had a serious bug leading to memory corruption. I've also added a message dumper, which turns out to be incredibly useful during debugging.
* Stack related fixes for valgrind platform test.Garrett D'Amore2017-01-01
|
* Final purge of old threading & synch stuff.Garrett D'Amore2017-01-01
|
* Pipe simplifications for thread management.Garrett D'Amore2017-01-01
| | | | | | | This may also address a race in closing down pipes. Now pipes are always registered with the socket. They also always have both a sender and receiver thread. If the protocol doesn't need one or the other, the stock thread just exits early.
* New thread infrastructure -- not used anywhere yet, but tested.Garrett D'Amore2017-01-01
|
* Condvars on MacOS X (even 10.12) don't work with monotonic times.Garrett D'Amore2016-12-27
|
* Substantial fixes for listen & dialers.Garrett D'Amore2016-12-25
| | | | | | | | | | At this point listening and dialing operations appear to function properly. As part of this I had to break the close logic up since otherwise we had a loop trying to reap a thread from itself. So there is now a separate reaper thread for pipes per-socket. I also changed lists to be a bit more rigid, and allocations now zero memory initially. (We had bugs due to uninitialized memory, and rather than hunt them all down, lets just init them to sane zero values.)
* Mix the PID into our random number seed.Garrett D'Amore2016-12-25
|
* Fix for incorrect nni_usleep(), found with newly created platform tests.Garrett D'Amore2016-12-23
|
* Fix incorrect timeouts on platforms without clock_gettime().Garrett D'Amore2016-12-23
|
* Try to satisfy uncrustify version differences. Make test use verbose by ↵Garrett D'Amore2016-12-23
| | | | default.
* Uncrustify fixes.Garrett D'Amore2016-12-23
|
* Initial tests -- open & close work (no pipes or endpoints yet).Garrett D'Amore2016-12-23
|
* Endpoint dialer implemented.Garrett D'Amore2016-12-22
|
* Inline locks (fewer allocs), simpler absolute times for wakeups. nn_sock_recv.Garrett D'Amore2016-12-22
|
* Synchronization enhancements - inproc & msgqueue. Absolute waits...Garrett D'Amore2016-12-22
|
* Start of work to inline mutexes and condition variables.Garrett D'Amore2016-12-22
|
* Use C99 structure initializers FTW. Various other changes.Garrett D'Amore2016-12-22
|
* Older Darwin lacks clock_gettime(); we need poll.h to use poll().Garrett D'Amore2016-12-21
|
* Ditch our own snprintf/vsnprintf (C99). Symbol naming fixes for inproc.Garrett D'Amore2016-12-21
|
* Uncrustify configuration, and shorter copyright banners, plus reformatGarrett D'Amore2016-12-21
| | | | code with uncrustify. (Minor adjustments.) No more arguments!
* More robust platform definition support.Garrett D'Amore2016-12-14
| | | | | | | The idea is that someday it will be possible to just concatenate the entire set of source files into a single giant source file, for systems that want to work this way. As a result, the build system now compiles every file, although some of them will not have any definitions.
* More comments, and detection of fork-reentrancy. Much effort was spentGarrett D'Amore2016-12-13
| | | | trying to come to a fork-safe solution, but ultimately we gave up.
* Added threading primitives, more complete transport API.Garrett D'Amore2016-12-12
|
* New inproc transport.Garrett D'Amore2016-12-12
| | | | Lots of supporting changes.
* New msgqueue implementation, use CLOCK_MONOTONIC if available.Garrett D'Amore2016-12-11
| | | | Start of socket definitions.
* Condition variable implementation.Garrett D'Amore2016-12-10
|
* Initial commit. This is not going to be useful to you for anything.Garrett D'Amore2016-12-10