| Commit message (Collapse) | Author | Age | ||
|---|---|---|---|---|
| ... | ||||
| * | fix copyrights | Garrett D'Amore | 2023-04-23 | |
| | | ||||
| * | fixes #1658 Possible use-after-free in dialer | Garrett D'Amore | 2023-04-23 | |
| | | ||||
| * | fixes #1657 Use after free in listener (data race) | Garrett D'Amore | 2023-04-23 | |
| | | ||||
| * | fixes #1619 Expose NNG_MAX_EXPIRE_THREADS via CMake | Garrett D'Amore | 2023-04-19 | |
| | | ||||
| * | Birth pipe object with non-zero reference. | Garrett D'Amore | 2023-04-19 | |
| | | ||||
| * | fixes #1610 nni_sock_open() in socket.c: use after free | Garrett D'Amore | 2023-04-19 | |
| | | ||||
| * | Socket close may leak messages. | Garrett D'Amore | 2023-02-07 | |
| | | | | | | | We try to move the msgq close up earler. While here we can stop dropping and reacquiring the lock -- this is likely left over and may lead to races. | |||
| * | Revert "fixes 1543 (#1616)" | Garrett D'Amore | 2023-02-05 | |
| | | | | | This reverts commit 8461c7207b440f5ba8c51b2236fcfa178f415a6f. | |||
| * | Don't bump error counters for NNG_ECLOSED. | Garrett D'Amore | 2023-02-05 | |
| | | ||||
| * | fixes 1543 (#1616) | josh salit | 2023-02-05 | |
| | | | | fixes #1543 by aborting tasks that may have been prepped, but not yet started. | |||
| * | Update idhash.c (#1638) | shikokuchuo | 2023-02-05 | |
| | | | | Fix for UBSAN error. As `id_reg_map` is initialised as NULL and passing NULL to `memcpy()` is undefined. Should make no difference to compiled code. Purely to appease the automated checks I have to deal with on my side. Thanks! | |||
| * | Silence NNI_ASSERT warnings when building with NDEBUG (#1621) (#1622) | Ruben Valls | 2022-10-18 | |
| | | ||||
| * | Fixes compiling when NNG_ENABLE_STATS is OFF (#1600) | Ruben Valls | 2022-06-20 | |
| | | | | The socket member `st_name` is not available when NNG_ENABLE_STATS is OFF. | |||
| * | fixes #1574 Non-blocking version of nng_aio_wait / nng_aio_result | Garrett D'Amore | 2022-04-18 | |
| | | | | | | | | This introduces a new API, nng_aio_busy(), that can be used to query the status of the aio without blocking. Some minor documentation fixes are included. | |||
| * | fixes #1577 nng_stat_find_socket always returns the same stats node | Garrett D'Amore | 2022-04-16 | |
| | | | | | | Also this fixes problems with uninitialized socket names, and the socket name stat not being adjusted correctly when set via API. | |||
| * | Replace nni_aio_prov_set_extra with nni_aio_prov_set_data. | Garrett D'Amore | 2021-12-31 | |
| | | | | | | | This takes one less parameter, and is simpler. It will let us reclaim the aio_prov_extra data space as well, so that we can use it for other purposes. | |||
| * | Fix leak of messages lost during device abort. | Garrett D'Amore | 2021-12-29 | |
| | | ||||
| * | don't use deprecated functions in tests (#1560) | Edward Rudd | 2021-12-29 | |
| | | ||||
| * | Introduce nng_device_aio(). | Garrett D'Amore | 2021-12-27 | |
| | | | | | | | This function is like nng_device(), but runs asynchronously. Also, this fixes #1503 nng_device causes nng_close to blocking | |||
| * | Socket and context initialization never fails. | Garrett D'Amore | 2021-12-25 | |
| | | | | | | This makes these functions entirely bullet proof, and eliminates yet more error handling cases. | |||
| * | Provide a tiny buf for lmq buffer by default. | Garrett D'Amore | 2021-12-25 | |
| | | | | | | This allows us to make nni_lmq_init() non-failing. (Although the buffer size requested at initialization might not be granted.) | |||
| * | Fix possible crash in allocation failure path for LMQ. | Garrett D'Amore | 2021-12-25 | |
| | | ||||
| * | Pollables can be completely inline. | Garrett D'Amore | 2021-12-25 | |
| | | | | | | This eliminates more failure paths, and brings us still closer to eliminating the possibility of failure during socket init. | |||
| * | Bus aio's can be inline. | Garrett D'Amore | 2021-12-25 | |
| | | ||||
| * | fixes #1552 Several warnings compiling in Windows with Visual Studio 2019 | Garrett D'Amore | 2021-12-25 | |
| | | | | | | Note that one of these warning is a real bug that would prevent TLS from functioning properly on Windows. | |||
| * | Could use GCC atomics for older versions of GCC. | Garrett D'Amore | 2021-12-08 | |
| | | | | | | | | | This should help greatly with performance on older systems such as CentOS 7 and GCC 4.8. Though, such folks really should update to newer compilers. Folks running version of GCC earlier than 4.7 will still pay a rather significant performance penalty, as they still implement atomics with a global mutex. | |||
| * | ID map static initialization support. | Garrett D'Amore | 2021-12-06 | |
| | | | | | This also arranges to clean up the maps at nng_fini time. | |||
| * | Static condvar initialization. | Garrett D'Amore | 2021-12-05 | |
| | | ||||
| * | Use static initialization for lists and mutexes. | Garrett D'Amore | 2021-12-05 | |
| | | | | | | | 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'Amore | 2021-12-05 | |
| | | | | | | 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_insert | Garrett D'Amore | 2021-12-02 | |
| | | ||||
| * | Fix incorrect elides of deprecated code. | Garrett D'Amore | 2021-11-29 | |
| | | | | | | | (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 #1346 windows ipc winsec fails frequently in CI/CD | Garrett D'Amore | 2021-11-02 | |
| | | ||||
| * | Remove unused eq_len member. | Garrett D'Amore | 2021-10-11 | |
| | | ||||
| * | Minor cleanups. | Garrett D'Amore | 2021-09-06 | |
| | | ||||
| * | Eliminate the pipe mutex and use atomic for pipe closed. | Garrett D'Amore | 2021-09-06 | |
| | | | | | | | | | | | 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'Amore | 2021-09-06 | |
| | | ||||
| * | fixes #1498 Endpoint close/shutdown could be synchronous (#1499) | Garrett D'Amore | 2021-09-04 | |
| | | ||||
| * | Minor cleanups. | Garrett D'Amore | 2021-08-21 | |
| | | ||||
| * | Remove unused prototypes. | Garrett D'Amore | 2021-08-21 | |
| | | ||||
| * | Move protocol.c into SP tree (it is SP specific.) | Garrett D'Amore | 2021-08-21 | |
| | | | | | Also, remove an extraneous initialization call. | |||
| * | fixes #1488 aio expiration list performance work needed | Garrett D'Amore | 2021-08-09 | |
| | | | | | | | 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. | |||
| * | * FIX #1486 by waking up latest aio each time. (#1487) | JaylinYu | 2021-08-09 | |
| | | ||||
| * | Remove separate protocol initialization step. | Garrett D'Amore | 2021-08-07 | |
| | | | | | | Nothing is using this, but it adds complexity and also requires additional lock activity each time a socket is opened. | |||
| * | Minor format and spelling in comments. | Garrett D'Amore | 2021-07-31 | |
| | | ||||
| * | Remove extra wrapper for platform clock support. | Garrett D'Amore | 2021-07-22 | |
| | | ||||
| * | Fix some unused variables. | Garrett D'Amore | 2021-07-22 | |
| | | ||||
| * | fixes #1475 nni_aio_begin should not dispatch task on stopped aio | Garrett D'Amore | 2021-07-22 | |
| | | ||||
| * | Fix the wrong ratio when expire queue shrink. (#1470) | wangha | 2021-07-16 | |
| | | ||||
| * | Minor clang-tidy fixes. | Garrett D'Amore | 2021-07-11 | |
| | | ||||
