| Commit message (Collapse) | Author | Age | ||
|---|---|---|---|---|
| ... | ||||
| * | windows ipc: check the return value of nni_aio_schedule | Garrett D'Amore | 2024-12-03 | |
| | | | | | | This hopefully addresses a long standing bug with Windows IPC sometimes hanging on shut down. | |||
| * | tests: void instead of empty parameter list | Garrett D'Amore | 2024-12-01 | |
| | | ||||
| * | http server: sprintf is considered deprecated | Garrett D'Amore | 2024-12-01 | |
| | | | | | | | | | | | Because it is typically associated with insecure code, use of sprintf is discouraged. Note that our usage was actually quite careful and not insecure, but its mere presence raises concern especially by parties who are unwilling or unable to assess the actual code for correctness. A better choice here would be strlcat, but strlcat is not universally available. | |||
| * | http server: fix race condition for server->closed | Garrett D'Amore | 2024-12-01 | |
| | | ||||
| * | tests: convert files test to NUTS | Garrett D'Amore | 2024-12-01 | |
| | | ||||
| * | tests: convert ipcsupp tests to NUTS | Garrett D'Amore | 2024-11-30 | |
| | | ||||
| * | tests: convert tcpsupp test to NUTS | Garrett D'Amore | 2024-11-30 | |
| | | | | | Also renamed this to tcp_stream_test. | |||
| * | tests: convert scalability test to NUTS | Garrett D'Amore | 2024-11-30 | |
| | | ||||
| * | tests: convert nonblock test to NUTS | Garrett D'Amore | 2024-11-30 | |
| | | ||||
| * | tests: convert synch test to NUTS. | Garrett D'Amore | 2024-11-30 | |
| | | | | | | While here we added a test for nng_cv_wake1 to demonstrate it does not fall afoul of the thundering herd. | |||
| * | tests: fix sign bug in comparing times | Garrett D'Amore | 2024-11-30 | |
| | | ||||
| * | tests: Allow more time for larger messages | Garrett D'Amore | 2024-11-30 | |
| | | ||||
| * | tests: incompletely initialized address in has_ipv6 check | Garrett D'Amore | 2024-11-30 | |
| | | ||||
| * | tests: fix type mismatch on htonl | Garrett D'Amore | 2024-11-30 | |
| | | ||||
| * | tests: convert ws transport test to NUTS | Garrett D'Amore | 2024-11-30 | |
| | | ||||
| * | tests: time collection overhead might be zero for coarse timers (Windows) | Garrett D'Amore | 2024-11-30 | |
| | | ||||
| * | tests: convert TCPv6 transport test to NUTS (and consolidate with v4) | Garrett D'Amore | 2024-11-30 | |
| | | ||||
| * | tests: better reporting of timing overhead | Garrett D'Amore | 2024-11-30 | |
| | | ||||
| * | tests: add a couple more transport specific tests | Garrett D'Amore | 2024-11-30 | |
| | | ||||
| * | tests: tcp test converted to NUTS | Garrett D'Amore | 2024-11-30 | |
| | | | | | | TCPv6 not done yet since that needs special work to be conditionalized. Also tcpsupp remains to be converted. | |||
| * | tests: inproc converted to NUTS | Garrett D'Amore | 2024-11-30 | |
| | | | | | | | | | This actually represents a conversion of the transport tests implemented in Convey terms to NUTS. As part of this, have implemented a simple round trip performance test, using PAIR. The rest of the transport tests will shortly be converted to this as well. | |||
| * | tests: Convert device test to NUTS | Garrett D'Amore | 2024-11-30 | |
| | | | | | | This also adds more tests for additional test cases (aio, and more validations of incompatible device configurations). | |||
| * | tests: req stress test converted to NUTS | Garrett D'Amore | 2024-11-30 | |
| | | ||||
| * | tests: multistress test converted to NUTS | Garrett D'Amore | 2024-11-30 | |
| | | ||||
| * | tests: Convert pipe test suite to NUTS | Garrett D'Amore | 2024-11-30 | |
| | | ||||
| * | tests: fix data race getting results | Garrett D'Amore | 2024-11-30 | |
| | | ||||
| * | sockfd: inline aio structs | Garrett D'Amore | 2024-11-24 | |
| | | ||||
| * | tls: don't use an extra resolver step in the listener. | Garrett D'Amore | 2024-11-24 | |
| | | | | | This was centralized in the stream layer a while ago. | |||
| * | sockfd: we can use NULL for dialer ops | Garrett D'Amore | 2024-11-24 | |
| | | ||||
| * | Remove all the get_ptr option accessors. | Garrett D'Amore | 2024-11-24 | |
| | | | | | | Temporarily we have removed access to the peer alt names, but that was never used and was not tested (it also didn't work with WolfSSL.) | |||
| * | Remove all the set_ptr option functions. | Garrett D'Amore | 2024-11-24 | |
| | | | | | | This hopefully ensures that we have good typed functions instead of just passing around pointers blithely. | |||
| * | Remove the NNG_OPT_IPC_SECURITY_DESCRIPTOR option. | Garrett D'Amore | 2024-11-24 | |
| | | | | | | | | | This is now replaced with nng_listener_set_security_descriptor and nng_stream_listener_set_security_descriptor functions. We may elect to remove these entirely, but for named pipe users they are probably still quite useful. Moving towards UNIX domain sockets would obsolete this functionality. | |||
| * | http: constify some functions | Garrett D'Amore | 2024-11-24 | |
| | | ||||
| * | ws: add some more test cases | Garrett D'Amore | 2024-11-24 | |
| | | ||||
| * | Remove string option functions from sockets and contexts. | Garrett D'Amore | 2024-11-24 | |
| | | | | | | These are not needed anymore, and the semantics of string accessors is brittle, so we want to eliminate these as much as possible. | |||
| * | socket: remove unused nng_socket_get_ptr and nng_socket_set_ptr. | Garrett D'Amore | 2024-11-24 | |
| | | | | | | | | | We will be removing these pointer based functions more generally as we we have type-specific accessors instead. We also removed the context versions of these functions, which were not previously documented. | |||
| * | option parser: test for no clustering support | Garrett D'Amore | 2024-11-24 | |
| | | ||||
| * | options parser: Add a couple of tests for ambiguous and missing arguments | Garrett D'Amore | 2024-11-23 | |
| | | ||||
| * | Converted options supplemental test to NUTS | Garrett D'Amore | 2024-11-23 | |
| | | ||||
| * | ipc: Add NNG_OPT_REMADDR for dialer. | Garrett D'Amore | 2024-11-23 | |
| | | ||||
| * | tls: add a mutual authentication test | Garrett D'Amore | 2024-11-23 | |
| | | | | | | | | | Also, make it clearer that TLS keys and certificates can only be set once on a configuration. (mbedTLS makes this confusing!) This mutual test is only fully validated on mbed, because wolfSSL seems to not properly validate this in many configurations. | |||
| * | tls: include error code from mbed in log messages | Garrett D'Amore | 2024-11-23 | |
| | | | | | Also add a test case for mismatch verify. | |||
| * | socket: fds can never be negative | Garrett D'Amore | 2024-11-23 | |
| | | ||||
| * | tests: add listener and dialer tests | Garrett D'Amore | 2024-11-23 | |
| | | ||||
| * | listener: add listener_create_url and listen_url | Garrett D'Amore | 2024-11-23 | |
| | | ||||
| * | dialers: add nng_dial_url and nng_dialer_create_url | Garrett D'Amore | 2024-11-23 | |
| | | | | | | This allows a URL object to be used for dialing, which may be easier than using a string if you already have the URL object. | |||
| * | Add nni_url_clone_inline. | Garrett D'Amore | 2024-11-23 | |
| | | | | | | The idea is to allow nng_dialer_create_url() and such to avoid having to reparse a URL that we already have. | |||
| * | url: add huge url tests | Garrett D'Amore | 2024-11-23 | |
| | | ||||
| * | Remove u_rawurl member of URL. | Garrett D'Amore | 2024-11-23 | |
| | | | | | | This means that most URLs can now be used without any allocations needed. It eliminates some failure paths. | |||
| * | core: remove NNG_OPT_URL entirely | Garrett D'Amore | 2024-11-23 | |
| | | ||||
