c/libuv
1
0
Fork 0
mirror of https://github.com/libuv/libuv synced 2025-03-28 21:13:16 +00:00
Code Issues Packages Projects Releases Wiki Activity
libuv/test/test-list.h

834 lines
25 KiB
C
Raw Normal View History

Add copyright headers, license file. closes #17.
2011-04-18 11:10:18 -07:00
/* Copyright Joyent, Inc. and other Node contributors. All rights reserved.
*
* Permission is hereby granted, free of charge, to any person obtaining a copy
* of this software and associated documentation files (the "Software"), to
* deal in the Software without restriction, including without limitation the
* rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
* sell copies of the Software, and to permit persons to whom the Software is
* furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice shall be included in
* all copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
* AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
* IN THE SOFTWARE.
*/
unix: allow nesting of kqueue fds in uv_poll_start kqueue file descriptors don't support ioctl(FIONBIO) (or any other ioctl for that matter) so retry using fcntl(F_GETFL) + fcntl(F_SETFL) when we receive a ENOTTY error. Fixes: https://github.com/libuv/libuv/issues/883 PR-URL: https://github.com/libuv/libuv/pull/885 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2016-05-23 20:03:30 +02:00
#include "uv.h"
platform api
2011-12-13 01:29:27 -08:00
TEST_DECLARE (platform_output)
test: add callback order test Ensure that idle callbacks run before other callbacks.
2012-05-22 16:04:09 +02:00
TEST_DECLARE (callback_order)
windows: run close callbacks after polling for i/o Makes the uv-win behavior compatible with uv-unix. Fixes #796.
2013-05-29 09:51:04 +02:00
TEST_DECLARE (close_order)
unix, windows: make uv_run_once() return a bool The return value of uv_run_once() now signals if it needs to be called again. Fixes #427.
2012-05-23 20:30:59 +02:00
TEST_DECLARE (run_once)
Implemented uv_run2 Allows for running the event loop in 3 modes: * default: loop runs until the refcount drops to zero * once: poll for events only once and block until one is handled * nowait: poll for events only once but don't block if there are no pending events
2012-12-18 16:10:11 +01:00
TEST_DECLARE (run_nowait)
unix, windows: add uv_loop_alive() function Useful to know when the the event loop is empty, this can't be done with uv_run() without possibly blocking, or running some events (which might empty the event loop as a side-effect).
2013-05-17 13:29:26 -07:00
TEST_DECLARE (loop_alive)
unix, windows: add uv_loop_init and uv_loop_close These functions supersede uv_loop_new and uv_loop_delete. uv_loop_init initialized a user allocated loop and uv_loop_close removes all associated resources a loop uses after it has finished execution. uv_loop_new and uv_loop_delete are now deprecated.
2014-02-17 22:37:20 +01:00
TEST_DECLARE (loop_close)
unix, windows: add uv_stop, stop running event loop
2013-01-16 09:36:20 +01:00
TEST_DECLARE (loop_stop)
unix, windows: always update loop time Fixes #846
2013-12-10 09:13:34 +01:00
TEST_DECLARE (loop_update_time)
windows: add uv_backend_timeout
2014-07-31 21:35:32 -04:00
TEST_DECLARE (loop_backend_timeout)
linux: fix epoll_pwait() sigmask size calculation Revisit the fix from commit b705b53. The problem with using sigset_t and _NSIG is that the size of sigset_t and the value of _NSIG depend on what headers libuv picks up first, <signal.h> or <asm/signal.h>. With the former, sizeof(sigset_t) = 128; with the latter, it's 8. Simply sidestep the issue by calculating the signal mask as a 64 bits integer, without using sigset_t or _NSIG. PR-URL: https://github.com/libuv/libuv/pull/83 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2014-12-25 14:11:55 +01:00
TEST_DECLARE (loop_configure)
test: add test for closing and recreating default loop
2014-09-08 21:47:50 +02:00
TEST_DECLARE (default_loop_close)
unix, windows: add thread barrier support
2012-10-09 21:44:20 +09:00
TEST_DECLARE (barrier_1)
TEST_DECLARE (barrier_2)
TEST_DECLARE (barrier_3)
Add support for condition variables on all platforms
2012-10-05 13:03:03 +02:00
TEST_DECLARE (condvar_1)
TEST_DECLARE (condvar_2)
TEST_DECLARE (condvar_3)
TEST_DECLARE (condvar_4)
TEST_DECLARE (condvar_5)
unix, windows: add semaphore functions
2012-06-03 03:17:21 +02:00
TEST_DECLARE (semaphore_1)
TEST_DECLARE (semaphore_2)
TEST_DECLARE (semaphore_3)
Add uv_is_tty()
2011-09-15 15:16:53 -07:00
TEST_DECLARE (tty)
win,tty: fix read stop in line mode Closing the handle does not make ReadConsoleW exit reliably on Windows 7 and above. Thus, after switching from line to raw mode, keypresses were held until enter was pressed. This makes ReadConsoleW exit by writing a return keypress to its input buffer, similar to what was already done for raw mode. Fixes: https://github.com/libuv/libuv/issues/852 PR-URL: https://github.com/libuv/libuv/pull/866 Reviewed-by: Bert Belder <bertbelder@gmail.com> Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2016-05-11 13:03:30 +01:00
#ifdef _WIN32
TEST_DECLARE (tty_raw)
Revert "Revert "win,tty: add support for ANSI codes in win10 v1511"" To bring back support for ANSI codes in win10 v1511. This reverts commit 8cbabaa81709562de68bcac0554e4e7af2f72cf6. PR-URL: https://github.com/libuv/libuv/pull/1143 Refs: https://github.com/libuv/libuv/pull/1138 Refs: https://github.com/libuv/libuv/pull/889 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com> Reviewed-By: Imran Iqbal <imran@imraniqbal.org>
2016-11-16 16:16:54 +01:00
TEST_DECLARE (tty_empty_write)
win,tty: fix MultiByteToWideChar output buffer Make sure there's enough room in the output buffer by dynamically allocating memory in case the size of the buffer needs to be greater than 8192 characters. PR-URL: https://github.com/libuv/libuv/pull/1143 Refs: https://github.com/libuv/libuv/pull/1138 Refs: https://github.com/libuv/libuv/pull/889 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com> Reviewed-By: Imran Iqbal <imran@imraniqbal.org>
2016-11-16 16:17:12 +01:00
TEST_DECLARE (tty_large_write)
win,tty: fix read stop in line mode Closing the handle does not make ReadConsoleW exit reliably on Windows 7 and above. Thus, after switching from line to raw mode, keypresses were held until enter was pressed. This makes ReadConsoleW exit by writing a return keypress to its input buffer, similar to what was already done for raw mode. Fixes: https://github.com/libuv/libuv/issues/852 PR-URL: https://github.com/libuv/libuv/pull/866 Reviewed-by: Bert Belder <bertbelder@gmail.com> Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2016-05-11 13:03:30 +01:00
#endif
unix: reject non-tty fds in uv_tty_init() Reject file descriptors that correspond to regular files or char/block devices. Such file descriptors are incompatible with epoll and trigger a run-time assert. Fixes: https://github.com/libuv/libuv/issues/255 PR-URL: https://github.com/libuv/libuv/pull/259 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2015-03-11 16:05:56 +01:00
TEST_DECLARE (tty_file)
unix: open ttyname instead of /dev/tty Find the real name of the tty using ttyname_r(3) instead of opening "/dev/tty" which causes trouble if the fd doesn't point to the controlling terminal. PR-URL: https://github.com/libuv/libuv/pull/779 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl> Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2016-03-23 23:18:53 +01:00
TEST_DECLARE (tty_pty)
windows: stdio over non-overlapped pipes
2011-10-19 00:48:38 -07:00
TEST_DECLARE (stdio_over_pipes)
inet: allow scopeid in uv_inet_pton We already support it in `uv_ip6_addr` anyway. See https://github.com/joyent/node/issues/7395
2014-04-14 15:22:08 +04:00
TEST_DECLARE (ip6_pton)
windows: when sharing a server socket, only call listen in the parent process
2011-10-25 13:07:29 -07:00
TEST_DECLARE (ipc_listen_before_write)
TEST_DECLARE (ipc_listen_after_write)
Disable ipc_send_recv_pipe test on Windows It's not supported yet
2012-04-12 03:33:53 +02:00
#ifndef _WIN32
test: test cross-process handle send/recv
2012-03-09 03:16:08 +01:00
TEST_DECLARE (ipc_send_recv_pipe)
pipe: enable inprocess uv_write2 on Windows When duplicating the socket handle being sent the target process id is defaulted to the current process id. This enables uv_write2 to be used for thread-clustering in addition to process-clustering on multi-threaded programming languages. The ipc tests are updated to run in two modes. In the _inproc mode the echo work is done on a second thread instead of in a second process. An internal function int uv_current_pid() is added to the windows specific code which caches the value of GetCurrentProcessId(). This means uv_write2 does not call GetCurrentProcessId() every inprocess send. Refs: https://github.com/joyent/libuv/issues/926 PR-URL: https://github.com/libuv/libuv/pull/540 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2015-10-28 11:40:51 -07:00
TEST_DECLARE (ipc_send_recv_pipe_inprocess)
Disable ipc_send_recv_pipe test on Windows It's not supported yet
2012-04-12 03:33:53 +02:00
#endif
test: test cross-process handle send/recv
2012-03-09 03:16:08 +01:00
TEST_DECLARE (ipc_send_recv_tcp)
pipe: enable inprocess uv_write2 on Windows When duplicating the socket handle being sent the target process id is defaulted to the current process id. This enables uv_write2 to be used for thread-clustering in addition to process-clustering on multi-threaded programming languages. The ipc tests are updated to run in two modes. In the _inproc mode the echo work is done on a second thread instead of in a second process. An internal function int uv_current_pid() is added to the windows specific code which caches the value of GetCurrentProcessId(). This means uv_write2 does not call GetCurrentProcessId() every inprocess send. Refs: https://github.com/joyent/libuv/issues/926 PR-URL: https://github.com/libuv/libuv/pull/540 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2015-10-28 11:40:51 -07:00
TEST_DECLARE (ipc_send_recv_tcp_inprocess)
windows: Enable passing of TCP connections over IPC
2012-03-08 12:02:50 -08:00
TEST_DECLARE (ipc_tcp_connection)
unix,win: make on_alloc_cb failures more resilient Treat both the base being NULL or the length being 0 as ENOBUFS. PR-URL: https://github.com/libuv/libuv/pull/997 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
2016-08-16 00:15:54 +02:00
TEST_DECLARE (tcp_alloc_cb_fail)
Named pipes implementation for Windows
2011-07-01 17:54:17 -07:00
TEST_DECLARE (tcp_ping_pong)
TEST_DECLARE (tcp_ping_pong_v6)
TEST_DECLARE (pipe_ping_pong)
Test for not immediately calling oio_accept from accept_cb Closes #10.
2011-04-22 00:45:49 +02:00
TEST_DECLARE (delayed_accept)
windows: don't fail uv_listen with UV_EALREADY if already listening. fixes #227.
2011-10-26 13:32:15 -07:00
TEST_DECLARE (multiple_listen)
unix: fix tcp write after bad connect freezing If the connect wouldn't go off (no such tcp remote or any other failure), the subsequent writes made would not be called. Now we call the writes in the queue with ECANCELED if the connect fails. Fix #1432
2014-08-21 10:59:35 +02:00
#ifndef _WIN32
TEST_DECLARE (tcp_write_after_connect)
#endif
Write-a-lot test
2011-04-22 05:12:10 +02:00
TEST_DECLARE (tcp_writealot)
test: add test for deferred TCP write failure The test exhibits that write failures are deferred to the callback. Refs: https://github.com/libuv/libuv/issues/339 PR_URL: https://github.com/libuv/libuv/pull/344
2015-05-05 23:09:23 +02:00
TEST_DECLARE (tcp_write_fail)
stream: introduce uv_try_write(...) `uv_try_write(stream, buf, size)` acts like `uv_write()`, but without queueing actual write until UV_POLLOUT (or IOCP completion). This is useful for doing writes using on-stack `uv_write_t` requests. fix #1025
2013-12-06 19:10:47 +04:00
TEST_DECLARE (tcp_try_write)
unix: guarantee write queue cb execution order in streams The problem was that the canceled write callbacks were being called before the already written callbacks, even though they were queued after them.
2014-07-06 14:43:59 +02:00
TEST_DECLARE (tcp_write_queue_order)
unix, windows: add uv_tcp_open and uv_udp_open
2012-09-17 12:15:12 +02:00
TEST_DECLARE (tcp_open)
unix, win: prevent replacing fd in uv_{udp,tcp,pipe}_t PR-URL: https://github.com/libuv/libuv/pull/400 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
2015-06-18 00:41:44 +02:00
TEST_DECLARE (tcp_open_twice)
test: rename test tcp_dont_connect_after_write Rename test tcp_dont_connect_after_write to tcp_connect_error_after_write, teensy tad more descriptive.
2012-06-04 17:18:02 +02:00
TEST_DECLARE (tcp_connect_error_after_write)
unix: fix event loop stall Stop libev from entering the epoll_wait/kevent/port_getn/etc. syscall when there are no active handles left, that will block indefinitely.
2012-06-04 23:18:32 +02:00
TEST_DECLARE (tcp_shutdown_after_write)
Add uv_pipe_bind tests
2011-07-21 02:00:12 +02:00
TEST_DECLARE (tcp_bind_error_addrinuse)
TEST_DECLARE (tcp_bind_error_addrnotavail_1)
TEST_DECLARE (tcp_bind_error_addrnotavail_2)
TEST_DECLARE (tcp_bind_error_fault)
TEST_DECLARE (tcp_bind_error_inval)
TEST_DECLARE (tcp_bind_localhost_ok)
unix, windows: validate flags on uv_udp|tcp_bind fixes #1247
2014-04-22 22:45:58 +02:00
TEST_DECLARE (tcp_bind_invalid_flags)
Add uv_pipe_bind tests
2011-07-21 02:00:12 +02:00
TEST_DECLARE (tcp_listen_without_bind)
test: enable tcp_connect_error_fault test
2012-05-17 11:44:05 +02:00
TEST_DECLARE (tcp_connect_error_fault)
test: add tcp connect timeout test
2012-05-17 11:43:33 +02:00
TEST_DECLARE (tcp_connect_timeout)
test: add tcp 'close on failed connect' test Demonstrates temporary event loop stall with uv-unix. The issue is that pending requests aren't processed until the next event (I/O, timeout, etc.) happens. See #446, #447 and #448.
2012-06-06 16:00:53 +02:00
TEST_DECLARE (tcp_close_while_connecting)
unix: bring back uv__stream_destroy() This is the revised version of reverted commit 431195c.
2011-09-09 23:53:09 +02:00
TEST_DECLARE (tcp_close)
unix, win: add ability to create tcp/udp sockets early Introduce two new APIs: int uv_tcp_init_ex(uv_loop_t*, uv_tcp_t* handle, int flags) int uv_udp_init_ex(uv_loop_t*, uv_udp_t* handle, int flags) The lower 8 bits of the flags field are used for the socket domain. AF_INET, AF_INET6 and AF_UNSPEC are supported. If AF_UNSPEC is specified the socket is created lazily, just like uv_{tcp,udp}_init. Some Windows notes: getsockname fails with WSAEINVAL if the socket is not bound. This could potentially be improved by detecting the socket family and filling the sockaddr_in/6 struct manually. bind returns WSAEFAULT if we try to bind a socket to the wrong family. Unix returns EINVAL. PR-URL: https://github.com/libuv/libuv/pull/400 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
2015-06-17 00:34:58 +02:00
TEST_DECLARE (tcp_create_early)
TEST_DECLARE (tcp_create_early_bad_bind)
TEST_DECLARE (tcp_create_early_bad_domain)
TEST_DECLARE (tcp_create_early_accept)
test: disable tcp_close_accept on Windows It always was a Unix only test
2014-08-05 00:14:26 +02:00
#ifndef _WIN32
unix: fix reopened fd bug When fd is closed and new one (with the same number) is opened inside kqueue/epoll/port loop's callback - stale events might invoke callbacks on wrong watchers. Check if watcher was changed after invocation and invalidate all events with the same fd. fix #826
2013-10-25 12:56:37 +04:00
TEST_DECLARE (tcp_close_accept)
stream: use SO_OOBINLINE on OS X In the collaboration with Ben Noordhuis <info@bnoordhuis.nl> and Saúl Ibarra Corretgé <saghul@gmail.com>. Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com> Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
2015-02-24 16:16:22 +03:00
TEST_DECLARE (tcp_oob)
test: disable tcp_close_accept on Windows It always was a Unix only test
2014-08-05 00:14:26 +02:00
#endif
unix: add TCP keepalive and no-delay control knobs
2011-10-21 15:37:47 -07:00
TEST_DECLARE (tcp_flags)
windows: convert WSAECONNABORTED to EOF
2011-11-07 15:48:33 -08:00
TEST_DECLARE (tcp_write_to_half_open_connection)
test: add 'unexpected read' tcp test Regression test that verifies that the event loop doesn't busy loop when the server sends a message and the client isn't reading.
2012-06-30 02:59:29 +02:00
TEST_DECLARE (tcp_unexpected_read)
unix: fix event loop stall Fix a rather obscure bug where the event loop stalls when an I/O watcher is stopped while an artificial event, generated with uv__io_feed(), is pending.
2012-12-13 16:37:12 +01:00
TEST_DECLARE (tcp_read_stop)
Add uv_pipe_bind tests
2011-07-21 02:00:12 +02:00
TEST_DECLARE (tcp_bind6_error_addrinuse)
TEST_DECLARE (tcp_bind6_error_addrnotavail)
TEST_DECLARE (tcp_bind6_error_fault)
TEST_DECLARE (tcp_bind6_error_inval)
TEST_DECLARE (tcp_bind6_localhost_ok)
unix,win: make on_alloc_cb failures more resilient Treat both the base being NULL or the length being 0 as ENOBUFS. PR-URL: https://github.com/libuv/libuv/pull/997 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
2016-08-16 00:15:54 +02:00
TEST_DECLARE (udp_alloc_cb_fail)
unix, windows: do not set SO_REUSEADDR by default on udp Add UV_UDP_REUSEADDR flag instead, which can be passed to uv_udp_bind. If the udp handle is unbound when uv_udp_set_memberhsip or uv_udp_set_multicast_interface is called, the handle will be bound with UV_UDP_REUSEADDR set.
2014-03-31 09:56:37 +02:00
TEST_DECLARE (udp_bind)
TEST_DECLARE (udp_bind_reuseaddr)
unix, win: add ability to create tcp/udp sockets early Introduce two new APIs: int uv_tcp_init_ex(uv_loop_t*, uv_tcp_t* handle, int flags) int uv_udp_init_ex(uv_loop_t*, uv_udp_t* handle, int flags) The lower 8 bits of the flags field are used for the socket domain. AF_INET, AF_INET6 and AF_UNSPEC are supported. If AF_UNSPEC is specified the socket is created lazily, just like uv_{tcp,udp}_init. Some Windows notes: getsockname fails with WSAEINVAL if the socket is not bound. This could potentially be improved by detecting the socket family and filling the sockaddr_in/6 struct manually. bind returns WSAEFAULT if we try to bind a socket to the wrong family. Unix returns EINVAL. PR-URL: https://github.com/libuv/libuv/pull/400 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
2015-06-17 00:34:58 +02:00
TEST_DECLARE (udp_create_early)
TEST_DECLARE (udp_create_early_bad_bind)
TEST_DECLARE (udp_create_early_bad_domain)
Add UDP support to libuv.
2011-08-19 23:22:30 +02:00
TEST_DECLARE (udp_send_and_recv)
test: add test for uv_udp_send() fix Test that the fast path will continue writing. On failure, this test will timeout. PR-URL: https://github.com/libuv/libuv/pull/1308 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com> Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
2017-04-18 17:59:32 -06:00
TEST_DECLARE (udp_send_hang_loop)
unix: remove incorrect assert After 41891222bc landed it's possible that uv__udp_sendmsg is called even if there are no pending write nor write completed requests: 1. User calls uv_udp_send and the request is sent immediately. The request is the added to the completed queue and we 'feed' the uv__io handle so that we process the completed request in the next iteration. 2. User calls uv_udp_send again but the request is not completed immediately, so it's queued in the write_queue. 3. The uv__io handle gets a UV__POLLOUT event and uv__udp_sendmsg is run, which completes the send request and puts it in the write_completed_queue. Afterwards, uv__udp_run_completed is executed and the write_completed queue is drained. 4. At this point, the uv__io handle was made pending in step 3, in uv__udp_sendmsg, but we no longer have requests to write or to complete, so we skip processing.
2014-07-11 22:42:55 +02:00
TEST_DECLARE (udp_send_immediate)
windows: fix buffer leak after failed udp send Fixes #1426 Signed-off-by: Saúl Ibarra Corretgé <saghul@gmail.com>
2014-08-27 21:50:48 +02:00
TEST_DECLARE (udp_send_unreachable)
add uv_udp_set_membership for unix multicast support - test-udp-multicast-join tests that multicast packets can be received - stub src/win/udp.c until support added
2011-09-29 19:29:15 -04:00
TEST_DECLARE (udp_multicast_join)
unix, windows: add IPv6 support to uv_udp_set_membership Reworked from initial version by @snoj
2014-03-17 10:06:25 +01:00
TEST_DECLARE (udp_multicast_join6)
test: add multicast TTL test
2012-01-19 11:07:28 -05:00
TEST_DECLARE (udp_multicast_ttl)
unix, win: add uv_udp_set_multicast_interface()
2013-01-22 10:47:31 -08:00
TEST_DECLARE (udp_multicast_interface)
unix, windows: add IPv6 support for uv_udp_multicast_interface
2014-03-28 00:39:09 +01:00
TEST_DECLARE (udp_multicast_interface6)
Add UDP support to libuv.
2011-08-19 23:22:30 +02:00
TEST_DECLARE (udp_dgram_too_big)
TEST_DECLARE (udp_dual_stack)
TEST_DECLARE (udp_ipv6_only)
test: add UDP broadcast/multicast/ttl tests
2012-01-28 02:16:00 +01:00
TEST_DECLARE (udp_options)
unix: fix ttl, multicast ttl and loop options on IPv6 Fixes #93 PR-URL: https://github.com/libuv/libuv/pull/99 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
2015-01-01 13:15:57 +01:00
TEST_DECLARE (udp_options6)
unix, windows: be consistent when binding implictly in UDP Only these functions will trigger an implicit binding of a UDP handle: - uv_udp_send - uv_udp_recv_start - uv_udp_set_membership All other functions will return UV_EBADF in case the socket was not bound. Note: currently the socket is created and bound at the same time. This may change in the future.
2014-05-07 11:12:19 +02:00
TEST_DECLARE (udp_no_autobind)
unix, windows: add uv_tcp_open and uv_udp_open
2012-09-17 12:15:12 +02:00
TEST_DECLARE (udp_open)
unix, win: prevent replacing fd in uv_{udp,tcp,pipe}_t PR-URL: https://github.com/libuv/libuv/pull/400 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
2015-06-18 00:41:44 +02:00
TEST_DECLARE (udp_open_twice)
unix, win: add uv_udp_try_send The function returns UV_EAGAIN if the queue is not empty and the message couldn't be sent immediately.
2014-06-30 00:01:10 +02:00
TEST_DECLARE (udp_try_send)
Add uv_pipe_bind tests
2011-07-21 02:00:12 +02:00
TEST_DECLARE (pipe_bind_error_addrinuse)
TEST_DECLARE (pipe_bind_error_addrnotavail)
TEST_DECLARE (pipe_bind_error_inval)
win: fix replacing pipe handle for pipe servers On Windows we create multiple pipe handles (system handles) which are attached to pending accept requests. Each of these will take turns in replacing the reference in handle->handle, so make sure we allow for that **only** for pipe servers. PR-URL: https://github.com/libuv/libuv/pull/488 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
2015-08-20 10:49:21 +02:00
TEST_DECLARE (pipe_connect_multiple)
Add uv_pipe_bind tests
2011-07-21 02:00:12 +02:00
TEST_DECLARE (pipe_listen_without_bind)
Fix UNIX pipe connect error reporting, add test
2011-11-03 15:46:36 -07:00
TEST_DECLARE (pipe_connect_bad_name)
windows: return UV_ENOTSOCK when doing uv_pipe_connect to a file
2012-03-01 12:11:12 -08:00
TEST_DECLARE (pipe_connect_to_file)
unix: don't block for io if any io handle is primed PR-URL: https://github.com/libuv/libuv/pull/347 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
2015-05-08 09:33:47 +02:00
TEST_DECLARE (pipe_connect_on_prepare)
unix, windows: add uv_pipe_getsockname
2013-12-17 00:15:54 +01:00
TEST_DECLARE (pipe_getsockname)
TEST_DECLARE (pipe_getsockname_abstract)
windows: make uv_read_stop immediately stop reading This implements locking around the blocking call to ReadFile to get around a Windows kernel bug where a blocking ReadFile operation on a stream can deadlock the thread. This allows uv_read_stop to immediately cancel a pending IO operation, and allows uv_pipe_getsockname to "pause" any pending read (from libuv) while it retrieves the sockname information. If unsupported by the OS (pre-Vista), this reverts to the old (e.g. deadlock-prone) behavior Closes #1313
2014-07-15 01:49:11 -04:00
TEST_DECLARE (pipe_getsockname_blocking)
win: fix setting pipe pending instances after bind It's an error to do so, so just ignore it. The test would cause an invalid memory access if the fix is undone. PR-URL: https://github.com/libuv/libuv/pull/488 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
2015-08-20 10:53:20 +02:00
TEST_DECLARE (pipe_pending_instances)
pipe: allow queueing pending handles Introduce `int uv_pipe_pending_count(uv_pipe_t*)` and `uv_handle_type uv_pipe_pending_type(uv_pipe_t*)`. They should be used in IPC pipe's read cb to accept incoming handles: int count = uv_pipe_pending_count(pipe); int i; for (i = 0; i < count; i++) { uv_handle_type type = uv_pipe_pending_type(pipe); /* ... */ uv_accept(...); }
2013-12-22 14:45:00 +04:00
TEST_DECLARE (pipe_sendmsg)
test: add pipe-server-close test Add a regression test for the pipe server close issue on Windows, which was reported in joyent/node#6749 and fixed in 7b16a3f.
2013-12-20 19:40:02 -08:00
TEST_DECLARE (pipe_server_close)
Test for not immediately calling oio_accept from accept_cb Closes #10.
2011-04-22 00:45:49 +02:00
TEST_DECLARE (connection_fail)
API Change: uv_close only called by user - never automatically Add test that failing on_connect callback does not trigger on_close.
2011-06-02 09:19:56 +02:00
TEST_DECLARE (connection_fail_doesnt_auto_close)
Tests: verify that shutdown_cb is always called
2012-03-09 04:29:45 +01:00
TEST_DECLARE (shutdown_close_tcp)
TEST_DECLARE (shutdown_close_pipe)
Add test to assert behavior of uv_tcp_t after shutdown Handles should *never* call the close callback without the user having first issued a uv_close() command. This test is broken on windows. See #52.
2011-06-02 11:31:42 +02:00
TEST_DECLARE (shutdown_eof)
unix: correct error when calling uv_shutdown twice Before this one of the requests was left as a ghost in the system. See https://github.com/JuliaLang/julia/issues/5793
2014-02-14 22:27:45 -05:00
TEST_DECLARE (shutdown_twice)
More extensive test-callback-stack. Fails on windows now.
2011-05-04 17:10:33 +02:00
TEST_DECLARE (callback_stack)
unix,win: add uv_os_{get,set,unset}env() These functions are used to create, retrieve, update, and delete environment variables. Fixes: https://github.com/libuv/libuv/issues/1198 PR-URL: https://github.com/libuv/libuv/pull/1234 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com> Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
2017-02-28 00:21:26 -05:00
TEST_DECLARE (env_vars)
error reporting: return sensible messages for synthetic errors
2011-10-06 01:00:56 +02:00
TEST_DECLARE (error_message)
unix,win: add uv_translate_sys_error() public API uv_translate_sys_error() was a private function for Windows. This commit adds an equivalent function on other platforms, and exposes it as public API. Exposing this is useful in scenarios where the application uses both libuv functions and platform-specific system calls and wants to report errors uniformly as libuv errors. Fixes: https://github.com/libuv/libuv/issues/79 PR-URL: https://github.com/libuv/libuv/pull/1060 Reviewed-By: Colin Ihrig <cjihrig@gmail.com> Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2016-09-18 15:18:48 -04:00
TEST_DECLARE (sys_error)
Update tests & benchmarks to use the new timer api
2011-05-17 01:17:48 +02:00
TEST_DECLARE (timer)
unix: make uv_timer_init() initialize repeat uv_timer_get_repeat() should return 0 for timers that haven't been started.
2013-03-04 22:17:50 -06:00
TEST_DECLARE (timer_init)
Test for uv_timer_again ,~get_repeat, ~set_repeat
2011-05-19 01:58:07 +02:00
TEST_DECLARE (timer_again)
unix: update timer if already active uv_timer_start() no longer returns an error when the timer is already active, now it just updates the timer. Consistent with the uv-win implementation. Fixes #425.
2012-05-26 02:09:35 +02:00
TEST_DECLARE (timer_start_twice)
unix,windows: fix timer order in case of same timeout Compare start_id of timer handles when they have the same timeout. start_id is allocated with loop->timer_counter in uv_timer_start.
2013-02-07 13:59:17 +09:00
TEST_DECLARE (timer_order)
unix: make timers handle large timeouts This commit fixes two closely related integer overflow bugs: * Timers with a timeout > INT_MAX cause uv__next_timeout() to return a negative value. * Timers with very large timeouts (close or equal to ULLONG_MAX) run on the next tick. In both cases, clamp the values to prevent the overflow from happening. Fixes joyent/node#5101.
2013-03-21 14:47:38 +01:00
TEST_DECLARE (timer_huge_timeout)
TEST_DECLARE (timer_huge_repeat)
unix, windows: run expired timers in run-once mode Before this commit, creating an event loop, starting a timer and calling uv_run(UV_RUN_ONCE) blocked in uv_run() until the timer expired - but didn't actually run the timer.
2013-05-25 10:37:08 +02:00
TEST_DECLARE (timer_run_once)
test: add 'start timer from check handle' test Check that a timer that is started from a check handle gets picked up correctly, i.e. that it influences the timeout used in the next tick of the event loop.
2013-06-26 13:02:39 +02:00
TEST_DECLARE (timer_from_check)
unix, windows: don't allow a NULL callback on timers
2014-09-04 14:17:59 +02:00
TEST_DECLARE (timer_null_callback)
win: wait for full timeout duration uv_poll should wait for at least the full timeout duration when there is nothing else to do. This was not happening because GetQueuedCompletionStatus can occasionally return up to 15ms early. The added test reproduces https://github.com/joyent/node/blob/d13d7f74d794340ac5e126cfb4ce507fe0f803d5/test/simple/test-timers-first-fire.js on libuv, being flaky before this fix. Fixes: https://github.com/joyent/node/issues/8960 PR-URL: https://github.com/libuv/libuv/pull/385 Reviewed-By: Alexis Campailla <alexis@janeasystems.com> Reviewed-by: Bert Belder <bertbelder@gmail.com>
2015-06-02 20:02:55 +01:00
TEST_DECLARE (timer_early_check)
Test: active idle watcher should not block other events
2011-07-21 02:51:26 +02:00
TEST_DECLARE (idle_starvation)
Test for oio_prepare/oio_check/oio_idle
2011-05-08 04:44:55 +02:00
TEST_DECLARE (loop_handles)
test: add lots of refcount tests
2012-01-14 00:11:27 +01:00
TEST_DECLARE (get_loadavg)
unix, windows: add uv_walk() Lets the libuv user iterate over the open handles. Mostly intended as a debugging tool or a post-hoc cleanup mechanism.
2012-05-30 01:51:28 +02:00
TEST_DECLARE (walk_handles)
test: add regression test for 29fdb3471
2013-10-30 21:43:23 +04:00
TEST_DECLARE (watcher_cross_stop)
Add reference count tests Broken on UNIX.
2011-05-20 08:22:19 -07:00
TEST_DECLARE (ref)
TEST_DECLARE (idle_ref)
TEST_DECLARE (async_ref)
TEST_DECLARE (prepare_ref)
TEST_DECLARE (check_ref)
Test that loop refs going down to zero in a prepare callback does not hang the event loop
2011-07-21 14:53:40 +02:00
TEST_DECLARE (unref_in_prepare_cb)
test: add lots of refcount tests
2012-01-14 00:11:27 +01:00
TEST_DECLARE (timer_ref)
TEST_DECLARE (timer_ref2)
TEST_DECLARE (fs_event_ref)
unix, windows: add stat() based file watcher Monitors a file path for changes. Supersedes ev_stat.
2012-06-14 07:30:08 +02:00
TEST_DECLARE (fs_poll_ref)
test: add lots of refcount tests
2012-01-14 00:11:27 +01:00
TEST_DECLARE (tcp_ref)
TEST_DECLARE (tcp_ref2)
windows: closing handles should always keep the loop alive This makes the tcp-ref2 and udp-ref2 tests pass again. Also adds another reference count test.
2012-10-25 15:10:16 +02:00
TEST_DECLARE (tcp_ref2b)
test: add lots of refcount tests
2012-01-14 00:11:27 +01:00
TEST_DECLARE (tcp_ref3)
Tests: verify that uv_write and uv_shutdown ref the event loop
2012-03-09 04:26:51 +01:00
TEST_DECLARE (tcp_ref4)
test: add lots of refcount tests
2012-01-14 00:11:27 +01:00
TEST_DECLARE (udp_ref)
TEST_DECLARE (udp_ref2)
TEST_DECLARE (udp_ref3)
TEST_DECLARE (pipe_ref)
TEST_DECLARE (pipe_ref2)
TEST_DECLARE (pipe_ref3)
Tests: verify that uv_write and uv_shutdown ref the event loop
2012-03-09 04:26:51 +01:00
TEST_DECLARE (pipe_ref4)
unix: read on stream even when UV__POLLHUP set. This fixes a SmartOS specific issue that happens when reading from a stream that is the reading end of a pipe that has been closed by the parent process. In this case, a UV__POLLHUP event would be set on the stream and would prevent the event loop from closing it. As a result, the event loop would think there are stil handles open, and leave the process hanging. Fixes #1419.
2014-08-14 16:38:56 -07:00
#ifndef _WIN32
TEST_DECLARE (pipe_close_stdout_read_stdin)
#endif
unix: implement uv_stream_set_blocking() Commit 393c1c5 ("unix: set non-block mode in uv_{pipe,tcp,udp}_open") causes a regression in the io.js cluster module. The io.js documentation states that `worker.send()` and `process.send()` are synchronous but they no longer were after upgrading to libuv v1.2.1. The reason they are synchronous is because of backpressure - or rather, lack of backpressure: a slow consumer eventually causes a fast producer to run out of memory because the backlog of pending messages in the producer can grow unchecked. Ergo, implement uv_stream_set_blocking() on UNIX platforms to let io.js enable the old blocking behavior for pipes again. Refs: https://github.com/iojs/io.js/issues/760 PR-URL: https://github.com/libuv/libuv/pull/187 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2015-02-09 12:53:52 +01:00
TEST_DECLARE (pipe_set_non_blocking)
test: add lots of refcount tests
2012-01-14 00:11:27 +01:00
TEST_DECLARE (process_ref)
unix, windows: add uv_has_ref() function
2013-04-17 00:16:42 +02:00
TEST_DECLARE (has_ref)
test: add test for uv_is_active and uv_is_closing
2012-08-27 23:57:48 +02:00
TEST_DECLARE (active)
unix: add uv_backend_fd() and uv_backend_timeout() This can be used in conjuction with uv_run_once() to poll in one thread and run the event loop's event callbacks in another. Useful for embedding libuv's event loop in another event loop.
2012-11-26 14:20:11 +04:00
TEST_DECLARE (embed)
Add test-async
2011-05-09 04:29:22 +02:00
TEST_DECLARE (async)
unix, windows: allow NULL async callback Allow a NULL callback so the user doesn't have to provide a dummy when the actual event is processed by e.g. a check handle callback.
2013-08-12 08:42:49 +02:00
TEST_DECLARE (async_null_cb)
fs: don't nullify req->bufs on EINTR uv__fs_buf_iter currently sets req->bufs to NULL after it is done, but if the operation fails with EINTR then it will be retried, at which point it expects the bufs to not be NULL, causing a seg fault as in https://github.com/nodejs/node/issues/4291. uv__fs_buf_iter should not set req->bufs to NULL if the operation fails with EINTR. Also, when it sets req->bufs to NULL, it should set req->nbufs to 0 as well, so we don't have the messy situation of a positive nbufs with no actual bufs. PR-URL: https://github.com/libuv/libuv/pull/661 Reviewed-By: Fedor Indutny <fedor@indutny.com>
2015-12-19 22:21:07 -08:00
TEST_DECLARE (eintr_handling)
Adds uv_get_exepath API Only works on Linux, Mac, Windows currently.
2011-05-20 18:05:37 -07:00
TEST_DECLARE (get_currentexe)
platform api
2011-12-13 01:29:27 -08:00
TEST_DECLARE (process_title)
uv_cwd + uv_chdir
2011-11-30 15:31:39 -08:00
TEST_DECLARE (cwd_and_chdir)
os: implement memory bindings * us_get_free_memory * us_get_total_memory
2011-10-01 15:45:42 +07:00
TEST_DECLARE (get_memory)
unix,win: add uv_get_passwd() This commit adds the uv_get_passwd() function, which returns a subset of the current effective user's password file entry. Refs: https://github.com/libuv/libuv/issues/11 Fixes: https://github.com/libuv/libuv/issues/731 PR-URL: https://github.com/libuv/libuv/pull/742 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl> Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2016-03-01 15:41:01 -05:00
TEST_DECLARE (get_passwd)
unix, windows: add uv_fileno Returns the platform specific file descriptor for handles that are backed by one. The datatype is abstracted as uv_os_fd_t, which maps to int on Unices and HANDLE on Windows. Users can use this function to set specific socket options, for example, in a non portable way. This function is essentially a shotgun, you better be careful with whatever you do with it, don't blame me if you used it to get the fd of a stream, close it yourself and expect things to Just Work.
2014-08-22 09:57:43 +02:00
TEST_DECLARE (handle_fileno)
unix,win: add uv_os_homedir() PR-URL: https://github.com/libuv/libuv/pull/350 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl> Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2015-05-12 09:17:19 -04:00
TEST_DECLARE (homedir)
unix,win: add uv_os_tmpdir() PR-URL: https://github.com/libuv/libuv/pull/672 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2016-01-04 16:52:44 -05:00
TEST_DECLARE (tmpdir)
Add test for uv_get_hrtime
2011-06-17 11:47:35 +02:00
TEST_DECLARE (hrtime)
unix: getaddrinfo_cb status arg should be 0 or -1 The documentation in uv.h states that the status argument to the uv_getaddrinfo() callback is either 0 or -1 but uv-unix actually passed it the addrinfo error code. Rectify that and add a regression test. Fixes #754.
2013-03-26 22:29:06 +01:00
TEST_DECLARE (getaddrinfo_fail)
unix, win: add synchronous uv_get{addr,name}info PR-URL: https://github.com/libuv/libuv/pull/156 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl> Reviewed-By: Bert Belder <bertbelder@gmail.com>
2015-01-20 09:43:57 +01:00
TEST_DECLARE (getaddrinfo_fail_sync)
Separate uv_getaddrinfo tests
2011-06-25 15:33:52 +02:00
TEST_DECLARE (getaddrinfo_basic)
unix, win: add synchronous uv_get{addr,name}info PR-URL: https://github.com/libuv/libuv/pull/156 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl> Reviewed-By: Bert Belder <bertbelder@gmail.com>
2015-01-20 09:43:57 +01:00
TEST_DECLARE (getaddrinfo_basic_sync)
Separate uv_getaddrinfo tests
2011-06-25 15:33:52 +02:00
TEST_DECLARE (getaddrinfo_concurrent)
unix, windows: getnameinfo implementation Closes #852
2014-05-11 12:51:08 +02:00
TEST_DECLARE (getnameinfo_basic_ip4)
unix, win: add synchronous uv_get{addr,name}info PR-URL: https://github.com/libuv/libuv/pull/156 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl> Reviewed-By: Bert Belder <bertbelder@gmail.com>
2015-01-20 09:43:57 +01:00
TEST_DECLARE (getnameinfo_basic_ip4_sync)
unix, windows: getnameinfo implementation Closes #852
2014-05-11 12:51:08 +02:00
TEST_DECLARE (getnameinfo_basic_ip6)
test: add udp getsockname test
2011-08-25 02:30:10 +02:00
TEST_DECLARE (getsockname_tcp)
TEST_DECLARE (getsockname_udp)
Allow tests and benchmarks to use the same helpers. Closes #21.
2011-04-19 14:51:10 +02:00
TEST_DECLARE (fail_always)
TEST_DECLARE (pass_always)
unix, windows: add uv_recv_buffer_size and uv_send_buffer_size
2014-07-07 14:56:11 +02:00
TEST_DECLARE (socket_buffer_size)
unix: retrieve execve() errors in process.c Make the forked child process send the errno to its parent process when it fails to spawn a new process.
2012-08-08 18:28:16 +02:00
TEST_DECLARE (spawn_fails)
unix: reap child on execvp() failure Reap the forked process when execvp() fails. Fixes leaking zombie child processes. PR-URL: https://github.com/libuv/libuv/pull/274 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl> Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2015-03-17 10:01:53 -05:00
#ifndef _WIN32
TEST_DECLARE (spawn_fails_check_for_waitpid_cleanup)
#endif
Start uv_spawn() Unix only at the moment. Lacks test for stdio
2011-07-20 17:55:26 -07:00
TEST_DECLARE (spawn_exit_code)
Add spawn_stdout test
2011-07-25 17:11:23 -07:00
TEST_DECLARE (spawn_stdout)
Windows: spawn child processes
2011-07-29 18:30:07 -07:00
TEST_DECLARE (spawn_stdin)
test: test for uv_spawn with stdio_count == 3
2012-05-31 22:56:23 +04:00
TEST_DECLARE (spawn_stdio_greater_than_3)
test: stdin/out/err should be set to /dev/null when ignored
2012-06-02 21:34:21 +02:00
TEST_DECLARE (spawn_ignored_stdio)
Windows: spawn child processes
2011-07-29 18:30:07 -07:00
TEST_DECLARE (spawn_and_kill)
process: implement UV_PROCESS_DETACHED flag for uv_spawn
2012-05-31 17:19:27 -07:00
TEST_DECLARE (spawn_detached)
test: process should not wait for stdio before terminating
2011-11-23 07:35:45 -08:00
TEST_DECLARE (spawn_and_kill_with_std)
test: uv_process_kill(proc, 0) should not kill the process
2011-10-11 21:00:56 +02:00
TEST_DECLARE (spawn_and_ping)
unix: do not set environ unless one is provided Currently, `uv_spawn` will set `environ` to the value of `options.env`, even if `options.env` is `NULL`. This results in child processes for whom `environ == NULL`, which can cause a variety of unexpected issues.
2012-11-01 10:55:06 -04:00
TEST_DECLARE (spawn_preserve_env)
Test for the uv_spawn's setuid and setgid options
2012-04-27 21:05:15 +02:00
TEST_DECLARE (spawn_setuid_fails)
TEST_DECLARE (spawn_setgid_fails)
windows: allow specifying FDs to be inherited by a child process Previously the only option was to create a pipe or an ipc channel. This patch makes it possible to inherit a handle that is already open in the parent process. There is also room for setting more than just stdin, stdout and stderr, although this is not supported yet.
2012-05-15 11:26:16 -07:00
TEST_DECLARE (spawn_stdout_to_file)
process: close stdio after dup2'ing it Thus allow passing the same file descriptor as the source of multiple stdios. Committed with the help and code parts from: * Sam Roberts <sam@strongloop.com> fix #1074
2014-01-20 19:30:08 +04:00
TEST_DECLARE (spawn_stdout_and_stderr_to_file)
test: spawn child replacing std{out,err} to stderr Refs: https://github.com/libuv/libuv/pull/224 PR-URL: https://github.com/libuv/libuv/pull/309 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
2015-03-20 09:50:01 +01:00
TEST_DECLARE (spawn_stdout_and_stderr_to_file2)
unix: fix swapping fds order in uv_spawn Alternative implementation (and test) to https://github.com/libuv/libuv/pull/226 Fixes: https://github.com/joyent/libuv/issues/1084 PR-URL: https://github.com/libuv/libuv/pull/309 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
2015-04-08 00:17:12 +02:00
TEST_DECLARE (spawn_stdout_and_stderr_to_file_swap)
unix: auto-unref spawn handle on process exit Consistent, for better or worse, with uv-win. Fixes #718.
2013-02-22 14:39:39 +01:00
TEST_DECLARE (spawn_auto_unref)
process: don't close stdio fds during spawn This is needed when closed stdio fd is reused for uv_spawn pipe. Fixes #1211
2014-03-23 12:48:06 +02:00
TEST_DECLARE (spawn_closed_process_io)
windows: read the PATH env var of the child The unix and windows process implementations diverge in their behavior when dealing with subprocesses that are spawned with a relative path. With unix the *child's* PATH environment variable is read, whereas with windows the *parent's* environment variable is read. This commit brings the two implementation in line with respect to their behavior of reading PATH by having both read the *child's* PATH environment variable. This involves looking into the user-provided environment on windows and extracting the PATH variable specifically so it can be inspected later on.
2014-06-24 15:39:01 -07:00
TEST_DECLARE (spawn_reads_child_path)
win: remove UV_HANDLE_CONNECTED UV_HANDLE_CONNECTED was defined, but never used anywhere - outside this if condition inside uv__stdio_create. So this test can't be true. UV_HANDLE_CONNECTION was meant. A test was also added verifying the behaviour. PR-URL: https://github.com/libuv/libuv/pull/404 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2015-06-19 16:30:24 +00:00
TEST_DECLARE (spawn_inherit_streams)
unix, windows: add stat() based file watcher Monitors a file path for changes. Supersedes ev_stat.
2012-06-14 07:30:08 +02:00
TEST_DECLARE (fs_poll)
unix, windows: add uv_fs_poll_getpath
2014-02-24 22:51:59 +01:00
TEST_DECLARE (fs_poll_getpath)
uv_kill
2011-11-02 14:05:22 -07:00
TEST_DECLARE (kill)
fs: add UV_ENOENT error code, add test
2011-09-06 00:16:41 +02:00
TEST_DECLARE (fs_file_noent)
unix: map `ENAMETOOLONG` to `UV_ENAMETOOLONG` With tests. Closes #295
2012-01-22 02:01:54 +01:00
TEST_DECLARE (fs_file_nametoolong)
test: test if `UV_ELOOP` mapping works
2012-01-25 12:39:44 +01:00
TEST_DECLARE (fs_file_loop)
uv_fs & uv_work APIs
2011-08-19 16:46:42 -07:00
TEST_DECLARE (fs_file_async)
TEST_DECLARE (fs_file_sync)
unix: fix uv_fs_write when using an empty buffer
2014-02-26 23:59:29 +01:00
TEST_DECLARE (fs_file_write_null_buffer)
uv_fs & uv_work APIs
2011-08-19 16:46:42 -07:00
TEST_DECLARE (fs_async_dir)
TEST_DECLARE (fs_async_sendfile)
unix, windows: add uv_fs_mkdtemp
2014-07-13 22:42:14 +03:00
TEST_DECLARE (fs_mkdtemp)
Add test for uv_fs_fstat, implement on unix.
2011-09-01 14:15:01 -07:00
TEST_DECLARE (fs_fstat)
unix, windows: add uv_fs_access()
2014-10-02 22:16:21 -04:00
TEST_DECLARE (fs_access)
add test fs_chmod, implement uv_fs_fchmod and uv_fs_chmod on unix
2011-09-01 15:36:01 -07:00
TEST_DECLARE (fs_chmod)
win: fix unlink of readonly files Unix unlinks files even if they are not writable, windows version should behave similarly. Fixes: https://github.com/joyent/node/issues/3006 PR-URL: https://github.com/libuv/libuv/pull/327 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2015-04-21 19:44:30 +01:00
TEST_DECLARE (fs_unlink_readonly)
windows: implement missing fs functions
2011-09-03 11:06:08 -07:00
TEST_DECLARE (fs_chown)
windows: uv_fs_link + uv_fs_symlink
2011-09-03 16:31:35 -07:00
TEST_DECLARE (fs_link)
test: add uv_fs_readlink test
2012-09-30 03:26:48 +02:00
TEST_DECLARE (fs_readlink)
fs: add uv_fs_realpath() Equivalent to realpath(3), returns the full resolved absolute path of a file or directory. PR-URL: https://github.com/libuv/libuv/pull/531 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2015-09-19 00:40:34 +03:00
TEST_DECLARE (fs_realpath)
windows: uv_fs_link + uv_fs_symlink
2011-09-03 16:31:35 -07:00
TEST_DECLARE (fs_symlink)
windows: support junctions with uv_fs_synlink based on @piscisaureus implementation
2012-05-08 15:44:11 -07:00
TEST_DECLARE (fs_symlink_dir)
unix: implement uv_fs_futime, add tests for uv_fs_utime and uv_fs_futime
2011-09-06 02:52:23 +02:00
TEST_DECLARE (fs_utime)
TEST_DECLARE (fs_futime)
windows: honor O_APPEND in uv_fs_open
2011-11-04 12:41:23 -07:00
TEST_DECLARE (fs_file_open_append)
windows: fix error reporting for uv_fs_ functions
2011-09-28 12:46:37 -07:00
TEST_DECLARE (fs_stat_missing_path)
test: fs: add tests for read EOF This fix was merged without tests: https://github.com/philips/libuv/tree/fix-read-on-windows-to-handle-eof So take tests from igorzi: https://github.com/igorzi/libuv/commit/46024bf33dcc7fc922378fe0d8b4f59f4c2cd605
2012-02-28 17:27:18 -08:00
TEST_DECLARE (fs_read_file_eof)
windows: file watcher
2011-09-16 12:04:36 -07:00
TEST_DECLARE (fs_event_watch_dir)
win: add support for recursive file watching Original patch by @ghostoy, modified by @bpasero and yours truly. Refs: https://github.com/joyent/libuv/pull/1473 Refs: https://github.com/libuv/libuv/pull/198 PR-URL: https://github.com/libuv/libuv/pull/421 Reviewed-By: Bert Belder <bertbelder@gmail.com>
2015-07-02 01:04:11 +02:00
TEST_DECLARE (fs_event_watch_dir_recursive)
windows: file watcher
2011-09-16 12:04:36 -07:00
TEST_DECLARE (fs_event_watch_file)
win: compare entire filename in watch events This commit causes Window file watching events to compare the entire file path when filtering events. This fixes a bug where incomplete path comparisons would cause invalid events to be raised. Refs: https://github.com/libuv/libuv/pull/682 PR-URL: https://github.com/libuv/libuv/pull/924 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2016-06-22 13:14:04 -04:00
TEST_DECLARE (fs_event_watch_file_exact_path)
test: add failing fs_event test Watches the same file twice. Fails on Linux with a segmentation fault.
2012-07-28 14:40:01 +02:00
TEST_DECLARE (fs_event_watch_file_twice)
windows: file watcher
2011-09-16 12:04:36 -07:00
TEST_DECLARE (fs_event_watch_file_current_dir)
win: fix watching root files When passing "\\?\C:" to CreateFile, it opens the drive rather than the root directory. So include the trailing backslash in the directory name. Fixes: https://github.com/nodejs/node/issues/4643 PR-URL: https://github.com/libuv/libuv/pull/689 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2016-01-14 19:21:47 +02:00
#ifdef _WIN32
TEST_DECLARE (fs_event_watch_file_root_dir)
#endif
win,sunos: stop handle on uv_fs_event_start() err This commit stops the handle in uv_fs_event_start() when an error occurs. Fixes: https://github.com/libuv/libuv/issues/1253 PR-URL: https://github.com/libuv/libuv/pull/1257 Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
2017-03-16 21:20:11 -04:00
TEST_DECLARE (fs_event_watch_invalid_path)
sunos: properly disarm PORT_LOADED fsevent watcher Fixes a segmentation fault when the watched entity is changed after the watcher has been closed.
2012-12-12 19:47:58 -08:00
TEST_DECLARE (fs_event_no_callback_after_close)
windows: don't emit fs-event callback after uv_fs_event handle is closed
2011-11-02 19:09:18 -07:00
TEST_DECLARE (fs_event_no_callback_on_close)
unix: properly disarm kqueue fs watcher Fixes "Assertion failed: (revents == EV_LIBUV_KQUEUE_HACK), function uv__fs_event, file ../src/unix/kqueue.c, line 58."
2011-09-27 15:04:45 +02:00
TEST_DECLARE (fs_event_immediate_close)
Test: verify that no callbacks are made after closing an fs event watcher
2012-03-27 22:56:34 +02:00
TEST_DECLARE (fs_event_close_with_pending_event)
Test for the uv_spawn's setuid and setgid options
2012-04-27 21:05:15 +02:00
TEST_DECLARE (fs_event_close_in_callback)
fsevents: fix clever rescheduling There're could be a situation, where one fsevents handle gets created and another one is destroyed simultaneously. In such cases `fsevent_need_reschedule` will be set to 1 twice and reset only once, leaving handle destructor hanging in uv_sem_wait().
2013-10-02 14:47:14 +04:00
TEST_DECLARE (fs_event_start_and_close)
test: test fsevents error reporting Create a lot of `FSEventStream`s to make `FSEventStreamStart()` fail. Fetch the error code and verify that it is correct.
2013-11-01 23:00:58 +04:00
TEST_DECLARE (fs_event_error_reporting)
unix, windows: add uv_fs_event_getpath
2014-02-24 00:51:13 +01:00
TEST_DECLARE (fs_event_getpath)
fs: rename uv_fs_readdir to uv_fs_scandir Closes #1431
2014-09-25 10:09:58 +02:00
TEST_DECLARE (fs_scandir_empty_dir)
fs: fix crash in uv_fs_scandir_next Handle the case when it's called after the request has ended, or in case it ended with an error. PR-URL: https://github.com/libuv/libuv/pull/1181 Reviewed-By: Colin Ihrig <cjihrig@gmail.com> Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com> Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2017-01-05 15:16:48 -08:00
TEST_DECLARE (fs_scandir_non_existent_dir)
fs: rename uv_fs_readdir to uv_fs_scandir Closes #1431
2014-09-25 10:09:58 +02:00
TEST_DECLARE (fs_scandir_file)
windows: enable uv_fs_open to open directories
2011-10-12 10:36:14 -07:00
TEST_DECLARE (fs_open_dir)
windows: uv_fs_rename to replace the new file if it exists fixes #283
2011-12-30 15:46:17 -08:00
TEST_DECLARE (fs_rename_to_existing_file)
windows: fix fs_write with nbufs > 1 and offset When multiple bufs are specified, overlapped needs to be advanced manually between each write. Without this, each buf will be written to the same position (if offset is specified.) ReadFile() automatically advances, but WriteFile() does not.
2014-10-19 20:48:51 -07:00
TEST_DECLARE (fs_write_multiple_bufs)
unix,win: return EINVAL on nullptr args in uv_fs_{read,write} PR-URL: https://github.com/libuv/libuv/pull/470 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2015-08-11 12:38:57 +02:00
TEST_DECLARE (fs_read_write_null_arguments)
unix: allow nbufs > IOV_MAX in uv_fs_{read,write} This allows writing and reading any amount of buffers, regardless of what IOV_MAX may be defined as. It also moves the IOV_MAX test from stream to core. This is based on the excellent work of @bwijen in #269. Refs: https://github.com/libuv/libuv/pull/269 PR-URL: https://github.com/libuv/libuv/pull/448 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2015-08-04 17:07:54 +09:00
TEST_DECLARE (fs_write_alotof_bufs)
TEST_DECLARE (fs_write_alotof_bufs_with_offset)
uv_fs & uv_work APIs
2011-08-19 16:46:42 -07:00
TEST_DECLARE (threadpool_queue_work_simple)
unix, windows: disallow NULL worker in thread pool See #629.
2012-11-24 08:25:32 +09:00
TEST_DECLARE (threadpool_queue_work_einval)
test: eio callbacks should run in their owning threads
2011-12-22 03:42:54 +01:00
TEST_DECLARE (threadpool_multiple_event_loops)
unix: add uv_cancel()
2012-11-26 03:12:38 +01:00
TEST_DECLARE (threadpool_cancel_getaddrinfo)
unix: allow uv_getnameinfo to be cancelled
2014-06-18 02:00:07 +02:00
TEST_DECLARE (threadpool_cancel_getnameinfo)
unix: add uv_cancel()
2012-11-26 03:12:38 +01:00
TEST_DECLARE (threadpool_cancel_work)
TEST_DECLARE (threadpool_cancel_fs)
unix: ensure done_cb gets called after uv_cancel() Wake up the event loop with uv_async_send() when a request is cancelled. Ensures the done_cb is run on the next tick of the event loop. Not sending a wakeup signal results in the done_cb not getting called until another request completes, which may be a long time coming when it's the only request in the queue or when other requests are executing long-running jobs. Fixes #669.
2012-12-28 15:10:04 +01:00
TEST_DECLARE (threadpool_cancel_single)
unix, windows: add thread-local storage API Uses the pthread_key_{create,delete} and pthread_{get,set}specific functions on UNIX platforms, Tls{Alloc,Free} and Tls{Get,Set}Value on Windows. Fixes #904.
2013-08-25 03:12:41 +02:00
TEST_DECLARE (thread_local_storage)
osx: set the default thread stack size to RLIMIT_STACK Fixes: https://github.com/libuv/libuv/issues/669 PR-URL: https://github.com/libuv/libuv/pull/671 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
2016-01-04 12:32:22 +01:00
TEST_DECLARE (thread_stack_size)
Wrap platform mutex and rwlock APIs. Read/write locks are emulated with critical sections on Windows XP and Vista because those platforms don't have a (complete) native read/write lock API.
2011-11-14 21:04:53 +01:00
TEST_DECLARE (thread_mutex)
TEST_DECLARE (thread_rwlock)
test: add tests for the uv_rwlock implementation PR-URL: https://github.com/libuv/libuv/pull/525 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2015-09-13 16:58:40 +02:00
TEST_DECLARE (thread_rwlock_trylock)
Wrap platform thread APIs.
2011-11-21 16:12:36 +01:00
TEST_DECLARE (thread_create)
unix, windows: add uv_thread_equal
2014-10-04 22:01:03 -07:00
TEST_DECLARE (thread_equal)
win, unix: add uv_dlerror() and uv_dlerror_free()
2012-02-25 01:32:31 +09:00
TEST_DECLARE (dlerror)
Test: add tests for uv_poll
2012-04-25 00:26:21 +02:00
TEST_DECLARE (poll_duplex)
TEST_DECLARE (poll_unidirectional)
test: verify that closing an active poll handle doesn't hang
2012-06-01 18:30:17 +02:00
TEST_DECLARE (poll_close)
unix: report errors for unpollable fds Libuv would abort() when trying to watch a file descriptor that is not compatible with epoll-style polling; file descriptors referring to on-disk files fall into this category. File descriptors that libuv creates itself are not an issue but external ones that come in through the uv_poll_init() API are. Make uv_poll_init() check whether the file descriptor is accepted by the underlying system call and return an error when it's not. Fixes: https://github.com/libuv/libuv/issues/658 PR-URL: https://github.com/libuv/libuv/pull/659 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2015-12-20 02:29:11 +01:00
TEST_DECLARE (poll_bad_fdtype)
unix: allow nesting of kqueue fds in uv_poll_start kqueue file descriptors don't support ioctl(FIONBIO) (or any other ioctl for that matter) so retry using fcntl(F_GETFL) + fcntl(F_SETFL) when we receive a ENOTTY error. Fixes: https://github.com/libuv/libuv/issues/883 PR-URL: https://github.com/libuv/libuv/pull/885 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2016-05-23 20:03:30 +02:00
#ifdef __linux__
TEST_DECLARE (poll_nested_epoll)
#endif
#ifdef UV_HAVE_KQUEUE
TEST_DECLARE (poll_nested_kqueue)
#endif
Merge branch 'v0.10' Conflicts: ChangeLog build.mk src/unix/darwin.c src/unix/getaddrinfo.c src/version.c test/test-list.h
2014-06-25 10:29:53 +02:00
unix, windows: detect errors in uv_ip4/6_addr
2013-12-13 01:01:38 +08:00
TEST_DECLARE (ip4_addr)
linux,darwin,win: link-local IPv6 addresses Modified uv_ip6_addr() to fill sin6_scope_id for link-local addresses. Fixes #271 Conflicts: build.mk
2013-03-14 22:20:32 +01:00
TEST_DECLARE (ip6_addr_link_local)
Merge branch 'v0.10' Conflicts: ChangeLog build.mk src/unix/darwin.c src/unix/getaddrinfo.c src/version.c test/test-list.h
2014-06-25 10:29:53 +02:00
test: test that closing a poll handle doesn't corrupt the stack This is a regression test for an issue that was originally reported in https://github.com/libuv/libuv/pull/36, and fixed in cd89452. PR-URL: https://github.com/libuv/libuv/pull/48 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2014-12-09 22:01:16 +01:00
TEST_DECLARE (poll_close_doesnt_corrupt_stack)
windows: fix handling closed socket while poll handle is closing fixes #1278 (backport from master for v0.10 branch)
2014-05-12 16:48:52 +02:00
TEST_DECLARE (poll_closesocket)
test: silence build warnings PR-URL: https://github.com/libuv/libuv/pull/893 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl> Reviewed-By: Colin Ihrig <cjihrig@gmail.com> Reviewed-By: Johan Bergström <bugs@bergstroem.nu> Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2016-05-30 08:36:36 +00:00
#ifdef _WIN32
windows: remove dependency on rpcrt4 and ole32 libs. fixes https://github.com/joyent/libuv/issues/118
2011-08-01 20:10:55 -07:00
TEST_DECLARE (spawn_detect_pipe_name_collisions_on_windows)
win,test: fix shared library build Ifdef out the tests that rely on internal symbols when making a shared build. PR-URL: https://github.com/libuv/libuv/pull/444 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
2015-07-15 23:04:22 +02:00
#if !defined(USING_UV_SHARED)
Support for unescaped arguments, suitable for use with cmd /c. Robust argument escaping that hopefully matches Windows' algorithm for unescaping.
2011-08-05 00:14:17 +01:00
TEST_DECLARE (argument_escaping)
Include important Windows environmental variables even when a blank/custom environment is specified.
2011-08-09 04:14:06 +01:00
TEST_DECLARE (environment_creation)
win,test: fix shared library build Ifdef out the tests that rely on internal symbols when making a shared build. PR-URL: https://github.com/libuv/libuv/pull/444 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
2015-07-15 23:04:22 +02:00
#endif
windows: add tests for uv_tcp_simultaneous_accepts
2011-11-01 01:13:13 -07:00
TEST_DECLARE (listen_with_simultaneous_accepts)
TEST_DECLARE (listen_no_simultaneous_accepts)
windows: fix stat("c:\\") regression
2012-01-27 18:15:10 +01:00
TEST_DECLARE (fs_stat_root)
windows: Fix an infinite loop in uv_spawn The PATH-parsing code for windows erroneously contained an infinite loop when the PATH started with a leading semicolon. Each iteration of the loop usually bumped over the separator, but if the first character was a semicolon then it would never skip it, causing the infinite loop. Closes #909
2014-03-31 14:16:07 -07:00
TEST_DECLARE (spawn_with_an_odd_path)
windows: relay TCP bind errors via ipc This is the libuv side of the fix for Node's cluster module on Windows. https://github.com/joyent/node/issues/7691 Windows and Unix return certain socket errors (i.e. EADDRINUSE) at different times: bind on Windows, and listen on Unix. In an effort to hide this difference, libuv on Windows stores such errors in the bind_error field of uv_tcp_t, to defer raising it at listen time. This worked fine except for the case in which a socket is shared in a Node cluster and a bind error occurs. A previous attempt to fix this ( https://github.com/joyent/libuv/commit/d1e6be1460f555a1f8a4063d7642696aa7238769 https://github.com/joyent/node/commit/3da36fe00e5d85414031ae812e473f16629d8645 ) was flawed becaused in an attempt to relay the error at the JS level it caused the master to start accepting connections. With this new approach, libuv itself is relaying the bind errors, providing for a uniform behavior of uv_tcp_listen.
2014-07-29 16:08:01 +02:00
TEST_DECLARE (ipc_listen_after_bind_twice)
win,signal: fix potential deadlock Trying to remove the controller from the controller handle itself leads to deadlock. Fix it by always having the global signal handler engaged. Ref: https://github.com/nodejs/node/issues/10165 PR-URL: https://github.com/libuv/libuv/pull/1168 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2016-12-12 11:34:42 +01:00
TEST_DECLARE (win32_signum_number)
Test for the uv_spawn's setuid and setgid options
2012-04-27 21:05:15 +02:00
#else
unix: fix accept() EMFILE error handling Fix a bug that was introduced in commit 3ee4d3f ("unix, windows: return error codes directly") and add a regression test for good measure. Hat tip to Fedor for pointing out the issue. Fixes #1007.
2013-11-25 01:30:11 +01:00
TEST_DECLARE (emfile)
linux: handle EPOLLHUP without EPOLLIN/EPOLLOUT Work around an epoll quirk where it sometimes reports just the EPOLLERR or EPOLLHUP event. In order to force the event loop to move forward, we merge in the read/write events that the watcher is interested in; uv__read() and uv__write() will then deal with the error or hangup in the usual fashion. Fixes #982.
2013-11-08 05:15:55 +01:00
TEST_DECLARE (close_fd)
test: build spawn_fs_open only on unixes
2014-02-05 13:35:37 +04:00
TEST_DECLARE (spawn_fs_open)
Test for the uv_spawn's setuid and setgid options
2012-04-27 21:05:15 +02:00
TEST_DECLARE (spawn_setuid_setgid)
windows: don't run signal tests They are dummy tests on windows, running them is misleading. Also remove duplicate 'test-signal.c' entry from uv.gyp.
2012-10-12 00:11:27 +02:00
TEST_DECLARE (we_get_signal)
TEST_DECLARE (we_get_signals)
signal: add uv_signal_start_oneshot method It behaves as `uv_signal_start` but it resets the signal handler as soon as the signal is received. Fixes: https://github.com/libuv/libuv/issues/1104 PR-URL: https://github.com/libuv/libuv/pull/1106 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl> Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com> Reviewed-By: Bartosz Sosnowski <bartosz@janeasystems.com>
2016-10-23 14:47:31 +02:00
TEST_DECLARE (we_get_signal_one_shot)
TEST_DECLARE (we_get_signals_mixed)
unix: support signal handlers outside the main loop
2012-10-16 17:42:30 +02:00
TEST_DECLARE (signal_multiple_loops)
linux: always deregister closing fds from epoll If the same file description is open in two different processes, then closing the file descriptor is not sufficient to deregister it from the epoll instance (as described in epoll(7)), resulting in spurious events that cause the event loop to spin repeatedly. So always explicitly deregister it. Fixes #1099.
2014-02-05 18:36:24 -05:00
TEST_DECLARE (closed_fd_events)
windows: remove dependency on rpcrt4 and ole32 libs. fixes https://github.com/joyent/libuv/issues/118
2011-08-01 20:10:55 -07:00
#endif
stream: fix small nit in select hack, add test
2013-04-26 23:43:28 +04:00
#ifdef __APPLE__
TEST_DECLARE (osx_select)
darwin: allocate enough space for select() hack `fd_set`s are way too small for `select()` hack when stream's fd is bigger than 1023. Make `fd_set`s a part of `uv__stream_select_t` structure. fix #1461
2014-09-07 23:56:15 +01:00
TEST_DECLARE (osx_select_many_fds)
stream: fix small nit in select hack, add test
2013-04-26 23:43:28 +04:00
#endif
Named pipes implementation for Windows
2011-07-01 17:54:17 -07:00
HELPER_DECLARE (tcp4_echo_server)
HELPER_DECLARE (tcp6_echo_server)
test: add udp4_echo_server helper
2012-01-13 19:15:50 +01:00
HELPER_DECLARE (udp4_echo_server)
Named pipes implementation for Windows
2011-07-01 17:54:17 -07:00
HELPER_DECLARE (pipe_echo_server)
First shot at test harness
2011-04-05 01:43:17 +02:00
test: test that uv_close() doesn't corrupt QUEUE The test adds all types of handles, that were known to be affected by QUEUE_REMOVE()-within-QUEUE_FOREACH() bug, to a loop. It then calls uv_close() to trigger QUEUE_REMOVE()-within-QUEUE_FOREACH() case and checks whether a particular QUEUE is corrupted or not. Restrict the test to Linux only for now as it fails on other platforms for various reasons. PR-URL: https://github.com/libuv/libuv/pull/621 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2015-09-30 12:47:02 -04:00
TEST_DECLARE (queue_foreach_delete)
Misc. cleanups
2011-06-25 12:35:49 +02:00
unix: make loops and watchers usable after fork() Added the uv_loop_fork() API that must be called in a child process to continue using an existing loop. Internally this calls a uv__io_fork function for each supported platform, similar to the way uv__platform_loop_init works. After this call, existing and new IO, async and signal watchers will contiue working as before on all platforms, as will the threadpool (although any threads it was using are of course gone). On Linux and BSDs that use kqueue, existing and new fsevent watchers will also continue to work as expected. On OS X, though, directory fsevents will not be able to use the optimized CoreFoundation path if they had already been used in the parent process, instead falling back to the kqueue path used on other BSDs. Existing fsevent watchers will not function on AIX or SunOS. This could be relatively easily fixed by someone with AIX knowledge in the future, but SunOS will require some additional work to keep track if the watchers. A new test file, test/test-fork.c, was added to contain fork-related tests to verify functionality in the child process. PR-URL: https://github.com/libuv/libuv/pull/846 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl> Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2016-04-22 14:16:31 -05:00
#ifndef _WIN32
TEST_DECLARE (fork_timer)
TEST_DECLARE (fork_socketpair)
TEST_DECLARE (fork_socketpair_started)
TEST_DECLARE (fork_signal_to_child)
TEST_DECLARE (fork_signal_to_child_closed)
TEST_DECLARE (fork_fs_events_child)
TEST_DECLARE (fork_fs_events_child_dir)
TEST_DECLARE (fork_fs_events_file_parent_child)
TEST_DECLARE (fork_threadpool_queue_work_simple)
#endif
Benchmark runner
2011-04-19 04:26:32 +02:00
TASK_LIST_START
test: support flexibly setting custom task options Add a single TEST_ENTRY_CUSTOM hook that can be used to override task entry defaults. Different tests can have different timeouts depending on what is appropriate for each test. A separate TEST_OUTPUT_ENTRY hook is no longer necessary. In order to support per-task timeouts, the timeout field has been moved into the task_entry_t struct. The default (5000) is now set as part of TEST_ENTRY.
2014-01-31 10:57:30 -06:00
TEST_ENTRY_CUSTOM (platform_output, 0, 1, 5000)
platform api
2011-12-13 01:29:27 -08:00
Disable test-callback-order It's not correct.
2012-05-22 17:13:04 +02:00
#if 0
test: add callback order test Ensure that idle callbacks run before other callbacks.
2012-05-22 16:04:09 +02:00
TEST_ENTRY (callback_order)
Disable test-callback-order It's not correct.
2012-05-22 17:13:04 +02:00
#endif
windows: run close callbacks after polling for i/o Makes the uv-win behavior compatible with uv-unix. Fixes #796.
2013-05-29 09:51:04 +02:00
TEST_ENTRY (close_order)
unix, windows: make uv_run_once() return a bool The return value of uv_run_once() now signals if it needs to be called again. Fixes #427.
2012-05-23 20:30:59 +02:00
TEST_ENTRY (run_once)
Implemented uv_run2 Allows for running the event loop in 3 modes: * default: loop runs until the refcount drops to zero * once: poll for events only once and block until one is handled * nowait: poll for events only once but don't block if there are no pending events
2012-12-18 16:10:11 +01:00
TEST_ENTRY (run_nowait)
unix, windows: add uv_loop_alive() function Useful to know when the the event loop is empty, this can't be done with uv_run() without possibly blocking, or running some events (which might empty the event loop as a side-effect).
2013-05-17 13:29:26 -07:00
TEST_ENTRY (loop_alive)
unix, windows: add uv_loop_init and uv_loop_close These functions supersede uv_loop_new and uv_loop_delete. uv_loop_init initialized a user allocated loop and uv_loop_close removes all associated resources a loop uses after it has finished execution. uv_loop_new and uv_loop_delete are now deprecated.
2014-02-17 22:37:20 +01:00
TEST_ENTRY (loop_close)
unix, windows: add uv_stop, stop running event loop
2013-01-16 09:36:20 +01:00
TEST_ENTRY (loop_stop)
unix, windows: always update loop time Fixes #846
2013-12-10 09:13:34 +01:00
TEST_ENTRY (loop_update_time)
windows: add uv_backend_timeout
2014-07-31 21:35:32 -04:00
TEST_ENTRY (loop_backend_timeout)
linux: fix epoll_pwait() sigmask size calculation Revisit the fix from commit b705b53. The problem with using sigset_t and _NSIG is that the size of sigset_t and the value of _NSIG depend on what headers libuv picks up first, <signal.h> or <asm/signal.h>. With the former, sizeof(sigset_t) = 128; with the latter, it's 8. Simply sidestep the issue by calculating the signal mask as a 64 bits integer, without using sigset_t or _NSIG. PR-URL: https://github.com/libuv/libuv/pull/83 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2014-12-25 14:11:55 +01:00
TEST_ENTRY (loop_configure)
test: add test for closing and recreating default loop
2014-09-08 21:47:50 +02:00
TEST_ENTRY (default_loop_close)
unix, windows: add thread barrier support
2012-10-09 21:44:20 +09:00
TEST_ENTRY (barrier_1)
TEST_ENTRY (barrier_2)
TEST_ENTRY (barrier_3)
Add support for condition variables on all platforms
2012-10-05 13:03:03 +02:00
TEST_ENTRY (condvar_1)
TEST_ENTRY (condvar_2)
TEST_ENTRY (condvar_3)
TEST_ENTRY (condvar_4)
TEST_ENTRY (condvar_5)
unix, windows: add semaphore functions
2012-06-03 03:17:21 +02:00
TEST_ENTRY (semaphore_1)
TEST_ENTRY (semaphore_2)
TEST_ENTRY (semaphore_3)
Disable test-callback-order It's not correct.
2012-05-22 17:13:04 +02:00
Fix UNIX pipe connect error reporting, add test
2011-11-03 15:46:36 -07:00
TEST_ENTRY (pipe_connect_bad_name)
windows: return UV_ENOTSOCK when doing uv_pipe_connect to a file
2012-03-01 12:11:12 -08:00
TEST_ENTRY (pipe_connect_to_file)
unix: don't block for io if any io handle is primed PR-URL: https://github.com/libuv/libuv/pull/347 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
2015-05-08 09:33:47 +02:00
TEST_ENTRY (pipe_connect_on_prepare)
Fix UNIX pipe connect error reporting, add test
2011-11-03 15:46:36 -07:00
test: add pipe-server-close test Add a regression test for the pipe server close issue on Windows, which was reported in joyent/node#6749 and fixed in 7b16a3f.
2013-12-20 19:40:02 -08:00
TEST_ENTRY (pipe_server_close)
unix: read on stream even when UV__POLLHUP set. This fixes a SmartOS specific issue that happens when reading from a stream that is the reading end of a pipe that has been closed by the parent process. In this case, a UV__POLLHUP event would be set on the stream and would prevent the event loop from closing it. As a result, the event loop would think there are stil handles open, and leave the process hanging. Fixes #1419.
2014-08-14 16:38:56 -07:00
#ifndef _WIN32
TEST_ENTRY (pipe_close_stdout_read_stdin)
#endif
unix: implement uv_stream_set_blocking() Commit 393c1c5 ("unix: set non-block mode in uv_{pipe,tcp,udp}_open") causes a regression in the io.js cluster module. The io.js documentation states that `worker.send()` and `process.send()` are synchronous but they no longer were after upgrading to libuv v1.2.1. The reason they are synchronous is because of backpressure - or rather, lack of backpressure: a slow consumer eventually causes a fast producer to run out of memory because the backlog of pending messages in the producer can grow unchecked. Ergo, implement uv_stream_set_blocking() on UNIX platforms to let io.js enable the old blocking behavior for pipes again. Refs: https://github.com/iojs/io.js/issues/760 PR-URL: https://github.com/libuv/libuv/pull/187 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2015-02-09 12:53:52 +01:00
TEST_ENTRY (pipe_set_non_blocking)
Add uv_is_tty()
2011-09-15 15:16:53 -07:00
TEST_ENTRY (tty)
win,tty: fix read stop in line mode Closing the handle does not make ReadConsoleW exit reliably on Windows 7 and above. Thus, after switching from line to raw mode, keypresses were held until enter was pressed. This makes ReadConsoleW exit by writing a return keypress to its input buffer, similar to what was already done for raw mode. Fixes: https://github.com/libuv/libuv/issues/852 PR-URL: https://github.com/libuv/libuv/pull/866 Reviewed-by: Bert Belder <bertbelder@gmail.com> Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2016-05-11 13:03:30 +01:00
#ifdef _WIN32
TEST_ENTRY (tty_raw)
Revert "Revert "win,tty: add support for ANSI codes in win10 v1511"" To bring back support for ANSI codes in win10 v1511. This reverts commit 8cbabaa81709562de68bcac0554e4e7af2f72cf6. PR-URL: https://github.com/libuv/libuv/pull/1143 Refs: https://github.com/libuv/libuv/pull/1138 Refs: https://github.com/libuv/libuv/pull/889 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com> Reviewed-By: Imran Iqbal <imran@imraniqbal.org>
2016-11-16 16:16:54 +01:00
TEST_ENTRY (tty_empty_write)
win,tty: fix MultiByteToWideChar output buffer Make sure there's enough room in the output buffer by dynamically allocating memory in case the size of the buffer needs to be greater than 8192 characters. PR-URL: https://github.com/libuv/libuv/pull/1143 Refs: https://github.com/libuv/libuv/pull/1138 Refs: https://github.com/libuv/libuv/pull/889 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com> Reviewed-By: Imran Iqbal <imran@imraniqbal.org>
2016-11-16 16:17:12 +01:00
TEST_ENTRY (tty_large_write)
win,tty: fix read stop in line mode Closing the handle does not make ReadConsoleW exit reliably on Windows 7 and above. Thus, after switching from line to raw mode, keypresses were held until enter was pressed. This makes ReadConsoleW exit by writing a return keypress to its input buffer, similar to what was already done for raw mode. Fixes: https://github.com/libuv/libuv/issues/852 PR-URL: https://github.com/libuv/libuv/pull/866 Reviewed-by: Bert Belder <bertbelder@gmail.com> Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2016-05-11 13:03:30 +01:00
#endif
unix: reject non-tty fds in uv_tty_init() Reject file descriptors that correspond to regular files or char/block devices. Such file descriptors are incompatible with epoll and trigger a run-time assert. Fixes: https://github.com/libuv/libuv/issues/255 PR-URL: https://github.com/libuv/libuv/pull/259 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2015-03-11 16:05:56 +01:00
TEST_ENTRY (tty_file)
unix: open ttyname instead of /dev/tty Find the real name of the tty using ttyname_r(3) instead of opening "/dev/tty" which causes trouble if the fd doesn't point to the controlling terminal. PR-URL: https://github.com/libuv/libuv/pull/779 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl> Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2016-03-23 23:18:53 +01:00
TEST_ENTRY (tty_pty)
windows: stdio over non-overlapped pipes
2011-10-19 00:48:38 -07:00
TEST_ENTRY (stdio_over_pipes)
inet: allow scopeid in uv_inet_pton We already support it in `uv_ip6_addr` anyway. See https://github.com/joyent/node/issues/7395
2014-04-14 15:22:08 +04:00
TEST_ENTRY (ip6_pton)
windows: when sharing a server socket, only call listen in the parent process
2011-10-25 13:07:29 -07:00
TEST_ENTRY (ipc_listen_before_write)
TEST_ENTRY (ipc_listen_after_write)
Disable ipc_send_recv_pipe test on Windows It's not supported yet
2012-04-12 03:33:53 +02:00
#ifndef _WIN32
test: test cross-process handle send/recv
2012-03-09 03:16:08 +01:00
TEST_ENTRY (ipc_send_recv_pipe)
pipe: enable inprocess uv_write2 on Windows When duplicating the socket handle being sent the target process id is defaulted to the current process id. This enables uv_write2 to be used for thread-clustering in addition to process-clustering on multi-threaded programming languages. The ipc tests are updated to run in two modes. In the _inproc mode the echo work is done on a second thread instead of in a second process. An internal function int uv_current_pid() is added to the windows specific code which caches the value of GetCurrentProcessId(). This means uv_write2 does not call GetCurrentProcessId() every inprocess send. Refs: https://github.com/joyent/libuv/issues/926 PR-URL: https://github.com/libuv/libuv/pull/540 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2015-10-28 11:40:51 -07:00
TEST_ENTRY (ipc_send_recv_pipe_inprocess)
Disable ipc_send_recv_pipe test on Windows It's not supported yet
2012-04-12 03:33:53 +02:00
#endif
test: test cross-process handle send/recv
2012-03-09 03:16:08 +01:00
TEST_ENTRY (ipc_send_recv_tcp)
pipe: enable inprocess uv_write2 on Windows When duplicating the socket handle being sent the target process id is defaulted to the current process id. This enables uv_write2 to be used for thread-clustering in addition to process-clustering on multi-threaded programming languages. The ipc tests are updated to run in two modes. In the _inproc mode the echo work is done on a second thread instead of in a second process. An internal function int uv_current_pid() is added to the windows specific code which caches the value of GetCurrentProcessId(). This means uv_write2 does not call GetCurrentProcessId() every inprocess send. Refs: https://github.com/joyent/libuv/issues/926 PR-URL: https://github.com/libuv/libuv/pull/540 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2015-10-28 11:40:51 -07:00
TEST_ENTRY (ipc_send_recv_tcp_inprocess)
windows: Enable passing of TCP connections over IPC
2012-03-08 12:02:50 -08:00
TEST_ENTRY (ipc_tcp_connection)
Add tcp reference count tests
2011-09-26 22:51:08 -07:00
unix,win: make on_alloc_cb failures more resilient Treat both the base being NULL or the length being 0 as ENOBUFS. PR-URL: https://github.com/libuv/libuv/pull/997 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
2016-08-16 00:15:54 +02:00
TEST_ENTRY (tcp_alloc_cb_fail)
Named pipes implementation for Windows
2011-07-01 17:54:17 -07:00
TEST_ENTRY (tcp_ping_pong)
TEST_HELPER (tcp_ping_pong, tcp4_echo_server)
First shot at test harness
2011-04-05 01:43:17 +02:00
Named pipes implementation for Windows
2011-07-01 17:54:17 -07:00
TEST_ENTRY (tcp_ping_pong_v6)
TEST_HELPER (tcp_ping_pong_v6, tcp6_echo_server)
TEST_ENTRY (pipe_ping_pong)
TEST_HELPER (pipe_ping_pong, pipe_echo_server)
win: IPv6 connect
2011-07-01 17:04:05 -07:00
Test for not immediately calling oio_accept from accept_cb Closes #10.
2011-04-22 00:45:49 +02:00
TEST_ENTRY (delayed_accept)
windows: don't fail uv_listen with UV_EALREADY if already listening. fixes #227.
2011-10-26 13:32:15 -07:00
TEST_ENTRY (multiple_listen)
Test for not immediately calling oio_accept from accept_cb Closes #10.
2011-04-22 00:45:49 +02:00
unix: fix tcp write after bad connect freezing If the connect wouldn't go off (no such tcp remote or any other failure), the subsequent writes made would not be called. Now we call the writes in the queue with ECANCELED if the connect fails. Fix #1432
2014-08-21 10:59:35 +02:00
#ifndef _WIN32
TEST_ENTRY (tcp_write_after_connect)
#endif
Write-a-lot test
2011-04-22 05:12:10 +02:00
TEST_ENTRY (tcp_writealot)
Named pipes implementation for Windows
2011-07-01 17:54:17 -07:00
TEST_HELPER (tcp_writealot, tcp4_echo_server)
Write-a-lot test
2011-04-22 05:12:10 +02:00
test: add test for deferred TCP write failure The test exhibits that write failures are deferred to the callback. Refs: https://github.com/libuv/libuv/issues/339 PR_URL: https://github.com/libuv/libuv/pull/344
2015-05-05 23:09:23 +02:00
TEST_ENTRY (tcp_write_fail)
TEST_HELPER (tcp_write_fail, tcp4_echo_server)
stream: introduce uv_try_write(...) `uv_try_write(stream, buf, size)` acts like `uv_write()`, but without queueing actual write until UV_POLLOUT (or IOCP completion). This is useful for doing writes using on-stack `uv_write_t` requests. fix #1025
2013-12-06 19:10:47 +04:00
TEST_ENTRY (tcp_try_write)
unix: guarantee write queue cb execution order in streams The problem was that the canceled write callbacks were being called before the already written callbacks, even though they were queued after them.
2014-07-06 14:43:59 +02:00
TEST_ENTRY (tcp_write_queue_order)
unix, windows: add uv_tcp_open and uv_udp_open
2012-09-17 12:15:12 +02:00
TEST_ENTRY (tcp_open)
TEST_HELPER (tcp_open, tcp4_echo_server)
unix, win: prevent replacing fd in uv_{udp,tcp,pipe}_t PR-URL: https://github.com/libuv/libuv/pull/400 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
2015-06-18 00:41:44 +02:00
TEST_ENTRY (tcp_open_twice)
unix, windows: add uv_tcp_open and uv_udp_open
2012-09-17 12:15:12 +02:00
unix: fix event loop stall Stop libev from entering the epoll_wait/kevent/port_getn/etc. syscall when there are no active handles left, that will block indefinitely.
2012-06-04 23:18:32 +02:00
TEST_ENTRY (tcp_shutdown_after_write)
TEST_HELPER (tcp_shutdown_after_write, tcp4_echo_server)
test: rename test tcp_dont_connect_after_write Rename test tcp_dont_connect_after_write to tcp_connect_error_after_write, teensy tad more descriptive.
2012-06-04 17:18:02 +02:00
TEST_ENTRY (tcp_connect_error_after_write)
Add uv_pipe_bind tests
2011-07-21 02:00:12 +02:00
TEST_ENTRY (tcp_bind_error_addrinuse)
TEST_ENTRY (tcp_bind_error_addrnotavail_1)
TEST_ENTRY (tcp_bind_error_addrnotavail_2)
TEST_ENTRY (tcp_bind_error_fault)
TEST_ENTRY (tcp_bind_error_inval)
TEST_ENTRY (tcp_bind_localhost_ok)
unix, windows: validate flags on uv_udp|tcp_bind fixes #1247
2014-04-22 22:45:58 +02:00
TEST_ENTRY (tcp_bind_invalid_flags)
Add uv_pipe_bind tests
2011-07-21 02:00:12 +02:00
TEST_ENTRY (tcp_listen_without_bind)
test: enable tcp_connect_error_fault test
2012-05-17 11:44:05 +02:00
TEST_ENTRY (tcp_connect_error_fault)
test: add tcp connect timeout test
2012-05-17 11:43:33 +02:00
TEST_ENTRY (tcp_connect_timeout)
test: add tcp 'close on failed connect' test Demonstrates temporary event loop stall with uv-unix. The issue is that pending requests aren't processed until the next event (I/O, timeout, etc.) happens. See #446, #447 and #448.
2012-06-06 16:00:53 +02:00
TEST_ENTRY (tcp_close_while_connecting)
unix: bring back uv__stream_destroy() This is the revised version of reverted commit 431195c.
2011-09-09 23:53:09 +02:00
TEST_ENTRY (tcp_close)
unix, win: add ability to create tcp/udp sockets early Introduce two new APIs: int uv_tcp_init_ex(uv_loop_t*, uv_tcp_t* handle, int flags) int uv_udp_init_ex(uv_loop_t*, uv_udp_t* handle, int flags) The lower 8 bits of the flags field are used for the socket domain. AF_INET, AF_INET6 and AF_UNSPEC are supported. If AF_UNSPEC is specified the socket is created lazily, just like uv_{tcp,udp}_init. Some Windows notes: getsockname fails with WSAEINVAL if the socket is not bound. This could potentially be improved by detecting the socket family and filling the sockaddr_in/6 struct manually. bind returns WSAEFAULT if we try to bind a socket to the wrong family. Unix returns EINVAL. PR-URL: https://github.com/libuv/libuv/pull/400 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
2015-06-17 00:34:58 +02:00
TEST_ENTRY (tcp_create_early)
TEST_ENTRY (tcp_create_early_bad_bind)
TEST_ENTRY (tcp_create_early_bad_domain)
TEST_ENTRY (tcp_create_early_accept)
test: disable tcp_close_accept on Windows It always was a Unix only test
2014-08-05 00:14:26 +02:00
#ifndef _WIN32
unix: fix reopened fd bug When fd is closed and new one (with the same number) is opened inside kqueue/epoll/port loop's callback - stale events might invoke callbacks on wrong watchers. Check if watcher was changed after invocation and invalidate all events with the same fd. fix #826
2013-10-25 12:56:37 +04:00
TEST_ENTRY (tcp_close_accept)
stream: use SO_OOBINLINE on OS X In the collaboration with Ben Noordhuis <info@bnoordhuis.nl> and Saúl Ibarra Corretgé <saghul@gmail.com>. Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com> Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
2015-02-24 16:16:22 +03:00
TEST_ENTRY (tcp_oob)
test: disable tcp_close_accept on Windows It always was a Unix only test
2014-08-05 00:14:26 +02:00
#endif
unix: add TCP keepalive and no-delay control knobs
2011-10-21 15:37:47 -07:00
TEST_ENTRY (tcp_flags)
windows: convert WSAECONNABORTED to EOF
2011-11-07 15:48:33 -08:00
TEST_ENTRY (tcp_write_to_half_open_connection)
test: add 'unexpected read' tcp test Regression test that verifies that the event loop doesn't busy loop when the server sends a message and the client isn't reading.
2012-06-30 02:59:29 +02:00
TEST_ENTRY (tcp_unexpected_read)
Add uv_pipe_bind tests
2011-07-21 02:00:12 +02:00
unix: fix event loop stall Fix a rather obscure bug where the event loop stalls when an I/O watcher is stopped while an artificial event, generated with uv__io_feed(), is pending.
2012-12-13 16:37:12 +01:00
TEST_ENTRY (tcp_read_stop)
TEST_HELPER (tcp_read_stop, tcp4_echo_server)
Add uv_pipe_bind tests
2011-07-21 02:00:12 +02:00
TEST_ENTRY (tcp_bind6_error_addrinuse)
TEST_ENTRY (tcp_bind6_error_addrnotavail)
TEST_ENTRY (tcp_bind6_error_fault)
TEST_ENTRY (tcp_bind6_error_inval)
TEST_ENTRY (tcp_bind6_localhost_ok)
unix,win: make on_alloc_cb failures more resilient Treat both the base being NULL or the length being 0 as ENOBUFS. PR-URL: https://github.com/libuv/libuv/pull/997 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
2016-08-16 00:15:54 +02:00
TEST_ENTRY (udp_alloc_cb_fail)
unix, windows: do not set SO_REUSEADDR by default on udp Add UV_UDP_REUSEADDR flag instead, which can be passed to uv_udp_bind. If the udp handle is unbound when uv_udp_set_memberhsip or uv_udp_set_multicast_interface is called, the handle will be bound with UV_UDP_REUSEADDR set.
2014-03-31 09:56:37 +02:00
TEST_ENTRY (udp_bind)
TEST_ENTRY (udp_bind_reuseaddr)
unix, win: add ability to create tcp/udp sockets early Introduce two new APIs: int uv_tcp_init_ex(uv_loop_t*, uv_tcp_t* handle, int flags) int uv_udp_init_ex(uv_loop_t*, uv_udp_t* handle, int flags) The lower 8 bits of the flags field are used for the socket domain. AF_INET, AF_INET6 and AF_UNSPEC are supported. If AF_UNSPEC is specified the socket is created lazily, just like uv_{tcp,udp}_init. Some Windows notes: getsockname fails with WSAEINVAL if the socket is not bound. This could potentially be improved by detecting the socket family and filling the sockaddr_in/6 struct manually. bind returns WSAEFAULT if we try to bind a socket to the wrong family. Unix returns EINVAL. PR-URL: https://github.com/libuv/libuv/pull/400 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
2015-06-17 00:34:58 +02:00
TEST_ENTRY (udp_create_early)
TEST_ENTRY (udp_create_early_bad_bind)
TEST_ENTRY (udp_create_early_bad_domain)
Add UDP support to libuv.
2011-08-19 23:22:30 +02:00
TEST_ENTRY (udp_send_and_recv)
test: add test for uv_udp_send() fix Test that the fast path will continue writing. On failure, this test will timeout. PR-URL: https://github.com/libuv/libuv/pull/1308 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com> Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
2017-04-18 17:59:32 -06:00
TEST_ENTRY (udp_send_hang_loop)
unix: remove incorrect assert After 41891222bc landed it's possible that uv__udp_sendmsg is called even if there are no pending write nor write completed requests: 1. User calls uv_udp_send and the request is sent immediately. The request is the added to the completed queue and we 'feed' the uv__io handle so that we process the completed request in the next iteration. 2. User calls uv_udp_send again but the request is not completed immediately, so it's queued in the write_queue. 3. The uv__io handle gets a UV__POLLOUT event and uv__udp_sendmsg is run, which completes the send request and puts it in the write_completed_queue. Afterwards, uv__udp_run_completed is executed and the write_completed queue is drained. 4. At this point, the uv__io handle was made pending in step 3, in uv__udp_sendmsg, but we no longer have requests to write or to complete, so we skip processing.
2014-07-11 22:42:55 +02:00
TEST_ENTRY (udp_send_immediate)
windows: fix buffer leak after failed udp send Fixes #1426 Signed-off-by: Saúl Ibarra Corretgé <saghul@gmail.com>
2014-08-27 21:50:48 +02:00
TEST_ENTRY (udp_send_unreachable)
Add UDP support to libuv.
2011-08-19 23:22:30 +02:00
TEST_ENTRY (udp_dgram_too_big)
TEST_ENTRY (udp_dual_stack)
TEST_ENTRY (udp_ipv6_only)
test: add UDP broadcast/multicast/ttl tests
2012-01-28 02:16:00 +01:00
TEST_ENTRY (udp_options)
unix: fix ttl, multicast ttl and loop options on IPv6 Fixes #93 PR-URL: https://github.com/libuv/libuv/pull/99 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
2015-01-01 13:15:57 +01:00
TEST_ENTRY (udp_options6)
unix, windows: be consistent when binding implictly in UDP Only these functions will trigger an implicit binding of a UDP handle: - uv_udp_send - uv_udp_recv_start - uv_udp_set_membership All other functions will return UV_EBADF in case the socket was not bound. Note: currently the socket is created and bound at the same time. This may change in the future.
2014-05-07 11:12:19 +02:00
TEST_ENTRY (udp_no_autobind)
unix, windows: add IPv6 support for uv_udp_multicast_interface
2014-03-28 00:39:09 +01:00
TEST_ENTRY (udp_multicast_interface)
TEST_ENTRY (udp_multicast_interface6)
add uv_udp_set_membership for unix multicast support - test-udp-multicast-join tests that multicast packets can be received - stub src/win/udp.c until support added
2011-09-29 19:29:15 -04:00
TEST_ENTRY (udp_multicast_join)
unix, windows: add IPv6 support to uv_udp_set_membership Reworked from initial version by @snoj
2014-03-17 10:06:25 +01:00
TEST_ENTRY (udp_multicast_join6)
test: add multicast TTL test
2012-01-19 11:07:28 -05:00
TEST_ENTRY (udp_multicast_ttl)
unix, win: add uv_udp_try_send The function returns UV_EAGAIN if the queue is not empty and the message couldn't be sent immediately.
2014-06-30 00:01:10 +02:00
TEST_ENTRY (udp_try_send)
Add UDP support to libuv.
2011-08-19 23:22:30 +02:00
unix, windows: add uv_tcp_open and uv_udp_open
2012-09-17 12:15:12 +02:00
TEST_ENTRY (udp_open)
TEST_HELPER (udp_open, udp4_echo_server)
unix, win: prevent replacing fd in uv_{udp,tcp,pipe}_t PR-URL: https://github.com/libuv/libuv/pull/400 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
2015-06-18 00:41:44 +02:00
TEST_ENTRY (udp_open_twice)
unix, windows: add uv_tcp_open and uv_udp_open
2012-09-17 12:15:12 +02:00
Add uv_pipe_bind tests
2011-07-21 02:00:12 +02:00
TEST_ENTRY (pipe_bind_error_addrinuse)
TEST_ENTRY (pipe_bind_error_addrnotavail)
TEST_ENTRY (pipe_bind_error_inval)
win: fix replacing pipe handle for pipe servers On Windows we create multiple pipe handles (system handles) which are attached to pending accept requests. Each of these will take turns in replacing the reference in handle->handle, so make sure we allow for that **only** for pipe servers. PR-URL: https://github.com/libuv/libuv/pull/488 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
2015-08-20 10:49:21 +02:00
TEST_ENTRY (pipe_connect_multiple)
Add uv_pipe_bind tests
2011-07-21 02:00:12 +02:00
TEST_ENTRY (pipe_listen_without_bind)
unix, windows: add uv_pipe_getsockname
2013-12-17 00:15:54 +01:00
TEST_ENTRY (pipe_getsockname)
TEST_ENTRY (pipe_getsockname_abstract)
windows: make uv_read_stop immediately stop reading This implements locking around the blocking call to ReadFile to get around a Windows kernel bug where a blocking ReadFile operation on a stream can deadlock the thread. This allows uv_read_stop to immediately cancel a pending IO operation, and allows uv_pipe_getsockname to "pause" any pending read (from libuv) while it retrieves the sockname information. If unsupported by the OS (pre-Vista), this reverts to the old (e.g. deadlock-prone) behavior Closes #1313
2014-07-15 01:49:11 -04:00
TEST_ENTRY (pipe_getsockname_blocking)
win: fix setting pipe pending instances after bind It's an error to do so, so just ignore it. The test would cause an invalid memory access if the fix is undone. PR-URL: https://github.com/libuv/libuv/pull/488 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
2015-08-20 10:53:20 +02:00
TEST_ENTRY (pipe_pending_instances)
pipe: allow queueing pending handles Introduce `int uv_pipe_pending_count(uv_pipe_t*)` and `uv_handle_type uv_pipe_pending_type(uv_pipe_t*)`. They should be used in IPC pipe's read cb to accept incoming handles: int count = uv_pipe_pending_count(pipe); int i; for (i = 0; i < count; i++) { uv_handle_type type = uv_pipe_pending_type(pipe); /* ... */ uv_accept(...); }
2013-12-22 14:45:00 +04:00
TEST_ENTRY (pipe_sendmsg)
IPv6 support Fixes #70. Fixes #19.
2011-06-26 22:34:04 -06:00
Test for not immediately calling oio_accept from accept_cb Closes #10.
2011-04-22 00:45:49 +02:00
TEST_ENTRY (connection_fail)
API Change: uv_close only called by user - never automatically Add test that failing on_connect callback does not trigger on_close.
2011-06-02 09:19:56 +02:00
TEST_ENTRY (connection_fail_doesnt_auto_close)
Test for not immediately calling oio_accept from accept_cb Closes #10.
2011-04-22 00:45:49 +02:00
Tests: verify that shutdown_cb is always called
2012-03-09 04:29:45 +01:00
TEST_ENTRY (shutdown_close_tcp)
TEST_HELPER (shutdown_close_tcp, tcp4_echo_server)
TEST_ENTRY (shutdown_close_pipe)
TEST_HELPER (shutdown_close_pipe, pipe_echo_server)
Add test to assert behavior of uv_tcp_t after shutdown Handles should *never* call the close callback without the user having first issued a uv_close() command. This test is broken on windows. See #52.
2011-06-02 11:31:42 +02:00
TEST_ENTRY (shutdown_eof)
Named pipes implementation for Windows
2011-07-01 17:54:17 -07:00
TEST_HELPER (shutdown_eof, tcp4_echo_server)
Add test to assert behavior of uv_tcp_t after shutdown Handles should *never* call the close callback without the user having first issued a uv_close() command. This test is broken on windows. See #52.
2011-06-02 11:31:42 +02:00
unix: correct error when calling uv_shutdown twice Before this one of the requests was left as a ghost in the system. See https://github.com/JuliaLang/julia/issues/5793
2014-02-14 22:27:45 -05:00
TEST_ENTRY (shutdown_twice)
TEST_HELPER (shutdown_twice, tcp4_echo_server)
More extensive test-callback-stack. Fails on windows now.
2011-05-04 17:10:33 +02:00
TEST_ENTRY (callback_stack)
Named pipes implementation for Windows
2011-07-01 17:54:17 -07:00
TEST_HELPER (callback_stack, tcp4_echo_server)
Add ol_close_cb stack test and fix it for Windows
2011-04-05 04:24:03 +02:00
unix,win: add uv_os_{get,set,unset}env() These functions are used to create, retrieve, update, and delete environment variables. Fixes: https://github.com/libuv/libuv/issues/1198 PR-URL: https://github.com/libuv/libuv/pull/1234 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com> Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
2017-02-28 00:21:26 -05:00
TEST_ENTRY (env_vars)
error reporting: return sensible messages for synthetic errors
2011-10-06 01:00:56 +02:00
TEST_ENTRY (error_message)
unix,win: add uv_translate_sys_error() public API uv_translate_sys_error() was a private function for Windows. This commit adds an equivalent function on other platforms, and exposes it as public API. Exposing this is useful in scenarios where the application uses both libuv functions and platform-specific system calls and wants to report errors uniformly as libuv errors. Fixes: https://github.com/libuv/libuv/issues/79 PR-URL: https://github.com/libuv/libuv/pull/1060 Reviewed-By: Colin Ihrig <cjihrig@gmail.com> Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2016-09-18 15:18:48 -04:00
TEST_ENTRY (sys_error)
error reporting: return sensible messages for synthetic errors
2011-10-06 01:00:56 +02:00
Update tests & benchmarks to use the new timer api
2011-05-17 01:17:48 +02:00
TEST_ENTRY (timer)
unix: make uv_timer_init() initialize repeat uv_timer_get_repeat() should return 0 for timers that haven't been started.
2013-03-04 22:17:50 -06:00
TEST_ENTRY (timer_init)
Test for uv_timer_again ,~get_repeat, ~set_repeat
2011-05-19 01:58:07 +02:00
TEST_ENTRY (timer_again)
unix: update timer if already active uv_timer_start() no longer returns an error when the timer is already active, now it just updates the timer. Consistent with the uv-win implementation. Fixes #425.
2012-05-26 02:09:35 +02:00
TEST_ENTRY (timer_start_twice)
unix,windows: fix timer order in case of same timeout Compare start_id of timer handles when they have the same timeout. start_id is allocated with loop->timer_counter in uv_timer_start.
2013-02-07 13:59:17 +09:00
TEST_ENTRY (timer_order)
unix: make timers handle large timeouts This commit fixes two closely related integer overflow bugs: * Timers with a timeout > INT_MAX cause uv__next_timeout() to return a negative value. * Timers with very large timeouts (close or equal to ULLONG_MAX) run on the next tick. In both cases, clamp the values to prevent the overflow from happening. Fixes joyent/node#5101.
2013-03-21 14:47:38 +01:00
TEST_ENTRY (timer_huge_timeout)
TEST_ENTRY (timer_huge_repeat)
unix, windows: run expired timers in run-once mode Before this commit, creating an event loop, starting a timer and calling uv_run(UV_RUN_ONCE) blocked in uv_run() until the timer expired - but didn't actually run the timer.
2013-05-25 10:37:08 +02:00
TEST_ENTRY (timer_run_once)
test: add 'start timer from check handle' test Check that a timer that is started from a check handle gets picked up correctly, i.e. that it influences the timeout used in the next tick of the event loop.
2013-06-26 13:02:39 +02:00
TEST_ENTRY (timer_from_check)
unix, windows: don't allow a NULL callback on timers
2014-09-04 14:17:59 +02:00
TEST_ENTRY (timer_null_callback)
win: wait for full timeout duration uv_poll should wait for at least the full timeout duration when there is nothing else to do. This was not happening because GetQueuedCompletionStatus can occasionally return up to 15ms early. The added test reproduces https://github.com/joyent/node/blob/d13d7f74d794340ac5e126cfb4ce507fe0f803d5/test/simple/test-timers-first-fire.js on libuv, being flaky before this fix. Fixes: https://github.com/joyent/node/issues/8960 PR-URL: https://github.com/libuv/libuv/pull/385 Reviewed-By: Alexis Campailla <alexis@janeasystems.com> Reviewed-by: Bert Belder <bertbelder@gmail.com>
2015-06-02 20:02:55 +01:00
TEST_ENTRY (timer_early_check)
Test for uv_timer_again ,~get_repeat, ~set_repeat
2011-05-19 01:58:07 +02:00
Test: active idle watcher should not block other events
2011-07-21 02:51:26 +02:00
TEST_ENTRY (idle_starvation)
Add reference count tests Broken on UNIX.
2011-05-20 08:22:19 -07:00
TEST_ENTRY (ref)
TEST_ENTRY (idle_ref)
unix, windows: add stat() based file watcher Monitors a file path for changes. Supersedes ev_stat.
2012-06-14 07:30:08 +02:00
TEST_ENTRY (fs_poll_ref)
Add reference count tests Broken on UNIX.
2011-05-20 08:22:19 -07:00
TEST_ENTRY (async_ref)
TEST_ENTRY (prepare_ref)
TEST_ENTRY (check_ref)
Test that loop refs going down to zero in a prepare callback does not hang the event loop
2011-07-21 14:53:40 +02:00
TEST_ENTRY (unref_in_prepare_cb)
test: add lots of refcount tests
2012-01-14 00:11:27 +01:00
TEST_ENTRY (timer_ref)
TEST_ENTRY (timer_ref2)
TEST_ENTRY (fs_event_ref)
TEST_ENTRY (tcp_ref)
TEST_ENTRY (tcp_ref2)
windows: closing handles should always keep the loop alive This makes the tcp-ref2 and udp-ref2 tests pass again. Also adds another reference count test.
2012-10-25 15:10:16 +02:00
TEST_ENTRY (tcp_ref2b)
test: add lots of refcount tests
2012-01-14 00:11:27 +01:00
TEST_ENTRY (tcp_ref3)
TEST_HELPER (tcp_ref3, tcp4_echo_server)
Tests: verify that uv_write and uv_shutdown ref the event loop
2012-03-09 04:26:51 +01:00
TEST_ENTRY (tcp_ref4)
TEST_HELPER (tcp_ref4, tcp4_echo_server)
test: add lots of refcount tests
2012-01-14 00:11:27 +01:00
TEST_ENTRY (udp_ref)
TEST_ENTRY (udp_ref2)
TEST_ENTRY (udp_ref3)
TEST_HELPER (udp_ref3, udp4_echo_server)
TEST_ENTRY (pipe_ref)
TEST_ENTRY (pipe_ref2)
TEST_ENTRY (pipe_ref3)
TEST_HELPER (pipe_ref3, pipe_echo_server)
Tests: verify that uv_write and uv_shutdown ref the event loop
2012-03-09 04:26:51 +01:00
TEST_ENTRY (pipe_ref4)
TEST_HELPER (pipe_ref4, pipe_echo_server)
test: add lots of refcount tests
2012-01-14 00:11:27 +01:00
TEST_ENTRY (process_ref)
unix, windows: add uv_has_ref() function
2013-04-17 00:16:42 +02:00
TEST_ENTRY (has_ref)
Add reference count tests Broken on UNIX.
2011-05-20 08:22:19 -07:00
Test for oio_prepare/oio_check/oio_idle
2011-05-08 04:44:55 +02:00
TEST_ENTRY (loop_handles)
unix, windows: add uv_walk() Lets the libuv user iterate over the open handles. Mostly intended as a debugging tool or a post-hoc cleanup mechanism.
2012-05-30 01:51:28 +02:00
TEST_ENTRY (walk_handles)
Test for oio_prepare/oio_check/oio_idle
2011-05-08 04:44:55 +02:00
test: add regression test for 29fdb3471
2013-10-30 21:43:23 +04:00
TEST_ENTRY (watcher_cross_stop)
test: add test for uv_is_active and uv_is_closing
2012-08-27 23:57:48 +02:00
TEST_ENTRY (active)
unix: add uv_backend_fd() and uv_backend_timeout() This can be used in conjuction with uv_run_once() to poll in one thread and run the event loop's event callbacks in another. Useful for embedding libuv's event loop in another event loop.
2012-11-26 14:20:11 +04:00
TEST_ENTRY (embed)
Add test-async
2011-05-09 04:29:22 +02:00
TEST_ENTRY (async)
unix, windows: allow NULL async callback Allow a NULL callback so the user doesn't have to provide a dummy when the actual event is processed by e.g. a check handle callback.
2013-08-12 08:42:49 +02:00
TEST_ENTRY (async_null_cb)
fs: don't nullify req->bufs on EINTR uv__fs_buf_iter currently sets req->bufs to NULL after it is done, but if the operation fails with EINTR then it will be retried, at which point it expects the bufs to not be NULL, causing a seg fault as in https://github.com/nodejs/node/issues/4291. uv__fs_buf_iter should not set req->bufs to NULL if the operation fails with EINTR. Also, when it sets req->bufs to NULL, it should set req->nbufs to 0 as well, so we don't have the messy situation of a positive nbufs with no actual bufs. PR-URL: https://github.com/libuv/libuv/pull/661 Reviewed-By: Fedor Indutny <fedor@indutny.com>
2015-12-19 22:21:07 -08:00
TEST_ENTRY (eintr_handling)
Add test-async
2011-05-09 04:29:22 +02:00
Adds uv_get_exepath API Only works on Linux, Mac, Windows currently.
2011-05-20 18:05:37 -07:00
TEST_ENTRY (get_currentexe)
platform api
2011-12-13 01:29:27 -08:00
TEST_ENTRY (process_title)
uv_cwd + uv_chdir
2011-11-30 15:31:39 -08:00
TEST_ENTRY (cwd_and_chdir)
os: implement memory bindings * us_get_free_memory * us_get_total_memory
2011-10-01 15:45:42 +07:00
TEST_ENTRY (get_memory)
unix,win: add uv_get_passwd() This commit adds the uv_get_passwd() function, which returns a subset of the current effective user's password file entry. Refs: https://github.com/libuv/libuv/issues/11 Fixes: https://github.com/libuv/libuv/issues/731 PR-URL: https://github.com/libuv/libuv/pull/742 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl> Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2016-03-01 15:41:01 -05:00
TEST_ENTRY (get_passwd)
os: implement loadavg (not working on cygwin/win)
2011-10-02 00:48:42 +07:00
TEST_ENTRY (get_loadavg)
unix, windows: add uv_fileno Returns the platform specific file descriptor for handles that are backed by one. The datatype is abstracted as uv_os_fd_t, which maps to int on Unices and HANDLE on Windows. Users can use this function to set specific socket options, for example, in a non portable way. This function is essentially a shotgun, you better be careful with whatever you do with it, don't blame me if you used it to get the fd of a stream, close it yourself and expect things to Just Work.
2014-08-22 09:57:43 +02:00
TEST_ENTRY (handle_fileno)
unix,win: add uv_os_homedir() PR-URL: https://github.com/libuv/libuv/pull/350 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl> Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2015-05-12 09:17:19 -04:00
TEST_ENTRY (homedir)
unix,win: add uv_os_tmpdir() PR-URL: https://github.com/libuv/libuv/pull/672 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2016-01-04 16:52:44 -05:00
TEST_ENTRY (tmpdir)
Add test for uv_get_hrtime
2011-06-17 11:47:35 +02:00
TEST_ENTRY (hrtime)
Separate uv_getaddrinfo tests
2011-06-25 15:33:52 +02:00
test: set getaddrinfo_fail timeout to 10000 getaddrinfo_fail can sometimes take slightly longer than the default timeout of 5000. Set its timeout to 10000 to compensate, but leave the timeout of all other tests at the default.
2014-02-01 00:50:23 +04:00
TEST_ENTRY_CUSTOM (getaddrinfo_fail, 0, 0, 10000)
unix, win: add synchronous uv_get{addr,name}info PR-URL: https://github.com/libuv/libuv/pull/156 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl> Reviewed-By: Bert Belder <bertbelder@gmail.com>
2015-01-20 09:43:57 +01:00
TEST_ENTRY (getaddrinfo_fail_sync)
test: set getaddrinfo_fail timeout to 10000 getaddrinfo_fail can sometimes take slightly longer than the default timeout of 5000. Set its timeout to 10000 to compensate, but leave the timeout of all other tests at the default.
2014-02-01 00:50:23 +04:00
Separate uv_getaddrinfo tests
2011-06-25 15:33:52 +02:00
TEST_ENTRY (getaddrinfo_basic)
unix, win: add synchronous uv_get{addr,name}info PR-URL: https://github.com/libuv/libuv/pull/156 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl> Reviewed-By: Bert Belder <bertbelder@gmail.com>
2015-01-20 09:43:57 +01:00
TEST_ENTRY (getaddrinfo_basic_sync)
Separate uv_getaddrinfo tests
2011-06-25 15:33:52 +02:00
TEST_ENTRY (getaddrinfo_concurrent)
Add test for uv_get_hrtime
2011-06-17 11:47:35 +02:00
unix, windows: getnameinfo implementation Closes #852
2014-05-11 12:51:08 +02:00
TEST_ENTRY (getnameinfo_basic_ip4)
unix, win: add synchronous uv_get{addr,name}info PR-URL: https://github.com/libuv/libuv/pull/156 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl> Reviewed-By: Bert Belder <bertbelder@gmail.com>
2015-01-20 09:43:57 +01:00
TEST_ENTRY (getnameinfo_basic_ip4_sync)
unix, windows: getnameinfo implementation Closes #852
2014-05-11 12:51:08 +02:00
TEST_ENTRY (getnameinfo_basic_ip6)
test: add udp getsockname test
2011-08-25 02:30:10 +02:00
TEST_ENTRY (getsockname_tcp)
TEST_ENTRY (getsockname_udp)
uv_getsockname
2011-07-11 13:45:16 -07:00
Test: add tests for uv_poll
2012-04-25 00:26:21 +02:00
TEST_ENTRY (poll_duplex)
TEST_ENTRY (poll_unidirectional)
test: verify that closing an active poll handle doesn't hang
2012-06-01 18:30:17 +02:00
TEST_ENTRY (poll_close)
unix: report errors for unpollable fds Libuv would abort() when trying to watch a file descriptor that is not compatible with epoll-style polling; file descriptors referring to on-disk files fall into this category. File descriptors that libuv creates itself are not an issue but external ones that come in through the uv_poll_init() API are. Make uv_poll_init() check whether the file descriptor is accepted by the underlying system call and return an error when it's not. Fixes: https://github.com/libuv/libuv/issues/658 PR-URL: https://github.com/libuv/libuv/pull/659 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2015-12-20 02:29:11 +01:00
TEST_ENTRY (poll_bad_fdtype)
unix: allow nesting of kqueue fds in uv_poll_start kqueue file descriptors don't support ioctl(FIONBIO) (or any other ioctl for that matter) so retry using fcntl(F_GETFL) + fcntl(F_SETFL) when we receive a ENOTTY error. Fixes: https://github.com/libuv/libuv/issues/883 PR-URL: https://github.com/libuv/libuv/pull/885 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2016-05-23 20:03:30 +02:00
#ifdef __linux__
TEST_ENTRY (poll_nested_epoll)
#endif
#ifdef UV_HAVE_KQUEUE
TEST_ENTRY (poll_nested_kqueue)
#endif
Test: add tests for uv_poll
2012-04-25 00:26:21 +02:00
unix, windows: add uv_recv_buffer_size and uv_send_buffer_size
2014-07-07 14:56:11 +02:00
TEST_ENTRY (socket_buffer_size)
unix: retrieve execve() errors in process.c Make the forked child process send the errno to its parent process when it fails to spawn a new process.
2012-08-08 18:28:16 +02:00
TEST_ENTRY (spawn_fails)
unix: reap child on execvp() failure Reap the forked process when execvp() fails. Fixes leaking zombie child processes. PR-URL: https://github.com/libuv/libuv/pull/274 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl> Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2015-03-17 10:01:53 -05:00
#ifndef _WIN32
TEST_ENTRY (spawn_fails_check_for_waitpid_cleanup)
#endif
Start uv_spawn() Unix only at the moment. Lacks test for stdio
2011-07-20 17:55:26 -07:00
TEST_ENTRY (spawn_exit_code)
Add spawn_stdout test
2011-07-25 17:11:23 -07:00
TEST_ENTRY (spawn_stdout)
Windows: spawn child processes
2011-07-29 18:30:07 -07:00
TEST_ENTRY (spawn_stdin)
test: test for uv_spawn with stdio_count == 3
2012-05-31 22:56:23 +04:00
TEST_ENTRY (spawn_stdio_greater_than_3)
test: stdin/out/err should be set to /dev/null when ignored
2012-06-02 21:34:21 +02:00
TEST_ENTRY (spawn_ignored_stdio)
Windows: spawn child processes
2011-07-29 18:30:07 -07:00
TEST_ENTRY (spawn_and_kill)
process: implement UV_PROCESS_DETACHED flag for uv_spawn
2012-05-31 17:19:27 -07:00
TEST_ENTRY (spawn_detached)
test: process should not wait for stdio before terminating
2011-11-23 07:35:45 -08:00
TEST_ENTRY (spawn_and_kill_with_std)
test: uv_process_kill(proc, 0) should not kill the process
2011-10-11 21:00:56 +02:00
TEST_ENTRY (spawn_and_ping)
unix: do not set environ unless one is provided Currently, `uv_spawn` will set `environ` to the value of `options.env`, even if `options.env` is `NULL`. This results in child processes for whom `environ == NULL`, which can cause a variety of unexpected issues.
2012-11-01 10:55:06 -04:00
TEST_ENTRY (spawn_preserve_env)
Test for the uv_spawn's setuid and setgid options
2012-04-27 21:05:15 +02:00
TEST_ENTRY (spawn_setuid_fails)
TEST_ENTRY (spawn_setgid_fails)
windows: allow specifying FDs to be inherited by a child process Previously the only option was to create a pipe or an ipc channel. This patch makes it possible to inherit a handle that is already open in the parent process. There is also room for setting more than just stdin, stdout and stderr, although this is not supported yet.
2012-05-15 11:26:16 -07:00
TEST_ENTRY (spawn_stdout_to_file)
process: close stdio after dup2'ing it Thus allow passing the same file descriptor as the source of multiple stdios. Committed with the help and code parts from: * Sam Roberts <sam@strongloop.com> fix #1074
2014-01-20 19:30:08 +04:00
TEST_ENTRY (spawn_stdout_and_stderr_to_file)
test: spawn child replacing std{out,err} to stderr Refs: https://github.com/libuv/libuv/pull/224 PR-URL: https://github.com/libuv/libuv/pull/309 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
2015-03-20 09:50:01 +01:00
TEST_ENTRY (spawn_stdout_and_stderr_to_file2)
unix: fix swapping fds order in uv_spawn Alternative implementation (and test) to https://github.com/libuv/libuv/pull/226 Fixes: https://github.com/joyent/libuv/issues/1084 PR-URL: https://github.com/libuv/libuv/pull/309 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
2015-04-08 00:17:12 +02:00
TEST_ENTRY (spawn_stdout_and_stderr_to_file_swap)
unix: auto-unref spawn handle on process exit Consistent, for better or worse, with uv-win. Fixes #718.
2013-02-22 14:39:39 +01:00
TEST_ENTRY (spawn_auto_unref)
process: don't close stdio fds during spawn This is needed when closed stdio fd is reused for uv_spawn pipe. Fixes #1211
2014-03-23 12:48:06 +02:00
TEST_ENTRY (spawn_closed_process_io)
windows: read the PATH env var of the child The unix and windows process implementations diverge in their behavior when dealing with subprocesses that are spawned with a relative path. With unix the *child's* PATH environment variable is read, whereas with windows the *parent's* environment variable is read. This commit brings the two implementation in line with respect to their behavior of reading PATH by having both read the *child's* PATH environment variable. This involves looking into the user-provided environment on windows and extracting the PATH variable specifically so it can be inspected later on.
2014-06-24 15:39:01 -07:00
TEST_ENTRY (spawn_reads_child_path)
win: remove UV_HANDLE_CONNECTED UV_HANDLE_CONNECTED was defined, but never used anywhere - outside this if condition inside uv__stdio_create. So this test can't be true. UV_HANDLE_CONNECTION was meant. A test was also added verifying the behaviour. PR-URL: https://github.com/libuv/libuv/pull/404 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2015-06-19 16:30:24 +00:00
TEST_ENTRY (spawn_inherit_streams)
unix, windows: add stat() based file watcher Monitors a file path for changes. Supersedes ev_stat.
2012-06-14 07:30:08 +02:00
TEST_ENTRY (fs_poll)
unix, windows: add uv_fs_poll_getpath
2014-02-24 22:51:59 +01:00
TEST_ENTRY (fs_poll_getpath)
uv_kill
2011-11-02 14:05:22 -07:00
TEST_ENTRY (kill)
unix, windows: preliminary signal handler support * a no-op on Windows for now * only supports the main loop on UNIX (again, for now)
2012-07-10 02:13:49 +02:00
test: test that closing a poll handle doesn't corrupt the stack This is a regression test for an issue that was originally reported in https://github.com/libuv/libuv/pull/36, and fixed in cd89452. PR-URL: https://github.com/libuv/libuv/pull/48 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2014-12-09 22:01:16 +01:00
TEST_ENTRY (poll_close_doesnt_corrupt_stack)
windows: fix handling closed socket while poll handle is closing fixes #1278 (backport from master for v0.10 branch)
2014-05-12 16:48:52 +02:00
TEST_ENTRY (poll_closesocket)
test: silence build warnings PR-URL: https://github.com/libuv/libuv/pull/893 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl> Reviewed-By: Colin Ihrig <cjihrig@gmail.com> Reviewed-By: Johan Bergström <bugs@bergstroem.nu> Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2016-05-30 08:36:36 +00:00
#ifdef _WIN32
windows: remove dependency on rpcrt4 and ole32 libs. fixes https://github.com/joyent/libuv/issues/118
2011-08-01 20:10:55 -07:00
TEST_ENTRY (spawn_detect_pipe_name_collisions_on_windows)
win,test: fix shared library build Ifdef out the tests that rely on internal symbols when making a shared build. PR-URL: https://github.com/libuv/libuv/pull/444 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
2015-07-15 23:04:22 +02:00
#if !defined(USING_UV_SHARED)
Support for unescaped arguments, suitable for use with cmd /c. Robust argument escaping that hopefully matches Windows' algorithm for unescaping.
2011-08-05 00:14:17 +01:00
TEST_ENTRY (argument_escaping)
Include important Windows environmental variables even when a blank/custom environment is specified.
2011-08-09 04:14:06 +01:00
TEST_ENTRY (environment_creation)
win,test: fix shared library build Ifdef out the tests that rely on internal symbols when making a shared build. PR-URL: https://github.com/libuv/libuv/pull/444 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
2015-07-15 23:04:22 +02:00
# endif
windows: add tests for uv_tcp_simultaneous_accepts
2011-11-01 01:13:13 -07:00
TEST_ENTRY (listen_with_simultaneous_accepts)
TEST_ENTRY (listen_no_simultaneous_accepts)
windows: fix stat("c:\\") regression
2012-01-27 18:15:10 +01:00
TEST_ENTRY (fs_stat_root)
windows: Fix an infinite loop in uv_spawn The PATH-parsing code for windows erroneously contained an infinite loop when the PATH started with a leading semicolon. Each iteration of the loop usually bumped over the separator, but if the first character was a semicolon then it would never skip it, causing the infinite loop. Closes #909
2014-03-31 14:16:07 -07:00
TEST_ENTRY (spawn_with_an_odd_path)
windows: relay TCP bind errors via ipc This is the libuv side of the fix for Node's cluster module on Windows. https://github.com/joyent/node/issues/7691 Windows and Unix return certain socket errors (i.e. EADDRINUSE) at different times: bind on Windows, and listen on Unix. In an effort to hide this difference, libuv on Windows stores such errors in the bind_error field of uv_tcp_t, to defer raising it at listen time. This worked fine except for the case in which a socket is shared in a Node cluster and a bind error occurs. A previous attempt to fix this ( https://github.com/joyent/libuv/commit/d1e6be1460f555a1f8a4063d7642696aa7238769 https://github.com/joyent/node/commit/3da36fe00e5d85414031ae812e473f16629d8645 ) was flawed becaused in an attempt to relay the error at the JS level it caused the master to start accepting connections. With this new approach, libuv itself is relaying the bind errors, providing for a uniform behavior of uv_tcp_listen.
2014-07-29 16:08:01 +02:00
TEST_ENTRY (ipc_listen_after_bind_twice)
win,signal: fix potential deadlock Trying to remove the controller from the controller handle itself leads to deadlock. Fix it by always having the global signal handler engaged. Ref: https://github.com/nodejs/node/issues/10165 PR-URL: https://github.com/libuv/libuv/pull/1168 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2016-12-12 11:34:42 +01:00
TEST_ENTRY (win32_signum_number)
Test for the uv_spawn's setuid and setgid options
2012-04-27 21:05:15 +02:00
#else
unix: fix accept() EMFILE error handling Fix a bug that was introduced in commit 3ee4d3f ("unix, windows: return error codes directly") and add a regression test for good measure. Hat tip to Fedor for pointing out the issue. Fixes #1007.
2013-11-25 01:30:11 +01:00
TEST_ENTRY (emfile)
linux: handle EPOLLHUP without EPOLLIN/EPOLLOUT Work around an epoll quirk where it sometimes reports just the EPOLLERR or EPOLLHUP event. In order to force the event loop to move forward, we merge in the read/write events that the watcher is interested in; uv__read() and uv__write() will then deal with the error or hangup in the usual fashion. Fixes #982.
2013-11-08 05:15:55 +01:00
TEST_ENTRY (close_fd)
test: build spawn_fs_open only on unixes
2014-02-05 13:35:37 +04:00
TEST_ENTRY (spawn_fs_open)
Test for the uv_spawn's setuid and setgid options
2012-04-27 21:05:15 +02:00
TEST_ENTRY (spawn_setuid_setgid)
windows: don't run signal tests They are dummy tests on windows, running them is misleading. Also remove duplicate 'test-signal.c' entry from uv.gyp.
2012-10-12 00:11:27 +02:00
TEST_ENTRY (we_get_signal)
TEST_ENTRY (we_get_signals)
signal: add uv_signal_start_oneshot method It behaves as `uv_signal_start` but it resets the signal handler as soon as the signal is received. Fixes: https://github.com/libuv/libuv/issues/1104 PR-URL: https://github.com/libuv/libuv/pull/1106 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl> Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com> Reviewed-By: Bartosz Sosnowski <bartosz@janeasystems.com>
2016-10-23 14:47:31 +02:00
TEST_ENTRY (we_get_signal_one_shot)
TEST_ENTRY (we_get_signals_mixed)
unix: support signal handlers outside the main loop
2012-10-16 17:42:30 +02:00
TEST_ENTRY (signal_multiple_loops)
linux: always deregister closing fds from epoll If the same file description is open in two different processes, then closing the file descriptor is not sufficient to deregister it from the epoll instance (as described in epoll(7)), resulting in spurious events that cause the event loop to spin repeatedly. So always explicitly deregister it. Fixes #1099.
2014-02-05 18:36:24 -05:00
TEST_ENTRY (closed_fd_events)
windows: remove dependency on rpcrt4 and ole32 libs. fixes https://github.com/joyent/libuv/issues/118
2011-08-01 20:10:55 -07:00
#endif
Start uv_spawn() Unix only at the moment. Lacks test for stdio
2011-07-20 17:55:26 -07:00
stream: fix small nit in select hack, add test
2013-04-26 23:43:28 +04:00
#ifdef __APPLE__
TEST_ENTRY (osx_select)
darwin: allocate enough space for select() hack `fd_set`s are way too small for `select()` hack when stream's fd is bigger than 1023. Make `fd_set`s a part of `uv__stream_select_t` structure. fix #1461
2014-09-07 23:56:15 +01:00
TEST_ENTRY (osx_select_many_fds)
stream: fix small nit in select hack, add test
2013-04-26 23:43:28 +04:00
#endif
fs: add UV_ENOENT error code, add test
2011-09-06 00:16:41 +02:00
TEST_ENTRY (fs_file_noent)
unix: map `ENAMETOOLONG` to `UV_ENAMETOOLONG` With tests. Closes #295
2012-01-22 02:01:54 +01:00
TEST_ENTRY (fs_file_nametoolong)
test: test if `UV_ELOOP` mapping works
2012-01-25 12:39:44 +01:00
TEST_ENTRY (fs_file_loop)
uv_fs & uv_work APIs
2011-08-19 16:46:42 -07:00
TEST_ENTRY (fs_file_async)
TEST_ENTRY (fs_file_sync)
unix: fix uv_fs_write when using an empty buffer
2014-02-26 23:59:29 +01:00
TEST_ENTRY (fs_file_write_null_buffer)
uv_fs & uv_work APIs
2011-08-19 16:46:42 -07:00
TEST_ENTRY (fs_async_dir)
TEST_ENTRY (fs_async_sendfile)
unix, windows: add uv_fs_mkdtemp
2014-07-13 22:42:14 +03:00
TEST_ENTRY (fs_mkdtemp)
Add test for uv_fs_fstat, implement on unix.
2011-09-01 14:15:01 -07:00
TEST_ENTRY (fs_fstat)
unix, windows: add uv_fs_access()
2014-10-02 22:16:21 -04:00
TEST_ENTRY (fs_access)
add test fs_chmod, implement uv_fs_fchmod and uv_fs_chmod on unix
2011-09-01 15:36:01 -07:00
TEST_ENTRY (fs_chmod)
win: fix unlink of readonly files Unix unlinks files even if they are not writable, windows version should behave similarly. Fixes: https://github.com/joyent/node/issues/3006 PR-URL: https://github.com/libuv/libuv/pull/327 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2015-04-21 19:44:30 +01:00
TEST_ENTRY (fs_unlink_readonly)
windows: implement missing fs functions
2011-09-03 11:06:08 -07:00
TEST_ENTRY (fs_chown)
unix: implement uv_fs_futime, add tests for uv_fs_utime and uv_fs_futime
2011-09-06 02:52:23 +02:00
TEST_ENTRY (fs_utime)
TEST_ENTRY (fs_futime)
test: add uv_fs_readlink test
2012-09-30 03:26:48 +02:00
TEST_ENTRY (fs_readlink)
fs: add uv_fs_realpath() Equivalent to realpath(3), returns the full resolved absolute path of a file or directory. PR-URL: https://github.com/libuv/libuv/pull/531 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2015-09-19 00:40:34 +03:00
TEST_ENTRY (fs_realpath)
unix: implement uv_fs_futime, add tests for uv_fs_utime and uv_fs_futime
2011-09-06 02:52:23 +02:00
TEST_ENTRY (fs_symlink)
windows: support junctions with uv_fs_synlink based on @piscisaureus implementation
2012-05-08 15:44:11 -07:00
TEST_ENTRY (fs_symlink_dir)
windows: fix error reporting for uv_fs_ functions
2011-09-28 12:46:37 -07:00
TEST_ENTRY (fs_stat_missing_path)
test: fs: add tests for read EOF This fix was merged without tests: https://github.com/philips/libuv/tree/fix-read-on-windows-to-handle-eof So take tests from igorzi: https://github.com/igorzi/libuv/commit/46024bf33dcc7fc922378fe0d8b4f59f4c2cd605
2012-02-28 17:27:18 -08:00
TEST_ENTRY (fs_read_file_eof)
windows: honor O_APPEND in uv_fs_open
2011-11-04 12:41:23 -07:00
TEST_ENTRY (fs_file_open_append)
windows: file watcher
2011-09-16 12:04:36 -07:00
TEST_ENTRY (fs_event_watch_dir)
win: add support for recursive file watching Original patch by @ghostoy, modified by @bpasero and yours truly. Refs: https://github.com/joyent/libuv/pull/1473 Refs: https://github.com/libuv/libuv/pull/198 PR-URL: https://github.com/libuv/libuv/pull/421 Reviewed-By: Bert Belder <bertbelder@gmail.com>
2015-07-02 01:04:11 +02:00
TEST_ENTRY (fs_event_watch_dir_recursive)
windows: file watcher
2011-09-16 12:04:36 -07:00
TEST_ENTRY (fs_event_watch_file)
win: compare entire filename in watch events This commit causes Window file watching events to compare the entire file path when filtering events. This fixes a bug where incomplete path comparisons would cause invalid events to be raised. Refs: https://github.com/libuv/libuv/pull/682 PR-URL: https://github.com/libuv/libuv/pull/924 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2016-06-22 13:14:04 -04:00
TEST_ENTRY (fs_event_watch_file_exact_path)
test: add failing fs_event test Watches the same file twice. Fails on Linux with a segmentation fault.
2012-07-28 14:40:01 +02:00
TEST_ENTRY (fs_event_watch_file_twice)
windows: file watcher
2011-09-16 12:04:36 -07:00
TEST_ENTRY (fs_event_watch_file_current_dir)
win: fix watching root files When passing "\\?\C:" to CreateFile, it opens the drive rather than the root directory. So include the trailing backslash in the directory name. Fixes: https://github.com/nodejs/node/issues/4643 PR-URL: https://github.com/libuv/libuv/pull/689 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2016-01-14 19:21:47 +02:00
#ifdef _WIN32
TEST_ENTRY (fs_event_watch_file_root_dir)
#endif
win,sunos: stop handle on uv_fs_event_start() err This commit stops the handle in uv_fs_event_start() when an error occurs. Fixes: https://github.com/libuv/libuv/issues/1253 PR-URL: https://github.com/libuv/libuv/pull/1257 Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
2017-03-16 21:20:11 -04:00
TEST_ENTRY (fs_event_watch_invalid_path)
sunos: properly disarm PORT_LOADED fsevent watcher Fixes a segmentation fault when the watched entity is changed after the watcher has been closed.
2012-12-12 19:47:58 -08:00
TEST_ENTRY (fs_event_no_callback_after_close)
windows: don't emit fs-event callback after uv_fs_event handle is closed
2011-11-02 19:09:18 -07:00
TEST_ENTRY (fs_event_no_callback_on_close)
unix: properly disarm kqueue fs watcher Fixes "Assertion failed: (revents == EV_LIBUV_KQUEUE_HACK), function uv__fs_event, file ../src/unix/kqueue.c, line 58."
2011-09-27 15:04:45 +02:00
TEST_ENTRY (fs_event_immediate_close)
Test: verify that no callbacks are made after closing an fs event watcher
2012-03-27 22:56:34 +02:00
TEST_ENTRY (fs_event_close_with_pending_event)
Test: verify that uv_close from an fs event callback works properly
2012-03-27 22:52:53 +02:00
TEST_ENTRY (fs_event_close_in_callback)
fsevents: fix clever rescheduling There're could be a situation, where one fsevents handle gets created and another one is destroyed simultaneously. In such cases `fsevent_need_reschedule` will be set to 1 twice and reset only once, leaving handle destructor hanging in uv_sem_wait().
2013-10-02 14:47:14 +04:00
TEST_ENTRY (fs_event_start_and_close)
test: test fsevents error reporting Create a lot of `FSEventStream`s to make `FSEventStreamStart()` fail. Fetch the error code and verify that it is correct.
2013-11-01 23:00:58 +04:00
TEST_ENTRY (fs_event_error_reporting)
unix, windows: add uv_fs_event_getpath
2014-02-24 00:51:13 +01:00
TEST_ENTRY (fs_event_getpath)
fs: rename uv_fs_readdir to uv_fs_scandir Closes #1431
2014-09-25 10:09:58 +02:00
TEST_ENTRY (fs_scandir_empty_dir)
fs: fix crash in uv_fs_scandir_next Handle the case when it's called after the request has ended, or in case it ended with an error. PR-URL: https://github.com/libuv/libuv/pull/1181 Reviewed-By: Colin Ihrig <cjihrig@gmail.com> Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com> Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2017-01-05 15:16:48 -08:00
TEST_ENTRY (fs_scandir_non_existent_dir)
fs: rename uv_fs_readdir to uv_fs_scandir Closes #1431
2014-09-25 10:09:58 +02:00
TEST_ENTRY (fs_scandir_file)
windows: enable uv_fs_open to open directories
2011-10-12 10:36:14 -07:00
TEST_ENTRY (fs_open_dir)
windows: uv_fs_rename to replace the new file if it exists fixes #283
2011-12-30 15:46:17 -08:00
TEST_ENTRY (fs_rename_to_existing_file)
windows: fix fs_write with nbufs > 1 and offset When multiple bufs are specified, overlapped needs to be advanced manually between each write. Without this, each buf will be written to the same position (if offset is specified.) ReadFile() automatically advances, but WriteFile() does not.
2014-10-19 20:48:51 -07:00
TEST_ENTRY (fs_write_multiple_bufs)
unix: allow nbufs > IOV_MAX in uv_fs_{read,write} This allows writing and reading any amount of buffers, regardless of what IOV_MAX may be defined as. It also moves the IOV_MAX test from stream to core. This is based on the excellent work of @bwijen in #269. Refs: https://github.com/libuv/libuv/pull/269 PR-URL: https://github.com/libuv/libuv/pull/448 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2015-08-04 17:07:54 +09:00
TEST_ENTRY (fs_write_alotof_bufs)
TEST_ENTRY (fs_write_alotof_bufs_with_offset)
unix,win: return EINVAL on nullptr args in uv_fs_{read,write} PR-URL: https://github.com/libuv/libuv/pull/470 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2015-08-11 12:38:57 +02:00
TEST_ENTRY (fs_read_write_null_arguments)
uv_fs & uv_work APIs
2011-08-19 16:46:42 -07:00
TEST_ENTRY (threadpool_queue_work_simple)
unix, windows: disallow NULL worker in thread pool See #629.
2012-11-24 08:25:32 +09:00
TEST_ENTRY (threadpool_queue_work_einval)
test: fix threadpool_multiple_event_loops on PPC On PPC (linux and AIX) uv_thread_join (which is just a call to pthread_join) takes quite a while. Increased the timeout of this specific test on PPC so that there is ample time for all threads to join back. The fs_do and getaddrinfo_do calls do not take up much time. Also removing the ifdef for AIX around fs_do since it did nothing. Fixes: https://github.com/libuv/libuv/issues/687 PR-URL: https://github.com/libuv/libuv/pull/737 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl> Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2016-02-25 11:51:31 -05:00
#if defined(__PPC__) || defined(__PPC64__) /* For linux PPC and AIX */
/* pthread_join takes a while, especially on AIX.
* Therefore being gratuitous with timeout.
*/
TEST_ENTRY_CUSTOM (threadpool_multiple_event_loops, 0, 0, 120000)
#else
test: eio callbacks should run in their owning threads
2011-12-22 03:42:54 +01:00
TEST_ENTRY (threadpool_multiple_event_loops)
test: fix threadpool_multiple_event_loops on PPC On PPC (linux and AIX) uv_thread_join (which is just a call to pthread_join) takes quite a while. Increased the timeout of this specific test on PPC so that there is ample time for all threads to join back. The fs_do and getaddrinfo_do calls do not take up much time. Also removing the ifdef for AIX around fs_do since it did nothing. Fixes: https://github.com/libuv/libuv/issues/687 PR-URL: https://github.com/libuv/libuv/pull/737 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl> Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2016-02-25 11:51:31 -05:00
#endif
unix: add uv_cancel()
2012-11-26 03:12:38 +01:00
TEST_ENTRY (threadpool_cancel_getaddrinfo)
unix: allow uv_getnameinfo to be cancelled
2014-06-18 02:00:07 +02:00
TEST_ENTRY (threadpool_cancel_getnameinfo)
unix: add uv_cancel()
2012-11-26 03:12:38 +01:00
TEST_ENTRY (threadpool_cancel_work)
TEST_ENTRY (threadpool_cancel_fs)
unix: ensure done_cb gets called after uv_cancel() Wake up the event loop with uv_async_send() when a request is cancelled. Ensures the done_cb is run on the next tick of the event loop. Not sending a wakeup signal results in the done_cb not getting called until another request completes, which may be a long time coming when it's the only request in the queue or when other requests are executing long-running jobs. Fixes #669.
2012-12-28 15:10:04 +01:00
TEST_ENTRY (threadpool_cancel_single)
unix, windows: add thread-local storage API Uses the pthread_key_{create,delete} and pthread_{get,set}specific functions on UNIX platforms, Tls{Alloc,Free} and Tls{Get,Set}Value on Windows. Fixes #904.
2013-08-25 03:12:41 +02:00
TEST_ENTRY (thread_local_storage)
osx: set the default thread stack size to RLIMIT_STACK Fixes: https://github.com/libuv/libuv/issues/669 PR-URL: https://github.com/libuv/libuv/pull/671 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
2016-01-04 12:32:22 +01:00
TEST_ENTRY (thread_stack_size)
Wrap platform mutex and rwlock APIs. Read/write locks are emulated with critical sections on Windows XP and Vista because those platforms don't have a (complete) native read/write lock API.
2011-11-14 21:04:53 +01:00
TEST_ENTRY (thread_mutex)
TEST_ENTRY (thread_rwlock)
test: add tests for the uv_rwlock implementation PR-URL: https://github.com/libuv/libuv/pull/525 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2015-09-13 16:58:40 +02:00
TEST_ENTRY (thread_rwlock_trylock)
Wrap platform thread APIs.
2011-11-21 16:12:36 +01:00
TEST_ENTRY (thread_create)
unix, windows: add uv_thread_equal
2014-10-04 22:01:03 -07:00
TEST_ENTRY (thread_equal)
win, unix: add uv_dlerror() and uv_dlerror_free()
2012-02-25 01:32:31 +09:00
TEST_ENTRY (dlerror)
unix, windows: detect errors in uv_ip4/6_addr
2013-12-13 01:01:38 +08:00
TEST_ENTRY (ip4_addr)
linux,darwin,win: link-local IPv6 addresses Modified uv_ip6_addr() to fill sin6_scope_id for link-local addresses. Fixes #271 Conflicts: build.mk
2013-03-14 22:20:32 +01:00
TEST_ENTRY (ip6_addr_link_local)
test: test that uv_close() doesn't corrupt QUEUE The test adds all types of handles, that were known to be affected by QUEUE_REMOVE()-within-QUEUE_FOREACH() bug, to a loop. It then calls uv_close() to trigger QUEUE_REMOVE()-within-QUEUE_FOREACH() case and checks whether a particular QUEUE is corrupted or not. Restrict the test to Linux only for now as it fails on other platforms for various reasons. PR-URL: https://github.com/libuv/libuv/pull/621 Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2015-09-30 12:47:02 -04:00
TEST_ENTRY (queue_foreach_delete)
unix: make loops and watchers usable after fork() Added the uv_loop_fork() API that must be called in a child process to continue using an existing loop. Internally this calls a uv__io_fork function for each supported platform, similar to the way uv__platform_loop_init works. After this call, existing and new IO, async and signal watchers will contiue working as before on all platforms, as will the threadpool (although any threads it was using are of course gone). On Linux and BSDs that use kqueue, existing and new fsevent watchers will also continue to work as expected. On OS X, though, directory fsevents will not be able to use the optimized CoreFoundation path if they had already been used in the parent process, instead falling back to the kqueue path used on other BSDs. Existing fsevent watchers will not function on AIX or SunOS. This could be relatively easily fixed by someone with AIX knowledge in the future, but SunOS will require some additional work to keep track if the watchers. A new test file, test/test-fork.c, was added to contain fork-related tests to verify functionality in the child process. PR-URL: https://github.com/libuv/libuv/pull/846 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl> Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
2016-04-22 14:16:31 -05:00
#ifndef _WIN32
TEST_ENTRY (fork_timer)
TEST_ENTRY (fork_socketpair)
TEST_ENTRY (fork_socketpair_started)
TEST_ENTRY (fork_signal_to_child)
TEST_ENTRY (fork_signal_to_child_closed)
TEST_ENTRY (fork_fs_events_child)
TEST_ENTRY (fork_fs_events_child_dir)
TEST_ENTRY (fork_fs_events_file_parent_child)
TEST_ENTRY (fork_threadpool_queue_work_simple)
#endif
Disable fail-always and pass-always
2011-05-03 16:56:26 -07:00
#if 0
/* These are for testing the test runner. */
First shot at test harness
2011-04-05 01:43:17 +02:00
TEST_ENTRY (fail_always)
TEST_ENTRY (pass_always)
Disable fail-always and pass-always
2011-05-03 16:56:26 -07:00
#endif
Benchmark runner
2011-04-19 04:26:32 +02:00
TASK_LIST_END
Reference in New Issue Copy Permalink