Commit Graph
187 Commits
Author SHA1 Message Date
Zoltan Varga 2b5078c306 [bcl] Fix the RenameTpThread test, renaming a tp thread should not fail. 2014-09-23 18:40:34 -04:00
Miguel de Icaza 0699e4da36 Fix warnings in mscorlib's Test suite + bring a couple more tests + fix thread rename test 2014-09-23 10:56:04 -04:00
Zoltan Varga 16d9d07821 [runtime] Allow renaming of threadpool threads. Fixes #23206. 2014-09-22 17:07:26 -04:00
Alexander Köplinger 48e112d652 [Cleanup] Removed TARGET_JVM
It was once part of a special Java target profile, but this is no longer used today.
2014-08-08 03:28:34 +02:00
Zoltan Varga dd2b784606 [bcl] Increase some test timeouts to avoid failures when running with llvm. 2014-07-08 00:46:31 +02:00
Rodrigo Kumpera a89b144a7d [corlib] Fix the 4.5 test suite. 2014-02-12 19:12:03 -05:00
Rodrigo Kumpera 51d2b75664 [corlib] Fix 4.0 test suite compilation. 2014-02-12 18:53:10 -05:00
Marek Safar 661af380c3 [corlib] Fixes race in disposing CancellationTokenSource. Fixes #17256 2014-01-17 21:04:08 +01:00
Marek Safar dce005f81f [corlib] Thread.GetNamedDataSlot allocates slot on failure. Fixes #16788 2014-01-01 15:28:40 +01:00
Andrea Canciani 0ac0c15211 Add testcases for callback and exception ordering
Translated from the F# testcases available here
https://github.com/fsharp/fsharp/blob/19507b98bc6a7995e9c19c40f044fdfeb578ef77/src/fsharp/FSharp.Core.Unittests/FSharp.Core/Microsoft.FSharp.Control/Cancellation.fs#L100
2013-12-31 22:01:27 +01:00
Andrea Canciani 33178ca69c Fix CancellationTokenSourceTest.ReEntrantRegistrationTest
This test run on .NET would result in both "unregister" and "register"
set to a "true" value. In order to ensure that the "unregister = true"
fragment is not executed, its disposal needs to be registered after,
because on .NET the CancellationTokenSource has a LIFO behavior.

This change simplifies the code and makes the expected behavior
consistent with that of .NET.
2013-12-31 22:00:34 +01:00
Sebastien Pouliot e4dcfc71ff [tests] Fix some corlib unit tests to run successfully on iOS 2013-12-10 21:48:09 -05:00
Zoltan Varga e652c6d7f2 [corlib] Disable a test which causes a wapi_unref_full () warning. 2013-12-09 19:56:47 +01:00
Bart Vries e3c37d1042 LogicalSet/GetData implementation in CallContext 2013-11-25 16:59:59 +01:00
Rodrigo Kumpera 6b71c73790 Fix net_2_0 test. 2013-10-09 12:58:44 -04:00
Marek Safar f410e545e2 Allow to queue multicast delegates in thread-pool 2013-10-09 11:26:34 +02:00
Sebastien Pouliot 7885214216 Use more detailled assert messages in some, randomly failing, Timer unit tests (to help track #14305) 2013-10-02 16:28:12 -04:00
Zoltan Varga 119df901c2 Disable a corlib test which fails on some test runners. 2013-09-04 17:32:01 +02:00
Alistair Leslie-Hughes ccbc9e3a8f Correct TrySetApartmentstate tests
License: This patch is under the MIT/X11 license.
2013-07-24 14:27:56 +10:00
Marek Safar b166ec8b48 Fixes failing tests 2013-07-20 09:18:58 +02:00
Sebastien Pouliot 0b09bfc495 Disable TestDisposeOnCallback for MONOTOUCH, the test does not really work as the state is null (and throws in a separate thread) 2013-07-19 09:38:58 -04:00
Alistair Leslie-Hughes 070c3abecf Allow TrySetApartmentState to set the same state multiple times.
License: This patch is under the MIT/X11 license.
2013-07-17 15:22:54 +10:00
Rodrigo Kumpera 7745e11b76 Fix 2.0 corlib test suite. 2013-05-28 14:00:09 -04:00
Marek Safar c48573e1c5 Add more argument checks to RegisterWaitForSingleObject 2013-05-25 10:23:12 +02:00
Zoltan Varga 0ff40ae70a Implement Volatile.Read/Write<T> and Interlocked.MemoryBarrier (). 2013-03-07 02:41:46 +01:00
Marek Safar 0526659302 Disable more security context tests on mobile 2013-03-06 15:50:22 +01:00
Marek Safar 99b72a868b Fix failing tests 2013-02-26 23:19:34 +01:00
Marek Safar 762b050917 Merge tests from mobile branch 2013-02-26 16:28:36 +01:00
Marek Safar 915b61a66d Mobile tests update 2013-02-15 16:35:41 +01:00
Marek Safar d6af35c68f Fix more mobile tests 2013-02-12 20:02:07 +01:00
Rodrigo Kumpera 1b92a4148c Since MOBILE implies NET_4_* now, take this into account and simplify our ifdefs. 2013-02-11 18:42:18 -05:00
Marek Safar afa79435aa Update for mobile profile 2013-02-07 16:34:21 +01:00
Yann ROBIN 5c2e8d62de fix code according to mono coding guidelines 2012-12-10 22:37:38 +01:00
Yann ROBIN 2238ba7dc0 add LogicalCallContext to ExecutionContext
Save LogicalCallContext when Capture is called

Restore LogicalCallContext on Run.

see msdn for behavior of ExecutionContext http://msdn.microsoft.com/en-us/library/h84136b2.aspx
2012-12-07 17:36:24 +01:00
Rolf Bjarne Kvinge 87810b523a Make WaitHandle.Wait* behave the same wrt WAIT_IO_COMPLETION. Fixes #8566.
Make the all check for exceptions if interrupted, and wait again
with a re-calculated timeout if no exceptions were to be thrown.

This fixes a deadlock in WaitAny, because WaitAny wasn't checking
for exceptions, just re-calculating the timeout. It also makes
the methods behave the same wrt other APC reasons.
2012-11-28 00:08:11 +01:00
Petr Onderka d7f2593b19 new CancellationTokenRegistration().Dispose() shouldn't throw 2012-08-19 23:58:16 +02:00
Jeremie Laval ed68f88573 [corlib] Dont use blocking wait call in CancellationTokenSource test 2012-07-23 16:49:55 +01:00
Jérémie Laval 75222a62c5 [corlib] Linked token callback to the proxy CancellationTokenSource shouldn't throw ODE in concurrent situations. 2012-07-23 12:48:40 +01:00
Jérémie Laval 253c5fb1d3 [corlib] Dispose linked tokens connections when disposing the resulting CancellationTokenSource
Based on baulig's patch: https://github.com/baulig/mono/commit/3ee4f7600b8efe61f7a701a93fffed8373b7b4b3
2012-07-23 11:48:55 +01:00
Martin Baulig 6354893b8c Fix race condition in TimerTest.TestChange().
Set the first interval to 10 milleseconds, not just one.  Then read
the bucket.count again after changing the interval, the callback may
fire a few times while we execute the Assert() and t.Change() callls.
2012-06-26 18:52:07 +02:00
Marek Safar 1041e924c1 Fix few tests 2012-06-25 19:35:13 +01:00
Marek Safar d5a78c50c1 Fix CancellationTokenSource cancellation during Dispose 2012-06-11 16:02:39 +01:00
Marek Safar 0b01be241c Improve failing test 2012-05-30 19:03:51 +01:00
Jeremie Laval b0dcb61630 [corlib] Fix NullRef when using dispose a CancellationTokenRegistration when the initial source has been disposed 2012-05-16 11:50:17 +01:00
Konrad M. Kruczynski 73ca5b067f Another fix and unit test added.
Now all three possible cases are tested (and working).
2012-04-23 15:44:16 +02:00
Konrad M. Kruczynski 83c06e2b22 Fixes as per the Rodrigo's comment.
Wait now throws an exception in case of cancellation token was
cancelled.
2012-04-23 12:02:09 +02:00
Konrad M. Kruczynski e6e63c0aa6 Fix for bug in ManualResetEventSlim.
The problem occured when one waited for an event with the
cancellation token and the wait was successful but not
immediate. In that case Wait returned false instead of true,
because of wait handle number mismatch. Also the unit test was
added.
2012-04-22 22:24:14 +02:00
Jeremie Laval 7167f57c4d Rework CancellationTokenSource to remove locking. Fix #4173.
Issue reported by MartinG, unit test code based on his repro.
2012-04-10 12:39:03 +01:00
Jeremie Laval b8abfaae4e Update copyrights 2012-04-10 12:39:02 +01:00
Jeremie Laval a3713f1843 Add consistency unit test for ManualResetEventSlim between internal state and WaitHandle abstraction 2012-04-10 12:39:02 +01:00