e753641ef1
Adding ability to simulate EWOULDBLOCK/SignalReadyToSend.
...
Calling VirtualSocketServer::SetSendingBlocked(true) will simulate the
network interface being blocked, and SetSendingBlocked(false) will
simulate it being unblocked, resulting in SignalReadyToSend if
appropriate.
I plan to use this to write tests for upper layers of code that deal
with EWOULDBLOCK/SignalReadyToSend.
Also doing some minor housekeeping in this CL (using RTC_DCHECK,
renaming variables, etc.).
R=pthatcher@webrtc.org , skvlad@webrtc.org
Review URL: https://codereview.webrtc.org/2284903002 .
Cr-Commit-Position: refs/heads/master@{#14170}
2016-09-09 20:16:25 +00:00
5b53b0eeb4
Remove timing dependent packet checks from socket tests.
...
This avoids flaky tests and the check itself doesn't provide much value anyway.
TBR=pthatcher@webrtc.org
Review URL: https://codereview.webrtc.org/2005993002 .
Cr-Commit-Position: refs/heads/master@{#12855}
2016-05-23 20:56:22 +00:00
9131efdb30
Read recv timestamps from socket (posix only).
...
This helps a lot on Android devices where the user threads can be scheduled with low priority when the app is in the background, causing spurious significantly delayed before a packet can be read from the socket. With this patch the timestamp is taken by the kernel when the packet actually arrives.
R=juberti@chromium.org
TBR=juberti@webrtc.org
BUG=webrtc:5773
Review URL: https://codereview.webrtc.org/1944683002 .
Cr-Commit-Position: refs/heads/master@{#12850}
2016-05-23 16:19:37 +00:00
2b3bf6b7d5
Re-enabling socket tests that were previously flaky.
...
It's assumed that these tests were flaky because they used a non-virtual
resource (sockets) and were being run in parallel. So this should be
fixed now that these tests are now not being run in parallel, thanks to
this CL: https://codereview.webrtc.org/1426643003
BUG=webrtc:4923
R=pthatcher@webrtc.org
Review URL: https://codereview.webrtc.org/1982733002 .
Cr-Commit-Position: refs/heads/master@{#12817}
2016-05-19 21:57:40 +00:00
0c4e06b4c6
Use suffixed {uint,int}{8,16,32,64}_t types.
...
Removes the use of uint8, etc. in favor of uint8_t.
BUG=webrtc:5024
R=henrik.lundin@webrtc.org , henrikg@webrtc.org , perkj@webrtc.org , solenberg@webrtc.org , stefan@webrtc.org , tina.legrand@webrtc.org
Review URL: https://codereview.webrtc.org/1362503003 .
Cr-Commit-Position: refs/heads/master@{#10196}
2015-10-07 10:23:32 +00:00
ec499beaf5
Increase testclient timeout from 1 to 5 seconds
...
BUG=4182
R=pthatcher@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/38839004
Cr-Commit-Position: refs/heads/master@{#8285}
git-svn-id: http://webrtc.googlecode.com/svn/trunk@8285 4adac7df-926f-26a2-2b94-8c16560cd09d
2015-02-07 22:38:16 +00:00
f048872e91
Adds a modified copy of talk/base to webrtc/base. It is the first step in
...
migrating talk/base to webrtc/base.
BUG=N/A
R=niklas.enbom@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/17479005
git-svn-id: http://webrtc.googlecode.com/svn/trunk@6129 4adac7df-926f-26a2-2b94-8c16560cd09d
2014-05-13 18:00:26 +00:00
e9a604accd
Revert 6107 "Adds a modified copy of talk/base to webrtc/base. I..."
...
This breaks Chromium FYI builds and prevent roll of webrtc/libjingle to Chrome.
http://chromegw.corp.google.com/i/chromium.webrtc.fyi/builders/Win%20Builder/builds/457
> Adds a modified copy of talk/base to webrtc/base. It is the first step in migrating talk/base to webrtc/base.
>
> BUG=N/A
> R=andrew@webrtc.org , wu@webrtc.org
>
> Review URL: https://webrtc-codereview.appspot.com/12199004
TBR=henrike@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/14479004
git-svn-id: http://webrtc.googlecode.com/svn/trunk@6116 4adac7df-926f-26a2-2b94-8c16560cd09d
2014-05-13 08:15:48 +00:00
2c7d1b39b9
Adds a modified copy of talk/base to webrtc/base. It is the first step in migrating talk/base to webrtc/base.
...
BUG=N/A
R=andrew@webrtc.org , wu@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/12199004
git-svn-id: http://webrtc.googlecode.com/svn/trunk@6107 4adac7df-926f-26a2-2b94-8c16560cd09d
2014-05-12 18:03:09 +00:00
