generic-library/webrtc
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity
6,527 Commits 1 Branch 0 Tags 108 MiB
fded02c164
Commit Graph

6527 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
buildbot@webrtc.org
c31651d847 (Auto)update libjingle 71378257-> 71410012 
git-svn-id: http://webrtc.googlecode.com/svn/trunk@6726 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-18 08:22:39 +00:00
kwiberg@webrtc.org
e364ac902f AudioBuffer: Optimize const accesses to arrays that autoconvert int16<->float 
Specifically, when someone asks for a const pointer to the int16
version of the array, there's no need to invalidate the float version
of that array, and vice versa. (But obviously, invalidation still has
to happen when someone asks for a non-const pointer.)

R=aluebs@webrtc.org, andrew@webrtc.org, minyue@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/18809004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6725 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-18 07:50:29 +00:00
andrew@webrtc.org
c145668dc8 Reduce runtime of RingBufferTest by a factor of 100. 
This test was needlessly long.

TBR=pbos

Review URL: https://webrtc-codereview.appspot.com/15029004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6724 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-17 23:16:44 +00:00
wu@webrtc.org
4f5da030f1 Use _numMixedParticipants instead of audioFrameList->size() to determine if there're more than one participants. 
There are two audioFrameLists. The previous check wouldn't work correctly if each list had a single member.

TEST=chrome https://apprtc.appspot.com/?debug=loopback&video=false and verify e2e delay stats
BUG=
R=andrew@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/15019004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6723 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-17 22:19:21 +00:00
mallinath@webrtc.org
aa93611375 Connect to the turn server if address cannot be resolved by the browser by using 
unresolved address. This case is only considered for TCP sockets. P2P layer will
assume socket will do the resolve by using a proxy.

BUG=3384
R=jiayl@webrtc.org, juberti@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/13829004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6722 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-17 21:55:04 +00:00
mallinath@webrtc.org
e5995aadd5 Assigning a priority to TURN server list passed to PeerConnection. First entry in the TURN server list will get the highest priotity and so forth. 
This priority will be used in calculating the candidate priority generated from the server. This will allow candidate generated from server to have unique priority.

BUG=3223
R=jiayl@webrtc.org, juberti@webrtc.org, pthatcher@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/16549004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6721 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-17 18:23:52 +00:00
jiayl@webrtc.org
e10d28cf14 fix 
git-svn-id: http://webrtc.googlecode.com/svn/trunk@6720 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-17 17:07:49 +00:00
stefan@webrtc.org
8b94e3da0f Fix issue where padding is sent before media with undefined timestamps if not abs-send-time is enabled. 
This broke bandwidth estimation for calls without abs-send-time is enabled, but where RTX was.

BUG=
R=pbos@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/16929004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6719 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-17 16:10:14 +00:00
aluebs@webrtc.org
4065988108 Remove unused ExperimentalNS API in AudioProcessing 
R=andrew@webrtc.org, kwiberg@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/12909004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6718 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-17 11:32:09 +00:00
kwiberg@webrtc.org
2b6bc8d84f AudioBuffer: Eliminate the SplitChannelBuffer class 
It's just a container for two IFChannelBuffers, and doesn't earn its
keep. The main problem is that the number of methods it needs that
just forward calls to either of its two IFChannelBuffers was already
large, and was about to grow.

R=aluebs@webrtc.org, minyue@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/16919004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6717 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-17 09:46:37 +00:00
pbos@webrtc.org
5301b0f1fc Move additional state into WebRtcVideoSendStream. 
Prevents having two places where codecs etc. are set up and allows us to
avoid creating the underlying VideoSendStream before send codecs are
set up.

BUG=1788
R=pthatcher@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/20939004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6716 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-17 08:51:46 +00:00
aluebs@webrtc.org
2561d52460 Simplify AudioBuffer::mixed_low_pass_data API 
R=andrew@webrtc.org, kwiberg@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/21869004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6715 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-17 08:27:39 +00:00
kwiberg@webrtc.org
af93fc08a1 AudioBuffer: Let ChannelBuffer handle bounds checking of channel parameter 
R=aluebs@webrtc.org, minyue@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/13019004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6714 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-17 08:18:33 +00:00
kwiberg@webrtc.org
2ade42bd96 Add unit test for MediaFile WAV file writing 
R=aluebs@webrtc.org, andrew@webrtc.org, minyue@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/16029004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6713 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-17 08:11:32 +00:00
tkchin@webrtc.org
4a472fb18d Fixes up rtc so that it compiles on iOS 8 SDK. 
Adds support for UIInterfaceOrientationUnknown (new with in SDK) and makes it the same as
UIInterfaceOrientationPortrait.

R=noahric@google.com, tkchin@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/13029004

Patch from David Maclachlan <dmaclach@chromium.org>.

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6712 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-17 00:21:59 +00:00
wu@webrtc.org
52eddec71b Revert 6707 "Add support of multiple STUN servers in UDPPort." 
Reason:
Breaks the build on callclient.cc.

> Add support of multiple STUN servers in UDPPort.
> Now UDPPort signals PortComplete or PortError when the Bind requests for all STUN servers are responded or failed. If any STUN bind is successful, PortComplete is signaled; otherwise, PortError is signaled.
> 
> I discovered a bug in SocketAddress while working on this. It didn't consider two addresses unequal if they have unresolved IP and different hosts. It's fixed now.
> 
> BUG=3310
> R=mallinath@webrtc.org
> 
> Review URL: https://webrtc-codereview.appspot.com/13879004

TBR=jiayl@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/14999004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6711 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-17 00:03:24 +00:00
minyue@webrtc.org
c56ae63ea6 r6709 lacks a change in BUILD.gn 
BUG=
R=marpan@google.com, marpan@webrtc.org, pbos@webrtc.org, wu@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/21919004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6710 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-16 22:18:49 +00:00
minyue@webrtc.org
74aaf29a0f Raw packet loss rate reported by RTP_RTCP module may vary too drastically over time. This CL is to add a filter to the value in VoE before lending it to audio coding module. 
The filter is an exponential filter borrowed from video coding module.

The method is written in a new class called PacketLossProtector (not sure if the name is nice), which can be used in the future for more sophisticated logic.

BUG=
R=henrika@webrtc.org, stefan@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/20809004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6709 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-16 21:28:26 +00:00
wu@webrtc.org
4c3e9917e7 Make sure b lines appear before all the a lines. Per RFC 4566, the order of media description should be: 
m=  (media name and transport address)
  i=* (media title)
  c=* (connection information -- optional if included at
       session level)
  b=* (zero or more bandwidth information lines)
  k=* (encryption key)
  a=* (zero or more media attribute lines)

BUG=2260
R=jiayl@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/15889004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6708 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-16 21:03:13 +00:00
jiayl@webrtc.org
46fb331bc5 Add support of multiple STUN servers in UDPPort. 
Now UDPPort signals PortComplete or PortError when the Bind requests for all STUN servers are responded or failed. If any STUN bind is successful, PortComplete is signaled; otherwise, PortError is signaled.

I discovered a bug in SocketAddress while working on this. It didn't consider two addresses unequal if they have unresolved IP and different hosts. It's fixed now.

BUG=3310
R=mallinath@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/13879004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6707 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-16 20:55:31 +00:00
tkchin@webrtc.org
2e3c97ddf5 Compile-time guard for iOS7 specific property. 
BUG=3487
R=glaznev@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/17969004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6706 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-16 19:59:05 +00:00
buildbot@webrtc.org
a8d8ad2be6 (Auto)update libjingle 71240799-> 71250251 
git-svn-id: http://webrtc.googlecode.com/svn/trunk@6705 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-16 14:23:08 +00:00
stefan@webrtc.org
4070b1db53 Print an info log instead of return an error if an external encoder is de-registered, but no corresponding internal encoder can be registered automatically. 
This is not an error case if for instance an external h.264 encoder is registered, but no internal implementation exists.

R=pbos@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/13009004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6704 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-16 11:20:40 +00:00
pbos@webrtc.org
63c60ed224 Remove old padding path in RTPSender. 
Removing RTPSender::SendPaddingAccordingToBitrate() as well as a couple
of arguments from SendPadData().

BUG=
R=stefan@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/14989004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6703 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-16 09:37:29 +00:00
kwiberg@webrtc.org
efb81d8d1f int16<->float conversions: Use size_t for array length argument, not int 
size_t is more appropriate for array lengths, since int might
theoretically be too small for a really large array. But more
importantly, if the caller's value is naturally of type size_t and the
function requires an int, VC++ will trigger warning C4267
(http://msdn.microsoft.com/en-us/library/6kck0s93.aspx) because the
implicit cast might be lossy, forcing the caller to do a manual cast.
Typing the function with size_t in the first place resolves the
problem.

R=aluebs@webrtc.org, andrew@webrtc.org, minyue@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/21909004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6702 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-16 08:36:52 +00:00
kwiberg@webrtc.org
0fa6366ed1 Define convenient FATAL_ERROR() and FATAL_ERROR_IF() macros 
R=henrike@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/16079004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6701 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-16 08:34:58 +00:00
kwiberg@webrtc.org
e8ea33ccb1 nrsh1 is written before tmp321 is read, so needs to be earlyclobber 
Otherwise, the compiler is allowed to put them in the same register
under the assumption that all inputs are read before any
(non-earlyclobber) output is written, which in this case would result
in nrsh2 being corrupted.

BUG=3439
R=aluebs@webrtc.org, ljubomir.papuga@gmail.com

Review URL: https://webrtc-codereview.appspot.com/16089004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6700 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-16 08:26:48 +00:00
pbos@webrtc.org
38ce7d03d8 Implement unittest for SetSendCodecsChangesExistingStreams. 
BUG=1788
R=pbos@webrtc.org, wu@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/19869004

Patch from Changbin Shao <changbin.shao@intel.com>.

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6699 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-16 08:01:38 +00:00
jiayl@webrtc.org
bac5f0fb56 Fix an invalid memory access due to typo in win/cursor.cc. 
BUG=crbug/391468
R=sergeyu@chromium.org

Review URL: https://webrtc-codereview.appspot.com/19949004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6698 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-15 20:32:03 +00:00
tkchin@webrtc.org
122caa51b1 After an audio interruption the audio unit no longer invokes its render callback, which results in a loss of audio. Restarting the audio unit post interruption fixes the issue. 
CL also replaces deprecated AudioSession calls with equivalent AVAudioSession ones.

BUG=3487
R=glaznev@webrtc.org, noahric@chromium.org

Review URL: https://webrtc-codereview.appspot.com/21769004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6697 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-15 20:20:47 +00:00
tommi@webrtc.org
47218956fc Minor refactoring of StatsCollector. 
* Make GetTimeNow a static method in the cc file.
* Make GetTransportIdFromProxy a static method as well and not a class method.

The second change is in preparation of removing the proxy_to_transport_ member variable which isn't needed and is just a copy from the session stats.

R=xians@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/20959004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6696 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-15 19:22:37 +00:00
tkchin@webrtc.org
42fe4350fe Remove Thread::RunningForChannelManager(). 
I haven't heard of this failing, so it should be safe to remove. Let me know if this isn't the case.

BUG=3388
R=andrew@webrtc.org, wu@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/18659004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6695 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-15 17:52:43 +00:00
stefan@webrtc.org
89fd1e8e99 Improvements to the pacer where it lost some budget due to truncation errors. 
With this CL the resolution is increased to microseconds and proper rounding
is done in the Process() function. This means that we will be allowed to send
more than prior to r6664 as we previously truncated away parts of our budget.

We will also not lose budget due to inaccurate calculations in
TimeUntilNextProcess(), which was a regression in r6664.

BUG=cr/393950
TEST=out/Debug/webrtc_perf_tests --gtest_filter=RampUpTest.Simulcast
R=pbos@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/20949004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6694 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-15 16:40:38 +00:00
pbos@webrtc.org
376b4ea93f Fix breakage introduced by r6691. 
ModuleRtpRtcpImpl returned incorrectly on RemoteNTP as the
RTCPReceiver::NTP changed return type.

BUG=
TBR=stefan@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/18799004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6693 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-15 15:51:33 +00:00
pbos@webrtc.org
2f4b14e3f3 Make RTCP sender report send media bytes. 
r6654 changed RtpSender::Bytes() to return the number of bytes sent
instead of number of media bytes. This is used by VideoEngine for stats.
This change broke RTCP which sends this same count as the number of
payload bytes sent (excluding headers and padding).

BUG=
R=stefan@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/14959004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6691 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-15 15:25:39 +00:00
kwiberg@webrtc.org
ffa8dcab1e Eliminate unnecessary #include 
R=henrika@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/14979004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6690 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-15 12:50:13 +00:00
kwiberg@webrtc.org
324f63ca38 rtc::Fatal output: Print space between # and message 
R=henrika@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/17949004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6689 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-15 11:41:05 +00:00
pbos@webrtc.org
bc73871251 Remove the VPM denoiser. 
The VPM denoiser give bad results, is slow and has not been used in
practice. Instead we use the VP8 denoiser. Testing this denoiser takes
up a lot of runtime on linux_memcheck (about 4 minutes) which we can do
without.

BUG=
R=stefan@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/16069004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6688 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-15 09:50:40 +00:00
tommi@webrtc.org
2adc51c86e Handle the case if an unusually long peer name is provided in the peerconnection example. 
R=xians@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/21899004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6687 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-15 08:56:07 +00:00
pbos@webrtc.org
cb859ecd3b Replace strcpy with talk_base::strcpyn. 
Cpplint reports error 'Almost always, snprintf is better than strcpy'
when checking code styles. The function talk_base::strcpyn() is a better
option than strcpy().

BUG=1788
R=pbos@webrtc.org, wu@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/12919004

Patch from Changbin Shao <changbin.shao@intel.com>.

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6686 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-15 08:28:20 +00:00
fbarchard@google.com
6823479ad3 Roll libyuv from 1033 to 1035 to get cpuid fix for AVX2 that avoids misdetect causing a crash in AVX2 code on cpus that do not have AVX2. 
BUG=libyuv:343
TESTED=libyuv try bots pass
R=andrew@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/20929004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6685 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-14 23:27:05 +00:00
fgalligan@google.com
d873540101 Roll chromium 282462:282879. 
Pick up the libvpx roll:
https://codereview.chromium.org/387003005/

R=andrew@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/19969004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6684 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-14 23:14:48 +00:00
henrike@webrtc.org
92a9bacf9a Rebase webrtc/base with r6682 version of talk/base: 
cls ported: r6671, r6672, r6679 (reverts and unreverts in r6680, r6682).
svn diff -r 6656:6682 http://webrtc.googlecode.com/svn/trunk/talk/base >
6682.diff
sed -i.bak "s/talk_base/rtc/g" 6682.diff
sed -i.bak "s/#ifdef WIN32/#if defined(WEBRTC_WIN)/g" 6682.diff
sed -i.bak "s/#if defined(WIN32)/#if defined(WEBRTC_WIN)/g" 6682.diff
patch -p0 -i 6682.diff

BUG=3379
TBR=tommi@webrtc.org,jiayl@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/14969004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6683 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-14 22:03:57 +00:00
henrike@webrtc.org
1b84116417 Add a facility to the Thread class to catch blocking regressions. 
This facility should be used in methods that run on known threads
(e.g. signaling, worker) and do not have blocking thread syncronization
operations via the Thread class such as Invoke, Sleep, etc.

This is a reland of an already reviewed cl (r6679) that got reverted by mistake.

TBR=xians@google.com,tommi@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/21889004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6682 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-14 21:42:39 +00:00
tkchin@webrtc.org
b038c72369 Enable SCTP compile for iOS. 
Chromium's been updated to pull a version of usrsctplib that will compile correctly. This update DEPS to point at new revision and turn on the compile time flags for iOS sctp.

BUG=3211
R=wu@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/13929004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6681 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-14 20:24:09 +00:00
buildbot@webrtc.org
aac14973aa (Auto)update libjingle 71116846-> 71117224 
git-svn-id: http://webrtc.googlecode.com/svn/trunk@6680 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-14 20:22:21 +00:00
tommi@webrtc.org
5be649fcfc Add a facility to the Thread class to catch blocking regressions. 
This facility should be used in methods that run on known threads
(e.g. signaling, worker) and do not have blocking thread syncronization
operations via the Thread class such as Invoke, Sleep, etc.

This is a reland of an already reviewed cl that got reverted by mistake.

TBR=xians@google.com

Review URL: https://webrtc-codereview.appspot.com/12999004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6679 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-14 20:21:36 +00:00
tommi@webrtc.org
242068d58c A step towards changing StatsReport::Value::name to an enum. 
The stats reporting code does a lot of unnecessary string copying.
This is a step in the direction of removing that and forcing use of only known constants.

This is a reland of an already reviewed cl that got reverted by mistake.

TBR=xians@google.com

Review URL: https://webrtc-codereview.appspot.com/12989004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6678 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-14 20:19:56 +00:00
tommi@webrtc.org
03505bcb7a Make StatsCollector depend on always having a valid session pointer. 
This is required since the session pointer is currently used on multiple threads but there's no synchronization code to guard it.
I'm removing the set_session() method and session() getter since they would cause problems if used without synchronization.

This is a reland of an already reviewed cl that got reverted by mistake.

TBR=xians@google.com

Review URL: https://webrtc-codereview.appspot.com/13959004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6677 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-14 20:15:26 +00:00
tommi@webrtc.org
b5348c64bb Minor refactoring of the session classes. 
Make member variables that never change and are touched on multiple threads, const.
Move implementations of setters/getters of variables that can change, into the cc file in preparation of adding thread correctness checks.

This is a relanding of a cl already reviewed but got reverted by mistake.

TBR=xians@google.com

Review URL: https://webrtc-codereview.appspot.com/12979004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6676 4adac7df-926f-26a2-2b94-8c16560cd09d
 2014-07-14 20:11:49 +00:00