1) Function responsible for receiving feedback, digesting data and deciding switch scenarios.
2) Function which enters Startup mode.
3) Function which exits Startup mode.
4) Function which calculates, whether or not full bandwidth is reached.
BUG=webrtc:7713
Review-Url: https://codereview.webrtc.org/2924603002
Cr-Commit-Position: refs/heads/master@{#18901}
I used a command like this to update the paths:
perl -pi -e "s/webrtc\/base/webrtc\/rtc_base/g" `find webrtc/rtc_base -name "*.cc" -o -name "*.h"`
The only manual edit is to add an include of webrtc/rtc_base/checks.h in
webrtc/modules/audio_device/android/opensles_common.h, which likely
was needed due to changed include paths due to 'git cl format'.
BUG=webrtc:7634
NOTRY=True
NOPRESUBMIT=True
Review-Url: https://codereview.webrtc.org/2969653002
Cr-Commit-Position: refs/heads/master@{#18871}
Reason for revert:
Resetting the estimate means that we need to start gathering data from scratch again. The combination of
1) DelayBasedEstimator not reacting to overuse unless there is a valid estimate of the acknowledged bitrate, and
2) AcknowledgedBitrateEstimator needing a significant amount of time/data to obtain an provide an estimate
causes poor performance in simulations/tests. It is not clear whether this will affect real networks negatively, but I suggest reverting this to be on the safe side.
See also https://bugs.chromium.org/p/webrtc/issues/detail?id=7884
Original issue's description:
> Test and fix for huge bwe drop after alr state.
>
> BUG=webrtc:7746
>
> Review-Url: https://codereview.webrtc.org/2931873002
> Cr-Commit-Position: refs/heads/master@{#18692}
> Committed: 37aa8ba616TBR=solenberg@webrtc.org,kwiberg@webrtc.org,minyue@webrtc.org,holmer@chromium.org,philipel@webrtc.org,oprypin@webrtc.org,holmer@google.com,stefan@webrtc.org,tschumim@webrtc.org
# Not skipping CQ checks because original CL landed more than 1 days ago.
BUG=webrtc:7746
Review-Url: https://codereview.webrtc.org/2964213002
Cr-Commit-Position: refs/heads/master@{#18866}
Change plotting of detector state from offset and gamma to T and threshold.
BUG=None
Review-Url: https://codereview.webrtc.org/2933243003
Cr-Commit-Position: refs/heads/master@{#18585}
Specifically, just like SafeMin() and SafeMax() it handles all
combinations of integer and all
combinations of floating-point arguments by picking a
result type that is guaranteed to be able to hold the result.
This CL also replaces a bunch of std::min + std:max call pairs with
calls to SafeClamp()---the ones that could easily be found by grep
because "min" and "max" were on the same line. :-)
BUG=webrtc:7459
Review-Url: https://codereview.webrtc.org/2808513003
Cr-Commit-Position: refs/heads/master@{#18542}
Code should implement BBR which is the congestion controlling algorithm. BBR tries to estimate two values bottle-neck bandwidth(bw) and round trip time(rtt),then use these two values to set two control parameters pacing rate(pacing_rate),the rate at which data should be sent and congestion window size (cwnd), cwnd is the upper bound for data in flight,data_in_flight <= cwnd at all time.
BBR has four modes:
1)Startup-ramping up throughput discovering estimated bw.
2)Drain-after Startup decrease throughput to drain queues.
3)Probe Bandwidth-most of the time BBR should be in this mode,
sending data at the rate of estimated bw, while sometimes trying to discover new bandwidth.
4)Probe Rtt-in this mode BBR tries to discover new rtt for the connection.
The key moment in BBR is when we receive feedback from the receiver,as this is the only moment which should effect our two estimators. At this moment all the switches between modes should happen, except switch to ProbeRtt mode (switching to ProbeRtt mode should happen when current min_rtt value expires).
This cl serves to emphasize the structure of Bbr, when switches happen and what key classes/functions should be implemented for proper functionality.
BUG=webrtc:7713
NOTRY=True
Review-Url: https://codereview.webrtc.org/2904183002
Cr-Commit-Position: refs/heads/master@{#18444}
Finally we are able to remove this class entirely, along with the last
vestiges of it's use. I've also removed some legacy files that were only
used for windows XP support.
BUG=webrtc:7035
Review-Url: https://codereview.webrtc.org/2790533002
Cr-Commit-Position: refs/heads/master@{#17480}
Mark ATTRIBUTE_UNUSED as deprecated since it only works with GCC and clang. I am not removing it now since typedefs.h is (perhaps incorrectly?) considered a public interface.
BUG=webrtc:7228
Review-Url: https://codereview.webrtc.org/2756483002
Cr-Commit-Position: refs/heads/master@{#17291}
In order to not make this CL too large I have broken it down into at least two
steps. Previous CL: https://codereview.chromium.org/2628563003/
webrtc::PacedSender::Process <--- previous CL start here
webrtc::PacedSender::SendPacket
webrtc::PacketRouter::TimeToSendPacket
webrtc::ModuleRtpRtcpImpl::TimeToSendPacket <--- previous CL end here, this Cl start here
webrtc::RTPSender::TimeToSendPacket
webrtc::RTPSender::PrepareAndSendPacket
webrtc::RTPSender::AddPacketToTransportFeedback
webrtc::TransportFeedbackAdapter::AddPacket
webrtc::SendTimeHistory::AddAndRemoveOld <--- this CL end here
BUG=webrtc:6822
Review-Url: https://codereview.webrtc.org/2708873003
Cr-Commit-Position: refs/heads/master@{#16796}
In order to not make this CL too large I have broken it down into at least two steps. In this CL we only propagate the pacing information part of the way:
webrtc::PacedSender::Process <--- propagate from here
webrtc::PacedSender::SendPacket
webrtc::PacketRouter::TimeToSendPacket
webrtc::ModuleRtpRtcpImpl::TimeToSendPacket <--- to here
webrtc::RTPSender::TimeToSendPacket
webrtc::RTPSender::PrepareAndSendPacket
webrtc::RTPSender::AddPacketToTransportFeedback
webrtc::TransportFeedbackAdapter::AddPacket
webrtc::SendTimeHistory::AddAndRemoveOld <--- goal is to propagte it here
BUG=webrtc:6822
Review-Url: https://codereview.webrtc.org/2628563003
Cr-Commit-Position: refs/heads/master@{#16664}
Bulk of the changes were done using
git grep -l '#include "webrtc/base/common.h"' | \
xargs sed -i '\,^#include.*webrtc/base/common\.h,d'
followed by adding back the include in the few places where it is
still needed, and in one case (pseudotcp.cc) instead deleting its use
of RTC_UNUSED.
BUG=webrtc:6424
Review-Url: https://codereview.webrtc.org/2644103002
Cr-Commit-Position: refs/heads/master@{#16263}
Theil and Sen's estimator essentially looks at the line through every pair of points and selects the median slope. This is robust to corruption of up to 29% of the data points.
Wire up new estimator to field trial experiment. Add unit and integration tests. Results are promising.
BUG=webrtc:6728
Review-Url: https://codereview.webrtc.org/2512693002
Cr-Commit-Position: refs/heads/master@{#15508}
AdaptiveVideoSource is used in testing/simulations of the bandwidth estimator.
Nada's reaction to delay depends on the current bitrate and the configured max rate in a non-intuituve way. Increase the starting bitrate to compensate for the increased max bitrate. This is only used in unit tests.
BUG=webrtc:6807
# Presubmit warns about a lint error in bwe.h that's unrelated to my change. Fixing it is beyond the scope of this CL.
NOPRESUBMIT=true
Review-Url: https://codereview.webrtc.org/2542843003
Cr-Commit-Position: refs/heads/master@{#15364}
Parse the estimation parameters from the field trial string.
BUG=webrtc:6690
Review-Url: https://codereview.webrtc.org/2489323002
Cr-Commit-Position: refs/heads/master@{#15126}
The new format for plot lines is PLOT <plot_no> <var_name>:<ssrc>@<alg_name> <time> <value>. The var_name is no longer prefixed by the context/tag (which most of the time was just the same as the test name.)
Update plot_dynamics.py script (which didn't work) to visualize the new BWE_TEST_LOGGING_PLOT lines.
BUG=webrtc:6621
Review-Url: https://codereview.webrtc.org/2456373002
Cr-Commit-Position: refs/heads/master@{#14983}
Rename kFullSendSideEstimator -> kSendSideEstimator and add new class SendSideBweSender (controlled by kSendSideEstimator) that actually uses the send side BWE.
Move the mock to logging/rtc_event_log/mock.
Allow congestion_controller, remote_bitrate_estimator and audio to depend on loggging/rtc_event_log
BUG=webrtc:6526
NOPRESUBMIT=True
Review-Url: https://codereview.webrtc.org/2431093003
Cr-Commit-Position: refs/heads/master@{#14772}
This includes if RTCP is received, but the number of packets received by the
other end hasn't increased.
Further, if no RTCP is received for more than 3 feedback intervals (3 seconds)
we start reducing the estimate by 20%. This is put under an experiment.
BUG=webrtc:6238
R=terelius@webrtc.org
Review URL: https://codereview.webrtc.org/2262213002 .
Cr-Commit-Position: refs/heads/master@{#14306}
This patch enables bwe related variable logging to the command line.
This is useful to test congestion control algorithm over real networks.
NOTRY=true
Review-Url: https://codereview.webrtc.org/2296253002
Cr-Commit-Position: refs/heads/master@{#14209}
Also lowering the min bitrate for simulations to be able to better capture this issue in the BweFeedbackTest.Choke200kbps30kbps200kbps performance test.
BUG=webrtc:6105
NOTRY=true
NOPRESUBMIT=true
Review-Url: https://codereview.webrtc.org/2201093006
Cr-Commit-Position: refs/heads/master@{#13639}
