Confirmed as occurring on windows as well:

Windows 10 build 1703

Tested on reported version Dev #62.0.3178.0 on Mac 10.12.6, Windows 7, Ubuntu 14.04 and able to reproduce the issue. However, issue is not reproduced on latest Canary and Stable.

Using the reverse bisect providing the bisect results,

Good Build: 62.0.3179.0  (Revision: 492477).
Bad Build: 62.0.3178.0  (Revision: 492239).

You are probably looking for a change made after 492340 (known good), but no later than 492341 (first known bad).

CHANGELOG URL:
The script might not always return single CL as suspect as some perf builds might get missing due to failure.
https://chromium.googlesource.com/chromium/src/+log/6fae0475a09388e17c07ba6fed383bb1e0eead2a..2320aa28460a4fd3da9e803b1fd4130958115aa7

Review ON: https://chromium-review.googlesource.com/603848

@maxmorin --  Could you please look into the issue, and merge the changes made to Dev.

Thanks.

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/9d8da709a1c467142a153d87b10ac6f18f36955f

commit 9d8da709a1c467142a153d87b10ac6f18f36955f
Author: Max Morin <maxmorin@chromium.org>
Date: Fri Aug 11 09:25:49 2017

[3178 merge] Roll WebRTC 19232:19254 (19 commits)

Changes: https://chromium.googlesource.com/external/webrtc/trunk/webrtc.git/+log/262f12e..375c817

$ git log 262f12e..375c817 --date=short --no-merges --format=%ad %ae %s
2017-08-07 brandtr@webrtc.org Minor improvements to VideoProcessor and corresponding test.
2017-08-07 brandtr@webrtc.org Rename WEBRTC_VIDEOPROCESSOR_H264_TESTS define to WEBRTC_USE_H264.
2017-08-07 asapersson@webrtc.org Remove unused members in MediaOptimization.
2017-08-04 deadbeef@webrtc.org Make Port (and subclasses) fully "Network"-based, instead of IP-based.
2017-08-04 kthelgason@webrtc.org Destroy compression session instead of reset it on release.
2017-08-04 philipel@webrtc.org Reland of Fix off-by-one bugs in video_coding::PacketBuffer when the buffer is filled with a single frame. (patchset #1 id:1 of https://codereview.chromium.org/2990183002/ )
2017-08-04 eladalon@webrtc.org Reland of SSRC and RSID may only refer to one sink each in RtpDemuxer (patchset #1 id:1 of https://codereview.webrtc.org/2993633002/ )
2017-08-04 srte@webrtc.org Renamed fields in common_types.h/RtcpStatistics.
2017-08-04 mflodman@webrtc.org Remove video_coding/codecs/OWNERS.
2017-08-04 magjed@webrtc.org ObjC: Support non-native frames in encoder
2017-08-03 deadbeef@webrtc.org Ignore "b=AS:-1" instead of treating as a hard error.
2017-08-03 zstein@webrtc.org JNI wrapper for PeerConnection::SetBitrate.
2017-08-03 mbonadei@webrtc.org Removing unused declared arg
2017-08-03 deadbeef@webrtc.org Relanding: Break peerconnection_jni.cc into multiple files, in "pc" directory.
2017-08-03 zhihuang@webrtc.org Revert of SSRC and RSID may only refer to one sink each in RtpDemuxer (patchset #15 id:280001 of https://codereview.webrtc.org/2968693002/ )
2017-08-03 mflodman@webrtc.org Rename ViEEncoder to VideoStreamEncoder
2017-08-03 eladalon@webrtc.org Mark ~DirectTransport with "override."
2017-08-03 mbonadei@webrtc.org Tracking mock_process_thread with a GN target
2017-08-03 eladalon@webrtc.org SSRC and RSID may only refer to one sink each in RtpDemuxer

TBR=maxmorin@chromium.org
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_chromium_archive_rel_ng;master.tryserver.chromium.mac:mac_chromium_archive_rel_ng

(cherry picked from commit 2320aa28460a4fd3da9e803b1fd4130958115aa7)

Bug:  753687 
Change-Id: Ia6abf9078a49ff2c70f05ea08b5e69ee3f20b9e1
Reviewed-on: https://chromium-review.googlesource.com/603848
Reviewed-by: Max Morin <maxmorin@chromium.org>
Commit-Queue: Max Morin <maxmorin@chromium.org>
Cr-Original-Commit-Position: refs/heads/master@{#492341}
Reviewed-on: https://chromium-review.googlesource.com/611986
Cr-Commit-Position: refs/branch-heads/3178@{#3}
Cr-Branched-From: cdd15784955039742fe9a8235581922d41b82d78-refs/heads/master@{#492239}
[modify] https://crrev.com/9d8da709a1c467142a153d87b10ac6f18f36955f/DEPS

Apparently, that's not how it works for DEPS changes. Will try again.

Ok, I made a CL for merging to the dev branch, but it requires manual review. Bug will be updated when it lands.

Never merged to dev, but I assume this problem doesn't reproduce now? Reopen if it does.