All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=709364

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyeAsSE1N0b3BwYWdlQWxlcnRQYXJlbnQiSldlYlJUQ1BlcmYvd2VicnRjLWxpbnV4LWxhcmdlLXRlc3RzL2lzYWNfZml4X3Rlc3QvaXNhYy90aW1lX3Blcl8xMG1zX2ZyYW1lDAsSDVN0b3BwYWdlQWxlcnQY6IQBDA


Bot(s) for this bug's original alert(s):

webrtc-linux-large-tests

hlundin@: Could you please take a look to see whether this increase makes sense.

This is not an increase, it is a data stoppage alert.

kjellander@, can you help me figure out why this data stopped? The printout of the metric "time_per_10ms_frame" happens in webrtc/modules/audio_coding/codecs/isac/fix/test/kenny.cc, which is included in the isac_fix_test binary. Do we not run that on but any more? Who killed Kenny?

Also, I cannot access the bot stdout from the data points in the graph.

This has been reported in https://bugs.chromium.org/p/webrtc/issues/detail?id=7307 since March 8. Still not fixed though.

Of course. Thanks for connecting the dots for me.

The following revision refers to this bug:
  https://chromium.googlesource.com/external/webrtc.git/+/49cad02cf38dc54f8ba0bdab28c3dbd21f1d4e81

commit 49cad02cf38dc54f8ba0bdab28c3dbd21f1d4e81
Author: kwiberg <kwiberg@webrtc.org>
Date: Mon Apr 10 09:29:33 2017

Ignore some UBSan errors

They proved to be too difficult to fix properly, so we revert the
saturation fixes that were done in
https://codereview.webrtc.org/2729573002 and
https://codereview.webrtc.org/2746903002, and ignore them instead.

BUG= webrtc:7307 ,  chromium:709364 ,  chromium:693868 

Review-Url: https://codereview.webrtc.org/2809483002
Cr-Commit-Position: refs/heads/master@{#17612}

[modify] https://crrev.com/49cad02cf38dc54f8ba0bdab28c3dbd21f1d4e81/webrtc/common_audio/signal_processing/include/spl_inl.h
[modify] https://crrev.com/49cad02cf38dc54f8ba0bdab28c3dbd21f1d4e81/webrtc/modules/audio_coding/codecs/isac/fix/source/entropy_coding.c

This should be fixed as of the CL in comment #9.

Checked the graph, and it is now alive again.