All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=671351

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICg55K1rAkM


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

linux-release

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8994081985399628544

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8994081973447932976

=== PERF REGRESSION ===


=== Auto-CCing suspected CL author chrome-cron@google.com ===

Hi chrome-cron@google.com, the bisect results pointed to your CL, please take a look at the
results.


===== BISECT JOB RESULTS =====
Status: completed


===== SUSPECTED CL(s) =====
Subject : Updating trunk VERSION from 2930.0 to 2931.0
Author  : chrome-cron
Commit description:
  
Cr-Commit-Position: refs/heads/master@{#434300}
Commit  : b4651de830072d7a2de382f08947f76aa9b4fd39
Date    : Thu Nov 24 04:06:13 2016


===== TESTED REVISIONS =====
Revision         Mean      Std Dev    N  Good?
chromium@433543  0.175625  0.0320917  8  good
chromium@433992  0.18325   0.0245662  8  good
chromium@434217  0.184125  0.0284055  8  good
chromium@434273  0.1695    0.0272029  8  good
chromium@434287  0.1805    0.012      8  good
chromium@434294  0.198125  0.0568232  8  good
chromium@434298  0.181125  0.0165189  8  good
chromium@434299  0.179     0.060564   8  good
chromium@434300  0.6366    0.0500719  5  bad    <--
chromium@434301  0.621     0.04998    5  bad
chromium@434329  0.6064    0.090737   5  bad
chromium@434441  0.6238    0.0484644  5  bad

Bisect job ran on: linux_perf_bisect
Bug ID: 671351

Test Command: ./src/out/Release/load_library_perf_tests --single-process-tests
Test Metric: time_to_load_library/libwidevinecdmadapter.so
Relative Change: 255.19%

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/linux_perf_bisect/builds/6885
Job details: https://chromeperf.appspot.com/buildbucket_job_status/8994081985399628544


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5295742942445568

| O O | Visit http://www.chromium.org/developers/speed-infra/perf-bug-faq
|  X  | for more information addressing perf regression bugs. For feedback,
| / \ | file a bug with component Tests>AutoBisect.  Thank you!

===== BISECT JOB RESULTS =====
Status: completed


===== SUSPECTED CL(s) =====
Subject : Updating trunk VERSION from 2930.0 to 2931.0
Author  : chrome-cron
Commit description:
  
Cr-Commit-Position: refs/heads/master@{#434300}
Commit  : b4651de830072d7a2de382f08947f76aa9b4fd39
Date    : Thu Nov 24 04:06:13 2016


===== TESTED REVISIONS =====
Revision         Mean      Std Dev     N  Good?
chromium@433543  0.18325   0.0204817   8  good
chromium@433992  0.221625  0.110163    8  good
chromium@434217  0.186875  0.0199719   8  good
chromium@434273  0.1785    0.0300333   8  good
chromium@434287  0.185375  0.0463883   8  good
chromium@434294  0.17625   0.00674537  8  good
chromium@434298  0.18275   0.0261821   8  good
chromium@434299  0.187375  0.0332246   8  good
chromium@434300  0.6176    0.0148728   5  bad    <--
chromium@434301  0.6372    0.0582477   5  bad
chromium@434329  0.6144    0.0103537   5  bad
chromium@434441  0.6618    0.00684105  5  bad

Bisect job ran on: linux_perf_bisect
Bug ID: 671351

Test Command: ./src/out/Release/load_library_perf_tests --single-process-tests
Test Metric: time_to_load_library/libwidevinecdmadapter.so
Relative Change: 261.15%

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/linux_perf_bisect/builds/6886
Job details: https://chromeperf.appspot.com/buildbucket_job_status/8994081973447932976


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6387789648625664

| O O | Visit http://www.chromium.org/developers/speed-infra/perf-bug-faq
|  X  | for more information addressing perf regression bugs. For feedback,
| / \ | file a bug with component Tests>AutoBisect.  Thank you!

We checked in a new CDM (version 1.4.8.962) into 2931.0. The CDM has some non-trivial changes which should be the cause of this regression.

That being said, the time to load the CDM is still super short on Linux; it went from 0.2ms on the old CDM to 0.6ms on the new CDM. With this, I don't feel it's worth extensive investigation. I'll mark this as WontFix.

hmchen: Just in case, do you have any clue what could contribute to this increase? If you don't see anything obvious, don't bother to investigate.

One possible reason may be the random number generator initialization.