All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=856361

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=7b8d36e05a70e613b31a9cc75f38fe1a887c3d261dfbb9e918cfc1b60190defa


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

mojo-linux-perf

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/14dcc71f240000

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/148e3877240000

😿 Pinpoint job stopped with an error.
https://pinpoint-dot-chromeperf.appspot.com/job/14dcc71f240000

All of the runs failed. The most common error (4/20 runs) was:
  main at /b/s/w/iTraceback (most recent call last):

😿 Pinpoint job stopped with an error.
https://pinpoint-dot-chromeperf.appspot.com/job/148e3877240000

All of the runs failed. The most common error (5/20 runs) was:
  main at /b/s/Traceback (most recent call last):

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/118a84af240000

📍 Found significant differences after each of 2 commits.
https://pinpoint-dot-chromeperf.appspot.com/job/118a84af240000

Remove blanks facility from the library by behdad@behdad.org
https://chromium.googlesource.com/external/fontconfig/+/8f4c4d278d013f6cc69ba7d7bf0f8aed11398dfb
331 → No values

Reland "Reland "Update fontconfig to 6cc99d6a"" by thomasanderson@chromium.org
https://chromium.googlesource.com/chromium/src/+/a0c1584a2fade0146b7cd3380abe42feda02ad7c
No values → 152.6

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions

#c7 doesn't seem to be related. Let's try again.

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/14fc100b240000

📍 Found significant differences after each of 2 commits.
https://pinpoint-dot-chromeperf.appspot.com/job/14fc100b240000

Remove blanks facility from the library by behdad@behdad.org
https://chromium.googlesource.com/external/fontconfig/+/8f4c4d278d013f6cc69ba7d7bf0f8aed11398dfb
324.9 → No values

Reland "Reland "Update fontconfig to 6cc99d6a"" by thomasanderson@chromium.org
https://chromium.googlesource.com/chromium/src/+/a0c1584a2fade0146b7cd3380abe42feda02ad7c
No values → 163.7

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions

behdad@ I've noticed a few of these issues recently.

The fontconfig commit in question is the one that removes <blank>.  On some older OSes like Trusty (which the Chrome CI runs on), this message gets printed to the console "Fontconfig warning: "/etc/fonts/fonts.conf", line 146: blank doesn't take any effect anymore. please remove it from your fonts.conf".

Not sure why removing this would cause a performance regression, though.

Re #c11: I think the problem is chromium doesn't build between those two commits.
e.g. With the following error message:
```
ninja: Entering directory `/b/c/b/Linux_Builder_Perf/src/out/Release'
ninja: error: '../../third_party/fontconfig/src/src/fcblanks.c', needed by 'obj/third_party/fontconfig/fontconfig/fcblanks.o', missing and no known rule to make it
```

Thus the bisect bot cannot go deep into it, and assumed your commit to be the culprit.

Took some look but still not sure how 'external/fontconfig/' works...
+A few people appeared between commits mentioned in #c10.

Note:
This is actually a loading performance improvement. The bug title says 'regression' because we didn't see the similar improvement with Network Service feature enabled, and we'd like to understand how to get the similar improvement within our feature.

I'm not sure there's anything to be figured out here.  This sounds like bot noise to me.

WontFix according to #c14.