All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=607963

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICglIziswoM


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

Google Chrome Linux x64

rkc: looks like https://chromium.googlesource.com/chromium/src/+/a7a149d373a58c0e324353175feb0cefb0cb9c8f regress the Chrome size by 100K. Can you verify this?

Friendly Perf Sheriff Ping.
rkc@, could you please check this, if you are not the right person to handle this, please revert back and/or cc anyone who might know. 

Hey Prasad, I did take a look but nothing seemed to jump out. I'll have to spend more time removing parts of code to figure out what is causing this. I plan on working on this early next week.

So the only way I see this is if I am building with symbols on the Chrome binary. In that case, yes, symbols do increase the size by about 100k, which is expected since the CL does add a lot of code.

When compiling without symbols, these are the sizes I see,
At commit ae57706988fb492736785ff91a29478f03accc8b,
166434232

At commit 451dea5a213091d3bc3c90da0433d325739ec5a6,
166436600

A total difference of 2k. If there is any new data that shows otherwise (100k size increase, with a build without symbols), please re-open this. Please do include the GN flags needed to do the build though. This is what I am getting with a default release linux non-component build.