All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=645378

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICg8beOoQoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgsbeiogkM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICg8bf9qQoM


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

linux
mac

hablich@, it looks like the V8 roll caused a ~250kb size regression on mac and linux. Can you help route this appropriately?

+cc mac/linux size owners as fyi

Is it possible to reproduce this on my machine? How do I need to build? Or is a normal Release build sufficient?

Try a non-component, release build.

machenbach@, we should be able to backtrack this via our build bucket binaries, right?

Meant Google Storage.

Clearly shows up with mac standalone:

Last revision where it's smaller:
https://chromium.googlesource.com/v8/v8/+/7b2a211b8e3bcf9422b2f1e87b729f8eff5207e8
https://isolateserver.appspot.com/browse?namespace=default-gzip&hash=ce5486c445ea6e6d51c5c436dc7827487db2b5c4

Range where it regressed:
https://chromium.googlesource.com/v8/v8/+log/7b2a211b8e3bcf9422b2f1e87b729f8eff5207e8..494e2493f2850d19d800d15890d26c077ed4ca3a
https://isolateserver.appspot.com/browse?namespace=default-gzip&hash=79a8554adeef307870488f2203daadd23dd1999c

It's probably one of the two (CC'ing authors):
[turbofan] put src/types.[h/cc] into src/compiler/types.[h/cc] by mvstanton
[RuntimeCallStats] Move tracing runtime instrumentation closer to the original version.

The second had a revert attempt, which didn't land. You could also build d8 on mac to repro.

ruby -e '`nm -S out/Release/d8 | grep " t " | cut -d" " -f2-`
    .lines().map{|l| l=l.split(" "); l[0]=l[0].to_i(16); l}
    .sort{|a,b| b[0]<=>a[0]}
    .each{|l| puts l.join(" ")}'

should to the trick to figure out the major offenders (which I can fix once v8 compiles...)

Seems like it is https://chromium.googlesource.com/v8/v8/+/e5ba156d887e08f3ae3d29aa35e669f19a7dcb97 according to my local build.

80143 t _ZN2v88internal8compiler13CodeGenerator23AssembleArchInstructionEPNS1_11InstructionE
60717 t _ZN2v88internal11ObjectStats4DumpERSt18basic_stringstreamIcSt11char_traitsIcESaIcEE
55190 t _ZN2v88internal11interpreter11Interpreter10InitializeEv
42026 t _ZN2v88internal11ObjectStats9PrintJSONEPKc
41903 t _ZN2v88internal11ObjectStats21CheckpointObjectStatsEv
36458 t _ZN2v88internal7Genesis16InitializeGlobalENS0_6HandleINS0_14JSGlobalObjectEEENS2_INS0_10JSFunctionEEENS0_17GlobalContextTypeE
32398 t _ZN2v88internal8compiler26MachineOperatorGlobalCacheC2Ev
31929 t _ZN2v88internal16RuntimeCallStats4DumpEv
30252 t _ZN2v88internal8compiler5Typer7Visitor6ReduceEPNS1_4NodeE
28014 t _ZN2v88internal8compiler22RepresentationSelector9VisitNodeEPNS1_4NodeENS1_10TruncationEPNS1_18SimplifiedLoweringE
27103 t _ZN2v88internal16RuntimeCallStatsC2Ev
23316 t _ZN2v88internal8CountersC2EPNS0_7IsolateE
23316 t _ZN2v88internal8CountersC1EPNS0_7IsolateE
21565 t _ZN2v88internal8Builtins5SetUpEPNS0_7IsolateEb
19121 t _ZN2v88internal16RuntimeCallStats5PrintERSo
17487 t _ZN6icu_5617DateFormatSymbols14initializeDataERKNS_6LocaleEPKcR10UErrorCodea
17353 t _ZN2v88internal4Heap20CreateInitialObjectsEv
14802 t _ZN2v88internal8compiler22MachineOperatorReducer6ReduceEPNS1_4NodeE
14667 t _ZN2v88internalL27Stats_Runtime_RegExpReplaceEiPPNS0_6ObjectEPNS0_7IsolateE
14566 t _ZN2v88internal19StatisticsExtension11GetCountersERKNS_20FunctionCallbackInfoINS_5ValueEEE
14266 t _ZN2v88internal21Runtime_RegExpReplaceEiPPNS0_6ObjectEPNS0_7IsolateE
14000 t _ZN2v88internal4wasm12_GLOBAL__N_113ModuleDecoder12DecodeModuleEPNS1_10WasmModuleEb
13727 t _ZN2v88internal8compiler13Int64Lowering9LowerNodeEPNS1_4NodeE
13480 t _ZN2v88internal8compiler19InstructionSelector9VisitNodeEPNS1_4NodeE
13200 t _ZN2v88internal8compiler29SimplifiedOperatorGlobalCacheC2Ev

The following revision refers to this bug:
  https://chromium.googlesource.com/v8/v8.git/+/f8654ce9f612ff3dc57e0b2d4d27fc9ea1ee0aab

commit f8654ce9f612ff3dc57e0b2d4d27fc9ea1ee0aab
Author: cbruni <cbruni@chromium.org>
Date: Mon Oct 17 11:09:39 2016

Do not inline object-stats functions to reduce binary size.

BUG= chromium:645378 

Review-Url: https://codereview.chromium.org/2422143002
Cr-Commit-Position: refs/heads/master@{#40355}

[modify] https://crrev.com/f8654ce9f612ff3dc57e0b2d4d27fc9ea1ee0aab/src/counters.cc
[modify] https://crrev.com/f8654ce9f612ff3dc57e0b2d4d27fc9ea1ee0aab/src/counters.h
[modify] https://crrev.com/f8654ce9f612ff3dc57e0b2d4d27fc9ea1ee0aab/src/heap/object-stats.cc
[modify] https://crrev.com/f8654ce9f612ff3dc57e0b2d4d27fc9ea1ee0aab/src/heap/object-stats.h

Please merge to 5.5 if the regression recovered.

[Automated comment] No milestone found on Merge-Request (i.e. merge-request-# label).

[Automated comment] There appears to be on-going work (i.e. bugroid changes), needs manual review.

[Automated comment] There appears to be on-going work (i.e. bugroid changes), needs manual review.

[Automated comment] There appears to be on-going work (i.e. bugroid changes), needs manual review.

[Automated comment] There appears to be on-going work (i.e. bugroid changes), needs manual review.

[Automated comment] There appears to be on-going work (i.e. bugroid changes), needs manual review.

[Automated comment] There appears to be on-going work (i.e. bugroid changes), needs manual review.

[Automated comment] There appears to be on-going work (i.e. bugroid changes), needs manual review.

[Automated comment] There appears to be on-going work (i.e. bugroid changes), needs manual review.

[Automated comment] There appears to be on-going work (i.e. bugroid changes), needs manual review.

[Automated comment] There appears to be on-going work (i.e. bugroid changes), needs manual review.

[Automated comment] There appears to be on-going work (i.e. bugroid changes), needs manual review.

[Automated comment] There appears to be on-going work (i.e. bugroid changes), needs manual review.

[Automated comment] There appears to be on-going work (i.e. bugroid changes), needs manual review.

[Automated comment] There appears to be on-going work (i.e. bugroid changes), needs manual review.

[Automated comment] There appears to be on-going work (i.e. bugroid changes), needs manual review.

[Automated comment] There appears to be on-going work (i.e. bugroid changes), needs manual review.

[Automated comment] There appears to be on-going work (i.e. bugroid changes), needs manual review.

[Automated comment] There appears to be on-going work (i.e. bugroid changes), needs manual review.

dimu: Your bot is going crazy. Can you please stop it? ^^^

[Automated comment] There appears to be on-going work (i.e. bugroid changes), needs manual review.

Is this resolved? The slider on Chromeperf does not work so I am not sure if the regression was recovered. It seems so but not sure.

Looking at the graphs in https://chromeperf.appspot.com/report?sid=1d1a5ae1c0dea3208b1873fc4d1850a42cd1aa3e102aa7a9934bf5360b49d0c5&start_rev=413443&end_rev=430229 I can see a downtick at reversion 417487 which involves runtime call stats changes.

I've been using the following script to find sizes of functions in the final binary.

ruby -e '`nm -S out/Release/d8 | grep " t " | cut -d" " -f2-`
    .lines().map{|l| l=l.split(" "); l[0]=l[0].to_i(16); l}
    .sort{|a,b| b[0]<=>a[0]}
    .each{|l| puts l.join(" ")}'