Chromium Code Reviews
chromiumcodereview-hr@appspot.gserviceaccount.com (chromiumcodereview-hr) | Please choose your nickname with Settings | Help | Chromium Project | Gerrit Changes | Sign out
(785)

Issues Search
    My Issues | Starred     Open | Closed | All

Issue 1053063003: Make --always-opt also optimize top-level code. (Closed)

Created:
5 years, 8 months ago by Michael Starzinger
Modified:
5 years, 8 months ago
Reviewers:
titzer, Toon Verwaest
CC:
v8-dev
Base URL:
https://chromium.googlesource.com/v8/v8.git@master
Target Ref:
refs/pending/heads/master
Project:
v8
Visibility:
Public.

Description

Make --always-opt also optimize top-level code. This enables eager optimization of top-level code with TurboFan and extends test coverage by triggering it with the --always-opt flag. Script contexts are now also properly allocated in TurboFan. R=titzer@chromium.org Committed: https://crrev.com/2d281e71ac49058ef5696f4255ff1fa30796bc5e Cr-Commit-Position: refs/heads/master@{#27633}

Patch Set 1 #

Patch Set 2 : Skip failing tests. #

Patch Set 3 : Skip failing tests. #

Patch Set 4 : Skip failing tests. #

Patch Set 5 : Skip failing tests. #

Total comments: 5
Unified diffs Side-by-side diffs Delta from patch set Stats (+122 lines, -37 lines) Patch
M src/arm/lithium-codegen-arm.cc View 1 chunk +1 line, -0 lines 0 comments Download
M src/arm64/lithium-codegen-arm64.cc View 1 chunk +1 line, -0 lines 0 comments Download
M src/bailout-reason.h View 2 chunks +2 lines, -1 line 0 comments Download
M src/compiler.cc View 1 chunk +18 lines, -0 lines 0 comments Download
M src/compiler/ast-graph-builder.h View 1 chunk +3 lines, -2 lines 0 comments Download
M src/compiler/ast-graph-builder.cc View 4 chunks +24 lines, -8 lines 0 comments Download
M src/compiler/js-generic-lowering.cc View 1 chunk +1 line, -1 line 0 comments Download
M src/compiler/linkage.cc View 1 chunk +0 lines, -1 line 0 comments Download
M src/compiler/operator-properties.cc View 1 chunk +1 line, -0 lines 0 comments Download
M src/factory.cc View 3 chunks +4 lines, -9 lines 0 comments Download
M src/hydrogen.cc View 2 chunks +6 lines, -1 line 3 comments Download
M src/ia32/lithium-codegen-ia32.cc View 1 chunk +1 line, -0 lines 0 comments Download
M src/mips/lithium-codegen-mips.cc View 1 chunk +1 line, -0 lines 0 comments Download
M src/x64/lithium-codegen-x64.cc View 1 chunk +1 line, -0 lines 0 comments Download
M test/cctest/cctest.status View 1 chunk +11 lines, -3 lines 0 comments Download
M test/cctest/test-api.cc View 2 chunks +7 lines, -2 lines 0 comments Download
M test/cctest/test-debug.cc View 3 chunks +7 lines, -5 lines 0 comments Download
M test/cctest/test-parsing.cc View 2 chunks +4 lines, -2 lines 0 comments Download
M test/cctest/test-thread-termination.cc View 3 chunks +3 lines, -0 lines 0 comments Download
M test/mjsunit/mjsunit.status View 1 2 3 4 3 chunks +14 lines, -1 line 0 comments Download
M test/mozilla/mozilla.status View 1 1 chunk +11 lines, -0 lines 2 comments Download
M test/unittests/compiler/js-operator-unittest.cc View 1 chunk +1 line, -1 line 0 comments Download

Messages

Total messages: 13 (2 generated)
Michael Starzinger
5 years, 8 months ago (2015-04-07 12:55:55 UTC) #1
titzer
On 2015/04/07 12:55:55, Michael Starzinger wrote:
5 years, 8 months ago (2015-04-07 14:43:29 UTC) #2
titzer
https://codereview.chromium.org/1053063003/diff/80001/test/mozilla/mozilla.status File test/mozilla/mozilla.status (right): https://codereview.chromium.org/1053063003/diff/80001/test/mozilla/mozilla.status#newcode62 test/mozilla/mozilla.status:62: 'ecma_3/Statements/regress-324650': [PASS, NO_VARIANTS], disable in turbofan variant only?
5 years, 8 months ago (2015-04-07 14:43:36 UTC) #3
titzer
On 2015/04/07 14:43:36, titzer wrote: > https://codereview.chromium.org/1053063003/diff/80001/test/mozilla/mozilla.status > File test/mozilla/mozilla.status (right): > > https://codereview.chromium.org/1053063003/diff/80001/test/mozilla/mozilla.status#newcode62 > ...
5 years, 8 months ago (2015-04-07 14:45:13 UTC) #4
Michael Starzinger
https://codereview.chromium.org/1053063003/diff/80001/test/mozilla/mozilla.status File test/mozilla/mozilla.status (right): https://codereview.chromium.org/1053063003/diff/80001/test/mozilla/mozilla.status#newcode62 test/mozilla/mozilla.status:62: 'ecma_3/Statements/regress-324650': [PASS, NO_VARIANTS], On 2015/04/07 14:43:36, titzer wrote: > ...
5 years, 8 months ago (2015-04-07 14:45:56 UTC) #5
commit-bot: I haz the power
CQ is trying da patch. Follow status at https://chromium-cq-status.appspot.com/patch-status/1053063003/80001
5 years, 8 months ago (2015-04-07 15:42:07 UTC) #7
commit-bot: I haz the power
Committed patchset #5 (id:80001)
5 years, 8 months ago (2015-04-07 15:44:27 UTC) #8
commit-bot: I haz the power
Patchset 5 (id:??) landed as https://crrev.com/2d281e71ac49058ef5696f4255ff1fa30796bc5e Cr-Commit-Position: refs/heads/master@{#27633}
5 years, 8 months ago (2015-04-07 15:44:32 UTC) #9
Toon Verwaest
https://codereview.chromium.org/1053063003/diff/80001/src/hydrogen.cc File src/hydrogen.cc (right): https://codereview.chromium.org/1053063003/diff/80001/src/hydrogen.cc#newcode4226 src/hydrogen.cc:4226: return false; Doesn't this entirely disable top-level OSR? If ...
5 years, 8 months ago (2015-04-07 20:55:05 UTC) #11
Michael Starzinger
https://codereview.chromium.org/1053063003/diff/80001/src/hydrogen.cc File src/hydrogen.cc (right): https://codereview.chromium.org/1053063003/diff/80001/src/hydrogen.cc#newcode4226 src/hydrogen.cc:4226: return false; On 2015/04/07 20:55:05, Toon Verwaest wrote: > ...
5 years, 8 months ago (2015-04-07 23:01:03 UTC) #12
Michael Starzinger
5 years, 8 months ago (2015-04-08 12:29:18 UTC) #13
Message was sent while issue was closed. 
https://codereview.chromium.org/1053063003/diff/80001/src/hydrogen.cc
File src/hydrogen.cc (right):

https://codereview.chromium.org/1053063003/diff/80001/src/hydrogen.cc#newcode...
src/hydrogen.cc:4226: return false;
On 2015/04/07 23:01:03, Michael Starzinger wrote:
> On 2015/04/07 20:55:05, Toon Verwaest wrote:
> > Doesn't this entirely disable top-level OSR?
> > 
> > If I'm not mistaken (didn't bisect), it seems like this is causing a large
> > sunspider-bitwise-and regression; 2.1 -> 8.9ms.
> 
> It might be that this needs to be guarded by an additional "heap_slots > 0"
> check, I will investigate this tomorrow. Thanks for pointing this out.

For the record, addressed by: https://codereview.chromium.org/1063373003/

Powered by Google App Engine
This is Rietveld 408576698