| Summary: | Repatch should support setters and plant calls to them directly | ||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Product: | WebKit | Reporter: | Filip Pizlo <fpizlo> | ||||||||||||||||
| Component: | JavaScriptCore | Assignee: | Filip Pizlo <fpizlo> | ||||||||||||||||
| Status: | RESOLVED FIXED | ||||||||||||||||||
| Severity: | Normal | CC: | barraclough, buildbot, ggaren, mark.lam, mhahnenberg, msaboff, oliver, ossy, rniwa, sam | ||||||||||||||||
| Priority: | P2 | ||||||||||||||||||
| Version: | 528+ (Nightly build) | ||||||||||||||||||
| Hardware: | All | ||||||||||||||||||
| OS: | All | ||||||||||||||||||
| Bug Depends on: | 131336 | ||||||||||||||||||
| Bug Blocks: | 129588 | ||||||||||||||||||
| Attachments: |
|
||||||||||||||||||
|
Description
Filip Pizlo
2014-03-25 15:46:09 PDT
Created attachment 228800 [details]
almost done
Created attachment 228802 [details]
the patch
This is a 12x speed-up on setter microbenchmarks. Comment on attachment 228802 [details]
the patch
r=me
Can you beef up setter.js to force a cached setter to miss the cache due to a redefined setter?
(In reply to comment #4) > (From update of attachment 228802 [details]) > r=me > > Can you beef up setter.js to force a cached setter to miss the cache due to a redefined setter? Sure! Comment on attachment 228802 [details] the patch Attachment 228802 [details] did not pass mac-wk2-ews (mac-wk2): Output: http://webkit-queues.appspot.com/results/5760488755429376 New failing tests: platform/mac/fast/scrolling/scroll-iframe-latched-mainframe.html platform/mac/fast/scrolling/scroll-select-latched-mainframe.html js/regress/setter.html platform/mac/fast/scrolling/scroll-div-latched-mainframe.html Created attachment 228808 [details]
Archive of layout-test-results from webkit-ews-12 for mac-mountainlion-wk2
The attached test failures were seen while running run-webkit-tests on the mac-wk2-ews.
Bot: webkit-ews-12 Port: mac-mountainlion-wk2 Platform: Mac OS X 10.8.5
Comment on attachment 228802 [details] the patch Attachment 228802 [details] did not pass mac-ews (mac): Output: http://webkit-queues.appspot.com/results/5040769945567232 New failing tests: js/regress/setter.html Created attachment 228809 [details]
Archive of layout-test-results from webkit-ews-04 for mac-mountainlion
The attached test failures were seen while running run-webkit-tests on the mac-ews.
Bot: webkit-ews-04 Port: mac-mountainlion Platform: Mac OS X 10.8.5
Comment on attachment 228802 [details] the patch Attachment 228802 [details] did not pass mac-wk2-ews (mac-wk2): Output: http://webkit-queues.appspot.com/results/6551012919738368 New failing tests: platform/mac/fast/scrolling/scroll-iframe-latched-mainframe.html platform/mac/fast/scrolling/scroll-select-latched-mainframe.html js/regress/setter.html platform/mac/fast/scrolling/scroll-div-latched-mainframe.html Created attachment 228810 [details]
Archive of layout-test-results from webkit-ews-11 for mac-mountainlion-wk2
The attached test failures were seen while running run-webkit-tests on the mac-wk2-ews.
Bot: webkit-ews-11 Port: mac-mountainlion-wk2 Platform: Mac OS X 10.8.5
Comment on attachment 228802 [details] the patch Attachment 228802 [details] did not pass mac-ews (mac): Output: http://webkit-queues.appspot.com/results/6604913685561344 New failing tests: js/regress/setter.html Created attachment 228813 [details]
Archive of layout-test-results from webkit-ews-08 for mac-mountainlion
The attached test failures were seen while running run-webkit-tests on the mac-ews.
Bot: webkit-ews-08 Port: mac-mountainlion Platform: Mac OS X 10.8.5
Created attachment 228862 [details]
patch for landing
Landed in http://trac.webkit.org/changeset/166945 (In reply to comment #15) > Landed in http://trac.webkit.org/changeset/166945 It broke 18 tests on 32 bit: http://build.webkit.org/builders/Apple%20Mavericks%2032-bit%20JSC%20%28BuildAndTest%29/builds/248 ** The following JSC stress test failures have been introduced: mozilla-tests.yaml/ecma/ExecutionContexts/10.1.6.js.mozilla-baseline mozilla-tests.yaml/ecma/ExecutionContexts/10.1.6.js.mozilla-dfg-eager-no-cjit-validate-phases mozilla-tests.yaml/ecma/Expressions/11.6.1-1.js.mozilla mozilla-tests.yaml/ecma/Expressions/11.6.1-1.js.mozilla-baseline mozilla-tests.yaml/ecma/Expressions/11.6.1-2.js.mozilla-baseline mozilla-tests.yaml/ecma/Expressions/11.6.2-1.js.mozilla-baseline mozilla-tests.yaml/ecma/Types/8.6.2.1-1.js.mozilla-baseline stress/setter.js.default stress/setter.js.no-llint stress/setter.js.always-trigger-copy-phase stress/setter.js.no-cjit-validate-phases regress/script-tests/assign-custom-setter-polymorphic.js.no-llint regress/script-tests/assign-custom-setter.js.no-llint regress/script-tests/assign-custom-setter.js.dfg-eager regress/script-tests/assign-custom-setter.js.dfg-eager-no-cjit-validate jsc-layout-tests.yaml/js/script-tests/for-in-cached.js.layout jsc-layout-tests.yaml/js/script-tests/for-in-cached.js.layout-no-llint jsc-layout-tests.yaml/js/script-tests/for-in-cached.js.layout-no-cjit Results for JSC stress tests: 18 failures found. (In reply to comment #16) > (In reply to comment #15) > > Landed in http://trac.webkit.org/changeset/166945 > It broke 18 tests on 32 bit: > http://build.webkit.org/builders/Apple%20Mavericks%2032-bit%20JSC%20%28BuildAndTest%29/builds/248 > > ** The following JSC stress test failures have been introduced: > mozilla-tests.yaml/ecma/ExecutionContexts/10.1.6.js.mozilla-baseline > mozilla-tests.yaml/ecma/ExecutionContexts/10.1.6.js.mozilla-dfg-eager-no-cjit-validate-phases > mozilla-tests.yaml/ecma/Expressions/11.6.1-1.js.mozilla > mozilla-tests.yaml/ecma/Expressions/11.6.1-1.js.mozilla-baseline > mozilla-tests.yaml/ecma/Expressions/11.6.1-2.js.mozilla-baseline > mozilla-tests.yaml/ecma/Expressions/11.6.2-1.js.mozilla-baseline > mozilla-tests.yaml/ecma/Types/8.6.2.1-1.js.mozilla-baseline > stress/setter.js.default > stress/setter.js.no-llint > stress/setter.js.always-trigger-copy-phase > stress/setter.js.no-cjit-validate-phases > regress/script-tests/assign-custom-setter-polymorphic.js.no-llint > regress/script-tests/assign-custom-setter.js.no-llint > regress/script-tests/assign-custom-setter.js.dfg-eager > regress/script-tests/assign-custom-setter.js.dfg-eager-no-cjit-validate > jsc-layout-tests.yaml/js/script-tests/for-in-cached.js.layout > jsc-layout-tests.yaml/js/script-tests/for-in-cached.js.layout-no-llint > jsc-layout-tests.yaml/js/script-tests/for-in-cached.js.layout-no-cjit > > Results for JSC stress tests: > 18 failures found. I am looking. (In reply to comment #17) > (In reply to comment #16) > > (In reply to comment #15) > > > Landed in http://trac.webkit.org/changeset/166945 > > It broke 18 tests on 32 bit: > > http://build.webkit.org/builders/Apple%20Mavericks%2032-bit%20JSC%20%28BuildAndTest%29/builds/248 > > > > ** The following JSC stress test failures have been introduced: > > mozilla-tests.yaml/ecma/ExecutionContexts/10.1.6.js.mozilla-baseline > > mozilla-tests.yaml/ecma/ExecutionContexts/10.1.6.js.mozilla-dfg-eager-no-cjit-validate-phases > > mozilla-tests.yaml/ecma/Expressions/11.6.1-1.js.mozilla > > mozilla-tests.yaml/ecma/Expressions/11.6.1-1.js.mozilla-baseline > > mozilla-tests.yaml/ecma/Expressions/11.6.1-2.js.mozilla-baseline > > mozilla-tests.yaml/ecma/Expressions/11.6.2-1.js.mozilla-baseline > > mozilla-tests.yaml/ecma/Types/8.6.2.1-1.js.mozilla-baseline > > stress/setter.js.default > > stress/setter.js.no-llint > > stress/setter.js.always-trigger-copy-phase > > stress/setter.js.no-cjit-validate-phases > > regress/script-tests/assign-custom-setter-polymorphic.js.no-llint > > regress/script-tests/assign-custom-setter.js.no-llint > > regress/script-tests/assign-custom-setter.js.dfg-eager > > regress/script-tests/assign-custom-setter.js.dfg-eager-no-cjit-validate > > jsc-layout-tests.yaml/js/script-tests/for-in-cached.js.layout > > jsc-layout-tests.yaml/js/script-tests/for-in-cached.js.layout-no-llint > > jsc-layout-tests.yaml/js/script-tests/for-in-cached.js.layout-no-cjit > > > > Results for JSC stress tests: > > 18 failures found. > > I am looking. Does not repro in release for me. Trying debug now... (In reply to comment #18) > (In reply to comment #17) > > (In reply to comment #16) > > > (In reply to comment #15) > > > > Landed in http://trac.webkit.org/changeset/166945 > > > It broke 18 tests on 32 bit: > > > http://build.webkit.org/builders/Apple%20Mavericks%2032-bit%20JSC%20%28BuildAndTest%29/builds/248 > > > > > > ** The following JSC stress test failures have been introduced: > > > mozilla-tests.yaml/ecma/ExecutionContexts/10.1.6.js.mozilla-baseline > > > mozilla-tests.yaml/ecma/ExecutionContexts/10.1.6.js.mozilla-dfg-eager-no-cjit-validate-phases > > > mozilla-tests.yaml/ecma/Expressions/11.6.1-1.js.mozilla > > > mozilla-tests.yaml/ecma/Expressions/11.6.1-1.js.mozilla-baseline > > > mozilla-tests.yaml/ecma/Expressions/11.6.1-2.js.mozilla-baseline > > > mozilla-tests.yaml/ecma/Expressions/11.6.2-1.js.mozilla-baseline > > > mozilla-tests.yaml/ecma/Types/8.6.2.1-1.js.mozilla-baseline > > > stress/setter.js.default > > > stress/setter.js.no-llint > > > stress/setter.js.always-trigger-copy-phase > > > stress/setter.js.no-cjit-validate-phases > > > regress/script-tests/assign-custom-setter-polymorphic.js.no-llint > > > regress/script-tests/assign-custom-setter.js.no-llint > > > regress/script-tests/assign-custom-setter.js.dfg-eager > > > regress/script-tests/assign-custom-setter.js.dfg-eager-no-cjit-validate > > > jsc-layout-tests.yaml/js/script-tests/for-in-cached.js.layout > > > jsc-layout-tests.yaml/js/script-tests/for-in-cached.js.layout-no-llint > > > jsc-layout-tests.yaml/js/script-tests/for-in-cached.js.layout-no-cjit > > > > > > Results for JSC stress tests: > > > 18 failures found. > > > > I am looking. > > Does not repro in release for me. Trying debug now... Oh yuck!! That assertion probably affects 64-bit too. It's a dumb assertion, will land fix shortly... (In reply to comment #19) > (In reply to comment #18) > > (In reply to comment #17) > > > (In reply to comment #16) > > > > (In reply to comment #15) > > > > > Landed in http://trac.webkit.org/changeset/166945 > > > > It broke 18 tests on 32 bit: > > > > http://build.webkit.org/builders/Apple%20Mavericks%2032-bit%20JSC%20%28BuildAndTest%29/builds/248 > > > > > > > > ** The following JSC stress test failures have been introduced: > > > > mozilla-tests.yaml/ecma/ExecutionContexts/10.1.6.js.mozilla-baseline > > > > mozilla-tests.yaml/ecma/ExecutionContexts/10.1.6.js.mozilla-dfg-eager-no-cjit-validate-phases > > > > mozilla-tests.yaml/ecma/Expressions/11.6.1-1.js.mozilla > > > > mozilla-tests.yaml/ecma/Expressions/11.6.1-1.js.mozilla-baseline > > > > mozilla-tests.yaml/ecma/Expressions/11.6.1-2.js.mozilla-baseline > > > > mozilla-tests.yaml/ecma/Expressions/11.6.2-1.js.mozilla-baseline > > > > mozilla-tests.yaml/ecma/Types/8.6.2.1-1.js.mozilla-baseline > > > > stress/setter.js.default > > > > stress/setter.js.no-llint > > > > stress/setter.js.always-trigger-copy-phase > > > > stress/setter.js.no-cjit-validate-phases > > > > regress/script-tests/assign-custom-setter-polymorphic.js.no-llint > > > > regress/script-tests/assign-custom-setter.js.no-llint > > > > regress/script-tests/assign-custom-setter.js.dfg-eager > > > > regress/script-tests/assign-custom-setter.js.dfg-eager-no-cjit-validate > > > > jsc-layout-tests.yaml/js/script-tests/for-in-cached.js.layout > > > > jsc-layout-tests.yaml/js/script-tests/for-in-cached.js.layout-no-llint > > > > jsc-layout-tests.yaml/js/script-tests/for-in-cached.js.layout-no-cjit > > > > > > > > Results for JSC stress tests: > > > > 18 failures found. > > > > > > I am looking. > > > > Does not repro in release for me. Trying debug now... > > Oh yuck!! That assertion probably affects 64-bit too. It's a dumb assertion, will land fix shortly... Fixed in http://trac.webkit.org/changeset/166952 |