Bug 175347 – JSTests/slowMicrobenchmarks/spread-small-array.js fails with JSC_validateExceptionChecks=1

• Status: RESOLVED
• Resolution: FIXED
• Component: JavaScriptCore

[Robin Morisset]

This is due to finishCreation not explicitely checking for exceptions between the calls to setConstantRegisters and setConstantIdentifiersSetRegisters.

[Robin Morisset]

Created attachment 317626 [details]
Proposed patch

[Build Bot]

Attachment 317626 [details] did not pass style-queue:


ERROR: Source/JavaScriptCore/ChangeLog:4:  Line contains tab character.  [whitespace/tab] [5]
ERROR: Source/JavaScriptCore/ChangeLog:6:  Line contains tab character.  [whitespace/tab] [5]
ERROR: Source/JavaScriptCore/ChangeLog:7:  Line contains tab character.  [whitespace/tab] [5]
ERROR: Source/JavaScriptCore/ChangeLog:8:  Line contains tab character.  [whitespace/tab] [5]
ERROR: Source/JavaScriptCore/ChangeLog:9:  Line contains tab character.  [whitespace/tab] [5]
ERROR: Source/JavaScriptCore/ChangeLog:10:  Line contains tab character.  [whitespace/tab] [5]
ERROR: Source/JavaScriptCore/bytecode/CodeBlock.cpp:881:  Missing space after ,  [whitespace/comma] [3]
ERROR: Source/JavaScriptCore/bytecode/CodeBlock.cpp:883:  Missing space after ,  [whitespace/comma] [3]
ERROR: Source/JavaScriptCore/bytecode/CodeBlock.cpp:889:  Missing space after ,  [whitespace/comma] [3]
ERROR: Source/JavaScriptCore/bytecode/CodeBlock.cpp:922:  Missing space after ,  [whitespace/comma] [3]
Total errors found: 10 in 4 files


If any of these errors are false positives, please file a bug against check-webkit-style.

[Robin Morisset]

Created attachment 317631 [details]
Same patch, with the whitespace issues fixed

[Saam Barati]

Comment on attachment 317631 [details]
Same patch, with the whitespace issues fixed

View in context: https://bugs.webkit.org/attachment.cgi?id=317631&action=review

r=me with changelog fix

> Source/JavaScriptCore/ChangeLog:10
> +        This is done by making finishCreation explicitely check for exceptions after setConstantRegister and setConstantIdentifiersSetRegisters.
> +        I chose to have this check replace the boolean returned previously by these functions for readability. The performance impact should be
> +        negligible considering how much more finishCreation does.
> +        This fix then caused another issue to appear as it was now clear that finishCreation can throw. And since it is called by ProgramCodeBlock::create(),
> +        FunctionCodeBlock::create() and friends, that are in turn called by ScriptExecutable::newCodeBlockFor, this last function also required a few tweaks.

Style nit: This explanation should go below the "Reviewed by ..."

[Build Bot]

Attachment 317631 [details] did not pass style-queue:


ERROR: Source/JavaScriptCore/bytecode/CodeBlock.cpp:881:  Extra space before )  [whitespace/parens] [2]
ERROR: Source/JavaScriptCore/bytecode/CodeBlock.cpp:883:  Extra space before )  [whitespace/parens] [2]
ERROR: Source/JavaScriptCore/bytecode/CodeBlock.cpp:889:  Extra space before )  [whitespace/parens] [2]
ERROR: Source/JavaScriptCore/bytecode/CodeBlock.cpp:922:  Extra space before )  [whitespace/parens] [2]
Total errors found: 4 in 4 files


If any of these errors are false positives, please file a bug against check-webkit-style.

[Robin Morisset]

Created attachment 317634 [details]
Same patch, with Changelog fix

[Build Bot]

Attachment 317634 [details] did not pass style-queue:


ERROR: Source/JavaScriptCore/bytecode/CodeBlock.cpp:881:  Extra space before )  [whitespace/parens] [2]
ERROR: Source/JavaScriptCore/bytecode/CodeBlock.cpp:883:  Extra space before )  [whitespace/parens] [2]
ERROR: Source/JavaScriptCore/bytecode/CodeBlock.cpp:889:  Extra space before )  [whitespace/parens] [2]
ERROR: Source/JavaScriptCore/bytecode/CodeBlock.cpp:922:  Extra space before )  [whitespace/parens] [2]
Total errors found: 4 in 4 files


If any of these errors are false positives, please file a bug against check-webkit-style.

[Mark Lam]

Comment on attachment 317634 [details]
Same patch, with Changelog fix

View in context: https://bugs.webkit.org/attachment.cgi?id=317634&action=review

> Source/JavaScriptCore/bytecode/CodeBlock.cpp:881
> +        RETURN_IF_EXCEPTION(scope, );

This looks weird.  Let's do this instead:
     RETURN_IF_EXCEPTION(scope, void());

> Source/JavaScriptCore/bytecode/CodeBlock.cpp:883
> +        RETURN_IF_EXCEPTION(scope, );

Ditto.

> Source/JavaScriptCore/bytecode/CodeBlock.cpp:889
> +            RETURN_IF_EXCEPTION(scope, );

Ditto.

> Source/JavaScriptCore/bytecode/CodeBlock.cpp:922
> +                RETURN_IF_EXCEPTION(scope, );

Ditto.

[Robin Morisset]

Created attachment 317638 [details]
Same patch, with Mark's suggestion

[Robin Morisset]

Created attachment 317641 [details]
exception_scope => exceptionScope, as suggested by Mark

[Mark Lam]

Comment on attachment 317641 [details]
exception_scope => exceptionScope, as suggested by Mark

View in context: https://bugs.webkit.org/attachment.cgi?id=317641&action=review

> Source/JavaScriptCore/bytecode/CodeBlock.cpp:404
> +    auto exceptionScope = DECLARE_THROW_SCOPE(vm);

can we call this throwScope instead since it comes from DECLARE_THROW_SCOPE?

[Robin Morisset]

Created attachment 317648 [details]
exceptionScope => throwScope

[WebKit Commit Bot]

Comment on attachment 317648 [details]
exceptionScope => throwScope

Clearing flags on attachment: 317648

Committed r220432: <http://trac.webkit.org/changeset/220432>

[WebKit Commit Bot]

All reviewed patches have been landed.  Closing bug.