Fix clang build error in JITOpcodes32_64.cpp

[David Kilzer (:ddkilzer)]

JavaScriptCore/jit/JITOpcodes32_64.cpp:741:36:{741:9-741:35}: error: operator '?:' has lower precedence than '+'; '+' will be evaluated first [-Werror,-Wparentheses,3]
     map(m_bytecodeOffset + dynamic ? OPCODE_LENGTH(op_resolve_global_dynamic) : OPCODE_LENGTH(op_resolve_global), dst, regT1, regT0);
         ~~~~~~~~~~~~~~~~~~~~~~~~~~ ^
JavaScriptCore/jit/JITOpcodes32_64.cpp:741:36: note: place parentheses around the '+' expression to silence this warning [3] 
     map(m_bytecodeOffset + dynamic ? OPCODE_LENGTH(op_resolve_global_dynamic) : OPCODE_LENGTH(op_resolve_global), dst, regT1, regT0);
                                    ^           
         (                         )           
fix-it:"JavaScriptCore/jit/JITOpcodes32_64.cpp":{741:9-741:9}:"("
fix-it:"JavaScriptCore/jit/JITOpcodes32_64.cpp":{741:35-741:35}:")"
JavaScriptCore/jit/JITOpcodes32_64.cpp:741:36:{741:28-741:94}: note: place parentheses around the '?:' expression to evaluate it first [3]
     map(m_bytecodeOffset + dynamic ? OPCODE_LENGTH(op_resolve_global_dynamic) : OPCODE_LENGTH(op_resolve_global), dst, regT1, regT0);
                            ~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1 error generated.

[David Kilzer (:ddkilzer)]

Committed r90232: <http://trac.webkit.org/changeset/90232>

[David Kilzer (:ddkilzer)]

Created attachment 99482 [details]
Patch

[Darin Adler]

Looks to me like clang correctly spotted a programming error here, so there was a symptom of this too.

[David Kilzer (:ddkilzer)]

This code was originally introduced in r58991 in JavaScriptCore/jit/JITOpcodes.cpp.

<http://trac.webkit.org/changeset/58991>

[Alexey Proskuryakov]

How could this be landed without review, and without a regression test?

[Alexey Proskuryakov]

I have an answer to my question - probably because it was originally believed that it's only a build fix.

[David Kilzer (:ddkilzer)]

<rdar://problem/9713309>