This causes amusing things to happen when the callsite index overlaps with the bytecode index of some handler.
Created attachment 268577 [details] the patch
Comment on attachment 268577 [details] the patch lol, sounds amusing, indeed r=me
Landed in http://trac.webkit.org/changeset/194786 Now the test failure list is: ** The following JSC stress test failures have been introduced: regress/script-tests/call-spread-apply.js.ftl-no-cjit-no-inline-validate regress/script-tests/call-spread-call.js.ftl-no-cjit-no-inline-validate regress/script-tests/varargs-strict-mode.js.ftl-no-cjit-no-inline-validate stress/load-varargs-then-inlined-call-and-exit-strict.js.ftl-no-cjit-no-inline-validate stress/op_negate.js.ftl-no-cjit stress/reflect-apply.js.ftl-eager-no-cjit stress/tail-call-varargs-no-stack-overflow.js.ftl-eager stress/tail-call-varargs-no-stack-overflow.js.ftl-eager-no-cjit stress/tail-call-varargs-no-stack-overflow.js.ftl-no-cjit-no-inline-validate stress/tail-call-varargs-no-stack-overflow.js.ftl-no-cjit-no-put-stack-validate stress/tail-call-varargs-no-stack-overflow.js.ftl-no-cjit-small-pool stress/tail-call-varargs-no-stack-overflow.js.ftl-no-cjit-validate stress/varargs-varargs-inlined-exit-strict-mode.js.ftl-no-cjit-no-inline-validate