[Webkit-unassigned] [Bug 255508] New: LLInt should update stackTraceLimit in GlobalObject

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=255508

            Bug ID: 255508
           Summary: LLInt should update stackTraceLimit in GlobalObject
           Product: WebKit
           Version: WebKit Local Build
          Hardware: PC
                OS: Linux
            Status: NEW
          Severity: Normal
          Priority: P2
         Component: JavaScriptCore
          Assignee: webkit-unassigned at lists.webkit.org
          Reporter: entryhii at gmail.com

function f0(a1) {
  try{
    Error.stackTraceLimit = a1; // put_by_id
    throw Error()
  }
  catch(e)
  {
    return e.stack;
  }
}
for (let v14 = 0; v14 < 2; v14++) {
  print(f0(v14));
}

With the above script as input to JSC, run JSC with the following parameters:
./jsc test.js --useConcurrentJIT=0 

JSC print two "undefined". The corrent result should be an "undefined" and a stackTrace. The problem may be in LLInt. 

The number of loops is 2. Therefore, LLInt will execute `put_by_id` twice, the first execution is slow_path, and `stackTraceLimit` in `globalObject` will be updated, but the slow_path is not used for the second time, and `stackTraceLimit` in `globalObject` is not updated. `getStackTrace` in Error.cpp will use `stackTraceLimit`, but stackTraceLimit is always 0, so two "undefined" are printed. 

Safari console also prints wrong result, V8 prints correct result.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-unassigned/attachments/20230417/b95881a1/attachment-0001.htm>