701248 - Assertion failure: ((js::SrcNoteType)(((*(sn) >> 3) >= SRC_XDELTA) ? SRC_XDELTA : *(sn) >> 3)) == SRC_DESTRUCT, at jsopcode.cpp:3543

Status: VERIFIED FIXED

(Core :: JavaScript Engine, defect)

Description

[Christian Holler (:decoder)]

The following test asserts on mozilla-central revision c60535115ea1 (no options required):


function F(x, y) {
  var { j, y, p, a } = testSyntax("#1=<a>b</a>", true) =  this;
}
var src = F.toSource(-1)+"\n";


Could be related to a decompiler patch that landed recently, Cc'ing Luke and Waldo.

Comment 1

[Jeff Walden [:Waldo]]

Reduced somewhat (potentially morphing, but I don't quite think so), I get this:

  (function F(x) { var {x} = this; }).toSource(-1);

That implicates bug 685321, which I really need to get to soon.  :-\

Comment 2

[Daniel Veditz [:dveditz]]

"depends on" rather than "dupe" in case of morphing, but if a patch for that bug fixes this one we can dupe at that time. Please include both Christian's original and your variant as regression tests when you check in.

Comment 3

[David Mandelin [:dmandelin]]

Jeff, recording you as assignee since you're working on the likely dup.

Comment 5

[Daniel Veditz [:dveditz]]

Would like decoder to verify that the fix for bug 685321 really does fix his testcase.

Comment 6

[Christian Holler (:decoder)]

Confirmed that the test in comment 0 no longer reproduces on trunk.

Comment 7

[Johnny Stenback (:jst)]

Fixed in 10 by bug 685321.

Comment 8

[Mihaela Velimiroviciu (:mihaelav)]

OS: Ubuntu 11.04 64bit
I have built the js from mozilla-beta-52cf2b0c8439 and run the tests from comment #0 and comment #1 and no error was displayed.

Comment 9

[Christian Holler (:decoder)]

Automatically extracted testcase for this bug was committed:

https://hg.mozilla.org/mozilla-central/rev/efaf8960a929