452338 - TM: "Assertion failure: obj2 == obj" modifying __count__

Status: RESOLVED FIXED

(Core :: JavaScript Engine, defect)

Description

[Jesse Ruderman]

./js -j
js> for (var j = 0; j < 4; ++j) __count__ = 3;

Assertion failure: obj2 == obj, at jstracer.cpp:4982

I'm using a js shell compiled from the tracemonkey branch.

Comment 1

[Blake Kaplan (:mrbkap) (inactive)]

Attachment: Fix

It doesn't make sense to run the count getter on set. A silent failure seems to preserve the old behavior and we do less work now.

Comment 2

[Brendan Eich [:brendan]]

Comment on attachment 335812 [details] [diff] [review]
Fix

>Bug 452338 - Abort on the weird case where we find a global name on the prototype of the scope chain.
>
>diff --git a/js/src/jsobj.cpp b/js/src/jsobj.cpp
>--- a/js/src/jsobj.cpp
>+++ b/js/src/jsobj.cpp
>@@ -139,8 +139,8 @@ static JSPropertySpec object_props[] = {
>                                                   obj_getSlot,  obj_setSlot},
>     {js_parent_str,JSSLOT_PARENT,JSPROP_READONLY|JSPROP_PERMANENT|JSPROP_SHARED,
>                                                   obj_getSlot,  obj_setSlot},
>-    {js_count_str, 0,            JSPROP_PERMANENT|JSPROP_SHARED,
>-                                                  obj_getCount, obj_getCount},
>+    {js_count_str, 0,            JSPROP_PERMANENT|JSPROP_READONLY|JSPROP_SHARED,

Should match the READONLY|PERMANENT|SHARED order for parent, eh?

>+        ABORT_TRACE("found a non-direct property on the global object");

I've taken to saying "JSOP_BINDNAME found a ..." and the like.

r=me with these.

/be

Comment 3

[Blake Kaplan (:mrbkap) (inactive)]

Attachment: For checkin

I'm going to end up checking this in on tracemonkey.

Comment 4

[Blake Kaplan (:mrbkap) (inactive)]

http://hg.mozilla.org/tracemonkey/rev/2206c4127fad

Comment 5

[Bob Clary [:bc] (inactive)]

/cvsroot/mozilla/js/tests/js1_5/extensions/regress-452338.js,v  <--  regress-452338.js
initial revision: 1.1

http://hg.mozilla.org/mozilla-central/rev/8db3af41e55a