Subj : Re: Exception in SpiderMonkey v1.6 pr 6a
To   : netscape.public.mozilla.jseng
From : Rob Swindell
Date : Mon May 02 2005 07:21 pm


"Brendan Eich" <brendan@meer.net> wrote in message
news:42765FEE.1040003@meer.net...
> Rob Swindell wrote:
> > Here's a call-stack from a recent crash of a script using js32.dll
> > (JavaScript-C 1.5 pre-release 6a 2004-06-09):
> >
> > js_LookupPropertyWithFlags(JSContext * 0x00a32010, JSObject *
0x015a2cf8,
> > long 0x01cbc3f0, unsigned int 0x00000000, JSObject * * 0x0012e39c,
> > JSProperty * * 0x0012e318) line 2442 + 34 bytes
> > js_LookupProperty(JSContext * 0x00a32010, JSObject * 0x015a2cf8, long
> > 0x01cbc3f0, JSObject * * 0x0012e39c, JSProperty * * 0x0012e318) line
2417 +
> > 27 bytes
> > js_Interpret(JSContext * 0x00a32010, unsigned char * 0x015f2913, long *
> > 0x0012e4ec) line 2426 + 43 bytes
>
> In RC6a, this line is the starred one below:
>
>                  str = js_ConcatStrings(cx, str, str2);
> *               if (!str) {
>                      ok = JS_FALSE;
>                      goto out;
>                  }
>
>
> Have you hacked on jsinterp.c by any chance?

Nope. I have jsinterp.c rev 3.162, where line 2462 is starred below:

            /* Skip properties not owned by obj when looking from origobj.
*/
*          ok = OBJ_LOOKUP_PROPERTY(cx, origobj, fid, &obj2, &prop);
            if (!ok)
                goto out;

BTW, it'd be nice if the source files had CVS tags (e.g. "$Id: $")
indicating the revision.

> > js_Invoke(JSContext * 0x00a32010, unsigned int 0x00000001, unsigned int
> > 0x00000000) line 1313 + 22 bytes
> > js_Interpret(JSContext * 0x00a32010, unsigned char * 0x015aa75f, long *
> > 0x0012f5b0) line 3566 + 15 bytes
> > js_Execute(JSContext * 0x00a32010, JSObject * 0x010e93a8, JSScript *
> > 0x016772c0, JSStackFrame * 0x00000000, unsigned int 0x00000000, long *
> > 0x0012f6cc) line 1523 + 19 bytes
> > JS_ExecuteScript(JSContext * 0x00a32010, JSObject * 0x010e93a8, JSScript
*
> > 0x016772c0, long * 0x0012f6cc) line 3630 + 25 bytes
> > js_exec(const char * 0x00a50c21, char * * 0x00a50bd4) line 721 + 36
bytes
> > main(int 0x0000000a, char * * 0x00a50bb0, char * * 0x00a50c58) line 976
+ 25
> > bytes
> > mainCRTStartup() line 206 + 25 bytes
> > KERNEL32! 7c59893d()
> >
> > Line 2442 in jsobj.c is simple a call to this (rather extensive) macro:
> >
> >     CHECK_FOR_FUNNY_INDEX(id);
> >
> > Any clues as the cause/solution would be appreciate,
>
>
> It's really impossible to say without accurate source line numbers
> referring to source I can check out, along with debugger information
> such as id's value in hex, memory it references, etc.

Here are the local variable contents at the time of the exception:

+       cs      0x00000000
+       cx      0x00a32010
        flags   0x00000000
        format  0x00000000
        id      0x01cbc3f0
        newresolve      0x0012d414
+       obj     0x015a2cf8
+       *objp   0x010e93a8
        ok      0x7ffdf000
+       pc      0x00000001 ""
+       *propp  0x01d7bcec

I realize now that my js32.dll built from a CVS snapshot where
JS_GetImplementationVersion() returns "JavaScript-C 1.5 pre-release 6a
2004-06-09", is not *actually* SpiderMonkey 1.5 PR 6a. I was under the
(apparently incorrect) assumption that cvs-mirror.mozilla.org only contained
released source and that the string returned from
JS_GetImplementationVersion() would actually reflect the version built from
the MAIN/HEAD rev. :-(

Can I suggest tagging the released SpiderMonkey files with a common label
(say "JS_LATEST_RELEASE") which will allow someone to get the latest
*released* version of the SpiderMonkey code? When a new release is made,
they are relabeed as JS_LATEST_RELEASE (which moves the label). Currently,
you would have to search for the lateset release label (e.g. "JS_150_RC6a")
and get based on that label or rely on the versioned-tarball from the FTP
server. :-(

Now I have to decide if I should revert to the actual 1.5 pr 6a code or use
the latest from CVS. Any suggestions?

> Find me on  irc.mozilla.org #developers some time.

I'll try to do that.

-Rob

.