On 8/16/06, Dave Methvin <[EMAIL PROTECTED]> wrote: > Yipes! I just did a quick test with IE6 and SVN 170. With nothing other than > the jQuery.js include and an empty body, it's leaking about 300KB for each > reload measured by Task Manager. FF doesn't leak at all. Can anyone else > repro that?
It was corrected with the patch of SVN 175. http://proj.jquery.com/dev/svn/rev/175/ > > We found this problem on svn rev.172, and examined the cause. Then it > turned out to be the following line: > > jQuery.event.add( window, "load", jQuery.ready ); > > > In the case of Firefox 1.5, IE and Opera 9, it seems to work without > this line. So we got to avoid the leak with the following patch: > > --- jquery-svn172.js 2006-08-11 10:34:25.871263749 +0900 > +++ jquery-svn172-patched.js 2006-08-11 10:34:25.870263917 +0900 > @@ -3895,11 +3895,11 @@ > jQuery.ready(); > } > }, 10); > + } else { > + // A fallback to window.onload, that will always work > + jQuery.event.add( window, "load", jQuery.ready ); > } > > - // A fallback to window.onload, that will always work > - jQuery.event.add( window, "load", jQuery.ready ); > - > } > jQuery.fn.extend({ > > > This improvement can be seen by monitoring the memory usage. The upper > image indicates the usage before fixing, and the lower one is for the > after. It consumed over 100M before the fix, but afterward it is > stabilized with only consuming less than 10MB. ---- Taku Sano _______________________________________________ jQuery mailing list [email protected] http://jquery.com/discuss/
