[Mono-bugs] [Bug 81663][Wis] Changed - Performance: Delegate optimization, DLR and IronPython

[bugzilla-daemon at bugzilla.ximian.com]

Please do not reply to this email- if you want to comment on the bug, go to the
URL shown below and enter your comments there.

Changed by vargaz at gmail.com.

http://bugzilla.ximian.com/show_bug.cgi?id=81663

--- shadow/81663	2007-05-23 13:22:44.000000000 -0400
+++ shadow/81663.tmp.15280	2007-05-23 18:07:53.000000000 -0400
@@ -102,6 +102,13 @@
 based on IronPython 2.0. So I recompiled IronPython 2.0a1 with the fix
 from seo and indeed the performance drop is significant (down to 40070.2).
 In the new profile delegate-invoke methods have more weight (there are
 two, taking 1.3 and 1.2 % of total time, for a total of about 6%).
 The main difference, though is in time spent in the GC 16+% vs 6% it
 had in 1.x versions.
+
+------- Additional Comments From vargaz at gmail.com  2007-05-23 18:07 -------
+Here is a patch to implement parts of Paolo's ideas. It creates three
+copies of the Invoke method: non-multicast-no-target,
+non-multicast-with-target and multicast. Strangely, it actually
+decreases performance on a micro benchmark for me on amd64.
+