[Mono-bugs] [Bug 68723][Nor] Changed - Performance on long arith

bugzilla-daemon@bugzilla.ximian.com bugzilla-daemon@bugzilla.ximian.com
Sun, 24 Oct 2004 20:49:19 -0400 (EDT)

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 bmaurer@users.sf.net.


--- shadow/68723	2004-10-24 15:31:55.000000000 -0400
+++ shadow/68723.tmp.12353	2004-10-24 20:49:19.000000000 -0400
@@ -101,6 +101,18 @@
 ------- Additional Comments From writeonlymemory@gmail.com  2004-10-24 15:31 -------
 So you believe it is just the long arithmetic, not the recursion?
 That's something I hadn't thought of.
+------- Additional Comments From bmaurer@users.sf.net  2004-10-24 20:49 -------
+yeah. Feel free to make a test case that shows otherwise...
+The only thing that is different in msft that might make a difference
+here is that they do not do the push ebp calling conv. They use things
+as offsets from esp, and use ebp as another register. This makes
+recursion a tad faster. However, division, multiplication, etc are
+orders of magnitude more expensive than push ebp.
+I checked in the debugger that msft is not doing any fancy
+recursion->loop opts.