http://bugzilla.novell.com/show_bug.cgi?id=550968
http://bugzilla.novell.com/show_bug.cgi?id=550968#c5 Rolf Bjarne Kvinge <[email protected]> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEEDINFO |NEW Info Provider|[email protected] | --- Comment #5 from Rolf Bjarne Kvinge <[email protected]> 2010-02-16 13:58:53 CET --- For the desktop it looks like MS caches assembly loading failures. See attached test case, disabling inline for the TestB method works, commenting the NoInlining attribute out makes it throw a FileNotFoundException (the LoadB method is still reached though, the exception happens when TestB is about to execute). For silverlight the story is a bit different: we need a way to load an assembly dynamically (using a stream/byte array) in the way the first test case does it. Here are a few references: http://msdn.microsoft.com/en-us/magazine/dd483293.aspx http://www.wintellect.com/CS/blogs/jprosise/archive/2008/10/22/cool-silverlight-trick-3.aspx Note that both these articles use [MethodImpl(MethodImplOptions.NoInlining)] to prevent inlining, but the msdn article about AssemblyPart.Load doesn't use it: http://msdn.microsoft.com/en-us/library/system.windows.assemblypart.load(VS.95).aspx nor do the Silverlight tests we got from MS which are failing due to this issue. I haven't been able to make a test case that fails without that attribute either, so apparently it's not required. Also note that AppDomain.AssemblyResolve isn't available for user code. -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are the QA contact for the bug. _______________________________________________ mono-bugs maillist - [email protected] http://lists.ximian.com/mailman/listinfo/mono-bugs
