[Mono-devel-list] Loading of assemblies by runtime and reflection

Zoltan Varga vargaz at gmail.com
Mon Oct 25 15:38:46 EDT 2004


  I would like to fix this but I can't reproduce it either with
current CVS or with
the 1.0 branch.


On Mon, 25 Oct 2004 21:16:01 +0200, Kamil Skalski <nazgul at nemerle.org> wrote:
> Hi!
> I am trying to make some investigation on
> http://bugzilla.ximian.com/show_bug.cgi?id=61152
> so I tracked what is called by Assembly.LoadFrom. It seems that this is the
> same code, which is used by runtime to load referenced assemblies
> (mono_assembly_load_from in metadata/assembly.c).
> Can anybody comment if it is true? I don't think that erasure of runtime
> loaded assemblies by reflecting some with the same name is a good behaviour.
> But even if it is, is shouldn't cause setting system_assembly in socket-io.c
> to NULL.
> I wanted to do some more debugging, but I have some problems running mono cvs
> version:
> checking how to run the C++ preprocessor... /lib/cpp
> configure: error: C++ preprocessor "/lib/cpp" fails sanity check
> See `config.log' for more details.
> The config.log file ends with some invalid preprocessor directives:
> #define VERSION "1.1.1"
> #endif
> #ifdef __cplusplus
> void exit (int);
> It's probably some issue with automake generating bad configure script...
> Sorry for this probably bogus message, but the 61152 problem is open now for a
> few months and IMHO it seems to be quite critical issue. Especially if it
> occurs in some more sofisticated scenarios with reflecting assemblies with
> various versions.
> Just to put some positive comments, you guys are doing great job. According to
> our summary (http://nemerle.org/wiki/index.php/Runtime_issues), mono has
> currently less open bugs interfering with our project than MS.NET.
> --
> Kamil Skalski
> http://nemerle.org developer
> _______________________________________________
> Mono-devel-list mailing list
> Mono-devel-list at lists.ximian.com
> http://lists.ximian.com/mailman/listinfo/mono-devel-list

More information about the Mono-devel-list mailing list