On 9/18/07, Scott Balmos <[EMAIL PROTECTED]> wrote: > Again, didn't want to dilute the list with too many discussions at once. > I've got ideas for a driver framework, VFS layer, etc when we get there. > One thing I might caution is whether additional stubs to the AOT > *really* need to be made, or if such hardware access can be accomplished > with a simple wrapped private unsafe method and a pointer. Remember > we've also currently got X86.Asm.* and friends for direct assembly writing.
It might be usefull to have sub-classes or structs as well (somehow) When i was messing around with some scheduler ideas i realized that the data you'd want to store for each thread is (partly) processor specific.. I guess we could use a void*.. but that wouldn't really be managed would it? A stub class/struct might help keep things clean and avoid a whole list of tiny sub-functions ------------------------------------------------------------------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ _______________________________________________ SharpOS-Developers mailing list SharpOS-Developers@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/sharpos-developers
