Hello All, Jason is right I think. I spoke with Alexandre the other night about implementing a Win32k.sys driver for WINE so we could use GDI32/User32 and it might be do-able it wont be somthing that is possible for at least another year at the rate we are sharing code.
Wines user32 depends on a few direct Wineserver calls and has quite a bit of old design issues from the Win16 days that will need to be fixed first. Not to mention Unixisms in a few places and of course ReactOS Win32k-User32 commuincation system uses our own system calls so that would need to be implemented in another way. I think the method we have currently for advapi32,kernel32,gdi32,ntdll and user32 are the best at the moment. We try to rip the source file as a whole such as the controls and string functions in user32 so we can easyly diff back to Winehq. For the other pure Win32 such as comctl32, shell32, ole*32, shlwapi and friends this is not a issues as we can just recompile without forking but for the lower level dlls we are going to need to keep forking. =(
Thanks Steven
--- Jason Filby jasonfilby@yahoo.com wrote:
Hi Dimitrie
User32 and gdi32 are likely to be a problem as the bulk of our implementation is in kernel mode win32k.sys.
Regards Jason
--- "Dimitrie O. Paun" dpaun@rogers.com wrote:
On October 14, 2003 03:11 pm, Jason Filby wrote:
I agree that forking should be avoided at all costs. Of course it will be unavoidable for lower level DLLs such as user32.
You are right, ReactOS will probably need it's own ntdll, and maybe kernel. But for user32 and gdi32, I am hoping you can use Wine's versions, and just implement a ReactOS driver for them...
-- Dimi.
Do you Yahoo!? The New Yahoo! Shopping - with improved product search http://shopping.yahoo.com _______________________________________________ Ros-kernel mailing list Ros-kernel@reactos.com http://reactos.geldorp.nl:8080/mailman/listinfo/ros-kernel
__________________________________ Do you Yahoo!? The New Yahoo! Shopping - with improved product search http://shopping.yahoo.com