Subject:
Re: [nanogui] Nano-X and win32
From:
"Cedric Lescop" ####@####.####
Date:
1 Jul 2008 09:31:23 -0000
Message-Id: <5efa11530807010231i2fb62912ydc35119e8152ce2f@mail.gmail.com>
Thank you for answering, I think I understand better now :)
The 2 things I want to do are:
- To write an application using Nano-X libs running on an embedded ARM
(without running any OS for the moment).
So, if I well understand you: I'll compile up with LINK_APP_TO_SERVER set to
have only one application. Basically, the ARM will run this application
meaning the client will have access to the server (and the GrXXX functions).
I'll need to write the screen driver for the framebuffer blitting. And, I
need to "set up the config file for the target". (I already have a software
to cross-compile application running on the ARM)
- Then, I want to implement the same "behavior" under MS Windows. I know,
sorry, I develop under MS Windows :s
So, if I want to compile up Nano-X with another software, I still set
LINK_APP_TO_SERVER to have one apllication only. And, I'll need to write a
screen driver for accessing the framebuffer under MS Windows. Can I have a
look to the driver you built few years ago for Microwindows?
I still have a question though :p
How can I compile up Microwindows with Nano-X when I'm developping under MS
Windows? For example, "unistd.h" isn't part of MS Windows... I read
somewhere that I can include "io.h" and "process.h" instead...
Regards,
Cedric
On Mon, Jun 30, 2008 at 9:58 PM, Greg Haerr ####@####.#### wrote:
> : Ok, so if I want to use Nano-X under Windows I need to compile it need
> : compile it with its win32 API.
> : Then, if I want to compile it under Windows for a Linux platform, I need
> to
> : compile it with NanoX libs (X11).
> : Is it going to give me the same "result"? That's what people call
> : "cross-compile"?
>
> If you're trying to compile up the microwindows distribution
> using Windows (say using Cygwin), this has nothing to
> do with win32 or nano-X APIs, as these APIs are
> implemented by microwindows for ITS applications.
>
> : > > I can compile the demo. (because of the prototypes in header
> files.)
> : > > But then, the linker gives me this error: undefined reference
> : > > to 'GrOpen',
> : > > 'GrNewWindow", "GrSelectEvents", etc..
> : > >
> : > > How/where can I find these functions?
> : > > - libnano-X.a (error: "could not read symbols: File in wrong
> : > > format")
>
> All the GrXXX functions are in libnano-X.a. Fix your Makefile
> or linker commands, the above errors tells you've done
> something wrong, since the other libraries are ok.
>
> : > > I also found the client.c which open aconnection to the
> : > > server, do I needto
> : > > implement this "communication" to use Nano-X under Window?
>
> If you want to RUN nano-X on Windows, then, you have two
> choices, you can either link the applications to the server,
> LINK_APP_TO_SERVER, or you'll have to implement
> a method for interprocess communication, which should work
> with some changes to the client.c code.
>
> However, there are a number of other issues, since Windows
> doesn't allow direct access to the framebuffer in the same way
> as Linux. I built a Windows screen driver for microwindows
> some years ago, along with a number of other changes, if
> this is what you're looking for. I didn't use Cygwin for
> this, but built a native port instead. (yes, I know we already
> have win32 on win32, but I guess it was for the fun of it,
> at the time I was using Windows quite a bit)
>
> If you're cross-compiling for X11 on another platform using
> Windows tools, then not much should require change, set the
> config file for the target just like on any system.
>
> Regards,
>
> Greg
>
>
