Date: Thu, 16 Oct 2008 20:27:34 +0400 From: "Ilya Kasnacheev" <ilya.kasnacheev@gmail.com> To: sdl@lists.libsdl.org Subject: [SDL] SDL for Windows CE: a few GAPI patches Hi *! I've just ported a POWDER roguelike ( http://www.zincland.com/powder/ ) to Windows CE (PDAs, Windows Mobile/Pocket PC). To do that, I had to get libsdl working. Thanks for the awesome project files, it built without a hitch. Nevertheless, I've found quite a few bugs in Windows CE (GAPI) SDL implementation, which I've solved and now present as a serie of patches. I'll try carefully annotate them. Please annotate them so I can work toward accepting them into the main source tree since without them SDL isn't really working on Windows CE (I wonder why nobody fixed them before, btw: why isn't SDL popular as a way to develop Windows CE games? Where are no ports?) These changes can't be considered flawless, but they can be considered working because I've yet to hear complains about things I fixed and POWDER build for Windows CE is now considered stable. Note: my comments start with !!, delete them before applying. diff -bru SDL-1.2.13/src/video/gapi/SDL_gapivideo.c SDL-1.2.13-new/src/video/gapi/SDL_gapivideo.c --- SDL-1.2.13/src/video/gapi/SDL_gapivideo.c 2007-12-31 07:48:00.000000000 +0300 +++ SDL-1.2.13-new/src/video/gapi/SDL_gapivideo.c 2008-10-16 20:02:11.000000000 +0400 @@ -643,6 +643,7 @@ } gapi->userOrientation = SDL_ORIENTATION_UP; + gapi->systemOrientation = SDL_ORIENTATION_UP; video->flags = SDL_FULLSCREEN; /* Clear flags, GAPI supports fullscreen only */ /* GAPI or VGA? */ @@ -661,18 +662,21 @@ } /* detect user landscape mode */ - if( (width > height) && (GetSystemMetrics(SM_CXSCREEN) < GetSystemMetrics(SM_CYSCREEN))) + if( (width > height) && (gapi->gxProperties.cxWidth < gapi->gxProperties.cyHeight)) gapi->userOrientation = SDL_ORIENTATION_RIGHT; + if(GetSystemMetrics(SM_CYSCREEN) < GetSystemMetrics(SM_CXSCREEN)) + gapi->systemOrientation = SDL_ORIENTATION_RIGHT; + /* shall we apply hires fix? for example when we do not use hires resource */ gapi->hiresFix = 0; - if( gapi->userOrientation == SDL_ORIENTATION_RIGHT ) + if( gapi->systemOrientation == gapi->userOrientation ) { - if( (width > GetSystemMetrics(SM_CYSCREEN)) || (height > GetSystemMetrics(SM_CXSCREEN))) + if( (width > GetSystemMetrics(SM_CXSCREEN)) || (height > GetSystemMetrics(SM_CYSCREEN))) gapi->hiresFix = 1; } else - if( (width > GetSystemMetrics(SM_CXSCREEN)) || (height > GetSystemMetrics(SM_CYSCREEN))) - if( !((width == GetSystemMetrics(SM_CYSCREEN)) && (height == GetSystemMetrics(SM_CXSCREEN)))) // user portrait, device landscape + if( (width > GetSystemMetrics(SM_CYSCREEN)) || (height > GetSystemMetrics(SM_CXSCREEN))) +// if( !((width == gapi->gxProperties.cyHeight) && (height == gapi->gxProperties.cxWidth))) // user portrait, device landscape gapi->hiresFix = 1; switch( gapi->userOrientation ) !! It used to query system metrics which return dimensions according to screen !! orientation, which can really be portrait, left landscape or right landscape. !! This is presumably incorrect because we couldn't care less about user mode !! dimensions - all we want are the GAPI framebuffer dimensions, which only match !! user dimensions in one of possible orientations. !! There's a fair dose of cargo cult programming involved in this fix, but it !! used to work only in one orientation (portrait for PDAs, where frame-buffer !! have same orientation as user screen), and now it works on all orientations. @@ -742,21 +746,30 @@ WIN_FlushMessageQueue(); /* Open GAPI display */ - if( !gapi->useVga && this->hidden->useGXOpenDisplay ) + if( !gapi->useVga && this->hidden->useGXOpenDisplay && !this->hidden->alreadyGXOpened ) + { + this->hidden->alreadyGXOpened = 1; if( !gapi->gxFunc.GXOpenDisplay(SDL_Window, GX_FULLSCREEN) ) { SDL_SetError("Couldn't initialize GAPI"); return(NULL); } + } #if REPORT_VIDEO_INFO printf("Video properties:\n"); printf("display bpp: %d\n", gapi->gxProperties.cBPP); printf("display width: %d\n", gapi->gxProperties.cxWidth); printf("display height: %d\n", gapi->gxProperties.cyHeight); + printf("system display width: %d\n", GetSystemMetrics(SM_CXSCREEN)); + printf("system display height: %d\n", GetSystemMetrics(SM_CYSCREEN)); printf("x pitch: %d\n", gapi->gxProperties.cbxPitch); printf("y pitch: %d\n", gapi->gxProperties.cbyPitch); printf("gapi flags: 0x%x\n", gapi->gxProperties.ffFormat); + printf("user orientation: %d\n", gapi->userOrientation); + printf("system orientation: %d\n", gapi->userOrientation); + printf("gapi orientation: %d\n", gapi->gapiOrientation); + if( !gapi->useVga && this->hidden->useGXOpenDisplay && gapi->needUpdate) { !! Previous version used to call gapi->gxFunc.GXOpenDisplay each time the video !! mode would be changed. You shouldn't, because this call has a meaning "Lock the !! GAPI framebuffer, designate it as busy", so the second call will fail (it is !! already locked/busy). !! Testing might not find that because most programs set up the video mode only !! once, but POWDER does this once in a while, so it crashed when in 320x240 mode !! (640x480 mode doesn't use that code, it worked fine). diff -bru SDL-1.2.13/src/video/gapi/SDL_gapivideo.h SDL-1.2.13-new/src/video/gapi/SDL_gapivideo.h --- SDL-1.2.13/src/video/gapi/SDL_gapivideo.h 2007-12-31 07:48:00.000000000 +0300 +++ SDL-1.2.13-new/src/video/gapi/SDL_gapivideo.h 2008-10-16 20:02:11.000000000 +0400 @@ -132,12 +132,17 @@ #define NUM_MODELISTS 4 /* 8, 16, 24, and 32 bits-per-pixel */ int SDL_nummodes[NUM_MODELISTS]; SDL_Rect **SDL_modelist[NUM_MODELISTS]; + // The orientation of the video mode user wants to get + // Probably restricted to UP and RIGHT enum SDL_ScreenOrientation userOrientation; int invert; char hiresFix; // using hires mode without defining hires resource // -------------- int useGXOpenDisplay; /* use GXOpenDispplay */ + int alreadyGXOpened; int w, h; + // The orientation of GAPI framebuffer. + // Never changes on the same device. enum SDL_ScreenOrientation gapiOrientation; void *buffer; // may be 8, 16, 24, 32 bpp @@ -153,6 +158,10 @@ int startOffset; // in bytes int useVga; int suspended; // do not pu anything into video memory + // The orientation of the system, as defined by SM_CXSCREEN and SM_CYSCREEN + // User can change it by using 'screen layout' in system options + // Restricted to UP or RIGHT + enum SDL_ScreenOrientation systemOrientation; }; !! This is a flag variable, see the previous comment !! And yet another orientation: now we have to keep three of them in mind. diff -bru SDL-1.2.13/src/video/wincommon/SDL_sysevents.c SDL-1.2.13-new/src/video/wincommon/SDL_sysevents.c --- SDL-1.2.13/src/video/wincommon/SDL_sysevents.c 2007-12-31 07:48:02.000000000 +0300 +++ SDL-1.2.13-new/src/video/wincommon/SDL_sysevents.c 2008-10-16 20:02:12.000000000 +0400 @@ -160,10 +160,22 @@ #endif */ } break; + // FIXME: Older version used just SDL_VideoSurface->(w, h) + // w and h are "clipped" while x and y are "raw", which caused + // x in former and y in latter case to be clipped in a wrong direction, + // thus offsetting the coordinate on 2 x clip pixels + // (like, 128 for 640 -> 512 clipping). + // We will now try to extract and use raw values. + // The way to do that RIGHT is do (orientation-dependent) clipping before + // doing this transform, but it's hardly possible. + + // SEE SDL_mouse.c /ClipOffset to understand these calculations. case SDL_ORIENTATION_RIGHT: if (!SDL_VideoSurface) break; - rotatedX = SDL_VideoSurface->w - *y; + rotatedX = (2 * ((SDL_VideoSurface->offset%SDL_VideoSurface->pitch)/ + SDL_VideoSurface->format->BytesPerPixel)) + + SDL_VideoSurface->w - *y; rotatedY = *x; *x = rotatedX; *y = rotatedY; @@ -172,7 +184,8 @@ if (!SDL_VideoSurface) break; rotatedX = *y; - rotatedY = SDL_VideoSurface->h - *x; + rotatedY = (2 * (SDL_VideoSurface->offset/SDL_VideoSurface->pitch)) + + SDL_VideoSurface->h - *x; *x = rotatedX; *y = rotatedY; break; !! That's the trickest part, hence the long comment. !! GAPI would really support only 320x240 or 640x480 mode, if application !! requested the different screen size (as POWDER did, wishing 256x192), then SDL !! is going to grab the first mode that fits the requested, and pad the screen !! with black bars (as they do with wide-screen films). !! It would also get, say, 240x320 mode, and to turn it into 256x192 it would !! need to rotate mouse clicks. !! It worked, but one bug slipped through: it would receive mouse clicks !! unpadded, then rotate them, and then pad the black bars. The problem is: rotate !! is done by GAPI driver while padding is done by SDL core. SDL core doesn't know !! anything about rotating, so it would pad one of dimensions incorrectly. I understand that some of my claims (or code) might seem unbacked, but you can always grab the POWDER binary, compile your own libsdl with one or more of those fixes turned off, and see how weird it would misbehave. I can even supply you with those custom builds of libsdl if you don't want to set up the build environment for windows ce, you'll just need a PDA or a smartphone with it. I plan to take care of SDL on Windows CE as long as I maintain the POWDER port. POWDER is good for that because it: Employs both padded (with centered image, black bars) and unpadded (image occupies full screen) graphics; initializes video more than once; uses both 320x240 and 640x480 video; uses both stylus and buttons. There's still a list of unresolved issues which I'm planning to fix: 1) Arrow buttons on PDA return weird scancodes compared to PC, this caused the game to misbehave before I've fixed that. You can see it on those diagrams: http://wrar.name/upload/powder-htc.png http://wrar.name/upload/powder-pda.png 2) SDL (or underlying windows) doesn't care to rotate arrow presses when we're in a low-res GAPI mode, but it will rotate them in VGA mode (because of different screen orientations, the same arrow buttons can suddently mean different directions). Solution: we should stick to GAPI user orientation (the orientation the program supposedly wants) and rotate the keys on our own. _______________________________________________ SDL mailing list SDL@lists.libsdl.org http://lists.libsdl.org/listinfo.cgi/sdl-libsdl.org

Simple DirectMedia Layer CREDITS
Thanks to everyone who made this possible, including:

* Cliff Matthews, for giving me a reason to start this project. :)
 -- Executor rocks!  *grin*

* Scott Call, for making a home for SDL on the 'Net... Thanks! :)

* The Linux Fund, C Magazine, Educational Technology Resources Inc.,
  Gareth Noyce, Jesse Pavel, Keith Kitchin, Jeremy Horvath, Thomas Nicholson,
  Hans-Peter Gygax, the Eternal Lands Development Team, Lars Brubaker,
  and Phoenix Kokido for financial contributions

* Gaëtan de Menten for writing the PHP and SQL behind the SDL website

* Tim Jones for the new look of the SDL website

* Marco Kraus for setting up SDL merchandise

* Martin Donlon for his work on the SDL Documentation Project

* Ryan Gordon for helping everybody out and keeping the dream alive. :)

* Mattias Engdegård, for help with the Solaris port and lots of other help

* Max Watson, Matt Slot, and Kyle for help with the MacOS Classic port

* Stan Shebs, for the initial Mac OS X port

* Eric Wing, Max Horn, and Darrell Walisser for unflagging work on the Mac OS X port

* Patrick Trainor, Jim Boucher, and Mike Gorchak for the QNX Neutrino port

* Carsten Griwodz for the AIX port

* Gabriele Greco, for the Amiga port

* Patrice Mandin, for the Atari port

* Hannu Viitala for the EPOC port

* Marcus Mertama for the S60 port.

* Peter Valchev for nagging me about the OpenBSD port until I got it right. :)

* Kent B Mein, for a place to do the IRIX port

* Ash, for a place to do the OSF/1 Alpha port

* David Sowsy, for help with the BeOS port

* Eugenia Loli, for endless work on porting SDL games to BeOS

* Jon Taylor for the GGI front-end

* Paulus Esterhazy, for the Visual C++ testing and libraries

* Brenda Tantzen, for Metrowerks CodeWarrior on MacOS

* Chris Nentwich, for the Hermes assembly blitters

* Michael Vance and Jim Kutter for the X11 OpenGL support

* Stephane Peter, for the AAlib front-end and multi-threaded timer idea.

* Jon Atkins for SDL_image, SDL_mixer and SDL_net documentation

* Peter Wiklund, for the 1998 winning SDL logo,
  and Arto Hamara, Steven Wong, and Kent Mein for other logo entries.

* Arne Claus, for the 2004 winning SDL logo,
  and Shandy Brown, Jac, Alex Lyman, Mikkel Gjoel, #Guy, Jonas Hartmann,
  Daniel Liljeberg,  Ronald Sowa, DocD, Pekka Jaervinen, Patrick Avella,
  Erkki Kontilla, Levon Gavalian, Hal Emerich, David Wiktorsson,
  S. Schury and F. Hufsky, Ciska de Ruyver, Shredweat, Tyler Montbriand,
  Martin Andersson, Merlyn Wysard, Fernando Ibanez, David Miller,
  Andre Bommele, lovesby.com, Francisco Camenforte Torres, and David Igreja
  for other logo entries.

* Bob Pendleton and David Olofson for being long time contributors to
  the SDL mailing list.

* Everybody at Loki Software, Inc. for their great contributions!

 And a big hand to everyone else who gave me appreciation, advice,
 and suggestions, especially the good folks on the SDL mailing list.

THANKS! :)

  -- Sam Lantinga			<slouken@libsdl.org>