Re: UPDATED: video mode changes

[Lionel Ulmer]

Just a small nitpick :
> - XF86VidModeModeInfo *mode;
> - DWORD dwBpp = GetSystemMetrics(SM_WINE_BPP);
> - if (dwBpp == 24) dwBpp = 32;

I would prefer this change to be done directly in the 'GetDeviceCaps'
function of the X11DRV to always return 32 in case the scren depth is 24. I
know for sure that this fixes at least one application and should not hurt
others :-)

A part from that, the patch looks nice... If it breaks apps on Desktop mode,
we will fix it then by re-adding the 'dummy' modes :-)

Lionel

UPDATED: video mode changes

[Alex Pasadyn]

Lionel Ulmer wrote:
> Just a small nitpick :
> > - XF86VidModeModeInfo *mode;
> > - DWORD dwBpp = GetSystemMetrics(SM_WINE_BPP);
> > - if (dwBpp == 24) dwBpp = 32;
>
> I would prefer this change to be done directly in the 'GetDeviceCaps'
> function of the X11DRV to always return 32 in case the scren depth is 24. I
> know for sure that this fixes at least one application and should not hurt
> others :-)

I tried changing the return for case BITSPIXEL in X11DRV_GetDeviceCaps, 
and now I get all sorts of errors from X11DRV_CreateBitmap:
Trying to make bitmap with planes=1, bpp=32

That shows up because the screen_depth global variable is still set to 
24, but the application gets 32 as the depth.

Interestingly, the applications I tried all still seem to work, even 
though they complain loudly.

Is there another variable or check that should change somewhere?

Alex

UPDATED: video mode changes

[Lionel Ulmer]

> I tried changing the return for case BITSPIXEL in X11DRV_GetDeviceCaps, 
> and now I get all sorts of errors from X11DRV_CreateBitmap:
> Trying to make bitmap with planes=1, bpp=32

Oh well, submit your patch as it is now. The 24 / 32 'problem' needs a
specific fix anyway.

A work around for the issue would be to essentially consider 24 and 32 modes
as identical in your code (this way one is sure to never 'accidently' refuse
a mode due to this feature).

Lionel