On Jul 15, 2010, at 13:30, Jon TURNEY wrote: > > > On 15/07/2010 17:28, Jamey Sharp wrote: >> The settings for each platform in rootlessConfig.h were misleading. On >> Apple builds, ROOTLESS_PROTECT_ALPHA was defined to TRUE; on Windows it >> was defined to NO. Both cases enabled alpha channel protection though, >> because the test in rootlessGC.c was a simple #ifdef. >> >> Colin Harrison says that for Xming builds he's always patched it to >> define ROOTLESS_PROTECT_ALPHA to TRUE for Windows as well, and that >> unconditionally enabling alpha-channel protection would be appropriate. > > I'm not sure I understand this well enough to know if this is wrong or right. > > It's unclear to me what this protected alpha channel is supposed to be used > for. If it's for use by the native window manager in alpha blending the > window into the native desktop, that's something we make no attempt to do on > Windows, and so there's no point in doing the extra work in rootless to > protect that alpha channel. > > Or maybe I've completely misunderstood the purposes of this...
The purpose is to prevent fb from messing up the alpha and ensuring that the resulting bitmap is usable by the native window manager. >> +variants. These are provided in rootless/safeAlpha. > > rootless/safeAlpha no longer exists, being removed some time ago, so this > text is still wrong :-) yeah, I think those bits moved into miPaintWindow ... _______________________________________________ [email protected]: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailman/listinfo/xorg-devel
