Robert Hildinger wrote: > On the platform that I am working on, which is ARM1176 based, I am > trying to implement a custom alpha blending blitter for the case of ARGB > to ARGB blends. This platform does not have any type of hardware > acceleration, so everything is done in software. My initial pass at > adding this custom blender was through the gfxdriver API, but I’m > running into issues with surface pool allocation / negotiation. > > Is the gfxdriver API really the best way to do this, or would it be > better to hook gfxcard.c to allow external functions to be called in > place of gBlit?
Writing a gfxdriver to implement CPU based acceleration should be fine. What kind of issues are you running into? Do you have surface pools without virtual addresses? -- Best regards, Denis Oliver Kropp .------------------------------------------. | DirectFB - Hardware accelerated graphics | | http://www.directfb.org/ | "------------------------------------------" _______________________________________________ directfb-dev mailing list directfb-dev@directfb.org http://mail.directfb.org/cgi-bin/mailman/listinfo/directfb-dev
