Daniel Borgmann <[EMAIL PROTECTED]> writes: > The idea seems to be to use memory mapping and a translator so > applications can just write to this translator to access the video > memory. The kernel would than directly talk to the > hardware, for example through a VESA driver. Is that correct?
More or less. One alternative is to have the *translator* talk directly to the hardware, so that you don't have to hack the kernel (you'd still need to run the translator as root, though). Moving device drivers in general out from the kernel and into userspace programs is a long term plan. Regards, /Niels -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
