On Thu, Feb 02, 2012 at 02:27:18PM +0000, Alan Cox wrote:
I see. For example the i810 also has a framebuffer driver. Do you see a way to fix this except writing a kms driver for all legacy devices? Otherwise I would leave the pci part untouched and only keep the platform/USB pieces which I'm admittedly more interested in.
Which is obsolete and unmaintained. More of a problem would be the various ati framebuffer drivers.
I would like to see Linux move to the situation where if there is a driver for a given device its either one or the other not one and some legacy code which is just extra work.
Doesn't need to be "all KMS" - but for any given card either/or seems perfectly reasonable.
The big thing that is needed is someone crazy enough to write a KMS driver to replace vesa/uvesafb and the like.
I am currently working on a layer which provides all the necessary drm glue code for simple framebuffer devices. My test driver is a Freescale i.MX framebuffer driver. This currently has < 500 loc which is less than the corresponding driver under drivers/video. I am not crazy enough to implement a vesa KMS driver, but my hope is that these tasks can become quite easy with such a layer.
Sascha