Dave wrote the AST driver as well. The G200 driver is based on the Matthew Garret's Cirrus KMS driver for QEMU. The Cirrus driver is based on my GLINT KMS driver, so the G200 driver is as well.
Well its bits of both really.
I wrote AST, then took the MGA and next cirrus code from mjg based on your code, I then transmuted them and caluclated their load bearing tangents. (Sorry watching peppa pig), I then remodeled the mga/cirrus code to look like my AST code and used the same memory management scheme as my AST code, so all 3 drivers nearly look the same apart from the crtc mode setting and gpu init functions :-)
Wow, the G200 is a 14-year-old graphics card and its seeing new device driver development work in the open source community.
I wish we could say that for all these chips around... I wonder what that driver actually does and if Matrox ever published enough docs to do more than modesetting. Anyway, havin KMS-drivers is always a good thing.
- flicker free boot
- faster VT switch
- kernel errors visible if X hangs
Now, which of these is important in a server?
RH/Fedora actually plans to ship no UMS drivers in the future, so they're writing KMS drivers for the hardware they care about. Without xf86-video-mga (say, if you're not running X), the only other option is matroxfb which doesn't support the G200.
Also, you can do things like power down the graphics chip during DPMS to save energy. And since it's a server that's not typically attached to a monitor, that's basically always.
I could point out that HD7xxx has had KMS support for a while (ie a whole pile more love already) but I imagine what you really mean is "if only radeonsi got 1000x more love than 14 year old hardware instead of the 100x or so it gets today" so that the high level of functionality expected for the new card could be implemented as quickly as the modesetting-only expectations for 14 year old hardware" ?