Matrox DRM/KMS Driver Gets New Hardware Support In Linux 4.3

[phoronix]

Phoronix: Matrox DRM/KMS Driver Gets New Hardware Support In Linux 4.3

It's not often that there's improvements in the Matrox "mgag200" DRM/KMS driver to talk about, but there is this morning...

http://www.phoronix.com/scan.php?pag...x-Linux-4.3-HW

[Drago]

Just to answer some peoples questions in advance: Matrox chips are extremely popular on the server boards. That is why David ( thus RedHat ) are investing in this old chipset.

[karasu]

Matrox, that's a name I haven't heard in a long long time.

[schmidtbag]

Just to answer some peoples questions in advance: Matrox chips are extremely popular on the server boards. That is why David ( thus RedHat ) are investing in this old chipset.

I personally was aware Matrox was popular in servers. What I don't understand is why bother supporting it at all. I figure using the vesa or framebuffer drivers would probably perform better at this rate.

[chithanh]

Because -vesa and userspace modesetting drivers have no place in the brave new world of UEFI secure boot.

[jonnor]

Because -vesa and userspace modesetting drivers have no place in the brave new world of UEFI secure boot.

Really, why is that?

[uid313]

Doesn't the generic driver already do this?

The xf86-video-modesetting driver.

[adler187]

Doesn't the generic driver already do this?

The xf86-video-modesetting driver.

Well, the DRM/KMS code is in the kernel, while the xf86-video-modesetting is in userspace. I'm assuming the xf86-video-modesetting will make use of the generic KMS support that David just added.

[mjg59]

I personally was aware Matrox was popular in servers. What I don't understand is why bother supporting it at all. I figure using the vesa or framebuffer drivers would probably perform better at this rate.

VESA doesn't work on UEFI systems. There's no performance benefit in using this driver over a straight framebuffer one, but that would mean having to support both xf86-video-fbdev and xf86-video-modesetting. Plus someone would have to write the kernel framebuffer driver for these chips.

[airlied]

Phoronix: Matrox DRM/KMS Driver Gets New Hardware Support In Linux 4.3

It's not often that there's improvements in the Matrox "mgag200" DRM/KMS driver to talk about, but there is this morning...

http://www.phoronix.com/scan.php?pag...x-Linux-4.3-HW

These aren't workstation class boards, nothing even close. They don't have real acceleration features, they may have a small DMA engine.

There is no value in writing anything else for them other than what they have.

Maybe you are confusing them with the old Matrox Gxxx boards or Pxxx boards, but hey research is hard.

Dave.