On 12/05/2014 01:38 AM, Armin K. wrote: > On 05.12.2014 01:06, Ken Moffat wrote: >> On Fri, Dec 05, 2014 at 12:30:20AM +0100, Armin K. wrote: >>> On 05.12.2014 00:08, Ken Moffat wrote: >>>> >>>> 1.4 The modesetting driver >>>> >>>> This requires CONFIG_FB and CONFIG_DRM_BOCHS. Using >>>> CONFIG_DRM_CIRRUS_QEMU is recommended, although it appears to work >>>> (with lower fps in glxgears, so higher cpu usage) without that. >>>> >>> >>> It's a generic driver and in theory it should work with any hardware >>> that has a KMS driver. It has also been merged into Xserver-1.17, much >>> like Glamor was to 1.16 so I propose that you postpone the addition >>> (just to avoid the unnecessary work). >>> >> >> Thanks for that. I was aware that it should work with any hardware >> that has a kms driver, but for x86 in practice I think that limits >> it to qemu - I assume chip-specific drivers will always be >> preferred, so intel, nouveau, radeon are unlikely to use it? > > I believe they can use it, but you'd get bad performance compared to the > hardware specific (and optimized) DDX-es. It's also a base for new > hardware specific DDX drivers that have a KMS driver. Another thing that > I am not 100% sure of is that the driver that was mergeed into xserver > tree is hooked up to glamor, so it may use OpenGL to accelerate 2D > graphics where supported. >
Now I have stumbled upon something very interesting. It seems that modesetting driver was performing better than the ati driver according to the openbenchmarking benchmark done by someone. http://openbenchmarking.org/result/1412165-LI-1412162LI31 Given that it uses glamor for 2D acceleration over 3D driver where supported, it's not a big surprise. -- Note: My last name is not Krejzi.
signature.asc
Description: OpenPGP digital signature
-- http://lists.linuxfromscratch.org/listinfo/blfs-dev FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
