I looked through the changelog in qgit and this patch file was included in a commit titled "Add Mir patches"
I know absolutely nothing about EGL or DRI2, but the bulk of the patch adds a new platform file (src/egl/drivers/dri2/platform_mir.c).
But let's just for argument's sake assume that the final default design of the Steambox features an APU from AMD and a customized Ubuntu LTS running with Xorg. That means that AMD actually doesn't have to improve much of their Linux drivers, they only have to improve the drivers for the Steambox, read: this specific APU, this specific kernel, this specific Xorg version, no Mir, no Wayland, no other kernels. They only have to fix issues for that to make Valve happy, why should they care for the ordinary Linux user in the consumer market? The rest of their effort can easily go to the professional cards and may be there, as a byproduct, you will see improvements for the consumer market.
Seriously, AMD now is not even able to keep pace with the [sarcasm]astonishing speed[/sarcasm] of even Slackware's development, why do you think that there will be something improved in the future just because a fixed platform is introduced?
Here's how I think its gonna play out. The driver makers are gonna write drivers only for either wayland or mir and the other side is gonna cave in and adopt. For all the imbeciles who claim that competition is good and doesn't cause fragmentation think of this scenario. If company A needs x number of users per OS to port their software to that OS and if that OS suddenly split into two OSes or display servers in our case, do you think they will magically support both out of the goodness of their hearts. Just look at Apple. Most OS X software is a second citizen to the Windows offerings because that's where the money is.
Basically, now that they have the converts, they want to splinter their religion from the main branch... (yeah, I'm using the weird zealotry for an argument here.)
I found out! Embrace, Extend and Extinguish? :cool:
Mir vs Wayland support in toolkits will look SAME. No technical reasons that would make one or second "better".
And Upstart was BEST solution, that get support from across Linux landscape. systemd started latter AFTER some devs decided that Upstart in not perfect. (Just like now some devs decided that Wayland isn't perfect ;) ) Heck Upstart landed in RHEL, that is probably best know "solid software" badge in Linux world.