Xserver driver merging pros & cons

[Matt Turner]

On Thu, Sep 15, 2011 at 11:45 AM, Jesse Barnes <jbarnes at virtuousgeek.org> wrote:
> Pros:
>  1) easier to propagate API changes across drivers (just like Linux)
>     1a) thus easier to change ABI

I suppose that's true. How often are we breaking the ABI though? How
often do we break the ABI and it doesn't get fixed in the drivers we
care about within a day or so? It's strange to sort of pretend that we
care about other drivers simply for the point of this argument.

I see in other emails that we're not even talking about moving older
drivers into the tree, so I'm even a bit more confused.

>  2) developers focused on driver development now have more incentive
>     to make sure the server works well so regular releases can still
>     happen (i.e. more people working on blockers whether driver or not
>     as releases approach)

In theory. I'm sort of skeptical that suddenly dropping driver code
into the X server will incentivize fixing X server blockers enough to
actually do it.

>  3) allows removal of driver compat code for various server versions
>     3a) thus removes combinations of driver+server that developers
>         have to support & test

But, enterprise hates this. It's hard for me to see this as a big win
when we know it makes a lot of other people's jobs harder. That, and
Keith said at least Intel would keep the ifdef checks in their driver.

>  4) increased test coverage for the server as users wanting current
>     driver code will be building new servers too

Maybe. I'd hate to force users to do that just to get new hardware
support or a simple driver bug fix.

As far as Gentoo goes, this would help the problem of updating the X
server and then having your drivers not work. I'm just not sure the
benefits are worth the costs though.

Matt