[Nix-dev] Dropping --with-gallium-drivers=i965 from mesa (was: Experiences with Nvidia Optimus?)
mathijs at bluescreen303.nl
Sat Jan 21 11:12:56 CET 2012
On Sat, Jan 21, 2012 at 5:05 AM, Yury G. Kudryashov
<urkud.urkud at gmail.com> wrote:
> Mathijs Kwik wrote:
>> On Fri, Jan 20, 2012 at 7:17 PM, Peter Simons <simons at cryp.to> wrote:
>>> Hi guys,
>>> > In a recent mesa upgrade, someone enabled the experimental gallium3d
>>> > driver for our i965-based cards. [...] Intel's open-source driver has
>>> > full support for 3d already, and intel stated they are not
>>> > interrested in maintaining a second implementation. I don't know
>>> > about i915, but i965's gallium driver rotted to the state that it was
>>> > thrown out of mesa trunk in november 2011 . Taking i965 out fixed
>>> > my issues, and all is well again, with acceleration and 3d.
>>> > I attached a patch (2 commits), the first commit upgrades dri2proto
>>> > so the intel 2_17_0 driver (which is already in nix) compiles. The
>>> > second commit removed i965 gallium from mesa.
>>> updating mesa causes significant re-builds, so we should wait a couple
>>> of days to give everyone a chance to express his or her concerns about
>>> that change.
>> I don't remember a vote + few days wait for turning it on?!?
>> Someone just decided to include an unmaintained, already dropped
>> upstream 2 months ago, experimental driver, breaking things and
>> causing instability.
> I sent a patch to the list about a week before applying it. Search for
> "libdrm and mesa u, pgrade" from 07 January. OTOH, I have nothing against
> reverting this part.
Ah, yes you are right. I didn't read it at the time as I'm all in
favor of upgrading mesa/libdrm/xorg.
I didn't realize upgrading meant enabling more drivers though, but
indeed if I had checked your patch, I would have seen it.
But as already stated, I don't mind stuff breaking. I do mind slow
procedures for changes :)
In this case, the i965 driver is unusable and broken, so just taking
that out seems best.
I do however think that for i915 users, gallium isn't that useful
either (compared to the normal intel driver). It's experimental and
somewhat incomplete, without heavy gains yet.
For ati/nvidia open source drivers, gallium drivers _are_ the way to
go, disabling them would just put them back to 2d.
Perhaps turning gallium drivers into options (defaulting to disabled
for intel, enabled for cards that have no alternative) is best?
> Yury G. Kudryashov,
> mailto: urkud at mccme.ru
> nix-dev mailing list
> nix-dev at lists.science.uu.nl
More information about the nix-dev