F16: Sandy Bridge -> lags, missing effects, ui-crashes

[Josh Boyer]

On Thu, May 31, 2012 at 1:02 PM, drago01 <drago01 at gmail.com> wrote:
> On Wed, May 30, 2012 at 10:32 PM, Reindl Harald <h.reindl at thelounge.net> wrote:
>>
>>
>> Am 30.05.2012 10:52, schrieb drago01:
>>> On Wed, May 30, 2012 at 2:36 AM, Jared K. Smith
>>> <jsmith at fedoraproject.org> wrote:
>>>> On Tue, May 29, 2012 at 5:29 PM, Reindl Harald <h.reindl at thelounge.net> wrote:
>>>>> Am 29.05.2012 22:45, schrieb Roberto Ragusa:
>>>>> i think i have to mention taht VT-d is active in the BIOS and
>>>>> i was wondering that it was stable because the combination of the
>>>>> i915 options with VT-d (hardware supported virtualization IO) is
>>>>> classified to be unstable
>>>>
>>>> Yes, that's a possible culprit. I've had massive problems with VT-d
>>>> enabled on both a Thinkpad T510 and on a Thinkpad X220.  I don't
>>>> pretend to understand what advantages VT-d is *supposed* to give me,
>>>> but it's the first thing I turn off in the BIOS.  In fact, on the
>>>> T510, I couldn't even get an installation to complete without turning
>>>> it off.
>>>
>>> You could just do boot with intel_iommu=igfx_off to verify that
>>
>> sounds not really smart if you are using the integrated graphics
>> engine as only graphics card which is fast enough for KDE4 3D
>> effects and all the things you need if you are not a gamer :-)
>
> You misunderstood this completely this won't "turn off your GPU" but
> just disables DMAR for it ...
> You shouldn't notice any difference neither in KDE nor in games.

You're aware that doesn't actually do anything in Fedora kernels,
right?  We've been carrying a patch called
linux-2.6-intel-iommu-igfx.patch for quite some time that removes the
igfx_off variable and replaces it with an igfx_on variable and defaults
to having the DMAR disabled without igfx_on being specified.  In fact:

From: drago01 <drago01 at gmail.com>
Date: Wed, 26 Oct 2011 13:37:27 -0400
Subject: [PATCH] Default to igfx_off


josh