[Intel-gfx] [PATCH] drm/i915: Fix dynamic allocation of physical handles

[Chris Wilson]

On Thu, May 15, 2014 at 12:56:48PM +0300, Jani Nikula wrote:
> On Wed, 14 May 2014, Chris Wilson <chris at chris-wilson.co.uk> wrote:
> > A single object may be referenced by multiple registers fundamentally
> > breaking the static allotment of ids in the current design. When the
> > object is used the second time, the physical address of the first
> > assignment is relinquished and a second one granted. However, the
> > hardware is still reading (and possibly writing) to the old physical
> > address now returned to the system. Eventually hilarity will ensue, but
> > in the short term, it just means that cursors are broken when using more
> > than one pipe.
> >
> > v2: Fix up leak of pci handle when handling an error during attachment,
> > and avoid a double kmap/kunmap. (Ville)
> > Rebase against -fixes.
> >
> > Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=77351
> > Signed-off-by: Chris Wilson <chris at chris-wilson.co.uk>
> > Cc: Ville Syrjälä <ville.syrjala at linux.intel.com>
> > Cc: Jani Nikula <jani.nikula at linux.intel.com>
> > Cc: stable at vger.kernel.org
> > ---
> >  drivers/gpu/drm/i915/i915_dma.c      |   1 -
> >  drivers/gpu/drm/i915/i915_drv.h      |  24 +--
> >  drivers/gpu/drm/i915/i915_gem.c      | 319 ++++++++++++++---------------------
> >  drivers/gpu/drm/i915/intel_display.c |  11 +-
> >  drivers/gpu/drm/i915/intel_overlay.c |  12 +-
> >  5 files changed, 136 insertions(+), 231 deletions(-)
> 
> With this diffstat, this late in rc, bug referring to 855GM (is this
> required on others?!), I want r-b/t-b on this.

It affects nearly all gen2/gen3 machines. You can punt it to Daniel, but
keep cc: stable, as I want to build upon this patch to improve phys
objs, so having it heading upstream is essential.
-Chris

-- 
Chris Wilson, Intel Open Source Technology Centre