Hi all,
The following two patches reduce the size of struct drm_i915_gem_object by
48 bytes (on 64bit). No impact in benchmarking whatsoever. Still, given how
easy it was and furthermore that small bos seem to be on the rise (again),
I think this is a worthwhile optimization.
Please review and me
This lru tracks fences, not objects, so move it to where it belongs.
As a side effect, this nicely shrinks drm_i915_gem_object by two
pointers.
Signed-off-by: Daniel Vetter
---
drivers/gpu/drm/i915/i915_drv.h |4 +---
drivers/gpu/drm/i915/i915_gem.c | 33 +
This saves a whooping 8 dwords. Zero functional changes. Because
some of the refcounts are rather tightly calculated, I've put
BUG_ONs in the code to check for overflows.
Signed-off-by: Daniel Vetter
---
drivers/gpu/drm/i915/i915_drv.h | 87 --
drivers/gpu/d
On Wed, 2010-04-28 at 08:30 +0200, Tino Keitel wrote:
> Your X server uses the VESA driver. This list is about the intel
> driver. As 1024x600 is not a resolution defined by VESA, it won't be
> used by the VESA driver. If you switch to the intel driver, it should
> work (and also give reasonable a
Adam,
Again, thanks for your help...
On Tue, 27 Apr 2010, Adam Jackson wrote:
No, that suggests that you have a KMS driver, just that it's buggy.
It'd be called i915.ko.
At this point you probably need to consider it a Debian kernel bug. I'd
try a 2.6.34 build, if they have any.
I build 2
On Wed, 2010-04-28 at 10:15 +0800, Zhenyu Wang wrote:
> On 2010.04.23 16:16:12 -0400, Adam Jackson wrote:
> > Multifunction SDVO cards stopped working after 14571b4, and would report
> > something that looked remarkably like an ADD2 SPD ROM instead of EDID.
> > This appears to be because DDC bus se
