In Haswell while disabling a pipe, we need to disable the DDI control as
well along with the PIPECONF. Otherwise we will hit assertions during crtc
disable
Signed-off-by: Shobhit Kumar
---
drivers/gpu/drm/i915/intel_display.c |7 +++
1 files changed, 7 insertions(+), 0 deletions(-)
diff
Summary
We finished a new round of kernel testing. During this QA circle, 6 new bugs
are still open, 5 more bugs have been fixed, and besides 4 old open bugs.
Specially we checked all the 3-pipe combinations which we have on Ivybridge
platform, all the 6 combinations work.
Test Environment
The v
On Thu, 7 Jun 2012 21:08:49 +0200
Daniel Vetter wrote:
> On Thu, Jun 07, 2012 at 09:12:42AM -0700, Ben Widawsky wrote:
> > Use the new libdrm functionality to actually do timed waits on the sync
> > object.
> >
> > This patch is missing the configure.ac update to check for the correct
> > libdrm
On Thu, Jun 07, 2012 at 09:12:42AM -0700, Ben Widawsky wrote:
> Use the new libdrm functionality to actually do timed waits on the sync
> object.
>
> This patch is missing the configure.ac update to check for the correct
> libdrm supporting this function. As of now, libdrm has not yet received
> t
On Tue, 05 Jun 2012 16:37:20 -0700
Kenneth Graunke wrote:
> On 06/04/2012 02:42 PM, Ben Widawsky wrote:
> > Setting myself up for a late night crying session once again. Most of the
> > people reading this probably know the history and reasons for the patches.
> > If
> > not, you can search the
Use the new libdrm functionality to actually do timed waits on the sync
object.
This patch is missing the configure.ac update to check for the correct
libdrm supporting this function. As of now, libdrm has not yet received
the version bump. That's mostly why this patch is "RFC"
Since intel_client
int drm_intel_gem_bo_wait(drm_intel_bo *bo, uint64_t timeout_ns)
This should bump the libdrm version. We're waiting for context support
so we can do both features in one bump.
v2: don't return remaining timeout amount
use get param and fallback for older kernels
v3: only doing getparam at init
p
drm/i915 now takes care itself of setting up the gtt for
these chips.
Signed-off-by: Daniel Vetter
---
drivers/char/agp/intel-agp.c | 11 ---
1 files changed, 0 insertions(+), 11 deletions(-)
diff --git a/drivers/char/agp/intel-agp.c b/drivers/char/agp/intel-agp.c
index c98c568..39c4a
When drm/i915 is in control of the gtt, we need to call
the enable function at all the relevant places ourselves.
Signed-off-by: Daniel Vetter
---
drivers/char/agp/intel-gtt.c|3 ++-
drivers/gpu/drm/i915/i915_gem.c |3 +++
include/drm/intel-gtt.h |2 ++
3 files changed, 7
We need this thing much earlier, and it doesn't make sense
in the hw enabling function intel_enable_gtt - this does not
change over a suspend/resume cycle ...
Signed-off-by: Daniel Vetter
---
drivers/char/agp/intel-gtt.c | 20 ++--
1 files changed, 10 insertions(+), 10 deletion
We won't enabled agp unconditionally. Furthermore we do have
the right checks for agp support in place in our ->load
function.
The only thing this variable still does is enforce the module
load order we want (and I'm not even sure whether it succeeds
for that). Hence just use it in a harmless plac
To be able to directly set up the intel-gtt code from drm/i915 and
avoid setting up the fake-agp driver we need to prepare a few things:
- pass both the bridge and gpu pci_dev to the probe function and add
code to handle the gpu pdev both being present (for drm/i915) and
not present (fake agp).
We need it for all things ums (which essentially only means up to
gen5) and to support b0rked XvMC userspace on gen3.
Signed-Off-by: Daniel Vetter
---
drivers/gpu/drm/i915/i915_dma.c | 21 -
1 files changed, 12 insertions(+), 9 deletions(-)
diff --git a/drivers/gpu/drm/i91
We only need it to fake the agp interface and don't actually
use it in the driver anywhere. Hence conditionalize that.
This is just a prep patch to eventually disable the fake agp
driver on gen6+.
Signed-off-by: Daniel Vetter
---
drivers/char/agp/intel-gtt.c |8 +---
1 files changed, 5
For that to work we need to export the base address of the gtt
mmio window from intel-gtt. Also replace all other uses of
dev->agp by values we already have at hand.
Signed-off-by: Daniel Vetter
---
drivers/char/agp/intel-gtt.c|5 ++---
drivers/gpu/drm/i915/i915_dma.c |
This is a leftover from the conversion of the i81x fake agp driver
over to the new intel-gtt code layoute.
Signed-Off-by: Daniel Vetter
---
drivers/char/agp/intel-gtt.c |3 ---
1 files changed, 0 insertions(+), 3 deletions(-)
diff --git a/drivers/char/agp/intel-gtt.c b/drivers/char/agp/inte
All drivers that use agp call the agp_init function now directly from
their ->load callback. All other parts check for dev->agp anyway to
check whether agp is available.
This flag has therefore outlived its usefullness, kill it.
Signed-Off-by: Daniel Vetter
---
drivers/gpu/drm/i810/i810_drv.c
Only i810 and i915 use this. But now that the driver's ->load function
is responsible for setting up agp, we can simply move this check in
there.
Signed-Off-by: Daniel Vetter
---
drivers/gpu/drm/drm_pci.c |6 +-
drivers/gpu/drm/i810/i810_dma.c |5 +
drivers/gpu/drm/i810/i81
I've done an audit of everything which is being set up between the
place where drm_pci_agp_init is called currently and where the
driver's ->load function is called. Nothing seems to depend upon
dev->agp being set up correctly.
This is one big step into squashing this giant midlayer mistaken.
Thou
Hi all,
With these patches there's no more /dev/agpgart fake agp interface for drm/i915,
at least not for snb and later.
The first 3 patches rework the drm core to move agp initialization into drivers.
A nice bonus is that these remove the mid-layer stench quite a bit from drm ...
The later patc
On Thu, 7 Jun 2012 09:10:08 +0200
Daniel Vetter wrote:
> On Wed, Jun 06, 2012 at 04:42:11PM -0700, Ben Widawsky wrote:
> > int drm_intel_gem_bo_wait(drm_intel_bo *bo, uint64_t timeout_ns)
> >
> > This should bump the libdrm version. We're waiting for context support
> > so we can do both feature
On Wed, Jun 06, 2012 at 04:42:11PM -0700, Ben Widawsky wrote:
> int drm_intel_gem_bo_wait(drm_intel_bo *bo, uint64_t timeout_ns)
>
> This should bump the libdrm version. We're waiting for context support
> so we can do both features in one bump.
>
> v2: don't return remaining timeout amount
> use
22 matches
Mail list logo
