Ben Widawsky writes:
> Compile tested only.
>
> Signed-off-by: Ben Widawsky
> ---
> drivers/gpu/drm/i915/intel_lrc.c | 3 ++-
> 1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/i915/intel_lrc.c
> b/drivers/gpu/drm/i915/intel_lrc.c
> index 42a7be1..a9bc207 100644
On to, 2015-12-17 at 21:10 +, Chris Wilson wrote:
> On Thu, Dec 17, 2015 at 09:25:50PM +0200, Joonas Lahtinen wrote:
> > Add helpers to reduce the amount of noise. Use the i915.debug
> > parameter
> > introduced in the previous patch to decide on whether to display
> > all
> > debug messages or
== Summary ==
Built on ac2305b6c91b9a84cc12566016ece257c3ebcba3 drm-intel-nightly:
2015y-12m-17d-16h-19m-23s UTC integration manifest
Test igt@kms_flip@basic-flip-vs-wf_vblank on bsw-nuc-2 pass -> dmesg-warn
Test igt@kms_flip@basic-flip-vs-wf_vblank on skl-i7k-2 dmesg-fail -> dmesg-warn
Test igt
== Summary ==
HEAD is now at ac2305b drm-intel-nightly: 2015y-12m-17d-16h-19m-23s UTC
integration manifest
Applying: drm/i915: HWSTAM is not a thing on SKL+
Repository lacks necessary blobs to fall back on 3-way merge.
Cannot fall back to three-way merge.
Patch failed at 0001 drm/i915: HWSTAM is
On Thu, 2015-12-17 at 15:16 +0200, Ville Syrjälä wrote:
> On Thu, Dec 17, 2015 at 02:46:03PM +0200, Mika Kahola wrote:
> > Disable DP fast link training if DP link configuration
> > changes. If one of the DP link parameters i.e. link
> > bandwidth, lane count, rate selection, port clock or bpp
>
>
Use dev_priv rather than dev pointer where applicable. Remove plenty of
unnecessary temp variables. No functional changes.
Signed-off-by: Jani Nikula
---
drivers/gpu/drm/i915/intel_panel.c | 156 ++---
1 file changed, 58 insertions(+), 98 deletions(-)
diff --git
== Summary ==
HEAD is now at ac2305b drm-intel-nightly: 2015y-12m-17d-16h-19m-23s UTC
integration manifest
Applying: drm/i915: edp resume/On time optimization.
Repository lacks necessary blobs to fall back on 3-way merge.
Cannot fall back to three-way merge.
Patch failed at 0001 drm/i915: edp res
Jani Nikula writes:
> Use dev_priv rather than dev pointer where applicable. Remove plenty of
> unnecessary temp variables. No functional changes.
>
> Signed-off-by: Jani Nikula
Reviewed-by: Mika Kuoppala
> ---
> drivers/gpu/drm/i915/intel_panel.c | 156
> ++-
On Fri, Dec 18, 2015 at 10:51:35AM +0200, Jani Nikula wrote:
> Use dev_priv rather than dev pointer where applicable. Remove plenty of
> unnecessary temp variables. No functional changes.
>
> Signed-off-by: Jani Nikula
> ---
> drivers/gpu/drm/i915/intel_panel.c | 156
> ++---
On Thu, Dec 17, 2015 at 08:59:42PM +, Chris Wilson wrote:
> On Thu, Dec 17, 2015 at 09:25:47PM +0200, Joonas Lahtinen wrote:
> > +#ifndef _I915_PARAMS_H_
> > +#define _I915_PARAMS_H_
> > +
> Is it worth saying
>
> #include /* for __read_mostly */
>
> or am I alone in not knowing that?
Anywa
On Thu, Dec 17, 2015 at 09:25:46PM +0200, Joonas Lahtinen wrote:
> Using __stringify(x) instead of #x adds support for macros as
> a parameter and reduces runtime overhead.
>
> Slightly increases the .text size but should not matter.
>
> Cc: Rob Clark
> Acked-by: Daniel Vetter
> Signed-off-by:
From: Dhanya
This patch will verify color correction capability of a display driver.
Gamma/CSC/De-gamma for SKL/BXT supported.
Signed-off-by: Dhanya
---
tests/.gitignore | 1 +
tests/Makefile.sources | 1 +
tests/kms_color.c | 999
On Fri, 18 Dec 2015, Mika Kuoppala wrote:
> Jani Nikula writes:
>
>> Use dev_priv rather than dev pointer where applicable. Remove plenty of
>> unnecessary temp variables. No functional changes.
>>
>> Signed-off-by: Jani Nikula
>
> Reviewed-by: Mika Kuoppala
Pushed to drm-intel-next-queued, th
== Summary ==
Built on ac2305b6c91b9a84cc12566016ece257c3ebcba3 drm-intel-nightly:
2015y-12m-17d-16h-19m-23s UTC integration manifest
Test igt@kms_flip@basic-flip-vs-wf_vblank on byt-nuc dmesg-warn -> skip
Test igt@kms_flip@basic-flip-vs-wf_vblank on skl-i7k-2 dmesg-fail -> dmesg-warn
Test igt@k
On Fri, 18 Dec 2015, Chris Wilson wrote:
> On Fri, Dec 18, 2015 at 10:51:35AM +0200, Jani Nikula wrote:
>> Use dev_priv rather than dev pointer where applicable. Remove plenty of
>> unnecessary temp variables. No functional changes.
>>
>> Signed-off-by: Jani Nikula
>> ---
>> drivers/gpu/drm/i91
From: Dhanya
This patch will verify color correction capability of a display driver.
Gamma/CSC/De-gamma for SKL/BXT supported.
Signed-off-by: Dhanya
---
tests/.gitignore | 1 +
tests/Makefile.sources | 1 +
tests/kms_color.c | 997
On Thu, 17 Dec 2015, yu@intel.com wrote:
> From: Alex Dai
>
> The device struct_mutex needs to be held before releasing any GEM
> objects allocated by GuC.
This is indeed so, but your patch subject needs to say it fixes an
actual bug rather than a "warning message problem" which makes one thi
On Thu, 17 Dec 2015, Imre Deak wrote:
> On Thu, 2015-12-17 at 09:49 -0800, Ben Widawsky wrote:
>> +> if (IS_GEN9(ring->dev))
>
> Nitpick: INTEL_INFO()->gen == 9 is the preferred way.
Oh? Since when and says who?
BR,
Jani.
--
Jani Nikula, Intel Open Source Technology Center
___
On to, 2015-12-17 at 21:02 +, Chris Wilson wrote:
> On Thu, Dec 17, 2015 at 09:25:49PM +0200, Joonas Lahtinen wrote:
> > Add i915.debug parameter to control output in cases where the debug
> > output
> > amount will be massive and as such is unsuitable for everyday use,
> > but still
> > signif
On Thu, 17 Dec 2015, Radhakrishna Sripada
wrote:
> Original value of 32 blocks is not sufficient when using cursor size of
> 256x256 causing FIFO underruns when the reworked wm
> caluclations in
>
> commit 024c9045221fe45482863c47c4b4c47d37f97cbf
> Author: Matt Roper
> Date: Thu Sep 24 15:53:1
One particularly stressful scenario consists of many independent tasks
all competing for GPU time and waiting upon the results (e.g. realtime
transcoding of many, many streams). One bottleneck in particular is that
each client waits on its own results, but every client is woken up after
every batch
On Thu, 17 Dec 2015, Daniel Vetter wrote:
> git 2.5 gained native support for worktrees, so let's use them.
>
> Since this is fairly new I don't think we should switch the general
> dim setup over to using it (for e.g. the maintainer-tools and other
> auxiliary branches) since 2.5 was only release
In normal cases, RC6 promotion timer is 1700us. This will result in more
time spent in C1 state. For more residency in C6 in case of media workloads,
this is changed to 250us. Not doing this for 3D workloads as too many C6-C0
transition delays can result in performance impact. Tracking the media
wo
On Fri, Dec 18, 2015 at 03:44:36PM +0530, Namrta Salonie wrote:
> In normal cases, RC6 promotion timer is 1700us. This will result in more
> time spent in C1 state. For more residency in C6 in case of media workloads,
> this is changed to 250us. Not doing this for 3D workloads as too many C6-C0
> t
== Summary ==
HEAD is now at 3c71db8 drm-intel-nightly: 2015y-12m-18d-09h-42m-49s UTC
integration manifest
Applying: drm/i915/vlv: Modifying RC6 Promotion timer for Media workloads.
Repository lacks necessary blobs to fall back on 3-way merge.
Cannot fall back to three-way merge.
Patch failed at
On Thu, Dec 17, 2015 at 07:04:33PM +0200, Imre Deak wrote:
> Since
>
> commit 357597e51dd1aa3cf764d322abf89217e3dcd7bb
> Author: Imre Deak
> Date: Tue Nov 10 06:12:22 2015 +0200
>
> drm/i915: remove HAS_RUNTIME_PM check from RPM get/put/assert helpers
>
> this file is writeable also on pl
On pe, 2015-12-18 at 11:59 +0200, Jani Nikula wrote:
> On Thu, 17 Dec 2015, Imre Deak wrote:
> > On Thu, 2015-12-17 at 09:49 -0800, Ben Widawsky wrote:
> > > +> if (IS_GEN9(ring->dev))
> >
> > Nitpick: INTEL_INFO()->gen == 9 is the preferred way.
>
> Oh? Since when and says who?
On Wed, Dec 16, 2015 at 07:06:20PM -0800, Radhakrishna Sripada wrote:
> Original value of 32 blocks is not sufficient when using cursor size of
> 256x256 causing FIFO underruns when the reworked wm
> caluclations in
>
> commit 024c9045221fe45482863c47c4b4c47d37f97cbf
> Author: Matt Roper
> Date:
Using __stringify(x) instead of #x adds support for macros as
a parameter and reduces runtime overhead.
Slightly increases the .text size but should not matter.
v2:
- Define I915_STATE_WARN_ON though I915_STATE_WARN
(Bikeshed inspiration by Chris)
Cc: Rob Clark
Cc: Chris Wilson
Acked-by: Dan
Take advantage of WARN return value to simplify the flow.
Cc: Rob Clark
Cc: Chris Wilson
Reported-by: Chris Wilson
Signed-off-by: Joonas Lahtinen
---
drivers/gpu/drm/i915/i915_drv.h | 15 +--
1 file changed, 5 insertions(+), 10 deletions(-)
diff --git a/drivers/gpu/drm/i915/i915_
Joonas Lahtinen (2):
drm/i915: Simplify _STATE_ debug macros
drm/i915: Compile time concatenate WARN_ON macro strings
drivers/gpu/drm/i915/i915_drv.h | 23 ++-
1 file changed, 6 insertions(+), 17 deletions(-)
--
2.4.3
___
Inte
From: Tvrtko Ursulin
Same effect for slightly less source code and resulting binary.
Signed-off-by: Tvrtko Ursulin
---
drivers/gpu/drm/i915/intel_lrc.c | 15 ++-
1 file changed, 6 insertions(+), 9 deletions(-)
Compile tested only!
diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/d
On Fri, Dec 18, 2015 at 12:39:17PM +0200, Joonas Lahtinen wrote:
> Using __stringify(x) instead of #x adds support for macros as
> a parameter and reduces runtime overhead.
>
> Slightly increases the .text size but should not matter.
>
> v2:
> - Define I915_STATE_WARN_ON though I915_STATE_WARN
>
Avoid flooding end users kernel log with warnings and mark recurring
warnings as such. This is a case when multiple code paths trigger a
warning and only displaying it once masks multiple errors in CI but
displaying it each time would make end user system log overflow
easily.
v2:
- Rebase for new
Wakeref series introduced multiple WARN_ONCE clauses which might be
triggered from several sources (hangcheck or cache flusing for example),
this series adds an i915.recur_warnings option and WARN_RECUR macro.
The WARN_ONCE calls are converted into WARN_RECUR which always WARNs if
recur_warnings o
Add i915.recur_warnings parameter to control output in cases where the warning
is of recurring type and is potentially called from multiple paths. Using just
WARN_ONCE would mask out other calling paths but one, this is not desireable
on developer machine or CI system, but is a compromise to be mad
Otherwise usage in the i915 debug macros yields problems due to
i915_drv.h <-> i915_trace.h <-> intel_drv.h include loops.
v2:
- Document not-so-obvious need for linux/cache.h (Chris)
Reviewed-by: Chris Wilson
Signed-off-by: Joonas Lahtinen
---
drivers/gpu/drm/i915/i915_drv.h| 40 ++---
Move all the bool variables to the end as per the comment.
Reviewed-by: Chris Wilson
Signed-off-by: Joonas Lahtinen
---
drivers/gpu/drm/i915/i915_params.h | 8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/drivers/gpu/drm/i915/i915_params.h
b/drivers/gpu/drm/i915/i915_
Using __stringify(x) instead of #x adds support for macros as
a parameter and compile-time concatenation reduces the runtime
overhead.
Slightly increases the .text size but should not matter.
v2:
- Define I915_STATE_WARN_ON though I915_STATE_WARN
(Bikeshed inspiration by Chris)
v3:
- More spec
Joonas Lahtinen (2):
drm/i915: Simplify _STATE_ debug macros
drm/i915: Compile-time concatenate WARN_ON macro strings
drivers/gpu/drm/i915/i915_drv.h | 23 ++-
1 file changed, 6 insertions(+), 17 deletions(-)
--
2.4.3
___
Inte
Take advantage of WARN return value to simplify the flow.
Cc: Rob Clark
Reviewed-by: Chris Wilson
Reported-by: Chris Wilson
Signed-off-by: Joonas Lahtinen
---
drivers/gpu/drm/i915/i915_drv.h | 15 +--
1 file changed, 5 insertions(+), 10 deletions(-)
diff --git a/drivers/gpu/drm/i
== Summary ==
Built on 3c71db89fc37b6061ce6b070ce73f89155da5f20 drm-intel-nightly:
2015y-12m-18d-09h-42m-49s UTC integration manifest
Test igt@kms_flip@basic-flip-vs-wf_vblank on ivb-t430s pass -> dmesg-warn
Test igt@kms_flip@basic-flip-vs-wf_vblank on byt-nuc dmesg-warn -> skip
Test igt@kms_fli
On Fri, Dec 18, 2015 at 01:08:17PM +0200, Joonas Lahtinen wrote:
> Add i915.recur_warnings parameter to control output in cases where the warning
> is of recurring type and is potentially called from multiple paths. Using just
> WARN_ONCE would mask out other calling paths but one, this is not desi
== Summary ==
Built on 3c71db89fc37b6061ce6b070ce73f89155da5f20 drm-intel-nightly:
2015y-12m-18d-09h-42m-49s UTC integration manifest
Test igt@kms_flip@basic-flip-vs-wf_vblank on ivb-t430s pass -> dmesg-warn
Test igt@kms_flip@basic-flip-vs-wf_vblank on byt-nuc dmesg-warn -> pass
Test igt@kms_pip
From: Tvrtko Ursulin
We can rely on context complete interrupt to wake up the waiters
apart in the case where requests are merged into a single ELSP
submission. In this case we inject MI_USER_INTERRUPTS in the
ring buffer to ensure prompt wake-ups.
This optimization has the effect on for example
On 18/12/15 10:39, Joonas Lahtinen wrote:
Using __stringify(x) instead of #x adds support for macros as
a parameter and reduces runtime overhead.
Slightly increases the .text size but should not matter.
v2:
- Define I915_STATE_WARN_ON though I915_STATE_WARN
(Bikeshed inspiration by Chris)
C
On pe, 2015-12-18 at 12:03 +, Dave Gordon wrote:
> On 18/12/15 10:39, Joonas Lahtinen wrote:
> > Using __stringify(x) instead of #x adds support for macros as
> > a parameter and reduces runtime overhead.
> >
> > Slightly increases the .text size but should not matter.
> >
> > v2:
> > - Defin
== Summary ==
Built on 3c71db89fc37b6061ce6b070ce73f89155da5f20 drm-intel-nightly:
2015y-12m-18d-09h-42m-49s UTC integration manifest
Test igt@kms_flip@basic-flip-vs-wf_vblank on ivb-t430s pass -> dmesg-warn
Test igt@kms_flip@basic-flip-vs-wf_vblank on bsw-nuc-2 pass -> dmesg-warn
Test igt@kms_p
v3:
- Fixed potential OOPS scenario with modulo in condition
(Dave Gordon)
Joonas Lahtinen (2):
drm/i915: Simplify _STATE_ debug macros
drm/i915: Compile-time concatenate WARN_ON macro strings
drivers/gpu/drm/i915/i915_drv.h | 23 ++-
1 file changed, 6 insertions(+), 17
Take advantage of WARN return value to simplify the flow.
Cc: Rob Clark
Reviewed-by: Chris Wilson
Reported-by: Chris Wilson
Signed-off-by: Joonas Lahtinen
---
drivers/gpu/drm/i915/i915_drv.h | 15 +--
1 file changed, 5 insertions(+), 10 deletions(-)
diff --git a/drivers/gpu/drm/i
Using __stringify(x) instead of #x adds support for macros as
a parameter and compile-time concatenation reduces the runtime
overhead.
Slightly increases the .text size but should not matter.
v2:
- Define I915_STATE_WARN_ON though I915_STATE_WARN
(Bikeshed inspiration by Chris)
v3:
- More spec
On Fri, Dec 18, 2015 at 11:59:41AM +, Tvrtko Ursulin wrote:
> From: Tvrtko Ursulin
>
> We can rely on context complete interrupt to wake up the waiters
> apart in the case where requests are merged into a single ELSP
> submission. In this case we inject MI_USER_INTERRUPTS in the
> ring buffer
== Summary ==
HEAD is now at 3c71db8 drm-intel-nightly: 2015y-12m-18d-09h-42m-49s UTC
integration manifest
Applying: drm/i915/bdw+: Do not emit user interrupts when not needed
Using index info to reconstruct a base tree...
M drivers/gpu/drm/i915/intel_lrc.c
Falling back to patching base and
On Fri, Dec 18, 2015 at 11:59:41AM +, Tvrtko Ursulin wrote:
> From: Tvrtko Ursulin
>
> We can rely on context complete interrupt to wake up the waiters
> apart in the case where requests are merged into a single ELSP
> submission. In this case we inject MI_USER_INTERRUPTS in the
> ring buffer
== Summary ==
Built on 3c71db89fc37b6061ce6b070ce73f89155da5f20 drm-intel-nightly:
2015y-12m-18d-09h-42m-49s UTC integration manifest
Test igt@kms_flip@basic-flip-vs-wf_vblank on ivb-t430s pass -> dmesg-warn
Test igt@kms_flip@basic-flip-vs-wf_vblank on bsw-nuc-2 pass -> dmesg-warn
Test igt@gem_s
On 18/12/15 12:28, Chris Wilson wrote:
On Fri, Dec 18, 2015 at 11:59:41AM +, Tvrtko Ursulin wrote:
From: Tvrtko Ursulin
We can rely on context complete interrupt to wake up the waiters
apart in the case where requests are merged into a single ELSP
submission. In this case we inject MI_USE
On pe, 2015-12-18 at 12:32 +0200, David Weinehall wrote:
> On Thu, Dec 17, 2015 at 07:04:33PM +0200, Imre Deak wrote:
> > Since
> >
> > commit 357597e51dd1aa3cf764d322abf89217e3dcd7bb
> > Author: Imre Deak
> > Date: Tue Nov 10 06:12:22 2015 +0200
> >
> > drm/i915: remove HAS_RUNTIME_PM che
== Summary ==
Built on 3c71db89fc37b6061ce6b070ce73f89155da5f20 drm-intel-nightly:
2015y-12m-18d-09h-42m-49s UTC integration manifest
Test gem_mmap:
Subgroup basic:
dmesg-warn -> PASS (byt-nuc)
Test gem_mmap_gtt:
Subgroup basic-write:
dmesg-w
== Summary ==
Built on 3c71db89fc37b6061ce6b070ce73f89155da5f20 drm-intel-nightly:
2015y-12m-18d-09h-42m-49s UTC integration manifest
Test gem_mmap:
Subgroup basic:
dmesg-warn -> PASS (byt-nuc)
Test gem_mmap_gtt:
Subgroup basic-write:
dmesg-w
The workarounds for disabling hdc invalidation and also forcing
context to be non coherent, are advised to be used up until rev D0.
However as it was found that rev F0, without the
WaForceEnableNonCoherent might system hang if the mesa
tried to use coherent mode.
As these two workarounds are abou
> Looking at dmesg I don't see the stuck pageflip anymore. It looks like the
> DDX is simply falling over itself, with now blame to the kernel here. Or
> if there is it's a new/different bug. So we need to restart triage and
> check whether this was introduced by a ddx or kernel upgrade.
> -Daniel
On Fri, Dec 18, 2015 at 01:51:38PM +, Tvrtko Ursulin wrote:
>
> On 18/12/15 12:28, Chris Wilson wrote:
> >On Fri, Dec 18, 2015 at 11:59:41AM +, Tvrtko Ursulin wrote:
> >>From: Tvrtko Ursulin
> >>
> >>We can rely on context complete interrupt to wake up the waiters
> >>apart in the case wh
Swap the order of context & engine cleanup, so that it is now
contexts, then engines.
This allows the context clean up code to do things like confirm
that ring->dev->struct_mutex is locked without a NULL pointer
dereference.
This came about as a result of the 'intel_ring_initialized() must
be simpl
Mika Kuoppala writes:
> The workarounds for disabling hdc invalidation and also forcing
> context to be non coherent, are advised to be used up until rev D0.
>
> However as it was found that rev F0, without the
> WaForceEnableNonCoherent might system hang if the mesa
> tried to use coherent mode.
Use the first retired request on a new context to unpin
the old context. This ensures that the hw context remains
bound until it has been written back to by the GPU.
Now that the context is pinned until later in the request/context
lifecycle, it no longer needs to be pinned from context_queue to
re
== Summary ==
Built on da33ddbb03875d7e1e5f06ac445368f97dcc3488 drm-intel-nightly:
2015y-12m-18d-13h-53m-06s UTC integration manifest
Test gem_mmap_gtt:
Subgroup basic-write-no-prefault:
pass -> DMESG-WARN (bsw-nuc-2)
Test kms_flip:
Subgroup basic-flip-vs-mo
On Fri, Dec 18, 2015 at 12:35:47PM +0200, Ville Syrjälä wrote:
> On Wed, Dec 16, 2015 at 07:06:20PM -0800, Radhakrishna Sripada wrote:
> > Original value of 32 blocks is not sufficient when using cursor size of
> > 256x256 causing FIFO underruns when the reworked wm
> > caluclations in
> >
> > com
== Summary ==
HEAD is now at da33ddb drm-intel-nightly: 2015y-12m-18d-13h-53m-06s UTC
integration manifest
Applying: drm/i915: Extend LRC pinning to cover GPU context writeback
Repository lacks necessary blobs to fall back on 3-way merge.
Cannot fall back to three-way merge.
Patch failed at 0001
On Fri, Dec 18, 2015 at 06:58:58AM -0800, Matt Roper wrote:
> On Fri, Dec 18, 2015 at 12:35:47PM +0200, Ville Syrjälä wrote:
> > On Wed, Dec 16, 2015 at 07:06:20PM -0800, Radhakrishna Sripada wrote:
> > > Original value of 32 blocks is not sufficient when using cursor size of
> > > 256x256 causing
On Fri, Dec 18, 2015 at 05:10:12PM +0200, Ville Syrjälä wrote:
> On Fri, Dec 18, 2015 at 06:58:58AM -0800, Matt Roper wrote:
> > On Fri, Dec 18, 2015 at 12:35:47PM +0200, Ville Syrjälä wrote:
> > > On Wed, Dec 16, 2015 at 07:06:20PM -0800, Radhakrishna Sripada wrote:
> > > > Original value of 32 bl
Hi,
A couple of people told me I should be able to set a blob property to
NULL from user space using a blob id 0. I couldn't get that to work
(probably nobody's doing this at the moment). Here is a patch to make
that work.
Cheers,
-
Lionel
Lionel Landwerlin (1):
drm/atomic: allow setting a bl
Signed-off-by: Lionel Landwerlin
---
drivers/gpu/drm/drm_atomic.c | 12 +++-
1 file changed, 7 insertions(+), 5 deletions(-)
diff --git a/drivers/gpu/drm/drm_atomic.c b/drivers/gpu/drm/drm_atomic.c
index 65f007a..b8c90a4 100644
--- a/drivers/gpu/drm/drm_atomic.c
+++ b/drivers/gpu/drm/drm
On Fri, Dec 18, 2015 at 07:14:17AM -0800, Matt Roper wrote:
> On Fri, Dec 18, 2015 at 05:10:12PM +0200, Ville Syrjälä wrote:
> > On Fri, Dec 18, 2015 at 06:58:58AM -0800, Matt Roper wrote:
> > > On Fri, Dec 18, 2015 at 12:35:47PM +0200, Ville Syrjälä wrote:
> > > > On Wed, Dec 16, 2015 at 07:06:20P
== Summary ==
HEAD is now at da33ddb drm-intel-nightly: 2015y-12m-18d-13h-53m-06s UTC
integration manifest
Applying: drm/atomic: allow setting a blob to NULL using id = 0
Repository lacks necessary blobs to fall back on 3-way merge.
Cannot fall back to three-way merge.
Patch failed at 0001 drm/at
On Fri, Dec 18, 2015 at 07:14:17AM -0800, Matt Roper wrote:
> On Fri, Dec 18, 2015 at 05:10:12PM +0200, Ville Syrjälä wrote:
> > On Fri, Dec 18, 2015 at 06:58:58AM -0800, Matt Roper wrote:
> > > On Fri, Dec 18, 2015 at 12:35:47PM +0200, Ville Syrjälä wrote:
> > > > On Wed, Dec 16, 2015 at 07:06:20P
On Fri, Dec 18, 2015 at 10:37:26AM +0200, Mika Kahola wrote:
> On Thu, 2015-12-17 at 15:16 +0200, Ville Syrjälä wrote:
> > On Thu, Dec 17, 2015 at 02:46:03PM +0200, Mika Kahola wrote:
> > > Disable DP fast link training if DP link configuration
> > > changes. If one of the DP link parameters i.e. l
On 17/12/15 19:00, Dave Gordon wrote:
On 17/12/15 12:27, Chris Wilson wrote:
On Thu, Dec 17, 2015 at 11:09:54AM +, Nick Hoath wrote:
[snip!]
If you want to make such a big change, eliminate the default_ctx from
execlists.
-Chris
No, we need the default (or global) context for idling the
With kernel 4.3.x (tested x=2,3) the screen goes blank as soon as the
system boots (both for VGA console and graphics). The only way to get
it to turn on is to suspend and resume the system.
The system is a Dell XPS 13 (latest rev) running OpenSuse Leap 42.1 and
this is the lspci output:
00:00.0
On Fri, Dec 11, 2015 at 11:32:59AM +, Chris Wilson wrote:
> Limit busywaiting only to the request currently being processed by the
> GPU. If the request is not currently being processed by the GPU, there
> is a very low likelihood of it being completed within the 2 microsecond
> spin timeout an
On 18/12/15 12:27, Joonas Lahtinen wrote:
Take advantage of WARN return value to simplify the flow.
Cc: Rob Clark
Reviewed-by: Chris Wilson
Reported-by: Chris Wilson
Signed-off-by: Joonas Lahtinen
---
drivers/gpu/drm/i915/i915_drv.h | 15 +--
1 file changed, 5 insertions(+), 1
On Fri, Dec 11, 2015 at 03:13:37PM +0100, Michał Winiarski wrote:
> According to bspec, some parts of HW require the addresses to be in
> a canonical form, where bits [63:48] == [47]. Let's convert addresses to
> canonical form prior to relocating and return converted offsets to
> userspace. We als
Hi Dave,
Seems I lied in my last drm-misc pull request and suddenly there's a big
pile of random stuff. Boris dug out Thierry's drm-trivial branch and
resubmitted everything since that branch didn't really work out.
On top of that Nicolas' changes to drm_dev_set_unique - this might
conflict with
[Paging danvet to the bottom paragraphs re client-cap ...]
Hi Lionel,
I've still got quite a few concerns about the implementation as it
stands. Some are minor quibbles (e.g. can't unset blob IDs), some are
larger design issues, some are rehashed comments from previous series,
and some are new now
Hi,
On 17 December 2015 at 18:57, Lionel Landwerlin
wrote:
> @@ -289,22 +289,30 @@ static const struct intel_device_info
> intel_haswell_m_info = {
> static const struct intel_device_info intel_broadwell_d_info = {
> HSW_FEATURES,
> .gen = 8,
> + .num_samples_after_ctm = B
Hi,
On 17 December 2015 at 18:57, Lionel Landwerlin
wrote:
> @@ -311,7 +312,9 @@ static const struct intel_device_info
> intel_cherryview_info = {
> .gen = 8, .num_pipes = 3,
> .need_gfx_hws = 1, .has_hotplug = 1,
> .ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_R
Hi,
On 17 December 2015 at 18:57, Lionel Landwerlin
wrote:
> void intel_crtc_attach_color_properties(struct drm_crtc *crtc)
> {
> + struct drm_device *dev = crtc->dev;
> + struct drm_mode_config *config = &dev->mode_config;
> + struct drm_mode_object *mode_obj = &crtc->base;
>
On 18 December 2015 at 15:17, Lionel Landwerlin
wrote:
> Signed-off-by: Lionel Landwerlin
> ---
> drivers/gpu/drm/drm_atomic.c | 12 +++-
> 1 file changed, 7 insertions(+), 5 deletions(-)
>
> diff --git a/drivers/gpu/drm/drm_atomic.c b/drivers/gpu/drm/drm_atomic.c
> index 65f007a..b8c90a
Hi,
On 18 December 2015 at 15:17, Lionel Landwerlin
wrote:
> A couple of people told me I should be able to set a blob property to
> NULL from user space using a blob id 0. I couldn't get that to work
> (probably nobody's doing this at the moment). Here is a patch to make
> that work.
You can't
On 18/12/15 09:49, Jani Nikula wrote:
On Fri, 18 Dec 2015, Chris Wilson wrote:
On Fri, Dec 18, 2015 at 10:51:35AM +0200, Jani Nikula wrote:
Use dev_priv rather than dev pointer where applicable. Remove plenty of
unnecessary temp variables. No functional changes.
Signed-off-by: Jani Nikula
--
From: Ville Syrjälä
Turns out CHV pipe C was glued on somewhat poorly, and there's something
wrong with the cursor. If the cursor straddles the left screen edge,
and is then moved away from the edge or disabled, the pipe will often
underrun. If enough underruns are triggered quickly enough the pi
From: Ville Syrjälä
The extra_long_opts passed to igt_*_parse_opts() isn't modified,
so let's make it const.
Signed-off-by: Ville Syrjälä
---
lib/igt_core.c | 6 +++---
lib/igt_core.h | 4 ++--
2 files changed, 5 insertions(+), 5 deletions(-)
diff --git a/lib/igt_core.c b/lib/igt_core.c
index
From: Ville Syrjälä
It's nice to see just how many components the crc claims to have
when the count don't match what we expect.
Signed-off-by: Ville Syrjälä
---
lib/igt_debugfs.c | 9 -
1 file changed, 4 insertions(+), 5 deletions(-)
diff --git a/lib/igt_debugfs.c b/lib/igt_debugfs.c
From: Ville Syrjälä
Add support for reading the CRC in non-blocking mode. Useful for tests
that want to start the CRC capture, then do a bunch of operations, then
collect however many CRCs that got generated. The current
igt_pipe_crc_new() + igt_pipe_crc_get_crcs() method would block until
it get
From: Ville Syrjälä
Signed-off-by: Ville Syrjälä
---
tests/kms_pipe_crc_basic.c | 33 +
1 file changed, 29 insertions(+), 4 deletions(-)
diff --git a/tests/kms_pipe_crc_basic.c b/tests/kms_pipe_crc_basic.c
index a3292c225203..eb8e15e7f413 100644
--- a/tests/kms_
From: Ville Syrjälä
Several tests do one or more of the followin:
* igt_create_fb() + igt_paint_test_pattern()
* igt_create_color_fb() + igt_paint_test_pattern()
* igt_create_fb() + igt_paint_image()
Extract them into new helpes: igt_create_pattern_fb(),
igt_create_color_pattern_fb(), igt_create
From: Ville Syrjälä
The test tries to anger CHV pipe C cursor by walking the edges of the
screen while moving the cursor across the screen edge.
The actual hw issue only occurs on pipe C, and only on the left screen
edge. The testcase can walk all the edges though, and on all pipes, just
so I co
On Fri, Dec 18, 2015 at 10:17:49AM +0200, Mika Kuoppala wrote:
> Ben Widawsky writes:
>
> > Compile tested only.
> >
> > Signed-off-by: Ben Widawsky
> > ---
> > drivers/gpu/drm/i915/intel_lrc.c | 3 ++-
> > 1 file changed, 2 insertions(+), 1 deletion(-)
> >
> > diff --git a/drivers/gpu/drm/i915
== Summary ==
Built on deb496ede4fc5e7affe15640a113ea9b5f1189e0 drm-intel-nightly:
2015y-12m-18d-16h-12m-24s UTC integration manifest
Test gem_storedw_loop:
Subgroup basic-render:
dmesg-warn -> PASS (skl-i5k-2)
Test kms_flip:
Subgroup basic-flip-vs-modeset:
Hello,
is it possible to switch vsync from internal LCD (laptop) to external monitor
when playing movie via VA-API? I can drop vsync with internal LCD and keep
vsync only with external monitor. I tried to use --primary for excernal
monitor when enabling output but no luck.
Is there any way to sy
On 15 December 2015 at 09:14, David Weinehall
wrote:
> Move power management related code to a separate library.
> Initially this is done only for workarounds that apply to external
> components. Modify the users of such workarounds accordingly.
> This currently involves HD audio and SATA link po
1 - 100 of 117 matches
Mail list logo
