This is called from execlist context init which we need to be unlocked.
Commit f89823c21224 ("drm/i915/perf: Implement
I915_PERF_ADD/REMOVE_CONFIG interface") added a lockdep assert to this
path for unclear reasons, remove it again!
Fixes: 701f8231a2fe ("drm/i915/perf: prune OA configs")
Signed-of
As we cannot have an exclusive stream set if the perf has not been
initialized, we only need to check for that exclusive stream.
Signed-off-by: Chris Wilson
Cc: Lionel Landwerlin
Cc: Matthew Auld
---
drivers/gpu/drm/i915/i915_perf.c | 7 ++-
1 file changed, 2 insertions(+), 5 deletions(-)
Quoting Chris Wilson (2017-08-10 18:57:41)
> Another case where we need to call sysfs_attr_init() to setup the
> internal lockdep class prior to use:
Just a reminder that lockdep is currently disabled in CI due to this bug.
-Chris
___
Intel-gfx mailing l
When FBC is enabled for linear, legacy Y-tiled and Yf-tiled
surfaces on gen9, the cfb stride must be programmed by SW as
cfb_stride = ceiling[(at least plane width in pixels)/
(32 * compression limit factor)] * 8
v2: Minor fix for a build error
v3: Fixed subject, register nam
On Thu, 2017-08-10 at 19:45 +0200, Stefan Assmann wrote:
> On 2017-08-10 07:50, Rodrigo Vivi wrote:
> > I'm not sure if this is really the case and I don't believe
> > this is the real fix for the bug mentioned here, but since
> > I don't see a reliable path when mst_port is set and when
> > mode_v
Cleanup the code. Map the pins in accordance to
individual platforms rather than according to ports.
Create separate functions for platforms.
v2:
- Add missing condition for CoffeeLake. Make platform
specific functions static. Add function
i915_ddc_pin_mapping().
Sugested-by Ville Syrjala
Cc:
On 10 August 2017 at 18:57, Chris Wilson wrote:
> Another case where we need to call sysfs_attr_init() to setup the
> internal lockdep class prior to use:
>
> [9.325229] BUG: key 880168bc7bb0 not in .data!
> [9.325240] DEBUG_LOCKS_WARN_ON(1)
> [9.325250] [ cut here ]---
On 10 August 2017 at 18:57, Chris Wilson wrote:
> This is called from execlist context init which we need to be unlocked.
> Commit f89823c21224 ("drm/i915/perf: Implement
> I915_PERF_ADD/REMOVE_CONFIG interface") added a lockdep assert to this
> path for unclear reasons, remove it again!
>
> Fixes
On 8/10/2017 1:32 AM, Arkadiusz Hiler wrote:
On Wed, Aug 09, 2017 at 10:00:16AM -0700, Belgaumkar, Vinay wrote:
On 8/9/2017 7:32 AM, Arkadiusz Hiler wrote:
On Tue, Aug 08, 2017 at 03:09:00PM -0700, Vinay Belgaumkar wrote:
This is an RFC for adding documentation to IGT subtests. Each subtest
On 10 August 2017 at 18:57, Chris Wilson wrote:
> As we cannot have an exclusive stream set if the perf has not been
> initialized, we only need to check for that exclusive stream.
>
> Signed-off-by: Chris Wilson
> Cc: Lionel Landwerlin
> Cc: Matthew Auld
Reviewed-by: Matthew Auld
On Thu, 2017-08-10 at 05:47 +, Navare, Manasi D wrote:
> We currently assume port A is connected to a DP sink when VBT is absent,
> instead assume it is connected to an eDP sink, which seems like a more common
> configuration. Although I don't have data to back this up, it is still just
> as
The first contention point we hit using syncobjs from concurrent
threads is the spinlock guarding drm_syncobj_find(). We use an RCU-safe
idr to store the syncobj, so if we employ RCU to free drm_syncobj and be
careful in acquiring the reference (as we may now observe zombie
synocbjs in the idr) we
On Wed, Aug 9, 2017 at 4:13 PM, Joe Kniss wrote:
> On Wed, Aug 9, 2017 at 12:14 PM, Noralf Trønnes wrote:
>>
>> Den 09.08.2017 01.42, skrev Joe Kniss:
>>>
>>> Because all drivers currently use gem objects for framebuffer planes,
>>> the virtual create_handle() is not required. This change adds a
On Thu, Aug 10, 2017 at 9:13 AM, Jani Nikula wrote:
> On Thu, 10 Aug 2017, Rodrigo Vivi wrote:
>> On Thu, Aug 10, 2017 at 5:29 AM, Jani Nikula wrote:
>>> Remove old warning about docproc directive that's not supported in the
>>> Sphinx toolchain.
>>
>> I got confused... the warning is not suppor
On Thu, 10 Aug 2017, "Pandiyan, Dhinakaran"
wrote:
> On Thu, 2017-08-10 at 05:47 +, Navare, Manasi D wrote:
>> We currently assume port A is connected to a DP sink when VBT is absent,
>> instead assume it is connected to an eDP sink, which seems like a more
>> common configuration. Although
On Thu, 10 Aug 2017, Patchwork wrote:
> == Series Details ==
>
> Series: drm/i915: add register macro definition style guide (rev2)
> URL : https://patchwork.freedesktop.org/series/28370/
> State : failure
>
> == Summary ==
>
> Series 28370v2 drm/i915: add register macro definition style guide
>
On Thu, Aug 10, 2017 at 12:39:46PM +0300, Jani Nikula wrote:
> Way back when the WaveDrom stuff was added, the scripts could only be
> accessed over http. This caused issues with sites served over https and
> modern browsers rightly complaining about mixed content. This was worked
> around by downl
On Thu, Aug 10, 2017 at 12:39:49PM +0300, Jani Nikula wrote:
> Based on Sphinx-quickstart, with existing and generated Makefiles merged
> together.
>
> This makes the rst2html and Sphinx builds work side by side. Plain
> 'make' continues to use rst2html, and 'make html' and friends use
> Sphinx. T
Quoting Matthew Auld (2017-08-10 20:02:22)
> On 10 August 2017 at 18:57, Chris Wilson wrote:
> > As we cannot have an exclusive stream set if the perf has not been
> > initialized, we only need to check for that exclusive stream.
> >
> > Signed-off-by: Chris Wilson
> > Cc: Lionel Landwerlin
> >
On Thu, Aug 10, 2017 at 9:23 AM, Chris Wilson wrote:
> Quoting Rodrigo Vivi (2017-08-10 15:50:43)
>> I'm not sure if this is really the case and I don't believe
>> this is the real fix for the bug mentioned here, but since
>> I don't see a reliable path when mst_port is set and when
>> mode_valid
On Thu, Aug 10, 2017 at 4:36 AM, Joonas Lahtinen
wrote:
> On ke, 2017-08-09 at 13:07 -0700, Rodrigo Vivi wrote:
>> A missing part to EU slice power gating is the
>> debugfs interface. This patch actually should have been
>> squashed to the initial EU slice power gating one.
>>
>> v2: Initial patch
On 08/08/17 05:30, Michal Wajdeczko wrote:
During debug we may want to investigate all communication
from the Guc. Add proper tracing macros in debug config.
v2: convert remaining DRM_DEBUG into new CT_DEBUG (Michal)
v3: use dedicated Kconfig (Daniele)
Signed-off-by: Michal Wajdeczko
Cc: Dani
On Wed, Aug 9, 2017 at 9:44 PM, Pandiyan, Dhinakaran
wrote:
> On Wed, 2017-08-02 at 12:26 -0700, Rodrigo Vivi wrote:
>> The idea is to have an unique place to decide the pin-port
>> per platform.
>>
>> So let's create this function now without any functional
>> change.
>
> This seems to change the
On 8/10/2017 3:26 AM, Petri Latvala wrote:
The current documentation for tests is limited to a single string per
test binary. This patch adds support for documenting individual
subtests.
The syntax for subtest documentation is:
igt_document_subtest("Frob knobs to see if one of the "
"Frequencies over 5.4 GHz only supported on certain
DDI ports and SKUs, and requires Vccio >= 0.95V."
More specifically, for current CNL SKUs available
(CNL-U and CNL-Y) we have:
DDI A - 5.4G eDP
DDI B - 8.1G DP
DDI C - 8.1G DP
DDI D - 5.4G DP
v2: Rebase on top of source_rates changes.
v3: Addre
Different from SKL we don't need ctrl1 and cfgcr2, but
we need to dump cfgcr0 and cfgcr1 instead.
v2: rebase and commit message
Cc: Clint Taylor
Cc: Mika Kahola
Signed-off-by: Rodrigo Vivi
Reviewed-by: Mika Kahola
---
drivers/gpu/drm/i915/intel_dpll_mgr.c | 11 ++-
1 file changed, 10
On 8/10/2017 3:26 AM, Petri Latvala wrote:
testdisplay has no subtests so hook them to existing subtest handling.
Signed-off-by: Petri Latvala
---
tests/testdisplay.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/tests/testdisplay.c b/tests/testdisplay.c
index f2a41fa..85b118f 1006
On 8/10/2017 3:26 AM, Petri Latvala wrote:
Signed-off-by: Petri Latvala
---
tests/igt_command_line.sh | 12 +++-
1 file changed, 11 insertions(+), 1 deletion(-)
diff --git a/tests/igt_command_line.sh b/tests/igt_command_line.sh
index 7f80fc8..b3a794c 100755
--- a/tests/igt_command_
On Thu, 2017-08-10 at 14:23 -0700, Rodrigo Vivi wrote:
> On Wed, Aug 9, 2017 at 9:44 PM, Pandiyan, Dhinakaran
> wrote:
> > On Wed, 2017-08-02 at 12:26 -0700, Rodrigo Vivi wrote:
> >> The idea is to have an unique place to decide the pin-port
> >> per platform.
> >>
> >> So let's create this functi
Signed-off-by: Jason Ekstrand
---
tests/syncobj_basic.c | 77 +--
1 file changed, 19 insertions(+), 58 deletions(-)
diff --git a/tests/syncobj_basic.c b/tests/syncobj_basic.c
index a7a6742..b08dd2d 100644
--- a/tests/syncobj_basic.c
+++ b/tests/syn
Signed-off-by: Jason Ekstrand
---
lib/Makefile.sources | 2 +
lib/igt_syncobj.c| 186 +++
lib/igt_syncobj.h| 66 ++
3 files changed, 254 insertions(+)
create mode 100644 lib/igt_syncobj.c
create mode 100644 lib/igt_synco
Signed-off-by: Jason Ekstrand
---
lib/igt_vgem.c | 25 -
lib/igt_vgem.h | 1 +
2 files changed, 25 insertions(+), 1 deletion(-)
diff --git a/lib/igt_vgem.c b/lib/igt_vgem.c
index 21cccb3..069c6dd 100644
--- a/lib/igt_vgem.c
+++ b/lib/igt_vgem.c
@@ -101,8 +101,9 @@ void *
This adds both trivial error-checking tests as well as more complex
tests which actually test whether or not waits do what they're supposed
to do. They only currently work on i915 but it should be simple to hook
them up for other drivers by simply implementing the little function
pointer hook prov
On 10 August 2017 at 20:29, Joe Kniss wrote:
> On Wed, Aug 9, 2017 at 4:13 PM, Joe Kniss wrote:
>> On Wed, Aug 9, 2017 at 12:14 PM, Noralf Trønnes wrote:
>>>
>>> Den 09.08.2017 01.42, skrev Joe Kniss:
Because all drivers currently use gem objects for framebuffer planes,
the virtua
Verified this as per the Bspec and it looks good.
Manasi
On Thu, Aug 10, 2017 at 03:40:08PM -0700, Rodrigo Vivi wrote:
> "Frequencies over 5.4 GHz only supported on certain
> DDI ports and SKUs, and requires Vccio >= 0.95V."
>
> More specifically, for current CNL SKUs available
> (CNL-U and CNL-
DPCD 600h - SET_POWER & SET_DP_PWR_VOLTAGE defines power state
101 = Set Main-Link for local Sink device and all downstream Sink
devices to D3 (power-down mode), keep AUX block fully powered, ready to
reply within a Response Timeout period of 300us.
This state is useful in a MST dock + MST monito
Dell monitor with a built-in MST branch does not light up on boot when
connected to a Thinkpad dock. The monitor also does not wake up after
Suspend-to-Idle or dpms off in this configuration. Keeping the AUX
powered up in the D3 power state solved this problem.
This partially fixes
Bugzilla: https
On Thu, 10 Aug 2017, Rodrigo Vivi wrote:
> On Thu, Aug 10, 2017 at 12:39:49PM +0300, Jani Nikula wrote:
>> Based on Sphinx-quickstart, with existing and generated Makefiles merged
>> together.
>>
>> This makes the rst2html and Sphinx builds work side by side. Plain
>> 'make' continues to use rst2
On Fri, 11 Aug 2017, Dhinakaran Pandiyan wrote:
> DPCD 600h - SET_POWER & SET_DP_PWR_VOLTAGE defines power state
>
> 101 = Set Main-Link for local Sink device and all downstream Sink
> devices to D3 (power-down mode), keep AUX block fully powered, ready to
> reply within a Response Timeout period
Hi,
On Thursday 10 August 2017 02:22 AM, Rodrigo Vivi wrote:
From: Paulo Zanoni
They're slightly different than the gen 9 calculations.
v2: Remove TODO comment. Code matches recent spec.
Cc: Mahesh Kumar
Cc: Maarten Lankhorst
Signed-off-by: Paulo Zanoni
Signed-off-by: Rodrigo Vivi
---
101 - 140 of 140 matches
Mail list logo
