On Thu, 2017-09-14 at 11:09 -0700, Vinay Belgaumkar wrote:
> Added the missing IGT_TEST_DESCRIPTION and some subtest
> descriptions.
>
> v2: Removed duplication, addressed comments, cc'd test author
>
> v3: Only comment abstract code, change some igt_info to igt_debug.
> Changed description t
On Fri, 2017-09-15 at 07:34 +, Szwichtenberg, Radoslaw wrote:
> On Thu, 2017-09-14 at 11:09 -0700, Vinay Belgaumkar wrote:
> > Added the missing IGT_TEST_DESCRIPTION and some subtest
> > descriptions.
> >
> > v2: Removed duplication, addressed comments, cc'd test author
> >
> > v3: Only comme
On 15/09/2017 01:00, Rogozhkin, Dmitry V wrote:
On Wed, 2017-09-13 at 11:34 +0100, Tvrtko Ursulin wrote:
+static int i915_pmu_event_init(struct perf_event *event)
+{
+ struct drm_i915_private *i915 =
+ container_of(event->pmu, typeof(*i915), pmu.base);
+ int cpu, ret;
Quoting Tvrtko Ursulin (2017-09-15 07:56:00)
>
> On 14/09/2017 21:36, Ville Syrjälä wrote:
> > On Mon, Sep 11, 2017 at 04:25:50PM +0100, Tvrtko Ursulin wrote:
> >> From: Tvrtko Ursulin
> >>
> >> Extract code from i915_energy_uJ (debugfs) so it can be used by
> >> other callers in future patches.
Thanks Chris. LGTM.
Minor inputs below
On 9/14/2017 3:28 PM, Chris Wilson wrote:
An interesting discussion regarding "hybrid interrupt polling" for NVMe
came to the conclusion that the ideal busyspin before sleeping was half
of the expected request latency (and better if it was already halfway
On 14/09/2017 20:49, Chris Wilson wrote:
Quoting Tvrtko Ursulin (2017-09-11 16:25:50)
diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index 60461f49936b..ff67df8d99fa 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -9369,3 +93
Quoting Michał Winiarski (2017-09-11 12:33:41)
> On Mon, Sep 11, 2017 at 09:41:25AM +0100, Chris Wilson wrote:
> > To silence the critcs:
> >
> > [56532.161115] workqueue: PF_MEMALLOC task 36(khugepaged) is flushing
> > !WQ_MEM_RECLAIM i915-userptr-release: (null)
> > [56532.161138]
On 14/09/2017 20:57, Chris Wilson wrote:
Quoting Tvrtko Ursulin (2017-09-13 11:34:17)
From: Tvrtko Ursulin
If only a subset of events is enabled we can afford to suspend
the sampling timer when the GPU is idle and so save some cycles
and power.
v2: Rebase and limit timer even more.
v3: Rebas
Quoting Kamble, Sagar A (2017-09-15 10:15:31)
> Thanks Chris. LGTM.
>
> Minor inputs below
>
>
> On 9/14/2017 3:28 PM, Chris Wilson wrote:
> > An interesting discussion regarding "hybrid interrupt polling" for NVMe
> > came to the conclusion that the ideal busyspin before sleeping was half
> > o
On 14/09/2017 21:16, Chris Wilson wrote:
Quoting Tvrtko Ursulin (2017-09-11 16:25:55)
From: Tvrtko Ursulin
Track total time requests have been executing on the hardware.
We add new kernel API to allow software tracking of time GPU
engines are spending executing requests.
Both per-engine and
On 14/09/2017 21:17, Chris Wilson wrote:
Quoting Tvrtko Ursulin (2017-09-11 16:25:56)
From: Tvrtko Ursulin
Export the stats added in the previous patch in debugfs.
Number of active clients reading this data is tracked and the
static key is only enabled whilst there are some.
Userspace is in
On 14/09/2017 21:26, Chris Wilson wrote:
Quoting Tvrtko Ursulin (2017-09-11 16:25:58)
From: Tvrtko Ursulin
Other kernel users might want to look at total GPU busyness
in order to implement things like package power distribution
algorithms more efficiently.
Who are we exporting these symbols
On 14/09/2017 21:22, Chris Wilson wrote:
Quoting Tvrtko Ursulin (2017-09-13 13:18:19)
From: Tvrtko Ursulin
This reduces the cost of the software engine busyness tracking
to a single no-op instruction when there are no listeners.
v2: Rebase and some comments.
v3: Rebase.
Signed-off-by: Tvrtk
Quoting Ville Syrjälä (2017-09-14 18:42:14)
> On Thu, Sep 14, 2017 at 05:42:13PM +0100, Chris Wilson wrote:
> > Commit 1bf6ad622b9b ("drm/vblank: drop the mode argument from
> > drm_calc_vbltimestamp_from_scanoutpos") removed the use of in_vbl, but
> > did not remove the local variable. Do so now.
Quoting Michal Wajdeczko (2017-09-14 16:08:02)
> As we now use same name for public module param and its local
> representation we can simplify param definition macro.
>
> Changes done with Coccinelle:
>
> @@
> declarer name module_param_named;
> declarer name module_param_named_unsafe;
> declare
Quoting Michal Wajdeczko (2017-09-14 16:08:03)
> We're always specifying description of each module param in
> separate macro. Let's combine description into our main macro.
> Started with Coccinelle, followed by minor cleanup.
>
> @match1@
> declarer name MODULE_PARM_DESC;
> identifier n;
> const
On 14/09/2017 10:58, Chris Wilson wrote:
An interesting discussion regarding "hybrid interrupt polling" for NVMe
came to the conclusion that the ideal busyspin before sleeping was half
of the expected request latency (and better if it was already halfway
through that request). This suggested tha
On Thu, Sep 14, 2017 at 09:37:37PM +0100, Chris Wilson wrote:
> Quoting Ville Syrjala (2017-09-14 16:17:31)
> > From: Ville Syrjälä
> >
> > i830 seems to occasionally forget the PIPESTAT enable bits when
> > we read the register. These aren't the only registers on i830 that
> > have problems with
On 15/09/2017 09:51, Chris Wilson wrote:
Quoting Tvrtko Ursulin (2017-09-15 07:56:00)
On 14/09/2017 21:36, Ville Syrjälä wrote:
On Mon, Sep 11, 2017 at 04:25:50PM +0100, Tvrtko Ursulin wrote:
From: Tvrtko Ursulin
Extract code from i915_energy_uJ (debugfs) so it can be used by
other callers
On Tue, Sep 12, 2017 at 04:57:29PM -0700, Dhinakaran Pandiyan wrote:
> The other instances of link training are protected with
> connection_mutex, so do the same in check_mst_status() too.
>
> Signed-off-by: Dhinakaran Pandiyan
> ---
> drivers/gpu/drm/i915/intel_dp.c | 4
> 1 file changed,
On Thu, Sep 14, 2017 at 12:59:35PM -0700, James Ausmus wrote:
> Make intel_dp_add_mst_connector handle error returns from the drm_ calls.
> Add intel_connector_destroy to support cleanup on the error path.
That name makes one think that it could be plugged into the connector
.destroy() hook. So I'
Quoting Tvrtko Ursulin (2017-09-15 11:01:02)
>
> On 14/09/2017 10:58, Chris Wilson wrote:
> > An interesting discussion regarding "hybrid interrupt polling" for NVMe
> > came to the conclusion that the ideal busyspin before sleeping was half
> > of the expected request latency (and better if it wa
> -Original Message-
> From: Intel-gfx [mailto:intel-gfx-boun...@lists.freedesktop.org] On Behalf Of
> Vidya Srinivas
> Sent: Thursday, September 14, 2017 5:42 PM
> To: intel-gfx@lists.freedesktop.org
> Cc: Srinivas, Vidya
> Subject: [Intel-gfx] [PATCH 1/2] drm/i915: Enable scanline read f
On Fri, Sep 15, 2017 at 11:07:21AM +0100, Tvrtko Ursulin wrote:
>
> On 15/09/2017 09:51, Chris Wilson wrote:
> > Quoting Tvrtko Ursulin (2017-09-15 07:56:00)
> >>
> >> On 14/09/2017 21:36, Ville Syrjälä wrote:
> >>> On Mon, Sep 11, 2017 at 04:25:50PM +0100, Tvrtko Ursulin wrote:
> From: Tvrtk
Quoting Ville Syrjälä (2017-09-15 11:34:03)
> On Fri, Sep 15, 2017 at 11:07:21AM +0100, Tvrtko Ursulin wrote:
> >
> > On 15/09/2017 09:51, Chris Wilson wrote:
> > > Quoting Tvrtko Ursulin (2017-09-15 07:56:00)
> > >>
> > >> On 14/09/2017 21:36, Ville Syrjälä wrote:
> > >>> On Mon, Sep 11, 2017 at
On 15/09/2017 11:38, Chris Wilson wrote:
Quoting Ville Syrjälä (2017-09-15 11:34:03)
On Fri, Sep 15, 2017 at 11:07:21AM +0100, Tvrtko Ursulin wrote:
On 15/09/2017 09:51, Chris Wilson wrote:
Quoting Tvrtko Ursulin (2017-09-15 07:56:00)
On 14/09/2017 21:36, Ville Syrjälä wrote:
On Mon, Sep
On Fri, 15 Sep 2017, Chris Wilson wrote:
> Quoting Michal Wajdeczko (2017-09-14 16:08:03)
>> We're always specifying description of each module param in
>> separate macro. Let's combine description into our main macro.
>> Started with Coccinelle, followed by minor cleanup.
>>
>> @match1@
>> decla
On Thu, Sep 14, 2017 at 03:55:54PM +0100, Chris Wilson wrote:
> Quoting Chris Wilson (2017-09-14 15:54:35)
> > Quoting Ville Syrjälä (2017-09-14 15:50:14)
> > Leaving
> > > > drm/i915: Remove duplicated irq_preinstall/uninstall hooks
> > which I can't remember enough and will need to read again.
On Thu, Sep 14, 2017 at 05:20:47PM -0700, Jordan Justen wrote:
> On 2017-09-14 12:13:08, Ville Syrjälä wrote:
> > I don't actually know what the plan is w.r.t. installing the lib, so
> > I'll leave this one for someone else to ack/nack.
The plan is to install the lib.
The concerns with installing
On Tue, Sep 12, 2017 at 11:19:26AM +0300, Jani Nikula wrote:
> Handle debugfs override edid and firmware edid at the low level to
> transparently and completely replace the real edid. Previously, we
> practically only used the modes from the override EDID, and none of the
> other data, such as audi
On Tue, Sep 12, 2017 at 11:19:27AM +0300, Jani Nikula wrote:
> If drm_kms_helper.edid_firmware module parameter is set at
> drm_kms_helper probe time, update the new drm.edid_firmware parameter
> for backwards compatibility.
>
> The drm_kms_helper.edid_firmware is now read-only in sysfs to prevent
If we see the seqno stop progressing, we abandon the test for fear that
the GPU died following the reset. However, during test teardown we still
wait for the GPU to idle before continuing, but we have already
confirmed that the GPU is dead. Furthermore, since we are inside a reset
test, we have dis
On Fri, Sep 15, 2017 at 02:56:04PM +0300, Petri Latvala wrote:
> On Thu, Sep 14, 2017 at 05:20:47PM -0700, Jordan Justen wrote:
> > On 2017-09-14 12:13:08, Ville Syrjälä wrote:
> > > I don't actually know what the plan is w.r.t. installing the lib, so
> > > I'll leave this one for someone else to a
Quoting Chris Wilson (2017-09-15 14:09:29)
> If we see the seqno stop progressing, we abandon the test for fear that
> the GPU died following the reset. However, during test teardown we still
> wait for the GPU to idle before continuing, but we have already
> confirmed that the GPU is dead. Further
== Series Details ==
Series: drm/i915/selftests: Try to recover from a wedged GPU during reset tests
URL : https://patchwork.freedesktop.org/series/30419/
State : success
== Summary ==
Series 30419v1 drm/i915/selftests: Try to recover from a wedged GPU during
reset tests
https://patchwork.fre
From: Ville Syrjälä
Install the assembler and disassemebler binaries, and the accompanying
pkg-config file. Change libbrw into a static library since we don't want
to install that.
Signed-off-by: Ville Syrjälä
---
assembler/meson.build | 21 +
1 file changed, 17 insertions(
From: Ville Syrjälä
Tell meson about the dependency on version.h. Avoids the compiler
falling over on account of IGT_GIT_SHA1 not being there.
Signed-off-by: Ville Syrjälä
---
lib/meson.build | 8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/lib/meson.build b/lib/meso
== Series Details ==
Series: meson: Install the (dis)assembler
URL : https://patchwork.freedesktop.org/series/30420/
State : success
== Summary ==
IGT patchset tested on top of latest successful build
93f72b23156589964b87bb9ca7c94c104976303c meson: Process intel_aubdump.in into
intel_aubdump
== Series Details ==
Series: meson: Install the (dis)assembler (rev2)
URL : https://patchwork.freedesktop.org/series/30420/
State : success
== Summary ==
IGT patchset tested on top of latest successful build
93f72b23156589964b87bb9ca7c94c104976303c meson: Process intel_aubdump.in into
intel_a
On Wed, Aug 16, 2017 at 05:46:07PM +0300, Imre Deak wrote:
> Atm, on GEN9 big core platforms before saving the hibernation image we
> uninitialize the display, disabling power wells manually, while before
> restoring the image we keep things powered (letting HW/DMC power down
> things as needed). T
== Series Details ==
Series: drm/i915/selftests: Try to recover from a wedged GPU during reset tests
URL : https://patchwork.freedesktop.org/series/30419/
State : success
== Summary ==
Test perf:
Subgroup polling:
pass -> FAIL (shard-hsw) fdo#102252 +1
fdo#
Hi,
so these two patches add i915 module parameters to globally override
how the driver handles dithering and gamma/csc conversion.
They serve two purposes: First as debug aid and "airbag" for working
around potential precision problems in getting pixels from rendering
to the display outputs. Thi
i915.enable_dithering allows to force dithering on all outputs
on (=1) or off (=0). The default is -1 for current automatic
per-pipe selection.
This is useful for debugging and for special case scenarios,
e.g., providing simulated 10 bpc output on 8 bpc digital sinks
if a 10 bpc framebuffer + rend
The new module parameter enable_hw_color_correction defaults to
true, to retain the current behaviour. If set to false, it will
disable all hardware color correction, like gamma/degamma and
csc.
This is useful for debugging gamma table / csc precision problems,
and to ensure unmodified pixel passt
== Series Details ==
Series: meson: Install the (dis)assembler
URL : https://patchwork.freedesktop.org/series/30420/
State : warning
== Summary ==
Test kms_frontbuffer_tracking:
Subgroup fbc-1p-indfb-fliptrack:
pass -> DMESG-WARN (shard-hsw)
Test kms_atomic_transi
== Series Details ==
Series: series starting with [1/2] drm/i915: Add module parameter to force
en-/disable dithering.
URL : https://patchwork.freedesktop.org/series/30428/
State : success
== Summary ==
Series 30428v1 series starting with [1/2] drm/i915: Add module parameter to
force en-/dis
Break up request coalescing by submitting a queue of requests from
different contexts. This makes sure that we mark all ready but not yet
submitted-to-hw requests as completed/EIO upon wedging.
Signed-off-by: Chris Wilson
Cc: Michał Winiarski
---
tests/gem_eio.c | 72 +++
Den 15.09.2017 00.29, skrev Laurent Pinchart:
Hi Noralf,
On Wednesday, 13 September 2017 18:19:22 EEST Noralf Trønnes wrote:
Den 13.09.2017 07.09, skrev Laurent Pinchart:
On Monday, 11 September 2017 17:31:54 EEST Noralf Trønnes wrote:
Hi,
I want to start out by saying that this patchset is
On Fri, Sep 15, 2017 at 05:10:25PM +0100, Chris Wilson wrote:
> Break up request coalescing by submitting a queue of requests from
> different contexts. This makes sure that we mark all ready but not yet
> submitted-to-hw requests as completed/EIO upon wedging.
>
> Signed-off-by: Chris Wilson
> C
When wedging the hw, we want to mark all in-flight requests as -EIO.
This is made slightly more complex by execlists who store the ready but
not yet submitted-to-hw requests on a private queue (an rbtree
priolist). Call into execlists to cancel not only the ELSP tracking for
the submitted requests,
Quoting Michał Winiarski (2017-09-15 17:47:27)
> On Fri, Sep 15, 2017 at 05:10:25PM +0100, Chris Wilson wrote:
> > Break up request coalescing by submitting a queue of requests from
> > different contexts. This makes sure that we mark all ready but not yet
> > submitted-to-hw requests as completed/
== Series Details ==
Series: meson: Install the (dis)assembler (rev2)
URL : https://patchwork.freedesktop.org/series/30420/
State : success
== Summary ==
Test kms_setmode:
Subgroup basic:
fail -> PASS (shard-hsw) fdo#99912
Test perf:
Subgroup polling
Quoting Chris Wilson (2017-09-15 17:49:16)
> When wedging the hw, we want to mark all in-flight requests as -EIO.
> This is made slightly more complex by execlists who store the ready but
> not yet submitted-to-hw requests on a private queue (an rbtree
> priolist). Call into execlists to cancel not
Quoting Chris Wilson (2017-09-15 17:49:16)
> diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.h
> b/drivers/gpu/drm/i915/intel_ringbuffer.h
> index abf171c3cb9c..04fc50c993bf 100644
> --- a/drivers/gpu/drm/i915/intel_ringbuffer.h
> +++ b/drivers/gpu/drm/i915/intel_ringbuffer.h
> @@ -306,6 +306,1
On Thu, Sep 14, 2017 at 02:50:09PM +0100, Chris Wilson wrote:
> If we wedged one engine with unready requests to a second engine
> (blocked by waiting on requests from the first, using a dma-fence),
> check that we propagate the -EIO to those in-flight requests.
>
> v2: check all ABI engines
>
>
== Series Details ==
Series: drm/i915: Cancel all ready but queued requests when wedging
URL : https://patchwork.freedesktop.org/series/30434/
State : success
== Summary ==
Series 30434v1 drm/i915: Cancel all ready but queued requests when wedging
https://patchwork.freedesktop.org/api/1.0/seri
On Fri, Sep 15, 2017 at 3:18 AM, Ville Syrjälä
wrote:
> On Thu, Sep 14, 2017 at 12:59:35PM -0700, James Ausmus wrote:
>> Make intel_dp_add_mst_connector handle error returns from the drm_ calls.
>> Add intel_connector_destroy to support cleanup on the error path.
>
> That name makes one think that
On Fri, Sep 15, 2017 at 05:59:30PM +0100, Chris Wilson wrote:
> Quoting Chris Wilson (2017-09-15 17:49:16)
> > diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.h
> > b/drivers/gpu/drm/i915/intel_ringbuffer.h
> > index abf171c3cb9c..04fc50c993bf 100644
> > --- a/drivers/gpu/drm/i915/intel_ringbuf
Quoting Michał Winiarski (2017-09-15 18:16:16)
> On Fri, Sep 15, 2017 at 05:59:30PM +0100, Chris Wilson wrote:
> > Quoting Chris Wilson (2017-09-15 17:49:16)
> > > diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.h
> > > b/drivers/gpu/drm/i915/intel_ringbuffer.h
> > > index abf171c3cb9c..04fc50c
On Thu, Sep 14, 2017 at 02:50:56PM +0100, Chris Wilson wrote:
> Some overlap with gem_exec_fence, but confirm that light for waiting on
> a fence (both native and external), that a request queued is also
> flagged as EIO upon wedging.
There's something wrong with the construction of this sentence.
When wedging the hw, we want to mark all in-flight requests as -EIO.
This is made slightly more complex by execlists who store the ready but
not yet submitted-to-hw requests on a private queue (an rbtree
priolist). Call into execlists to cancel not only the ELSP tracking for
the submitted requests,
== Series Details ==
Series: series starting with [1/2] drm/i915: Add module parameter to force
en-/disable dithering.
URL : https://patchwork.freedesktop.org/series/30428/
State : success
== Summary ==
Test perf:
Subgroup polling:
pass -> FAIL (shard-hsw)
Make intel_dp_add_mst_connector handle error returns from the drm_ calls.
Add intel_connector_free to support cleanup on the error path.
v2: Rename new function to avoid confusion, and simplify error
paths (Ville)
Signed-off-by: James Ausmus
---
Note that this patch is compile/boot tested only
== Series Details ==
Series: drm/i915: Cancel all ready but queued requests when wedging (rev2)
URL : https://patchwork.freedesktop.org/series/30434/
State : success
== Summary ==
Series 30434v2 drm/i915: Cancel all ready but queued requests when wedging
https://patchwork.freedesktop.org/api/1
From: Ville Syrjälä
A register dumper that can't run with the driver loaded is useless most
of the time.
Signed-off-by: Ville Syrjälä
---
tools/intel_watermark.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools/intel_watermark.c b/tools/intel_watermark.c
index d8c78480
From: Ville Syrjälä
Let's print the linetime watermarks un usecs. Might make it easier to
spot bogus values.
Signed-off-by: Ville Syrjälä
---
tools/intel_watermark.c | 6 --
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/tools/intel_watermark.c b/tools/intel_watermark.c
inde
From: Dhinakaran Pandiyan
Added support to print SKL watermark and DDB registers.
v2: Printed raw register data, renamed planes and combined two printf()'s
(Ville)
Signed-off-by: Dhinakaran Pandiyan
[vsyrjala: make it build, fix indentation, etc.]
Signed-off-by: Ville Syrjälä
---
tools/intel
From: Ville Syrjälä
Signed-off-by: Ville Syrjälä
---
tools/intel_watermark.c | 15 +++
1 file changed, 15 insertions(+)
diff --git a/tools/intel_watermark.c b/tools/intel_watermark.c
index 24ecaa780be5..ce920192295b 100644
--- a/tools/intel_watermark.c
+++ b/tools/intel_watermark.c
From: Ville Syrjälä
Signed-off-by: Ville Syrjälä
---
tools/intel_watermark.c | 12 ++--
1 file changed, 6 insertions(+), 6 deletions(-)
diff --git a/tools/intel_watermark.c b/tools/intel_watermark.c
index c08379d66e51..4f18e8df1810 100644
--- a/tools/intel_watermark.c
+++ b/tools/intel
From: Ville Syrjälä
Print hex numbers with "0x" prefix, and make the output a bit more
compact.
Signed-off-by: Ville Syrjälä
---
tools/intel_watermark.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/tools/intel_watermark.c b/tools/intel_watermark.c
index 4f18e8df181
From: Dhinakaran Pandiyan
Having registers for nonexistent planes in the dumpo might end up being
rather confusing. Try to only include real planes.
Signed-off-by: Ville Syrjälä
---
tools/intel_watermark.c | 40 +---
1 file changed, 33 insertions(+), 7 delet
On Thu, Sep 14, 2017 at 09:12:50AM +, Patchwork wrote:
> == Series Details ==
>
> Series: drm/i915: Miscellaneous fixes to reduce dependency for I915_MAX_PIPES
> URL : https://patchwork.freedesktop.org/series/30336/
> State : warning
>
> == Summary ==
>
> Test kms_cursor_legacy:
>
== Series Details ==
Series: drm/i915: Handle drm-layer errors in intel_dp_add_mst_connector (rev2)
URL : https://patchwork.freedesktop.org/series/30384/
State : success
== Summary ==
Series 30384v2 drm/i915: Handle drm-layer errors in intel_dp_add_mst_connector
https://patchwork.freedesktop.o
On Fri, Sep 15, 2017 at 10:49:09AM -0700, James Ausmus wrote:
> Make intel_dp_add_mst_connector handle error returns from the drm_ calls.
> Add intel_connector_free to support cleanup on the error path.
>
> v2: Rename new function to avoid confusion, and simplify error
> paths (Ville)
>
> Signed-
On Fri, Sep 15, 2017 at 06:31:00PM +0100, Chris Wilson wrote:
> When wedging the hw, we want to mark all in-flight requests as -EIO.
> This is made slightly more complex by execlists who store the ready but
> not yet submitted-to-hw requests on a private queue (an rbtree
> priolist). Call into exec
The patch looks good for eDP link training optimizations.
Reviewed-by: Manasi Navare
Manasi
On Tue, Aug 22, 2017 at 09:34:46AM -0700, Jim Bride wrote:
> This set of changes has some history to them. There were several attempts
> to add what was called "fast link training" to i915, which actual
On Thu, 2017-09-14 at 15:26 -0700, Ausmus, James wrote:
> On Tue, Sep 12, 2017 at 4:57 PM, Dhinakaran Pandiyan
> wrote:
> > The other instances of link training are protected with
> > connection_mutex, so do the same in check_mst_status() too.
>
> We don't seem to be taking connection_mutex arou
== Series Details ==
Series: series starting with [1/7] tools/intel_watermark: Don't require master
URL : https://patchwork.freedesktop.org/series/30442/
State : success
== Summary ==
IGT patchset tested on top of latest successful build
93f72b23156589964b87bb9ca7c94c104976303c meson: Process
On Thu, 2017-09-14 at 15:36 -0700, Ausmus, James wrote:
> On Tue, Sep 12, 2017 at 4:57 PM, Dhinakaran Pandiyan
> wrote:
> > intel_dp_can_mst() performs these checks.
> >
> > Signed-off-by: Dhinakaran Pandiyan
> > ---
> > drivers/gpu/drm/i915/intel_dp.c | 6 --
> > 1 file changed, 6 deletions
On Thu, Sep 14, 2017 at 03:26:37PM -0700, Ausmus, James wrote:
> On Tue, Sep 12, 2017 at 4:57 PM, Dhinakaran Pandiyan
> wrote:
> > The other instances of link training are protected with
> > connection_mutex, so do the same in check_mst_status() too.
>
> We don't seem to be taking connection_mute
On Thu, Aug 24, 2017 at 10:15:42AM -0700, Manasi Navare wrote:
> On Thu, Aug 24, 2017 at 03:33:27PM +0300, Ville Syrjälä wrote:
> > On Thu, Aug 17, 2017 at 08:03:04PM -0700, Manasi Navare wrote:
> > > In the channel EQ retry loop, we break from the loop in case
> > > of failure to get link status o
== Series Details ==
Series: drm/i915: Cancel all ready but queued requests when wedging (rev2)
URL : https://patchwork.freedesktop.org/series/30434/
State : warning
== Summary ==
Test perf:
Subgroup polling:
pass -> FAIL (shard-hsw) fdo#102252
Test kms_plan
On Fri, Sep 15, 2017 at 11:38 AM, Manasi Navare
wrote:
> On Thu, Sep 14, 2017 at 03:26:37PM -0700, Ausmus, James wrote:
>> On Tue, Sep 12, 2017 at 4:57 PM, Dhinakaran Pandiyan
>> wrote:
>> > The other instances of link training are protected with
>> > connection_mutex, so do the same in check_mst
== Series Details ==
Series: drm/i915: Handle drm-layer errors in intel_dp_add_mst_connector (rev2)
URL : https://patchwork.freedesktop.org/series/30384/
State : success
== Summary ==
Test perf:
Subgroup polling:
pass -> FAIL (shard-hsw) fdo#102252
Test kms_
Make intel_dp_add_mst_connector handle error returns from the drm_ calls.
Add intel_connector_free to support cleanup on the error path.
v2: Rename new function to avoid confusion, and simplify error
paths (Ville)
v3: Indentation fixup, style fixes (Ville)
Signed-off-by: James Ausmus
---
Note
On Fri, Sep 15, 2017 at 11:09 AM, Ville Syrjälä
wrote:
> On Fri, Sep 15, 2017 at 10:49:09AM -0700, James Ausmus wrote:
>> Make intel_dp_add_mst_connector handle error returns from the drm_ calls.
>> Add intel_connector_free to support cleanup on the error path.
>>
>> v2: Rename new function to avo
== Series Details ==
Series: drm/i915: Handle drm-layer errors in intel_dp_add_mst_connector (rev3)
URL : https://patchwork.freedesktop.org/series/30384/
State : success
== Summary ==
Series 30384v3 drm/i915: Handle drm-layer errors in intel_dp_add_mst_connector
https://patchwork.freedesktop.o
== Series Details ==
Series: series starting with [1/7] tools/intel_watermark: Don't require master
URL : https://patchwork.freedesktop.org/series/30442/
State : failure
== Summary ==
Test gem_flink_race:
Subgroup flink_close:
pass -> FAIL (shard-hsw) fdo#10
Reviewed-by: Jordan Justen
On 2017-09-15 06:35:52, Ville Syrjala wrote:
> From: Ville Syrjälä
>
> Install the assembler and disassemebler binaries, and the accompanying
> pkg-config file. Change libbrw into a static library since we don't want
> to install that.
>
> Signed-off-by: Ville Syrjäl
Reviewed-by: Jordan Justen
On 2017-09-15 06:59:54, Ville Syrjala wrote:
> From: Ville Syrjälä
>
> Tell meson about the dependency on version.h. Avoids the compiler
> falling over on account of IGT_GIT_SHA1 not being there.
>
> Signed-off-by: Ville Syrjälä
> ---
> lib/meson.build | 8
Prepared intel_auth_huc to separate HuC specific functionality
from GuC send action. Created new header intel_huc.h to group
HuC specific declarations.
Cc: Michal Wajdeczko
Cc: Michał Winiarski
Signed-off-by: Sagar Arun Kamble
---
drivers/gpu/drm/i915/intel_guc.c | 19 +++
driv
Cc: Michal Wajdeczko
Cc: Michał Winiarski
Signed-off-by: Sagar Arun Kamble
---
drivers/gpu/drm/i915/i915_drv.h | 1 +
drivers/gpu/drm/i915/intel_guc.h | 126 +++
drivers/gpu/drm/i915/intel_uc.h | 125 --
3 files changed
Create intel_guc.c and move guc communication init functionality from
intel_uc.c. Prepared new initialization function intel_guc_init_early.
Moved below functions to intel_guc.c.
1. intel_guc_send_nop
2. gen8_guc_raise_irq
And below functions to intel_guc.h.
1. intel_guc_send
2. intel_guc_notify
C
s/guc_init_send_regs/intel_guc_init_send_regs.
Calling intel_guc_init_send_regs from intel_uc_init_hw.
Moved below functions to intel_guc.c from intel_uc.c
1. guc_send_regs
2. intel_guc_init_send_regs
3. intel_guc_send_mmio
Cc: Michal Wajdeczko
Cc: Michał Winiarski
Signed-off-by: Sagar Arun Kamb
Also put the declaration under the i915_guc_submission section in
intel_uc.h.
Cc: Michal Wajdeczko
Cc: Michał Winiarski
Signed-off-by: Sagar Arun Kamble
---
drivers/gpu/drm/i915/i915_drv.c| 6 +++---
drivers/gpu/drm/i915/i915_gem.c| 2 +-
drivers/gpu/drm/i915/i915_guc
This series is based on https://patchwork.freedesktop.org/series/30345/.
Sagar Arun Kamble (11):
drm/i915/guc: Pass intel_guc to intel_guc_suspend/resume instead of
drm_i915_private
drm/i915/guc: Create intel_guc.c for defining GuC specific
functionality
drm/i915/guc: Move guc_send_*
Cc: Michal Wajdeczko
Cc: Michał Winiarski
Signed-off-by: Sagar Arun Kamble
---
drivers/gpu/drm/i915/intel_guc.c | 16
drivers/gpu/drm/i915/intel_guc.h | 1 +
drivers/gpu/drm/i915/intel_uc.c | 16
drivers/gpu/drm/i915/intel_uc.h | 1 -
4 files changed, 17 in
guc_ggtt_invalidate/guc_interrupts should be disabled towards
end of reset/suspend post GuC suspension as these are setup back again
during recovery/resume.
Prepared helpers intel_guc_pause and intel_guc_unpause that will do
teardown/bringup of this setup along with suspension/resumption of GuC if
With GuC v9, new type of Default/critical logging in GuC to enable
capturing minimal important logs in production systems efficiently.
This patch enables this logging in GuC by default always. It should
be noted that streaming support with half-full interrupt mechanism
that is present for normal lo
Functionality needed to disable GuC interrupts and cleanup the
runtime/relay data structures is already covered in the unload path
via intel_guc_fini_hw and intel_guc_cleanup hence remove
i915_guc_log_unregister
v2: Removed the function i915_guc_log_unregister.
Cc: Michal Wajdeczko
Signed-off-by
1 - 100 of 107 matches
Mail list logo
