Re: [Intel-gfx] [PATCH v5 4/4] drm/i915/perf: add support for Coffeelake GT2

On 04/09/17 10:25, Matthew Auld wrote: On 30 August 2017 at 17:12, Lionel Landwerlin wrote: Add the test configuration & timestamp frequency for Coffeelake GT2. Signed-off-by: Lionel Landwerlin Do we not want to also disable the

[Intel-gfx] ✗ Fi.CI.IGT: warning for drm/i915: Speed up DMC firmware loading (rev3)

== Series Details == Series: drm/i915: Speed up DMC firmware loading (rev3) URL : https://patchwork.freedesktop.org/series/29688/ State : warning == Summary == Test tools_test: Subgroup tools_test: pass -> DMESG-WARN (shard-hsw) shard-hswtotal:2265

[Intel-gfx] ✗ Fi.CI.IGT: warning for drm/i915: Speed up DMC firmware loading (rev2)

== Series Details == Series: drm/i915: Speed up DMC firmware loading (rev2) URL : https://patchwork.freedesktop.org/series/29688/ State : warning == Summary == Test kms_cursor_crc: Subgroup cursor-256x256-offscreen: pass -> SKIP (shard-hsw) Test perf:

[Intel-gfx] ✓ Fi.CI.BAT: success for drm/i915: Speed up DMC firmware loading (rev3)

== Series Details == Series: drm/i915: Speed up DMC firmware loading (rev3) URL : https://patchwork.freedesktop.org/series/29688/ State : success == Summary == Series 29688v3 drm/i915: Speed up DMC firmware loading https://patchwork.freedesktop.org/api/1.0/series/29688/revisions/3/mbox/ Test

Re: [Intel-gfx] [PATCH v3] drm/i915: Speed up DMC firmware loading

On Mon, Sep 04, 2017 at 08:15:57PM +0100, Chris Wilson wrote: > Quoting Chris Wilson (2017-09-04 20:14:32) > > Quoting David Weinehall (2017-09-04 20:08:06) > > > Currently we're doing: > > > > > > 1. acquire lock > > > 2. write word to hardware > > > 3. release lock > > > 4. repeat from 1 > > >

Re: [Intel-gfx] [PATCH v3] drm/i915: Speed up DMC firmware loading

Quoting Chris Wilson (2017-09-04 20:14:32) > Quoting David Weinehall (2017-09-04 20:08:06) > > Currently we're doing: > > > > 1. acquire lock > > 2. write word to hardware > > 3. release lock > > 4. repeat from 1 > > > > to load the DMC firmware. Due to the cost of acquiring/releasing a lock, >

Re: [Intel-gfx] [PATCH v3] drm/i915: Speed up DMC firmware loading

Quoting David Weinehall (2017-09-04 20:08:06) > Currently we're doing: > > 1. acquire lock > 2. write word to hardware > 3. release lock > 4. repeat from 1 > > to load the DMC firmware. Due to the cost of acquiring/releasing a lock, > and the size of the DMC firmware, this slows down DMC loading

Re: [Intel-gfx] [PATCH v2] drm/i915: Speed up DMC firmware loading

On Mon, Sep 04, 2017 at 07:55:56PM +0100, Chris Wilson wrote: > Quoting David Weinehall (2017-09-04 19:38:04) > > v2: Per feedback from Chris & Ville there's no need to do the whole > > forcewake dance, so lose that bit (Chris, Ville) > > > @@ -251,9 +253,20 @@ void

[Intel-gfx] [PATCH v3] drm/i915: Speed up DMC firmware loading

Currently we're doing: 1. acquire lock 2. write word to hardware 3. release lock 4. repeat from 1 to load the DMC firmware. Due to the cost of acquiring/releasing a lock, and the size of the DMC firmware, this slows down DMC loading a lot. This patch simply acquires the lock, writes the entire

[Intel-gfx] ✓ Fi.CI.BAT: success for drm/i915: Speed up DMC firmware loading (rev2)

== Series Details == Series: drm/i915: Speed up DMC firmware loading (rev2) URL : https://patchwork.freedesktop.org/series/29688/ State : success == Summary == Series 29688v2 drm/i915: Speed up DMC firmware loading https://patchwork.freedesktop.org/api/1.0/series/29688/revisions/2/mbox/ Test

Re: [Intel-gfx] [PATCH v2] drm/i915: Speed up DMC firmware loading

Quoting David Weinehall (2017-09-04 19:38:04) > v2: Per feedback from Chris & Ville there's no need to do the whole > forcewake dance, so lose that bit (Chris, Ville) > @@ -251,9 +253,20 @@ void intel_csr_load_program(struct drm_i915_private > *dev_priv) > return; > }

Re: [Intel-gfx] [PATCH v4 1/2] drm/i915: Track minimum acceptable cdclk instead of "minimum dotclock"

Op 04-09-17 om 17:58 schreef Ville Syrjälä: > On Mon, Sep 04, 2017 at 12:39:25PM +0200, Maarten Lankhorst wrote: >> Op 31-08-17 om 20:48 schreef Ville Syrjälä: >>> On Wed, Aug 30, 2017 at 09:57:03PM +0300, ville.syrj...@linux.intel.com >>> wrote: From: Ville Syrjälä

[Intel-gfx] [PATCH v2] drm/i915: Speed up DMC firmware loading

Currently we're doing: 1. acquire lock 2. write word to hardware 3. release lock 4. repeat from 1 to load the DMC firmware. Due to the cost of acquiring/releasing a lock, and the size of the DMC firmware, this slows down DMC loading a lot. This patch simply acquires the lock, writes the entire

Re: [Intel-gfx] [PATCH i-g-t v2] tests/kms_frontbuffer_tracking: increase FBC wait timeout to 5s

Em Seg, 2017-09-04 às 11:45 +0100, Chris Wilson escreveu: > Quoting Paulo Zanoni (2017-09-01 20:12:01) > > Em Sex, 2017-08-25 às 14:11 +0100, Chris Wilson escreveu: > > > Quoting Lofstedt, Marta (2017-08-25 13:50:16) > > > > > > > > > > > > > -Original Message- > > > > > From: Lofstedt,

Re: [Intel-gfx] [PATCH i-g-t] tests/kms_frontbuffer_tracking: convert macros to functions

Em Seg, 2017-09-04 às 10:00 +0200, Daniel Vetter escreveu: > On Fri, Sep 01, 2017 at 04:44:38PM -0300, Paulo Zanoni wrote: > > Em Seg, 2017-08-14 às 11:25 +0200, Daniel Vetter escreveu: > > > Macros that should be C functions but aren't are really hard to > > > read and confusing. Convert them

Re: [Intel-gfx] [PATCH] drm/i915/cnl: WaFbcSkipSegments

Em Sex, 2017-09-01 às 14:31 -0700, Rodrigo Vivi escreveu: > On Fri, Sep 1, 2017 at 2:04 PM, Paulo Zanoni m> wrote: > > Em Ter, 2017-08-29 às 16:08 -0700, Rodrigo Vivi escreveu: > > > Skip compressing 1 segment at the end of the frame, > > > avoid a pixel count mismatch

Re: [Intel-gfx] [PATCH] drm/atomic: Move drm_crtc_commit to drm_crtc_state, v4.

Hi Maarten, [auto build test ERROR on drm/drm-next] [also build test ERROR on next-20170904] [cannot apply to v4.13] [if your patch is applied to the wrong git tree, please drop us a note to help improve the system] url: https://github.com/0day-ci/linux/commits/Maarten-Lankhorst/drm-atomic

Re: [Intel-gfx] [PATCH] drm/i915: Wake up the device for the fbdev setup

On Sun, Sep 03, 2017 at 01:18:40PM +0100, Chris Wilson wrote: > Quoting ville.syrj...@linux.intel.com (2017-09-01 20:54:56) > > From: Ville Syrjälä > > > > Our fbdev setup requires the device to be awake for access > > through the GTT. If one boots without

Re: [Intel-gfx] [PATCH 2/2] drm/i915: io unmap functions want __iomem

On Fri, Sep 01, 2017 at 07:53:05PM +0100, Chris Wilson wrote: > Quoting ville.syrj...@linux.intel.com (2017-09-01 18:12:52) > > From: Ville Syrjälä > > > > Don't cast away the __iomem from the io_mapping functions so that > > sparse won't be so unhappy when we pass

Re: [Intel-gfx] [PATCH 2/2] drm/i915: Fail addfb ioctl if color and CCS buffers overlap

On Thu, Aug 31, 2017 at 04:52:15PM -0300, Gabriel Krisman Bertazi wrote: > With this patch the new testcase igt@kms_ccs@pipe-X-invalid-ccs-offset > succeeds. I don't think we actually want to reject overlap. I had a patch for that years ago, but I decided to drop it because people might want to

[Intel-gfx] ✗ Fi.CI.IGT: failure for drm/atomic: Fix use-after-free with unbound connectors/planes. (rev3)

== Series Details == Series: drm/atomic: Fix use-after-free with unbound connectors/planes. (rev3) URL : https://patchwork.freedesktop.org/series/29538/ State : failure == Summary == Test kms_atomic_transition: Subgroup plane-use-after-nonblocking-unbind-fencing: fail

Re: [Intel-gfx] [PATCH 0/4] RFC: drm: Allow driver-specific ioctls to be registered

Indeed, we argued at first to let the driver handle the ioctls directly, but we would like to use the DRM interface if possible. On Mon, Sep 4, 2017 at 6:26 PM, Chris Wilson wrote: > Quoting Marius Vlad (2017-09-04 16:16:41) > > From: Marius Vlad

Re: [Intel-gfx] [PATCH 0/4] RFC: drm: Allow driver-specific ioctls to be registered

There isn't any dark plot behind it. For instance, in our use case, a DPU (Display Process Unit) which has a blit feature (using DRM_RENDER_ALLOW) can be implemented cleanly in a separate driver and not being dependent on the DRM core driver. If the blit feature is present/enabled, we can

Re: [Intel-gfx] [PATCH v4 1/2] drm/i915: Track minimum acceptable cdclk instead of "minimum dotclock"

On Mon, Sep 04, 2017 at 12:39:25PM +0200, Maarten Lankhorst wrote: > Op 31-08-17 om 20:48 schreef Ville Syrjälä: > > On Wed, Aug 30, 2017 at 09:57:03PM +0300, ville.syrj...@linux.intel.com > > wrote: > >> From: Ville Syrjälä > >> > >> Make the min_pixclk thing less

Re: [Intel-gfx] [PATCH v3] drm/i915: Add interface to reserve fence registers for vGPU

Quoting Zhenyu Wang (2017-09-04 16:04:03) > On 2017.09.04 11:01:41 +0100, Chris Wilson wrote: > > Quoting changbin...@intel.com (2017-09-04 09:01:01) > > > From: Changbin Du > > > > > > In the past, vGPU alloc fence registers by walking through mm.fence_list > > > to find

Re: [Intel-gfx] [PATCH 0/4] RFC: drm: Allow driver-specific ioctls to be registered

Quoting Marius Vlad (2017-09-04 16:16:41) > From: Marius Vlad > > Currently driver-specific ioctls have to be declared static and are confined > to > DRM core driver. This patch series provides the means to remove those > constrains > and allow to register

Re: [Intel-gfx] [PATCH 0/4] RFC: drm: Allow driver-specific ioctls to be registered

On Mon, Sep 04, 2017 at 06:16:41PM +0300, Marius Vlad wrote: > From: Marius Vlad > > Currently driver-specific ioctls have to be declared static and are confined > to > DRM core driver. This patch series provides the means to remove those > constrains > and allow to

[Intel-gfx] ✓ Fi.CI.BAT: success for drm/atomic: Fix use-after-free with unbound connectors/planes. (rev3)

== Series Details == Series: drm/atomic: Fix use-after-free with unbound connectors/planes. (rev3) URL : https://patchwork.freedesktop.org/series/29538/ State : success == Summary == Series 29538v3 drm/atomic: Fix use-after-free with unbound connectors/planes.

Re: [Intel-gfx] [PATCH v3] drm/i915: Add interface to reserve fence registers for vGPU

On 2017.09.04 11:01:41 +0100, Chris Wilson wrote: > Quoting changbin...@intel.com (2017-09-04 09:01:01) > > From: Changbin Du > > > > In the past, vGPU alloc fence registers by walking through mm.fence_list > > to find fence which pin_count = 0 and vma is empty. vGPU may

[Intel-gfx] [PATCH] drm/atomic: Move drm_crtc_commit to drm_crtc_state, v4.

Most code only cares about the current commit or previous commit. Fortuantely we already have a place to track those. Move it to drm_crtc_state where it belongs. :) The per-crtc commit_list is kept for places where we have to look deeper than the current or previous commit for checking whether to

Re: [Intel-gfx] [PATCH 1/2] drm/i915/dsi: Send SHUTDOWN only for v3+ VBT's

On Mon, Sep 04, 2017 at 10:59:32AM +0300, Mika Kahola wrote: > On Fri, 2017-09-01 at 16:43 +0300, Ville Syrjälä wrote: > > On Fri, Sep 01, 2017 at 10:51:00AM +0300, Mika Kahola wrote: > > > > > > According to spec we should send SHUTDOWN before > > > MIPI_SEQ_DISPLAY_OFF for > > > v3+ VBT's.

Re: [Intel-gfx] [PATCH i-g-t v2] kms_rotation_crc: 90 degree flip test is not a stress test

On 04/09/2017 15:43, Daniel Vetter wrote: On Mon, Sep 04, 2017 at 03:36:56PM +0100, Tvrtko Ursulin wrote: On 04/09/2017 15:27, Tvrtko Ursulin wrote: On 07/08/2017 16:53, Daniel Vetter wrote: On Fri, Aug 04, 2017 at 09:43:41AM +0100, Tvrtko Ursulin wrote: From: Tvrtko Ursulin

Re: [Intel-gfx] [PATCH i-g-t 12/12] RFC: meson build system support

On Mon, Sep 04, 2017 at 04:11:17PM +0300, Jani Nikula wrote: > On Sat, 02 Sep 2017, Daniel Vetter wrote: > > +if cc.has_member('struct sysinfo', 'totalram', > > + prefix : '#include ') > > + config_h.set('HAVE_STRUCT_SYSINFO_TOTALRAM', 1) > > +endif > > + > >

Re: [Intel-gfx] [PATCH] drm/i915: Skip waking the device to service pwrite

On Mon, Sep 04, 2017 at 11:18:07AM +0100, Chris Wilson wrote: > Quoting Daniel Vetter (2017-09-04 09:12:12) > > On Wed, Aug 30, 2017 at 06:48:19PM +0100, Chris Wilson wrote: > > > If the device is in runtime suspend, resuming takes time and reduces our > > > powersaving. If this was for a small

Re: [Intel-gfx] [PATCH i-g-t v2] kms_rotation_crc: 90 degree flip test is not a stress test

On Mon, Sep 04, 2017 at 03:36:56PM +0100, Tvrtko Ursulin wrote: > > On 04/09/2017 15:27, Tvrtko Ursulin wrote: > > On 07/08/2017 16:53, Daniel Vetter wrote: > > > On Fri, Aug 04, 2017 at 09:43:41AM +0100, Tvrtko Ursulin wrote: > > > > From: Tvrtko Ursulin > > > > > > >

Re: [Intel-gfx] [PATCH i-g-t v2] kms_rotation_crc: 90 degree flip test is not a stress test

On 04/09/2017 15:27, Tvrtko Ursulin wrote: On 07/08/2017 16:53, Daniel Vetter wrote: On Fri, Aug 04, 2017 at 09:43:41AM +0100, Tvrtko Ursulin wrote: From: Tvrtko Ursulin To the best of my recollection the page flipping test was added simply to start exercising page

Re: [Intel-gfx] [PATCH i-g-t v2] kms_rotation_crc: 90 degree flip test is not a stress test

On 07/08/2017 16:53, Daniel Vetter wrote: On Fri, Aug 04, 2017 at 09:43:41AM +0100, Tvrtko Ursulin wrote: From: Tvrtko Ursulin To the best of my recollection the page flipping test was added simply to start exercising page flips with 90/270 rotation. There is no

Re: [Intel-gfx] [PATCH v4 4/4] drm/i915: Acquire PUNIT->PMIC bus for intel_uncore_forcewake_reset()

Thomas, Ingo, Peter - Can I have your ack on merging the below patch via drm/i915? Unfortunately it wasn't originally sent to the proper mailing list. You can see the full series it's part of at [1]. BR, Jani. [1] https://patchwork.freedesktop.org/series/29058/ On Sun, 20 Aug 2017, Hans de

Re: [Intel-gfx] [PATCH i-g-t 12/12] RFC: meson build system support

On Sat, 02 Sep 2017, Daniel Vetter wrote: > +if cc.has_member('struct sysinfo', 'totalram', > + prefix : '#include ') > + config_h.set('HAVE_STRUCT_SYSINFO_TOTALRAM', 1) > +endif > + > +add_project_arguments('-D_GNU_SOURCE', language : 'c') Just something

[Intel-gfx] ✗ Fi.CI.IGT: failure for drm/atomic: Fix use-after-free with unbound connectors/planes. (rev2)

== Series Details == Series: drm/atomic: Fix use-after-free with unbound connectors/planes. (rev2) URL : https://patchwork.freedesktop.org/series/29538/ State : failure == Summary == Test kms_setmode: Subgroup basic: pass -> FAIL (shard-hsw) fdo#99912 Test

Re: [Intel-gfx] [PATCH 03/23] drm/i915/gemfs: enable THP

On 29 August 2017 at 15:49, Joonas Lahtinen wrote: > On Mon, 2017-08-21 at 19:34 +0100, Matthew Auld wrote: >> Enable transparent-huge-pages through gemfs by mounting with >> huge=within_size. >> >> Signed-off-by: Matthew Auld >> Cc:

Re: [Intel-gfx] ✓ Fi.CI.IGT: success for series starting with [1/8] drm/i915: Try harder to finish the idle-worker

Quoting Patchwork (2017-09-03 15:17:57) > == Series Details == > > Series: series starting with [1/8] drm/i915: Try harder to finish the > idle-worker > URL : https://patchwork.freedesktop.org/series/29764/ > State : success > > == Summary == > > Test kms_sysfs_edid_timing: >

Re: [Intel-gfx] [RFC PATCH 1/5] drm/i915: Always wait for flip_done

Quoting Maarten Lankhorst (2017-08-30 13:17:48) > The next commit removes the wait for flip_done in in > drm_atomic_helper_commit_cleanup_done, but we need it for the tests > to pass. Instead of using complicated vblank tracking which ends > up being ignored anyway, call the correct atomic helper.

[Intel-gfx] ✓ Fi.CI.BAT: success for drm/atomic: Fix use-after-free with unbound connectors/planes. (rev2)

== Series Details == Series: drm/atomic: Fix use-after-free with unbound connectors/planes. (rev2) URL : https://patchwork.freedesktop.org/series/29538/ State : success == Summary == Series 29538v2 drm/atomic: Fix use-after-free with unbound connectors/planes.

Re: [Intel-gfx] [PATCH 1/2] drm/i915: Silence sparse by using gfp_t

Quoting Joonas Lahtinen (2017-09-04 07:50:38) > On Fri, 2017-09-01 at 15:57 +0100, Chris Wilson wrote: > > Sparse enforces that GFP flags are only manipulated inside gfp_t locals. > > > > Fixes: 4d470f7359c4 ("drm/i915: Avoid undefined behaviour of "u32 >> 32"") > > Signed-off-by: Chris Wilson

Re: [Intel-gfx] [PATCH] drm/i915: Try harder to finish the idle-worker

Quoting Daniel Vetter (2017-09-04 09:35:49) > On Fri, Sep 01, 2017 at 03:11:23PM +0100, Chris Wilson wrote: > > If a worker requeues itself, it may switch to a different kworker pool, > > which flush_work() considers as complete. To be strict, we then need to > > keep flushing the work until it is

[Intel-gfx] ✗ Fi.CI.IGT: warning for drm/i915: Add interface to reserve fence registers for vGPU (rev3)

== Series Details == Series: drm/i915: Add interface to reserve fence registers for vGPU (rev3) URL : https://patchwork.freedesktop.org/series/29523/ State : warning == Summary == Test kms_flip: Subgroup wf_vblank-vs-dpms-interruptible: pass -> DMESG-WARN

Re: [Intel-gfx] [PATCH] drm/i915: Try harder to finish the idle-worker

Quoting Mika Kuoppala (2017-09-04 11:19:09) > Chris Wilson writes: > > +/* > > + * Wait until the work is finally complete, even if it tries to postpone > > + * by requeueing itself. Note, that if the worker never cancels itself, > > + * we will spin forever. > > + */ >

[Intel-gfx] [PATCH v2 4/6] drm/atomic: Return commit in drm_crtc_commit_get for better annotation

This will allow code to do x->commit = drm_crtc_commit_get(commit), making it clearer where references are used. Signed-off-by: Maarten Lankhorst --- drivers/gpu/drm/drm_atomic_helper.c | 3 +-- include/drm/drm_atomic.h| 6 +- 2 files changed,

[Intel-gfx] [PATCH v2 5/6] drm/atomic: Fix freeing connector/plane state too early by tracking commits, v3.

Currently we neatly track the crtc state, but forget to look at plane/connector state. When doing a nonblocking modeset, immediately followed by a setprop before the modeset completes, the setprop will see the modesets new state as the old state and free it. This has to be solved by waiting for

[Intel-gfx] [PATCH v2 3/6] drm/atomic: Remove waits in drm_atomic_helper_commit_cleanup_done, v2.

When commit synchronization through drm_crtc_commit was first introduced, we tried to solve the problem of the flip_done needing a reference count by blocking in cleanup_done. This has been changed by commit 24835e442f28 ("drm: reference count event->completion") which made the waits here no

[Intel-gfx] [PATCH v2 6/6] drm/atomic: Make async plane update checks work as intended, v2.

By always keeping track of the last commit in plane_state, we know whether there is an active update on the plane or not. With that information we can reject the fast update, and force the slowpath to be used as was originally intended. We cannot use plane_state->crtc->state here, because this

[Intel-gfx] [PATCH v2 2/6] drm/atomic: Move drm_crtc_commit to drm_crtc_state, v3.

Most code only cares about the current commit or previous commit. Fortuantely we already have a place to track those. Move it to drm_crtc_state where it belongs. :) The per-crtc commit_list is kept for places where we have to look deeper than the current or previous commit for checking whether to

[Intel-gfx] [PATCH v2 0/6] drm/atomic: Fix use-after-free with unbound connectors/planes.

New iteration with all feedback hopefully addressed. Maarten Lankhorst (6): drm/i915: Always wait for flip_done, v2. drm/atomic: Move drm_crtc_commit to drm_crtc_state, v3. drm/atomic: Remove waits in drm_atomic_helper_commit_cleanup_done, v2. drm/atomic: Return commit in

[Intel-gfx] [PATCH v2 1/6] drm/i915: Always wait for flip_done, v2.

The next commit removes the wait for flip_done in in drm_atomic_helper_commit_cleanup_done, but we need it for the tests to pass. Instead of using complicated vblank tracking which ends up being ignored anyway, call the correct atomic helper. :) Changes since v1: - Always call

Re: [Intel-gfx] [PATCH i-g-t v2] tests/kms_frontbuffer_tracking: increase FBC wait timeout to 5s

Quoting Paulo Zanoni (2017-09-01 20:12:01) > Em Sex, 2017-08-25 às 14:11 +0100, Chris Wilson escreveu: > > Quoting Lofstedt, Marta (2017-08-25 13:50:16) > > > > > > > > > > -Original Message- > > > > From: Lofstedt, Marta > > > > Sent: Friday, August 25, 2017 2:54 PM > > > > To: 'Chris

Re: [Intel-gfx] [PATCH v4 1/2] drm/i915: Track minimum acceptable cdclk instead of "minimum dotclock"

Op 31-08-17 om 20:48 schreef Ville Syrjälä: > On Wed, Aug 30, 2017 at 09:57:03PM +0300, ville.syrj...@linux.intel.com wrote: >> From: Ville Syrjälä >> >> Make the min_pixclk thing less confusing by changing it to track >> the minimum acceptable cdclk frequency

Re: [Intel-gfx] [PATCH i-g-t 02/12] build: Nuke #ifdef HAVE_CONFIG_H cargo-cult

On Mon, Sep 04, 2017 at 12:26:25PM +0200, Szwichtenberg, Radoslaw wrote: > On Mon, 2017-09-04 at 12:27 +0300, Arkadiusz Hiler wrote: > > On Mon, Sep 04, 2017 at 10:45:19AM +0200, Daniel Vetter wrote: > > > On Mon, Sep 04, 2017 at 11:28:09AM +0300, Arkadiusz Hiler wrote: > > > > On Sat, Sep 02,

Re: [Intel-gfx] [PATCH i-g-t 02/12] build: Nuke #ifdef HAVE_CONFIG_H cargo-cult

On Mon, 2017-09-04 at 12:27 +0300, Arkadiusz Hiler wrote: > On Mon, Sep 04, 2017 at 10:45:19AM +0200, Daniel Vetter wrote: > > On Mon, Sep 04, 2017 at 11:28:09AM +0300, Arkadiusz Hiler wrote: > > > On Sat, Sep 02, 2017 at 07:03:56PM +0200, Daniel Vetter wrote: > > > > We have it. Daniel Stone said

Re: [Intel-gfx] [PATCH] drm/i915: Try harder to finish the idle-worker

Chris Wilson writes: > If a worker requeues itself, it may switch to a different kworker pool, > which flush_work() considers as complete. To be strict, we then need to > keep flushing the work until it is no longer pending. > > References:

Re: [Intel-gfx] [PATCH] drm/i915: Skip waking the device to service pwrite

Quoting Daniel Vetter (2017-09-04 09:12:12) > On Wed, Aug 30, 2017 at 06:48:19PM +0100, Chris Wilson wrote: > > If the device is in runtime suspend, resuming takes time and reduces our > > powersaving. If this was for a small write into an object, that resume > > will take longer than any savings

Re: [Intel-gfx] [PATCH v3] drm/i915: Add interface to reserve fence registers for vGPU

Quoting changbin...@intel.com (2017-09-04 09:01:01) > From: Changbin Du > > In the past, vGPU alloc fence registers by walking through mm.fence_list > to find fence which pin_count = 0 and vma is empty. vGPU may not find > enough fence registers this way. Because a fence

Re: [Intel-gfx] [PATCH i-g-t 02/12] build: Nuke #ifdef HAVE_CONFIG_H cargo-cult

On Mon, Sep 04, 2017 at 10:45:19AM +0200, Daniel Vetter wrote: > On Mon, Sep 04, 2017 at 11:28:09AM +0300, Arkadiusz Hiler wrote: > > On Sat, Sep 02, 2017 at 07:03:56PM +0200, Daniel Vetter wrote: > > > We have it. Daniel Stone said it comes from the X11 transition to the > > > modular build. > >

Re: [Intel-gfx] [PATCH v5 4/4] drm/i915/perf: add support for Coffeelake GT2

On 30 August 2017 at 17:12, Lionel Landwerlin wrote: > Add the test configuration & timestamp frequency for Coffeelake GT2. > > Signed-off-by: Lionel Landwerlin Do we not want to also disable the clock-ratio-change reports? Also can

Re: [Intel-gfx] [PATCH i-g-t] scripts/run-tests.sh: Use piglit's --ignore-missing

On Mon, Sep 04, 2017 at 10:33:32AM +0200, Daniel Vetter wrote: > On Fri, Sep 01, 2017 at 04:09:03PM +0300, Arkadiusz Hiler wrote: > > Recently we added a number of chamelium tests to the fast-feedback testlist. > > > > Chemelium is build-optional - requires `./configure --enable-chamelium`. > >

Re: [Intel-gfx] [PATCH i-g-t 12/12] RFC: meson build system support

On Sat, Sep 02, 2017 at 07:04:06PM +0200, Daniel Vetter wrote: > Why? > > Because it's fast. > > Like really, really fast. > > Some data (from a snb laptop, so rather lower-powered): > > - Incremental build after $ touch lib/igt_core.c with meson: 0.6s > It notices that the symbol list of

[Intel-gfx] [PATCH i-g-t 3/7] meson: Add some compiler flags to reduce warnings.

From: Eric Anholt These warnings are apparently new compared to the autotools build. We can fix the things they complain about later, if we want. Signed-off-by: Eric Anholt Signed-off-by: Daniel Vetter --- meson.build | 11

[Intel-gfx] [PATCH i-g-t 7/7] meson: add manpage support

It seems like meson doesn't want you to string together targets like make does, but wants it all in one step. So another little shell script it is. Signed-off-by: Daniel Vetter --- man/defs.rst.in | 5 + man/meson.build | 45

[Intel-gfx] [PATCH i-g-t 5/7] meson: Use static libs to handle IGT_LOG_DOMAIN.

From: Eric Anholt It means that compiler errors in the .c files take you to the source place in your editor, not a preprocessed temporary. v2: Add the library deps, fails linking otherwise. Signed-off-by: Eric Anholt (v1) Signed-off-by: Daniel Vetter

[Intel-gfx] [PATCH i-g-t 1/7] lib/ioctl_wrappers: make the valgrind wrapper always emit a statement:w

gcc complains otherwise about empty ; statements ... Signed-off-by: Daniel Vetter --- lib/ioctl_wrappers.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/lib/ioctl_wrappers.c b/lib/ioctl_wrappers.c index b4d6210d5942..48750427a0c1 100644 ---

[Intel-gfx] [PATCH i-g-t 6/7] meson: detect cc flags

Somehow my gcc has a different idea of what no-implicit-fallthrough should look like than the one Eric used. fixup compiler flags Signed-off-by: Daniel Vetter --- meson.build | 22 ++ 1 file changed, 14 insertions(+), 8 deletions(-) diff --git

[Intel-gfx] [PATCH i-g-t 4/7] meson: Don't build the igt audio test without gsl available.

From: Eric Anholt Signed-off-by: Eric Anholt Signed-off-by: Daniel Vetter --- tests/meson.build | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/meson.build b/tests/meson.build index 73833758be0e..4dd5a9c9d4c7

[Intel-gfx] [PATCH i-g-t 2/7] tests/kms_plane: Appease gcc -Wempty-body

Not exactly sure what's the point, but oh well. Signed-off-by: Daniel Vetter --- tests/kms_plane.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/tests/kms_plane.c b/tests/kms_plane.c index 927d5d37fece..812497500d2d 100644 --- a/tests/kms_plane.c

[Intel-gfx] ✓ Fi.CI.BAT: success for drm/i915: Add interface to reserve fence registers for vGPU (rev3)

== Series Details == Series: drm/i915: Add interface to reserve fence registers for vGPU (rev3) URL : https://patchwork.freedesktop.org/series/29523/ State : success == Summary == Series 29523v3 drm/i915: Add interface to reserve fence registers for vGPU

Re: [Intel-gfx] [PATCH i-g-t 1/5] Add support for subtest-specific documentation

On Mon, Sep 04, 2017 at 08:42:58AM +, Szwichtenberg, Radoslaw wrote: > On Fri, 2017-08-11 at 10:20 +0200, Daniel Vetter wrote: > > On Thu, Aug 10, 2017 at 01:26:47PM +0300, Petri Latvala wrote: > > > The current documentation for tests is limited to a single string per > > > test binary. This

Re: [Intel-gfx] [PATCH i-g-t 1/3] Use PATH_MAX to fix some sprintf-into-short-buffers warnings.

On Sat, Sep 02, 2017 at 09:38:51PM -0700, Eric Anholt wrote: > Signed-off-by: Eric Anholt > --- > > This little series cleans up many compiler warnings I saw when testing > danvet's meson branch. On the series: Reviewed-by: Daniel Vetter I've pulled in

Re: [Intel-gfx] [PATCH i-g-t 01/12] build: Define _GNU_SOURCE in Makefile.am

On Sun, Sep 03, 2017 at 10:27:03AM +0100, Chris Wilson wrote: > Quoting Daniel Vetter (2017-09-02 18:03:55) > > In meson I want to just set this everywhere (no reason not to), and > > doing so will allow us to clean up a few things. > > GNU_SOURCE is defined in config.h, which should be included

Re: [Intel-gfx] [PATCH i-g-t 02/12] build: Nuke #ifdef HAVE_CONFIG_H cargo-cult

On Mon, Sep 04, 2017 at 11:28:09AM +0300, Arkadiusz Hiler wrote: > On Sat, Sep 02, 2017 at 07:03:56PM +0200, Daniel Vetter wrote: > > We have it. Daniel Stone said it comes from the X11 transition to the > > modular build. > > > > Signed-off-by: Daniel Vetter > > It's

Re: [Intel-gfx] [PATCH i-g-t 1/5] Add support for subtest-specific documentation

On Fri, 2017-08-11 at 10:20 +0200, Daniel Vetter wrote: > On Thu, Aug 10, 2017 at 01:26:47PM +0300, 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

Re: [Intel-gfx] [PATCH i-g-t 02/12] build: Nuke #ifdef HAVE_CONFIG_H cargo-cult

On Sun, Sep 03, 2017 at 10:29:06AM +0100, Chris Wilson wrote: > Quoting Daniel Vetter (2017-09-02 18:03:56) > > We have it. Daniel Stone said it comes from the X11 transition to the > > modular build. > > What? It's from AC_CONFIG_HEADERS([config.h]) > > We tell autoconf to build it so it

Re: [Intel-gfx] [PATCH] drm/i915: Try harder to finish the idle-worker

On Fri, Sep 01, 2017 at 03:11:23PM +0100, Chris Wilson wrote: > If a worker requeues itself, it may switch to a different kworker pool, > which flush_work() considers as complete. To be strict, we then need to > keep flushing the work until it is no longer pending. > > References:

Re: [Intel-gfx] [PATCH i-g-t] scripts/run-tests.sh: Use piglit's --ignore-missing

On Fri, Sep 01, 2017 at 04:09:03PM +0300, Arkadiusz Hiler wrote: > Recently we added a number of chamelium tests to the fast-feedback testlist. > > Chemelium is build-optional - requires `./configure --enable-chamelium`. > > To mitigate issue with piglit exiting abruptly due to the (possibly) >

Re: [Intel-gfx] [PATCH i-g-t v2] tests/gem_flink_basic: Add documentation for subtests

On Thu, Aug 31, 2017 at 02:33:23PM -0700, Vinay Belgaumkar wrote: > Added the missing IGT_TEST_DESCRIPTION and some subtest > descriptions. Trying to establish a method to document > subtests, it should describe the feature being tested > rather than how. The HOW part can, if needed, be >

Re: [Intel-gfx] [PATCH i-g-t 02/12] build: Nuke #ifdef HAVE_CONFIG_H cargo-cult

On Sat, Sep 02, 2017 at 07:03:56PM +0200, Daniel Vetter wrote: > We have it. Daniel Stone said it comes from the X11 transition to the > modular build. > > Signed-off-by: Daniel Vetter It's also used for Android builds, which lack the generated config.h. But is this

Re: [Intel-gfx] [PATCH 1/2] drm: Fix example comment of format modifier blob

On Thu, Aug 31, 2017 at 01:00:35PM -0700, Ben Widawsky wrote: > On 17-08-31 16:52:14, Gabriel Krisman Bertazi wrote: > > To represent formats 98-102, the supported formats mask must be > > 0x7c and not 0x3c. > > > > Signed-off-by: Gabriel Krisman Bertazi

Re: [Intel-gfx] [PATCH 2/2] drm/i915: Fail addfb ioctl if color and CCS buffers overlap

On Thu, Aug 31, 2017 at 04:52:15PM -0300, Gabriel Krisman Bertazi wrote: > With this patch the new testcase igt@kms_ccs@pipe-X-invalid-ccs-offset > succeeds. > > Signed-off-by: Gabriel Krisman Bertazi Do we have igts for this? If so, please add a Testcase: line. -Daniel

Re: [Intel-gfx] [PATCH v3] drm/i915: Keep the device awake whilst in the GTT domain

On Fri, Sep 01, 2017 at 01:16:06PM +0100, Chris Wilson wrote: > Since runtime suspend is very harsh on GTT mmappings (they all get > zapped on suspend) keep the device awake while the buffer remains in > the GTT domain. However, userspace can control the domain and > although there is a soft

Re: [Intel-gfx] [PATCH] drm/i915: Disable DRRS when PSR is enabled

On Thu, Aug 31, 2017 at 05:37:31PM +, Sripada, Radhakrishna wrote: > > > > -Original Message- > > From: Vivi, Rodrigo > > Sent: Wednesday, August 30, 2017 5:59 PM > > To: Sripada, Radhakrishna > > Cc: intel-gfx@lists.freedesktop.org; Pandiyan,

Re: [Intel-gfx] [PATCH] drm/i915: Skip waking the device to service pwrite

On Wed, Aug 30, 2017 at 06:48:19PM +0100, Chris Wilson wrote: > If the device is in runtime suspend, resuming takes time and reduces our > powersaving. If this was for a small write into an object, that resume > will take longer than any savings in using the indirect GGTT access to > avoid the cpu

Re: [Intel-gfx] [PATCH i-g-t] lib/igt_aux: Allow sysfs open to fail when setting suspend/resume delay

On Wed, Aug 30, 2017 at 09:15:03PM +0300, Arkadiusz Hiler wrote: > On Wed, Aug 30, 2017 at 04:56:09PM +0300, Paul Kocialkowski wrote: > > This removes the igt_require condition on the sysfs open call used to > > write the suspend/resume delay so that it is allowed to fail. Intsead, > > the code

[Intel-gfx] [PATCH v3] drm/i915: Add interface to reserve fence registers for vGPU

From: Changbin Du In the past, vGPU alloc fence registers by walking through mm.fence_list to find fence which pin_count = 0 and vma is empty. vGPU may not find enough fence registers this way. Because a fence can be bind to vma even though it is not in using. We have

Re: [Intel-gfx] [PATCH i-g-t] tests/kms_frontbuffer_tracking: convert macros to functions

On Fri, Sep 01, 2017 at 04:44:38PM -0300, Paulo Zanoni wrote: > Em Seg, 2017-08-14 às 11:25 +0200, Daniel Vetter escreveu: > > Macros that should be C functions but aren't are really hard to > > read and confusing. Convert them over. > > > > v2: Clean up commit message and keep printing the line

Re: [Intel-gfx] [PATCH 1/2] drm/i915/dsi: Send SHUTDOWN only for v3+ VBT's

On Fri, 2017-09-01 at 16:43 +0300, Ville Syrjälä wrote: > On Fri, Sep 01, 2017 at 10:51:00AM +0300, Mika Kahola wrote: > > > > According to spec we should send SHUTDOWN before > > MIPI_SEQ_DISPLAY_OFF for > > v3+ VBT's. Testing with VBT v3 the current implementation yields > > the > > following

Re: [Intel-gfx] [RFC PATCH 1/5] drm/i915: Always wait for flip_done

Op 04-09-17 om 09:30 schreef Daniel Vetter: > On Thu, Aug 31, 2017 at 05:18:06PM +0200, Maarten Lankhorst wrote: >> Op 30-08-17 om 15:40 schreef Daniel Vetter: >>> On Wed, Aug 30, 2017 at 02:54:28PM +0200, Maarten Lankhorst wrote: Op 30-08-17 om 14:43 schreef Daniel Vetter: > On Wed, Aug

Re: [Intel-gfx] [RFC PATCH 5/5] drm/atomic: Make async plane update checks work as intended.

On Sat, Sep 02, 2017 at 03:59:22PM -0300, Gustavo Padovan wrote: > 2017-08-30 Daniel Vetter : > > > On Wed, Aug 30, 2017 at 02:17:52PM +0200, Maarten Lankhorst wrote: > > > By always keeping track of the last commit in plane_state, we know > > > whether there is an active update

Re: [Intel-gfx] [RFC PATCH 1/5] drm/i915: Always wait for flip_done

On Thu, Aug 31, 2017 at 05:18:06PM +0200, Maarten Lankhorst wrote: > Op 30-08-17 om 15:40 schreef Daniel Vetter: > > On Wed, Aug 30, 2017 at 02:54:28PM +0200, Maarten Lankhorst wrote: > >> Op 30-08-17 om 14:43 schreef Daniel Vetter: > >>> On Wed, Aug 30, 2017 at 02:17:48PM +0200, Maarten Lankhorst

Re: [Intel-gfx] [PATCH i-g-t v2] tests/gem_flink_basic: Add documentation for subtests

On Fri, Sep 01, 2017 at 12:55:37PM +0100, Arkadiusz Hiler wrote: > On Thu, Aug 31, 2017 at 02:33:23PM -0700, Vinay Belgaumkar wrote: > > Added the missing IGT_TEST_DESCRIPTION and some subtest > > descriptions. Trying to establish a method to document > > Hey Vinay, > > Please add appropriate

Re: [Intel-gfx] [PATCH 1/2] drm/i915: Silence sparse by using gfp_t

On Fri, 2017-09-01 at 15:57 +0100, Chris Wilson wrote: > Sparse enforces that GFP flags are only manipulated inside gfp_t locals. > > Fixes: 4d470f7359c4 ("drm/i915: Avoid undefined behaviour of "u32 >> 32"") > Signed-off-by: Chris Wilson > Cc: Joonas Lahtinen

Re: [Intel-gfx] [RFC PATCH 4/4] drm/i915: Expose RPCS (SSEU) configuration to userspace

On Fri, 2017-09-01 at 18:12 +0100, Lionel Landwerlin wrote: > From: Chris Wilson > > We want to allow userspace to reconfigure the subslice configuration for > its own use case. To do so, we expose a context parameter to allow > adjustment of the RPCS register stored

Re: [Intel-gfx] [PATCH v14 0/7] drm/i915/gvt: Dma-buf support for GVT-g

Hi, > > However, I see VFIO_DEVICE_QUERY_GFX_PLANE failures which I think > > should > > not be there.  When the guest didn't define a plane yet I get "No > > such device" > > errors instead of a plane_info struct with fields (drm_format, > > width, height, size) > > set to zero.  I also see

  1   2   >