Re: [Intel-gfx] linux-next: manual merge of the mali-dp tree with the drm-misc tree

Hi Liviu, On Thu, 3 Nov 2016 17:19:58 + Liviu Dudau wrote: > > I have revamped the mali-dp tree and rebased it on the newer > version of drm-next (which includes the drm-misc change) and pushed the > updated patch in my tree. Thanks for that. However, several of the

Re: [Intel-gfx] [PATCH v8 04/12] drm/i915: return EACCES for check_cmd() failures

On Thu, 2016-10-27 at 19:14 -0700, Robert Bragg wrote: > check_cmd() is checking whether a command adheres to certain > restrictions that ensure it's safe to execute within a privileged batch > buffer. Returning false implies a privilege problem, not that the > command is invalid. > > The

Re: [Intel-gfx] [PATCH] drm/i915/dp: Update connector status for DP MST hotplugs

Hi Dhinakaran, [auto build test ERROR on drm-intel/for-linux-next] [also build test ERROR on v4.9-rc3 next-20161028] [if your patch is applied to the wrong git tree, please drop us a note to help improve the system] url:

[Intel-gfx] ✗ Fi.CI.BAT: warning for drm/i915/dp: Update connector status for DP MST hotplugs (rev2)

== Series Details == Series: drm/i915/dp: Update connector status for DP MST hotplugs (rev2) URL : https://patchwork.freedesktop.org/series/14821/ State : warning == Summary == Series 14821v2 drm/i915/dp: Update connector status for DP MST hotplugs

[Intel-gfx] [PATCH] drm/i915/dp: Update connector status for DP MST hotplugs

Hotplugging a monitor in DP MST configuration triggers short pulses. Although the short pulse handling path ends up in the MST hotplug function, we do not perform a detect before sending the hotplug uvent. This leads to the connector status not being updated when the userspace calls

[Intel-gfx] [PATCH] drm/i915/dp: Update connector status for DP MST hotplugs

Hotplugging a monitor in DP MST configuration triggers short pulses. Although the short pulse handling path ends up in the MST hotplug function, we do not perform a detect before sending the hotplug uvent. This leads to the connector status not being updated when the userspace calls

Re: [Intel-gfx] [PATCH v8 10/12] drm/i915: add oa_event_min_timer_exponent sysctl

On Wed, Nov 2, 2016 at 6:29 AM, sourab gupta wrote: > On Thu, 2016-10-27 at 19:14 -0700, Robert Bragg wrote: > > The minimal sampling period is now configurable via a > > dev.i915.oa_min_timer_exponent sysctl parameter. > > > > Following the precedent set by perf, the

Re: [Intel-gfx] [RFC PATCH v2 0/8] Add support for Legacy Hdmi audio

On Sat, Oct 1, 2016 at 2:22 AM, Jerome Anand wrote: > Legacy Hdmi audio drivers are added. > Added support for audio/ gfx interface using irq chip framework Just discussed this with Rakesh here at LPC and also with Mark Brown and since earlier this years there's no a

Re: [Intel-gfx] [PATCH 1/2] drm/i915: Make sure engines are idle during GPU idling in LR mode

On Thu, Nov 03, 2016 at 10:57:23PM +0200, Imre Deak wrote: > On Thu, 2016-11-03 at 18:59 +, Chris Wilson wrote: > > On Thu, Nov 03, 2016 at 06:19:37PM +0200, Imre Deak wrote: > > > We assume that the GPU is idle once receiving the seqno via the last > > > request's user interrupt. In execlist

Re: [Intel-gfx] ✓ Fi.CI.BAT: success for drm/i915: Fix test on inputs for vma_compare()

On Thu, Nov 03, 2016 at 08:45:31PM -, Patchwork wrote: > == Series Details == > > Series: drm/i915: Fix test on inputs for vma_compare() > URL : https://patchwork.freedesktop.org/series/14798/ > State : success > > == Summary == > > Series 14798v1 drm/i915: Fix test on inputs for

Re: [Intel-gfx] [PATCH 1/2] drm/i915: Make sure engines are idle during GPU idling in LR mode

On Thu, 2016-11-03 at 18:59 +, Chris Wilson wrote: > On Thu, Nov 03, 2016 at 06:19:37PM +0200, Imre Deak wrote: > > We assume that the GPU is idle once receiving the seqno via the last > > request's user interrupt. In execlist mode the corresponding context > > completed interrupt can be

[Intel-gfx] ✓ Fi.CI.BAT: success for drm/i915: Fix test on inputs for vma_compare()

== Series Details == Series: drm/i915: Fix test on inputs for vma_compare() URL : https://patchwork.freedesktop.org/series/14798/ State : success == Summary == Series 14798v1 drm/i915: Fix test on inputs for vma_compare()

Re: [Intel-gfx] [PATCH] drm/i915: Fix test on inputs for vma_compare()

On 3 November 2016 at 20:08, Chris Wilson wrote: > When supplying a view to vma_compare() it is required that the supplied > i915_address_space is the global GTT. I tested the VMA instead (which is > the current position in the rbtree and maybe from any address space). >

[Intel-gfx] [PATCH] drm/i915: Fix test on inputs for vma_compare()

When supplying a view to vma_compare() it is required that the supplied i915_address_space is the global GTT. I tested the VMA instead (which is the current position in the rbtree and maybe from any address space). Reported-by: Matthew Auld Tested-by: Matthew Auld

Re: [Intel-gfx] [PATCH 08/12] drm/i915/guc: Cache the client mapping

On Thu, Nov 03, 2016 at 04:37:24PM +, Tvrtko Ursulin wrote: > > On 02/11/2016 17:50, Chris Wilson wrote: > >Use i915_gem_object_pin_map() for the guc client's lifetime to replace > >the peristent kmap + frequent kmap_atomic with a permanent vmapping. > >This avoids taking the obj->mm.lock

Re: [Intel-gfx] [PATCH 06/12] drm/i915/scheduler: Execute requests in order of priorities

On Thu, Nov 03, 2016 at 04:21:25PM +, Tvrtko Ursulin wrote: > >+static void update_priorities(struct i915_priotree *pt, int prio) > >+{ > >+struct drm_i915_gem_request *request = > >+container_of(pt, struct drm_i915_gem_request, priotree); > >+struct intel_engine_cs *engine

Re: [Intel-gfx] [PATCH 2/2] drm/i915: Add assert for no pending GPU requests during resume in LR mode

On Thu, Nov 03, 2016 at 06:19:38PM +0200, Imre Deak wrote: > During resume we will reset the SW/HW tracking for each ring head/tail > pointers and so are not prepared to replay any pending requests (as > opposed to GPU reset time). Add an assert for this. A bit belated. You could just put

Re: [Intel-gfx] [PATCH v4 3/8] drm/i915: Decode system memory bandwidth

Em Qui, 2016-10-13 às 16:28 +0530, Kumar, Mahesh escreveu: > This patch adds support to decode system memory bandwidth > which will be used for arbitrated display memory percentage > calculation in GEN9 based system. > > Changes from v1: >  - Address comments from Paulo >  - implement decode

Re: [Intel-gfx] [PATCH 1/2] drm/i915: Make sure engines are idle during GPU idling in LR mode

On Thu, Nov 03, 2016 at 06:19:37PM +0200, Imre Deak wrote: > We assume that the GPU is idle once receiving the seqno via the last > request's user interrupt. In execlist mode the corresponding context > completed interrupt can be delayed though and until this latter > interrupt arrives we consider

Re: [Intel-gfx] [PATCH] drm: make drm_get_format_name thread-safe

On Sun, Aug 14, 2016 at 8:02 PM, Eric Engestrom wrote: > Signed-off-by: Eric Engestrom > --- > > I moved the main bits to be the first diffs, shouldn't affect anything > when applying the patch, but I wanted to ask: > I don't like the hard-coded `32` the

Re: [Intel-gfx] [PATCH 2/2] drm/i915: Reinit polling before hpd when resuming

On Thu, Nov 03, 2016 at 11:42:38AM -0400, Lyude wrote: > Now that we don't run the connector reprobing from i915_drm_resume(), we > need to make it so we don't have to wait for reprobing to finish so that > we actually speed things up. In order to do this, we need to make sure > that

Re: [Intel-gfx] [PATCH 1/2] drm/i915: Remove redundant reprobe in i915_drm_resume

On Thu, Nov 03, 2016 at 11:42:37AM -0400, Lyude wrote: > Weine's investigation on benchmarking the suspend/resume process pointed > out a lot of the time in suspend/resume is being spent reprobing. While > the reprobing process is a lengthy one for good reason, we don't need to > hold up the

Re: [Intel-gfx] [PATCH 16/19] drm/i915: Use new atomic iterator macros in fbc

On Thu, Nov 03, 2016 at 07:55:20PM +0200, Ville Syrjälä wrote: > On Thu, Nov 03, 2016 at 03:45:43PM -0200, Paulo Zanoni wrote: > > Em Qui, 2016-11-03 às 18:45 +0200, Ville Syrjälä escreveu: > > > On Mon, Oct 17, 2016 at 02:37:15PM +0200, Maarten Lankhorst wrote: > > > > > > > > Signed-off-by:

Re: [Intel-gfx] [PATCH 16/19] drm/i915: Use new atomic iterator macros in fbc

On Thu, Nov 03, 2016 at 03:45:43PM -0200, Paulo Zanoni wrote: > Em Qui, 2016-11-03 às 18:45 +0200, Ville Syrjälä escreveu: > > On Mon, Oct 17, 2016 at 02:37:15PM +0200, Maarten Lankhorst wrote: > > > > > > Signed-off-by: Maarten Lankhorst > > > > > > --- > > >

Re: [Intel-gfx] [PATCH 17/19] drm/i915: Use new atomic iterator macros in wm code

Em Qui, 2016-11-03 às 18:49 +0200, Ville Syrjälä escreveu: > On Mon, Oct 17, 2016 at 02:37:16PM +0200, Maarten Lankhorst wrote: > > > > Signed-off-by: Maarten Lankhorst > > > > --- > >  drivers/gpu/drm/i915/intel_pm.c | 12 ++-- > >  1 file changed, 6

Re: [Intel-gfx] [PATCH 16/19] drm/i915: Use new atomic iterator macros in fbc

Em Qui, 2016-11-03 às 18:45 +0200, Ville Syrjälä escreveu: > On Mon, Oct 17, 2016 at 02:37:15PM +0200, Maarten Lankhorst wrote: > > > > Signed-off-by: Maarten Lankhorst > > > > --- > >  drivers/gpu/drm/i915/intel_fbc.c | 6 +++--- > >  1 file changed, 3

Re: [Intel-gfx] [PATCH 06/12] drm/i915/scheduler: Execute requests in order of priorities

On Thu, Nov 03, 2016 at 04:21:25PM +, Tvrtko Ursulin wrote: > >+rb = rb_next(rb); > >+rb_erase(>priotree.node, >execlist_queue); > >+RB_CLEAR_NODE(>priotree.node); > >+cursor->priotree.priority = INT_MAX; > >+ > >+/* We keep the

Re: [Intel-gfx] linux-next: manual merge of the mali-dp tree with the drm-misc tree

On Tue, Oct 25, 2016 at 11:20:44AM +1100, Stephen Rothwell wrote: > Hi Liviu, Hi Stephen, > > Today's linux-next merge of the mali-dp tree got a conflict in: > > drivers/gpu/drm/arm/malidp_planes.c > > between commit: > > ea0e1ce20f73 ("drm/arm: Use per-plane rotation property") > >

Re: [Intel-gfx] [PATCH 19/19] drm/atomic: Rename atomic oldnew iterator

On Mon, Oct 17, 2016 at 02:37:18PM +0200, Maarten Lankhorst wrote: > With the old users of for_each_obj_in_state gone, we can rename > for_each_oldnew_obj_in_state back to that name. It's slightly less > ugly. > > Signed-off-by: Maarten Lankhorst Simple

Re: [Intel-gfx] [PATCH 15/19] drm/i915: Use new atomic iterator macros in ddi

On Mon, Oct 17, 2016 at 02:37:14PM +0200, Maarten Lankhorst wrote: > Also make the function static, it's only used inside intel_ddi.c > > Signed-off-by: Maarten Lankhorst > --- > drivers/gpu/drm/i915/intel_ddi.c | 4 ++-- > drivers/gpu/drm/i915/intel_drv.h | 2

Re: [Intel-gfx] [PATCH 18/19] drm/i915: Use new atomic iterator macros in display code

On Mon, Oct 17, 2016 at 02:37:17PM +0200, Maarten Lankhorst wrote: > Signed-off-by: Maarten Lankhorst > --- > drivers/gpu/drm/i915/intel_display.c | 156 > ++- > 1 file changed, 80 insertions(+), 76 deletions(-) > > diff --git

Re: [Intel-gfx] [PATCH 07/12] drm/i915/scheduler: Boost priorities for flips

On Thu, Nov 03, 2016 at 04:29:52PM +, Tvrtko Ursulin wrote: > > On 02/11/2016 17:50, Chris Wilson wrote: > >Boost the priority of any rendering required to show the next pageflip > >as we want to avoid missing the vblank by being delayed by invisible > >workload. We prioritise avoiding jank

Re: [Intel-gfx] [PATCH 16/19] drm/i915: Use new atomic iterator macros in fbc

On Mon, Oct 17, 2016 at 02:37:15PM +0200, Maarten Lankhorst wrote: > Signed-off-by: Maarten Lankhorst > --- > drivers/gpu/drm/i915/intel_fbc.c | 6 +++--- > 1 file changed, 3 insertions(+), 3 deletions(-) > > diff --git a/drivers/gpu/drm/i915/intel_fbc.c >

Re: [Intel-gfx] [PATCH 1/2] drm/i915: Make sure engines are idle during GPU idling in LR mode

On 03/11/2016 16:19, Imre Deak wrote: We assume that the GPU is idle once receiving the seqno via the last request's user interrupt. In execlist mode the corresponding context completed interrupt can be delayed though and until this latter interrupt arrives we consider the request to be pending

Re: [Intel-gfx] [PATCH 17/19] drm/i915: Use new atomic iterator macros in wm code

On Mon, Oct 17, 2016 at 02:37:16PM +0200, Maarten Lankhorst wrote: > Signed-off-by: Maarten Lankhorst > --- > drivers/gpu/drm/i915/intel_pm.c | 12 ++-- > 1 file changed, 6 insertions(+), 6 deletions(-) > > diff --git a/drivers/gpu/drm/i915/intel_pm.c

Re: [Intel-gfx] [PATCH 12/19] drm/msm: Use new atomic iterator macros

On Mon, Oct 17, 2016 at 02:37:11PM +0200, Maarten Lankhorst wrote: > Signed-off-by: Maarten Lankhorst > --- > drivers/gpu/drm/msm/mdp/mdp4/mdp4_kms.c | 4 ++-- > drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c | 6 +++--- > drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.h

Re: [Intel-gfx] [PATCH i-g-t 3/4 v5] tests/drv_module_reload: Convert sh script to C version.

On Thu, 03 Nov 2016, Marius Vlad wrote: > v5: > - reworked gem_info to gem_sanitychecks (Chris Wilson) > - remove subgroups/subtests for gem_exec_store and gem_sanitycheck > (Chris Wilson) > > v4: > - adjust test to make use of lib/igt_kmod > - replaced SW_FINISH with

[Intel-gfx] ✗ Fi.CI.BAT: warning for series starting with [1/2] drm/i915: Make sure engines are idle during GPU idling in LR mode

== Series Details == Series: series starting with [1/2] drm/i915: Make sure engines are idle during GPU idling in LR mode URL : https://patchwork.freedesktop.org/series/14789/ State : warning == Summary == Series 14789v1 Series without cover letter

Re: [Intel-gfx] [PATCH 1/2] drm/i915: Remove redundant reprobe in i915_drm_resume

On Thu, 2016-11-03 at 16:25 +, Chris Wilson wrote: > On Thu, Nov 03, 2016 at 12:11:55PM -0400, Lyude Paul wrote: > > > > On Thu, 2016-11-03 at 12:11 -0400, Lyude Paul wrote: > > > > > > On Thu, 2016-11-03 at 16:02 +, Chris Wilson wrote: > > > > > > > > > > > > On Thu, Nov 03, 2016 at

Re: [Intel-gfx] [PATCH 14/19] drm/mediatek: Use new atomic iterator macros

On Mon, Oct 17, 2016 at 02:37:13PM +0200, Maarten Lankhorst wrote: > Signed-off-by: Maarten Lankhorst Patches 13-14 Reviewed-by: Ville Syrjälä > --- > drivers/gpu/drm/mediatek/mtk_drm_drv.c | 4 ++-- > 1 file changed, 2

Re: [Intel-gfx] [PATCH 08/12] drm/i915/guc: Cache the client mapping

On 02/11/2016 17:50, Chris Wilson wrote: Use i915_gem_object_pin_map() for the guc client's lifetime to replace the peristent kmap + frequent kmap_atomic with a permanent vmapping. This avoids taking the obj->mm.lock mutex whilst inside irq context later. Signed-off-by: Chris Wilson

Re: [Intel-gfx] [PATCH 07/12] drm/i915/scheduler: Boost priorities for flips

On 02/11/2016 17:50, Chris Wilson wrote: Boost the priority of any rendering required to show the next pageflip as we want to avoid missing the vblank by being delayed by invisible workload. We prioritise avoiding jank and jitter in the GUI over starving background tasks. Signed-off-by: Chris

Re: [Intel-gfx] [PATCH 1/2] drm/i915: Remove redundant reprobe in i915_drm_resume

On Thu, Nov 03, 2016 at 12:11:55PM -0400, Lyude Paul wrote: > On Thu, 2016-11-03 at 12:11 -0400, Lyude Paul wrote: > > On Thu, 2016-11-03 at 16:02 +, Chris Wilson wrote: > > > > > > On Thu, Nov 03, 2016 at 11:42:37AM -0400, Lyude wrote: > > > > > > > > > > > > Weine's investigation on

[Intel-gfx] [PATCH i-g-t 3/4 v5] tests/drv_module_reload: Convert sh script to C version.

v5: - reworked gem_info to gem_sanitychecks (Chris Wilson) - remove subgroups/subtests for gem_exec_store and gem_sanitycheck (Chris Wilson) v4: - adjust test to make use of lib/igt_kmod - replaced SW_FINISH with SET_CACHEING (Chris Wilson) v3: - fix passing boolean value as flags to

[Intel-gfx] [PATCH i-g-t 4/4 v5] tests/kms_sysfs_edid_timing: Convert sh to C version.

v3: - use igt_mean for accounting (Chris Wilson) - make it Intel-agnostic when searching for connectors (Chris Wilson) v2: - don't read cached values (Chris Wilson) - warn on per connector, and fail per mean (Chris Wilson) These are synthetic: 10ms per connector, and 50ms for all.

[Intel-gfx] [PATCH i-g-t 1/4 v5] lib/{igt_sysfs, igt_aux}: Make available to other users kick_fbcon() (unbind_fbcon()), and added helpers to lib/igt_aux, lib/igt_kmod.

lib/igt_aux: Added igt_pkill helper. lib/igt_kmod: Added load/unload kmod helpers. v5: - added igt_i915_driver_{load/unload}. - added kick_snd_hda_intel() to match current tests/drv_module_reload_basic and integrated into igt_i915_driver_load/unload. - added gtk-doc section for lib/igt_kmod v4:

[Intel-gfx] [PATCH i-g-t 2/4 v5] lib/igt_gvt: Make use of libkmod helpers and fix reading gvt parameter.

v2: - use igt_sysfs_get_boolean() to get gvt status (Chris Wilson) - do not hard-fail when i915 module could not be loaded/unloaded (Chris Wilson) Signed-off-by: Marius Vlad --- lib/igt_gvt.c | 37 ++--- tests/gvt_basic.c | 2 +- 2

[Intel-gfx] [PATCH i-g-t 0/4 v5] Convert sh scripts to C variants.

Changes since last version include rebasing to account for modifications in tests/drv_module_reload_basic and address comments. This series adds some library support to help converting sh scripts to C version. Based on that I've converted drv_module_reload_basic and kms_sysfs_edid_timing. Other

Re: [Intel-gfx] [PATCH 07/19] drm/atomic: Fix atomic helpers to use the new iterator macros.

On Mon, Oct 17, 2016 at 02:37:06PM +0200, Maarten Lankhorst wrote: > Signed-off-by: Maarten Lankhorst > --- > drivers/gpu/drm/drm_atomic_helper.c | 302 > +++- > 1 file changed, 157 insertions(+), 145 deletions(-) > > diff

Re: [Intel-gfx] [PATCH 06/12] drm/i915/scheduler: Execute requests in order of priorities

Hi, 1st pass comments only for now. On 02/11/2016 17:50, Chris Wilson wrote: Track the priority of each request and use it to determine the order in which we submit requests to the hardware via execlists. The priority of the request is determined by the user (eventually via the context) but

[Intel-gfx] [PATCH 1/2] drm/i915: Make sure engines are idle during GPU idling in LR mode

We assume that the GPU is idle once receiving the seqno via the last request's user interrupt. In execlist mode the corresponding context completed interrupt can be delayed though and until this latter interrupt arrives we consider the request to be pending on the ELSP submit port. This can cause

[Intel-gfx] [PATCH 2/2] drm/i915: Add assert for no pending GPU requests during resume in LR mode

During resume we will reset the SW/HW tracking for each ring head/tail pointers and so are not prepared to replay any pending requests (as opposed to GPU reset time). Add an assert for this. Cc: Chris Wilson Cc: Mika Kuoppala Signed-off-by:

[Intel-gfx] ✓ Fi.CI.BAT: success for Small fixes to speed up resume

== Series Details == Series: Small fixes to speed up resume URL : https://patchwork.freedesktop.org/series/14787/ State : success == Summary == Series 14787v1 Small fixes to speed up resume https://patchwork.freedesktop.org/api/1.0/series/14787/revisions/1/mbox/ fi-bdw-5557u total:241

Re: [Intel-gfx] [PATCH 1/2] drm/i915: Remove redundant reprobe in i915_drm_resume

On Thu, 2016-11-03 at 12:11 -0400, Lyude Paul wrote: > On Thu, 2016-11-03 at 16:02 +, Chris Wilson wrote: > > > > On Thu, Nov 03, 2016 at 11:42:37AM -0400, Lyude wrote: > > > > > > > > > Weine's investigation on benchmarking the suspend/resume process pointed > > > out a lot of the time in

Re: [Intel-gfx] [PATCH 1/2] drm/i915: Remove redundant reprobe in i915_drm_resume

On Thu, 2016-11-03 at 16:02 +, Chris Wilson wrote: > On Thu, Nov 03, 2016 at 11:42:37AM -0400, Lyude wrote: > > > > Weine's investigation on benchmarking the suspend/resume process pointed > > out a lot of the time in suspend/resume is being spent reprobing. While > > the reprobing process is

Re: [Intel-gfx] [PATCH 1/2] drm/i915: Remove redundant reprobe in i915_drm_resume

On Thu, Nov 03, 2016 at 11:42:37AM -0400, Lyude wrote: > Weine's investigation on benchmarking the suspend/resume process pointed > out a lot of the time in suspend/resume is being spent reprobing. While > the reprobing process is a lengthy one for good reason, we don't need to > hold up the

Re: [Intel-gfx] [PATCH 1/2] Revert "drm: Add and handle new aspect ratios in DRM layer"

On Thu, Nov 3, 2016 at 1:47 PM, Sharma, Shashank wrote: > Hi Ville, > (-dri-devel) > I would appreciate an internal discussion before going to dri-devel. Nack on internal discssion. We do development in the open. > What did this patch break ? > This is exposed in the

Re: [Intel-gfx] [PATCH 06/19] drm/atomic: Use new atomic iterator macros.

On Mon, Oct 17, 2016 at 02:37:05PM +0200, Maarten Lankhorst wrote: > Signed-off-by: Maarten Lankhorst > --- > drivers/gpu/drm/drm_atomic.c | 10 +- > 1 file changed, 5 insertions(+), 5 deletions(-) > > diff --git a/drivers/gpu/drm/drm_atomic.c

[Intel-gfx] [PATCH 0/2] Small fixes to speed up resume

Recently David Weinehall has been investigating how we can make the resume process in i915 faster, and poked me to take a look at why we were taking so long while reprobing. As it turns out the main issue is just that we hold up the entire resume process while we reprobe connectors, which isn't

[Intel-gfx] [PATCH 1/2] drm/i915: Remove redundant reprobe in i915_drm_resume

Weine's investigation on benchmarking the suspend/resume process pointed out a lot of the time in suspend/resume is being spent reprobing. While the reprobing process is a lengthy one for good reason, we don't need to hold up the entire suspend/resume process while we wait for it to finish.

[Intel-gfx] [PATCH 2/2] drm/i915: Reinit polling before hpd when resuming

Now that we don't run the connector reprobing from i915_drm_resume(), we need to make it so we don't have to wait for reprobing to finish so that we actually speed things up. In order to do this, we need to make sure that i915_drm_resume() doesn't get blocked by i915_hpd_poll_init_work() while

Re: [Intel-gfx] [PATCH 05/19] drm/atomic: Make add_affected_connectors look at crtc_state.

On Mon, Oct 17, 2016 at 02:37:04PM +0200, Maarten Lankhorst wrote: > This kills another dereference of connector->state. connector_mask > holds all unchanged connectors at least and any changed connectors > are already in state anyway. > > Signed-off-by: Maarten Lankhorst

Re: [Intel-gfx] [PATCH 04/19] drm/blend: Use new atomic iterator macros.

On Mon, Oct 17, 2016 at 02:37:03PM +0200, Maarten Lankhorst wrote: > Signed-off-by: Maarten Lankhorst Patches 02-04 looks sane to me, so for them Reviewed-by: Ville Syrjälä > --- > drivers/gpu/drm/drm_blend.c | 22

Re: [Intel-gfx] [PATCH 01/19] drm/atomic: Add new iterators over all state

On Wed, Nov 02, 2016 at 09:28:46AM +0100, Maarten Lankhorst wrote: > Op 01-11-16 om 14:41 schreef Ville Syrjälä: > > On Tue, Nov 01, 2016 at 02:34:00PM +0100, Maarten Lankhorst wrote: > >> Op 01-11-16 om 14:09 schreef Ville Syrjälä: > >>> On Mon, Oct 17, 2016 at 02:37:00PM +0200, Maarten Lankhorst

Re: [Intel-gfx] [PATCH v3 1/3] lib: add igt_dummyload

On Thu, Nov 03, 2016 at 11:40:36AM +0200, Abdiel Janulgue wrote: > A lot of igt testcases need some GPU workload to make sure a race > window is big enough. Unfortunately having a fixed amount of > workload leads to spurious test failures or overtly long runtimes > on some fast/slow platforms.

Re: [Intel-gfx] ✗ Fi.CI.BAT: warning for drm/i915: Tidy slab cache allocations (rev2)

On 02/11/2016 16:45, Patchwork wrote: == Series Details == Series: drm/i915: Tidy slab cache allocations (rev2) URL : https://patchwork.freedesktop.org/series/14731/ State : warning == Summary == Series 14731v2 drm/i915: Tidy slab cache allocations

Re: [Intel-gfx] [PATCH 03/12] drm/i915: Remove engine->execlist_lock

On 03/11/2016 12:28, Chris Wilson wrote: On Thu, Nov 03, 2016 at 10:47:54AM +, Tvrtko Ursulin wrote: On 02/11/2016 17:50, Chris Wilson wrote: @@ -600,9 +598,8 @@ static void intel_lrc_irq_handler(unsigned long data) static void execlists_submit_request(struct drm_i915_gem_request

Re: [Intel-gfx] [PATCH 09/12] HACK drm/i915/scheduler: emulate a scheduler for guc

On 02/11/2016 17:50, Chris Wilson wrote: This emulates execlists on top of the GuC in order to defer submission of requests to the hardware. This deferral allows time for high priority requests to gazump their way to the head of the queue, however it nerfs the GuC by converting it back into a

Re: [Intel-gfx] [PATCH 03/12] drm/i915: Remove engine->execlist_lock

On Thu, Nov 03, 2016 at 10:47:54AM +, Tvrtko Ursulin wrote: > > On 02/11/2016 17:50, Chris Wilson wrote: > >@@ -600,9 +598,8 @@ static void intel_lrc_irq_handler(unsigned long data) > > static void execlists_submit_request(struct drm_i915_gem_request *request) > > { > > struct

Re: [Intel-gfx] [PATCH 05/12] drm/i915/scheduler: Record all dependencies upon request construction

On Thu, Nov 03, 2016 at 11:03:47AM +, Tvrtko Ursulin wrote: > > On 02/11/2016 17:50, Chris Wilson wrote: > >The scheduler needs to know the dependencies of each request for the > >lifetime of the request, as it may choose to reschedule the requests at > >any time and must ensure the

Re: [Intel-gfx] [PATCH 02/12] drm/i915: Defer transfer onto execution timeline to actual hw submission

On Thu, Nov 03, 2016 at 10:51:14AM +, Chris Wilson wrote: > Yes. Worse is that the 2 comes from having different paths to this point > with their own nesting pattern. Not any more, that was a leftover from one version that managed to nest signaling/execution. The issue is

Re: [Intel-gfx] [PATCH 05/12] drm/i915/scheduler: Record all dependencies upon request construction

On 02/11/2016 17:50, Chris Wilson wrote: The scheduler needs to know the dependencies of each request for the lifetime of the request, as it may choose to reschedule the requests at any time and must ensure the dependency tree is not broken. This is in additional to using the fence to only

Re: [Intel-gfx] ✓ Fi.CI.BAT: success for drm/i915: Introduce HAS_64BIT_RELOC (rev3)

On to, 2016-11-03 at 09:16 +, Patchwork wrote: > == Series Details == > > Series: drm/i915: Introduce HAS_64BIT_RELOC (rev3) > URL   : https://patchwork.freedesktop.org/series/14730/ > State : success > > == Summary == Committing the patch, thanks for the review. As discussed in IRC, the

Re: [Intel-gfx] [PATCH 02/12] drm/i915: Defer transfer onto execution timeline to actual hw submission

On Thu, Nov 03, 2016 at 10:34:26AM +, Tvrtko Ursulin wrote: > > On 02/11/2016 17:50, Chris Wilson wrote: > >Defer the transfer from the client's timeline onto the execution > >timeline from the point of readiness to the point of actual submission. > >For example, in execlists, a request is

Re: [Intel-gfx] [PATCH 04/12] drm/i915/scheduler: Signal the arrival of a new request

On 02/11/2016 17:50, Chris Wilson wrote: The start of the scheduler, add a hook into request submission for the scheduler to see the arrival of new requests and prepare its runqueues. Signed-off-by: Chris Wilson --- drivers/gpu/drm/i915/i915_drv.c | 4

Re: [Intel-gfx] [PATCH 03/12] drm/i915: Remove engine->execlist_lock

On 02/11/2016 17:50, Chris Wilson wrote: The execlist_lock is now completely subsumed by the engine->timeline->lock, and so we can remove the redundant layer of locking. Signed-off-by: Chris Wilson --- drivers/gpu/drm/i915/i915_debugfs.c | 4 ++--

Re: [Intel-gfx] [PATCH 01/12] drm/i915: Split request submit/execute phase into two

On 02/11/2016 17:50, Chris Wilson wrote: In order to support deferred scheduling, we need to differentiate between when the request is ready to run (i.e. the submit fence is signaled) and when the request is actually run (a new execute fence). This is typically split between the request itself

Re: [Intel-gfx] [PATCH 02/12] drm/i915: Defer transfer onto execution timeline to actual hw submission

On 02/11/2016 17:50, Chris Wilson wrote: Defer the transfer from the client's timeline onto the execution timeline from the point of readiness to the point of actual submission. For example, in execlists, a request is finally submitted to hardware when the hardware is ready, and only put onto

Re: [Intel-gfx] [PATCH] drm/i915: Fix pages pin counting around swizzle quirk

On Wed, Nov 02, 2016 at 09:43:54AM +, Chris Wilson wrote: > commit bc0629a76726 ("drm/i915: Track pages pinned due to swizzling > quirk") fixed one problem, but revealed a whole lot more. The root cause > of the pin count mismatch for the swizzle quirk (for L-shaped memory on > gen3/4) was

[Intel-gfx] [PATCH v3 3/3] igt/kms_flip: Use new igt_spin_batch

Cc: Chris Wilson Cc: Daniel Vetter Signed-off-by: Abdiel Janulgue --- tests/kms_flip.c | 185 ++- 1 file changed, 4 insertions(+), 181 deletions(-) diff --git

[Intel-gfx] [PATCH v3 2/3] igt/gem_wait: Use new igt_spin_batch

Cc: Chris Wilson Cc: Daniel Vetter Signed-off-by: Abdiel Janulgue --- tests/gem_wait.c | 125 --- 1 file changed, 7 insertions(+), 118 deletions(-) diff --git

[Intel-gfx] [PATCH v3 1/3] lib: add igt_dummyload

A lot of igt testcases need some GPU workload to make sure a race window is big enough. Unfortunately having a fixed amount of workload leads to spurious test failures or overtly long runtimes on some fast/slow platforms. This library contains functionality to submit GPU workloads that should

Re: [Intel-gfx] [PATCH 0/7] drm/i915: Add pipe scaler for Gen9 in atomic path

Hi, Please review the patches and comments. On 10/28/2016 6:05 PM, Nabendu Maiti wrote: Attaching old discussion thread for easy reference. On Tue, Jan 05, 2016 at 05:18:40PM +0200, Ville Syrjälä wrote: On Tue, Jan 05, 2016 at 03:50:38PM +0100, Daniel Vetter wrote: > On Mon, Jan 04,

Re: [Intel-gfx] [RFC PATCH] get_maintainer: Look for arbitrary letter prefixes in sections

On Thu, 2016-11-03 at 02:16 -0700, Joe Perches wrote: > Yes, it's been reported and should be fixed in -mm. > The fix should show up in -next in a little bit. Great. > For now, try: > $ sed -i -e 's/\xA0/ /g' scripts/get_maintainer.pl Thanks, Paul Bolle

Re: [Intel-gfx] [RFC PATCH] get_maintainer: Look for arbitrary letter prefixes in sections

On Thu, 2016-11-03 at 10:07 +0100, Paul Bolle wrote: > On Mon, 2016-10-24 at 11:05 -0700, Joe Perches wrote: > > Jani Nikula proposes patches to add a few new letter prefixes > > for "B:" bug reporting and "C:" maintainer chatting to the > > various sections of MAINTAINERS. > > > > Add a generic

[Intel-gfx] ✓ Fi.CI.BAT: success for drm/i915: Introduce HAS_64BIT_RELOC (rev3)

== Series Details == Series: drm/i915: Introduce HAS_64BIT_RELOC (rev3) URL : https://patchwork.freedesktop.org/series/14730/ State : success == Summary == Series 14730v3 drm/i915: Introduce HAS_64BIT_RELOC https://patchwork.freedesktop.org/api/1.0/series/14730/revisions/3/mbox/ Test

Re: [Intel-gfx] [RFC PATCH] get_maintainer: Look for arbitrary letter prefixes in sections

On Mon, 2016-10-24 at 11:05 -0700, Joe Perches wrote: > Jani Nikula proposes patches to add a few new letter prefixes > for "B:" bug reporting and "C:" maintainer chatting to the > various sections of MAINTAINERS. > > Add a generic mechanism to get_maintainer.pl to find sections that > have any

Re: [Intel-gfx] [PATCH] drm/i915/sw_fence: Replace private use of wq->flags with bit zero in wq->private

On Thu, Nov 03, 2016 at 08:33:58AM +, Tvrtko Ursulin wrote: > On 02/11/2016 17:34, Chris Wilson wrote: > >It looks ok, but I just don't see the point. wq->flags is private to the > >wq->func callback. > > A very superficial skim shows that wake_up_common at least looks at > the flags. So I

Re: [Intel-gfx] [PATCH v3] drm/i915: Introduce HAS_64BIT_RELOC

On Thu, Nov 03, 2016 at 10:39:46AM +0200, Joonas Lahtinen wrote: > Move has_64bit_reloc into dev_priv->info. This will make it visible > in the feature listing debug output. > > v2: > - Keep the struct member to keep GCC fragile but happy (Chris) > v3: > - More detailed commit message (Chris) > -

[Intel-gfx] [PATCH v3] drm/i915: Introduce HAS_64BIT_RELOC

Move has_64bit_reloc into dev_priv->info. This will make it visible in the feature listing debug output. v2: - Keep the struct member to keep GCC fragile but happy (Chris) v3: - More detailed commit message (Chris) - Include forgotten CHV and BXT (Chris) Cc: Chris Wilson

Re: [Intel-gfx] [PATCH] drm/i915/sw_fence: Replace private use of wq->flags with bit zero in wq->private

On 02/11/2016 17:34, Chris Wilson wrote: On Wed, Nov 02, 2016 at 05:00:28PM +, Tvrtko Ursulin wrote: From: Tvrtko Ursulin Use of an un-allocated bit in flags is making me nervous so I thought to use the bit zero of the private pointer instead. That should be