From: "Pandiyan, Dhinakaran"
The total vcpi time slots is always 63 and does not depend on the link BW,
remove total_slots from MST topology manager struct. The next change is to
remove total_pbn which is hardcoded to 2560. The total PBN that the
topology manager allocates from depends on the lin
From: "Pandiyan, Dhinakaran"
The avail_slots member in the MST topology manager is never updated to
reflect the available vcpi slots. The check is effectively against
total slots, 63. So, let's make that check obvious and remove
avail_slots. While at it, make debug messages more descriptive.
Cc:
From: "Pandiyan, Dhinakaran"
Having a ->atomic_release callback is useful to release shared resources
that get allocated in compute_config(). This function is expected to be
called in the atomic_check() phase before new resources are acquired.
v3: Use the new 'for_each_oldnew_connector_in_state(
From: "Pandiyan, Dhinakaran"
drm_dp_atomic_find_vcpi_slots() should be called from ->atomic_check() to
check there are sufficient vcpi slots for a mode and to add that to the
state. This should be followed by a call to drm_dp_mst_allocate_vcpi()
in ->atomic_commit() to initialize a struct vcpi fo
From: "Pandiyan, Dhinakaran"
It is necessary to track states for objects other than connector, crtc
and plane for atomic modesets. But adding objects like DP MST link
bandwidth to drm_atomic_state would mean that a non-core object will be
modified by the core helper functions for swapping and cle
From: "Pandiyan, Dhinakaran"
drm_dp_mst_allocate_vcpi() apart from setting up the vcpi structure,
also finds if there are enough slots available. This check is a duplicate
of that implemented in drm_dp_mst_find_vcpi_slots(). Let's move this check
out and reuse the existing drm_dp_mst_find_vcpi_sl
From: "Pandiyan, Dhinakaran"
Use the added helpers to track MST link bandwidth for atomic modesets.
Link bw is acquired in the ->atomic_check() phase when CRTCs are being
enabled with drm_atomic_find_vcpi_slots() instead of drm_find_vcpi_slots().
Similarly, link bw is released during ->atomic_che
Link bandwidth is a shared resource between multiple displays in DP MST
configurations. For atomic modesetting drivers, checking if there is
sufficient link bandwidth for a mode needs to be done during the
atomic_check phase to avoid failed modesets when multiple CRTC's and
connectors are involved.
From: "Pandiyan, Dhinakaran"
Link bandwidth is shared between multiple display streams in DP MST
configurations. The DP MST topology manager structure maintains the
shared link bandwidth for a primary link directly connected to the GPU. For
atomic modesetting drivers, checking if there is suffici
== Series Details ==
Series: Adding driver-private objects to atomic state (rev3)
URL : https://patchwork.freedesktop.org/series/19355/
State : success
== Summary ==
Series 19355v3 Adding driver-private objects to atomic state
https://patchwork.freedesktop.org/api/1.0/series/19355/revisions/3/
Hi Greg,
On Mon, Mar 13, 2017 at 07:40:50AM +0100, Daniel Vetter wrote:
> On Sun, Mar 12, 2017 at 11:01 PM, Greg KH wrote:
> > So if a commit says "cherry-pick", I guess I can always assume it's safe
> > to add, right? If not, _then_ I have to run the "search backwards"
> > logic, right?
> >
> >
On Wed, 15 Mar 2017, Chris Wilson wrote:
> On Wed, Mar 15, 2017 at 05:09:04PM +0200, ville.syrj...@linux.intel.com wrote:
>> From: Ville Syrjälä
>>
>> Add a command for extracting various tags (eg. Reviwed-by:) from
>> emails. You can give the comamnd a rangeish to add the tags from
>> the same
On Thu, Mar 16, 2017 at 11:50:06AM +0530, sourab.gu...@intel.com wrote:
> From: Sourab Gupta
>
> This patch extends the i915 perf framework to handle the perf sample
> collection for any given gpu engine. Particularly, the support
> for collecting timestamp sample type is added, which can be req
On Thu, Mar 16, 2017 at 11:50:11AM +0530, sourab.gu...@intel.com wrote:
> +static int copy_mmio_list(struct drm_i915_private *dev_priv,
> + void __user *mmio)
> +{
> + void __user *mmio_list = ((u8 __user *)mmio + 4);
> + u32 num_mmio;
> + int ret;
> +
> +
On Wed, 15 Mar 2017, ville.syrj...@linux.intel.com wrote:
> From: Ville Syrjälä
>
> Currently ILK-BDW explicitly disable LP1+ watermarks from their
> .init_clock_gating() hooks. Unfortunately that hook gets called way too
> late since by that time we've already initialized all the watermark
> stat
On Thu, 16 Mar 2017, sourab.gu...@intel.com wrote:
Please Cc LKML for patches which touch subsystems managed on LKML. Also
CC'ing people on the 0/N letter would be helpful, so we can see what this
is about.
> For the drivers to be able to use the cross timestamp framework,
> they need the inform
On Thu, Mar 16, 2017 at 11:44:10AM +0530, sourab.gu...@intel.com wrote:
> @@ -3593,6 +3670,7 @@ void i915_oa_init_reg_state(struct intel_engine_cs
> *engine,
> void i915_oa_update_reg_state(struct intel_engine_cs *engine,
> struct i915_gem_context *ctx,
>
On Thu, Mar 16, 2017 at 11:44:15AM +0530, sourab.gu...@intel.com wrote:
> From: Sourab Gupta
>
> This patch enables userspace to specify tags (per workload), provided via
> execbuffer ioctl, which could be added to OA reports, to help associate
> reports with the corresponding workloads.
>
> The
On Thu, Mar 16, 2017 at 11:44:08AM +0530, sourab.gu...@intel.com wrote:
> From: Sourab Gupta
>
> This patch adds a new ctx getparam ioctl parameter, which can be used to
> retrieve ctx unique id by userspace.
>
> This can be used by userspace to map the OA reports received in the
> i915 perf sam
On Thu, Mar 16, 2017 at 12:10:30AM -0700, Dhinakaran Pandiyan wrote:
> From: "Pandiyan, Dhinakaran"
>
> Having a ->atomic_release callback is useful to release shared resources
> that get allocated in compute_config(). This function is expected to be
> called in the atomic_check() phase before ne
On Thu, 2017-03-16 at 00:49 -0700, Chris Wilson wrote:
> On Thu, Mar 16, 2017 at 11:50:06AM +0530, sourab.gu...@intel.com wrote:
> > From: Sourab Gupta
> >
> > This patch extends the i915 perf framework to handle the perf sample
> > collection for any given gpu engine. Particularly, the support
On Thu, Mar 16, 2017 at 11:44:10AM +0530, sourab.gu...@intel.com wrote:
> +static void i915_perf_command_stream_hook_oa(struct i915_perf_stream *stream,
> + struct drm_i915_gem_request *request)
> +{
> + struct drm_i915_private *dev_priv = request->i915;
> +
On 15/03/2017 22:22, Chris Wilson wrote:
Tvrtko spotted a stale reference to b->lock (now b->rb_lock) so review
the comments and try to improve them in passing.
Signed-off-by: Chris Wilson
Cc: Tvrtko Ursulin
Cc: Mika Kuoppala
---
drivers/gpu/drm/i915/intel_breadcrumbs.c | 25 +++
On Wed, Mar 15, 2017 at 09:23:30AM +0100, Maarten Lankhorst wrote:
> Op 14-03-17 om 07:50 schreef Daniel Vetter:
> > On Mon, Mar 13, 2017 at 12:08:19PM +0100, Maarten Lankhorst wrote:
> >> Op 13-03-17 om 11:15 schreef Daniel Vetter:
> >>> On Thu, Mar 09, 2017 at 03:52:04PM +0100, Maarten Lankhorst
On Wed, Mar 15, 2017 at 01:51:36PM +0200, Jani Nikula wrote:
> Helper to rebase a branch onto drm-tip.
>
> Signed-off-by: Jani Nikula
I like. Ack.
-Daniel
> ---
> dim | 23 +++
> dim.rst | 6 ++
> 2 files changed, 29 insertions(+)
>
> diff --git a/dim b/dim
> inde
On Thu, Mar 16, 2017 at 08:40:04AM +, Tvrtko Ursulin wrote:
>
> On 15/03/2017 22:22, Chris Wilson wrote:
> >Tvrtko spotted a stale reference to b->lock (now b->rb_lock) so review
> >the comments and try to improve them in passing.
> >
> >Signed-off-by: Chris Wilson
> >Cc: Tvrtko Ursulin
> >C
On Thu, 2017-03-16 at 01:10 -0700, Chris Wilson wrote:
> On Thu, Mar 16, 2017 at 11:44:10AM +0530, sourab.gu...@intel.com wrote:
> > @@ -3593,6 +3670,7 @@ void i915_oa_init_reg_state(struct intel_engine_cs
> > *engine,
> > void i915_oa_update_reg_state(struct intel_engine_cs *engine,
> >
On 15/03/2017 14:17, Patchwork wrote:
== Series Details ==
Series: drm/i915/uc: Rename intel_uc_fw.fw to .type
URL : https://patchwork.freedesktop.org/series/21285/
State : success
== Summary ==
Series 21285v1 drm/i915/uc: Rename intel_uc_fw.fw to .type
https://patchwork.freedesktop.org/api
On Thu, 2017-03-16 at 01:31 -0700, Chris Wilson wrote:
> On Thu, Mar 16, 2017 at 11:44:10AM +0530, sourab.gu...@intel.com wrote:
> > +static void i915_perf_command_stream_hook_oa(struct i915_perf_stream
> > *stream,
> > + struct drm_i915_gem_request *request)
> >
On 15/03/2017 13:37, Michal Wajdeczko wrote:
After negative guc fw selection we could leave guc
submission flag still turned on. Reorder some checks
to cover this case. While here, fix info message and
return early if there is no Guc.
Signed-off-by: Michal Wajdeczko
Cc: Arkadiusz Hiler
Cc: Jo
On Thu, Mar 16, 2017 at 02:24:55PM +0530, sourab gupta wrote:
> On Thu, 2017-03-16 at 01:10 -0700, Chris Wilson wrote:
> > On Thu, Mar 16, 2017 at 11:44:10AM +0530, sourab.gu...@intel.com wrote:
> > > @@ -3593,6 +3670,7 @@ void i915_oa_init_reg_state(struct intel_engine_cs
> > > *engine,
> > > vo
On Wed, Mar 15, 2017 at 08:40:26PM +, Chris Wilson wrote:
> Avoid adding to the waitqueue and reprobing the current vblank if the
> caller is only querying the current vblank sequence and timestamp, where
> we know that the wait would return immediately.
>
> v2: Add CRTC identifier to debug me
Chris Wilson writes:
> Compute the offset of the end of the crc32 field using offsetofend()
> rather than open-coding.
>
> Signed-off-by: Chris Wilson
> Cc: Zhenyu Wang
> Cc: Zhi Wang
Reviewed-by: Mika Kuoppala
> ---
> drivers/gpu/drm/i915/gvt/firmware.c | 2 +-
> 1 file changed, 1 inserti
On Thu, Mar 16, 2017 at 09:40:05AM +0200, Jani Nikula wrote:
> On Wed, 15 Mar 2017, Chris Wilson wrote:
> > On Wed, Mar 15, 2017 at 05:09:04PM +0200, ville.syrj...@linux.intel.com
> > wrote:
> >> From: Ville Syrjälä
> >>
> >> Add a command for extracting various tags (eg. Reviwed-by:) from
> >>
On Thu, 2017-03-16 at 02:03 -0700, Chris Wilson wrote:
> On Thu, Mar 16, 2017 at 02:24:55PM +0530, sourab gupta wrote:
> > On Thu, 2017-03-16 at 01:10 -0700, Chris Wilson wrote:
> > > On Thu, Mar 16, 2017 at 11:44:10AM +0530, sourab.gu...@intel.com wrote:
> > > > @@ -3593,6 +3670,7 @@ void i915_oa_
On 15/03/2017 14:01, Chris Wilson wrote:
lockdep doesn't like us taking the mm->mmap_sem inside the get_pages
callback for a couple of reasons. The straightforward deadlock:
[13755.434059] =
[13755.434061] [ INFO: possible recursive locking detected ]
On Thu, Mar 16, 2017 at 03:22:03PM +0530, sourab gupta wrote:
> On Thu, 2017-03-16 at 02:03 -0700, Chris Wilson wrote:
> > On Thu, Mar 16, 2017 at 02:24:55PM +0530, sourab gupta wrote:
> > > On Thu, 2017-03-16 at 01:10 -0700, Chris Wilson wrote:
> > > > On Thu, Mar 16, 2017 at 11:44:10AM +0530, sou
> -Original Message-
> From: Nikula, Jani
> Sent: Thursday, February 16, 2017 9:03 PM
> To: Chauhan, Madhav ; intel-
> g...@lists.freedesktop.org
> Cc: Conselvan De Oliveira, Ander ;
> Shankar, Uma ; Mukherjee, Indranil
> ; Sharma, Shashank
> ; Chauhan, Madhav
> ; ville.syrj...@linux.intel.
On Wed, Mar 15, 2017 at 11:08:56AM +0100, Maarten Lankhorst wrote:
> As a proof of concept, first try to convert intel_tv, which is a rarely
> used connector. It has 5 properties, tv format and 4 margins.
>
> I'm less certain about the state behavior itself, should we pass a size
> parameter to in
On Wed, Mar 15, 2017 at 09:09:31AM -0700, Jason Ekstrand wrote:
> +#ifndef DRM_IOCTL_I915_GEM_EXECBUFFER2_WR
> +#define DRM_IOCTL_I915_GEM_EXECBUFFER2_WRDRM_IOWR(DRM_COMMAND_BASE +
> DRM_I915_GEM_EXECBUFFER2_WR, struct drm_i915_gem_execbuffer2)
> +#endif
If DRM_IOCTL_I915_GEM_EXECBUFFER2_WR i
On ke, 2017-03-15 at 11:49 +, Chris Wilson wrote:
> > From: Michal Wajdeczko
>
> Manual pointer manipulation is error prone. Let compiler calculate
> right offsets for us in case we need to change ads layout.
>
> v2: don't call it object (Chris)
> v3: restyle offset assignments (Chris)
> v4:
Chris Wilson writes:
> On Wed, Mar 15, 2017 at 05:42:59PM +0200, Mika Kuoppala wrote:
>> Plan is to make generic residency calculation utility
>> function for usage outside of sysfs. As a first step
>> move residency calculation into intel_pm.c
>>
>> Signed-off-by: Mika Kuoppala
> Reviewed-by:
guc_addon_create() makes the assumption that it need only kmap the
initial page in order to write all of the configuration data used by the
guc. Confusingly it also allocates many scratch pages in the same vma
and passes that to the guc. Reassure the reader that all is well with a
BUILD_BUG_ON() th
On Thu, Mar 16, 2017 at 10:55:06AM +, Chris Wilson wrote:
> guc_addon_create() makes the assumption that it need only kmap the
> initial page in order to write all of the configuration data used by the
> guc. Confusingly it also allocates many scratch pages in the same vma
> and passes that to
On Thu, Mar 16, 2017 at 10:58:20AM +, Chris Wilson wrote:
> On Thu, Mar 16, 2017 at 10:55:06AM +, Chris Wilson wrote:
> > guc_addon_create() makes the assumption that it need only kmap the
> > initial page in order to write all of the configuration data used by the
> > guc. Confusingly it a
On Thu, Mar 16, 2017 at 12:06:20PM +0100, Michal Wajdeczko wrote:
> On Thu, Mar 16, 2017 at 10:58:20AM +, Chris Wilson wrote:
> > On Thu, Mar 16, 2017 at 10:55:06AM +, Chris Wilson wrote:
> > > guc_addon_create() makes the assumption that it need only kmap the
> > > initial page in order to
On Thu, Mar 16, 2017 at 09:56:19AM +, Tvrtko Ursulin wrote:
> >Fixes: 1c8782dd313e ("drm/i915/userptr: Disallow wrapping GTT into a
> >userptr")
> >Testcase: igt/gem_userptr_blits/map-fixed-invalidate-gup
> >Testcase: igt/gem_userptr_blits/dmabuf-sync
> >Signed-off-by: Chris Wilson
> >Cc: Mic
On Thu, Mar 16, 2017 at 11:18:12AM +, Chris Wilson wrote:
> On Thu, Mar 16, 2017 at 12:06:20PM +0100, Michal Wajdeczko wrote:
> > On Thu, Mar 16, 2017 at 10:58:20AM +, Chris Wilson wrote:
> > > On Thu, Mar 16, 2017 at 10:55:06AM +, Chris Wilson wrote:
> > > > guc_addon_create() makes th
Hi Ville,
On Wednesday 15 March 2017 03:53 PM, Ville Syrjälä wrote:
On Wed, Mar 15, 2017 at 12:29:35PM +0530, Praveen Paneri wrote:
FBC is only enabled for X-tiled framebuffers but there are
quite a few cases where we tend to use Y-tiled framebuffers.
So enabling it for non X-tiled framebuffers
guc_addon_create() makes the assumption that it need only kmap the
initial page in order to write all of the configuration data used by the
guc. Confusingly it also allocates many scratch pages in the same vma
and passes that to the guc. Reassure the reader that all is well with a
BUILD_BUG_ON() th
== Series Details ==
Series: drm/i915/guc: Document that the ads blob entries only lie within the
first page
URL : https://patchwork.freedesktop.org/series/21372/
State : success
== Summary ==
Series 21372v1 drm/i915/guc: Document that the ads blob entries only lie within
the first page
http
== Series Details ==
Series: drm/i915/guc: Document that the ads blob entries only lie within the
first page (rev2)
URL : https://patchwork.freedesktop.org/series/21372/
State : failure
== Summary ==
Series 21372v2 drm/i915/guc: Document that the ads blob entries only lie within
the first pa
When FBC is enabled for linear, legacy Y-tiled and Yf-tiled
surfaces on gen9, the cfb stride must be programmed by SW as
cfb_stride = ceiling[(at least plane width in pixels)/
(32 * compression limit factor)] * 8
v2: Minor fix for a build error
v3: Fixed subject, register na
Op 16-03-17 om 11:48 schreef Daniel Vetter:
> On Wed, Mar 15, 2017 at 11:08:56AM +0100, Maarten Lankhorst wrote:
>> As a proof of concept, first try to convert intel_tv, which is a rarely
>> used connector. It has 5 properties, tv format and 4 margins.
>>
>> I'm less certain about the state behavio
== Series Details ==
Series: drm/i915: Enable FBC for non X-tiled FBs (rev3)
URL : https://patchwork.freedesktop.org/series/21264/
State : success
== Summary ==
Series 21264v3 drm/i915: Enable FBC for non X-tiled FBs
https://patchwork.freedesktop.org/api/1.0/series/21264/revisions/3/mbox/
Tes
On Thu, Mar 16, 2017 at 11:41:51AM +, Chris Wilson wrote:
> guc_addon_create() makes the assumption that it need only kmap the
> initial page in order to write all of the configuration data used by the
> guc. Confusingly it also allocates many scratch pages in the same vma
> and passes that to
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 simple execlist (where the CPU has
to wake up
---
drivers/gpu/drm/i915/i915_params.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/i915/i915_params.c
b/drivers/gpu/drm/i915/i915_params.c
index b6a7e363d076..9dcc8a0dd3c0 100644
--- a/drivers/gpu/drm/i915/i915_params.c
+++ b/drivers/gpu/drm/i915/i915_p
On Thu, Mar 16, 2017 at 6:14 AM, wrote:
> From: Sourab Gupta
>
> This series adds framework for collection of OA reports associated with the
> render command stream, which are collected around batchbuffer boundaries.
>
> Refloating the series rebased on Robert's latest patch set for
> 'Enabling
On pe, 2017-03-10 at 08:28 -0800, Oscar Mateo wrote:
> Starting with intel_guc_loader, down to intel_guc_submission
> and finally to intel_guc_log.
>
> v2:
> - Null execbuf client outside guc_client_free (Daniele)
> - Assert if things try to get allocated twice (Daniele/Joonas)
> - Null guc-
On Thu, 2017-03-16 at 03:09 -0700, Chris Wilson wrote:
> On Thu, Mar 16, 2017 at 03:22:03PM +0530, sourab gupta wrote:
> > On Thu, 2017-03-16 at 02:03 -0700, Chris Wilson wrote:
> > > On Thu, Mar 16, 2017 at 02:24:55PM +0530, sourab gupta wrote:
> > > > On Thu, 2017-03-16 at 01:10 -0700, Chris Wils
On Thu, 16 Mar 2017, "Chauhan, Madhav" wrote:
>> -Original Message-
>> From: Nikula, Jani
>> Sent: Thursday, February 16, 2017 9:03 PM
>> To: Chauhan, Madhav ; intel-
>> g...@lists.freedesktop.org
>> Cc: Conselvan De Oliveira, Ander ;
>> Shankar, Uma ; Mukherjee, Indranil
>> ; Sharma, Shas
Since obj->active_count is only updated upon retirement, if we see an
active object in the batch pool, double check that is still active
before deciding to allocate a new object.
Signed-off-by: Chris Wilson
---
drivers/gpu/drm/i915/i915_gem_batch_pool.c | 37 ++
1 fil
The major scaling bottleneck in execbuffer is the processing of the
execobjects. Creating an auxiliary list is inefficient when compared to
using the execobject array we already have allocated.
Reservation is then split into phases. As we lookup up the VMA, we
try and bind it back into active loca
The advent of full-ppgtt lead to an extra indirection between the object
and its binding. That extra indirection has a noticeable impact on how
fast we can convert from the user handles to our internal vma for
execbuffer. In order to bypass the extra indirection, we use a
resizable hashtable to jum
Introduce a new execobject.flag (EXEC_OBJECT_CAPTURE) that userspace may
use to indicate that it wants the contents of this buffer preserved in
the error state (/sys/class/drm/cardN/error) following a GPU hang
involving this batch.
Use this at your discretion, the contents of the error state. alth
Combine the two slightly overlapping parameter structures we pass around
the execbuffer routines into one.
Signed-off-by: Chris Wilson
---
drivers/gpu/drm/i915/i915_gem_execbuffer.c | 548 -
1 file changed, 232 insertions(+), 316 deletions(-)
diff --git a/drivers/gpu
When choosing a slot for an execbuffer, we ideally want to use the same
address as last time (so that we don't have to rebind it) and the same
address as expected by the user (so that we don't have to fixup any
relocations pointing to it). If we first try to bind the incoming
execbuffer->offset fro
Currently the vma has one link member that is used for both holding its
place in the execbuf reservation list, and in any eviction list. This
dual property is quite tricky and error prone.
Signed-off-by: Chris Wilson
Reviewed-by: Mika Kuoppala
---
drivers/gpu/drm/i915/i915_gem_evict.c | 14
Same series as before (and so on...), resent to keep Joonas's inbox full.
-Chris
___
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
This has the benefit of not requiring us to manipulate the
vma->exec_link list when tearing down the execbuffer, and is a
marginally cheaper test to detect the user error.
Signed-off-by: Chris Wilson
---
drivers/gpu/drm/i915/i915_gem_evict.c | 17 ++-
drivers/gpu/drm/i915/i915_gem_execb
If the user requires patching of their batch or auxiliary buffers, we
currently make the alterations on the cpu. If they are active on the GPU
at the time, we wait under the struct_mutex for them to finish executing
before we rewrite the contents. This happens if shared relocation trees
are used be
The only time we need to emit a flush inside request emission is after
an execbuffer, for which we can use the full __i915_add_request(). All
other instances want the simpler i915_add_request() without flushing, so
remove the useless helper.
Signed-off-by: Chris Wilson
---
drivers/gpu/drm/i915/g
On pe, 2017-03-10 at 08:28 -0800, Oscar Mateo wrote:
> When initializing the GuC log struct, there is an object we need to
> allocate always, since the GuC needs its address at fw load time.
> The rest is only needed during runtime, in the sense that we only
> create if we actually enable GuC loggi
This simply hides the EAGAIN caused by userptr when userspace causes
resource contention. However, it is quite beneficial with highly
contended userptr users as we avoid repeating the setup costs and
kernel-user context switches.
Signed-off-by: Chris Wilson
Reviewed-by: Michał Winiarski
---
dri
Currently, the last object in the execlist is the always the batch.
However, when building the batch buffer we often know the batch object
first and if we can use the first slot in the execlist we can emit
relocation instructions relative to it immediately and avoid a separate
pass to adjust the re
i915_gem_stolen_list_info() sneakily takes advantage of the
obj->obj_exec_link to save itself from having to allocate. Enough of the
subterfuge, just allocate an array of pointers and sort them instead of
the list.
Signed-off-by: Chris Wilson
Reviewed-by: Mika Kuoppala
---
drivers/gpu/drm/i915/
Use a priority stored in the context as the initial value when
submitting a request. This allows us to change the default priority on a
per-context basis, allowing different contexts to be favoured with GPU
time at the expense of lower importance work. The user can adjust the
context's priority via
We can simplify our tracking of pending writes in an execbuf to the
single bit in the vma->exec_entry->flags, but that requires the
relocation function knowing the object's vma. Pass it along.
Signed-off-by: Chris Wilson
---
drivers/gpu/drm/i915/i915_gem_execbuffer.c | 100 --
On Thu, 2017-03-16 at 15:10 +0200, Jani Nikula wrote:
> On Thu, 16 Mar 2017, "Chauhan, Madhav" wrote:
> > > -Original Message-
> > > From: Nikula, Jani
> > > Sent: Thursday, February 16, 2017 9:03 PM
> > > To: Chauhan, Madhav ; intel-
> > > g...@lists.freedesktop.org
> > > Cc: Conselvan De
On pe, 2017-03-10 at 08:28 -0800, Oscar Mateo wrote:
> Doorbell release flow requires that we wait for GEN8_DRB_VALID bit to go
> to zero after updating db_status before we call the GuC to release the
> doorbell.
>
> Kudos to Daniele for finding this out.
>
> Cc: Joonas Lahtinen
> Cc: Daniele Ce
On Thu, Mar 16, 2017 at 06:42:21PM +0530, sourab gupta wrote:
> On Thu, 2017-03-16 at 03:09 -0700, Chris Wilson wrote:
> > On Thu, Mar 16, 2017 at 03:22:03PM +0530, sourab gupta wrote:
> > > On Thu, 2017-03-16 at 02:03 -0700, Chris Wilson wrote:
> > > > On Thu, Mar 16, 2017 at 02:24:55PM +0530, sou
On Thu, 16 Mar 2017, Ander Conselvan De Oliveira wrote:
> On Thu, 2017-03-16 at 15:10 +0200, Jani Nikula wrote:
>> On Thu, 16 Mar 2017, "Chauhan, Madhav" wrote:
>> > > -Original Message-
>> > > From: Nikula, Jani
>> > > Sent: Thursday, February 16, 2017 9:03 PM
>> > > To: Chauhan, Madhav
On Thu, 16 Mar 2017, Ville Syrjälä wrote:
> On Thu, Mar 16, 2017 at 09:40:05AM +0200, Jani Nikula wrote:
>> On Wed, 15 Mar 2017, Chris Wilson wrote:
>> > On Wed, Mar 15, 2017 at 05:09:04PM +0200, ville.syrj...@linux.intel.com
>> > wrote:
>> >> From: Ville Syrjälä
>> >>
>> >> Add a command for
== Series Details ==
Series: series starting with [1/2] drm/i915/scheduler: emulate a scheduler for
guc
URL : https://patchwork.freedesktop.org/series/21376/
State : success
== Summary ==
Series 21376v1 Series without cover letter
https://patchwork.freedesktop.org/api/1.0/series/21376/revisio
On Tue, 2017-03-14 at 13:01 -0700, Manasi Navare wrote:
> From: "Navare, Manasi D"
>
> Display stream compression is supported on DP 1.4 DP
> devices. This patch adds the corersponding DPCD
> register definitions for DSC.
>
> v3:
> * Add some SHIFTS and MASKS for uniformity (Jani Nikula)
> v2:
>
== Series Details ==
Series: series starting with [01/15] drm/i915: Copy user requested buffers into
the error state
URL : https://patchwork.freedesktop.org/series/21377/
State : warning
== Summary ==
Series 21377v1 Series without cover letter
https://patchwork.freedesktop.org/api/1.0/series/
On Thu, Mar 16, 2017 at 08:38:30AM +0100, Daniel Vetter wrote:
> Hi Greg,
>
> On Mon, Mar 13, 2017 at 07:40:50AM +0100, Daniel Vetter wrote:
> > On Sun, Mar 12, 2017 at 11:01 PM, Greg KH
> > wrote:
> > > So if a commit says "cherry-pick", I guess I can always assume it's safe
> > > to add, right
Chris Wilson writes:
> When manually overwriting the HWS, rather than assume irq_seqno_barrier
> does the right thing, we can explicitly flush the cacheline instead.
> This avoids us calling the engine->irq_seqno_barrier() from an illegal
> context:
>
> [ 1472.651797] BUG: scheduling while atomic
On Thu, Mar 16, 2017 at 6:14 AM, wrote:
> From: Sourab Gupta
>
> This patch exposes a new sample source field to userspace. This field can
> be populated to specify the origin of the OA report.
> Currently, the OA samples are being generated only periodically, and hence
> there's only source fla
On to, 2017-03-16 at 13:31 +0100, Michal Wajdeczko wrote:
> On Thu, Mar 16, 2017 at 11:41:51AM +, Chris Wilson wrote:
> >
> > guc_addon_create() makes the assumption that it need only kmap the
> > initial page in order to write all of the configuration data used by the
> > guc. Confusingly it
On Thu, Mar 16, 2017 at 04:15:35PM +0200, Mika Kuoppala wrote:
> Chris Wilson writes:
>
> > When manually overwriting the HWS, rather than assume irq_seqno_barrier
> > does the right thing, we can explicitly flush the cacheline instead.
> > This avoids us calling the engine->irq_seqno_barrier() f
On Thu, 16 Mar 2017, Greg KH wrote:
> And again, you all are the only ones that have this issue. You might
> find a handfull of patches for stable that come in twice in the rest of
> the kernel, but your "little" driver dwarfs that by an order of
> magnitude. I really think you are doing it wron
Do an early read of the execlists' queue before we take the spinlock and
start checking. This is safe as the first writer to the execlists queue
will cause the tasklet to be run again after a memory barrier.
Signed-off-by: Chris Wilson
Cc: Tvrtko Ursulin
Cc: Joonas Lahtinen
Cc: Michał Winiarski
On Thu, Mar 16, 2017 at 01:05:06PM +0100, Maarten Lankhorst wrote:
> Op 16-03-17 om 11:48 schreef Daniel Vetter:
> > On Wed, Mar 15, 2017 at 11:08:56AM +0100, Maarten Lankhorst wrote:
> >> As a proof of concept, first try to convert intel_tv, which is a rarely
> >> used connector. It has 5 properti
== Series Details ==
Series: drm/i915: Skip execlists_dequeue() early if the list is empty
URL : https://patchwork.freedesktop.org/series/21382/
State : failure
== Summary ==
Series 21382v1 drm/i915: Skip execlists_dequeue() early if the list is empty
https://patchwork.freedesktop.org/api/1.0/
Op 16-03-17 om 15:56 schreef Daniel Vetter:
> On Thu, Mar 16, 2017 at 01:05:06PM +0100, Maarten Lankhorst wrote:
>> Op 16-03-17 om 11:48 schreef Daniel Vetter:
>>> On Wed, Mar 15, 2017 at 11:08:56AM +0100, Maarten Lankhorst wrote:
As a proof of concept, first try to convert intel_tv, which is
On Thu, Mar 16, 2017 at 04:27:42PM +0100, Maarten Lankhorst wrote:
> Op 16-03-17 om 15:56 schreef Daniel Vetter:
> > On Thu, Mar 16, 2017 at 01:05:06PM +0100, Maarten Lankhorst wrote:
> >> Op 16-03-17 om 11:48 schreef Daniel Vetter:
> >>> On Wed, Mar 15, 2017 at 11:08:56AM +0100, Maarten Lankhorst
The vblank code really wants to look at crtc->state without having to
take a ww_mutex. One option might be to take one of the vblank locks
right when assigning crtc->state, which would ensure that the vblank
code doesn't race and access freed memory.
But userspace tends to poke the vblank_ioctl to
This should be impossible, but let's assert that we do not pin a context
4 billion times before retiring!
Signed-off-by: Chris Wilson
Cc: Tvrtko Ursulin
Cc: Joonas Lahtinen
---
drivers/gpu/drm/i915/intel_lrc.c| 1 +
drivers/gpu/drm/i915/intel_ringbuffer.c | 1 +
2 files changed, 2 inse
1 - 100 of 191 matches
Mail list logo
