== Series Details ==
Series: series starting with [1/1] drm/i915: Do RPM Wake during GuC/HuC status
read
URL : https://patchwork.freedesktop.org/series/19034/
State : failure
== Summary ==
Series 19034v1 Series without cover letter
https://patchwork.freedesktop.org/api/1.0/series/19034/revisi
On Fri, Feb 03, 2017 at 01:00:18PM +0530, Sagar Arun Kamble wrote:
> HUC_STATUS, GUC_STATUS, SOFT_SCRATCH registers are read in debugfs.
> This patch covers those accesses by RPM get/put.
See also I915_PARAM_HUC_STATUS.
> Cc: Arkadiusz Hiler
> Cc: Anusha Srivatsa
> Signed-off-by: Sagar Arun Ka
On Thu, Feb 02, 2017 at 07:27:45AM -0800, Oscar Mateo wrote:
> From: Michal Wajdeczko
>
> The GuC descriptor is big in size. If we use local definition of
> guc_desc we have a chance to overflow stack. Use allocated one.
>
> v2: Rebased
> v3: Split
> v4: Handle ENOMEM, cover all uses of guc_cont
HUC_STATUS, GUC_STATUS, SOFT_SCRATCH registers are read in debugfs.
This patch covers those accesses by RPM get/put.
Cc: Arkadiusz Hiler
Cc: Anusha Srivatsa
Signed-off-by: Sagar Arun Kamble
---
drivers/gpu/drm/i915/i915_debugfs.c | 6 ++
1 file changed, 6 insertions(+)
diff --git a/drive
== Series Details ==
Series: drm/i915/bxt: Add MST support when do DPLL calculation (rev3)
URL : https://patchwork.freedesktop.org/series/17815/
State : failure
== Summary ==
Series 17815v3 drm/i915/bxt: Add MST support when do DPLL calculation
https://patchwork.freedesktop.org/api/1.0/series/
From: "Lee, Shawn C"
Add the missing INTEL_OUTPUT_DP_MST case in bxt_get_dpll()
to correctly initialize the crtc_state and port plls when
link training a DP MST monitor on BXT/APL devices.
Fixes: a277ca7dc01d ("drm/i915: Split bxt_ddi_pll_select()")
Bugs: https://bugs.freedesktop.org/show_bug.cg
Regards
Shashank
On 2/2/2017 11:43 PM, Thierry Reding wrote:
On Thu, Feb 02, 2017 at 11:08:22AM +0530, Sharma, Shashank wrote:
Regards
Shashank
On 2/1/2017 10:02 PM, Thierry Reding wrote:
On Wed, Feb 01, 2017 at 06:14:39PM +0530, Shashank Sharma wrote:
HDMI 2.0 spec mandates scrambling f
== Series Details ==
Series: drm/i915/guc: Make intel_guc_send a function pointer
URL : https://patchwork.freedesktop.org/series/19024/
State : failure
== Summary ==
Series 19024v1 drm/i915/guc: Make intel_guc_send a function pointer
https://patchwork.freedesktop.org/api/1.0/series/19024/revis
== Series Details ==
Series: drm/i915/guc: Dynamically alloc GuC descriptor
URL : https://patchwork.freedesktop.org/series/19022/
State : success
== Summary ==
Series 19022v1 drm/i915/guc: Dynamically alloc GuC descriptor
https://patchwork.freedesktop.org/api/1.0/series/19022/revisions/1/mbox/
From: Michal Wajdeczko
Prepare for an alternate GuC communication interface.
v2: Make a few functions static and name them correctly while we are at it
(Oscar)
Signed-off-by: Michel Thierry
Signed-off-by: Michal Wajdeczko
Signed-off-by: Oscar Mateo
---
drivers/gpu/drm/i915/intel_uc.c | 25
From: Michal Wajdeczko
The GuC descriptor is big in size. If we use local definition of
guc_desc we have a chance to overflow stack. Use allocated one.
v2: Rebased
v3: Split
v4: Handle ENOMEM, cover all uses of guc_context_desc, use kzalloc (Oscar)
Signed-off-by: Deepak S
Signed-off-by: Michal
== Series Details ==
Series: series starting with [v5] drm: Improve drm_mm search (and fix topdown
allocation) with rbtrees (rev2)
URL : https://patchwork.freedesktop.org/series/18987/
State : success
== Summary ==
Series 18987v2 Series without cover letter
https://patchwork.freedesktop.org/a
== Series Details ==
Series: drm/i915: Drain the freed state from the tail of the next commit
URL : https://patchwork.freedesktop.org/series/19012/
State : failure
== Summary ==
Series 19012v1 drm/i915: Drain the freed state from the tail of the next commit
https://patchwork.freedesktop.org/ap
The drm_mm range manager claimed to support top-down insertion, but it
was neither searching for the top-most hole that could fit the
allocation request nor fitting the request to the hole correctly.
In order to search the range efficiently, we create a secondary index
for the holes using either t
tree: git://anongit.freedesktop.org/drm-intel drm-intel-nightly
head: 0f01216949002d20b9dc6d300c82df5ffa59e9a7
commit: e4563f6ba71792c77aeccb2092cc23149b44e642 [1076/1086] drm: Rely on
mode_config data for fb_helper initialization
config: i386-randconfig-x0-02030244 (attached as .config)
compi
If we have any residual freed atomic state from earlier commits, flush
the freed list after performing the current modeset. This prevents the
freed list from ever-growing if userspace manages to starve the kernel
threads (i.e. we are never able to run our free state worker and
eventually the system
On Fri, May 27, 2016 at 10:19 AM, Chris Bainbridge
wrote:
> On 25 May 2016 at 08:31, Sedat Dilek wrote:
>> Hi Daniel,
>>
>> with latest Linus Git I see this with my Intel SandyBridge GPU...
>>
>> [ 17.629014] [drm:intel_cpu_fifo_underrun_irq_handler [i915]]
>> *ERROR* CPU pipe A FIFO underrun
>
tree: git://anongit.freedesktop.org/drm/drm-tip drm-tip
head: 36aab92fa2091dd54ea983752aa40e427e83d113
commit: e4563f6ba71792c77aeccb2092cc23149b44e642 [1066/1073] drm: Rely on
mode_config data for fb_helper initialization
config: i386-randconfig-x007-201705 (attached as .config)
compiler: gcc
vmwgfx part: Reviewed-by: Sinclair Yeh
On Thu, Feb 02, 2017 at 11:44:33AM +, Chris Wilson wrote:
> The drm_mm range manager claimed to support top-down insertion, but it
> was neither searching for the top-most hole that could fit the
> allocation request nor fitting the request to the hole
On Wed, Feb 01, 2017 at 11:40:06AM -0800, Manasi Navare wrote:
> Are you planning on submitting a v2 for these pretty soon
> that can make it to patchwork/
>
> Regards
> Manasi
>
> On Thu, Jan 26, 2017 at 09:44:14PM +0200, Jani Nikula wrote:
> > This is kind of version 1½ of [1], basically just r
On Thu, Feb 02, 2017 at 11:08:22AM +0530, Sharma, Shashank wrote:
> Regards
>
> Shashank
>
>
> On 2/1/2017 10:02 PM, Thierry Reding wrote:
> > On Wed, Feb 01, 2017 at 06:14:39PM +0530, Shashank Sharma wrote:
> > > HDMI 2.0 spec mandates scrambling for modes with pixel clock higher
> > > than 340
Reviewed-by: Robert Foss
On 2017-02-02 06:12 AM, Mika Kahola wrote:
To be more suitable for BAT testing, let's modify the test such a way
that it only uses the maximum number of available planes. This reduces
the total number of subtests.
v2: Fix broken kms_plane_multiple --list-subtests (Petr
On Thu, Feb 02, 2017 at 10:58:43AM +0530, Sharma, Shashank wrote:
> Thanks for the review Thierry. My comments inline.
>
> Regards
> Shashank
> On 2/1/2017 9:40 PM, Thierry Reding wrote:
> > On Wed, Feb 01, 2017 at 06:14:38PM +0530, Shashank Sharma wrote:
> > > This patch does following:
> > > - A
== Series Details ==
Series: series starting with [01/19] drm/i915: Micro-optimise
i915_get_ggtt_vma_pages()
URL : https://patchwork.freedesktop.org/series/18997/
State : success
== Summary ==
Series 18997v1 Series without cover letter
https://patchwork.freedesktop.org/api/1.0/series/18997/re
On Thu, Feb 02, 2017 at 10:42:48AM +0200, Jani Nikula wrote:
> On Thu, 02 Feb 2017, Manasi Navare wrote:
> > On Thu, Jan 26, 2017 at 09:44:26PM +0200, Jani Nikula wrote:
> >> Localize link_rate_index to the if block, and rename to just index to
> >> reduce indent.
> >>
> >> Cc: Manasi Navare
> >
== Series Details ==
Series: drm/i915: remove 512GB allocation warning
URL : https://patchwork.freedesktop.org/series/18995/
State : success
== Summary ==
Series 18995v1 drm/i915: remove 512GB allocation warning
https://patchwork.freedesktop.org/api/1.0/series/18995/revisions/1/mbox/
fi-bdw-
On 02/02/2017 17:05, Chris Wilson wrote:
On Thu, Feb 02, 2017 at 04:39:49PM +, Tvrtko Ursulin wrote:
On 02/02/2017 16:10, Chris Wilson wrote:
On Thu, Feb 02, 2017 at 03:57:43PM +, Tvrtko Ursulin wrote:
On 02/02/2017 15:32, Chris Wilson wrote:
On Thu, Feb 02, 2017 at 03:02:32PM +000
On Thu, Feb 02, 2017 at 04:39:49PM +, Tvrtko Ursulin wrote:
>
> On 02/02/2017 16:10, Chris Wilson wrote:
> >On Thu, Feb 02, 2017 at 03:57:43PM +, Tvrtko Ursulin wrote:
> >>
> >>On 02/02/2017 15:32, Chris Wilson wrote:
> >>>On Thu, Feb 02, 2017 at 03:02:32PM +, Chris Wilson wrote:
> >>>
== Series Details ==
Series: Still call-traces after suspend-resume (pm? i915? cpu/hotplug?) (rev2)
URL : https://patchwork.freedesktop.org/series/18871/
State : success
== Summary ==
Series 18871v2 Still call-traces after suspend-resume (pm? i915? cpu/hotplug?)
https://patchwork.freedesktop.o
On 02/02/2017 16:10, Chris Wilson wrote:
On Thu, Feb 02, 2017 at 03:57:43PM +, Tvrtko Ursulin wrote:
On 02/02/2017 15:32, Chris Wilson wrote:
On Thu, Feb 02, 2017 at 03:02:32PM +, Chris Wilson wrote:
Improve the sg iteration and in hte process eliminate a bug in
miscomputing the pml4
Hi,
On 2 February 2017 at 07:41, Maarten Lankhorst
wrote:
> i915 is pretty much feature complete. Support for atomic i915-specific
> connector properties is still missing; those properties can (for now)
> only be set through the legacy ioctl.
>
> ILK style watermarks and gen9+ watermarks are hand
On Wed, Jan 04, 2017 at 08:42:30PM +0200, ville.syrj...@linux.intel.com wrote:
> From: Ville Syrjälä
>
> DRM_UT_CORE generates way too much noise usually, so having the
> framebuffer init failures use DRM_UT_CORE is a pain when trying to
> find out the reason why you failed in creating a framebuf
On Wed, Jan 04, 2017 at 08:42:29PM +0200, ville.syrj...@linux.intel.com wrote:
> From: Ville Syrjälä
>
> intel_fill_fb_info() should pass the correct plane index to
> _intel_compute_tile_offset() once we start to care about the AUX
> surface.
>
> Signed-off-by: Ville Syrjälä
This changes how x
On Thu, Feb 02, 2017 at 03:57:43PM +, Tvrtko Ursulin wrote:
>
> On 02/02/2017 15:32, Chris Wilson wrote:
> >On Thu, Feb 02, 2017 at 03:02:32PM +, Chris Wilson wrote:
> >>Improve the sg iteration and in hte process eliminate a bug in
> >>miscomputing the pml4 length as orig_nents< >>full le
On 02/02/2017 15:32, Chris Wilson wrote:
On Thu, Feb 02, 2017 at 03:02:32PM +, Chris Wilson wrote:
Improve the sg iteration and in hte process eliminate a bug in
miscomputing the pml4 length as orig_nents<
which fixes a corner case of 0c40ce130e38
Fixes: 0c40ce130e38 ("drm/i915: Trim the o
Hi,
> >
> > Add basic somewhere in the test names?
>
> Why? Does something still parse basic in the test name and add it to a test
No
> set? Shouldn't that now be pulling from these lists instead?
You are right, yes , we control only static list and addition to it through
reviews, Petri, more
Hi Dave, here's Maarten's backport of the vma fixes for v4.10.
BR,
Jani.
The following changes since commit 566cf877a1fcb6d6dc0126b076aad062054c2637:
Linux 4.10-rc6 (2017-01-29 14:25:17 -0800)
are available in the git repository at:
git://anongit.freedesktop.org/git/drm-intel
tags/topic/
On Thu, Feb 02, 2017 at 03:02:32PM +, Chris Wilson wrote:
> Improve the sg iteration and in hte process eliminate a bug in
> miscomputing the pml4 length as orig_nents< full length of the sg table.
>
which fixes a corner case of 0c40ce130e38
Fixes: 0c40ce130e38 ("drm/i915: Trim the object sg
On Wed, Jan 04, 2017 at 08:42:28PM +0200, ville.syrj...@linux.intel.com wrote:
> From: Ville Syrjälä
>
> Based on empirical evidence the display engine (at least) always
> treats Yf tiles as 128Bx32. Currently we're assuming the tile dimensions
> change based on the pixel format. Let's adjust our
== Series Details ==
Series: drm/i915: Recreate internal objects with single page segments if dmar
fails
URL : https://patchwork.freedesktop.org/series/18991/
State : success
== Summary ==
Series 18991v1 drm/i915: Recreate internal objects with single page segments if
dmar fails
https://patc
Nice! I actually was already thinking of bringing up the fact we should
just be turning this on by default now, especially so we can see atomic
start getting some real use.
So, I'm more then happy to say I support flipping the switch :)
Reviewed-by: Lyude
On Thu, 2017-02-02 at 08:41 +0100, Maart
The plan in the near-future is to allow requests to be removed from the
signaler. We can no longer then rely on holding a reference to the
request for the duration it is in the signaling tree, and instead must
obtain a reference to the request for the current operation using RCU.
Signed-off-by: Ch
On 2 February 2017 at 15:02, Chris Wilson wrote:
> Very large objects are expected, so don't WARN the user if they are
> using them!
>
> Signed-off-by: Chris Wilson
Let's just go with yours then :)
Reviewed-by: Matthew Auld
___
Intel-gfx mailing list
On Thu, Feb 02, 2017 at 02:55:00PM +, Matthew Auld wrote:
> Now that we have selftests in place exercising truly huge allocations
> we will start to hit the 512GB warning, so now seems like a good time to
> remove it.
Hey! I wrote userspace tests to hit this WARN() a couple of years ago ;)
-Ch
Use a local variable to avoid having to type out the full name of the
gpu_error wait_queue.
Signed-off-by: Chris Wilson
---
drivers/gpu/drm/i915/i915_gem_request.c | 8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/drivers/gpu/drm/i915/i915_gem_request.c
b/drivers/gpu/d
Combine the common code for the pair of waiters into a single function.
Signed-off-by: Chris Wilson
---
drivers/gpu/drm/i915/i915_gem_request.c | 21 -
1 file changed, 12 insertions(+), 9 deletions(-)
diff --git a/drivers/gpu/drm/i915/i915_gem_request.c
b/drivers/gpu/drm/i9
On reflection, we are only using the execute fence as a waitqueue on the
global_seqno and not using it for dependency tracking between fences
(unlike the submit and dma fences). By only treating it as a waitqueue,
we can then treat it similar to the other waitqueues durin submit,
making the code si
If we preempt a request and remove it from the execution queue, we need
to undo its global seqno and restart any waiters.
Signed-off-by: Chris Wilson
---
drivers/gpu/drm/i915/intel_breadcrumbs.c | 63 ++--
drivers/gpu/drm/i915/intel_ringbuffer.h | 1 +
2 files chang
Now that the code is getting simpler, we can reduce the indentation when
waiting for the global_seqno.
Signed-off-by: Chris Wilson
---
drivers/gpu/drm/i915/i915_gem_request.c | 33 ++---
1 file changed, 14 insertions(+), 19 deletions(-)
diff --git a/drivers/gpu/drm/i
Add ourselves to the gpu error waitqueue earllier on, even before we
determine we have to wait on the seqno. This is so that we can then
share the waitqueue between stages in subsequent patches.
Signed-off-by: Chris Wilson
---
drivers/gpu/drm/i915/i915_gem_request.c | 10 +-
1 file chang
Add a mock selftest to preempt a request and check that we cancel it,
requeue the request and then complete its execution.
Signed-off-by: Chris Wilson
---
drivers/gpu/drm/i915/selftests/i915_gem_request.c | 59 +++
drivers/gpu/drm/i915/selftests/mock_request.c | 19 ++
As we handoff the GPU reset to the waiter, we need to check we don't
miss a wakeup if it has already been sent prior to us starting the wait.
Signed-off-by: Chris Wilson
---
drivers/gpu/drm/i915/i915_gem_request.c | 7 +--
1 file changed, 5 insertions(+), 2 deletions(-)
diff --git a/drivers
Replace the global device seqno with one for each engine, and account
for in-flight seqno on each separately. This is consistent with
dma-fence as each timeline has separate fence-contexts for each engine
and a seqno is only ordered within a fence-context (i.e. seqno do not
need to be ordered wrt
After the request is cancelled, we then need to remove it from the
global execution timeline and return it to the context timeline, the
inverse of submit_request().
Signed-off-by: Chris Wilson
---
drivers/gpu/drm/i915/i915_gem_request.c | 58 +++-
drivers/gpu/drm/i91
A request is assigned a global seqno only when it is on the hardware
execution queue. The global seqno can be used to maintain a list of
requests on the same engine in retirement order, for example for
constructing a priority queue for waiting. Prior to its execution, or
if it is subsequently remov
If we change the wait_queue_t from using the autoremove_wake_function to
the default_wake_function, we no longer have to restore the wait_queue_t
entry on the wait_queue_head_t list after being woken up by it, as we
are unusual in sleeping multiple times on the same wait_queue_t.
Signed-off-by: Ch
It had only one callsite and existed to keep the code clearer. Now
having shared the wait-on-error between phases and with plans to change
the wait-for-execute in the next few patches, remove the out of line
wait loop and move it into the main body of i915_wait_request.
Signed-off-by: Chris Wilson
On Thu, Feb 02, 2017 at 02:33:54PM +0100, Lucas Stach wrote:
> Am Donnerstag, den 02.02.2017, 11:44 + schrieb Chris Wilson:
> > @@ -192,6 +188,8 @@ static int etnaviv_iommu_find_iova(struct etnaviv_iommu
> > *mmu,
> > list_del_init(&m->scan_node);
> > }
> >
>
The barrier here is not required - we apply the barrier before the range
is ever reused by the GPU instead.
Signed-off-by: Chris Wilson
---
drivers/gpu/drm/i915/i915_gem_gtt.c | 3 ---
1 file changed, 3 deletions(-)
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c
b/drivers/gpu/drm/i915/i915_g
We never assign or use the ppgtt->enable() callback, so remove it.
Signed-off-by: Chris Wilson
---
drivers/gpu/drm/i915/i915_gem_gtt.h | 1 -
1 file changed, 1 deletion(-)
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.h
b/drivers/gpu/drm/i915/i915_gem_gtt.h
index 5af5a5d53d93..caaa2c5c23be 10
The tracepoints are now entirely synonymous with binding and unbinding the
VMA (and the tracepoints there).
Signed-off-by: Chris Wilson
---
drivers/gpu/drm/i915/i915_gem_gtt.c | 4 --
drivers/gpu/drm/i915/i915_trace.h | 97 -
2 files changed, 101 deletions(
The aliasing_ppgtt is a regular ppgtt, and we can use the regular
i915_ppgtt_put() to properly tear it down.
Signed-off-by: Chris Wilson
---
drivers/gpu/drm/i915/i915_gem_gtt.c | 53 +++--
1 file changed, 15 insertions(+), 38 deletions(-)
diff --git a/drivers/gpu
We want to reload the PDP (and flush the TLB) when the addresses are
changed.
Signed-off-by: Chris Wilson
---
drivers/gpu/drm/i915/i915_gem_gtt.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c
b/drivers/gpu/drm/i915/i915_gem_gtt.c
ind
Improve the sg iteration and in hte process eliminate a bug in
miscomputing the pml4 length as orig_nents<
---
drivers/gpu/drm/i915/i915_gem_gtt.c | 157 +++-
1 file changed, 82 insertions(+), 75 deletions(-)
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c
b/driv
We only operate on known extents (both for alloc/clear) and so we can use
both the knowledge of the bind/unbind range along with the knowledge of
the existing pagetable to avoid having to allocate temporary and
auxiliary bitmaps.
Signed-off-by: Chris Wilson
---
drivers/gpu/drm/i915/i915_gem_gtt.
We only operate on known extents (both for alloc/clear) and so we can use
both the knowledge of the bind/unbind range along with the knowledge of
the existing pagetable to avoid having to allocate temporary and
auxiliary bitmaps.
Signed-off-by: Chris Wilson
---
drivers/gpu/drm/i915/i915_gem_gtt.
We only operate on known extents (both for alloc/clear) and so we can use
both the knowledge of the bind/unbind range along with the knowledge of
the existing pagetable to avoid having to allocate temporary and
auxiliary bitmaps.
Signed-off-by: Chris Wilson
---
drivers/gpu/drm/i915/i915_gem_gtt.
The predominant VMA class is normal GTT, so allow gcc to emphasize that
path and avoid unnecessary stack movement.
Signed-off-by: Chris Wilson
---
drivers/gpu/drm/i915/i915_gem_gtt.c | 61 +++--
1 file changed, 32 insertions(+), 29 deletions(-)
diff --git a/drive
We only operate on known extents (both for alloc/clear) and so we can use
both the knowledge of the bind/unbind range along with the knowledge of
the existing pagetable to avoid having to allocate temporary and
auxiliary bitmaps.
Signed-off-by: Chris Wilson
---
drivers/gpu/drm/i915/i915_gem_gtt.
Upon creation of the va range, it is initialised to point at scratch.
Signed-off-by: Chris Wilson
---
drivers/gpu/drm/i915/i915_gem_gtt.c | 5 -
1 file changed, 5 deletions(-)
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c
b/drivers/gpu/drm/i915/i915_gem_gtt.c
index e777ef19893e..991a15e
Similar to how we already split the bind_vma for ggtt/aliasing_gtt, also
split up the unbind for symmetry.
Signed-off-by: Chris Wilson
---
drivers/gpu/drm/i915/i915_gem_gtt.c | 28
1 file changed, 20 insertions(+), 8 deletions(-)
diff --git a/drivers/gpu/drm/i915/i9
In the future, we need to call allocate_va_range on the aliasing-ppgtt
which means moving the call down from the vma into the vm (which is
more appropriate for calling the vm function).
Signed-off-by: Chris Wilson
---
drivers/gpu/drm/i915/i915_gem_gtt.c | 10 +-
drivers/gpu/drm/i915/i915
Signed-off-by: Chris Wilson
---
drivers/gpu/drm/i915/i915_gem_gtt.c | 34 --
1 file changed, 12 insertions(+), 22 deletions(-)
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c
b/drivers/gpu/drm/i915/i915_gem_gtt.c
index 991a15efe478..66c3540eb387 100644
--- a/dri
Inline the address computation to avoid the vfunc call for every page.
We still have to pay the high overhead of sg_page_iter_next(), but now
at least GCC can optimise the inner most loop, giving a significant
boost to some thrashing Unreal Engine workloads.
Signed-off-by: Chris Wilson
---
drive
As these are now both plain and simple kmap_atomic/kunmap_atomic pairs,
we can remove the wrappers for a small gain of clarity (in particular,
not hiding the atomic critical sections!).
Signed-off-by: Chris Wilson
---
drivers/gpu/drm/i915/i915_gem_gtt.c | 69 ++---
We flush the entire page every time we update a few bytes, making the
update of a page table many, many times slower than is required. If we
create a WC map of the page for our updates, we can avoid the clflush
but incur additional cost for creating the pagetable. We amoritize that
cost by reusing
Very large objects are expected, so don't WARN the user if they are
using them!
Signed-off-by: Chris Wilson
---
drivers/gpu/drm/i915/i915_gem_gtt.c | 4
1 file changed, 4 deletions(-)
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c
b/drivers/gpu/drm/i915/i915_gem_gtt.c
index 2428a5fe532e
Now that we have selftests in place exercising truly huge allocations
we will start to hit the 512GB warning, so now seems like a good time to
remove it.
Cc: Joonas Lahtinen
Cc: Chris Wilson
Signed-off-by: Matthew Auld
---
drivers/gpu/drm/i915/i915_gem_gtt.c | 4
1 file changed, 4 deletio
On Thu, Feb 2, 2017 at 6:44 AM, Chris Wilson wrote:
> The drm_mm range manager claimed to support top-down insertion, but it
> was neither searching for the top-most hole that could fit the
> allocation request nor fitting the request to the hole correctly.
>
> In order to search the range efficie
On Wed, 01 Feb 2017, "Vivi, Rodrigo" wrote:
> Reviewed-by: Rodrigo Vivi
Pushed to drm-intel-next-queued, thanks for the review.
BR,
Jani.
>
>> On Feb 1, 2017, at 5:46 AM, Nikula, Jani wrote:
>>
>> Apparently there are machines out there with Skylake CPU and KabyPoint
>> PCH. Judging from ou
== Series Details ==
Series: series starting with [v4,1/2] drm: Improve drm_mm search (and fix
topdown allocation) with rbtrees
URL : https://patchwork.freedesktop.org/series/18987/
State : failure
== Summary ==
Series 18987v1 Series without cover letter
https://patchwork.freedesktop.org/api/
Op 02-02-17 om 14:44 schreef Chris Wilson:
> On Thu, Feb 02, 2017 at 02:30:19PM +0100, Maarten Lankhorst wrote:
>> Op 02-02-17 om 10:18 schreef Chris Wilson:
>>> These are meant to be fast and sensitive to new (and old) bugs...
>>>
>>> Signed-off-by: Chris Wilson
>>> Cc: Petri Latvala
>>> ---
>>>
== Series Details ==
Series: drm/i915/scheduler: emulate a scheduler for guc (rev4)
URL : https://patchwork.freedesktop.org/series/18909/
State : success
== Summary ==
Series 18909v4 drm/i915/scheduler: emulate a scheduler for guc
https://patchwork.freedesktop.org/api/1.0/series/18909/revision
On Thu, Feb 02, 2017 at 02:30:19PM +0100, Maarten Lankhorst wrote:
> Op 02-02-17 om 10:18 schreef Chris Wilson:
> > These are meant to be fast and sensitive to new (and old) bugs...
> >
> > Signed-off-by: Chris Wilson
> > Cc: Petri Latvala
> > ---
> > tests/intel-ci/fast-feedback.testlist | 19 +
On Fri, Jan 27, 2017 at 09:54:01PM +0200, Jani Nikula wrote:
> On Fri, 27 Jan 2017, "Vivi, Rodrigo" wrote:
> > Reviewed-by: Rodrigo Vivi
> >
> > Jani this is a case for cc:stable?
>
> Yes, if we know old kernels can handle SRV GT3.
>
> BR,
> Jani.
I don't think this particular device is differ
On Wed, Jan 04, 2017 at 08:42:27PM +0200, ville.syrj...@linux.intel.com wrote:
> From: Ville Syrjälä
>
> To make life easier let's allow skl_plane_stride() to be called for the
> AUX surface even when there is no AUX surface. Avoids special cases in
> the callers.
>
> Signed-off-by: Ville Syrjäl
On Thu, Feb 02, 2017 at 12:41:42PM +, Tvrtko Ursulin wrote:
>
> On 01/02/2017 11:34, Chris Wilson wrote:
> >On Wed, Feb 01, 2017 at 11:17:39AM +, Tvrtko Ursulin wrote:
>
> >>>+
> >>>+ for (npages = npages_funcs; *npages; npages++) {
> >>>+ prandom
On Wed, Jan 04, 2017 at 08:42:26PM +0200, ville.syrj...@linux.intel.com wrote:
> From: Ville Syrjälä
>
> Let's try to keep the alignment requirements in one place, and so
> towards that end let's move the AUX_DIST alignment handling into
> intel_surf_alignment() alongside the main surface alignme
Am Donnerstag, den 02.02.2017, 11:44 + schrieb Chris Wilson:
> The drm_mm range manager claimed to support top-down insertion, but it
> was neither searching for the top-most hole that could fit the
> allocation request nor fitting the request to the hole correctly.
>
> In order to search the
Op 02-02-17 om 10:18 schreef Chris Wilson:
> These are meant to be fast and sensitive to new (and old) bugs...
>
> Signed-off-by: Chris Wilson
> Cc: Petri Latvala
> ---
> tests/intel-ci/fast-feedback.testlist | 19 +++
> 1 file changed, 19 insertions(+)
>
> diff --git a/tests/int
On Wed, Jan 04, 2017 at 08:42:25PM +0200, ville.syrj...@linux.intel.com wrote:
> From: Ville Syrjälä
>
> Now that framebuffers can be used even before calling
> drm_framebuffer_init() we can start to plumb them into more places,
> instead of passing individual pieces for fb metadata.
>
> Signed-
Chris Wilson writes:
> Check that we can reset the GPU and continue executing from the next
> request.
>
> Signed-off-by: Chris Wilson
> ---
> drivers/gpu/drm/i915/i915_drv.h| 4 +-
> drivers/gpu/drm/i915/intel_hangcheck.c | 4 +
> .../gpu/drm/i915/selftests/
If we fail to dma-map the object, the most common cause is lack of space
inside the SW-IOTLB due to fragmentation. If we recreate the_sg_table
using segments of PAGE_SIZE (and single page allocations), we may succeed
in remapping the scatterlist.
First became a significant problem for the mock sel
On Thu, Feb 02, 2017 at 01:10:57PM +, Matthew Auld wrote:
> On 2 February 2017 at 09:08, Chris Wilson wrote:
> > + err = -EINVAL;
> > + mutex_lock(&i915->drm.struct_mutex);
> > + err = i915_gem_object_attach_phys(obj, PAGE_SIZE);
> > + mutex_unlock(&i915->drm.struct_mut
Joonas Lahtinen writes:
> On ke, 2017-02-01 at 16:50 +, Chris Wilson wrote:
>> They include useful material such as what mode the VM address space is
>> running in, what submission mode, extra quirks, etc.
>>
>> Signed-off-by: Chris Wilson
>
>
>
>> +static void err_print_params(struct dr
On 2 February 2017 at 09:08, Chris Wilson wrote:
> The phys object is a rarely used device (only very old machines require
> a chunk of physically contiguous pages for a few hardware interactions).
> As such, it is not exercised by CI and to combat that we want to add a
> test that exercises the p
On Thu, Feb 02, 2017 at 12:49:58PM +, Tvrtko Ursulin wrote:
>
> On 02/02/2017 09:08, Chris Wilson wrote:
> >+static bool wait_for_ready(struct igt_wakeup *w)
> >+{
> >+DEFINE_WAIT(ready);
> >+
> >+if (atomic_dec_and_test(w->done))
> >+wake_up_atomic_t(w->done);
> >+
> >+
On 02/02/2017 09:08, Chris Wilson wrote:
Exercise creating rotated VMA and checking the page order within.
v2: Be more creative in rotated params
v3: ...
Signed-off-by: Chris Wilson
---
drivers/gpu/drm/i915/selftests/i915_vma.c | 179 ++
1 file changed, 179 ins
On 02/02/2017 09:08, Chris Wilson wrote:
Third retroactive test, make sure that the seqno waiters are woken.
v2: Smattering of comments, rearrange code
v3: Fix assert.
Signed-off-by: Chris Wilson
---
drivers/gpu/drm/i915/selftests/intel_breadcrumbs.c | 201 +
1 file ch
On 01/02/2017 11:34, Chris Wilson wrote:
On Wed, Feb 01, 2017 at 11:17:39AM +, Tvrtko Ursulin wrote:
+
+ for (npages = npages_funcs; *npages; npages++) {
+ prandom_seed_state(&prng,
+ i915
1 - 100 of 196 matches
Mail list logo