Two exception handling register moves are faster to inline than a call
to __copy_user_ll(). We already apply the conversion for a get_user()
call, so for symmetry we should also apply the optimisation to
copy_from_user.
Signed-off-by: Chris Wilson
Cc: Thomas Gleixner
Cc: Ingo Molnar
Cc: "H. Pet
For known compile-time fixed sizes, teach x86-32 copy_to_user() to
convert them to the simpler put_user and inline it similar to the
optimisation applied to copy_from_user() and already used by x86-64.
Signed-off-by: Chris Wilson
Cc: Thomas Gleixner
Cc: Ingo Molnar
Cc: "H. Peter Anvin"
---
ar
Extend the list of replacements for compile-time known sizes to include
6/12 byte copies. These expand to two movs (along with their exception
table) and are cheaper to inline than the function call (similar to the
10 byte copy already handled).
Signed-off-by: Chris Wilson
Cc: Thomas Gleixner
Cc
== Series Details ==
Series: series starting with [1/3] x86-32: Teach copy_from_user to unroll
.size=6/8
URL : https://patchwork.freedesktop.org/series/25148/
State : success
== Summary ==
Series 25148v1 Series without cover letter
https://patchwork.freedesktop.org/api/1.0/series/25148/revisi
Op 26-05-17 om 17:15 schreef Mahesh Kumar:
> A display resolution is only supported if it meets all the restrictions
> below for Maximum Pipe Pixel Rate.
>
> The display resolution must fit within the maximum pixel rate output
> from the pipe. Make sure that the display pipe is able to feed pixels
On Wed, May 31, 2017 at 06:01:10PM -0700, Michel Thierry wrote:
> On 5/31/2017 12:05 PM, Chris Wilson wrote:
> >Commit 7c3f86b6dc51 ("drm/i915: Invalidate the guc ggtt TLB upon
> >insertion") added the restoration of the invalidation routine after the
> >GuC was disabled, but missed that the GuC wa
When we enable the GuC, we enable an alternative mechanism for doing
post-GGTT update invalidation. Likewise, when we disable the GuC, we
restore the previous method. Assert that we change between known
endpoints, so that we can catch if we accidentally clobber some other
gen and if we change the i
On Sat, May 27, 2017 at 04:38:51PM +0800, Xiaoguang Chen wrote:
> dmabuf for GVT-g can be exported to users who can use the dmabuf to show
> the desktop of vm which use intel vgpu.
>
> Currently we provide query and create new dmabuf operations.
>
> Users of dmabuf can cache some created dmabufs
== Series Details ==
Series: drm/i915/guc: Assert that we switch between known ggtt->invalidate
functions
URL : https://patchwork.freedesktop.org/series/25150/
State : success
== Summary ==
Series 25150v1 drm/i915/guc: Assert that we switch between known
ggtt->invalidate functions
https://pa
On Thu, Jun 01, 2017 at 03:15:13PM +0900, Inki Dae wrote:
>
>
> 2017년 05월 31일 17:45에 Daniel Vetter 이(가) 쓴 글:
> > On Tue, May 30, 2017 at 09:03:34AM +0900, Inki Dae wrote:
> >> Hi Daniel,
> >>
> >> 2017년 05월 24일 23:51에 Daniel Vetter 이(가) 쓴 글:
> >>> Only in the load failure path, where the hardware
On Wed, May 31, 2017 at 07:51:59PM +0100, Matthew Auld wrote:
> + max_page_size = BIT(fls64(obj->mm.page_sizes.sg)-1);
> +
> + /* If were are actually dealing with a single page-size, mark it so */
> + if (IS_ALIGNED(obj->mm.page_sizes.phys, max_page_size))
> + obj->mm.page_
On Thu, Jun 1, 2017 at 2:56 AM, kbuild test robot wrote:
> Hi Andy,
>
> [auto build test ERROR on pm/linux-next]
> [also build test ERROR on v4.12-rc3 next-20170531]
> [if your patch is applied to the wrong git tree, please drop us a note to
> help improve the system]
Dropped ahead. As cover let
On Wed, May 31, 2017 at 07:52:01PM +0100, Matthew Auld wrote:
> We can't mix 64K and 4K pte's in the same page-table, so for now we
> align 64K objects to 2M to avoid any potential mixing. This is
> potentially wasteful but in reality shouldn't be too bad since this only
> applies to the virtual ad
On Wed, May 31, 2017 at 07:52:04PM +0100, Matthew Auld wrote:
> When SW enables the use of 2M/1G pages, it must disable the GTT cache.
>
> Signed-off-by: Matthew Auld
> Cc: Joonas Lahtinen
> ---
> drivers/gpu/drm/i915/intel_pm.c | 12 ++--
> 1 file changed, 6 insertions(+), 6 deletions(
On Thu, Jun 01, 2017 at 10:04:46AM +0100, Chris Wilson wrote:
> When we enable the GuC, we enable an alternative mechanism for doing
> post-GGTT update invalidation. Likewise, when we disable the GuC, we
> restore the previous method. Assert that we change between known
> endpoints, so that we can
On Thu, Jun 01, 2017 at 12:03:18PM +0200, Michal Wajdeczko wrote:
> On Thu, Jun 01, 2017 at 10:04:46AM +0100, Chris Wilson wrote:
> > When we enable the GuC, we enable an alternative mechanism for doing
> > post-GGTT update invalidation. Likewise, when we disable the GuC, we
> > restore the previou
On ke, 2017-05-31 at 20:05 +0100, Chris Wilson wrote:
> Commit 7c3f86b6dc51 ("drm/i915: Invalidate the guc ggtt TLB upon
> insertion") added the restoration of the invalidation routine after the
> GuC was disabled, but missed that the GuC was unconditionally disabled
> when not used. This then over
On to, 2017-06-01 at 12:03 +0200, Michal Wajdeczko wrote:
> On Thu, Jun 01, 2017 at 10:04:46AM +0100, Chris Wilson wrote:
> >
> > void i915_ggtt_disable_guc(struct drm_i915_private *i915)
> > {
> > - if (i915->ggtt.invalidate == guc_ggtt_invalidate)
> > - i915->ggtt.invalidate = gen
On ke, 2017-05-31 at 19:51 +0100, Matthew Auld wrote:
> Enable transparent-huge-pages through gemfs by mounting with
> huge=within_size.
>
> Signed-off-by: Matthew Auld
> Cc: Joonas Lahtinen
> Cc: Chris Wilson
> @@ -41,6 +42,28 @@ struct vfsmount *i915_gemfs_create(void)
>
> gemfs_mn
Seems that GLK has a dotclock that's twice the display clock.
skl_max_scale checks for IS_GEMINILAKE, so perform the same check here.
While at it, change the DRM_ERROR to DEBUG_KMS.
Fixes: 73b0ca8ec76d ("drm/i915/skl+: consider max supported plane pixel
rate while scaling")
Cc: Mahesh Kumar
Sign
On to, 2017-06-01 at 10:04 +0100, Chris Wilson wrote:
> When we enable the GuC, we enable an alternative mechanism for doing
> post-GGTT update invalidation. Likewise, when we disable the GuC, we
> restore the previous method. Assert that we change between known
> endpoints, so that we can catch if
hmm... didn't considered 2 pixels per clock.
thanks.
Reviewed-by: Mahesh Kumar
On Thursday 01 June 2017 04:04 PM, Maarten Lankhorst wrote:
Seems that GLK has a dotclock that's twice the display clock.
skl_max_scale checks for IS_GEMINILAKE, so perform the same check here.
While at it, chang
On ke, 2017-05-31 at 19:51 +0100, Matthew Auld wrote:
> Not a fully blown gemfs, just our very own tmpfs kernel mount. Doing so
> moves us away from the shmemfs shm_mnt, and gives us the much needed
> flexibility to do things like set our own mount options, namely huge=
> which should allow us to e
== Series Details ==
Series: drm/i915/glk: Fix dotclock calculation in skl_check_pipe_max_pixel_rate
URL : https://patchwork.freedesktop.org/series/25155/
State : success
== Summary ==
Series 25155v1 drm/i915/glk: Fix dotclock calculation in
skl_check_pipe_max_pixel_rate
https://patchwork.fre
Op 01-06-17 om 12:52 schreef Mahesh Kumar:
>
> hmm... didn't considered 2 pixels per clock.
>
> thanks.
>
> Reviewed-by: Mahesh Kumar
>
Thanks, fix pushed. :)
___
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailma
On ke, 2017-05-31 at 19:51 +0100, Matthew Auld wrote:
> In preparation for huge gtt pages expose a page_size_mask as part of the
> device info, to indicate the page sizes supported by the HW. Currently
> only 4K is supported.
>
> Signed-off-by: Matthew Auld
> Cc: Joonas Lahtinen
> Cc: Mika Kuop
On Thu, Jun 01, 2017 at 01:44:07PM +0300, Joonas Lahtinen wrote:
> On to, 2017-06-01 at 10:04 +0100, Chris Wilson wrote:
> > When we enable the GuC, we enable an alternative mechanism for doing
> > post-GGTT update invalidation. Likewise, when we disable the GuC, we
> > restore the previous method.
On Wednesday 31 May 2017 10:16 PM, kbuild test robot wrote:
Hi Shobhit,
[auto build test ERROR on drm-intel/for-linux-next]
[also build test ERROR on v4.12-rc3 next-20170531]
[if your patch is applied to the wrong git tree, please drop us a note to help
improve the system]
Guess this needs to
On Thu, Jun 01, 2017 at 01:24:55PM +0300, Joonas Lahtinen wrote:
> On to, 2017-06-01 at 12:03 +0200, Michal Wajdeczko wrote:
> > On Thu, Jun 01, 2017 at 10:04:46AM +0100, Chris Wilson wrote:
> > >
> > > void i915_ggtt_disable_guc(struct drm_i915_private *i915)
> > > {
> > > - if (i915->ggtt.inv
With gcc7, the conditional usage of (port == PORT_A ? PORT_C : PORT_A)
triggers -Werror=int-in-bool-context which breaks the build. Instead,
use a temporary port_other variable that avoids hitting this error.
% gcc --version
gcc (SUSE Linux) 7.1.1 20170517 [gcc-7-branch revision 248152]
Copy
On to, 2017-06-01 at 12:01 +0100, Chris Wilson wrote:
> On Thu, Jun 01, 2017 at 01:44:07PM +0300, Joonas Lahtinen wrote:
> >
> > On to, 2017-06-01 at 10:04 +0100, Chris Wilson wrote:
> > >
> > > When we enable the GuC, we enable an alternative mechanism for doing
> > > post-GGTT update invalidati
On Wed, May 31, 2017 at 01:33:47PM +0100, Lionel Landwerlin wrote:
> u32 gen7_latched_oastatus1;
> + u32 ctx_oactxctrl_offset;
> + u32 ctx_flexeu0_offset;
> +
> + /* The RPT_ID/reason field for Gen8+ includes a bit
>
On Wed, May 31, 2017 at 01:33:55PM +0100, Lionel Landwerlin wrote:
> @@ -2493,6 +2499,44 @@ struct drm_i915_private {
> const struct i915_oa_format *oa_formats;
> int n_builtin_sets;
> } oa;
> +
> + struct {
> +
On Thu, 01 Jun 2017, Aleksa Sarai wrote:
> With gcc7, the conditional usage of (port == PORT_A ? PORT_C : PORT_A)
> triggers -Werror=int-in-bool-context which breaks the build. Instead,
> use a temporary port_other variable that avoids hitting this error.
>
> % gcc --version
> gcc (SUSE Linux)
On Wed, May 31, 2017 at 01:33:55PM +0100, Lionel Landwerlin wrote:
> static void
> -free_oa_buffer(struct drm_i915_private *i915)
> +free_sseu_buffer(struct drm_i915_private *i915)
> {
> - mutex_lock(&i915->drm.struct_mutex);
> + i915_gem_object_unpin_map(i915->perf.sseu_buffer.vma->obj);
== Series Details ==
Series: i915: fix build on gcc7
URL : https://patchwork.freedesktop.org/series/25158/
State : success
== Summary ==
Series 25158v1 i915: fix build on gcc7
https://patchwork.freedesktop.org/api/1.0/series/25158/revisions/1/mbox/
Test gem_exec_flush:
Subgroup basic-
On Thu, 01 Jun 2017, Joonas Lahtinen wrote:
> On ke, 2017-05-31 at 19:51 +0100, Matthew Auld wrote:
>> In preparation for huge gtt pages expose a page_size_mask as part of the
>> device info, to indicate the page sizes supported by the HW. Currently
>> only 4K is supported.
>>
>> Signed-off-by:
This is probably already fixed in drm-next by
commit 0ad4dc887d4168448e8c801aa4edd8fe1e0bd534
Author: Hans de Goede
Date: Thu May 18 13:06:44 2017 +0200
drm/i915: Fix new -Wint-in-bool-context gcc compiler warning
> which I also think is a more sensible fix than this one.
Ah, I didn'
On 01/06/17 09:02, Daniel Kasak wrote:
Any news? Seems every TV I bump into these days has Miracast support ...
Sorry, it must be frustrating. Some code needs to be re-implemented
because it was GPL-based and we need it as MIT. Until I have some time
to do this, this will probably not move fo
On Wed, May 31, 2017 at 05:34:42PM +, Patchwork wrote:
> == Series Details ==
>
> Series: drm/i915/ddi: Avoid long delays during system suspend / eDP disabling
> URL : https://patchwork.freedesktop.org/series/25116/
> State : success
>
> == Summary ==
>
> Series 25116v1 drm/i915/ddi: Avoid
On 1 June 2017 at 11:49, Joonas Lahtinen
wrote:
> On ke, 2017-05-31 at 19:51 +0100, Matthew Auld wrote:
>> Not a fully blown gemfs, just our very own tmpfs kernel mount. Doing so
>> moves us away from the shmemfs shm_mnt, and gives us the much needed
>> flexibility to do things like set our own mo
On Wed, 31 May 2017, Ville Syrjälä wrote:
> On Wed, May 31, 2017 at 08:05:35PM +0300, Imre Deak wrote:
>> Atm disabling either DP or eDP outputs can generate a spurious short
>> pulse interrupt. The reason is that after disabling the port the source
>> will stop sending a valid stream data, while
On Wed, 31 May 2017, Clint Taylor wrote:
> On 05/31/2017 03:16 AM, Jani Nikula wrote:
>> Print DID not VID on the DID error path. Looks like a copy-paste error
>> from the VID error path. Clarify and clean up error logging, making them
>> distinguishable from each other, while at it.
>
> Reviewed-
On Thu, Jun 01, 2017 at 09:36:58AM -, Patchwork wrote:
> == Series Details ==
>
> Series: drm/i915/guc: Assert that we switch between known ggtt->invalidate
> functions
> URL : https://patchwork.freedesktop.org/series/25150/
> State : success
>
> == Summary ==
>
> Series 25150v1 drm/i915/
On Wed, May 31, 2017 at 05:01:53PM +0300, Mika Kuoppala wrote:
> Chris Wilson writes:
>
> > On Tue, May 30, 2017 at 03:33:41PM +0300, Mika Kuoppala wrote:
> >> Chris Wilson writes:
> >>
> >> > As another precaution when testing whether the CS engine is actually
> >> > idle, also inspect the rin
On Mon, May 22, 2017 at 08:55:09AM +0300, Joonas Lahtinen wrote:
> On su, 2017-05-21 at 13:40 +0100, Chris Wilson wrote:
> > The assertion that we want to make before disabling the pin of the pages
> > for the unknown swizzling quirk is that the quirk is indeed active.
> >
> > Fixes: 2c3a3f44dc13
I tried __GFP_NORETRY in the belief that __GFP_RECLAIM was effective. It
struggles with handling reclaim via kswapd (through inconsistency within
throttle_direct_reclaim() and even then the race between multiple
allocators makes the two step of reclaim then allocate fragile), and as
our buffers are
Commit 24f8e00a8a2e ("drm/i915: Prefer to report ENOMEM rather than
incur the oom for gfx allocations") made the bold decision to try and
avoid the oomkiller by reporting -ENOMEM to userspace if our allocation
failed after attempting to free enough buffer objects. In short, it
appears we were givin
In commit 5763ff04dc4e ("drm/i915: Avoid GPU stalls from kswapd") we
stopped direct reclaim and kswapd from triggering GPU/client stalls
whilst running (by restricting the objects they could reap to be idle).
However with abusive GPU usage, it becomes quite easy to starve kswapd
of memory and prev
On Thu, Jun 01, 2017 at 12:34:13PM +0200, Maarten Lankhorst wrote:
> Seems that GLK has a dotclock that's twice the display clock.
> skl_max_scale checks for IS_GEMINILAKE, so perform the same check here.
>
> While at it, change the DRM_ERROR to DEBUG_KMS.
>
> Fixes: 73b0ca8ec76d ("drm/i915/skl+:
On Wed, 31 May 2017, Rodrigo Vivi wrote:
> Split out BXT and CNP's setup_backlight(),enable_backlight(),
> disable_backlight() and hz_to_pwm() into
> two separate functions instead of reusing BXT function.
>
> Reuse set_backlight() and get_backlight() since they have
> no reference to the utility
On Thu, Jun 01, 2017 at 03:55:13PM +0300, Jani Nikula wrote:
> On Wed, 31 May 2017, Ville Syrjälä wrote:
> > On Wed, May 31, 2017 at 08:05:35PM +0300, Imre Deak wrote:
> >> Atm disabling either DP or eDP outputs can generate a spurious short
> >> pulse interrupt. The reason is that after disabling
On Wed, May 31, 2017 at 10:41:52PM +0300, Andy Shevchenko wrote:
> acpi_evaluate_dsm() and friends take a pointer to a raw buffer of 16
> bytes. Instead we convert them to use guid_t type. At the same time we
> convert current users.
>
> acpi_str_to_uuid() becomes useless after the conversion and
On Thu, 01 Jun 2017, Ville Syrjälä wrote:
> On Thu, Jun 01, 2017 at 03:55:13PM +0300, Jani Nikula wrote:
>> On Wed, 31 May 2017, Ville Syrjälä wrote:
>> > On Wed, May 31, 2017 at 08:05:35PM +0300, Imre Deak wrote:
>> >> Atm disabling either DP or eDP outputs can generate a spurious short
>> >> pu
On Thu, Jun 01, 2017 at 04:58:50PM +0300, Ville Syrjälä wrote:
> On Thu, Jun 01, 2017 at 03:55:13PM +0300, Jani Nikula wrote:
> > On Wed, 31 May 2017, Ville Syrjälä wrote:
> > > On Wed, May 31, 2017 at 08:05:35PM +0300, Imre Deak wrote:
> > >> Atm disabling either DP or eDP outputs can generate a
== Series Details ==
Series: series starting with [1/3] drm/i915: Allow kswapd to pause the device
whilst reaping
URL : https://patchwork.freedesktop.org/series/25164/
State : success
== Summary ==
Series 25164v1 Series without cover letter
https://patchwork.freedesktop.org/api/1.0/series/251
On Wed, May 31, 2017 at 12:41 PM, Andy Shevchenko
wrote:
> acpi_evaluate_dsm() and friends take a pointer to a raw buffer of 16
> bytes. Instead we convert them to use guid_t type. At the same time we
> convert current users.
>
> acpi_str_to_uuid() becomes useless after the conversion and it's saf
On Wed, May 31, 2017 at 12:41 PM, Andy Shevchenko
wrote:
> There are new types and helpers that are supposed to be used in new code.
>
> As a preparation to get rid of legacy types and API functions do
> the conversion here.
>
> Cc: Dan Williams
> Signed-off-by: Andy Shevchenko
> ---
> drivers/
From: Ville Syrjälä
This series eliminates the problematic load detect abuse for the
pipe A quirk. My main motivations were to isolate these quirks
more from atomic to avoid regressions, and to save a bit of extra
power. I believe I cooked this up a few years ago but never posted
it. In the meant
From: Ville Syrjälä
If intel_crtc_disable_noatomic() were to ever get called during resume
e'd end up deadlocking since resume has its own acqcuire_ctx but
intel_crtc_disable_noatomic() still tries to use the
mode_config.acquire_ctx. Pass down the correct acquire ctx from the top.
Cc: sta...@vge
From: Ville Syrjälä
The magic "enable the DPLL three times" sequence feels like it
deserves a loop.
Signed-off-by: Ville Syrjälä
---
drivers/gpu/drm/i915/intel_display.c | 15 ++-
1 file changed, 6 insertions(+), 9 deletions(-)
diff --git a/drivers/gpu/drm/i915/intel_display.c
b
From: Ville Syrjälä
830 more or less requires both pipes and DPLLs to remain on as long
as either pipe is needed. However, when neither pipe is actually needed,
we can save a bit of power by turning everything off. To do that we add
a new "power well" that turns both pipes and DPLLs on and off in
From: Ville Syrjälä
Pass down the correct acquire context to the pipe A quirk load detect
hack during display resume. Avoids deadlocking the entire thing.
Cc: sta...@vger.kernel.org
Cc: Maarten Lankhorst
Fixes: e2c8b8701e2d ("drm/i915: Use atomic helpers for suspend, v2.")
Signed-off-by: Ville
From: Ville Syrjälä
The pipe A force quirk shouldn't needed except on 830. So let's nuke it
for the Toshiba Protege R-205/S-209 945 machines. This quirk pre-dates
KMS so it's usefulness is doubtful at best now.
Unfortunately the original bug report [1] isn't very helpful since it
doesn't describ
From: Ville Syrjälä
With 830 the only thing needing pipe quirks, we can just drop the quirk
defines and replace the checks with IS_I830() checks.
Signed-off-by: Ville Syrjälä
---
drivers/gpu/drm/i915/i915_drv.h | 2 -
drivers/gpu/drm/i915/intel_display.c | 92
From: Ville Syrjälä
The pipe A force quirk shouldn't needed except on 830. So let's nuke it
for the IBM Thinkpad T60 945 machines. This quirk pre-dates
KMS so it's usefulness is doubtful at best now.
The original bug report [1] describes the symptoms as "system hang on
closing T60 panel lid", an
On Thu, 01 Jun 2017, Imre Deak wrote:
> On Thu, Jun 01, 2017 at 04:58:50PM +0300, Ville Syrjälä wrote:
>> On Thu, Jun 01, 2017 at 03:55:13PM +0300, Jani Nikula wrote:
>> > On Wed, 31 May 2017, Ville Syrjälä wrote:
>> > > On Wed, May 31, 2017 at 08:05:35PM +0300, Imre Deak wrote:
>> > >> Atm disab
On Thu, Jun 01, 2017 at 05:36:12PM +0300, ville.syrj...@linux.intel.com wrote:
> From: Ville Syrjälä
>
> Note that I'm entirely removing the few remaining pipe A quirks
> for non-830 platforms as they predate KMS and the hardware
> really shouldn't need them.
> drm/i915: Drop pipe A quirk for
On Thu, 01 Jun 2017, ville.syrj...@linux.intel.com wrote:
> From: Ville Syrjälä
>
> If intel_crtc_disable_noatomic() were to ever get called during resume
> e'd end up deadlocking since resume has its own acqcuire_ctx but
> intel_crtc_disable_noatomic() still tries to use the
> mode_config.acquire
On Thu, 01 Jun 2017, ville.syrj...@linux.intel.com wrote:
> From: Ville Syrjälä
>
> Pass down the correct acquire context to the pipe A quirk load detect
> hack during display resume. Avoids deadlocking the entire thing.
Have we seen this in the wild? References?
BR,
Jani.
>
> Cc: sta...@vger.k
On Thu, 01 Jun 2017, ville.syrj...@linux.intel.com wrote:
> From: Ville Syrjälä
>
> The magic "enable the DPLL three times" sequence feels like it
> deserves a loop.
I think the copy-paste unrolled loop is more fun. ;)
Reviewed-by: Jani Nikula
> Signed-off-by: Ville Syrjälä
> ---
> drivers/
On Thu, Jun 01, 2017 at 05:36:16PM +0300, ville.syrj...@linux.intel.com wrote:
> From: Ville Syrjälä
>
> 830 more or less requires both pipes and DPLLs to remain on as long
> as either pipe is needed. However, when neither pipe is actually needed,
> we can save a bit of power by turning everythin
On Thu, Jun 01, 2017 at 05:36:15PM +0300, ville.syrj...@linux.intel.com wrote:
> From: Ville Syrjälä
>
> The magic "enable the DPLL three times" sequence feels like it
> deserves a loop.
Hmm, I thought once upon a time we figured out what the magic was about.
Or was that another loop?
Reviewed
== Series Details ==
Series: drm/i915: Pipe A quirk rework
URL : https://patchwork.freedesktop.org/series/25169/
State : success
== Summary ==
Series 25169v1 drm/i915: Pipe A quirk rework
https://patchwork.freedesktop.org/api/1.0/series/25169/revisions/1/mbox/
fi-bdw-5557u total:278 pass
If the atomic commit doesn't include any new plane, there is no need to
choose a new CRTC for FBC, and the intel_fbc_choose_crtc() will bail out
early. Although, if the FBC setup failed in the previous commit, if the
current commit doesn't include new plane update, it tries to overwrite
no_fbc_rea
On 05/24/2017 04:52 PM, Daniel Vetter wrote:
> Seems entirely cargo-culted.
>
> Cc: Benjamin Gaignard
> Cc: Vincent Abriou
> Signed-off-by: Daniel Vetter
> ---
> drivers/gpu/drm/sti/sti_drv.c | 1 -
> 1 file changed, 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/sti/sti_drv.c b/drivers/g
== Series Details ==
Series: drm: i915: Preserve old FBC status for update without new planes (rev2)
URL : https://patchwork.freedesktop.org/series/24635/
State : success
== Summary ==
Series 24635v2 drm: i915: Preserve old FBC status for update without new planes
https://patchwork.freedesktop
On Thu, Jun 01, 2017 at 05:47:45PM +0300, Jani Nikula wrote:
> On Thu, 01 Jun 2017, ville.syrj...@linux.intel.com wrote:
> > From: Ville Syrjälä
> >
> > Pass down the correct acquire context to the pipe A quirk load detect
> > hack during display resume. Avoids deadlocking the entire thing.
>
> H
On Thu, Jun 01, 2017 at 05:48:16PM +0300, Jani Nikula wrote:
> On Thu, 01 Jun 2017, ville.syrj...@linux.intel.com wrote:
> > From: Ville Syrjälä
> >
> > If intel_crtc_disable_noatomic() were to ever get called during resume
> > e'd end up deadlocking since resume has its own acqcuire_ctx but
> > i
Please ignore what I wrote. Looks like I was a bit hasty in my
judgement. The code just doesn't read easily but seems correct, other
than the array size.
Harry
On 2017-05-31 04:17 PM, Harry Wentland wrote:
On 2017-05-31 09:32 AM, Harry Wentland wrote:
On 2017-05-31 05:37 AM, Daniel Vetter wr
On Thu, Jun 01, 2017 at 03:48:51PM +0100, Chris Wilson wrote:
> On Thu, Jun 01, 2017 at 05:36:15PM +0300, ville.syrj...@linux.intel.com wrote:
> > From: Ville Syrjälä
> >
> > The magic "enable the DPLL three times" sequence feels like it
> > deserves a loop.
>
> Hmm, I thought once upon a time
On Thu, Jun 01, 2017 at 03:46:43PM +0100, Chris Wilson wrote:
> On Thu, Jun 01, 2017 at 05:36:16PM +0300, ville.syrj...@linux.intel.com wrote:
> > From: Ville Syrjälä
> >
> > 830 more or less requires both pipes and DPLLs to remain on as long
> > as either pipe is needed. However, when neither pi
On Thu, 2017-06-01 at 02:15 +, Pandiyan, Dhinakaran wrote:
> On Tue, 2017-05-30 at 15:42 -0700, Rodrigo Vivi wrote:
> > Split out BXT and CNP's setup_backlight(),enable_backlight(),
> > disable_backlight() and hz_to_pwm() into
> > two separate functions instead of reusing BXT function.
> >
> >
On Thu, 1 Jun 2017 03:01:28 +
"Chen, Xiaoguang" wrote:
> Hi Kirti,
>
> >-Original Message-
> >From: Kirti Wankhede [mailto:kwankh...@nvidia.com]
> >Sent: Thursday, June 01, 2017 1:23 AM
> >To: Chen, Xiaoguang ; Gerd Hoffmann
> >; alex.william...@redhat.com; ch...@chris-wilson.co.uk;
For ease of use (i.e. avoiding a few checks and function calls), store
the object's cache coherency next to the cache is dirty bit.
Signed-off-by: Chris Wilson
Cc: Dongwon Kim
Cc: Matt Roper
Tested-by: Dongwon Kim
---
drivers/gpu/drm/i915/i915_gem.c | 14 +++---
drive
Currently, we only mark the CPU cache as dirty if we skip a clflush.
This leads to some confusion where we have to ask if the object is in
the write domain or missed a clflush. If we always mark the cache as
dirty, this becomes a much simply question to answer.
The goal remains to do as few clflus
On Thu, Jun 01, 2017 at 07:05:41PM +0300, Ville Syrjälä wrote:
> On Thu, Jun 01, 2017 at 05:48:16PM +0300, Jani Nikula wrote:
> > On Thu, 01 Jun 2017, ville.syrj...@linux.intel.com wrote:
> > > From: Ville Syrjälä
> > >
> > > If intel_crtc_disable_noatomic() were to ever get called during resume
>
== Series Details ==
Series: series starting with [1/2] drm/i915: Mark CPU cache as dirty on every
transition for CPU writes
URL : https://patchwork.freedesktop.org/series/25173/
State : success
== Summary ==
Series 25173v1 Series without cover letter
https://patchwork.freedesktop.org/api/1.0
Op 01-06-17 om 15:52 schreef Ville Syrjälä:
> On Thu, Jun 01, 2017 at 12:34:13PM +0200, Maarten Lankhorst wrote:
>> Seems that GLK has a dotclock that's twice the display clock.
>> skl_max_scale checks for IS_GEMINILAKE, so perform the same check here.
>>
>> While at it, change the DRM_ERROR to DEB
Based on your clarification the second option feels like the right choice, with
a relevant comment in code. Like you said, we get to retain BXT register
definitions and clarify that the register is on a PCH for CNP.
-DK
-Original Message-
From: Vivi, Rodrigo
Sent: Thursday, June 1, 20
From: Ville Syrjälä
The number of compressed segments has been available ever since
FBC2 was introduced in g4x, it just moved from the STATUS register
into STATUS2 on IVB.
For FBC1 if we really wanted the number of compressed segments we'd
have to trawl through the tags, but in this case since t
On Sat, 27 May 2017 16:38:52 +0800
Xiaoguang Chen wrote:
> User space should create the management fd for the dma-buf operation first.
> Then user can query the plane information and create dma-buf if necessary
> using the management fd.
>
> Signed-off-by: Xiaoguang Chen
> ---
> drivers/gpu/dr
From: Ville Syrjälä
Remove the SNB PCH refclock init call from the runtime resume handler.
I don't think it was actually needed even when we had SNB runtime PM,
and if definitely isn't needed ever since SNB runtime PM was nuked in
commit d4c5636e7447 ("drm/i915: Remove runtime PM for SNB").
Cc:
== Series Details ==
Series: drm/i915: Implement fbc_status "Compressing" info for all platforms
URL : https://patchwork.freedesktop.org/series/25176/
State : success
== Summary ==
Series 25176v1 drm/i915: Implement fbc_status "Compressing" info for all
platforms
https://patchwork.freedesktop
On 6/1/2017 10:08 PM, Alex Williamson wrote:
> On Thu, 1 Jun 2017 03:01:28 +
> "Chen, Xiaoguang" wrote:
>
>> Hi Kirti,
>>
>>> -Original Message-
>>> From: Kirti Wankhede [mailto:kwankh...@nvidia.com]
>>> Sent: Thursday, June 01, 2017 1:23 AM
>>> To: Chen, Xiaoguang ; Gerd Hoffmann
>
== Series Details ==
Series: drm/i915: Remove dead code from runtime resume handler
URL : https://patchwork.freedesktop.org/series/25182/
State : success
== Summary ==
Series 25182v1 drm/i915: Remove dead code from runtime resume handler
https://patchwork.freedesktop.org/api/1.0/series/25182/r
From: Ville Syrjälä
IVB+ have the cursor "FBC" feature, meaning they support a
somewhat limited form of non-square cursors. Let's test that.
Signed-off-by: Ville Syrjälä
---
tests/kms_cursor_crc.c | 8 +++-
1 file changed, 7 insertions(+), 1 deletion(-)
diff --git a/tests/kms_cursor_crc.c
On Wed, May 31, 2017 at 09:06:12PM +0530, Sharma, Shashank wrote:
> Regards
>
> Shashank
>
>
> On 5/31/2017 6:11 PM, Ville Syrjälä wrote:
> > On Tue, May 30, 2017 at 10:00:12PM +0530, Sharma, Shashank wrote:
> >> Regards
> >>
> >> Shashank
> >>
> >>
> >> On 5/30/2017 9:43 PM, Ville Syrjälä wrote
From: Chris Wilson
---
integration-manifest | 26 ++
1 file changed, 26 insertions(+)
create mode 100644 integration-manifest
diff --git a/integration-manifest b/integration-manifest
new file mode 100644
index 000..bb20fc4
--- /dev/null
+++ b/integration-manifest
@@
From: Chris Wilson
The assertion that we want to make before disabling the pin of the pages
for the unknown swizzling quirk is that the quirk is indeed active, and
that the quirk is disabled before we do apply it to the pages.
Fixes: 2c3a3f44dc13 ("drm/i915: Fix pages pin counting around swizzle
1 - 100 of 127 matches
Mail list logo
