== Series Details ==
Series: WRPLL fixes for HDMI 2.0 on Cannonlake. (rev3)
URL : https://patchwork.freedesktop.org/series/33823/
State : warning
== Summary ==
Test kms_cursor_legacy:
Subgroup cursora-vs-flipa-legacy:
pass -> SKIP (shard-hsw)
Test kms_frontb
== Series Details ==
Series: drm/i915/selftests: Always initialise err (rev2)
URL : https://patchwork.freedesktop.org/series/33822/
State : success
== Summary ==
Test kms_setmode:
Subgroup basic:
pass -> FAIL (shard-hsw) fdo#99912
Test kms_flip:
Subg
== Series Details ==
Series: series starting with [v4] lib: Attempt to load the module for a missing
device (rev4)
URL : https://patchwork.freedesktop.org/series/33774/
State : warning
== Summary ==
Test drv_module_reload:
Subgroup basic-reload-inject:
dmesg-warn -> PA
== Series Details ==
Series: drm/i915: Add might_sleep() check to wait_for()
URL : https://patchwork.freedesktop.org/series/33833/
State : success
== Summary ==
Test kms_flip:
Subgroup dpms-vs-vblank-race-interruptible:
pass -> FAIL (shard-hsw) fdo#103060
Te
== Series Details ==
Series: series starting with [1/8] drm/i915/selftests: Increase size for mock
ringbuffer
URL : https://patchwork.freedesktop.org/series/33830/
State : success
== Summary ==
Test perf:
Subgroup polling:
fail -> PASS (shard-hsw) fdo#10225
== Series Details ==
Series: WRPLL fixes for HDMI 2.0 on Cannonlake. (rev3)
URL : https://patchwork.freedesktop.org/series/33823/
State : success
== Summary ==
Series 33823v3 WRPLL fixes for HDMI 2.0 on Cannonlake.
https://patchwork.freedesktop.org/api/1.0/series/33823/revisions/3/mbox/
Test
== Series Details ==
Series: igt/kms_rotation_crc: Add RGB565 90 degree test for gen>9 (rev2)
URL : https://patchwork.freedesktop.org/series/33132/
State : success
== Summary ==
Test drv_module_reload:
Subgroup basic-reload:
pass -> DMESG-WARN (shard-hsw) fdo#1027
== Series Details ==
Series: WRPLL fixes for HDMI 2.0 on Cannonlake. (rev2)
URL : https://patchwork.freedesktop.org/series/33823/
State : success
== Summary ==
Series 33823v2 WRPLL fixes for HDMI 2.0 on Cannonlake.
https://patchwork.freedesktop.org/api/1.0/series/33823/revisions/2/mbox/
Test
I confess I never fully understood that previous calculation,
so this is not a "fix". But let's simplify this math
so poor brains like mine can read and make some sense of
it in the future.
v2: Don't follow the spec since that gives invalid
values and it is also confusing. This Ville's
ver
== Series Details ==
Series: lib: Ask the kernel to quiesce the GPU (rev3)
URL : https://patchwork.freedesktop.org/series/31850/
State : failure
== Summary ==
IGT patchset tested on top of latest successful build
f370d5903689f37620e006f004a91d6bdeac7c16 lib/i915: Query semaphore status using
== Series Details ==
Series: drm/i915/selftests: Always initialise err
URL : https://patchwork.freedesktop.org/series/33822/
State : success
== Summary ==
Test perf:
Subgroup polling:
fail -> PASS (shard-hsw) fdo#102252
Test kms_busy:
Subgroup extend
Spec describe all values in MHz. We handle our
clocks in KHz. This includes the best_dco_centrality that was
forgot in the same unity as spec. Consequently we couldn't
get a good divider for high frequenies. Hence HDMI 2.0 wasn't
working.
Spec tells 99 for initial best_dco_centrality meaning t
I tested the patch from Ville Syrjälä's email and it works fine on top of
4.14 for me.
Thanks,
Neil
On Thu, Nov 9, 2017 at 1:16 PM, Eric Anholt wrote:
> Daniel Vetter writes:
>
> > On Wed, Nov 08, 2017 at 02:21:08PM -0800, Eric Anholt wrote:
> >> Ville Syrjälä writes:
> >>
> >> > On Wed, Nov
On Tue, Nov 14, 2017 at 11:47:57AM -0800, Rodrigo Vivi wrote:
> Accordingly to spec "If Kdiv != 2, then Qdiv must be 1."
> but we already handle qdiv values properly and this case here
> should be spurious. But instead of blindly replacing let's
> warn loudly instead. Because it means something was
== Series Details ==
Series: drm/i915: Prevent overflow of execbuf.buffer_count and num_cliprects
URL : https://patchwork.freedesktop.org/series/33840/
State : warning
== Summary ==
Series 33840v1 drm/i915: Prevent overflow of execbuf.buffer_count and
num_cliprects
https://patchwork.freedeskt
On Tue, Nov 14, 2017 at 09:29:57PM +, Rodrigo Vivi wrote:
> On Tue, Nov 14, 2017 at 08:46:38PM +, Ville Syrjälä wrote:
> > On Tue, Nov 14, 2017 at 10:22:27PM +0200, Ville Syrjälä wrote:
> > > On Tue, Nov 14, 2017 at 11:47:58AM -0800, Rodrigo Vivi wrote:
> > > > I confess I never fully under
== Series Details ==
Series: drm/i915/selftests: Markup __iomem for igt_gem_coherency
URL : https://patchwork.freedesktop.org/series/33819/
State : success
== Summary ==
Test perf:
Subgroup polling:
fail -> PASS (shard-hsw) fdo#102252
Test kms_busy:
== Series Details ==
Series: drm/i915/selftests: Always initialise err (rev2)
URL : https://patchwork.freedesktop.org/series/33822/
State : success
== Summary ==
Series 33822v2 drm/i915/selftests: Always initialise err
https://patchwork.freedesktop.org/api/1.0/series/33822/revisions/2/mbox/
T
We check whether the multiplies will overflow prior to calling
kmalloc_array so that we can respond with -EINVAL for the invalid user
arguments rather than treating it as an -ENOMEM that would otherwise
occur. However, as Dan Carpenter pointed out, we did an addition on the
unsigned int prior to pa
== Series Details ==
Series: series starting with [1/2] drm/i915: Drop the unbound page cache upon
idling
URL : https://patchwork.freedesktop.org/series/33817/
State : warning
== Summary ==
Test kms_busy:
Subgroup extended-modeset-hang-newfb-with-reset-render-a:
pass
On Tue, 2017-11-14 at 22:33 +, Chris Wilson wrote:
> smatch does not track initialised values as well as gcc, and this
> triggers many warnings by smatch not presented by gcc. Silence smatch by
> initialising the error values to -ENODEV, which we use to denote
> internal errors. (If we see a se
== Series Details ==
Series: series starting with [v4] lib: Attempt to load the module for a missing
device (rev4)
URL : https://patchwork.freedesktop.org/series/33774/
State : success
== Summary ==
IGT patchset tested on top of latest successful build
f370d5903689f37620e006f004a91d6bdeac7c16
smatch does not track initialised values as well as gcc, and this
triggers many warnings by smatch not presented by gcc. Silence smatch by
initialising the error values to -ENODEV, which we use to denote
internal errors. (If we see a selftest fail with a silent -ENODEV, we
know smatch was right!)
== Series Details ==
Series: drm/i915: Add might_sleep() check to wait_for()
URL : https://patchwork.freedesktop.org/series/33833/
State : success
== Summary ==
Series 33833v1 drm/i915: Add might_sleep() check to wait_for()
https://patchwork.freedesktop.org/api/1.0/series/33833/revisions/1/mbo
== Series Details ==
Series: drm/i915: Fix kerneldocs for intel_audio.c
URL : https://patchwork.freedesktop.org/series/33816/
State : warning
== Summary ==
Test perf:
Subgroup polling:
fail -> PASS (shard-hsw) fdo#102252
Test kms_busy:
Subgroup exten
On Tue, 2017-11-14 at 19:47 +, Chris Wilson wrote:
> smatch does not track initialised values as well as gcc, and this
> triggers many warnings by smatch not presented by gcc. Silence smatch by
> initialising the error values to -ENODEV, which we use to denote
> internal errors. (If we see a s
== Series Details ==
Series: series starting with [1/8] drm/i915/selftests: Increase size for mock
ringbuffer
URL : https://patchwork.freedesktop.org/series/33830/
State : success
== Summary ==
Series 33830v1 series starting with [1/8] drm/i915/selftests: Increase size for
mock ringbuffer
ht
We should long past the time of trying to use wait_for() from inside
atomic contexts, so add a might_sleep() check to prevent misuse.
Signed-off-by: Chris Wilson
Cc: Tvrtko Ursulin
---
drivers/gpu/drm/i915/intel_drv.h | 11 ++-
1 file changed, 2 insertions(+), 9 deletions(-)
diff --git
Since the introduction of debugfs/i915_drop_caches, we have offered the
ability to wait upon all outstanding batches. This is more efficient and
less error prone (one example is the use of context priorities, we have
to idle at the lowest in order not to jump over any low priority tasks
we want to
== Series Details ==
Series: drm/uapi: Validate mode flags/type, and deprecate some of them
URL : https://patchwork.freedesktop.org/series/33810/
State : failure
== Summary ==
Test kms_busy:
Subgroup extended-modeset-hang-newfb-with-reset-render-c:
pass -> DMESG-W
On Tue, Nov 14, 2017 at 06:44:51PM -0200, Paulo Zanoni wrote:
> Em Ter, 2017-11-14 às 22:34 +0200, Ville Syrjälä escreveu:
> > On Tue, Nov 14, 2017 at 06:12:41PM -0200, Paulo Zanoni wrote:
> > > Em Qui, 2017-11-02 às 17:17 +0200, Ville Syrjala escreveu:
> > > > From: Ville Syrjälä
> > > >
> > > >
If we asked to open a particular chipset and we find no matching device,
try again after attempting to load its module. Previously we only did
this for vgem, which is not automatically probed during boot, but if we
want to leave the module unloaded we have to try harder when we need the
device.
v2
Execlists and legacy ringbuffer submission are no longer feature
comparable (execlists now offer greater functionality that should
overcome their performance hit) and obsoletes the unsafe module
parameter, i.e. comparing the two modes of execution is no longer
useful, so remove the debug tool.
Sig
I should have admitted defeat long ago as there has been a rare but
persistent error on Sandybridge where semaphore signaling did not
propagate to the waiter, leading to a GPU hang.
With the work on fence signaling for v4.9, the impact of using CPU driven
signaling was greatly reduced wrt to the l
During request construction, after pinning the context we know whether
or not we have to emit a context switch. So move this common operation
from every caller into i915_gem_request_alloc() itself.
v2: Always submit the request if we emitted some commands during request
construction, as typically
We don't actually emit any commands into the ringbuffer, so we set it
very small. However, an upcoming change centralises the wait-for-space
into i915_gem_request_alloc() and that imposes a minimum size upon all
ringbuffers (mock or real) of MIN_SPACE_FOR_ADD_REQUEST. Grow the
mock ringbuffer such
At the start of building a request, we would wait for roughly enough
space to fit the average request (to reduce the likelihood of having to
wait and abort partway through request construction). To achieve we
would try to begin a 0-length command packet, this just adds extra
confusion so make the w
Since removing the module parameter to force selection of ringbuffer
emission for gen8, the code is defunct. Remove it.
To put the difference into perspective, a couple of microbenchmarks
(bdw i7-5557u, 20170324):
ring execlists
exec continuous nops
Having disabled the broken semaphores on Sandybridge, there is no need
for a modparam any more, so remove it in favour of a simple
HAS_LEGACY_SEMAPHORES() guard.
Signed-off-by: Chris Wilson
Cc: Joonas Lahtinen
Reviewed-by: Maarten Lankhorst
---
drivers/gpu/drm/i915/i915_drv.c | 7 +---
As the semaphores is just part of the engine, include it with the
general pretty printer universally used for debugging.
Signed-off-by: Chris Wilson
Cc: Joonas Lahtinen
---
drivers/gpu/drm/i915/i915_debugfs.c| 32
drivers/gpu/drm/i915/intel_engine_cs.c | 9
On Tue, Nov 14, 2017 at 08:46:38PM +, Ville Syrjälä wrote:
> On Tue, Nov 14, 2017 at 10:22:27PM +0200, Ville Syrjälä wrote:
> > On Tue, Nov 14, 2017 at 11:47:58AM -0800, Rodrigo Vivi wrote:
> > > I confess I never fully understood that previous calculation,
> > > so maybe this is cannot be call
On Mon, Oct 16, 2017 at 02:47:14PM +, Michal Wajdeczko wrote:
> Remaining functions in intel_guc_loader.c were focused around
> GuC firmware. Rename them to match object-verb pattern and
> rename file itself.
Error: Cannot open file ../drivers/gpu/drm/i915/intel_guc_loader.c
== Series Details ==
Series: series starting with [v3] lib: Attempt to load the module for a missing
device (rev3)
URL : https://patchwork.freedesktop.org/series/33774/
State : failure
== Summary ==
IGT patchset tested on top of latest successful build
f370d5903689f37620e006f004a91d6bdeac7c16
On Tue, Nov 14, 2017 at 06:59:21PM -0200, Paulo Zanoni wrote:
> Em Qui, 2017-11-02 às 17:17 +0200, Ville Syrjala escreveu:
> > From: Ville Syrjälä
> >
> > While I have no solid proof that ILK follows the ELK path when it
> > comes to the stolen memory reserved area, there are some hints that
> >
On Tue, Nov 14, 2017 at 11:47:55AM -0800, Rodrigo Vivi wrote:
> - 64 bits is not needed for afe_clock now we don't convert
> that to Hz.
> - 16 bits is not enough for all dco stuff.
> - unsigned is not relevant/needed for all divisors values.
>
Yup great catch, DCO stuff needs 3 bytes so we nee
== Series Details ==
Series: igt/kms_rotation_crc: Add RGB565 90 degree test for gen>9 (rev2)
URL : https://patchwork.freedesktop.org/series/33132/
State : success
== Summary ==
IGT patchset tested on top of latest successful build
f370d5903689f37620e006f004a91d6bdeac7c16 lib/i915: Query semap
On Tue, 2017-11-14 at 21:11 +0200, Ville Syrjala wrote:
> From: Ville Syrjälä
>
> Fix copy/paste fail in kerneldocs for intel_audio_codec_disable().
>
> Cc: Daniel Vetter
Reviewed-by: Dhinakaran Pandiyan
> Suggested-by: Daniel Vetter
> Signed-off-by: Ville Syrjälä
> ---
> drivers/gpu/d
Em Qui, 2017-11-02 às 17:17 +0200, Ville Syrjala escreveu:
> From: Ville Syrjälä
>
> While I have no solid proof that ILK follows the ELK path when it
> comes to the stolen memory reserved area, there are some hints that
> it might be the case. Unfortunately my ILK doesn't have this enabled,
> an
On Tue, Nov 14, 2017 at 11:47:54AM -0800, Rodrigo Vivi wrote:
> No functional change. Just starting the wrpll fixes
> with a clean-up to make units a bit more clear.
>
> Cc: Mika Kahola
> Cc: Manasi Navare
> Cc: James Ausmus
> Signed-off-by: Rodrigo Vivi
> ---
> drivers/gpu/drm/i915/intel_dpl
== Series Details ==
Series: WRPLL fixes for HDMI 2.0 on Cannonlake.
URL : https://patchwork.freedesktop.org/series/33823/
State : failure
== Summary ==
Series 33823v1 WRPLL fixes for HDMI 2.0 on Cannonlake.
https://patchwork.freedesktop.org/api/1.0/series/33823/revisions/1/mbox/
Test kms_fli
== Series Details ==
Series: drm/i915: Mark the userptr invalidate workqueue as WQ_MEM_RECLAIM
URL : https://patchwork.freedesktop.org/series/33804/
State : success
== Summary ==
shard-hswtotal:2584 pass:1472 dwarn:3 dfail:1 fail:9 skip:1099
time:9499s
Blacklisted hosts:
shard-a
On Tue, Nov 14, 2017 at 10:22:27PM +0200, Ville Syrjälä wrote:
> On Tue, Nov 14, 2017 at 11:47:58AM -0800, Rodrigo Vivi wrote:
> > I confess I never fully understood that previous calculation,
> > so maybe this is cannot be called a "fix". But let's follow
> > the math that is written on Spec so we
Em Ter, 2017-11-14 às 22:34 +0200, Ville Syrjälä escreveu:
> On Tue, Nov 14, 2017 at 06:12:41PM -0200, Paulo Zanoni wrote:
> > Em Qui, 2017-11-02 às 17:17 +0200, Ville Syrjala escreveu:
> > > From: Ville Syrjälä
> > >
> > > Apparently there are some machines that put semi-sensible looking
> > > v
== Series Details ==
Series: igt/gem_exec_capture: Exercise readback of userptr (rev2)
URL : https://patchwork.freedesktop.org/series/31480/
State : warning
== Summary ==
Test drv_module_reload:
Subgroup basic-reload:
dmesg-warn -> PASS (shard-hsw) fdo#102707
Test
On Tue, 2017-11-14 at 20:32 +0200, Ville Syrjala wrote:
> @@ -253,8 +251,11 @@ struct drm_display_mode {
>* - DRM_MODE_TYPE_USERDEF: Mode defined via kernel command line
>*
>* Plus a big list of flags which shouldn't be used at all, but are
> - * still around since th
On Tue, Nov 14, 2017 at 08:22:27PM +, Ville Syrjälä wrote:
> On Tue, Nov 14, 2017 at 11:47:58AM -0800, Rodrigo Vivi wrote:
> > I confess I never fully understood that previous calculation,
> > so maybe this is cannot be called a "fix". But let's follow
> > the math that is written on Spec so we
Quoting Paulo Zanoni (2017-11-14 20:29:49)
> Em Ter, 2017-11-14 às 20:19 +, Chris Wilson escreveu:
> > Only fbc actually depends on stolen allocation to
> > function, and no one complains if fbc is disabled. (There's a sketch
> > out
> > there that we could use a contiguous allocation for fbc
== Series Details ==
Series: igt/gem_ctx_switch: Do a warmup pass over all contexts (rev2)
URL : https://patchwork.freedesktop.org/series/33515/
State : success
== Summary ==
Test drv_module_reload:
Subgroup basic-reload-inject:
pass -> DMESG-WARN (shard-hsw) fdo#
On Tue, Nov 14, 2017 at 06:33:41PM -0200, Paulo Zanoni wrote:
> Em Qui, 2017-11-02 às 17:17 +0200, Ville Syrjala escreveu:
> > From: Ville Syrjälä
> >
> > Now that we should be properly filtering out the cases when the
> > stolen
> > reserved area is disabled, let's convert the debug message abou
== Series Details ==
Series: dma-buf/fence: Fix lock inversion within dma-fence-array
URL : https://patchwork.freedesktop.org/series/33799/
State : success
== Summary ==
Test kms_setmode:
Subgroup basic:
pass -> FAIL (shard-hsw) fdo#99912
fdo#99912 https://
Verified this from the Bspec that central frequency should
be left at default at 8400MHz which is value 3 in cfgcr1.central_freq.
Looks good to me.
Reviewed-by: Manasi Navare
On Tue, Nov 14, 2017 at 11:47:53AM -0800, Rodrigo Vivi wrote:
> "Display software must leave this field at the default va
== Series Details ==
Series: series starting with [1/2] lib/gt: Mark up engine classes
URL : https://patchwork.freedesktop.org/series/33777/
State : warning
== Summary ==
Test kms_setmode:
Subgroup basic:
pass -> FAIL (shard-hsw) fdo#99912
Test drv_module_re
On Tue, Nov 14, 2017 at 06:12:41PM -0200, Paulo Zanoni wrote:
> Em Qui, 2017-11-02 às 17:17 +0200, Ville Syrjala escreveu:
> > From: Ville Syrjälä
> >
> > Apparently there are some machines that put semi-sensible looking
> > values
> > into the stolen "reserved" base and size, except those values
Em Qui, 2017-11-02 às 17:17 +0200, Ville Syrjala escreveu:
> From: Ville Syrjälä
>
> Now that we should be properly filtering out the cases when the
> stolen
> reserved area is disabled, let's convert the debug message about a
> misplaced reserved area into an error.
Worst that could happen is t
== Series Details ==
Series: tools/error_decode: Print ASCII user buffers (rev2)
URL : https://patchwork.freedesktop.org/series/33673/
State : warning
== Summary ==
Test kms_busy:
Subgroup extended-modeset-hang-oldfb-with-reset-render-c:
pass -> DMESG-WARN (shard-
Em Ter, 2017-11-14 às 20:19 +, Chris Wilson escreveu:
> Quoting Paulo Zanoni (2017-11-14 20:12:41)
> > Em Qui, 2017-11-02 às 17:17 +0200, Ville Syrjala escreveu:
> > > From: Ville Syrjälä
> > >
> > > Apparently there are some machines that put semi-sensible looking
> > > values
> > > into the
== Series Details ==
Series: lib: Dump /sys/kernel/debug/suspend_stats after suspend failure (rev2)
URL : https://patchwork.freedesktop.org/series/33560/
State : success
== Summary ==
Test drv_module_reload:
Subgroup basic-reload:
dmesg-warn -> PASS (shard-hsw) fd
== Series Details ==
Series: drm/i915/selftests: Always initialise err
URL : https://patchwork.freedesktop.org/series/33822/
State : success
== Summary ==
Series 33822v1 drm/i915/selftests: Always initialise err
https://patchwork.freedesktop.org/api/1.0/series/33822/revisions/1/mbox/
Test cha
On Tue, Nov 14, 2017 at 12:09:39PM -0800, Rodrigo Vivi wrote:
> On Tue, Nov 14, 2017 at 08:00:31PM +, Ville Syrjälä wrote:
> > On Tue, Nov 14, 2017 at 11:47:56AM -0800, Rodrigo Vivi wrote:
> > > Spec describe all values in MHz. We handle our
> > > clocks in KHz. This includes the best_dco_centr
== Series Details ==
Series: YCBCR 4:2:0 output support for LSPCON
URL : https://patchwork.freedesktop.org/series/33794/
State : warning
== Summary ==
Test kms_plane:
Subgroup plane-panning-bottom-right-suspend-pipe-a-planes:
pass -> SKIP (shard-hsw)
Test km
== Series Details ==
Series: drm/i915: Misc. PMIC bus access notifier fixes (rev2)
URL : https://patchwork.freedesktop.org/series/32274/
State : success
== Summary ==
Test kms_busy:
Subgroup extended-modeset-hang-newfb-with-reset-render-b:
dmesg-warn -> PASS (shar
On Tue, Nov 14, 2017 at 11:47:58AM -0800, Rodrigo Vivi wrote:
> I confess I never fully understood that previous calculation,
> so maybe this is cannot be called a "fix". But let's follow
> the math that is written on Spec so we have get more confident
> this is what hardware expect.
>
> Cc: Mika
== Series Details ==
Series: drm/i915: Remove pre-production pooled-EU w/a for Broxton
URL : https://patchwork.freedesktop.org/series/33781/
State : success
== Summary ==
Test kms_busy:
Subgroup extended-modeset-hang-newfb-with-reset-render-b:
dmesg-warn -> PASS (
Quoting Paulo Zanoni (2017-11-14 20:12:41)
> Em Qui, 2017-11-02 às 17:17 +0200, Ville Syrjala escreveu:
> > From: Ville Syrjälä
> >
> > Apparently there are some machines that put semi-sensible looking
> > values
> > into the stolen "reserved" base and size, except those values are
> > actually
>
Am 14.11.2017 um 17:27 schrieb Chris Wilson:
Ages ago Rob Clark noted,
"Currently with fence-array, we have a potential deadlock situation. If
we fence_add_callback() on an array-fence, the array-fence's lock is
acquired first, and in it's ->enable_signaling() callback, it will install
cbs on i
Em Qui, 2017-11-02 às 17:17 +0200, Ville Syrjala escreveu:
> From: Ville Syrjälä
>
> Apparently there are some machines that put semi-sensible looking
> values
> into the stolen "reserved" base and size, except those values are
> actually
> outside the stolen memory. There is a bit in the registe
On Tue, Nov 14, 2017 at 08:00:31PM +, Ville Syrjälä wrote:
> On Tue, Nov 14, 2017 at 11:47:56AM -0800, Rodrigo Vivi wrote:
> > Spec describe all values in MHz. We handle our
> > clocks in KHz. This includes the best_dco_centrality that was
> > forgot in the same unity as spec. Consequently we c
== Series Details ==
Series: drm/i915/selftests: Markup __iomem for igt_gem_coherency
URL : https://patchwork.freedesktop.org/series/33819/
State : success
== Summary ==
Series 33819v1 drm/i915/selftests: Markup __iomem for igt_gem_coherency
https://patchwork.freedesktop.org/api/1.0/series/338
On Tue, Nov 14, 2017 at 11:47:56AM -0800, Rodrigo Vivi wrote:
> Spec describe all values in MHz. We handle our
> clocks in KHz. This includes the best_dco_centrality that was
> forgot in the same unity as spec. Consequently we couldn't
> get a good divider for high frequenies. Hence HDMI 2.0 wasn't
Quoting Sagar Arun Kamble (2017-11-14 19:47:05)
>
>
> On 11/15/2017 1:01 AM, Chris Wilson wrote:
> > Quoting Michal Wajdeczko (2017-11-14 19:23:24)
> >> On Tue, 14 Nov 2017 19:27:26 +0100, Chris Wilson
> >> wrote:
> >>
> >>> Quoting Sagar Arun Kamble (2017-11-14 18:19:01)
>
> On 11/14/
== Series Details ==
Series: series starting with [1/2] drm/i915: Drop the unbound page cache upon
idling
URL : https://patchwork.freedesktop.org/series/33817/
State : success
== Summary ==
Series 33817v1 series starting with [1/2] drm/i915: Drop the unbound page cache
upon idling
https://pa
I confess I never fully understood that previous calculation,
so maybe this is cannot be called a "fix". But let's follow
the math that is written on Spec so we have get more confident
this is what hardware expect.
Cc: Mika Kahola
Cc: Manasi Navare
Cc: James Ausmus
Signed-off-by: Rodrigo Vivi
No functional change. Just starting the wrpll fixes
with a clean-up to make units a bit more clear.
Cc: Mika Kahola
Cc: Manasi Navare
Cc: James Ausmus
Signed-off-by: Rodrigo Vivi
---
drivers/gpu/drm/i915/intel_dpll_mgr.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git
Starting on GLK we support HDMI 2.0. So this patch only
extend the work Shashank has made to GLK to CNL.
v2: The version that compiles :/
Cc: Paulo Zanoni
Cc: Shashank Sharma
Cc: Manasi Navare
Signed-off-by: Rodrigo Vivi
---
drivers/gpu/drm/i915/intel_hdmi.c | 7 ---
1 file changed, 4 in
Accordingly to spec "If Kdiv != 2, then Qdiv must be 1."
but we already handle qdiv values properly and this case here
should be spurious. But instead of blindly replacing let's
warn loudly instead. Because it means something was really
wrong on initial setup.
Cc: Mika Kahola
Cc: Manasi Navare
C
- 64 bits is not needed for afe_clock now we don't convert
that to Hz.
- 16 bits is not enough for all dco stuff.
- unsigned is not relevant/needed for all divisors values.
Cc: Mika Kahola
Cc: Manasi Navare
Cc: James Ausmus
Signed-off-by: Rodrigo Vivi
---
drivers/gpu/drm/i915/intel_dpll_mgr
"Display software must leave this field at the default value.
It no longer needs to be configured as part of PLL programming."
We respect this already and we are setting up the default
one line below: "DPLL_CFGCR1_CENTRAL_FREQ".
Also we don't touch anywhere else this central_freq for cnl.
So let'
Spec describe all values in MHz. We handle our
clocks in KHz. This includes the best_dco_centrality that was
forgot in the same unity as spec. Consequently we couldn't
get a good divider for high frequenies. Hence HDMI 2.0 wasn't
working.
This patch also replaces the use of "* KHz(1)" with the val
With these fixes and clean-ups around wrpll plus
extending HDMI 2.0 from GLK to CNL we could finally
get a functional HDMI 2.0 display with 4k60Hz.
Thanks,
Rodrigo.
Rodrigo Vivi (7):
drm/i915/cnl: Remove spurious central_freq.
drm/i915/cnl: Remove useless conversion.
drm/i915/cnl: Fix, simp
smatch does not track initialised values as well as gcc, and this
triggers many warnings by smatch not presented by gcc. Silence smatch by
initialising the error values to -ENODEV, which we use to denote
internal errors. (If we see a selftest fail with a silent -ENODEV, we
know smatch was right!)
On 11/15/2017 1:01 AM, Chris Wilson wrote:
Quoting Michal Wajdeczko (2017-11-14 19:23:24)
On Tue, 14 Nov 2017 19:27:26 +0100, Chris Wilson
wrote:
Quoting Sagar Arun Kamble (2017-11-14 18:19:01)
On 11/14/2017 5:53 PM, Michal Wajdeczko wrote:
On Mon, 13 Nov 2017 09:48:11 +0100, Sagar Arun
On Tue, Nov 14, 2017 at 1:32 PM, Ville Syrjala
wrote:
> From: Ville Syrjälä
>
> These days DRM_MODE_TYPE_USERDEF is used to flag modes defined via the
> kernel command line. Update the docs to reflect that fact.
>
> Signed-off-by: Ville Syrjälä
Acked-by: Alex Deucher
> ---
> include/drm/drm_
== Series Details ==
Series: drm/i915: Fix kerneldocs for intel_audio.c
URL : https://patchwork.freedesktop.org/series/33816/
State : success
== Summary ==
Series 33816v1 drm/i915: Fix kerneldocs for intel_audio.c
https://patchwork.freedesktop.org/api/1.0/series/33816/revisions/1/mbox/
Test k
Quoting Michal Wajdeczko (2017-11-14 19:23:24)
> On Tue, 14 Nov 2017 19:27:26 +0100, Chris Wilson
> wrote:
>
> > Quoting Sagar Arun Kamble (2017-11-14 18:19:01)
> >>
> >>
> >> On 11/14/2017 5:53 PM, Michal Wajdeczko wrote:
> >> > On Mon, 13 Nov 2017 09:48:11 +0100, Sagar Arun Kamble
> >> > wro
On Tue, 14 Nov 2017 19:27:26 +0100, Chris Wilson
wrote:
Quoting Sagar Arun Kamble (2017-11-14 18:19:01)
On 11/14/2017 5:53 PM, Michal Wajdeczko wrote:
> On Mon, 13 Nov 2017 09:48:11 +0100, Sagar Arun Kamble
> wrote:
>> -static void i915_guc_irq_handler(unsigned long data)
>> +static void
On Tue, Nov 14, 2017 at 1:32 PM, Ville Syrjala
wrote:
> From: Ville Syrjälä
>
> Currently we don't sanity check the 3D stereo flags for modes filled out
> by the kernel. Move the check from drm_mode_convert_umode() into
> drm_mode_validate_basic() so that we get the same check going both ways.
>
On Tue, Nov 14, 2017 at 1:32 PM, Ville Syrjala
wrote:
> From: Ville Syrjälä
>
> BUILTIN, CRTC_C, CLOCK_C, and DEFULT mode types are unused. Let's
> refuse to generate them or accept them from userspace either. A
> cursory check didn't reveal any userspace code that would depend
> on these.
>
> Cc
Silence sparse warnings by using __iomem markup and io accessors.
Signed-off-by: Chris Wilson
---
drivers/gpu/drm/i915/selftests/i915_gem_coherency.c | 16
1 file changed, 8 insertions(+), 8 deletions(-)
diff --git a/drivers/gpu/drm/i915/selftests/i915_gem_coherency.c
b/driver
On Tue, Nov 14, 2017 at 1:32 PM, Ville Syrjala
wrote:
> From: Ville Syrjälä
>
> No idea what the DRM_MODE_TYPE_CLOCK_CRTC_C define is supposed to
> achieve. Totally unused so kill if off.
>
> Signed-off-by: Ville Syrjälä
Reviewed-by: Alex Deucher
> ---
> include/drm/drm_modes.h | 3 ---
> 1
On Tue, Nov 14, 2017 at 1:43 PM, Chris Wilson wrote:
> Quoting Ville Syrjala (2017-11-14 18:32:54)
>> From: Ville Syrjälä
>>
>> For some reason drm_mode_set_crtcinfo() does nothing of the mode has
> s/of/if/
With the typo fixed:
Reviewed-by: Alex Deucher
>
>> the DRM_MODE_TYPE_BUILTIN flag set
1 - 100 of 224 matches
Mail list logo
