Existing DDB algorithm divide the DDB wrt data rate,
hence the planes with the less height but same width
will be allocated less blocks and watermark are based
on width which requires more DDB. With this data the flip
may fail.
In new DDB algorithm, the DDB is divided based on
watermark requiremen
On ma, 2016-10-03 at 17:20 +0300, Jani Nikula wrote:
> Only try removing intel_ips if it's actually loaded, and let the errors
> through to the logs if removal fails.
>
> Signed-off-by: Jani Nikula
Reviewed-by: Joonas Lahtinen
Regards, Joonas
--
Joonas Lahtinen
Open Source Technology Center
I
On ma, 2016-10-03 at 17:20 +0300, Jani Nikula wrote:
> Only try removing snd_hda_intel if it's actually loaded, and let the
> errors through to the logs if removal fails. This is a clue if i915
> removal fails later.
>
> Signed-off-by: Jani Nikula
Reviewed-by: Joonas Lahtinen
Regards, Joonas
-
On ma, 2016-10-03 at 17:20 +0300, Jani Nikula wrote:
> Add a helper for checking whether a module is reloaded, using
> lsmod. Also make the grep stricter than before.
>
> Signed-off-by: Jani Nikula
Reviewed-by: Joonas Lahtinen
Regards, Joonas
--
Joonas Lahtinen
Open Source Technology Center
I
On 3 October 2016 at 23:39, Daniel Vetter wrote:
> Hi Dave,
>
> As promised another pull request. A bit late because flu.
> - generic pipe crc from Tomeu, required a backmerge to apply
> - fixes for my fumbled drm_plane.c extraction
> - display_info cleanup/fixes from Ville
> - misc stuff all over
On Mon, 2016-10-03 at 13:50 +0300, Jani Nikula wrote:
> On Fri, 30 Sep 2016, Rodrigo Vivi wrote:
> > According to spec: "KBL re-uses SKL values, except where
> > specific KBL values are listed."
> >
> > And recently spec has changed adding different table for Display Port only.
> > But for all SKU
== Series Details ==
Series: drm/i915/dp/psr: Enable n + 1 active frame capture for PSR
URL : https://patchwork.freedesktop.org/series/13249/
State : warning
== Summary ==
Series 13249v1 drm/i915/dp/psr: Enable n + 1 active frame capture for PSR
https://patchwork.freedesktop.org/api/1.0/series
On Thu, Sep 29, 2016 at 04:17:06PM -0700, Manasi Navare wrote:
> On Thu, Sep 29, 2016 at 09:05:01AM -0700, Manasi Navare wrote:
> > On Thu, Sep 29, 2016 at 06:48:43PM +0300, Jani Nikula wrote:
> > > On Thu, 29 Sep 2016, Ville Syrjälä wrote:
> > > > On Thu, Sep 29, 2016 at 12:44:19PM +0100, Chris W
On some system + panel combinations the vblank interval is too
short for PSR setup to occur. While reading the eDP v1.3 spec
I saw that Figure 4-7 describes a solution to this very problem
by configuring PSR to capture the second active frame after the
PSR entry indication in order to provide more
== Series Details ==
Series: HuC Loading Patches
URL : https://patchwork.freedesktop.org/series/13244/
State : warning
== Summary ==
Series 13244v1 HuC Loading Patches
https://patchwork.freedesktop.org/api/1.0/series/13244/revisions/1/mbox/
Test drv_module_reload_basic:
skip
From: Peter Antoine
This patch returns the GuC status to the caller. It is used so
that the userspace knows if the GuC has been loaded.
v4: rebase.
v5: rebased.
Signed-off-by: Anusha Srivatsa
Signed-off-by: Peter Antoine
---
drivers/gpu/drm/i915/i915_drv.c | 4
drivers/gpu/drm/
From: Peter Antoine
Rename some of the GuC fw loading code to make them more general. We
will utilise them for HuC loading as well.
s/intel_guc_fw/intel_uc_fw/g
s/GUC_FIRMWARE/UC_FIRMWARE/g
Struct intel_guc_fw is renamed to intel_uc_fw. Prefix of tts members,
such as 'guc' or 'guc_fw'
From: Peter Antoine
The HuC authentication is done by host2guc call. The HuC RSA keys
are sent to GuC for authentication.
v2: rebased on top of drm-intel-nightly.
changed name format and upped version 1.7.
v3: rebased on top of drm-intel-nightly.
v4: changed wait_for_automic to wait_for
v5:
From: Peter Antoine
The HuC loading process is similar to GuC. The intel_uc_fw_fetch()
is used for both cases.
HuC loading needs to be before GuC loading. The WOPCM setting must
be done early before loading any of them.
v2: rebased on-top of drm-intel-nightly.
removed if(HAS_GUC()) before t
From: Peter Antoine
This patch will allow for getparams to return the status of the HuC.
As the HuC has to be validated by the GuC this patch uses the validated
status to show when the HuC is loaded and ready for use. You cannot use
the loaded status as with the GuC as the HuC is verified after i
From: Peter Antoine
Add debugfs entry for HuC loading status check.
v2: rebase on-top of drm-intel-nightly.
v3: rebased again.
v7: rebased.
v8: rebased.
Tested-by: Xiang Haihao
Signed-off-by: Anusha Srivatsa
Signed-off-by: Alex Dai
Signed-off-by: Peter Antoine
Reviewed-by: Dave Gordon
---
From: Peter Antoine
This patch adds the HuC Loading for the BXT.
Version 1.7 of the HuC firmware.
v2: rebased.
v3: rebased.
changed file name to match the install package format.
v7: rebased.
v8: rebased.
Signed-off-by: Anusha Srivatsa
Signed-off-by: Peter Antoine
Reviewed-by: David Gordo
From: Peter Antoine
HuC firmware css header has almost exactly same definition as GuC
firmware except for the sw_version. Also, add a new member fw_type
into intel_uc_fw to indicate what kind of fw it is. So, the loader
will pull right sw_version from header.
v2: rebased on-top of drm-intel-nigh
These patches add HuC loading support. The userspace
patches that check for a fully loaded HuC firmware and use
it can be found at:
https://lists.freedesktop.org/archives/libva/2016-September/004554.html
https://lists.freedesktop.org/archives/libva/2016-September/004555.html
v2: rebased.
Peter
On Mon, Oct 3, 2016 at 3:06 PM, Petri Latvala wrote:
> Add myself, remove Daniel by request.
>
> Signed-off-by: Petri Latvala
> ---
>
> For the record.
>
> Daniel requested to be removed from the list (Daniel, please confirm,
> that's how I understood), and I'm going to be maintaining IGT with
>
== Series Details ==
Series: series starting with [1/8] drm/i915: Share the computation of ring size
for RING_CTL register
URL : https://patchwork.freedesktop.org/series/13231/
State : warning
== Summary ==
Series 13231v1 Series without cover letter
https://patchwork.freedesktop.org/api/1.0/s
Some of the Intel platforms have odd numbers of LUT entries and we
need to tests a couple of values around the expected result. Bring
back the CRC equal function we need that doesn't trigger an assert
right away, while we still assert if we can't find a correct result in
the outter loop.
v2: updat
Thes might leave us some more breadcrumbs on module removal errors.
BR,
Jani.
Jani Nikula (3):
igt/drv_module_reload_basic: add helper for checking module reloaded
igt/drv_module_reload_basic: let snd_hda_intel removal errors through
igt/drv_module_reload_basic: let intel_ips removal errors
Only try removing snd_hda_intel if it's actually loaded, and let the
errors through to the logs if removal fails. This is a clue if i915
removal fails later.
Signed-off-by: Jani Nikula
---
tests/drv_module_reload_basic | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/tests/
Add a helper for checking whether a module is reloaded, using
lsmod. Also make the grep stricter than before.
Signed-off-by: Jani Nikula
---
tests/drv_module_reload_basic | 8 +++-
1 file changed, 7 insertions(+), 1 deletion(-)
diff --git a/tests/drv_module_reload_basic b/tests/drv_module_r
Only try removing intel_ips if it's actually loaded, and let the errors
through to the logs if removal fails.
Signed-off-by: Jani Nikula
---
tests/drv_module_reload_basic | 6 --
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/tests/drv_module_reload_basic b/tests/drv_module_re
== Series Details ==
Series: series starting with [CI,1/3] drm/i915: Just clear the mmiodebug before
a register access
URL : https://patchwork.freedesktop.org/series/13230/
State : warning
== Summary ==
Series 13230v1 Series without cover letter
https://patchwork.freedesktop.org/api/1.0/serie
On Mon, Oct 03, 2016 at 04:14:39PM +0300, Mika Kuoppala wrote:
> Chris Wilson writes:
>
> > Check that there was not a late recovery between us declaring the GPU
> > hung and processing the reset. If the GPU did recover by itself, let the
> > request remain on the active list and see if it hangs
On Mon, Oct 3, 2016 at 11:35 AM, Jani Nikula
wrote:
>>220/**
>>221 * @DRM_FB_HELPER_DEFAULT_OPS:
>
> Superfluous @.
Stefan, pls fix this and run
$ make htmldocs
to make sure the output looks correct. Cargo-culting kerneldoc without
checking the output doesn't work too we
Hi Dave,
As promised another pull request. A bit late because flu.
- generic pipe crc from Tomeu, required a backmerge to apply
- fixes for my fumbled drm_plane.c extraction
- display_info cleanup/fixes from Ville
- misc stuff all over
Cheers, Daniel
The following changes since commit c0d5fb4d0
On Mon, 03 Oct 2016, Petri Latvala wrote:
> Add myself, remove Daniel by request.
>
> Signed-off-by: Petri Latvala
Acked-by: Jani Nikula
> ---
>
> For the record.
>
> Daniel requested to be removed from the list (Daniel, please confirm,
> that's how I understood), and I'm going to be maintain
Chris Wilson writes:
> Check that there was not a late recovery between us declaring the GPU
> hung and processing the reset. If the GPU did recover by itself, let the
> request remain on the active list and see if it hangs again!
>
Did you see this in action? Makes sense to recheck
after reset.
Chris Wilson writes:
> Whilst we reset the GPU, we want to prevent execlists from submitting
> new work (which it does via an interrupt handler). To achieve this we
> disable the irq (and drain the irq tasklet) around the reset. When we
> enable it again afters, the interrupt queue should be empt
Chris Wilson writes:
> After a GPU reset, we want to replay our queue of requests. However, the
> GPU reset clobbered the state and we only fixup the state for the
> guitly
s/guitly/guilty.
Reviewed-by: Mika Kuoppala
> request - and engines deemed innocent we try to leave untouched so that
>
Add myself, remove Daniel by request.
Signed-off-by: Petri Latvala
---
For the record.
Daniel requested to be removed from the list (Daniel, please confirm,
that's how I understood), and I'm going to be maintaining IGT with
Marius until Marius steps down.
MAINTAINERS | 2 +-
1 file changed, 1
Chris Wilson writes:
> Since both legacy and execlists want to poopulate the RING_CTL
> register,
A bit too graphic, lets switch to populate.
Reviewed-by: Mika Kuoppala
> share the computation of the right bits for the ring->size. We can then
> stop masking errors and explicitly forbid them d
On Mon, Oct 03, 2016 at 01:52:39PM +0100, Chris Wilson wrote:
> Since both legacy and execlists want to poopulate the RING_CTL register,
> share the computation of the right bits for the ring->size. We can then
> stop masking errors and explicitly forbid them during creation!
>
> Signed-off-by: Ch
On Mon, Oct 03, 2016 at 01:44:14PM +0100, Tvrtko Ursulin wrote:
> >+static void i915_gem_cleanup_engine(struct intel_engine_cs *engine)
>
> I am aware this is a late comment, but wanted to say that the name
> above is not ideal since we have both i915_gem_cleanup_engines and
> dev_priv->gt.cleanup
On Braswell, at least, we observe that the context image is written in
multiple phases. The first phase is to clear the register state, and
subsequently rewrite it. A GPU reset at the right moment can interrupt
the context update leaving it corrupt, and our update of the RING_HEAD
is not sufficient
Knowing where the RINGs are pointing is extremely useful in diagnosing
if the engines are executing the ringbuffers you expect - and igt may be
suppressing the usual method of looking in the GPU error state.
Signed-off-by: Chris Wilson
Cc: Mika Kuoppala
---
drivers/gpu/drm/i915/i915_debugfs.c
Signed-off-by: Chris Wilson
---
drivers/gpu/drm/i915/i915_debugfs.c | 12
1 file changed, 12 insertions(+)
diff --git a/drivers/gpu/drm/i915/i915_debugfs.c
b/drivers/gpu/drm/i915/i915_debugfs.c
index 9c95ce73f2aa..39b76c24c84f 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/
Whilst we reset the GPU, we want to prevent execlists from submitting
new work (which it does via an interrupt handler). To achieve this we
disable the irq (and drain the irq tasklet) around the reset. When we
enable it again afters, the interrupt queue should be empty and we can
reinitialise from
After a GPU reset, we want to replay our queue of requests. However, the
GPU reset clobbered the state and we only fixup the state for the guitly
request - and engines deemed innocent we try to leave untouched so that
we recover as completely as possible. However, we need to clear the sw
tracking o
Since both legacy and execlists want to poopulate the RING_CTL register,
share the computation of the right bits for the ring->size. We can then
stop masking errors and explicitly forbid them during creation!
Signed-off-by: Chris Wilson
Cc: Mika Kuoppala
---
drivers/gpu/drm/i915/i915_reg.h
Include the position of the active request in the ring, and display that
alongside the current RING registers (on a GPU hang).
Signed-off-by: Chris Wilson
Cc: Mika Kuoppala
---
drivers/gpu/drm/i915/i915_drv.h | 3 +++
drivers/gpu/drm/i915/i915_gpu_error.c | 9 +++--
2 files changed, 1
Check that there was not a late recovery between us declaring the GPU
hung and processing the reset. If the GPU did recover by itself, let the
request remain on the active list and see if it hangs again!
Signed-off-by: Chris Wilson
Cc: Mika Kuoppala
---
drivers/gpu/drm/i915/i915_gem.c | 3 +++
When we enable the per-register access mmiodebug, it is to detect which
access is illegal. Reporting on earlier untraced access outside of the
mmiodebug does not help debugging (as the suspicion is immediately put
upon the current register which is not at fault)!
References: https://bugs.freedeskt
When decoding the semaphores inside hangcheck, we need to use the hw-id
and not the local array index.
Fixes: de1add360522 ("drm/i915: Decouple execbuf uAPI ...")
Testcase: igt/gem_exec_whisper/hang # gen6-7
Signed-off-by: Chris Wilson
Cc: Mika Kuoppala
Cc: Tvrtko Ursulin
Cc: Daniel Vetter
Cc:
We use obj->phys_handle to choose the pread/pwrite path, but as
obj->phys_handle is a union with obj->userptr, we then mistakenly use
the phys_handle path for userptr objects within pread/pwrite.
Testcase: igt/gem_userptr_blits/forbidden-operations
Bugzilla: https://bugs.freedesktop.org/show_bug.c
On 07/09/2016 15:45, Chris Wilson wrote:
Update reset path in preparation for engine reset which requires
identification of incomplete requests and associated context and fixing
their state so that engine can resume correctly after reset.
The request that caused the hang will be skipped and hea
Chris Wilson writes:
> Whilst we reset the GPU, we want to prevent execlists from submitting
> new work (which it does via an interrupt handler). To achieve this we
> disable the irq (and drain the irq tasklet) around the reset. When we
> enable it again afters, the interrupt queue should be empt
Chris Wilson writes:
> On Braswell, at least, we observe that the context image is written in
> multiple phases. The first phase is to clear the register state, and
> subsequently rewrite it. A GPU reset at the right moment can interrupt
> the context update leaving it corrupt, and our update of
On Mon, 2016-10-03 at 10:55 +0300, ville.syrj...@linux.intel.com wrote:
> From: Ville Syrjälä
>
> We can't rely on connector->status in the detect() hook if the long hpd
> was already handled by the dig_port_work as that won't update
> connector->status. Thus we have to defer the long hpd handlin
On Wed, Sep 21, 2016 at 07:47:37PM +0530, Maiti, Nabendu Bikash wrote:
> Hi,
>
>
> On 9/20/2016 1:55 PM, Ville Syrjälä wrote:
> > On Tue, Aug 30, 2016 at 10:30:54AM +0530, Nabendu Maiti wrote:
> >> Following patch series add pipe scaler functionality in atomic path.The
> >> pipe
> >> scaler can
On Fri, 30 Sep 2016, Rodrigo Vivi wrote:
> According to spec: "KBL re-uses SKL values, except where
> specific KBL values are listed."
>
> And recently spec has changed adding different table for Display Port only.
> But for all SKUs (H,S,U,Y) we have slightly different values.
>
> Cc: Manasi Nava
Chris Wilson writes:
> When we enable the per-register access mmiodebug, it is to detect which
> access is illegal. Reporting on earlier untraced access outside of the
> mmiodebug does not help debugging (as the suspicion is immediately put
> upon the current register which is not at fault)!
>
>
On Mon, Oct 03, 2016 at 01:09:31PM +0300, Mika Kuoppala wrote:
> Chris Wilson writes:
>
> > When we enable the per-register access mmiodebug, it is to detect which
> > access is illegal. Reporting on earlier untraced access outside of the
> > mmiodebug does not help debugging (as the suspicion is
Chris Wilson writes:
> When we enable the per-register access mmiodebug, it is to detect which
> access is illegal. Reporting on earlier untraced access outside of the
> mmiodebug does not help debugging (as the suspicion is immediately put
> upon the current register which is not at fault)!
>
I
On Fri, 30 Sep 2016, kbuild test robot wrote:
> tree: git://anongit.freedesktop.org/drm-intel topic/drm-misc
> head: 6c4d6f9f997c5dafccb54b52167f0c4d0ea37874
> commit: 6c4d6f9f997c5dafccb54b52167f0c4d0ea37874 [3/3] drm/fb-helper: add
> DRM_FB_HELPER_DEFAULT_OPS for fb_ops
> reproduce: make ht
== Series Details ==
Series: Forcewake binary search & code shrink (rev6)
URL : https://patchwork.freedesktop.org/series/13080/
State : warning
== Summary ==
Series 13080v6 Forcewake binary search & code shrink
https://patchwork.freedesktop.org/api/1.0/series/13080/revisions/6/mbox/
Test kms_
From: Tvrtko Ursulin
Also verify the order at runtime. This was we can start using
binary search on it in a following patch.
v2: Add comment on the sorted array and only check it when
debug option is enabled.
v3: Use IS_ENABLED. (Chris Wilson)
Signed-off-by: Tvrtko Ursulin
Reviewed-by: Jo
From: Tvrtko Ursulin
If we store this in the uncore structure we are on a good way to
show more commonality between the per-platform implementations.
v2: Constify table pointer and correct coding style. (Chris Wilson)
v3: Rebase.
Signed-off-by: Tvrtko Ursulin
Reviewed-by: Chris Wilson
---
dr
From: Tvrtko Ursulin
Sorting the tables (verified at runtime to help during
development) is another prerequisite for interesting
work which will follow.
v2:
* Remove const away cast and improve comments. (Chris Wilson)
* Check tables only when debug option is enabled.
v3: Use IS_ENABLED. (Chr
On 03/10/2016 09:05, Joonas Lahtinen wrote:
On pe, 2016-09-30 at 19:08 +0100, Chris Wilson wrote:
On Fri, Sep 30, 2016 at 06:48:48PM +0100, Tvrtko Ursulin wrote:
+static int mmio_reg_cmp(const void *key, const void *elt)
+{
+ u32 offset = (u32)(unsigned long)key;
+ i915_reg_t *reg
== Series Details ==
Series: series starting with [1/2] drm/i915: Move long hpd handling into the
hotplug work
URL : https://patchwork.freedesktop.org/series/13225/
State : success
== Summary ==
Series 13225v1 Series without cover letter
https://patchwork.freedesktop.org/api/1.0/series/13225/
On pe, 2016-09-30 at 19:08 +0100, Chris Wilson wrote:
> On Fri, Sep 30, 2016 at 06:48:48PM +0100, Tvrtko Ursulin wrote:
> > +static int mmio_reg_cmp(const void *key, const void *elt)
> > +{
> > + u32 offset = (u32)(unsigned long)key;
> > + i915_reg_t *reg = (i915_reg_t *)elt;
> > +
> > + if
On pe, 2016-09-30 at 18:48 +0100, Tvrtko Ursulin wrote:
>
> +find_fw_domain(u32 offset, const struct intel_forcewake_range *ranges,
> + unsigned int num_ranges)
Gotta remind that generally "fw" is associated with firmware.
"forcewake" might be appropriate word in function names too.
>
From: Ville Syrjälä
We can't rely on connector->status in the detect() hook if the long hpd
was already handled by the dig_port_work as that won't update
connector->status. Thus we have to defer the long hpd handling entirely
until the hotplug work runs to avoid the double long hpd handling
the "
From: Ville Syrjälä
Allow returning "connected" or "unknown" connector status for DP branch
devices that don't have an EDID. Currently we'd claim the thing as
"disconnected" if there is no EDID.
This stuff used to broken already, I think, but it got more broken by
commit f21a21983ef1 ("drm/i915:
On pe, 2016-09-30 at 11:38 +0100, Chris Wilson wrote:
> dma_buf_export() adds a reference to the owning module to the dmabuf (to
> prevent the driver from being unloaded whilst a third party still refers
> to the dmabuf). However, drm_gem_prime_export() was passing its own
> THIS_MODULE (i.e. drm.k
On la, 2016-10-01 at 20:42 +0100, Chris Wilson wrote:
> When decoding the semaphores inside hangcheck, we need to use the hw-id
> and not the local array index.
>
> Fixes: de1add360522 ("drm/i915: Decouple execbuf uAPI ...")
> Testcase: igt/gem_exec_whisper/hang # gen6-7
> Signed-off-by: Chris Wil
On pe, 2016-09-30 at 18:31 +0100, Chris Wilson wrote:
> We use obj->phys_handle to choose the pread/pwrite path, but as
> obj->phys_handle is a union with obj->userptr, we then mistakenly use
> the phys_handle path for userptr objects within pread/pwrite.
>
> Testcase: igt/gem_userptr_blits/forbid
72 matches
Mail list logo