An important property for multi-client systems is that each client gets
a 'fair' allotment of system time. (Where fairness is at the whim of the
context properties, such as priorities.) This test forks N independent
clients (albeit they happen to share a single vm), and does an equal
amount of work
An important property for multi-client systems is that each client gets
a 'fair' allotment of system time. (Where fairness is at the whim of the
context properties, such as priorities.) This test forks N independent
clients (albeit they happen to share a single vm), and does an equal
amount of work
An important property for multi-client systems is that each client gets
a 'fair' allotment of system time. (Where fairness is at the whim of the
context properties, such as priorities.) This test forks N independent
clients (albeit they happen to share a single vm), and does an equal
amount of work
Chris Wilson writes:
> Quoting Mika Kuoppala (2020-06-01 15:56:55)
>> Chris Wilson writes:
>>
>> > As we rewrite the batches on the fly to implement the non-preemptible
>> > lock, we need to tell Tigerlake to read the batch afresh each time.
>> > Amusingly, the disable is a part of an arb-check
Chris Wilson writes:
> Pull the routines for writing CS packets out of intel_ring_submission
> into their own files. These are low level operations for building CS
> instructions, rather than the logic for filling the global ring buffer
> with requests, and we will wnat to reuse them outside of t
Chris Wilson writes:
> Use the central mechanism for recording and verifying that we restore
> the w/a for the older devices as well.
>
> Signed-off-by: Chris Wilson
Reviewed-by: Mika Kuoppala
> ---
> .../gpu/drm/i915/gt/intel_ring_submission.c | 28 -
> drivers/gpu/drm/i91
Chris Wilson writes:
> An important property for multi-client systems is that each client gets
> a 'fair' allotment of system time. (Where fairness is at the whim of the
> context properties, such as priorities.) This test forks N independent
> clients (albeit they happen to share a single vm), a
Quoting Mika Kuoppala (2020-06-02 10:18:34)
> Chris Wilson writes:
>
> > An important property for multi-client systems is that each client gets
> > a 'fair' allotment of system time. (Where fairness is at the whim of the
> > context properties, such as priorities.) This test forks N independent
On Sat, May 30, 2020 at 06:22:58AM +0300, Laurent Pinchart wrote:
> Hi Daniel,
>
> Thank you for the patch.
>
> On Wed, May 27, 2020 at 11:53:32AM +0200, Daniel Vetter wrote:
> > Only when vblanks are supported ofc.
> >
> > Some drivers do this already, but most unfortunately missed it. This
> >
Op 12-05-2020 om 11:08 schreef Christian König:
> Am 12.05.20 um 10:59 schrieb Daniel Vetter:
>> But only for non-zero timeout, to avoid false positives.
>>
>> One question here is whether the might_sleep should be unconditional,
>> or only for real timeouts. I'm not sure, so went with the more
>>
Only when vblanks are supported ofc.
Some drivers do this already, but most unfortunately missed it. This
opens up bugs after driver load, before the crtc is enabled for the
first time. syzbot spotted this when loading vkms as a secondary
output. Given how many drivers are buggy it's best to solve
This is already done as part of the drm_atomic_helper_shutdown(),
and in that case only for the crtc which are actually on.
Signed-off-by: Daniel Vetter
Cc: Liviu Dudau
Cc: Brian Starkey
Cc:
---
drivers/gpu/drm/arm/malidp_drv.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/drivers/gpu/drm
This is already taken care of by drm_atomic_helper_shutdown(), and
in that case only for the CRTC which are actually on.
Only tricky bit here is that we kill the interrupt handling before we
shut down crtc, so need to reorder that.
Signed-off-by: Daniel Vetter
Cc: Liviu Dudau
Cc: Brian Starkey
This is already done as part of the drm_atomic_helper_shutdown(),
and in that case only for the crtc which are actually on.
v2: I overlooked that malidp also needs to have it's interrupt shut
down reordered.
Signed-off-by: Daniel Vetter
Cc: Liviu Dudau
Cc: Brian Starkey
---
drivers/gpu/drm/ar
For reasons that be, the HW only allows usersace to read its own
CTX_TIMESTAMP [context local HW runtime] on rcs. Make it available for
all by adding it to the whitelists.
Signed-off-by: Chris Wilson
---
This probably means the change occurred in the glk/cfl timeframe...
---
drivers/gpu/drm/i915
---
tests/intel-ci/fast-feedback.testlist | 163 +-
1 file changed, 3 insertions(+), 160 deletions(-)
diff --git a/tests/intel-ci/fast-feedback.testlist
b/tests/intel-ci/fast-feedback.testlist
index 04f6affcf..9cf460894 100644
--- a/tests/intel-ci/fast-feedback.testlist
+
An important property for multi-client systems is that each client gets
a 'fair' allotment of system time. (Where fairness is at the whim of the
context properties, such as priorities.) This test forks N independent
clients (albeit they happen to share a single vm), and does an equal
amount of work
== Series Details ==
Series: drm/i915/tgl: Add Wa_1409371443
URL : https://patchwork.freedesktop.org/series/77892/
State : success
== Summary ==
CI Bug Log - changes from CI_DRM_8568_full -> Patchwork_17837_full
Summary
---
**SUCCESS
== Series Details ==
Series: drm/i915/dsi: Dont forget to clean up the connector on error (rev3)
URL : https://patchwork.freedesktop.org/series/77011/
State : failure
== Summary ==
CI Bug Log - changes from CI_DRM_8568_full -> Patchwork_17838_full
==
== Series Details ==
Series: series starting with [1/3] drm/atomic-helper: reset vblank on crtc
reset (rev2)
URL : https://patchwork.freedesktop.org/series/77908/
State : warning
== Summary ==
$ dim checkpatch origin/drm-tip
c9cd581c2623 drm/atomic-helper: reset vblank on crtc reset
-:247: WA
Cometlake is small refresh of Coffeelake, but since we have found out a
difference in the plaforms, we need to identify the separate platforms.
Since we previously took Coffeelake/Cometlake as identical, update all
IS_COFFEELAKE() to also include IS_COMETLAKE().
Signed-off-by: Chris Wilson
---
For reasons that be, the HW only allows usersace to read its own
CTX_TIMESTAMP [context local HW runtime] on rcs. Make it available for
all by adding it to the whitelists.
v2: The change took effect from Cometlake.
Signed-off-by: Chris Wilson
---
drivers/gpu/drm/i915/gt/intel_workarounds.c | 25
On 2020-06-01 at 18:19:44 +0530, Shankar, Uma wrote:
>
>
> > -Original Message-
> > From: Intel-gfx On Behalf Of
> > Anshuman Gupta
> > Sent: Monday, June 1, 2020 3:45 PM
> > To: intel-gfx@lists.freedesktop.org
> > Cc: sta...@vger.kernel.org
> > Subject: [Intel-gfx] [RFC] drm/i915: lpsp
The pins on the Bay Trail SoC have separate input-buffer and output-buffer
enable bits and a read of the level bit of the value register will always
return the value from the input-buffer.
The BIOS of a device may configure a pin in output-only mode, only enabling
the output buffer, and write 1 to
Cometlake is small refresh of Coffeelake, but since we have found out a
difference in the plaforms, we need to identify the separate platforms.
Since we previously took Coffeelake/Cometlake as identical, update all
IS_COFFEELAKE() to also include IS_COMETLAKE().
Signed-off-by: Chris Wilson
---
== Series Details ==
Series: series starting with [1/3] drm/atomic-helper: reset vblank on crtc
reset (rev2)
URL : https://patchwork.freedesktop.org/series/77908/
State : success
== Summary ==
CI Bug Log - changes from CI_DRM_8571 -> Patchwork_17839
===
On 2020-06-01 at 17:11:32 +0300, Ville Syrjälä wrote:
> On Mon, Jun 01, 2020 at 03:45:16PM +0530, Anshuman Gupta wrote:
> > Gen12 hw are failing to enable lpsp configuration due to PG3 was left on
> > due to valid usgae count of POWER_DOMAIN_AUDIO.
> > It is not required to get POWER_DOMAIN_AUDIO r
Nirmoy please keep in mind that your current implementation doesn't
fully solve the issue the test case is exercising.
In other words what you have implement is fast skipping of fragmented
address space for bottom-up and top-down.
But what this test here exercises is the fast skipping of alig
Hi Daniel,
On Tue, Jun 02, 2020 at 11:51:39AM +0200, Daniel Vetter wrote:
> This is already done as part of the drm_atomic_helper_shutdown(),
> and in that case only for the crtc which are actually on.
>
I'm pretty sure that it didn't used to be the case when I wrote the code
and I was hitting wa
== Series Details ==
Series: drm/i915/gt: Make the CTX_TIMESTAMP readable on !rcs
URL : https://patchwork.freedesktop.org/series/77910/
State : failure
== Summary ==
CI Bug Log - changes from CI_DRM_8571 -> Patchwork_17840
Summary
---
On Tue, Jun 02, 2020 at 11:51:40AM +0200, Daniel Vetter wrote:
> This is already taken care of by drm_atomic_helper_shutdown(), and
> in that case only for the CRTC which are actually on.
>
> Only tricky bit here is that we kill the interrupt handling before we
> shut down crtc, so need to reorder
Hi Daniel,
On Tue, Jun 02, 2020 at 11:55:05AM +0200, Daniel Vetter wrote:
> This is already done as part of the drm_atomic_helper_shutdown(),
> and in that case only for the crtc which are actually on.
>
> v2: I overlooked that malidp also needs to have it's interrupt shut
> down reordered.
Got
> -Original Message-
> From: Gupta, Anshuman
> Sent: Tuesday, June 2, 2020 5:37 PM
> To: Shankar, Uma
> Cc: intel-gfx@lists.freedesktop.org; sta...@vger.kernel.org
> Subject: Re: [Intel-gfx] [RFC] drm/i915: lpsp with hdmi/dp outputs
>
> On 2020-06-01 at 18:19:44 +0530, Shankar, Uma wro
== Series Details ==
Series: series starting with drm/i915: Identify Cometlake platform (rev2)
URL : https://patchwork.freedesktop.org/series/77916/
State : warning
== Summary ==
$ dim checkpatch origin/drm-tip
b57ebeadf460 drm/i915: Identify Cometlake platform
-:354: CHECK:MACRO_ARG_REUSE: Ma
== Series Details ==
Series: series starting with drm/i915: Identify Cometlake platform (rev2)
URL : https://patchwork.freedesktop.org/series/77916/
State : warning
== Summary ==
$ dim sparse --fast origin/drm-tip
Sparse version: v0.6.0
Fast mode used, each commit won't be checked separately.
== Series Details ==
Series: series starting with drm/i915: Identify Cometlake platform (rev2)
URL : https://patchwork.freedesktop.org/series/77916/
State : failure
== Summary ==
CI Bug Log - changes from CI_DRM_8571 -> Patchwork_17841
Summ
== Series Details ==
Series: pinctrl: baytrail: Fix pin being driven low for a while on
gpiod_get(..., GPIOD_OUT_HIGH)
URL : https://patchwork.freedesktop.org/series/77917/
State : failure
== Summary ==
CI Bug Log - changes from CI_DRM_8571 -> Patchwork_17842
=
Quoting Zhenyu Wang (2020-05-28 06:35:59)
>
> Hi,
>
> Here's two queued warning fixes for gvt-next. One is for clang warning
> on debug only function and another one from coccicheck to use ARRAY_SIZE.
Pulled now, thanks for the PR.
Regards, Joonas
>
> Thanks
> --
> The following changes since
Cometlake is small refresh of Coffeelake, but since we have found out a
difference in the plaforms, we need to identify the separate platforms.
Since we previously took Coffeelake/Cometlake as identical, update all
IS_COFFEELAKE() to also include IS_COMETLAKE().
Signed-off-by: Chris Wilson
---
For reasons that be, the HW only allows usersace to read its own
CTX_TIMESTAMP [context local HW runtime] on rcs. Make it available for
all by adding it to the whitelists.
v2: The change took effect from Cometlake.
v3: Ignore timestamps that autoincrement when validating the whitelist
Signed-off-
Am 02.06.20 um 16:13 schrieb Nirmoy:
Hi Christian,
On 6/2/20 2:47 PM, Christian König wrote:
Nirmoy please keep in mind that your current implementation doesn't
fully solve the issue the test case is exercising.
In other words what you have implement is fast skipping of fragmented
address sp
== Series Details ==
Series: series starting with [1/2] drm/i915: Identify Cometlake platform
URL : https://patchwork.freedesktop.org/series/77922/
State : warning
== Summary ==
$ dim checkpatch origin/drm-tip
cb7a71f01dd4 drm/i915: Identify Cometlake platform
-:367: CHECK:MACRO_ARG_REUSE: Mac
== Series Details ==
Series: series starting with [1/2] drm/i915: Identify Cometlake platform
URL : https://patchwork.freedesktop.org/series/77922/
State : warning
== Summary ==
$ dim sparse --fast origin/drm-tip
Sparse version: v0.6.0
Fast mode used, each commit won't be checked separately.
-
The reset member in i915_params was previously changed to unsigned, but
this failed to change the actual module parameter.
Fixes: aae970d8454b ("drm/i915: Mark i915.reset as unsigned")
Cc: Chris Wilson
Cc: Mika Kuoppala
Signed-off-by: Jani Nikula
---
drivers/gpu/drm/i915/i915_params.c | 2 +-
Quoting Jani Nikula (2020-06-02 16:11:26)
> The reset member in i915_params was previously changed to unsigned, but
> this failed to change the actual module parameter.
>
> Fixes: aae970d8454b ("drm/i915: Mark i915.reset as unsigned")
> Cc: Chris Wilson
> Cc: Mika Kuoppala
> Signed-off-by: Jani
On Tue, Jun 02, 2020 at 02:21:30PM +0200, Hans de Goede wrote:
> The pins on the Bay Trail SoC have separate input-buffer and output-buffer
> enable bits and a read of the level bit of the value register will always
> return the value from the input-buffer.
>
> The BIOS of a device may configure a
On Tue, Jun 02, 2020 at 06:23:17PM +0300, Andy Shevchenko wrote:
> On Tue, Jun 02, 2020 at 02:21:30PM +0200, Hans de Goede wrote:
...
> Wouldn't be simple below fix the issue?
>
> @@ -1171,14 +1171,10 @@ static int byt_gpio_direction_input(struct gpio_chip
> *chip, unsigned int offset)
> stati
== Series Details ==
Series: series starting with [1/2] drm/i915: Identify Cometlake platform
URL : https://patchwork.freedesktop.org/series/77922/
State : failure
== Summary ==
CI Bug Log - changes from CI_DRM_8572 -> Patchwork_17843
Summa
On Fri, 2020-05-22 at 13:26 -0700, Vivek Kasireddy wrote:
> If an error is encountered during the DSI initialization setup, the
> drm connector object also needs to be cleaned up along with the encoder.
> The error can happen due to a missing mode in the VBT or for other
> reasons.
>
> v2: Rephras
== Series Details ==
Series: drm/i915/params: fix i915.reset module param type
URL : https://patchwork.freedesktop.org/series/77923/
State : warning
== Summary ==
$ dim checkpatch origin/drm-tip
b693251369b3 drm/i915/params: fix i915.reset module param type
-:25: CHECK:PARENTHESIS_ALIGNMENT: A
On Tue, 2020-06-02 at 11:22 +, Patchwork wrote:
> == Series Details ==
>
> Series: drm/i915/dsi: Dont forget to clean up the connector on error (rev3)
> URL : https://patchwork.freedesktop.org/series/77011/
> State : failure
>
> == Summary ==
>
> CI Bug Log - changes from CI_DRM_8568_full
For reasons that be, the HW only allows usersace to read its own
CTX_TIMESTAMP [context local HW runtime] on rcs. Make it available for
all by adding it to the whitelists.
v2: The change took effect from Cometlake.
v3: Ignore timestamps that autoincrement when validating the whitelist
Signed-off-
== Series Details ==
Series: drm/i915/params: fix i915.reset module param type
URL : https://patchwork.freedesktop.org/series/77923/
State : success
== Summary ==
CI Bug Log - changes from CI_DRM_8572 -> Patchwork_17844
Summary
---
*
== Series Details ==
Series: series starting with [1/2] drm/i915: Identify Cometlake platform (rev2)
URL : https://patchwork.freedesktop.org/series/77922/
State : warning
== Summary ==
$ dim checkpatch origin/drm-tip
a71c14019fde drm/i915: Identify Cometlake platform
-:367: CHECK:MACRO_ARG_REU
== Series Details ==
Series: series starting with [1/2] drm/i915: Identify Cometlake platform (rev2)
URL : https://patchwork.freedesktop.org/series/77922/
State : warning
== Summary ==
$ dim sparse --fast origin/drm-tip
Sparse version: v0.6.0
Fast mode used, each commit won't be checked separa
Hi Christian,
On 6/2/20 2:47 PM, Christian König wrote:
Nirmoy please keep in mind that your current implementation doesn't
fully solve the issue the test case is exercising.
In other words what you have implement is fast skipping of fragmented
address space for bottom-up and top-down.
But
== Series Details ==
Series: series starting with [1/2] drm/i915: Identify Cometlake platform (rev2)
URL : https://patchwork.freedesktop.org/series/77922/
State : success
== Summary ==
CI Bug Log - changes from CI_DRM_8573 -> Patchwork_17845
Hi All,
My desktop (Intel(R) Xeon(R) CPU E3-1245 v5 @ 3.50GHz) is spewing tons
and tons of:
[ 778.461227] [drm:gen8_de_irq_handler [i915]] *ERROR* Fault errors on pipe B:
0x0080
[ 778.477763] [drm:gen8_de_irq_handler [i915]] *ERROR* Fault errors on pipe B:
0x0080
[ 778.577718] [drm:g
Hi Peter
Please file a bug by follow this instructions:
https://gitlab.freedesktop.org/drm/intel/-/wikis/How-to-file-i915-bugs
On Tue, 2020-06-02 at 19:46 +0200, Peter Zijlstra wrote:
> Hi All,
>
> My desktop (Intel(R) Xeon(R) CPU E3-1245 v5 @ 3.50GHz) is spewing tons
> and tons of:
>
> [ 778.
On Tue, Jun 02, 2020 at 06:08:03PM +, Souza, Jose wrote:
> Hi Peter
> Please file a bug by follow this instructions:
> https://gitlab.freedesktop.org/drm/intel/-/wikis/How-to-file-i915-bugs
*sigh*, top posting and webforms :-(
Steps to reproduce: Boot into X
How often: Always
uname -r: 5.6
From: Clint Taylor
Set GS Timer to 224. Combine with Wa_1604555607 due to register FF_MODE2
not being able to be read.
Cc: Caz Yokoyama
Cc: Matt Atwood
Signed-off-by: Clint Taylor
---
drivers/gpu/drm/i915/gt/intel_workarounds.c | 8
drivers/gpu/drm/i915/i915_reg.h | 2 ++
== Series Details ==
Series: series starting with [1/3] drm/atomic-helper: reset vblank on crtc
reset (rev2)
URL : https://patchwork.freedesktop.org/series/77908/
State : failure
== Summary ==
CI Bug Log - changes from CI_DRM_8571_full -> Patchwork_17839_full
=
On Tue, 2020-06-02 at 15:05 +0100, Chris Wilson wrote:
> Cometlake is small refresh of Coffeelake, but since we have found out a
> difference in the plaforms, we need to identify the separate platforms.
>
> Since we previously took Coffeelake/Cometlake as identical, update all
> IS_COFFEELAKE() to
On Tue, 2020-06-02 at 16:48 +0100, Chris Wilson wrote:
> For reasons that be, the HW only allows usersace to read its own
> CTX_TIMESTAMP [context local HW runtime] on rcs. Make it available for
> all by adding it to the whitelists.
>
> v2: The change took effect from Cometlake.
> v3: Ignore times
== Series Details ==
Series: drm/i915/tgl: Implement WA_16011163337
URL : https://patchwork.freedesktop.org/series/77933/
State : success
== Summary ==
CI Bug Log - changes from CI_DRM_8573 -> Patchwork_17846
Summary
---
**SUCCESS**
Quoting clinton.a.tay...@intel.com (2020-06-02 20:25:01)
> From: Clint Taylor
>
> Set GS Timer to 224. Combine with Wa_1604555607 due to register FF_MODE2
> not being able to be read.
>
> Cc: Caz Yokoyama
> Cc: Matt Atwood
> Signed-off-by: Clint Taylor
> ---
> drivers/gpu/drm/i915/gt/intel_w
As latest update we have now 2 voltage swing tables for DP over DKL
PHY with only one difference in Level 0 pre-emphasis 3.
So with 2 tables for DP is time to have one single function to return
all DKL voltage swing tables.
BSpec: 49292
Cc: Khaled Almahallawy
Signed-off-by: José Roberto de Souza
== Series Details ==
Series: drm/i915/tgl: Add HBR and HBR2+ voltage swing table
URL : https://patchwork.freedesktop.org/series/77934/
State : success
== Summary ==
CI Bug Log - changes from CI_DRM_8573 -> Patchwork_17847
Summary
---
We infrequently use the direct i915 backpointer from the i915_request,
so do we really need to waste the space in the struct for it? 8 bytes
from the most frequently allocated struct vs an 3 bytes and pointer
chasing in using rq->engine->i915?
Signed-off-by: Chris Wilson
---
drivers/gpu/drm/i915
== Series Details ==
Series: drm/i915: Drop i915_request.i915 backpointer
URL : https://patchwork.freedesktop.org/series/77936/
State : success
== Summary ==
CI Bug Log - changes from CI_DRM_8573 -> Patchwork_17848
Summary
---
**SUCC
== Series Details ==
Series: drm/i915/params: fix i915.reset module param type
URL : https://patchwork.freedesktop.org/series/77923/
State : failure
== Summary ==
CI Bug Log - changes from CI_DRM_8572_full -> Patchwork_17844_full
Summary
--
Hi all,
Today's linux-next merge of the drm-intel-fixes tree got a conflict in:
drivers/gpu/drm/i915/gt/intel_lrc.c
between commit:
f53ae29c0ea1 ("drm/i915/gt: Include a few tracek for timeslicing")
from Linus' tree and commit:
00febf644648 ("drm/i915/gt: Incorporate the virtual engine
Hi Daniel,
Thank you for the patch.
May I remind you about the -v option to git-format-patch ? :-) Seriously
speaking, it really helps review.
On Tue, Jun 02, 2020 at 11:51:38AM +0200, Daniel Vetter wrote:
> Only when vblanks are supported ofc.
>
> Some drivers do this already, but most unfortu
== Series Details ==
Series: series starting with [1/2] drm/i915: Identify Cometlake platform (rev2)
URL : https://patchwork.freedesktop.org/series/77922/
State : success
== Summary ==
CI Bug Log - changes from CI_DRM_8573_full -> Patchwork_17845_full
==
> -Original Message-
> From: Intel-gfx On Behalf Of Chris
> Wilson
> Sent: Tuesday, June 02, 2020 3:10 PM
> To: intel-gfx@lists.freedesktop.org
> Cc: Chris Wilson
> Subject: [Intel-gfx] [PATCH] drm/i915: Drop i915_request.i915 backpointer
>
> We infrequently use the direct i915 backpo
== Series Details ==
Series: drm/i915/tgl: Implement WA_16011163337
URL : https://patchwork.freedesktop.org/series/77933/
State : failure
== Summary ==
CI Bug Log - changes from CI_DRM_8573_full -> Patchwork_17846_full
Summary
---
**
== Series Details ==
Series: drm/i915/tgl: Add HBR and HBR2+ voltage swing table
URL : https://patchwork.freedesktop.org/series/77934/
State : success
== Summary ==
CI Bug Log - changes from CI_DRM_8573_full -> Patchwork_17847_full
Summary
== Series Details ==
Series: drm/i915: Drop i915_request.i915 backpointer
URL : https://patchwork.freedesktop.org/series/77936/
State : success
== Summary ==
CI Bug Log - changes from CI_DRM_8573_full -> Patchwork_17848_full
Summary
---
78 matches
Mail list logo
