== Series Details ==
Series: drm/i915: Check for fused or unused pipes
URL : https://patchwork.freedesktop.org/series/35389/
State : success
== Summary ==
Series 35389v1 drm/i915: Check for fused or unused pipes
https://patchwork.freedesktop.org/api/1.0/series/35389/revisions/1/mbox/
Test deb
It is possoble to get very mysterious results from kms_sysfs_edid timing
Since there is no try-bot for IGT I am using IGT PW to get more data.
Marta Lofstedt (2):
test/kms_sysfs_edid_timing : Increase THRESHOLD_PER_CONNECTOR and
THRESHOLD_TOTAL
HACK to get kms_sysfs_edid_timing run on more
== Series Details ==
Series: drm/i915: Check for fused or unused pipes
URL : https://patchwork.freedesktop.org/series/35389/
State : success
== Summary ==
Test kms_flip:
Subgroup dpms-vs-vblank-race:
pass -> DMESG-FAIL (shard-hsw) fdo#103060
Subgroup vblan
On Friday, December 15, 2017 9:59:02 AM PST Mika Kahola wrote:
> In case of fused or unused pipes, return early with a warning when reading
> information for encoder.
>
> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=103206
> Reported-by: Thomas Gleixner
> Signed-off-by: Jani Nikula
> S
On Fri, 15 Dec 2017, Dhinakaran Pandiyan wrote:
> On Friday, December 15, 2017 9:59:02 AM PST Mika Kahola wrote:
>> In case of fused or unused pipes, return early with a warning when reading
>> information for encoder.
>>
>> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=103206
>> Reporte
We want to exclude any GGTT objects from being present on our internal
lists to avoid the deadlock we may run into with our requirement for
struct_mutex during invalidate. However, if the gup_fast fails, we put
the userptr onto the workqueue and mark it as active, so that we
remember to serialise t
On 15/12/2017 09:27, Chris Wilson wrote:
We want to exclude any GGTT objects from being present on our internal
lists to avoid the deadlock we may run into with our requirement for
struct_mutex during invalidate. However, if the gup_fast fails, we put
the userptr onto the workqueue and mark it a
Quoting Chris Wilson (2017-12-15 09:27:27)
> We want to exclude any GGTT objects from being present on our internal
> lists to avoid the deadlock we may run into with our requirement for
> struct_mutex during invalidate. However, if the gup_fast fails, we put
> the userptr onto the workqueue and ma
== Series Details ==
Series: drm/i915/userptr: Probe vma range before gup
URL : https://patchwork.freedesktop.org/series/35394/
State : success
== Summary ==
Series 35394v1 drm/i915/userptr: Probe vma range before gup
https://patchwork.freedesktop.org/api/1.0/series/35394/revisions/1/mbox/
Te
On Fri, 2017-12-15 at 11:27 +0200, Jani Nikula wrote:
> On Fri, 15 Dec 2017, Dhinakaran Pandiyan om> wrote:
> >
> > On Friday, December 15, 2017 9:59:02 AM PST Mika Kahola wrote:
> > >
> > > In case of fused or unused pipes, return early with a warning
> > > when reading
> > > information for en
We want to exclude any GGTT objects from being present on our internal
lists to avoid the deadlock we may run into with our requirement for
struct_mutex during invalidate. However, if the gup_fast fails, we put
the userptr onto the workqueue and mark it as active, so that we
remember to serialise t
We want to exclude any GGTT objects from being present on our internal
lists to avoid the deadlock we may run into with our requirement for
struct_mutex during invalidate. However, if the gup_fast fails, we put
the userptr onto the workqueue and mark it as active, so that we
remember to serialise t
On Friday, December 15, 2017 11:27:02 AM PST Jani Nikula wrote:
> On Fri, 15 Dec 2017, Dhinakaran Pandiyan
wrote:
> > On Friday, December 15, 2017 9:59:02 AM PST Mika Kahola wrote:
> >> In case of fused or unused pipes, return early with a warning when
> >> reading
> >> information for encoder.
>
On Fri, 15 Dec 2017, Dhinakaran Pandiyan wrote:
> On Friday, December 15, 2017 11:27:02 AM PST Jani Nikula wrote:
>> On Fri, 15 Dec 2017, Dhinakaran Pandiyan
> wrote:
>> > On Friday, December 15, 2017 9:59:02 AM PST Mika Kahola wrote:
>> >> In case of fused or unused pipes, return early with a w
Internal objects consistent of scratch pages not subject to the
persistence guarantees of user facing objects. They are used for
example, in ring buffers where they are only required for temporary
storage of commands that will be rewritten every time. As they are
temporary constructs, quietly repor
4K modes testing by using dummy EDID data has never been working
properly on boxes with DP++ (dual-mode) adaptors. The reason for
this is that those modes got pruned during hdmi mode validation.
intel_hdmi_mode_valid returns CLOCK_HIGH because the pixel clock
reported by the 4k mode is higher than
On Thu, 2017-12-14 at 20:55 -0800, Yaodong Li wrote:
> On 12/14/2017 03:43 AM, Joonas Lahtinen wrote:
> > On Wed, 2017-12-13 at 14:59 -0800, Yaodong Li wrote:
> > > On 12/13/2017 01:34 PM, Michal Wajdeczko wrote:
> > > > On Wed, 13 Dec 2017 19:19:06 +0100, Yaodong Li
> > > > wrote:
> > > >
> > >
Quoting Chris Wilson (2017-12-15 09:53:45)
> We want to exclude any GGTT objects from being present on our internal
> lists to avoid the deadlock we may run into with our requirement for
> struct_mutex during invalidate. However, if the gup_fast fails, we put
> the userptr onto the workqueue and ma
On Thu, Dec 14, 2017 at 09:11:01PM -0500, Alex Deucher wrote:
> On Thu, Dec 14, 2017 at 3:30 PM, Daniel Vetter wrote:
> > diff --git a/Documentation/gpu/drm-kms.rst b/Documentation/gpu/drm-kms.rst
> > index 420025bd6a9b..cbba93483aec 100644
> > --- a/Documentation/gpu/drm-kms.rst
> > +++ b/Documen
On Thu, 2017-12-14 at 19:57 +, Chris Wilson wrote:
> Why open code a printf/write loop, when we already have a function
> providing it?
>
> Signed-off-by: Chris Wilson
> Cc: Joonas Lahtinen
Reviewed-by: Joonas Lahtinen
Will merge the patch.
Regards, Joonas
--
Joonas Lahtinen
Open Source
== Series Details ==
Series: drm/i915/userptr: Probe vma range before gup (rev3)
URL : https://patchwork.freedesktop.org/series/35394/
State : success
== Summary ==
Series 35394v3 drm/i915/userptr: Probe vma range before gup
https://patchwork.freedesktop.org/api/1.0/series/35394/revisions/3/mb
Quoting Joonas Lahtinen (2017-12-15 10:31:08)
> On Thu, 2017-12-14 at 19:57 +, Chris Wilson wrote:
> > Why open code a printf/write loop, when we already have a function
> > providing it?
> >
> > Signed-off-by: Chris Wilson
> > Cc: Joonas Lahtinen
>
> Reviewed-by: Joonas Lahtinen
>
> Will
== Series Details ==
Series: drm/i915/userptr: Probe vma range before gup
URL : https://patchwork.freedesktop.org/series/35394/
State : warning
== Summary ==
Test gem_userptr_blits:
Subgroup map-fixed-invalidate-busy-gup:
pass -> DMESG-WARN (shard-snb)
Quoting Chris Wilson (2017-12-15 10:09:02)
> We want to exclude any GGTT objects from being present on our internal
> lists to avoid the deadlock we may run into with our requirement for
> struct_mutex during invalidate. However, if the gup_fast fails, we put
> the userptr onto the workqueue and ma
On Fri, 2017-12-15 at 10:17 +, Chris Wilson wrote:
> Internal objects consistent of scratch pages not subject to the
> persistence guarantees of user facing objects. They are used for
> example, in ring buffers where they are only required for temporary
> storage of commands that will be rewrit
== Series Details ==
Series: drm/i915: Allow internal page allocations to fail
URL : https://patchwork.freedesktop.org/series/35397/
State : success
== Summary ==
Series 35397v1 drm/i915: Allow internal page allocations to fail
https://patchwork.freedesktop.org/api/1.0/series/35397/revisions/1
== Series Details ==
Series: drm/i915: Ignore TMDS clock limit for DP++ when EDID override is set
(rev2)
URL : https://patchwork.freedesktop.org/series/35338/
State : success
== Summary ==
Series 35338v2 drm/i915: Ignore TMDS clock limit for DP++ when EDID override is
set
https://patchwork.f
We want to exclude any GGTT objects from being present on our internal
lists to avoid the deadlock we may run into with our requirement for
struct_mutex during invalidate. However, if the gup_fast fails, we put
the userptr onto the workqueue and mark it as active, so that we
remember to serialise t
Quoting Chris Wilson (2017-12-07 16:59:36)
> Quoting Michal Wajdeczko (2017-12-07 16:52:46)
> > In some cases debugfs or sysfs may return errors that we
> > want to check. Return -errno from helper functions to make
> > asserts easier.
> >
> > v2: don't forget about EOF ret=0 (Chris)
> > small
== Series Details ==
Series: drm/i915/userptr: Probe vma range before gup (rev3)
URL : https://patchwork.freedesktop.org/series/35394/
State : failure
== Summary ==
Test kms_plane:
Subgroup plane-panning-bottom-right-suspend-pipe-b-planes:
pass -> DMESG-WARN (shar
On Fri, Dec 08, 2017 at 11:30:07PM +, Patchwork wrote:
> == Series Details ==
>
> Series: igt/core_suspend: Exercise igt_system_suspend_autoresume() (rev2)
> URL : https://patchwork.freedesktop.org/series/31986/
> State : success
>
> == Summary ==
https://intel-gfx-ci.01.org/tree/drm-tip/I
Quoting Arkadiusz Hiler (2017-12-15 12:06:01)
> On Fri, Dec 08, 2017 at 11:30:07PM +, Patchwork wrote:
> > == Series Details ==
> >
> > Series: igt/core_suspend: Exercise igt_system_suspend_autoresume() (rev2)
> > URL : https://patchwork.freedesktop.org/series/31986/
> > State : success
> >
Quoting Chris Wilson (2017-09-13 11:39:14)
> Both gem_linear_blits and gem_tiled_blit do not request the full 48b
> GTT layout for their objects, restricting themselves to 4G. The
> underlying test that they trigger eviction is unaffected by this
> restriction, so we can simply reduce their memory
== Series Details ==
Series: drm/i915/userptr: Probe vma range before gup (rev4)
URL : https://patchwork.freedesktop.org/series/35394/
State : success
== Summary ==
Series 35394v4 drm/i915/userptr: Probe vma range before gup
https://patchwork.freedesktop.org/api/1.0/series/35394/revisions/4/mb
== Series Details ==
Series: drm/i915: Allow internal page allocations to fail
URL : https://patchwork.freedesktop.org/series/35397/
State : warning
== Summary ==
Test kms_atomic:
Subgroup plane_cursor_legacy:
pass -> SKIP (shard-hsw)
Test gem_tiled_swapping
== Series Details ==
Series: igt/pm_rps: Always allocate spin[0]
URL : https://patchwork.freedesktop.org/series/35176/
State : success
== Summary ==
IGT patchset tested on top of latest successful build
417b29e2fbc97f99455a9e4bd22270bab0a4628f lib/igt_sysfs: Let
igt_sysfs_read|write return -e
Hi,
This is a known issue. Could you try out this patch to see if that
would fix this issue for you?
https://patchwork.freedesktop.org/series/35389/
On Fri, 2017-12-15 at 17:08 +0530, Jaswinder Singh Rajput wrote:
> Hello friends,
>
> I am getting multiple warnings in i915/intel_audio.c . I am
== Series Details ==
Series: drm/i915: Ignore TMDS clock limit for DP++ when EDID override is set
(rev2)
URL : https://patchwork.freedesktop.org/series/35338/
State : success
== Summary ==
Test kms_frontbuffer_tracking:
Subgroup fbc-1p-primscrn-cur-indfb-move:
pass
Quoting Joonas Lahtinen (2017-12-04 10:45:13)
> On Tue, 2017-11-28 at 17:29 +, Chris Wilson wrote:
> > Quoting Michal Wajdeczko (2017-11-28 17:01:15)
> > > In commit 3f6ae7b19 ("igt/kms_frontbuffer_tracking: Keep the debugfs
> > > dir around") we introduced custom variant of __igt_debugfs_read
It is possoble to get very mysterious results from kms_sysfs_edid timing
Since there is no try-bot for IGT I am using IGT PW to get more data.
Marta Lofstedt (2):
test/kms_sysfs_edid_timing : Increase THRESHOLD_PER_CONNECTOR and
THRESHOLD_TOTAL
HACK to get kms_sysfs_edid_timing run on more
The discussion on how to fix the issues has been stagnant for a
long time, see the Bugzilla below and:
https://patchwork.freedesktop.org/patch/170429/
After only changing display cable the result of my machine went
from always failing to hit a WARN. This is also illustrated on
CI-shards, where the
Signed-off-by: Marta Lofstedt
---
tests/intel-ci/fast-feedback.testlist | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tests/intel-ci/fast-feedback.testlist
b/tests/intel-ci/fast-feedback.testlist
index f71a16bc..a6c3f614 100644
--- a/tests/intel-ci/fast-feedback.testlist
++
A context encompasses the driver's view of process related state, and
encapsulates the logical GPU state where available. Each context is
currently equivalent to a process in CPU terms. Like with processes,
sometimes the user wants a lighter encapsulation that shares some state
with the parent proc
Signed-off-by: Chris Wilson
Cc: Joonas Lahtinen
Cc: Tvrtko Ursulin
Cc: Mika Kuoppala
---
tests/Makefile.sources | 1 +
tests/gem_ctx_shared.c | 294 +
2 files changed, 295 insertions(+)
create mode 100644 tests/gem_ctx_shared.c
diff --git a/t
== Series Details ==
Series: kms_sysfs_edid timing and hack to run for BAT
URL : https://patchwork.freedesktop.org/series/35404/
State : success
== Summary ==
IGT patchset tested on top of latest successful build
c0be3310715e2f744b892c51f09e62273bcc8e57 tests/kms_frontbuffer_tracking:
Correct
We may have fused or unused pipes in our system. Let's check that the pipe
in question is within limits of accessible pipes. In case, that we are not
able to access the pipe, we return early with a warning.
v2: Rephrasing of the commit message (Jani)
Bugzilla: https://bugs.freedesktop.org/show_bu
Quoting Pandiyan, Dhinakaran (2017-12-14 22:03:49)
> On Wed, 2017-12-13 at 18:28 +, Chris Wilson wrote:
> > Just printk the string, or at least do not double up on the newlines!
> >
>
> Reviewed-by: Dhinakaran Pandiyan
And pushed, thanks for the review.
-Chris
__
== Series Details ==
Series: drm/i915: Extend CREATE_CONTEXT to allow inheritance ala clone()
URL : https://patchwork.freedesktop.org/series/35405/
State : success
== Summary ==
Series 35405v1 drm/i915: Extend CREATE_CONTEXT to allow inheritance ala clone()
https://patchwork.freedesktop.org/ap
On Fri, 2017-12-15 at 19:20 +0530, Jaswinder Singh Rajput wrote:
> On Fri, 2017-12-15 at 14:44 +0200, Mika Kahola wrote:
> >
> > Hi,
> >
> > This is a known issue. Could you try out this patch to see if that
> > would fix this issue for you?
> >
> > https://patchwork.freedesktop.org/series/35389
== Series Details ==
Series: drm/i915/userptr: Probe vma range before gup (rev4)
URL : https://patchwork.freedesktop.org/series/35394/
State : failure
== Summary ==
Test gem_userptr_blits:
Subgroup map-fixed-invalidate-overlap-busy-gup:
pass -> FAIL (shard-s
Instead of trying different seq_puts messages, lets use common
-ENODEV error code to indicate missing/unsupported feature.
v2: don't forget about guc_log_control fops (Sagar)
Suggested-by: Chris Wilson
Signed-off-by: Michal Wajdeczko
Cc: Chris Wilson
Cc: Joonas Lahtinen
Cc: Sagar Arun Kamble
We want to exclude any GGTT objects from being present on our internal
lists to avoid the deadlock we may run into with our requirement for
struct_mutex during invalidate. However, if the gup_fast fails, we put
the userptr onto the workqueue and mark it as active, so that we
remember to serialise t
Start the spinner first; before starting the measurement, and sleep for
a known duration. In particular, CI once saw a measurement for busyness
greater than the intended batch_duration!
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=104241
Signed-off-by: Chris Wilson
Cc: Tvrtko Ursulin
-
== Series Details ==
Series: drm/i915: Check for fused or unused pipes (rev2)
URL : https://patchwork.freedesktop.org/series/35389/
State : success
== Summary ==
Series 35389v2 drm/i915: Check for fused or unused pipes
https://patchwork.freedesktop.org/api/1.0/series/35389/revisions/2/mbox/
T
Create all the spinners before starting the sampler and then measure how
long we sleep.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=104160
Signed-off-by: Chris Wilson
Cc: Tvrtko Ursulin
---
tests/perf_pmu.c | 43 +++
1 file changed, 23 insertio
== Series Details ==
Series: drm/i915: Unifying debugfs return codes for unsupported features (rev2)
URL : https://patchwork.freedesktop.org/series/34552/
State : success
== Summary ==
Series 34552v2 drm/i915: Unifying debugfs return codes for unsupported features
https://patchwork.freedesktop
== Series Details ==
Series: kms_sysfs_edid timing and hack to run for BAT
URL : https://patchwork.freedesktop.org/series/35404/
State : success
== Summary ==
Test gem_tiled_swapping:
Subgroup non-threaded:
dmesg-warn -> PASS (shard-hsw) fdo#104218 +1
Test kms_fli
Don't rely on the timer being precise when we can sleep for a known
duration.
Signed-off-by: Chris Wilson
Cc: Tvrtko Ursulin
---
tests/perf_pmu.c | 10 +-
1 file changed, 5 insertions(+), 5 deletions(-)
diff --git a/tests/perf_pmu.c b/tests/perf_pmu.c
index e35aca899..1197cae4e 100644
== Series Details ==
Series: drm/i915: Extend CREATE_CONTEXT to allow inheritance ala clone()
URL : https://patchwork.freedesktop.org/series/35405/
State : success
== Summary ==
Test gem_tiled_swapping:
Subgroup non-threaded:
dmesg-warn -> PASS (shard-hsw) fdo#104
On Fri, 15 Dec 2017, Mika Kahola wrote:
> We may have fused or unused pipes in our system. Let's check that the pipe
> in question is within limits of accessible pipes. In case, that we are not
> able to access the pipe, we return early with a warning.
>
> v2: Rephrasing of the commit message (Jan
Don't rely on the timer being precise when we can sleep for a known
duration.
Signed-off-by: Chris Wilson
Cc: Tvrtko Ursulin
---
tests/perf_pmu.c | 36
1 file changed, 24 insertions(+), 12 deletions(-)
diff --git a/tests/perf_pmu.c b/tests/perf_pmu.c
index
== Series Details ==
Series: drm/i915/userptr: Probe vma range before gup (rev5)
URL : https://patchwork.freedesktop.org/series/35394/
State : success
== Summary ==
Series 35394v5 drm/i915/userptr: Probe vma range before gup
https://patchwork.freedesktop.org/api/1.0/series/35394/revisions/5/mb
When monitoring the GPU with i915 perf, reports are tagged with a hw
id. When also tracking the requests using the kernel tracepoints, if
we include the hw_id from i915_gem_context, this allows us to
correlate a process with hw id fields in the OA reports.
Signed-off-by: Lionel Landwerlin
---
dr
== Series Details ==
Series: igt: Exercise creating context with shared GTT
URL : https://patchwork.freedesktop.org/series/35406/
State : success
== Summary ==
IGT patchset tested on top of latest successful build
c0be3310715e2f744b892c51f09e62273bcc8e57 tests/kms_frontbuffer_tracking:
Correc
Quoting Lionel Landwerlin (2017-12-15 15:51:36)
> When monitoring the GPU with i915 perf, reports are tagged with a hw
> id. When also tracking the requests using the kernel tracepoints, if
> we include the hw_id from i915_gem_context, this allows us to
> correlate a process with hw id fields in th
Don't rely on the timer being precise when we can sleep for a known
duration.
Signed-off-by: Chris Wilson
Cc: Tvrtko Ursulin
---
tests/perf_pmu.c | 10 +-
1 file changed, 5 insertions(+), 5 deletions(-)
diff --git a/tests/perf_pmu.c b/tests/perf_pmu.c
index 96091b853..b3b289656 100644
Sleep for a known duration. In particular, CI once saw a measurement for
busyness greater than the intended batch_duration!
v2: Go back to starting pmu sampling outside of spinner; the GPU should
be idle.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=104241
Signed-off-by: Chris Wilson
C
Don't rely on the timer being precise when we can sleep for a known
duration.
Signed-off-by: Chris Wilson
Cc: Tvrtko Ursulin
---
tests/perf_pmu.c | 38 ++
1 file changed, 26 insertions(+), 12 deletions(-)
diff --git a/tests/perf_pmu.c b/tests/perf_pmu.c
inde
Create all the spinners before starting the sampler and then measure how
long we sleep.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=104160
Signed-off-by: Chris Wilson
Cc: Tvrtko Ursulin
---
tests/perf_pmu.c | 45 +
1 file changed, 25 insert
On 15/12/17 16:08, Chris Wilson wrote:
Quoting Lionel Landwerlin (2017-12-15 15:51:36)
When monitoring the GPU with i915 perf, reports are tagged with a hw
id. When also tracking the requests using the kernel tracepoints, if
we include the hw_id from i915_gem_context, this allows us to
correlate
== Series Details ==
Series: drm/i915/trace: add hw_id to gem requests trace points
URL : https://patchwork.freedesktop.org/series/35420/
State : success
== Summary ==
Series 35420v1 drm/i915/trace: add hw_id to gem requests trace points
https://patchwork.freedesktop.org/api/1.0/series/35420/r
* Lucas De Marchi wrote:
> CFL was missing from intel_early_ids[]. The PCI ID needs to be there to
> allow the memory region to be stolen, otherwise we could have RAM being
> arbitrarily overwritten if for example we keep using the UEFI framebuffer,
> depending on how BIOS has set up the e820 ma
== Series Details ==
Series: drm/i915: Check for fused or unused pipes (rev2)
URL : https://patchwork.freedesktop.org/series/35389/
State : success
== Summary ==
Test kms_frontbuffer_tracking:
Subgroup fbc-1p-offscren-pri-shrfb-draw-render:
fail -> PASS (sha
== Series Details ==
Series: igt/perf_pmu: Tighten busy measurement
URL : https://patchwork.freedesktop.org/series/35412/
State : success
== Summary ==
IGT patchset tested on top of latest successful build
c0be3310715e2f744b892c51f09e62273bcc8e57 tests/kms_frontbuffer_tracking:
Correctly hand
Use the normal batch_duration_ns and display the sampled frequency:
Frequency: min=100, max=750, boost=750 MHz
Min frequency: requested 100.0, actual 100.0
Max frequency: requested 755.6, actual 755.6
Signed-off-by: Chris Wilson
Cc: Tvrtko Ursulin
---
tests/perf_pmu.c |
== Series Details ==
Series: igt/perf_pmu: Tighten measurements for most-busy
URL : https://patchwork.freedesktop.org/series/35414/
State : success
== Summary ==
IGT patchset tested on top of latest successful build
c0be3310715e2f744b892c51f09e62273bcc8e57 tests/kms_frontbuffer_tracking:
Corr
On 15/12/2017 16:18, Lionel Landwerlin wrote:
On 15/12/17 16:08, Chris Wilson wrote:
Quoting Lionel Landwerlin (2017-12-15 15:51:36)
When monitoring the GPU with i915 perf, reports are tagged with a hw
id. When also tracking the requests using the kernel tracepoints, if
we include the hw_id fr
== Series Details ==
Series: drm/i915: Unifying debugfs return codes for unsupported features (rev2)
URL : https://patchwork.freedesktop.org/series/34552/
State : failure
== Summary ==
Test gem_exec_suspend:
Subgroup basic-s3:
incomplete -> PASS (shard-hsw) fdo#10
== Series Details ==
Series: igt/perf_pmu: Measure the reference batch for busy-check-all
URL : https://patchwork.freedesktop.org/series/35416/
State : failure
== Summary ==
IGT patchset tested on top of latest successful build
c0be3310715e2f744b892c51f09e62273bcc8e57 tests/kms_frontbuffer_tra
== Series Details ==
Series: igt/perf_pmu: Measure the reference batch for all-busy-check-all
URL : https://patchwork.freedesktop.org/series/35417/
State : failure
== Summary ==
IGT patchset tested on top of latest successful build
c0be3310715e2f744b892c51f09e62273bcc8e57 tests/kms_frontbuffer
== Series Details ==
Series: series starting with [v2,1/4] igt/perf_pmu: Tighten busy measurement
URL : https://patchwork.freedesktop.org/series/35421/
State : success
== Summary ==
IGT patchset tested on top of latest successful build
c0be3310715e2f744b892c51f09e62273bcc8e57 tests/kms_frontbu
It doesn't look like tracepoint have any guarantee to have always the
same ordering of their parameter. Instead of relying on a predefined
regexp, let's split the parameters on commas and access the values
through a map.
Signed-off-by: Lionel Landwerlin
---
scripts/trace.pl | 30
When monitoring the GPU with i915 perf, reports are tagged with a hw
id. When also tracking the requests using the kernel tracepoints, if
we include the hw_id from i915_gem_context, this allows us to
correlate a process with hw id fields in the OA reports.
v2: Place hw_id at the end of the tracepo
== Series Details ==
Series: drm/i915/userptr: Probe vma range before gup (rev5)
URL : https://patchwork.freedesktop.org/series/35394/
State : success
== Summary ==
Test kms_frontbuffer_tracking:
Subgroup fbc-suspend:
pass -> INCOMPLETE (shard-hsw) fdo#103540 +2
On 15/12/2017 16:10, Chris Wilson wrote:
Sleep for a known duration. In particular, CI once saw a measurement for
busyness greater than the intended batch_duration!
v2: Go back to starting pmu sampling outside of spinner; the GPU should
be idle.
Bugzilla: https://bugs.freedesktop.org/show_bug.
== Series Details ==
Series: igt/perf_pmu: Speed up frequency measurement
URL : https://patchwork.freedesktop.org/series/35429/
State : warning
== Summary ==
IGT patchset tested on top of latest successful build
c0be3310715e2f744b892c51f09e62273bcc8e57 tests/kms_frontbuffer_tracking:
Correctl
On 15/12/2017 17:05, Chris Wilson wrote:
Use the normal batch_duration_ns and display the sampled frequency:
Frequency: min=100, max=750, boost=750 MHz
Min frequency: requested 100.0, actual 100.0
Max frequency: requested 755.6, actual 755.6
Signed-off-by: Chris Wilson
== Series Details ==
Series: drm/i915/trace: add hw_id to gem requests trace points (rev2)
URL : https://patchwork.freedesktop.org/series/35420/
State : success
== Summary ==
Series 35420v2 drm/i915/trace: add hw_id to gem requests trace points
https://patchwork.freedesktop.org/api/1.0/series/
== Series Details ==
Series: igt: Exercise creating context with shared GTT
URL : https://patchwork.freedesktop.org/series/35406/
State : success
== Summary ==
Warning: bzip IGTPW_680/shard-snb5/results13.json.bz2 wasn't in correct JSON
format
Test gem_exec_suspend:
Subgroup basic-s3:
Since commit: drm/i915/scheduler: Support user-defined priorities, the
driver support an extra context param to set context's priority. Add
tests for that interface and update invalid tests.
Signed-off-by: Antonio Argenziano
Cc: Chris Wilson
Cc: Michal Winiarski
---
tests/gem_ctx_param.c | 77
On 12/15/2017 02:21 AM, Joonas Lahtinen wrote:
On Thu, 2017-12-14 at 20:55 -0800, Yaodong Li wrote:
On 12/14/2017 03:43 AM, Joonas Lahtinen wrote:
On Wed, 2017-12-13 at 14:59 -0800, Yaodong Li wrote:
On 12/13/2017 01:34 PM, Michal Wajdeczko wrote:
On Wed, 13 Dec 2017 19:19:06 +0100, Yaodong L
== Series Details ==
Series: trace.pl: handle request tracepoint fields regardless of their order
URL : https://patchwork.freedesktop.org/series/35434/
State : success
== Summary ==
IGT patchset tested on top of latest successful build
c0be3310715e2f744b892c51f09e62273bcc8e57 tests/kms_frontbu
== Series Details ==
Series: tests/gem_ctx_param: Update invalid param
URL : https://patchwork.freedesktop.org/series/35438/
State : success
== Summary ==
IGT patchset tested on top of latest successful build
c0be3310715e2f744b892c51f09e62273bcc8e57 tests/kms_frontbuffer_tracking:
Correctly h
== Series Details ==
Series: igt/perf_pmu: Tighten busy measurement
URL : https://patchwork.freedesktop.org/series/35412/
State : failure
== Summary ==
Test kms_flip:
Subgroup flip-vs-panning-vs-hang:
dmesg-warn -> PASS (shard-snb) fdo#103821
Subgroup vbla
On 13/12/17 05:26, Chris Wilson wrote:
Quoting Petri Latvala (2017-12-13 12:58:14)
Since the test is considered to be HW focused, meaning that it doesn't
really exercise the driver but rather an HW feature, it is placed in
tests/hw-tests.
Do we really want to keep this test at all? It's an i
Signed-off-by: Adam Jackson
---
tools/intel_vbt_decode.c | 4 ++--
tools/intel_vbt_defs.h | 6 +++---
2 files changed, 5 insertions(+), 5 deletions(-)
diff --git a/tools/intel_vbt_decode.c b/tools/intel_vbt_decode.c
index 9d90c69d..ce7da2c0 100644
--- a/tools/intel_vbt_decode.c
+++ b/tools/int
== Series Details ==
Series: igt/perf_pmu: Tighten measurements for most-busy
URL : https://patchwork.freedesktop.org/series/35414/
State : failure
== Summary ==
Test kms_flip:
Subgroup blt-wf_vblank-vs-modeset-interruptible:
incomplete -> PASS (shard-hsw)
== Series Details ==
Series: series starting with [v2,1/4] igt/perf_pmu: Tighten busy measurement
URL : https://patchwork.freedesktop.org/series/35421/
State : failure
== Summary ==
Test kms_flip:
Subgroup flip-vs-panning:
pass -> DMESG-WARN (shard-hsw) fdo#103540
Quoting Petri Latvala (2017-12-13 12:58:15)
> From: Antonio Argenziano
>
> When writing to an invalid memory location, the HW should be clever
> enough to silently discard the write without disrupting execution.
> gem_bad_address aim at just that. The test has been updated to move away
> from the
Quoting Tvrtko Ursulin (2017-12-15 18:16:52)
>
> On 15/12/2017 16:10, Chris Wilson wrote:
> > Sleep for a known duration. In particular, CI once saw a measurement for
> > busyness greater than the intended batch_duration!
> >
> > v2: Go back to starting pmu sampling outside of spinner; the GPU sh
1 - 100 of 137 matches
Mail list logo
