All of lore.kernel.org
 help / color / mirror / Atom feed
* [Buildroot] glmark2 regression on mx6
@ 2017-04-20 22:32 Fabio Estevam
  2017-04-21  8:19 ` Gary Bisson
  0 siblings, 1 reply; 5+ messages in thread
From: Fabio Estevam @ 2017-04-20 22:32 UTC (permalink / raw)
  To: buildroot

Hi Gary,

I have just tested glmark2 with Buildroot master on imx6 and I got the
following error:

# glmark2-es2-drm
** Failed to set swap interval. Results may be bounded above by refresh rate.
=======================================================
    glmark2 2014.03
=======================================================
    OpenGL Information
    GL_VENDOR:     etnaviv
    GL_RENDERER:   Gallium 0.4 on Vivante GC2000 rev 5108
    GL_VERSION:    OpenGL ES 2.0 Mesa 17.0.3
=======================================================
** Failed to set swap interval. Results may be bounded above by refresh rate.
[build] use-vbo=false:Segmentation fault

Before I start investigating this problem, I would like to know if you
have already seen this issue or if you have any suggestions.

Thanks,

Fabio Estevam

^ permalink raw reply	[flat|nested] 5+ messages in thread

* [Buildroot] glmark2 regression on mx6
  2017-04-20 22:32 [Buildroot] glmark2 regression on mx6 Fabio Estevam
@ 2017-04-21  8:19 ` Gary Bisson
  2017-04-21 13:58   ` Fabio Estevam
  0 siblings, 1 reply; 5+ messages in thread
From: Gary Bisson @ 2017-04-21  8:19 UTC (permalink / raw)
  To: buildroot

Hi Fabio,

On Fri, Apr 21, 2017 at 12:32 AM, Fabio Estevam <festevam@gmail.com> wrote:
> Hi Gary,
>
> I have just tested glmark2 with Buildroot master on imx6 and I got the
> following error:
>
> # glmark2-es2-drm
> ** Failed to set swap interval. Results may be bounded above by refresh rate.
> =======================================================
>     glmark2 2014.03
> =======================================================
>     OpenGL Information
>     GL_VENDOR:     etnaviv
>     GL_RENDERER:   Gallium 0.4 on Vivante GC2000 rev 5108
>     GL_VERSION:    OpenGL ES 2.0 Mesa 17.0.3

Never tried Mesa 17.0.3. What's the kernel version you're using?

> =======================================================
> ** Failed to set swap interval. Results may be bounded above by refresh rate.

That swap interval message I've seen in the past (and discarded it).

> [build] use-vbo=false:Segmentation fault

Segfault however I've never seen. Is kmscube running fine with this setup?

> Before I start investigating this problem, I would like to know if you
> have already seen this issue or if you have any suggestions.

Does this happen for each test? Have you tried specifying one test like:
glmark2-es2-drm -b desktop:effect=shadow:windows=4:shadow-size=5
glmark2-es2-drm -b jellyfish
glmark2-es2-drm -b ideas

Is the segfault happening on each example above?

Other than this I don't have much suggestion except asking on IRC
(#etnaviv), the channel is very active lately (Android work).

Regards,
Gary

^ permalink raw reply	[flat|nested] 5+ messages in thread

* [Buildroot] glmark2 regression on mx6
  2017-04-21  8:19 ` Gary Bisson
@ 2017-04-21 13:58   ` Fabio Estevam
  2017-05-31  3:53     ` mjp112
  0 siblings, 1 reply; 5+ messages in thread
From: Fabio Estevam @ 2017-04-21 13:58 UTC (permalink / raw)
  To: buildroot

Hi Gary,

On Fri, Apr 21, 2017 at 5:19 AM, Gary Bisson
<gary.bisson@boundarydevices.com> wrote:

> Never tried Mesa 17.0.3. What's the kernel version you're using?

Also tried Mesa 17.0.4 with the same result.

Tested kernels 4.9, 4.10 and 4.11-rc and the glmark2 segfault always happen.

>
>> =======================================================
>> ** Failed to set swap interval. Results may be bounded above by refresh rate.
>
> That swap interval message I've seen in the past (and discarded it).
>
>> [build] use-vbo=false:Segmentation fault
>
> Segfault however I've never seen. Is kmscube running fine with this setup?
>
>> Before I start investigating this problem, I would like to know if you
>> have already seen this issue or if you have any suggestions.
>
> Does this happen for each test? Have you tried specifying one test like:
> glmark2-es2-drm -b desktop:effect=shadow:windows=4:shadow-size=5
> glmark2-es2-drm -b jellyfish
> glmark2-es2-drm -b ideas
>
> Is the segfault happening on each example above?

Yes, it happens in all of these examples as well.

In the last case there is also some warnings before the segfault:

# glmark2-es2-drm -b ideas
** Failed to set swap interval. Results may be bounded above by refresh rate.
=======================================================
    glmark2 2014.03
=======================================================
    OpenGL Information
    GL_VENDOR:     etnaviv
    GL_RENDERER:   Gallium 0.4 on Vivante GC2000 rev 5108
    GL_VERSION:    OpenGL ES 2.0 Mesa 17.0.4
=======================================================
** Failed to set swap interval. Results may be bounded above by refresh rate.
[ideas] <default>:[   79.619057] etnaviv-gpu 130000.gpu: hangcheck
detected gpu lockup!
[   79.625325] etnaviv-gpu 130000.gpu:      completed fence: 61
[   79.631058] etnaviv-gpu 130000.gpu:      active fence: 62
[   79.637492] etnaviv-gpu 130000.gpu: hangcheck recover!
[   83.116894] etnaviv-gpu 130000.gpu: hangcheck detected gpu lockup!
[   83.123160] etnaviv-gpu 130000.gpu:      completed fence: 62
[   83.128892] etnaviv-gpu 130000.gpu:      active fence: 63
[   83.134440] etnaviv-gpu 130000.gpu: hangcheck recover!
Segmentation fault

> Other than this I don't have much suggestion except asking on IRC
> (#etnaviv), the channel is very active lately (Android work).

Will report to the etnaviv folks, thanks.

^ permalink raw reply	[flat|nested] 5+ messages in thread

* [Buildroot] glmark2 regression on mx6
  2017-04-21 13:58   ` Fabio Estevam
@ 2017-05-31  3:53     ` mjp112
  2017-05-31 21:21       ` Fabio Estevam
  0 siblings, 1 reply; 5+ messages in thread
From: mjp112 @ 2017-05-31  3:53 UTC (permalink / raw)
  To: buildroot

I am seeing the exaclty same results..

My setup is : imx6q kernel 4.12-rc3 and
   OpenGL Information
    GL_VENDOR:     etnaviv
    GL_RENDERER:   Gallium 0.4 on Vivante GC2000 rev 5108
    GL_VERSION:    OpenGL ES 2.0 Mesa 17.0.5

Any ideas what is causing this? 




--
View this message in context: http://buildroot-busybox.2317881.n4.nabble.com/glmark2-regression-on-mx6-tp162973p166062.html
Sent from the Buildroot (busybox) mailing list archive at Nabble.com.

^ permalink raw reply	[flat|nested] 5+ messages in thread

* [Buildroot] glmark2 regression on mx6
  2017-05-31  3:53     ` mjp112
@ 2017-05-31 21:21       ` Fabio Estevam
  0 siblings, 0 replies; 5+ messages in thread
From: Fabio Estevam @ 2017-05-31 21:21 UTC (permalink / raw)
  To: buildroot

On Wed, May 31, 2017 at 12:53 AM, mjp112 <mika.penttila@gmail.com> wrote:
> I am seeing the exaclty same results..
>
> My setup is : imx6q kernel 4.12-rc3 and
>    OpenGL Information
>     GL_VENDOR:     etnaviv
>     GL_RENDERER:   Gallium 0.4 on Vivante GC2000 rev 5108
>     GL_VERSION:    OpenGL ES 2.0 Mesa 17.0.5
>
> Any ideas what is causing this?

I haven't had a chance to debug this issue yet.

Could you please create a new thread reporting the problem to
etnaviv at lists.freedesktop.org and dri-devel at lists.freedesktop.org?

^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2017-05-31 21:21 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2017-04-20 22:32 [Buildroot] glmark2 regression on mx6 Fabio Estevam
2017-04-21  8:19 ` Gary Bisson
2017-04-21 13:58   ` Fabio Estevam
2017-05-31  3:53     ` mjp112
2017-05-31 21:21       ` Fabio Estevam

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.