On 2017-11-03 05:15 AM, Nicolai Hähnle wrote:
Hi Andres,
Thank you for doing this. I haven't read the series in detail yet, but I
have a bunch of questions and comments.
Thanks for all the feedback.
I find it very unfortunate to have both fence and semaphore objects in
the Gallium inter
Hi Andres,
Thank you for doing this. I haven't read the series in detail yet, but I
have a bunch of questions and comments.
I find it very unfortunate to have both fence and semaphore objects in
the Gallium interface. The two even become mixed up in the amdgpu
winsys. My first thought is tha
That's actually a pretty good suggestion, as it would help me get my feet
wet on piglit changes with something simple.
Thanks,
Andres
On Nov 2, 2017 1:24 AM, "Timothy Arceri" wrote:
> On 02/11/17 15:12, Andres Rodriguez wrote:
>
>> It was tested against steamvr using an opengl vr client (hellov
On 02/11/17 15:12, Andres Rodriguez wrote:
It was tested against steamvr using an opengl vr client (hellovr_opengl
for simplicity).
Hi Andres,
Full tests would be great but at the very least we should start with
some piglit tests that exercise the error paths.
Tim
Regards,
Andres
On 20
It was tested against steamvr using an opengl vr client (hellovr_opengl
for simplicity).
Regards,
Andres
On 2017-11-02 12:07 AM, Jason Ekstrand wrote:
Out of curiosity, how has this series been tested beyond regression
testing?
On November 1, 2017 20:57:57 Andres Rodriguez wrote:
This s
Out of curiosity, how has this series been tested beyond regression testing?
On November 1, 2017 20:57:57 Andres Rodriguez wrote:
This series adds radeonsi support for GL_EXT_semaphore.
GL_EXT_semaphore is used by steam's vrclient to synchronize access to shared
surfaces (vulkan <-> gl inter
This series adds radeonsi support for GL_EXT_semaphore.
GL_EXT_semaphore is used by steam's vrclient to synchronize access to shared
surfaces (vulkan <-> gl interop). It allows our gl vrclients to enqueue their
framebuffer surface without waiting for a glFinish().
If anyone has any suggestions on
