Sil Vilerino

Hi @evelikov , thank you for pointing this out. Please see inline below. > Greetings, while I'm no maintainer here are bunch of high-level notes: > > * please split...

@dvrogozh > @sivileri Can you, please, clarify the intent: in which applications or services this will be used? Overall there are the following questions which arise: > > 1. Who...

@evelikov @dvrogozh I'm seeing a failure in the freebsd build in the last couple of days but doesn't seem to be related to the code itself. For example, [the latest...

> Well, the only thing I can tell right now is that you are incredibly lucky considering that #625 which I merged just few minutes before you rebased and hit...

@evelikov @dvrogozh (cc @XinfengZhang) Along with this pull request, there are projects hosted for [an open source driver being developed for Windows](https://gitlab.freedesktop.org/sivileri/mesa/-/commits/va_win_testing/) as well as [an open source sample app...

> @sivileri : you should have been brought this up front with the references in PR description. Not seeing that part and not knowing it even exists was the major...

Thanks for approving @dvrogozh , could you please (or any other maintainer) merge this to master? I don't seem to have permission to do so myself. 

Exported two more functions in libva.def, that were used by the google tests in https://github.com/intel/libva-utils/pull/283, also ported the tests to windows in addition to vainfo now. vaMaxNumConfigAttributes vaBufferTypeStr

> Just compared the va.def against the exported symbols on my Arch machine, here is the diff with some inline notes: > > ```diff > vaAcquireBufferHandle > -VA_API_0.32.0 ## version...

Actually, looks like this also affects other codecs like H264/HEVC encoding with VAAPI, could we add such check for those as well?