Introduction and updates from NVIDIA

[James Jones]

On 05/11/2016 02:31 PM, Daniel Stone wrote:
> Hi James,
>
> On 11 May 2016 at 21:43, James Jones <jajones at nvidia.com> wrote:
>> On 05/04/2016 08:56 AM, Daniel Stone wrote:
>>> Right - but as with the point I was making below, GBM _right now_ is
>>> more capable than Streams _right now_. GBM right now would require API
>>> additions to match EGLStreams + EGLSwitch + Streams/KMS-interop, but
>>> the last two aren't written either, so. (More below.)
>>
>> The current behavior that enables this, where basically all Wayland buffers
>> must be allocated as scanout-capable, isn't reasonable on NVIDIA hardware.
>> The requirements for scanout are too onerous.
>
> I think we're talking past each other, so I'd like to pare the
> discussion down to these two sentences, and my two resultant points,
> for now:
>
> I posit that the Streams proposal you (plural) have put forward is, at
> best, no better at meeting these criteria:
>    - there is currently no support for direct scanout from client
> buffers in Streams, so it must always pessimise towards GPU
> composition
>    - GBM stacks can obviously do the same: implement a no-op
> gbm_bo_import, and have your client always allocate non-scanout
> buffers - presto, you've matched Streams
>
> I posit that GBM _can_ match the capability of a hypothetical
> EGLStreams/EGLSwitch implementation. Current _implementations_ of GBM
> cannot, but I posit that it is not a limitation of the API it exposes,
> and unlike Streams, the capability can be plumbed in with no new
> external API required.
>
> These seem pretty fundamental, so ... am I missing something? :\ If
> so, can you please outline fairly specifically how you think
> non-Streams implementations are not capable of meeting the criteria in
> your two sentences?

I respect the need to rein in the discussion, but I think several 
substantive aspects have been lost here.  I typed up a much longer 
response below, but I'll try to summarize in 4 sentences:

GBM could match the allocation aspects of streams used in Miguel's first 
round of patches.  However, I disagree that its core API is sufficient 
to match the allocation capabilities of EGLStream+EGLSwitch where all 
producing and consuming devices+engines are known at allocation time. 
Further, streams have additional equally valuable functionality beyond 
allocation that GBM does not seem intended to address.  Absent 
agreement, I believe co-existence of EGLStreams and GBM+wl_drm in 
Wayland/Weston is a reasonable path forward in the short term.

The longer version:

GBM alone can not perform as well as EGLStreams unless it is extended 
into something more or less the same as EGLStreams, where it knows 
exactly what engines are being used to produce the buffer content (along 
with their current configuration), and exactly what 
engines/configuration are being used to consume it.  This implies 
allocating against multiple specific objects, rather than a device and a 
set of allocation modifier flags, and/or importing an external 
allocation and hoping it meets the current requirements.  From what I 
can see, GBM fundamentally understands at most the consumer side of the 
equation.

Suppose however, GBM was taught everything streams know implicitly about 
all users of the buffers at allocation time.  After allocation, GBM is 
done with its job, but streams & drivers aren't.

The act of transitioning a buffer from optimal "producer mode" to 
optimal "consumer mode" relies on all the device & config information as 
well, meaning it would need to be fed into the graphics driver (EGL or 
whatever window system binding is used) by each window system the 
graphics driver was running on to achieve equivalent capabilities to 
EGLStream.

Fundamentally, the API-level view of individual graphics buffers as raw 
globally coherent & accessible stores of pixels with static layout is 
flawed.  Images on a GPU are more of a mutating spill space for a 
collection of state describing the side effects of various commands than 
a 2D array of pixels.  Forcing GPUs to resolve an image to a 2D array of 
pixels in any particular layout can be very inefficient.  The 
GL+GLX/EGL/etc. driver model hides this well, but it breaks down in a 
few cases like EGLImage and GLX_EXT_texture_from_pixmap, the former not 
really living up to its implied potential because of this, and the 
latter mostly working only because it has a very limited domain where 
things can be shared, but still requires a lot of platform-specific code 
to support properly.  Vulkan brings a lot more of this out into the open 
with its very explicit image state transitions and limitations on which 
engines can access an image in any given state, but that's just within 
the Vulkan API itself (I.e., strictly on a single GPU and optionally an 
associated display engine within the same driver & process) so far.

The EGLStream encapsulation takes into consideration the new use cases 
EGLImage, GBM, etc. were intended to address, and restores what I 
believe to be the minimal amount of the traditional GL+GLX/EGL/etc. 
model, while still allowing as much of the flexibility of the "a bunch 
of buffers" mental model as possible.  We can re-invent that with GBM 
API adjustments, a set of restrictions on how the buffers it allocates 
can be used, and another layer of metadata being pumped into drivers on 
top of that, but I suspect we'd wind up with something that looks very 
similar to streams.

We're both delving into future developments and hypotheticals to some 
degree here.  If we can't agree now on which direction is best, I 
believe the right solution is to allow the two to co-exist and compete 
collegially until the benefits of one or the other become more apparent. 
  The Wayland protocol and Weston compositor were designed in a manner 
that makes this as painless as possible.  It's not like we're going to 
get a ton of Wayland clients that suddenly rely on EGLStream.  At worst, 
streams lose out and some dead code needs to be deleted from any 
compositors that adopted them.  As we discussed, there is some 
maintenance cost to having two paths, but I believe it is reasonably 
contained.

Thanks,
-James

> Cheers,
> Daniel
>