This normally gets printed by libplacebo itself when initializing the
context, but due to the way our code is structured (for convenience) we
don't have the log hook enabled by the time this function call is
relevant. So instead just print it manually as an easier work-around
than restructuring the code.
This commit rips out the entire mpv vulkan implementation in favor of
exposing lightweight wrappers on top of libplacebo instead, which
provides much of the same except in a more up-to-date and polished form.
This (finally) unifies the code base between mpv and libplacebo, which
is something I've been hoping to do for a long time.
Note: The ra_pl wrappers are abstract enough from the actual libplacebo
device type that we can in theory re-use them for other devices like
d3d11 or even opengl in the future, so I moved them to a separate
directory for the time being. However, the rest of the code is still
vulkan-specific, so I've kept the "vulkan" naming and file paths, rather
than introducing a new `--gpu-api` type. (Which would have been ended up
with significantly more code duplicaiton)
Plus, the code and functionality is similar enough that for most users
this should just be a straight-up drop-in replacement.
Note: This commit excludes some changes; specifically, the updates to
context_win and hwdec_cuda are deferred to separate commits for
authorship reasons.
Makes performance slightly better when using multiple queues by avoiding
unnecessary semaphores due to bad queue selection.
Also remove an aeons-old workaround for an nvidia bug that only ever
existed in the earliest beta vulkan drivers anyway.
I was inconsistent about this originally, as the functionality was
moved into the core spec in 1.1 and so both suffixed and unsuffixed
versions of everything exist and can be mixed together.
There's no reason to fail to build with 1.0.39+ so I'm fixing the
names.
The CUDA/Vulkan interop works on the basis of memory being exported
from Vulkan and then imported by CUDA. To enable this, we add a way
to declare a buffer as being intended for export, and then add a
function to do the export.
For now, we support the fd and Handle based exports on Linux and
Windows respectively. There are others, which we can support when
a need arises.
Also note that this is just for exporting buffers, rather than
textures (VkImages). Image import on the CUDA side is supposed to
work, but it is currently buggy and waiting for a new driver release.
Finally, at least with my nvidia hardware and drivers, everything
seems to work even if we don't initialise the buffer with the right
exportability options. Nevertheless I'm enforcing it so that we're
following the spec.
Instead of enabling every feature under the sun, make an effort to just
whitelist the ones we actually might use. Turns out the extended storage
format support is needed for some of the storage formats we use, in
particular rgba16.
The queue family index and the queue info index are not necessarily the
same, so we're forced to do a check based on the queue family index
itself.
Fixes#5049
Async compute in particular seems to cause problems on some drivers, and
even when supprted the benefits are not that massive from the tests I
have seen, so it's probably safe to keep off by default.
Async transfer on the other hand seems to work better and offers a more
substantial improvement, so it's kept on.
This gets confused by e.g. SPARSE_BIT on the TRANSFER_BIT, leading to
situations where "more specialized" is ambiguous and the logic breaks
down. So to fix it, only compare the subset we care about.
Instead of using a single primary queue, we generate multiple
vk_cmdpools and pick the right one dynamically based on the intent.
This has a number of immediate benefits:
1. We can use async texture uploads
2. We can use the DMA engine for buffer updates
3. We can benefit from async compute on AMD GPUs
Unfortunately, the major downside is that due to the lack of QF
ownership tracking, we need to use CONCURRENT sharing for all resources
(buffers *and* images!). In theory, we could try figuring out a way to
get rid of the concurrent sharing for buffers (which is only needed for
compute shader UBOs), but even so, the concurrent sharing mode doesn't
really seem to have a significant impact over here (nvidia). It's
possible that other platforms may disagree.
Our deadlock-avoidance strategy is stupidly simple: Just flush the
command every time we need to switch queues, and make sure all
submission and callbacks happen in FIFO order. This required lifting the
cmds_pending and cmds_queued out from vk_cmdpool to mpvk_ctx, and some
functions died/got moved as a result, but that's a relatively minor
change.
On my hardware this is a fairly significant performance boost, mainly
due to async transfers. (Nvidia doesn't expose separate compute queues
anyway). On AMD, this should be a performance boost as well due to async
compute.
This combines VkSemaphores and VkEvents into a common umbrella
abstraction which can resolve to either.
We aggressively try to prefer VkEvents over VkSemaphores whenever the
conditions are met (1. we can unsignal the semaphore, i.e. it comes from
the same frame; and 2. it comes from the same queue).
Instead of being submitted immediately, commands are appended into an
internal submission queue, and the actual submission is done once per
frame (at the same time as queue cycling). Again, the benefits are not
immediately obvious because nothing benefits from this yet, but it will
make more sense for an upcoming vk_signal mechanism.
This also cleans up the way the ra_vk submission interacts with the
synchronization/callbacks from the ra_vk_ctx. Although currently, the
way the dependency is signalled is a bit hacky: normally it would be
associated with the ra_tex itself and waited on in the appropriate stage
implicitly. But that code is just temporary, so I'm keeping it in there
for a better commit order.
Instead of associating a single VkSemaphore with every command buffer
and allowing the user to ad-hoc wait on it during submission, make the
raw semaphores-to-signal array work like the raw semaphores-to-wait-on
array. Doesn't really provide a clear benefit yet, but it's required for
upcoming modifications.
1. No more static arrays (deps / callbacks / queues / cmds)
2. Allows safely recording multiple commands at the same time
3. Uses resources optimally by never over-allocating commands
FlagBits is just the name of the enum. The actual data type representing
a combination of these flags follows the *Flags convention. (The
relevant difference is that the latter is defined to be uint32_t instead
of left implicit)
For consistency, use *Flags everywhere instead of randomly switching
between *Flags and *FlagBits.
Also fix a wrong type name on `stageFlags`, pointed out by @atomnuker
In addition to the built-in nvidia compiler, we now also support a
backend based on libshaderc. shaderc is sort of like glslang except it
has a C API and is available as a dynamic library.
The generated SPIR-V is now cached alongside the VkPipeline in the
cached_program. We use a special cache header to ensure validity of this
cache before passing it blindly to the vulkan implementation, since
passing invalid SPIR-V can cause all sorts of nasty things. It's also
designed to self-invalidate if the compiler gets better, by offering a
catch-all `int compiler_version` that implementations can use as a cache
invalidation marker.
This time based on ra/vo_gpu. 2017 is the year of the vulkan desktop!
Current problems / limitations / improvement opportunities:
1. The swapchain/flipping code violates the vulkan spec, by assuming
that the presentation queue will be bounded (in cases where rendering
is significantly faster than vsync). But apparently, there's simply
no better way to do this right now, to the point where even the
stupid cube.c examples from LunarG etc. do it wrong.
(cf. https://github.com/KhronosGroup/Vulkan-Docs/issues/370)
2. The memory allocator could be improved. (This is a universal
constant)
3. Could explore using push descriptors instead of descriptor sets,
especially since we expect to switch descriptors semi-often for some
passes (like interpolation). Probably won't make a difference, but
the synchronization overhead might be a factor. Who knows.
4. Parallelism across frames / async transfer is not well-defined, we
either need to use a better semaphore / command buffer strategy or a
resource pooling layer to safely handle cross-frame parallelism.
(That said, I gave resource pooling a try and was not happy with the
result at all - so I'm still exploring the semaphore strategy)
5. We aggressively use pipeline barriers where events would offer a much
more fine-grained synchronization mechanism. As a result of this, we
might be suffering from GPU bubbles due to too-short dependencies on
objects. (That said, I'm also exploring the use of semaphores as a an
ordering tactic which would allow cross-frame time slicing in theory)
Some minor changes to the vo_gpu and infrastructure, but nothing
consequential.
NOTE: For safety, all use of asynchronous commands / multiple command
pools is currently disabled completely. There are some left-over relics
of this in the code (e.g. the distinction between dev_poll and
pool_poll), but that is kept in place mostly because this will be
re-extended in the future (vulkan rev 2).
The queue count is also currently capped to 1, because of the lack of
cross-frame semaphores means we need the implicit synchronization from
the same-queue semantics to guarantee a correct result.
Niklas Haas
Philip Langdale
Rostislav Pehlivanov