A begin/end sequence is something like (it's all macros based):
radeon_begin(cs);
radeon_emit(PKT3(PKT3_DRAW_INDEX_AUTO, 1, cmd_buffer->state.predicating));
radeon_emit(vertex_count);
radeon_emit(V_0287F0_DI_SRC_SEL_AUTO_INDEX | use_opaque);
radeon_end();
This is loosely based on RadeonSI (see !8653 (a0978fff)) and it seems
indeed faster overall.
The main goal of this rework is to re-use the same logic as RadeonSI
for paired packets on GFX12 (also GFX11 dGPUs) because it's supposed
to be way faster, especially on GFX12 where the CP is slow. The other
goal is to share more cmdbuf emission between both drivers in the near
future.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34229>
With all the RT-enabled driver setting this field, we can now have the
runtime use it instead of calling into the driver's vfunc.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34026>
Forcing a dedicated sparse queue is problematic in real-world scenarios.
In the current implicit sync world for sparse updates, we can rely on
submission order.
For use cases where an application can take advantage of the separate
sparse queue to do "async" updates, the existing implementation works
well, but problems arise when trying to implement D3D-style submission
ordering. E.g., when a game does sparse on a graphics or compute queue,
we need to guarantee that previous submissions, sparse update and future
submissions are properly ordered.
The Vulkan way of implementing this is to:
- Signal graphics queue to timeline N (i.e. last submission made)
- Wait on timeline N on the sparse queue
- Do sparse updates
- Signal timeline N + 1 on sparse queue
- Wait for timeline N + 1 on graphics queue (can be deferred until next
graphics submit)
This causes an unavoidable bubble in GPU execution, since the
existing sparse queue ends up doing:
- Wait pending signal. The implication here is that all previous GPU
work must have been submitted.
- Do VM operations on CPU timeline
- Wait for semaphores to signal (this is required for signal ordering)
- ... GPU is meanwhile stalling in a bubble due to GPU -> CPU -> GPU roundtrip.
- Signal semaphore on CPU (unblocks GPU work)
Letting the GPU go idle here is not great, and we can be screwed over by bad thread scheduling.
Another knock-on effect is that the graphics queue is now forced into
using a thread for submissions. This is because when the graphics queue
wants to wait for timeline N + 1, the sparse queue may not have
signalled the timeline yet on CPU, so effectively, we have created a
wait-before-signal situation internally in RADV. Throwing another thread
under the bus is not great either.
Just letting the queue in question support sparse binding solves all
these issues and I don't see a path forward where the D3D use case can
be solved in a separate queue world.
It is also friendlier to the ecosystem at large. RADV is the only driver
I know of that insists on separate sparse queues and multiple games
assume that graphics queue can support sparse.
Signed-off-by: Hans-Kristian Arntzen <post@arntzen-software.no>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33166>
This used to abort (see the previous commit) when the hardware wasn't
able to sample all SPM counters because the BO was too small. The SPM
BO can now be resized like the SQTT BO.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31883>
The major differences compared to the NV extensions are:
- support for the sequence index as push constants
- support for draw with count tokens (note that DrawID is zero for
normal draws)
- support for raytracing
- support for IES (only compute is supported for now)
- improved preprocessing support with the state command buffer param
The NV DGC extensions were only enabled for vkd3d-proton and it will
maintain both paths for a while, so they can be replaced by the EXT.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31383>
vkd3d-proton always sets the acceleration structure size to be the
whole buffer size. Because of that, allocating read back buffers
for all acceleration structures causes a system with a finite amount
of RAM to OOM.
This is solved by allocating read back buffers on build where the
required size is known.
Reviewed-by: Friedrich Vock <friedrich.vock@gmx.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29537>
This commit adds an implementation of removesuffix so we don't
need the 'str' one which was added in 3.9.
Acked-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Acked-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28831>
DebugMarkerSetObjectNameEXT is just a less powerful version of
SetDebugUtilsObjectNameEXT. Fixes the objectType cast warning as well.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27975>
The submit_sub_index field is used by RGP to determine the number of
submits. Previously, it was incorrectly reporting the same number of
submits than command buffers.
Fixes: 88cbe32048 ("radv: add support for RGP queue events")
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27439>
This only dumps the begin tokens. Tokens are written to a buffer
containing a 12 byte header at the beginning.
We use an intermediate format for the ray history tokens because the RRA
format is very inefficient.
Reviewed-by: Friedrich Vock <friedrich.vock@gmx.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25548>
This way, we can see exactly which dispatches belong to acceleration
structure builds and which dispatches were caused by the app. We loose
the red coloring in RGP for this, but we get way more information.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26697>
Detroit: Become Human has an optimization issue where the same BO is
mapped and unmapped on a per-frame basis, which leads to massive
overhead in the kernel, both creating the mapping and taking page
faults.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26942>
