This could return the graphics DCC pipeline if it was created before,
and crash or potentially hang the GPU.
Found this while working on in-progress VKCTS coverage.
Cc: mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39689>
This was an oversight of VK_KHR_dynamic_rendering_local_read which has
been addressed by VK_KHR_maintenance10 which introduced new flags to
give more information to implementations.
The Vulkan spec says:
"VK_RENDERING_ATTACHMENT_INPUT_ATTACHMENT_FEEDBACK_BIT_KHR is
intended to give implementations similar information as a subpass
where an attachment could be used as both a color attachment and
input attachment. Some implementations require extra work to make
this scenario work beyond just considering the image layouts.
Implementations which have no such considerations may treat this
flag as a noop. The primary use case for this flag is to enable
feedback loops inside a single shader."
"Applications are encouraged to use
VK_RENDERING_LOCAL_READ_CONCURRENT_ACCESS_CONTROL_BIT_KHR if
maintenance10 is available and they use feedback loops with
VK_KHR_dynamic_rendering_local_read. Feedback loops are still
allowed when not using the rendering flag, but the performance
implication was an oversight in the original definition of
VK_KHR_dynamic_rendering_local_read."
Because it's clearly defined by the Vulkan spec, let's just pessimize
always to avoid relying on some shaders state which require to do very
late decompression passes. This will allow us to do more cleanups and
optimizations related to the framebuffer. Also note that DRLR is still
a niche feature.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39538>
If the rendering state is inherited in the secondary, otherwise nothing
wait for the pending flushes after a decompression pass. One more
argument to stop delaying this.
Fixes
dEQP-VK.renderpasses.dynamic_rendering.partial_secondary_cmd_buff.local_read.*
Cc: mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39678>
Avoiding NaNs should have the same effect but it's good practice to not
rely on float OPs for correctness.
Fixes: 95a89f7 ("radv: Report smaller bvh sizes when possible")
Reviewed-by: Natalie Vock <natalie.vock@gmx.de>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39640>
It will lower txf and buffer image loads to load_buffer_amd.
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39474>
It was incorrect to mark chit/miss arguments as discardable without
the equivalent in the traversal shader. Also, tail calls with modified
parameters that aren't marked discardable are incorrect.
This could lead to random corruption by clobbering parameter values
across two levels of nested calls: A Raygen shader calls traversal,
expecting e.g. the ray tMax parameter to be preserved. Traversal
overwrites the parameter's register with the hit t and tail-calls chit,
which immediately returns to raygen. Now the raygen shader still has the
clobbered tMax (which is actually the ray hit t) - if it calls traversal
multiple times, the second traversal iteration may use the previous
ray's hit t as tMax instead of the intended value.
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39579>
There were two issues here:
1. Tail calls where the tail-callee receives modified parameters are
hazardous and only work if the parameter is return or discardable.
Otherwise, the caller of the function that executes the tail-call may
not expect some of the parameters to be clobbered.
2. There was also an indexing confusion with the call instruction vs.
call signature parameters. The call instruction has not been adapted
to the new lowered signatures, where the system args are prepended. To
make things clearer, split the loop into two, one iterating over
parameters in the call signature and one for parameters of the call
instruction.
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39579>
Also change to use H265 constant for maxDpbSlots (both values for H264 and H265
are the same).
Fixes: ee535aa039 ("radv: video: rework maxActiveReferenceSlot/MaxDpbSlots")
Reviewed-by: Benjamin Cheng <benjamin.cheng@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39609>
This is just wrong if the secondary uses ESO because the emitted
pipelines would be NULL in the secondary, but if the app re-binds
the same pipeline in the primary it would consider it as already
emitted. A sequence like this would break:
CmdBindPipeline(compute)
CmdDispatch()
CmdExecuteCommands() --> with ESO compute
CmdBindPipeline(compute)
CmdDispatch()
This tracking is probably useless anyways because it's unlikely that
apps will rebind the same pipeline right after CmdExecuteCommands() but
let's keep it because this is a bugfix.
Fixes
dEQP-VK.api.command_buffers.pipeline_shader_object_mix_with_secondaries.
Cc: mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39587>
It might be that the radv_pipeline_cache_lookup_nir_handle() in
radv_ray_tracing_pipeline_cache_search() fails but we will later need the
NIR. If rt_stages[i].shader was non-NULL, then we would not have created
the NIR.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Backport-to: 25.2
Reviewed-by: Konstantin Seurer <konstantin.seurer@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38263>
cef8eff74d ("radv/video: Override H265 SPS unaligned resolutions")
fixes the case where app specifies resolution with lower than required
alignment. But in case of higher alignment, the stream is still not
going to be correctly decodable.
Use size from session params to set the coded size, instead of using
codedExtent of input image.
Only use codedExtent to calculate padding.
Fixes dEQP-VK.video.encode.h265.quantization_map_delta*
Reviewed-by: Benjamin Cheng <benjamin.cheng@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39529>
HiZ must only be cleared when the full HiZ workaround is enabled. This
means that the previous slow clear draw would disable HiZ because it
hits the conditions (ie. depth/stencil enable and depth writes enabled).
So, the draw and the dispatch can run in parallel by moving the barrier
earlier.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39433>
Needs to consider the base offset, otherwise it's resolving to the
first 3D slice.
Fixes very recent VKCTS coverage dEQP-VK.pipeline.*.multisample.m10_resolve.*.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39393>
Some apps (old FFmpeg, contemporary CTS) send down pMi{Col,Row}Starts in
SB units, not MI units. Instead of dependening on those values which
could be unreliable, derive the tile sizes in SB using other parameters.
Cc: mesa-stable
Reviewed-by: Ruijing Dong <ruijing.dong@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39492>
radv_dump_shader_stats() printed stats for every shader with a certain
stage, and we called this function each time an RT shader is compiled.
This means we could repeat the stats for a shader.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39484>
