This is actually a no-op on AMD, so we really don't want to lower it to
something more complicated. There may be a more efficient way to do
this on Intel too. In addition, in the future we'll want to use this for
lowering boolean reduce operations, where the inverse ballot will
operate on the backend's "natural" ballot type as indicated by
options->ballot_bit_size, instead of uvec4 as produced by SPIR-V. In
total, there are now three possible lowerings we may have to perform:
- inverse_ballot with source type of uvec4 from SPIR-V to inverse_ballot
with natural source type, when the backend supports inverse_ballot
natively.
- inverse_ballot with source type of uvec4 from SPIR-V to arithmetic,
when the backend doesn't support inverse_ballot.
- inverse_ballot with natural source type from reduce operation, when
the backend doesn't support inverse_ballot.
Previously we just did the second lowering unconditionally in vtn, but
it's just a combination of the first and third. We add support here for
the first and third lowerings in nir_lower_subgroups, instead of simply
moving the second lowering, to avoid unnecessary churn.
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25123>
In Vulkan, UBOs are lowered by nir_lower_explicit_io, and the ubo_base_agx
sysval is unused (since it doesn't handle descriptor sets). That makes the UBO
lowering GL-only and hence belongs with the GL driver rather than the compiler.
This lets us delete the ubo_base_agx sysval.
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24847>
Doing a descriptor crawl with binding tables requires a real binding table in
the shader, which won't work for VK or merged shader stages in GL. Instead,
let's lower anything that needs a crawl to bindless in the driver, so the
compiler code doesn't need to know anything about descriptor binding models.
That gets rid of the texture_base_agx sysval, which is problematic when there
are multiple descriptor sets worth of textures.
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24847>
For merging shader stages, it will be useful to express a load from an explicit
GL "descriptor set", so we can represent things like UBO loads with merged
shaders where UBOs can come from either stage. To do so, we add an intrinsic
representing a load from the driver's uniform tables, indexed like "descriptor
sets" with "bindings".
In principle, a layered GL-on-Vulkan implementation would use literal descriptor
sets for each stage, so I feel comfortable with the analogy here.
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24847>
There are two new variable modes:
- nir_var_mem_node_payload
- nir_var_mem_node_payload_in
Also add a few more intrinsics and some shader info.
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24512>
Push costants in Zink are not flat indexed like in vulkan drivers which
makes the `nir_intrinsic_load_push_constant` intrinsic inappropiate.
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24401>
sed + ninja clang-format + fix up spacing for common code.
If you are unhappy that I did not manually change the whitespace of your driver,
you need to enable clang-format for it so the formatting would happen
automatically.
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Acked-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24428>
Non of the know etnaviv GPUs support this feature in hardware
and the binary blob provides sizes via uniforms too.
Signed-off-by: Christian Gmeiner <cgmeiner@igalia.com>
Acked-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24217>
This intrinsic (vec2 tess_coord) is generally useful for non-r600 backends.
Promote it.
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24159>
Read-after-write hazards require special handling on AGX, since image loads are
implemented with texturing. Add intrinsics to handle these hazards.
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24148>
Note the writemask handling is chosen for consistency with the rest of NIR. In
every other instance, writemask=w requires a vec4 source. This is hardcoded into
nir_validate and nir_print as what it means to have a writemask.
More importantly, consistency with how register writemasks currently work.
nir_print hides it, but r0.w = fneg ssa_1.x is actually a vec4 instruction with
source ssa_1.xxxx. As a silly example nir_dest_num_components(that) = 4 in the
old model. I realize this is quite strange coming from a scalar ISA, but it's
perfectly natural for the class of vec4 hardware for which this was designed. In
that hardware, conceptually all instructions are vec4`, so the sequence "fneg
ssa_1 and write to channel w" is implemented as "fneg a vec4 with ssa_1.x in the
last component and write that vec4 out but mask to write only the w channel".
Isn't this inefficient? It can be. To save power, Midgard has scalar ALUs in
addition to vec4 ALUs. Those details are confined to the backend VLIW scheduler;
the instruction selection is still done as vec4. This mechanism has little in
common with AMD's SALUs. Midgard has a wave size of 1, with special hacks for
derivatives.
As a result, all backends consuming register writemasks are expecting this
pattern of code. Changing the store to take a vec1 instead of a vec4 would
require changing every backend to reswizzle the sources to resurrect the vec4. I
started typing a branch to do this yesterday, but it made a mess of both Midgard
and nir-to-tgsi. Without any good reason to think it'd actually help
performance, I abandoned the idea. Getting all 15 backends converted to the
helpers is enough of a challenge without forcing 10 backends to reswizzle their
sources too.
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23089>
On some architectures, gl_FragCoord.xy is available as an integer but
gl_FragCoord.zw requires interpolation. Add dedicated intrinsics so we can
lower it all in NIR.
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23836>
sample_mask_agx corresponds directly to the hardware's 2-source instruction, but
it's hard to use correctly and even harder to legalize after the fact, since
it's responsible for not only discard but also late depth/stencil testing. For
our various high-level lowering passes, it's easier to use a one-source discard
(where we don't have to worry about sample masks), which the compiler will
internally lower to the two-source instruction. Introduce such an instruction.
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Asahi Lina <lina@asahilina.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23832>
Derefs have index-based access semantics, which means we don't need
custom intrinsics to encode an index instead of a byte offset.
Remove the "masked" store intrinsics and just emit the pair of atomics
directly. This massively reduces duplication between scratch, shared,
and constant, while also moving more things into nir so more optimizations
can be done.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23173>
There's a few changes in here that are very inter-related.
First, we stop lowering load_deref on shader_temp to load_ptr_dxil,
and just leave it as load_deref. In order for that to work, we need
the derefs to be in a shape that's acceptable to DXIL, so the only
current producer of shader_temp loads (the CLC frontend) needs to
run some lowering passes on them first.
The DXIL backend is augmented to just write out deref indices while
walking a deref chain, which will get combined in the load op into
a GEP instruction. For non-mesh/raytracing shaders, these are required
to be single-level scalar arrays, but the complexity here is preparation
for when we don't need to do that anymore.
Additionally, the const lookups are changed from using a hash table
to just putting an index on the variable.
All of this together is enough to enable the authored-forever-ago test
which uses indirect array access into a const packed struct. The
load_ptr_dxil handling didn't deal with packed structs / unaligned
accesses, but now that we're in a logical address space with derefs
instead of physical, there's no alignment to deal with anymore and
the fact that it's packed goes out the window.
This removes one custom DXIL intrinsic.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23173>
This is a piece of cake with unified atomics :-) This will let us do our
addressing math tricks nice and easily.
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Acked-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23529>
For VK_KHR_fragment_shader_barycentric, AMD needs to know the primitive
topology in the fragment shader but with fast-link GPL this is unknown
at compile time and it needs to be passed dynamically.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16742>
This introduces new intrinsics nir_intrinsic_load_barycentric_coord_xxx
with 3-components instead of expanding the existing ones that are
supposed to interpolate input varyings, while BaryCoord is a sysval
on most hardware.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23254>
Now that radeonsi support pass desc to ssbo atomic ops,
we can use ssbo atomic instead. aco does not implement
nir_buffer_atomic_add either.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23096>
Intel HW has multiple ways to access resources like UBO/SSBO/images :
- binding tables : a small ~240 heap of surfaces
- bindless surfaces : a 64Mb heap of surfaces up to Gfx12+, 4Gb on Gfx12.5+
- surfaces : a 4Gb heap on Gfx12.5+ (mostly unused at the moment,
only available through the LSC)
For samplers, we have 2 options since Gfx11+ :
- samplers indexed from the Dynamic State Heap (4Gb)
- samplers indexed from the Bindless Sampler Heap (4Gb)
Additionally our whole push constant promotion mechanism is based
around binding table indices. This is problematic if you want to also
promote to push constants things that would be accessed through the
bindless heap.
To solve this issue, we introduce a new intrinsic that will cary a
block index that is not based off the binding table index nor the
bindless table offset.
We will also use this intrinsic to identify whether the buffer/surface
index in load_ubo/load_ssbo/store_ssbo/etc... is relative to the
binding table or the bindless heap.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21645>
sample_mask_agx maps to the AGX instruction used to write out a sample mask.
api_sample_mask_agx is a system value that returns the value of glSampleMask
(or its Vulkan equivalent), used to lower glSampleMask (etc).
This is distinct from sample_mask_in, which we map to the hardware thing and
AND with this as a lowering.
sample_positions_agx is a system value returning the sample positions in a
packed fixed-point format matching the hardware register, used to lower
gl_SamplePositions.
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Asahi Lina <lina@asahilina.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23040>
Some hardware has an instruction to load the address of a texel in a writeable
image, given the coordinates ("LEA_IMAGE"). This operation is defined only for
uncompressed images, but it is well-defined regardless of the underlying
twiddling. As such, it is not expected to be produced by APIs but is useful for
internal lowering when it is known that images will be uncompressed (e.g.
because image_store does not support compression on the hardware).
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23120>
I was scratching my head about this for a few minutes until I found the answer
in spirv_to_nir. Hopefully this saves someone else some head scratching in turn.
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23120>
To lower smooth lines conditionally in fragment shaders for RADV
because the line rasterization mode in Vulkan can be dynamic.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21587>
The intrinsics are now totally dead and can be removed.
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Emma Anholt <emma@anholt.net>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23036>
committed has to be a constant so there is no need to have a src and
depend on constant folding to remove the i2b.
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Acked-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22963>
Currently, we have an atomic intrinsic for each combination of memory type
(global, shared, image, etc) and atomic operation (add, sub, etc). So for m
types of memory supported by the driver and n atomic opcodes, the driver has to
handle O(mn) intrinsics. This makes a total mess in every single backend I've
looked at, without fail.
It would be a lot nicer to unify the intrinsics. There are two obvious ways:
1. Make the memory type a constant index, keep different intrinsics for
different operations. The problem with this is that different memory types
imply different intrinsic signatures (number of sources, etc). As an
example, it doesn't make sense to unify global_atomic_amd with
global_atomic_2x32, as an example. The first takes 3 scalar sources, the
second takes 1 vector and 1 scalar. Also, in any single backend, there are a
lot more operations than there are memory types.
2. Make the opcode a constant index, keep different intrinsics for different
operations. This works well, with one exception: compswap and fcompswap
take an extra argument that other atomics don't, so there's an extra axis of
variation for the intrinsic signatures.
So, the solution is to have 2 intrinsics for each memory type -- for atomics
taking 1 argument and atomics taking 2 respectively. Both of these intrinsics
take an nir_atomic_op enum to describe its operation. We don't use a nir_op for
this purpose, as there are some atomics (cmpxchg, inc_wrap, etc) that don't
cleanly map to any ALU op and it would be weird to force it.
The plan is to transition to these new opcodes gradually. This series adds a
lowering pass producing these opcodes from the existing opcodes, so that
backends can opt-in to the new forms one-by-one. Then we can convert backends
separately without any cross-tree flag day. Once everything is converted, we can
convert the producers and core NIR as a flag day, but we have far fewer
producers than backends so this should be fine. Finally we can drop the old
stuff.
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rob Clark <robclark@freedesktop.org>
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22914>
When rendering a scaled tile, we need to use the original, hardware
FragCoord when accessing input attachments that are on-tile (i.e. were
rendered to in a previous subpass) because they are also scaled in the
same way that FragCoord is scaled. For input attachments that aren't
already on-tile, however, we need to use the fixed gl_FragCoord. Add a
new intrinsic and a bitfield of input attachments which should use it.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20304>
