Some Metal devices have a bug with depth array textures using comparison
with explicit LoD, where the LoD given will be biased by some amount.
For these devices, we can use a gradient instead, which does not exhibit
this problem. As with the fragment demote workaround, this is only
expected to be needed until the bug is fixed in Metal.
- Add CompilerMSL::Options::argument_buffers_tier as an enumeration to
allow calling app to specify platform argument buffer tier capabilities.
- Support iOS writable images in Tier2 argument buffers when specified.
Tier capabilities based on recommendations from Apple engineering.
Some Metal devices have a bug where storage resources can still be
written to even if the fragment is discarded. This is obviously a bug in
Metal, but bothering Apple to fix it will only fix it for newer
versions; therefore, a workaround is needed for older versions. I have
made this an option so that, in case the bug is ever fixed, the
workaround can be disabled.
This workaround is simple: if a fragment shader may discard its fragment
and writes to a storage resource, a variable representing the
`HelperInvocation` built-in is created and passed to all functions. The
flag is checked on all resource writes; writes do not occur when
`HelperInvocation` is `true`. This relies on the earlier workaround to
update `HelperInvocation` when the fragment is discarded.
Fixes at least 3 failures in the CTS.
Some Metal devices have a bug where `simd_is_helper_thread()` won't
return true after a fragment has been discarded. We can work around this
by manually setting `gl_HelperInvocation` upon discarding a fragment.
This is fairly unintrusive, so it is enabled by default. I've made it an
option so that, when the bug is fixed, we can disable it.
Flattening doesn't play well with dynamic indices. In this case, it's
better to leave it as an array of structs.
(I wanted to do this for named blocks generally. Trouble is, the builtin
`gl_out` block is *also* a named block...)
Fixes six more CTS tests, under
`dEQP-VK.tessellation.user_defined_io.per_patch_block_array.*`.
Using vertex-style stage input is complex, and it doesn't support
nesting of structures or arrays. By using raw buffer input instead, we
get this support "for free," and everything becomes much simpler.
Arguably, this is the way I should've done this in the first place.
Eventually, I'd like to make this the default, and then remove the
option altogether. (And I still need to do that with
`multi_patch_workgroup`...)
Should help fix 66 tests in the Vulkan CTS, under the following trees:
- `dEQP-VK.pipeline.*.interface_matching.*`
- `dEQP-VK.tessellation.user_defined_io.*`
- `dEQP-VK.clipping.user_defined.*`
This is analogous to the existing support for fixing up shader inputs.
It is intended to be used with tessellation to add implicit builtins
that are read from a later stage, despite not being written in an
earlier stage. (Believe it or not, this is in fact legal in Vulkan.)
Helps fix 8 CTS tests under `dEQP-VK.pipeline.*.no_position`. (Eight
other tests work solely by accident without this change.)
Makes codegen from typical D3D emulation SPIR-V more readable.
Also makes cross compilation with NotEqual more sensible.
It's very rare to actually need the strict NaN-checks in practice.
Also, glslang now emits UnordNotEqual by default it seems, so give up
trying to assume OrdNotEqual. Harmonize for UnordNotEqual as the sane
default.
In Metal, the `[[position]]` input to a fragment shader remains at
fragment center, even at sample rate, like OpenGL and Direct3D. In
Vulkan, however, when the fragment shader runs at sample rate, the
`FragCoord` builtin moves to the sample position in the framebuffer,
instead of the fragment center. To account for this difference, adjust
the `FragCoord`, if present, by the sample position. The -0.5 offset is
because the fragment center is at (0.5, 0.5).
Also, add an option to force sample-rate shading in a fragment shader.
Since Metal has no explicit control for this, this is done by adding a
dummy `[[sample_id]]` which is otherwise unused, if none is already
present. This is intended to be used from e.g. MoltenVK when a
pipeline's `minSampleShading` value is nonzero.
Instead of checking if any `Input` variables have `Sample`
interpolation, I've elected to check that the `SampleRateShading`
capability is present. Since `SampleId`, `SamplePosition`, and the
`Sample` interpolation decoration require this cap, this should be
equivalent for any valid SPIR-V module. If this isn't acceptable, let me
know.
Add support for declaring a fixed subgroup size. Metal, like Vulkan with
`VK_EXT_subgroup_size_control`, allows the thread execution width to
vary depending on factors such as register usage. Unfortunately, this
breaks several tests that depend on the subgroup size being what the
device says it is. So we'll fix the subgroup size at the size the device
declares. The extra invocations in the subgroup will appear to be
inactive. Because of this, the ballot mask builtins are now ANDed with
the active subgroup mask.
Add support for emulating a subgroup of size 1. This is intended to be
used by Vulkan Portability implementations (e.g. MoltenVK) when the
hardware/software combo provides insufficient support for subgroups.
Luckily for us, Vulkan 1.1 only requires that the subgroup size be at
least 1.
Add support for quadgroup and SIMD-group functions which were added to
iOS in Metal 2.2 and 2.3. This will allow clients to take advantage of
expanded quadgroup and SIMD-group support in recent Metal versions and
on recent Apple GPUs (families 6 and 7).
Gut emulation of subgroup builtins in fragment shaders. It turns out
codegen for the SIMD-group functions in fragment wasn't implemented for
AMD on Mojave; it's a safe bet that it wasn't implemented for the other
drivers either. Subgroup support in fragment shaders now requires Metal
2.2.
Fix reversed coordinates: `y` should be used to calculate the row
address. Align row address to the row stride.
I've made the row alignment a function constant; this makes it possible
to override it at pipeline compile time.
Honestly, I don't know how this worked at all for Epic. It definitely
didn't work in the CTS prior to this.
These need to use arrayed texture types, or Metal will complain when
binding the resource. The target layer is addressed relative to the
Layer output by the vertex pipeline, or to the ViewIndex if in a
multiview pipeline. Unlike with the s/t coordinates, Vulkan does not
forbid non-zero layer coordinates here, though this cannot be expressed
in Vulkan GLSL.
Supporting 3D textures will require additional work. Part of the problem
is that Metal does not allow texture views to subset a 3D texture, so we
need some way to pass the base depth to the shader.
Some older iOS devices don't support layered rendering. In that case,
don't set `[[render_target_array_index]]`, because the compiler will
reject the shader in that case. The client will then have to unroll the
render pass manually.
This should hopefully reduce underutilization of the GPU, especially on
GPUs where the thread execution width is greater than the number of
control points.
This also simplifies initialization by reading the buffer directly
instead of using Metal's vertex-attribute-in-compute support. It turns
out the only way in which shader stages are allowed to differ in their
interfaces is in the number of components per vector; the base type must
be the same. Since we are using the raw buffer instead of attributes, we
can now also emit arrays and matrices directly into the buffer, instead
of flattening them and then unpacking them. Structs are still flattened,
however; this is due to the need to handle vectors with fewer components
than were output, and I think handling this while also directly emitting
structs could get ugly.
Another advantage of this scheme is that the extra invocations needed to
read the attributes when there were more input than output points are
now no more. The number of threads per workgroup is now lcm(SIMD-size,
output control points). This should ensure we always process a whole
number of patches per workgroup.
To avoid complexity handling indices in the tessellation control shader,
I've also changed the way vertex shaders for tessellation are handled.
They are now compute kernels using Metal's support for vertex-style
stage input. This lets us always emit vertices into the buffer in order
of vertex shader execution. Now we no longer have to deal with indexing
in the tessellation control shader. This also fixes a long-standing
issue where if an index were greater than the number of vertices to
draw, the vertex shader would wind up writing outside the buffer, and
the vertex would be lost.
This is a breaking change, and I know SPIRV-Cross has other clients, so
I've hidden this behind an option for now. In the future, I want to
remove this option and make it the default.
Metal is picky about interface matching. If the types don't match
exactly, down to the number of vector components, Metal fails pipline
compilation. To support pipelines where the number of components
consumed by the fragment shader is less than that produced by the vertex
shader, we have to fix up the fragment shader to accept all the
components produced.
Like with `point_size` when not rendering points, Metal complains when
writing to a variable using the `[[depth]]` qualifier when no depth
buffer be attached. In that case, we must avoid emitting `FragDepth`,
just like with `PointSize`.
I assume it will also complain if there be no stencil attachment and the
shader write to `[[stencil]]`, or it write to `[[color(n)]]` but there
be no color attachment at n.
Here, the inline uniform block is explicit: we instantiate the buffer
block itself in the argument buffer, instead of a pointer to the buffer.
I just hope this will work with the `MTLArgumentDescriptor` API...
Note that Metal recursively assigns individual members of embedded
structs IDs. This means for automatic assignment that we have to
calculate the binding stride for a given buffer block. For MoltenVK,
we'll simply increment the ID by the size of the inline uniform block.
Then the later IDs will never conflict with the inline uniform block. We
can get away with this because Metal doesn't require that IDs be
contiguous, only monotonically increasing.
If there are enough members in an IAB, we cannot use the constant
address space as MSL compiler complains about there being too many
members. Support emitting the device address space instead.
