Commit Graph
163 Commits
Author SHA1 Message Date
Alexandre Mommers 92a1702c3c [naga wgsl-in] Add support for unsigned types when calling textureLoad with the level parameter. (#7058) 2025-02-24 21:13:06 +00:00
Samson 2f255edc60 [naga] Use const ctx instead of global ctx for type resolution (#6935)
Signed-off-by: sagudev <16504129+sagudev@users.noreply.github.com>
2025-02-24 16:24:37 +01:00
Connor FitzgeraldandErich Gubler 473dd2fdd9 Refactor Snapshots to Be Fully Configuration Based (#7153)
Co-authored-by: Erich Gubler <erichdongubler@gmail.com>
2025-02-21 18:22:54 -05:00
Jim Blandy 5dad2c4cd4 [naga] Apply module compaction to GLSL tests. (#7186) 2025-02-19 20:59:06 -05:00
Connor Fitzgerald e590555a8c test: use spirv-as instead of pre-assembled SPIR-V (#7157) 2025-02-17 14:48:09 -05:00
Jamie Nicol d625d083c3 [naga wgsl-in] Implement any() and all() during const evaluation (#7166) 2025-02-17 14:13:49 -05:00
Jamie Nicol 6cf4b710b8 [naga spv-out] Avoid undefined behaviour for integer division and modulo
Integer division or modulo is undefined behaviour in SPIR-V when the
divisor is zero, or when the dividend is the most negative number
representable by the result type and the divisor is negative one.

This patch makes us avoid this undefined behaviour and instead ensures
we adhere to the WGSL spec in these cases: for divisions the
expression evaluates to the value of the dividend, and for modulos the
expression evaluates to zero.

Similarily to how we handle these cases for the MSL and HLSL backends,
prior to emitting each function we emit code for any helper functions
required by that function's expressions. In this case that is helper
functions for integer division and modulo. Then, when emitting the
actual function's body, if we encounter an expression which needs
wrapped we instead emit a function call to the helper.
2025-02-14 15:17:40 -08:00
Jamie Nicol b6186ba332 [naga hlsl-out] Avoid undefined behaviour for integer division, modulo, negation, and abs
Emit helper functions for MathFunction::Abs and UnaryOperator::Negate
with a signed integer scalar or vector operand. And for
BinaryOperator::Divide and BinaryOperator::Modulo with signed or
unsigned integer scalar or vector operands.

Abs and Negate are written to avoid signed integer overflow when the
operand equals INT_MIN. This is achieved by bitcasting the value to
unsigned, using the negation operator, then bitcasting the result back
to signed. As HLSL's bitcast functions asint() and asuint() only work
for 32-bit types, we only use this workaround in such cases.

Division and Modulo avoid undefined behaviour for INT_MIN / -1 and
divide-by-zero by using 1 for the divisor instead. Additionally we
avoid undefined behaviour when using the modulo operator on operands
of mixed signedness by using the equation from the WGSL spec, using
division, subtraction and multiplication, rather than HLSL's modulus
operator.
2025-02-14 15:17:40 -08:00
Jamie Nicol 88f6e9b8f8 [naga hlsl-out] Avoid undefined behaviour for signed integer addition, subtraction, and multiplication
Though not explicitly specified one way or the other, we have been
informed by DirectX engineers that signed integer overflow may be
undefined behaviour in some cases. To avoid this, we therefore bitcast
signed operands to unsigned prior to performing addition, subtraction,
or multiplication, then bitcast the result back to signed. As signed
types are represented as two's complement, this gives the correct
result whilst avoid any potential undefined behaviour.

Unfortunately HLSL's bitcast functions asint() and asuint() only work
for the 32-bit int and uint types. We therefore only apply this
workaround for 32-bit signed arithmetic. Support for other bit widths
could be added in the future, but extra care must be taken when
converting from unsigned to signed to avoid undefined or
implemented-defined behaviour.
2025-02-14 15:17:40 -08:00
Jamie Nicol 30363ef493 [naga hlsl-out] Wrap int literals in constructor-style cast
There is no literal suffix in HLSL for the integer type. We can,
however, wrap integer literals in a constructor style cast. This
avoids ambiguity when passing literals to overloaded functions, which
we'll make use of in the subsequent patch in this PR.
2025-02-14 15:17:40 -08:00
Jamie Nicol 8807c83363 [naga msl-out] Avoid undefined behaviour for integer division, modulo, negation, and abs
Adds infrastructure to the MSL backend for emitting helper functions,
based upon the existing HLSL backend equivalent. Emit helper functions
for MathFunction::Abs and UnaryOperator::Negate with a signed integer
scalar or vector operand. And for BinaryOperator::Divide and
BinaryOperator::Modulo with signed or unsigned integer scalar or
vector operands.

Abs and Negate are written to avoid signed integer overflow when the
operand equals INT_MIN. This is achieved by bitcasting the value to
unsigned and using the negation operator, then bitcasting the result
back to signed.

Division and Modulo avoid undefined bevaviour for INT_MIN / -1 and
divide-by-zero by using 1 for the divisor instead. Additionally we
avoid undefined behaviour when using the modulo operator when either
or both operands are negative by using the equation from the WGSL
spec, using division, subtraction and multiplication, rather than
MSL's modulus operator.

Lastly, as the usage of the wrapper function for unary integer
negation makes the negation_avoids_prefix_decrement() testcase less
interesting, we extend it to additionally test negating floats.
2025-02-14 15:17:40 -08:00
Jamie Nicol 4bb09e1d3c [naga wgsl-in] Ensure constant evaluation correctly handles Composes of vector ZeroValues (#7138) 2025-02-14 13:18:12 -05:00
Vecvec 5af9e30009 Fix HLSL single scalar loads (#7104) 2025-02-12 18:05:32 +00:00
Dzmitry Malyshau 189c97c88a Implement ray query candidate intersection generation and confirmation (#7047) 2025-02-12 16:23:29 +01:00
Jamie Nicol a546e60d6d [naga wgsl-in] Ensure correct array stride when converting types during const evaluation (#7112)
When converting the underlying scalar type of an array during const
evaluation, we currently use the resulting base type's size as the array
stride for the resulting type. For certain types, this may not match the
required alignment and will therefore result in a validation error.

For example, `array<vec3<f32>, N>` should have a stride of 16. But if
declared with an abstract initializer, eg `array(vec3(0.0))` we will
incorrectly determine the stride to be 12.

To solve this, we use the proc::Layouter struct to determine the
required array stride during const evaluation. To avoid repeating
layouting work, we reuse the Lowerer's layouter, passing it through the
various *Contexts through to the ConstantEvaluator.
2025-02-12 15:46:13 +01:00
Jamie Nicol c07fab2c12 [naga wgsl-in] Allow abstract literals to be used as return values
When lowering a return statement, call expression_for_abstract()
rather than expression() to avoid concretizing the return value. Then,
if the function has a return type, call try_automatic_conversions() to
attempt to convert our return value to the correct type.

This has the unfortunate side effect that some errors that would have
been caught by the validator are instead encountered as conversion
errors by the parser. This may result in a slightly less descriptive
error message in some cases. (See the change to the invalid_functions()
test, for example.)
2025-02-07 11:44:03 -08:00
Vecvec d34707ec3f Rename instance_id & instance_custom_index to instance_index & instance_custom_data (#6780) 2025-02-05 18:23:22 +01:00
Jamie NicolandTeodor Tanasoaia 4e7d892317 [naga msl-out hlsl-out] Improve workaround for infinite loops causing undefined behaviour (#6929)
Co-authored-by: Teodor Tanasoaia <28601907+teoxoy@users.noreply.github.com>
2025-01-31 15:27:52 -05:00
Vecvec 7cde4707ec Fix hlsl storage format generation (#6993) 2025-01-31 16:31:03 +01:00
Jamie Nicol 6c8d0b061c [naga] Restore return statement at end of functions without return types (#7021) 2025-01-29 13:54:49 -05:00
Jamie Nicol b245b358f5 [naga] Don't treat Emit statements as return statements during ensure_block_returns() (#7013) 2025-01-28 10:48:40 -05:00
SamsonandErich Gubler 04aff59a14 [naga] implement pointer_composite_access WGSL language extension (#6913)
Co-authored-by: Erich Gubler <erichdongubler@gmail.com>
2025-01-23 13:29:17 -05:00
Teodor Tanasoaia a8bd1fe1ac [d3d12] bound check dynamic buffers (#6931)
To achieve this for dynamic storage buffers we changed the way we bind them. They now go in root tables and we pass the offset via a root constant.
2025-01-23 17:18:36 +00:00
Jamie Nicol 0282d61f91 [naga hlsl-out] Handle array types for function return values and calls (#6971) 2025-01-23 16:56:18 +01:00
Connor Fitzgerald 436f716715 Unconditionally Generate Bindless Samplers in DX12 (#6766) 2025-01-21 14:24:03 -05:00
b626020769 [naga] parse and validate @must_use attribute (#6801)
* feat: parse and enforce `@must_use` attribute

* feat: refuse `@must_use` of void function

* generated naga/tests/out/ir

* revert extraneous changes

* fix clippy

* update repeated attribute error to trunk format

* struct members can't have `@must_use`

* review fixes

* Remove must_use from FunctionResult and revert the related changes

---------

Co-authored-by: turbocrime <turbocrime@users.noreply.github.com>
Co-authored-by: Jamie Nicol <jnicol@mozilla.com>
2025-01-20 18:09:36 +00:00
Jamie Nicol 1f939e1c44 [naga msl-out] Avoid undefined behaviour due to signed integer overflow (#6959)
Signed integer overflow is undefined behaviour in MSL. However, signed
integers are defined to be two's complement. This allows us to cast
signed values to their corresponding unsigned type, perform the
arithmetic on these unsigned values (which has defined overflow
behaviour) then cast the result back to signed.

Care must be taken when emitting the isign polyfill, which uses
metal::select(). We must ensure the -1, 0, and 1 literals used as
inputs to select() have the correct width, else bitcasting the output
of select() will fail due to mismatched widths.
2025-01-20 18:11:27 +01:00
VecvecandConnor Fitzgerald bdef8c0407 [spirv] Stop naga causing undefined behavior in rayQueryGet*Intersection (#6752)
Co-authored-by: Connor Fitzgerald <connorwadefitzgerald@gmail.com>
2025-01-17 00:04:33 +00:00
atlv be95178709 64 bit image atomics (#5537) 2025-01-15 08:05:13 -05:00
sagudev ea5d3b953f Add test
Signed-off-by: sagudev <16504129+sagudev@users.noreply.github.com>
2025-01-15 11:14:17 +01:00
VecvecandConnor Fitzgerald 21de7f7773 Support DXR in wgpu-hal & naga. (#6777)
Co-authored-by: Connor Fitzgerald <connorwadefitzgerald@gmail.com>
2025-01-14 21:28:37 -05:00
18471d8e78 Image atomics support (#6706)
* Image atomics support

* Address feedback

* fix merge

* Fixes

* Add a couple tests

* Update wgpu-types/src/lib.rs

Co-authored-by: Teodor Tanasoaia <28601907+teoxoy@users.noreply.github.com>

* feedback

* feedback

* glsl

* glsl fix

* fix glsl

* fix fix

* fix fix fic

* fix?

* fix

---------

Co-authored-by: Connor Fitzgerald <connorwadefitzgerald@gmail.com>
Co-authored-by: Teodor Tanasoaia <28601907+teoxoy@users.noreply.github.com>
2025-01-13 15:37:12 +00:00
Asher Jingkong ChenandTeodor Tanasoaia 198762e5cc feat: Add 32-bit floating-point atomics (SHADER_FLOAT32_ATOMIC) (#6234)
* feat: Add 32-bit floating-point atomics

* Current supported platforms: Metal
* Platforms to support in the future: Vulkan

Related issues or PRs:

* gfx-rs/wgpu#1020

* Add changelog

* Edit changelog

* feat: Add 32-bit float atomics support for Vulkan (SPIR-V shaders)

* atomicSub for f32 in the previous commits is removed.

* Update test

* chore: doc type link

* refactor: Revise float atomics on msl and spv

* Make branches tidy
* Also revise old codes
* Ensure the implementations are supported by Metal and Vulkan backends

* refactor: Renaming flt32 atomics to float32 atomics

* chore: Add link to Vulkan feature

* fix: cargo fmt

* chore: hack comment

* Revert changelog

* Fix: Cargo advisory

* Update wgpu-hal/src/metal/adapter.rs

Co-authored-by: Teodor Tanasoaia <28601907+teoxoy@users.noreply.github.com>

* Update naga/src/lib.rs

Co-authored-by: Teodor Tanasoaia <28601907+teoxoy@users.noreply.github.com>

* Adjust feature flag position

---------

Co-authored-by: Teodor Tanasoaia <28601907+teoxoy@users.noreply.github.com>
2025-01-09 16:03:47 +00:00
Connor Fitzgerald a8a91737b2 Switch Binding Arrays on Metal to Argument Buffers (#6751) 2025-01-07 16:00:56 -05:00
Jamie Nicol 15095122a2 [naga] Handle bitshifts for abstract lhs operands during const evaluation (#6865) 2025-01-06 17:33:25 -05:00
Jim Blandy a84b6ca579 [naga] Add tests for automatic conversions in assignments.
Add tests for automatic conversions of the right-hand sides of
assignment expressions.
2025-01-06 17:29:58 -05:00
Jim Blandy 04f0de6d64 [naga] Add some tests to in/abstract-types-var.wgsl.
Fill out a few more cases for variable type inference and automatic
conversions of initializers.
2025-01-06 17:29:58 -05:00
Erich Gubler 9d9b99adb6 fix: make unpack4x{I,U}8 output exprs. composable in {hlsl,msl}-out (#6773) 2024-12-23 09:10:18 -08:00
Schell Carl ScivallyandJim Blandy ea7556853d [naga spv-in] Support atomics in fields of global structs
Handle SPIR-V input that performs atomic operations on structs that
have more than one field. Track which fields of which struct types are
used by atomic operations on which global variables, and then give
those global variables new types in which exactly those fields have
had their `Scalar` leaf types changed to `Atomic`.

Add snapshot tests.

Co-authored-by: Jim Blandy <jimb@red-bean.com>
2024-12-17 07:45:41 -08:00
Kent Slaney 5bec461fb5 Override-expressions for Fixed Size Arrays (#6654) 2024-12-10 15:43:22 +01:00
Schell Carl Scivally 8b93a71129 chore: [spv-in] atomics snapshot tests (#6692)
* chore: [spv-in] clean up atomic upgrade tests

* add output test files

* update changelog

* remove extraneous snapshot out files
2024-12-09 19:54:02 -08:00
Jamie Nicol 90859b4ea1 [naga wgsl-in] Attempt automatic conversion for arguments to user defined function calls (#6577)
When lowering arguments for a user-defined function call, avoid
concretizing the argument types. Instead make use of the existing
`try_automatic_conversions()` machinery to attempt to convert each
argument to the type expected by the function. This is straightforward
as user-defined functions only have a single overload.

This additionally changes an argument type in the test
parse_pointers() from `ptr<private>` to `ptr<function>`. The former is
invalid code which is indeed caught by the validator, but the test
only asserts that parsing succeeds, not validation. With this patch,
this error is now caught during parsing which caused the test to fail.
2024-12-09 16:36:17 +00:00
Kent Slaney b56960b8e8 Add Pipeline Overrides for workgroup_size (#6635) 2024-12-06 11:44:05 +01:00
Dzmitry Malyshau 314e196c90 Implement candidate intersections 2024-11-27 11:49:32 +01:00
Jasper St. Pierre 7fff667c5a [glsl-in] Don't reinterpret function arguments twice in the normal case
We already lowered the function expression if it's a normal in argument; no need to do it again. This eliminates an unused variable and expression. With chained function calls, this could lead to a lot of waste.

There's still an extra unused expression in the case of an out/inout argument; ideally we'd remove these expressions, but it might be tricky.

Fixes #6602
2024-11-26 11:32:15 -08:00
Sludge a9f14c7a90 Allow 64-bit hex literals and const-evaluate unary ops (#6616)
* Allow 64-bit hex literals

* Implement const evaluation for unary 64-bit ops

* Test 64-bit hex literals

* Add changelog entry
2024-11-25 18:54:12 -08:00
Jamie Nicol 00a6032eb7 [naga spv-out] Handle nested arrays when adding matrix decorations
Previously we only checked whether the outermost array's subtype was a
Matrix when determining whether to add ColMajor and MatrixStride
decorations, meaning arrays of arrays of matrices would not be
decorated.
2024-11-20 06:38:39 -08:00
Jim Blandy 0b82776947 [naga msl-out] Avoid UB by making all loops bounded.
In MSL output, avoid undefined behavior due to unbounded loops by
adding an unpredictable, never-actually-taken `break` to the bottom of
each loop body, rather than adding an unpredictable,
never-actually-taken branch over each loop.

This will probably have more of a performance impact, because it
affects each iteration of the loop, but unlike branching over the
loop, which leaves infinite loops (and thus undefined behavior) in the
output, this actually ensures that no loop presented to Metal is
unbounded, so that there is no undefined behavior present that the
optimizer could use to make unwelcome inferences.

Fixes #6528.
2024-11-18 14:10:50 -08:00
Erich Gubler e59f00399e fix(wgsl-in): include user and unknown rules in diagnostic(…) tracking 2024-11-18 17:10:07 -05:00
Jamie Nicol 6f528e2ebc [wgsl-in] Reject invalid texture types for textureSampleBias (#6566)
Non-sampled textures are not allowed, and neither are 1 dimensional
textures.
2024-11-18 17:37:52 +01:00