| Commit message (Collapse) | Author | Age |
| ... | |
| |
|
|
|
|
|
|
|
|
|
|
|
|
| |
(#7737)
* Initial plan
* Add validation and clear error message for float generic parameters
Co-authored-by: csyonghe <2652293+csyonghe@users.noreply.github.com>
---------
Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>
Co-authored-by: csyonghe <2652293+csyonghe@users.noreply.github.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Emit special diagnostic for invalid pointer taking operations
* Update source/slang/slang-diagnostic-defs.h
Co-authored-by: ArielG-NV <159081215+ArielG-NV@users.noreply.github.com>
* Add OperatorAddressOf KnownBuiltin modifier
* update error message for non-l-value assignment
* update the diagnostics in the tests
* Use enum based KnownBuiltinDeclName
* format code (#7772)
Co-authored-by: slangbot <186143334+slangbot@users.noreply.github.com>
---------
Co-authored-by: ArielG-NV <159081215+ArielG-NV@users.noreply.github.com>
Co-authored-by: slangbot <ellieh+slangbot@nvidia.com>
Co-authored-by: slangbot <186143334+slangbot@users.noreply.github.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
| |
When you track IR creation with `_debugUID` or `_slangIRAllocBreak`,
you had to set a break point near `_slangIRAllocBreak` is used
and unset it after changing the value of `_slangIRAllocBreak`.
Then, when you re-run slangc for more debuggin, you have to set the
break-point again, because it was unset.
With this change, you can keep the break-point in the new line, and
it will hit only once per debugging session.
Not a big improvement, but I found this useful because it requires less
manual handling of the break-points.
|
| |
|
|
| |
* Add Claude local setting files to .gitignore
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Fix `slang-generate` segfault when parsing `(...)`
Currently the following code causes a failure:
* `myFunc($((int)val))`
* This fails since we parse it as `int)val` due to trying to find a body with repeating `(`.
The primary issue here is that auto-format prefers `$((...))` over `$( (...))`, making this an annoying bug.
* only read first char with `(`
* format code
---------
Co-authored-by: slangbot <186143334+slangbot@users.noreply.github.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Initial plan
* Implement enum-based KnownBuiltin system to replace string comparisons
Co-authored-by: csyonghe <2652293+csyonghe@users.noreply.github.com>
* Add test for enum-based KnownBuiltin system and verify functionality
Co-authored-by: csyonghe <2652293+csyonghe@users.noreply.github.com>
* Implement enum-based KnownBuiltin system with direct integer values
Co-authored-by: csyonghe <2652293+csyonghe@users.noreply.github.com>
* Fix IntVal access and update tests for new enum-based KnownBuiltin system
Co-authored-by: csyonghe <2652293+csyonghe@users.noreply.github.com>
* Replace hardcoded KnownBuiltin integers with preprocessor enum syntax
- Updated all KnownBuiltin attributes to use $( (int)KnownBuiltinDeclName::EnumValue) syntax
- Added space between parentheses to avoid preprocessor bug: $( (int) instead of $((int)
- Updated both core.meta.slang and hlsl.meta.slang files
- Eliminates preprocessor-time integer conversion, baking enum values directly into meta files
- Maintains same functionality while using type-safe enum references
Co-authored-by: csyonghe <2652293+csyonghe@users.noreply.github.com>
* Fix IDifferentiablePtr KnownBuiltin mapping regression
Co-authored-by: csyonghe <2652293+csyonghe@users.noreply.github.com>
* Remove unused IDifferentiablePtrType enum case from KnownBuiltinDeclName
Co-authored-by: csyonghe <2652293+csyonghe@users.noreply.github.com>
* Clean up temporary AST dump files from testing
Co-authored-by: csyonghe <2652293+csyonghe@users.noreply.github.com>
* Replace hardcoded integer with descriptive constant in KnownBuiltin test
Replace the hardcoded [KnownBuiltin(0)] with a descriptive named constant
GEOMETRY_STREAM_APPEND_BUILTIN to improve code readability and maintainability.
The test now clearly indicates which builtin enum value is being tested.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-authored-by: Gangzheng Tong <gtong-nv@users.noreply.github.com>
---------
Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>
Co-authored-by: csyonghe <2652293+csyonghe@users.noreply.github.com>
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: Gangzheng Tong <gtong-nv@users.noreply.github.com>
Co-authored-by: Gangzheng Tong <tonggangzheng@gmail.com>
|
| | |
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
| |
* Fix language server crash.
* Fix tests.
* Fix.
* Revert changes.
|
| |
|
| |
Co-authored-by: Ellie Hermaszewska <ellieh@nvidia.com>
|
| |
|
|
|
|
|
|
|
| |
Fixes #7733
Copy gl_NumWorkGroups into hlsl.meta.slang as WorkgroupCount function so
that it can be used for GLSL and SPIR-V targets without GLSL syntax.
Also change WorkgroupSize function to allow use with mesh shading capability.
Update pipeline/rasterization/mesh/task-simple.slang to test it in task and mesh stages.
|
| |
|
|
|
|
|
|
|
| |
* Update claude instructions to help with debugging.
* Increase limit.
* Simplify prompt.
* increase limit.
|
| |
|
|
|
| |
* Fixup address spaces after inlining.
* add -O0
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
* enable coop vec tests
* disable tests which fail with reason
* disable other failing tests
* disable failing test
* add to expected fail list
* sort list
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Initial plan
* Fix unnecessary Int64 SPIRV capability usage in pointer marshalling
Replace uint64-based pointer marshalling with uint2-based approach to avoid requiring Int64 capability in SPIRV output. This affects both basic type marshalling and resource handle marshalling for pointer types.
Co-authored-by: csyonghe <2652293+csyonghe@users.noreply.github.com>
* Replace test cases with user-provided test case
Co-authored-by: csyonghe <2652293+csyonghe@users.noreply.github.com>
* Fix test case to avoid unrelated pointer casting operations that require Int64
Co-authored-by: csyonghe <2652293+csyonghe@users.noreply.github.com>
---------
Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>
Co-authored-by: csyonghe <2652293+csyonghe@users.noreply.github.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Fix segfault with Ptr<T> extension using 'This' type reference
Set LookupOptions::NoDeref when looking up target type members within
extension declarations to prevent automatic pointer dereferencing that
was causing breadcrumb handling issues and segfaults.
Fixes #7656
Co-authored-by: Yong He <csyonghe@users.noreply.github.com>
* Enhance test for ptr-extension-this-type to verify correctness with interpreter
- Changed function to return sizeof(This) - sizeof(Ptr<int>) + 1
- Modified test to use interpreter with filecheck to verify result equals 1
- Added main() function that calls the extension method and prints result
- Verifies that 'This' correctly refers to Ptr<int> in extension context
Co-authored-by: Yong He <csyonghe@users.noreply.github.com>
---------
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: Yong He <csyonghe@users.noreply.github.com>
|
| |
|
|
|
|
|
|
|
|
|
|
| |
* Use stdout for --help text instead of stderr
* format code (#13)
Co-authored-by: slangbot <186143334+slangbot@users.noreply.github.com>
---------
Co-authored-by: slangbot <ellieh+slangbot@nvidia.com>
Co-authored-by: slangbot <186143334+slangbot@users.noreply.github.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Add DebugGlobalVariable instructions to SPIR-V output
Implements generation of DebugGlobalVariable instructions for global variables
like Texture2D and SamplerState in SPIR-V debug information output. Adds debug
type support for texture and sampler types using DebugTypeComposite.
* Handle two more types for DebugGlobalVariable
kIROp_RaytracingAccelerationStructureType and
kIROp_SamplerComparisonStateType had to be handled in
`emitDebugTypeImpl()`
* Fix format
* Refactor debug type emission to reduce duplication
Use IRSamplerStateTypeBase type check and fallback pattern instead of
separate cases for each opcode type.
* Fix compiler warning
* Simplify `emitDebugTypeImpl()` more
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Check vectorValue before accessing
* Fix metal segfault by using IRElementExtract for general vector handling
Address review comments by replacing IRMakeVector-specific code with
IRElementExtract to handle any vector instruction type (IRIntCast, etc).
This makes the code more robust and fixes cases where float2(1,2)
creates IRIntCast instead of IRMakeVector.
Co-authored-by: Yong He <csyonghe@users.noreply.github.com>
* Fix sign comparison warning in metal legalize
Cast originalElementCount->getValue() to UInt to avoid comparison between signed and unsigned integers.
Co-authored-by: Yong He <csyonghe@users.noreply.github.com>
---------
Co-authored-by: Yong He <yonghe@outlook.com>
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: Yong He <csyonghe@users.noreply.github.com>
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Fix IEEE 754 NaN comparisons in constant folding
Added proper NaN handling in SCCP optimization pass to follow IEEE 754 standard:
- NaN \!= any value returns true
- All other NaN comparisons return false
- Added double precision NaN detection support
- Fixed type detection to check operands instead of result type
* Avoid differentiating NaN and non-NaN cases
* format code (#76)
---------
Co-authored-by: slangbot <ellieh+slangbot@nvidia.com>
|
| |
|
|
|
|
|
| |
This PR modified CLAUDE.md to avoid common mistakes that may waste the tokens.
Currently what we have is a default CLAUDE.md that gets generated automatically with "/init" and some of the information are outdated.
Removing the claude file edit hook, because ./extras/formatting.sh doesn't take an individual file name as its input.
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
* Ensure generic constraints are checked before inner extension.
* Add warning for non-standard generic extension.
* Fix tests.
* Fix test.
* Ban interface types from equality constraints.
* Fix.
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Update falcor perf test CI
* Add slang-gfx build back since Falcor is still using it
* format code (#7704)
Co-authored-by: slangbot <186143334+slangbot@users.noreply.github.com>
---------
Co-authored-by: slangbot <ellieh+slangbot@nvidia.com>
Co-authored-by: slangbot <186143334+slangbot@users.noreply.github.com>
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Initial plan
* Implement int16_t/uint16_t support for WGSL target
Co-authored-by: csyonghe <2652293+csyonghe@users.noreply.github.com>
* Change int16_t/uint16_t to emit proper diagnostics instead of auto-promoting
Co-authored-by: csyonghe <2652293+csyonghe@users.noreply.github.com>
* Implement diagnoseOnce to prevent duplicate diagnostics
Co-authored-by: csyonghe <2652293+csyonghe@users.noreply.github.com>
* Optimize diagnoseOnce to use HashSet::add() return value
Co-authored-by: csyonghe <2652293+csyonghe@users.noreply.github.com>
* Apply code formatting to slang-emit-c-like.h
Co-authored-by: csyonghe <2652293+csyonghe@users.noreply.github.com>
* Fix diagnoseOnce to use all parameters instead of first parameter only
Co-authored-by: csyonghe <2652293+csyonghe@users.noreply.github.com>
---------
Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>
Co-authored-by: csyonghe <2652293+csyonghe@users.noreply.github.com>
|
| |
|
|
|
|
|
|
|
| |
* Only append --help value categories if explictly specified
* Add consistent help-category list, omit help-category category
* Fix formatting
* Add missing indent
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Add Claude Code Assistant for using LLMs
add claude.md files for context to the LLM
* format code
* Update claude.yml
add pull request target
* Force test claude.yml
* Remove sensitive flag - Update claude.yml
* Update claude.yml
Reverted the direct_prompt
* format code (#7674)
Co-authored-by: slangbot <186143334+slangbot@users.noreply.github.com>
* Fix Claude workflow based on PR review feedback
- Remove unused pull_request_review_comment trigger
- Increase timeout from 30 to 60 minutes
- Change fetch-depth from 0 to 2 for minimal git history
- Add recursive submodule checkout following copilot pattern
- Add environment setup with libx11-dev dependency
- Switch to release build config following copilot pattern
- Increase max_turns from 10 to 20 for longer conversations
- Update custom instructions with build commands and OS info
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* Update workflow - configure debug build
* Update max turns to 50 for M size work
* Add Claude Code settings with auto-formatting hook
- Create .claude/settings.json with PostToolUse hook
- Automatically runs ./extras/formatting.sh after Write/Edit/MultiEdit operations
- Ensures code formatting consistency when Claude makes changes
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
---------
Co-authored-by: slangbot <186143334+slangbot@users.noreply.github.com>
Co-authored-by: slangbot <ellieh+slangbot@nvidia.com>
Co-authored-by: Claude <noreply@anthropic.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
emitter (#7691)
* Initial plan
* Fix SV_VertexID and SV_InstanceID crash with -preserve-params in SPIR-V emitter
Co-authored-by: csyonghe <2652293+csyonghe@users.noreply.github.com>
* Fix transformation layer to handle unused builtin params properly
This ensures that builtin decorations are correctly transformed for unused parameters with -preserve-params, preventing the SPIR-V emitter from seeing un-translated builtins like sv_vertexid.
Co-authored-by: csyonghe <2652293+csyonghe@users.noreply.github.com>
* Format code according to project style
Co-authored-by: csyonghe <2652293+csyonghe@users.noreply.github.com>
---------
Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>
Co-authored-by: csyonghe <2652293+csyonghe@users.noreply.github.com>
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Initial plan
* Add diagnostic for unreferenced generic parameters in extensions
Co-authored-by: csyonghe <2652293+csyonghe@users.noreply.github.com>
* Use GenericTypeParamDeclBase and eliminate intermediate list
Co-authored-by: csyonghe <2652293+csyonghe@users.noreply.github.com>
* Update test to use filecheck format for better verification
Co-authored-by: csyonghe <2652293+csyonghe@users.noreply.github.com>
* Update extension validation to use getMembersOfType for better constraint collection
Co-authored-by: csyonghe <2652293+csyonghe@users.noreply.github.com>
* Add temporary workaround for inheritance constraints but issue persists
Co-authored-by: csyonghe <2652293+csyonghe@users.noreply.github.com>
* Fix inheritance constraint validation by moving to SemanticsDeclConformancesVisitor
Co-authored-by: csyonghe <2652293+csyonghe@users.noreply.github.com>
* Fix inheritance constraint validation by removing inheritance declaration checks
Co-authored-by: csyonghe <2652293+csyonghe@users.noreply.github.com>
* Move extension generic parameter validation back to SemanticsDeclBasesVisitor
Co-authored-by: csyonghe <2652293+csyonghe@users.noreply.github.com>
* Fix test.
* format code (#7671)
Co-authored-by: slangbot <186143334+slangbot@users.noreply.github.com>
* Fix and update.
* Refine fix.
---------
Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>
Co-authored-by: csyonghe <2652293+csyonghe@users.noreply.github.com>
Co-authored-by: Yong He <yonghe@outlook.com>
Co-authored-by: slangbot <ellieh+slangbot@nvidia.com>
Co-authored-by: slangbot <186143334+slangbot@users.noreply.github.com>
|
| |
|
|
|
|
|
|
|
|
|
| |
close #6286.
This PR is to improve the diagnostics for no_diff usage.
In a differentiable function, any calls to a non-diff function with constant arguments should not require no_diff attribute.
This PR adds this extra check at `checkAutoDiffUsages` where it checks the differentiability on IR.
In a differentiable method, we will force to use `[NoDiffThis]` attribute if there is access to non-differentiable `This` type. Once this access is detected we will report a warning to bring users attention that this access won't generate any derivative, they have to use `[NoDiffThis]` to suppress that warning.
This PR adds this check at type checking stage, because it's the easiest way to find out all the `This` accesses.
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Port first gfx unit test to slang-rhi
* port triangle example to use slang-rhi
* port platform-test to slang-rhi
* Update platform-test to throttle mouse move events
* port gpu-printing example to use slang-rhi
* port model-viewer example to use slang-rhi
* port ray-tracing example to use slang-rhi
* port ray-tracing pipeline example to use slang-rhi
* port reflection parameter blocks example to use slang-rhi
* port shader-object example to use slang-rhi
* port shader-toy example to use slang-rhi
* Port most of tests to slang-rhi
* port link-time-constant-array-size to use slang-rhi
* Fix tests and find matching tests in slang-rhi
* port autodiff-texture
* remove gfx target; port nv-aftermath-example
* update include path for shader-cursor.h
* Disabled 2 more ported tests
* fix build error
* remove gfx test
* put slang-rhi (static-lib) before slang (shared)
* format code (#7621)
Co-authored-by: slangbot <186143334+slangbot@users.noreply.github.com>
* add debug callback
* format code (#7649)
Co-authored-by: slangbot <186143334+slangbot@users.noreply.github.com>
* Address review comments; revert back to use SLANG_CHECK_MSG
---------
Co-authored-by: slangbot <ellieh+slangbot@nvidia.com>
Co-authored-by: slangbot <186143334+slangbot@users.noreply.github.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* stable names
* tests, options and ci for stable names
* Add back compat design document
* fix warnings
* formatting
* comment
* neaten
* regenerate command line reference
* consolidate ci scripts
* faster ci
* remove libreadline
* Move new function to end of interface
---------
Co-authored-by: slangbot <186143334+slangbot@users.noreply.github.com>
|
| |
|
|
|
|
|
| |
* Generate "OpExecutionMode PointMode" for tessellation shaders instead of the incorrect geometry and mesh shader specific "OpExecutionMode OutputPoints".
* Add a test case verifying the OpExecutionMode is correct.
Fixes #7660
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
constraints (#7578)
* fix problem
* cleanup comment
* format code
* make change more restrictive
* format code
* push logic update
* format code
* push test fix
* make test more general
---------
Co-authored-by: slangbot <186143334+slangbot@users.noreply.github.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Add error for forward references in generic constraints
Change addresses issue #6545.
The slang compiler's type checker is unable to support cases where
a generic type parameter is referenced as a constraint before it is
declared. For example code like:
````
interface IFoo<T : IFloat>
{
}
void bar<Foo : IFoo<T>, T : IFloat>()
{
}
````
Is not supported, but will currently report a generic error like,
"(0): error 99999: Slang compilation aborted due to an exception
of class Slang::InternalError: unexpected: generic type constraint
during lowering"
This change adds a new check for this kind of code and reports an
error like,
"error 30117: generic constraint for parameter 'Foo' references type
parameter 'T' before it is declared"
when detected. Basic testing of this error is also added in a new
diagnostic test.
* Add error for forward refs in generic constiants update 1
* Add error for forward refs in generic constraints update 2
Revised algo in checkForwardReferencesInGenericConstraint to run
in O(n) instead of the previous O(n^2) using HashSets.
* Add error for forward refs in generic constraints update 3
-Update logic adding referenced decl's in collectReferencedDecls.
-Simplified error string logic.
* Add error for forward refs in generic constraints update 4
-Declare collectReferencedDecls in slang-check.h such that it can
be used as a general utility function.
* format code
---------
Co-authored-by: slangbot <186143334+slangbot@users.noreply.github.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Add category list at end of slangc -h output
* Use value category for categories
* format code (#12)
Co-authored-by: slangbot <186143334+slangbot@users.noreply.github.com>
* Regenerate cmdline reference
---------
Co-authored-by: slangbot <ellieh+slangbot@nvidia.com>
Co-authored-by: slangbot <186143334+slangbot@users.noreply.github.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The bug here is subtle enough that it was missed during development of the example, and it it unlikely to be hit by typical applications.
All that said, it is good to ensure that this example does things correctly, since we are pointing developers to it as an example of how to correctly extract information using the reflection API.
In order to get the bug to surface, one can target Vulkan/SPIR-V using input code like:
```
struct Small
{
float4 v;
Texture2D t;
}
struct Big
{
float4 u;
ConstantBuffer<Small> smallCB;
ParameterBlock<Small> smallPB;
}
ConstantBuffer<Big> bigCB;
```
Without this change, the example would display an incorrect cumulative offset for `bigCB.smallPB.t`.
The incorrectly-computed offset showed that the `binding`s consumed by `bigCB.smallCB` (for both its constant buffer and its `t` field) were being included in the offset computed for the `t` within the parameter block (when the outer parameter block should have caused those offsets to be ignored).
The underlying bug in the example is that when recursively printing the cumulative offsets for the "container" and "content" part of a buffer/block, the example was including the buffer/block in the `AccessPath` when printing cumulative offsets for the "content" part, but not for the "container" part.
The fix is just to shift the order of things and use the same `AccessPath` when printing both the "container" and "content" parts of a buffer/block.
It is difficult to be 100% certain that this change is correct, and that there are no corner cases where the new code gives unexpected/wrong results.
I have manually tested the example code by throwing a variety of shader parameter declarations at it, and the new logic seems robust.
We are probably at the point where the code of this example probably needs to be turned into a stand-alone tool that we can use as part of our automated testing.
That will have to wait for a future change; the goal of this change is to just get the simple bug fixe checked in.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Improve slang-test output verbosity control
This commit improves the existing command-line argument for slang-test,
"-v". Previously it printed more information when "-v" was used.
This commit adds a new option to silence the information output so that
LLM processes less tokens when things are working as expected.
* format code (#74)
---------
Co-authored-by: slangbot <ellieh+slangbot@nvidia.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* bottleneck ir module reading and writing
* compute/simple working
* more complex tests working
* neaten
* factor out SourceLoc serialization
* document changes
* Appease clang
* Correct name serialization
* remove unnecessary code
* neaten
* neaten
|
| |
|
|
|
| |
* Language server: sort completion candidate by relevance.
* Small adjustment.
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Initial plan
* Implement copysign forward and backward derivatives
Co-authored-by: csyonghe <2652293+csyonghe@users.noreply.github.com>
* Fix copysign test format to use expected.txt file
Co-authored-by: csyonghe <2652293+csyonghe@users.noreply.github.com>
* Add wgsl support to copysign and fix y==0 derivative case
Co-authored-by: csyonghe <2652293+csyonghe@users.noreply.github.com>
* Add wgsl support to copysign helper functions
Co-authored-by: csyonghe <2652293+csyonghe@users.noreply.github.com>
* Fix copysign derivative to return 0 when either x or y is 0
Co-authored-by: csyonghe <2652293+csyonghe@users.noreply.github.com>
---------
Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>
Co-authored-by: csyonghe <2652293+csyonghe@users.noreply.github.com>
|