gpuweb / WSL

Web Shading Language

https://gpuweb.github.io/WSL/

gpuweb/WSL Issues

Consider allowing multiple setters for a particular field
Closed 4 years ago
Buffer packing rules are unspecified
Closed 4 years ago
Shader stage variable matching is unspecified
Closed 4 years ago
We are too restrictive about the possible behaviors for out-of-bounds accesses
Closed 4 years ago
The reference implementation allow arbitrary types as indices
Closed 4 years ago
can't find semantics of shift operator <<
Closed 4 years ago
Consider making anders more flexible
Closed 4 years ago
Support constant global variables
Closed 4 years ago
The reference implementation allows operator&[] to take an array reference as first argument
Closed 4 years ago
Decide on the resolution of overloading in presence of literals
Closed 4 years ago
Do-while validation rules are a little overzealous
Closed 4 years ago6
Operator and greedy matching
Closed 4 years ago
WHLSL lacks swizzling
Closed 4 years ago1
if WSL is a compilation target, it needs a way to carry source-level info
Closed 4 years ago
Consider not supporting boolean matrixes
Closed 4 years ago1
order of application of dynamic reduction rules
Closed 4 years ago
suggest supporting hex float literals: to aid testing corner cases
Closed 4 years ago
typo: funciton
Closed 4 years ago
WSL refers to "contiguous" sets of bytes, but that's an unnecessary overspecification
Closed 4 years ago
WSL needs a plan for extensibility without collisions with user identifiers
Closed 4 years ago
please add a table of contents to the generated spec
Closed 4 years ago
reducing to an abstract left-value, e.g. fail case and other questions
Closed 4 years ago
order of evaluation of function call arguments
Closed 4 years ago
Are #INF and -1.#INF intentionally excluded? (HLSL compatibility)
Closed 4 years ago
Decide on the type of the second argument of ldexp
Closed 4 years ago1
Decide on the possible results of atan
Closed 4 years ago
generated spec lacks version info
Closed 4 years ago
it appears that "A-5" fails to parse as an expression
Closed 4 years ago
scientific notation is not permitted in float literals
Closed 4 years ago
Cleanup the rules on Textures
Closed 4 years ago
Decide on the precision of floating-point operations, and how they can be optimized
Closed 4 years ago1
Should we allow user-defined operator++/operator-- or not?
Closed 4 years ago
Decide whether we want to support acosh/asinh/atanh
Closed 4 years ago
Some matrix constructors are missing
Closed 4 years ago
remove spec text about having to check enums when loaded
Closed 4 years ago
Fix spec about samplers
Closed 4 years ago
Should we support Texture2DMS<T>, TextureDepth2DMS<float> ?
Closed 4 years ago
device writes in vertex shaders
Closed 4 years ago
OpEntryPoint with duplicate names
Closed 4 years ago1
The reference implementation treats post-{increment/decrement} as pre-{incr/decr} when there is a field setter
Closed 5 years ago1
Decide on the shadowing of enums by local variables
Closed 5 years ago1
Keywords are not reserved in the reference implementation
Closed 5 years ago1
Replace ReturnChecker and UnreachableCodeChecker with the Behavior machinery from the spec
Closed 5 years ago1
"void" is probably allowed as a variable type or as a parameter type
Closed 5 years ago3
firstbithigh, firstbitlow, clz, ctz..
Closed 5 years ago3
The reference implementation allows getters/setters on the same fields that already have address takers
Closed 5 years ago3
Fix the combination of setters and post-{increment/decrement}
Closed 5 years ago1
Should the result of assignment, +=, -=, ..., ++, and -- be a left-value?
Closed 5 years ago1
The spec is wrong about continue in for loops
Closed 5 years ago
Remove the vector type syntax vector<blah, N>
Closed 5 years ago2