slang.git/tests, branch master

Fix infinite loop in SPIRVLegalizationContext::processWorkList (#8712)

2025-10-17T01:22:39+00:00

When slangc is invoked with -g, a source shader that has static infinite loop can generate IR that have branch to a block that contains a branch to the first block that contains the first branch, resulting in infinite loop. Change SPIRVLegalizationContext::processWorkList to only add branch target to work list via its parent, this avoids the infinite loop above. Also change addToWorkList to stop addUsersToWorkList, users should be added explicitly by logic for specific insts. Add regression test as tests/spirv/infinite-loop.slang Fixes #8669

Inline global constants for shader style CPU targets (#8686)

2025-10-16T18:53:10+00:00

On the shader-host-callable target, test `gh-4874.slang` generates IR that contains global constants referencing global params. These need to get inlined into functions, as otherwise `introduceExplicitGlobalContext()` will fail with "no outer func at use site for global", making the test crash the compiler.

Fix wrong diagnostic when checking identical casting expr. (#8727)

2025-10-16T18:52:38+00:00

`SemanticsVisitor::CheckInvokeExprWithCheckedOperands` made several references to `expr` parameter in its `inout` parameter l-value-ness validation logic to access arguments, which is wrong because `expr` is not necessarily the same as `result`/`invoke` (the result of calling `ResolveInvoke()` in the first line of the function. Changing it to `invoke` for consistency. Also add a special case logic to return early in case the resolved invoke expr is `argument[0]` when the original invoke expr is `T(funcThatReturnsT())`. Closes #8659.

Fix use of variadic generics with [Differentiable]. (#8736)

2025-10-16T18:23:13+00:00

There was a bug that causes the compiler failing to treat a `no_diff TypePack` as a type pack, and thus diagnose an error when resolving the following call. The fix is to unwrap any ModifiedType wrappers in `IsTypePack()` check.

Fix WGSL bitshift test typo (#8720)

2025-10-16T10:42:57+00:00

Random drive-by test fix, this was reading past the end of the buffer but usually succeeded because the expected result is 0.

Immutable access qualifier for pointers and use `__ldg` on cuda. (#8710)

2025-10-16T03:59:47+00:00

This PR implements `Access.Immutable` to allow pointers to immutable data. The new type `ImmutablePtr` is defined as an alias of `Ptr`. By forming a immutable pointer, the programmer is conveying to the compiler that the data at the pointer address will never change during the execution of the current program. Therefore loads from immutable pointers can be deduplicated by the compiler, and will translate to `__ldg` when generating code for CUDA. The SPIRV backend is not changed in this PR, since the current SPIRV spec makes it very difficult to specify loads from immutable address without generating tons of wrappers and boilerplate type declarations. We would like to see the spec evolved a bit to around its support of `NonWritable` physical storage pointers or immutable loads before we attempt to express such immutability in SPIRV. For now we simply emit ordinary pointers and loads when generating spirv. --------- Co-authored-by: slangbot <186143334+slangbot@users.noreply.github.com>

Fix segfault on arrays of structs containing parameter blocks (#8555)

2025-10-13T14:14:45+00:00

Closes https://github.com/shader-slang/slang/issues/8154 However there is further design work to do on implementing the "NonAddressableType" suggestion

Add diagnostic for cyclic #include. (#8679)

2025-10-11T00:31:01+00:00

Allow entry points with missing numthreads on CPU targets (#8678)

2025-10-10T23:01:51+00:00

Several tests have compute entry points without a `[numthreads(x,y,z)]` decoration. Currently, none of these tests run on the CPU target, as they crash the compiler. I took a look at the SPIR-V emitter, which falls back to a workgroup size of (1,1,1): https://github.com/shader-slang/slang/blob/1e0908bd7107dfbdac912b693c3ab9bd6e1dc8b3/source/slang/slang-ir-spirv-legalize.cpp#L1635-L1643 To match this behaviour, this PR implements a fallback solution that makes `emitCalcGroupExtents()` emit (1,1,1). This PR is both a question and a suggestion; I'm not sure the approach here is at all reasonable. Personally, I'd just like to explicitly add `[numthreads(1,1,1)]` to all such tests, but I don't know if it's actually legal and supported to not have a `numthreads`. So the implementation here is a bit conservative. I ran across these when I went through tests for the upcoming LLVM target. These were the final blockers to get all autodiff and language-features tests passing (not counting the ones using things like wave intrinsics and barriers etc.)

Fix `specializeRTTIObject` to use non-zero RTTI value to work with `Optional`. (#8677)

2025-10-10T21:34:56+00:00

Closes #8673. The issue is that we use the RTTI field of an existential to check if it is null. We have the logic to help the user to fill in a non-zero value for the RTTI field when such an object is filled from the host. However, when there is slang code creating an existential value, we still have old logic in the compiler that just fills in 0 for the RTTI field, causing an `Optional.hasValue` to always return false in such cases.