-
Notifications
You must be signed in to change notification settings - Fork 197
- #217 · nihalpasham opened
on Jun 14, 2026
Issues
is:issue state:open
is:issue state:open
Issue creation is restricted in this repository
Search results
- Status: Open.#313 In NVlabs/cuda-oxide;
Device-side SetDiscriminant is unsupported for both direct-tag and niche-encoded enums
bugSomething isn't workingSomething isn't workingcodegenDevice code-generation pipeline (Rust MIR to IR to PTX)Device code-generation pipeline (Rust MIR to IR to PTX)enhancementNew feature or requestNew feature or requestIR-loweringLowering between dialects (dialect-mir to LLVM dialect)Lowering between dialects (dialect-mir to LLVM dialect)TBDNot yet triaged; scope or disposition undecidedNot yet triaged; scope or disposition undecidedStatus: Open.#306 In NVlabs/cuda-oxide;Resolve TODOs for MIR to LLVM conversion tests
IR-loweringLowering between dialects (dialect-mir to LLVM dialect)Lowering between dialects (dialect-mir to LLVM dialect)QoL-improvementQuality-of-life improvement; not a correctness fixQuality-of-life improvement; not a correctness fixtestsAdds or improves test coverageAdds or improves test coveragetrackingUmbrella tracking issueUmbrella tracking issueStatus: Open.#303 In NVlabs/cuda-oxide;Select PTX ISA features from the requested target architecture
bugSomething isn't workingSomething isn't workingbuild-relatedBuild scripts, toolchain discovery, or compilation setupBuild scripts, toolchain discovery, or compilation setupcodegenDevice code-generation pipeline (Rust MIR to IR to PTX)Device code-generation pipeline (Rust MIR to IR to PTX)Status: Open.#289 In NVlabs/cuda-oxide;Derive DeviceCopy for enums with valid zero bit patterns
enhancementNew feature or requestNew feature or requesthost-apisHost-side runtime APIs (cuda-host, cuda-core, cuda-async)Host-side runtime APIs (cuda-host, cuda-core, cuda-async)safetyMemory safety, soundness, or undefined behaviorMemory safety, soundness, or undefined behaviorStatus: Open.#285 In NVlabs/cuda-oxide;Tracking: Ampere SM80+ device intrinsics
codegenDevice code-generation pipeline (Rust MIR to IR to PTX)Device code-generation pipeline (Rust MIR to IR to PTX)cuda-featureA CUDA hardware/toolkit feature (TMA, tcgen05, WGMMA)A CUDA hardware/toolkit feature (TMA, tcgen05, WGMMA)enhancementNew feature or requestNew feature or requestepicMulti-deliverable initiativeMulti-deliverable initiativeintrinsicsDevice intrinsics and libdevice math mappingsDevice intrinsics and libdevice math mappingsIR-loweringLowering between dialects (dialect-mir to LLVM dialect)Lowering between dialects (dialect-mir to LLVM dialect)testsAdds or improves test coverageAdds or improves test coveragetrackingUmbrella tracking issueUmbrella tracking issueStatus: Open.#274 In NVlabs/cuda-oxide;Tracking: full device debug metadata for variables and types
codegenDevice code-generation pipeline (Rust MIR to IR to PTX)Device code-generation pipeline (Rust MIR to IR to PTX)debug infoDebug information in generated PTX/IR (line/source maps)Debug information in generated PTX/IR (line/source maps)enhancementNew feature or requestNew feature or requestIR-loweringLowering between dialects (dialect-mir to LLVM dialect)Lowering between dialects (dialect-mir to LLVM dialect)llvm-exportTextual LLVM/NVVM IR exporter (llvm-export crate)Textual LLVM/NVVM IR exporter (llvm-export crate)trackingUmbrella tracking issueUmbrella tracking issueStatus: Open.#226 In NVlabs/cuda-oxide;- Status: Open.#217 In NVlabs/cuda-oxide;
Windows MSVC support fork and signed release artifacts
build-relatedBuild scripts, toolchain discovery, or compilation setupBuild scripts, toolchain discovery, or compilation setupenhancementNew feature or requestNew feature or requeston-holdPaused: valid but deferred until a prerequisite decision or dependency resolvesPaused: valid but deferred until a prerequisite decision or dependency resolvesplatform-supportOS / target platform support (Windows, macOS, non-Linux toolchains)OS / target platform support (Windows, macOS, non-Linux toolchains)TBDNot yet triaged; scope or disposition undecidedNot yet triaged; scope or disposition undecidedStatus: Open.#139 In NVlabs/cuda-oxide;Is
index_1dsafe whereblockDim.y> 1?intrinsicsDevice intrinsics and libdevice math mappingsDevice intrinsics and libdevice math mappingssafetyMemory safety, soundness, or undefined behaviorMemory safety, soundness, or undefined behaviorTBDNot yet triaged; scope or disposition undecidedNot yet triaged; scope or disposition undecidedStatus: Open.#115 In NVlabs/cuda-oxide;Support const generics to kernel entry point
codegenDevice code-generation pipeline (Rust MIR to IR to PTX)Device code-generation pipeline (Rust MIR to IR to PTX)enhancementNew feature or requestNew feature or requestStatus: Open.#110 In NVlabs/cuda-oxide;CUDA Graph API
enhancementNew feature or requestNew feature or requesthost-apisHost-side runtime APIs (cuda-host, cuda-core, cuda-async)Host-side runtime APIs (cuda-host, cuda-core, cuda-async)Status: Open.#107 In NVlabs/cuda-oxide;