What happened
Following the getting-started guide on a GTX 1070 (Pascal, sm_61) with CUDA 12.9, cargo run builds successfully but fails at runtime with:
No indication of why the PTX is invalid.
Root cause
cuda_builder::CudaBuilder defaults to NvvmArch::Compute75 (crates/nvvm/src/lib.rs):
/// This default value of 7.5 corresponds to Turing and later devices. We default to this
/// because it is the minimum supported by CUDA 13.0 while being in the middle of the range
/// supported by CUDA 12.x.
#[default]
Compute75,
This is a deliberate, documented choice — but it means the getting-started example silently generates PTX targeting sm_75, which the CUDA driver refuses to load on any GPU below Turing (Pascal, Maxwell, Volta — all still fully supported for building/running on CUDA 12.x, just not compiling for them by default). The failure surfaces as a generic cust::error::CudaError::InvalidPtx with nothing pointing at the arch mismatch, which is a confusing first-run experience for anyone following the guide on older hardware.
Suggestions (either would help; happy to send a PR for whichever is preferred)
- Guide fix: add a line to the getting-started example / Windows setup section noting that
.arch(NvvmArch::ComputeXX) should be set explicitly to match the target GPU's actual compute capability, with a pointer to Compute Capability Gating and to nvidia-smi --query-gpu=compute_cap --format=csv for finding it.
- Better diagnostics: have
cust::Module::from_ptx (or the PTX loader) detect a CUDA_ERROR_INVALID_PTX / CUDA_ERROR_NO_BINARY_FOR_GPU outcome and, when possible, compare the PTX's .target smXX line against the current device's actual compute capability, surfacing a clearer error like "PTX built for sm_75 but device is sm_61 — rebuild with .arch(NvvmArch::Compute61)".
Either fix would have saved a fair amount of debugging time here since the error gives no hint that architecture — not toolchain or driver version — is the issue.
Environment
- GPU: GTX 1070 (Pascal,
sm_61)
- Driver: 582.28
- CUDA Toolkit: 12.9.2
- Rust nightly: pinned per
rust-toolchain.toml at the commit in question
- OS: Windows 11
What happened
Following the getting-started guide on a GTX 1070 (Pascal,
sm_61) with CUDA 12.9,cargo runbuilds successfully but fails at runtime with:No indication of why the PTX is invalid.
Root cause
cuda_builder::CudaBuilderdefaults toNvvmArch::Compute75(crates/nvvm/src/lib.rs):This is a deliberate, documented choice — but it means the getting-started example silently generates PTX targeting
sm_75, which the CUDA driver refuses to load on any GPU below Turing (Pascal, Maxwell, Volta — all still fully supported for building/running on CUDA 12.x, just not compiling for them by default). The failure surfaces as a genericcust::error::CudaError::InvalidPtxwith nothing pointing at the arch mismatch, which is a confusing first-run experience for anyone following the guide on older hardware.Suggestions (either would help; happy to send a PR for whichever is preferred)
.arch(NvvmArch::ComputeXX)should be set explicitly to match the target GPU's actual compute capability, with a pointer to Compute Capability Gating and tonvidia-smi --query-gpu=compute_cap --format=csvfor finding it.cust::Module::from_ptx(or the PTX loader) detect aCUDA_ERROR_INVALID_PTX/CUDA_ERROR_NO_BINARY_FOR_GPUoutcome and, when possible, compare the PTX's.target smXXline against the current device's actual compute capability, surfacing a clearer error like "PTX built for sm_75 but device is sm_61 — rebuild with.arch(NvvmArch::Compute61)".Either fix would have saved a fair amount of debugging time here since the error gives no hint that architecture — not toolchain or driver version — is the issue.
Environment
sm_61)rust-toolchain.tomlat the commit in question