Skip to content

feat(cuda): accept a pre-linked CUBIN in load_module_bytes #7

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
haixuanTao wants to merge 1 commit into
base: main
Choose a base branch
from
Open

feat(cuda): accept a pre-linked CUBIN in load_module_bytes #7

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
13 changes: 10 additions & 3 deletions crates/khal/src/backend/cuda.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -271,9 +271,16 @@ impl Backend for Cuda {
}
}

// Expect PTX text bytes.
let ptx_str = std::str::from_utf8(bytes).map_err(|_| CudaBackendError::InvalidPtx)?;
let ptx = cudarc::nvrtc::Ptx::from_src(ptx_str.to_string());
// Accept either PTX text or a pre-linked CUBIN (detected by ELF magic).
// A cubin is required when the module references symbols the driver JIT
// cannot resolve on its own (e.g. libdevice `__nv_*` math), which a
// toolchain links into a self-contained binary ahead of time.
let ptx = if bytes.starts_with(&[0x7f, b'E', b'L', b'F']) {
cudarc::nvrtc::Ptx::from_binary(bytes.to_vec())
} else {
let ptx_str = std::str::from_utf8(bytes).map_err(|_| CudaBackendError::InvalidPtx)?;
cudarc::nvrtc::Ptx::from_src(ptx_str.to_string())
};
let module = self.ctx.load_module(ptx)?;

// Cache the loaded module.
Expand Down
Loading