[clang] [llvm] [Clang][SYCL] Compile device code once per IGCA target (PR #239010)
Yury Plyakhin via cfe-commits
cfe-commits at lists.llvm.org
Thu Nov 12 18:10:33 PST 2026
================
@@ -70,6 +70,10 @@ class LLVM_LIBRARY_VISIBILITY SYCLToolChain : public ToolChain {
SYCLInstallationDetector SYCLInstallation;
};
+/// \return the name SYCL device code for \p Arch is compiled for: the IGCA
+/// target of an Intel GPU, e.g. "igca_30r" for "xe-mtl-u", or \p Arch itself.
----------------
YuriPlyakhin wrote:
```suggestion
/// \return the architecture to compile SYCL device code for, given \p Arch
/// from --offload-arch: the IGCA target if \p Arch is an Intel GPU, e.g.
/// "igca_30r" for "xe-mtl-u", and \p Arch itself otherwise.
```
https://github.com/llvm/llvm-project/pull/239010
More information about the cfe-commits
mailing list