[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 18:10:31 PST 2026


================
@@ -1,21 +1,56 @@
 /// Tests for Intel GPU names in --offload-arch.
 
-/// An Intel GPU name names a device job.
+/// An Intel GPU is compiled for its IGCA target, e.g. xe-bmg-g21 for igca_40r.
----------------
YuriPlyakhin wrote:

maybe:
```suggestion
/// IGCA target is used to compile for real device specified, e.g. igca_40r for xe-bmg-g21.
```

https://github.com/llvm/llvm-project/pull/239010


More information about the cfe-commits mailing list