aaron.ballman added a comment. Mostly grammar related review comments, nothing substantive.
================ Comment at: clang/docs/SYCLSupport.md:816 +### Address spaces handling + ---------------- ================ Comment at: clang/docs/SYCLSupport.md:818 + +SYCL specification uses C++ classes to represent pointers to disjoint memory +regions on an accelerator to enable compilation with standard C++ toolchain and ---------------- ================ Comment at: clang/docs/SYCLSupport.md:819 +SYCL specification uses C++ classes to represent pointers to disjoint memory +regions on an accelerator to enable compilation with standard C++ toolchain and +SYCL compiler toolchain. Section 3.8.2 of SYCL 2020 specification defines ---------------- ================ Comment at: clang/docs/SYCLSupport.md:825 + +The main address space semantic difference of SYCL mode from OpenCL is that +SYCL doesn't perform address space qualifier inference detailed in ---------------- ================ Comment at: clang/docs/SYCLSupport.md:826 +The main address space semantic difference of SYCL mode from OpenCL is that +SYCL doesn't perform address space qualifier inference detailed in +[OpenCL C v3.0 s6.7.8](https://www.khronos.org/registry/OpenCL/specs/3.0-unified/html/OpenCL_C.html#addr-spaces-inference). ---------------- ================ Comment at: clang/docs/SYCLSupport.md:827 +SYCL doesn't perform address space qualifier inference detailed in +[OpenCL C v3.0 s6.7.8](https://www.khronos.org/registry/OpenCL/specs/3.0-unified/html/OpenCL_C.html#addr-spaces-inference). + ---------------- ================ Comment at: clang/docs/SYCLSupport.md:831 +OpenMP/CUDA/HIP, SYCL uses clang's "default" address space for types with no +address space attributes. This design has two important features: keeps the type system consistent with C++ on one hand and enable tools for emitting device code aligned with SPIR memory model (and other GPU targets). + ---------------- You should also re-wrap for 80 col (whenever we finish talking about what words to add). ================ Comment at: clang/docs/SYCLSupport.md:839 + +SYCL device compiler turns into + ---------------- ================ Comment at: clang/docs/SYCLSupport.md:845 + +OpenCL compiler turn into + ---------------- ================ Comment at: clang/docs/SYCLSupport.md:851 + +Changing variable type has massive and destructive effect in C++. For instance +this does not compile in C++ for OpenCL mode: ---------------- ================ Comment at: clang/docs/SYCLSupport.md:899-900 + +Depending on the compiler mode `multi_ptr` will either decorate internal data +with address space attribute or not. + ---------------- ================ Comment at: clang/docs/SYCLSupport.md:902 + +To utilize existing clang's functionality, we re-use following OpenCL address +space attributes for decoration pointers implementation: ---------------- ================ Comment at: clang/docs/SYCLSupport.md:903 +To utilize existing clang's functionality, we re-use following OpenCL address +space attributes for decoration pointers implementation: + ---------------- ================ Comment at: clang/docs/SYCLSupport.md:914 + +Default address space represents "Generic-memory", which is a virtual address +space which overlaps the global, local and private address spaces. SYCL mode ---------------- ================ Comment at: clang/docs/SYCLSupport.md:915 +Default address space represents "Generic-memory", which is a virtual address +space which overlaps the global, local and private address spaces. SYCL mode +enables conversion to/from default address space from/to address space ---------------- ================ Comment at: clang/docs/SYCLSupport.md:916 +space which overlaps the global, local and private address spaces. SYCL mode +enables conversion to/from default address space from/to address space +attributed type. ---------------- ================ Comment at: clang/docs/SYCLSupport.md:919 + +SPIR target allocates SYCL namespace scope variables in global address space. + ---------------- Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D99488/new/ https://reviews.llvm.org/D99488 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
