[SYCL][AMDGPU] Add flag to enable cross-address space syncscopes for atomics on AMD

[GeorgeWeb]

TODO: Add description.

[github-actions[bot]]

:warning: C/C++ code formatter, clang-format found issues in your code. :warning:

You can test this locally with the following command:

git-clang-format --diff 3288a66d48d5aee7412ad12118794f28e6634550 92462f0eb06ff354c3d97e6cdee50e8932a1f846 -- clang/lib/CodeGen/TargetInfo.h clang/lib/CodeGen/Targets/AMDGPU.cpp clang/lib/Driver/ToolChains/Clang.cpp clang/lib/Frontend/CompilerInvocation.cpp

View the diff from clang-format here.

diff --git a/clang/lib/Frontend/CompilerInvocation.cpp b/clang/lib/Frontend/CompilerInvocation.cpp
index 878be1e598..2843f38244 100644
--- a/clang/lib/Frontend/CompilerInvocation.cpp
+++ b/clang/lib/Frontend/CompilerInvocation.cpp
@@ -2180,7 +2180,7 @@ bool CompilerInvocation::ParseCodeGenArgs(CodeGenOptions &Opts, ArgList &Args,
   if (Arg *A = Args.getLastArg(
           options::OPT_mamdgpu_cross_addr_space_atomic_memory_ordering)) {
     if (Opts.EnableCrossAddressSpaceAtomicMemoryOrdering &&
-      LangOptsRef.SYCLIsDevice && !T.isAMDGPU()) {
+        LangOptsRef.SYCLIsDevice && !T.isAMDGPU()) {
       unsigned int DiagID = diag::warn_drv_unsupported_opt_for_target;
       Diags.Report(DiagID) << A->getSpelling() << T.str();
     }