[clang] [llvm] [AMDGPU] Add OpenCL-specific fence address space masks (PR #78572)

[Matt Arsenault via llvm-commits]

================
@@ -69,6 +69,7 @@ BUILTIN(__builtin_amdgcn_iglp_opt, "vIi", "n")
 BUILTIN(__builtin_amdgcn_s_dcache_inv, "v", "n")
 BUILTIN(__builtin_amdgcn_buffer_wbinvl1, "v", "n")
 BUILTIN(__builtin_amdgcn_fence, "vUicC*", "n")
+BUILTIN(__builtin_amdgcn_masked_fence, "vUiUicC*", "n")
----------------
arsenm wrote:

Should put some documentation in LanguageExtensions for the arguments 

https://github.com/llvm/llvm-project/pull/78572