[llvm] [AMDGPU] Document amdgpu-as in AMDGPUUsage (PR #94335)

[Sameer Sahasrabuddhe via llvm-commits]

================
@@ -5969,6 +5969,31 @@ following sections:
 * :ref:`amdgpu-amdhsa-memory-model-gfx942`
 * :ref:`amdgpu-amdhsa-memory-model-gfx10-gfx11`
 
+.. _amdgpu-fence-as:
+
+Fence and Address Spaces
+++++++++++++++++++++++++++++++
+
+LLVM fences do not have address space information, thus, fence
+codegen usually needs to be conservative and fence all address spaces.
+
+In the case of OpenCL, where synchronization can only happen in the
+same address space, this can result in extra unnecessary waits.
+For instance, a fence that is supposed to only target local memory will
----------------
ssahasra wrote:

Use verb "synchronize" instead of "target".

https://github.com/llvm/llvm-project/pull/94335