[llvm] eb954b0 - [AMDGPU] Drop docs for invalid load-release and store-acquire operations (#202338)

[via llvm-commits]

Author: ritter-x2a
Date: 2026-06-09T06:59:00Z
New Revision: eb954b0bb91f3a9374a00279650e1a0977871062

URL: https://github.com/llvm/llvm-project/commit/eb954b0bb91f3a9374a00279650e1a0977871062
DIFF: https://github.com/llvm/llvm-project/commit/eb954b0bb91f3a9374a00279650e1a0977871062.diff

LOG: [AMDGPU] Drop docs for invalid load-release and store-acquire operations (#202338)

The LangRef says "release and acq_rel orderings are not valid on load
instructions" [for loads](https://llvm.org/docs/LangRef.html#load-instruction)
and "acquire and acq_rel orderings aren't valid on store instructions"
[for stores](https://llvm.org/docs/LangRef.html#store-instruction).
Providing them in textual IR is diagnosed with an error.

Therefore, we should not define semantics for these invalid constructs.

Part of LCOMPILER-2273.

Added: 
    

Modified: 
    llvm/docs/AMDGPUUsage.rst

Removed: 
    


################################################################################
diff  --git a/llvm/docs/AMDGPUUsage.rst b/llvm/docs/AMDGPUUsage.rst
index 121b66a657f5d..ed4c759fd24f8 100644
--- a/llvm/docs/AMDGPUUsage.rst
+++ b/llvm/docs/AMDGPUUsage.rst
@@ -7214,15 +7214,6 @@ orderings (``acquire``, ``release``, ``acq_rel``, or ``seq_cst``).
 The memory model does not support the region address space which is treated as
 non-atomic.
 
-Acquire memory ordering is not meaningful on store atomic instructions and is
-treated as non-atomic.
-
-Release memory ordering is not meaningful on load atomic instructions and is
-treated as non-atomic.
-
-Acquire-release memory ordering is not meaningful on load or store atomic
-instructions and is treated as acquire and release respectively.
-
 The memory order also adds the single thread optimization constraints defined in
 table
 :ref:`amdgpu-amdhsa-memory-model-single-thread-optimization-constraints-table`.