AMDGPU: Add subtarget feature for global atomic fadd denormal support (llvm#96443)

Not sure what the behavior for gfx90a is. The SPG says it always flushes.
The instruction documentation says it does not.

Author: arsenm

llvm/lib/Target/AMDGPU/AMDGPU.td

@@ -788,6 +788,13 @@ def FeatureFlatAtomicFaddF32Inst
   "Has flat_atomic_add_f32 instruction"
 >;
 
+def FeatureMemoryAtomicFAddF32DenormalSupport
+  : SubtargetFeature<"memory-atomic-fadd-f32-denormal-support",
+  "HasMemoryAtomicFaddF32DenormalSupport",
+  "true",
+  "global/flat/buffer atomic fadd for float supports denormal handling"
+>;
+
 def FeatureAgentScopeFineGrainedRemoteMemoryAtomics
   : SubtargetFeature<"agent-scope-fine-grained-remote-memory-atomics",
   "HasAgentScopeFineGrainedRemoteMemoryAtomics",
@@ -1427,7 +1434,8 @@ def FeatureISAVersion9_4_Common : FeatureSet<
    FeatureKernargPreload,
    FeatureAtomicFMinFMaxF64GlobalInsts,
    FeatureAtomicFMinFMaxF64FlatInsts,
-   FeatureAgentScopeFineGrainedRemoteMemoryAtomics
+   FeatureAgentScopeFineGrainedRemoteMemoryAtomics,
+   FeatureMemoryAtomicFAddF32DenormalSupport
    ]>;
 
 def FeatureISAVersion9_4_0 : FeatureSet<
@@ -1539,7 +1547,8 @@ def FeatureISAVersion11_Common : FeatureSet<
    FeatureFlatAtomicFaddF32Inst,
    FeatureImageInsts,
    FeaturePackedTID,
-   FeatureVcmpxPermlaneHazard]>;
+   FeatureVcmpxPermlaneHazard,
+   FeatureMemoryAtomicFAddF32DenormalSupport]>;
 
 // There are few workarounds that need to be
 // added to all targets. This pessimizes codegen
@@ -1631,7 +1640,9 @@ def FeatureISAVersion12 : FeatureSet<
    FeatureScalarDwordx3Loads,
    FeatureDPPSrc1SGPR,
    FeatureMaxHardClauseLength32,
-   Feature1_5xVGPRs]>;
+   Feature1_5xVGPRs,
+   FeatureMemoryAtomicFAddF32DenormalSupport
+   ]>;
 
 def FeatureISAVersion12_Generic: FeatureSet<
   !listconcat(FeatureISAVersion12.Features,

llvm/lib/Target/AMDGPU/GCNSubtarget.h

@@ -167,6 +167,7 @@ class GCNSubtarget final : public AMDGPUGenSubtargetInfo,
   bool HasAtomicFlatPkAdd16Insts = false;
   bool HasAtomicFaddRtnInsts = false;
   bool HasAtomicFaddNoRtnInsts = false;
+  bool HasMemoryAtomicFaddF32DenormalSupport = false;
   bool HasAtomicBufferGlobalPkAddF16NoRtnInsts = false;
   bool HasAtomicBufferGlobalPkAddF16Insts = false;
   bool HasAtomicCSubNoRtnInsts = false;
@@ -872,6 +873,12 @@ class GCNSubtarget final : public AMDGPUGenSubtargetInfo,
 
   bool hasFlatAtomicFaddF32Inst() const { return HasFlatAtomicFaddF32Inst; }
 
+  /// \return true if the target's flat, global, and buffer atomic fadd for
+  /// float supports denormal handling.
+  bool hasMemoryAtomicFaddF32DenormalSupport() const {
+    return HasMemoryAtomicFaddF32DenormalSupport;
+  }
+
   /// \return true if atomic operations targeting fine-grained memory work
   /// correctly at device scope, in allocations in host or peer PCIe device
   /// memory.