Translate atomicrmw fsub into FNegate and AtomicFAddEXT (#1780)

Signed-off-by: Sarnie, Nick <nick.sarnie@intel.com>

Original commit:
KhronosGroup/SPIRV-LLVM-Translator@f95a6d7

Author: sarnex

llvm-spirv/lib/SPIRV/SPIRVWriter.cpp

@@ -2236,16 +2236,15 @@ LLVMToSPIRVBase::transValueWithoutDecoration(Value *V, SPIRVBasicBlock *BB,
     AtomicRMWInst::BinOp Op = ARMW->getOperation();
     bool SupportedAtomicInst =
         AtomicRMWInst::isFPOperation(Op)
-            ? (Op == AtomicRMWInst::FAdd || Op == AtomicRMWInst::FMin ||
-               Op == AtomicRMWInst::FMax)
+            ? (Op == AtomicRMWInst::FAdd || Op == AtomicRMWInst::FSub ||
+               Op == AtomicRMWInst::FMin || Op == AtomicRMWInst::FMax)
             : Op != AtomicRMWInst::Nand;
     if (!BM->getErrorLog().checkError(
             SupportedAtomicInst, SPIRVEC_InvalidInstruction, V,
             "Atomic " + AtomicRMWInst::getOperationName(Op).str() +
                 " is not supported in SPIR-V!\n"))
       return nullptr;
 
-    spv::Op OC = LLVMSPIRVAtomicRmwOpCodeMap::map(Op);
     AtomicOrderingCABI Ordering = llvm::toCABI(ARMW->getOrdering());
     auto MemSem = OCLMemOrderMap::map(static_cast<OCLMemOrderKind>(Ordering));
     std::vector<Value *> Operands(4);
@@ -2259,9 +2258,18 @@ LLVMToSPIRVBase::transValueWithoutDecoration(Value *V, SPIRVBasicBlock *BB,
     Operands[1] = getUInt32(M, spv::ScopeDevice);
     Operands[2] = getUInt32(M, MemSem);
     Operands[3] = ARMW->getValOperand();
-    std::vector<SPIRVId> Ops = BM->getIds(transValue(Operands, BB));
+    std::vector<SPIRVValue *> OpVals = transValue(Operands, BB);
+    std::vector<SPIRVId> Ops = BM->getIds(OpVals);
     SPIRVType *Ty = transType(ARMW->getType());
 
+    spv::Op OC;
+    if (Op == AtomicRMWInst::FSub) {
+      // Implement FSub through FNegate and AtomicFAddExt
+      Ops[3] = BM->addUnaryInst(OpFNegate, Ty, OpVals[3], BB)->getId();
+      OC = OpAtomicFAddEXT;
+    } else
+      OC = LLVMSPIRVAtomicRmwOpCodeMap::map(Op);
+
     return mapValue(V, BM->addInstTemplate(OC, Ops, BB, Ty));
   }
 

llvm-spirv/test/extensions/EXT/SPV_EXT_shader_atomic_float_/atomicrmw_fsub_double.ll

@@ -0,0 +1,38 @@
+; RUN: llvm-as -opaque-pointers=0 < %s -o %t.bc
+; RUN: llvm-spirv -opaque-pointers=0 --spirv-ext=+SPV_EXT_shader_atomic_float_add %t.bc -o %t.spv
+; RUN: spirv-val %t.spv
+; RUN: llvm-spirv -to-text %t.spv -o - | FileCheck --check-prefix=CHECK-SPIRV %s
+
+; RUN: llvm-spirv -r %t.spv -o %t.rev.bc
+; RUN: llvm-dis -opaque-pointers=0 %t.rev.bc -o - | FileCheck %s --check-prefixes=CHECK-LLVM
+
+; CHECK-SPIRV-DAG: Extension "SPV_EXT_shader_atomic_float_add"
+; CHECK-SPIRV-DAG: Capability AtomicFloat64AddEXT
+; CHECK-SPIRV: TypeInt [[Int:[0-9]+]] 32 0
+; CHECK-SPIRV-DAG: Constant [[Int]] [[Scope_Device:[0-9]+]] 1 {{$}}
+; CHECK-SPIRV-DAG: Constant [[Int]] [[MemSem_SequentiallyConsistent:[0-9]+]] 16
+; CHECK-SPIRV: TypeFloat [[Double:[0-9]+]] 64
+; CHECK-SPIRV: Variable {{[0-9]+}} [[DoublePointer:[0-9]+]]
+; CHECK-SPIRV: Constant [[Double]] [[DoubleValue:[0-9]+]] 0 1078263808
+
+target datalayout = "e-i64:64-v16:16-v24:32-v32:32-v48:64-v96:128-v192:256-v256:256-v512:512-v1024:1024"
+target triple = "spir64"
+
+@f = common dso_local local_unnamed_addr addrspace(1) global double 0.000000e+00, align 8
+
+; Function Attrs: nounwind
+define dso_local spir_func void @test_atomicrmw_fadd() local_unnamed_addr #0 {
+entry:
+ %0 = atomicrmw fsub double addrspace(1)* @f, double 42.000000e+00 seq_cst
+; CHECK-SPIRV: FNegate [[Double]] [[NegateValue:[0-9]+]] [[DoubleValue]]
+; CHECK-SPIRV: AtomicFAddEXT [[Double]] {{[0-9]+}} [[DoublePointer]] [[Scope_Device]] [[MemSem_SequentiallyConsistent]] [[NegateValue]]
+; CHECK-LLVM: [[FNegateLLVM:%[0-9]+]] = fneg double 4.200000e+01
+; CHECK-LLVM: call spir_func double {{.*}}atomic_add{{.*}}(double addrspace(1)* @f, double [[FNegateLLVM]])
+  ret void
+}
+
+attributes #0 = { nounwind "correctly-rounded-divide-sqrt-fp-math"="false" "disable-tail-calls"="false" "frame-pointer"="all" "less-precise-fpmad"="false" "min-legal-vector-width"="0" "no-infs-fp-math"="false" "no-jump-tables"="false" "no-nans-fp-math"="false" "no-signed-zeros-fp-math"="false" "no-trapping-math"="false" "stack-protector-buffer-size"="8" "unsafe-fp-math"="false" "use-soft-float"="false" }
+
+!llvm.module.flags = !{!0}
+
+!0 = !{i32 1, !"wchar_size", i32 4}

llvm-spirv/test/extensions/EXT/SPV_EXT_shader_atomic_float_/atomicrmw_fsub_float.ll

@@ -0,0 +1,38 @@
+; RUN: llvm-as -opaque-pointers=0 < %s -o %t.bc
+; RUN: llvm-spirv -opaque-pointers=0 --spirv-ext=+SPV_EXT_shader_atomic_float_add %t.bc -o %t.spv
+; RUN: spirv-val %t.spv
+; RUN: llvm-spirv -to-text %t.spv -o - | FileCheck --check-prefix=CHECK-SPIRV %s
+
+; RUN: llvm-spirv -r %t.spv -o %t.rev.bc
+; RUN: llvm-dis -opaque-pointers=0 %t.rev.bc -o - | FileCheck %s --check-prefixes=CHECK-LLVM
+
+; CHECK-SPIRV-DAG: Extension "SPV_EXT_shader_atomic_float_add"
+; CHECK-SPIRV-DAG: Capability AtomicFloat32AddEXT
+; CHECK-SPIRV: TypeInt [[Int:[0-9]+]] 32 0
+; CHECK-SPIRV-DAG: Constant [[Int]] [[Scope_Device:[0-9]+]] 1 {{$}}
+; CHECK-SPIRV-DAG: Constant [[Int]] [[MemSem_SequentiallyConsistent:[0-9]+]] 16
+; CHECK-SPIRV: TypeFloat [[Float:[0-9]+]] 32
+; CHECK-SPIRV: Variable {{[0-9]+}} [[FPPointer:[0-9]+]]
+; CHECK-SPIRV: Constant [[Float]] [[FPValue:[0-9]+]] 1109917696
+
+target datalayout = "e-i64:64-v16:16-v24:32-v32:32-v48:64-v96:128-v192:256-v256:256-v512:512-v1024:1024"
+target triple = "spir64"
+
+@f = common dso_local local_unnamed_addr addrspace(1) global float 0.000000e+00, align 4
+
+; Function Attrs: nounwind
+define dso_local spir_func void @test_atomicrmw_fadd() local_unnamed_addr #0 {
+entry:
+ %0 = atomicrmw fsub float addrspace(1)* @f, float 42.000000e+00 seq_cst
+; CHECK-SPIRV: FNegate [[Float]] [[NegateValue:[0-9]+]] [[FPValue]]
+; CHECK-SPIRV: AtomicFAddEXT [[Float]] {{[0-9]+}} [[FPPointer]] [[Scope_Device]] [[MemSem_SequentiallyConsistent]] [[NegateValue]]
+; CHECK-LLVM: [[FNegateLLVM:%[0-9]+]] = fneg float 4.200000e+01
+; CHECK-LLVM: call spir_func float {{.*}}atomic_add{{.*}}(float addrspace(1)* @f, float [[FNegateLLVM]])
+  ret void
+}
+
+attributes #0 = { nounwind "correctly-rounded-divide-sqrt-fp-math"="false" "disable-tail-calls"="false" "frame-pointer"="all" "less-precise-fpmad"="false" "min-legal-vector-width"="0" "no-infs-fp-math"="false" "no-jump-tables"="false" "no-nans-fp-math"="false" "no-signed-zeros-fp-math"="false" "no-trapping-math"="false" "stack-protector-buffer-size"="8" "unsafe-fp-math"="false" "use-soft-float"="false" }
+
+!llvm.module.flags = !{!0}
+
+!0 = !{i32 1, !"wchar_size", i32 4}