Skip to content

[Clang][CodeGen] Fix type for atomic float incdec operators #107075

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Sep 4, 2024
Merged

[Clang][CodeGen] Fix type for atomic float incdec operators #107075

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
25 changes: 14 additions & 11 deletions clang/lib/CodeGen/CGExprScalar.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2865,19 +2865,22 @@ ScalarExprEmitter::EmitScalarPrePostIncDec(const UnaryOperator *E, LValue LV,
llvm::AtomicOrdering::SequentiallyConsistent);
return isPre ? Builder.CreateBinOp(op, old, amt) : old;
}
// Special case for atomic increment/decrement on floats
// Special case for atomic increment/decrement on floats.
// Bail out non-power-of-2-sized floating point types (e.g., x86_fp80).
if (type->isFloatingType()) {
llvm::AtomicRMWInst::BinOp aop =
isInc ? llvm::AtomicRMWInst::FAdd : llvm::AtomicRMWInst::FSub;
llvm::Instruction::BinaryOps op =
isInc ? llvm::Instruction::FAdd : llvm::Instruction::FSub;
llvm::Value *amt = llvm::ConstantFP::get(
VMContext, llvm::APFloat(static_cast<float>(1.0)));
llvm::AtomicRMWInst *old =
CGF.emitAtomicRMWInst(aop, LV.getAddress(), amt,
llvm::AtomicOrdering::SequentiallyConsistent);
llvm::Type *Ty = ConvertType(type);
if (llvm::has_single_bit(Ty->getScalarSizeInBits())) {
llvm::AtomicRMWInst::BinOp aop =
isInc ? llvm::AtomicRMWInst::FAdd : llvm::AtomicRMWInst::FSub;
llvm::Instruction::BinaryOps op =
isInc ? llvm::Instruction::FAdd : llvm::Instruction::FSub;
llvm::Value *amt = llvm::ConstantFP::get(Ty, 1.0);
llvm::AtomicRMWInst *old =
CGF.emitAtomicRMWInst(aop, LV.getAddress(), amt,
llvm::AtomicOrdering::SequentiallyConsistent);

return isPre ? Builder.CreateBinOp(op, old, amt) : old;
return isPre ? Builder.CreateBinOp(op, old, amt) : old;
}
}
value = EmitLoadOfLValue(LV, E->getExprLoc());
input = value;
Expand Down
112 changes: 40 additions & 72 deletions clang/test/CodeGen/AMDGPU/amdgpu-atomic-float.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -99,20 +99,16 @@ float test_float_pre_inc()
// SAFE-NEXT: [[ENTRY:.*:]]
// SAFE-NEXT: [[RETVAL:%.*]] = alloca double, align 8, addrspace(5)
// SAFE-NEXT: [[RETVAL_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[RETVAL]] to ptr
// SAFE-NEXT: [[TMP0:%.*]] = atomicrmw fadd ptr addrspacecast (ptr addrspace(1) @test_double_post_inc.n to ptr), float 1.000000e+00 seq_cst, align 8
// SAFE-NEXT: store float [[TMP0]], ptr [[RETVAL_ASCAST]], align 8
// SAFE-NEXT: [[TMP1:%.*]] = load double, ptr [[RETVAL_ASCAST]], align 8
// SAFE-NEXT: ret double [[TMP1]]
// SAFE-NEXT: [[TMP0:%.*]] = atomicrmw fadd ptr addrspacecast (ptr addrspace(1) @test_double_post_inc.n to ptr), double 1.000000e+00 seq_cst, align 8
// SAFE-NEXT: ret double [[TMP0]]
//
// UNSAFE-LABEL: define dso_local double @test_double_post_inc(
// UNSAFE-SAME: ) #[[ATTR0]] {
// UNSAFE-NEXT: [[ENTRY:.*:]]
// UNSAFE-NEXT: [[RETVAL:%.*]] = alloca double, align 8, addrspace(5)
// UNSAFE-NEXT: [[RETVAL_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[RETVAL]] to ptr
// UNSAFE-NEXT: [[TMP0:%.*]] = atomicrmw fadd ptr addrspacecast (ptr addrspace(1) @test_double_post_inc.n to ptr), float 1.000000e+00 seq_cst, align 8, !amdgpu.no.fine.grained.memory [[META3]], !amdgpu.ignore.denormal.mode [[META3]]
// UNSAFE-NEXT: store float [[TMP0]], ptr [[RETVAL_ASCAST]], align 8
// UNSAFE-NEXT: [[TMP1:%.*]] = load double, ptr [[RETVAL_ASCAST]], align 8
// UNSAFE-NEXT: ret double [[TMP1]]
// UNSAFE-NEXT: [[TMP0:%.*]] = atomicrmw fadd ptr addrspacecast (ptr addrspace(1) @test_double_post_inc.n to ptr), double 1.000000e+00 seq_cst, align 8, !amdgpu.no.fine.grained.memory [[META3]]
// UNSAFE-NEXT: ret double [[TMP0]]
//
double test_double_post_inc()
{
Expand All @@ -125,20 +121,16 @@ double test_double_post_inc()
// SAFE-NEXT: [[ENTRY:.*:]]
// SAFE-NEXT: [[RETVAL:%.*]] = alloca double, align 8, addrspace(5)
// SAFE-NEXT: [[RETVAL_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[RETVAL]] to ptr
// SAFE-NEXT: [[TMP0:%.*]] = atomicrmw fsub ptr addrspacecast (ptr addrspace(1) @test_double_post_dc.n to ptr), float 1.000000e+00 seq_cst, align 8
// SAFE-NEXT: store float [[TMP0]], ptr [[RETVAL_ASCAST]], align 8
// SAFE-NEXT: [[TMP1:%.*]] = load double, ptr [[RETVAL_ASCAST]], align 8
// SAFE-NEXT: ret double [[TMP1]]
// SAFE-NEXT: [[TMP0:%.*]] = atomicrmw fsub ptr addrspacecast (ptr addrspace(1) @test_double_post_dc.n to ptr), double 1.000000e+00 seq_cst, align 8
// SAFE-NEXT: ret double [[TMP0]]
//
// UNSAFE-LABEL: define dso_local double @test_double_post_dc(
// UNSAFE-SAME: ) #[[ATTR0]] {
// UNSAFE-NEXT: [[ENTRY:.*:]]
// UNSAFE-NEXT: [[RETVAL:%.*]] = alloca double, align 8, addrspace(5)
// UNSAFE-NEXT: [[RETVAL_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[RETVAL]] to ptr
// UNSAFE-NEXT: [[TMP0:%.*]] = atomicrmw fsub ptr addrspacecast (ptr addrspace(1) @test_double_post_dc.n to ptr), float 1.000000e+00 seq_cst, align 8, !amdgpu.no.fine.grained.memory [[META3]]
// UNSAFE-NEXT: store float [[TMP0]], ptr [[RETVAL_ASCAST]], align 8
// UNSAFE-NEXT: [[TMP1:%.*]] = load double, ptr [[RETVAL_ASCAST]], align 8
// UNSAFE-NEXT: ret double [[TMP1]]
// UNSAFE-NEXT: [[TMP0:%.*]] = atomicrmw fsub ptr addrspacecast (ptr addrspace(1) @test_double_post_dc.n to ptr), double 1.000000e+00 seq_cst, align 8, !amdgpu.no.fine.grained.memory [[META3]]
// UNSAFE-NEXT: ret double [[TMP0]]
//
double test_double_post_dc()
{
Expand All @@ -151,22 +143,18 @@ double test_double_post_dc()
// SAFE-NEXT: [[ENTRY:.*:]]
// SAFE-NEXT: [[RETVAL:%.*]] = alloca double, align 8, addrspace(5)
// SAFE-NEXT: [[RETVAL_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[RETVAL]] to ptr
// SAFE-NEXT: [[TMP0:%.*]] = atomicrmw fsub ptr addrspacecast (ptr addrspace(1) @test_double_pre_dc.n to ptr), float 1.000000e+00 seq_cst, align 8
// SAFE-NEXT: [[TMP1:%.*]] = fsub float [[TMP0]], 1.000000e+00
// SAFE-NEXT: store float [[TMP1]], ptr [[RETVAL_ASCAST]], align 8
// SAFE-NEXT: [[TMP2:%.*]] = load double, ptr [[RETVAL_ASCAST]], align 8
// SAFE-NEXT: ret double [[TMP2]]
// SAFE-NEXT: [[TMP0:%.*]] = atomicrmw fsub ptr addrspacecast (ptr addrspace(1) @test_double_pre_dc.n to ptr), double 1.000000e+00 seq_cst, align 8
// SAFE-NEXT: [[TMP1:%.*]] = fsub double [[TMP0]], 1.000000e+00
// SAFE-NEXT: ret double [[TMP1]]
//
// UNSAFE-LABEL: define dso_local double @test_double_pre_dc(
// UNSAFE-SAME: ) #[[ATTR0]] {
// UNSAFE-NEXT: [[ENTRY:.*:]]
// UNSAFE-NEXT: [[RETVAL:%.*]] = alloca double, align 8, addrspace(5)
// UNSAFE-NEXT: [[RETVAL_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[RETVAL]] to ptr
// UNSAFE-NEXT: [[TMP0:%.*]] = atomicrmw fsub ptr addrspacecast (ptr addrspace(1) @test_double_pre_dc.n to ptr), float 1.000000e+00 seq_cst, align 8, !amdgpu.no.fine.grained.memory [[META3]]
// UNSAFE-NEXT: [[TMP1:%.*]] = fsub float [[TMP0]], 1.000000e+00
// UNSAFE-NEXT: store float [[TMP1]], ptr [[RETVAL_ASCAST]], align 8
// UNSAFE-NEXT: [[TMP2:%.*]] = load double, ptr [[RETVAL_ASCAST]], align 8
// UNSAFE-NEXT: ret double [[TMP2]]
// UNSAFE-NEXT: [[TMP0:%.*]] = atomicrmw fsub ptr addrspacecast (ptr addrspace(1) @test_double_pre_dc.n to ptr), double 1.000000e+00 seq_cst, align 8, !amdgpu.no.fine.grained.memory [[META3]]
// UNSAFE-NEXT: [[TMP1:%.*]] = fsub double [[TMP0]], 1.000000e+00
// UNSAFE-NEXT: ret double [[TMP1]]
//
double test_double_pre_dc()
{
Expand All @@ -179,22 +167,18 @@ double test_double_pre_dc()
// SAFE-NEXT: [[ENTRY:.*:]]
// SAFE-NEXT: [[RETVAL:%.*]] = alloca double, align 8, addrspace(5)
// SAFE-NEXT: [[RETVAL_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[RETVAL]] to ptr
// SAFE-NEXT: [[TMP0:%.*]] = atomicrmw fadd ptr addrspacecast (ptr addrspace(1) @test_double_pre_inc.n to ptr), float 1.000000e+00 seq_cst, align 8
// SAFE-NEXT: [[TMP1:%.*]] = fadd float [[TMP0]], 1.000000e+00
// SAFE-NEXT: store float [[TMP1]], ptr [[RETVAL_ASCAST]], align 8
// SAFE-NEXT: [[TMP2:%.*]] = load double, ptr [[RETVAL_ASCAST]], align 8
// SAFE-NEXT: ret double [[TMP2]]
// SAFE-NEXT: [[TMP0:%.*]] = atomicrmw fadd ptr addrspacecast (ptr addrspace(1) @test_double_pre_inc.n to ptr), double 1.000000e+00 seq_cst, align 8
// SAFE-NEXT: [[TMP1:%.*]] = fadd double [[TMP0]], 1.000000e+00
// SAFE-NEXT: ret double [[TMP1]]
//
// UNSAFE-LABEL: define dso_local double @test_double_pre_inc(
// UNSAFE-SAME: ) #[[ATTR0]] {
// UNSAFE-NEXT: [[ENTRY:.*:]]
// UNSAFE-NEXT: [[RETVAL:%.*]] = alloca double, align 8, addrspace(5)
// UNSAFE-NEXT: [[RETVAL_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[RETVAL]] to ptr
// UNSAFE-NEXT: [[TMP0:%.*]] = atomicrmw fadd ptr addrspacecast (ptr addrspace(1) @test_double_pre_inc.n to ptr), float 1.000000e+00 seq_cst, align 8, !amdgpu.no.fine.grained.memory [[META3]], !amdgpu.ignore.denormal.mode [[META3]]
// UNSAFE-NEXT: [[TMP1:%.*]] = fadd float [[TMP0]], 1.000000e+00
// UNSAFE-NEXT: store float [[TMP1]], ptr [[RETVAL_ASCAST]], align 8
// UNSAFE-NEXT: [[TMP2:%.*]] = load double, ptr [[RETVAL_ASCAST]], align 8
// UNSAFE-NEXT: ret double [[TMP2]]
// UNSAFE-NEXT: [[TMP0:%.*]] = atomicrmw fadd ptr addrspacecast (ptr addrspace(1) @test_double_pre_inc.n to ptr), double 1.000000e+00 seq_cst, align 8, !amdgpu.no.fine.grained.memory [[META3]]
// UNSAFE-NEXT: [[TMP1:%.*]] = fadd double [[TMP0]], 1.000000e+00
// UNSAFE-NEXT: ret double [[TMP1]]
//
double test_double_pre_inc()
{
Expand All @@ -207,20 +191,16 @@ double test_double_pre_inc()
// SAFE-NEXT: [[ENTRY:.*:]]
// SAFE-NEXT: [[RETVAL:%.*]] = alloca half, align 2, addrspace(5)
// SAFE-NEXT: [[RETVAL_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[RETVAL]] to ptr
// SAFE-NEXT: [[TMP0:%.*]] = atomicrmw fadd ptr addrspacecast (ptr addrspace(1) @test__Float16_post_inc.n to ptr), float 1.000000e+00 seq_cst, align 2
// SAFE-NEXT: store float [[TMP0]], ptr [[RETVAL_ASCAST]], align 2
// SAFE-NEXT: [[TMP1:%.*]] = load half, ptr [[RETVAL_ASCAST]], align 2
// SAFE-NEXT: ret half [[TMP1]]
// SAFE-NEXT: [[TMP0:%.*]] = atomicrmw fadd ptr addrspacecast (ptr addrspace(1) @test__Float16_post_inc.n to ptr), half 0xH3C00 seq_cst, align 2
// SAFE-NEXT: ret half [[TMP0]]
//
// UNSAFE-LABEL: define dso_local half @test__Float16_post_inc(
// UNSAFE-SAME: ) #[[ATTR0]] {
// UNSAFE-NEXT: [[ENTRY:.*:]]
// UNSAFE-NEXT: [[RETVAL:%.*]] = alloca half, align 2, addrspace(5)
// UNSAFE-NEXT: [[RETVAL_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[RETVAL]] to ptr
// UNSAFE-NEXT: [[TMP0:%.*]] = atomicrmw fadd ptr addrspacecast (ptr addrspace(1) @test__Float16_post_inc.n to ptr), float 1.000000e+00 seq_cst, align 2, !amdgpu.no.fine.grained.memory [[META3]], !amdgpu.ignore.denormal.mode [[META3]]
// UNSAFE-NEXT: store float [[TMP0]], ptr [[RETVAL_ASCAST]], align 2
// UNSAFE-NEXT: [[TMP1:%.*]] = load half, ptr [[RETVAL_ASCAST]], align 2
// UNSAFE-NEXT: ret half [[TMP1]]
// UNSAFE-NEXT: [[TMP0:%.*]] = atomicrmw fadd ptr addrspacecast (ptr addrspace(1) @test__Float16_post_inc.n to ptr), half 0xH3C00 seq_cst, align 2, !amdgpu.no.fine.grained.memory [[META3]]
// UNSAFE-NEXT: ret half [[TMP0]]
//
_Float16 test__Float16_post_inc()
{
Expand All @@ -233,20 +213,16 @@ _Float16 test__Float16_post_inc()
// SAFE-NEXT: [[ENTRY:.*:]]
// SAFE-NEXT: [[RETVAL:%.*]] = alloca half, align 2, addrspace(5)
// SAFE-NEXT: [[RETVAL_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[RETVAL]] to ptr
// SAFE-NEXT: [[TMP0:%.*]] = atomicrmw fsub ptr addrspacecast (ptr addrspace(1) @test__Float16_post_dc.n to ptr), float 1.000000e+00 seq_cst, align 2
// SAFE-NEXT: store float [[TMP0]], ptr [[RETVAL_ASCAST]], align 2
// SAFE-NEXT: [[TMP1:%.*]] = load half, ptr [[RETVAL_ASCAST]], align 2
// SAFE-NEXT: ret half [[TMP1]]
// SAFE-NEXT: [[TMP0:%.*]] = atomicrmw fsub ptr addrspacecast (ptr addrspace(1) @test__Float16_post_dc.n to ptr), half 0xH3C00 seq_cst, align 2
// SAFE-NEXT: ret half [[TMP0]]
//
// UNSAFE-LABEL: define dso_local half @test__Float16_post_dc(
// UNSAFE-SAME: ) #[[ATTR0]] {
// UNSAFE-NEXT: [[ENTRY:.*:]]
// UNSAFE-NEXT: [[RETVAL:%.*]] = alloca half, align 2, addrspace(5)
// UNSAFE-NEXT: [[RETVAL_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[RETVAL]] to ptr
// UNSAFE-NEXT: [[TMP0:%.*]] = atomicrmw fsub ptr addrspacecast (ptr addrspace(1) @test__Float16_post_dc.n to ptr), float 1.000000e+00 seq_cst, align 2, !amdgpu.no.fine.grained.memory [[META3]]
// UNSAFE-NEXT: store float [[TMP0]], ptr [[RETVAL_ASCAST]], align 2
// UNSAFE-NEXT: [[TMP1:%.*]] = load half, ptr [[RETVAL_ASCAST]], align 2
// UNSAFE-NEXT: ret half [[TMP1]]
// UNSAFE-NEXT: [[TMP0:%.*]] = atomicrmw fsub ptr addrspacecast (ptr addrspace(1) @test__Float16_post_dc.n to ptr), half 0xH3C00 seq_cst, align 2, !amdgpu.no.fine.grained.memory [[META3]]
// UNSAFE-NEXT: ret half [[TMP0]]
//
_Float16 test__Float16_post_dc()
{
Expand All @@ -259,22 +235,18 @@ _Float16 test__Float16_post_dc()
// SAFE-NEXT: [[ENTRY:.*:]]
// SAFE-NEXT: [[RETVAL:%.*]] = alloca half, align 2, addrspace(5)
// SAFE-NEXT: [[RETVAL_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[RETVAL]] to ptr
// SAFE-NEXT: [[TMP0:%.*]] = atomicrmw fsub ptr addrspacecast (ptr addrspace(1) @test__Float16_pre_dc.n to ptr), float 1.000000e+00 seq_cst, align 2
// SAFE-NEXT: [[TMP1:%.*]] = fsub float [[TMP0]], 1.000000e+00
// SAFE-NEXT: store float [[TMP1]], ptr [[RETVAL_ASCAST]], align 2
// SAFE-NEXT: [[TMP2:%.*]] = load half, ptr [[RETVAL_ASCAST]], align 2
// SAFE-NEXT: ret half [[TMP2]]
// SAFE-NEXT: [[TMP0:%.*]] = atomicrmw fsub ptr addrspacecast (ptr addrspace(1) @test__Float16_pre_dc.n to ptr), half 0xH3C00 seq_cst, align 2
// SAFE-NEXT: [[TMP1:%.*]] = fsub half [[TMP0]], 0xH3C00
// SAFE-NEXT: ret half [[TMP1]]
//
// UNSAFE-LABEL: define dso_local half @test__Float16_pre_dc(
// UNSAFE-SAME: ) #[[ATTR0]] {
// UNSAFE-NEXT: [[ENTRY:.*:]]
// UNSAFE-NEXT: [[RETVAL:%.*]] = alloca half, align 2, addrspace(5)
// UNSAFE-NEXT: [[RETVAL_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[RETVAL]] to ptr
// UNSAFE-NEXT: [[TMP0:%.*]] = atomicrmw fsub ptr addrspacecast (ptr addrspace(1) @test__Float16_pre_dc.n to ptr), float 1.000000e+00 seq_cst, align 2, !amdgpu.no.fine.grained.memory [[META3]]
// UNSAFE-NEXT: [[TMP1:%.*]] = fsub float [[TMP0]], 1.000000e+00
// UNSAFE-NEXT: store float [[TMP1]], ptr [[RETVAL_ASCAST]], align 2
// UNSAFE-NEXT: [[TMP2:%.*]] = load half, ptr [[RETVAL_ASCAST]], align 2
// UNSAFE-NEXT: ret half [[TMP2]]
// UNSAFE-NEXT: [[TMP0:%.*]] = atomicrmw fsub ptr addrspacecast (ptr addrspace(1) @test__Float16_pre_dc.n to ptr), half 0xH3C00 seq_cst, align 2, !amdgpu.no.fine.grained.memory [[META3]]
// UNSAFE-NEXT: [[TMP1:%.*]] = fsub half [[TMP0]], 0xH3C00
// UNSAFE-NEXT: ret half [[TMP1]]
//
_Float16 test__Float16_pre_dc()
{
Expand All @@ -287,22 +259,18 @@ _Float16 test__Float16_pre_dc()
// SAFE-NEXT: [[ENTRY:.*:]]
// SAFE-NEXT: [[RETVAL:%.*]] = alloca half, align 2, addrspace(5)
// SAFE-NEXT: [[RETVAL_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[RETVAL]] to ptr
// SAFE-NEXT: [[TMP0:%.*]] = atomicrmw fadd ptr addrspacecast (ptr addrspace(1) @test__Float16_pre_inc.n to ptr), float 1.000000e+00 seq_cst, align 2
// SAFE-NEXT: [[TMP1:%.*]] = fadd float [[TMP0]], 1.000000e+00
// SAFE-NEXT: store float [[TMP1]], ptr [[RETVAL_ASCAST]], align 2
// SAFE-NEXT: [[TMP2:%.*]] = load half, ptr [[RETVAL_ASCAST]], align 2
// SAFE-NEXT: ret half [[TMP2]]
// SAFE-NEXT: [[TMP0:%.*]] = atomicrmw fadd ptr addrspacecast (ptr addrspace(1) @test__Float16_pre_inc.n to ptr), half 0xH3C00 seq_cst, align 2
// SAFE-NEXT: [[TMP1:%.*]] = fadd half [[TMP0]], 0xH3C00
// SAFE-NEXT: ret half [[TMP1]]
//
// UNSAFE-LABEL: define dso_local half @test__Float16_pre_inc(
// UNSAFE-SAME: ) #[[ATTR0]] {
// UNSAFE-NEXT: [[ENTRY:.*:]]
// UNSAFE-NEXT: [[RETVAL:%.*]] = alloca half, align 2, addrspace(5)
// UNSAFE-NEXT: [[RETVAL_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[RETVAL]] to ptr
// UNSAFE-NEXT: [[TMP0:%.*]] = atomicrmw fadd ptr addrspacecast (ptr addrspace(1) @test__Float16_pre_inc.n to ptr), float 1.000000e+00 seq_cst, align 2, !amdgpu.no.fine.grained.memory [[META3]], !amdgpu.ignore.denormal.mode [[META3]]
// UNSAFE-NEXT: [[TMP1:%.*]] = fadd float [[TMP0]], 1.000000e+00
// UNSAFE-NEXT: store float [[TMP1]], ptr [[RETVAL_ASCAST]], align 2
// UNSAFE-NEXT: [[TMP2:%.*]] = load half, ptr [[RETVAL_ASCAST]], align 2
// UNSAFE-NEXT: ret half [[TMP2]]
// UNSAFE-NEXT: [[TMP0:%.*]] = atomicrmw fadd ptr addrspacecast (ptr addrspace(1) @test__Float16_pre_inc.n to ptr), half 0xH3C00 seq_cst, align 2, !amdgpu.no.fine.grained.memory [[META3]]
// UNSAFE-NEXT: [[TMP1:%.*]] = fadd half [[TMP0]], 0xH3C00
// UNSAFE-NEXT: ret half [[TMP1]]
//
_Float16 test__Float16_pre_inc()
{
Expand Down
Loading