llvmbot wrote:
<!--LLVM PR SUMMARY COMMENT-->
@llvm/pr-subscribers-clang-codegen
Author: Andrzej Warzyński (banach-space)
<details>
<summary>Changes</summary>
Updates the logic in `CodeGenFunction::EmitAArch64BuiltinExpr` so that
we always start with the general code and we only fall-back to
specialised cases (i.e. `switch` stmts) for intrinsics for which the
general code does no apply.
BEFORE (only high-level:
```cpp
Value *CodeGenFunction::EmitAArch64BuiltinExpr() {
(...)
/// 1. SWITCH STMT FOR NON-OVERLOADED INTRINSIS
switch (BuiltinID) {
default break:
case NEON::BI__builtin_neon_vabsh_f16:
(...)
}
/// 2. GENERAL CODE
Builtin = findARMVectorIntrinsicInMap(AArch64SIMDIntrinsicMap, BuiltinID,
AArch64SIMDIntrinsicsProvenSorted);
if (Builtin)
return EmitCommonNeonBuiltinExpr(
Builtin->BuiltinID, Builtin->LLVMIntrinsic,
Builtin->AltLLVMIntrinsic,
Builtin->NameHint, Builtin->TypeModifier, E, Ops,
/*never use addresses*/ Address::invalid(), Address::invalid(), Arch);
if (Value *V = EmitAArch64TblBuiltinExpr(*this, BuiltinID, E, Ops, Arch))
return V;
/// 3. SWITCH STMT FOR THE REMAINING INTRINSIS
switch (BuiltinID) {
default return nullptr:
case NEON::BI__builtin_neon_vbsl_v:
(...)
}
}
```
AFTER:
```cpp
Value *CodeGenFunction::EmitAArch64BuiltinExpr() {
/// 1. GENERAL CODE
Builtin = findARMVectorIntrinsicInMap(AArch64SIMDIntrinsicMap, BuiltinID,
AArch64SIMDIntrinsicsProvenSorted);
if (Builtin)
return EmitCommonNeonBuiltinExpr(
Builtin->BuiltinID, Builtin->LLVMIntrinsic,
Builtin->AltLLVMIntrinsic,
Builtin->NameHint, Builtin->TypeModifier, E, Ops,
/*never use addresses*/ Address::invalid(), Address::invalid(), Arch);
if (Value *V = EmitAArch64TblBuiltinExpr(*this, BuiltinID, E, Ops, Arch))
return V;
/// 2. SWITCH STMT FOR NON-OVERLOADED INTRINSIS
switch (BuiltinID) {
default break:
case NEON::BI__builtin_neon_vabsh_f16:
(...)
}
/// 3. SWITCH STMT FOR THE REMAINING INTRINSIS
switch (BuiltinID) {
default return nullptr:
case NEON::BI__builtin_neon_vbsl_v:
(...)
}
}
```
In addition:
* Remove `vaddq_p128` from `AArch64SIMDIntrinsicMap`. Its not required
there (it's an array for intrinsics for which the general code
works, but that's not the case for `vaddq_p128`).
* Extracted the declaration of `Int` so that it can be re-used.
---
Full diff: https://github.com/llvm/llvm-project/pull/180966.diff
1 Files Affected:
- (modified) clang/lib/CodeGen/TargetBuiltins/ARM.cpp (+18-23)
``````````diff
diff --git a/clang/lib/CodeGen/TargetBuiltins/ARM.cpp
b/clang/lib/CodeGen/TargetBuiltins/ARM.cpp
index 06538756b91a8..0f01ab3539110 100644
--- a/clang/lib/CodeGen/TargetBuiltins/ARM.cpp
+++ b/clang/lib/CodeGen/TargetBuiltins/ARM.cpp
@@ -915,7 +915,6 @@ static const ARMVectorIntrinsicInfo
AArch64SIMDIntrinsicMap[] = {
NEONMAP1(vabsq_v, aarch64_neon_abs, 0),
NEONMAP0(vadd_v),
NEONMAP0(vaddhn_v),
- NEONMAP0(vaddq_p128),
NEONMAP0(vaddq_v),
NEONMAP1(vaesdq_u8, aarch64_crypto_aesd, 0),
NEONMAP1(vaeseq_u8, aarch64_crypto_aese, 0),
@@ -5930,6 +5929,21 @@ Value *CodeGenFunction::EmitAArch64BuiltinExpr(unsigned
BuiltinID,
bool usgn = Type.isUnsigned();
bool quad = Type.isQuad();
+ unsigned Int;
+
+ // Not all intrinsics handled by the common case work for AArch64 yet, so
only
+ // defer to common code if it's been added to our special map.
+ Builtin = findARMVectorIntrinsicInMap(AArch64SIMDIntrinsicMap, BuiltinID,
+ AArch64SIMDIntrinsicsProvenSorted);
+
+ if (Builtin)
+ return EmitCommonNeonBuiltinExpr(
+ Builtin->BuiltinID, Builtin->LLVMIntrinsic, Builtin->AltLLVMIntrinsic,
+ Builtin->NameHint, Builtin->TypeModifier, E, Ops,
+ /*never use addresses*/ Address::invalid(), Address::invalid(), Arch);
+
+ if (Value *V = EmitAArch64TblBuiltinExpr(*this, BuiltinID, E, Ops, Arch))
+ return V;
// Handle non-overloaded intrinsics first.
switch (BuiltinID) {
@@ -6003,7 +6017,6 @@ Value *CodeGenFunction::EmitAArch64BuiltinExpr(unsigned
BuiltinID,
case NEON::BI__builtin_neon_vcvtnh_s16_f16:
case NEON::BI__builtin_neon_vcvtph_s16_f16:
case NEON::BI__builtin_neon_vcvth_s16_f16: {
- unsigned Int;
llvm::Type *InTy = Int16Ty;
llvm::Type* FTy = HalfTy;
llvm::Type *Tys[2] = {InTy, FTy};
@@ -6037,7 +6050,6 @@ Value *CodeGenFunction::EmitAArch64BuiltinExpr(unsigned
BuiltinID,
case NEON::BI__builtin_neon_vcalth_f16:
case NEON::BI__builtin_neon_vcageh_f16:
case NEON::BI__builtin_neon_vcagth_f16: {
- unsigned Int;
llvm::Type* InTy = Int32Ty;
llvm::Type* FTy = HalfTy;
llvm::Type *Tys[2] = {InTy, FTy};
@@ -6058,7 +6070,6 @@ Value *CodeGenFunction::EmitAArch64BuiltinExpr(unsigned
BuiltinID,
}
case NEON::BI__builtin_neon_vcvth_n_s16_f16:
case NEON::BI__builtin_neon_vcvth_n_u16_f16: {
- unsigned Int;
llvm::Type* InTy = Int32Ty;
llvm::Type* FTy = HalfTy;
llvm::Type *Tys[2] = {InTy, FTy};
@@ -6075,7 +6086,6 @@ Value *CodeGenFunction::EmitAArch64BuiltinExpr(unsigned
BuiltinID,
}
case NEON::BI__builtin_neon_vcvth_n_f16_s16:
case NEON::BI__builtin_neon_vcvth_n_f16_u16: {
- unsigned Int;
llvm::Type* FTy = HalfTy;
llvm::Type* InTy = Int32Ty;
llvm::Type *Tys[2] = {FTy, InTy};
@@ -6486,7 +6496,7 @@ Value *CodeGenFunction::EmitAArch64BuiltinExpr(unsigned
BuiltinID,
}
case NEON::BI__builtin_neon_vqshld_n_u64:
case NEON::BI__builtin_neon_vqshld_n_s64: {
- unsigned Int = BuiltinID == NEON::BI__builtin_neon_vqshld_n_u64
+ Int = BuiltinID == NEON::BI__builtin_neon_vqshld_n_u64
? Intrinsic::aarch64_neon_uqshl
: Intrinsic::aarch64_neon_sqshl;
Ops.push_back(EmitScalarExpr(E->getArg(1)));
@@ -6495,7 +6505,7 @@ Value *CodeGenFunction::EmitAArch64BuiltinExpr(unsigned
BuiltinID,
}
case NEON::BI__builtin_neon_vrshrd_n_u64:
case NEON::BI__builtin_neon_vrshrd_n_s64: {
- unsigned Int = BuiltinID == NEON::BI__builtin_neon_vrshrd_n_u64
+ Int = BuiltinID == NEON::BI__builtin_neon_vrshrd_n_u64
? Intrinsic::aarch64_neon_urshl
: Intrinsic::aarch64_neon_srshl;
Ops.push_back(EmitScalarExpr(E->getArg(1)));
@@ -6505,7 +6515,7 @@ Value *CodeGenFunction::EmitAArch64BuiltinExpr(unsigned
BuiltinID,
}
case NEON::BI__builtin_neon_vrsrad_n_u64:
case NEON::BI__builtin_neon_vrsrad_n_s64: {
- unsigned Int = BuiltinID == NEON::BI__builtin_neon_vrsrad_n_u64
+ Int = BuiltinID == NEON::BI__builtin_neon_vrsrad_n_u64
? Intrinsic::aarch64_neon_urshl
: Intrinsic::aarch64_neon_srshl;
Ops[1] = Builder.CreateBitCast(Ops[1], Int64Ty);
@@ -6695,21 +6705,6 @@ Value *CodeGenFunction::EmitAArch64BuiltinExpr(unsigned
BuiltinID,
if (!Ty)
return nullptr;
- // Not all intrinsics handled by the common case work for AArch64 yet, so
only
- // defer to common code if it's been added to our special map.
- Builtin = findARMVectorIntrinsicInMap(AArch64SIMDIntrinsicMap, BuiltinID,
- AArch64SIMDIntrinsicsProvenSorted);
-
- if (Builtin)
- return EmitCommonNeonBuiltinExpr(
- Builtin->BuiltinID, Builtin->LLVMIntrinsic, Builtin->AltLLVMIntrinsic,
- Builtin->NameHint, Builtin->TypeModifier, E, Ops,
- /*never use addresses*/ Address::invalid(), Address::invalid(), Arch);
-
- if (Value *V = EmitAArch64TblBuiltinExpr(*this, BuiltinID, E, Ops, Arch))
- return V;
-
- unsigned Int;
bool ExtractLow = false;
bool ExtendLaneArg = false;
switch (BuiltinID) {
``````````
</details>
https://github.com/llvm/llvm-project/pull/180966
_______________________________________________
cfe-commits mailing list
[email protected]
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
