[clang] 62a251f - [Clang][BFloat16] Upgrade __bf16 by supporting increment/decrement operations

[Jun Sha via cfe-commits]

Author: Jun Sha (Joshua)
Date: 2023-07-28T16:21:24+08:00
New Revision: 62a251f824f63a56563b246035b9bd24078aa98b

URL: https://github.com/llvm/llvm-project/commit/62a251f824f63a56563b246035b9bd24078aa98b
DIFF: https://github.com/llvm/llvm-project/commit/62a251f824f63a56563b246035b9bd24078aa98b.diff

LOG: [Clang][BFloat16] Upgrade __bf16 by supporting increment/decrement operations

Since __bf16 has been upgraded from a storage-only type to an arithmetic type in https://reviews.llvm.org/rGe62175736551abf40a3410bc246f58e650eb8158, it should support all the basic arithmetic operations like other float types, including increment and decrement.

Reviewed By: pengfei

Differential Revision: https://reviews.llvm.org/D152768

Added: 
    

Modified: 
    clang/lib/CodeGen/CGExprScalar.cpp
    clang/test/Sema/arm-bfloat.cpp

Removed: 
    


################################################################################
diff  --git a/clang/lib/CodeGen/CGExprScalar.cpp b/clang/lib/CodeGen/CGExprScalar.cpp
index fe1a59b21f3801..33defb9c87e90f 100644
--- a/clang/lib/CodeGen/CGExprScalar.cpp
+++ b/clang/lib/CodeGen/CGExprScalar.cpp
@@ -2764,8 +2764,8 @@ ScalarExprEmitter::EmitScalarPrePostIncDec(const UnaryOperator *E, LValue LV,
       amt = llvm::ConstantFP::get(VMContext,
                                   llvm::APFloat(static_cast<double>(amount)));
     else {
-      // Remaining types are Half, LongDouble, __ibm128 or __float128. Convert
-      // from float.
+      // Remaining types are Half, Bfloat16, LongDouble, __ibm128 or __float128. 
+      // Convert from float.
       llvm::APFloat F(static_cast<float>(amount));
       bool ignored;
       const llvm::fltSemantics *FS;
@@ -2775,6 +2775,8 @@ ScalarExprEmitter::EmitScalarPrePostIncDec(const UnaryOperator *E, LValue LV,
         FS = &CGF.getTarget().getFloat128Format();
       else if (value->getType()->isHalfTy())
         FS = &CGF.getTarget().getHalfFormat();
+      else if (value->getType()->isBFloatTy())
+        FS = &CGF.getTarget().getBFloat16Format();
       else if (value->getType()->isPPC_FP128Ty())
         FS = &CGF.getTarget().getIbm128Format();
       else

diff  --git a/clang/test/Sema/arm-bfloat.cpp b/clang/test/Sema/arm-bfloat.cpp
index 34bea9b2ec9725..6a3eacd331d5c4 100644
--- a/clang/test/Sema/arm-bfloat.cpp
+++ b/clang/test/Sema/arm-bfloat.cpp
@@ -19,6 +19,8 @@ void test(bool b) {
   bf16 - bf16;
   bf16 * bf16;
   bf16 / bf16;
+  ++bf16;
+  --bf16;
 
   __fp16 fp16;