[clang] [IRGen][AArch64][RISCV] Generalize bitcast between i1 predicate vector and i8 fixed vector. (PR #76548)

[via cfe-commits]

llvmbot wrote:


<!--LLVM PR SUMMARY COMMENT-->

@llvm/pr-subscribers-clang-codegen

Author: Craig Topper (topperc)

<details>
<summary>Changes</summary>

Instead of only handling vscale x 16 x i1 predicate vectors, handle any scalable i1 vector where the known minimum is divisible by 8.

This will be used on RISC-V where we have multiple sizes of predicate types.

Though I do wonder if we have the bitcast on the wrong side. Maybe we should be casting the fixed type to an i1 fixed vector? That would better handle cases where the mininum elements isn't a multiple of 8, but vscale*min_elements is.

---
Full diff: https://github.com/llvm/llvm-project/pull/76548.diff


1 Files Affected:

- (modified) clang/lib/CodeGen/CGExprScalar.cpp (+14-10) 


``````````diff
diff --git a/clang/lib/CodeGen/CGExprScalar.cpp b/clang/lib/CodeGen/CGExprScalar.cpp
index 6adf99531e30e3..ace4f743c77596 100644
--- a/clang/lib/CodeGen/CGExprScalar.cpp
+++ b/clang/lib/CodeGen/CGExprScalar.cpp
@@ -2136,14 +2136,16 @@ Value *ScalarExprEmitter::VisitCastExpr(CastExpr *CE) {
     // bitcast.
     if (const auto *FixedSrc = dyn_cast<llvm::FixedVectorType>(SrcTy)) {
       if (const auto *ScalableDst = dyn_cast<llvm::ScalableVectorType>(DstTy)) {
-        // If we are casting a fixed i8 vector to a scalable 16 x i1 predicate
+        // If we are casting a fixed i8 vector to a scalable i1 predicate
         // vector, use a vector insert and bitcast the result.
         bool NeedsBitCast = false;
-        auto PredType = llvm::ScalableVectorType::get(Builder.getInt1Ty(), 16);
         llvm::Type *OrigType = DstTy;
-        if (ScalableDst == PredType &&
-            FixedSrc->getElementType() == Builder.getInt8Ty()) {
-          DstTy = llvm::ScalableVectorType::get(Builder.getInt8Ty(), 2);
+        if (ScalableDst->getElementType()->isIntegerTy(1) &&
+            ScalableDst->getElementCount().isKnownMultipleOf(8) &&
+            FixedSrc->getElementType()->isIntegerTy(8)) {
+          DstTy = llvm::VectorType::get(
+              FixedSrc->getElementType(),
+              ScalableDst->getElementCount().divideCoefficientBy(8));
           ScalableDst = cast<llvm::ScalableVectorType>(DstTy);
           NeedsBitCast = true;
         }
@@ -2164,12 +2166,14 @@ Value *ScalarExprEmitter::VisitCastExpr(CastExpr *CE) {
     // bitcast.
     if (const auto *ScalableSrc = dyn_cast<llvm::ScalableVectorType>(SrcTy)) {
       if (const auto *FixedDst = dyn_cast<llvm::FixedVectorType>(DstTy)) {
-        // If we are casting a scalable 16 x i1 predicate vector to a fixed i8
+        // If we are casting a scalable i1 predicate vector to a fixed i8
         // vector, bitcast the source and use a vector extract.
-        auto PredType = llvm::ScalableVectorType::get(Builder.getInt1Ty(), 16);
-        if (ScalableSrc == PredType &&
-            FixedDst->getElementType() == Builder.getInt8Ty()) {
-          SrcTy = llvm::ScalableVectorType::get(Builder.getInt8Ty(), 2);
+        if (ScalableSrc->getElementType()->isIntegerTy(1) &&
+            ScalableSrc->getElementCount().isKnownMultipleOf(8) &&
+            FixedDst->getElementType()->isIntegerTy(8)) {
+          SrcTy = llvm::VectorType::get(
+              FixedDst->getElementType(),
+              ScalableSrc->getElementCount().divideCoefficientBy(8));
           ScalableSrc = cast<llvm::ScalableVectorType>(SrcTy);
           Src = Builder.CreateBitCast(Src, SrcTy);
         }

``````````

</details>


https://github.com/llvm/llvm-project/pull/76548