[clang] [CIR] Skip generation of a continue block when flattening TernaryOp (PR #142165)

[via cfe-commits]

llvmbot wrote:


<!--LLVM PR SUMMARY COMMENT-->

@llvm/pr-subscribers-clangir

Author: Morris Hafner (mmha)

<details>
<summary>Changes</summary>

We used to insert a continue Block at the end of a flattened ternary op that only contained a branch to the remaing operation of the remaining Block. This patch removes that continue block and changes the true/false blocks to directly jump to the remaining ops.

With this patch the CIR now generates exactly the same LLVM IR as the original codegen.

This upstreams llvm/clangir#<!-- -->1651.

---
Full diff: https://github.com/llvm/llvm-project/pull/142165.diff


3 Files Affected:

- (modified) clang/lib/CIR/Dialect/Transforms/FlattenCFG.cpp (+15-16) 
- (modified) clang/test/CIR/Lowering/ternary.cir (-2) 
- (modified) clang/test/CIR/Transforms/ternary.cir (-4) 


``````````diff
diff --git a/clang/lib/CIR/Dialect/Transforms/FlattenCFG.cpp b/clang/lib/CIR/Dialect/Transforms/FlattenCFG.cpp
index 26e5c0572f12e..6081a436f5c29 100644
--- a/clang/lib/CIR/Dialect/Transforms/FlattenCFG.cpp
+++ b/clang/lib/CIR/Dialect/Transforms/FlattenCFG.cpp
@@ -16,9 +16,11 @@
 #include "mlir/IR/Block.h"
 #include "mlir/IR/Builders.h"
 #include "mlir/IR/PatternMatch.h"
+#include "mlir/IR/ValueRange.h"
 #include "mlir/Support/LogicalResult.h"
 #include "mlir/Transforms/DialectConversion.h"
 #include "mlir/Transforms/GreedyPatternRewriteDriver.h"
+#include "clang/AST/DeclBase.h"
 #include "clang/CIR/Dialect/IR/CIRDialect.h"
 #include "clang/CIR/Dialect/Passes.h"
 #include "clang/CIR/MissingFeatures.h"
@@ -491,15 +493,7 @@ class CIRTernaryOpFlattening : public mlir::OpRewritePattern<cir::TernaryOp> {
     Location loc = op->getLoc();
     Block *condBlock = rewriter.getInsertionBlock();
     Block::iterator opPosition = rewriter.getInsertionPoint();
-    Block *remainingOpsBlock = rewriter.splitBlock(condBlock, opPosition);
-    llvm::SmallVector<mlir::Location, 2> locs;
-    // Ternary result is optional, make sure to populate the location only
-    // when relevant.
-    if (op->getResultTypes().size())
-      locs.push_back(loc);
-    Block *continueBlock =
-        rewriter.createBlock(remainingOpsBlock, op->getResultTypes(), locs);
-    rewriter.create<cir::BrOp>(loc, remainingOpsBlock);
+    auto *remainingOpsBlock = rewriter.splitBlock(condBlock, opPosition);
 
     Region &trueRegion = op.getTrueRegion();
     Block *trueBlock = &trueRegion.front();
@@ -508,24 +502,29 @@ class CIRTernaryOpFlattening : public mlir::OpRewritePattern<cir::TernaryOp> {
     auto trueYieldOp = dyn_cast<cir::YieldOp>(trueTerminator);
 
     rewriter.replaceOpWithNewOp<cir::BrOp>(trueYieldOp, trueYieldOp.getArgs(),
-                                           continueBlock);
-    rewriter.inlineRegionBefore(trueRegion, continueBlock);
+                                           remainingOpsBlock);
+    rewriter.inlineRegionBefore(trueRegion, remainingOpsBlock);
 
-    Block *falseBlock = continueBlock;
     Region &falseRegion = op.getFalseRegion();
+    Block *falseBlock = &falseRegion.front();
 
-    falseBlock = &falseRegion.front();
     mlir::Operation *falseTerminator = falseRegion.back().getTerminator();
     rewriter.setInsertionPointToEnd(&falseRegion.back());
     auto falseYieldOp = dyn_cast<cir::YieldOp>(falseTerminator);
     rewriter.replaceOpWithNewOp<cir::BrOp>(falseYieldOp, falseYieldOp.getArgs(),
-                                           continueBlock);
-    rewriter.inlineRegionBefore(falseRegion, continueBlock);
+                                           remainingOpsBlock);
+    rewriter.inlineRegionBefore(falseRegion, remainingOpsBlock);
 
     rewriter.setInsertionPointToEnd(condBlock);
     rewriter.create<cir::BrCondOp>(loc, op.getCond(), trueBlock, falseBlock);
 
-    rewriter.replaceOp(op, continueBlock->getArguments());
+    if (auto rt = op.getResultTypes(); rt.size()) {
+      iterator_range args = remainingOpsBlock->addArguments(rt, op.getLoc());
+      SmallVector<mlir::Value, 2> values;
+      llvm::copy(args, std::back_inserter(values));
+      rewriter.replaceOpUsesWithinBlock(op, values, remainingOpsBlock);
+    }
+    rewriter.eraseOp(op);
 
     // Ok, we're done!
     return mlir::success();
diff --git a/clang/test/CIR/Lowering/ternary.cir b/clang/test/CIR/Lowering/ternary.cir
index 247c6ae3a1e17..c2226cd92ece7 100644
--- a/clang/test/CIR/Lowering/ternary.cir
+++ b/clang/test/CIR/Lowering/ternary.cir
@@ -25,6 +25,4 @@ module  {
 // LLVM:   br label %[[M]]
 // LLVM: [[M]]:
 // LLVM:   [[R:%[[:alnum:]]+]] = phi i32 [ 1, %[[B2]] ], [ 0, %[[B1]] ]
-// LLVM:   br label %[[B3:[[:alnum:]]+]]
-// LLVM: [[B3]]:
 // LLVM:   ret i32 [[R]]
diff --git a/clang/test/CIR/Transforms/ternary.cir b/clang/test/CIR/Transforms/ternary.cir
index 67ef7f95a6b52..0c22268495697 100644
--- a/clang/test/CIR/Transforms/ternary.cir
+++ b/clang/test/CIR/Transforms/ternary.cir
@@ -37,8 +37,6 @@ module {
 // CHECK:    %6 = cir.const #cir.int<5> : !s32i
 // CHECK:    cir.br ^bb3(%6 : !s32i)
 // CHECK:  ^bb3(%7: !s32i):  // 2 preds: ^bb1, ^bb2
-// CHECK:    cir.br ^bb4
-// CHECK:  ^bb4:  // pred: ^bb3
 // CHECK:    cir.store %7, %1 : !s32i, !cir.ptr<!s32i>
 // CHECK:    %8 = cir.load %1 : !cir.ptr<!s32i>, !s32i
 // CHECK:    cir.return %8 : !s32i
@@ -60,8 +58,6 @@ module {
 // CHECK: ^bb2:  // pred: ^bb0
 // CHECK:   cir.br ^bb3
 // CHECK: ^bb3:  // 2 preds: ^bb1, ^bb2
-// CHECK:   cir.br ^bb4
-// CHECK: ^bb4:  // pred: ^bb3
 // CHECK:   cir.return
 // CHECK: }
 

``````````

</details>


https://github.com/llvm/llvm-project/pull/142165