[llvm] [EarlyCSE] Support CSE for commutative intrinsics with over 2 args (PR #67255)

via llvm-commits llvm-commits at lists.llvm.org
Sat Sep 23 23:47:46 PDT 2023


https://github.com/XChy updated https://github.com/llvm/llvm-project/pull/67255

>From 7988e4b7bce25e968b442f82c360a0afd9bca8ec Mon Sep 17 00:00:00 2001
From: XChy <xxs_chy at outlook.com>
Date: Sun, 24 Sep 2023 13:40:20 +0800
Subject: [PATCH] [EarlyCSE] Support CSE for commutative intrinsics with over 2
 args

---
 llvm/lib/Transforms/Scalar/EarlyCSE.cpp  | 19 ++++++++-----
 llvm/test/Transforms/EarlyCSE/commute.ll | 36 ++++--------------------
 2 files changed, 18 insertions(+), 37 deletions(-)

diff --git a/llvm/lib/Transforms/Scalar/EarlyCSE.cpp b/llvm/lib/Transforms/Scalar/EarlyCSE.cpp
index 4990fa9f8b5ea36..06379f04f872e0d 100644
--- a/llvm/lib/Transforms/Scalar/EarlyCSE.cpp
+++ b/llvm/lib/Transforms/Scalar/EarlyCSE.cpp
@@ -314,14 +314,17 @@ static unsigned getHashValueImpl(SimpleValue Val) {
          "Invalid/unknown instruction");
 
   // Handle intrinsics with commutative operands.
-  // TODO: Extend this to handle intrinsics with >2 operands where the 1st
-  //       2 operands are commutative.
   auto *II = dyn_cast<IntrinsicInst>(Inst);
-  if (II && II->isCommutative() && II->arg_size() == 2) {
+  if (II && II->isCommutative() && II->arg_size() >= 2) {
     Value *LHS = II->getArgOperand(0), *RHS = II->getArgOperand(1);
     if (LHS > RHS)
       std::swap(LHS, RHS);
-    return hash_combine(II->getOpcode(), LHS, RHS);
+    auto HashCode = hash_combine(II->getOpcode(), LHS, RHS);
+    if (II->arg_size() > 2)
+      HashCode =
+          hash_combine(HashCode, hash_combine_range(II->value_op_begin() + 2,
+                                                    II->value_op_end()));
+    return HashCode;
   }
 
   // gc.relocate is 'special' call: its second and third operands are
@@ -396,13 +399,15 @@ static bool isEqualImpl(SimpleValue LHS, SimpleValue RHS) {
            LHSCmp->getSwappedPredicate() == RHSCmp->getPredicate();
   }
 
-  // TODO: Extend this for >2 args by matching the trailing N-2 args.
   auto *LII = dyn_cast<IntrinsicInst>(LHSI);
   auto *RII = dyn_cast<IntrinsicInst>(RHSI);
   if (LII && RII && LII->getIntrinsicID() == RII->getIntrinsicID() &&
-      LII->isCommutative() && LII->arg_size() == 2) {
+      LII->isCommutative() && LII->arg_size() >= 2) {
     return LII->getArgOperand(0) == RII->getArgOperand(1) &&
-           LII->getArgOperand(1) == RII->getArgOperand(0);
+           LII->getArgOperand(1) == RII->getArgOperand(0) &&
+           (LII->arg_size() == 2 ||
+            std::equal(LII->arg_begin() + 2, LII->arg_end(),
+                       RII->arg_begin() + 2));
   }
 
   // See comment above in `getHashValue()`.
diff --git a/llvm/test/Transforms/EarlyCSE/commute.ll b/llvm/test/Transforms/EarlyCSE/commute.ll
index 6aaaf992e441423..015ea8b3e28763c 100644
--- a/llvm/test/Transforms/EarlyCSE/commute.ll
+++ b/llvm/test/Transforms/EarlyCSE/commute.ll
@@ -999,14 +999,10 @@ define i4 @smin_umin(i4 %a, i4 %b) {
   ret i4 %o
 }
 
-; TODO: handle >2 args
-
 define i16 @smul_fix(i16 %a, i16 %b) {
 ; CHECK-LABEL: @smul_fix(
 ; CHECK-NEXT:    [[X:%.*]] = call i16 @llvm.smul.fix.i16(i16 [[A:%.*]], i16 [[B:%.*]], i32 3)
-; CHECK-NEXT:    [[Y:%.*]] = call i16 @llvm.smul.fix.i16(i16 [[B]], i16 [[A]], i32 3)
-; CHECK-NEXT:    [[O:%.*]] = or i16 [[X]], [[Y]]
-; CHECK-NEXT:    ret i16 [[O]]
+; CHECK-NEXT:    ret i16 [[X]]
 ;
   %x = call i16 @llvm.smul.fix.i16(i16 %a, i16 %b, i32 3)
   %y = call i16 @llvm.smul.fix.i16(i16 %b, i16 %a, i32 3)
@@ -1014,14 +1010,10 @@ define i16 @smul_fix(i16 %a, i16 %b) {
   ret i16 %o
 }
 
-; TODO: handle >2 args
-
 define i16 @umul_fix(i16 %a, i16 %b, i32 %s) {
 ; CHECK-LABEL: @umul_fix(
 ; CHECK-NEXT:    [[X:%.*]] = call i16 @llvm.umul.fix.i16(i16 [[A:%.*]], i16 [[B:%.*]], i32 1)
-; CHECK-NEXT:    [[Y:%.*]] = call i16 @llvm.umul.fix.i16(i16 [[B]], i16 [[A]], i32 1)
-; CHECK-NEXT:    [[O:%.*]] = or i16 [[X]], [[Y]]
-; CHECK-NEXT:    ret i16 [[O]]
+; CHECK-NEXT:    ret i16 [[X]]
 ;
   %x = call i16 @llvm.umul.fix.i16(i16 %a, i16 %b, i32 1)
   %y = call i16 @llvm.umul.fix.i16(i16 %b, i16 %a, i32 1)
@@ -1029,14 +1021,10 @@ define i16 @umul_fix(i16 %a, i16 %b, i32 %s) {
   ret i16 %o
 }
 
-; TODO: handle >2 args
-
 define <3 x i16> @smul_fix_sat(<3 x i16> %a, <3 x i16> %b) {
 ; CHECK-LABEL: @smul_fix_sat(
 ; CHECK-NEXT:    [[X:%.*]] = call <3 x i16> @llvm.smul.fix.sat.v3i16(<3 x i16> [[A:%.*]], <3 x i16> [[B:%.*]], i32 2)
-; CHECK-NEXT:    [[Y:%.*]] = call <3 x i16> @llvm.smul.fix.sat.v3i16(<3 x i16> [[B]], <3 x i16> [[A]], i32 2)
-; CHECK-NEXT:    [[O:%.*]] = or <3 x i16> [[X]], [[Y]]
-; CHECK-NEXT:    ret <3 x i16> [[O]]
+; CHECK-NEXT:    ret <3 x i16> [[X]]
 ;
   %x = call <3 x i16> @llvm.smul.fix.sat.v3i16(<3 x i16> %a, <3 x i16> %b, i32 2)
   %y = call <3 x i16> @llvm.smul.fix.sat.v3i16(<3 x i16> %b, <3 x i16> %a, i32 2)
@@ -1044,14 +1032,10 @@ define <3 x i16> @smul_fix_sat(<3 x i16> %a, <3 x i16> %b) {
   ret <3 x i16> %o
 }
 
-; TODO: handle >2 args
-
 define <3 x i16> @umul_fix_sat(<3 x i16> %a, <3 x i16> %b) {
 ; CHECK-LABEL: @umul_fix_sat(
 ; CHECK-NEXT:    [[X:%.*]] = call <3 x i16> @llvm.umul.fix.sat.v3i16(<3 x i16> [[A:%.*]], <3 x i16> [[B:%.*]], i32 3)
-; CHECK-NEXT:    [[Y:%.*]] = call <3 x i16> @llvm.umul.fix.sat.v3i16(<3 x i16> [[B]], <3 x i16> [[A]], i32 3)
-; CHECK-NEXT:    [[O:%.*]] = or <3 x i16> [[X]], [[Y]]
-; CHECK-NEXT:    ret <3 x i16> [[O]]
+; CHECK-NEXT:    ret <3 x i16> [[X]]
 ;
   %x = call <3 x i16> @llvm.umul.fix.sat.v3i16(<3 x i16> %a, <3 x i16> %b, i32 3)
   %y = call <3 x i16> @llvm.umul.fix.sat.v3i16(<3 x i16> %b, <3 x i16> %a, i32 3)
@@ -1085,14 +1069,10 @@ define i16 @umul_fix_scale(i16 %a, i16 %b, i32 %s) {
   ret i16 %o
 }
 
-; TODO: handle >2 args
-
 define float @fma(float %a, float %b, float %c) {
 ; CHECK-LABEL: @fma(
 ; CHECK-NEXT:    [[X:%.*]] = call float @llvm.fma.f32(float [[A:%.*]], float [[B:%.*]], float [[C:%.*]])
-; CHECK-NEXT:    [[Y:%.*]] = call float @llvm.fma.f32(float [[B]], float [[A]], float [[C]])
-; CHECK-NEXT:    [[R:%.*]] = fdiv nnan float [[X]], [[Y]]
-; CHECK-NEXT:    ret float [[R]]
+; CHECK-NEXT:    ret float 1.000000e+00
 ;
   %x = call float @llvm.fma.f32(float %a, float %b, float %c)
   %y = call float @llvm.fma.f32(float %b, float %a, float %c)
@@ -1113,14 +1093,10 @@ define float @fma_different_add_ops(float %a, float %b, float %c, float %d) {
   ret float %r
 }
 
-; TODO: handle >2 args
-
 define <2 x double> @fmuladd(<2 x double> %a, <2 x double> %b, <2 x double> %c) {
 ; CHECK-LABEL: @fmuladd(
 ; CHECK-NEXT:    [[X:%.*]] = call <2 x double> @llvm.fmuladd.v2f64(<2 x double> [[A:%.*]], <2 x double> [[B:%.*]], <2 x double> [[C:%.*]])
-; CHECK-NEXT:    [[Y:%.*]] = call <2 x double> @llvm.fmuladd.v2f64(<2 x double> [[B]], <2 x double> [[A]], <2 x double> [[C]])
-; CHECK-NEXT:    [[R:%.*]] = fdiv nnan <2 x double> [[X]], [[Y]]
-; CHECK-NEXT:    ret <2 x double> [[R]]
+; CHECK-NEXT:    ret <2 x double> <double 1.000000e+00, double 1.000000e+00>
 ;
   %x = call <2 x double> @llvm.fmuladd.v2f64(<2 x double> %a, <2 x double> %b, <2 x double> %c)
   %y = call <2 x double> @llvm.fmuladd.v2f64(<2 x double> %b, <2 x double> %a, <2 x double> %c)



More information about the llvm-commits mailing list