[InstCombine] Fold a shifty implementation of clamp-to-allones.

Summary:
Fold
or(ashr(subNSW(Y, X), ScalarSizeInBits(Y)-1), X)
into
X s> Y ? -1 : X

https://rise4fun.com/Alive/d8Ab

clamp255 is a common operator in image processing, can be implemented
in a shifty way "(255 - X) >> 31 | X & 255". Fold shift into select
enables more optimization, e.g., vmin generation for ARM target.

Reviewers: lebedev.ri, efriedma, spatel, kparzysz, bcahoon

Reviewed By: lebedev.ri

Subscribers: kristof.beyls, hiraditya, llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D67800

llvm-svn: 372678

2 files changed, 39 insertions, 35 deletions

diff --git a/llvm/lib/Transforms/InstCombine/InstCombineAndOrXor.cpp b/llvm/lib/Transforms/InstCombine/InstCombineAndOrXor.cpp
index 490faddaefb..cde4cc1e842 100644
--- a/llvm/lib/Transforms/InstCombine/InstCombineAndOrXor.cpp
+++ b/llvm/lib/Transforms/InstCombine/InstCombineAndOrXor.cpp
@@ -2677,6 +2677,21 @@ Instruction *InstCombiner::visitOr(BinaryOperator &I) {
     }
   }
 
+  // or(ashr(subNSW(Y, X), ScalarSizeInBits(Y)-1), X)  --> X s> Y ? -1 : X.
+  {
+    Value *X, *Y;
+    const APInt *ShAmt;
+    Type *Ty = I.getType();
+    if (match(&I, m_c_Or(m_OneUse(m_AShr(m_NSWSub(m_Value(Y), m_Value(X)),
+                                         m_APInt(ShAmt))),
+                         m_Deferred(X))) &&
+        *ShAmt == Ty->getScalarSizeInBits() - 1) {
+      Value *NewICmpInst = Builder.CreateICmpSGT(X, Y);
+      return SelectInst::Create(NewICmpInst, ConstantInt::getAllOnesValue(Ty),
+                                X);
+    }
+  }
+
   return nullptr;
 }
 
diff --git a/llvm/test/Transforms/InstCombine/sub-ashr-or-to-icmp-select.ll b/llvm/test/Transforms/InstCombine/sub-ashr-or-to-icmp-select.ll
index 630a14fd242..937bff62dc7 100644
--- a/llvm/test/Transforms/InstCombine/sub-ashr-or-to-icmp-select.ll
+++ b/llvm/test/Transforms/InstCombine/sub-ashr-or-to-icmp-select.ll
@@ -12,9 +12,8 @@
 
 define i32 @clamp255_i32(i32 %x) {
 ; CHECK-LABEL: @clamp255_i32(
-; CHECK-NEXT:    [[SUB:%.*]] = sub nsw i32 255, [[X:%.*]]
-; CHECK-NEXT:    [[SHR:%.*]] = ashr i32 [[SUB]], 31
-; CHECK-NEXT:    [[OR:%.*]] = or i32 [[SHR]], [[X]]
+; CHECK-NEXT:    [[TMP1:%.*]] = icmp slt i32 [[X:%.*]], 255
+; CHECK-NEXT:    [[OR:%.*]] = select i1 [[TMP1]], i32 [[X]], i32 255
 ; CHECK-NEXT:    [[AND:%.*]] = and i32 [[OR]], 255
 ; CHECK-NEXT:    ret i32 [[AND]]
 ;
@@ -27,9 +26,8 @@ define i32 @clamp255_i32(i32 %x) {
 
 define i8 @sub_ashr_or_i8(i8 %x, i8 %y) {
 ; CHECK-LABEL: @sub_ashr_or_i8(
-; CHECK-NEXT:    [[SUB:%.*]] = sub nsw i8 [[Y:%.*]], [[X:%.*]]
-; CHECK-NEXT:    [[SHR:%.*]] = ashr i8 [[SUB]], 7
-; CHECK-NEXT:    [[OR:%.*]] = or i8 [[SHR]], [[X]]
+; CHECK-NEXT:    [[TMP1:%.*]] = icmp sgt i8 [[X:%.*]], [[Y:%.*]]
+; CHECK-NEXT:    [[OR:%.*]] = select i1 [[TMP1]], i8 -1, i8 [[X]]
 ; CHECK-NEXT:    ret i8 [[OR]]
 ;
   %sub = sub nsw i8 %y, %x
@@ -40,9 +38,8 @@ define i8 @sub_ashr_or_i8(i8 %x, i8 %y) {
 
 define i16 @sub_ashr_or_i16(i16 %x, i16 %y) {
 ; CHECK-LABEL: @sub_ashr_or_i16(
-; CHECK-NEXT:    [[SUB:%.*]] = sub nsw i16 [[Y:%.*]], [[X:%.*]]
-; CHECK-NEXT:    [[SHR:%.*]] = ashr i16 [[SUB]], 15
-; CHECK-NEXT:    [[OR:%.*]] = or i16 [[SHR]], [[X]]
+; CHECK-NEXT:    [[TMP1:%.*]] = icmp sgt i16 [[X:%.*]], [[Y:%.*]]
+; CHECK-NEXT:    [[OR:%.*]] = select i1 [[TMP1]], i16 -1, i16 [[X]]
 ; CHECK-NEXT:    ret i16 [[OR]]
 ;
   %sub = sub nsw i16 %y, %x
@@ -53,9 +50,8 @@ define i16 @sub_ashr_or_i16(i16 %x, i16 %y) {
 
 define i32 @sub_ashr_or_i32(i32 %x, i32 %y) {
 ; CHECK-LABEL: @sub_ashr_or_i32(
-; CHECK-NEXT:    [[SUB:%.*]] = sub nsw i32 [[Y:%.*]], [[X:%.*]]
-; CHECK-NEXT:    [[SHR:%.*]] = ashr i32 [[SUB]], 31
-; CHECK-NEXT:    [[OR:%.*]] = or i32 [[SHR]], [[X]]
+; CHECK-NEXT:    [[TMP1:%.*]] = icmp sgt i32 [[X:%.*]], [[Y:%.*]]
+; CHECK-NEXT:    [[OR:%.*]] = select i1 [[TMP1]], i32 -1, i32 [[X]]
 ; CHECK-NEXT:    ret i32 [[OR]]
 ;
   %sub = sub nsw i32 %y, %x
@@ -66,9 +62,8 @@ define i32 @sub_ashr_or_i32(i32 %x, i32 %y) {
 
 define i64 @sub_ashr_or_i64(i64 %x, i64 %y) {
 ; CHECK-LABEL: @sub_ashr_or_i64(
-; CHECK-NEXT:    [[SUB:%.*]] = sub nsw i64 [[Y:%.*]], [[X:%.*]]
-; CHECK-NEXT:    [[SHR:%.*]] = ashr i64 [[SUB]], 63
-; CHECK-NEXT:    [[OR:%.*]] = or i64 [[SHR]], [[X]]
+; CHECK-NEXT:    [[TMP1:%.*]] = icmp sgt i64 [[X:%.*]], [[Y:%.*]]
+; CHECK-NEXT:    [[OR:%.*]] = select i1 [[TMP1]], i64 -1, i64 [[X]]
 ; CHECK-NEXT:    ret i64 [[OR]]
 ;
   %sub = sub nsw i64 %y, %x
@@ -81,9 +76,8 @@ define i64 @sub_ashr_or_i64(i64 %x, i64 %y) {
 
 define i32 @sub_ashr_or_i32_nuw_nsw(i32 %x, i32 %y) {
 ; CHECK-LABEL: @sub_ashr_or_i32_nuw_nsw(
-; CHECK-NEXT:    [[SUB:%.*]] = sub nuw nsw i32 [[Y:%.*]], [[X:%.*]]
-; CHECK-NEXT:    [[SHR:%.*]] = ashr i32 [[SUB]], 31
-; CHECK-NEXT:    [[OR:%.*]] = or i32 [[SHR]], [[X]]
+; CHECK-NEXT:    [[TMP1:%.*]] = icmp sgt i32 [[X:%.*]], [[Y:%.*]]
+; CHECK-NEXT:    [[OR:%.*]] = select i1 [[TMP1]], i32 -1, i32 [[X]]
 ; CHECK-NEXT:    ret i32 [[OR]]
 ;
   %sub = sub nuw nsw i32 %y, %x
@@ -96,9 +90,8 @@ define i32 @sub_ashr_or_i32_nuw_nsw(i32 %x, i32 %y) {
 
 define i32 @sub_ashr_or_i32_commute(i32 %x, i32 %y) {
 ; CHECK-LABEL: @sub_ashr_or_i32_commute(
-; CHECK-NEXT:    [[SUB:%.*]] = sub nsw i32 [[Y:%.*]], [[X:%.*]]
-; CHECK-NEXT:    [[SHR:%.*]] = ashr i32 [[SUB]], 31
-; CHECK-NEXT:    [[OR:%.*]] = or i32 [[SHR]], [[X]]
+; CHECK-NEXT:    [[TMP1:%.*]] = icmp sgt i32 [[X:%.*]], [[Y:%.*]]
+; CHECK-NEXT:    [[OR:%.*]] = select i1 [[TMP1]], i32 -1, i32 [[X]]
 ; CHECK-NEXT:    ret i32 [[OR]]
 ;
   %sub = sub nsw i32 %y, %x
@@ -111,9 +104,8 @@ define i32 @sub_ashr_or_i32_commute(i32 %x, i32 %y) {
 
 define <4 x i32> @sub_ashr_or_i32_vec(<4 x i32> %x, <4 x i32> %y) {
 ; CHECK-LABEL: @sub_ashr_or_i32_vec(
-; CHECK-NEXT:    [[SUB:%.*]] = sub nsw <4 x i32> [[Y:%.*]], [[X:%.*]]
-; CHECK-NEXT:    [[SHR:%.*]] = ashr <4 x i32> [[SUB]], <i32 31, i32 31, i32 31, i32 31>
-; CHECK-NEXT:    [[OR:%.*]] = or <4 x i32> [[SHR]], [[X]]
+; CHECK-NEXT:    [[TMP1:%.*]] = icmp sgt <4 x i32> [[X:%.*]], [[Y:%.*]]
+; CHECK-NEXT:    [[OR:%.*]] = select <4 x i1> [[TMP1]], <4 x i32> <i32 -1, i32 -1, i32 -1, i32 -1>, <4 x i32> [[X]]
 ; CHECK-NEXT:    ret <4 x i32> [[OR]]
 ;
   %sub = sub nsw <4 x i32> %y, %x
@@ -124,9 +116,8 @@ define <4 x i32> @sub_ashr_or_i32_vec(<4 x i32> %x, <4 x i32> %y) {
 
 define <4 x i32> @sub_ashr_or_i32_vec_nuw_nsw(<4 x i32> %x, <4 x i32> %y) {
 ; CHECK-LABEL: @sub_ashr_or_i32_vec_nuw_nsw(
-; CHECK-NEXT:    [[SUB:%.*]] = sub nuw nsw <4 x i32> [[Y:%.*]], [[X:%.*]]
-; CHECK-NEXT:    [[SHR:%.*]] = ashr <4 x i32> [[SUB]], <i32 31, i32 31, i32 31, i32 31>
-; CHECK-NEXT:    [[OR:%.*]] = or <4 x i32> [[SHR]], [[X]]
+; CHECK-NEXT:    [[TMP1:%.*]] = icmp sgt <4 x i32> [[X:%.*]], [[Y:%.*]]
+; CHECK-NEXT:    [[OR:%.*]] = select <4 x i1> [[TMP1]], <4 x i32> <i32 -1, i32 -1, i32 -1, i32 -1>, <4 x i32> [[X]]
 ; CHECK-NEXT:    ret <4 x i32> [[OR]]
 ;
   %sub = sub nuw nsw <4 x i32> %y, %x
@@ -137,9 +128,8 @@ define <4 x i32> @sub_ashr_or_i32_vec_nuw_nsw(<4 x i32> %x, <4 x i32> %y) {
 
 define <4 x i32> @sub_ashr_or_i32_vec_commute(<4 x i32> %x, <4 x i32> %y) {
 ; CHECK-LABEL: @sub_ashr_or_i32_vec_commute(
-; CHECK-NEXT:    [[SUB:%.*]] = sub nsw <4 x i32> [[Y:%.*]], [[X:%.*]]
-; CHECK-NEXT:    [[SHR:%.*]] = ashr <4 x i32> [[SUB]], <i32 31, i32 31, i32 31, i32 31>
-; CHECK-NEXT:    [[OR:%.*]] = or <4 x i32> [[SHR]], [[X]]
+; CHECK-NEXT:    [[TMP1:%.*]] = icmp sgt <4 x i32> [[X:%.*]], [[Y:%.*]]
+; CHECK-NEXT:    [[OR:%.*]] = select <4 x i1> [[TMP1]], <4 x i32> <i32 -1, i32 -1, i32 -1, i32 -1>, <4 x i32> [[X]]
 ; CHECK-NEXT:    ret <4 x i32> [[OR]]
 ;
   %sub = sub nsw <4 x i32> %y, %x
@@ -154,8 +144,8 @@ define i32 @sub_ashr_or_i32_extra_use_sub(i32 %x, i32 %y, i32* %p) {
 ; CHECK-LABEL: @sub_ashr_or_i32_extra_use_sub(
 ; CHECK-NEXT:    [[SUB:%.*]] = sub nsw i32 [[Y:%.*]], [[X:%.*]]
 ; CHECK-NEXT:    store i32 [[SUB]], i32* [[P:%.*]], align 4
-; CHECK-NEXT:    [[SHR:%.*]] = ashr i32 [[SUB]], 31
-; CHECK-NEXT:    [[OR:%.*]] = or i32 [[SHR]], [[X]]
+; CHECK-NEXT:    [[TMP1:%.*]] = icmp slt i32 [[Y]], [[X]]
+; CHECK-NEXT:    [[OR:%.*]] = select i1 [[TMP1]], i32 -1, i32 [[X]]
 ; CHECK-NEXT:    ret i32 [[OR]]
 ;
   %sub = sub nsw i32 %y, %x
@@ -167,9 +157,8 @@ define i32 @sub_ashr_or_i32_extra_use_sub(i32 %x, i32 %y, i32* %p) {
 
 define i32 @sub_ashr_or_i32_extra_use_or(i32 %x, i32 %y, i32* %p) {
 ; CHECK-LABEL: @sub_ashr_or_i32_extra_use_or(
-; CHECK-NEXT:    [[SUB:%.*]] = sub nsw i32 [[Y:%.*]], [[X:%.*]]
-; CHECK-NEXT:    [[SHR:%.*]] = ashr i32 [[SUB]], 31
-; CHECK-NEXT:    [[OR:%.*]] = or i32 [[SHR]], [[X]]
+; CHECK-NEXT:    [[TMP1:%.*]] = icmp sgt i32 [[X:%.*]], [[Y:%.*]]
+; CHECK-NEXT:    [[OR:%.*]] = select i1 [[TMP1]], i32 -1, i32 [[X]]
 ; CHECK-NEXT:    store i32 [[OR]], i32* [[P:%.*]], align 4
 ; CHECK-NEXT:    ret i32 [[OR]]
 ;