From 92d0c1c1290aa4e570cce83cf671d4d6d2d6d9c3 Mon Sep 17 00:00:00 2001
From: Sanjay Patel <spatel@rotateright.com>
Date: Sun, 15 Jul 2018 14:52:16 +0000
Subject: [PATCH] [InstSimplify] fold minnum/maxnum with NaN arg

This fold is repeated/misplaced in instcombine, but I'm
not sure if it's safe to remove that yet because some
other folds appear to be asserting that the transform
has occurred within instcombine itself.

This isn't the best fix for PR37776, but it probably
hides the bug with the given code example:
https://bugs.llvm.org/show_bug.cgi?id=37776

We have another test to demonstrate the more general bug.

llvm-svn: 337127
---
 llvm/lib/Analysis/InstructionSimplify.cpp     |  8 +++++++
 .../InstSimplify/floating-point-arithmetic.ll | 24 +++++++------------
 2 files changed, 16 insertions(+), 16 deletions(-)

diff --git a/llvm/lib/Analysis/InstructionSimplify.cpp b/llvm/lib/Analysis/InstructionSimplify.cpp
index 87b1fc5b8ee7..2eed5fc64187 100644
--- a/llvm/lib/Analysis/InstructionSimplify.cpp
+++ b/llvm/lib/Analysis/InstructionSimplify.cpp
@@ -4725,6 +4725,14 @@ static Value *SimplifyIntrinsic(Function *F, IterTy ArgBegin, IterTy ArgEnd,
           return LHS;
       }
       return nullptr;
+    case Intrinsic::maxnum:
+    case Intrinsic::minnum:
+      // If one argument is NaN, return the other argument.
+      if (match(LHS, m_NaN()))
+        return RHS;
+      if (match(RHS, m_NaN()))
+        return LHS;
+      return nullptr;
     default:
       return nullptr;
     }
diff --git a/llvm/test/Transforms/InstSimplify/floating-point-arithmetic.ll b/llvm/test/Transforms/InstSimplify/floating-point-arithmetic.ll
index ead4bdbfe5dd..afd375f3b13e 100644
--- a/llvm/test/Transforms/InstSimplify/floating-point-arithmetic.ll
+++ b/llvm/test/Transforms/InstSimplify/floating-point-arithmetic.ll
@@ -474,8 +474,7 @@ declare <2 x double> @llvm.maxnum.v2f64(<2 x double>, <2 x double>)
 
 define double @maxnum_nan_op0(double %x) {
 ; CHECK-LABEL: @maxnum_nan_op0(
-; CHECK-NEXT:    [[R:%.*]] = call double @llvm.maxnum.f64(double 0x7FF8000000000000, double [[X:%.*]])
-; CHECK-NEXT:    ret double [[R]]
+; CHECK-NEXT:    ret double [[X:%.*]]
 ;
   %r = call double @llvm.maxnum.f64(double 0x7ff8000000000000, double %x)
   ret double %r
@@ -483,8 +482,7 @@ define double @maxnum_nan_op0(double %x) {
 
 define double @maxnum_nan_op1(double %x) {
 ; CHECK-LABEL: @maxnum_nan_op1(
-; CHECK-NEXT:    [[R:%.*]] = call double @llvm.maxnum.f64(double [[X:%.*]], double 0x7FF800000000DEAD)
-; CHECK-NEXT:    ret double [[R]]
+; CHECK-NEXT:    ret double [[X:%.*]]
 ;
   %r = call double @llvm.maxnum.f64(double %x, double 0x7ff800000000dead)
   ret double %r
@@ -492,8 +490,7 @@ define double @maxnum_nan_op1(double %x) {
 
 define double @minnum_nan_op0(double %x) {
 ; CHECK-LABEL: @minnum_nan_op0(
-; CHECK-NEXT:    [[R:%.*]] = call double @llvm.minnum.f64(double 0x7FF8000DEAD00000, double [[X:%.*]])
-; CHECK-NEXT:    ret double [[R]]
+; CHECK-NEXT:    ret double [[X:%.*]]
 ;
   %r = call double @llvm.minnum.f64(double 0x7ff8000dead00000, double %x)
   ret double %r
@@ -501,8 +498,7 @@ define double @minnum_nan_op0(double %x) {
 
 define double @minnum_nan_op1(double %x) {
 ; CHECK-LABEL: @minnum_nan_op1(
-; CHECK-NEXT:    [[R:%.*]] = call double @llvm.minnum.f64(double [[X:%.*]], double 0x7FF800DEAD00DEAD)
-; CHECK-NEXT:    ret double [[R]]
+; CHECK-NEXT:    ret double [[X:%.*]]
 ;
   %r = call double @llvm.minnum.f64(double %x, double 0x7ff800dead00dead)
   ret double %r
@@ -510,8 +506,7 @@ define double @minnum_nan_op1(double %x) {
 
 define <2 x double> @maxnum_nan_op0_vec(<2 x double> %x) {
 ; CHECK-LABEL: @maxnum_nan_op0_vec(
-; CHECK-NEXT:    [[R:%.*]] = call <2 x double> @llvm.maxnum.v2f64(<2 x double> <double 0x7FF8000000000000, double undef>, <2 x double> [[X:%.*]])
-; CHECK-NEXT:    ret <2 x double> [[R]]
+; CHECK-NEXT:    ret <2 x double> [[X:%.*]]
 ;
   %r = call <2 x double> @llvm.maxnum.v2f64(<2 x double> <double 0x7ff8000000000000, double undef>, <2 x double> %x)
   ret <2 x double> %r
@@ -519,8 +514,7 @@ define <2 x double> @maxnum_nan_op0_vec(<2 x double> %x) {
 
 define <2 x double> @maxnum_nan_op1_vec(<2 x double> %x) {
 ; CHECK-LABEL: @maxnum_nan_op1_vec(
-; CHECK-NEXT:    [[R:%.*]] = call <2 x double> @llvm.maxnum.v2f64(<2 x double> [[X:%.*]], <2 x double> <double 0x7FF800000000DEAD, double 0x7FF8FFFFFFFFFFFF>)
-; CHECK-NEXT:    ret <2 x double> [[R]]
+; CHECK-NEXT:    ret <2 x double> [[X:%.*]]
 ;
   %r = call <2 x double> @llvm.maxnum.v2f64(<2 x double> %x, <2 x double> <double 0x7ff800000000dead, double 0x7ff8ffffffffffff>)
   ret <2 x double> %r
@@ -528,8 +522,7 @@ define <2 x double> @maxnum_nan_op1_vec(<2 x double> %x) {
 
 define <2 x double> @minnum_nan_op0_vec(<2 x double> %x) {
 ; CHECK-LABEL: @minnum_nan_op0_vec(
-; CHECK-NEXT:    [[R:%.*]] = call <2 x double> @llvm.minnum.v2f64(<2 x double> <double undef, double 0x7FF8000DEAD00000>, <2 x double> [[X:%.*]])
-; CHECK-NEXT:    ret <2 x double> [[R]]
+; CHECK-NEXT:    ret <2 x double> [[X:%.*]]
 ;
   %r = call <2 x double> @llvm.minnum.v2f64(<2 x double> <double undef, double 0x7ff8000dead00000>, <2 x double> %x)
   ret <2 x double> %r
@@ -537,8 +530,7 @@ define <2 x double> @minnum_nan_op0_vec(<2 x double> %x) {
 
 define <2 x double> @minnum_nan_op1_vec(<2 x double> %x) {
 ; CHECK-LABEL: @minnum_nan_op1_vec(
-; CHECK-NEXT:    [[R:%.*]] = call <2 x double> @llvm.minnum.v2f64(<2 x double> [[X:%.*]], <2 x double> <double 0x7FF800DEAD00DEAD, double 0x7FF800DEAD00DEAD>)
-; CHECK-NEXT:    ret <2 x double> [[R]]
+; CHECK-NEXT:    ret <2 x double> [[X:%.*]]
 ;
   %r = call <2 x double> @llvm.minnum.v2f64(<2 x double> %x, <2 x double> <double 0x7ff800dead00dead, double 0x7ff800dead00dead>)
   ret <2 x double> %r