From f0903de1aa7a77a1d1cf9ecdf17e0c16ca135eaa Mon Sep 17 00:00:00 2001
From: Simon Pilgrim <llvm-dev@redking.me.uk>
Date: Wed, 29 Apr 2020 16:47:10 +0100
Subject: [PATCH] [x86] Enable bypassing 64-bit division on generic x86-64

This is currently enabled for Intel big cores from Sandy Bridge onward, as well as Atom, Silvermont, and KNL, due to 64-bit division being so slow on these cores. AMD cores can do this in hardware (use 32-bit division based on input operand width), so it's not a win there. But since the majority of x86 CPUs benefit from this optimization, and since the potential upside is significantly greater than the downside, we should enable this for the generic x86-64 target.

Patch By: @atdt

Reviewed By: @craig.topper, @RKSimon

Differential Revision: https://reviews.llvm.org/D75567
---
 llvm/lib/Target/X86/X86.td                         |  1 +
 llvm/test/CodeGen/X86/bypass-slow-division-tune.ll | 22 ++++++++++++++++++++++
 2 files changed, 23 insertions(+)

diff --git a/llvm/lib/Target/X86/X86.td b/llvm/lib/Target/X86/X86.td
index c7990ba..921c779 100644
--- a/llvm/lib/Target/X86/X86.td
+++ b/llvm/lib/Target/X86/X86.td
@@ -1260,6 +1260,7 @@ def : ProcessorModel<"x86-64", SandyBridgeModel, [
   FeatureNOPL,
   Feature64Bit,
   FeatureSlow3OpsLEA,
+  FeatureSlowDivide64,
   FeatureSlowIncDec,
   FeatureMacroFusion,
   FeatureInsertVZEROUPPER
diff --git a/llvm/test/CodeGen/X86/bypass-slow-division-tune.ll b/llvm/test/CodeGen/X86/bypass-slow-division-tune.ll
index 75a00dd..8369a44 100644
--- a/llvm/test/CodeGen/X86/bypass-slow-division-tune.ll
+++ b/llvm/test/CodeGen/X86/bypass-slow-division-tune.ll
@@ -66,9 +66,20 @@ define i64 @div64(i64 %a, i64 %b) {
 ; X64-LABEL: div64:
 ; X64:       # %bb.0: # %entry
 ; X64-NEXT:    movq %rdi, %rax
+; X64-NEXT:    movq %rdi, %rcx
+; X64-NEXT:    orq %rsi, %rcx
+; X64-NEXT:    shrq $32, %rcx
+; X64-NEXT:    je .LBB1_1
+; X64-NEXT:  # %bb.2:
 ; X64-NEXT:    cqto
 ; X64-NEXT:    idivq %rsi
 ; X64-NEXT:    retq
+; X64-NEXT:  .LBB1_1:
+; X64-NEXT:    # kill: def $eax killed $eax killed $rax
+; X64-NEXT:    xorl %edx, %edx
+; X64-NEXT:    divl %esi
+; X64-NEXT:    # kill: def $eax killed $eax def $rax
+; X64-NEXT:    retq
 ;
 ; SLM-LABEL: div64:
 ; SLM:       # %bb.0: # %entry
@@ -178,9 +189,20 @@ define i64 @div64_hugews(i64 %a, i64 %b) {
 ; X64-LABEL: div64_hugews:
 ; X64:       # %bb.0:
 ; X64-NEXT:    movq %rdi, %rax
+; X64-NEXT:    movq %rdi, %rcx
+; X64-NEXT:    orq %rsi, %rcx
+; X64-NEXT:    shrq $32, %rcx
+; X64-NEXT:    je .LBB4_1
+; X64-NEXT:  # %bb.2:
 ; X64-NEXT:    cqto
 ; X64-NEXT:    idivq %rsi
 ; X64-NEXT:    retq
+; X64-NEXT:  .LBB4_1:
+; X64-NEXT:    # kill: def $eax killed $eax killed $rax
+; X64-NEXT:    xorl %edx, %edx
+; X64-NEXT:    divl %esi
+; X64-NEXT:    # kill: def $eax killed $eax def $rax
+; X64-NEXT:    retq
 ;
 ; SLM-LABEL: div64_hugews:
 ; SLM:       # %bb.0:
-- 
2.7.4