diff --git a/llvm/lib/Target/X86/X86TargetTransformInfo.h b/llvm/lib/Target/X86/X86TargetTransformInfo.h
--- a/llvm/lib/Target/X86/X86TargetTransformInfo.h
+++ b/llvm/lib/Target/X86/X86TargetTransformInfo.h
@@ -177,6 +177,10 @@
                                          Align Alignment,
                                          TTI::TargetCostKind CostKind,
                                          const Instruction *I);
+  InstructionCost getPointersChainCost(ArrayRef<const Value *> Ptrs,
+                                       const Value *Base,
+                                       const TTI::PointersChainInfo &Info,
+                                       TTI::TargetCostKind CostKind);
   InstructionCost getAddressComputationCost(Type *PtrTy, ScalarEvolution *SE,
                                             const SCEV *Ptr);
 
diff --git a/llvm/lib/Target/X86/X86TargetTransformInfo.cpp b/llvm/lib/Target/X86/X86TargetTransformInfo.cpp
--- a/llvm/lib/Target/X86/X86TargetTransformInfo.cpp
+++ b/llvm/lib/Target/X86/X86TargetTransformInfo.cpp
@@ -4899,6 +4899,23 @@
   return Cost + LT.first;
 }
 
+InstructionCost X86TTIImpl::getPointersChainCost(
+    ArrayRef<const Value *> Ptrs, const Value *Base,
+    const TTI::PointersChainInfo &Info, TTI::TargetCostKind CostKind) {
+  if (Info.isSameBase() && Info.isKnownStride()) {
+    // If all the pointers have known stride all the differences are translated
+    // into constants. X86 memory addressing allows encoding it into
+    // displacement. So we just need to take the base GEP cost.
+    if (const auto *BaseGEP = dyn_cast<GetElementPtrInst>(Base)) {
+      SmallVector<const Value *> Indices(BaseGEP->indices());
+      return getGEPCost(BaseGEP->getSourceElementType(),
+                        BaseGEP->getPointerOperand(), Indices, CostKind);
+    }
+    return TTI::TCC_Free;
+  }
+  return BaseT::getPointersChainCost(Ptrs, Base, Info, CostKind);
+}
+
 InstructionCost X86TTIImpl::getAddressComputationCost(Type *Ty,
                                                       ScalarEvolution *SE,
                                                       const SCEV *Ptr) {
diff --git a/llvm/test/Transforms/SLPVectorizer/X86/remark_horcost.ll b/llvm/test/Transforms/SLPVectorizer/X86/remark_horcost.ll
--- a/llvm/test/Transforms/SLPVectorizer/X86/remark_horcost.ll
+++ b/llvm/test/Transforms/SLPVectorizer/X86/remark_horcost.ll
@@ -86,7 +86,7 @@
   ; YAML-NEXT: Function:        foo
   ; YAML-NEXT: Args:
   ; YAML-NEXT:   - String:          'Stores SLP vectorized with cost '
-  ; YAML-NEXT:   - Cost:            '-14'
+  ; YAML-NEXT:   - Cost:            '-5'
   ; YAML-NEXT:   - String:          ' and with tree size '
   ; YAML-NEXT:   - TreeSize:        '4'
 
diff --git a/llvm/test/Transforms/SLPVectorizer/X86/remark_not_all_parts.ll b/llvm/test/Transforms/SLPVectorizer/X86/remark_not_all_parts.ll
--- a/llvm/test/Transforms/SLPVectorizer/X86/remark_not_all_parts.ll
+++ b/llvm/test/Transforms/SLPVectorizer/X86/remark_not_all_parts.ll
@@ -62,7 +62,7 @@
   ; YAML-NEXT: Function:        foo
   ; YAML-NEXT: Args:
   ; YAML-NEXT:   - String:          'Stores SLP vectorized with cost '
-  ; YAML-NEXT:   - Cost:            '-4'
+  ; YAML-NEXT:   - Cost:            '-1'
   ; YAML-NEXT:   - String:          ' and with tree size '
   ; YAML-NEXT:   - TreeSize:        '4'