diff --git a/llvm/lib/Transforms/InstCombine/InstCombineShifts.cpp b/llvm/lib/Transforms/InstCombine/InstCombineShifts.cpp
--- a/llvm/lib/Transforms/InstCombine/InstCombineShifts.cpp
+++ b/llvm/lib/Transforms/InstCombine/InstCombineShifts.cpp
@@ -71,10 +71,11 @@
 //
 // There are many variants to this pattern:
 //   a)  (x & ((1 << MaskShAmt) - 1)) << ShiftShAmt
+//   b)  (x & (~(-1 << maskNbits))) << shiftNbits
 // All these patterns can be simplified to just:
 //   x << ShiftShAmt
 // iff:
-//   a) (MaskShAmt+ShiftShAmt) u>= bitwidth(x)
+//   a,b) (MaskShAmt+ShiftShAmt) u>= bitwidth(x)
 static Instruction *
 dropRedundantMaskingOfLeftShiftInput(BinaryOperator *OuterShift,
                                      const SimplifyQuery &SQ) {
@@ -86,9 +87,11 @@
 
   // ((1 << MaskShAmt) - 1)
   auto MaskA = m_Add(m_Shl(m_One(), m_Value(MaskShAmt)), m_AllOnes());
+  // (~(-1 << maskNbits))
+  auto MaskB = m_Xor(m_Shl(m_AllOnes(), m_Value(MaskShAmt)), m_AllOnes());
 
   Value *X;
-  if (!match(Masked, m_c_And(MaskA, m_Value(X))))
+  if (!match(Masked, m_c_And(m_CombineOr(MaskA, MaskB), m_Value(X))))
     return nullptr;
 
   // Can we simplify (MaskShAmt+ShiftShAmt) ?
diff --git a/llvm/test/Transforms/InstCombine/redundant-left-shift-input-masking-variant-b.ll b/llvm/test/Transforms/InstCombine/redundant-left-shift-input-masking-variant-b.ll
--- a/llvm/test/Transforms/InstCombine/redundant-left-shift-input-masking-variant-b.ll
+++ b/llvm/test/Transforms/InstCombine/redundant-left-shift-input-masking-variant-b.ll
@@ -25,7 +25,7 @@
 ; CHECK-NEXT:    call void @use32(i32 [[T1]])
 ; CHECK-NEXT:    call void @use32(i32 [[T2]])
 ; CHECK-NEXT:    call void @use32(i32 [[T3]])
-; CHECK-NEXT:    [[T4:%.*]] = shl i32 [[T2]], [[T3]]
+; CHECK-NEXT:    [[T4:%.*]] = shl i32 [[X]], [[T3]]
 ; CHECK-NEXT:    ret i32 [[T4]]
 ;
   %t0 = shl i32 -1, %nbits
@@ -50,7 +50,7 @@
 ; CHECK-NEXT:    call void @use32(i32 [[T1]])
 ; CHECK-NEXT:    call void @use32(i32 [[T2]])
 ; CHECK-NEXT:    call void @use32(i32 [[T3]])
-; CHECK-NEXT:    [[T4:%.*]] = shl i32 [[T2]], [[T3]]
+; CHECK-NEXT:    [[T4:%.*]] = shl i32 [[X]], [[T3]]
 ; CHECK-NEXT:    ret i32 [[T4]]
 ;
   %t0 = shl i32 -1, %nbits
@@ -77,7 +77,7 @@
 ; CHECK-NEXT:    call void @use32(i32 [[T2]])
 ; CHECK-NEXT:    call void @use32(i32 [[T3]])
 ; CHECK-NEXT:    call void @use32(i32 [[T4]])
-; CHECK-NEXT:    [[T5:%.*]] = shl i32 [[T3]], [[T4]]
+; CHECK-NEXT:    [[T5:%.*]] = shl i32 [[X]], [[T4]]
 ; CHECK-NEXT:    ret i32 [[T5]]
 ;
   %t0 = add i32 %nbits, 1
@@ -109,7 +109,7 @@
 ; CHECK-NEXT:    call void @use3xi32(<3 x i32> [[T2]])
 ; CHECK-NEXT:    call void @use3xi32(<3 x i32> [[T3]])
 ; CHECK-NEXT:    call void @use3xi32(<3 x i32> [[T4]])
-; CHECK-NEXT:    [[T5:%.*]] = shl <3 x i32> [[T3]], [[T4]]
+; CHECK-NEXT:    [[T5:%.*]] = shl <3 x i32> [[X]], [[T4]]
 ; CHECK-NEXT:    ret <3 x i32> [[T5]]
 ;
   %t0 = add <3 x i32> %nbits, <i32 0, i32 0, i32 0>
@@ -138,7 +138,7 @@
 ; CHECK-NEXT:    call void @use3xi32(<3 x i32> [[T2]])
 ; CHECK-NEXT:    call void @use3xi32(<3 x i32> [[T3]])
 ; CHECK-NEXT:    call void @use3xi32(<3 x i32> [[T4]])
-; CHECK-NEXT:    [[T5:%.*]] = shl <3 x i32> [[T3]], [[T4]]
+; CHECK-NEXT:    [[T5:%.*]] = shl <3 x i32> [[X]], [[T4]]
 ; CHECK-NEXT:    ret <3 x i32> [[T5]]
 ;
   %t0 = add <3 x i32> %nbits, <i32 -1, i32 0, i32 1>
@@ -166,7 +166,7 @@
 ; CHECK-NEXT:    call void @use3xi32(<3 x i32> [[T2]])
 ; CHECK-NEXT:    call void @use3xi32(<3 x i32> [[T3]])
 ; CHECK-NEXT:    call void @use3xi32(<3 x i32> [[T4]])
-; CHECK-NEXT:    [[T5:%.*]] = shl <3 x i32> [[T3]], [[T4]]
+; CHECK-NEXT:    [[T5:%.*]] = shl <3 x i32> [[X]], [[T4]]
 ; CHECK-NEXT:    ret <3 x i32> [[T5]]
 ;
   %t0 = add <3 x i32> %nbits, <i32 0, i32 undef, i32 0>
@@ -198,7 +198,7 @@
 ; CHECK-NEXT:    call void @use32(i32 [[T1]])
 ; CHECK-NEXT:    call void @use32(i32 [[T2]])
 ; CHECK-NEXT:    call void @use32(i32 [[T3]])
-; CHECK-NEXT:    [[T4:%.*]] = shl i32 [[T2]], [[T3]]
+; CHECK-NEXT:    [[T4:%.*]] = shl i32 [[X]], [[T3]]
 ; CHECK-NEXT:    ret i32 [[T4]]
 ;
   %x = call i32 @gen32()
@@ -260,7 +260,7 @@
 ; CHECK-NEXT:    call void @use32(i32 [[T3]])
 ; CHECK-NEXT:    call void @use32(i32 [[T4]])
 ; CHECK-NEXT:    call void @use32(i32 [[T5]])
-; CHECK-NEXT:    [[T6:%.*]] = shl i32 [[T4]], [[T5]]
+; CHECK-NEXT:    [[T6:%.*]] = shl i32 [[T1]], [[T5]]
 ; CHECK-NEXT:    ret i32 [[T6]]
 ;
   %t0 = shl i32 -1, %nbits0
@@ -291,7 +291,7 @@
 ; CHECK-NEXT:    call void @use32(i32 [[T1]])
 ; CHECK-NEXT:    call void @use32(i32 [[T2]])
 ; CHECK-NEXT:    call void @use32(i32 [[T3]])
-; CHECK-NEXT:    [[T4:%.*]] = shl nuw i32 [[T2]], [[T3]]
+; CHECK-NEXT:    [[T4:%.*]] = shl i32 [[X]], [[T3]]
 ; CHECK-NEXT:    ret i32 [[T4]]
 ;
   %t0 = shl i32 -1, %nbits
@@ -316,7 +316,7 @@
 ; CHECK-NEXT:    call void @use32(i32 [[T1]])
 ; CHECK-NEXT:    call void @use32(i32 [[T2]])
 ; CHECK-NEXT:    call void @use32(i32 [[T3]])
-; CHECK-NEXT:    [[T4:%.*]] = shl nsw i32 [[T2]], [[T3]]
+; CHECK-NEXT:    [[T4:%.*]] = shl i32 [[X]], [[T3]]
 ; CHECK-NEXT:    ret i32 [[T4]]
 ;
   %t0 = shl i32 -1, %nbits
@@ -341,7 +341,7 @@
 ; CHECK-NEXT:    call void @use32(i32 [[T1]])
 ; CHECK-NEXT:    call void @use32(i32 [[T2]])
 ; CHECK-NEXT:    call void @use32(i32 [[T3]])
-; CHECK-NEXT:    [[T4:%.*]] = shl nuw nsw i32 [[T2]], [[T3]]
+; CHECK-NEXT:    [[T4:%.*]] = shl i32 [[X]], [[T3]]
 ; CHECK-NEXT:    ret i32 [[T4]]
 ;
   %t0 = shl i32 -1, %nbits