diff --git a/llvm/lib/Target/RISCV/RISCVISelDAGToDAG.cpp b/llvm/lib/Target/RISCV/RISCVISelDAGToDAG.cpp
--- a/llvm/lib/Target/RISCV/RISCVISelDAGToDAG.cpp
+++ b/llvm/lib/Target/RISCV/RISCVISelDAGToDAG.cpp
@@ -2308,6 +2308,44 @@
   return false;
 }
 
+// Can we benefit from compressed load/store instructions by inferring MemVT
+// from all uses of this address?
+static bool benefitFromCompressedLoadStore(SDValue Addr, uint64_t Offset,
+                                           const RISCVSubtarget *Subtarget) {
+  if (!isShiftedInt<5, 2>(Offset) && !isShiftedInt<5, 3>(Offset))
+    return false;
+  if (!Subtarget->hasStdExtCOrZca())
+    return false;
+
+  for (auto *Use : Addr->uses()) {
+    if (Use->getOpcode() != ISD::LOAD && Use->getOpcode() != ISD::STORE)
+      continue;
+    // Don't allow stores of the value. It must be used as the address.
+    if (Use->getOpcode() == ISD::STORE &&
+        cast<StoreSDNode>(Use)->getValue() == Addr)
+      continue;
+
+    EVT VT = cast<MemSDNode>(Use)->getMemoryVT();
+
+    if (VT == MVT::i32 && Subtarget->hasStdExtCOrZca() &&
+        isShiftedInt<5, 2>(Offset))
+      return true;
+    if (VT == MVT::i64 && Subtarget->hasStdExtCOrZca() &&
+        isShiftedInt<5, 3>(Offset))
+      return true;
+    if (VT == MVT::f32 &&
+        (Subtarget->hasStdExtC() || Subtarget->hasStdExtZcf()) &&
+        isShiftedInt<5, 2>(Offset))
+      return true;
+    if (VT == MVT::f64 &&
+        (Subtarget->hasStdExtC() || Subtarget->hasStdExtZcd()) &&
+        isShiftedInt<5, 3>(Offset))
+      return true;
+  }
+
+  return false;
+}
+
 bool RISCVDAGToDAGISel::SelectAddrRegImm(SDValue Addr, SDValue &Base,
                                          SDValue &Offset) {
   if (SelectAddrFrameIndex(Addr, Base, Offset))
@@ -2322,10 +2360,55 @@
     return true;
   }
 
-  if (CurDAG->isBaseWithConstantOffset(Addr)) {
+  // Decompose the addimm chain into (Base, Offset) pairs
+  SmallVector<std::pair<SDValue, int64_t>, 4> AddrDecomposition;
+  {
+    SDValue CurAddr = Addr;
+    int64_t CurOffset = 0;
+    while (CurAddr.getOpcode() == ISD::ADD &&
+           isa<ConstantSDNode>(CurAddr.getOperand(1))) {
+      SDValue CurBase = CurAddr.getOperand(0);
+      CurOffset += cast<ConstantSDNode>(CurAddr.getOperand(1))->getSExtValue();
+      AddrDecomposition.push_back(std::make_pair(CurBase, CurOffset));
+      CurAddr = CurBase;
+    }
+    if (AddrDecomposition.empty() && CurDAG->isBaseWithConstantOffset(Addr)) {
+      AddrDecomposition.push_back(std::make_pair(
+          Addr.getOperand(0),
+          cast<ConstantSDNode>(Addr.getOperand(1))->getSExtValue()));
+    }
+  }
+
+  if (!AddrDecomposition.empty()) {
+    bool IsWorthFoldingAdd = isWorthFoldingAdd(Addr);
+
+    uint32_t BestCost = std::numeric_limits<uint32_t>::max();
+    Base = Addr.getOperand(0);
     int64_t CVal = cast<ConstantSDNode>(Addr.getOperand(1))->getSExtValue();
+    for (auto [CurBase, CurOffset] : AddrDecomposition) {
+      uint32_t CurCost = std::numeric_limits<uint32_t>::max();
+      if (isInt<12>(CurOffset)) {
+        CurCost =
+            benefitFromCompressedLoadStore(Addr, CurOffset, Subtarget) ? 2 : 4;
+      } else if (IsWorthFoldingAdd) {
+        int64_t Adj = CurOffset < 0 ? -2048 : 2047;
+        int64_t OffsetVal = CurOffset - Adj;
+        if (isInt<12>(OffsetVal)) {
+          CurCost = benefitFromCompressedLoadStore(Addr, OffsetVal, Subtarget)
+                        ? 2
+                        : 4;
+        }
+      }
+
+      // Prefer deeper pair when they cost the same.
+      if (CurCost <= BestCost) {
+        BestCost = CurCost;
+        Base = CurBase;
+        CVal = CurOffset;
+      }
+    }
+
     if (isInt<12>(CVal)) {
-      Base = Addr.getOperand(0);
       if (Base.getOpcode() == RISCVISD::ADD_LO) {
         SDValue LoOperand = Base.getOperand(1);
         if (auto *GA = dyn_cast<GlobalAddressSDNode>(LoOperand)) {
@@ -2353,23 +2436,21 @@
       Offset = CurDAG->getTargetConstant(CVal, DL, VT);
       return true;
     }
-  }
 
-  // Handle ADD with large immediates.
-  if (Addr.getOpcode() == ISD::ADD && isa<ConstantSDNode>(Addr.getOperand(1))) {
-    int64_t CVal = cast<ConstantSDNode>(Addr.getOperand(1))->getSExtValue();
+    // Handle ADD chain with large immediates.
     assert(!isInt<12>(CVal) && "simm12 not already handled?");
 
-    // Handle immediates in the range [-4096,-2049] or [2048, 4094]. We can use
+    // Handle immediates in the range [-4096, -2049] or [2048, 4094]. We can use
     // an ADDI for part of the offset and fold the rest into the load/store.
     // This mirrors the AddiPair PatFrag in RISCVInstrInfo.td.
-    if (isInt<12>(CVal / 2) && isInt<12>(CVal - CVal / 2)) {
-      int64_t Adj = CVal < 0 ? -2048 : 2047;
+    int64_t Adj = CVal < 0 ? -2048 : 2047;
+    int64_t OffsetVal = CVal - Adj;
+    if (IsWorthFoldingAdd && isInt<12>(OffsetVal)) {
       Base = SDValue(
-          CurDAG->getMachineNode(RISCV::ADDI, DL, VT, Addr.getOperand(0),
+          CurDAG->getMachineNode(RISCV::ADDI, DL, VT, Base,
                                  CurDAG->getTargetConstant(Adj, DL, VT)),
           0);
-      Offset = CurDAG->getTargetConstant(CVal - Adj, DL, VT);
+      Offset = CurDAG->getTargetConstant(OffsetVal, DL, VT);
       return true;
     }
 
@@ -2379,13 +2460,13 @@
     // stores that can fold the lo12 bits. Otherwise, the ADD will get iseled
     // separately with the full materialized immediate creating extra
     // instructions.
-    if (isWorthFoldingAdd(Addr) &&
-        selectConstantAddr(CurDAG, DL, VT, Subtarget, Addr.getOperand(1), Base,
-                           Offset)) {
+    SDValue AddrBase = Base;
+    SDValue AddrOffset = CurDAG->getTargetConstant(CVal, DL, VT);
+    if (IsWorthFoldingAdd && selectConstantAddr(CurDAG, DL, VT, Subtarget,
+                                                AddrOffset, Base, Offset)) {
       // Insert an ADD instruction with the materialized Hi52 bits.
-      Base = SDValue(
-          CurDAG->getMachineNode(RISCV::ADD, DL, VT, Addr.getOperand(0), Base),
-          0);
+      Base = SDValue(CurDAG->getMachineNode(RISCV::ADD, DL, VT, AddrBase, Base),
+                     0);
       return true;
     }
   }
diff --git a/llvm/test/CodeGen/RISCV/split-offsets.ll b/llvm/test/CodeGen/RISCV/split-offsets.ll
--- a/llvm/test/CodeGen/RISCV/split-offsets.ll
+++ b/llvm/test/CodeGen/RISCV/split-offsets.ll
@@ -1,43 +1,47 @@
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
 ; RUN: llc -mtriple=riscv32 -verify-machineinstrs < %s \
-; RUN:   | FileCheck %s -check-prefix=RV32I
+; RUN:   | FileCheck %s -check-prefixes=CHECK,RV32,RV32I
 ; RUN: llc -mtriple=riscv64 -verify-machineinstrs < %s \
-; RUN:   | FileCheck %s -check-prefix=RV64I
+; RUN:   | FileCheck %s -check-prefixes=CHECK,RV64,RV64I
+; RUN: llc -mtriple=riscv32 -mattr=+c -verify-machineinstrs < %s \
+; RUN:   | FileCheck %s -check-prefixes=CHECK,RV32,RV32C
+; RUN: llc -mtriple=riscv64 -mattr=+c -verify-machineinstrs < %s \
+; RUN:   | FileCheck %s -check-prefixes=CHECK,RV64,RV64C
 
 ; Check that memory accesses to array elements with large offsets have those
 ; offsets split into a base offset, plus a smaller offset that is folded into
 ; the memory operation. We should also only compute that base offset once,
 ; since it can be shared for all memory operations in this test.
 define void @test1(ptr %sp, ptr %t, i32 %n) {
-; RV32I-LABEL: test1:
-; RV32I:       # %bb.0: # %entry
-; RV32I-NEXT:    lw a0, 0(a0)
-; RV32I-NEXT:    lui a2, 20
-; RV32I-NEXT:    addi a2, a2, -1920
-; RV32I-NEXT:    add a1, a1, a2
-; RV32I-NEXT:    add a0, a0, a2
-; RV32I-NEXT:    li a2, 2
-; RV32I-NEXT:    sw a2, 0(a0)
-; RV32I-NEXT:    li a3, 1
-; RV32I-NEXT:    sw a3, 4(a0)
-; RV32I-NEXT:    sw a3, 0(a1)
-; RV32I-NEXT:    sw a2, 4(a1)
-; RV32I-NEXT:    ret
+; RV32-LABEL: test1:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    lw a0, 0(a0)
+; RV32-NEXT:    lui a2, 20
+; RV32-NEXT:    addi a2, a2, -1920
+; RV32-NEXT:    add a1, a1, a2
+; RV32-NEXT:    add a0, a0, a2
+; RV32-NEXT:    li a2, 2
+; RV32-NEXT:    sw a2, 0(a0)
+; RV32-NEXT:    li a3, 1
+; RV32-NEXT:    sw a3, 4(a0)
+; RV32-NEXT:    sw a3, 0(a1)
+; RV32-NEXT:    sw a2, 4(a1)
+; RV32-NEXT:    ret
 ;
-; RV64I-LABEL: test1:
-; RV64I:       # %bb.0: # %entry
-; RV64I-NEXT:    ld a0, 0(a0)
-; RV64I-NEXT:    lui a2, 20
-; RV64I-NEXT:    addiw a2, a2, -1920
-; RV64I-NEXT:    add a1, a1, a2
-; RV64I-NEXT:    add a0, a0, a2
-; RV64I-NEXT:    li a2, 2
-; RV64I-NEXT:    sw a2, 0(a0)
-; RV64I-NEXT:    li a3, 1
-; RV64I-NEXT:    sw a3, 4(a0)
-; RV64I-NEXT:    sw a3, 0(a1)
-; RV64I-NEXT:    sw a2, 4(a1)
-; RV64I-NEXT:    ret
+; RV64-LABEL: test1:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    ld a0, 0(a0)
+; RV64-NEXT:    lui a2, 20
+; RV64-NEXT:    addiw a2, a2, -1920
+; RV64-NEXT:    add a1, a1, a2
+; RV64-NEXT:    add a0, a0, a2
+; RV64-NEXT:    li a2, 2
+; RV64-NEXT:    sw a2, 0(a0)
+; RV64-NEXT:    li a3, 1
+; RV64-NEXT:    sw a3, 4(a0)
+; RV64-NEXT:    sw a3, 0(a1)
+; RV64-NEXT:    sw a2, 4(a1)
+; RV64-NEXT:    ret
 entry:
   %s = load ptr, ptr %sp
   %gep0 = getelementptr [65536 x i32], ptr %s, i64 0, i32 20000
@@ -53,48 +57,48 @@
 
 ; Ditto. Check it when the GEPs are not in the entry block.
 define void @test2(ptr %sp, ptr %t, i32 %n) {
-; RV32I-LABEL: test2:
-; RV32I:       # %bb.0: # %entry
-; RV32I-NEXT:    li a3, 0
-; RV32I-NEXT:    lw a0, 0(a0)
-; RV32I-NEXT:    lui a4, 20
-; RV32I-NEXT:    addi a4, a4, -1920
-; RV32I-NEXT:    add a1, a1, a4
-; RV32I-NEXT:    add a0, a0, a4
-; RV32I-NEXT:    blez a2, .LBB1_2
-; RV32I-NEXT:  .LBB1_1: # %while_body
-; RV32I-NEXT:    # =>This Inner Loop Header: Depth=1
-; RV32I-NEXT:    addi a4, a3, 1
-; RV32I-NEXT:    sw a4, 0(a0)
-; RV32I-NEXT:    sw a3, 4(a0)
-; RV32I-NEXT:    sw a4, 0(a1)
-; RV32I-NEXT:    sw a3, 4(a1)
-; RV32I-NEXT:    mv a3, a4
-; RV32I-NEXT:    blt a4, a2, .LBB1_1
-; RV32I-NEXT:  .LBB1_2: # %while_end
-; RV32I-NEXT:    ret
+; RV32-LABEL: test2:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    li a3, 0
+; RV32-NEXT:    lw a0, 0(a0)
+; RV32-NEXT:    lui a4, 20
+; RV32-NEXT:    addi a4, a4, -1920
+; RV32-NEXT:    add a1, a1, a4
+; RV32-NEXT:    add a0, a0, a4
+; RV32-NEXT:    blez a2, .LBB1_2
+; RV32-NEXT:  .LBB1_1: # %while_body
+; RV32-NEXT:    # =>This Inner Loop Header: Depth=1
+; RV32-NEXT:    addi a4, a3, 1
+; RV32-NEXT:    sw a4, 0(a0)
+; RV32-NEXT:    sw a3, 4(a0)
+; RV32-NEXT:    sw a4, 0(a1)
+; RV32-NEXT:    sw a3, 4(a1)
+; RV32-NEXT:    mv a3, a4
+; RV32-NEXT:    blt a4, a2, .LBB1_1
+; RV32-NEXT:  .LBB1_2: # %while_end
+; RV32-NEXT:    ret
 ;
-; RV64I-LABEL: test2:
-; RV64I:       # %bb.0: # %entry
-; RV64I-NEXT:    li a3, 0
-; RV64I-NEXT:    ld a0, 0(a0)
-; RV64I-NEXT:    lui a4, 20
-; RV64I-NEXT:    addiw a4, a4, -1920
-; RV64I-NEXT:    add a1, a1, a4
-; RV64I-NEXT:    add a0, a0, a4
-; RV64I-NEXT:    sext.w a2, a2
-; RV64I-NEXT:    blez a2, .LBB1_2
-; RV64I-NEXT:  .LBB1_1: # %while_body
-; RV64I-NEXT:    # =>This Inner Loop Header: Depth=1
-; RV64I-NEXT:    addiw a4, a3, 1
-; RV64I-NEXT:    sw a4, 0(a0)
-; RV64I-NEXT:    sw a3, 4(a0)
-; RV64I-NEXT:    sw a4, 0(a1)
-; RV64I-NEXT:    sw a3, 4(a1)
-; RV64I-NEXT:    mv a3, a4
-; RV64I-NEXT:    blt a4, a2, .LBB1_1
-; RV64I-NEXT:  .LBB1_2: # %while_end
-; RV64I-NEXT:    ret
+; RV64-LABEL: test2:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    li a3, 0
+; RV64-NEXT:    ld a0, 0(a0)
+; RV64-NEXT:    lui a4, 20
+; RV64-NEXT:    addiw a4, a4, -1920
+; RV64-NEXT:    add a1, a1, a4
+; RV64-NEXT:    add a0, a0, a4
+; RV64-NEXT:    sext.w a2, a2
+; RV64-NEXT:    blez a2, .LBB1_2
+; RV64-NEXT:  .LBB1_1: # %while_body
+; RV64-NEXT:    # =>This Inner Loop Header: Depth=1
+; RV64-NEXT:    addiw a4, a3, 1
+; RV64-NEXT:    sw a4, 0(a0)
+; RV64-NEXT:    sw a3, 4(a0)
+; RV64-NEXT:    sw a4, 0(a1)
+; RV64-NEXT:    sw a3, 4(a1)
+; RV64-NEXT:    mv a3, a4
+; RV64-NEXT:    blt a4, a2, .LBB1_1
+; RV64-NEXT:  .LBB1_2: # %while_end
+; RV64-NEXT:    ret
 entry:
   %s = load ptr, ptr %sp
   br label %while_cond
@@ -122,27 +126,27 @@
 ; instructions. Make sure we use an offset and common base for each of the
 ; stores.
 define void @test3(ptr %t) {
-; RV32I-LABEL: test3:
-; RV32I:       # %bb.0: # %entry
-; RV32I-NEXT:    lui a1, 20
-; RV32I-NEXT:    addi a1, a1, -1920
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    li a1, 2
-; RV32I-NEXT:    sw a1, 4(a0)
-; RV32I-NEXT:    li a1, 3
-; RV32I-NEXT:    sw a1, 8(a0)
-; RV32I-NEXT:    ret
+; RV32-LABEL: test3:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    lui a1, 20
+; RV32-NEXT:    addi a1, a1, -1920
+; RV32-NEXT:    add a0, a0, a1
+; RV32-NEXT:    li a1, 2
+; RV32-NEXT:    sw a1, 4(a0)
+; RV32-NEXT:    li a1, 3
+; RV32-NEXT:    sw a1, 8(a0)
+; RV32-NEXT:    ret
 ;
-; RV64I-LABEL: test3:
-; RV64I:       # %bb.0: # %entry
-; RV64I-NEXT:    lui a1, 20
-; RV64I-NEXT:    addiw a1, a1, -1920
-; RV64I-NEXT:    add a0, a0, a1
-; RV64I-NEXT:    li a1, 2
-; RV64I-NEXT:    sw a1, 4(a0)
-; RV64I-NEXT:    li a1, 3
-; RV64I-NEXT:    sw a1, 8(a0)
-; RV64I-NEXT:    ret
+; RV64-LABEL: test3:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    lui a1, 20
+; RV64-NEXT:    addiw a1, a1, -1920
+; RV64-NEXT:    add a0, a0, a1
+; RV64-NEXT:    li a1, 2
+; RV64-NEXT:    sw a1, 4(a0)
+; RV64-NEXT:    li a1, 3
+; RV64-NEXT:    sw a1, 8(a0)
+; RV64-NEXT:    ret
 entry:
   %splitgep = getelementptr i8, ptr %t, i64 80000
   %0 = getelementptr i8, ptr %splitgep, i64 4
@@ -154,27 +158,15 @@
 
 ; Test from PR62734.
 define void @test4(ptr %dest) {
-; RV32I-LABEL: test4:
-; RV32I:       # %bb.0:
-; RV32I-NEXT:    addi a0, a0, 2047
-; RV32I-NEXT:    addi a1, a0, 1
-; RV32I-NEXT:    li a2, 1
-; RV32I-NEXT:    sb a2, 1(a0)
-; RV32I-NEXT:    sb a2, 1(a1)
-; RV32I-NEXT:    sb a2, 2(a1)
-; RV32I-NEXT:    sb a2, 3(a1)
-; RV32I-NEXT:    ret
-;
-; RV64I-LABEL: test4:
-; RV64I:       # %bb.0:
-; RV64I-NEXT:    addi a0, a0, 2047
-; RV64I-NEXT:    addi a1, a0, 1
-; RV64I-NEXT:    li a2, 1
-; RV64I-NEXT:    sb a2, 1(a0)
-; RV64I-NEXT:    sb a2, 1(a1)
-; RV64I-NEXT:    sb a2, 2(a1)
-; RV64I-NEXT:    sb a2, 3(a1)
-; RV64I-NEXT:    ret
+; CHECK-LABEL: test4:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a0, a0, 2047
+; CHECK-NEXT:    li a1, 1
+; CHECK-NEXT:    sb a1, 1(a0)
+; CHECK-NEXT:    sb a1, 2(a0)
+; CHECK-NEXT:    sb a1, 3(a0)
+; CHECK-NEXT:    sb a1, 4(a0)
+; CHECK-NEXT:    ret
   %p1 = getelementptr i8, ptr %dest, i32 2048
   store i8 1, ptr %p1
   %p2 = getelementptr i8, ptr %dest, i32 2049
@@ -185,3 +177,147 @@
   store i8 1, ptr %p4
   ret void
 }
+
+; Don't fold addimm chains when we can benefit from compressed load/store instructions.
+define void @test5(ptr %dest) {
+; RV32I-LABEL: test5:
+; RV32I:       # %bb.0:
+; RV32I-NEXT:    addi a0, a0, 2047
+; RV32I-NEXT:    li a1, 1
+; RV32I-NEXT:    sw a1, 1953(a0)
+; RV32I-NEXT:    sw a1, 1957(a0)
+; RV32I-NEXT:    sw a1, 1961(a0)
+; RV32I-NEXT:    sw a1, 1965(a0)
+; RV32I-NEXT:    ret
+;
+; RV64I-LABEL: test5:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    addi a0, a0, 2047
+; RV64I-NEXT:    li a1, 1
+; RV64I-NEXT:    sw a1, 1953(a0)
+; RV64I-NEXT:    sw a1, 1957(a0)
+; RV64I-NEXT:    sw a1, 1961(a0)
+; RV64I-NEXT:    sw a1, 1965(a0)
+; RV64I-NEXT:    ret
+;
+; RV32C-LABEL: test5:
+; RV32C:       # %bb.0:
+; RV32C-NEXT:    addi a0, a0, 2047
+; RV32C-NEXT:    addi a0, a0, 1953
+; RV32C-NEXT:    li a1, 1
+; RV32C-NEXT:    sw a1, 0(a0)
+; RV32C-NEXT:    sw a1, 4(a0)
+; RV32C-NEXT:    sw a1, 8(a0)
+; RV32C-NEXT:    sw a1, 12(a0)
+; RV32C-NEXT:    ret
+;
+; RV64C-LABEL: test5:
+; RV64C:       # %bb.0:
+; RV64C-NEXT:    addi a0, a0, 2047
+; RV64C-NEXT:    addi a0, a0, 1953
+; RV64C-NEXT:    li a1, 1
+; RV64C-NEXT:    sw a1, 0(a0)
+; RV64C-NEXT:    sw a1, 4(a0)
+; RV64C-NEXT:    sw a1, 8(a0)
+; RV64C-NEXT:    sw a1, 12(a0)
+; RV64C-NEXT:    ret
+  %p1 = getelementptr i32, ptr %dest, i32 1000
+  store i32 1, ptr %p1
+  %p2 = getelementptr i32, ptr %dest, i32 1001
+  store i32 1, ptr %p2
+  %p3 = getelementptr i32, ptr %dest, i32 1002
+  store i32 1, ptr %p3
+  %p4 = getelementptr i32, ptr %dest, i32 1003
+  store i32 1, ptr %p4
+  ret void
+}
+
+; FIXME: We can reuse a1 to emit compressed load/store instructions.
+define void @test6(ptr %dest) {
+; RV32I-LABEL: test6:
+; RV32I:       # %bb.0:
+; RV32I-NEXT:    addi a1, a0, 2047
+; RV32I-NEXT:    li a2, 1
+; RV32I-NEXT:    sw a2, 2040(a0)
+; RV32I-NEXT:    sw a2, 2044(a0)
+; RV32I-NEXT:    sw a2, 1(a1)
+; RV32I-NEXT:    sw a2, 5(a1)
+; RV32I-NEXT:    ret
+;
+; RV64I-LABEL: test6:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    addi a1, a0, 2047
+; RV64I-NEXT:    li a2, 1
+; RV64I-NEXT:    sw a2, 2040(a0)
+; RV64I-NEXT:    sw a2, 2044(a0)
+; RV64I-NEXT:    sw a2, 1(a1)
+; RV64I-NEXT:    sw a2, 5(a1)
+; RV64I-NEXT:    ret
+;
+; RV32C-LABEL: test6:
+; RV32C:       # %bb.0:
+; RV32C-NEXT:    addi a1, a0, 2047
+; RV32C-NEXT:    addi a1, a1, 1
+; RV32C-NEXT:    li a2, 1
+; RV32C-NEXT:    sw a2, 2040(a0)
+; RV32C-NEXT:    sw a2, 2044(a0)
+; RV32C-NEXT:    sw a2, 0(a1)
+; RV32C-NEXT:    sw a2, 4(a1)
+; RV32C-NEXT:    ret
+;
+; RV64C-LABEL: test6:
+; RV64C:       # %bb.0:
+; RV64C-NEXT:    addi a1, a0, 2047
+; RV64C-NEXT:    addi a1, a1, 1
+; RV64C-NEXT:    li a2, 1
+; RV64C-NEXT:    sw a2, 2040(a0)
+; RV64C-NEXT:    sw a2, 2044(a0)
+; RV64C-NEXT:    sw a2, 0(a1)
+; RV64C-NEXT:    sw a2, 4(a1)
+; RV64C-NEXT:    ret
+  %p1 = getelementptr i32, ptr %dest, i32 510
+  store i32 1, ptr %p1
+  %p2 = getelementptr i32, ptr %dest, i32 511
+  store i32 1, ptr %p2
+  %p3 = getelementptr i32, ptr %dest, i32 512
+  store i32 1, ptr %p3
+  %p4 = getelementptr i32, ptr %dest, i32 513
+  store i32 1, ptr %p4
+  ret void
+}
+
+; Don't fold addimm chains when the offset is large.
+define void @test7(ptr %dest) {
+; RV32-LABEL: test7:
+; RV32:       # %bb.0:
+; RV32-NEXT:    lui a1, 10
+; RV32-NEXT:    addi a1, a1, -960
+; RV32-NEXT:    add a0, a0, a1
+; RV32-NEXT:    li a1, 1
+; RV32-NEXT:    sw a1, 0(a0)
+; RV32-NEXT:    sw a1, 4(a0)
+; RV32-NEXT:    sw a1, 8(a0)
+; RV32-NEXT:    sw a1, 12(a0)
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: test7:
+; RV64:       # %bb.0:
+; RV64-NEXT:    lui a1, 10
+; RV64-NEXT:    addiw a1, a1, -960
+; RV64-NEXT:    add a0, a0, a1
+; RV64-NEXT:    li a1, 1
+; RV64-NEXT:    sw a1, 0(a0)
+; RV64-NEXT:    sw a1, 4(a0)
+; RV64-NEXT:    sw a1, 8(a0)
+; RV64-NEXT:    sw a1, 12(a0)
+; RV64-NEXT:    ret
+  %p1 = getelementptr i32, ptr %dest, i32 10000
+  store i32 1, ptr %p1
+  %p2 = getelementptr i32, ptr %dest, i32 10001
+  store i32 1, ptr %p2
+  %p3 = getelementptr i32, ptr %dest, i32 10002
+  store i32 1, ptr %p3
+  %p4 = getelementptr i32, ptr %dest, i32 10003
+  store i32 1, ptr %p4
+  ret void
+}