BPF address space insn (#84410)

This commit aims to support BPF arena kernel side
[feature](https://lore.kernel.org/bpf/20240209040608.98927-1-alexei.starovoitov@gmail.com/):
- arena is a memory region accessible from both BPF program and
userspace;
- base pointers for this memory region differ between kernel and user
spaces;
- `dst_reg = addr_space_cast(src_reg, dst_addr_space, src_addr_space)`
translates src_reg, a pointer in src_addr_space to dst_reg, equivalent
pointer in dst_addr_space, {src,dst}_addr_space are immediate constants;
- number 0 is assigned to kernel address space;
- number 1 is assigned to user address space.

On the LLVM side, the goal is to make load and store operations on arena
pointers "transparent" for BPF programs:
- assume that pointers with non-zero address space are pointers to
  arena memory;
- assume that arena is identified by address space number;
- assume that address space zero corresponds to kernel address space;
- assume that every BPF-side load or store from arena is done via
pointer in user address space, thus convert base pointers using
`addr_space_cast(src_reg, 0, 1)`;

Only load, store, cmpxchg and atomicrmw IR instructions are handled by
this transformation.

For example, the following C code:

```c
   #define __as __attribute__((address_space(1)))
   void copy(int __as *from, int __as *to) { *to = *from; }
```

Compiled to the following IR:

```llvm
    define void @copy(ptr addrspace(1) %from, ptr addrspace(1) %to) {
    entry:
      %0 = load i32, ptr addrspace(1) %from, align 4
      store i32 %0, ptr addrspace(1) %to, align 4
      ret void
    }
```

Is transformed to:

```llvm
    %to2 = addrspacecast ptr addrspace(1) %to to ptr     ;; !
    %from1 = addrspacecast ptr addrspace(1) %from to ptr ;; !
    %0 = load i32, ptr %from1, align 4, !tbaa !3
    store i32 %0, ptr %to2, align 4, !tbaa !3
    ret void
```

And compiled as:

```asm
    r2 = addr_space_cast(r2, 0, 1)
    r1 = addr_space_cast(r1, 0, 1)
    r1 = *(u32 *)(r1 + 0)
    *(u32 *)(r2 + 0) = r1
    exit
```

Co-authored-by: Eduard Zingerman <eddyz87@gmail.com>

21 files changed, 615 insertions, 0 deletions

diff --git a/clang/lib/Basic/Targets/BPF.cpp b/clang/lib/Basic/Targets/BPF.cpp
index e3fbbb7..26a54f6 100644
--- a/clang/lib/Basic/Targets/BPF.cpp
+++ b/clang/lib/Basic/Targets/BPF.cpp
@@ -35,6 +35,9 @@ void BPFTargetInfo::getTargetDefines(const LangOptions &Opts,
     Builder.defineMacro("__BPF_CPU_VERSION__", "0");
     return;
   }
+
+  Builder.defineMacro("__BPF_FEATURE_ARENA_CAST");
+
   if (CPU.empty() || CPU == "generic" || CPU == "v1") {
     Builder.defineMacro("__BPF_CPU_VERSION__", "1");
     return;
diff --git a/clang/test/Preprocessor/bpf-predefined-macros.c b/clang/test/Preprocessor/bpf-predefined-macros.c
index ff4d00a..fea24d1 100644
--- a/clang/test/Preprocessor/bpf-predefined-macros.c
+++ b/clang/test/Preprocessor/bpf-predefined-macros.c
@@ -61,6 +61,9 @@ int r;
 #ifdef __BPF_FEATURE_ST
 int s;
 #endif
+#ifdef __BPF_FEATURE_ARENA_CAST
+int t;
+#endif
 
 // CHECK: int b;
 // CHECK: int c;
@@ -90,6 +93,11 @@ int s;
 // CPU_V4: int r;
 // CPU_V4: int s;
 
+// CPU_V1: int t;
+// CPU_V2: int t;
+// CPU_V3: int t;
+// CPU_V4: int t;
+
 // CPU_GENERIC: int g;
 
 // CPU_PROBE: int f;
diff --git a/llvm/lib/Target/BPF/AsmParser/BPFAsmParser.cpp b/llvm/lib/Target/BPF/AsmParser/BPFAsmParser.cpp
index 0d1eef6..3145bc3 100644
--- a/llvm/lib/Target/BPF/AsmParser/BPFAsmParser.cpp
+++ b/llvm/lib/Target/BPF/AsmParser/BPFAsmParser.cpp
@@ -271,6 +271,7 @@ public:
         .Case("xchg32_32", true)
         .Case("cmpxchg_64", true)
         .Case("cmpxchg32_32", true)
+        .Case("addr_space_cast", true)
         .Default(false);
   }
 };
diff --git a/llvm/lib/Target/BPF/BPF.h b/llvm/lib/Target/BPF/BPF.h
index 5c77d18..bbdbdbb 100644
--- a/llvm/lib/Target/BPF/BPF.h
+++ b/llvm/lib/Target/BPF/BPF.h
@@ -66,6 +66,14 @@ public:
   static bool isRequired() { return true; }
 };
 
+class BPFASpaceCastSimplifyPass
+    : public PassInfoMixin<BPFASpaceCastSimplifyPass> {
+public:
+  PreservedAnalyses run(Function &F, FunctionAnalysisManager &AM);
+
+  static bool isRequired() { return true; }
+};
+
 class BPFAdjustOptPass : public PassInfoMixin<BPFAdjustOptPass> {
 public:
   PreservedAnalyses run(Module &M, ModuleAnalysisManager &AM);
diff --git a/llvm/lib/Target/BPF/BPFASpaceCastSimplifyPass.cpp b/llvm/lib/Target/BPF/BPFASpaceCastSimplifyPass.cpp
new file mode 100644
index 0000000..f87b299
--- /dev/null
+++ b/llvm/lib/Target/BPF/BPFASpaceCastSimplifyPass.cpp
@@ -0,0 +1,92 @@
+//===-- BPFASpaceCastSimplifyPass.cpp - BPF addrspacecast simplications --===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "BPF.h"
+#include <optional>
+
+#define DEBUG_TYPE "bpf-aspace-simplify"
+
+using namespace llvm;
+
+namespace {
+
+struct CastGEPCast {
+  AddrSpaceCastInst *OuterCast;
+
+  // Match chain of instructions:
+  //   %inner = addrspacecast N->M
+  //   %gep   = getelementptr %inner, ...
+  //   %outer = addrspacecast M->N %gep
+  // Where I is %outer.
+  static std::optional<CastGEPCast> match(Value *I) {
+    auto *OuterCast = dyn_cast<AddrSpaceCastInst>(I);
+    if (!OuterCast)
+      return std::nullopt;
+    auto *GEP = dyn_cast<GetElementPtrInst>(OuterCast->getPointerOperand());
+    if (!GEP)
+      return std::nullopt;
+    auto *InnerCast = dyn_cast<AddrSpaceCastInst>(GEP->getPointerOperand());
+    if (!InnerCast)
+      return std::nullopt;
+    if (InnerCast->getSrcAddressSpace() != OuterCast->getDestAddressSpace())
+      return std::nullopt;
+    if (InnerCast->getDestAddressSpace() != OuterCast->getSrcAddressSpace())
+      return std::nullopt;
+    return CastGEPCast{OuterCast};
+  }
+
+  static PointerType *changeAddressSpace(PointerType *Ty, unsigned AS) {
+    return Ty->get(Ty->getContext(), AS);
+  }
+
+  // Assuming match(this->OuterCast) is true, convert:
+  //   (addrspacecast M->N (getelementptr (addrspacecast N->M ptr) ...))
+  // To:
+  //   (getelementptr ptr ...)
+  GetElementPtrInst *rewrite() {
+    auto *GEP = cast<GetElementPtrInst>(OuterCast->getPointerOperand());
+    auto *InnerCast = cast<AddrSpaceCastInst>(GEP->getPointerOperand());
+    unsigned AS = OuterCast->getDestAddressSpace();
+    auto *NewGEP = cast<GetElementPtrInst>(GEP->clone());
+    NewGEP->setName(GEP->getName());
+    NewGEP->insertAfter(OuterCast);
+    NewGEP->setOperand(0, InnerCast->getPointerOperand());
+    auto *GEPTy = cast<PointerType>(GEP->getType());
+    NewGEP->mutateType(changeAddressSpace(GEPTy, AS));
+    OuterCast->replaceAllUsesWith(NewGEP);
+    OuterCast->eraseFromParent();
+    if (GEP->use_empty())
+      GEP->eraseFromParent();
+    if (InnerCast->use_empty())
+      InnerCast->eraseFromParent();
+    return NewGEP;
+  }
+};
+
+} // anonymous namespace
+
+PreservedAnalyses BPFASpaceCastSimplifyPass::run(Function &F,
+                                                 FunctionAnalysisManager &AM) {
+  SmallVector<CastGEPCast, 16> WorkList;
+  bool Changed = false;
+  for (BasicBlock &BB : F) {
+    for (Instruction &I : BB)
+      if (auto It = CastGEPCast::match(&I))
+        WorkList.push_back(It.value());
+    Changed |= !WorkList.empty();
+
+    while (!WorkList.empty()) {
+      CastGEPCast InsnChain = WorkList.pop_back_val();
+      GetElementPtrInst *NewGEP = InsnChain.rewrite();
+      for (User *U : NewGEP->users())
+        if (auto It = CastGEPCast::match(U))
+          WorkList.push_back(It.value());
+    }
+  }
+  return Changed ? PreservedAnalyses::none() : PreservedAnalyses::all();
+}
diff --git a/llvm/lib/Target/BPF/BPFCheckAndAdjustIR.cpp b/llvm/lib/Target/BPF/BPFCheckAndAdjustIR.cpp
index 81effc9..edd59aa 100644
--- a/llvm/lib/Target/BPF/BPFCheckAndAdjustIR.cpp
+++ b/llvm/lib/Target/BPF/BPFCheckAndAdjustIR.cpp
@@ -14,6 +14,8 @@
 //     optimizations are done and those builtins can be removed.
 //   - remove llvm.bpf.getelementptr.and.load builtins.
 //   - remove llvm.bpf.getelementptr.and.store builtins.
+//   - for loads and stores with base addresses from non-zero address space
+//     cast base address to zero address space (support for BPF arenas).
 //
 //===----------------------------------------------------------------------===//
 
@@ -55,6 +57,7 @@ private:
   bool removeCompareBuiltin(Module &M);
   bool sinkMinMax(Module &M);
   bool removeGEPBuiltins(Module &M);
+  bool insertASpaceCasts(Module &M);
 };
 } // End anonymous namespace
 
@@ -416,11 +419,124 @@ bool BPFCheckAndAdjustIR::removeGEPBuiltins(Module &M) {
   return Changed;
 }
 
+// Wrap ToWrap with cast to address space zero:
+// - if ToWrap is a getelementptr,
+//   wrap it's base pointer instead and return a copy;
+// - if ToWrap is Instruction, insert address space cast
+//   immediately after ToWrap;
+// - if ToWrap is not an Instruction (function parameter
+//   or a global value), insert address space cast at the
+//   beginning of the Function F;
+// - use Cache to avoid inserting too many casts;
+static Value *aspaceWrapValue(DenseMap<Value *, Value *> &Cache, Function *F,
+                              Value *ToWrap) {
+  auto It = Cache.find(ToWrap);
+  if (It != Cache.end())
+    return It->getSecond();
+
+  if (auto *GEP = dyn_cast<GetElementPtrInst>(ToWrap)) {
+    Value *Ptr = GEP->getPointerOperand();
+    Value *WrappedPtr = aspaceWrapValue(Cache, F, Ptr);
+    auto *GEPTy = cast<PointerType>(GEP->getType());
+    auto *NewGEP = GEP->clone();
+    NewGEP->insertAfter(GEP);
+    NewGEP->mutateType(GEPTy->getPointerTo(0));
+    NewGEP->setOperand(GEP->getPointerOperandIndex(), WrappedPtr);
+    NewGEP->setName(GEP->getName());
+    Cache[ToWrap] = NewGEP;
+    return NewGEP;
+  }
+
+  IRBuilder IB(F->getContext());
+  if (Instruction *InsnPtr = dyn_cast<Instruction>(ToWrap))
+    IB.SetInsertPoint(*InsnPtr->getInsertionPointAfterDef());
+  else
+    IB.SetInsertPoint(F->getEntryBlock().getFirstInsertionPt());
+  auto *PtrTy = cast<PointerType>(ToWrap->getType());
+  auto *ASZeroPtrTy = PtrTy->getPointerTo(0);
+  auto *ACast = IB.CreateAddrSpaceCast(ToWrap, ASZeroPtrTy, ToWrap->getName());
+  Cache[ToWrap] = ACast;
+  return ACast;
+}
+
+// Wrap a pointer operand OpNum of instruction I
+// with cast to address space zero
+static void aspaceWrapOperand(DenseMap<Value *, Value *> &Cache, Instruction *I,
+                              unsigned OpNum) {
+  Value *OldOp = I->getOperand(OpNum);
+  if (OldOp->getType()->getPointerAddressSpace() == 0)
+    return;
+
+  Value *NewOp = aspaceWrapValue(Cache, I->getFunction(), OldOp);
+  I->setOperand(OpNum, NewOp);
+  // Check if there are any remaining users of old GEP,
+  // delete those w/o users
+  for (;;) {
+    auto *OldGEP = dyn_cast<GetElementPtrInst>(OldOp);
+    if (!OldGEP)
+      break;
+    if (!OldGEP->use_empty())
+      break;
+    OldOp = OldGEP->getPointerOperand();
+    OldGEP->eraseFromParent();
+  }
+}
+
+// Support for BPF arenas:
+// - for each function in the module M, update pointer operand of
+//   each memory access instruction (load/store/cmpxchg/atomicrmw)
+//   by casting it from non-zero address space to zero address space, e.g:
+//
+//   (load (ptr addrspace (N) %p) ...)
+//     -> (load (addrspacecast ptr addrspace (N) %p to ptr))
+//
+// - assign section with name .arena.N for globals defined in
+//   non-zero address space N
+bool BPFCheckAndAdjustIR::insertASpaceCasts(Module &M) {
+  bool Changed = false;
+  for (Function &F : M) {
+    DenseMap<Value *, Value *> CastsCache;
+    for (BasicBlock &BB : F) {
+      for (Instruction &I : BB) {
+        unsigned PtrOpNum;
+
+        if (auto *LD = dyn_cast<LoadInst>(&I))
+          PtrOpNum = LD->getPointerOperandIndex();
+        else if (auto *ST = dyn_cast<StoreInst>(&I))
+          PtrOpNum = ST->getPointerOperandIndex();
+        else if (auto *CmpXchg = dyn_cast<AtomicCmpXchgInst>(&I))
+          PtrOpNum = CmpXchg->getPointerOperandIndex();
+        else if (auto *RMW = dyn_cast<AtomicRMWInst>(&I))
+          PtrOpNum = RMW->getPointerOperandIndex();
+        else
+          continue;
+
+        aspaceWrapOperand(CastsCache, &I, PtrOpNum);
+      }
+    }
+    Changed |= !CastsCache.empty();
+  }
+  // Merge all globals within same address space into single
+  // .arena.<addr space no> section
+  for (GlobalVariable &G : M.globals()) {
+    if (G.getAddressSpace() == 0 || G.hasSection())
+      continue;
+    SmallString<16> SecName;
+    raw_svector_ostream OS(SecName);
+    OS << ".arena." << G.getAddressSpace();
+    G.setSection(SecName);
+    // Prevent having separate section for constants
+    G.setConstant(false);
+  }
+  return Changed;
+}
+
 bool BPFCheckAndAdjustIR::adjustIR(Module &M) {
   bool Changed = removePassThroughBuiltin(M);
   Changed = removeCompareBuiltin(M) || Changed;
   Changed = sinkMinMax(M) || Changed;
   Changed = removeGEPBuiltins(M) || Changed;
+  Changed = insertASpaceCasts(M) || Changed;
   return Changed;
 }
 
diff --git a/llvm/lib/Target/BPF/BPFInstrInfo.td b/llvm/lib/Target/BPF/BPFInstrInfo.td
index 82d3470..7198e94 100644
--- a/llvm/lib/Target/BPF/BPFInstrInfo.td
+++ b/llvm/lib/Target/BPF/BPFInstrInfo.td
@@ -420,6 +420,35 @@ let Predicates = [BPFHasMovsx] in {
 }
 }
 
+def ADDR_SPACE_CAST
+    : ALU_RR<BPF_ALU64, BPF_MOV, 64,
+             (outs GPR:$dst),
+             (ins GPR:$src, i64imm:$dst_as, i64imm:$src_as),
+             "$dst = addr_space_cast($src, $dst_as, $src_as)",
+             []> {
+  bits<64> dst_as;
+  bits<64> src_as;
+
+  let Inst{47-32} = 1;
+  let Inst{31-16} = dst_as{15-0};
+  let Inst{15-0} = src_as{15-0};
+}
+
+def SrcAddrSpace : SDNodeXForm<addrspacecast, [{
+  return CurDAG->getTargetConstant(
+    cast<AddrSpaceCastSDNode>(N)->getSrcAddressSpace(),
+    SDLoc(N), MVT::i64);
+}]>;
+
+def DstAddrSpace : SDNodeXForm<addrspacecast, [{
+  return CurDAG->getTargetConstant(
+    cast<AddrSpaceCastSDNode>(N)->getDestAddressSpace(),
+    SDLoc(N), MVT::i64);
+}]>;
+
+def : Pat<(addrspacecast:$this GPR:$src),
+          (ADDR_SPACE_CAST $src, (DstAddrSpace $this), (SrcAddrSpace $this))>;
+
 def FI_ri
     : TYPE_LD_ST<BPF_IMM.Value, BPF_DW.Value,
                  (outs GPR:$dst),
diff --git a/llvm/lib/Target/BPF/BPFTargetMachine.cpp b/llvm/lib/Target/BPF/BPFTargetMachine.cpp
index 08ac4b2..5f26bec 100644
--- a/llvm/lib/Target/BPF/BPFTargetMachine.cpp
+++ b/llvm/lib/Target/BPF/BPFTargetMachine.cpp
@@ -121,6 +121,10 @@ void BPFTargetMachine::registerPassBuilderCallbacks(
           FPM.addPass(BPFPreserveStaticOffsetPass(false));
           return true;
         }
+        if (PassName == "bpf-aspace-simplify") {
+          FPM.addPass(BPFASpaceCastSimplifyPass());
+          return true;
+        }
         return false;
       });
   PB.registerPipelineStartEPCallback(
@@ -135,6 +139,7 @@ void BPFTargetMachine::registerPassBuilderCallbacks(
   PB.registerPeepholeEPCallback([=](FunctionPassManager &FPM,
                                     OptimizationLevel Level) {
     FPM.addPass(SimplifyCFGPass(SimplifyCFGOptions().hoistCommonInsts(true)));
+    FPM.addPass(BPFASpaceCastSimplifyPass());
   });
   PB.registerScalarOptimizerLateEPCallback(
       [=](FunctionPassManager &FPM, OptimizationLevel Level) {
diff --git a/llvm/lib/Target/BPF/CMakeLists.txt b/llvm/lib/Target/BPF/CMakeLists.txt
index d88e7ad..cb21ed0 100644
--- a/llvm/lib/Target/BPF/CMakeLists.txt
+++ b/llvm/lib/Target/BPF/CMakeLists.txt
@@ -24,6 +24,7 @@ add_llvm_target(BPFCodeGen
   BPFAbstractMemberAccess.cpp
   BPFAdjustOpt.cpp
   BPFAsmPrinter.cpp
+  BPFASpaceCastSimplifyPass.cpp
   BPFCheckAndAdjustIR.cpp
   BPFFrameLowering.cpp
   BPFInstrInfo.cpp
diff --git a/llvm/test/CodeGen/BPF/addr-space-auto-casts.ll b/llvm/test/CodeGen/BPF/addr-space-auto-casts.ll
new file mode 100644
index 0000000..08e11e8
--- /dev/null
+++ b/llvm/test/CodeGen/BPF/addr-space-auto-casts.ll
@@ -0,0 +1,78 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4
+; RUN: opt --bpf-check-and-opt-ir -S -mtriple=bpf-pc-linux < %s | FileCheck %s
+
+define void @simple_store(ptr addrspace(272) %foo) {
+; CHECK-LABEL: define void @simple_store(
+; CHECK-SAME: ptr addrspace(272) [[FOO:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[FOO1:%.*]] = addrspacecast ptr addrspace(272) [[FOO]] to ptr
+; CHECK-NEXT:    [[ADD_PTR2:%.*]] = getelementptr inbounds i8, ptr [[FOO1]], i64 16
+; CHECK-NEXT:    store volatile i32 57005, ptr [[ADD_PTR2]], align 4
+; CHECK-NEXT:    [[ADD_PTR13:%.*]] = getelementptr inbounds i8, ptr [[FOO1]], i64 12
+; CHECK-NEXT:    store volatile i32 48879, ptr [[ADD_PTR13]], align 4
+; CHECK-NEXT:    ret void
+;
+entry:
+  %add.ptr = getelementptr inbounds i8, ptr addrspace(272) %foo, i64 16
+  store volatile i32 57005, ptr addrspace(272) %add.ptr, align 4
+  %add.ptr1 = getelementptr inbounds i8, ptr addrspace(272) %foo, i64 12
+  store volatile i32 48879, ptr addrspace(272) %add.ptr1, align 4
+  ret void
+}
+
+define void @separate_addr_store(ptr addrspace(272) %foo, ptr addrspace(272) %bar) {
+; CHECK-LABEL: define void @separate_addr_store(
+; CHECK-SAME: ptr addrspace(272) [[FOO:%.*]], ptr addrspace(272) [[BAR:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[BAR3:%.*]] = addrspacecast ptr addrspace(272) [[BAR]] to ptr
+; CHECK-NEXT:    [[FOO1:%.*]] = addrspacecast ptr addrspace(272) [[FOO]] to ptr
+; CHECK-NEXT:    [[ADD_PTR2:%.*]] = getelementptr inbounds i8, ptr [[FOO1]], i64 16
+; CHECK-NEXT:    store volatile i32 57005, ptr [[ADD_PTR2]], align 4
+; CHECK-NEXT:    [[ADD_PTR14:%.*]] = getelementptr inbounds i8, ptr [[BAR3]], i64 12
+; CHECK-NEXT:    store volatile i32 48879, ptr [[ADD_PTR14]], align 4
+; CHECK-NEXT:    ret void
+;
+entry:
+  %add.ptr = getelementptr inbounds i8, ptr addrspace(272) %foo, i64 16
+  store volatile i32 57005, ptr addrspace(272) %add.ptr, align 4
+  %add.ptr1 = getelementptr inbounds i8, ptr addrspace(272) %bar, i64 12
+  store volatile i32 48879, ptr addrspace(272) %add.ptr1, align 4
+  ret void
+}
+
+define i32 @simple_load(ptr addrspace(272) %foo) {
+; CHECK-LABEL: define i32 @simple_load(
+; CHECK-SAME: ptr addrspace(272) [[FOO:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[FOO1:%.*]] = addrspacecast ptr addrspace(272) [[FOO]] to ptr
+; CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr [[FOO1]], align 4
+; CHECK-NEXT:    ret i32 [[TMP0]]
+;
+entry:
+  %0 = load i32, ptr addrspace(272) %foo, align 4
+  ret i32 %0
+}
+
+define { i32, i1 } @simple_cmpxchg(ptr addrspace(1) %i) {
+; CHECK-LABEL: define { i32, i1 } @simple_cmpxchg(
+; CHECK-SAME: ptr addrspace(1) [[I:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[I1:%.*]] = addrspacecast ptr addrspace(1) [[I]] to ptr
+; CHECK-NEXT:    [[A:%.*]] = cmpxchg ptr [[I1]], i32 7, i32 42 monotonic monotonic, align 4
+; CHECK-NEXT:    ret { i32, i1 } [[A]]
+;
+entry:
+  %a = cmpxchg ptr addrspace(1) %i, i32 7, i32 42 monotonic monotonic, align 4
+  ret { i32, i1 } %a
+}
+
+define void @simple_atomicrmw(ptr addrspace(1) %p) {
+; CHECK-LABEL: define void @simple_atomicrmw(
+; CHECK-SAME: ptr addrspace(1) [[P:%.*]]) {
+; CHECK-NEXT:    [[P1:%.*]] = addrspacecast ptr addrspace(1) [[P]] to ptr
+; CHECK-NEXT:    [[A:%.*]] = atomicrmw add ptr [[P1]], i64 42 monotonic, align 8
+; CHECK-NEXT:    ret void
+;
+  %a = atomicrmw add ptr addrspace(1) %p, i64 42 monotonic, align 8
+  ret void
+}
diff --git a/llvm/test/CodeGen/BPF/addr-space-cast.ll b/llvm/test/CodeGen/BPF/addr-space-cast.ll
new file mode 100644
index 0000000..ad2860d
--- /dev/null
+++ b/llvm/test/CodeGen/BPF/addr-space-cast.ll
@@ -0,0 +1,22 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 4
+; RUN: llc -march=bpfel -mcpu=v4 -filetype=asm -show-mc-encoding < %s | FileCheck %s
+
+define ptr addrspace(1) @foo(ptr %p) {
+; CHECK-LABEL: foo:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    r0 = addr_space_cast(r1, 1, 0) # encoding: [0xbf,0x10,0x01,0x00,0x00,0x00,0x01,0x00]
+; CHECK-NEXT:    exit                           # encoding: [0x95,0x00,0x00,0x00,0x00,0x00,0x00,0x00]
+entry:
+  %0 = addrspacecast ptr %p to ptr addrspace(1)
+  ret ptr addrspace(1) %0
+}
+
+define ptr @bar(ptr addrspace(1) %p) {
+; CHECK-LABEL: bar:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    r0 = addr_space_cast(r1, 0, 1) # encoding: [0xbf,0x10,0x01,0x00,0x01,0x00,0x00,0x00]
+; CHECK-NEXT:    exit                           # encoding: [0x95,0x00,0x00,0x00,0x00,0x00,0x00,0x00]
+entry:
+  %0 = addrspacecast ptr addrspace(1) %p to ptr
+  ret ptr %0
+}
diff --git a/llvm/test/CodeGen/BPF/addr-space-gep-chain.ll b/llvm/test/CodeGen/BPF/addr-space-gep-chain.ll
new file mode 100644
index 0000000..3ac85fb
--- /dev/null
+++ b/llvm/test/CodeGen/BPF/addr-space-gep-chain.ll
@@ -0,0 +1,25 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4
+; RUN: opt --bpf-check-and-opt-ir -S -mtriple=bpf-pc-linux < %s | FileCheck %s
+
+define void @test(ptr addrspace(1) %p) {
+; CHECK-LABEL: define void @test(
+; CHECK-SAME:    ptr addrspace(1) [[P:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[P1:%.*]] = addrspacecast ptr addrspace(1) [[P]] to ptr
+; CHECK-NEXT:    [[A2:%.*]] = getelementptr inbounds i8, ptr [[P1]], i64 8
+; CHECK-NEXT:    [[B3:%.*]] = getelementptr inbounds i8, ptr [[A2]], i64 16
+; CHECK-NEXT:    [[C4:%.*]] = getelementptr inbounds i8, ptr [[B3]], i64 24
+; CHECK-NEXT:    [[D5:%.*]] = getelementptr inbounds i8, ptr [[C4]], i64 32
+; CHECK-NEXT:    store i64 11, ptr [[C4]], align 8
+; CHECK-NEXT:    store i64 22, ptr [[D5]], align 8
+; CHECK-NEXT:    ret void
+;
+entry:
+  %a = getelementptr inbounds i8, ptr addrspace(1) %p, i64 8
+  %b = getelementptr inbounds i8, ptr addrspace(1) %a, i64 16
+  %c = getelementptr inbounds i8, ptr addrspace(1) %b, i64 24
+  %d = getelementptr inbounds i8, ptr addrspace(1) %c, i64 32
+  store i64 11, ptr addrspace(1) %c, align 8
+  store i64 22, ptr addrspace(1) %d, align 8
+  ret void
+}
diff --git a/llvm/test/CodeGen/BPF/addr-space-globals.ll b/llvm/test/CodeGen/BPF/addr-space-globals.ll
new file mode 100644
index 0000000..878ba0d
--- /dev/null
+++ b/llvm/test/CodeGen/BPF/addr-space-globals.ll
@@ -0,0 +1,30 @@
+; RUN: llc -march=bpfel -mcpu=v4 < %s | FileCheck %s
+
+; Generated from the following C code:
+;
+;   #define __as __attribute__((address_space(272)))
+;   __as const char a[2] = {1,2};
+;   __as char b[2] = {3,4};
+;   __as char c[2];
+;
+; Using the following command:
+;
+;   clang --target=bpf -O2 -S -emit-llvm -o t.ll t.c
+
+
+@a = dso_local local_unnamed_addr addrspace(272) constant [2 x i8] [i8 1, i8 2], align 1
+@b = dso_local local_unnamed_addr addrspace(272) global [2 x i8] [i8 3, i8 4], align 1
+@c = dso_local local_unnamed_addr addrspace(272) global [2 x i8] zeroinitializer, align 1
+
+; Verify that a,b,c reside in the same section
+
+; CHECK:     .section .arena.272,"aw",@progbits
+; CHECK-NOT: .section
+; CHECK:     .globl  a
+; CHECK:     .ascii  "\001\002"
+; CHECK-NOT: .section
+; CHECK:     .globl  b
+; CHECK:     .ascii  "\003\004"
+; CHECK-NOT: .section
+; CHECK:     .globl  c
+; CHECK:     .zero   2
diff --git a/llvm/test/CodeGen/BPF/addr-space-globals2.ll b/llvm/test/CodeGen/BPF/addr-space-globals2.ll
new file mode 100644
index 0000000..d1e2318
--- /dev/null
+++ b/llvm/test/CodeGen/BPF/addr-space-globals2.ll
@@ -0,0 +1,25 @@
+; RUN: llc -march=bpfel -mcpu=v4 < %s | FileCheck %s
+
+; Generated from the following C code:
+;
+;   __attribute__((address_space(1))) char a[2] = {1,2};
+;   __attribute__((address_space(2))) char b[2] = {3,4};
+;
+; Using the following command:
+;
+;   clang --target=bpf -O2 -S -emit-llvm -o t.ll t.c
+
+@a = dso_local local_unnamed_addr addrspace(1) global [2 x i8] [i8 1, i8 2], align 1
+@b = dso_local local_unnamed_addr addrspace(2) global [2 x i8] [i8 3, i8 4], align 1
+
+; Verify that a,b reside in separate sections
+
+; CHECK:     .section .arena.1,"aw",@progbits
+; CHECK-NOT: .section
+; CHECK:     .globl  a
+; CHECK:     .ascii  "\001\002"
+
+; CHECK:     .section .arena.2,"aw",@progbits
+; CHECK-NOT: .section
+; CHECK:     .globl  b
+; CHECK:     .ascii  "\003\004"
diff --git a/llvm/test/CodeGen/BPF/addr-space-phi.ll b/llvm/test/CodeGen/BPF/addr-space-phi.ll
new file mode 100644
index 0000000..6d28b07
--- /dev/null
+++ b/llvm/test/CodeGen/BPF/addr-space-phi.ll
@@ -0,0 +1,53 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4
+; RUN: opt --bpf-check-and-opt-ir -S -mtriple=bpf-pc-linux < %s | FileCheck %s
+
+; Generated from the following C code:
+;
+;   #define __uptr __attribute__((address_space(1)))
+;
+;   extern int __uptr *magic1();
+;   extern int __uptr *magic2();
+;
+;   void test(long i) {
+;     int __uptr *a;
+;
+;     if (i > 42)
+;       a = magic1();
+;     else
+;       a = magic2();
+;     a[5] = 7;
+;   }
+;
+; Using the following command:
+;
+;   clang --target=bpf -O2 -S -emit-llvm -o t.ll t.c
+
+define void @test(i64 noundef %i) {
+; CHECK:       if.end:
+; CHECK-NEXT:    [[A_0:%.*]] = phi ptr addrspace(1)
+; CHECK-NEXT:    [[A_01:%.*]] = addrspacecast ptr addrspace(1) [[A_0]] to ptr
+; CHECK-NEXT:    [[ARRAYIDX2:%.*]] = getelementptr inbounds i32, ptr [[A_01]], i64 5
+; CHECK-NEXT:    store i32 7, ptr [[ARRAYIDX2]], align 4
+; CHECK-NEXT:    ret void
+;
+entry:
+  %cmp = icmp sgt i64 %i, 42
+  br i1 %cmp, label %if.then, label %if.else
+
+if.then:                                          ; preds = %entry
+  %call = tail call ptr addrspace(1) @magic1()
+  br label %if.end
+
+if.else:                                          ; preds = %entry
+  %call1 = tail call ptr addrspace(1) @magic2()
+  br label %if.end
+
+if.end:                                           ; preds = %if.else, %if.then
+  %a.0 = phi ptr addrspace(1) [ %call, %if.then ], [ %call1, %if.else ]
+  %arrayidx = getelementptr inbounds i32, ptr addrspace(1) %a.0, i64 5
+  store i32 7, ptr addrspace(1) %arrayidx, align 4
+  ret void
+}
+
+declare ptr addrspace(1) @magic1(...)
+declare ptr addrspace(1) @magic2(...)
diff --git a/llvm/test/CodeGen/BPF/addr-space-simplify-1.ll b/llvm/test/CodeGen/BPF/addr-space-simplify-1.ll
new file mode 100644
index 0000000..32d6728
--- /dev/null
+++ b/llvm/test/CodeGen/BPF/addr-space-simplify-1.ll
@@ -0,0 +1,19 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4
+; RUN: opt -passes=bpf-aspace-simplify -mtriple=bpf-pc-linux -S < %s | FileCheck %s
+
+; Check that bpf-aspace-simplify pass removes unnecessary (for BPF)
+; address space casts for cast M->N -> GEP -> cast N->M chain.
+
+define dso_local ptr addrspace(1) @test (ptr addrspace(1) %p) {
+; CHECK-LABEL: define dso_local ptr addrspace(1) @test(
+; CHECK-SAME: ptr addrspace(1) [[P:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[B1:%.*]] = getelementptr inbounds i8, ptr addrspace(1) [[P]], i64 8
+; CHECK-NEXT:    ret ptr addrspace(1) [[B1]]
+;
+  entry:
+  %a = addrspacecast ptr addrspace(1) %p to ptr
+  %b = getelementptr inbounds i8, ptr %a, i64 8
+  %c = addrspacecast ptr %b to ptr addrspace(1)
+  ret ptr addrspace(1) %c
+}
diff --git a/llvm/test/CodeGen/BPF/addr-space-simplify-2.ll b/llvm/test/CodeGen/BPF/addr-space-simplify-2.ll
new file mode 100644
index 0000000..a296555
--- /dev/null
+++ b/llvm/test/CodeGen/BPF/addr-space-simplify-2.ll
@@ -0,0 +1,21 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4
+; RUN: opt -passes=bpf-aspace-simplify -mtriple=bpf-pc-linux -S < %s | FileCheck %s
+
+; Check that bpf-aspace-simplify pass does not change
+; chain 'cast M->N -> GEP -> cast N->K'.
+
+define dso_local ptr addrspace(2) @test (ptr addrspace(1) %p) {
+; CHECK-LABEL: define dso_local ptr addrspace(2) @test(
+; CHECK-SAME: ptr addrspace(1) [[P:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[A:%.*]] = addrspacecast ptr addrspace(1) [[P]] to ptr
+; CHECK-NEXT:    [[B:%.*]] = getelementptr inbounds i8, ptr [[A]], i64 8
+; CHECK-NEXT:    [[C:%.*]] = addrspacecast ptr [[B]] to ptr addrspace(2)
+; CHECK-NEXT:    ret ptr addrspace(2) [[C]]
+;
+  entry:
+  %a = addrspacecast ptr addrspace(1) %p to ptr
+  %b = getelementptr inbounds i8, ptr %a, i64 8
+  %c = addrspacecast ptr %b to ptr addrspace(2)
+  ret ptr addrspace(2) %c
+}
diff --git a/llvm/test/CodeGen/BPF/addr-space-simplify-3.ll b/llvm/test/CodeGen/BPF/addr-space-simplify-3.ll
new file mode 100644
index 0000000..a7736c4
--- /dev/null
+++ b/llvm/test/CodeGen/BPF/addr-space-simplify-3.ll
@@ -0,0 +1,26 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4
+; RUN: opt -passes=bpf-aspace-simplify -mtriple=bpf-pc-linux -S < %s | FileCheck %s
+
+; Check that when bpf-aspace-simplify pass modifies chain
+; 'cast M->N -> GEP -> cast N->M' it does not remove GEP,
+; when that GEP is used by some other instruction.
+
+define dso_local ptr addrspace(1) @test (ptr addrspace(1) %p) {
+; CHECK-LABEL: define dso_local ptr addrspace(1) @test(
+; CHECK-SAME: ptr addrspace(1) [[P:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[A:%.*]] = addrspacecast ptr addrspace(1) [[P]] to ptr
+; CHECK-NEXT:    [[B:%.*]] = getelementptr inbounds i8, ptr [[A]], i64 8
+; CHECK-NEXT:    [[B1:%.*]] = getelementptr inbounds i8, ptr addrspace(1) [[P]], i64 8
+; CHECK-NEXT:    call void @sink(ptr [[B]])
+; CHECK-NEXT:    ret ptr addrspace(1) [[B1]]
+;
+  entry:
+  %a = addrspacecast ptr addrspace(1) %p to ptr
+  %b = getelementptr inbounds i8, ptr %a, i64 8
+  %c = addrspacecast ptr %b to ptr addrspace(1)
+  call void @sink(ptr %b)
+  ret ptr addrspace(1) %c
+}
+
+declare dso_local void @sink(ptr)
diff --git a/llvm/test/CodeGen/BPF/addr-space-simplify-4.ll b/llvm/test/CodeGen/BPF/addr-space-simplify-4.ll
new file mode 100644
index 0000000..b2c384b
--- /dev/null
+++ b/llvm/test/CodeGen/BPF/addr-space-simplify-4.ll
@@ -0,0 +1,21 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4
+; RUN: opt -passes=bpf-aspace-simplify -mtriple=bpf-pc-linux -S < %s | FileCheck %s
+
+; Check that bpf-aspace-simplify pass simplifies chain
+; 'cast K->M -> cast M->N -> GEP -> cast N->M -> cast M->K'.
+
+define dso_local ptr addrspace(2) @test (ptr addrspace(2) %p) {
+; CHECK-LABEL: define dso_local ptr addrspace(2) @test(
+; CHECK-SAME: ptr addrspace(2) [[P:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[C12:%.*]] = getelementptr inbounds i8, ptr addrspace(2) [[P]], i64 8
+; CHECK-NEXT:    ret ptr addrspace(2) [[C12]]
+;
+  entry:
+  %a = addrspacecast ptr addrspace(2) %p to ptr addrspace(1)
+  %b = addrspacecast ptr addrspace(1) %a to ptr
+  %c = getelementptr inbounds i8, ptr %b, i64 8
+  %d = addrspacecast ptr %c to ptr addrspace(1)
+  %e = addrspacecast ptr addrspace (1) %d to ptr addrspace(2)
+  ret ptr addrspace(2) %e
+}
diff --git a/llvm/test/CodeGen/BPF/addr-space-simplify-5.ll b/llvm/test/CodeGen/BPF/addr-space-simplify-5.ll
new file mode 100644
index 0000000..b62d253
--- /dev/null
+++ b/llvm/test/CodeGen/BPF/addr-space-simplify-5.ll
@@ -0,0 +1,25 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4
+; RUN: opt -passes=bpf-aspace-simplify -mtriple=bpf-pc-linux -S < %s | FileCheck %s
+
+; Check that bpf-aspace-simplify pass removes unnecessary (for BPF)
+; address space casts for cast M->N -> GEP -> cast N->M chain,
+; where chain is split between several BBs.
+
+define dso_local ptr addrspace(1) @test (ptr addrspace(1) %p) {
+; CHECK-LABEL: define dso_local ptr addrspace(1) @test(
+; CHECK-SAME: ptr addrspace(1) [[P:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    br label [[EXIT:%.*]]
+; CHECK:       exit:
+; CHECK-NEXT:    [[B1:%.*]] = getelementptr inbounds i8, ptr addrspace(1) [[P]], i64 8
+; CHECK-NEXT:    ret ptr addrspace(1) [[B1]]
+;
+entry:
+  %a = addrspacecast ptr addrspace(1) %p to ptr
+  %b = getelementptr inbounds i8, ptr %a, i64 8
+  br label %exit
+
+exit:
+  %c = addrspacecast ptr %b to ptr addrspace(1)
+  ret ptr addrspace(1) %c
+}
diff --git a/llvm/test/CodeGen/BPF/assembler-disassembler.s b/llvm/test/CodeGen/BPF/assembler-disassembler.s
index 2bc7421..991d6ed 100644
--- a/llvm/test/CodeGen/BPF/assembler-disassembler.s
+++ b/llvm/test/CodeGen/BPF/assembler-disassembler.s
@@ -289,3 +289,10 @@ r0 = *(u32*)skb[42]
 r0 = *(u8*)skb[r1]
 r0 = *(u16*)skb[r1]
 r0 = *(u32*)skb[r1]
+
+// CHECK: bf 10 01 00 01 00 00 00	r0 = addr_space_cast(r1, 0x0, 0x1)
+// CHECK: bf 21 01 00 00 00 01 00	r1 = addr_space_cast(r2, 0x1, 0x0)
+// CHECK: bf 43 01 00 2a 00 07 00	r3 = addr_space_cast(r4, 0x7, 0x2a)
+r0 = addr_space_cast(r1, 0, 1)
+r1 = addr_space_cast(r2, 1, 0)
+r3 = addr_space_cast(r4, 7, 42)