diff --git a/llvm/docs/LangRef.rst b/llvm/docs/LangRef.rst
index 0e87a8e2ace0e..f169ab941c457 100644
--- a/llvm/docs/LangRef.rst
+++ b/llvm/docs/LangRef.rst
@@ -14139,6 +14139,41 @@ Semantics:
 """"""""""
 See description of '``llvm.instrprof.increment``' intrinsic.
 
+'``llvm.instrprof.callsite``' Intrinsic
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Syntax:
+"""""""
+
+::
+
+      declare void @llvm.instrprof.callsite(ptr <name>, i64 <hash>,
+                                            i32 <num-counters>,
+                                            i32 <index>, ptr <callsite>)
+
+Overview:
+"""""""""
+
+.. FIXME: detail when it's emitted once the support is added
+
+The '``llvm.instrprof.callsite``' intrinsic should be emitted before a callsite
+that's not to a "fake" callee (like another intrinsic or asm).
+
+Arguments:
+""""""""""
+The first 4 arguments are similar to ``llvm.instrprof.increment``. The indexing
+is specific to callsites, meaning callsites are indexed from 0, independent from
+the indexes used by the other intrinsics (such as 
+``llvm.instrprof.increment[.step]``).
+
+The last argument is the called value of the callsite this intrinsic precedes.
+
+Semantics:
+""""""""""
+.. FIXME: detail how when the lowering pass is added.
+
+This is lowered by contextual profiling.
+
 '``llvm.instrprof.timestamp``' Intrinsic
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 
diff --git a/llvm/include/llvm/IR/IntrinsicInst.h b/llvm/include/llvm/IR/IntrinsicInst.h
index 4f22720f1c558..081e72cc82b2c 100644
--- a/llvm/include/llvm/IR/IntrinsicInst.h
+++ b/llvm/include/llvm/IR/IntrinsicInst.h
@@ -1435,6 +1435,7 @@ class InstrProfInstBase : public IntrinsicInst {
     case Intrinsic::instrprof_cover:
     case Intrinsic::instrprof_increment:
     case Intrinsic::instrprof_increment_step:
+    case Intrinsic::instrprof_callsite:
     case Intrinsic::instrprof_timestamp:
     case Intrinsic::instrprof_value_profile:
       return true;
@@ -1519,6 +1520,21 @@ class InstrProfIncrementInstStep : public InstrProfIncrementInst {
   }
 };
 
+/// This represents the llvm.instrprof.callsite intrinsic.
+/// It is structurally like the increment or step counters, hence the
+/// inheritance relationship, albeit somewhat tenuous (it's not 'counting' per
+/// se)
+class InstrProfCallsite : public InstrProfCntrInstBase {
+public:
+  static bool classof(const IntrinsicInst *I) {
+    return I->getIntrinsicID() == Intrinsic::instrprof_callsite;
+  }
+  static bool classof(const Value *V) {
+    return isa<IntrinsicInst>(V) && classof(cast<IntrinsicInst>(V));
+  }
+  Value *getCallee() const;
+};
+
 /// This represents the llvm.instrprof.timestamp intrinsic.
 class InstrProfTimestampInst : public InstrProfCntrInstBase {
 public:
diff --git a/llvm/include/llvm/IR/Intrinsics.td b/llvm/include/llvm/IR/Intrinsics.td
index 1d20f7e1b1985..a14e9dedef8c9 100644
--- a/llvm/include/llvm/IR/Intrinsics.td
+++ b/llvm/include/llvm/IR/Intrinsics.td
@@ -914,6 +914,11 @@ def int_instrprof_increment_step : Intrinsic<[],
                                         [llvm_ptr_ty, llvm_i64_ty,
                                          llvm_i32_ty, llvm_i32_ty, llvm_i64_ty]>;
 
+// Callsite instrumentation for contextual profiling
+def int_instrprof_callsite : Intrinsic<[],
+                                        [llvm_ptr_ty, llvm_i64_ty,
+                                         llvm_i32_ty, llvm_i32_ty, llvm_ptr_ty]>;
+
 // A timestamp for instrumentation based profiling.
 def int_instrprof_timestamp : Intrinsic<[], [llvm_ptr_ty, llvm_i64_ty,
                                              llvm_i32_ty, llvm_i32_ty]>;
diff --git a/llvm/lib/IR/IntrinsicInst.cpp b/llvm/lib/IR/IntrinsicInst.cpp
index 89403e1d7fcb4..8faeb4e9951f7 100644
--- a/llvm/lib/IR/IntrinsicInst.cpp
+++ b/llvm/lib/IR/IntrinsicInst.cpp
@@ -291,6 +291,12 @@ Value *InstrProfIncrementInst::getStep() const {
   return ConstantInt::get(Type::getInt64Ty(Context), 1);
 }
 
+Value *InstrProfCallsite::getCallee() const {
+  if (isa<InstrProfCallsite>(this))
+    return getArgOperand(4);
+  return nullptr;
+}
+
 std::optional<RoundingMode> ConstrainedFPIntrinsic::getRoundingMode() const {
   unsigned NumOperands = arg_size();
   Metadata *MD = nullptr;
diff --git a/llvm/unittests/IR/IntrinsicsTest.cpp b/llvm/unittests/IR/IntrinsicsTest.cpp
index 3fa4b2cf73b6b..dddd2f73d4446 100644
--- a/llvm/unittests/IR/IntrinsicsTest.cpp
+++ b/llvm/unittests/IR/IntrinsicsTest.cpp
@@ -81,6 +81,7 @@ TEST_F(IntrinsicsTest, InstrProfInheritance) {
   __ISA(InstrProfCoverInst, InstrProfCntrInstBase);
   __ISA(InstrProfIncrementInst, InstrProfCntrInstBase);
   __ISA(InstrProfIncrementInstStep, InstrProfIncrementInst);
+  __ISA(InstrProfCallsite, InstrProfCntrInstBase);
   __ISA(InstrProfTimestampInst, InstrProfCntrInstBase);
   __ISA(InstrProfValueProfileInst, InstrProfCntrInstBase);
   __ISA(InstrProfMCDCBitmapInstBase, InstrProfInstBase);
@@ -94,6 +95,7 @@ TEST_F(IntrinsicsTest, InstrProfInheritance) {
           {Intrinsic::instrprof_cover, isInstrProfCoverInst},
           {Intrinsic::instrprof_increment, isInstrProfIncrementInst},
           {Intrinsic::instrprof_increment_step, isInstrProfIncrementInstStep},
+          {Intrinsic::instrprof_callsite, isInstrProfCallsite},
           {Intrinsic::instrprof_mcdc_condbitmap_update,
            isInstrProfMCDCCondBitmapUpdate},
           {Intrinsic::instrprof_mcdc_parameters,