Minor extension to llvm.experimental.patchpoint: don't require a call.

If a null call target is provided, don't emit a dummy call. This
allows the runtime to reserve as little nop space as it needs without
the requirement of emitting a call.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@194676 91177308-0d34-0410-b5e6-96231b3b80d8

3 files changed, 33 insertions, 12 deletions

diff --git a/lib/CodeGen/SelectionDAG/SelectionDAGBuilder.cpp b/lib/CodeGen/SelectionDAG/SelectionDAGBuilder.cpp
index 98e067ebd4..e80b63021f 100644
--- a/lib/CodeGen/SelectionDAG/SelectionDAGBuilder.cpp
+++ b/lib/CodeGen/SelectionDAG/SelectionDAGBuilder.cpp
@@ -6826,7 +6826,7 @@ void SelectionDAGBuilder::visitStackmap(const CallInst &CI) {
 /// \brief Lower llvm.experimental.patchpoint directly to its target opcode.
 void SelectionDAGBuilder::visitPatchpoint(const CallInst &CI) {
   // void|i64 @llvm.experimental.patchpoint.void|i64(i32 <id>,
-  //                                                 i32 <numNopBytes>,
+  //                                                 i32 <numBytes>,
   //                                                 i8* <target>,
   //                                                 i32 <numArgs>,
   //                                                 [Args...],
diff --git a/lib/Target/X86/X86MCInstLower.cpp b/lib/Target/X86/X86MCInstLower.cpp
index c718b1cceb..5c3f9af8bf 100644
--- a/lib/Target/X86/X86MCInstLower.cpp
+++ b/lib/Target/X86/X86MCInstLower.cpp
@@ -765,6 +765,8 @@ static void LowerSTACKMAP(MCStreamer &OutStreamer,
     OutStreamer.EmitInstruction(MCInstBuilder(X86::NOOP));
 }
 
+// Lower a patchpoint of the form:
+// [<def>], <id>, <numBytes>, <target>, <numArgs>
 static void LowerPATCHPOINT(MCStreamer &OutStreamer,
                             X86MCInstLower &MCInstLowering,
                             StackMaps &SM,
@@ -813,17 +815,20 @@ static void LowerPATCHPOINT(MCStreamer &OutStreamer,
                     getStackMapEndMOP(MI.operands_begin(), MI.operands_end()),
                     isAnyRegCC && hasDef);
 
-  // Emit MOV to materialize the target address and the CALL to target.
-  // This is encoded with 12-13 bytes, depending on which register is used.
-  // We conservatively assume that it is 12 bytes and emit in worst case one
-  // extra NOP byte.
-  unsigned EncodedBytes = 12;
-  OutStreamer.EmitInstruction(MCInstBuilder(X86::MOV64ri)
-                              .addReg(MI.getOperand(ScratchIdx).getReg())
-                              .addImm(MI.getOperand(StartIdx + 2).getImm()));
-  OutStreamer.EmitInstruction(MCInstBuilder(X86::CALL64r)
-                              .addReg(MI.getOperand(ScratchIdx).getReg()));
-
+  unsigned EncodedBytes = 0;
+  int64_t CallTarget = MI.getOperand(StartIdx + 2).getImm();
+  if (CallTarget) {
+    // Emit MOV to materialize the target address and the CALL to target.
+    // This is encoded with 12-13 bytes, depending on which register is used.
+    // We conservatively assume that it is 12 bytes and emit in worst case one
+    // extra NOP byte.
+    EncodedBytes = 12;
+    OutStreamer.EmitInstruction(MCInstBuilder(X86::MOV64ri)
+                                .addReg(MI.getOperand(ScratchIdx).getReg())
+                                .addImm(CallTarget));
+    OutStreamer.EmitInstruction(MCInstBuilder(X86::CALL64r)
+                                .addReg(MI.getOperand(ScratchIdx).getReg()));
+  }
   // Emit padding.
   unsigned NumNOPBytes = MI.getOperand(StartIdx + 1).getImm();
   assert(NumNOPBytes >= EncodedBytes &&
diff --git a/test/CodeGen/X86/patchpoint.ll b/test/CodeGen/X86/patchpoint.ll
index 5b96916804..0e79a38dd6 100644
--- a/test/CodeGen/X86/patchpoint.ll
+++ b/test/CodeGen/X86/patchpoint.ll
@@ -79,6 +79,22 @@ entry:
   ret i64 10
 }
 
+; Test small patchpoints that don't emit calls.
+define void @small_patchpoint_codegen(i64 %p1, i64 %p2, i64 %p3, i64 %p4) {
+entry:
+; CHECK-LABEL: small_patchpoint_codegen:
+; CHECK:      Ltmp
+; CHECK:      nop
+; CHECK-NEXT: nop
+; CHECK-NEXT: nop
+; CHECK-NEXT: nop
+; CHECK-NEXT: nop
+; CHECK-NEXT: popq
+; CHECK-NEXT: ret
+  %result = tail call i64 (i32, i32, i8*, i32, ...)* @llvm.experimental.patchpoint.i64(i32 5, i32 5, i8* null, i32 2, i64 %p1, i64 %p2)
+  ret void
+}
+
 declare void @llvm.experimental.stackmap(i32, i32, ...)
 declare void @llvm.experimental.patchpoint.void(i32, i32, i8*, i32, ...)
 declare i64 @llvm.experimental.patchpoint.i64(i32, i32, i8*, i32, ...)