[llvm] 2d3d2e7 - MIR: Fix parser crash on syntax error on first character

[Matt Arsenault via llvm-commits]

Author: Matt Arsenault
Date: 2021-02-18T18:59:08-05:00
New Revision: 2d3d2e78d08578f4699cc268b7c292c3ff97e7fe

URL: https://github.com/llvm/llvm-project/commit/2d3d2e78d08578f4699cc268b7c292c3ff97e7fe
DIFF: https://github.com/llvm/llvm-project/commit/2d3d2e78d08578f4699cc268b7c292c3ff97e7fe.diff

LOG: MIR: Fix parser crash on syntax error on first character

This was calling the diagnostic printer before the context member was
initialized.

Added: 
    llvm/test/CodeGen/MIR/Generic/first-character-parse-error.mir

Modified: 
    llvm/lib/CodeGen/MIRParser/MIRParser.cpp

Removed: 
    


################################################################################
diff  --git a/llvm/lib/CodeGen/MIRParser/MIRParser.cpp b/llvm/lib/CodeGen/MIRParser/MIRParser.cpp
index 2d8613b41311..7dccd5da9c9a 100644
--- a/llvm/lib/CodeGen/MIRParser/MIRParser.cpp
+++ b/llvm/lib/CodeGen/MIRParser/MIRParser.cpp
@@ -51,9 +51,9 @@ namespace llvm {
 /// file.
 class MIRParserImpl {
   SourceMgr SM;
+  LLVMContext &Context;
   yaml::Input In;
   StringRef Filename;
-  LLVMContext &Context;
   SlotMapping IRSlots;
   std::unique_ptr<PerTargetMIParsingState> Target;
 
@@ -176,10 +176,11 @@ MIRParserImpl::MIRParserImpl(std::unique_ptr<MemoryBuffer> Contents,
                              StringRef Filename, LLVMContext &Context,
                              std::function<void(Function &)> Callback)
     : SM(),
+      Context(Context),
       In(SM.getMemoryBuffer(SM.AddNewSourceBuffer(std::move(Contents), SMLoc()))
              ->getBuffer(),
          nullptr, handleYAMLDiag, this),
-      Filename(Filename), Context(Context), ProcessIRFunction(Callback) {
+      Filename(Filename), ProcessIRFunction(Callback) {
   In.setContext(&In);
 }
 

diff  --git a/llvm/test/CodeGen/MIR/Generic/first-character-parse-error.mir b/llvm/test/CodeGen/MIR/Generic/first-character-parse-error.mir
new file mode 100644
index 000000000000..00a01058dc8c
--- /dev/null
+++ b/llvm/test/CodeGen/MIR/Generic/first-character-parse-error.mir
@@ -0,0 +1,14 @@
+:# RUN: not llc -run-pass=none %s -o - 2>&1 | FileCheck %s
+
+# The : before the run comment is syntactically invalid. This used to
+# crash in the SourceMgr diagnostic printer because it was called
+# before the LLVMContext was initialized.
+
+# CHECK: error: YAML:1:1: Unrecognized character while tokenizing.
+
+---
+name: foo
+body: |
+  bb.0:
+
+...