[PATCH] D18046: [X86] Providing correct unwind info in function epilogue

Keno Fischer via llvm-commits llvm-commits at lists.llvm.org
Sat Apr 23 10:19:15 PDT 2016


loladiro added a comment.

Two more comments I found while testing this out.


================
Comment at: lib/Target/X86/X86CFIInstrInserter.cpp:105
@@ +104,3 @@
+
+  if (!MF.getMMI().hasDebugInfo())
+    return false;
----------------
`&& !MF->getFunction()->needsUnwindTableEntry()` would be good. That's what's used elsewhere.

================
Comment at: lib/Target/X86/X86CFIInstrInserter.cpp:377
@@ +376,3 @@
+    if (!TFL->hasFP(MF) && (MBBI.getOpcode() == X86::ADD32ri8 ||
+                            MBBI.getOpcode() == X86::ADD64ri8 ||
+                            MBBI.getOpcode() == X86::ADD32ri)) {
----------------
Missing `X86::ADD64ri32`. A sufficiently large alloca would probably work for a test.


Repository:
  rL LLVM

http://reviews.llvm.org/D18046





More information about the llvm-commits mailing list