[PATCH] D17211: Consider regmasks when computing register-based DBG_VALUE live ranges

[Adrian Prantl via llvm-commits]

aprantl added inline comments.

================
Comment at: lib/CodeGen/AsmPrinter/DbgValueHistoryCalculator.cpp:165
@@ +164,3 @@
+                 ++AI)
+              Regs.set(*AI);
+          } else if (MO.isRegMask()) {
----------------
Together with the code, could you also inline the relevant bits of the doxygen comment?
Would be nice to have a line describing each case in the if-statement.

================
Comment at: lib/CodeGen/AsmPrinter/DbgValueHistoryCalculator.cpp:195
@@ +194,3 @@
+                clobberRegisterUses(RegVars, *AI, Result, MI);
+          } else if (MO.isRegMask()) {
+            for (int I = ChangingRegs.find_first(); I != -1;
----------------
same here. More comments appreciated.

================
Comment at: lib/CodeGen/LiveDebugValues.cpp:229
@@ +228,3 @@
+      // list SP as preserved.  While the debug info may be off for an
+      // instruction or two around callee-cleanup calls, tranfering the
+      // DEBUG_VALUE across the call is still a better user experience.
----------------
typo :-)

================
Comment at: test/DebugInfo/X86/array.ll:19
@@ -18,4 +18,3 @@
 ;
-; CHECK:     ##DEBUG_VALUE: main:array <- [%R{{.*}}+0]
-; CHECK:     ##DEBUG_VALUE: main:array <- [%R{{.*}}+0]
-; CHECK:     ##DEBUG_VALUE: main:array <- [%R{{.*}}+0]
+; FIXME: If we described this location as RSP-relative instead of RDI-relative
+; the live range would be larger.
----------------
Is there a place in the LLVM sources we could move/clone this comment to?


http://reviews.llvm.org/D17211