Subzero: Use a proper RegNumT type instead of int32_t/SizeT.

src/IceRegistersX8632.h

Index: src/IceRegistersX8632.h
diff --git a/src/IceRegistersX8632.h b/src/IceRegistersX8632.h
index 5d71aa70a7612564922d319ae6eb4f906645b23c..bb614e0c905bff72733a57d42778aed2272e82ff 100644
--- a/src/IceRegistersX8632.h
+++ b/src/IceRegistersX8632.h
@@ -80,10 +80,10 @@ public:
         Encoded_Not_X87STReg = -1
   };
 
-  static inline X87STRegister getEncodedSTReg(int32_t RegNum) {
-    assert(Encoded_X87ST_First <= RegNum);
-    assert(RegNum <= Encoded_X87ST_Last);
-    return X87STRegister(RegNum);
+  static inline X87STRegister getEncodedSTReg(uint32_t X87RegNum) {
+    assert(int(Encoded_X87ST_First) <= int(X87RegNum));

[Eric Holk, 2016/02/08 19:37:10]

Why cast both of these to int? Where they both unsigned to start with? If so,
why not leave them that way?

[Jim Stichnoth, 2016/02/09 19:33:39]

The int() cast is because otherwise g++ warns that the unsigned comparison
against 0 will always succeed, and -Werror makes the build fail. :(

The x87 stuff isn't really part of the RegNumT enum space.  It's kind of a hack
because we basically use the x87 floating point stack purely for the x86-32
function call ABI which says that FP values are returned in st(0).

[Eric Holk, 2016/02/10 01:11:30]

Ah, okay. That makes sense.

+    assert(X87RegNum <= Encoded_X87ST_Last);
+    return X87STRegister(X87RegNum);
   }
 };