Start adding an integrated assembler.

src/llvm2ice.cpp

Index: src/llvm2ice.cpp
diff --git a/src/llvm2ice.cpp b/src/llvm2ice.cpp
index a4a96b9808309c73e6ae1cd63d915fd9067e53cd..412b76642b101f1e6a65a1084e41eac75f4fe3c1 100644
--- a/src/llvm2ice.cpp
+++ b/src/llvm2ice.cpp
@@ -108,6 +108,9 @@ static cl::opt<bool>
 BuildOnRead("build-on-read",
             cl::desc("Build ICE instructions when reading bitcode"),
             cl::init(false));
+static cl::opt<bool>
+    DisableIntegratedAssembler("no-integrated-as",

[Jim Stichnoth, 2014/09/04 20:57:48]

Can the sense of this be reversed for better clarity?  E.g.

static cl::opt<bool> UseIntegratedAssembler("integrated-as", cl::init(true));
...
Flags.UseIntegratedAssembler = UseIntegratedAssembler;

[jvoung (off chromium), 2014/09/08 21:39:00]

Done.

I started with UseIntegratedAssembler cl::init(false), but that made testing
more difficult, since I needed to remember exactly which tests get affected and
add the flag to test.

+                               cl::desc("Do not use integrated assembler"));
 
 int main(int argc, char **argv) {
 
@@ -138,6 +141,7 @@ int main(int argc, char **argv) {
   Flags.DisableGlobals = DisableGlobals;
   Flags.FunctionSections = FunctionSections;
   Flags.UseSandboxing = UseSandboxing;
+  Flags.UseIntegratedAssembler = !DisableIntegratedAssembler;
 
   Ice::GlobalContext Ctx(Ls, Os, VMask, TargetArch, OptLevel, TestPrefix,
                          Flags);