Use the integrated assembler as default on PowerPC

This was already done in clang, this commit now uses the integrated
assembler as default when using LLVM tools directly.

A number of test cases using inline asm had to be adapted, either by
updating the expected output, or by using -no-integrated-as (for such
tests that deliberately use an invalid instruction in inline asm).

llvm-svn: 225819
diff --git a/llvm/test/CodeGen/PowerPC/crsave.ll b/llvm/test/CodeGen/PowerPC/crsave.ll
index a9b4b36..602ba94 100644
--- a/llvm/test/CodeGen/PowerPC/crsave.ll
+++ b/llvm/test/CodeGen/PowerPC/crsave.ll
@@ -6,7 +6,7 @@
 define i32 @test_cr2() nounwind uwtable {
 entry:
   %ret = alloca i32, align 4
-  %0 = call i32 asm sideeffect "\0A\09mtcr $4\0A\09cmp 2,$2,$1\0A\09mfcr $0", "=r,r,r,r,r,~{cr2}"(i32 1, i32 2, i32 3, i32 0) nounwind
+  %0 = call i32 asm sideeffect "\0A\09mtcr $4\0A\09cmpw 2,$2,$1\0A\09mfcr $0", "=r,r,r,r,r,~{cr2}"(i32 1, i32 2, i32 3, i32 0) nounwind
   store i32 %0, i32* %ret, align 4
   call void @foo()
   %1 = load i32* %ret, align 4
@@ -35,7 +35,7 @@
 define i32 @test_cr234() nounwind {
 entry:
   %ret = alloca i32, align 4
-  %0 = call i32 asm sideeffect "\0A\09mtcr $4\0A\09cmp 2,$2,$1\0A\09cmp 3,$2,$2\0A\09cmp 4,$2,$3\0A\09mfcr $0", "=r,r,r,r,r,~{cr2},~{cr3},~{cr4}"(i32 1, i32 2, i32 3, i32 0) nounwind
+  %0 = call i32 asm sideeffect "\0A\09mtcr $4\0A\09cmpw 2,$2,$1\0A\09cmpw 3,$2,$2\0A\09cmpw 4,$2,$3\0A\09mfcr $0", "=r,r,r,r,r,~{cr2},~{cr3},~{cr4}"(i32 1, i32 2, i32 3, i32 0) nounwind
   store i32 %0, i32* %ret, align 4
   call void @foo()
   %1 = load i32* %ret, align 4