Change printfln to printfn. Make LLVM 15 tests default.

test/test_suite/floats/convert_float.c3t

@@ -13,12 +13,12 @@ define void @foo_test() #0 {
 entry:
   %x = alloca float, align 4
   %y = alloca double, align 8
-  store float 1.200000e+01, float* %x, align 4
-  %0 = load float, float* %x, align 4
+  store float 1.200000e+01, ptr %x, align 4
+  %0 = load float, ptr %x, align 4
   %fmul = fmul float %0, -1.000000e+00
-  store float %fmul, float* %x, align 4
-  %1 = load float, float* %x, align 4
+  store float %fmul, ptr %x, align 4
+  %1 = load float, ptr %x, align 4
   %fpfpext = fpext float %1 to double
-  store double %fpfpext, double* %y, align 8
+  store double %fpfpext, ptr %y, align 8
   ret void
-}
+}

test/test_suite/floats/inf_nan.c3t

@@ -9,18 +9,18 @@ fn void main()
     double x = (double)(z);
 }
 
-// #expect: fe.ll
+/* #expect: fe.ll
 
   %z = alloca float, align 4
   %z2 = alloca double, align 8
   %g = alloca float, align 4
   %g2 = alloca double, align 8
   %x = alloca double, align 8
-  store float 0x7FF0000000000000, float* %z, align 4
-  store double 0x7FF0000000000000, double* %z2, align 8
-  store float 0x7FF8000000000000, float* %g, align 4
-  store double 0x7FF8000000000000, double* %g2, align 8
-  %0 = load float, float* %z, align 4
+  store float 0x7FF0000000000000, ptr %z, align 4
+  store double 0x7FF0000000000000, ptr %z2, align 8
+  store float 0x7FF8000000000000, ptr %g, align 4
+  store double 0x7FF8000000000000, ptr %g2, align 8
+  %0 = load float, ptr %z, align 4
   %fpfpext = fpext float %0 to double
-  store double %fpfpext, double* %x, align 8
+  store double %fpfpext, ptr %x, align 8
   ret void

test/test_suite/floats/mod.c3t

@@ -0,0 +1,14 @@
+// #target: macos-x64
+
+fn double x(double a, double b)
+{
+	return a % b;
+}
+
+/* #expect: mod.ll
+
+define double @mod_x(double %0, double %1) #0 {
+entry:
+  %fmod = frem double %0, %1
+  ret double %fmod
+}