[AMDGPU][SplitModule] Fix a couple of issues

A static analysis tool found that ModuleCost could be zero, so would
perform divide by zero when being printed. Perhaps this is unreachable
in practice, but the fix is straightforward enough and unlikely to be a
performance concern.

The same tool warned that a division was always being performed in
integer division, so was either 0.0 or 1.0. This doesn't seem
intentional, so has been fixed to return a true ratio using
floating-point division. This has a knock-on effect on how a test was
splitting modules.

Author: frasercrmck

llvm/lib/Target/AMDGPU/AMDGPUSplitModule.cpp

@@ -149,7 +149,8 @@ static constexpr unsigned InvalidPID = -1;
 /// \param Dem denominator
 /// \returns a printable object to print (Num/Dem) using "%0.2f".
 static auto formatRatioOf(CostType Num, CostType Dem) {
-  return format("%0.2f", (static_cast<double>(Num) / Dem) * 100);
+  CostType DemOr1 = Dem ? Dem : 1;
+  return format("%0.2f", (static_cast<double>(Num) / DemOr1) * 100);
 }
 
 /// Checks whether a given function is non-copyable.
@@ -1101,7 +1102,7 @@ void RecursiveSearchSplitting::pickPartition(unsigned Depth, unsigned Idx,
         // Check if the amount of code in common makes it worth it.
         assert(SimilarDepsCost && Entry.CostExcludingGraphEntryPoints);
         const double Ratio =
-            SimilarDepsCost / Entry.CostExcludingGraphEntryPoints;
+            (double)SimilarDepsCost / Entry.CostExcludingGraphEntryPoints;
         assert(Ratio >= 0.0 && Ratio <= 1.0);
         if (LargeFnOverlapForMerge > Ratio) {
           // For debug, just print "L", so we'll see "L3=P3" for instance, which

llvm/test/tools/llvm-split/AMDGPU/large-kernels-merging.ll

@@ -15,19 +15,20 @@
 ; Also check w/o large kernels processing to verify they are indeed handled
 ; differently.
 
-; P0 is empty
-; CHECK0: declare
+; CHECK0: define internal void @HelperC()
+; CHECK0: define amdgpu_kernel void @C
 
-; CHECK1: define internal void @HelperC()
-; CHECK1: define amdgpu_kernel void @C
+; CHECK1: define internal void @large2()
+; CHECK1: define internal void @large1()
+; CHECK1: define internal void @large0()
+; CHECK1: define internal void @HelperB()
+; CHECK1: define amdgpu_kernel void @B
 
 ; CHECK2: define internal void @large2()
 ; CHECK2: define internal void @large1()
 ; CHECK2: define internal void @large0()
 ; CHECK2: define internal void @HelperA()
-; CHECK2: define internal void @HelperB()
 ; CHECK2: define amdgpu_kernel void @A
-; CHECK2: define amdgpu_kernel void @B
 
 ; NOLARGEKERNELS-CHECK0: define internal void @HelperC()
 ; NOLARGEKERNELS-CHECK0: define amdgpu_kernel void @C