ThinLTO: sort inputs and schedule by decreasing size

This is a compile time optimization: keeping a large file to process
at the end hurts parallelism.
The heurisitic used right now is the input buffer size, however we
may want to consider the number of functions to import or the
different number of files to load for importing as well.

From: Mehdi Amini <[email protected]>

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@269684 91177308-0d34-0410-b5e6-96231b3b80d8

Conflicts:
	lib/LTO/ThinLTOCodeGenerator.cpp

Author: mehdi_amini

lib/LTO/ThinLTOCodeGenerator.cpp

@@ -52,6 +52,8 @@
 #include "llvm/Transforms/ObjCARC.h"
 #include "llvm/Transforms/Utils/FunctionImportUtils.h"
 
+#include <numeric>
+
 using namespace llvm;
 
 #define DEBUG_TYPE "thinlto"
@@ -881,11 +883,24 @@ void ThinLTOCodeGenerator::run() {
   for (auto &DefinedGVSummaries : ModuleToDefinedGVSummaries)
     ExportLists[DefinedGVSummaries.first()];
 
+  // Compute the ordering we will process the inputs: the rough heuristic here
+  // is to sort them per size so that the largest module get schedule as soon as
+  // possible. This is purely a compile-time optimization.
+  std::vector<int> ModulesOrdering;
+  ModulesOrdering.resize(Modules.size());
+  std::iota(ModulesOrdering.begin(), ModulesOrdering.end(), 0);
+  std::sort(ModulesOrdering.begin(), ModulesOrdering.end(),
+            [&](int LeftIndex, int RightIndex) {
+              auto LSize = Modules[LeftIndex].getBufferSize();
+              auto RSize = Modules[RightIndex].getBufferSize();
+              return LSize > RSize;
+            });
+
   // Parallel optimizer + codegen
   {
     ThreadPool Pool(getNumCores());
-    int count = 0;
-    for (auto &ModuleBuffer : Modules) {
+    for (auto IndexCount : ModulesOrdering) {
+      auto &ModuleBuffer = Modules[IndexCount];
       Pool.async([&](int count) {
         auto ModuleIdentifier = ModuleBuffer.getBufferIdentifier();
         auto &ExportList = ExportLists[ModuleIdentifier];
@@ -937,8 +952,7 @@ void ThinLTOCodeGenerator::run() {
 
         OutputBuffer = CacheEntry.write(std::move(OutputBuffer));
         ProducedBinaries[count] = std::move(OutputBuffer);
-      }, count);
-      count++;
+      }, IndexCount);
     }
   }