commit a0c0857e7a2398237d00a13ad9ac7078c0148578
author Wei Mi <wmi@google.com> Mon Jun 11 22:40:43 2018 +0000
committer Wei Mi <wmi@google.com> Mon Jun 11 22:40:43 2018 +0000
tree 7ed375920f8708f1cc9ef6ca32651cb7c166abef
parent 19b22d406d5d37428cb62f0f8b0de79afbe99b63

[SampleFDO] Add a new compact binary format for sample profile.

Name table occupies a big chunk of size in current binary format sample profile.
In order to reduce its size, the patch changes the sample writer/reader to
save/restore MD5Hash of names in the name table. Sample annotation phase will
also use MD5Hash of name to query samples accordingly.

Experiment shows compact binary format can reduce the size of sample profile by
2/3 compared with binary format generally.

Differential Revision: https://reviews.llvm.org/D47955

llvm-svn: 334447

llvm/lib/Transforms/IPO/SampleProfile.cpp

diff --git a/llvm/lib/Transforms/IPO/SampleProfile.cpp b/llvm/lib/Transforms/IPO/SampleProfile.cpp
index 1c53a82..357a7d1 100644
--- a/llvm/lib/Transforms/IPO/SampleProfile.cpp
+++ b/llvm/lib/Transforms/IPO/SampleProfile.cpp
@@ -638,6 +638,8 @@
   if (FS == nullptr)
     return nullptr;
 
+  std::string CalleeGUID;
+  CalleeName = getRepInFormat(CalleeName, Reader->getFormat(), CalleeGUID);
   return FS->findFunctionSamplesAt(LineLocation(FunctionSamples::getOffset(DIL),
                                                 DIL->getBaseDiscriminator()),
                                    CalleeName);
@@ -753,6 +755,7 @@
     Function &F, DenseSet<GlobalValue::GUID> &InlinedGUIDs) {
   DenseSet<Instruction *> PromotedInsns;
   bool Changed = false;
+  bool isCompact = (Reader->getFormat() == SPF_Compact_Binary);
   while (true) {
     bool LocalChanged = false;
     SmallVector<Instruction *, 10> CIS;
@@ -784,7 +787,8 @@
         for (const auto *FS : findIndirectCallFunctionSamples(*I, Sum)) {
           if (IsThinLTOPreLink) {
             FS->findInlinedFunctions(InlinedGUIDs, F.getParent(),
-                                     PSI->getOrCompHotCountThreshold());
+                                     PSI->getOrCompHotCountThreshold(),
+                                     isCompact);
             continue;
           }
           auto CalleeFunctionName = FS->getName();
@@ -793,7 +797,9 @@
           // clone the caller first, and inline the cloned caller if it is
           // recursive. As llvm does not inline recursive calls, we will
           // simply ignore it instead of handling it explicitly.
-          if (CalleeFunctionName == F.getName())
+          std::string FGUID;
+          auto Fname = getRepInFormat(F.getName(), Reader->getFormat(), FGUID);
+          if (CalleeFunctionName == Fname)
             continue;
 
           const char *Reason = "Callee function not available";
@@ -823,7 +829,8 @@
           LocalChanged = true;
       } else if (IsThinLTOPreLink) {
         findCalleeFunctionSamples(*I)->findInlinedFunctions(
-            InlinedGUIDs, F.getParent(), PSI->getOrCompHotCountThreshold());
+            InlinedGUIDs, F.getParent(), PSI->getOrCompHotCountThreshold(),
+            isCompact);
       }
     }
     if (LocalChanged) {