[BOLT] Only parse probes for profiled functions in profile-write-pseudo-probes mode

[aaupov]

Implement selective probe parsing for profiled functions only when
emitting probe information to YAML profile as suggested in
#102904 (review)

For a large binary, this reduces probe parsing

• processing time from 10.5925s to 5.6295s,
• peak RSS from 10.54 to 7.98 GiB.

[llvmbot]

@llvm/pr-subscribers-bolt

Author: Amir Ayupov (aaupov)

Changes

---

Full diff: https://github.com/llvm/llvm-project/pull/106365.diff

1 Files Affected:

• (modified) bolt/lib/Rewrite/PseudoProbeRewriter.cpp (+7-3)

diff --git a/bolt/lib/Rewrite/PseudoProbeRewriter.cpp b/bolt/lib/Rewrite/PseudoProbeRewriter.cpp
index 228913e6ea1f39..89a7fddbb5d2af 100644
--- a/bolt/lib/Rewrite/PseudoProbeRewriter.cpp
+++ b/bolt/lib/Rewrite/PseudoProbeRewriter.cpp
@@ -72,7 +72,8 @@ class PseudoProbeRewriter final : public MetadataRewriter {
 
   /// Parse .pseudo_probe_desc section and .pseudo_probe section
   /// Setup Pseudo probe decoder
-  void parsePseudoProbe();
+  /// If \p ProfiledOnly is set, only parse records for functions with profile.
+  void parsePseudoProbe(bool ProfiledOnly = false);
 
   /// PseudoProbe decoder
   std::shared_ptr<MCPseudoProbeDecoder> ProbeDecoderPtr;
@@ -92,7 +93,7 @@ class PseudoProbeRewriter final : public MetadataRewriter {
 
 Error PseudoProbeRewriter::preCFGInitializer() {
   if (opts::ProfileWritePseudoProbes)
-    parsePseudoProbe();
+    parsePseudoProbe(true);
 
   return Error::success();
 }
@@ -105,7 +106,7 @@ Error PseudoProbeRewriter::postEmitFinalizer() {
   return Error::success();
 }
 
-void PseudoProbeRewriter::parsePseudoProbe() {
+void PseudoProbeRewriter::parsePseudoProbe(bool ProfiledOnly) {
   MCPseudoProbeDecoder &ProbeDecoder(*ProbeDecoderPtr);
   PseudoProbeDescSection = BC.getUniqueSectionByName(".pseudo_probe_desc");
   PseudoProbeSection = BC.getUniqueSectionByName(".pseudo_probe");
@@ -136,6 +137,7 @@ void PseudoProbeRewriter::parsePseudoProbe() {
   MCPseudoProbeDecoder::Uint64Map FuncStartAddrs;
   SmallVector<StringRef, 3> Suffixes({".llvm.", ".destroy", ".resume"});
   for (const BinaryFunction *F : BC.getAllBinaryFunctions()) {
+    bool HasProfile = F->hasProfileAvailable();
     for (const MCSymbol *Sym : F->getSymbols()) {
       StringRef SymName = NameResolver::restore(Sym->getName());
       if (std::optional<StringRef> CommonName =
@@ -144,6 +146,8 @@ void PseudoProbeRewriter::parsePseudoProbe() {
       }
       uint64_t GUID = Function::getGUID(SymName);
       FuncStartAddrs[GUID] = F->getAddress();
+      if (ProfiledOnly && HasProfile)
+        GuidFilter.insert(GUID);
     }
   }
   Contents = PseudoProbeSection->getContents();

[wlei-llvm]

LGTM, thanks.