[PATCH] D109637: [llvm-profgen] Ignore broken LBR samples

[Hongtao Yu via Phabricator via llvm-commits]

hoy created this revision.
Herald added subscribers: modimo, wenlei.
hoy requested review of this revision.
Herald added a project: LLVM.
Herald added a subscriber: llvm-commits.

Perf script can sometimes give disordered LBR samples like below.

       b022500
      32de0044
      3386e1d1
  7f118e05720c
  7f118df2d81f

0x2a0b9622/0x2a0b9610/P/-/-/1  0x2a0b79ff/0x2a0b9618/P/-/-/2  0x2a0b7a4a/0x2a0b79e8/P/-/-/1  0x2a0b7a33/0x2a0b7a46/P/-/-/1  0x2a0b7a42/0x2a0b7a23/P/-/-/1  0x2a0b7a21/0x2a0b7a37/P/-/-/2  0x2a0b79e6/0x2a0b7a07/P/-/-/1  0x2a0b79d4/0x2a0b79dc/P/-/-/2  0x2a0b7a03/0x2a0b79aa/P/-/-/1  0x2a0b79a8/0x2a0b7a00/P/-/-/234  0x2a0b9613/0x2a0b7930/P/-/-/1  0x2a0b9622/0x2a0b9610/P/-/-/1  0x2a0b79ff/0x2a0b9618/P/-/-/2  0x2a0b7a4a/0x2aWarning:
Processed 10263226 events and lost 1 chunks!

Note that the last LBR record 0x2a0b7a4a/0x2aWarning: . Currently llvm-profgen does not detect that and as a result an uninitialized branch target value will be used. The uninitialized value can cause creepy instruction ranges created which which in turn will result in a completely wrong profile. An example is like

.... @ _ZN5folly13loadUnalignedIsEET_PKv]:18446744073709551615:18446744073709551615
 1: 18446744073709551615
 !CFGChecksum: 4294967295
 !Attributes: 0


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D109637

Files:
  llvm/tools/llvm-profgen/PerfReader.cpp
  llvm/tools/llvm-profgen/PerfReader.h
  llvm/tools/llvm-profgen/ProfileGenerator.cpp


Index: llvm/tools/llvm-profgen/ProfileGenerator.cpp
===================================================================
--- llvm/tools/llvm-profgen/ProfileGenerator.cpp
+++ llvm/tools/llvm-profgen/ProfileGenerator.cpp
@@ -183,6 +183,7 @@
   for (auto Item : Ranges) {
     uint64_t Begin = Item.first.first;
     uint64_t End = Item.first.second;
+    assert(Begin <= End && "invalid instruction range");
     uint64_t Count = Item.second;
     if (Boundaries.find(Begin) == Boundaries.end())
       Boundaries[Begin] = BoundaryPoint();
@@ -194,7 +195,7 @@
   }
 
   uint64_t BeginAddress = UINT64_MAX;
-  int Count = 0;
+  uint64_t Count = 0;
   for (auto Item : Boundaries) {
     uint64_t Address = Item.first;
     BoundaryPoint &Point = Item.second;
@@ -208,6 +209,7 @@
       assert((BeginAddress != UINT64_MAX) &&
              "First boundary point cannot be 'end' point");
       DisjointRanges[{BeginAddress, Address}] = Count;
+      assert(Count >= Point.EndCount && "mismatched live ranges");
       Count -= Point.EndCount;
       BeginAddress = Address + 1;
     }
Index: llvm/tools/llvm-profgen/PerfReader.h
===================================================================
--- llvm/tools/llvm-profgen/PerfReader.h
+++ llvm/tools/llvm-profgen/PerfReader.h
@@ -373,6 +373,7 @@
   BranchSample BranchCounter;
 
   void recordRangeCount(uint64_t Start, uint64_t End, uint64_t Repeat) {
+    assert(Start <= End && "invalid instruction range");
     RangeCounter[{Start, End}] += Repeat;
   }
   void recordBranchCount(uint64_t Source, uint64_t Target, uint64_t Repeat) {
Index: llvm/tools/llvm-profgen/PerfReader.cpp
===================================================================
--- llvm/tools/llvm-profgen/PerfReader.cpp
+++ llvm/tools/llvm-profgen/PerfReader.cpp
@@ -422,8 +422,11 @@
     Token.split(Addresses, "/");
     uint64_t Src;
     uint64_t Dst;
-    Addresses[0].substr(2).getAsInteger(16, Src);
-    Addresses[1].substr(2).getAsInteger(16, Dst);
+
+    // Stop at broken LBR records.
+    if (Addresses[0].substr(2).getAsInteger(16, Src) ||
+        Addresses[1].substr(2).getAsInteger(16, Dst))
+      break;
 
     bool SrcIsInternal = Binary->addressIsCode(Src);
     bool DstIsInternal = Binary->addressIsCode(Dst);


-------------- next part --------------
A non-text attachment was scrubbed...
Name: D109637.372033.patch
Type: text/x-patch
Size: 2255 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20210910/d0ec07b6/attachment.bin>