[scudo] Don't track free/use stats for transfer batches.

The Scudo C unit tests are currently non-hermetic. In particular, adding
or removing a transfer batch is a global state of the allocator that
persists between tests. This can cause flakiness in
ScudoWrappersCTest.MallInfo, because the creation or teardown of a batch
causes mallinfo's uordblks or fordblks to move up or down by the size of
a transfer batch on malloc/free.

It's my opinion that uordblks and fordblks should track the statistics
related to the user's malloc() and free() usage, and not the state of
the internal allocator structures. Thus, excluding the transfer batches
from stat collection does the trick and makes these tests pass.

Repro instructions of the bug:
 1. ninja ./projects/compiler-rt/lib/scudo/standalone/tests/ScudoCUnitTest-x86_64-Test
 2. ./projects/compiler-rt/lib/scudo/standalone/tests/ScudoCUnitTest-x86_64-Test --gtest_filter=ScudoWrappersCTest.MallInfo

Reviewed By: cryptoad

Differential Revision: https://reviews.llvm.org/D101653

diff --git a/compiler-rt/lib/scudo/standalone/local_cache.h b/compiler-rt/lib/scudo/standalone/local_cache.h
index 43cbc68..5003937 100644 (file)
--- a/compiler-rt/lib/scudo/standalone/local_cache.h
+++ b/compiler-rt/lib/scudo/standalone/local_cache.h
@@ -135,6 +135,7 @@ private:
   struct PerClass {
     u32 Count;
     u32 MaxCount;
+    // Note: ClassSize is zero for the transfer batch.
     uptr ClassSize;
     CompactPtrT Chunks[2 * TransferBatch::MaxNumCached];
   };
@@ -154,7 +155,13 @@ private:
       PerClass *P = &PerClassArray[I];
       const uptr Size = SizeClassAllocator::getSizeByClassId(I);
       P->MaxCount = 2 * TransferBatch::getMaxCached(Size);
-      P->ClassSize = Size;
+      if (I != BatchClassId) {
+        P->ClassSize = Size;
+      } else {
+        // ClassSize in this struct is only used for malloc/free stats, which
+        // should only track user allocations, not internal movements.
+        P->ClassSize = 0;
+      }
     }
   }