[compiler-rt] [scudo] Allow to resize allocation ring buffer (PR #82683)

[via llvm-commits]

================
@@ -1594,12 +1600,33 @@ class Allocator {
     RB->StackDepotSize = StackDepotSize;
     RB->RawStackDepotMap = DepotMap;
 
-    atomic_store(&RingBufferAddress, reinterpret_cast<uptr>(RB),
-                 memory_order_release);
+    swapOutRingBuffer(RB);
     static_assert(sizeof(AllocationRingBuffer) %
                           alignof(typename AllocationRingBuffer::Entry) ==
                       0,
                   "invalid alignment");
+    return true;
+  }
+
+  void swapOutRingBuffer(AllocationRingBuffer *NewRB) {
+    // To allow resizeRingBuffer to be called in a multi-threaded context by apps,
+    // we do not actually unmap, but only madvise(DONTNEED) the pages. That way,
+    // straggler threads will not crash.
----------------
ChiaHungDuan wrote:

The case I'm concerning is like, one thread is writing to the old buffer and another thread is calling DONTNEED on the old buffer. Both are writing ("either repopulate the page or substitute a zero page" is still a writing) to the same memory and they are not atomic operations. This is why I think it's an undefined behavior. 

About the wasting (leaking) pages, right, this is another concern I have. Even it's only for debugging, intentionally leaking memory in a memory allocator seems weird to me. In the perspective of maintenance, it's confusing  people who are not familiar with this.

I know it's unlikely to cause problem by this but I would prefer considering any approaches that are safer and reasonable. 

https://github.com/llvm/llvm-project/pull/82683