Here is a real patch.
-------- Weitergeleitete Nachricht --------
Betreff: Bug in Support/Allocator.h
Datum: Mon, 11 May 2015 17:25:45 +0200
Von: Kal <b17c0de at gmail.com>
An: LLVM Developers Mailing List (llvmdev at cs.uiuc.edu)
<llvmdev at cs.uiuc.edu>
Hi,
llvm::BumpPtrAllocatorImpl::Reset() looks like it has a bug.
There is this check at the top:
if (Slabs.empty())
return;
But what if you don't have any normal Slabs allocated but only
CustomSizedSlabs.
I think this should be:
if (Slabs.empty() && CustomSizedSlabs.empty())
return;
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20150511/0805fe5b/attachment.html>
-------------- next part --------------
Index: Allocator.h
==================================================================---
Allocator.h (revision 236999)
+++ Allocator.h (working copy)
@@ -187,17 +187,16 @@
/// \brief Deallocate all but the current slab and reset the current pointer
/// to the beginning of it, freeing all memory allocated so far.
void Reset() {
- if (Slabs.empty())
- return;
+ if (!Slabs.empty()) {
+ // Reset the state.
+ BytesAllocated = 0;
+ CurPtr = (char *)Slabs.front();
+ End = CurPtr + SlabSize;
- // Reset the state.
- BytesAllocated = 0;
- CurPtr = (char *)Slabs.front();
- End = CurPtr + SlabSize;
-
- // Deallocate all but the first slab, and all custome sized slabs.
- DeallocateSlabs(std::next(Slabs.begin()), Slabs.end());
- Slabs.erase(std::next(Slabs.begin()), Slabs.end());
+ // Deallocate all but the first slab, and all custome sized slabs.
+ DeallocateSlabs(std::next(Slabs.begin()), Slabs.end());
+ Slabs.erase(std::next(Slabs.begin()), Slabs.end());
+ }
DeallocateCustomSizedSlabs();
CustomSizedSlabs.clear();
}