[Lldb-commits] [lldb] [lldb] Memory cache: no overlap and read across cache entries (PR #222688)

[Felipe de Azevedo Piovezan via lldb-commits]

================
@@ -12,12 +12,97 @@
 #include "lldb/Utility/RangeMap.h"
 #include "lldb/lldb-private.h"
 #include "llvm/ADT/ArrayRef.h"
+#include "llvm/ADT/DenseMap.h"
 #include "llvm/ADT/SmallVector.h"
 #include <map>
+#include <memory>
 #include <mutex>
 #include <vector>
 
 namespace lldb_private {
+
+/// A set of whole, aligned cache lines, keyed by line index.  A key names a
----------------
felipepiovezan wrote:

(Feel free to push back if you think my jet-lagged self is overthinking this too much)

I think I understand why you phrased it this way, but IMO "whole line" sounds a bit off. 
There are three properties we are capturing here: A) entries are aligned, B) entries have a fixed size, C) all entries have the same size. When we say "line", we are capturing all three of those (because we're making an analogy to hardware caches)

So I would rephrase this entire comment slightly:

```
A set of cache entries, all of which are aligned and have the same size. Entries cannot be partially filled.
```

How they are keyed, or whether a length is stored in each entry, is an implementation detail. I would move these to the `using Collection = ` line.

https://github.com/llvm/llvm-project/pull/222688