🔗 Memory Pools
- Goal: Reduce memory fragmentation and provide detailed statistics
- Status: Done.
- Version: 2.0
- More:
🔗 Details
MemPools is a set of pooled memory allocators running on top of malloc(). It’s purpose is to reduce memory fragmentation and provide detailed statistics on memory consumption.
Preferably all memory allocations in Squid should be done using MemPools or one of the types built on top of it (i.e. cbdata).
MemPools are currently half-migrated towards proper C++, having been converted from C functions to static members of a C++ class. This leaves some issues open, such as initialization order.
Also, with the current advancements in malloc implementations one may want to link Squid against an alternaive malloc implementation:
🔗 Public API
See http://www.squid-cache.org/Doc/code/namespaceMem.html and http://www.squid-cache.org/Doc/code/group__MemPoolsAPI.html for the public API definitions.
🔗 MEMPROXY_CLASS Macro
This macro defines pooled new and delete operators for the class in which it is used. It should be your first choice of how to integrate a C++ class in Squid for dynamic allocation. Other API mechanisms are possible, but are designed for special use cases.
For easy reading and code maintenance it should be placed at the top of the class definition in the private area before any other API details and followed by an empty line then the ‘public:’ section definition.
class Foo
{
MEMPROXY_CLASS(Foo);
public:
...
};
Classes which use the CBDATA_CLASS macro must not also use MEMPROXY_CLASS. That includes use in the direct line of inheritence within a class hierarchy.
Categories: Feature
Navigation: Site Search, Site Pages, Categories, 🔼 go up