From mboxrd@z Thu Jan 1 00:00:00 1970 Date: Tue, 28 Nov 2006 23:08:37 -0800 From: Andrew Morton Subject: Re: Slab: Remove kmem_cache_t Message-Id: <20061128230837.48fcc34f.akpm@osdl.org> In-Reply-To: <456D2B8E.4060802@yahoo.com.au> References: <456D0757.6050903@yahoo.com.au> <456D0FC4.4050704@yahoo.com.au> <20061128200619.67080e11.akpm@osdl.org> <456D1D82.3060001@yahoo.com.au> <20061128222409.cda8cd5e.akpm@osdl.org> <456D2B8E.4060802@yahoo.com.au> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-linux-mm@kvack.org Return-Path: To: Nick Piggin Cc: Christoph Lameter , linux-mm@kvack.org, Linus Torvalds List-ID: On Wed, 29 Nov 2006 17:41:18 +1100 Nick Piggin wrote: > > Well, you'd just do > > > > extern struct kmem_cache *wozzle; > > > > because you "know" that struct kmem_cache == kmem_cache_t. The compiler > > will swallow it all. > > > > Do I need to explain how much that sucks? > > > > Well the only code that is doing this is presumably some slab internal > stuff. And that does "know" that struct kmem_cache == kmem_cache_t. > Actually, once struct kmem_cache gets moved into slab.h, I would be > interested to know what remaining forward dependencies are needed at > all. Christoph? > > To be clear: this won't be some random driver or subsystem code (or > even anything outside of mm/slab.c, hopefully) that is doing this, > will it? Yes, it will. Any module which calls kmem_cache_create() needs to save its return value into some storage. That storage has type `struct kmem_cache *', or kmem_cache_t *. And, btw, in neither case does that kmem_cache_create() caller need to know what's inside `struct kmem_cache'. -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org