Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757932AbXFVO5K (ORCPT ); Fri, 22 Jun 2007 10:57:10 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1753701AbXFVO46 (ORCPT ); Fri, 22 Jun 2007 10:56:58 -0400 Received: from waste.org ([66.93.16.53]:51766 "EHLO waste.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753647AbXFVO45 (ORCPT ); Fri, 22 Jun 2007 10:56:57 -0400 Date: Fri, 22 Jun 2007 09:56:35 -0500 From: Matt Mackall To: Yoshinori Sato , Nick Piggin Cc: lkml Subject: Re: [PATCH] SLOB allocator imcompatible SLAB Message-ID: <20070622145635.GX11115@waste.org> References: MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.13 (2006-08-11) Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2680 Lines: 87 On Fri, Jun 22, 2007 at 05:08:07PM +0900, Yoshinori Sato wrote: > Because the page which SLOB allocator got does not have PG_slab, This is for a NOMMU system? You're using an old kernel with an old version of SLOB. SLOB in newer kernels actually sets per-page flags. Nick, can you see any reason not to s/PG_active/PG_slab/ in the current code? > Signed-off-by: Yoshinori Sato > > diff --git a/mm/slob.c b/mm/slob.c > index 71976c5..d10bcda 100644 > --- a/mm/slob.c > +++ b/mm/slob.c > @@ -73,6 +73,21 @@ static DEFINE_SPINLOCK(block_lock); > static void slob_free(void *b, int size); > static void slob_timer_cbk(void); > > +static inline void set_slabflags(const void *ptr, int order) > +{ > + int i; > + struct page *page = virt_to_page(ptr); > + for (i = 0; i < (1 << order); i++) > + __SetPageSlab(page++); > +} > + > +static inline void clear_slabflags(const void *ptr, int order) > +{ > + int i; > + struct page *page = virt_to_page(ptr); > + for (i = 0; i < (1 << order); i++) > + __ClearPageSlab(page++); > +} > > static void *slob_alloc(size_t size, gfp_t gfp, int align) > { > @@ -180,6 +195,7 @@ void *__kmalloc(size_t size, gfp_t gfp) > bb->pages = (void *)__get_free_pages(gfp, bb->order); > > if (bb->pages) { > + set_slabflags(bb->pages, bb->order); > spin_lock_irqsave(&block_lock, flags); > bb->next = bigblocks; > bigblocks = bb; > @@ -240,6 +256,7 @@ void kfree(const void *block) > if (bb->pages == block) { > *last = bb->next; > spin_unlock_irqrestore(&block_lock, flags); > + clear_slabflags(block, bb->order); > free_pages((unsigned long)block, bb->order); > slob_free(bb, sizeof(bigblock_t)); > return; > @@ -323,9 +340,11 @@ void *kmem_cache_alloc(struct kmem_cache *c, gfp_t flags) > > if (c->size < PAGE_SIZE) > b = slob_alloc(c->size, flags, c->align); > - else > + else { > b = (void *)__get_free_pages(flags, get_order(c->size)); > - > + if (b) > + set_slabflags(b, get_order(c->size)); > + } > if (c->ctor) > c->ctor(b, c, 0); > > @@ -347,8 +366,10 @@ static void __kmem_cache_free(void *b, int size) > { > if (size < PAGE_SIZE) > slob_free(b, size); > - else > + else { > + clear_slabflags(b, get_order(size)); > free_pages((unsigned long)b, get_order(size)); > + } > } > > static void kmem_rcu_free(struct rcu_head *head) -- Mathematics is the supreme nostalgia of our time. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/