From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from hastur.corp.sgi.com (hastur.corp.sgi.com [198.149.32.33]) by omx1.americas.sgi.com (8.12.10/8.12.9/linux-outbound_gateway-1.1) with ESMTP id j9DMcbxT028635 for ; Thu, 13 Oct 2005 17:38:38 -0500 Received: from spindle.corp.sgi.com (spindle.corp.sgi.com [198.29.75.13]) by hastur.corp.sgi.com (8.12.9/8.12.10/SGI_generic_relay-1.2) with ESMTP id j9DMcCeS202766246 for ; Thu, 13 Oct 2005 15:38:12 -0700 (PDT) Received: from schroedinger.engr.sgi.com (schroedinger.engr.sgi.com [163.154.5.55]) by spindle.corp.sgi.com (SGI-8.12.5/8.12.9/generic_config-1.2) with ESMTP id j9DMcasT95548564 for ; Thu, 13 Oct 2005 15:38:36 -0700 (PDT) Message-ID: <434EDDCA.9010001@austin.ibm.com> Date: Thu, 13 Oct 2005 17:20:58 -0500 From: Joel Schopp Reply-To: jschopp@austin.ibm.com MIME-Version: 1.0 Subject: Re: [Lhms-devel] [PATCH] Page eviction support in vmscan.c References: In-Reply-To: Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit ReSent-To: linux-mm@kvack.org ReSent-Message-ID: Sender: owner-linux-mm@kvack.org Return-Path: To: Christoph Lameter Cc: lhms-devel@lists.sourceforge.net, linux-mm@vger.kernel.org, akpm@osdl.org List-ID: > This patch adds functions that allow the eviction of pages to swap space. > Page eviction may be useful to migrate pages, suspend programs or for > ummapping single pages (useful for faulty pages or pages with soft ECC > failures) I'm curious what use motivated you to write it. I think for migration it would usually make more sense to let the swapper free up LRU memory and then do a memory to memory migration. But I'm not really a migration expert > swapout_pages does its best to swapout the pages and does multiple passes over the list. > However, swapout_pages may not be able to evict all pages for a variety of reasons. Have you thought about using this in combination with the fragmentation avoidance patches Mel has been posting? __GFP_USER flag that adds would go a long way toward determining what can and can't be swapped out. We use that for migration with great success. I'd assume the criteria for swapout and migration are pretty similar. > /* > + * Swapout evicts the pages on the list to swap space. > + * This is essentially a dumbed down version of shrink_list Have you thought about reusing code from shrink list without duplicating it? That is a whole lot of duplicated code to maintain twice. > + if (PageDirty(page)) { > + /* Page is dirty, try to write it out here */ > + switch(pageout(page, mapping)) { > + case PAGE_KEEP: > + case PAGE_ACTIVATE: > + goto retry_later_locked; > + case PAGE_SUCCESS: > + goto retry_later; > + case PAGE_CLEAN: > + ; /* try to free the page below */ > + } > + } Tabs vs spaces? > + > + list_del(&page->lru); > + unlock_page(page); > + put_page(page); > + continue; Tabs vs spaces? -- 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