From mboxrd@z Thu Jan 1 00:00:00 1970 Date: Mon, 22 Mar 2004 01:46:52 +0100 From: Andrea Arcangeli Subject: Re: [RFC][PATCH 1/3] radix priority search tree - objrmap complexity fix Message-ID: <20040322004652.GF3649@dualathlon.random> References: Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: Sender: owner-linux-mm@kvack.org Return-Path: To: Rajesh Venkatasubramanian Cc: akpm@osdl.org, torvalds@osdl.org, hugh@veritas.com, mbligh@aracnet.com, riel@redhat.com, mingo@elte.hu, linux-kernel@vger.kernel.org, linux-mm@kvack.org List-ID: On Sun, Mar 21, 2004 at 05:10:45PM -0500, Rajesh Venkatasubramanian wrote: > http://marc.theaimsgroup.com/?l=linux-kernel&m=107966438414248 > > Andrea says the system may hang, however, in this case system > does not hang. It's a live lock, not a deadlock. I didn't wait more than a few minutes every time before declaring the kernel broken and rebooting the machine. still if the prio_tree fixed my problem it means at the very least it reduced the contention on the locks a lot ;) It would be curious to test it after changing the return 1 to return 0 in the page_referenced trylock failures? the results looks great, thanks. what about the cost of a tree rebalance, is that O(log(N)) like with the rbtrees? -- 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: aart@kvack.org