From mboxrd@z Thu Jan 1 00:00:00 1970 Date: Mon, 9 Oct 2000 20:47:51 +0200 (CEST) From: Ingo Molnar Reply-To: mingo@elte.hu Subject: Re: [PATCH] VM fix for 2.4.0-test9 & OOM handler In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-linux-mm@kvack.org Return-Path: To: Rik van Riel Cc: Marco Colombo , MM mailing list , linux-kernel@vger.kernel.org, Linus Torvalds List-ID: On Mon, 9 Oct 2000, Rik van Riel wrote: > In that case the time the process has been running and the > CPU time used will save the process if it's been running for > a long time. 'importance' is not something we can measure reliably within the kernel. And assuming that a niced, not long-running process is unimportant misses the bus as well. What if i just started an important simulation before going to vacation for two weeks? > would you really care if a simulation would be killed after > 5 minutes? [...] yes, i would. I would probably end up not using nice values. Please, Rik, dont penalize an unrelated kernel feature! > [...] The objective is to destroy the least amount of work, which > means giving a bonus to processes which have used a lot of CPU time > already ... regardless of nice value. your OOM code does not follow this objective: + /* + * Niced processes are most likely less important, so double + * their badness points. + */ + if (p->nice > 0) + points *= 2; Niced processes *can be just as important*. > If you have a better algorithm, feel free to send patches. yes. Please remove the above part. Ingo -- 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.eu.org/Linux-MM/