From: Richard McRoberts <rdm@bamboo.verinet.com> Subject: Re: [PATCH] Re: 2.1.xxx makes Electric Fence 22x slower To: haible@ilog.fr Date: Thu, 3 Sep 1998 19:27:01 -0600 (MDT) Bruno's patch gives excellent performance in my Electric Fence example. Here is a summary of all the results to date. They include two fuzzy hash patches from Andrey V. Savochkin, and today's AVL Tree patch from Bruno Haible. I have continued to perform the tests as consistently as possible -- immediately after a fresh reboot in each case. The two numbers in each line are elapsed times in seconds for the application linked (1) without Electric Fence, and (2) with Electric Fence, respectively. 2.0.34 : 3.82, 14.61 2.1.117, no patches : 3.71, 325.85 2.1.111, 1st fuzzy hash patch : 3.76, 29.73 2.1.119, 2nd fuzzy hash patch : 3.71, 20.88 2.1.119, AVL Tree patch : 3.73, 13.93 The decimal digits shown are not all significant, so please don't infer too much precision from them. I am aware that the question, of how (or possibly whether?) this problem should be solved, is controversial. Certainly it isn't my intention to be inflammatory in presenting this data. I'm just an interested onlooker (and user) who is grateful for the willingness and expertise you guys are bringing to this issue. Thanks, Richard D. McRoberts Loveland, Colorado USA rdm@verinet.com - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/faq.html