From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 26690 invoked by alias); 2 Oct 2006 00:18:10 -0000 Received: (qmail 26682 invoked by uid 22791); 2 Oct 2006 00:18:09 -0000 X-Spam-Status: No, hits=-2.3 required=5.0 tests=AWL,BAYES_00,FORGED_RCVD_HELO,TW_CP X-Spam-Check-By: sourceware.org Received: from tomts13.bellnexxia.net (HELO tomts13-srv.bellnexxia.net) (209.226.175.34) by sourceware.org (qpsmtpd/0.31) with ESMTP; Mon, 02 Oct 2006 00:18:05 +0000 Received: from krystal.dyndns.org ([65.95.39.115]) by tomts13-srv.bellnexxia.net (InterMail vM.5.01.06.13 201-253-122-130-113-20050324) with ESMTP id <20061002001801.FMVU29052.tomts13-srv.bellnexxia.net@krystal.dyndns.org> for ; Sun, 1 Oct 2006 20:18:01 -0400 Received: from localhost (localhost [127.0.0.1]) (uid 1000) by krystal.dyndns.org with local; Sun, 01 Oct 2006 20:07:32 -0400 id 00321904.45205844.000073E0 Date: Mon, 02 Oct 2006 00:18:00 -0000 From: Mathieu Desnoyers To: Nicholas Miell Cc: Martin Bligh , "Frank Ch. Eigler" , Masami Hiramatsu , prasanna@in.ibm.com, Andrew Morton , Ingo Molnar , Paul Mundt , linux-kernel , Jes Sorensen , Tom Zanussi , Richard J Moore , Michel Dagenais , Christoph Hellwig , Greg Kroah-Hartman , Thomas Gleixner , William Cohen , ltt-dev@shafik.org, systemtap@sources.redhat.com, Alan Cox , Jeremy Fitzhardinge , Karim Yaghmour , Pavel Machek , Joe Perches , "Randy.Dunlap" , "Jose R. Santos" Subject: Re: Performance analysis of Linux Kernel Markers 0.20 for 2.6.17 Message-ID: <20061002000731.GA22337@Krystal> References: <20060930180157.GA25761@Krystal> <1159642933.2355.1.camel@entropy> <20061001034212.GB13527@Krystal> <1159676382.2355.13.camel@entropy> <20061001153317.GB24313@Krystal> <1159747060.2355.21.camel@entropy> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline In-Reply-To: <1159747060.2355.21.camel@entropy> X-Editor: vi X-Info: http://krystal.dyndns.org:8080 X-Operating-System: Linux/2.4.32-grsec (i686) X-Uptime: 20:05:28 up 39 days, 21:14, 4 users, load average: 0.21, 0.23, 0.24 User-Agent: Mutt/1.5.13 (2006-08-11) X-IsSubscribed: yes Mailing-List: contact systemtap-help@sourceware.org; run by ezmlm Precedence: bulk List-Subscribe: List-Post: List-Help: , Sender: systemtap-owner@sourceware.org X-SW-Source: 2006-q4/txt/msg00004.txt.bz2 * Nicholas Miell (nmiell@comcast.net) wrote: > On Sun, 2006-10-01 at 11:33 -0400, Mathieu Desnoyers wrote: > > * Nicholas Miell (nmiell@comcast.net) wrote: > > > On Sat, 2006-09-30 at 23:42 -0400, Mathieu Desnoyers wrote: > > > > * Nicholas Miell (nmiell@comcast.net) wrote: > > > > > > > > > > Has anyone done any performance measurements with the "regular function > > > > > call replaced by a NOP" type of marker? > > > > > > > > > > > > > Here it is (on the same setup as the other tests : Pentium 4, 3 GHz) : > > > > > > > > * Execute an empty loop > > > > > > > > - Without marker > > > > NR_LOOPS : 10000000 > > > > time delta (cycles): 15026497 > > > > cycles per loop : 1.50 > > > > > > > > - With 5 NOPs > > > > NR_LOOPS : 100000 > > > > time delta (cycles): 300157 > > > > cycles per loop : 3.00 > > > > added cycles per loop for nops : 3.00-1.50 = 1.50 > > > > > > > > > > > > * Execute a loop of memcpy 4096 bytes > > > > > > > > - Without marker > > > > NR_LOOPS : 10000 > > > > time delta (cycles): 12981555 > > > > cycles per loop : 1298.16 > > > > > > > > - With 5 NOPs > > > > NR_LOOPS : 10000 > > > > time delta (cycles): 12983925 > > > > cycles per loop : 1298.39 > > > > added cycles per loop for nops : 0.23 > > > > > > > > > > > > If we compare this approach to the jump-over-call markers (in cycles per loop) : > > > > > > > > NOPs Jump over call generic Jump over call optimized > > > > empty loop 1.50 1.17 2.50 > > > > memcpy 0.23 2.12 0.07 > > > > > > > > > > > > > > > > Mathieu > > > > > > What about with two NOPs (".byte 0x66, 0x66, 0x90, 0x66, 0x90" - this > > > should work with everything) or one (".byte 0x0f, 0x1f, 0x44, 0x00, > > > 0x00" - AFAIK, this should work with P6 or newer). > > > > > > (Sorry, I should have mentioned this the first time.) > > > > > > > Hi, > > > > The tests I made were with : > > #define GENERIC_NOP1 ".byte 0x90\n" > > #define GENERIC_NOP4 ".byte 0x8d,0x74,0x26,0x00\n" > > #define GENERIC_NOP5 GENERIC_NOP1 GENERIC_NOP4 > > > > Now with the tests you ask for : > > > > * Execute an empty loop > > - 2 NOPs ".byte 0x66, 0x66, 0x90, 0x66, 0x90" > > NR_LOOPS : 100000 > > time delta (cycles): 200190 > > cycles per loop : 2.00 > > cycles per loop for nops : 2.00-1.50 = 0.50 > > > > - 1 NOP "0x0f, 0x1f, 0x44, 0x00, 0x00" > > NR_LOOPS : 100000 > > time delta (cycles): 300172 > > cycles per loop : 3.00 > > cycles per loop for nops : 3.00-1.50 = 2.50 > > > > > > * Execute a loop of memcpy 4096 bytes > > - 2 NOPs ".byte 0x66, 0x66, 0x90, 0x66, 0x90" > > NR_LOOPS : 10000 > > time delta (cycles): 12981293 > > cycles per loop : 1298.13 > > cycles per loop for nops : 1298.16-1298.13=0.03 > > > > - 1 NOP "0x0f, 0x1f, 0x44, 0x00, 0x00" > > NR_LOOPS : 10000 > > time delta (cycles): 12985590 > > cycles per loop : 1298.56 > > cycles per loop for nops : 0.43 > > > > To summarize in chart form: > > JoC JoCo 2NOP 1NOP > empty loop 1.17 2.50 0.50 2.50 > memcpy 2.12 0.07 0.03 0.43 > > JoC = Jump over call - generic > JoCo = Jump over call - optimized > 2NOP = "data16 data16 nop; data16 nop" > 1NOP = NOP with ModRM > > I left out your "nop; lea 0(%esi), %esi" because it isn't actually a NOP > (the CPU will do actual work even if it has no effect, and on AMD64, > that insn is "nop; lea 0(%rdi), %esi", which will truncate RDI+0 to fit > 32-bits.) > > The performance of NOP with ModRM doesn't suprise me -- AFAIK, only the > most recent of Intel CPUs actually special case that to be a true > no-work-done NOP. > > It'd be nice to see the results of "jump to an out-of-line call with the > jump replaced by a NOP", but even if it performs well (and it should, > the argument passing and stack alignment overhead won't be executed in > the disabled probe case), actually using it in practice would be > difficult without compiler support (call instructions are easy to find > thanks to their relocations, which local jumps don't have). > Hi, Just to make sure we see things the same way : the JoC approach is similar to the out-of-line call in that the argument passing and stack alignment are not executed when the probe is disabled. Mathieu OpenPGP public key: http://krystal.dyndns.org:8080/key/compudj.gpg Key fingerprint: 8CD5 52C3 8E3C 4140 715F BA06 3F25 A8FE 3BAE 9A68