This is the mail archive of the
systemtap@sourceware.org
mailing list for the systemtap project.
Re: Performance analysis of Linux Kernel Markers 0.20 for 2.6.17
- From: Mathieu Desnoyers <compudj at krystal dot dyndns dot org>
- To: Nicholas Miell <nmiell at comcast dot net>
- Cc: Martin Bligh <mbligh at google dot com>, "Frank Ch. Eigler" <fche at redhat dot com>, Masami Hiramatsu <masami dot hiramatsu dot pt at hitachi dot com>, prasanna at in dot ibm dot com, Andrew Morton <akpm at osdl dot org>, Ingo Molnar <mingo at elte dot hu>, Paul Mundt <lethal at linux-sh dot org>, linux-kernel <linux-kernel at vger dot kernel dot org>, Jes Sorensen <jes at sgi dot com>, Tom Zanussi <zanussi at us dot ibm dot com>, Richard J Moore <richardj_moore at uk dot ibm dot com>, Michel Dagenais <michel dot dagenais at polymtl dot ca>, Christoph Hellwig <hch at infradead dot org>, Greg Kroah-Hartman <gregkh at suse dot de>, Thomas Gleixner <tglx at linutronix dot de>, William Cohen <wcohen at redhat dot com>, ltt-dev at shafik dot org, systemtap at sources dot redhat dot com, Alan Cox <alan at lxorguk dot ukuu dot org dot uk>, Jeremy Fitzhardinge <jeremy at goop dot org>, Karim Yaghmour <karim at opersys dot com>, Pavel Machek <pavel at suse dot cz>, Joe Perches <joe at perches dot com>, "Randy.Dunlap" <rdunlap at xenotime dot net>, "Jose R. Santos" <jrs at us dot ibm dot com>
- Date: Sat, 30 Sep 2006 23:42:12 -0400
- Subject: Re: Performance analysis of Linux Kernel Markers 0.20 for 2.6.17
- References: <20060930180157.GA25761@Krystal> <1159642933.2355.1.camel@entropy>
* Nicholas Miell (nmiell@comcast.net) wrote:
> On Sat, 2006-09-30 at 14:01 -0400, Mathieu Desnoyers wrote:
> > Hi,
> >
> > Following the huge discussion thread about tracing/static vs dynamic
> > instrumentation/markers, a consensus seems to emerge about the need for a
> > marker system in the Linux kernel. The main issues this mechanism addresses are:
> >
> > - Identify code important to runtime data collection/analysis tools in tree so
> > that it follows the code changes naturally.
> > - Be visually appealing to kernel developers.
> > - Have a very low impact on the system performance.
> > - Integrate in the standard kernel infrastructure : use C and loadable modules.
> >
> > The time has come for some performance measurements of the Linux Kernel Markers,
> > which follows. I attach a PDF with tables and charts which condense these
> > results.
>
> 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
OpenPGP public key: http://krystal.dyndns.org:8080/key/compudj.gpg
Key fingerprint: 8CD5 52C3 8E3C 4140 715F BA06 3F25 A8FE 3BAE 9A68