This is the mail archive of the
systemtap@sourceware.org
mailing list for the systemtap project.
Re: [PATCH 0/11] LTTng-core (basic tracing infrastructure) 0.5.108
- From: Ingo Molnar <mingo at elte dot hu>
- To: Vara Prasad <prasadav at us dot ibm dot com>
- Cc: Ingo Molnar <mingo at redhat dot com>, Michel Dagenais <michel dot dagenais at polymtl dot ca>, Tom Zanussi <zanussi at us dot ibm dot com>, systemtap <systemtap at sourceware dot org>, Martin Hunt <hunt at redhat dot com>
- Date: Tue, 19 Sep 2006 11:58:56 +0200
- Subject: Re: [PATCH 0/11] LTTng-core (basic tracing infrastructure) 0.5.108
- References: <1158348954.5724.481.camel@localhost.localdomain> <450B0585.5070700@opersys.com> <1158351780.5724.507.camel@localhost.localdomain> <Pine.LNX.4.64.0609152236010.6761@scrub.home> <20060915204812.GA6909@elte.hu> <Pine.LNX.4.64.0609152314250.6761@scrub.home> <20060915215112.GB12789@elte.hu> <Pine.LNX.4.64.0609160018110.6761@scrub.home> <20060915231419.GA24731@elte.hu> <450FA531.7070504@us.ibm.com>
* Vara Prasad <prasadav@us.ibm.com> wrote:
> Ingo, can you please give some details of what you like in LTT ring
> buffer implementation. In systemtap we use standard relay interface
> from mainline kernel for bulk data transfer. If there is something
> that is good in LTT we certainly would like to use it.
yeah - but that code originally came from LTT (via a number of
cleanups). So yes, kernel/relayfs.c is the right one i think - and I
supported its inclusion strongly when Andrew was unwilling to merge it
due to "no in-kernel users". So i just wanted to highlight that my
opinion is not against LTT or tracing - in fact the opposite is true.
Ingo