[lttng-dev] Sub-buffer multiplicity on multi-core systems

[Jérémie Galarneau]

On Mon, Oct 7, 2013 at 5:28 PM, Thibault, Daniel
<Daniel.Thibault at drdc-rddc.gc.ca> wrote:
>    I understand that each processor can be understood as generating its own event stream (for a given event).  Assume a quad-core system.  At which point are the four event streams merged into one?  What I mean is: is the ring buffer which represents a kernel channel for a given session (for example) actually four ring buffers (one per CPU), or just one?  I suspect the merge occurs early (at the LTTng kernel module level), but I want to be sure.  (I understand that the four streams remain distinct in the logical CTF sense: what I'm wondering about here is the sub-buffer multiplicity)
>

In kernel tracing, ring buffers are separate per CPU, per channel.
Each ring buffer is flushed into its own CTF data stream file. These
streams are only "merged" by Babeltrace (or any other reader) while
reading the trace.

Regards,
Jérémie

> Daniel U. Thibault
> Protection des systèmes et contremesures (PSC) | Systems Protection & Countermeasures (SPC)
> Cyber sécurité pour les missions essentielles (CME) | Mission Critical Cyber Security (MCCS)
> R & D pour la défense Canada - Valcartier (RDDC Valcartier) | Defence R&D Canada - Valcartier (DRDC Valcartier)
> 2459 route de la Bravoure
> Québec QC  G3J 1X5
> CANADA
> Vox : (418) 844-4000 x4245
> Fax : (418) 844-4538
> NAC : 918V QSDJ <http://www.travelgis.com/map.asp?addr=918V%20QSDJ>
> Gouvernement du Canada | Government of Canada
> <http://www.valcartier.drdc-rddc.gc.ca/>
>
> _______________________________________________
> lttng-dev mailing list
> lttng-dev at lists.lttng.org
> http://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev



-- 
Jérémie Galarneau
EfficiOS Inc.
http://www.efficios.com