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

[Thibault, Daniel]

> >    Thanks for confirming the 1:1 correspondence between channels and files.
> >    I wanted to make sure I understood the ins and outs properly.
>
> Not quite,
>
> It's a 1:1 correspondence between ring buffer and file.
> 1:N correspondence between channel and ring buffers. (typically, N being the number of cpus) (therefore 1:N correspondence between channel and files)
> 1:M correspondence between ring buffer and sub-buffers. (M being the number of subbuffers per buffers, configurable on a per-channel basis)
> 1:1 correspondence between sub-buffer and CTF trace packet.
>
> Mathieu

   I meant 1:1 correspondence between channel+cpu and file, as you said (I should have stayed in bed this morning :-) ).

   In a situation where all user-space events are being traced per UID, aren't the ring buffers further duplicated if there are both 32- and 64-bit events?  It would seem so since you will get both a session/ust/uid/1000/64-bit/channel0_0 file and a session/ust/uid/1000/32-bit/channel0_0 file.

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/>