Skip to main content
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
[linuxtools-dev] Re: [dsdp-tcf-dev] Standard protocols/interfaces/formats for performance tools (TCF, LTTng, ...) 
> The focus currently

is using real-time trace data as the event source for analysis.  In the
future this will expand as we desire to be able to correlate analysis of
heterogeneous systems, e.g. embedded Linux + LTTng events on one
machine, correlated with real-time trace data collected from a bunch of
DSP's.
A posteriori traces synchronization is an important problem in such a setup. Benjamin Poirier contributed a very nice procedure to synchronize traces in the latest LTTV. 


-What protocols are currently in use that we might consider as a
starting point?  I see that the linuxtools project apparently has one
for transferring LTTng event data.  Are there any docs for this
protocol?
We extended TCF (Michael Sills-Lavoie) to handle raw binary data, offer zero copy transfer and accept plugin modules, and we built a module to list and activate tracepoints. This has been contributed and integrated into TCF and should be the base for LTTng and Eclipse Linux Tools 


-file formats: event log file formats is another obvious candidate for
standardization.  Mentor has a file format we use that was inspired by
LTTng's format but is optimized for extremely large real-time trace
logs.  I intend to throw this into the mix.  Any others we should think
about? (The LTTng format obviously...)
A number of formats are optimized for system independance, ease of parsing and other such considerations. However, for something like extremely large traces, an XML type format is obviously unsuitable. You want to minimize the overhead at trace creation time and have something very compact. If you have improvements over LTTng's format for very large traces, we could look and take the best of both!

Back to the top