Re: LTTng for SH architecture
Mathieu Desnoyers <[email protected]>
| Newsgroups | gmane.linux.kernel.tracing |
|---|---|
| Message-ID | <20070614150333.GA19903@Krystal> |
* Giuseppe Cavallaro ([email protected]) wrote: > Thanks Mathieu, > so I wonder if it's worth while continuing my work with the LTTng > 0.6.0version on the kernel > 2.6.17. > Indeed, looking at the latest patches (0.9.5) for the kernel 2.6.21 I've > seen that many pieces, of the LTT code, have been changed (for example, I've > found no LTT_FACILITY_CHECKSUM macro). > Applying the LTTng 0.9.5 patches to the 2.6.17 kernel, it's likely I'd > verify too many things (relay FS, common structures) etc. So I'd spend a > little while. > What do you think? > Well, in the long run, the best thing would be to work on a recent kernel to do the initial SH4 port, so it could be added to the current LTTng patches, and then do a backport of what you need to 2.6.17. You can refer to a few emails in the ltt-dev ML about what pieces must be taken from the recent kernel to the older ones for a correct backport. The advantage of this approach is that your work will afterward be maintained with the rest of LTTng. Mathieu > Giuseppe > > On 14/06/07, Mathieu Desnoyers <[email protected]> wrote: > > > >* Giuseppe Cavallaro ([email protected]) wrote: > >> Hi All, > >> I'd like to use LTTng on my kernel 2.6.17 (SH4 architecture). > >> Following the compatibility table published on the web, I've applied the > >> LTTng 0.6.0 patches on my kernel. > >> So I've started porting LTT on the SH4 but I've met some problems > >because > >> I'm not familiar with LTTng internals. > >> Looking at the new LTTng 0.9.5 version (for the kernel 2.6.21) and I've > >seen > >> there is a "first porting" on SH. > >> Where can I get more information about the LTTng internals? > >> Welcome advice. > >> > > > >Hi Giuseppe, > > > >See the Documentation sectino of the ltt.polymtl.ca website. > > > >I recommend that you look at the "lttng instrumentation" patches for > >each architectures. You'll find principally that include/asm-$ARCH/ltt.h > >hold the timestamping mechanism for each architecture (does SH4 have a > >64 or 32 bits timestamp counter ? (get_cycles())) > > > >You will also find various instrumentation points that are architecture > >specific that must be ported to SH4. > > > >I guess looking into other architecture's patches is the best way for > >you to learn how to do this port. > > > >If you run into problem please tell us. Also, it would be useful to have > >this port for the LTTng head so it can be integrated in the next > >releases. > > > >Thanks, > > > >Mathieu > > > >> Giuseppe > > > >> _______________________________________________ > >> Ltt-dev mailing list > >> [email protected] > >> http://listserv.shafik.org/mailman/listinfo/ltt-dev > > > > > >-- > >Mathieu Desnoyers > >Computer Engineering Ph.D. Student, Ecole Polytechnique de Montreal > >OpenPGP key fingerprint: 8CD5 52C3 8E3C 4140 715F BA06 3F25 A8FE 3BAE > >9A68 > > > _______________________________________________ > Ltt-dev mailing list > [email protected] > http://listserv.shafik.org/mailman/listinfo/ltt-dev -- Mathieu Desnoyers Computer Engineering Ph.D. Student, Ecole Polytechnique de Montreal OpenPGP key fingerprint: 8CD5 52C3 8E3C 4140 715F BA06 3F25 A8FE 3BAE 9A68