Re: linux 2.6.22.7

Mathieu Desnoyers <[email protected]> Thu, 20 Dec 2007 10:22:11 -0500
Newsgroups gmane.linux.kernel.tracing
Message-ID <20071220152211.GF22523@Krystal>
* karthikeyan.s ([email protected]) wrote:
> Hi again,
> Followed the instructions as such from quickstart guide. By the way, i have
> an arm core here.
> 
> ltt-control-0.43-10082007
> lttv --> 0.8.82
> lttng patch --> 0.9.5
> 

The versions seems compatible. I would look into LTTng 0.9.6 to see what
changed if I were you.

> Here is the message form dmesg
> LTT : ltt-facilities init
> LTT : Facility kernel_arch registered with id 86
> LTT : Facility kernel registered with id 211
> LTT unable to register probe kernel_sched_wait_task
> LTT unable to register probe kernel_sched_migrate_task

This is wrong. Is the instrumentation correct. I guess it's the "kernel"
probe that has missing ifdef CONFIG_SMP, and you are on UP. Either add
the ifdefs in ltt/probes/ltt-probe-kernel.c or remove those probe
registration.

> LTT : Facility list registered with id 145
> LTT : ltt-relay init
> ltt-control init
> LTT : ltt-facility-statedump init
> 
> ltt-control ltt_control_input
> ltt_control : trace trace
> Creating trace trace
> ltt-control ltt_control_input
> ltt_control : trace trace
> Start tracing trace
> Dumping facility kernel_arch
> Dumping facility list
> Dumping facility kernel
> ltt_statedump_start
> do_ltt_statedump
> do_ltt_statedump end
> ----
> ltt-control ltt_control_input
> ltt_control : trace trace
> Stop tracing trace
> ltt-control ltt_control_input
> ltt_control : trace trace
> Destroying trace trace
> LTT : cpu : 22954 events lost in cpu channel (cpu 0).

Hrm, this is wrong too. Too many events ? Too small buffers. Hrm.. On
ARM, the timestamping has changed since 2.6.21. You might want to look
at current include/asm-arm/ltt.h and see the diff.

> --------
> When i run textdump
> 
> lttv -m textDump -t /tmp/trace
> 
> I get
> ---
> ** (process:25057): WARNING **: Unknown facility 145 (0x91) in tracefile
> /control/network
> 
> ** (process:25057): WARNING **: Unknown facility 145 (0x91) in tracefile
> /control/processes
> 
> ** (process:25057): WARNING **: Unknown facility 211 (0xd3) in tracefile
> /control/modules
> 
> ** (process:25057): WARNING **: Unknown facility 145 (0x91) in tracefile
> /control/interrupts
> 
> ** (process:25057): WARNING **: Unknown facility 211 (0xd3) in tracefile
> /cpu
> Trace set contains 1 traces
> 
> 
> ** (process:25057): WARNING **: Unknown facility 145 (0x91) in tracefile
> /control/network
> 
> ** (process:25057): WARNING **: Unknown facility 145 (0x91) in tracefile
> /control/processes
> 
> ** (process:25057): WARNING **: Unknown facility 211 (0xd3) in tracefile
> /control/modules
> 
> ** (process:25057): WARNING **: Unknown facility 145 (0x91) in tracefile
> /control/interrupts
> 
> ** (process:25057): WARNING **: Unknown facility 211 (0xd3) in tracefile
> /cpu
> 
> ** ERROR **: file event.c: line 86 (ltt_event_facility): assertion failed:
> (facility->exists)

Could be caused by the "unable to register probe" messages, caused by
incorrect CONFIG_SMP ifdefs in probes.

> aborting...
> /usr/local/bin/lttv: line 15: 25057 Aborted                 $0.real $*
> ------
> 
> Any help is appreciated.
> 
> Thanks,
> karthik
> 
> On 12/17/07, karthikeyan. s <[email protected]> wrote:
> >
> >
> > Hi mathieu,
> > Thanks for the reply.
> >
> > I patched the 2.6.22.7 kernel with 2.6.21 patch.Patching was not an issue
> > but getting it to run is a problem. And i did not have to get any file from
> > the 2.6.22-mm* patches (most probable cause? ).
> >
> > Should I take a 2.6.22 or 2.6.23 patch and sort things out or use 2.6.21and go ahead?
> >
> > Thanks again!
> > karthik
> >
> >
> > On Nov 16, 2007 7:36 PM, Mathieu Desnoyers <[email protected]>
> > wrote:
> >
> > >  * karthikeyan.s ([email protected]) wrote:
> > > > Hi,
> > > > I am required to use 2.6.22.7 kernel.
> > > >
> > > > There are a bunch of patches available. But none seem to be for the
> > > mainline
> > > > kernel.
> > > >
> > > > Can you please suggest which patch to use, as i have already suffered
> > > a lot
> > > > on using a patch and reverting back a number of times
> > > >
> > > > 2.6.22-rc2-mm1
> > > > 2.6.22-rc4-mm2
> > > > 2.6.22.1-rt4
> > > > 2.6.22-rc2-mm1
> > > >
> > >
> > > There is no LTTng specifically for 2.6.22, a port will be needed. (which
> > > means fixing the rejects, grabbing some files from the -mm patches...)
> > >
> > > Mathieu
> > >
> > > > Thanks!
> > > > -karthik
> > >
> > > > _______________________________________________
> > > > 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
> > >
> >
> >
> >
> > --
> > ---
> > S. Karthikeyan | +919980814745
> > ---
> 
> 
> 
> 
> -- 
> ---
> S. Karthikeyan | +919980814745
> ---

> _______________________________________________
> 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