[lttng-dev] [BABELTRACE PATCH] Use default if there is no clock declaration
Mathieu Desnoyers
mathieu.desnoyers at efficios.com
Mon Feb 13 09:26:42 EST 2012
* Mathieu Desnoyers (mathieu.desnoyers at efficios.com) wrote:
> * Juha Niskanen (juniskane at gmail.com) wrote:
> > In absence of clock description in metadata, use a default clock
> > source which increments once per nanosecond.
> >
> > Without this patch, the converter crashes when given old traces
> > that are lacking clock description block.
>
> Merged (with tiny cosmetic fixes). Thanks !
FYI, with commit:
commit 02dc4610b1dccefaf545ba77378857ee6188ea33
Author: Mathieu Desnoyers <mathieu.desnoyers at efficios.com>
Date: Mon Feb 13 09:03:55 2012 -0500
lib: open trace error handling fix
When a trace from a trace collection path fails, consider this as a hard
error. Print errors to stderr rather than stdout.
The end goal is to never show an output if any of the trace specified
(or trace within a trace collection) fails to open.
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers at efficios.com>
opening unrelated traces into the same trace context will fail, because
those trace cannot be correlated. This will therefore not allow
Babeltrace to view LTTng-UST/LTTng-modules traces taken from older
2.0 prereleases.
I am adding a --clock-force-correlate to allow the user to force
babeltrace to accept those older traces, letting babeltrace assume that
correlation is OK. See:
commit 82ace6d6d97ecb15d093232c76c5d4a3c31e6948
Author: Mathieu Desnoyers <mathieu.desnoyers at efficios.com>
Date: Mon Feb 13 09:26:53 2012 -0500
Add --clock-force-correlate option
For support of older 2.0-pre traces.
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers at efficios.com>
Thanks,
Mathieu
>
> Mathieu
>
> >
> > Signed-off-by: Juha Niskanen <juniskane at gmail.com>
> > ---
> > .../ctf/metadata/ctf-visitor-generate-io-struct.c | 29 +++++++++++++++----
> > 1 files changed, 23 insertions(+), 6 deletions(-)
> >
> > diff --git a/formats/ctf/metadata/ctf-visitor-generate-io-struct.c b/formats/ctf/metadata/ctf-visitor-generate-io-struct.c
> > index 2d3361c..8de2cae 100644
> > --- a/formats/ctf/metadata/ctf-visitor-generate-io-struct.c
> > +++ b/formats/ctf/metadata/ctf-visitor-generate-io-struct.c
> > @@ -2302,6 +2302,20 @@ error:
> > }
> >
> > static
> > +void clock_default_monotonic(struct ctf_trace *trace)
> > +{
> > + struct ctf_clock *clock;
> > +
> > + clock = g_new0(struct ctf_clock, 1);
> > + /* Default clock frequency is set to 1000000000 */
> > + clock->freq = 1000000000ULL;
> > + clock->name = g_quark_from_string("monotonic");
> > +
> > + trace->single_clock = clock;
> > + g_hash_table_insert(trace->clocks, (gpointer) (unsigned long) clock->name, clock);
> > +}
> > +
> > +static
> > void clock_free(gpointer data)
> > {
> > struct ctf_clock *clock = data;
> > @@ -2516,12 +2530,15 @@ retry:
> > * declarations need to query clock hash table,
> > * so clock need to be treated first.
> > */
> > - cds_list_for_each_entry(iter, &node->u.root.clock, siblings) {
> > - ret = ctf_clock_visit(fd, depth + 1, iter,
> > - trace);
> > - if (ret) {
> > - fprintf(fd, "[error] %s: clock declaration error\n", __func__);
> > - goto error;
> > + if (cds_list_empty(&node->u.root.clock))
> > + clock_default_monotonic(trace);
> > + else {
> > + cds_list_for_each_entry(iter, &node->u.root.clock, siblings) {
> > + ret = ctf_clock_visit(fd, depth + 1, iter, trace);
> > + if (ret) {
> > + fprintf(fd, "[error] %s: clock declaration error\n", __func__);
> > + goto error;
> > + }
> > }
> > }
> > env_clock_done = 1;
> > --
> > 1.7.4.1
> >
> >
> > _______________________________________________
> > lttng-dev mailing list
> > lttng-dev at lists.lttng.org
> > http://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev
>
> --
> Mathieu Desnoyers
> Operating System Efficiency R&D Consultant
> EfficiOS Inc.
> http://www.efficios.com
>
> _______________________________________________
> lttng-dev mailing list
> lttng-dev at lists.lttng.org
> http://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev
--
Mathieu Desnoyers
Operating System Efficiency R&D Consultant
EfficiOS Inc.
http://www.efficios.com
More information about the lttng-dev
mailing list