Re: [PULL 1/2] perf/tool: Fix usage of trace events with '-' in trace system name.

[Jiri Olsa]

On Wed, Apr 16, 2014 at 03:31:15PM +0200, Christian Borntraeger wrote:
> On 16/04/14 15:28, Christian Borntraeger wrote:
> > On 16/04/14 15:00, Jiri Olsa wrote:
> >> On Wed, Apr 09, 2014 at 04:21:58PM +0200, Christian Borntraeger wrote:
> >>> From: Alexander Yarygin <yarygin@xxxxxxxxxxxxxxxxxx>
> >>>
> >>> Trace events potentially can have a '-' in their trace system name,
> >>> e.g. kvm on s390 defines kvm-s390:* tracepoints.
> >>> tools/perf could not parse them, because there was no rule for this:
> >>> $ sudo ./perf top -e "kvm-s390:*"
> >>> invalid or unsupported event: 'kvm-s390:*'
> >>>
> >>> This patch allows '-' to be a part of PE_NAME token, so tracepoints
> >>> with '-' can be parsed by the event_legacy_tracepoint rule.
> >>> Without the patch, perf will not accept such tracepoints in the -e
> >>> option.
> >>>
> >>> Signed-off-by: Alexander Yarygin <yarygin@xxxxxxxxxxxxxxxxxx>
> >>> Signed-off-by: Christian Borntraeger <borntraeger@xxxxxxxxxx>
> >>> ---
> >>>  tools/perf/util/parse-events.l |    2 +-
> >>>  1 file changed, 1 insertion(+), 1 deletion(-)
> >>>
> >>> diff --git a/tools/perf/util/parse-events.l b/tools/perf/util/parse-events.l
> >>> index 3432995..ca20da7 100644
> >>> --- a/tools/perf/util/parse-events.l
> >>> +++ b/tools/perf/util/parse-events.l
> >>> @@ -199,7 +199,7 @@ r{num_raw_hex}		{ return raw(yyscanner); }
> >>>  {num_hex}		{ return value(yyscanner, 16); }
> >>>  
> >>>  {modifier_event}	{ return str(yyscanner, PE_MODIFIER_EVENT); }
> >>> -{name}			{ return str(yyscanner, PE_NAME); }
> >>> +{name_minus}			{ return str(yyscanner, PE_NAME); }
> >>>  "/"			{ BEGIN(config); return '/'; }
> >>>  -			{ return '-'; }
> >>>  ,			{ BEGIN(event); return ','; }
> >>
> >> this breaks parsing of cache events like:
> >>
> >>   $ perf record -e 'L1-dcache-loads' ls
> >>
> >> also test 10 (same issue):
> >>   $ ./perf test 10
> >>   10: roundtrip evsel->name check                            : FAILED!
> >>
> >>
> >> it might be little tricky to fix, let me know if you
> >> have any troubles with that, I could look on it
> > 
> > Hmm, so do you prefer tackling this problem directly at  event_legacy_tracepoint, e.g.
> > like in
> > https://lkml.org/lkml/2014/3/24/364
> 
> A totally different approach would be to rename the kvm-s390 trace events to kvm_s390.
> 

hopefully that wouldn't be necessary ;-)

I checked the other approach and it works.. I'll see if I can find
some other way, if not I'll take that patch

thanks,
jirka
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/