| Message ID | 20221123195010.6859-12-mike.leach@linaro.org (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | coresight: Add new API to allocate trace source ID values | expand |
Peter On 23/11/2022 19:50, Mike Leach wrote: > CoreSight trace being updated to use the perf_report_aux_output_id() > in a similar way to intel-pt. > > This function in needs export visibility to allow it to be called from > kernel loadable modules, which CoreSight may configured to be built as. > > Signed-off-by: Mike Leach <mike.leach@linaro.org> > Acked-by: Suzuki K Poulose <suzuki.poulose@arm.com> > --- > kernel/events/core.c | 1 + > 1 file changed, 1 insertion(+) > > diff --git a/kernel/events/core.c b/kernel/events/core.c > index 4ec3717003d5..ad388552f1d5 100644 > --- a/kernel/events/core.c > +++ b/kernel/events/core.c > @@ -9231,6 +9231,7 @@ void perf_report_aux_output_id(struct perf_event *event, u64 hw_id) > > perf_output_end(&handle); > } > +EXPORT_SYMBOL_GPL(perf_report_aux_output_id); > Are you happy with me picking up this change through the coresight tree ? Suzuki > static int > __perf_event_account_interrupt(struct perf_event *event, int throttle)
Peter, On 24/11/2022 11:04, Suzuki Kuruppassery Poulose wrote: > Peter > > > On 23/11/2022 19:50, Mike Leach wrote: >> CoreSight trace being updated to use the perf_report_aux_output_id() >> in a similar way to intel-pt. >> >> This function in needs export visibility to allow it to be called from >> kernel loadable modules, which CoreSight may configured to be built as. >> >> Signed-off-by: Mike Leach <mike.leach@linaro.org> >> Acked-by: Suzuki K Poulose <suzuki.poulose@arm.com> >> --- >> kernel/events/core.c | 1 + >> 1 file changed, 1 insertion(+) >> >> diff --git a/kernel/events/core.c b/kernel/events/core.c >> index 4ec3717003d5..ad388552f1d5 100644 >> --- a/kernel/events/core.c >> +++ b/kernel/events/core.c >> @@ -9231,6 +9231,7 @@ void perf_report_aux_output_id(struct perf_event >> *event, u64 hw_id) >> perf_output_end(&handle); >> } >> +EXPORT_SYMBOL_GPL(perf_report_aux_output_id); > > Are you happy with me picking up this change through the coresight tree ? > > Suzuki > Please could you let me know your thoughts on this ? Coresight PMU would use this to inform the dynamically allocated trace id for a CPU to the perf via this HWID packet. If you are happy with the above change, I could pick it up via the coresight tree with your Ack. Otherwise, please could you pick it up ? Suzuki
Peter, Ingo, Arnaldo On 13/12/2022 10:08, Suzuki K Poulose wrote: > Peter, > > On 24/11/2022 11:04, Suzuki Kuruppassery Poulose wrote: >> Peter >> >> >> On 23/11/2022 19:50, Mike Leach wrote: >>> CoreSight trace being updated to use the perf_report_aux_output_id() >>> in a similar way to intel-pt. >>> >>> This function in needs export visibility to allow it to be called from >>> kernel loadable modules, which CoreSight may configured to be built as. >>> >>> Signed-off-by: Mike Leach <mike.leach@linaro.org> >>> Acked-by: Suzuki K Poulose <suzuki.poulose@arm.com> >>> --- >>> kernel/events/core.c | 1 + >>> 1 file changed, 1 insertion(+) >>> >>> diff --git a/kernel/events/core.c b/kernel/events/core.c >>> index 4ec3717003d5..ad388552f1d5 100644 >>> --- a/kernel/events/core.c >>> +++ b/kernel/events/core.c >>> @@ -9231,6 +9231,7 @@ void perf_report_aux_output_id(struct >>> perf_event *event, u64 hw_id) >>> perf_output_end(&handle); >>> } >>> +EXPORT_SYMBOL_GPL(perf_report_aux_output_id); >> >> Are you happy with me picking up this change through the coresight >> tree ? >> >> Suzuki >> > > Please could you let me know your thoughts on this ? Coresight PMU would > use this to inform the dynamically allocated trace id for a CPU to the > perf via this HWID packet. > > If you are happy with the above change, I could pick it up via the > coresight tree with your Ack. Otherwise, please could you pick it > up ? > Please could you let us know if this is acceptable and if so, how can this be pushed ? CoreSight PMU (which can be built as a module) is using these packets to advertise the traceID allocated for a given CPU bound ETM (which is made dynamic due to the limited number of ids available). The perf tool can use these packets to then decode and separate the trace data from shared buffer. As of now, the series to enable the dynamic trace id allocation is blocked on this change. Kind regards Suzuki > Suzuki >
On Wed, Nov 23, 2022 at 07:50:07PM +0000, Mike Leach wrote: > CoreSight trace being updated to use the perf_report_aux_output_id() > in a similar way to intel-pt. > > This function in needs export visibility to allow it to be called from > kernel loadable modules, which CoreSight may configured to be built as. > > Signed-off-by: Mike Leach <mike.leach@linaro.org> > Acked-by: Suzuki K Poulose <suzuki.poulose@arm.com> Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org> > --- > kernel/events/core.c | 1 + > 1 file changed, 1 insertion(+) > > diff --git a/kernel/events/core.c b/kernel/events/core.c > index 4ec3717003d5..ad388552f1d5 100644 > --- a/kernel/events/core.c > +++ b/kernel/events/core.c > @@ -9231,6 +9231,7 @@ void perf_report_aux_output_id(struct perf_event *event, u64 hw_id) > > perf_output_end(&handle); > } > +EXPORT_SYMBOL_GPL(perf_report_aux_output_id); > > static int > __perf_event_account_interrupt(struct perf_event *event, int throttle) > -- > 2.17.1 >
diff --git a/kernel/events/core.c b/kernel/events/core.c index 4ec3717003d5..ad388552f1d5 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -9231,6 +9231,7 @@ void perf_report_aux_output_id(struct perf_event *event, u64 hw_id) perf_output_end(&handle); } +EXPORT_SYMBOL_GPL(perf_report_aux_output_id); static int __perf_event_account_interrupt(struct perf_event *event, int throttle)