* [PATCH v3 1/2] perf evlist: Allow setting arbitrary leader
@ 2021-11-30 17:49 Ian Rogers
2021-11-30 17:49 ` [PATCH v3 2/2] perf parse-events: Architecture specific leader override Ian Rogers
2021-11-30 19:27 ` [PATCH v3 1/2] perf evlist: Allow setting arbitrary leader Arnaldo Carvalho de Melo
0 siblings, 2 replies; 3+ messages in thread
From: Ian Rogers @ 2021-11-30 17:49 UTC (permalink / raw)
To: Andi Kleen, Jiri Olsa, Namhyung Kim, John Garry, Kajol Jain,
Paul A . Clarke, Arnaldo Carvalho de Melo, Riccardo Mancini,
Kan Liang, Peter Zijlstra, Ingo Molnar, Mark Rutland,
Alexander Shishkin, linux-perf-users, linux-kernel, Vineet Singh
Cc: eranian, Ian Rogers
The leader of a group is the first, but allow it to be an arbitrary
list member so that for Intel topdown events slots may always be the
group leader.
v3. Switched list_entry to list_first_entry as suggested by Arnaldo
Carvalho de Melo <acme@kernel.org>.
Reviewed-by: Kajol Jain<kjain@linux.ibm.com>
Acked-by: Jiri Olsa <jolsa@redhat.com>
Signed-off-by: Ian Rogers <irogers@google.com>
---
tools/lib/perf/evlist.c | 15 +++++++++------
tools/lib/perf/include/internal/evlist.h | 2 +-
tools/perf/util/parse-events.c | 4 ++--
3 files changed, 12 insertions(+), 9 deletions(-)
diff --git a/tools/lib/perf/evlist.c b/tools/lib/perf/evlist.c
index e37dfad31383..245acbc53bd3 100644
--- a/tools/lib/perf/evlist.c
+++ b/tools/lib/perf/evlist.c
@@ -643,14 +643,14 @@ perf_evlist__next_mmap(struct perf_evlist *evlist, struct perf_mmap *map,
return overwrite ? evlist->mmap_ovw_first : evlist->mmap_first;
}
-void __perf_evlist__set_leader(struct list_head *list)
+void __perf_evlist__set_leader(struct list_head *list, struct perf_evsel *leader)
{
- struct perf_evsel *evsel, *leader;
+ struct perf_evsel *first, *last, *evsel;
- leader = list_entry(list->next, struct perf_evsel, node);
- evsel = list_entry(list->prev, struct perf_evsel, node);
+ first = list_first_entry(list, struct perf_evsel, node);
+ last = list_last_entry(list, struct perf_evsel, node);
- leader->nr_members = evsel->idx - leader->idx + 1;
+ leader->nr_members = last->idx - first->idx + 1;
__perf_evlist__for_each_entry(list, evsel)
evsel->leader = leader;
@@ -659,7 +659,10 @@ void __perf_evlist__set_leader(struct list_head *list)
void perf_evlist__set_leader(struct perf_evlist *evlist)
{
if (evlist->nr_entries) {
+ struct perf_evsel *first = list_entry(evlist->entries.next,
+ struct perf_evsel, node);
+
evlist->nr_groups = evlist->nr_entries > 1 ? 1 : 0;
- __perf_evlist__set_leader(&evlist->entries);
+ __perf_evlist__set_leader(&evlist->entries, first);
}
}
diff --git a/tools/lib/perf/include/internal/evlist.h b/tools/lib/perf/include/internal/evlist.h
index f366dbad6a88..6f74269a3ad4 100644
--- a/tools/lib/perf/include/internal/evlist.h
+++ b/tools/lib/perf/include/internal/evlist.h
@@ -127,5 +127,5 @@ int perf_evlist__id_add_fd(struct perf_evlist *evlist,
void perf_evlist__reset_id_hash(struct perf_evlist *evlist);
-void __perf_evlist__set_leader(struct list_head *list);
+void __perf_evlist__set_leader(struct list_head *list, struct perf_evsel *leader);
#endif /* __LIBPERF_INTERNAL_EVLIST_H */
diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c
index 5bfb6f892489..c2935aca4b67 100644
--- a/tools/perf/util/parse-events.c
+++ b/tools/perf/util/parse-events.c
@@ -1834,8 +1834,8 @@ void parse_events__set_leader(char *name, struct list_head *list,
if (parse_events__set_leader_for_uncore_aliase(name, list, parse_state))
return;
- __perf_evlist__set_leader(list);
- leader = list_entry(list->next, struct evsel, core.node);
+ leader = list_first_entry(list, struct evsel, core.node);
+ __perf_evlist__set_leader(list, &leader->core);
leader->group_name = name ? strdup(name) : NULL;
}
--
2.34.0.rc2.393.gf8c9666880-goog
^ permalink raw reply related [flat|nested] 3+ messages in thread
* [PATCH v3 2/2] perf parse-events: Architecture specific leader override
2021-11-30 17:49 [PATCH v3 1/2] perf evlist: Allow setting arbitrary leader Ian Rogers
@ 2021-11-30 17:49 ` Ian Rogers
2021-11-30 19:27 ` [PATCH v3 1/2] perf evlist: Allow setting arbitrary leader Arnaldo Carvalho de Melo
1 sibling, 0 replies; 3+ messages in thread
From: Ian Rogers @ 2021-11-30 17:49 UTC (permalink / raw)
To: Andi Kleen, Jiri Olsa, Namhyung Kim, John Garry, Kajol Jain,
Paul A . Clarke, Arnaldo Carvalho de Melo, Riccardo Mancini,
Kan Liang, Peter Zijlstra, Ingo Molnar, Mark Rutland,
Alexander Shishkin, linux-perf-users, linux-kernel, Vineet Singh
Cc: eranian, Ian Rogers
Currently topdown events must appear after a slots event:
$ perf stat -e '{slots,topdown-fe-bound}' /bin/true
Performance counter stats for '/bin/true':
3,183,090 slots
986,133 topdown-fe-bound
Reversing the events yields:
$ perf stat -e '{topdown-fe-bound,slots}' /bin/true
Error:
The sys_perf_event_open() syscall returned with 22 (Invalid argument) for event (topdown-fe-bound).
For metrics the order of events is determined by iterating over a
hashmap, and so slots isn't guaranteed to be first which can yield this
error.
Change the set_leader in parse-events, called when a group is closed, so
that rather than always making the first event the leader, if the slots
event exists then it is made the leader. It is then moved to the head of
the evlist otherwise it won't be opened in the correct order.
The result is:
$ perf stat -e '{topdown-fe-bound,slots}' /bin/true
Performance counter stats for '/bin/true':
3,274,795 slots
1,001,702 topdown-fe-bound
A problem with this approach is the slots event is identified by name,
names can be overwritten like 'cpu/slots,name=foo/' and this causes the
leader change to fail.
The change also modifies and fixes mixed groups like, with the change:
$ perf stat -e '{instructions,slots,topdown-fe-bound}' -a -- sleep 2
Performance counter stats for 'system wide':
5574985410 slots
971981616 instructions
1348461887 topdown-fe-bound
2.001263120 seconds time elapsed
Without the change:
$ perf stat -e '{instructions,slots,topdown-fe-bound}' -a -- sleep 2
Performance counter stats for 'system wide':
<not counted> instructions
<not counted> slots
<not supported> topdown-fe-bound
2.006247990 seconds time elapsed
Something that may be undesirable here is that the events are reordered
in the output.
v2. Move the list_move operation into parse_events__set_leader as
suggested by Jiri Olsa <jolsa@redhat.com>.
v3. Switched list_entry to list_first_entry as suggested by Arnaldo
Carvalho de Melo <acme@kernel.org>.
Reviewed-by: Kajol Jain<kjain@linux.ibm.com>
Acked-by: Jiri Olsa <jolsa@redhat.com>
Signed-off-by: Ian Rogers <irogers@google.com>
---
tools/perf/arch/x86/util/evlist.c | 17 +++++++++++++++++
tools/perf/util/evlist.h | 1 +
tools/perf/util/parse-events.c | 8 +++++++-
3 files changed, 25 insertions(+), 1 deletion(-)
diff --git a/tools/perf/arch/x86/util/evlist.c b/tools/perf/arch/x86/util/evlist.c
index 0b0951030a2f..f924246eff78 100644
--- a/tools/perf/arch/x86/util/evlist.c
+++ b/tools/perf/arch/x86/util/evlist.c
@@ -17,3 +17,20 @@ int arch_evlist__add_default_attrs(struct evlist *evlist)
else
return parse_events(evlist, TOPDOWN_L1_EVENTS, NULL);
}
+
+struct evsel *arch_evlist__leader(struct list_head *list)
+{
+ struct evsel *evsel, *first;
+
+ first = list_first_entry(list, struct evsel, core.node);
+
+ if (!pmu_have_event("cpu", "slots"))
+ return first;
+
+ __evlist__for_each_entry(list, evsel) {
+ if (evsel->pmu_name && !strcmp(evsel->pmu_name, "cpu") &&
+ evsel->name && strstr(evsel->name, "slots"))
+ return evsel;
+ }
+ return first;
+}
diff --git a/tools/perf/util/evlist.h b/tools/perf/util/evlist.h
index 97bfb8d0be4f..993437ffe429 100644
--- a/tools/perf/util/evlist.h
+++ b/tools/perf/util/evlist.h
@@ -110,6 +110,7 @@ int __evlist__add_default_attrs(struct evlist *evlist,
__evlist__add_default_attrs(evlist, array, ARRAY_SIZE(array))
int arch_evlist__add_default_attrs(struct evlist *evlist);
+struct evsel *arch_evlist__leader(struct list_head *list);
int evlist__add_dummy(struct evlist *evlist);
diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c
index c2935aca4b67..1faf3c19430b 100644
--- a/tools/perf/util/parse-events.c
+++ b/tools/perf/util/parse-events.c
@@ -1821,6 +1821,11 @@ parse_events__set_leader_for_uncore_aliase(char *name, struct list_head *list,
return ret;
}
+__weak struct evsel *arch_evlist__leader(struct list_head *list)
+{
+ return list_first_entry(list, struct evsel, core.node);
+}
+
void parse_events__set_leader(char *name, struct list_head *list,
struct parse_events_state *parse_state)
{
@@ -1834,9 +1839,10 @@ void parse_events__set_leader(char *name, struct list_head *list,
if (parse_events__set_leader_for_uncore_aliase(name, list, parse_state))
return;
- leader = list_first_entry(list, struct evsel, core.node);
+ leader = arch_evlist__leader(list);
__perf_evlist__set_leader(list, &leader->core);
leader->group_name = name ? strdup(name) : NULL;
+ list_move(&leader->core.node, list);
}
/* list_event is assumed to point to malloc'ed memory */
--
2.34.0.rc2.393.gf8c9666880-goog
^ permalink raw reply related [flat|nested] 3+ messages in thread
* Re: [PATCH v3 1/2] perf evlist: Allow setting arbitrary leader
2021-11-30 17:49 [PATCH v3 1/2] perf evlist: Allow setting arbitrary leader Ian Rogers
2021-11-30 17:49 ` [PATCH v3 2/2] perf parse-events: Architecture specific leader override Ian Rogers
@ 2021-11-30 19:27 ` Arnaldo Carvalho de Melo
1 sibling, 0 replies; 3+ messages in thread
From: Arnaldo Carvalho de Melo @ 2021-11-30 19:27 UTC (permalink / raw)
To: Ian Rogers
Cc: Andi Kleen, Jiri Olsa, Namhyung Kim, John Garry, Kajol Jain,
Paul A . Clarke, Riccardo Mancini, Kan Liang, Peter Zijlstra,
Ingo Molnar, Mark Rutland, Alexander Shishkin, linux-perf-users,
linux-kernel, Vineet Singh, eranian
Em Tue, Nov 30, 2021 at 09:49:44AM -0800, Ian Rogers escreveu:
> The leader of a group is the first, but allow it to be an arbitrary
> list member so that for Intel topdown events slots may always be the
> group leader.
>
> v3. Switched list_entry to list_first_entry as suggested by Arnaldo
> Carvalho de Melo <acme@kernel.org>.
Thanks, applied both patches.
- Arnaldo
> Reviewed-by: Kajol Jain<kjain@linux.ibm.com>
> Acked-by: Jiri Olsa <jolsa@redhat.com>
> Signed-off-by: Ian Rogers <irogers@google.com>
> ---
> tools/lib/perf/evlist.c | 15 +++++++++------
> tools/lib/perf/include/internal/evlist.h | 2 +-
> tools/perf/util/parse-events.c | 4 ++--
> 3 files changed, 12 insertions(+), 9 deletions(-)
>
> diff --git a/tools/lib/perf/evlist.c b/tools/lib/perf/evlist.c
> index e37dfad31383..245acbc53bd3 100644
> --- a/tools/lib/perf/evlist.c
> +++ b/tools/lib/perf/evlist.c
> @@ -643,14 +643,14 @@ perf_evlist__next_mmap(struct perf_evlist *evlist, struct perf_mmap *map,
> return overwrite ? evlist->mmap_ovw_first : evlist->mmap_first;
> }
>
> -void __perf_evlist__set_leader(struct list_head *list)
> +void __perf_evlist__set_leader(struct list_head *list, struct perf_evsel *leader)
> {
> - struct perf_evsel *evsel, *leader;
> + struct perf_evsel *first, *last, *evsel;
>
> - leader = list_entry(list->next, struct perf_evsel, node);
> - evsel = list_entry(list->prev, struct perf_evsel, node);
> + first = list_first_entry(list, struct perf_evsel, node);
> + last = list_last_entry(list, struct perf_evsel, node);
>
> - leader->nr_members = evsel->idx - leader->idx + 1;
> + leader->nr_members = last->idx - first->idx + 1;
>
> __perf_evlist__for_each_entry(list, evsel)
> evsel->leader = leader;
> @@ -659,7 +659,10 @@ void __perf_evlist__set_leader(struct list_head *list)
> void perf_evlist__set_leader(struct perf_evlist *evlist)
> {
> if (evlist->nr_entries) {
> + struct perf_evsel *first = list_entry(evlist->entries.next,
> + struct perf_evsel, node);
> +
> evlist->nr_groups = evlist->nr_entries > 1 ? 1 : 0;
> - __perf_evlist__set_leader(&evlist->entries);
> + __perf_evlist__set_leader(&evlist->entries, first);
> }
> }
> diff --git a/tools/lib/perf/include/internal/evlist.h b/tools/lib/perf/include/internal/evlist.h
> index f366dbad6a88..6f74269a3ad4 100644
> --- a/tools/lib/perf/include/internal/evlist.h
> +++ b/tools/lib/perf/include/internal/evlist.h
> @@ -127,5 +127,5 @@ int perf_evlist__id_add_fd(struct perf_evlist *evlist,
>
> void perf_evlist__reset_id_hash(struct perf_evlist *evlist);
>
> -void __perf_evlist__set_leader(struct list_head *list);
> +void __perf_evlist__set_leader(struct list_head *list, struct perf_evsel *leader);
> #endif /* __LIBPERF_INTERNAL_EVLIST_H */
> diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c
> index 5bfb6f892489..c2935aca4b67 100644
> --- a/tools/perf/util/parse-events.c
> +++ b/tools/perf/util/parse-events.c
> @@ -1834,8 +1834,8 @@ void parse_events__set_leader(char *name, struct list_head *list,
> if (parse_events__set_leader_for_uncore_aliase(name, list, parse_state))
> return;
>
> - __perf_evlist__set_leader(list);
> - leader = list_entry(list->next, struct evsel, core.node);
> + leader = list_first_entry(list, struct evsel, core.node);
> + __perf_evlist__set_leader(list, &leader->core);
> leader->group_name = name ? strdup(name) : NULL;
> }
>
> --
> 2.34.0.rc2.393.gf8c9666880-goog
--
- Arnaldo
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2021-11-30 19:27 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-11-30 17:49 [PATCH v3 1/2] perf evlist: Allow setting arbitrary leader Ian Rogers
2021-11-30 17:49 ` [PATCH v3 2/2] perf parse-events: Architecture specific leader override Ian Rogers
2021-11-30 19:27 ` [PATCH v3 1/2] perf evlist: Allow setting arbitrary leader Arnaldo Carvalho de Melo
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).