Skip to contentSkip to conversation

perf top: Use evsel's cpus to replace user_requested_cpus

PATCH Superseded [email protected] 3 participants 9 messages Dec 12, 2023 – Dec 13, 2023 perf events

v2 of 4 Superseded Latest is v4 →

lore
Share & tools
lore
#0 lore raw

PATCH v2 1/1 perf top: Use evsel's cpus to replace user_requested_cpus

From: Kan Liang <[email protected]>

perf top errors out on a hybrid machine
$perf top

Error:
The cycles:P event is not supported.

The perf top expects that the "cycles" is collected on all CPUs in the system. But for hybrid there is no single "cycles" event which can cover all CPUs. Perf has to split it into two cycles events, e.g., cpu_core/cycles/ and cpu_atom/cycles/. Each event has its own CPU mask. If a event is opened on the unsupported CPU. The open fails. That's the reason of the above error out.

Perf should only open the cycles event on the corresponding CPU. The commit ef91871c960e ("perf evlist: Propagate user CPU maps intersecting core PMU maps") intersect the requested CPU map with the CPU map of the PMU. Use the evsel's cpus to replace user_requested_cpus.

The evlist's threads are also propagated to the evsel's threads in __perf_evlist__propagate_maps(). For a system-wide event, perf appends a dummy event and assign it to the evsel's threads. For a per-thread event, the evlist's thread_map is assigned to the evsel's threads. The same as the other tools, e.g., perf record, using the evsel's threads when opening an event.

---

Changes since V1:
- Update the description
- Add Reviewed-by from Ian

 tools/perf/builtin-top.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
tools/perf/builtin-top.c +2 -2
index ea8c7eca5eee..cce9350177e2 100644
@@ -1027,8 +1027,8 @@ static int perf_top__start_counters(struct perf_top *top)
 
 	evlist__for_each_entry(evlist, counter) {
 try_again:
-		if (evsel__open(counter, top->evlist->core.user_requested_cpus,
-				     top->evlist->core.threads) < 0) {
+		if (evsel__open(counter, counter->core.cpus,
+				counter->core.threads) < 0) {
 
 			/*
 			 * Specially handle overwrite fall back.
Signature · 1 line
2.35.1
reviewer
#1 lore raw
Quoted 32 lines · On Tue, Dec 12, 2023 at 11:39 AM <[email protected]> wrote:

From: Kan Liang <[email protected]>

perf top errors out on a hybrid machine
$perf top

Error:
The cycles:P event is not supported.

The perf top expects that the "cycles" is collected on all CPUs in the system. But for hybrid there is no single "cycles" event which can cover all CPUs. Perf has to split it into two cycles events, e.g., cpu_core/cycles/ and cpu_atom/cycles/. Each event has its own CPU mask. If a event is opened on the unsupported CPU. The open fails. That's the reason of the above error out.

Perf should only open the cycles event on the corresponding CPU. The commit ef91871c960e ("perf evlist: Propagate user CPU maps intersecting core PMU maps") intersect the requested CPU map with the CPU map of the PMU. Use the evsel's cpus to replace user_requested_cpus.

The evlist's threads are also propagated to the evsel's threads in __perf_evlist__propagate_maps(). For a system-wide event, perf appends a dummy event and assign it to the evsel's threads. For a per-thread event, the evlist's thread_map is assigned to the evsel's threads. The same as the other tools, e.g., perf record, using the evsel's threads when opening an event.

Quoted 5 lines

---

Changes since V1:
- Update the description
- Add Reviewed-by from Ian

Thanks Kan, quick question. Does "perf top" on hybrid ask the user to select between the cycles event on cpu_atom and cpu_core? I'm wondering if there is some kind of missing "hybrid-merge" functionality like we have for perf stat.

Thanks,
Ian

Quoted 21 lines
 tools/perf/builtin-top.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
tools/perf/builtin-top.c +2 -3
index ea8c7eca5eee..cce9350177e2 100644
@@ -1027,8 +1027,8 @@ static int perf_top__start_counters(struct perf_top *top)
 
        evlist__for_each_entry(evlist, counter) {
 try_again:
-               if (evsel__open(counter, top->evlist->core.user_requested_cpus,
-                                    top->evlist->core.threads) < 0) {
+               if (evsel__open(counter, counter->core.cpus,
+                               counter->core.threads) < 0) {
 
                        /*
                         * Specially handle overwrite fall back.
--

2.35.1

author
#2 lore raw
Quoted 35 lines · Ian Rogers
Quoted 34 lines · On Tue, Dec 12, 2023 at 11:39 AM <[email protected]> wrote:

From: Kan Liang <[email protected]>

perf top errors out on a hybrid machine
$perf top

Error:
The cycles:P event is not supported.

The perf top expects that the "cycles" is collected on all CPUs in the system. But for hybrid there is no single "cycles" event which can cover all CPUs. Perf has to split it into two cycles events, e.g., cpu_core/cycles/ and cpu_atom/cycles/. Each event has its own CPU mask. If a event is opened on the unsupported CPU. The open fails. That's the reason of the above error out.

Perf should only open the cycles event on the corresponding CPU. The commit ef91871c960e ("perf evlist: Propagate user CPU maps intersecting core PMU maps") intersect the requested CPU map with the CPU map of the PMU. Use the evsel's cpus to replace user_requested_cpus.

The evlist's threads are also propagated to the evsel's threads in __perf_evlist__propagate_maps(). For a system-wide event, perf appends a dummy event and assign it to the evsel's threads. For a per-thread event, the evlist's thread_map is assigned to the evsel's threads. The same as the other tools, e.g., perf record, using the evsel's threads when opening an event.

---

Quoted 6 lines
Quoted 3 lines

Changes since V1:
- Update the description
- Add Reviewed-by from Ian

Thanks Kan, quick question. Does "perf top" on hybrid ask the user to select between the cycles event on cpu_atom and cpu_core?

Yes, but the event doesn't include the PMU information. We probably need a follow up patch to append the PMU name.

Available samples
385 cycles:P

903 cycles:P

Thanks,
Kan

Quoted 28 lines

I'm
wondering if there is some kind of missing "hybrid-merge" functionality like we have for perf stat.

Thanks,
Ian

Quoted 21 lines
 tools/perf/builtin-top.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
tools/perf/builtin-top.c +2 -3
index ea8c7eca5eee..cce9350177e2 100644
@@ -1027,8 +1027,8 @@ static int perf_top__start_counters(struct perf_top *top)
 
        evlist__for_each_entry(evlist, counter) {
 try_again:
-               if (evsel__open(counter, top->evlist->core.user_requested_cpus,
-                                    top->evlist->core.threads) < 0) {
+               if (evsel__open(counter, counter->core.cpus,
+                               counter->core.threads) < 0) {
 
                        /*
                         * Specially handle overwrite fall back.
--

2.35.1

reviewer
#3 lore raw
Quoted 46 lines · Kan
Quoted 35 lines · Ian Rogers
Quoted 34 lines · On Tue, Dec 12, 2023 at 11:39 AM <[email protected]> wrote:

From: Kan Liang <[email protected]>

perf top errors out on a hybrid machine
$perf top

Error:
The cycles:P event is not supported.

The perf top expects that the "cycles" is collected on all CPUs in the system. But for hybrid there is no single "cycles" event which can cover all CPUs. Perf has to split it into two cycles events, e.g., cpu_core/cycles/ and cpu_atom/cycles/. Each event has its own CPU mask. If a event is opened on the unsupported CPU. The open fails. That's the reason of the above error out.

Perf should only open the cycles event on the corresponding CPU. The commit ef91871c960e ("perf evlist: Propagate user CPU maps intersecting core PMU maps") intersect the requested CPU map with the CPU map of the PMU. Use the evsel's cpus to replace user_requested_cpus.

The evlist's threads are also propagated to the evsel's threads in __perf_evlist__propagate_maps(). For a system-wide event, perf appends a dummy event and assign it to the evsel's threads. For a per-thread event, the evlist's thread_map is assigned to the evsel's threads. The same as the other tools, e.g., perf record, using the evsel's threads when opening an event.

---

Quoted 6 lines
Quoted 3 lines

Changes since V1:
- Update the description
- Add Reviewed-by from Ian

Thanks Kan, quick question. Does "perf top" on hybrid ask the user to select between the cycles event on cpu_atom and cpu_core?

Quoted 7 lines

Yes, but the event doesn't include the PMU information. We probably need a follow up patch to append the PMU name.

Available samples
385 cycles:P

903 cycles:P

Thanks and agreed, it isn't possible to tell which is which PMU/CPU type at the moment. I tried the patch with perf top --stdio, there wasn't a choice of event and I can't tell what counter is being displayed. When I quit I also see:

```
exiting.
corrupted double-linked list
Aborted (core dumped)
```

but I wasn't able to repro this on a debuggable binary/system.

If my memory serves there was a patch where perf top was showing >1 event. It would be nice here to do some kind of hybrid merging rather than having to view each PMU's top separately.

Thanks,
Ian

Quoted 31 lines

Thanks,
Kan

Quoted 28 lines

I'm
wondering if there is some kind of missing "hybrid-merge" functionality like we have for perf stat.

Thanks,
Ian

Quoted 21 lines
 tools/perf/builtin-top.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
tools/perf/builtin-top.c +2 -3
index ea8c7eca5eee..cce9350177e2 100644
@@ -1027,8 +1027,8 @@ static int perf_top__start_counters(struct perf_top *top)
 
        evlist__for_each_entry(evlist, counter) {
 try_again:
-               if (evsel__open(counter, top->evlist->core.user_requested_cpus,
-                                    top->evlist->core.threads) < 0) {
+               if (evsel__open(counter, counter->core.cpus,
+                               counter->core.threads) < 0) {
 
                        /*
                         * Specially handle overwrite fall back.
--

2.35.1

maintainer
#4 lore raw
Quoted 65 lines · Ian Rogers
Quoted 46 lines · Kan
Quoted 35 lines · Ian Rogers
Quoted 34 lines · On Tue, Dec 12, 2023 at 11:39 AM <[email protected]> wrote:

From: Kan Liang <[email protected]>

perf top errors out on a hybrid machine
$perf top

Error:
The cycles:P event is not supported.

The perf top expects that the "cycles" is collected on all CPUs in the system. But for hybrid there is no single "cycles" event which can cover all CPUs. Perf has to split it into two cycles events, e.g., cpu_core/cycles/ and cpu_atom/cycles/. Each event has its own CPU mask. If a event is opened on the unsupported CPU. The open fails. That's the reason of the above error out.

Perf should only open the cycles event on the corresponding CPU. The commit ef91871c960e ("perf evlist: Propagate user CPU maps intersecting core PMU maps") intersect the requested CPU map with the CPU map of the PMU. Use the evsel's cpus to replace user_requested_cpus.

The evlist's threads are also propagated to the evsel's threads in __perf_evlist__propagate_maps(). For a system-wide event, perf appends a dummy event and assign it to the evsel's threads. For a per-thread event, the evlist's thread_map is assigned to the evsel's threads. The same as the other tools, e.g., perf record, using the evsel's threads when opening an event.

---

Quoted 6 lines
Quoted 3 lines

Changes since V1:
- Update the description
- Add Reviewed-by from Ian

Thanks Kan, quick question. Does "perf top" on hybrid ask the user to select between the cycles event on cpu_atom and cpu_core?

Quoted 7 lines

Yes, but the event doesn't include the PMU information. We probably need a follow up patch to append the PMU name.

Available samples
385 cycles:P

903 cycles:P

Thanks and agreed, it isn't possible to tell which is which PMU/CPU type at the moment. I tried the patch with perf top --stdio, there wasn't a choice of event and I can't tell what counter is being displayed. When I quit I also see:

```
exiting.
corrupted double-linked list
Aborted (core dumped)
```
Quoted 5 lines

but I wasn't able to repro this on a debuggable binary/system.

If my memory serves there was a patch where perf top was showing >1 event. It would be nice here to do some kind of hybrid merging rather than having to view each PMU's top separately.

Using event groups, but I noticed you removed the --group option. Maybe perf top can just use { ... } notation for explicit grouping, but it might be implicit like in the hybrid case.

Thanks,
Namhyung

author
#5 lore raw
Quoted 52 lines · Namhyung Kim
Quoted 51 lines · Ian Rogers
Quoted 49 lines · Kan
Quoted 35 lines · Ian Rogers
Quoted 34 lines · On Tue, Dec 12, 2023 at 11:39 AM <[email protected]> wrote:

From: Kan Liang <[email protected]>

perf top errors out on a hybrid machine
$perf top

Error:
The cycles:P event is not supported.

The perf top expects that the "cycles" is collected on all CPUs in the system. But for hybrid there is no single "cycles" event which can cover all CPUs. Perf has to split it into two cycles events, e.g., cpu_core/cycles/ and cpu_atom/cycles/. Each event has its own CPU mask. If a event is opened on the unsupported CPU. The open fails. That's the reason of the above error out.

Perf should only open the cycles event on the corresponding CPU. The commit ef91871c960e ("perf evlist: Propagate user CPU maps intersecting core PMU maps") intersect the requested CPU map with the CPU map of the PMU. Use the evsel's cpus to replace user_requested_cpus.

The evlist's threads are also propagated to the evsel's threads in __perf_evlist__propagate_maps(). For a system-wide event, perf appends a dummy event and assign it to the evsel's threads. For a per-thread event, the evlist's thread_map is assigned to the evsel's threads. The same as the other tools, e.g., perf record, using the evsel's threads when opening an event.

---

Quoted 6 lines
Quoted 3 lines

Changes since V1:
- Update the description
- Add Reviewed-by from Ian

Thanks Kan, quick question. Does "perf top" on hybrid ask the user to select between the cycles event on cpu_atom and cpu_core?

Yes, but the event doesn't include the PMU information. We probably need a follow up patch to append the PMU name.

Quoted 8 lines
Quoted 8 lines
Quoted 4 lines

Available samples
385 cycles:P

903 cycles:P

Thanks and agreed, it isn't possible to tell which is which PMU/CPU type at the moment. I tried the patch with perf top --stdio, there wasn't a choice of event

The perf top --stdio uses a dedicated display function, see perf_top__header_snprintf() in util/top.c

It only shows one event at a time. "E" is used to switch the event.

Quoted 2 lines
Quoted 2 lines

and I can't tell what counter is being
displayed.

For the hybrid case, I think we may display both PMU name and event name. For example,

Available samples
656 cycles:P cpu_atom

701 cycles:P cpu_core

tools/perf/ui/browsers/hists.c +0 -0
index f4812b226818..afc7a1d54fe4 100644
@@ -3433,8 +3433,10 @@ static void perf_evsel_menu__write(struct

ui_browser *browser,

        }

        nr_events = convert_unit(nr_events, &unit);

- printed = scnprintf(bf, sizeof(bf), "%lu%c%s%s", nr_events,
- unit, unit == ' ' ? "" : " ", ev_name);
+ printed = scnprintf(bf, sizeof(bf), "%lu%c%s%s %s", nr_events,
+ unit, unit == ' ' ? "" : " ", ev_name,
+ evsel->pmu ? evsel->pmu_name : "");
+

        ui_browser__printf(browser, "%s", bf);

        nr_events = evsel->evlist->stats.nr_events[PERF_RECORD_LOST];
Quoted 7 lines
Quoted 7 lines

When I quit I also see:

```
exiting.
corrupted double-linked list
Aborted (core dumped)
```

but I wasn't able to repro this on a debuggable binary/system.

I haven't see the issue yet.

Quoted 4 lines
Quoted 4 lines

If my memory serves there was a patch where perf top was showing >1 event. It would be nice here to do some kind of hybrid merging rather than having to view each PMU's top separately.

The current perf top doesn't merge when there are >1 event. sudo ./perf top -e "cycles,instructions"

Available samples
2K cycles

2K instructions

For now, I think we may just append a PMU name to distinguish the hybrid case.

We may implement the merging feature which impacts both hybrid and non-hybrid cases later separately.

Quoted 5 lines

Using event groups, but I noticed you removed the --group option. Maybe perf top can just use { ... } notation for explicit grouping, but it might be implicit like in the hybrid case.

Yes, if the events are from different PMUs, the perf tool will implicitly de-group the hybrid events. "{ ... }" may not help here.

Thanks,
Kan

reviewer
#6 lore raw
Quoted 73 lines · Kan
Quoted 52 lines · Namhyung Kim
Quoted 51 lines · Ian Rogers
Quoted 49 lines · Kan
Quoted 35 lines · Ian Rogers
Quoted 34 lines · On Tue, Dec 12, 2023 at 11:39 AM <[email protected]> wrote:

From: Kan Liang <[email protected]>

perf top errors out on a hybrid machine
$perf top

Error:
The cycles:P event is not supported.

The perf top expects that the "cycles" is collected on all CPUs in the system. But for hybrid there is no single "cycles" event which can cover all CPUs. Perf has to split it into two cycles events, e.g., cpu_core/cycles/ and cpu_atom/cycles/. Each event has its own CPU mask. If a event is opened on the unsupported CPU. The open fails. That's the reason of the above error out.

Perf should only open the cycles event on the corresponding CPU. The commit ef91871c960e ("perf evlist: Propagate user CPU maps intersecting core PMU maps") intersect the requested CPU map with the CPU map of the PMU. Use the evsel's cpus to replace user_requested_cpus.

The evlist's threads are also propagated to the evsel's threads in __perf_evlist__propagate_maps(). For a system-wide event, perf appends a dummy event and assign it to the evsel's threads. For a per-thread event, the evlist's thread_map is assigned to the evsel's threads. The same as the other tools, e.g., perf record, using the evsel's threads when opening an event.

---

Quoted 6 lines
Quoted 3 lines

Changes since V1:
- Update the description
- Add Reviewed-by from Ian

Thanks Kan, quick question. Does "perf top" on hybrid ask the user to select between the cycles event on cpu_atom and cpu_core?

Yes, but the event doesn't include the PMU information. We probably need a follow up patch to append the PMU name.

Quoted 8 lines
Quoted 8 lines
Quoted 4 lines

Available samples
385 cycles:P

903 cycles:P

Thanks and agreed, it isn't possible to tell which is which PMU/CPU type at the moment. I tried the patch with perf top --stdio, there wasn't a choice of event

The perf top --stdio uses a dedicated display function, see perf_top__header_snprintf() in util/top.c

It only shows one event at a time. "E" is used to switch the event.

Quoted 2 lines
Quoted 2 lines

and I can't tell what counter is being
displayed.

Quoted 7 lines

For the hybrid case, I think we may display both PMU name and event name. For example,

Available samples
656 cycles:P cpu_atom

701 cycles:P cpu_core

I think there would be more uniformity if we could do:
cpu/cycles/P
I'm just reminded of the stat output where sometimes you can get things like:
cpu/cycles/
or sometimes:
cycles [cpu]
It seems the slash style approach is the most uniform given it looks like what is written on the command line. Perhaps we need some kind of helper function to do this as reformatting the modifier is a pain.

Quoted 50 lines
tools/perf/ui/browsers/hists.c +4 -2
index f4812b226818..afc7a1d54fe4 100644
@@ -3433,8 +3433,10 @@ static void perf_evsel_menu__write(struct
 ui_browser *browser,
        }
 
        nr_events = convert_unit(nr_events, &unit);
-       printed = scnprintf(bf, sizeof(bf), "%lu%c%s%s", nr_events,
-                          unit, unit == ' ' ? "" : " ", ev_name);
+       printed = scnprintf(bf, sizeof(bf), "%lu%c%s%s %s", nr_events,
+                          unit, unit == ' ' ? "" : " ", ev_name,
+                          evsel->pmu ? evsel->pmu_name : "");
+
        ui_browser__printf(browser, "%s", bf);
 
        nr_events = evsel->evlist->stats.nr_events[PERF_RECORD_LOST];
Quoted 7 lines
Quoted 7 lines

When I quit I also see:

```
exiting.
corrupted double-linked list
Aborted (core dumped)
```

but I wasn't able to repro this on a debuggable binary/system.

I haven't see the issue yet.

Quoted 4 lines
Quoted 4 lines

If my memory serves there was a patch where perf top was showing >1 event. It would be nice here to do some kind of hybrid merging rather than having to view each PMU's top separately.

The current perf top doesn't merge when there are >1 event. sudo ./perf top -e "cycles,instructions"

Available samples
2K cycles

2K instructions

For now, I think we may just append a PMU name to distinguish the hybrid case.

We may implement the merging feature which impacts both hybrid and non-hybrid cases later separately.

Quoted 1 line
Quoted 7 lines
Quoted 4 lines

Using event groups, but I noticed you removed the --group option. Maybe perf top can just use { ... } notation for explicit grouping, but it might be implicit like in the hybrid case.

Yes, if the events are from different PMUs, the perf tool will implicitly de-group the hybrid events. "{ ... }" may not help here.

I think grouping may have been the situation where I saw >1 event displayed but I agree with Kan, we implicitly de-group events on different PMUs so it won't help here.

Thanks,
Ian

Quoted 2 lines

Thanks,
Kan

author
#7 lore raw
Quoted 72 lines · Ian Rogers
Quoted 64 lines · Kan
Quoted 47 lines · Namhyung Kim
Quoted 46 lines · Ian Rogers
Quoted 45 lines · Kan
Quoted 34 lines · Ian Rogers
Quoted 33 lines · On Tue, Dec 12, 2023 at 11:39 AM <[email protected]> wrote:

From: Kan Liang <[email protected]>

perf top errors out on a hybrid machine
$perf top

Error:
The cycles:P event is not supported.

The perf top expects that the "cycles" is collected on all CPUs in the system. But for hybrid there is no single "cycles" event which can cover all CPUs. Perf has to split it into two cycles events, e.g., cpu_core/cycles/ and cpu_atom/cycles/. Each event has its own CPU mask. If a event is opened on the unsupported CPU. The open fails. That's the reason of the above error out.

Perf should only open the cycles event on the corresponding CPU. The commit ef91871c960e ("perf evlist: Propagate user CPU maps intersecting core PMU maps") intersect the requested CPU map with the CPU map of the PMU. Use the evsel's cpus to replace user_requested_cpus.

The evlist's threads are also propagated to the evsel's threads in __perf_evlist__propagate_maps(). For a system-wide event, perf appends a dummy event and assign it to the evsel's threads. For a per-thread event, the evlist's thread_map is assigned to the evsel's threads. The same as the other tools, e.g., perf record, using the evsel's threads when opening an event.

---

Quoted 5 lines
Quoted 3 lines

Changes since V1:
- Update the description
- Add Reviewed-by from Ian

Thanks Kan, quick question. Does "perf top" on hybrid ask the user to select between the cycles event on cpu_atom and cpu_core?

Yes, but the event doesn't include the PMU information. We probably need a follow up patch to append the PMU name.

Quoted 7 lines
Quoted 7 lines
Quoted 4 lines

Available samples
385 cycles:P

903 cycles:P

Thanks and agreed, it isn't possible to tell which is which PMU/CPU type at the moment. I tried the patch with perf top --stdio, there wasn't a choice of event

The perf top --stdio uses a dedicated display function, see perf_top__header_snprintf() in util/top.c

It only shows one event at a time. "E" is used to switch the event.

Quoted 2 lines
Quoted 2 lines

and I can't tell what counter is being
displayed.

Quoted 7 lines

For the hybrid case, I think we may display both PMU name and event name. For example,

Available samples
656 cycles:P cpu_atom

701 cycles:P cpu_core

Quoted 9 lines

I think there would be more uniformity if we could do:
cpu/cycles/P
I'm just reminded of the stat output where sometimes you can get things like:
cpu/cycles/
or sometimes:
cycles [cpu]
It seems the slash style approach is the most uniform given it looks like what is written on the command line. Perhaps we need some kind of helper function to do this as reformatting the modifier is a pain.

Actually, we already have a helper in the perf record, record__uniquify_name().
I can move it to the util/record.c and let's perf top invoke it before the open as well. Then we can get this in the perf top.

Available samples
148 cpu_atom/cycles:P/

1K cpu_core/cycles:P/

It should be good enough to distinguish the events. I will send a patch to support it.

Thanks,
Kan

maintainer
#8 lore raw
In reply to Ian Rogers
Quoted 32 lines · Ian Rogers
Quoted 30 lines · Kan
Quoted 5 lines · Namhyung Kim
Quoted 3 lines · Ian Rogers

If my memory serves there was a patch where perf top was showing >1 event. It would be nice here to do some kind of hybrid merging rather than having to view each PMU's top separately.

The current perf top doesn't merge when there are >1 event. sudo ./perf top -e "cycles,instructions"

Available samples
2K cycles

2K instructions

For now, I think we may just append a PMU name to distinguish the hybrid case.

We may implement the merging feature which impacts both hybrid and non-hybrid cases later separately.

Quoted 5 lines

Using event groups, but I noticed you removed the --group option. Maybe perf top can just use { ... } notation for explicit grouping, but it might be implicit like in the hybrid case.

Quoted 6 lines
Quoted 2 lines

Yes, if the events are from different PMUs, the perf tool will implicitly de-group the hybrid events. "{ ... }" may not help here.

I think grouping may have been the situation where I saw >1 event displayed but I agree with Kan, we implicitly de-group events on different PMUs so it won't help here.

The --group option was implemented in perf report which means it doesn't matter if events are in different PMUs. It's just to display the results together.

Thanks,
Namhyung