perf stat: Add Topdown metrics events as default events
The Topdown Microarchitecture Analysis (TMA) Method is a structured
analysis methodology to identify critical performance bottlenecks in
out-of-order processors. From the Ice Lake and later platforms, the
Topdown information can be retrieved from the dedicated "metrics"
register, which isn't impacted by other events. Also, the Topdown
metrics support both per thread/process and per core measuring. Adding
Topdown metrics events as default events can enrich the default
measuring information, and would not cost any extra multiplexing.
Introduce arch_evlist__add_default_attrs() to allow architecture
specific default events. Add the Topdown metrics events in the X86
specific arch_evlist__add_default_attrs(). Other architectures can add
their own default events later separately.
With the patch:
$ perf stat sleep 1
Performance counter stats for 'sleep 1':
0.82 msec task-clock:u # 0.001 CPUs utilized
0 context-switches:u # 0.000 K/sec
0 cpu-migrations:u # 0.000 K/sec
61 page-faults:u # 0.074 M/sec
319,941 cycles:u # 0.388 GHz
242,802 instructions:u # 0.76 insn per cycle
54,380 branches:u # 66.028 M/sec
4,043 branch-misses:u # 7.43% of all branches
1,585,555 slots:u # 1925.189 M/sec
238,941 topdown-retiring:u # 15.0% retiring
410,378 topdown-bad-spec:u # 25.8% bad speculation
634,222 topdown-fe-bound:u # 39.9% frontend bound
304,675 topdown-be-bound:u # 19.2% backend bound
1.001791625 seconds time elapsed
0.000000000 seconds user
0.001572000 seconds sys
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jin Yao <yao.jin@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: http://lore.kernel.org/lkml/20210121133752.118327-1-kan.liang@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2021-01-21 05:37:52 -08:00
|
|
|
// SPDX-License-Identifier: GPL-2.0
|
perf stat: Remove evlist__add_default_attrs use strings
add_default_atttributes would add evsels by having pre-created
perf_event_attr, however, this needed fixing for hybrid as the
extended PMU type was necessary for each core PMU. The logic for this
was in an arch specific x86 function and wasn't present for ARM,
meaning that default events weren't being opened on all PMUs on
ARM. Change the creation of the default events to use parse_events and
strings as that will open the events on all PMUs.
Rather than try to detect events on PMUs before parsing, parse the
event but skip its output in stat-display.
The previous order of hardware events was: cycles,
stalled-cycles-frontend, stalled-cycles-backend, instructions. As
instructions is a more fundamental concept the order is changed to:
instructions, cycles, stalled-cycles-frontend, stalled-cycles-backend.
Closes: https://lore.kernel.org/lkml/CAP-5=fVABSBZnsmtRn1uF-k-G1GWM-L5SgiinhPTfHbQsKXb_g@mail.gmail.com/
Acked-by: Namhyung Kim <namhyung@kernel.org>
Signed-off-by: Ian Rogers <irogers@google.com>
[Don't display unsupported default events except 'cycles']
Acked-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: James Clark <james.clark@linaro.org>
Cc: Yang Jihong <yangjihong@bytedance.com>
Cc: Dominique Martinet <asmadeus@codewreck.org>
Cc: Colin Ian King <colin.i.king@gmail.com>
Cc: Howard Chu <howardchu95@gmail.com>
Cc: Ze Gao <zegao2021@gmail.com>
Cc: Yicong Yang <yangyicong@hisilicon.com>
Cc: Weilin Wang <weilin.wang@intel.com>
Cc: Will Deacon <will@kernel.org>
Cc: Mike Leach <mike.leach@linaro.org>
Cc: Jing Zhang <renyu.zj@linux.alibaba.com>
Cc: Yang Li <yang.lee@linux.alibaba.com>
Cc: Leo Yan <leo.yan@linux.dev>
Cc: ak@linux.intel.com
Cc: Athira Rajeev <atrajeev@linux.vnet.ibm.com>
Cc: linux-arm-kernel@lists.infradead.org
Cc: Sun Haiyong <sunhaiyong@loongson.cn>
Cc: John Garry <john.g.garry@oracle.com>
Link: https://lore.kernel.org/r/20240926144851.245903-4-james.clark@linaro.org
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
2024-09-26 15:48:34 +01:00
|
|
|
#include <string.h>
|
|
|
|
#include "../../../util/evlist.h"
|
|
|
|
#include "../../../util/evsel.h"
|
perf stat: Add topdown metrics in the default perf stat on the hybrid machine
Topdown metrics are missed in the default perf stat on the hybrid machine,
add Topdown metrics in default perf stat for hybrid systems.
Currently, we support the perf metrics Topdown for the p-core PMU in the
perf stat default, the perf metrics Topdown support for e-core PMU will be
implemented later separately. Refactor the code adds two x86 specific
functions. Widen the size of the event name column by 7 chars, so that all
metrics after the "#" become aligned again.
The perf metrics topdown feature is supported on the cpu_core of ADL. The
dedicated perf metrics counter and the fixed counter 3 are used for the
topdown events. Adding the topdown metrics doesn't trigger multiplexing.
Before:
# ./perf stat -a true
Performance counter stats for 'system wide':
53.70 msec cpu-clock # 25.736 CPUs utilized
80 context-switches # 1.490 K/sec
24 cpu-migrations # 446.951 /sec
52 page-faults # 968.394 /sec
2,788,555 cpu_core/cycles/ # 51.931 M/sec
851,129 cpu_atom/cycles/ # 15.851 M/sec
2,974,030 cpu_core/instructions/ # 55.385 M/sec
416,919 cpu_atom/instructions/ # 7.764 M/sec
586,136 cpu_core/branches/ # 10.916 M/sec
79,872 cpu_atom/branches/ # 1.487 M/sec
14,220 cpu_core/branch-misses/ # 264.819 K/sec
7,691 cpu_atom/branch-misses/ # 143.229 K/sec
0.002086438 seconds time elapsed
After:
# ./perf stat -a true
Performance counter stats for 'system wide':
61.39 msec cpu-clock # 24.874 CPUs utilized
76 context-switches # 1.238 K/sec
24 cpu-migrations # 390.968 /sec
52 page-faults # 847.097 /sec
2,753,695 cpu_core/cycles/ # 44.859 M/sec
903,899 cpu_atom/cycles/ # 14.725 M/sec
2,927,529 cpu_core/instructions/ # 47.690 M/sec
428,498 cpu_atom/instructions/ # 6.980 M/sec
581,299 cpu_core/branches/ # 9.470 M/sec
83,409 cpu_atom/branches/ # 1.359 M/sec
13,641 cpu_core/branch-misses/ # 222.216 K/sec
8,008 cpu_atom/branch-misses/ # 130.453 K/sec
14,761,308 cpu_core/slots/ # 240.466 M/sec
3,288,625 cpu_core/topdown-retiring/ # 22.3% retiring
1,323,323 cpu_core/topdown-bad-spec/ # 9.0% bad speculation
5,477,470 cpu_core/topdown-fe-bound/ # 37.1% frontend bound
4,679,199 cpu_core/topdown-be-bound/ # 31.7% backend bound
646,194 cpu_core/topdown-heavy-ops/ # 4.4% heavy operations # 17.9% light operations
1,244,999 cpu_core/topdown-br-mispredict/ # 8.4% branch mispredict # 0.5% machine clears
3,891,800 cpu_core/topdown-fetch-lat/ # 26.4% fetch latency # 10.7% fetch bandwidth
1,879,034 cpu_core/topdown-mem-bound/ # 12.7% memory bound # 19.0% Core bound
0.002467839 seconds time elapsed
Reviewed-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Xing Zhengjun <zhengjun.xing@linux.intel.com>
Acked-by: Ian Rogers <irogers@google.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://lore.kernel.org/r/20220721065706.2886112-6-zhengjun.xing@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2022-07-21 14:57:06 +08:00
|
|
|
#include "topdown.h"
|
2023-05-02 15:38:44 -07:00
|
|
|
#include "evsel.h"
|
perf stat: Add Topdown metrics events as default events
The Topdown Microarchitecture Analysis (TMA) Method is a structured
analysis methodology to identify critical performance bottlenecks in
out-of-order processors. From the Ice Lake and later platforms, the
Topdown information can be retrieved from the dedicated "metrics"
register, which isn't impacted by other events. Also, the Topdown
metrics support both per thread/process and per core measuring. Adding
Topdown metrics events as default events can enrich the default
measuring information, and would not cost any extra multiplexing.
Introduce arch_evlist__add_default_attrs() to allow architecture
specific default events. Add the Topdown metrics events in the X86
specific arch_evlist__add_default_attrs(). Other architectures can add
their own default events later separately.
With the patch:
$ perf stat sleep 1
Performance counter stats for 'sleep 1':
0.82 msec task-clock:u # 0.001 CPUs utilized
0 context-switches:u # 0.000 K/sec
0 cpu-migrations:u # 0.000 K/sec
61 page-faults:u # 0.074 M/sec
319,941 cycles:u # 0.388 GHz
242,802 instructions:u # 0.76 insn per cycle
54,380 branches:u # 66.028 M/sec
4,043 branch-misses:u # 7.43% of all branches
1,585,555 slots:u # 1925.189 M/sec
238,941 topdown-retiring:u # 15.0% retiring
410,378 topdown-bad-spec:u # 25.8% bad speculation
634,222 topdown-fe-bound:u # 39.9% frontend bound
304,675 topdown-be-bound:u # 19.2% backend bound
1.001791625 seconds time elapsed
0.000000000 seconds user
0.001572000 seconds sys
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jin Yao <yao.jin@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: http://lore.kernel.org/lkml/20210121133752.118327-1-kan.liang@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2021-01-21 05:37:52 -08:00
|
|
|
|
perf parse-events: Sort and group parsed events
This change is intended to be a no-op for most current cases, the
default sort order is the order the events were parsed. Where it
varies is in how groups are handled. Previously an uncore and core
event that are grouped would most often cause the group to be removed:
```
$ perf stat -e '{instructions,uncore_imc_free_running_0/data_total/}' -a sleep 1
WARNING: grouped events cpus do not match, disabling group:
anon group { instructions, uncore_imc_free_running_0/data_total/ }
...
```
However, when wildcards are used the events should be re-sorted and
re-grouped in parse_events__set_leader, but this currently fails for
simple examples:
```
$ perf stat -e '{uncore_imc_free_running/data_read/,uncore_imc_free_running/data_write/}' -a sleep 1
Performance counter stats for 'system wide':
<not counted> MiB uncore_imc_free_running/data_read/
<not counted> MiB uncore_imc_free_running/data_write/
1.000996992 seconds time elapsed
```
A futher failure mode, fixed in this patch, is to force topdown events
into a group.
This change moves sorting the evsels in the evlist after parsing. It
requires parsing to set up groups. First the evsels are sorted
respecting the existing groupings and parse order, but also reordering
to ensure evsels of the same PMU and group appear together. So that
software and aux events respect groups, their pmu_name is taken from
the group leader. The sorting is done with list_sort removing a memory
allocation.
After sorting a pass is done to correct the group leaders and for
topdown events ensuring they have a group leader.
This fixes the problems seen before:
```
$ perf stat -e '{uncore_imc_free_running/data_read/,uncore_imc_free_running/data_write/}' -a sleep 1
Performance counter stats for 'system wide':
727.42 MiB uncore_imc_free_running/data_read/
81.84 MiB uncore_imc_free_running/data_write/
1.000948615 seconds time elapsed
```
As well as making groups not fail for cases like:
```
$ perf stat -e '{imc_free_running_0/data_total/,imc_free_running_1/data_total/}' -a sleep 1
Performance counter stats for 'system wide':
256.47 MiB imc_free_running_0/data_total/
256.48 MiB imc_free_running_1/data_total/
1.001165442 seconds time elapsed
```
Signed-off-by: Ian Rogers <irogers@google.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Florian Fischer <florian.fischer@muhq.space>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: James Clark <james.clark@arm.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: John Garry <john.g.garry@oracle.com>
Cc: Kajol Jain <kjain@linux.ibm.com>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Kim Phillips <kim.phillips@amd.com>
Cc: Leo Yan <leo.yan@linaro.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ravi Bangoria <ravi.bangoria@amd.com>
Cc: Sean Christopherson <seanjc@google.com>
Cc: Steinar H. Gunderson <sesse@google.com>
Cc: Stephane Eranian <eranian@google.com>
Cc: Suzuki Poulouse <suzuki.poulose@arm.com>
Cc: Xing Zhengjun <zhengjun.xing@linux.intel.com>
Link: https://lore.kernel.org/r/20230312021543.3060328-2-irogers@google.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2023-03-11 18:15:40 -08:00
|
|
|
int arch_evlist__cmp(const struct evsel *lhs, const struct evsel *rhs)
|
perf parse-events: Architecture specific leader override
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.
Reviewed-by: Kajol Jain <kjain@linux.ibm.com>
Signed-off-by: Ian Rogers <irogers@google.com>
Acked-by: Jiri Olsa <jolsa@redhat.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: John Garry <john.garry@huawei.com>
Cc: Kajol Jain <kjain@linux.ibm.com>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Paul Clarke <pc@us.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Riccardo Mancini <rickyman7@gmail.com>
Cc: Stephane Eranian <eranian@google.com>
Cc: Vineet Singh <vineet.singh@intel.com>
Link: http://lore.kernel.org/lkml/20211130174945.247604-2-irogers@google.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2021-11-30 09:49:45 -08:00
|
|
|
{
|
2024-09-13 08:47:09 +00:00
|
|
|
/*
|
|
|
|
* Currently the following topdown events sequence are supported to
|
|
|
|
* move and regroup correctly.
|
|
|
|
*
|
|
|
|
* a. all events in a group
|
|
|
|
* perf stat -e "{instructions,topdown-retiring,slots}" -C0 sleep 1
|
|
|
|
* WARNING: events were regrouped to match PMUs
|
|
|
|
* Performance counter stats for 'CPU(s) 0':
|
|
|
|
* 15,066,240 slots
|
|
|
|
* 1,899,760 instructions
|
|
|
|
* 2,126,998 topdown-retiring
|
|
|
|
* b. all events not in a group
|
|
|
|
* perf stat -e "instructions,topdown-retiring,slots" -C0 sleep 1
|
|
|
|
* WARNING: events were regrouped to match PMUs
|
|
|
|
* Performance counter stats for 'CPU(s) 0':
|
|
|
|
* 2,045,561 instructions
|
|
|
|
* 17,108,370 slots
|
|
|
|
* 2,281,116 topdown-retiring
|
|
|
|
* c. slots event in a group but topdown metrics events outside the group
|
|
|
|
* perf stat -e "{instructions,slots},topdown-retiring" -C0 sleep 1
|
|
|
|
* WARNING: events were regrouped to match PMUs
|
|
|
|
* Performance counter stats for 'CPU(s) 0':
|
|
|
|
* 20,323,878 slots
|
|
|
|
* 2,634,884 instructions
|
|
|
|
* 3,028,656 topdown-retiring
|
|
|
|
* d. slots event and topdown metrics events in two groups
|
|
|
|
* perf stat -e "{instructions,slots},{topdown-retiring}" -C0 sleep 1
|
|
|
|
* WARNING: events were regrouped to match PMUs
|
|
|
|
* Performance counter stats for 'CPU(s) 0':
|
|
|
|
* 26,319,024 slots
|
|
|
|
* 2,427,791 instructions
|
|
|
|
* 2,683,508 topdown-retiring
|
2025-03-06 18:39:05 -08:00
|
|
|
* e. slots event and metrics event are not in a group and not adjacent
|
|
|
|
* perf stat -e "{instructions,slots},cycles,topdown-retiring" -C0 sleep 1
|
2024-09-13 08:47:09 +00:00
|
|
|
* WARNING: events were regrouped to match PMUs
|
2025-03-06 18:39:05 -08:00
|
|
|
* 68,433,522 slots
|
|
|
|
* 8,856,102 topdown-retiring
|
|
|
|
* 7,791,494 instructions
|
|
|
|
* 11,469,513 cycles
|
2024-09-13 08:47:09 +00:00
|
|
|
*/
|
2023-07-31 22:36:31 -07:00
|
|
|
if (topdown_sys_has_perf_metrics() &&
|
|
|
|
(arch_evsel__must_be_in_group(lhs) || arch_evsel__must_be_in_group(rhs))) {
|
perf parse-events: Sort and group parsed events
This change is intended to be a no-op for most current cases, the
default sort order is the order the events were parsed. Where it
varies is in how groups are handled. Previously an uncore and core
event that are grouped would most often cause the group to be removed:
```
$ perf stat -e '{instructions,uncore_imc_free_running_0/data_total/}' -a sleep 1
WARNING: grouped events cpus do not match, disabling group:
anon group { instructions, uncore_imc_free_running_0/data_total/ }
...
```
However, when wildcards are used the events should be re-sorted and
re-grouped in parse_events__set_leader, but this currently fails for
simple examples:
```
$ perf stat -e '{uncore_imc_free_running/data_read/,uncore_imc_free_running/data_write/}' -a sleep 1
Performance counter stats for 'system wide':
<not counted> MiB uncore_imc_free_running/data_read/
<not counted> MiB uncore_imc_free_running/data_write/
1.000996992 seconds time elapsed
```
A futher failure mode, fixed in this patch, is to force topdown events
into a group.
This change moves sorting the evsels in the evlist after parsing. It
requires parsing to set up groups. First the evsels are sorted
respecting the existing groupings and parse order, but also reordering
to ensure evsels of the same PMU and group appear together. So that
software and aux events respect groups, their pmu_name is taken from
the group leader. The sorting is done with list_sort removing a memory
allocation.
After sorting a pass is done to correct the group leaders and for
topdown events ensuring they have a group leader.
This fixes the problems seen before:
```
$ perf stat -e '{uncore_imc_free_running/data_read/,uncore_imc_free_running/data_write/}' -a sleep 1
Performance counter stats for 'system wide':
727.42 MiB uncore_imc_free_running/data_read/
81.84 MiB uncore_imc_free_running/data_write/
1.000948615 seconds time elapsed
```
As well as making groups not fail for cases like:
```
$ perf stat -e '{imc_free_running_0/data_total/,imc_free_running_1/data_total/}' -a sleep 1
Performance counter stats for 'system wide':
256.47 MiB imc_free_running_0/data_total/
256.48 MiB imc_free_running_1/data_total/
1.001165442 seconds time elapsed
```
Signed-off-by: Ian Rogers <irogers@google.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Florian Fischer <florian.fischer@muhq.space>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: James Clark <james.clark@arm.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: John Garry <john.g.garry@oracle.com>
Cc: Kajol Jain <kjain@linux.ibm.com>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Kim Phillips <kim.phillips@amd.com>
Cc: Leo Yan <leo.yan@linaro.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ravi Bangoria <ravi.bangoria@amd.com>
Cc: Sean Christopherson <seanjc@google.com>
Cc: Steinar H. Gunderson <sesse@google.com>
Cc: Stephane Eranian <eranian@google.com>
Cc: Suzuki Poulouse <suzuki.poulose@arm.com>
Cc: Xing Zhengjun <zhengjun.xing@linux.intel.com>
Link: https://lore.kernel.org/r/20230312021543.3060328-2-irogers@google.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2023-03-11 18:15:40 -08:00
|
|
|
/* Ensure the topdown slots comes first. */
|
2024-09-13 08:47:07 +00:00
|
|
|
if (arch_is_topdown_slots(lhs))
|
perf parse-events: Sort and group parsed events
This change is intended to be a no-op for most current cases, the
default sort order is the order the events were parsed. Where it
varies is in how groups are handled. Previously an uncore and core
event that are grouped would most often cause the group to be removed:
```
$ perf stat -e '{instructions,uncore_imc_free_running_0/data_total/}' -a sleep 1
WARNING: grouped events cpus do not match, disabling group:
anon group { instructions, uncore_imc_free_running_0/data_total/ }
...
```
However, when wildcards are used the events should be re-sorted and
re-grouped in parse_events__set_leader, but this currently fails for
simple examples:
```
$ perf stat -e '{uncore_imc_free_running/data_read/,uncore_imc_free_running/data_write/}' -a sleep 1
Performance counter stats for 'system wide':
<not counted> MiB uncore_imc_free_running/data_read/
<not counted> MiB uncore_imc_free_running/data_write/
1.000996992 seconds time elapsed
```
A futher failure mode, fixed in this patch, is to force topdown events
into a group.
This change moves sorting the evsels in the evlist after parsing. It
requires parsing to set up groups. First the evsels are sorted
respecting the existing groupings and parse order, but also reordering
to ensure evsels of the same PMU and group appear together. So that
software and aux events respect groups, their pmu_name is taken from
the group leader. The sorting is done with list_sort removing a memory
allocation.
After sorting a pass is done to correct the group leaders and for
topdown events ensuring they have a group leader.
This fixes the problems seen before:
```
$ perf stat -e '{uncore_imc_free_running/data_read/,uncore_imc_free_running/data_write/}' -a sleep 1
Performance counter stats for 'system wide':
727.42 MiB uncore_imc_free_running/data_read/
81.84 MiB uncore_imc_free_running/data_write/
1.000948615 seconds time elapsed
```
As well as making groups not fail for cases like:
```
$ perf stat -e '{imc_free_running_0/data_total/,imc_free_running_1/data_total/}' -a sleep 1
Performance counter stats for 'system wide':
256.47 MiB imc_free_running_0/data_total/
256.48 MiB imc_free_running_1/data_total/
1.001165442 seconds time elapsed
```
Signed-off-by: Ian Rogers <irogers@google.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Florian Fischer <florian.fischer@muhq.space>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: James Clark <james.clark@arm.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: John Garry <john.g.garry@oracle.com>
Cc: Kajol Jain <kjain@linux.ibm.com>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Kim Phillips <kim.phillips@amd.com>
Cc: Leo Yan <leo.yan@linaro.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ravi Bangoria <ravi.bangoria@amd.com>
Cc: Sean Christopherson <seanjc@google.com>
Cc: Steinar H. Gunderson <sesse@google.com>
Cc: Stephane Eranian <eranian@google.com>
Cc: Suzuki Poulouse <suzuki.poulose@arm.com>
Cc: Xing Zhengjun <zhengjun.xing@linux.intel.com>
Link: https://lore.kernel.org/r/20230312021543.3060328-2-irogers@google.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2023-03-11 18:15:40 -08:00
|
|
|
return -1;
|
2024-09-13 08:47:07 +00:00
|
|
|
if (arch_is_topdown_slots(rhs))
|
perf parse-events: Sort and group parsed events
This change is intended to be a no-op for most current cases, the
default sort order is the order the events were parsed. Where it
varies is in how groups are handled. Previously an uncore and core
event that are grouped would most often cause the group to be removed:
```
$ perf stat -e '{instructions,uncore_imc_free_running_0/data_total/}' -a sleep 1
WARNING: grouped events cpus do not match, disabling group:
anon group { instructions, uncore_imc_free_running_0/data_total/ }
...
```
However, when wildcards are used the events should be re-sorted and
re-grouped in parse_events__set_leader, but this currently fails for
simple examples:
```
$ perf stat -e '{uncore_imc_free_running/data_read/,uncore_imc_free_running/data_write/}' -a sleep 1
Performance counter stats for 'system wide':
<not counted> MiB uncore_imc_free_running/data_read/
<not counted> MiB uncore_imc_free_running/data_write/
1.000996992 seconds time elapsed
```
A futher failure mode, fixed in this patch, is to force topdown events
into a group.
This change moves sorting the evsels in the evlist after parsing. It
requires parsing to set up groups. First the evsels are sorted
respecting the existing groupings and parse order, but also reordering
to ensure evsels of the same PMU and group appear together. So that
software and aux events respect groups, their pmu_name is taken from
the group leader. The sorting is done with list_sort removing a memory
allocation.
After sorting a pass is done to correct the group leaders and for
topdown events ensuring they have a group leader.
This fixes the problems seen before:
```
$ perf stat -e '{uncore_imc_free_running/data_read/,uncore_imc_free_running/data_write/}' -a sleep 1
Performance counter stats for 'system wide':
727.42 MiB uncore_imc_free_running/data_read/
81.84 MiB uncore_imc_free_running/data_write/
1.000948615 seconds time elapsed
```
As well as making groups not fail for cases like:
```
$ perf stat -e '{imc_free_running_0/data_total/,imc_free_running_1/data_total/}' -a sleep 1
Performance counter stats for 'system wide':
256.47 MiB imc_free_running_0/data_total/
256.48 MiB imc_free_running_1/data_total/
1.001165442 seconds time elapsed
```
Signed-off-by: Ian Rogers <irogers@google.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Florian Fischer <florian.fischer@muhq.space>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: James Clark <james.clark@arm.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: John Garry <john.g.garry@oracle.com>
Cc: Kajol Jain <kjain@linux.ibm.com>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Kim Phillips <kim.phillips@amd.com>
Cc: Leo Yan <leo.yan@linaro.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ravi Bangoria <ravi.bangoria@amd.com>
Cc: Sean Christopherson <seanjc@google.com>
Cc: Steinar H. Gunderson <sesse@google.com>
Cc: Stephane Eranian <eranian@google.com>
Cc: Suzuki Poulouse <suzuki.poulose@arm.com>
Cc: Xing Zhengjun <zhengjun.xing@linux.intel.com>
Link: https://lore.kernel.org/r/20230312021543.3060328-2-irogers@google.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2023-03-11 18:15:40 -08:00
|
|
|
return 1;
|
2024-10-11 11:02:06 +00:00
|
|
|
|
2024-09-13 08:47:09 +00:00
|
|
|
/*
|
2024-10-11 11:02:06 +00:00
|
|
|
* Move topdown metrics events forward only when topdown metrics
|
|
|
|
* events are not in same group with previous slots event. If
|
|
|
|
* topdown metrics events are already in same group with slots
|
|
|
|
* event, do nothing.
|
2024-09-13 08:47:09 +00:00
|
|
|
*/
|
2025-03-06 18:39:04 -08:00
|
|
|
if (lhs->core.leader != rhs->core.leader) {
|
|
|
|
bool lhs_topdown = arch_is_topdown_metrics(lhs);
|
|
|
|
bool rhs_topdown = arch_is_topdown_metrics(rhs);
|
|
|
|
|
|
|
|
if (lhs_topdown && !rhs_topdown)
|
|
|
|
return -1;
|
|
|
|
if (!lhs_topdown && rhs_topdown)
|
|
|
|
return 1;
|
|
|
|
}
|
perf parse-events: Architecture specific leader override
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.
Reviewed-by: Kajol Jain <kjain@linux.ibm.com>
Signed-off-by: Ian Rogers <irogers@google.com>
Acked-by: Jiri Olsa <jolsa@redhat.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: John Garry <john.garry@huawei.com>
Cc: Kajol Jain <kjain@linux.ibm.com>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Paul Clarke <pc@us.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Riccardo Mancini <rickyman7@gmail.com>
Cc: Stephane Eranian <eranian@google.com>
Cc: Vineet Singh <vineet.singh@intel.com>
Link: http://lore.kernel.org/lkml/20211130174945.247604-2-irogers@google.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2021-11-30 09:49:45 -08:00
|
|
|
}
|
perf parse-events: Sort and group parsed events
This change is intended to be a no-op for most current cases, the
default sort order is the order the events were parsed. Where it
varies is in how groups are handled. Previously an uncore and core
event that are grouped would most often cause the group to be removed:
```
$ perf stat -e '{instructions,uncore_imc_free_running_0/data_total/}' -a sleep 1
WARNING: grouped events cpus do not match, disabling group:
anon group { instructions, uncore_imc_free_running_0/data_total/ }
...
```
However, when wildcards are used the events should be re-sorted and
re-grouped in parse_events__set_leader, but this currently fails for
simple examples:
```
$ perf stat -e '{uncore_imc_free_running/data_read/,uncore_imc_free_running/data_write/}' -a sleep 1
Performance counter stats for 'system wide':
<not counted> MiB uncore_imc_free_running/data_read/
<not counted> MiB uncore_imc_free_running/data_write/
1.000996992 seconds time elapsed
```
A futher failure mode, fixed in this patch, is to force topdown events
into a group.
This change moves sorting the evsels in the evlist after parsing. It
requires parsing to set up groups. First the evsels are sorted
respecting the existing groupings and parse order, but also reordering
to ensure evsels of the same PMU and group appear together. So that
software and aux events respect groups, their pmu_name is taken from
the group leader. The sorting is done with list_sort removing a memory
allocation.
After sorting a pass is done to correct the group leaders and for
topdown events ensuring they have a group leader.
This fixes the problems seen before:
```
$ perf stat -e '{uncore_imc_free_running/data_read/,uncore_imc_free_running/data_write/}' -a sleep 1
Performance counter stats for 'system wide':
727.42 MiB uncore_imc_free_running/data_read/
81.84 MiB uncore_imc_free_running/data_write/
1.000948615 seconds time elapsed
```
As well as making groups not fail for cases like:
```
$ perf stat -e '{imc_free_running_0/data_total/,imc_free_running_1/data_total/}' -a sleep 1
Performance counter stats for 'system wide':
256.47 MiB imc_free_running_0/data_total/
256.48 MiB imc_free_running_1/data_total/
1.001165442 seconds time elapsed
```
Signed-off-by: Ian Rogers <irogers@google.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Florian Fischer <florian.fischer@muhq.space>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: James Clark <james.clark@arm.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: John Garry <john.g.garry@oracle.com>
Cc: Kajol Jain <kjain@linux.ibm.com>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Kim Phillips <kim.phillips@amd.com>
Cc: Leo Yan <leo.yan@linaro.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ravi Bangoria <ravi.bangoria@amd.com>
Cc: Sean Christopherson <seanjc@google.com>
Cc: Steinar H. Gunderson <sesse@google.com>
Cc: Stephane Eranian <eranian@google.com>
Cc: Suzuki Poulouse <suzuki.poulose@arm.com>
Cc: Xing Zhengjun <zhengjun.xing@linux.intel.com>
Link: https://lore.kernel.org/r/20230312021543.3060328-2-irogers@google.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2023-03-11 18:15:40 -08:00
|
|
|
|
2024-07-20 02:20:56 -04:00
|
|
|
/* Retire latency event should not be group leader*/
|
|
|
|
if (lhs->retire_lat && !rhs->retire_lat)
|
|
|
|
return 1;
|
|
|
|
if (!lhs->retire_lat && rhs->retire_lat)
|
|
|
|
return -1;
|
|
|
|
|
perf parse-events: Sort and group parsed events
This change is intended to be a no-op for most current cases, the
default sort order is the order the events were parsed. Where it
varies is in how groups are handled. Previously an uncore and core
event that are grouped would most often cause the group to be removed:
```
$ perf stat -e '{instructions,uncore_imc_free_running_0/data_total/}' -a sleep 1
WARNING: grouped events cpus do not match, disabling group:
anon group { instructions, uncore_imc_free_running_0/data_total/ }
...
```
However, when wildcards are used the events should be re-sorted and
re-grouped in parse_events__set_leader, but this currently fails for
simple examples:
```
$ perf stat -e '{uncore_imc_free_running/data_read/,uncore_imc_free_running/data_write/}' -a sleep 1
Performance counter stats for 'system wide':
<not counted> MiB uncore_imc_free_running/data_read/
<not counted> MiB uncore_imc_free_running/data_write/
1.000996992 seconds time elapsed
```
A futher failure mode, fixed in this patch, is to force topdown events
into a group.
This change moves sorting the evsels in the evlist after parsing. It
requires parsing to set up groups. First the evsels are sorted
respecting the existing groupings and parse order, but also reordering
to ensure evsels of the same PMU and group appear together. So that
software and aux events respect groups, their pmu_name is taken from
the group leader. The sorting is done with list_sort removing a memory
allocation.
After sorting a pass is done to correct the group leaders and for
topdown events ensuring they have a group leader.
This fixes the problems seen before:
```
$ perf stat -e '{uncore_imc_free_running/data_read/,uncore_imc_free_running/data_write/}' -a sleep 1
Performance counter stats for 'system wide':
727.42 MiB uncore_imc_free_running/data_read/
81.84 MiB uncore_imc_free_running/data_write/
1.000948615 seconds time elapsed
```
As well as making groups not fail for cases like:
```
$ perf stat -e '{imc_free_running_0/data_total/,imc_free_running_1/data_total/}' -a sleep 1
Performance counter stats for 'system wide':
256.47 MiB imc_free_running_0/data_total/
256.48 MiB imc_free_running_1/data_total/
1.001165442 seconds time elapsed
```
Signed-off-by: Ian Rogers <irogers@google.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Florian Fischer <florian.fischer@muhq.space>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: James Clark <james.clark@arm.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: John Garry <john.g.garry@oracle.com>
Cc: Kajol Jain <kjain@linux.ibm.com>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Kim Phillips <kim.phillips@amd.com>
Cc: Leo Yan <leo.yan@linaro.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ravi Bangoria <ravi.bangoria@amd.com>
Cc: Sean Christopherson <seanjc@google.com>
Cc: Steinar H. Gunderson <sesse@google.com>
Cc: Stephane Eranian <eranian@google.com>
Cc: Suzuki Poulouse <suzuki.poulose@arm.com>
Cc: Xing Zhengjun <zhengjun.xing@linux.intel.com>
Link: https://lore.kernel.org/r/20230312021543.3060328-2-irogers@google.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2023-03-11 18:15:40 -08:00
|
|
|
/* Default ordering by insertion index. */
|
|
|
|
return lhs->core.idx - rhs->core.idx;
|
perf parse-events: Architecture specific leader override
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.
Reviewed-by: Kajol Jain <kjain@linux.ibm.com>
Signed-off-by: Ian Rogers <irogers@google.com>
Acked-by: Jiri Olsa <jolsa@redhat.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: John Garry <john.garry@huawei.com>
Cc: Kajol Jain <kjain@linux.ibm.com>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Paul Clarke <pc@us.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Riccardo Mancini <rickyman7@gmail.com>
Cc: Stephane Eranian <eranian@google.com>
Cc: Vineet Singh <vineet.singh@intel.com>
Link: http://lore.kernel.org/lkml/20211130174945.247604-2-irogers@google.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2021-11-30 09:49:45 -08:00
|
|
|
}
|
2025-07-18 20:05:16 -07:00
|
|
|
|
|
|
|
int arch_evlist__add_required_events(struct list_head *list)
|
|
|
|
{
|
|
|
|
struct evsel *pos, *metric_event = NULL;
|
|
|
|
int idx = 0;
|
|
|
|
|
|
|
|
if (!topdown_sys_has_perf_metrics())
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
list_for_each_entry(pos, list, core.node) {
|
|
|
|
if (arch_is_topdown_slots(pos)) {
|
|
|
|
/* Slots event already present, nothing to do. */
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
if (metric_event == NULL && arch_is_topdown_metrics(pos))
|
|
|
|
metric_event = pos;
|
|
|
|
idx++;
|
|
|
|
}
|
|
|
|
if (metric_event == NULL) {
|
|
|
|
/* No topdown metric events, nothing to do. */
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
return topdown_insert_slots_event(list, idx + 1, metric_event);
|
|
|
|
}
|