perf test: Enable system wide for metricgroups test
Ian Rogers
reviewerPATCH 1/2 perf test: Enable system wide for metricgroups test
Uncore events as group leaders fail in per-thread mode causing exit errors. Enable system-wide for metricgroup testing. This fixes the HPC metric group when tested on skylakex.
--- tools/perf/tests/shell/stat_all_metricgroups.sh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
tools/perf/tests/shell/stat_all_metricgroups.sh +1 -1
@@ -6,7 +6,7 @@ set -e for m in $(perf list --raw-dump metricgroups); do echo "Testing $m" - perf stat -M "$m" true + perf stat -M "$m" -a true done exit 0
Signature · 1 line
2.34.0.rc2.393.gf8c9666880-goog
Ian Rogers
reviewerPATCH 2/2 perf evsel: Improve error message for uncore events
When a group has multiple events and the leader fails it can yield errors like:
$ perf stat -e '{uncore_imc/cas_count_read/},instructions' /bin/true
Error:
The sys_perf_event_open() syscall returned with 22 (Invalid argument) for event (uncore_imc/cas_count_read/).
/bin/dmesg | grep -i perf may provide additional information.However, when not the group leader <not supported> is given:
$ perf stat -e '{instructions,uncore_imc/cas_count_read/}' /bin/true
...
1,619,057 instructions
<not supported> MiB uncore_imc/cas_count_read/This is necessary because get_group_fd will fail if the leader fails and is the direct result of the check on line 750 of builtin-stat.c in stat_handle_error that returns COUNTER_SKIP for the latter case.
This patch improves the error message to:
$ perf stat -e '{uncore_imc/cas_count_read/},instructions' /bin/true
Error:
Invalid event (uncore_imc/cas_count_read/) in per-thread mode, enable system wide with '-a'.--- tools/perf/util/evsel.c | 5 +++++ 1 file changed, 5 insertions(+)
tools/perf/util/evsel.c +5 -0
@@ -2950,6 +2950,11 @@ int evsel__open_strerror(struct evsel *evsel, struct target *target,return scnprintf (msg, size,"wrong clockid (%d)." , clockid);if (perf_missing_features.aux_output)return scnprintf (msg, size,"The 'aux_output' feature is not supported, update the kernel." ); +if ((evsel__leader (evsel) == evsel) && + (evsel->core.leader->nr_members >1 )) +return scnprintf (msg, size, +"Invalid event (%s) in per-thread mode, enable system wide with '-a'." , +evsel__name (evsel));break ;case ENODATA : return scnprintf(msg, size, "Cannot collect data source with the load latency event alone. "
Signature · 1 line
2.34.0.rc2.393.gf8c9666880-goog
Re: [PATCH 2/2] perf evsel: Improve error message for uncore events
Quoted 35 lines · Ian Rogers
When a group has multiple events and the leader fails it can yield errors like:
$ perf stat -e '{uncore_imc/cas_count_read/},instructions' /bin/true Error: The sys_perf_event_open() syscall returned with 22 (Invalid argument) for event (uncore_imc/cas_count_read/). /bin/dmesg | grep -i perf may provide additional information.However, when not the group leader <not supported> is given:
$ perf stat -e '{instructions,uncore_imc/cas_count_read/}' /bin/true ... 1,619,057 instructions <not supported> MiB uncore_imc/cas_count_read/This is necessary because
get_group_fdwill fail if the leader fails and is the direct result of the check on line 750 ofbuiltin-stat.cinstat_handle_errorthat returnsCOUNTER_SKIPfor the latter case.This patch improves the error message to:
$ perf stat -e '{uncore_imc/cas_count_read/},instructions' /bin/true Error: Invalid event (uncore_imc/cas_count_read/) in per-thread mode, enable system wide with '-a'.Signed-off-by Ian Rogers <[email protected]>--- tools/perf/util/evsel.c | 5 +++++ 1 file changed, 5 insertions(+)tools/perf/util/evsel.c +0
-0
tools/perf/util/evsel.c int evsel__open_strerror(struct evsel *evsel, struct target *target, · line 2957
Quoted 8 lines
tools/perf/util/evsel.c +5
-0@@ -2950,6 +2950,11 @@ int evsel__open_strerror(struct evsel *evsel, struct target *target,return scnprintf (msg, size,"wrong clockid (%d)." , clockid);if (perf_missing_features.aux_output)return scnprintf (msg, size,"The 'aux_output' feature is not supported, update the kernel." ); +if ((evsel__leader (evsel) == evsel) && + (evsel->core.leader->nr_members >1 )) +return scnprintf (msg, size, +"Invalid event (%s) in per-thread mode, enable system wide with '-a'." , +evsel__name (evsel));
should we rather check 'target' pointer for the per-thread mode? I'm not sure that per-thread mode will always be the case for the failure
jirka
Quoted 6 lines
tools/perf/util/evsel.c +0
-0break ;case ENODATA : return scnprintf(msg, size, "Cannot collect data source with the load latency event alone. "Signature · 1 line
2.34.0.rc2.393.gf8c9666880-goog
Ian Rogers
reviewerRe: [PATCH 2/2] perf evsel: Improve error message for uncore events
Quoted 32 lines · Jiri Olsa
Quoted 30 lines · Ian Rogers
When a group has multiple events and the leader fails it can yield errors like:
$ perf stat -e '{uncore_imc/cas_count_read/},instructions' /bin/true Error: The sys_perf_event_open() syscall returned with 22 (Invalid argument) for event (uncore_imc/cas_count_read/). /bin/dmesg | grep -i perf may provide additional information.However, when not the group leader <not supported> is given:
$ perf stat -e '{instructions,uncore_imc/cas_count_read/}' /bin/true ... 1,619,057 instructions <not supported> MiB uncore_imc/cas_count_read/This is necessary because
get_group_fdwill fail if the leader fails and is the direct result of the check on line 750 ofbuiltin-stat.cinstat_handle_errorthat returnsCOUNTER_SKIPfor the latter case.This patch improves the error message to:
$ perf stat -e '{uncore_imc/cas_count_read/},instructions' /bin/true Error: Invalid event (uncore_imc/cas_count_read/) in per-thread mode, enable system wide with '-a'.Signed-off-by Ian Rogers <[email protected]>--- tools/perf/util/evsel.c | 5 +++++ 1 file changed, 5 insertions(+)
Quoted 16 lines
tools/perf/util/evsel.c int evsel__open_strerror(struct evsel *evsel, struct target *target, · line 2957
Quoted 13 lines
tools/perf/util/evsel.c +5
-0@@ -2950,6 +2950,11 @@ int evsel__open_strerror(struct evsel *evsel, struct target *target,return scnprintf (msg, size,"wrong clockid (%d)." , clockid);if (perf_missing_features.aux_output)return scnprintf (msg, size,"The 'aux_output' feature is not supported, update the kernel." ); +if ((evsel__leader (evsel) == evsel) && + (evsel->core.leader->nr_members >1 )) +return scnprintf (msg, size, +"Invalid event (%s) in per-thread mode, enable system wide with '-a'." , +evsel__name (evsel));should we rather check 'target' pointer for the per-thread mode? I'm not sure that per-thread mode will always be the case for the failure
Unfortunately the target isn't populated at that point:
gdb --args perf stat -e '{uncore_imc/cas_count_write/},instructions' /bin/true (gdb) p *target
$2 = {pid = 0x0, tid = 0x0, cpu_list = 0x0, uid_str = 0x0, bpf_str = 0x0, uid = 4294967295, system_wide = false, uses_mmap = false,
default_per_cpu = false, per_thread = false, use_bpf = false, hybrid
= false, attr_map = 0x0}
#0 evsel__open_strerror (evsel=0x616000015680, target=0x5555586aa140 <target>, err=22, msg=0x7fffffff78d0 "]k\264WUU", size=8192)
at util/evsel.c:2857
#1 0x00005555561744e8 in stat_handle_error (counter=0x616000015680) at builtin-stat.c:771
#2 0x0000555556174f05 in __run_perf_stat (argc=1, argv=0x7fffffffe450, run_idx=0) at builtin-stat.c:852
#3 0x00005555561763e1 in run_perf_stat (argc=1, argv=0x7fffffffe450, run_idx=0) at builtin-stat.c:1048
#4 0x000055555617df82 in cmd_stat (argc=1, argv=0x7fffffffe450) at builtin-stat.c:2550
#5 0x00005555562f36b8 in run_builtin (p=0x5555586bad00 <commands+288>, argc=4, argv=0x7fffffffe450) at perf.c:313
#6 0x00005555562f3c11 in handle_internal_command (argc=4, argv=0x7fffffffe450) at perf.c:365
#7 0x00005555562f3fce in run_argv (argcp=0x7fffffffe230, argv=0x7fffffffe240) at perf.c:409
#8 0x00005555562f47bd in main (argc=4, argv=0x7fffffffe450) at perf.c:539
Thanks,
Ian
Quoted 9 lines
jirka
Quoted 6 lines
tools/perf/util/evsel.c +0
-1break ;case ENODATA : return scnprintf(msg, size, "Cannot collect data source with the load latency event alone. " --2.34.0.rc2.393.gf8c9666880-goog
Namhyung Kim
maintainerRe: [PATCH 2/2] perf evsel: Improve error message for uncore events
Hi Ian,
Quoted 51 lines · Ian Rogers
Quoted 32 lines · Jiri Olsa
Quoted 30 lines · Ian Rogers
When a group has multiple events and the leader fails it can yield errors like:
$ perf stat -e '{uncore_imc/cas_count_read/},instructions' /bin/true Error: The sys_perf_event_open() syscall returned with 22 (Invalid argument) for event (uncore_imc/cas_count_read/). /bin/dmesg | grep -i perf may provide additional information.However, when not the group leader <not supported> is given:
$ perf stat -e '{instructions,uncore_imc/cas_count_read/}' /bin/true ... 1,619,057 instructions <not supported> MiB uncore_imc/cas_count_read/This is necessary because
get_group_fdwill fail if the leader fails and is the direct result of the check on line 750 ofbuiltin-stat.cinstat_handle_errorthat returnsCOUNTER_SKIPfor the latter case.This patch improves the error message to:
$ perf stat -e '{uncore_imc/cas_count_read/},instructions' /bin/true Error: Invalid event (uncore_imc/cas_count_read/) in per-thread mode, enable system wide with '-a'.Signed-off-by Ian Rogers <[email protected]>--- tools/perf/util/evsel.c | 5 +++++ 1 file changed, 5 insertions(+)Quoted 16 lines
tools/perf/util/evsel.c int evsel__open_strerror(struct evsel *evsel, struct target *target, · line 2957
Quoted 13 lines
tools/perf/util/evsel.c +5
-0@@ -2950,6 +2950,11 @@ int evsel__open_strerror(struct evsel *evsel, struct target *target,return scnprintf (msg, size,"wrong clockid (%d)." , clockid);if (perf_missing_features.aux_output)return scnprintf (msg, size,"The 'aux_output' feature is not supported, update the kernel." ); +if ((evsel__leader (evsel) == evsel) && + (evsel->core.leader->nr_members >1 )) +return scnprintf (msg, size, +"Invalid event (%s) in per-thread mode, enable system wide with '-a'." , +evsel__name (evsel));should we rather check 'target' pointer for the per-thread mode? I'm not sure that per-thread mode will always be the case for the failure
Quoted 1 line
Unfortunately the target isn't populated at that point:
It might be populated properly, as in this case
it should have no target. I think you can use
!target__has_cpu().
Thanks,
Namhyung
Quoted 39 lines
gdb --args perf stat -e '{uncore_imc/cas_count_write/},instructions'
/bin/true(gdb) p *target
$2 = {pid =0x0, tid =0x0,cpu_list=0x0,uid_str=0x0,bpf_str=0x0, uid = 4294967295,system_wide= false,uses_mmap= false,
default_per_cpu= false,per_thread= false,use_bpf= false, hybrid
= false,attr_map=0x0}#0
evsel__open_strerror(evsel=0x616000015680, target=0x5555586aa140 <target>, err=22, msg=0x7fffffff78d0 "]k\264WUU", size=8192)at util/evsel.c:2857#1
0x00005555561744e8instat_handle_error(counter=0x616000015680) at builtin-stat.c:771
#20x0000555556174f05in__run_perf_stat(argc=1, argv=0x7fffffffe450, run_idx=0) at builtin-stat.c:852
#30x00005555561763e1inrun_perf_stat(argc=1, argv=0x7fffffffe450, run_idx=0) at builtin-stat.c:1048
#40x000055555617df82incmd_stat(argc=1, argv=0x7fffffffe450) at builtin-stat.c:2550
#50x00005555562f36b8inrun_builtin(p=0x5555586bad00 <commands+288>, argc=4, argv=0x7fffffffe450) at perf.c:313
#60x00005555562f3c11inhandle_internal_command(argc=4, argv=0x7fffffffe450) at perf.c:365
#70x00005555562f3fceinrun_argv(argcp=0x7fffffffe230, argv=0x7fffffffe240) at perf.c:409
#80x00005555562f47bdin main (argc=4, argv=0x7fffffffe450) at perf.c:539Thanks,
IanQuoted 9 lines
jirka
Quoted 6 lines
tools/perf/util/evsel.c +0
-1break ;case ENODATA : return scnprintf(msg, size, "Cannot collect data source with the load latency event alone. " --2.34.0.rc2.393.gf8c9666880-goog
Ian Rogers
reviewerRe: [PATCH 2/2] perf evsel: Improve error message for uncore events
Quoted 55 lines · Namhyung Kim
Hi Ian,
Quoted 51 lines · Ian Rogers
Quoted 48 lines · Jiri Olsa
Quoted 35 lines · Ian Rogers
When a group has multiple events and the leader fails it can yield errors like:
$ perf stat -e '{uncore_imc/cas_count_read/},instructions' /bin/true Error: The sys_perf_event_open() syscall returned with 22 (Invalid argument) for event (uncore_imc/cas_count_read/). /bin/dmesg | grep -i perf may provide additional information.However, when not the group leader <not supported> is given:
$ perf stat -e '{instructions,uncore_imc/cas_count_read/}' /bin/true ... 1,619,057 instructions <not supported> MiB uncore_imc/cas_count_read/This is necessary because
get_group_fdwill fail if the leader fails and is the direct result of the check on line 750 ofbuiltin-stat.cinstat_handle_errorthat returnsCOUNTER_SKIPfor the latter case.This patch improves the error message to:
$ perf stat -e '{uncore_imc/cas_count_read/},instructions' /bin/true Error: Invalid event (uncore_imc/cas_count_read/) in per-thread mode, enable system wide with '-a'.Signed-off-by Ian Rogers <[email protected]>--- tools/perf/util/evsel.c | 5 +++++ 1 file changed, 5 insertions(+)tools/perf/util/evsel.c +0
-0tools/perf/util/evsel.c int evsel__open_strerror(struct evsel *evsel, struct target *target, · line 2957
Quoted 8 lines
tools/perf/util/evsel.c +5
-0@@ -2950,6 +2950,11 @@ int evsel__open_strerror(struct evsel *evsel, struct target *target,return scnprintf (msg, size,"wrong clockid (%d)." , clockid);if (perf_missing_features.aux_output)return scnprintf (msg, size,"The 'aux_output' feature is not supported, update the kernel." ); +if ((evsel__leader (evsel) == evsel) && + (evsel->core.leader->nr_members >1 )) +return scnprintf (msg, size, +"Invalid event (%s) in per-thread mode, enable system wide with '-a'." , +evsel__name (evsel));should we rather check 'target' pointer for the per-thread mode? I'm not sure that per-thread mode will always be the case for the failure
Quoted 5 lines
Quoted 1 line
Unfortunately the target isn't populated at that point:
It might be populated properly, as in this case
it should have no target. I think you can use
!target__has_cpu().
Thanks :-) I'm not sure having !target__has_cpu is intention revealing. Perhaps changing the error to say that an invalid/unsupported event can't be the group leader, unless it is the only event, and then suggest system wide mode.
Thanks,
Ian
Quoted 43 lines
Thanks,
NamhyungQuoted 39 lines
gdb --args perf stat -e '{uncore_imc/cas_count_write/},instructions'
/bin/true(gdb) p *target
$2 = {pid =0x0, tid =0x0,cpu_list=0x0,uid_str=0x0,bpf_str=0x0, uid = 4294967295,system_wide= false,uses_mmap= false,
default_per_cpu= false,per_thread= false,use_bpf= false, hybrid
= false,attr_map=0x0}#0
evsel__open_strerror(evsel=0x616000015680, target=0x5555586aa140 <target>, err=22, msg=0x7fffffff78d0 "]k\264WUU", size=8192)at util/evsel.c:2857#1
0x00005555561744e8instat_handle_error(counter=0x616000015680) at builtin-stat.c:771
#20x0000555556174f05in__run_perf_stat(argc=1, argv=0x7fffffffe450, run_idx=0) at builtin-stat.c:852
#30x00005555561763e1inrun_perf_stat(argc=1, argv=0x7fffffffe450, run_idx=0) at builtin-stat.c:1048
#40x000055555617df82incmd_stat(argc=1, argv=0x7fffffffe450) at builtin-stat.c:2550
#50x00005555562f36b8inrun_builtin(p=0x5555586bad00 <commands+288>, argc=4, argv=0x7fffffffe450) at perf.c:313
#60x00005555562f3c11inhandle_internal_command(argc=4, argv=0x7fffffffe450) at perf.c:365
#70x00005555562f3fceinrun_argv(argcp=0x7fffffffe230, argv=0x7fffffffe240) at perf.c:409
#80x00005555562f47bdin main (argc=4, argv=0x7fffffffe450) at perf.c:539Thanks,
IanQuoted 9 lines
jirka
Quoted 6 lines
tools/perf/util/evsel.c +0
-1break ;case ENODATA : return scnprintf(msg, size, "Cannot collect data source with the load latency event alone. " --2.34.0.rc2.393.gf8c9666880-goog