Age | Commit message (Collapse) | Author | Files | Lines |
|
For a metric like:
EVENT1 if #smt_on else EVENT2
currently EVENT1 and EVENT2 will be measured and then when the metric is
reported EVENT1 or EVENT2 will be printed depending on the value from
smt_on() during the expr parsing. Computing both events is unnecessary and
can lead to multiplexing as discussed in this thread:
https://lore.kernel.org/lkml/[email protected]/
If the input is constant to certain operators like:
IDS1 if CONST else IDS2
then the result will be either IDS1 or IDS2 depending on CONST (which
may be evaluated from an entire expression), and so IDS1 or IDS2 may
be discarded avoiding events from being programmed.
Signed-off-by: Ian Rogers <[email protected]>
Tested-by: John Garry <[email protected]>
Acked-by: Jiri Olsa <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jin Yao <[email protected]>
Cc: Kajol Jain <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Paul Clarke <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Sandeep Dasgupta <[email protected]>
Cc: Stephane Eranian <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
When we're computing ID values, if we have constant values then compute
the constant result. For example:
1 + 2
Previously .val would be set to BOTTOM by union_expr, meaning that
all values are possible. With this change .val is set to 3.
Later changes will use the constant values to hopefully eliminate ID
values that don't need to be computed.
Signed-off-by: Ian Rogers <[email protected]>
Tested-by: John Garry <[email protected]>
Acked-by: Jiri Olsa <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jin Yao <[email protected]>
Cc: Kajol Jain <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Paul Clarke <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Sandeep Dasgupta <[email protected]>
Cc: Stephane Eranian <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Add a new option to parsing that the set of IDs being used should be
computed, this means every action needs to handle the compute_ids and
regular case. This means actions yield a new ids type is a set of ids or
the value being computed. Use the compute_ids case to replace find IDs
parsing.
Signed-off-by: Ian Rogers <[email protected]>
Tested-by: John Garry <[email protected]>
Acked-by: Jiri Olsa <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jin Yao <[email protected]>
Cc: Kajol Jain <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Paul Clarke <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Sandeep Dasgupta <[email protected]>
Cc: Stephane Eranian <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
A metric may be a constant value, for example, some SMT metrics are
constant 0 if #smt_on is 0. If we eliminate all the events then there is
no printing. Fix this by forcing metrics like this to have a
duration_time tool event, previously the metric would fail when parsing
the events with a parse error.
Signed-off-by: Ian Rogers <[email protected]>
Tested-by: John Garry <[email protected]>
Acked-by: Jiri Olsa <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jin Yao <[email protected]>
Cc: Kajol Jain <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Paul Clarke <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Sandeep Dasgupta <[email protected]>
Cc: Stephane Eranian <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
[ Reflow one __parse_events() call so that a ternary operation gets in a single line ]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Add utilities to new/free an ids hashmap, as well as to union. Add
testing of the union. Unioning hashmaps will be used when parsing the
metric, if a value is known then the hashmap is unnecessary, otherwise
we need to union together all the event ids to compute their values for
reporting.
Signed-off-by: Ian Rogers <[email protected]>
Tested-by: John Garry <[email protected]>
Acked-by: Jiri Olsa <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jin Yao <[email protected]>
Cc: Kajol Jain <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Paul Clarke <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Sandeep Dasgupta <[email protected]>
Cc: Stephane Eranian <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
A later change will remove the notion of other, rename the function to
expr__find_ids as this is what it populates.
Signed-off-by: Ian Rogers <[email protected]>
Tested-by: John Garry <[email protected]>
Acked-by: Jiri Olsa <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jin Yao <[email protected]>
Cc: Kajol Jain <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Paul Clarke <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Sandeep Dasgupta <[email protected]>
Cc: Stephane Eranian <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
No functional change, just modifying whitespace. This creates additional
space for adding logic to actions in later changes.
Signed-off-by: Ian Rogers <[email protected]>
Tested-by: John Garry <[email protected]>
Acked-by: Jiri Olsa <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jin Yao <[email protected]>
Cc: Kajol Jain <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Paul Clarke <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Sandeep Dasgupta <[email protected]>
Cc: Stephane Eranian <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
No functional change, switch the operators to use macros so that
additional complexity for constants can be added in a later change.
Signed-off-by: Ian Rogers <[email protected]>
Tested-by: John Garry <[email protected]>
Acked-by: Jiri Olsa <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jin Yao <[email protected]>
Cc: Kajol Jain <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Paul Clarke <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Sandeep Dasgupta <[email protected]>
Cc: Stephane Eranian <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
No functional change, so the type of expr remains <num>. A later patch
will change the computation to be an aggregate type and making this
change makes that later change smaller.
Signed-off-by: Ian Rogers <[email protected]>
Tested-by: John Garry <[email protected]>
Acked-by: Jiri Olsa <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jin Yao <[email protected]>
Cc: Kajol Jain <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Paul Clarke <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Sandeep Dasgupta <[email protected]>
Cc: Stephane Eranian <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
No functional change. Inlining d_ratio makes it easier to special case
for constants in a later patch.
Signed-off-by: Ian Rogers <[email protected]>
Tested-by: John Garry <[email protected]>
Acked-by: Jiri Olsa <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Kajol Jain <[email protected]>
Cc: Jin Yao <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Paul Clarke <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Sandeep Dasgupta <[email protected]>
Cc: Stephane Eranian <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
If during computing a metric an event (id) is missing the parsing
aborts. A later patch will make it so that events that aren't used in
the output are deliberately omitted, in which case we don't want the
abort. Modify the missing ID case to report NAN for these cases.
Reviewed-by: Andi Kleen <[email protected]>
Signed-off-by: Ian Rogers <[email protected]>
Tested-by: John Garry <[email protected]>
Acked-by: Jiri Olsa <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jin Yao <[email protected]>
Cc: Kajol Jain <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Paul Clarke <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Sandeep Dasgupta <[email protected]>
Cc: Stephane Eranian <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
A later change to parsing the ids out (in expr__find_other) will
potentially drop hashmaps and so it is more convenient to move
expr_parse_ctx to have a hashmap pointer rather than a struct value.
As this pointer must be freed, rather than just going out of scope, add
expr__ctx_new and expr__ctx_free to manage expr_parse_ctx memory.
Adjust use of struct expr_parse_ctx accordingly.
Reviewed-by: Andi Kleen <[email protected]>
Signed-off-by: Ian Rogers <[email protected]>
Tested-by: John Garry <[email protected]>
Acked-by: Jiri Olsa <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jin Yao <[email protected]>
Cc: Kajol Jain <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Paul Clarke <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Sandeep Dasgupta <[email protected]>
Cc: Stephane Eranian <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
To improve alias matching, remove the PMU name prefix from the
EventName. This will mean that the pmu code will merge aliases, such
that we no longer get a huge list of per-PMU events - see
perf_pmu_merge_alias().
Also make the following associated changes:
- Use "ConfigCode" rather than "EventCode", so the pmu code is not so
disagreeable about inconsistent event codes
- Add undocumented HHA event codes to allow alias merging (for those
events)
Signed-off-by: John Garry <[email protected]>
Acked-by: Ian Rogers <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Leo Yan <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Mathieu Poirier <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Shaokun Zhang <[email protected]>
Cc: Will Deacon <[email protected]>
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Add a new test event for a system event whose event member is in form
"config=".
Signed-off-by: John Garry <[email protected]>
Acked-by: Ian Rogers <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Leo Yan <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Mathieu Poirier <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Shaokun Zhang <[email protected]>
Cc: Will Deacon <[email protected]>
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Function compare_pmu_events() does not compare all struct pmu-events
members, so add tests for missing members "name", "event", "aggr_mod",
"event", "metric_constraint", and "metric_group", and re-order the tests
to match current struct pmu-events member ordering.
Also fix uncore_hisi_l3c_rd_hit_cpipe.event member, now that we're
actually testing it.
Signed-off-by: John Garry <[email protected]>
Acked-by: Ian Rogers <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Leo Yan <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Mathieu Poirier <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Shaokun Zhang <[email protected]>
Cc: Will Deacon <[email protected]>
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Some PMUs use "config=XXX" for eventcodes, like:
more /sys/bus/event_source/devices/hisi_sccl1_ddrc3/events/act_cmd
config=0x5
However jevents would give an alias with .event field "event=0x5" for
this event. This is handled without issue by the parse events code, but
the pmu alias code gets a bit confused, as it warns about assigning
"event=0x5" over "config=0x5" in perf_pmu_assign_str() when merging
aliases: ./perf stat -v -e act_cmd ... alias act_cmd differs in field
'value' ...
To make things a bit more straightforward, allow jevents to support
"config=XXX" as well, by supporting a "ConfigCode" field.
Signed-off-by: John Garry <[email protected]>
Acked-by: Ian Rogers <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Leo Yan <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Mathieu Poirier <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Shaokun Zhang <[email protected]>
Cc: Will Deacon <[email protected]>
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
For numeric terms, the config field may be NULL as it is not set from
the l+y parsing.
Fix by setting the term config from the term type name.
Also fix up the pmu-events test to set the alias strings to set the
period term properly, and fix up parse-events test to check the term
config string.
Signed-off-by: John Garry <[email protected]>
Acked-by: Ian Rogers <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Leo Yan <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Mathieu Poirier <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Shaokun Zhang <[email protected]>
Cc: Will Deacon <[email protected]>
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
libtraceevent has added more levels of debug printout and with changes
like:
https://lore.kernel.org/linux-trace-devel/[email protected]
previously generated output like "registering plugin" is no longer
displayed.
This change makes it so that if perf's verbose debug output is enabled
then the debug and info libtraceevent messages can be displayed.
The code is conditionally enabled based on the libtraceevent version as
discussed in the RFC:
https://lore.kernel.org/lkml/[email protected]/
v2. Is a rebase and handles the case of building without
LIBTRACEEVENT_DYNAMIC.
Signed-off-by: Ian Rogers <[email protected]>
Acked-by: Jiri Olsa <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Stephane Eranian <[email protected]>
Cc: Steven Rostedt (VMware) <[email protected]>
Cc: Tzvetomir Stoyanov (VMware) <[email protected]>
Cc: [email protected]
Link: http://lore.kernel.org/lkml/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
This will allow version specific support of libtracefs.
Signed-off-by: Ian Rogers <[email protected]>
Acked-by: Jiri Olsa <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Stephane Eranian <[email protected]>
Cc: Steven Rostedt (VMware) <[email protected]>
Cc: Tzvetomir Stoyanov (VMware) <[email protected]>
Cc: [email protected]
Link: http://lore.kernel.org/lkml/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
The definition is derived from pkg-config as discussed in:
https://lore.kernel.org/lkml/[email protected]/
The definition is computed using expr rather than passed to be computed
in C code, this avoids complications with quote in the variable
expansions.
For example see the target python/perf.so in Makefile.perf.
Signed-off-by: Ian Rogers <[email protected]>
Acked-by: Jiri Olsa <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Stephane Eranian <[email protected]>
Cc: Steven Rostedt (VMware) <[email protected]>
Cc: Tzvetomir Stoyanov (VMware) <[email protected]>
Cc: [email protected]
Link: http://lore.kernel.org/lkml/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Currently libtracefs isn't used by perf, but there are potential
improvements by using it as identified Steven Rostedt's e-mail:
https://lore.kernel.org/lkml/[email protected]/
This change is modelled on the dynamic libtraceevent patch by Michael
Petlan:
https://lore.kernel.org/linux-perf-users/[email protected]/
v3. Adds file missed in v1 and v2 spotted by Jiri Olsa.
Signed-off-by: Ian Rogers <[email protected]>
Acked-by: Jiri Olsa <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Stephane Eranian <[email protected]>
Cc: Steven Rostedt (VMware) <[email protected]>
Cc: Tzvetomir Stoyanov (VMware) <[email protected]>
Cc: [email protected]
Link: http://lore.kernel.org/lkml/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Iterate over the list of PMUs and run the 'true' workload on them. If
the event isn't printed then run the large 'perf bench internals
synthesize' workload and check the event is counted.
On a Skylake this test takes 1m15s mainly running the 'true' workload.
Suggested-by: John Garry <[email protected]>
Signed-off-by: Ian Rogers <[email protected]>
Reviewed-by: John Garry <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Jin Yao <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Paul Clarke <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Stephane Eranian <[email protected]>
Link: http://lore.kernel.org/lkml/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Test every metric and metricgroup with 'true' as a workload. For
metrics, check that we see the metric printed or get unsupported. If the
'true' workload executes too quickly retry with 'perf bench internals
synthesize'.
v3. Fix test condition (thanks to Paul A. Clarke <[email protected]>). Add a
fallback case of a larger workload so that we don't ignore "<not
counted>".
v2. Switched the workload to something faster.
Signed-off-by: Ian Rogers <[email protected]>
Reviewed-by: John Garry <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Jin Yao <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Paul Clarke <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Stephane Eranian <[email protected]>
Link: http://lore.kernel.org/lkml/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
The tools/perf/pmu-events/jevents.c file isn't being compiled with
-Werror and -Wextra, which will be the case soon, so before we turn
those compiler flags on, fix what it would flag.
Cc: Alexander Shishkin <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Like Xu <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
To: John Garry <[email protected]>
|
|
If the 'perf iostat' user specifies two or more iio_root_ports and also
specifies the cpu(s) by -C which is not *connected to all* the above iio
ports, the iostat_print_metric() will run into trouble:
For example:
$ perf iostat list
S0-uncore_iio_0<0000:16>
S1-uncore_iio_0<0000:97> # <--- CPU 1 is located in the socket S0
$ perf iostat 0000:16,0000:97 -C 1 -- ls
port Inbound Read(MB) Inbound Write(MB) Outbound Read(MB) Outbound
Write(MB) ../perf-iostat: line 12: 104418 Segmentation fault
(core dumped) perf stat --iostat$DELIMITER$*
The core-dump stack says, in the above corner case, the returned
(struct perf_counts_values *) count will be NULL, and the caller
iostat_print_metric() apparently doesn't not handle this case.
433 struct perf_counts_values *count = perf_counts(evsel->counts, die, 0);
434
435 if (count->run && count->ena) {
(gdb) p count
$1 = (struct perf_counts_values *) 0x0
The deeper reason is that there are actually no statistics from the user
specified pair "iostat 0000:X, -C (disconnected) Y ", but let's fix it with
minimum cost by adding a NULL check in the user space.
Fixes: f9ed693e8bc0e7de ("perf stat: Enable iostat mode for x86 platforms")
Signed-off-by: Like Xu <[email protected]>
Cc: Alexander Antonov <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Ian Rogers <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Stephane Eranian <[email protected]>
Link: http://lore.kernel.org/lkml/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
An iostate use case like "perf iostat 0000:16,0000:97 -- ls" should be
implemented to work in system-wide mode to ensure that the output from
print_header() is consistent with the user documentation perf-iostat.txt,
rather than incorrectly assuming that the kernel does not support it:
Error:
The sys_perf_event_open() syscall returned with 22 (Invalid argument) \
for event (uncore_iio_0/event=0x83,umask=0x04,ch_mask=0xF,fc_mask=0x07/).
/bin/dmesg | grep -i perf may provide additional information.
This error is easily fixed by assigning system-wide mode by default
for IOSTAT_RUN only when the target cpu_list is unspecified.
Fixes: f07952b179697771 ("perf stat: Basic support for iostat in perf")
Signed-off-by: Like Xu <[email protected]>
Cc: Alexander Antonov <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Ian Rogers <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Stephane Eranian <[email protected]>
Link: http://lore.kernel.org/lkml/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
This patch adds basic arch initialization and instruction associate
support for the riscv64 CPU architecture.
Example output:
$ perf annotate --stdio2
Samples: 122K of event 'task-clock:u', 4000 Hz, Event count (approx.): 30637250000, [percent: local period]
strcmp() /usr/lib64/libc-2.32.so
Percent
Disassembly of section .text:
0000000000069a30 <strcmp>:
__GI_strcmp():
const unsigned char *s2 = (const unsigned char *) p2;
unsigned char c1, c2;
do
{
c1 = (unsigned char) *s1++;
37.30 lbu a5,0(a0)
c2 = (unsigned char) *s2++;
1.23 addi a1,a1,1
c1 = (unsigned char) *s1++;
18.68 addi a0,a0,1
c2 = (unsigned char) *s2++;
1.37 lbu a4,-1(a1)
if (c1 == '\0')
18.71 ↓ beqz a5,18
return c1 - c2;
}
Signed-off-by: William Cohen <[email protected]>
Cc: Albert Ou <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Palmer Dabbelt <[email protected]>
Cc: Paul Walmsley <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: [email protected]
Link: http://lore.kernel.org/lkml/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
If there is no configuration file at first, the user can write any pair
of "key.subkey=value" to the newly created configuration file, while
value validation against a valid configurable key is *deferred* until
the next execution or the implied execution of "perf config ... ".
For example:
$ rm ~/.perfconfig
$ perf config call-graph.dump-size=65529
$ cat ~/.perfconfig
# this file is auto-generated.
[call-graph]
dump-size = 65529
$ perf config call-graph.dump-size=2048
callchain: Incorrect stack dump size (max 65528): 65529
Error: wrong config key-value pair call-graph.dump-size=65529
The user might expect that the second value 2048 is valid and can be
updated to the configuration file, but the error message is very
confusing because the first value 65529 is not reported as an error
during the last configuration.
It is recommended not to change the current behavior of delayed
validation (as more effort is needed), but to refine the original error
message to *clearly indicate* that the cause of the error is the
configuration file.
Signed-off-by: Like Xu <[email protected]>
Acked-by: Ian Rogers <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Link: http://lore.kernel.org/lkml/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Considering that perf and its subcommands have so many parameters, the
documentation is always the first stop for perf beginners. Fixing some
spelling errors will relax the eyes of some readers a little bit.
s/specicfication/specification/
s/caheline/cacheline/
s/tranasaction/transaction/
s/complan/complain/
s/sched_wakep/sched_wakeup/
s/possble/possible/
s/methology/methodology/
Signed-off-by: Like Xu <[email protected]>
Acked-by: Ian Rogers <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Jonathan Corbet <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Stephane Eranian <[email protected]>
Link: http://lore.kernel.org/lkml/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Relative path include works in the regular build due to -I paths but may
fail in other situations.
v2. Rebase. Comments on v1 were that we should handle include paths
differently and it is agreed that can be a sensible refactor but
beyond the scope of this change.
https://lore.kernel.org/lkml/[email protected]/
Signed-off-by: Ian Rogers <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: John Garry <[email protected]>
Cc: Leo Yan <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Mathieu Poirier <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Sergey Senozhatsky <[email protected]>
Cc: Stephane Eranian <[email protected]>
Cc: Will Deacon <[email protected]>
Cc: [email protected]
Link: http://lore.kernel.org/lkml/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
There is a spelling mistake in the description text, fix it.
Signed-off-by: Colin King <[email protected]>
Acked-by: Ian Rogers <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: [email protected]
Link: http://lore.kernel.org/lkml/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
This test occasionally fails on aarch64 when a sample is taken in
free@plt and it fails with "Bytes read differ from those read by
objdump".
This is because that symbol is near a section boundary in the elf file.
Despite the -z option to always output zeros, objdump uses
bfd_map_over_sections() to iterate through the elf file so it doesn't
see outside of the sections where these zeros are and can't print them.
For example this boundary proceeds free@plt in libc with a gap of 48
bytes between .plt and .text:
objdump -d -z --start-address=0x23cc8 --stop-address=0x23d08 libc-2.30.so
libc-2.30.so: file format elf64-littleaarch64
Disassembly of section .plt:
0000000000023cc8 <*ABS*+0x7fd00@plt+0x8>:
23cc8: 91018210 add x16, x16, #0x60
23ccc: d61f0220 br x17
Disassembly of section .text:
0000000000023d00 <abort@@GLIBC_2.17-0x98>:
23d00: a9bf7bfd stp x29, x30, [sp, #-16]!
23d04: 910003fd mov x29, sp
Taking a sample in free@plt is very rare because it is so small, but the
test can be forced to fail almost every time on any platform by linking
the test with a shared library that has a single empty function and
calling it in a loop.
The fix is to zero the buffers so that when there is a jump in the
addresses output by objdump, zeros are already filled in between.
Signed-off-by: James Clark <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Namhyung Kim <[email protected]>
Link: http://lore.kernel.org/lkml/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
To ensure the stack frames are on the stack tail calls optimizations
need to be inhibited. If your compiler supports an attribute use it,
otherwise use an asm volatile barrier.
The barrier fix was suggested here:
https://lore.kernel.org/lkml/[email protected]/
Tested with an optimized clang build and by forcing the asm barrier
route with an optimized clang build.
A GCC bug tracking a proper disable_tail_calls is:
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97831
Fixes: 9ae1e990f1ab ("perf tools: Remove broken __no_tail_call
attribute")
v2. is a rebase. The original fix patch generated quite a lot of
discussion over the right place for the fix:
https://lore.kernel.org/lkml/[email protected]/
The patch reflects my preference of it being near the use, so that
future code cleanups don't break this somewhat special usage.
Signed-off-by: Ian Rogers <[email protected]>
Acked-by: Jiri Olsa <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Ard Biesheuvel <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Miguel Ojeda <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Stephane Eranian <[email protected]>
Cc: [email protected]
Link: http://lore.kernel.org/lkml/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Part of hardware cache events are only available on one CPU PMU.
For example, 'L1-dcache-load-misses' is only available on cpu_core.
perf list should clearly report this info.
root@otcpl-adl-s-2:~# ./perf list
Before:
L1-dcache-load-misses [Hardware cache event]
L1-dcache-loads [Hardware cache event]
L1-dcache-stores [Hardware cache event]
L1-icache-load-misses [Hardware cache event]
L1-icache-loads [Hardware cache event]
LLC-load-misses [Hardware cache event]
LLC-loads [Hardware cache event]
LLC-store-misses [Hardware cache event]
LLC-stores [Hardware cache event]
branch-load-misses [Hardware cache event]
branch-loads [Hardware cache event]
dTLB-load-misses [Hardware cache event]
dTLB-loads [Hardware cache event]
dTLB-store-misses [Hardware cache event]
dTLB-stores [Hardware cache event]
iTLB-load-misses [Hardware cache event]
node-load-misses [Hardware cache event]
node-loads [Hardware cache event]
node-store-misses [Hardware cache event]
node-stores [Hardware cache event]
After:
L1-dcache-loads [Hardware cache event]
L1-dcache-stores [Hardware cache event]
L1-icache-load-misses [Hardware cache event]
LLC-load-misses [Hardware cache event]
LLC-loads [Hardware cache event]
LLC-store-misses [Hardware cache event]
LLC-stores [Hardware cache event]
branch-load-misses [Hardware cache event]
branch-loads [Hardware cache event]
cpu_atom/L1-icache-loads/ [Hardware cache event]
cpu_core/L1-dcache-load-misses/ [Hardware cache event]
cpu_core/node-load-misses/ [Hardware cache event]
cpu_core/node-loads/ [Hardware cache event]
dTLB-load-misses [Hardware cache event]
dTLB-loads [Hardware cache event]
dTLB-store-misses [Hardware cache event]
dTLB-stores [Hardware cache event]
iTLB-load-misses [Hardware cache event]
Now we can clearly see 'L1-dcache-load-misses' is only available
on cpu_core.
If without pmu prefix, it indicates the event is available on both
cpu_core and cpu_atom.
Signed-off-by: Jin Yao <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Jin Yao <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Kan Liang <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Link: http://lore.kernel.org/lkml/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Minor cleanup motivated by trying to separately fuzz test parse-events.
Signed-off-by: Ian Rogers <[email protected]>
Acked-by: Namhyung Kim <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Jin Yao <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Leo Yan <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Stephane Eranian <[email protected]>
Link: http://lore.kernel.org/lkml/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
As a full union is always sent, ensure all bytes of the union are
initialized with memset to avoid msan warnings of use of uninitialized
memory.
An example warning from the daemon test:
Uninitialized bytes in __interceptor_write at offset 71 inside [0x7ffd98da6280, 72)
==11602==WARNING: MemorySanitizer: use-of-uninitialized-value
#0 0x5597edccdbe4 in ion tools/lib/perf/lib.c:18:6
#1 0x5597edccdbe4 in writen tools/lib/perf/lib.c:47:9
#2 0x5597ed221d30 in send_cmd tools/perf/builtin-daemon.c:1376:22
#3 0x5597ed21b48c in cmd_daemon tools/perf/builtin-daemon.c
#4 0x5597ed1d6b67 in run_builtin tools/perf/perf.c:313:11
#5 0x5597ed1d6036 in handle_internal_command tools/perf/perf.c:365:8
#6 0x5597ed1d6036 in run_argv tools/perf/perf.c:409:2
#7 0x5597ed1d6036 in main tools/perf/perf.c:539:3
SUMMARY: MemorySanitizer: use-of-uninitialized-value tools/lib/perf/lib.c:18:6 in ion
Exiting
Signed-off-by: Ian Rogers <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Link: http://lore.kernel.org/lkml/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
To pick up fixes in the last pushed perf/urgent.
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Perf code re-implements libbpf's btf__load_from_kernel_by_id() API as
a weak function, presumably to dynamically link against old version of
libbpf shared library. Unfortunately this causes compilation warning
when perf is compiled against libbpf v0.6+.
For now, just ignore deprecation warning, but there might be a better
solution, depending on perf's needs.
Signed-off-by: Andrii Nakryiko <[email protected]>
Cc: Alexei Starovoitov <[email protected]>
Cc: Daniel Borkmann <[email protected]>
Cc: [email protected]
LPU-Reference: [email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
It's later supposed to be either a correct address or NULL. Without the
initialization, it may contain an undefined value which results in the
following segmentation fault:
# perf top --sort comm -g --ignore-callees=do_idle
terminates with:
#0 0x00007ffff56b7685 in __strlen_avx2 () from /lib64/libc.so.6
#1 0x00007ffff55e3802 in strdup () from /lib64/libc.so.6
#2 0x00005555558cb139 in hist_entry__init (callchain_size=<optimized out>, sample_self=true, template=0x7fffde7fb110, he=0x7fffd801c250) at util/hist.c:489
#3 hist_entry__new (template=template@entry=0x7fffde7fb110, sample_self=sample_self@entry=true) at util/hist.c:564
#4 0x00005555558cb4ba in hists__findnew_entry (hists=hists@entry=0x5555561d9e38, entry=entry@entry=0x7fffde7fb110, al=al@entry=0x7fffde7fb420,
sample_self=sample_self@entry=true) at util/hist.c:657
#5 0x00005555558cba1b in __hists__add_entry (hists=hists@entry=0x5555561d9e38, al=0x7fffde7fb420, sym_parent=<optimized out>, bi=bi@entry=0x0, mi=mi@entry=0x0,
sample=sample@entry=0x7fffde7fb4b0, sample_self=true, ops=0x0, block_info=0x0) at util/hist.c:288
#6 0x00005555558cbb70 in hists__add_entry (sample_self=true, sample=0x7fffde7fb4b0, mi=0x0, bi=0x0, sym_parent=<optimized out>, al=<optimized out>, hists=0x5555561d9e38)
at util/hist.c:1056
#7 iter_add_single_cumulative_entry (iter=0x7fffde7fb460, al=<optimized out>) at util/hist.c:1056
#8 0x00005555558cc8a4 in hist_entry_iter__add (iter=iter@entry=0x7fffde7fb460, al=al@entry=0x7fffde7fb420, max_stack_depth=<optimized out>, arg=arg@entry=0x7fffffff7db0)
at util/hist.c:1231
#9 0x00005555557cdc9a in perf_event__process_sample (machine=<optimized out>, sample=0x7fffde7fb4b0, evsel=<optimized out>, event=<optimized out>, tool=0x7fffffff7db0)
at builtin-top.c:842
#10 deliver_event (qe=<optimized out>, qevent=<optimized out>) at builtin-top.c:1202
#11 0x00005555558a9318 in do_flush (show_progress=false, oe=0x7fffffff80e0) at util/ordered-events.c:244
#12 __ordered_events__flush (oe=oe@entry=0x7fffffff80e0, how=how@entry=OE_FLUSH__TOP, timestamp=timestamp@entry=0) at util/ordered-events.c:323
#13 0x00005555558a9789 in __ordered_events__flush (timestamp=<optimized out>, how=<optimized out>, oe=<optimized out>) at util/ordered-events.c:339
#14 ordered_events__flush (how=OE_FLUSH__TOP, oe=0x7fffffff80e0) at util/ordered-events.c:341
#15 ordered_events__flush (oe=oe@entry=0x7fffffff80e0, how=how@entry=OE_FLUSH__TOP) at util/ordered-events.c:339
#16 0x00005555557cd631 in process_thread (arg=0x7fffffff7db0) at builtin-top.c:1114
#17 0x00007ffff7bb817a in start_thread () from /lib64/libpthread.so.0
#18 0x00007ffff5656dc3 in clone () from /lib64/libc.so.6
If you look at the frame #2, the code is:
488 if (he->srcline) {
489 he->srcline = strdup(he->srcline);
490 if (he->srcline == NULL)
491 goto err_rawdata;
492 }
If he->srcline is not NULL (it is not NULL if it is uninitialized rubbish),
it gets strdupped and strdupping a rubbish random string causes the problem.
Also, if you look at the commit 1fb7d06a509e, it adds the srcline property
into the struct, but not initializing it everywhere needed.
Committer notes:
Now I see, when using --ignore-callees=do_idle we end up here at line
2189 in add_callchain_ip():
2181 if (al.sym != NULL) {
2182 if (perf_hpp_list.parent && !*parent &&
2183 symbol__match_regex(al.sym, &parent_regex))
2184 *parent = al.sym;
2185 else if (have_ignore_callees && root_al &&
2186 symbol__match_regex(al.sym, &ignore_callees_regex)) {
2187 /* Treat this symbol as the root,
2188 forgetting its callees. */
2189 *root_al = al;
2190 callchain_cursor_reset(cursor);
2191 }
2192 }
And the al that doesn't have the ->srcline field initialized will be
copied to the root_al, so then, back to:
1211 int hist_entry_iter__add(struct hist_entry_iter *iter, struct addr_location *al,
1212 int max_stack_depth, void *arg)
1213 {
1214 int err, err2;
1215 struct map *alm = NULL;
1216
1217 if (al)
1218 alm = map__get(al->map);
1219
1220 err = sample__resolve_callchain(iter->sample, &callchain_cursor, &iter->parent,
1221 iter->evsel, al, max_stack_depth);
1222 if (err) {
1223 map__put(alm);
1224 return err;
1225 }
1226
1227 err = iter->ops->prepare_entry(iter, al);
1228 if (err)
1229 goto out;
1230
1231 err = iter->ops->add_single_entry(iter, al);
1232 if (err)
1233 goto out;
1234
That al at line 1221 is what hist_entry_iter__add() (called from
sample__resolve_callchain()) saw as 'root_al', and then:
iter->ops->add_single_entry(iter, al);
will go on with al->srcline with a bogus value, I'll add the above
sequence to the cset and apply, thanks!
Signed-off-by: Michael Petlan <[email protected]>
CC: Milian Wolff <[email protected]>
Cc: Jiri Olsa <[email protected]>
Fixes: 1fb7d06a509e ("perf report Use srcline from callchain for hist entries")
Link: https //lore.kernel.org/r/[email protected]
Reported-by: Juri Lelli <[email protected]>
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
set_print_ip_opts() was not being called when type != attr->type
because there is not a one-to-one relationship between output types
and attr->type. That resulted in ip not printing.
The attr_type() function is removed, and the match of attr->type to
output type is corrected.
Example on ADL using taskset to select an atom cpu:
# perf record -e cpu_atom/cpu-cycles/ taskset 0x1000 uname
Linux
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.003 MB perf.data (7 samples) ]
Before:
# perf script | head
taskset 428 [-01] 10394.179041: 1 cpu_atom/cpu-cycles/:
taskset 428 [-01] 10394.179043: 1 cpu_atom/cpu-cycles/:
taskset 428 [-01] 10394.179044: 11 cpu_atom/cpu-cycles/:
taskset 428 [-01] 10394.179045: 407 cpu_atom/cpu-cycles/:
taskset 428 [-01] 10394.179046: 16789 cpu_atom/cpu-cycles/:
taskset 428 [-01] 10394.179052: 676300 cpu_atom/cpu-cycles/:
uname 428 [-01] 10394.179278: 4079859 cpu_atom/cpu-cycles/:
After:
# perf script | head
taskset 428 10394.179041: 1 cpu_atom/cpu-cycles/: ffffffff95a0bb97 __intel_pmu_enable_all.constprop.48+0x47 ([kernel.kallsyms])
taskset 428 10394.179043: 1 cpu_atom/cpu-cycles/: ffffffff95a0bb97 __intel_pmu_enable_all.constprop.48+0x47 ([kernel.kallsyms])
taskset 428 10394.179044: 11 cpu_atom/cpu-cycles/: ffffffff95a0bb97 __intel_pmu_enable_all.constprop.48+0x47 ([kernel.kallsyms])
taskset 428 10394.179045: 407 cpu_atom/cpu-cycles/: ffffffff95a0bb97 __intel_pmu_enable_all.constprop.48+0x47 ([kernel.kallsyms])
taskset 428 10394.179046: 16789 cpu_atom/cpu-cycles/: ffffffff95a0bb97 __intel_pmu_enable_all.constprop.48+0x47 ([kernel.kallsyms])
taskset 428 10394.179052: 676300 cpu_atom/cpu-cycles/: 7f829ef73800 cfree+0x0 (/lib/libc-2.32.so)
uname 428 10394.179278: 4079859 cpu_atom/cpu-cycles/: ffffffff95bae912 vma_interval_tree_remove+0x1f2 ([kernel.kallsyms])
Signed-off-by: Adrian Hunter <[email protected]>
Reviewed-by: Kan Liang <[email protected]>
Cc: Jin Yao <[email protected]>
Cc: Jiri Olsa <[email protected]>
Link: http://lore.kernel.org/lkml/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Some x86 microarchitectures fuse a subset of cmp/test/ALU instructions
with branch instructions, and thus perf annotate highlight such valid
pairs as fused.
When annotated with source, perf uses struct disasm_line to contain
either source or instruction line from objdump output. Usually, a C
statement generates multiple instructions which include such
cmp/test/ALU + branch instruction pairs. But in case of assembly
function, each individual assembly source line generate one
instruction.
The 'perf annotate' instruction fusion logic assumes the previous
disasm_line as the previous instruction line, which is wrong because,
for assembly function, previous disasm_line contains source line. And
thus perf fails to highlight valid fused instruction pairs for assembly
functions.
Fix it by searching backward until we find an instruction line and
consider that disasm_line as fused with current branch instruction.
Before:
│ cmpq %rcx, RIP+8(%rsp)
0.00 │ cmp %rcx,0x88(%rsp)
│ je .Lerror_bad_iret <--- Source line
0.14 │ ┌──je b4 <--- Instruction line
│ │movl %ecx, %eax
After:
│ cmpq %rcx, RIP+8(%rsp)
0.00 │ ┌──cmp %rcx,0x88(%rsp)
│ │je .Lerror_bad_iret
0.14 │ ├──je b4
│ │movl %ecx, %eax
Reviewed-by: Jin Yao <[email protected]>
Signed-off-by: Ravi Bangoria <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Kim Phillips <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Namhyung Kim <[email protected]>
Link: https //lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Add an option to control the synthesizing behavior.
--synth <no|all|task|mmap|cgroup>
Fine-tune event synthesis: default=all
This can be useful when we know it doesn't need some synthesis like
in a specific usecase and/or when using pipe:
$ perf record -a --all-cgroups --synth cgroup -o- sleep 1 | \
> perf report -i- -s cgroup
Committer notes:
Added a clarification to the man page entry for --synth that this is
about pre-existing threads.
Signed-off-by: Namhyung Kim <[email protected]>
Acked-by: Jiri Olsa <[email protected]>
Cc: Adrian Hunter <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Ian Rogers <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jin Yao <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Stephane Eranian <[email protected]>
Link: https //lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
during record
Depending on the use case, it might require some kind of synthesizing
and some not. Make it controllable to turn off heavy operations like
MMAP for all tasks.
Currently all users are converted to enable all the synthesis by
default. It'll be updated in the later patch.
Signed-off-by: Namhyung Kim <[email protected]>
Acked-by: Jiri Olsa <[email protected]>
Cc: Adrian Hunter <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Ian Rogers <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jin Yao <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Stephane Eranian <[email protected]>
Link: https //lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Enum forward declarations aren't allowed as the size can't be implied.
Switch to just using an int. This fixes a clang warning:
In file included from tools/perf/bench/evlist-open-close.c:13:
tools/perf/bench/../util/parse-events.h:185:6: error: redeclaration of already-defined enum 'perf_tool_event' is a GNU extension [-Werror,-Wgnu-redeclared-enum]
enum perf_tool_event;
^
tools/perf/bench/../util/evsel.h:28:6: note: previous definition is here
enum perf_tool_event {
^
Signed-off-by: Ian Rogers <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Nathan Chancellor <[email protected]>
Cc: Nick Desaulniers <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Link: http://lore.kernel.org/lkml/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Perf code re-implements libbpf's btf__load_from_kernel_by_id() API as
a weak function, presumably to dynamically link against old version of
libbpf shared library. Unfortunately this causes compilation warning
when perf is compiled against libbpf v0.6+.
For now, just ignore deprecation warning, but there might be a better
solution, depending on perf's needs.
Signed-off-by: Andrii Nakryiko <[email protected]>
Cc: Alexei Starovoitov <[email protected]>
Cc: Daniel Borkmann <[email protected]>
Cc: [email protected]
LPU-Reference: [email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
As a part of libbpf 1.0 plan[0], this patch deprecates use of
bpf_map__resize in favour of bpf_map__set_max_entries.
Reference: https://github.com/libbpf/libbpf/issues/304
[0]: https://github.com/libbpf/libbpf/wiki/Libbpf:-the-road-to-v1.0#libbpfh-high-level-apis
Signed-off-by: Muhammad Falak R Wani <[email protected]>
Acked-by: Andrii Nakryiko <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Alexei Starovoitov <[email protected]>
Cc: Daniel Borkmann <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: John Fastabend <[email protected]>
Cc: KP Singh <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Martin KaFai Lau <[email protected]>
Cc: Muhammad Falak R Wani <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Quentin Monnet <[email protected]>
Cc: Song Liu <[email protected]>
Cc: Yu Kuai <[email protected]>
Link: http //lore.kernel.org/lkml/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
AMD family 15h and above microarchs fuse a subset of cmp/test/ALU
instructions with branch instructions[1][2]. Add perf annotate
fused instruction support for these microarchs.
Before:
│ testb $0x80,0x51(%rax)
│ ┌──jne 5b3
0.78 │ │ mov %r13,%rdi
│ │→ callq mark_page_accessed
1.08 │5b3:└─→mov 0x8(%r13),%rax
After:
│ ┌──testb $0x80,0x51(%rax)
│ ├──jne 5b3
0.78 │ │ mov %r13,%rdi
│ │→ callq mark_page_accessed
1.08 │5b3:└─→mov 0x8(%r13),%rax
[1] https://bugzilla.kernel.org/attachment.cgi?id=298553
[2] https://bugzilla.kernel.org/attachment.cgi?id=298555
Committer testing:
On a:
$ grep -m1 "model name" /proc/cpuinfo
model name : AMD Ryzen 9 3900X 12-Core Processor
$
Samples: 44K of event 'cycles', 4000 Hz, Event count (approx.): 7533249650
_int_malloc /usr/lib64/libc-2.33.so [Percent: local period]
Percent│ ┌──test %eax,%eax
│ ├──jne 884
│ │↓ jmpq 943
│ │ nop
│878:│ add $0x10,%rdx
0.64 │ │ add %eax,%eax
0.57 │ │↓ je cc9
0.77 │884:└─→test %esi,%eax
│ ↑ je 878
│ mov 0x18(%rdx),%r15
Reported-by: Kim Phillips <[email protected]>
Signed-off-by: Ravi Bangoria <[email protected]>
Tested-by: Arnaldo Carvalho de Melo <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Jin Yao <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Namhyung Kim <[email protected]>
Link: https //lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Some distributions, like debian, don't link perf with libbfd. Add a
build flag to make this configuration buildable and testable.
This was inspired by:
https://lore.kernel.org/linux-perf-users/[email protected]/T/#u
Signed-off-by: Ian Rogers <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: tony garnock-jones <[email protected]>
Link: http://lore.kernel.org/lkml/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Currently perf saves a build-id with size but old versions assumes the
size of 20. In case the build-id is less than 20 (like for MD5), it'd
fill the rest with 0s.
I saw a problem when old version of perf record saved a binary in the
build-id cache and new version of perf reads the data. The symbols
should be read from the build-id cache (as the path no longer has the
same binary) but it failed due to mismatch in the build-id.
symsrc__init: build id mismatch for /home/namhyung/.debug/.build-id/53/e4c2f42a4c61a2d632d92a72afa08f00000000/elf.
The build-id event in the data has 20 byte build-ids, but it saw a
different size (16) when it reads the build-id of the elf file in the
build-id cache.
$ readelf -n ~/.debug/.build-id/53/e4c2f42a4c61a2d632d92a72afa08f00000000/elf
Displaying notes found in: .note.gnu.build-id
Owner Data size Description
GNU 0x00000010 NT_GNU_BUILD_ID (unique build ID bitstring)
Build ID: 53e4c2f42a4c61a2d632d92a72afa08f
Let's fix this by allowing trailing zeros if the size is different.
Fixes: 39be8d0115b321ed ("perf tools: Pass build_id object to dso__build_id_equal()")
Signed-off-by: Namhyung Kim <[email protected]>
Acked-by: Jiri Olsa <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Ian Rogers <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Link: http://lore.kernel.org/lkml/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
A config terms list was spliced twice, resulting in a never-ending loop
when the list was traversed. Fix by using list_splice_init() and copying
and freeing the lists as necessary.
This patch also depends on patch "perf tools: Factor out
copy_config_terms() and free_config_terms()"
Example on ADL:
Before:
# perf record -e '{intel_pt//,cycles/aux-sample-size=4096/pp}' uname &
# jobs
[1]+ Running perf record -e "{intel_pt//,cycles/aux-sample-size=4096/pp}" uname
# perf top -E 10
PerfTop: 4071 irqs/sec kernel: 6.9% exact: 100.0% lost: 0/0 drop: 0/0 [4000Hz cycles], (all, 24 CPUs)
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
97.60% perf [.] __evsel__get_config_term
0.25% [kernel] [k] kallsyms_expand_symbol.constprop.13
0.24% perf [.] kallsyms__parse
0.15% [kernel] [k] _raw_spin_lock
0.14% [kernel] [k] number
0.13% [kernel] [k] advance_transaction
0.08% [kernel] [k] format_decode
0.08% perf [.] map__process_kallsym_symbol
0.08% perf [.] rb_insert_color
0.08% [kernel] [k] vsnprintf
exiting.
# kill %1
After:
# perf record -e '{intel_pt//,cycles/aux-sample-size=4096/pp}' uname &
Linux
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.060 MB perf.data ]
# perf script | head
perf-exec 604 [001] 1827.312293: psb: psb offs: 0 ffffffffb8415e87 pt_config_start+0x37 ([kernel.kallsyms])
perf-exec 604 1827.312293: 1 branches: ffffffffb856a3bd event_sched_in.isra.133+0xfd ([kernel.kallsyms]) => ffffffffb856a9a0 perf_pmu_nop_void+0x0 ([kernel.kallsyms])
perf-exec 604 1827.312293: 1 branches: ffffffffb856b10e merge_sched_in+0x26e ([kernel.kallsyms]) => ffffffffb856a2c0 event_sched_in.isra.133+0x0 ([kernel.kallsyms])
perf-exec 604 1827.312293: 1 branches: ffffffffb856a45d event_sched_in.isra.133+0x19d ([kernel.kallsyms]) => ffffffffb8568b80 perf_event_set_state.part.61+0x0 ([kernel.kallsyms])
perf-exec 604 1827.312293: 1 branches: ffffffffb8568b86 perf_event_set_state.part.61+0x6 ([kernel.kallsyms]) => ffffffffb85662a0 perf_event_update_time+0x0 ([kernel.kallsyms])
perf-exec 604 1827.312293: 1 branches: ffffffffb856a35c event_sched_in.isra.133+0x9c ([kernel.kallsyms]) => ffffffffb8567610 perf_log_itrace_start+0x0 ([kernel.kallsyms])
perf-exec 604 1827.312293: 1 branches: ffffffffb856a377 event_sched_in.isra.133+0xb7 ([kernel.kallsyms]) => ffffffffb8403b40 x86_pmu_add+0x0 ([kernel.kallsyms])
perf-exec 604 1827.312293: 1 branches: ffffffffb8403b86 x86_pmu_add+0x46 ([kernel.kallsyms]) => ffffffffb8403940 collect_events+0x0 ([kernel.kallsyms])
perf-exec 604 1827.312293: 1 branches: ffffffffb8403a7b collect_events+0x13b ([kernel.kallsyms]) => ffffffffb8402cd0 collect_event+0x0 ([kernel.kallsyms])
Fixes: 30def61f64bac5 ("perf parse-events Create two hybrid cache events")
Fixes: 94da591b1c7913 ("perf parse-events Create two hybrid raw events")
Fixes: 9cbfa2f64c04d9 ("perf parse-events Create two hybrid hardware events")
Signed-off-by: Adrian Hunter <[email protected]>
Acked-by: Jiri Olsa <[email protected]>
Cc: Jin Yao <[email protected]>
Cc: Kan Liang <[email protected]>
Link: https //lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|