Enhance the thread stack to output detailed information about paired
calls and returns.
The enhanced processing consumes sample information via
thread_stack__process() and outputs information about paired calls /
returns via a call-back.
While the call-back makes it possible for the facility to be used by
arbitrary tools, a subsequent patch will provide the information to
Python scripting via the db-export interface.
An important part of the call/return information is the
call path which provides a structure that defines a context
sensitive call graph.
Note that there are now two ways to use the thread stack.
For simply providing a call stack (like you would get from the perf
record -g option) the interface consists of thread_stack__event() and
thread_stack__sample().
Whereas the enhanced interface consists of call_return_processor__new()
and thread_stack__process().
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/1414678188-14946-5-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Pull perf/core improvements from Arnaldo Carvalho de Melo:
User visible changes:
- Don't open the DWARF info multiple times, keeping instead a dwfl handle
in struct dso, greatly speeding up 'perf report' on powerpc. (Sukadev Bhattiprolu)
- Introduce PARSE_OPT_DISABLED option flag and use it to avoid showing
undersired options in tools that provides frontends to 'perf record', like
sched, kvm, etc (Namhyung Kim)
Infrastructure changes:
- More Intel PT work, including a facility to export sample data (comms,
threads, symbol names, etc) in a database friendly way, with an script to use
this to create a postgresql database. (Adrian Hunter)
- Use make sure that thread->mg->machine points to the machine where
the thread exists (it was being set only for the kmaps kernel modules
case, do it as well for the mmaps) and use it to shorten function
signatures (Arnaldo Carvalho de Melo)
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Pull perf/urgent fixes from Arnaldo Carvalho de Melo:
- Fix report -F (abort, in_tx, mispredict, etc) segfaults for sample.data files
without branch info (Jiri Olsa)
- Add patch that should have went in a previous patchkit to use global cache
provided by libunwind (Namhyung Kim)
- Make CPUINFO_PROC an array to support different kernels, problem
detected when the information reported via /proc/cpuinfo changed on ARM (Wang Nan)
- 'perf probe' --demangle typo fix and a new --quiet option (Masami Hiramatsu)
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Add an index of the event identifiers, in preparation for Intel PT.
The event id (also called the sample id) is a unique number
allocated by the kernel to the event created by perf_event_open(). Events
can include the event id by having a sample type including PERF_SAMPLE_ID or
PERF_SAMPLE_IDENTIFIER.
Currently the main use of the event id is to match an event back to the
evsel to which it belongs i.e. perf_evlist__id2evsel()
The purpose of this patch is to make it possible to match an event back to
the mmap from which it was read. The reason that is useful is because the
mmap represents a time-ordered context (either for a cpu or for a thread).
Intel PT decodes trace information on that basis. In full-trace mode, that
information can be recorded when the Intel PT trace is read, but in
sample-mode the Intel PT trace data is embedded in a sample and it is in
that case that the "id index" is needed.
So the mmaps are numbered (idx) and the cpu and tid recorded against the id
by perf_evlist__set_sid_idx() which is called by perf_evlist__mmap_per_evsel().
That information is recorded on the perf.data file in the new "id index".
idx, cpu and tid are added to struct perf_sample_id (which is the node of
evlist's hash table to match ids to evsels). The information can be
retrieved using perf_evlist__id2sid(). Note however this all depends on
having a sample type including PERF_SAMPLE_ID or PERF_SAMPLE_IDENTIFIER,
otherwise ids are not recorded.
The "id index" is a synthesized event record which will be created when
Intel PT sampling is used by calling perf_event__synthesize_id_index().
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/1414417770-18602-2-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Add a Python script to export to a postgresql database.
The script requires the Python pyside module and the Qt PostgreSQL
driver. The packages needed are probably named "python-pyside" and
"libqt4-sql-psql"
The caller of the script must be able to create postgresql databases.
The script takes the database name as a parameter. The database and
database tables are created. Data is written to flat files which are
then imported using SQL COPY FROM.
Example:
$ perf record ls
...
$ perf script report export-to-postgresql lsdb
2014-02-14 10:55:38.631431 Creating database...
2014-02-14 10:55:39.291958 Writing to intermediate files...
2014-02-14 10:55:39.350280 Copying to database...
2014-02-14 10:55:39.358536 Removing intermediate files...
2014-02-14 10:55:39.358665 Adding primary keys
2014-02-14 10:55:39.658697 Adding foreign keys
2014-02-14 10:55:39.667412 Done
$ psql lsdb
lsdb-# \d
List of relations
Schema | Name | Type | Owner
--------+-----------------+-------+-------
public | comm_threads | table | acme
public | comms | table | acme
public | dsos | table | acme
public | machines | table | acme
public | samples | table | acme
public | samples_view | view | acme
public | selected_events | table | acme
public | symbols | table | acme
public | threads | table | acme
(9 rows)
lsdb-# \d samples
Table "public.samples"
Column | Type | Modifiers
---------------+---------+-----------
id | bigint | not null
evsel_id | bigint |
machine_id | bigint |
thread_id | bigint |
comm_id | bigint |
dso_id | bigint |
symbol_id | bigint |
sym_offset | bigint |
ip | bigint |
time | bigint |
cpu | integer |
to_dso_id | bigint |
to_symbol_id | bigint |
to_sym_offset | bigint |
to_ip | bigint |
period | bigint |
weight | bigint |
transaction | bigint |
data_src | bigint |
Indexes:
"samples_pkey" PRIMARY KEY, btree (id)
Foreign-key constraints:
"commfk" FOREIGN KEY (comm_id) REFERENCES comms(id)
"dsofk" FOREIGN KEY (dso_id) REFERENCES dsos(id)
"evselfk" FOREIGN KEY (evsel_id) REFERENCES selected_events(id)
"machinefk" FOREIGN KEY (machine_id) REFERENCES machines(id)
"symbolfk" FOREIGN KEY (symbol_id) REFERENCES symbols(id)
"threadfk" FOREIGN KEY (thread_id) REFERENCES threads(id)
"todsofk" FOREIGN KEY (to_dso_id) REFERENCES dsos(id)
"tosymbolfk" FOREIGN KEY (to_symbol_id) REFERENCES symbols(id)
lsdb-# \d samples_view
View "public.samples_view"
Column | Type | Modifiers
-------------------+-------------------------+-----------
id | bigint |
time | bigint |
cpu | integer |
pid | integer |
tid | integer |
command | character varying(16) |
event | character varying(80) |
ip_hex | text |
symbol | character varying(2048) |
sym_offset | bigint |
dso_short_name | character varying(256) |
to_ip_hex | text |
to_symbol | character varying(2048) |
to_sym_offset | bigint |
to_dso_short_name | character varying(256) |
lsdb=# select * from samples_view;
id| time |cpu | pid | tid |command| event | ip_hex | symbol |sym_off| dso_name|to_ip_hex|to_symbol|to_sym_off|to_dso_name
--+------------+----+------+------+-------+--------+---------------+---------------------+-------+---------+---------+---------+----------+----------
1 |12202825015 | -1 | 7339 | 7339 |:17339 | cycles | fffff8104d24a |native_write_msr_safe| 10 | [kernel]| 0 | unknown | 0| unknown
2 |12203258804 | -1 | 7339 | 7339 |:17339 | cycles | fffff8104d24a |native_write_msr_safe| 10 | [kernel]| 0 | unknown | 0| unknown
3 |12203988119 | -1 | 7339 | 7339 |:17339 | cycles | fffff8104d24a |native_write_msr_safe| 10 | [kernel]| 0 | unknown | 0| unknown
My notes (which may be out-of-date) on setting up postgresql so you can
create databases:
fedora:
$ sudo yum install postgresql postgresql-server python-pyside qt-postgresql
$ sudo su - postgres -c initdb
$ sudo service postgresql start
$ sudo su - postgres
$ createuser -s <your username>
I used the the unix user name in createuser.
If it fails, try createuser without -s and answer the following question
to allow your user to create tables:
Shall the new role be a superuser? (y/n) y
ubuntu:
$ sudo apt-get install postgresql
$ sudo su - postgres
$ createuser <your username>
Shall the new role be a superuser? (y/n) y
You may want to disable automatic startup. One way is to edit
/etc/postgresql/9.3/main/start.conf. Another is to disable the init
script e.g. sudo update-rc.d postgresql disable
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/1414061124-26830-8-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
This patch introduces an abstraction for exporting sample data in a
database-friendly way. The abstraction does not implement the actual
output. A subsequent patch takes this facility into use for extending
the script interface.
The abstraction is needed because static data like symbols, dsos, comms
etc need to be exported only once. That means allocating them a unique
identifier and recording it on each structure. The member 'db_id' is
used for that. 'db_id' is just a 64-bit sequence number.
Exporting centres around the db_export__sample() function which exports
the associated data structures if they have not yet been allocated a
db_id.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/1414061124-26830-6-git-send-email-adrian.hunter@intel.com
[ committer note: Stash db_id using symbol_conf.priv_size + symbol__priv() and foo->priv areas ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
'perf record' post-processes the event stream to create a list of
build-ids for object files for which sample events have been recorded.
That results in those object files being recorded in the build-id cache.
In the case of VDSO, perf tools reads it from memory and copies it into
a temporary file, which as decribed above, gets added to the build-id
cache.
Then when the perf.data file is processed by other tools, the build-id
of VDSO is listed in the perf.data file and the VDSO can be read from
the build-id cache. In that case the name of the map, the short name of
the DSO, and the entry in the build-id cache are all "[vdso]".
However, in the 64-bit case, there also can be 32-bit compatibility
VDSOs.
A previous patch added programs "perf-read-vdso32" and "perf
read-vdsox32".
This patch uses those programs to read the correct VDSO for a thread and
create a temporary file just as for the 64-bit VDSO.
The map name and the entry in the build-id cache are still "[vdso]" but
the DSO short name becomes "[vdso32]" and "[vdsox32]" respectively.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/1414061124-26830-16-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
When 'perf record' write headers, it calls write_xxx in
tools/perf/util/header.c, and check return value. It rolls back all
working only when return value is negative.
This patch ensures write_cpudesc() and write_total_mem() return negative number
when error. Without this patch, headers reported by 'perf report' header is
error in some platform. Following output is caputured on ARM, which doesn't
contain "Processor" field in /proc/cpuinfo. See "cpudesc", "total memory" and
"cmdline" field.
bash-4.2# perf record ls
...
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.001 MB perf.data (~36 samples) ]
bash-4.2# perf report --stdio --header
Error:
The perf.data file has no samples!
# ========
# captured on: Fri Sep 12 10:09:10 2014
# hostname : arma15el
# os release : 3.17.0+
# perf version : 3.10.53
# arch : armv7l
# nrcpus online : 4
# nrcpus avail : 1
# cpudesc : (null)
# total memory : 0 kB
# cmdline :
# event : name = cycles, type = 0, config = 0x0, config1 = 0x0, config2 = 0x0, excl_usr = 0, excl_kern = 0, excl_host = 0, excl_guest = 1, precise_ip = 0
# pmu mappings: not available
# ========
#
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Li Zefan <lizefan@huawei.com>
Cc: Namhyung Kim <namhyung.kim@lge.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Waiman Long <Waiman.Long@hp.com>
Link: http://lkml.kernel.org/r/1413428909-80017-1-git-send-email-wangnan0@huawei.com
Signed-off-by: Wang Nan <wangnan0@huawei.com>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The perf probe command has some exclusive options. Use new PARSE_OPT_EXCLUSIVE
flag to simplify the code and show more compact usage.
$ perf probe -l -a foo
Error: switch `a' cannot be used with switch `l'
usage: perf probe [<options>] 'PROBEDEF' ['PROBEDEF' ...]
or: perf probe [<options>] --add 'PROBEDEF' [--add 'PROBEDEF' ...]
or: perf probe [<options>] --del '[GROUP:]EVENT' ...
or: perf probe --list
or: perf probe [<options>] --line 'LINEDESC'
or: perf probe [<options>] --vars 'PROBEPOINT'
-a, --add <[EVENT=]FUNC[@SRC][+OFF|%return|:RL|;PT]|SRC:AL|SRC;PT [[NAME=]ARG ...]>
probe point definition, where
GROUP: Group name (optional)
EVENT: Event name
FUNC: Function name
OFF: Offset from function entry (in byte)
%return: Put the probe at function return
SRC: Source code path
RL: Relative line number from function entry.
AL: Absolute line number in file.
PT: Lazy expression of line code.
ARG: Probe argument (local variable name or
kprobe-tracer argument format.)
-l, --list list up current probe events
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Acked-by: Hemant Kumar <hemant@linux.vnet.ibm.com>
Acked-by: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Hemant Kumar <hemant@linux.vnet.ibm.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1413990949-13953-6-git-send-email-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The 'perf kvm stat record' tool is an alias of 'perf record' with
predefined kvm related options. All options that passed to 'perf kvm
stat record' are processed by the 'perf record' tool. So, 'perf kvm
stat record --help' prints help of usage for the 'perf record'
command. There are a few options useful for 'perf kvm stat record',
the rest either break kvm related output or don't change it.
Let's print safe for 'perf kvm stat record' options in addition to
general 'perf record' --help output.
With this patch, new output looks like below:
$ perf kvm stat record -h
usage: perf kvm stat record [<options>]
-p, --pid <pid> record events on existing process id
-t, --tid <tid> record events on existing thread id
-r, --realtime <n> collect data with this RT SCHED_FIFO priority
--no-buffering collect data without buffering
-a, --all-cpus system-wide collection from all CPUs
-C, --cpu <cpu> list of cpus to monitor
-c, --count <n> event period to sample
-o, --output <file> output file name
-i, --no-inherit child tasks do not inherit counters
-m, --mmap-pages <pages>
number of mmap data pages
-v, --verbose be more verbose (show counter open errors, etc)
-q, --quiet don't print any message
-s, --stat per thread counts
-D, --delay <n> ms to wait before starting measurement after program start
-u, --uid <user> user to profile
--per-thread use per-thread mmaps
$ perf kvm stat record -n sleep 1
Error: switch `n' is not usable
usage: perf kvm stat record [<options>]
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Acked-by: Hemant Kumar <hemant@linux.vnet.ibm.com>
Cc: Alexander Yarygin <yarygin@linux.vnet.ibm.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1413990949-13953-4-git-send-email-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cache the DWARF debug info for DSO so we don't have to rebuild it for each
address in the DSO.
Note that dso__new() uses calloc() so don't need to set dso->dwfl to NULL.
$ /tmp/perf.orig --version
perf version 3.18.rc1.gc2661b8
$ /tmp/perf.new --version
perf version 3.18.rc1.g402d62
$ perf stat -e cycles,instructions /tmp/perf.orig report -g > orig
Performance counter stats for '/tmp/perf.orig report -g':
6,428,177,183 cycles # 0.000 GHz
4,176,288,391 instructions # 0.65 insns per cycle
1.840666132 seconds time elapsed
$ perf stat -e cycles,instructions /tmp/perf.new report -g > new
Performance counter stats for '/tmp/perf.new report -g':
305,773,142 cycles # 0.000 GHz
276,048,272 instructions # 0.90 insns per cycle
0.087693543 seconds time elapsed
$ diff orig new
$
Changelog[v2]:
[Arnaldo Carvalho] Cache in existing global objects rather than create
new static/globals in functions.
Reported-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Sukadev Bhattiprolu <sukadev@linux.vnet.ibm.com>
Cc: Anton Blanchard <anton@au1.ibm.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: http://lkml.kernel.org/r/20141022000958.GB2228@us.ibm.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
After kernel 3.7 (commit b4b8f770eb),
/proc/cpuinfo replaces 'Processor' to 'model name'.
This patch makes CPUINFO_PROC to an array and provides two choices for
ARM, makes it compatible for different kernel version.
v1 -> v2: minor changes as suggested by Namhyung Kim:
- Doesn't pass @h and @evlist to __write_cpudesc;
- Coding style fix.
v2 -> v3:
- Rebase:
git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux.git perf/core
Signed-off-by: Wang Nan <wangnan0@huawei.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Li Zefan <lizefan@huawei.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: http://lkml.kernel.org/r/1414115126-7479-1-git-send-email-wangnan0@huawei.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
These patches:
86a349a28b ("perf/x86/intel: Add Broadwell core support")
c46e665f03 ("perf/x86: Add INST_RETIRED.ALL workarounds")
fdda3c4aac ("perf/x86/intel: Use Broadwell cache event list for Haswell")
introduced magic constants and unexplained changes:
https://lkml.org/lkml/2014/10/28/1128https://lkml.org/lkml/2014/10/27/325https://lkml.org/lkml/2014/8/27/546https://lkml.org/lkml/2014/10/28/546
Peter Zijlstra has attempted to help out, to clean up the mess:
https://lkml.org/lkml/2014/10/28/543
But has not received helpful and constructive replies which makes
me doubt wether it can all be finished in time until v3.18 is
released.
Despite various review feedback the author (Andi Kleen) has answered
only few of the review questions and has generally been uncooperative,
only giving replies when prompted repeatedly, and only giving minimal
answers instead of constructively explaining and helping along the effort.
That kind of behavior is not acceptable.
There's also a boot crash on Intel E5-1630 v3 CPUs reported for another
commit from Andi Kleen:
e735b9db12 ("perf/x86/intel/uncore: Add Haswell-EP uncore support")
https://lkml.org/lkml/2014/10/22/730
Which is not yet resolved. The uncore driver is independent in theory,
but the crash makes me worry about how well all these patches were
tested and makes me uneasy about the level of interminging that the
Broadwell and Haswell code has received by the commits above.
As a first step to resolve the mess revert the Broadwell client commits
back to the v3.17 version, before we run out of time and problematic
code hits a stable upstream kernel.
( If the Haswell-EP crash is not resolved via a simple fix then we'll have
to revert the Haswell-EP uncore driver as well. )
The Broadwell client series has to be submitted in a clean fashion, with
single, well documented changes per patch. If they are submitted in time
and are accepted during review then they can possibly go into v3.19 but
will need additional scrutiny due to the rocky history of this patch set.
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: eranian@google.com
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Link: http://lkml.kernel.org/r/1409683455-29168-3-git-send-email-andi@firstfloor.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Pull perf/urgent fix from Arnaldo Carvalho de Melo:
- Fix 'perf diff' initialization, calling the recently added hists__init()
function so that extra space is allocated per perf_evsel for the hists storage
that it also uses, just like report and top. (Kan Liang)
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
