perf documentation updates
cover perf documentation updates
While using perf I found that many of the commands had options that were not documented in their man pages. The following patches update all of the man pages to include the missing options each command supports. Additionally there are a few spelling fixes, and a fix for perf diff which had two options that used the same flag.
Shawn Bohrer (14):
perf: Document missing annotate options.
perf: Document missing buildid-list options.
perf: fix diff displacement and modules options short flag
perf: Document missing diff options
perf: Document missing kvm options
perf: Document missing lock options
perf: Fix spelling mistake in probe documentation
perf: Document missing record options
perf: Document missing report options
perf: Document missing sched options
perf: Document missing stat options
perf: Fix spelling mistake in test documentation
perf: Document missing top options
perf: Document missing trace options
tools/perf/Documentation/perf-annotate.txt | 37 +++++++++++++++++- tools/perf/Documentation/perf-buildid-list.txt | 3 + tools/perf/Documentation/perf-diff.txt | 19 +++++++++- tools/perf/Documentation/perf-kvm.txt | 8 +++- tools/perf/Documentation/perf-lock.txt | 15 +++++++ tools/perf/Documentation/perf-probe.txt | 2 +- tools/perf/Documentation/perf-record.txt | 17 ++++++-- tools/perf/Documentation/perf-report.txt | 49 ++++++++++++++++++++++-- tools/perf/Documentation/perf-sched.txt | 18 ++++++++- tools/perf/Documentation/perf-stat.txt | 34 +++++++++++++--- tools/perf/Documentation/perf-test.txt | 2 +- tools/perf/Documentation/perf-top.txt | 28 ++++++++++++-- tools/perf/Documentation/perf-trace.txt | 7 +++ tools/perf/builtin-diff.c | 2 +- 14 files changed, 214 insertions(+), 27 deletions(-)
Signature · 1 line
1.7.3.2
PATCH 1/14 perf: Document missing annotate options.
--- tools/perf/Documentation/perf-annotate.txt | 37 +++++++++++++++++++++++++++- 1 files changed, 36 insertions(+), 1 deletions(-)
tools/perf/Documentation/perf-annotate.txt +36 -1
@@ -24,12 +24,47 @@ OPTIONS --input=:: Input file name. (default: perf.data) +-d:: +--dsos=<dso[,dso...]>:: + Only consider symbols in these dsos. +-s:: +--symbol=<symbol>:: + Symbol to annotate. + +-f:: +--force:: + Don't complain, do it. + +-v:: +--verbose:: + Be more verbose. (Show symbol address, etc) + +-D:: +--dump-raw-trace:: + Dump raw trace in ASCII. + +-k:: +--vmlinux=<file>:: + vmlinux pathname. + +-m:: +--modules:: + Load module symbols. WARNING: use only with -k and LIVE kernel. + +-l:: +--print-line:: + Print matching source lines (may be slow). + +-P:: +--full-paths:: + Don't shorten the displayed pathnames. + --stdio:: Use the stdio interface. --tui:: Use the TUI interface Use of --tui requires a tty, if one is not present, as when piping to other commands, the stdio interface is used. This interfaces starts by centering on the line with more - samples, TAB/UNTAB cycles thru the lines with more samples. + samples, TAB/UNTAB cycles through the lines with more samples. SEE ALSO --------
Signature · 1 line
1.7.3.2
[tip:perf/core] perf annotate: Document missing options.
Commit-ID: 1968ad911d0769e58a484bb721d275814baf9f2f Gitweb: http://git.kernel.org/tip/1968ad911d0769e58a484bb721d275814baf9f2f Author: Shawn Bohrer <[email protected]> AuthorDate: Tue, 30 Nov 2010 19:57:09 -0600 Committer: Arnaldo Carvalho de Melo <[email protected]> CommitDate: Wed, 1 Dec 2010 18:22:45 -0200
perf annotate: Document missing options.
Cc: Ingo Molnar <[email protected]>
Cc: Paul Mackerras <[email protected]>
Cc: Peter Zijlstra <[email protected]>
LKML-Reference: <[email protected]>
--- tools/perf/Documentation/perf-annotate.txt | 37 +++++++++++++++++++++++++++- 1 files changed, 36 insertions(+), 1 deletions(-)
tools/perf/Documentation/perf-annotate.txt +36 -1
@@ -24,12 +24,47 @@ OPTIONS --input=:: Input file name. (default: perf.data) +-d:: +--dsos=<dso[,dso...]>:: + Only consider symbols in these dsos. +-s:: +--symbol=<symbol>:: + Symbol to annotate. + +-f:: +--force:: + Don't complain, do it. + +-v:: +--verbose:: + Be more verbose. (Show symbol address, etc) + +-D:: +--dump-raw-trace:: + Dump raw trace in ASCII. + +-k:: +--vmlinux=<file>:: + vmlinux pathname. + +-m:: +--modules:: + Load module symbols. WARNING: use only with -k and LIVE kernel. + +-l:: +--print-line:: + Print matching source lines (may be slow). + +-P:: +--full-paths:: + Don't shorten the displayed pathnames. + --stdio:: Use the stdio interface. --tui:: Use the TUI interface Use of --tui requires a tty, if one is not present, as when piping to other commands, the stdio interface is used. This interfaces starts by centering on the line with more - samples, TAB/UNTAB cycles thru the lines with more samples. + samples, TAB/UNTAB cycles through the lines with more samples. SEE ALSO --------
PATCH 2/14 perf: Document missing buildid-list options.
In reply to Shawn Bohrer--- tools/perf/Documentation/perf-buildid-list.txt | 3 +++ 1 files changed, 3 insertions(+), 0 deletions(-)
tools/perf/Documentation/perf-buildid-list.txt +3 -0
@@ -18,6 +18,9 @@ perf report. OPTIONS ------- +-H:: +--with-hits:: + Show only DSOs with hits. -i:: --input=:: Input file name. (default: perf.data)
Signature · 1 line
1.7.3.2
[tip:perf/core] perf buildid-list: Document missing options
Commit-ID: b6a535dbf615e168e796eec7318c77d1a3b7047f Gitweb: http://git.kernel.org/tip/b6a535dbf615e168e796eec7318c77d1a3b7047f Author: Shawn Bohrer <[email protected]> AuthorDate: Tue, 30 Nov 2010 19:57:10 -0600 Committer: Arnaldo Carvalho de Melo <[email protected]> CommitDate: Wed, 1 Dec 2010 18:22:45 -0200
perf buildid-list: Document missing options
Cc: Ingo Molnar <[email protected]>
Cc: Paul Mackerras <[email protected]>
Cc: Peter Zijlstra <[email protected]>
LKML-Reference: <[email protected]>
--- tools/perf/Documentation/perf-buildid-list.txt | 3 +++ 1 files changed, 3 insertions(+), 0 deletions(-)
tools/perf/Documentation/perf-buildid-list.txt +3 -0
@@ -18,6 +18,9 @@ perf report. OPTIONS ------- +-H:: +--with-hits:: + Show only DSOs with hits. -i:: --input=:: Input file name. (default: perf.data)
PATCH 3/14 perf: fix diff displacement and modules options short flag
In reply to Shawn BohrerThe --displacement and --modules options to perf diff both use -m as a short flag. Change --displacement to use -M since other perf commands use -m, --modules.
--- tools/perf/builtin-diff.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)
tools/perf/builtin-diff.c +1 -1
@@ -173,7 +173,7 @@ static const char * const diff_usage[] = {static const struct option options[] = {OPT_INCR ('v' ,"verbose" , &verbose,"be more verbose (show symbol address, etc)" ), -OPT_BOOLEAN ('m' ,"displacement" , &show_displacement, +OPT_BOOLEAN ('M' ,"displacement" , &show_displacement,"Show position displacement relative to baseline" ),OPT_BOOLEAN ('D' ,"dump-raw-trace" , &dump_trace,"dump raw trace in ASCII" ),
Signature · 1 line
1.7.3.2
[tip:perf/core] perf diff: Fix displacement and modules options short flag
Commit-ID: 342955593af08f185279e074b3d77719d2f23b82 Gitweb: http://git.kernel.org/tip/342955593af08f185279e074b3d77719d2f23b82 Author: Shawn Bohrer <[email protected]> AuthorDate: Tue, 30 Nov 2010 19:57:11 -0600 Committer: Arnaldo Carvalho de Melo <[email protected]> CommitDate: Wed, 1 Dec 2010 18:22:45 -0200
perf diff: Fix displacement and modules options short flag
The --displacement and --modules options to perf diff both use -m as a short flag. Change --displacement to use -M since other perf commands use -m, --modules.
Cc: Ingo Molnar <[email protected]>
Cc: Paul Mackerras <[email protected]>
Cc: Peter Zijlstra <[email protected]>
LKML-Reference: <[email protected]>
--- tools/perf/builtin-diff.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)
tools/perf/builtin-diff.c +1 -1
@@ -173,7 +173,7 @@ static const char * const diff_usage[] = {static const struct option options[] = {OPT_INCR ('v' ,"verbose" , &verbose,"be more verbose (show symbol address, etc)" ), -OPT_BOOLEAN ('m' ,"displacement" , &show_displacement, +OPT_BOOLEAN ('M' ,"displacement" , &show_displacement,"Show position displacement relative to baseline" ),OPT_BOOLEAN ('D' ,"dump-raw-trace" , &dump_trace,"dump raw trace in ASCII" ),
PATCH 4/14 perf: Document missing diff options
In reply to Shawn Bohrer--- tools/perf/Documentation/perf-diff.txt | 19 ++++++++++++++++++- 1 files changed, 18 insertions(+), 1 deletions(-)
tools/perf/Documentation/perf-diff.txt +18 -1
@@ -19,6 +19,18 @@ If no parameters are passed it will assume perf.data.old and perf.data. OPTIONS ------- +-M:: +--displacement:: + Show position displacement relative to baseline. + +-D:: +--dump-raw-trace:: + Dump raw trace in ASCII. + +-m:: +--modules:: + Load module symbols. WARNING: use only with -k and LIVE kernel + -d:: --dsos=:: Only consider symbols in these dsos. CSV that understands
@@ -42,7 +54,7 @@ OPTIONS --field-separator=:: Use a special separator character and don't pad with spaces, replacing - all occurances of this separator in symbol names (and other output) + all occurrences of this separator in symbol names (and other output) with a '.' character, that thus it's the only non valid separator. -v::
@@ -50,6 +62,11 @@ OPTIONS Be verbose, for instance, show the raw counts in addition to the diff. +-f:: +--force:: + Don't complain, do it. + + SEE ALSO -------- linkperf:perf-record[1]
Signature · 1 line
1.7.3.2
[tip:perf/core] perf diff: Document missing options
Commit-ID: 5ea4f857850450dc1442144a00547d6623d79d78 Gitweb: http://git.kernel.org/tip/5ea4f857850450dc1442144a00547d6623d79d78 Author: Shawn Bohrer <[email protected]> AuthorDate: Tue, 30 Nov 2010 19:57:12 -0600 Committer: Arnaldo Carvalho de Melo <[email protected]> CommitDate: Wed, 1 Dec 2010 18:22:46 -0200
perf diff: Document missing options
Cc: Ingo Molnar <[email protected]>
Cc: Paul Mackerras <[email protected]>
Cc: Peter Zijlstra <[email protected]>
LKML-Reference: <[email protected]>
--- tools/perf/Documentation/perf-diff.txt | 19 ++++++++++++++++++- 1 files changed, 18 insertions(+), 1 deletions(-)
tools/perf/Documentation/perf-diff.txt +18 -1
@@ -19,6 +19,18 @@ If no parameters are passed it will assume perf.data.old and perf.data. OPTIONS ------- +-M:: +--displacement:: + Show position displacement relative to baseline. + +-D:: +--dump-raw-trace:: + Dump raw trace in ASCII. + +-m:: +--modules:: + Load module symbols. WARNING: use only with -k and LIVE kernel + -d:: --dsos=:: Only consider symbols in these dsos. CSV that understands
@@ -42,7 +54,7 @@ OPTIONS --field-separator=:: Use a special separator character and don't pad with spaces, replacing - all occurances of this separator in symbol names (and other output) + all occurrences of this separator in symbol names (and other output) with a '.' character, that thus it's the only non valid separator. -v::
@@ -50,6 +62,11 @@ OPTIONS Be verbose, for instance, show the raw counts in addition to the diff. +-f:: +--force:: + Don't complain, do it. + + SEE ALSO -------- linkperf:perf-record[1]
PATCH 5/14 perf: Document missing kvm options
In reply to Shawn Bohrer--- tools/perf/Documentation/perf-kvm.txt | 8 +++++++- 1 files changed, 7 insertions(+), 1 deletions(-)
tools/perf/Documentation/perf-kvm.txt +7 -1
@@ -22,7 +22,7 @@ There are a couple of variants of perf kvm: a performance counter profile of guest os in realtime of an arbitrary workload. - 'perf kvm record <command>' to record the performance couinter profile + 'perf kvm record <command>' to record the performance counter profile of an arbitrary workload and save it into a perf data file. If both --host and --guest are input, the perf data file name is perf.data.kvm. If there is no --host but --guest, the file name is perf.data.guest.
@@ -40,6 +40,12 @@ There are a couple of variants of perf kvm: OPTIONS ------- +-i:: +--input=:: + Input file name. +-o:: +--output:: + Output file name. --host=:: Collect host side performance profile. --guest=::
Signature · 1 line
1.7.3.2
[tip:perf/core] perf kvm: Document missing options
Commit-ID: 5c0ef0ab077c8bdb224540fea60473439be1bdb4 Gitweb: http://git.kernel.org/tip/5c0ef0ab077c8bdb224540fea60473439be1bdb4 Author: Shawn Bohrer <[email protected]> AuthorDate: Tue, 30 Nov 2010 19:57:13 -0600 Committer: Arnaldo Carvalho de Melo <[email protected]> CommitDate: Wed, 1 Dec 2010 18:22:46 -0200
perf kvm: Document missing options
Cc: Ingo Molnar <[email protected]>
Cc: Paul Mackerras <[email protected]>
Cc: Peter Zijlstra <[email protected]>
LKML-Reference: <[email protected]>
--- tools/perf/Documentation/perf-kvm.txt | 8 +++++++- 1 files changed, 7 insertions(+), 1 deletions(-)
tools/perf/Documentation/perf-kvm.txt +7 -1
@@ -22,7 +22,7 @@ There are a couple of variants of perf kvm: a performance counter profile of guest os in realtime of an arbitrary workload. - 'perf kvm record <command>' to record the performance couinter profile + 'perf kvm record <command>' to record the performance counter profile of an arbitrary workload and save it into a perf data file. If both --host and --guest are input, the perf data file name is perf.data.kvm. If there is no --host but --guest, the file name is perf.data.guest.
@@ -40,6 +40,12 @@ There are a couple of variants of perf kvm: OPTIONS ------- +-i:: +--input=:: + Input file name. +-o:: +--output:: + Output file name. --host=:: Collect host side performance profile. --guest=::
PATCH 6/14 perf: Document missing lock options
In reply to Shawn Bohrer--- tools/perf/Documentation/perf-lock.txt | 15 +++++++++++++++ 1 files changed, 15 insertions(+), 0 deletions(-)
tools/perf/Documentation/perf-lock.txt +15 -0
@@ -24,6 +24,21 @@ and statistics with this 'perf lock' command. 'perf lock report' reports statistical data. +OPTIONS +------- + +-i:: +--input=<file>:: + Input file name. + +-v:: +--verbose:: + Be more verbose (show symbol address, etc). + +-D:: +--dump-raw-trace:: + Dump raw trace in ASCII. + SEE ALSO -------- linkperf:perf[1]
Signature · 1 line
1.7.3.2
[tip:perf/core] perf lock: Document missing options
Commit-ID: 4aace251519745977f4b5ddf625b630b4807be3a Gitweb: http://git.kernel.org/tip/4aace251519745977f4b5ddf625b630b4807be3a Author: Shawn Bohrer <[email protected]> AuthorDate: Tue, 30 Nov 2010 19:57:14 -0600 Committer: Arnaldo Carvalho de Melo <[email protected]> CommitDate: Wed, 1 Dec 2010 18:22:46 -0200
perf lock: Document missing options
Cc: Ingo Molnar <[email protected]>
Cc: Paul Mackerras <[email protected]>
Cc: Peter Zijlstra <[email protected]>
LKML-Reference: <[email protected]>
--- tools/perf/Documentation/perf-lock.txt | 15 +++++++++++++++ 1 files changed, 15 insertions(+), 0 deletions(-)
tools/perf/Documentation/perf-lock.txt +15 -0
@@ -24,6 +24,21 @@ and statistics with this 'perf lock' command. 'perf lock report' reports statistical data. +OPTIONS +------- + +-i:: +--input=<file>:: + Input file name. + +-v:: +--verbose:: + Be more verbose (show symbol address, etc). + +-D:: +--dump-raw-trace:: + Dump raw trace in ASCII. + SEE ALSO -------- linkperf:perf[1]
PATCH 7/14 perf: Fix spelling mistake in probe documentation
In reply to Shawn Bohrer--- tools/perf/Documentation/perf-probe.txt | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)
tools/perf/Documentation/perf-probe.txt +1 -1
@@ -115,7 +115,7 @@ Each probe argument follows below syntax. LINE SYNTAX ----------- -Line range is descripted by following syntax. +Line range is described by following syntax. "FUNC[:RLN[+NUM|-RLN2]]|SRC:ALN[+NUM|-ALN2]"
Signature · 1 line
1.7.3.2
Re: [PATCH 07/14] perf: Fix spelling mistake in probe documentation
Quoted 4 lines · Shawn Bohrer
Signed-off-by Shawn Bohrer <[email protected]>--- tools/perf/Documentation/perf-probe.txt | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)
Thanks for fixing :)
Quoted 14 lines
tools/perf/Documentation/perf-probe.txt +1
-1@@ -115,7 +115,7 @@ Each probe argument follows below syntax. LINE SYNTAX ----------- -Line range is descripted by following syntax. +Line range is described by following syntax. "FUNC[:RLN[+NUM|-RLN2]]|SRC:ALN[+NUM|-ALN2]"
Signature · 4 lines
Masami HIRAMATSU 2nd Dept. Linux Technology Center Hitachi, Ltd., Systems Development Laboratory E-mail: [email protected]
[tip:perf/core] perf probe: Fix spelling mistake in documentation
In reply to Shawn BohrerCommit-ID: 9d5b7f5b2b2c1ceade1fbbaefb2bd9167436329d Gitweb: http://git.kernel.org/tip/9d5b7f5b2b2c1ceade1fbbaefb2bd9167436329d Author: Shawn Bohrer <[email protected]> AuthorDate: Tue, 30 Nov 2010 19:57:15 -0600 Committer: Arnaldo Carvalho de Melo <[email protected]> CommitDate: Wed, 1 Dec 2010 18:22:46 -0200
perf probe: Fix spelling mistake in documentation
LKML-Reference: <[email protected]>
--- tools/perf/Documentation/perf-probe.txt | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)
tools/perf/Documentation/perf-probe.txt +1 -1
@@ -115,7 +115,7 @@ Each probe argument follows below syntax. LINE SYNTAX ----------- -Line range is descripted by following syntax. +Line range is described by following syntax. "FUNC[:RLN[+NUM|-RLN2]]|SRC:ALN[+NUM|-ALN2]"
PATCH 8/14 perf: Document missing record options
In reply to Shawn Bohrer--- tools/perf/Documentation/perf-record.txt | 17 +++++++++++++---- 1 files changed, 13 insertions(+), 4 deletions(-)
tools/perf/Documentation/perf-record.txt +13 -4
@@ -39,15 +39,24 @@ OPTIONS be passed as follows: '\mem:addr[:[r][w][x]]'. If you want to profile read-write accesses in 0x1000, just set 'mem:0x1000:rw'. + +--filter=<filter>:: + Event filter. + -a:: - System-wide collection. +--all-cpus:: + System-wide collection from all CPUs. -l:: Scale counter values. -p:: --pid=:: - Record events on existing pid. + Record events on existing process ID. + +-t:: +--tid=:: + Record events on existing thread ID. -r:: --realtime=::
@@ -109,8 +118,8 @@ Collect raw sample records from all opened counters (default for tracepoint coun -C:: --cpu:: -Collect samples only on the list of cpus provided. Multiple CPUs can be provided as a -comma-sperated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2. +Collect samples only on the list of CPUs provided. Multiple CPUs can be provided as a +comma-separated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2. In per-thread mode with inheritance mode on (default), samples are captured only when the thread executes on the designated CPUs. Default is to monitor all CPUs.
Signature · 1 line
1.7.3.2
[tip:perf/core] perf record: Document missing options
Commit-ID: 08dbd7e3fa2398910713b21399cca7c6e4b43011 Gitweb: http://git.kernel.org/tip/08dbd7e3fa2398910713b21399cca7c6e4b43011 Author: Shawn Bohrer <[email protected]> AuthorDate: Tue, 30 Nov 2010 19:57:16 -0600 Committer: Arnaldo Carvalho de Melo <[email protected]> CommitDate: Wed, 1 Dec 2010 18:22:47 -0200
perf record: Document missing options
Cc: Ingo Molnar <[email protected]>
Cc: Paul Mackerras <[email protected]>
Cc: Peter Zijlstra <[email protected]>
LKML-Reference: <[email protected]>
--- tools/perf/Documentation/perf-record.txt | 17 +++++++++++++---- 1 files changed, 13 insertions(+), 4 deletions(-)
tools/perf/Documentation/perf-record.txt +13 -4
@@ -39,15 +39,24 @@ OPTIONS be passed as follows: '\mem:addr[:[r][w][x]]'. If you want to profile read-write accesses in 0x1000, just set 'mem:0x1000:rw'. + +--filter=<filter>:: + Event filter. + -a:: - System-wide collection. +--all-cpus:: + System-wide collection from all CPUs. -l:: Scale counter values. -p:: --pid=:: - Record events on existing pid. + Record events on existing process ID. + +-t:: +--tid=:: + Record events on existing thread ID. -r:: --realtime=::
@@ -109,8 +118,8 @@ Collect raw sample records from all opened counters (default for tracepoint coun -C:: --cpu:: -Collect samples only on the list of cpus provided. Multiple CPUs can be provided as a -comma-sperated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2. +Collect samples only on the list of CPUs provided. Multiple CPUs can be provided as a +comma-separated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2. In per-thread mode with inheritance mode on (default), samples are captured only when the thread executes on the designated CPUs. Default is to monitor all CPUs.
PATCH 9/14 perf: Document missing report options
In reply to Shawn Bohrer--- tools/perf/Documentation/perf-report.txt | 49 +++++++++++++++++++++++++++-- 1 files changed, 45 insertions(+), 4 deletions(-)
tools/perf/Documentation/perf-report.txt +45 -4
@@ -20,6 +20,11 @@ OPTIONS -i:: --input=:: Input file name. (default: perf.data) + +-v:: +--verbose:: + Be more verbose. (show symbol address, etc) + -d:: --dsos=:: Only consider symbols in these dsos. CSV that understands
@@ -27,6 +32,10 @@ OPTIONS -n:: --show-nr-samples:: Show the number of samples for each symbol + +--showcpuutilization:: + Show sample percentage for different cpu modes. + -T:: --threads:: Show per-thread event counters
@@ -39,12 +48,24 @@ OPTIONS Only consider these symbols. CSV that understands file://filename entries. +-U:: +--hide-unresolved:: + Only display entries resolved to a symbol. + -s:: --sort=:: Sort by key(s): pid, comm, dso, symbol, parent. +-p:: +--parent=<regex>:: + regex filter to identify parent, see: '--sort parent' + +-x:: +--exclude-other:: + Only display entries with parent-match. + -w:: ---field-width=:: +--column-widths=<width[,width...]>:: Force each column width to the provided list, for large terminal readability.
@@ -52,19 +73,26 @@ OPTIONS --field-separator=:: Use a special separator character and don't pad with spaces, replacing - all occurances of this separator in symbol names (and other output) + all occurrences of this separator in symbol names (and other output) with a '.' character, that thus it's the only non valid separator. +-D:: +--dump-raw-trace:: + Dump raw trace in ASCII. + -g [type,min]:: --call-graph:: - Display callchains using type and min percent threshold. + Display call chains using type and min percent threshold. type can be either: - - flat: single column, linear exposure of callchains. + - flat: single column, linear exposure of call chains. - graph: use a graph tree, displaying absolute overhead rates. - fractal: like graph, but displays relative rates. Each branch of the tree is considered as a new profiled object. + Default: fractal,0.5. +--pretty=<key>:: + Pretty printing style. key: normal, raw + --stdio:: Use the stdio interface. --tui:: Use the TUI interface, that is integrated with annotate and allows
@@ -72,6 +100,19 @@ OPTIONS requires a tty, if one is not present, as when piping to other commands, the stdio interface is used. +-k:: +--vmlinux=<file>:: + vmlinux pathname + +-m:: +--modules:: + Load module symbols. WARNING: This should only be used with -k and + a LIVE kernel. + +-f:: +--force:: + Don't complain, do it. + SEE ALSO -------- linkperf:perf-stat[1]
Signature · 1 line
1.7.3.2
[tip:perf/core] perf report: Document missing options
Commit-ID: e04fffc321aeebab4962cfc120952272f2d1df98 Gitweb: http://git.kernel.org/tip/e04fffc321aeebab4962cfc120952272f2d1df98 Author: Shawn Bohrer <[email protected]> AuthorDate: Tue, 30 Nov 2010 19:57:17 -0600 Committer: Arnaldo Carvalho de Melo <[email protected]> CommitDate: Wed, 1 Dec 2010 18:22:47 -0200
perf report: Document missing options
Cc: Ingo Molnar <[email protected]>
Cc: Paul Mackerras <[email protected]>
Cc: Peter Zijlstra <[email protected]>
LKML-Reference: <[email protected]>
--- tools/perf/Documentation/perf-report.txt | 49 +++++++++++++++++++++++++++-- 1 files changed, 45 insertions(+), 4 deletions(-)
tools/perf/Documentation/perf-report.txt +45 -4
@@ -20,6 +20,11 @@ OPTIONS -i:: --input=:: Input file name. (default: perf.data) + +-v:: +--verbose:: + Be more verbose. (show symbol address, etc) + -d:: --dsos=:: Only consider symbols in these dsos. CSV that understands
@@ -27,6 +32,10 @@ OPTIONS -n:: --show-nr-samples:: Show the number of samples for each symbol + +--showcpuutilization:: + Show sample percentage for different cpu modes. + -T:: --threads:: Show per-thread event counters
@@ -39,12 +48,24 @@ OPTIONS Only consider these symbols. CSV that understands file://filename entries. +-U:: +--hide-unresolved:: + Only display entries resolved to a symbol. + -s:: --sort=:: Sort by key(s): pid, comm, dso, symbol, parent. +-p:: +--parent=<regex>:: + regex filter to identify parent, see: '--sort parent' + +-x:: +--exclude-other:: + Only display entries with parent-match. + -w:: ---field-width=:: +--column-widths=<width[,width...]>:: Force each column width to the provided list, for large terminal readability.
@@ -52,19 +73,26 @@ OPTIONS --field-separator=:: Use a special separator character and don't pad with spaces, replacing - all occurances of this separator in symbol names (and other output) + all occurrences of this separator in symbol names (and other output) with a '.' character, that thus it's the only non valid separator. +-D:: +--dump-raw-trace:: + Dump raw trace in ASCII. + -g [type,min]:: --call-graph:: - Display callchains using type and min percent threshold. + Display call chains using type and min percent threshold. type can be either: - - flat: single column, linear exposure of callchains. + - flat: single column, linear exposure of call chains. - graph: use a graph tree, displaying absolute overhead rates. - fractal: like graph, but displays relative rates. Each branch of the tree is considered as a new profiled object. + Default: fractal,0.5. +--pretty=<key>:: + Pretty printing style. key: normal, raw + --stdio:: Use the stdio interface. --tui:: Use the TUI interface, that is integrated with annotate and allows
@@ -72,6 +100,19 @@ OPTIONS requires a tty, if one is not present, as when piping to other commands, the stdio interface is used. +-k:: +--vmlinux=<file>:: + vmlinux pathname + +-m:: +--modules:: + Load module symbols. WARNING: This should only be used with -k and + a LIVE kernel. + +-f:: +--force:: + Don't complain, do it. + SEE ALSO -------- linkperf:perf-stat[1]
PATCH 10/14 perf: Document missing sched options
In reply to Shawn Bohrer--- tools/perf/Documentation/perf-sched.txt | 18 ++++++++++++++++-- 1 files changed, 16 insertions(+), 2 deletions(-)
tools/perf/Documentation/perf-sched.txt +16 -2
@@ -8,11 +8,11 @@ perf-sched - Tool to trace/measure scheduler properties (latencies) SYNOPSIS -------- [verse] -'perf sched' {record|latency|replay|trace} +'perf sched' {record|latency|map|replay|trace} DESCRIPTION ----------- -There are four variants of perf sched: +There are five variants of perf sched: 'perf sched record <command>' to record the scheduling events of an arbitrary workload.
@@ -30,8 +30,22 @@ There are four variants of perf sched: of the workload as it occurred when it was recorded - and can repeat it a number of times, measuring its performance.) + 'perf sched map' to print a textual context-switching outline of + workload captured via perf sched record. Columns stand for + individual CPUs, and the two-letter shortcuts stand for tasks that + are running on a CPU. A '*' denotes the CPU that had the event, and + a dot signals an idle CPU. + OPTIONS ------- +-i:: +--input=<file>:: + Input file name. (default: perf.data) + +-v:: +--verbose:: + Be more verbose. (show symbol address, etc) + -D:: --dump-raw-trace=:: Display verbose dump of the sched data.
Signature · 1 line
1.7.3.2
[tip:perf/core] perf sched: Document missing options
Commit-ID: 1eacc94a66ce347abbf75f223361b21461331383 Gitweb: http://git.kernel.org/tip/1eacc94a66ce347abbf75f223361b21461331383 Author: Shawn Bohrer <[email protected]> AuthorDate: Tue, 30 Nov 2010 19:57:18 -0600 Committer: Arnaldo Carvalho de Melo <[email protected]> CommitDate: Wed, 1 Dec 2010 18:22:48 -0200
perf sched: Document missing options
Cc: Ingo Molnar <[email protected]>
Cc: Paul Mackerras <[email protected]>
Cc: Peter Zijlstra <[email protected]>
LKML-Reference: <[email protected]>
--- tools/perf/Documentation/perf-sched.txt | 18 ++++++++++++++++-- 1 files changed, 16 insertions(+), 2 deletions(-)
tools/perf/Documentation/perf-sched.txt +16 -2
@@ -8,11 +8,11 @@ perf-sched - Tool to trace/measure scheduler properties (latencies) SYNOPSIS -------- [verse] -'perf sched' {record|latency|replay|trace} +'perf sched' {record|latency|map|replay|trace} DESCRIPTION ----------- -There are four variants of perf sched: +There are five variants of perf sched: 'perf sched record <command>' to record the scheduling events of an arbitrary workload.
@@ -30,8 +30,22 @@ There are four variants of perf sched: of the workload as it occurred when it was recorded - and can repeat it a number of times, measuring its performance.) + 'perf sched map' to print a textual context-switching outline of + workload captured via perf sched record. Columns stand for + individual CPUs, and the two-letter shortcuts stand for tasks that + are running on a CPU. A '*' denotes the CPU that had the event, and + a dot signals an idle CPU. + OPTIONS ------- +-i:: +--input=<file>:: + Input file name. (default: perf.data) + +-v:: +--verbose:: + Be more verbose. (show symbol address, etc) + -D:: --dump-raw-trace=:: Display verbose dump of the sched data.
PATCH 11/14 perf: Document missing stat options
In reply to Shawn Bohrer--- tools/perf/Documentation/perf-stat.txt | 34 +++++++++++++++++++++++++------ 1 files changed, 27 insertions(+), 7 deletions(-)
tools/perf/Documentation/perf-stat.txt +27 -7
@@ -8,8 +8,8 @@ perf-stat - Run a command and gather performance counter statistics SYNOPSIS -------- [verse] -'perf stat' [-e <EVENT> | --event=EVENT] [-S] [-a] <command> -'perf stat' [-e <EVENT> | --event=EVENT] [-S] [-a] -- <command> [<options>] +'perf stat' [-e <EVENT> | --event=EVENT] [-a] <command> +'perf stat' [-e <EVENT> | --event=EVENT] [-a] -- <command> [<options>] DESCRIPTION -----------
@@ -33,26 +33,46 @@ OPTIONS -i:: --no-inherit:: child tasks do not inherit counters + -p:: --pid=<pid>:: - stat events on existing pid + stat events on existing process id + +-t:: +--tid=<tid>:: + stat events on existing thread id -a:: - system-wide collection +--all-cpus:: + system-wide collection from all CPUs -c:: - scale counter values +--scale:: + scale/normalize counter values + +-r:: +--repeat=<n>:: + repeat command and print average + stddev (max: 100) -B:: +--big-num:: print large numbers with thousands' separators according to locale -C:: --cpu=:: -Count only on the list of cpus provided. Multiple CPUs can be provided as a -comma-sperated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2. +Count only on the list of CPUs provided. Multiple CPUs can be provided as a +comma-separated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2. In per-thread mode, this option is ignored. The -a option is still necessary to activate system-wide monitoring. Default is to count on all CPUs. +-n:: +--null:: + null run - don't start any counters + +-v:: +--verbose:: + be more verbose (show counter open errors, etc) + EXAMPLES --------
Signature · 1 line
1.7.3.2
[tip:perf/core] perf stat: Document missing options
Commit-ID: 8c207692fc8fa3ea1a5ff97ad698efb09a81975a Gitweb: http://git.kernel.org/tip/8c207692fc8fa3ea1a5ff97ad698efb09a81975a Author: Shawn Bohrer <[email protected]> AuthorDate: Tue, 30 Nov 2010 19:57:19 -0600 Committer: Arnaldo Carvalho de Melo <[email protected]> CommitDate: Wed, 1 Dec 2010 18:22:49 -0200
perf stat: Document missing options
Cc: Ingo Molnar <[email protected]>
Cc: Paul Mackerras <[email protected]>
Cc: Peter Zijlstra <[email protected]>
LKML-Reference: <[email protected]>
--- tools/perf/Documentation/perf-stat.txt | 34 +++++++++++++++++++++++++------ 1 files changed, 27 insertions(+), 7 deletions(-)
tools/perf/Documentation/perf-stat.txt +27 -7
@@ -8,8 +8,8 @@ perf-stat - Run a command and gather performance counter statistics SYNOPSIS -------- [verse] -'perf stat' [-e <EVENT> | --event=EVENT] [-S] [-a] <command> -'perf stat' [-e <EVENT> | --event=EVENT] [-S] [-a] -- <command> [<options>] +'perf stat' [-e <EVENT> | --event=EVENT] [-a] <command> +'perf stat' [-e <EVENT> | --event=EVENT] [-a] -- <command> [<options>] DESCRIPTION -----------
@@ -35,21 +35,33 @@ OPTIONS child tasks do not inherit counters -p:: --pid=<pid>:: - stat events on existing pid + stat events on existing process id + +-t:: +--tid=<tid>:: + stat events on existing thread id + -a:: - system-wide collection +--all-cpus:: + system-wide collection from all CPUs -c:: - scale counter values +--scale:: + scale/normalize counter values + +-r:: +--repeat=<n>:: + repeat command and print average + stddev (max: 100) -B:: +--big-num:: print large numbers with thousands' separators according to locale -C:: --cpu=:: -Count only on the list of cpus provided. Multiple CPUs can be provided as a -comma-sperated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2. +Count only on the list of CPUs provided. Multiple CPUs can be provided as a +comma-separated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2. In per-thread mode, this option is ignored. The -a option is still necessary to activate system-wide monitoring. Default is to count on all CPUs.
@@ -58,6 +70,14 @@ to activate system-wide monitoring. Default is to count on all CPUs. Do not aggregate counts across all monitored CPUs in system-wide mode (-a). This option is only valid in system-wide mode. +-n:: +--null:: + null run - don't start any counters + +-v:: +--verbose:: + be more verbose (show counter open errors, etc) + EXAMPLES --------
PATCH 12/14 perf: Fix spelling mistake in test documentation
In reply to Shawn Bohrer--- tools/perf/Documentation/perf-test.txt | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)
tools/perf/Documentation/perf-test.txt +1 -1
@@ -12,7 +12,7 @@ SYNOPSIS DESCRIPTION ----------- -This command does assorted sanity tests, initially thru linked routines but +This command does assorted sanity tests, initially through linked routines but also will look for a directory with more tests in the form of scripts. OPTIONS
Signature · 1 line
1.7.3.2
[tip:perf/core] perf test: Fix spelling mistake in documentation
Commit-ID: f68d6bd451782b58c2b563dc86f0a81ba106c30c Gitweb: http://git.kernel.org/tip/f68d6bd451782b58c2b563dc86f0a81ba106c30c Author: Shawn Bohrer <[email protected]> AuthorDate: Tue, 30 Nov 2010 19:57:20 -0600 Committer: Arnaldo Carvalho de Melo <[email protected]> CommitDate: Wed, 1 Dec 2010 18:22:49 -0200
perf test: Fix spelling mistake in documentation
Cc: Ingo Molnar <[email protected]>
Cc: Paul Mackerras <[email protected]>
Cc: Peter Zijlstra <[email protected]>
LKML-Reference: <[email protected]>
--- tools/perf/Documentation/perf-test.txt | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)
tools/perf/Documentation/perf-test.txt +1 -1
@@ -12,7 +12,7 @@ SYNOPSIS DESCRIPTION ----------- -This command does assorted sanity tests, initially thru linked routines but +This command does assorted sanity tests, initially through linked routines but also will look for a directory with more tests in the form of scripts. OPTIONS
PATCH 13/14 perf: Document missing top options
In reply to Shawn Bohrer--- tools/perf/Documentation/perf-top.txt | 28 ++++++++++++++++++++++++---- 1 files changed, 24 insertions(+), 4 deletions(-)
tools/perf/Documentation/perf-top.txt +24 -4
@@ -12,7 +12,7 @@ SYNOPSIS DESCRIPTION ----------- -This command generates and displays a performance counter profile in realtime. +This command generates and displays a performance counter profile in real time. OPTIONS
@@ -27,8 +27,8 @@ OPTIONS -C <cpu-list>:: --cpu=<cpu>:: -Monitor only on the list of cpus provided. Multiple CPUs can be provided as a -comma-sperated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2. +Monitor only on the list of CPUs provided. Multiple CPUs can be provided as a +comma-separated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2. Default is to monitor all CPUS. -d <seconds>::
@@ -50,6 +50,10 @@ Default is to monitor all CPUS. --count-filter=<count>:: Only display functions with more events than this. +-g:: +--group:: + Put the counters into a counter group. + -F <freq>:: --freq=<freq>:: Profile at this frequency.
@@ -68,7 +72,11 @@ Default is to monitor all CPUS. -p <pid>:: --pid=<pid>:: - Profile events on existing pid. + Profile events on existing Process ID. + +-t <tid>:: +--tid=<tid>:: + Profile events on existing thread ID. -r <priority>:: --realtime=<priority>::
@@ -78,6 +86,18 @@ Default is to monitor all CPUS. --sym-annotate=<symbol>:: Annotate this symbol. +-K:: +--hide_kernel_symbols:: + Hide kernel symbols. + +-U:: +--hide_user_symbols:: + Hide user symbols. + +-D:: +--dump-symtab:: + Dump the symbol table used for profiling. + -v:: --verbose:: Be more verbose (show counter open errors, etc).
Signature · 1 line
1.7.3.2
[tip:perf/core] perf top: Document missing options
Commit-ID: 2e7a988198b11877d844d147ec81c7caea82c314 Gitweb: http://git.kernel.org/tip/2e7a988198b11877d844d147ec81c7caea82c314 Author: Shawn Bohrer <[email protected]> AuthorDate: Tue, 30 Nov 2010 19:57:21 -0600 Committer: Arnaldo Carvalho de Melo <[email protected]> CommitDate: Wed, 1 Dec 2010 18:22:48 -0200
perf top: Document missing options
Cc: Ingo Molnar <[email protected]>
Cc: Paul Mackerras <[email protected]>
Cc: Peter Zijlstra <[email protected]>
LKML-Reference: <[email protected]>
--- tools/perf/Documentation/perf-top.txt | 28 ++++++++++++++++++++++++---- 1 files changed, 24 insertions(+), 4 deletions(-)
tools/perf/Documentation/perf-top.txt +24 -4
@@ -12,7 +12,7 @@ SYNOPSIS DESCRIPTION ----------- -This command generates and displays a performance counter profile in realtime. +This command generates and displays a performance counter profile in real time. OPTIONS
@@ -27,8 +27,8 @@ OPTIONS -C <cpu-list>:: --cpu=<cpu>:: -Monitor only on the list of cpus provided. Multiple CPUs can be provided as a -comma-sperated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2. +Monitor only on the list of CPUs provided. Multiple CPUs can be provided as a +comma-separated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2. Default is to monitor all CPUS. -d <seconds>::
@@ -50,6 +50,10 @@ Default is to monitor all CPUS. --count-filter=<count>:: Only display functions with more events than this. +-g:: +--group:: + Put the counters into a counter group. + -F <freq>:: --freq=<freq>:: Profile at this frequency.
@@ -68,7 +72,11 @@ Default is to monitor all CPUS. -p <pid>:: --pid=<pid>:: - Profile events on existing pid. + Profile events on existing Process ID. + +-t <tid>:: +--tid=<tid>:: + Profile events on existing thread ID. -r <priority>:: --realtime=<priority>::
@@ -78,6 +86,18 @@ Default is to monitor all CPUS. --sym-annotate=<symbol>:: Annotate this symbol. +-K:: +--hide_kernel_symbols:: + Hide kernel symbols. + +-U:: +--hide_user_symbols:: + Hide user symbols. + +-D:: +--dump-symtab:: + Dump the symbol table used for profiling. + -v:: --verbose:: Be more verbose (show counter open errors, etc).
PATCH 14/14 perf: Document missing trace options
In reply to Shawn Bohrer--- tools/perf/Documentation/perf-trace.txt | 7 +++++++ 1 files changed, 7 insertions(+), 0 deletions(-)
tools/perf/Documentation/perf-trace.txt +7 -0
@@ -104,6 +104,13 @@ OPTIONS normally don't - this option allows the latter to be run in system-wide mode. +-i:: +--input=:: + Input file name. + +-d:: +--debug-mode:: + Do various checks like samples ordering and lost events. SEE ALSO --------
Signature · 1 line
1.7.3.2
[tip:perf/core] perf trace: Document missing options
Commit-ID: 646420f1bcf6ddf3e150f92a5f2a8bd7d6057ff8 Gitweb: http://git.kernel.org/tip/646420f1bcf6ddf3e150f92a5f2a8bd7d6057ff8 Author: Shawn Bohrer <[email protected]> AuthorDate: Tue, 30 Nov 2010 19:57:22 -0600 Committer: Arnaldo Carvalho de Melo <[email protected]> CommitDate: Wed, 1 Dec 2010 18:22:48 -0200
perf trace: Document missing options
Cc: Ingo Molnar <[email protected]>
Cc: Paul Mackerras <[email protected]>
Cc: Peter Zijlstra <[email protected]>
LKML-Reference: <[email protected]>
--- tools/perf/Documentation/perf-trace.txt | 7 +++++++ 1 files changed, 7 insertions(+), 0 deletions(-)
tools/perf/Documentation/perf-trace.txt +7 -0
@@ -104,6 +104,13 @@ OPTIONS normally don't - this option allows the latter to be run in system-wide mode. +-i:: +--input=:: + Input file name. + +-d:: +--debug-mode:: + Do various checks like samples ordering and lost events. SEE ALSO --------