We need to use the per event info snapshoted at record time to
synthesize the events name, so do it just after reading the perf.data
headers, when we already processed the /sys events data, otherwise we'll
end up using the local /sys that only by sheer luck will have the same
tracepoint ID -> real event association.
Example:
# uname -a
Linux felicio.ghostprotocols.net 3.4.0-rc5+ #1 SMP Sat May 19 15:27:11 BRT 2012 x86_64 x86_64 x86_64 GNU/Linux
# perf record -e sched:sched_switch usleep 1
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.015 MB perf.data (~648 samples) ]
# cat /t/events/sched/sched_switch/id
279
# perf evlist -v
sched:sched_switch: sample_freq=1, type: 2, config: 279, size: 80, sample_type: 1159, read_format: 7, disabled: 1, inherit: 1, mmap: 1, comm: 1, enable_on_exec: 1, sample_id_all: 1, exclude_guest: 1
#
So on the above machine the sched:sched_switch has tracepoint id 279, but on
the machine were we'll analyse it it has a different id:
$ cat /t/events/sched/sched_switch/id
56
$ perf evlist -i /tmp/perf.data
kmem:mm_balancedirty_writeout
$ cat /t/events/kmem/mm_balancedirty_writeout/id
279
With this fix:
$ perf evlist -i /tmp/perf.data
sched:sched_switch
Reported-by: Dmitry Antipov <dmitry.antipov@linaro.org>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
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/n/tip-auwks8fpuhmrdpiefs55o5oz@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The following commit:
commit 56f3bae706
Author: Jim Cromie <jim.cromie@gmail.com>
Date: Wed Sep 7 17:14:00 2011 -0600
perf stat: Add --log-fd <N> option to redirect stderr elsewhere
introduced a bug in the way perf stat outputs the results by default,
i.e., without the --log-fd or --output option. It would default to
writing to file descriptor 0, i.e., stdin. Writing to stdin is allowed
and is equivalent to writing to stdout. However, there is a major
difference for any script that was already capturing the output of perf
stat via redirection:
perf stat >/tmp/log .... or perf stat 2>/tmp/log ....
They would not capture anything anymore. They would have to do:
perf stat 0>/tmp/log ...
This breaks compatibility with existing scripts and does not look very
natural.
This patch fixes the problem by looking at output_fd only when it was
modified by user (> 0). It also checks that the value if positive.
Passing --log-fd 0 is ignored.
I would also argue that defaulting to stderr for the results is not the
right thing to do, though this patch does not address this specific
issue.
Signed-off-by: Stephane Eranian <eranian@google.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Jim Cromie <jim.cromie@gmail.com>
Link: http://lkml.kernel.org/r/20120515111111.GA9870@quad
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Based on Jiri's latest attempt:
https://lkml.org/lkml/2012/5/16/61
Basically, adds_features should be byte swapped assuming unsigned
longs are either 8-bytes (u64) or 4-bytes (u32).
Fixes 32-bit ppc dumping 64-bit x86 feature data:
========
captured on: Sun May 20 19:23:23 2012
hostname : nxos-vdc-dev3
os release : 3.4.0-rc7+
perf version : 3.4.rc4.137.g978da3
arch : x86_64
nrcpus online : 16
nrcpus avail : 16
cpudesc : Intel(R) Xeon(R) CPU E5540 @ 2.53GHz
cpuid : GenuineIntel,6,26,5
total memory : 24680324 kB
...
Verified 64-bit x86 can still dump feature data for 32-bit ppc.
Signed-off-by: David Ahern <dsahern@gmail.com>
Reviewed-by: Jiri Olsa <jolsa@redhat.com>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/4FBBB539.5010805@gmail.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
We swap the sample_id_all header by u64 pointers. Some members of the
header happen to be 32 bit values. We need to handle them separatelly.
Together with other endianity patches, this change fixies perf report
discrepancies on origin and target systems as described in test 1 below,
e.g. following perf report diff:
...
0.12% ps [kernel.kallsyms] [k] clear_page
- 0.12% awk bash [.] alloc_word_desc
+ 0.12% awk bash [.] yyparse
0.11% beah-rhts-task libpython2.6.so.1.0 [.] 0x5560e
0.10% perf libc-2.12.so [.] __ctype_toupper_loc
- 0.09% rhts-test-runne bash [.] maybe_make_export_env
+ 0.09% rhts-test-runne bash [.] 0x385a0
0.09% ps [kernel.kallsyms] [k] page_fault
...
Note, running following to test perf endianity handling:
test 1)
- origin system:
# perf record -a -- sleep 10 (any perf record will do)
# perf report > report.origin
# perf archive perf.data
- copy the perf.data, report.origin and perf.data.tar.bz2
to a target system and run:
# tar xjvf perf.data.tar.bz2 -C ~/.debug
# perf report > report.target
# diff -u report.origin report.target
- the diff should produce no output
(besides some white space stuff and possibly different
date/TZ output)
test 2)
- origin system:
# perf record -ag -fo /tmp/perf.data -- sleep 1
- mount origin system root to the target system on /mnt/origin
- target system:
# perf script --symfs /mnt/origin -I -i /mnt/origin/tmp/perf.data \
--kallsyms /mnt/origin/proc/kallsyms
- complete perf.data header is displayed
Signed-off-by: Jiri Olsa <jolsa@redhat.com>
Reviewed-by: David Ahern <dsahern@gmail.com>
Tested-by: David Ahern <dsahern@gmail.com>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1338380624-7443-4-git-send-email-jolsa@redhat.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Adding endianity swapping for event header attached via sample_id_all.
Currently we dont do that and it's causing wrong data to be read when
running report on architecture with different endianity than the record.
The perf is currently able to process 32-bit PPC samples on 32-bit
and 64-bit x86.
Together with other endianity patches, this change fixies perf report
discrepancies on origin and target systems as described in test 1
below, e.g. following perf report diff:
...
0.12% ps [kernel.kallsyms] [k] clear_page
- 0.12% awk bash [.] alloc_word_desc
+ 0.12% awk bash [.] yyparse
0.11% beah-rhts-task libpython2.6.so.1.0 [.] 0x5560e
0.10% perf libc-2.12.so [.] __ctype_toupper_loc
- 0.09% rhts-test-runne bash [.] maybe_make_export_env
+ 0.09% rhts-test-runne bash [.] 0x385a0
0.09% ps [kernel.kallsyms] [k] page_fault
...
Note, running following to test perf endianity handling:
test 1)
- origin system:
# perf record -a -- sleep 10 (any perf record will do)
# perf report > report.origin
# perf archive perf.data
- copy the perf.data, report.origin and perf.data.tar.bz2
to a target system and run:
# tar xjvf perf.data.tar.bz2 -C ~/.debug
# perf report > report.target
# diff -u report.origin report.target
- the diff should produce no output
(besides some white space stuff and possibly different
date/TZ output)
test 2)
- origin system:
# perf record -ag -fo /tmp/perf.data -- sleep 1
- mount origin system root to the target system on /mnt/origin
- target system:
# perf script --symfs /mnt/origin -I -i /mnt/origin/tmp/perf.data \
--kallsyms /mnt/origin/proc/kallsyms
- complete perf.data header is displayed
Signed-off-by: Jiri Olsa <jolsa@redhat.com>
Reviewed-by: David Ahern <dsahern@gmail.com>
Tested-by: David Ahern <dsahern@gmail.com>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1338380624-7443-3-git-send-email-jolsa@redhat.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Currently we dont care about the file object's endianness. It's possible
we read buildid file object from different architecture than we are
currentlly running on. So we need to care about properly reading such
object's data - handle different endianness properly.
Adding:
needs_swap DSO field
dso__swap_init function to initialize DSO's needs_swap
DSO__SWAP to read the data with proper swaps
Together with other endianity patches, this change fixies perf report
discrepancies on origin and target systems as described in test 1 below,
e.g. following perf report diff:
...
0.12% ps [kernel.kallsyms] [k] clear_page
- 0.12% awk bash [.] alloc_word_desc
+ 0.12% awk bash [.] yyparse
0.11% beah-rhts-task libpython2.6.so.1.0 [.] 0x5560e
0.10% perf libc-2.12.so [.] __ctype_toupper_loc
- 0.09% rhts-test-runne bash [.] maybe_make_export_env
+ 0.09% rhts-test-runne bash [.] 0x385a0
0.09% ps [kernel.kallsyms] [k] page_fault
...
Note, running following to test perf endianity handling:
test 1)
- origin system:
# perf record -a -- sleep 10 (any perf record will do)
# perf report > report.origin
# perf archive perf.data
- copy the perf.data, report.origin and perf.data.tar.bz2
to a target system and run:
# tar xjvf perf.data.tar.bz2 -C ~/.debug
# perf report > report.target
# diff -u report.origin report.target
- the diff should produce no output
(besides some white space stuff and possibly different
date/TZ output)
test 1)
- origin system:
# perf record -ag -fo /tmp/perf.data -- sleep 1
- mount origin system root to the target system on /mnt/origin
- target system:
# perf script --symfs /mnt/origin -I -i /mnt/origin/tmp/perf.data \
--kallsyms /mnt/origin/proc/kallsyms
- complete perf.data header is displayed
Signed-off-by: Jiri Olsa <jolsa@redhat.com>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1338380624-7443-2-git-send-email-jolsa@redhat.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
$ perf script -i /tmp/perf.data
...
gcc 13623 544315.062858: context-switches:
ffffffff815f65c9 __schedule ([kernel.kallsyms])
ffffffff81087cea __cond_resched ([kernel.kallsyms])
ffffffff815f6b92 _cond_resched ([kernel.kallsyms])
ffffffff815fb87a do_page_fault ([kernel.kallsyms])
ffffffff815f8465 page_fault ([kernel.kallsyms])
2b7a71ea0303 _dl_lookup_symbol_x ([kernel.kallsyms])
2b7a71ea1eb5 _dl_relocate_object ([kernel.kallsyms])
2b7a71e99b2e dl_main ([kernel.kallsyms])
2b7a71eab7f4 _dl_sysdep_start ([kernel.kallsyms])
All DSO's in a callchain are printed as [kernel.kallsyms].
git bisect chased it to:
547a92e0ae is the first bad commit
commit 547a92e0ae
Author: Akihiro Nagai <akihiro.nagai.hw@hitachi.com>
Date: Mon Jan 30 13:42:57 2012 +0900
perf script: Unify the expressions indicating "unknown"
The perf script command uses various expressions to indicate "unknown".
It is unfriendly for user scripts to parse it. So, this patch unifies
the expressions to "[unknown]".
Looks like a copy-paste in that the other references use al.map but this one
should be node->map.
With this patch you get:
$ perf script -i /tmp/perf.data
...
gcc 13623 544315.062858: context-switches:
ffffffff815f65c9 __schedule ([kernel.kallsyms])
ffffffff81087cea __cond_resched ([kernel.kallsyms])
ffffffff815f6b92 _cond_resched ([kernel.kallsyms])
ffffffff815fb87a do_page_fault ([kernel.kallsyms])
ffffffff815f8465 page_fault ([kernel.kallsyms])
2b7a71ea0303 _dl_lookup_symbol_x (/lib64/ld-2.14.90.so)
2b7a71ea1eb5 _dl_relocate_object (/lib64/ld-2.14.90.so)
2b7a71e99b2e dl_main (/lib64/ld-2.14.90.so)
2b7a71eab7f4 _dl_sysdep_start (/lib64/ld-2.14.90.so)
Signed-off-by: David Ahern <dsahern@gmail.com>
Cc: Akihiro Nagai <akihiro.nagai.hw@hitachi.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1338353906-60706-1-git-send-email-dsahern@gmail.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
When no event is specified the tools use perf_evlist__add_default(), that will
call event_attr_init to initialize the KVM exclusion bits.
When the change was made to the tools so that by default guest samples would be
excluded, the changes were made just to the parsing routines and to
perf_evlist__add_default(), not to perf_evlist__add_attrs, that is used so far
just by perf stat to add multiple events, according to the level of detail
specified.
Recently the tools were changed to reconstruct the event name from all the
details in perf_event_attr, not just from .type and .config, but taking into
account all the feature bits (.exclude_{guest,host,user,kernel,etc},
.precise_ip, etc).
That is when we noticed that the default for perf stat wasn't the one for the
rest of the tools, i.e. the .exclude_guest bit wasn't being set.
I.e. the default, that doesn't call event_attr_init was showing the :HG
modifier:
$ perf stat usleep 1
Performance counter stats for 'usleep 1':
0.942119 task-clock # 0.454 CPUs utilized
1 context-switches # 0.001 M/sec
0 CPU-migrations # 0.000 K/sec
126 page-faults # 0.134 M/sec
693,193 cycles:HG # 0.736 GHz [40.11%]
407,461 stalled-cycles-frontend:HG # 58.78% frontend cycles idle [72.29%]
365,403 stalled-cycles-backend:HG # 52.71% backend cycles idle
465,982 instructions:HG # 0.67 insns per cycle
# 0.87 stalled cycles per insn
89,760 branches:HG # 95.275 M/sec
6,178 branch-misses:HG # 6.88% of all branches
0.002077228 seconds time elapsed
While if one explicitely specifies the same events, which will make the parsing code
to be called and thus event_attr_init is called:
$ perf stat -e task-clock,context-switches,migrations,page-faults,cycles,stalled-cycles-frontend,stalled-cycles-backend,instructions,branches,branch-misses usleep 1
Performance counter stats for 'usleep 1':
1.040349 task-clock # 0.500 CPUs utilized
2 context-switches # 0.002 M/sec
0 CPU-migrations # 0.000 K/sec
127 page-faults # 0.122 M/sec
587,966 cycles # 0.565 GHz [13.18%]
459,167 stalled-cycles-frontend # 78.09% frontend cycles idle
390,249 stalled-cycles-backend # 66.37% backend cycles idle
504,006 instructions # 0.86 insns per cycle
# 0.91 stalled cycles per insn
96,455 branches # 92.714 M/sec
6,522 branch-misses # 6.76% of all branches [96.12%]
0.002078681 seconds time elapsed
Fix it by introducing a perf_evlist__add_default_attrs method that will call
evlist_attr_init in all the perf_event_attr entries before adding the events.
Reported-by: Ingo Molnar <mingo@kernel.org>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
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/n/tip-4eysr236r0pgiyum9epwxw7s@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Merge back Linus's latest branch so that we pick up the uprobes changes.
( I tested this branch locally and while it's one from the middle of the
merge window it's a good one to base further work off. )
Signed-off-by: Ingo Molnar <mingo@kernel.org>
