linux.git/tools/perf/util/session.c, branch v2.6.39

perf session: Pass evsel in event_ops->sample()

2011-03-23T22:28:58+00:00

Resolving the sample->id to an evsel since the most advanced tools,
report and annotate, and the others will too when they evolve to
properly support multi-event perf.data files.

Good also because it does an extra validation, checking that the ID is
valid when present. When that is not the case, the overhead is just a
branch + function call (perf_evlist__id2evsel).

Cc: Frederic Weisbecker 
Cc: Ingo Molnar 
Cc: Mike Galbraith 
Cc: Paul Mackerras 
Cc: Peter Zijlstra 
Cc: Stephane Eranian 
Cc: Tom Zanussi 
LKML-Reference: 
Signed-off-by: Arnaldo Carvalho de Melo

perf script: Add support for dumping symbols

2011-03-14T20:06:50+00:00

Add option to dump symbols found in events.

e.g., perf script -f comm,pid,tid,time,trace,sym

swapper     0/0       537.037184: prev_comm=swapper prev_pid=0 prev_prio=120...
        ffffffff81030350 perf_trace_sched_switch ([kernel.kallsyms])
        ffffffff81382ac5 schedule ([kernel.kallsyms])
        ffffffff8100134a cpu_idle ([kernel.kallsyms])
        ffffffff81370b39 rest_init ([kernel.kallsyms])
        ffffffff81696c23 start_kernel ([kernel.kallsyms].init.text)
        ffffffff816962af x86_64_start_reservations ([kernel.kallsyms].init.text)
        ffffffff816963b9 x86_64_start_kernel ([kernel.kallsyms].init.text)

sshd  1675/1675    537.037309: prev_comm=sshd prev_pid=1675 prev_prio=120...
        ffffffff81030350 perf_trace_sched_switch ([kernel.kallsyms])
        ffffffff81382ac5 schedule ([kernel.kallsyms])
        ffffffff813837aa schedule_hrtimeout_range_clock ([kernel.kallsyms])
        ffffffff81383886 schedule_hrtimeout_range ([kernel.kallsyms])
        ffffffff8110c4f9 poll_schedule_timeout ([kernel.kallsyms])
        ffffffff8110cd20 do_select ([kernel.kallsyms])
        ffffffff8110ced8 core_sys_select ([kernel.kallsyms])
        ffffffff8110d00d sys_select ([kernel.kallsyms])
        ffffffff81002bc2 system_call ([kernel.kallsyms])
            7f1647e56e93 __GI_select (/lib64/libc-2.12.90.so)

netstat  1692/1692    537.038664: prev_comm=netstat prev_pid=1692 prev_prio=...
        ffffffff81030350 perf_trace_sched_switch ([kernel.kallsyms])
        ffffffff81382ac5 schedule ([kernel.kallsyms])
        ffffffff81002c3a sysret_careful ([kernel.kallsyms])
            7f7a6cd1b210 __GI___libc_read (/lib64/libc-2.12.90.so)

Acked-by: Frederic Weisbecker 
Cc: Frederic Weisbecker 
Cc: Ingo Molnar 
Cc: Paul Mackerras 
Cc: Peter Zijlstra 
Cc: Thomas Gleixner 
LKML-Reference: <1299734608-5223-6-git-send-email-daahern@cisco.com>
Signed-off-by: David Ahern 
Signed-off-by: Arnaldo Carvalho de Melo

perf session: Use evlist/evsel for managing perf.data attributes

2011-03-10T14:15:54+00:00

So that we can reuse things like the id to attr lookup routine
(perf_evlist__id2evsel) that uses a hash table instead of the linear
lookup done in the older perf_header_attr routines, etc.

Also to make evsels/evlist more pervasive an API, simplyfing using the
emerging perf lib.

cc: Arun Sharma 
Cc: Frederic Weisbecker 
Cc: Ingo Molnar 
Cc: Mike Galbraith 
Cc: Paul Mackerras 
Cc: Peter Zijlstra 
Cc: Stephane Eranian 
Cc: Tom Zanussi 
LKML-Reference: 
Signed-off-by: Arnaldo Carvalho de Melo

perf tools: Improve support for sessions with multiple events

2011-03-06T16:13:40+00:00

By creating an perf_evlist out of the attributes in the perf.data file
header, so that we can use evlists and evsels when reading recorded
sessions in addition to when we record sessions.

More work is needed to allow tools to allow the user to select which
events are wanted when browsing sessions, be it just one or a subset of
them, aggregated or showed at the same time but with different
indications on the UI to allow seeing workloads thru different views at
the same time.

But the overall goal/trend is to more uniformly use evsels and evlists.

Cc: Frederic Weisbecker 
Cc: Ingo Molnar 
Cc: Mike Galbraith 
Cc: Paul Mackerras 
Cc: Peter Zijlstra 
Cc: Stephane Eranian 
Cc: Tom Zanussi 
LKML-Reference: 
Signed-off-by: Arnaldo Carvalho de Melo

perf tools: Kill event_t typedef, use 'union perf_event' instead

2011-01-29T18:25:37+00:00

And move the event_t methods to the perf_event__ too.

No code changes, just namespace consistency.

Cc: Frederic Weisbecker 
Cc: Ingo Molnar 
Cc: Mike Galbraith 
Cc: Paul Mackerras 
Cc: Peter Zijlstra 
Cc: Stephane Eranian 
Cc: Tom Zanussi 
LKML-Reference: 
Signed-off-by: Arnaldo Carvalho de Melo

perf tools: Rename 'struct sample_data' to 'struct perf_sample'

2011-01-29T18:25:20+00:00

Making the namespace more uniform.

Cc: Frederic Weisbecker 
Cc: Ingo Molnar 
Cc: Mike Galbraith 
Cc: Paul Mackerras 
Cc: Peter Zijlstra 
Cc: Stephane Eranian 
Cc: Tom Zanussi 
LKML-Reference: 
Signed-off-by: Arnaldo Carvalho de Melo

perf tools: Move event__parse_sample to evsel.c

2011-01-24T15:17:56+00:00

To avoid linking more stuff in the python binding I'm working on, future
csets will make the sample type be taken from the evsel itself, but for
that we need to first have one file per cpu and per sample_type, not a
single perf.data file.

Cc: Frederic Weisbecker 
Cc: Ingo Molnar 
Cc: Mike Galbraith 
Cc: Paul Mackerras 
Cc: Peter Zijlstra 
Cc: Stephane Eranian 
Cc: Tom Zanussi 
LKML-Reference: 
Signed-off-by: Arnaldo Carvalho de Melo

perf callchain: Feed callchains into a cursor

2011-01-22T21:56:31+00:00

The callchains are fed with an array of a fixed size.
As a result we iterate over each callchains three times:

- 1st to resolve symbols
- 2nd to filter out context boundaries
- 3rd for the insertion into the tree

This also involves some pairs of memory allocation/deallocation
everytime we insert a callchain, for the filtered out array of
addresses and for the array of symbols that comes along.

Instead, feed the callchains through a linked list with persistent
allocations. It brings several pros like:

- Merge the 1st and 2nd iterations in one. That was possible before
but in a way that would involve allocating an array slightly taller
than necessary because we don't know in advance the number of context
boundaries to filter out.

- Much lesser allocations/deallocations. The linked list keeps
persistent empty entries for the next usages and is extendable at
will.

- Makes it easier for multiple sources of callchains to feed a
stacktrace together. This is deemed to pave the way for cfi based
callchains wherein traditional frame pointer based kernel
stacktraces will precede cfi based user ones, producing an overall
callchain which size is hardly predictable. This requirement
makes the static array obsolete and makes a linked list based
iterator a much more flexible fit.

Basic testing on a big perf file containing callchains (~ 176 MB)
has shown a throughput gain of about 11% with perf report.

Cc: Ingo Molnar 
Cc: Paul Mackerras 
Cc: Peter Zijlstra 
LKML-Reference: <1294977121-5700-2-git-send-email-fweisbec@gmail.com>
Signed-off-by: Frederic Weisbecker 
Signed-off-by: Arnaldo Carvalho de Melo

perf tools: Fix 64 bit integer format strings

2011-01-23T01:41:57+00:00

Using %L[uxd] has issues in some architectures, like on ppc64.  Fix it
by making our 64 bit integers typedefs of stdint.h types and using
PRI[ux]64 like, for instance, git does.

Reported by Denis Kirjanov that provided a patch for one case, I went
and changed all cases.

Reported-by: Denis Kirjanov 
Tested-by: Denis Kirjanov 
LKML-Reference: <20110120093246.GA8031@hera.kernel.org>
Cc: Denis Kirjanov 
Cc: Frederic Weisbecker 
Cc: Ingo Molnar 
Cc: Mike Galbraith 
Cc: Paul Mackerras 
Cc: Peter Zijlstra 
Cc: Pingtian Han 
Cc: Stephane Eranian 
Cc: Tom Zanussi 
Signed-off-by: Arnaldo Carvalho de Melo

perf session: Fix infinite loop in __perf_session__process_events

2011-01-11T00:23:08+00:00

In this if statement:

        if (head + event->header.size >= mmap_size) {
                if (mmaps[map_idx]) {
                        munmap(mmaps[map_idx], mmap_size);
                        mmaps[map_idx] = NULL;
                }

                page_offset = page_size * (head / page_size);
                file_offset += page_offset;
                head -= page_offset;
                goto remap;
        }

With, for instance, these values:

head=2992
event->header.size=48
mmap_size=3040

We end up endlessly looping back to remap. Off by one.

Problem introduced in 55b4462.

Reported-by: Linus Torvalds 
Reported-by: Ingo Molnar 
Reported-by: David Ahern 
Bisected-by: David Ahern 
Tested-by: David Ahern 
Cc: David Ahern 
Cc: Frederic Weisbecker 
Cc: Ingo Molnar 
Cc: Mike Galbraith 
Cc: Paul Mackerras 
Cc: Peter Zijlstra 
Cc: Stephane Eranian 
Cc: Thomas Gleixner 
Cc: Tom Zanussi 
LKML-Reference: 
Signed-off-by: Arnaldo Carvalho de Melo