linux-stable.git/tools/perf/util/trace-event-parse.c, branch linux-3.5.y

perf script: Fix format regression due to libtraceevent merge

2012-07-02T14:08:11+00:00

Consider the commands:
    perf record -e sched:sched_switch -fo /tmp/perf.data  -a -- sleep 1
    perf script -i /tmp/perf.data

In v3.4 the output has the form (lines wrapped here)
    perf 29214 [005] 821043.582596: sched_switch:
prev_comm=perf prev_pid=29214 prev_prio=120
prev_state=S ==> next_comm=swapper/5 next_pid=0 next_prio=120

In 3.5 that same line has become:
    perf 29214 [005] 821043.582596: sched_switch:
<...>-29214 [005]     0.000000000: sched_switch:
prev_comm=perf prev_pid=29214 prev_prio=120
prev_state=S ==> next_comm=swapper/5 next_pid=0 next_prio=120

Note the duplicates in the output -- pid, cpu, event name. With
this patch the v3.4 output is restored:
    perf 29214 [005] 821043.582596: sched_switch:
prev_comm=perf prev_pid=29214 prev_prio=120
prev_state=S ==> next_comm=swapper/5 next_pid=0 next_prio=120

v3:
Remove that pesky newline too. Output now matches v3.4 (pre-libtracevent).

v2:
Change print_trace_event function local to perf per Steve's comments.

Signed-off-by: David Ahern 
Acked-by: Steven Rostedt 
Cc: Frederic Weisbecker 
Cc: Namhyung Kim 
Cc: Peter Zijlstra 
Cc: Stephane Eranian 
Cc: Steven Rostedt 
Link: http://lkml.kernel.org/r/1339698977-68962-1-git-send-email-dsahern@gmail.com
Signed-off-by: Arnaldo Carvalho de Melo

parse-events: Rename struct record to struct pevent_record

2012-04-25T11:35:48+00:00

As libtraceevent will be a library, having struct record is far
too generic of a name to use. Renaming it to be consistent with the
rest of the functions will be a better long term solution.

Signed-off-by: Steven Rostedt 
Cc: Ingo Molnar 
Cc: Thomas Gleixner 
Cc: Peter Zijlstra 
Cc: Arnaldo Carvalho de Melo 
Cc: Steven Rostedt 
Cc: Borislav Petkov 
Cc: Jiri Olsa 
Cc: Arun Sharma 
Cc: Namhyung Kim 
Signed-off-by: Frederic Weisbecker

perf/events: Add flag to produce nsec output

2012-04-25T11:35:18+00:00

libtraceevent library prints out in usecs but perf wants to print out
in nsecs. Add a flag that lets the user decide to print out in usec
or nsec times.

Signed-off-by: Steven Rostedt 
Cc: Ingo Molnar 
Cc: Thomas Gleixner 
Cc: Peter Zijlstra 
Cc: Arnaldo Carvalho de Melo 
Cc: Steven Rostedt 
Cc: Borislav Petkov 
Cc: Jiri Olsa 
Cc: Arun Sharma 
Cc: Namhyung Kim 
Signed-off-by: Frederic Weisbecker

perf: Have perf use the new libtraceevent.a library

2012-04-25T11:28:48+00:00

The event parsing code in perf was originally copied from trace-cmd
but never was kept up-to-date with the changes that was done there.
The trace-cmd libtraceevent.a code is much more mature than what is
currently in perf.

This updates the code to use wrappers to handle the calls to the
new event parsing code. The new code requires a handle to be pass
around, which removes the global event variables and allows
more than one event structure to be read from different files
(and different machines).

But perf still has the old global events and the code throughout
perf does not yet have a nice way to pass around a handle.
A global 'pevent' has been made for perf and the old calls have
been created as wrappers to the new event parsing code that uses
the global pevent.

With this change, perf can later incorporate the pevent handle into
the perf structures and allow more than one file to be read and
compared, that contains different events.

Signed-off-by: Steven Rostedt 
Cc: Ingo Molnar 
Cc: Thomas Gleixner 
Cc: Peter Zijlstra 
Cc: Arnaldo Carvalho de Melo 
Cc: Steven Rostedt 
Cc: Borislav Petkov 
Cc: Jiri Olsa 
Cc: Arun Sharma 
Cc: Namhyung Kim 
Signed-off-by: Frederic Weisbecker

perf: Separate out trace-cmd parse-events from perf files

2012-04-25T10:28:18+00:00

Move the trace-event-parse.c code that originally came from trace-cmd into
their own files. The new file will be called trace-parse-events.c, as
the name of trace-cmd's file was parse-events.c too, but it conflicted
with the parse-events.c file in perf that parses the command line.

This tries to update the code with mimimal changes.

Perf specific code stays in the trace-event-parse.[ch] files and
the common parsing code is now in trace-parse-events.c and
trace-parse-events.h.

Signed-off-by: Steven Rostedt 
Cc: Ingo Molnar 
Cc: Thomas Gleixner 
Cc: Peter Zijlstra 
Cc: Arnaldo Carvalho de Melo 
Cc: Steven Rostedt 
Cc: Borislav Petkov 
Cc: Jiri Olsa 
Cc: Arun Sharma 
Cc: Namhyung Kim 
Signed-off-by: Frederic Weisbecker

perf tools: Simplify event_read_id exit path

2012-03-22T18:10:42+00:00

We're freeing the token in any case so simplify the exit path by
unifying it.

No functional change.

Signed-off-by: Borislav Petkov 
Link: http://lkml.kernel.org/r/1332339347-21342-1-git-send-email-bp@amd64.org
Signed-off-by: Arnaldo Carvalho de Melo

perf tools: Allow expressions in __print_symbolic() fields

2012-02-17T18:34:08+00:00

The __print_symbolic() function takes a sequence of key-value pairs for
pretty-printing a constant.  The new kvm:kvm_exit print fmt uses the
expression:

  __print_symbolic(..., { 0x040 + 1, "DB excp" }, ...)

Currently only atoms are supported and this print fmt fails to parse.
This patch adds support for expressions instead of just atoms so that
0x040 + 1 is parsed successfully.  Also add arg_num_eval() support for
the '+' operator.

Acked-by: Steven Rostedt 
Cc: Ingo Molnar 
Cc: Steven Rostedt 
Link: http://lkml.kernel.org/r/1315148939-14313-1-git-send-email-stefanha@linux.vnet.ibm.com
Signed-off-by: Stefan Hajnoczi 
Signed-off-by: Arnaldo Carvalho de Melo

perf tools: Remove unnecessary ctype.h inclusion

2012-01-30T20:37:35+00:00

There are unnecessary #include  out there, and they might cause
a nasty build failure in some environment. As we already have most of
ctype macros in util.h, just get rid of them.

A few of exceptions are util/symbol.c which needs isupper() macro util.h
doesn't provide and perl scripting support code which includes ctype.h
internally.

Suggested-by: Ingo Molnar 
Cc: Ingo Molnar 
Cc: Paul Mackerras 
Cc: Peter Zijlstra 
Link: http://lkml.kernel.org/r/1327827356-8786-4-git-send-email-namhyung@gmail.com
Signed-off-by: Namhyung Kim 
Signed-off-by: Arnaldo Carvalho de Melo

perf tools: Fix broken build by defining _GNU_SOURCE in Makefile

2012-01-24T22:26:33+00:00

When building on my Debian/mips system, util/util.c fails to build
because commit 1aed2671738785e8f5aea663a6fda91aa7ef59b5 (perf kvm: Do
guest-only counting by default) indirectly includes stdio.h before the
feature selection in util.h is done.  This prevents _GNU_SOURCE in
util.h from enabling the declaration of getline(), from now second
inclusion of stdio.h, and the build is broken.

There is another breakage in util/evsel.c caused by include ordering,
but I didn't fully track down the commit that caused it.

The root cause of all this is an inconsistent definition of _GNU_SOURCE,
so I move the definition into the Makefile so that it is passed to all
invocations of the compiler and used uniformly for all system header
files.  All other #define and #undef of _GNU_SOURCE are removed as they
cause conflicts with the definition passed to the compiler.

All the features.h definitions (_LARGEFILE64_SOURCE _FILE_OFFSET_BITS=64
and _GNU_SOURCE) are needed by the python glue code too, so they are
moved to BASIC_CFLAGS, and the misleading comments about BASIC_CFLAGS
are removed.

This gives me a clean build on x86_64 (fc12) and mips (Debian).

Cc: David Daney 
Cc: Ingo Molnar 
Cc: Joerg Roedel 
Cc: Namhyung Kim 
Cc: Paul Mackerras 
Cc: Peter Zijlstra 
Link: http://lkml.kernel.org/r/1326836461-11952-1-git-send-email-ddaney.cavm@gmail.com
Signed-off-by: David Daney 
Signed-off-by: Arnaldo Carvalho de Melo

perf: Fix parsing of __print_flags() in TP_printk()

2011-11-07T16:02:35+00:00

A update is made to the sched:sched_switch event that adds some
logic to the first parameter of the __print_flags() that shows the
state of tasks. This change cause perf to fail parsing the flags.

A simple fix is needed to have the parser be able to process ops
within the argument.

Cc: stable@vger.kernel.org
Reported-by: Andrew Vagin 
Signed-off-by: Steven Rostedt