Donate to e Foundation | Murena handsets with /e/OS | Own a part of Murena! Learn more

Commit ced30bc9 authored by Ingo Molnar's avatar Ingo Molnar
Browse files

Merge tag 'perf-core-for-mingo-20160310' of... 

Merge tag 'perf-core-for-mingo-20160310' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux

 into perf/core

Pull perf/core improvements and fixes from Arnaldo Carvalho de Melo:

User visible changes:

  - Implement 'perf stat --metric-only' (Andi Kleen)

  - Fix perf script python database export crash (Chris Phlipot)

Infrastructure changes:

  - perf top/report --hierarchy assorted fixes for problems introduced in this
    perf/core cycle (Namhyung Kim)

  - Support '~' operation in libtraceevent (Steven Rosted)

Build fixes:

  - Fix bulding of jitdump on opensuse on ubuntu systems when the DWARF
    devel files are not installed (Arnaldo Carvalho de Melo)

  - Do not try building jitdump on unsupported arches (Jiri Olsa)

Signed-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: default avatarIngo Molnar <mingo@kernel.org>
parents 3a99e6db 206cab65

tools/lib/traceevent/event-parse.c

+6 −0
Original line number Diff line number Diff line
@@ -2398,6 +2398,12 @@ static int arg_num_eval(struct print_arg *arg, long long *val) 
				break;

			*val = left + right;

			break;

		case '~':

			ret = arg_num_eval(arg->op.right, &right);

			if (!ret)

				break;

			*val = ~right;

			break;

		default:

			do_warning("unknown op '%s'", arg->op.op);

			ret = 0;

tools/perf/Documentation/perf-stat.txt

+27 −0
Original line number Diff line number Diff line
@@ -147,6 +147,10 @@ Print count deltas every N milliseconds (minimum: 10ms) 
The overhead percentage could be high in some cases, for instance with small, sub 100ms intervals.  Use with caution.

	example: 'perf stat -I 1000 -e cycles -a sleep 5'



--metric-only::

Only print computed metrics. Print them in a single line.

Don't show any raw values. Not supported with --per-thread.



--per-socket::

Aggregate counts per processor socket for system-wide mode measurements.  This

is a useful mode to detect imbalance between sockets.  To enable this mode,
@@ -219,6 +223,29 @@ $ perf stat -- make -j 


 Wall-clock time elapsed:   719.554352 msecs



CSV FORMAT

----------



With -x, perf stat is able to output a not-quite-CSV format output

Commas in the output are not put into "". To make it easy to parse

it is recommended to use a different character like -x \;



The fields are in this order:



	- optional usec time stamp in fractions of second (with -I xxx)

	- optional CPU, core, or socket identifier

	- optional number of logical CPUs aggregated

	- counter value

	- unit of the counter value or empty

	- event name

	- run time of counter

	- percentage of measurement time the counter was running

	- optional variance if multiple values are collected with -r

	- optional metric value

	- optional unit of metric



Additional metrics may be printed with all earlier fields being empty.



SEE ALSO

--------

linkperf:perf-top[1], linkperf:perf-list[1]

tools/perf/arch/arm/Makefile

+1 −0
Original line number Diff line number Diff line 
ifndef NO_DWARF

PERF_HAVE_DWARF_REGS := 1

endif

PERF_HAVE_JITDUMP := 1

tools/perf/arch/arm64/Makefile

+1 −0
Original line number Diff line number Diff line 
ifndef NO_DWARF

PERF_HAVE_DWARF_REGS := 1

endif

PERF_HAVE_JITDUMP := 1

tools/perf/arch/powerpc/Makefile

+1 −0
Original line number Diff line number Diff line
@@ -3,3 +3,4 @@ PERF_HAVE_DWARF_REGS := 1 
endif



HAVE_KVM_STAT_SUPPORT := 1

PERF_HAVE_JITDUMP := 1