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

Commit 3eb9ede2 authored by Ingo Molnar's avatar Ingo Molnar
Browse files

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

Merge tag 'perf-core-for-mingo' 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:

New features:

- Allow using trace events fields as sort order keys, making 'perf evlist --trace_fields'
  show those, and then the user can select a subset and use like:

    perf top -e sched:sched_switch -s prev_comm,next_comm

  That works as well in 'perf report' when handling files containing
  tracepoints.

  The default when just tracepoint events are found in a perf.data file is to
  format it like ftrace, using the libtraceevent formatters, plugins, etc (Namhyung Kim)

- Add support in 'perf script' to process 'perf stat record' generated files,
  culminating in a python perf script that calculates CPI (Cycles per
  Instruction) (Jiri Olsa)

- Show random perf tool tips in the 'perf report' bottom line (Namhyung Kim)

- perf report now defaults to --group if the perf.data file has grouped events, try it with:

  # perf record -e '{cycles,instructions}' -a sleep 1
  [ perf record: Woken up 1 times to write data ]
  [ perf record: Captured and wrote 1.093 MB perf.data (1247 samples) ]
  # perf report
  # Samples: 1K of event 'anon group { cycles, instructions }'
  # Event count (approx.): 1955219195
  #
  #       Overhead  Command     Shared Object      Symbol

     2.86%   0.22%  swapper     [kernel.kallsyms]  [k] intel_idle
     1.05%   0.33%  firefox     libxul.so          [.] js::SetObjectElement
     1.05%   0.00%  kworker/0:3 [kernel.kallsyms]  [k] gen6_ring_get_seqno
     0.88%   0.17%  chrome      chrome             [.] 0x0000000000ee27ab
     0.65%   0.86%  firefox     libxul.so          [.] js::ValueToId<(js::AllowGC)1>
     0.64%   0.23%  JS Helper   libxul.so          [.] js::SplayTree<js::jit::LiveRange*, js::jit::LiveRange>::splay
     0.62%   1.27%  firefox     libxul.so          [.] js::GetIterator
     0.61%   1.74%  firefox     libxul.so          [.] js::NativeSetProperty
     0.61%   0.31%  firefox     libxul.so          [.] js::SetPropertyByDefining

User visible fixes:

- Coect data mmaps so that the DWARF unwinder can handle usecases needing them,
  like softice (Jiri Olsa)

- Decay callchains in fractal mode, fixing up cases where 'perf top -g' would
  show entries with more than 100% (Namhyung Kim)

Infrastructure changes:

- Sync tools/lib with the lib/ in the kernel sources for find_bit.c and
  move bitmap.[ch] from tools/perf/util/ to tools/lib/ (Arnaldo Carvalho de Melo)

- No need to set attr.sample_freq in some 'perf test' entries that only
  want to deal with PERF_RECORD_ meta-events, improve a bit error output
  for CQM test (Arnaldo Carvalho de Melo)

- Fix python binding build, adding some missing object files now required
  due to cpumap using find_bit stuff (Arnaldo Carvalho de Melo)

- tools/build improvemnts (Jiri Olsa)

- Add more files to cscope/ctags databases (Jiri Olsa)

- Do not show 'trace' in 'perf help' if it is not compiled in (Jiri Olsa)

- Make perf_evlist__open() open evsels with their cpus and threads,
  like perf record does, making them consistent (Adrian Hunter)

- Fix pmu snapshot initialization bug (Stephane Eranian)

- Add missing headers in perf's MANIFEST (Wang Nan)

Signed-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: default avatarIngo Molnar <mingo@kernel.org>
parents 9cc2617d 775d8a1b

tools/build/Makefile.feature

+27 −14
Original line number Diff line number Diff line
@@ -122,13 +122,31 @@ define feature_print_text_code 
    MSG = $(shell printf '...%30s: %s' $(1) $(2))

endef



#

# generates feature value assignment for name, like:

#   $(call feature_assign,dwarf) == feature-dwarf=1

#

feature_assign = feature-$(1)=$(feature-$(1))



FEATURE_DUMP_FILENAME = $(OUTPUT)FEATURE-DUMP$(FEATURE_USER)

FEATURE_DUMP := $(foreach feat,$(FEATURE_DISPLAY),feature-$(feat)($(feature-$(feat))))

FEATURE_DUMP_FILE := $(shell touch $(FEATURE_DUMP_FILENAME); cat $(FEATURE_DUMP_FILENAME))

FEATURE_DUMP := $(shell touch $(FEATURE_DUMP_FILENAME); cat $(FEATURE_DUMP_FILENAME))



ifeq ($(dwarf-post-unwind),1)

  FEATURE_DUMP += dwarf-post-unwind($(dwarf-post-unwind-text))

feature_dump_check = $(eval $(feature_dump_check_code))

define feature_dump_check_code

  ifeq ($(findstring $(1),$(FEATURE_DUMP)),)

    $(2) := 1

  endif

endef



#

# First check if any test from FEATURE_DISPLAY

# and set feature_display := 1 if it does

$(foreach feat,$(FEATURE_DISPLAY),$(call feature_dump_check,$(call feature_assign,$(feat)),feature_display))



#

# Now also check if any other test changed,

# so we force FEATURE-DUMP generation

$(foreach feat,$(FEATURE_TESTS),$(call feature_dump_check,$(call feature_assign,$(feat)),feature_dump_changed))



# The $(feature_display) controls the default detection message

# output. It's set if:
@@ -137,13 +155,13 @@ endif 
# - one of the $(FEATURE_DISPLAY) is not detected

# - VF is enabled



ifneq ("$(FEATURE_DUMP)","$(FEATURE_DUMP_FILE)")

  $(shell echo "$(FEATURE_DUMP)" > $(FEATURE_DUMP_FILENAME))

  feature_display := 1

ifeq ($(feature_dump_changed),1)

  $(shell rm -f $(FEATURE_DUMP_FILENAME))

  $(foreach feat,$(FEATURE_TESTS),$(shell echo "$(call feature_assign,$(feat))" >> $(FEATURE_DUMP_FILENAME)))

endif



feature_display_check = $(eval $(feature_check_display_code))

define feature_display_check_code

define feature_check_display_code

  ifneq ($(feature-$(1)), 1)

    feature_display := 1

  endif
@@ -160,11 +178,6 @@ ifeq ($(feature_display),1) 
  $(info )

  $(info Auto-detecting system features:)

  $(foreach feat,$(FEATURE_DISPLAY),$(call feature_print_status,$(feat),))



  ifeq ($(dwarf-post-unwind),1)

    $(call feature_print_text,"DWARF post unwind library", $(dwarf-post-unwind-text))

  endif



  ifneq ($(feature_verbose),1)

    $(info )

  endif

tools/perf/util/include/linux/bitmap.h→tools/include/linux/bitmap.h

+2 −0
Original line number Diff line number Diff line
@@ -11,6 +11,8 @@ int __bitmap_weight(const unsigned long *bitmap, int bits); 
void __bitmap_or(unsigned long *dst, const unsigned long *bitmap1,

		 const unsigned long *bitmap2, int bits);



#define BITMAP_FIRST_WORD_MASK(start) (~0UL << ((start) & (BITS_PER_LONG - 1)))



#define BITMAP_LAST_WORD_MASK(nbits)					\

(									\

	((nbits) % BITS_PER_LONG) ?					\

tools/perf/util/bitmap.c→tools/lib/bitmap.c

+0 −0

File moved.

View file

tools/lib/bpf/Makefile

+4 −0
Original line number Diff line number Diff line
@@ -80,7 +80,11 @@ endif 
endif



ifeq ($(check_feat),1)

ifeq ($(FEATURES_DUMP),)

include $(srctree)/tools/build/Makefile.feature

else

include $(FEATURES_DUMP)

endif

endif



export prefix libdir src obj

tools/lib/util/find_next_bit.c→tools/lib/find_bit.c

+84 −0
Original line number Diff line number Diff line 
/* find_next_bit.c: fallback find next bit implementation

/* bit search implementation

 *

 * Copied from lib/find_next_bit.c to tools/lib/next_bit.c

 * Copied from lib/find_bit.c to tools/lib/find_bit.c

 *

 * Copyright (C) 2004 Red Hat, Inc. All Rights Reserved.

 * Written by David Howells (dhowells@redhat.com)

 *

 * Copyright (C) 2008 IBM Corporation

 * 'find_last_bit' is written by Rusty Russell <rusty@rustcorp.com.au>

 * (Inspired by David Howell's find_next_bit implementation)

 *

 * Rewritten by Yury Norov <yury.norov@gmail.com> to decrease

 * size and improve performance, 2015.

 *

 * This program is free software; you can redistribute it and/or

 * modify it under the terms of the GNU General Public License

 * as published by the Free Software Foundation; either version
@@ -12,52 +19,50 @@ 
 */



#include <linux/bitops.h>

#include <asm/types.h>

#include <asm/byteorder.h>

#include <linux/bitmap.h>

#include <linux/kernel.h>



#define BITOP_WORD(nr)		((nr) / BITS_PER_LONG)

#if !defined(find_next_bit)



#ifndef find_next_bit

/*

 * Find the next set bit in a memory region.

 * This is a common helper function for find_next_bit and

 * find_next_zero_bit.  The difference is the "invert" argument, which

 * is XORed with each fetched word before searching it for one bits.

 */

unsigned long find_next_bit(const unsigned long *addr, unsigned long size,

			    unsigned long offset)

static unsigned long _find_next_bit(const unsigned long *addr,

		unsigned long nbits, unsigned long start, unsigned long invert)

{

	const unsigned long *p = addr + BITOP_WORD(offset);

	unsigned long result = offset & ~(BITS_PER_LONG-1);

	unsigned long tmp;



	if (offset >= size)

		return size;

	size -= result;

	offset %= BITS_PER_LONG;

	if (offset) {

		tmp = *(p++);

		tmp &= (~0UL << offset);

		if (size < BITS_PER_LONG)

			goto found_first;

		if (tmp)

			goto found_middle;

		size -= BITS_PER_LONG;

		result += BITS_PER_LONG;

	if (!nbits || start >= nbits)

		return nbits;



	tmp = addr[start / BITS_PER_LONG] ^ invert;



	/* Handle 1st word. */

	tmp &= BITMAP_FIRST_WORD_MASK(start);

	start = round_down(start, BITS_PER_LONG);



	while (!tmp) {

		start += BITS_PER_LONG;

		if (start >= nbits)

			return nbits;



		tmp = addr[start / BITS_PER_LONG] ^ invert;

	}

	while (size & ~(BITS_PER_LONG-1)) {

		if ((tmp = *(p++)))

			goto found_middle;

		result += BITS_PER_LONG;

		size -= BITS_PER_LONG;



	return min(start + __ffs(tmp), nbits);

}

	if (!size)

		return result;

	tmp = *p;

#endif



found_first:

	tmp &= (~0UL >> (BITS_PER_LONG - size));

	if (tmp == 0UL)		/* Are any bits set? */

		return result + size;	/* Nope. */

found_middle:

	return result + __ffs(tmp);

#ifndef find_next_bit

/*

 * Find the next set bit in a memory region.

 */

unsigned long find_next_bit(const unsigned long *addr, unsigned long size,

			    unsigned long offset)

{

	return _find_next_bit(addr, size, offset, 0UL);

}

#endif
@@ -67,23 +72,13 @@ unsigned long find_next_bit(const unsigned long *addr, unsigned long size, 
 */

unsigned long find_first_bit(const unsigned long *addr, unsigned long size)

{

	const unsigned long *p = addr;

	unsigned long result = 0;

	unsigned long tmp;

	unsigned long idx;



	while (size & ~(BITS_PER_LONG-1)) {

		if ((tmp = *(p++)))

			goto found;

		result += BITS_PER_LONG;

		size -= BITS_PER_LONG;

	for (idx = 0; idx * BITS_PER_LONG < size; idx++) {

		if (addr[idx])

			return min(idx * BITS_PER_LONG + __ffs(addr[idx]), size);

	}

	if (!size)

		return result;



	tmp = (*p) & (~0UL >> (BITS_PER_LONG - size));

	if (tmp == 0UL)		/* Are any bits set? */

		return result + size;	/* Nope. */

found:

	return result + __ffs(tmp);

	return size;

}

#endif