[GIT PULL 00/18] perf/core improvements and fixes

[GIT PULL 00/18] perf/core improvements and fixes

[Arnaldo Carvalho de Melo]

Hi Ingo,

	Please consider pulling,

- Arnaldo

The following changes since commit 41e950c033b7df997d4b38653efe6554be9b96a7:

  Merge tag 'perf-core-for-mingo' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/urgent (2014-12-12 09:09:52 +0100)

are available in the git repository at:


  git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux.git tags/perf-core-for-mingo

for you to fetch changes up to 67195c75a87232f055ff415fc4624ef01f24fc3d:

  perf symbols: Fix use after free in filename__read_build_id (2014-12-16 13:38:28 -0300)

----------------------------------------------------------------
perf/core improvements and fixes:

User visible:

- The mmap address range for the ring buffer now is calculated using the
  contents of /proc/sys/kernel/perf_event_mlock_kb.

  This fixes an -EPERM case where 'trace' was trying to use more than what
  configured on perf_event_mlock_kb. (Arnaldo Carvalho de Melo)

Infrastructure:

- Move bitops definitions so that they match the header file hierarchy
  in the kernel sources where that code came from. (Arnaldo Carvalho de Melo)

- Adopt round{down,up}_pow_of_two from the kernel and use it instead of
  equivalent code, so that we reuse more kernel code and make tools/ look
  more like kernel source code, to encourage further contributions from
  kernel hackers (Arnaldo Carvalho de Melo)

- Fix use after free in filename__read_build_id (Mitchell Krome)

Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

----------------------------------------------------------------
Arnaldo Carvalho de Melo (17):
      perf evlist: Fixup brown paper bag on "hint" for --mmap-pages cmdline arg
      perf evlist: Clarify sterror_mmap variable names
      perf evlist: Improve the strerror_mmap method
      perf trace: Let the perf_evlist__mmap autosize the number of pages to use
      perf evlist: Do not use hard coded value for a mmap_pages default
      tools: Move __ffs implementation to tools/include/asm-generic/bitops/__ffs.h
      tools: Move code originally from linux/log2.h to tools/include/linux/
      tools: Move code originally from asm-generic/atomic.h into tools/include/asm-generic/
      tools: Whitespace prep patches for moving bitops.h
      tools lib: Move asm-generic/bitops/find.h code to tools/include and tools/lib
      tools: Introduce asm-generic/bitops.h
      tools: Move bitops.h from tools/perf/util to tools/
      tools: Adopt fls_long and deps
      tools: Adopt rounddown_pow_of_two and deps
      perf tools: Make the mmap length autotuning more robust
      tools: Adopt roundup_pow_of_two
      perf evlist: Use roundup_pow_of_two

Mitchell Krome (1):
      perf symbols: Fix use after free in filename__read_build_id

 tools/include/asm-generic/bitops.h        |  27 +++++
 tools/include/asm-generic/bitops/__ffs.h  |  43 +++++++
 tools/include/asm-generic/bitops/__fls.h  |   1 +
 tools/include/asm-generic/bitops/atomic.h |  22 ++++
 tools/include/asm-generic/bitops/find.h   |  33 ++++++
 tools/include/asm-generic/bitops/fls.h    |   1 +
 tools/include/asm-generic/bitops/fls64.h  |   1 +
 tools/include/linux/bitops.h              |  53 +++++++++
 tools/include/linux/log2.h                | 185 ++++++++++++++++++++++++++++++
 tools/lib/util/find_next_bit.c            |  89 ++++++++++++++
 tools/perf/MANIFEST                       |  13 ++-
 tools/perf/Makefile.perf                  |  14 ++-
 tools/perf/builtin-trace.c                |   2 +-
 tools/perf/util/evlist.c                  |  46 ++++++--
 tools/perf/util/include/linux/bitops.h    | 162 --------------------------
 tools/perf/util/symbol-minimal.c          |   8 +-
 tools/perf/util/util.h                    |  29 -----
 17 files changed, 522 insertions(+), 207 deletions(-)
 create mode 100644 tools/include/asm-generic/bitops.h
 create mode 100644 tools/include/asm-generic/bitops/__ffs.h
 create mode 100644 tools/include/asm-generic/bitops/__fls.h
 create mode 100644 tools/include/asm-generic/bitops/atomic.h
 create mode 100644 tools/include/asm-generic/bitops/find.h
 create mode 100644 tools/include/asm-generic/bitops/fls.h
 create mode 100644 tools/include/asm-generic/bitops/fls64.h
 create mode 100644 tools/include/linux/bitops.h
 create mode 100644 tools/include/linux/log2.h
 create mode 100644 tools/lib/util/find_next_bit.c
 delete mode 100644 tools/perf/util/include/linux/bitops.h

[PATCH 01/18] perf evlist: Fixup brown paper bag on "hint" for --mmap-pages cmdline arg

[Arnaldo Carvalho de Melo]

From: Arnaldo Carvalho de Melo <acme@redhat.com>

When failing due to asking for a number of mmap pages that is more than
the max, it was suggesting that an even bigger number of mmap pages
should be specified, doh, au contraire!

Before:

  [acme@ssdandy linux]$ trace -m 128 usleep 1
  Error:	Operation not permitted.
  Hint:	Check /proc/sys/kernel/perf_event_mlock_kb (516 kB) setting.
  Hint:	Tried using 516 kB.
  Hint:	Try using a bigger -m/--mmap-pages value.
  [acme@ssdandy linux]$

After:

  [acme@ssdandy linux]$ trace -m 128 usleep 1
  Error:	Operation not permitted.
  Hint:	Check /proc/sys/kernel/perf_event_mlock_kb (516 kB) setting.
  Hint:	Tried using 516 kB.
  Hint:	Try using a smaller -m/--mmap-pages value.
  [acme@ssdandy linux]$

And to (really) clarify what happens above, when what the user requests
is <= max and even then it fails, a changeset is being made to tell that
this is a per user limit, not per process (in the above example there
was another 'perf trace' running for this user, which was using all the
pages it could use).

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: David Ahern <dsahern@gmail.com>
Cc: Don Zickus <dzickus@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@kernel.org>
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-8qope8lxb898narnq5kmu2gf@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/evlist.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c
index bb5dfc5d1e75..de7515dd683a 100644
--- a/tools/perf/util/evlist.c
+++ b/tools/perf/util/evlist.c
@@ -1495,7 +1495,7 @@ int perf_evlist__strerror_mmap(struct perf_evlist *evlist, int err, char *buf, s
 		scnprintf(buf, size, "Error:\t%s.\n"
 				     "Hint:\tCheck /proc/sys/kernel/perf_event_mlock_kb (%d kB) setting.\n"
 				     "Hint:\tTried using %zd kB.\n"
-				     "Hint:\tTry using a bigger -m/--mmap-pages value.",
+				     "Hint:\tTry using a smaller -m/--mmap-pages value.",
 				     emsg, value, evlist->mmap_len / 1024);
 		break;
 	default:
-- 
1.9.3

[PATCH 02/18] perf evlist: Clarify sterror_mmap variable names

[Arnaldo Carvalho de Melo]

From: Arnaldo Carvalho de Melo <acme@redhat.com>

Prep patch for doing further checks like when the number of pages that
is being attempted is actually below /proc/sys/kernel/perf_event_mlock_kb but
the operation fails because the user doesn't have CAP_IPC_LOCK.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: David Ahern <dsahern@gmail.com>
Cc: Don Zickus <dzickus@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@kernel.org>
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-wetzlux7mzvofu5cuji5i71i@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/evlist.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c
index de7515dd683a..d661f252b891 100644
--- a/tools/perf/util/evlist.c
+++ b/tools/perf/util/evlist.c
@@ -1487,16 +1487,16 @@ int perf_evlist__strerror_open(struct perf_evlist *evlist __maybe_unused,
 int perf_evlist__strerror_mmap(struct perf_evlist *evlist, int err, char *buf, size_t size)
 {
 	char sbuf[STRERR_BUFSIZE], *emsg = strerror_r(err, sbuf, sizeof(sbuf));
-	int value;
+	int pages_attempted = evlist->mmap_len / 1024, pages_max_per_user;
 
 	switch (err) {
 	case EPERM:
-		sysctl__read_int("kernel/perf_event_mlock_kb", &value);
+		sysctl__read_int("kernel/perf_event_mlock_kb", &pages_max_per_user);
 		scnprintf(buf, size, "Error:\t%s.\n"
 				     "Hint:\tCheck /proc/sys/kernel/perf_event_mlock_kb (%d kB) setting.\n"
 				     "Hint:\tTried using %zd kB.\n"
 				     "Hint:\tTry using a smaller -m/--mmap-pages value.",
-				     emsg, value, evlist->mmap_len / 1024);
+				     emsg, pages_max_per_user, pages_attempted);
 		break;
 	default:
 		scnprintf(buf, size, "%s", emsg);
-- 
1.9.3

[PATCH 03/18] perf evlist: Improve the strerror_mmap method

[Arnaldo Carvalho de Melo]

From: Arnaldo Carvalho de Melo <acme@redhat.com>

Considering the per user locked pages limit, improve the message when a
user uses multiple simultaneous perf mmap calls:

When the request is more than the current maximum:

  [acme@ssdandy linux]$ trace -m 128 usleep 1
  Error: Operation not permitted.
  Hint:  Check /proc/sys/kernel/perf_event_mlock_kb (516 kB) setting.
  Hint:  Tried using 516 kB.
  Hint:  Try 'sudo sh -c "echo 1032 > /proc/sys/kernel/perf_event_mlock_kb"', or
  Hint:  Try using a smaller -m/--mmap-pages value.
  [acme@ssdandy linux]$

And when the limit is less than that:

  [acme@ssdandy linux]$ trace -m 512 usleep 1
  Error: Operation not permitted.
  Hint:  Check /proc/sys/kernel/perf_event_mlock_kb (2056 kB) setting.
  Hint:  Tried using 2052 kB.
  Hint:  Try using a smaller -m/--mmap-pages value.
  [acme@ssdandy linux]$

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: David Ahern <dsahern@gmail.com>
Cc: Don Zickus <dzickus@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@kernel.org>
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-yqdie3c8qvdgenwleri267d4@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/evlist.c | 17 +++++++++++++----
 1 file changed, 13 insertions(+), 4 deletions(-)

diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c
index d661f252b891..7847f3885081 100644
--- a/tools/perf/util/evlist.c
+++ b/tools/perf/util/evlist.c
@@ -1487,16 +1487,25 @@ int perf_evlist__strerror_open(struct perf_evlist *evlist __maybe_unused,
 int perf_evlist__strerror_mmap(struct perf_evlist *evlist, int err, char *buf, size_t size)
 {
 	char sbuf[STRERR_BUFSIZE], *emsg = strerror_r(err, sbuf, sizeof(sbuf));
-	int pages_attempted = evlist->mmap_len / 1024, pages_max_per_user;
+	int pages_attempted = evlist->mmap_len / 1024, pages_max_per_user, printed = 0;
 
 	switch (err) {
 	case EPERM:
 		sysctl__read_int("kernel/perf_event_mlock_kb", &pages_max_per_user);
-		scnprintf(buf, size, "Error:\t%s.\n"
+		printed += scnprintf(buf + printed, size - printed,
+				     "Error:\t%s.\n"
 				     "Hint:\tCheck /proc/sys/kernel/perf_event_mlock_kb (%d kB) setting.\n"
-				     "Hint:\tTried using %zd kB.\n"
-				     "Hint:\tTry using a smaller -m/--mmap-pages value.",
+				     "Hint:\tTried using %zd kB.\n",
 				     emsg, pages_max_per_user, pages_attempted);
+
+		if (pages_attempted >= pages_max_per_user) {
+			printed += scnprintf(buf + printed, size - printed,
+					     "Hint:\tTry 'sudo sh -c \"echo %d > /proc/sys/kernel/perf_event_mlock_kb\"', or\n",
+					     pages_max_per_user + pages_attempted);
+		}
+
+		printed += scnprintf(buf + printed, size - printed,
+				     "Hint:\tTry using a smaller -m/--mmap-pages value.");
 		break;
 	default:
 		scnprintf(buf, size, "%s", emsg);
-- 
1.9.3

[PATCH 04/18] perf trace: Let the perf_evlist__mmap autosize the number of pages to use

[Arnaldo Carvalho de Melo]

From: Arnaldo Carvalho de Melo <acme@redhat.com>

So that normal users can run 'trace', we were using a hardcoded 1024
pages value that was more than the default /proc/sys/kernel/perf_event_mlock_kb
setting.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: David Ahern <dsahern@gmail.com>
Cc: Don Zickus <dzickus@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@kernel.org>
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-3banj3yh0sjz41obxtgiel3a@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-trace.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c
index 327541e43c7f..badfabc6a01f 100644
--- a/tools/perf/builtin-trace.c
+++ b/tools/perf/builtin-trace.c
@@ -2485,7 +2485,7 @@ int cmd_trace(int argc, const char **argv, const char *prefix __maybe_unused)
 			.user_freq     = UINT_MAX,
 			.user_interval = ULLONG_MAX,
 			.no_buffering  = true,
-			.mmap_pages    = 1024,
+			.mmap_pages    = UINT_MAX,
 		},
 		.output = stdout,
 		.show_comm = true,
-- 
1.9.3

[PATCH 05/18] perf evlist: Do not use hard coded value for a mmap_pages default

[Arnaldo Carvalho de Melo]

From: Arnaldo Carvalho de Melo <acme@redhat.com>

So far what is in there by default is what we were using: 512KB + the
control page, but the admin may change that, and if it does to a smaller
value, all calls to tooling for non root users start failing, requiring
that the user manually set --mmap_pages/-m.

Use instead what is in /proc/sys/kernel/perf_event_mlock_kb.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: David Ahern <dsahern@gmail.com>
Cc: Don Zickus <dzickus@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@kernel.org>
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-2f6mtm8xu3wo5lhkql6jdblh@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/evlist.c | 20 ++++++++++++++++----
 1 file changed, 16 insertions(+), 4 deletions(-)

diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c
index 7847f3885081..ac808680e61c 100644
--- a/tools/perf/util/evlist.c
+++ b/tools/perf/util/evlist.c
@@ -893,10 +893,22 @@ out_unmap:
 
 static size_t perf_evlist__mmap_size(unsigned long pages)
 {
-	/* 512 kiB: default amount of unprivileged mlocked memory */
-	if (pages == UINT_MAX)
-		pages = (512 * 1024) / page_size;
-	else if (!is_power_of_2(pages))
+	if (pages == UINT_MAX) {
+		int max;
+
+		if (sysctl__read_int("kernel/perf_event_mlock_kb", &max) < 0) {
+			/*
+			 * Pick a once upon a time good value, i.e. things look
+			 * strange since we can't read a sysctl value, but lets not
+			 * die yet...
+			 */
+			max = 512;
+		} else {
+			max -= (page_size / 1024);
+		}
+
+		pages = (max * 1024) / page_size;
+	} else if (!is_power_of_2(pages))
 		return 0;
 
 	return (pages + 1) * page_size;
-- 
1.9.3

[PATCH 06/18] tools: Move __ffs implementation to tools/include/asm-generic/bitops/__ffs.h

[Arnaldo Carvalho de Melo]

From: Arnaldo Carvalho de Melo <acme@redhat.com>

To match the Linux kernel source code structure from where this code came from.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: David Ahern <dsahern@gmail.com>
Cc: Don Zickus <dzickus@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@kernel.org>
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-gubysnp4a8hd98lxoeruak13@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/include/asm-generic/bitops/__ffs.h | 43 ++++++++++++++++++++++++++++++++
 tools/perf/MANIFEST                      |  1 +
 tools/perf/Makefile.perf                 |  1 +
 tools/perf/util/include/linux/bitops.h   | 37 +--------------------------
 4 files changed, 46 insertions(+), 36 deletions(-)
 create mode 100644 tools/include/asm-generic/bitops/__ffs.h

diff --git a/tools/include/asm-generic/bitops/__ffs.h b/tools/include/asm-generic/bitops/__ffs.h
new file mode 100644
index 000000000000..c94175015a82
--- /dev/null
+++ b/tools/include/asm-generic/bitops/__ffs.h
@@ -0,0 +1,43 @@
+#ifndef _TOOLS_LINUX_ASM_GENERIC_BITOPS___FFS_H_
+#define _TOOLS_LINUX_ASM_GENERIC_BITOPS___FFS_H_
+
+#include <asm/types.h>
+
+/**
+ * __ffs - find first bit in word.
+ * @word: The word to search
+ *
+ * Undefined if no bit exists, so code should check against 0 first.
+ */
+static __always_inline unsigned long __ffs(unsigned long word)
+{
+	int num = 0;
+
+#if __BITS_PER_LONG == 64
+	if ((word & 0xffffffff) == 0) {
+		num += 32;
+		word >>= 32;
+	}
+#endif
+	if ((word & 0xffff) == 0) {
+		num += 16;
+		word >>= 16;
+	}
+	if ((word & 0xff) == 0) {
+		num += 8;
+		word >>= 8;
+	}
+	if ((word & 0xf) == 0) {
+		num += 4;
+		word >>= 4;
+	}
+	if ((word & 0x3) == 0) {
+		num += 2;
+		word >>= 2;
+	}
+	if ((word & 0x1) == 0)
+		num += 1;
+	return num;
+}
+
+#endif /* _TOOLS_LINUX_ASM_GENERIC_BITOPS___FFS_H_ */
diff --git a/tools/perf/MANIFEST b/tools/perf/MANIFEST
index 344c4d3d0a4a..39c08636357b 100644
--- a/tools/perf/MANIFEST
+++ b/tools/perf/MANIFEST
@@ -5,6 +5,7 @@ tools/lib/api
 tools/lib/symbol/kallsyms.c
 tools/lib/symbol/kallsyms.h
 tools/include/asm/bug.h
+tools/include/asm-generic/bitops/__ffs.h
 tools/include/linux/compiler.h
 tools/include/linux/hash.h
 tools/include/linux/export.h
diff --git a/tools/perf/Makefile.perf b/tools/perf/Makefile.perf
index 763e68fb5767..bfd86a32254a 100644
--- a/tools/perf/Makefile.perf
+++ b/tools/perf/Makefile.perf
@@ -232,6 +232,7 @@ LIB_H += ../include/linux/hash.h
 LIB_H += ../../include/linux/stringify.h
 LIB_H += util/include/linux/bitmap.h
 LIB_H += util/include/linux/bitops.h
+LIB_H += ../include/asm-generic/bitops/__ffs.h 
 LIB_H += ../include/linux/compiler.h
 LIB_H += util/include/linux/const.h
 LIB_H += util/include/linux/ctype.h
diff --git a/tools/perf/util/include/linux/bitops.h b/tools/perf/util/include/linux/bitops.h
index c3294163de17..762bb83333c0 100644
--- a/tools/perf/util/include/linux/bitops.h
+++ b/tools/perf/util/include/linux/bitops.h
@@ -52,42 +52,7 @@ static inline unsigned long hweight_long(unsigned long w)
 
 #define BITOP_WORD(nr)		((nr) / BITS_PER_LONG)
 
-/**
- * __ffs - find first bit in word.
- * @word: The word to search
- *
- * Undefined if no bit exists, so code should check against 0 first.
- */
-static __always_inline unsigned long __ffs(unsigned long word)
-{
-	int num = 0;
-
-#if BITS_PER_LONG == 64
-	if ((word & 0xffffffff) == 0) {
-		num += 32;
-		word >>= 32;
-	}
-#endif
-	if ((word & 0xffff) == 0) {
-		num += 16;
-		word >>= 16;
-	}
-	if ((word & 0xff) == 0) {
-		num += 8;
-		word >>= 8;
-	}
-	if ((word & 0xf) == 0) {
-		num += 4;
-		word >>= 4;
-	}
-	if ((word & 0x3) == 0) {
-		num += 2;
-		word >>= 2;
-	}
-	if ((word & 0x1) == 0)
-		num += 1;
-	return num;
-}
+#include <asm-generic/bitops/__ffs.h>
 
 typedef const unsigned long __attribute__((__may_alias__)) long_alias_t;
 
-- 
1.9.3

[PATCH 07/18] tools: Move code originally from linux/log2.h to tools/include/linux/

[Arnaldo Carvalho de Melo]

From: Arnaldo Carvalho de Melo <acme@redhat.com>

>From tools/perf/util/include/linux, so that it becomes accessible to
other tools/.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: David Ahern <dsahern@gmail.com>
Cc: Don Zickus <dzickus@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@kernel.org>
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-uqohgzilp3ebd3cbybnf3luc@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/include/linux/log2.h | 26 ++++++++++++++++++++++++++
 tools/perf/MANIFEST        |  3 ++-
 tools/perf/Makefile.perf   |  1 +
 tools/perf/util/evlist.c   |  1 +
 tools/perf/util/util.h     | 11 -----------
 5 files changed, 30 insertions(+), 12 deletions(-)
 create mode 100644 tools/include/linux/log2.h

diff --git a/tools/include/linux/log2.h b/tools/include/linux/log2.h
new file mode 100644
index 000000000000..141b7665d842
--- /dev/null
+++ b/tools/include/linux/log2.h
@@ -0,0 +1,26 @@
+/* Integer base 2 logarithm calculation
+ *
+ * Copyright (C) 2006 Red Hat, Inc. All Rights Reserved.
+ * Written by David Howells (dhowells@redhat.com)
+ *
+ * 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
+ * 2 of the License, or (at your option) any later version.
+ */
+
+#ifndef _TOOLS_LINUX_LOG2_H
+#define _TOOLS_LINUX_LOG2_H
+
+/*
+ *  Determine whether some value is a power of two, where zero is
+ * *not* considered a power of two.
+ */
+
+static inline __attribute__((const))
+bool is_power_of_2(unsigned long n)
+{
+	return (n != 0 && ((n & (n - 1)) == 0));
+}
+
+#endif /* _TOOLS_LINUX_LOG2_H */
diff --git a/tools/perf/MANIFEST b/tools/perf/MANIFEST
index 39c08636357b..db7827fd2518 100644
--- a/tools/perf/MANIFEST
+++ b/tools/perf/MANIFEST
@@ -7,8 +7,9 @@ tools/lib/symbol/kallsyms.h
 tools/include/asm/bug.h
 tools/include/asm-generic/bitops/__ffs.h
 tools/include/linux/compiler.h
-tools/include/linux/hash.h
 tools/include/linux/export.h
+tools/include/linux/hash.h
+tools/include/linux/log2.h
 tools/include/linux/types.h
 include/linux/const.h
 include/linux/perf_event.h
diff --git a/tools/perf/Makefile.perf b/tools/perf/Makefile.perf
index bfd86a32254a..bd6073e0dec7 100644
--- a/tools/perf/Makefile.perf
+++ b/tools/perf/Makefile.perf
@@ -234,6 +234,7 @@ LIB_H += util/include/linux/bitmap.h
 LIB_H += util/include/linux/bitops.h
 LIB_H += ../include/asm-generic/bitops/__ffs.h 
 LIB_H += ../include/linux/compiler.h
+LIB_H += ../include/linux/log2.h
 LIB_H += util/include/linux/const.h
 LIB_H += util/include/linux/ctype.h
 LIB_H += util/include/linux/kernel.h
diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c
index ac808680e61c..8eb92cf5d552 100644
--- a/tools/perf/util/evlist.c
+++ b/tools/perf/util/evlist.c
@@ -25,6 +25,7 @@
 
 #include <linux/bitops.h>
 #include <linux/hash.h>
+#include <linux/log2.h>
 
 static void perf_evlist__mmap_put(struct perf_evlist *evlist, int idx);
 static void __perf_evlist__munmap(struct perf_evlist *evlist, int idx);
diff --git a/tools/perf/util/util.h b/tools/perf/util/util.h
index 008b361b1758..25b22bbea066 100644
--- a/tools/perf/util/util.h
+++ b/tools/perf/util/util.h
@@ -269,17 +269,6 @@ void event_attr_init(struct perf_event_attr *attr);
 #define _STR(x) #x
 #define STR(x) _STR(x)
 
-/*
- *  Determine whether some value is a power of two, where zero is
- * *not* considered a power of two.
- */
-
-static inline __attribute__((const))
-bool is_power_of_2(unsigned long n)
-{
-	return (n != 0 && ((n & (n - 1)) == 0));
-}
-
 static inline unsigned next_pow2(unsigned x)
 {
 	if (!x)
-- 
1.9.3

[PATCH 08/18] tools: Move code originally from asm-generic/atomic.h into tools/include/asm-generic/

[Arnaldo Carvalho de Melo]

From: Arnaldo Carvalho de Melo <acme@redhat.com>

To match the Linux kernel source code structure from where this code came from.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: David Ahern <dsahern@gmail.com>
Cc: Don Zickus <dzickus@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@kernel.org>
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-1ldjhvioch1uczilno5e1epl@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/include/asm-generic/bitops/atomic.h | 22 ++++++++++++++++++++++
 tools/perf/MANIFEST                       |  1 +
 tools/perf/Makefile.perf                  |  1 +
 tools/perf/util/include/linux/bitops.h    | 16 +---------------
 4 files changed, 25 insertions(+), 15 deletions(-)
 create mode 100644 tools/include/asm-generic/bitops/atomic.h

diff --git a/tools/include/asm-generic/bitops/atomic.h b/tools/include/asm-generic/bitops/atomic.h
new file mode 100644
index 000000000000..4bccd7c3d5d6
--- /dev/null
+++ b/tools/include/asm-generic/bitops/atomic.h
@@ -0,0 +1,22 @@
+#ifndef _TOOLS_LINUX_ASM_GENERIC_BITOPS_ATOMIC_H_
+#define _TOOLS_LINUX_ASM_GENERIC_BITOPS_ATOMIC_H_
+
+#include <asm/types.h>
+
+static inline void set_bit(int nr, unsigned long *addr)
+{
+	addr[nr / __BITS_PER_LONG] |= 1UL << (nr % __BITS_PER_LONG);
+}
+
+static inline void clear_bit(int nr, unsigned long *addr)
+{
+	addr[nr / __BITS_PER_LONG] &= ~(1UL << (nr % __BITS_PER_LONG));
+}
+
+static __always_inline int test_bit(unsigned int nr, const unsigned long *addr)
+{
+	return ((1UL << (nr % __BITS_PER_LONG)) &
+		(((unsigned long *)addr)[nr / __BITS_PER_LONG])) != 0;
+}
+
+#endif /* _TOOLS_LINUX_ASM_GENERIC_BITOPS_ATOMIC_H_ */
diff --git a/tools/perf/MANIFEST b/tools/perf/MANIFEST
index db7827fd2518..67d48e20f7e1 100644
--- a/tools/perf/MANIFEST
+++ b/tools/perf/MANIFEST
@@ -5,6 +5,7 @@ tools/lib/api
 tools/lib/symbol/kallsyms.c
 tools/lib/symbol/kallsyms.h
 tools/include/asm/bug.h
+tools/include/asm-generic/bitops/atomic.h
 tools/include/asm-generic/bitops/__ffs.h
 tools/include/linux/compiler.h
 tools/include/linux/export.h
diff --git a/tools/perf/Makefile.perf b/tools/perf/Makefile.perf
index bd6073e0dec7..fc4082fdf6bf 100644
--- a/tools/perf/Makefile.perf
+++ b/tools/perf/Makefile.perf
@@ -232,6 +232,7 @@ LIB_H += ../include/linux/hash.h
 LIB_H += ../../include/linux/stringify.h
 LIB_H += util/include/linux/bitmap.h
 LIB_H += util/include/linux/bitops.h
+LIB_H += ../include/asm-generic/bitops/atomic.h 
 LIB_H += ../include/asm-generic/bitops/__ffs.h 
 LIB_H += ../include/linux/compiler.h
 LIB_H += ../include/linux/log2.h
diff --git a/tools/perf/util/include/linux/bitops.h b/tools/perf/util/include/linux/bitops.h
index 762bb83333c0..71f017dc6230 100644
--- a/tools/perf/util/include/linux/bitops.h
+++ b/tools/perf/util/include/linux/bitops.h
@@ -29,21 +29,7 @@
 	     (bit) < (size);					\
 	     (bit) = find_next_bit((addr), (size), (bit) + 1))
 
-static inline void set_bit(int nr, unsigned long *addr)
-{
-	addr[nr / BITS_PER_LONG] |= 1UL << (nr % BITS_PER_LONG);
-}
-
-static inline void clear_bit(int nr, unsigned long *addr)
-{
-	addr[nr / BITS_PER_LONG] &= ~(1UL << (nr % BITS_PER_LONG));
-}
-
-static __always_inline int test_bit(unsigned int nr, const unsigned long *addr)
-{
-	return ((1UL << (nr % BITS_PER_LONG)) &
-		(((unsigned long *)addr)[nr / BITS_PER_LONG])) != 0;
-}
+#include <asm-generic/bitops/atomic.h>
 
 static inline unsigned long hweight_long(unsigned long w)
 {
-- 
1.9.3

[PATCH 09/18] tools: Whitespace prep patches for moving bitops.h

[Arnaldo Carvalho de Melo]

From: Arnaldo Carvalho de Melo <acme@redhat.com>

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: David Ahern <dsahern@gmail.com>
Cc: Don Zickus <dzickus@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@kernel.org>
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-6xmwcvgm2rvoayv2mf9n5sf8@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/include/linux/bitops.h | 15 ++++++++-------
 1 file changed, 8 insertions(+), 7 deletions(-)

diff --git a/tools/perf/util/include/linux/bitops.h b/tools/perf/util/include/linux/bitops.h
index 71f017dc6230..58c4a2520ac2 100644
--- a/tools/perf/util/include/linux/bitops.h
+++ b/tools/perf/util/include/linux/bitops.h
@@ -10,13 +10,14 @@
 #endif
 
 #define BITS_PER_LONG __WORDSIZE
-#define BITS_PER_BYTE           8
-#define BITS_TO_LONGS(nr)       DIV_ROUND_UP(nr, BITS_PER_BYTE * sizeof(long))
-#define BITS_TO_U64(nr)         DIV_ROUND_UP(nr, BITS_PER_BYTE * sizeof(u64))
-#define BITS_TO_U32(nr)         DIV_ROUND_UP(nr, BITS_PER_BYTE * sizeof(u32))
-#define BITS_TO_BYTES(nr)       DIV_ROUND_UP(nr, BITS_PER_BYTE)
-#define BIT_WORD(nr)            ((nr) / BITS_PER_LONG)
-#define BIT_MASK(nr)            (1UL << ((nr) % BITS_PER_LONG))
+
+#define BIT_MASK(nr)		(1UL << ((nr) % BITS_PER_LONG))
+#define BIT_WORD(nr)		((nr) / BITS_PER_LONG)
+#define BITS_PER_BYTE		8
+#define BITS_TO_LONGS(nr)	DIV_ROUND_UP(nr, BITS_PER_BYTE * sizeof(long))
+#define BITS_TO_U64(nr)		DIV_ROUND_UP(nr, BITS_PER_BYTE * sizeof(u64))
+#define BITS_TO_U32(nr)		DIV_ROUND_UP(nr, BITS_PER_BYTE * sizeof(u32))
+#define BITS_TO_BYTES(nr)	DIV_ROUND_UP(nr, BITS_PER_BYTE)
 
 #define for_each_set_bit(bit, addr, size) \
 	for ((bit) = find_first_bit((addr), (size));		\
-- 
1.9.3

[PATCH 10/18] tools lib: Move asm-generic/bitops/find.h code to tools/include and tools/lib

[Arnaldo Carvalho de Melo]

From: Arnaldo Carvalho de Melo <acme@redhat.com>

To match the Linux kernel source code structure from where this code came from.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: David Ahern <dsahern@gmail.com>
Cc: Don Zickus <dzickus@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@kernel.org>
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-fkbma1h04ki0zzdmp0dpgfyy@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/include/asm-generic/bitops/find.h | 33 ++++++++++++
 tools/lib/util/find_next_bit.c          | 89 +++++++++++++++++++++++++++++++++
 tools/perf/MANIFEST                     |  3 ++
 tools/perf/Makefile.perf                |  5 ++
 tools/perf/util/include/linux/bitops.h  | 73 +--------------------------
 5 files changed, 131 insertions(+), 72 deletions(-)
 create mode 100644 tools/include/asm-generic/bitops/find.h
 create mode 100644 tools/lib/util/find_next_bit.c

diff --git a/tools/include/asm-generic/bitops/find.h b/tools/include/asm-generic/bitops/find.h
new file mode 100644
index 000000000000..31f51547fcd4
--- /dev/null
+++ b/tools/include/asm-generic/bitops/find.h
@@ -0,0 +1,33 @@
+#ifndef _TOOLS_LINUX_ASM_GENERIC_BITOPS_FIND_H_
+#define _TOOLS_LINUX_ASM_GENERIC_BITOPS_FIND_H_
+
+#ifndef find_next_bit
+/**
+ * find_next_bit - find the next set bit in a memory region
+ * @addr: The address to base the search on
+ * @offset: The bitnumber to start searching at
+ * @size: The bitmap size in bits
+ *
+ * Returns the bit number for the next set bit
+ * If no bits are set, returns @size.
+ */
+extern unsigned long find_next_bit(const unsigned long *addr, unsigned long
+		size, unsigned long offset);
+#endif
+
+#ifndef find_first_bit
+
+/**
+ * find_first_bit - find the first set bit in a memory region
+ * @addr: The address to start the search at
+ * @size: The maximum number of bits to search
+ *
+ * Returns the bit number of the first set bit.
+ * If no bits are set, returns @size.
+ */
+extern unsigned long find_first_bit(const unsigned long *addr,
+				    unsigned long size);
+
+#endif /* find_first_bit */
+
+#endif /*_TOOLS_LINUX_ASM_GENERIC_BITOPS_FIND_H_ */
diff --git a/tools/lib/util/find_next_bit.c b/tools/lib/util/find_next_bit.c
new file mode 100644
index 000000000000..41b44f65a79e
--- /dev/null
+++ b/tools/lib/util/find_next_bit.c
@@ -0,0 +1,89 @@
+/* find_next_bit.c: fallback find next bit implementation
+ *
+ * Copied from lib/find_next_bit.c to tools/lib/next_bit.c
+ *
+ * Copyright (C) 2004 Red Hat, Inc. All Rights Reserved.
+ * Written by David Howells (dhowells@redhat.com)
+ *
+ * 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
+ * 2 of the License, or (at your option) any later version.
+ */
+
+#include <linux/bitops.h>
+#include <asm/types.h>
+#include <asm/byteorder.h>
+
+#define BITOP_WORD(nr)		((nr) / BITS_PER_LONG)
+
+#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)
+{
+	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;
+	}
+	while (size & ~(BITS_PER_LONG-1)) {
+		if ((tmp = *(p++)))
+			goto found_middle;
+		result += BITS_PER_LONG;
+		size -= BITS_PER_LONG;
+	}
+	if (!size)
+		return result;
+	tmp = *p;
+
+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);
+}
+#endif
+
+#ifndef find_first_bit
+/*
+ * Find the first set bit in a memory region.
+ */
+unsigned long find_first_bit(const unsigned long *addr, unsigned long size)
+{
+	const unsigned long *p = addr;
+	unsigned long result = 0;
+	unsigned long tmp;
+
+	while (size & ~(BITS_PER_LONG-1)) {
+		if ((tmp = *(p++)))
+			goto found;
+		result += BITS_PER_LONG;
+		size -= BITS_PER_LONG;
+	}
+	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);
+}
+#endif
diff --git a/tools/perf/MANIFEST b/tools/perf/MANIFEST
index 67d48e20f7e1..bfd7e22bafad 100644
--- a/tools/perf/MANIFEST
+++ b/tools/perf/MANIFEST
@@ -4,8 +4,10 @@ tools/lib/traceevent
 tools/lib/api
 tools/lib/symbol/kallsyms.c
 tools/lib/symbol/kallsyms.h
+tools/lib/util/find_next_bit.c
 tools/include/asm/bug.h
 tools/include/asm-generic/bitops/atomic.h
+tools/include/asm-generic/bitops/find.h
 tools/include/asm-generic/bitops/__ffs.h
 tools/include/linux/compiler.h
 tools/include/linux/export.h
@@ -18,6 +20,7 @@ include/linux/rbtree.h
 include/linux/list.h
 include/linux/hash.h
 include/linux/stringify.h
+lib/find_next_bit.c
 lib/rbtree.c
 include/linux/swab.h
 arch/*/include/asm/unistd*.h
diff --git a/tools/perf/Makefile.perf b/tools/perf/Makefile.perf
index fc4082fdf6bf..2d799da4eacc 100644
--- a/tools/perf/Makefile.perf
+++ b/tools/perf/Makefile.perf
@@ -233,6 +233,7 @@ LIB_H += ../../include/linux/stringify.h
 LIB_H += util/include/linux/bitmap.h
 LIB_H += util/include/linux/bitops.h
 LIB_H += ../include/asm-generic/bitops/atomic.h 
+LIB_H += ../include/asm-generic/bitops/find.h 
 LIB_H += ../include/asm-generic/bitops/__ffs.h 
 LIB_H += ../include/linux/compiler.h
 LIB_H += ../include/linux/log2.h
@@ -338,6 +339,7 @@ LIB_OBJS += $(OUTPUT)util/event.o
 LIB_OBJS += $(OUTPUT)util/evlist.o
 LIB_OBJS += $(OUTPUT)util/evsel.o
 LIB_OBJS += $(OUTPUT)util/exec_cmd.o
+LIB_OBJS += $(OUTPUT)util/find_next_bit.o
 LIB_OBJS += $(OUTPUT)util/help.o
 LIB_OBJS += $(OUTPUT)util/kallsyms.o
 LIB_OBJS += $(OUTPUT)util/levenshtein.o
@@ -737,6 +739,9 @@ $(OUTPUT)util/kallsyms.o: ../lib/symbol/kallsyms.c $(OUTPUT)PERF-CFLAGS
 $(OUTPUT)util/rbtree.o: ../../lib/rbtree.c $(OUTPUT)PERF-CFLAGS
 	$(QUIET_CC)$(CC) -o $@ -c $(CFLAGS) -Wno-unused-parameter -DETC_PERFCONFIG='"$(ETC_PERFCONFIG_SQ)"' $<
 
+$(OUTPUT)util/find_next_bit.o: ../lib/util/find_next_bit.c $(OUTPUT)PERF-CFLAGS
+	$(QUIET_CC)$(CC) -o $@ -c $(CFLAGS) -Wno-unused-parameter -DETC_PERFCONFIG='"$(ETC_PERFCONFIG_SQ)"' $<
+
 $(OUTPUT)util/parse-events.o: util/parse-events.c $(OUTPUT)PERF-CFLAGS
 	$(QUIET_CC)$(CC) -o $@ -c $(CFLAGS) -Wno-redundant-decls $<
 
diff --git a/tools/perf/util/include/linux/bitops.h b/tools/perf/util/include/linux/bitops.h
index 58c4a2520ac2..292aadeb66c5 100644
--- a/tools/perf/util/include/linux/bitops.h
+++ b/tools/perf/util/include/linux/bitops.h
@@ -37,78 +37,7 @@ static inline unsigned long hweight_long(unsigned long w)
 	return sizeof(w) == 4 ? hweight32(w) : hweight64(w);
 }
 
-#define BITOP_WORD(nr)		((nr) / BITS_PER_LONG)
-
 #include <asm-generic/bitops/__ffs.h>
-
-typedef const unsigned long __attribute__((__may_alias__)) long_alias_t;
-
-/*
- * Find the first set bit in a memory region.
- */
-static inline unsigned long
-find_first_bit(const unsigned long *addr, unsigned long size)
-{
-	long_alias_t *p = (long_alias_t *) addr;
-	unsigned long result = 0;
-	unsigned long tmp;
-
-	while (size & ~(BITS_PER_LONG-1)) {
-		if ((tmp = *(p++)))
-			goto found;
-		result += BITS_PER_LONG;
-		size -= BITS_PER_LONG;
-	}
-	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);
-}
-
-/*
- * Find the next set bit in a memory region.
- */
-static inline unsigned long
-find_next_bit(const unsigned long *addr, unsigned long size, unsigned long offset)
-{
-	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;
-	}
-	while (size & ~(BITS_PER_LONG-1)) {
-		if ((tmp = *(p++)))
-			goto found_middle;
-		result += BITS_PER_LONG;
-		size -= BITS_PER_LONG;
-	}
-	if (!size)
-		return result;
-	tmp = *p;
-
-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);
-}
+#include <asm-generic/bitops/find.h>
 
 #endif
-- 
1.9.3

[PATCH 11/18] tools: Introduce asm-generic/bitops.h

[Arnaldo Carvalho de Melo]

From: Arnaldo Carvalho de Melo <acme@redhat.com>

In preparation for moving linux/bitops.h from tools/perf/util/ to
tools/include/.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: David Ahern <dsahern@gmail.com>
Cc: Don Zickus <dzickus@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@kernel.org>
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-2wuk8vahl7voz0ie55f07c9k@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/include/asm-generic/bitops.h     | 24 ++++++++++++++++++++++++
 tools/perf/MANIFEST                    |  3 ++-
 tools/perf/Makefile.perf               |  1 +
 tools/perf/util/include/linux/bitops.h | 17 ++++++++++-------
 4 files changed, 37 insertions(+), 8 deletions(-)
 create mode 100644 tools/include/asm-generic/bitops.h

diff --git a/tools/include/asm-generic/bitops.h b/tools/include/asm-generic/bitops.h
new file mode 100644
index 000000000000..6dfd9d5fd828
--- /dev/null
+++ b/tools/include/asm-generic/bitops.h
@@ -0,0 +1,24 @@
+#ifndef __TOOLS_ASM_GENERIC_BITOPS_H
+#define __TOOLS_ASM_GENERIC_BITOPS_H
+
+/*
+ * tools/ copied this from include/asm-generic/bitops.h, bit by bit as it needed
+ * some functions.
+ *
+ * For the benefit of those who are trying to port Linux to another
+ * architecture, here are some C-language equivalents.  You should
+ * recode these in the native assembly language, if at all possible.
+ *
+ * C language equivalents written by Theodore Ts'o, 9/26/92
+ */
+
+#include <asm-generic/bitops/__ffs.h>
+#include <asm-generic/bitops/find.h>
+
+#ifndef _TOOLS_LINUX_BITOPS_H_
+#error only <linux/bitops.h> can be included directly
+#endif
+
+#include <asm-generic/bitops/atomic.h>
+
+#endif /* __TOOLS_ASM_GENERIC_BITOPS_H */
diff --git a/tools/perf/MANIFEST b/tools/perf/MANIFEST
index bfd7e22bafad..86f115cdbb5b 100644
--- a/tools/perf/MANIFEST
+++ b/tools/perf/MANIFEST
@@ -7,8 +7,9 @@ tools/lib/symbol/kallsyms.h
 tools/lib/util/find_next_bit.c
 tools/include/asm/bug.h
 tools/include/asm-generic/bitops/atomic.h
-tools/include/asm-generic/bitops/find.h
 tools/include/asm-generic/bitops/__ffs.h
+tools/include/asm-generic/bitops/find.h
+tools/include/asm-generic/bitops.h
 tools/include/linux/compiler.h
 tools/include/linux/export.h
 tools/include/linux/hash.h
diff --git a/tools/perf/Makefile.perf b/tools/perf/Makefile.perf
index 2d799da4eacc..d9bf9685e9da 100644
--- a/tools/perf/Makefile.perf
+++ b/tools/perf/Makefile.perf
@@ -235,6 +235,7 @@ LIB_H += util/include/linux/bitops.h
 LIB_H += ../include/asm-generic/bitops/atomic.h 
 LIB_H += ../include/asm-generic/bitops/find.h 
 LIB_H += ../include/asm-generic/bitops/__ffs.h 
+LIB_H += ../include/asm-generic/bitops.h
 LIB_H += ../include/linux/compiler.h
 LIB_H += ../include/linux/log2.h
 LIB_H += util/include/linux/const.h
diff --git a/tools/perf/util/include/linux/bitops.h b/tools/perf/util/include/linux/bitops.h
index 292aadeb66c5..5237bc9f6266 100644
--- a/tools/perf/util/include/linux/bitops.h
+++ b/tools/perf/util/include/linux/bitops.h
@@ -1,5 +1,5 @@
-#ifndef _PERF_LINUX_BITOPS_H_
-#define _PERF_LINUX_BITOPS_H_
+#ifndef _TOOLS_LINUX_BITOPS_H_
+#define _TOOLS_LINUX_BITOPS_H_
 
 #include <linux/kernel.h>
 #include <linux/compiler.h>
@@ -19,6 +19,14 @@
 #define BITS_TO_U32(nr)		DIV_ROUND_UP(nr, BITS_PER_BYTE * sizeof(u32))
 #define BITS_TO_BYTES(nr)	DIV_ROUND_UP(nr, BITS_PER_BYTE)
 
+/*
+ * Include this here because some architectures need generic_ffs/fls in
+ * scope
+ *
+ * XXX: this needs to be asm/bitops.h, when we get to per arch optimizations
+ */
+#include <asm-generic/bitops.h>
+
 #define for_each_set_bit(bit, addr, size) \
 	for ((bit) = find_first_bit((addr), (size));		\
 	     (bit) < (size);					\
@@ -30,14 +38,9 @@
 	     (bit) < (size);					\
 	     (bit) = find_next_bit((addr), (size), (bit) + 1))
 
-#include <asm-generic/bitops/atomic.h>
-
 static inline unsigned long hweight_long(unsigned long w)
 {
 	return sizeof(w) == 4 ? hweight32(w) : hweight64(w);
 }
 
-#include <asm-generic/bitops/__ffs.h>
-#include <asm-generic/bitops/find.h>
-
 #endif
-- 
1.9.3

[PATCH 12/18] tools: Move bitops.h from tools/perf/util to tools/

[Arnaldo Carvalho de Melo]

From: Arnaldo Carvalho de Melo <acme@redhat.com>

So that we better mirror the kernel sources and make it available for
other tools.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: David Ahern <dsahern@gmail.com>
Cc: Don Zickus <dzickus@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@kernel.org>
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-mvfu6x753tksnto3t6412m93@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/include/linux/bitops.h           | 46 ++++++++++++++++++++++++++++++++++
 tools/perf/MANIFEST                    |  1 +
 tools/perf/Makefile.perf               |  2 +-
 tools/perf/util/include/linux/bitops.h | 46 ----------------------------------
 4 files changed, 48 insertions(+), 47 deletions(-)
 create mode 100644 tools/include/linux/bitops.h
 delete mode 100644 tools/perf/util/include/linux/bitops.h

diff --git a/tools/include/linux/bitops.h b/tools/include/linux/bitops.h
new file mode 100644
index 000000000000..5237bc9f6266
--- /dev/null
+++ b/tools/include/linux/bitops.h
@@ -0,0 +1,46 @@
+#ifndef _TOOLS_LINUX_BITOPS_H_
+#define _TOOLS_LINUX_BITOPS_H_
+
+#include <linux/kernel.h>
+#include <linux/compiler.h>
+#include <asm/hweight.h>
+
+#ifndef __WORDSIZE
+#define __WORDSIZE (__SIZEOF_LONG__ * 8)
+#endif
+
+#define BITS_PER_LONG __WORDSIZE
+
+#define BIT_MASK(nr)		(1UL << ((nr) % BITS_PER_LONG))
+#define BIT_WORD(nr)		((nr) / BITS_PER_LONG)
+#define BITS_PER_BYTE		8
+#define BITS_TO_LONGS(nr)	DIV_ROUND_UP(nr, BITS_PER_BYTE * sizeof(long))
+#define BITS_TO_U64(nr)		DIV_ROUND_UP(nr, BITS_PER_BYTE * sizeof(u64))
+#define BITS_TO_U32(nr)		DIV_ROUND_UP(nr, BITS_PER_BYTE * sizeof(u32))
+#define BITS_TO_BYTES(nr)	DIV_ROUND_UP(nr, BITS_PER_BYTE)
+
+/*
+ * Include this here because some architectures need generic_ffs/fls in
+ * scope
+ *
+ * XXX: this needs to be asm/bitops.h, when we get to per arch optimizations
+ */
+#include <asm-generic/bitops.h>
+
+#define for_each_set_bit(bit, addr, size) \
+	for ((bit) = find_first_bit((addr), (size));		\
+	     (bit) < (size);					\
+	     (bit) = find_next_bit((addr), (size), (bit) + 1))
+
+/* same as for_each_set_bit() but use bit as value to start with */
+#define for_each_set_bit_from(bit, addr, size) \
+	for ((bit) = find_next_bit((addr), (size), (bit));	\
+	     (bit) < (size);					\
+	     (bit) = find_next_bit((addr), (size), (bit) + 1))
+
+static inline unsigned long hweight_long(unsigned long w)
+{
+	return sizeof(w) == 4 ? hweight32(w) : hweight64(w);
+}
+
+#endif
diff --git a/tools/perf/MANIFEST b/tools/perf/MANIFEST
index 86f115cdbb5b..7729af9f417a 100644
--- a/tools/perf/MANIFEST
+++ b/tools/perf/MANIFEST
@@ -10,6 +10,7 @@ tools/include/asm-generic/bitops/atomic.h
 tools/include/asm-generic/bitops/__ffs.h
 tools/include/asm-generic/bitops/find.h
 tools/include/asm-generic/bitops.h
+tools/include/linux/bitops.h
 tools/include/linux/compiler.h
 tools/include/linux/export.h
 tools/include/linux/hash.h
diff --git a/tools/perf/Makefile.perf b/tools/perf/Makefile.perf
index d9bf9685e9da..55b9c97cafc2 100644
--- a/tools/perf/Makefile.perf
+++ b/tools/perf/Makefile.perf
@@ -231,7 +231,7 @@ LIB_H += ../../include/uapi/linux/const.h
 LIB_H += ../include/linux/hash.h
 LIB_H += ../../include/linux/stringify.h
 LIB_H += util/include/linux/bitmap.h
-LIB_H += util/include/linux/bitops.h
+LIB_H += ../include/linux/bitops.h
 LIB_H += ../include/asm-generic/bitops/atomic.h 
 LIB_H += ../include/asm-generic/bitops/find.h 
 LIB_H += ../include/asm-generic/bitops/__ffs.h 
diff --git a/tools/perf/util/include/linux/bitops.h b/tools/perf/util/include/linux/bitops.h
deleted file mode 100644
index 5237bc9f6266..000000000000
--- a/tools/perf/util/include/linux/bitops.h
+++ /dev/null
@@ -1,46 +0,0 @@
-#ifndef _TOOLS_LINUX_BITOPS_H_
-#define _TOOLS_LINUX_BITOPS_H_
-
-#include <linux/kernel.h>
-#include <linux/compiler.h>
-#include <asm/hweight.h>
-
-#ifndef __WORDSIZE
-#define __WORDSIZE (__SIZEOF_LONG__ * 8)
-#endif
-
-#define BITS_PER_LONG __WORDSIZE
-
-#define BIT_MASK(nr)		(1UL << ((nr) % BITS_PER_LONG))
-#define BIT_WORD(nr)		((nr) / BITS_PER_LONG)
-#define BITS_PER_BYTE		8
-#define BITS_TO_LONGS(nr)	DIV_ROUND_UP(nr, BITS_PER_BYTE * sizeof(long))
-#define BITS_TO_U64(nr)		DIV_ROUND_UP(nr, BITS_PER_BYTE * sizeof(u64))
-#define BITS_TO_U32(nr)		DIV_ROUND_UP(nr, BITS_PER_BYTE * sizeof(u32))
-#define BITS_TO_BYTES(nr)	DIV_ROUND_UP(nr, BITS_PER_BYTE)
-
-/*
- * Include this here because some architectures need generic_ffs/fls in
- * scope
- *
- * XXX: this needs to be asm/bitops.h, when we get to per arch optimizations
- */
-#include <asm-generic/bitops.h>
-
-#define for_each_set_bit(bit, addr, size) \
-	for ((bit) = find_first_bit((addr), (size));		\
-	     (bit) < (size);					\
-	     (bit) = find_next_bit((addr), (size), (bit) + 1))
-
-/* same as for_each_set_bit() but use bit as value to start with */
-#define for_each_set_bit_from(bit, addr, size) \
-	for ((bit) = find_next_bit((addr), (size), (bit));	\
-	     (bit) < (size);					\
-	     (bit) = find_next_bit((addr), (size), (bit) + 1))
-
-static inline unsigned long hweight_long(unsigned long w)
-{
-	return sizeof(w) == 4 ? hweight32(w) : hweight64(w);
-}
-
-#endif
-- 
1.9.3

[PATCH 13/18] tools: Adopt fls_long and deps

[Arnaldo Carvalho de Melo]

From: Arnaldo Carvalho de Melo <acme@redhat.com>

Will be used when adopting rounddown_pow_of_two.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: David Ahern <dsahern@gmail.com>
Cc: Don Zickus <dzickus@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@kernel.org>
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-9m0tt5300q1ygv51hejjas82@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/include/asm-generic/bitops.h       | 3 +++
 tools/include/asm-generic/bitops/__fls.h | 1 +
 tools/include/asm-generic/bitops/fls.h   | 1 +
 tools/include/asm-generic/bitops/fls64.h | 1 +
 tools/include/linux/bitops.h             | 7 +++++++
 tools/perf/MANIFEST                      | 3 +++
 tools/perf/Makefile.perf                 | 3 +++
 7 files changed, 19 insertions(+)
 create mode 100644 tools/include/asm-generic/bitops/__fls.h
 create mode 100644 tools/include/asm-generic/bitops/fls.h
 create mode 100644 tools/include/asm-generic/bitops/fls64.h

diff --git a/tools/include/asm-generic/bitops.h b/tools/include/asm-generic/bitops.h
index 6dfd9d5fd828..6eedba1f7732 100644
--- a/tools/include/asm-generic/bitops.h
+++ b/tools/include/asm-generic/bitops.h
@@ -13,6 +13,9 @@
  */
 
 #include <asm-generic/bitops/__ffs.h>
+#include <asm-generic/bitops/fls.h>
+#include <asm-generic/bitops/__fls.h>
+#include <asm-generic/bitops/fls64.h>
 #include <asm-generic/bitops/find.h>
 
 #ifndef _TOOLS_LINUX_BITOPS_H_
diff --git a/tools/include/asm-generic/bitops/__fls.h b/tools/include/asm-generic/bitops/__fls.h
new file mode 100644
index 000000000000..2218b9add4c1
--- /dev/null
+++ b/tools/include/asm-generic/bitops/__fls.h
@@ -0,0 +1 @@
+#include <../../../../include/asm-generic/bitops/__fls.h>
diff --git a/tools/include/asm-generic/bitops/fls.h b/tools/include/asm-generic/bitops/fls.h
new file mode 100644
index 000000000000..dbf711a28f71
--- /dev/null
+++ b/tools/include/asm-generic/bitops/fls.h
@@ -0,0 +1 @@
+#include <../../../../include/asm-generic/bitops/fls.h>
diff --git a/tools/include/asm-generic/bitops/fls64.h b/tools/include/asm-generic/bitops/fls64.h
new file mode 100644
index 000000000000..980b1f63c047
--- /dev/null
+++ b/tools/include/asm-generic/bitops/fls64.h
@@ -0,0 +1 @@
+#include <../../../../include/asm-generic/bitops/fls64.h>
diff --git a/tools/include/linux/bitops.h b/tools/include/linux/bitops.h
index 5237bc9f6266..26005a15e7e2 100644
--- a/tools/include/linux/bitops.h
+++ b/tools/include/linux/bitops.h
@@ -43,4 +43,11 @@ static inline unsigned long hweight_long(unsigned long w)
 	return sizeof(w) == 4 ? hweight32(w) : hweight64(w);
 }
 
+static inline unsigned fls_long(unsigned long l)
+{
+	if (sizeof(l) == 4)
+		return fls(l);
+	return fls64(l);
+}
+
 #endif
diff --git a/tools/perf/MANIFEST b/tools/perf/MANIFEST
index 7729af9f417a..0c6cc8de007f 100644
--- a/tools/perf/MANIFEST
+++ b/tools/perf/MANIFEST
@@ -8,7 +8,10 @@ tools/lib/util/find_next_bit.c
 tools/include/asm/bug.h
 tools/include/asm-generic/bitops/atomic.h
 tools/include/asm-generic/bitops/__ffs.h
+tools/include/asm-generic/bitops/__fls.h
 tools/include/asm-generic/bitops/find.h
+tools/include/asm-generic/bitops/fls64.h
+tools/include/asm-generic/bitops/fls.h
 tools/include/asm-generic/bitops.h
 tools/include/linux/bitops.h
 tools/include/linux/compiler.h
diff --git a/tools/perf/Makefile.perf b/tools/perf/Makefile.perf
index 55b9c97cafc2..6b7153f8b056 100644
--- a/tools/perf/Makefile.perf
+++ b/tools/perf/Makefile.perf
@@ -234,7 +234,10 @@ LIB_H += util/include/linux/bitmap.h
 LIB_H += ../include/linux/bitops.h
 LIB_H += ../include/asm-generic/bitops/atomic.h 
 LIB_H += ../include/asm-generic/bitops/find.h 
+LIB_H += ../include/asm-generic/bitops/fls64.h
+LIB_H += ../include/asm-generic/bitops/fls.h
 LIB_H += ../include/asm-generic/bitops/__ffs.h 
+LIB_H += ../include/asm-generic/bitops/__fls.h
 LIB_H += ../include/asm-generic/bitops.h
 LIB_H += ../include/linux/compiler.h
 LIB_H += ../include/linux/log2.h
-- 
1.9.3

[PATCH 14/18] tools: Adopt rounddown_pow_of_two and deps

[Arnaldo Carvalho de Melo]

From: Arnaldo Carvalho de Melo <acme@redhat.com>

Will be used to make sure we pass a power of two when automatically
setting up the perf_mmap addr range length, as the kernel code
validating input on /proc/sys/kernel/perf_event_mlock_kb accepts any
integer, if we plain use it to set up the mmap lenght, we may get an
EINVAL when passing a non power of two.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: David Ahern <dsahern@gmail.com>
Cc: Don Zickus <dzickus@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@kernel.org>
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-zflvep0q01dmkruf4o291l4p@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/include/linux/log2.h | 134 +++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 134 insertions(+)

diff --git a/tools/include/linux/log2.h b/tools/include/linux/log2.h
index 141b7665d842..990b138362d6 100644
--- a/tools/include/linux/log2.h
+++ b/tools/include/linux/log2.h
@@ -13,6 +13,30 @@
 #define _TOOLS_LINUX_LOG2_H
 
 /*
+ * deal with unrepresentable constant logarithms
+ */
+extern __attribute__((const, noreturn))
+int ____ilog2_NaN(void);
+
+/*
+ * non-constant log of base 2 calculators
+ * - the arch may override these in asm/bitops.h if they can be implemented
+ *   more efficiently than using fls() and fls64()
+ * - the arch is not required to handle n==0 if implementing the fallback
+ */
+static inline __attribute__((const))
+int __ilog2_u32(u32 n)
+{
+	return fls(n) - 1;
+}
+
+static inline __attribute__((const))
+int __ilog2_u64(u64 n)
+{
+	return fls64(n) - 1;
+}
+
+/*
  *  Determine whether some value is a power of two, where zero is
  * *not* considered a power of two.
  */
@@ -23,4 +47,114 @@ bool is_power_of_2(unsigned long n)
 	return (n != 0 && ((n & (n - 1)) == 0));
 }
 
+/*
+ * round down to nearest power of two
+ */
+static inline __attribute__((const))
+unsigned long __rounddown_pow_of_two(unsigned long n)
+{
+	return 1UL << (fls_long(n) - 1);
+}
+
+/**
+ * ilog2 - log of base 2 of 32-bit or a 64-bit unsigned value
+ * @n - parameter
+ *
+ * constant-capable log of base 2 calculation
+ * - this can be used to initialise global variables from constant data, hence
+ *   the massive ternary operator construction
+ *
+ * selects the appropriately-sized optimised version depending on sizeof(n)
+ */
+#define ilog2(n)				\
+(						\
+	__builtin_constant_p(n) ? (		\
+		(n) < 1 ? ____ilog2_NaN() :	\
+		(n) & (1ULL << 63) ? 63 :	\
+		(n) & (1ULL << 62) ? 62 :	\
+		(n) & (1ULL << 61) ? 61 :	\
+		(n) & (1ULL << 60) ? 60 :	\
+		(n) & (1ULL << 59) ? 59 :	\
+		(n) & (1ULL << 58) ? 58 :	\
+		(n) & (1ULL << 57) ? 57 :	\
+		(n) & (1ULL << 56) ? 56 :	\
+		(n) & (1ULL << 55) ? 55 :	\
+		(n) & (1ULL << 54) ? 54 :	\
+		(n) & (1ULL << 53) ? 53 :	\
+		(n) & (1ULL << 52) ? 52 :	\
+		(n) & (1ULL << 51) ? 51 :	\
+		(n) & (1ULL << 50) ? 50 :	\
+		(n) & (1ULL << 49) ? 49 :	\
+		(n) & (1ULL << 48) ? 48 :	\
+		(n) & (1ULL << 47) ? 47 :	\
+		(n) & (1ULL << 46) ? 46 :	\
+		(n) & (1ULL << 45) ? 45 :	\
+		(n) & (1ULL << 44) ? 44 :	\
+		(n) & (1ULL << 43) ? 43 :	\
+		(n) & (1ULL << 42) ? 42 :	\
+		(n) & (1ULL << 41) ? 41 :	\
+		(n) & (1ULL << 40) ? 40 :	\
+		(n) & (1ULL << 39) ? 39 :	\
+		(n) & (1ULL << 38) ? 38 :	\
+		(n) & (1ULL << 37) ? 37 :	\
+		(n) & (1ULL << 36) ? 36 :	\
+		(n) & (1ULL << 35) ? 35 :	\
+		(n) & (1ULL << 34) ? 34 :	\
+		(n) & (1ULL << 33) ? 33 :	\
+		(n) & (1ULL << 32) ? 32 :	\
+		(n) & (1ULL << 31) ? 31 :	\
+		(n) & (1ULL << 30) ? 30 :	\
+		(n) & (1ULL << 29) ? 29 :	\
+		(n) & (1ULL << 28) ? 28 :	\
+		(n) & (1ULL << 27) ? 27 :	\
+		(n) & (1ULL << 26) ? 26 :	\
+		(n) & (1ULL << 25) ? 25 :	\
+		(n) & (1ULL << 24) ? 24 :	\
+		(n) & (1ULL << 23) ? 23 :	\
+		(n) & (1ULL << 22) ? 22 :	\
+		(n) & (1ULL << 21) ? 21 :	\
+		(n) & (1ULL << 20) ? 20 :	\
+		(n) & (1ULL << 19) ? 19 :	\
+		(n) & (1ULL << 18) ? 18 :	\
+		(n) & (1ULL << 17) ? 17 :	\
+		(n) & (1ULL << 16) ? 16 :	\
+		(n) & (1ULL << 15) ? 15 :	\
+		(n) & (1ULL << 14) ? 14 :	\
+		(n) & (1ULL << 13) ? 13 :	\
+		(n) & (1ULL << 12) ? 12 :	\
+		(n) & (1ULL << 11) ? 11 :	\
+		(n) & (1ULL << 10) ? 10 :	\
+		(n) & (1ULL <<  9) ?  9 :	\
+		(n) & (1ULL <<  8) ?  8 :	\
+		(n) & (1ULL <<  7) ?  7 :	\
+		(n) & (1ULL <<  6) ?  6 :	\
+		(n) & (1ULL <<  5) ?  5 :	\
+		(n) & (1ULL <<  4) ?  4 :	\
+		(n) & (1ULL <<  3) ?  3 :	\
+		(n) & (1ULL <<  2) ?  2 :	\
+		(n) & (1ULL <<  1) ?  1 :	\
+		(n) & (1ULL <<  0) ?  0 :	\
+		____ilog2_NaN()			\
+				   ) :		\
+	(sizeof(n) <= 4) ?			\
+	__ilog2_u32(n) :			\
+	__ilog2_u64(n)				\
+ )
+
+
+/**
+ * rounddown_pow_of_two - round the given value down to nearest power of two
+ * @n - parameter
+ *
+ * round the given value down to the nearest power of two
+ * - the result is undefined when n == 0
+ * - this can be used to initialise global variables from constant data
+ */
+#define rounddown_pow_of_two(n)			\
+(						\
+	__builtin_constant_p(n) ? (		\
+		(1UL << ilog2(n))) :		\
+	__rounddown_pow_of_two(n)		\
+ )
+
 #endif /* _TOOLS_LINUX_LOG2_H */
-- 
1.9.3

[PATCH 15/18] perf tools: Make the mmap length autotuning more robust

[Arnaldo Carvalho de Melo]

From: Arnaldo Carvalho de Melo <acme@redhat.com>

If /proc/sys/kernel/perf_event_mlock_kb is not (power of 2 + PAGE_SIZE_in_kb)
and we let the perf tools do mmap length autosizing based on that, then, for
non-CAP_IPC_LOCK users when /proc/sys/kernel/perf_event_paranoid is > -1, then
we get an -EINVAL that ends up in:

  [acme@ssdandy linux]$ trace usleep 1
  Invalid argument
  [acme@ssdandy linux]$ perf record usleep 1
  failed to mmap with 22 (Invalid argument)

After this fix:

  [acme@ssdandy linux]$ trace usleep 1
  <SNIP>
   0.806 ( 0.006 ms): munmap(addr: 0x7f7e4740a000, len: 66467) = 0
   0.869 ( 0.002 ms): brk(                                   ) = 0x7bb000
   0.873 ( 0.003 ms): brk(brk: 0x7dc000                      ) = 0x7dc000
   0.877 ( 0.001 ms): brk(                                   ) = 0x7dc000
   0.953 ( 0.058 ms): nanosleep(rqtp: 0x7fff26ab9420         ) = 0
   0.959 ( 0.000 ms): exit_group(
  [acme@ssdandy linux]$ perf record usleep 1
  [ perf record: Woken up 1 times to write data ]
  [ perf record: Captured and wrote 0.017 MB perf.data (~759 samples) ]
  [acme@ssdandy linux]$

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: David Ahern <dsahern@gmail.com>
Cc: Don Zickus <dzickus@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@kernel.org>
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-6p6l5ou6jev6o7ymc4nn1n2a@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/evlist.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c
index 8eb92cf5d552..7d88327d94a9 100644
--- a/tools/perf/util/evlist.c
+++ b/tools/perf/util/evlist.c
@@ -909,6 +909,8 @@ static size_t perf_evlist__mmap_size(unsigned long pages)
 		}
 
 		pages = (max * 1024) / page_size;
+		if (!is_power_of_2(pages))
+			pages = rounddown_pow_of_two(pages);
 	} else if (!is_power_of_2(pages))
 		return 0;
 
-- 
1.9.3

[PATCH 16/18] tools: Adopt roundup_pow_of_two

[Arnaldo Carvalho de Melo]

From: Arnaldo Carvalho de Melo <acme@redhat.com>

To replace equivalent code used in the mmap_pages command line
parameter handling in tools/perf.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: David Ahern <dsahern@gmail.com>
Cc: Don Zickus <dzickus@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@kernel.org>
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-i44zs02xt4zexfxywpklo7km@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/include/linux/log2.h | 25 +++++++++++++++++++++++++
 1 file changed, 25 insertions(+)

diff --git a/tools/include/linux/log2.h b/tools/include/linux/log2.h
index 990b138362d6..41446668ccce 100644
--- a/tools/include/linux/log2.h
+++ b/tools/include/linux/log2.h
@@ -48,6 +48,15 @@ bool is_power_of_2(unsigned long n)
 }
 
 /*
+ * round up to nearest power of two
+ */
+static inline __attribute__((const))
+unsigned long __roundup_pow_of_two(unsigned long n)
+{
+	return 1UL << fls_long(n - 1);
+}
+
+/*
  * round down to nearest power of two
  */
 static inline __attribute__((const))
@@ -141,6 +150,22 @@ unsigned long __rounddown_pow_of_two(unsigned long n)
 	__ilog2_u64(n)				\
  )
 
+/**
+ * roundup_pow_of_two - round the given value up to nearest power of two
+ * @n - parameter
+ *
+ * round the given value up to the nearest power of two
+ * - the result is undefined when n == 0
+ * - this can be used to initialise global variables from constant data
+ */
+#define roundup_pow_of_two(n)			\
+(						\
+	__builtin_constant_p(n) ? (		\
+		(n == 1) ? 1 :			\
+		(1UL << (ilog2((n) - 1) + 1))	\
+				   ) :		\
+	__roundup_pow_of_two(n)			\
+ )
 
 /**
  * rounddown_pow_of_two - round the given value down to nearest power of two
-- 
1.9.3

[PATCH 17/18] perf evlist: Use roundup_pow_of_two

[Arnaldo Carvalho de Melo]

From: Arnaldo Carvalho de Melo <acme@redhat.com>

And remove the equivalent next_pow2{_l} functions.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: David Ahern <dsahern@gmail.com>
Cc: Don Zickus <dzickus@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@kernel.org>
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-hl9ct3wcbs5deai3v5ljmuws@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/evlist.c |  2 +-
 tools/perf/util/util.h   | 18 ------------------
 2 files changed, 1 insertion(+), 19 deletions(-)

diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c
index 7d88327d94a9..cbab1fb77b1d 100644
--- a/tools/perf/util/evlist.c
+++ b/tools/perf/util/evlist.c
@@ -948,7 +948,7 @@ static long parse_pages_arg(const char *str, unsigned long min,
 		/* leave number of pages at 0 */
 	} else if (!is_power_of_2(pages)) {
 		/* round pages up to next power of 2 */
-		pages = next_pow2_l(pages);
+		pages = roundup_pow_of_two(pages);
 		if (!pages)
 			return -EINVAL;
 		pr_info("rounding mmap pages size to %lu bytes (%lu pages)\n",
diff --git a/tools/perf/util/util.h b/tools/perf/util/util.h
index 25b22bbea066..be198ac27031 100644
--- a/tools/perf/util/util.h
+++ b/tools/perf/util/util.h
@@ -269,24 +269,6 @@ void event_attr_init(struct perf_event_attr *attr);
 #define _STR(x) #x
 #define STR(x) _STR(x)
 
-static inline unsigned next_pow2(unsigned x)
-{
-	if (!x)
-		return 1;
-	return 1ULL << (32 - __builtin_clz(x - 1));
-}
-
-static inline unsigned long next_pow2_l(unsigned long x)
-{
-#if BITS_PER_LONG == 64
-	if (x <= (1UL << 31))
-		return next_pow2(x);
-	return (unsigned long)next_pow2(x >> 32) << 32;
-#else
-	return next_pow2(x);
-#endif
-}
-
 size_t hex_width(u64 v);
 int hex2u64(const char *ptr, u64 *val);
 
-- 
1.9.3

[PATCH 18/18] perf symbols: Fix use after free in filename__read_build_id

[Arnaldo Carvalho de Melo]

From: Mitchell Krome <mitchellkrome@gmail.com>

In filename__read_build_id, phdr points to memory in buf, which gets realloced
before a call to fseek that uses phdr->p_offset. This change stores the value
of p_offset before buf is realloced, so the fseek can use the value safely.

Signed-off-by: Mitchell Krome <mitchellkrome@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20141216021612.GA7199@mitchell
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/symbol-minimal.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/tools/perf/util/symbol-minimal.c b/tools/perf/util/symbol-minimal.c
index fa585c63f56a..d7efb03b3f9a 100644
--- a/tools/perf/util/symbol-minimal.c
+++ b/tools/perf/util/symbol-minimal.c
@@ -129,6 +129,7 @@ int filename__read_build_id(const char *filename, void *bf, size_t size)
 
 		for (i = 0, phdr = buf; i < ehdr.e_phnum; i++, phdr++) {
 			void *tmp;
+			long offset;
 
 			if (need_swap) {
 				phdr->p_type = bswap_32(phdr->p_type);
@@ -140,12 +141,13 @@ int filename__read_build_id(const char *filename, void *bf, size_t size)
 				continue;
 
 			buf_size = phdr->p_filesz;
+			offset = phdr->p_offset;
 			tmp = realloc(buf, buf_size);
 			if (tmp == NULL)
 				goto out_free;
 
 			buf = tmp;
-			fseek(fp, phdr->p_offset, SEEK_SET);
+			fseek(fp, offset, SEEK_SET);
 			if (fread(buf, buf_size, 1, fp) != 1)
 				goto out_free;
 
@@ -178,6 +180,7 @@ int filename__read_build_id(const char *filename, void *bf, size_t size)
 
 		for (i = 0, phdr = buf; i < ehdr.e_phnum; i++, phdr++) {
 			void *tmp;
+			long offset;
 
 			if (need_swap) {
 				phdr->p_type = bswap_32(phdr->p_type);
@@ -189,12 +192,13 @@ int filename__read_build_id(const char *filename, void *bf, size_t size)
 				continue;
 
 			buf_size = phdr->p_filesz;
+			offset = phdr->p_offset;
 			tmp = realloc(buf, buf_size);
 			if (tmp == NULL)
 				goto out_free;
 
 			buf = tmp;
-			fseek(fp, phdr->p_offset, SEEK_SET);
+			fseek(fp, offset, SEEK_SET);
 			if (fread(buf, buf_size, 1, fp) != 1)
 				goto out_free;
 
-- 
1.9.3

Re: [PATCH 05/18] perf evlist: Do not use hard coded value for a mmap_pages default

[Jiri Olsa]

On Tue, Dec 16, 2014 at 01:57:07PM -0300, Arnaldo Carvalho de Melo wrote:
> From: Arnaldo Carvalho de Melo <acme@redhat.com>
> 
> So far what is in there by default is what we were using: 512KB + the
> control page, but the admin may change that, and if it does to a smaller
> value, all calls to tooling for non root users start failing, requiring
> that the user manually set --mmap_pages/-m.
> 
> Use instead what is in /proc/sys/kernel/perf_event_mlock_kb.
> 
> Cc: Adrian Hunter <adrian.hunter@intel.com>
> Cc: Borislav Petkov <bp@suse.de>
> Cc: David Ahern <dsahern@gmail.com>
> Cc: Don Zickus <dzickus@redhat.com>
> Cc: Frederic Weisbecker <fweisbec@gmail.com>
> Cc: Jiri Olsa <jolsa@redhat.com>
> Cc: Mike Galbraith <efault@gmx.de>
> Cc: Namhyung Kim <namhyung@kernel.org>
> 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-2f6mtm8xu3wo5lhkql6jdblh@git.kernel.org
> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
> ---
>  tools/perf/util/evlist.c | 20 ++++++++++++++++----
>  1 file changed, 16 insertions(+), 4 deletions(-)
> 
> diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c
> index 7847f3885081..ac808680e61c 100644
> --- a/tools/perf/util/evlist.c
> +++ b/tools/perf/util/evlist.c
> @@ -893,10 +893,22 @@ out_unmap:
>  
>  static size_t perf_evlist__mmap_size(unsigned long pages)
>  {
> -	/* 512 kiB: default amount of unprivileged mlocked memory */
> -	if (pages == UINT_MAX)
> -		pages = (512 * 1024) / page_size;
> -	else if (!is_power_of_2(pages))
> +	if (pages == UINT_MAX) {
> +		int max;
> +
> +		if (sysctl__read_int("kernel/perf_event_mlock_kb", &max) < 0) {
> +			/*
> +			 * Pick a once upon a time good value, i.e. things look
> +			 * strange since we can't read a sysctl value, but lets not
> +			 * die yet...
> +			 */
> +			max = 512;
> +		} else {
> +			max -= (page_size / 1024);
> +		}

so this way you depend on value in perf_event_mlock_kb being power 2,
otherwise:

# echo 1000 > /proc/sys/kernel/perf_event_mlock_kb
$ perf record ls
failed to mmap with 22 (Invalid argument)

maybe you could use the logic/code from parse_pages_arg function

jirka

Re: [PATCH 06/18] tools: Move __ffs implementation to tools/include/asm-generic/bitops/__ffs.h

[Jiri Olsa]

On Tue, Dec 16, 2014 at 01:57:08PM -0300, Arnaldo Carvalho de Melo wrote:

SNIP

> +	}
> +	if ((word & 0x1) == 0)
> +		num += 1;
> +	return num;
> +}
> +
> +#endif /* _TOOLS_LINUX_ASM_GENERIC_BITOPS___FFS_H_ */
> diff --git a/tools/perf/MANIFEST b/tools/perf/MANIFEST
> index 344c4d3d0a4a..39c08636357b 100644
> --- a/tools/perf/MANIFEST
> +++ b/tools/perf/MANIFEST
> @@ -5,6 +5,7 @@ tools/lib/api
>  tools/lib/symbol/kallsyms.c
>  tools/lib/symbol/kallsyms.h
>  tools/include/asm/bug.h
> +tools/include/asm-generic/bitops/__ffs.h
>  tools/include/linux/compiler.h
>  tools/include/linux/hash.h
>  tools/include/linux/export.h
> diff --git a/tools/perf/Makefile.perf b/tools/perf/Makefile.perf
> index 763e68fb5767..bfd86a32254a 100644
> --- a/tools/perf/Makefile.perf
> +++ b/tools/perf/Makefile.perf
> @@ -232,6 +232,7 @@ LIB_H += ../include/linux/hash.h
>  LIB_H += ../../include/linux/stringify.h
>  LIB_H += util/include/linux/bitmap.h
>  LIB_H += util/include/linux/bitops.h
> +LIB_H += ../include/asm-generic/bitops/__ffs.h 

whitespace at the EOLN ;-)

jirka

Re: [PATCH 10/18] tools lib: Move asm-generic/bitops/find.h code to tools/include and tools/lib

[Jiri Olsa]

On Tue, Dec 16, 2014 at 01:57:12PM -0300, Arnaldo Carvalho de Melo wrote:
> From: Arnaldo Carvalho de Melo <acme@redhat.com>
> 
> To match the Linux kernel source code structure from where this code came from.

SNIP

> +			goto found;
> +		result += BITS_PER_LONG;
> +		size -= BITS_PER_LONG;
> +	}
> +	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);
> +}
> +#endif
> diff --git a/tools/perf/MANIFEST b/tools/perf/MANIFEST
> index 67d48e20f7e1..bfd7e22bafad 100644
> --- a/tools/perf/MANIFEST
> +++ b/tools/perf/MANIFEST
> @@ -4,8 +4,10 @@ tools/lib/traceevent
>  tools/lib/api
>  tools/lib/symbol/kallsyms.c
>  tools/lib/symbol/kallsyms.h
> +tools/lib/util/find_next_bit.c

hum, I had the impression that we will move 'generic' things under 'api' to
be built separatelly.. so whats the current notion? ;-)

what kind of code belongs to 'lib/api/...' and what to 'lib/...' ?

thanks,
jirka

Re: [PATCH 10/18] tools lib: Move asm-generic/bitops/find.h code to tools/include and tools/lib

[Jiri Olsa]

On Tue, Dec 16, 2014 at 01:57:12PM -0300, Arnaldo Carvalho de Melo wrote:
> From: Arnaldo Carvalho de Melo <acme@redhat.com>
> 
> To match the Linux kernel source code structure from where this code came from.

SNIP

> diff --git a/tools/perf/MANIFEST b/tools/perf/MANIFEST
> index 67d48e20f7e1..bfd7e22bafad 100644
> --- a/tools/perf/MANIFEST
> +++ b/tools/perf/MANIFEST
> @@ -4,8 +4,10 @@ tools/lib/traceevent
>  tools/lib/api
>  tools/lib/symbol/kallsyms.c
>  tools/lib/symbol/kallsyms.h
> +tools/lib/util/find_next_bit.c
>  tools/include/asm/bug.h
>  tools/include/asm-generic/bitops/atomic.h
> +tools/include/asm-generic/bitops/find.h
>  tools/include/asm-generic/bitops/__ffs.h
>  tools/include/linux/compiler.h
>  tools/include/linux/export.h
> @@ -18,6 +20,7 @@ include/linux/rbtree.h
>  include/linux/list.h
>  include/linux/hash.h
>  include/linux/stringify.h
> +lib/find_next_bit.c
>  lib/rbtree.c
>  include/linux/swab.h
>  arch/*/include/asm/unistd*.h
> diff --git a/tools/perf/Makefile.perf b/tools/perf/Makefile.perf
> index fc4082fdf6bf..2d799da4eacc 100644
> --- a/tools/perf/Makefile.perf
> +++ b/tools/perf/Makefile.perf
> @@ -233,6 +233,7 @@ LIB_H += ../../include/linux/stringify.h
>  LIB_H += util/include/linux/bitmap.h
>  LIB_H += util/include/linux/bitops.h
>  LIB_H += ../include/asm-generic/bitops/atomic.h 
> +LIB_H += ../include/asm-generic/bitops/find.h 

whitespace at the EOLN ;-)

jirka

Re: [PATCH 15/18] perf tools: Make the mmap length autotuning more robust

[Jiri Olsa]

On Tue, Dec 16, 2014 at 01:57:17PM -0300, Arnaldo Carvalho de Melo wrote:
> From: Arnaldo Carvalho de Melo <acme@redhat.com>
> 
> If /proc/sys/kernel/perf_event_mlock_kb is not (power of 2 + PAGE_SIZE_in_kb)
> and we let the perf tools do mmap length autosizing based on that, then, for
> non-CAP_IPC_LOCK users when /proc/sys/kernel/perf_event_paranoid is > -1, then
> we get an -EINVAL that ends up in:
> 
>   [acme@ssdandy linux]$ trace usleep 1
>   Invalid argument
>   [acme@ssdandy linux]$ perf record usleep 1
>   failed to mmap with 22 (Invalid argument)
> 
> After this fix:
> 
>   [acme@ssdandy linux]$ trace usleep 1
>   <SNIP>
>    0.806 ( 0.006 ms): munmap(addr: 0x7f7e4740a000, len: 66467) = 0
>    0.869 ( 0.002 ms): brk(                                   ) = 0x7bb000
>    0.873 ( 0.003 ms): brk(brk: 0x7dc000                      ) = 0x7dc000
>    0.877 ( 0.001 ms): brk(                                   ) = 0x7dc000
>    0.953 ( 0.058 ms): nanosleep(rqtp: 0x7fff26ab9420         ) = 0
>    0.959 ( 0.000 ms): exit_group(
>   [acme@ssdandy linux]$ perf record usleep 1
>   [ perf record: Woken up 1 times to write data ]
>   [ perf record: Captured and wrote 0.017 MB perf.data (~759 samples) ]
>   [acme@ssdandy linux]$
> 
> Cc: Adrian Hunter <adrian.hunter@intel.com>
> Cc: Borislav Petkov <bp@suse.de>
> Cc: David Ahern <dsahern@gmail.com>
> Cc: Don Zickus <dzickus@redhat.com>
> Cc: Frederic Weisbecker <fweisbec@gmail.com>
> Cc: Jiri Olsa <jolsa@redhat.com>
> Cc: Mike Galbraith <efault@gmx.de>
> Cc: Namhyung Kim <namhyung@kernel.org>
> 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-6p6l5ou6jev6o7ymc4nn1n2a@git.kernel.org
> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
> ---
>  tools/perf/util/evlist.c | 2 ++
>  1 file changed, 2 insertions(+)
> 
> diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c
> index 8eb92cf5d552..7d88327d94a9 100644
> --- a/tools/perf/util/evlist.c
> +++ b/tools/perf/util/evlist.c
> @@ -909,6 +909,8 @@ static size_t perf_evlist__mmap_size(unsigned long pages)
>  		}
>  
>  		pages = (max * 1024) / page_size;
> +		if (!is_power_of_2(pages))
> +			pages = rounddown_pow_of_two(pages);

ok then ;-)

jirka

Re: [PATCH 05/18] perf evlist: Do not use hard coded value for a mmap_pages default

[Arnaldo Carvalho de Melo]

Em Wed, Dec 17, 2014 at 02:23:30PM +0100, Jiri Olsa escreveu:
> On Tue, Dec 16, 2014 at 01:57:07PM -0300, Arnaldo Carvalho de Melo wrote:
> > +	if (pages == UINT_MAX) {
> > +		int max;
> > +
> > +		if (sysctl__read_int("kernel/perf_event_mlock_kb", &max) < 0) {
> > +			/*
> > +			 * Pick a once upon a time good value, i.e. things look
> > +			 * strange since we can't read a sysctl value, but lets not
> > +			 * die yet...
> > +			 */
> > +			max = 512;
> > +		} else {
> > +			max -= (page_size / 1024);
> > +		}
 
> so this way you depend on value in perf_event_mlock_kb being power 2,
> otherwise:

> # echo 1000 > /proc/sys/kernel/perf_event_mlock_kb
> $ perf record ls
> failed to mmap with 22 (Invalid argument)

Right, I thought that perf_event_mlock_kb input was validated, realized
that it is not, see the following patchkit that makes it validate that
after reading, using rounddown_pow_of_two if not.

- Arnaldo
 
> maybe you could use the logic/code from parse_pages_arg function
> 
> jirka

Re: [PATCH 10/18] tools lib: Move asm-generic/bitops/find.h code to tools/include and tools/lib

[Arnaldo Carvalho de Melo]

Em Wed, Dec 17, 2014 at 02:42:23PM +0100, Jiri Olsa escreveu:
> On Tue, Dec 16, 2014 at 01:57:12PM -0300, Arnaldo Carvalho de Melo wrote:
> > +++ b/tools/perf/Makefile.perf
> > @@ -233,6 +233,7 @@ LIB_H += ../../include/linux/stringify.h
> >  LIB_H += util/include/linux/bitmap.h
> >  LIB_H += util/include/linux/bitops.h
> >  LIB_H += ../include/asm-generic/bitops/atomic.h 
> > +LIB_H += ../include/asm-generic/bitops/find.h 
> 
> whitespace at the EOLN ;-)

How could this slip thru? /me scratches head, will check, but I thought
that I catched these things with the .git hooks, humm, three new entries
in Makefile.perf has, three doesn't :-\ Will try to plug this in my
machine setups.

- Arnaldo

Re: [PATCH 10/18] tools lib: Move asm-generic/bitops/find.h code to tools/include and tools/lib

[Arnaldo Carvalho de Melo]

Em Wed, Dec 17, 2014 at 02:41:16PM +0100, Jiri Olsa escreveu:
> On Tue, Dec 16, 2014 at 01:57:12PM -0300, Arnaldo Carvalho de Melo wrote:
> > +++ b/tools/perf/MANIFEST
> > @@ -4,8 +4,10 @@ tools/lib/traceevent
> >  tools/lib/api
> >  tools/lib/symbol/kallsyms.c
> >  tools/lib/symbol/kallsyms.h
> > +tools/lib/util/find_next_bit.c
 
> hum, I had the impression that we will move 'generic' things under 'api' to
> be built separatelly.. so whats the current notion? ;-)
 
> what kind of code belongs to 'lib/api/...' and what to 'lib/...' ?

Well, the reasoning here was that if libkapi is for functions that
provide glue over kernel APIs, which a set of bitmap handling routines
is not, i.e. that is more generic and not strictly related to the
kernel.

- Arnaldo

Re: [PATCH 10/18] tools lib: Move asm-generic/bitops/find.h code to tools/include and tools/lib

[Arnaldo Carvalho de Melo]

Em Wed, Dec 17, 2014 at 11:02:35AM -0300, Arnaldo Carvalho de Melo escreveu:
> Em Wed, Dec 17, 2014 at 02:42:23PM +0100, Jiri Olsa escreveu:
> > On Tue, Dec 16, 2014 at 01:57:12PM -0300, Arnaldo Carvalho de Melo wrote:
> > > +++ b/tools/perf/Makefile.perf
> > > @@ -233,6 +233,7 @@ LIB_H += ../../include/linux/stringify.h
> > >  LIB_H += util/include/linux/bitmap.h
> > >  LIB_H += util/include/linux/bitops.h
> > >  LIB_H += ../include/asm-generic/bitops/atomic.h 
> > > +LIB_H += ../include/asm-generic/bitops/find.h 
> > 
> > whitespace at the EOLN ;-)
> 
> How could this slip thru? /me scratches head, will check, but I thought
> that I catched these things with the .git hooks, humm, three new entries
> in Makefile.perf has, three doesn't :-\ Will try to plug this in my
> machine setups.

I redid the offending csets and put them on a new signed tag:

  perf-core-for-mingo-2

And also force pushed perf/core, i.e. the original stuff is there,
available thru the perf-core-for-mingo signed tag, the fixed up stuff is
on perf/core and on this new signed tag.

Ingo, if you see this, please prefer the fixed up one,

Thanks,

- Arnaldo

Re: [GIT PULL 00/18] perf/core improvements and fixes

[Arnaldo Carvalho de Melo]

Em Tue, Dec 16, 2014 at 01:57:02PM -0300, Arnaldo Carvalho de Melo escreveu:
> Hi Ingo,
> 
> 	Please consider pulling,
> 
> - Arnaldo
> 
> The following changes since commit 41e950c033b7df997d4b38653efe6554be9b96a7:
> 
>   Merge tag 'perf-core-for-mingo' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/urgent (2014-12-12 09:09:52 +0100)
> 
> are available in the git repository at:
> 
> 
>   git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux.git tags/perf-core-for-mingo

Ingo,

	Please consider pulling the perf-core-for-mingo-2 signed tag
instead, it is exactly the same content modulo removal of some
whitespaces at the end of a few lines in tools/perf/Makefile.perf that
Jiri found while reading those patches,

Thanks,

- Arnaldo
 
> for you to fetch changes up to 67195c75a87232f055ff415fc4624ef01f24fc3d:
> 
>   perf symbols: Fix use after free in filename__read_build_id (2014-12-16 13:38:28 -0300)
> 
> ----------------------------------------------------------------
> perf/core improvements and fixes:
> 
> User visible:
> 
> - The mmap address range for the ring buffer now is calculated using the
>   contents of /proc/sys/kernel/perf_event_mlock_kb.
> 
>   This fixes an -EPERM case where 'trace' was trying to use more than what
>   configured on perf_event_mlock_kb. (Arnaldo Carvalho de Melo)
> 
> Infrastructure:
> 
> - Move bitops definitions so that they match the header file hierarchy
>   in the kernel sources where that code came from. (Arnaldo Carvalho de Melo)
> 
> - Adopt round{down,up}_pow_of_two from the kernel and use it instead of
>   equivalent code, so that we reuse more kernel code and make tools/ look
>   more like kernel source code, to encourage further contributions from
>   kernel hackers (Arnaldo Carvalho de Melo)
> 
> - Fix use after free in filename__read_build_id (Mitchell Krome)
> 
> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
> 
> ----------------------------------------------------------------
> Arnaldo Carvalho de Melo (17):
>       perf evlist: Fixup brown paper bag on "hint" for --mmap-pages cmdline arg
>       perf evlist: Clarify sterror_mmap variable names
>       perf evlist: Improve the strerror_mmap method
>       perf trace: Let the perf_evlist__mmap autosize the number of pages to use
>       perf evlist: Do not use hard coded value for a mmap_pages default
>       tools: Move __ffs implementation to tools/include/asm-generic/bitops/__ffs.h
>       tools: Move code originally from linux/log2.h to tools/include/linux/
>       tools: Move code originally from asm-generic/atomic.h into tools/include/asm-generic/
>       tools: Whitespace prep patches for moving bitops.h
>       tools lib: Move asm-generic/bitops/find.h code to tools/include and tools/lib
>       tools: Introduce asm-generic/bitops.h
>       tools: Move bitops.h from tools/perf/util to tools/
>       tools: Adopt fls_long and deps
>       tools: Adopt rounddown_pow_of_two and deps
>       perf tools: Make the mmap length autotuning more robust
>       tools: Adopt roundup_pow_of_two
>       perf evlist: Use roundup_pow_of_two
> 
> Mitchell Krome (1):
>       perf symbols: Fix use after free in filename__read_build_id
> 
>  tools/include/asm-generic/bitops.h        |  27 +++++
>  tools/include/asm-generic/bitops/__ffs.h  |  43 +++++++
>  tools/include/asm-generic/bitops/__fls.h  |   1 +
>  tools/include/asm-generic/bitops/atomic.h |  22 ++++
>  tools/include/asm-generic/bitops/find.h   |  33 ++++++
>  tools/include/asm-generic/bitops/fls.h    |   1 +
>  tools/include/asm-generic/bitops/fls64.h  |   1 +
>  tools/include/linux/bitops.h              |  53 +++++++++
>  tools/include/linux/log2.h                | 185 ++++++++++++++++++++++++++++++
>  tools/lib/util/find_next_bit.c            |  89 ++++++++++++++
>  tools/perf/MANIFEST                       |  13 ++-
>  tools/perf/Makefile.perf                  |  14 ++-
>  tools/perf/builtin-trace.c                |   2 +-
>  tools/perf/util/evlist.c                  |  46 ++++++--
>  tools/perf/util/include/linux/bitops.h    | 162 --------------------------
>  tools/perf/util/symbol-minimal.c          |   8 +-
>  tools/perf/util/util.h                    |  29 -----
>  17 files changed, 522 insertions(+), 207 deletions(-)
>  create mode 100644 tools/include/asm-generic/bitops.h
>  create mode 100644 tools/include/asm-generic/bitops/__ffs.h
>  create mode 100644 tools/include/asm-generic/bitops/__fls.h
>  create mode 100644 tools/include/asm-generic/bitops/atomic.h
>  create mode 100644 tools/include/asm-generic/bitops/find.h
>  create mode 100644 tools/include/asm-generic/bitops/fls.h
>  create mode 100644 tools/include/asm-generic/bitops/fls64.h
>  create mode 100644 tools/include/linux/bitops.h
>  create mode 100644 tools/include/linux/log2.h
>  create mode 100644 tools/lib/util/find_next_bit.c
>  delete mode 100644 tools/perf/util/include/linux/bitops.h
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/

Re: [GIT PULL 00/18] perf/core improvements and fixes

[Ingo Molnar]

* Arnaldo Carvalho de Melo <acme@kernel.org> wrote:

> Em Tue, Dec 16, 2014 at 01:57:02PM -0300, Arnaldo Carvalho de Melo escreveu:
> > Hi Ingo,
> > 
> > 	Please consider pulling,
> > 
> > - Arnaldo
> > 
> > The following changes since commit 41e950c033b7df997d4b38653efe6554be9b96a7:
> > 
> >   Merge tag 'perf-core-for-mingo' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/urgent (2014-12-12 09:09:52 +0100)
> > 
> > are available in the git repository at:
> > 
> > 
> >   git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux.git tags/perf-core-for-mingo
> 
> Ingo,
> 
> 	Please consider pulling the perf-core-for-mingo-2 signed tag
> instead, it is exactly the same content modulo removal of some
> whitespaces at the end of a few lines in tools/perf/Makefile.perf that
> Jiri found while reading those patches,

Pulled, thanks a lot!

	Ingo