Skip to content

cloud-kernel
cloud-kernel

openanolis / cloud-kernel
1 年多 前同步成功

通知 160
Star 36
Fork 7
cloud-kernel
cloud-kernel
切换分支/标签
  1. 20 4月, 2017 2 次提交
  3. 14 3月, 2017 1 次提交
    • H
      perf tools: Add PERF_RECORD_NAMESPACES to include namespaces related info · f3b3614a
      由 Hari Bathini 提交于 
      Introduce a new option to record PERF_RECORD_NAMESPACES events emitted
by the kernel when fork, clone, setns or unshare are invoked. And update
perf-record documentation with the new option to record namespace
events.

Committer notes:

Combined it with a later patch to allow printing it via 'perf report -D'
and be able to test the feature introduced in this patch. Had to move
here also perf_ns__name(), that was introduced in another later patch.

Also used PRIu64 and PRIx64 to fix the build in some enfironments wrt:

  util/event.c:1129:39: error: format '%lx' expects argument of type 'long unsigned int', but argument 6 has type 'long long unsigned int' [-Werror=format=]
     ret  += fprintf(fp, "%u/%s: %lu/0x%lx%s", idx
                                         ^
Testing it:

  # perf record --namespaces -a
  ^C[ perf record: Woken up 1 times to write data ]
  [ perf record: Captured and wrote 1.083 MB perf.data (423 samples) ]
  #
  # perf report -D
  <SNIP>
  3 2028902078892 0x115140 [0xa0]: PERF_RECORD_NAMESPACES 14783/14783 - nr_namespaces: 7
                [0/net: 3/0xf0000081, 1/uts: 3/0xeffffffe, 2/ipc: 3/0xefffffff, 3/pid: 3/0xeffffffc,
                 4/user: 3/0xeffffffd, 5/mnt: 3/0xf0000000, 6/cgroup: 3/0xeffffffb]

  0x1151e0 [0x30]: event: 9
  .
  . ... raw event: size 48 bytes
  .  0000:  09 00 00 00 02 00 30 00 c4 71 82 68 0c 7f 00 00  ......0..q.h....
  .  0010:  a9 39 00 00 a9 39 00 00 94 28 fe 63 d8 01 00 00  .9...9...(.c....
  .  0020:  03 00 00 00 00 00 00 00 ce c4 02 00 00 00 00 00  ................
  <SNIP>
        NAMESPACES events:          1
  <SNIP>
  #
Signed-off-by: NHari Bathini <hbathini@linux.vnet.ibm.com>
Acked-by: NJiri Olsa <jolsa@kernel.org>
Tested-by: NArnaldo Carvalho de Melo <acme@redhat.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Alexei Starovoitov <ast@fb.com>
Cc: Ananth N Mavinakayanahalli <ananth@linux.vnet.ibm.com>
Cc: Aravinda Prasad <aravinda@linux.vnet.ibm.com>
Cc: Brendan Gregg <brendan.d.gregg@gmail.com>
Cc: Daniel Borkmann <daniel@iogearbox.net>
Cc: Eric Biederman <ebiederm@xmission.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Sargun Dhillon <sargun@sargun.me>
Cc: Steven Rostedt <rostedt@goodmis.org>
Link: http://lkml.kernel.org/r/148891930386.25309.18412039920746995488.stgit@hbathini.in.ibm.comSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      f3b3614a
  5. 04 3月, 2017 1 次提交
  7. 15 2月, 2017 1 次提交
  9. 14 2月, 2017 1 次提交
  11. 12 1月, 2017 1 次提交
  13. 15 11月, 2016 1 次提交
    • J
      perf report: Add branch flag to callchain cursor node · 410024db
      由 Jin Yao 提交于 
      Since the branch ip has been added to call stack for easier browsing,
this patch adds more branch information. For example, add a flag to
indicate if this ip is a branch, and also add with the branch flag.

Then we can know if the cursor node represents a branch and know what
the branch flag it has.

The branch history code has a loop detection pass that removes loops. It
would be nice for knowing how many loops were removed then in next
steps, we can compute out the average number of iterations.

For example:

Before remove_loops(),
entry0: from = 0x100, to = 0x200
entry1: from = 0x300, to = 0x250
entry2: from = 0x300, to = 0x250
entry3: from = 0x300, to = 0x250
entry4: from = 0x700, to = 0x800

After remove_loops()
entry0: from = 0x100, to = 0x200
entry1: from = 0x300, to = 0x250
entry2: from = 0x700, to = 0x800

The original entry2 and entry3 are removed. So the number of iterations
(from = 0x300, to = 0x250) is equal to removed number + 1 (2 + 1).

iterations = removed number + 1;
average iteractions = Sum(iteractions) / number of samples

This formula ignores other cases, for example, iterations cross multiple
buffers and one buffer contains 2+ loops. Because in practice, it's good
enough.
Signed-off-by: NYao Jin <yao.jin@linux.intel.com>
Acked-by: NAndi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Kan Liang <kan.liang@intel.com>
Cc: Linux-kernel@vger.kernel.org
Cc: Yao Jin <yao.jin@linux.intel.com>
Link: http://lkml.kernel.org/n/1477876794-30749-2-git-send-email-yao.jin@linux.intel.com
[ Renamed 'iter' to 'nr_loop_iter' for clarity ]
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      410024db
  15. 03 10月, 2016 1 次提交
    • A
      perf tools: Experiment with cppcheck · 18ef15c6
      由 Arnaldo Carvalho de Melo 提交于 
      Experimenting a bit using cppcheck[1], a static checker brought to my
attention by Colin, reducing the scope of some variables, reducing the
line of source code lines in the process:

  $ cppcheck --enable=style tools/perf/util/thread.c
  Checking tools/perf/util/thread.c...
  [tools/perf/util/thread.c:17]: (style) The scope of the variable 'leader' can be reduced.
  [tools/perf/util/thread.c:133]: (style) The scope of the variable 'err' can be reduced.
  [tools/perf/util/thread.c:273]: (style) The scope of the variable 'err' can be reduced.

Will continue later, but these are already useful, keep them.

1: https://sourceforge.net/p/cppcheck/wiki/Home/

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Colin Ian King <colin.king@canonical.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-ixws7lbycihhpmq9cc949ti6@git.kernel.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      18ef15c6
  17. 05 9月, 2016 2 次提交
  19. 27 7月, 2016 1 次提交
  21. 22 6月, 2016 1 次提交
  23. 07 6月, 2016 1 次提交
    • H
      perf unwind: Move unwind__prepare_access from thread_new into thread__insert_map · 8132a2a8
      由 He Kuang 提交于 
      To determine the libunwind methods to use, we should get the
32bit/64bit information from maps of a thread. When a thread is newly
created, the information is not prepared. This patch moves
unwind__prepare_access() into thread__insert_map() so we can get the
information we need from maps. Meanwhile, let thread__insert_map()
return value and show messages on error.
Signed-off-by: NHe Kuang <hekuang@huawei.com>
Acked-by: NJiri Olsa <jolsa@kernel.org>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Ekaterina Tumanova <tumanova@linux.vnet.ibm.com>
Cc: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Kan Liang <kan.liang@intel.com>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Sukadev Bhattiprolu <sukadev@linux.vnet.ibm.com>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/r/1464924803-22214-5-git-send-email-hekuang@huawei.comSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      8132a2a8
  25. 20 5月, 2016 3 次提交
  27. 17 5月, 2016 2 次提交
  29. 06 5月, 2016 3 次提交
    • C
      perf callchain: Fix incorrect ordering of entries · 9919a65e
      由 Chris Phlipot 提交于 
      The existing implementation of thread__resolve_callchain, under certain
circumstances, can assemble callchain entries in the incorrect order.

The callchain entries are resolved incorrectly for a sample when all of
the following conditions are met:

1. callchain_param.order is set to ORDER_CALLER

2. thread__resolve_callchain_sample is able to resolve callchain entries
   for the sample.

3. unwind__get_entries is also able to resolve callchain entries for the
   sample.

The fix is accomplished by reversing the order in which
thread__resolve_callchain_sample and unwind__get_entries are called when
callchain_param.order is set to ORDER_CALLER.

Unwind specific code from thread__resolve_callchain is also moved into a
new static function to improve readability of the fix.

How to Reproduce the Existing Bug:

Modifying perf script to print call trees in the opposite order or
applying the remaining patches from this series and comparing the
results output from export-to-postgtresql.py are the easiest ways to see
the bug, however it can still be seen in current builds using perf
report.

Here is how i can reproduce the bug using perf report:

  # perf record --call-graph=dwarf stress -c 1 -t 5

when i run this command:

  # perf report --call-graph=flat,0,0,callee

This callchain, containing kernel (handle_irq_event, etc) and userspace
samples (__libc_start_main, etc) is contained in the output, which looks
correct (callee order):

                gen8_irq_handler
                handle_irq_event_percpu
                handle_irq_event
                handle_edge_irq
                handle_irq
                do_IRQ
                ret_from_intr
                __random
                rand
                0x558f2a04dded
                0x558f2a04c774
                __libc_start_main
                0x558f2a04dcd9

Now run this command using caller order:

  # perf report --call-graph=flat,0,0,caller

It is expected to see the exact reverse of the above when using caller
order (with "0x558f2a04dcd9" at the top and "gen8_irq_handler" at the
bottom) in the output, but it is nowhere to be found.

instead you see this:

                ret_from_intr
                do_IRQ
                handle_irq
                handle_edge_irq
                handle_irq_event
                handle_irq_event_percpu
                gen8_irq_handler
                0x558f2a04dcd9
                __libc_start_main
                0x558f2a04c774
                0x558f2a04dded
                rand
                __random

Notice how internally the kernel symbols are reversed and the user space
symbols are reversed, but the kernel symbols still appear above the user
space symbols.

if this patch is applied and perf script is re-run, you will see the
expected output (with "0x558f2a04dcd9" at the top and "gen8_irq_handler"
at the bottom):

                0x558f2a04dcd9
                __libc_start_main
                0x558f2a04c774
                0x558f2a04dded
                rand
                __random
                ret_from_intr
                do_IRQ
                handle_irq
                handle_edge_irq
                handle_irq_event
                handle_irq_event_percpu
                gen8_irq_handler
Signed-off-by: NChris Phlipot <cphlipot0@gmail.com>
Tested-by: NArnaldo Carvalho de Melo <acme@redhat.com>
Acked-by: NJiri Olsa <jolsa@kernel.org>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1461831551-12213-2-git-send-email-cphlipot0@gmail.comSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      9919a65e
    • J
      perf hists: Move sort__has_parent into struct perf_hpp_list · de7e6a7c
      由 Jiri Olsa 提交于 
      Now we have sort dimensions private for struct hists, we need to make
dimension booleans hists specific as well.

Moving sort__has_parent into struct perf_hpp_list.
Signed-off-by: NJiri Olsa <jolsa@kernel.org>
Cc: David Ahern <dsahern@gmail.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1462276488-26683-3-git-send-email-jolsa@kernel.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      de7e6a7c
    • A
      perf machine: Introduce number of threads member · d2c11034
      由 Arnaldo Carvalho de Melo 提交于 
      To be used, for instance, for pre-allocating an rb_tree array for
sorting by other keys besides the current pid one.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Milian Wolff <milian.wolff@kdab.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-ja0ifkwue7ttjhbwijn6g6eu@git.kernel.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      d2c11034
  31. 27 4月, 2016 1 次提交
  33. 19 4月, 2016 1 次提交
  35. 18 4月, 2016 2 次提交
  37. 15 4月, 2016 1 次提交
    • A
      perf callchain: Start moving away from global per thread cursors · 91d7b2de
      由 Arnaldo Carvalho de Melo 提交于 
      The recent perf_evsel__fprintf_callchain() move to evsel.c added several
new symbol requirements to the python binding, for instance:

  # perf test -v python
  16: Try 'import perf' in python, checking link problems      :
  --- start ---
  test child forked, pid 18030
  Traceback (most recent call last):
    File "<stdin>", line 1, in <module>
  ImportError: /tmp/build/perf/python/perf.so: undefined symbol:
  callchain_cursor
  test child finished with -1
  ---- end ----
  Try 'import perf' in python, checking link problems: FAILED!
  #

This would require linking against callchain.c to access to the global
callchain_cursor variables.

Since lots of functions already receive as a parameter a
callchain_cursor struct pointer, make that be the case for some more
function so that we can start phasing out usage of yet another global
variable.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-djko3097eyg2rn66v2qcqfvn@git.kernel.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      91d7b2de
  39. 23 3月, 2016 1 次提交
  41. 14 12月, 2015 1 次提交
  43. 11 12月, 2015 1 次提交
    • W
      perf tools: Clear struct machine during machine__init() · 93b0ba3c
      由 Wang Nan 提交于 
      There are so many test cases use stack allocated 'struct machine'.
Including:
  test__hists_link
  test__hists_filter
  test__mmap_thread_lookup
  test__thread_mg_share
  test__hists_output
  test__hists_cumulate

Also, in non-test code (for example, machine__new_host()) there are
code use 'malloc()' to alloc struct machine.

These are dangerous operations, cause some tests fail or hung in
machines__exit(). For example, in

 machines__exit ->
   machine__destroy_kernel_maps ->
     map_groups__remove ->
       maps__remove ->
         pthread_rwlock_wrlock

a incorrectly initialized lock causes unintended behavior.

This patch memset(0) that structure in machine__init() to ensure all
fields in 'struct machine' are initialized to zero.
Signed-off-by: NWang Nan <wangnan0@huawei.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Zefan Li <lizefan@huawei.com>
Cc: pi3orama@163.com
Link: http://lkml.kernel.org/r/1449541544-67621-17-git-send-email-wangnan0@huawei.com
[ Use memset, see 'man bzero' ]
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      93b0ba3c
  45. 10 12月, 2015 1 次提交
  47. 08 12月, 2015 1 次提交
  49. 27 11月, 2015 2 次提交
    • W
      perf machine: Adjust dso->long_name for offline module · c03d5184
      由 Wang Nan 提交于 
      Something unexpected may happen if copy statically linked perf to a
production environment:

  # ./perf probe -m ./mymodule.ko my_func
  [mymodule] with build id 326ab42550ef3d24944f53c817533728367effeb not found, continuing without symbols
  Failed to find symbol my_func in /home/wangnan/kmodule/mymodule.ko
    Error: Failed to add events.
  # ./perf buildid-cache -a ./mymodule.ko
  # ./perf probe -m ./mymodule.ko my_func
  Added new event:
    probe:my_func        (on my_func in /home/wangnan/kmodule/mymodule.ko)

  You can now use it in all perf tools, such as:

  	perf record -e probe:my_func -aR sleep 1

Where:

  # ldd ./perf
 	not a dynamic executable
  # strace -e open ./perf probe -m ./mymodule.ko my_func
  ...
  open("/home/wangnan/kmodule/mymodule.ko", O_RDONLY) = 3
  open("/home/wangnan/kmodule/../lib64/elfutils/libebl_x86_64.so", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory)
  ...
  open("/lib64/tls/libebl_x86_64.so", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory)
  open("/lib64/libebl_x86_64.so", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory)
  open("/usr/lib64/tls/libebl_x86_64.so", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory)
  open("/usr/lib64/libebl_x86_64.so", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory)
  open("[mymodule]", O_RDONLY)            = -1 ENOENT (No such file or directory)
  open("/home/wangnan/.debug/.build-id/32/6ab42550ef3d24944f53c817533728367effeb", O_RDONLY) = -1 ENOENT (No such file or directory)
  open("[mymodule]", O_RDONLY)            = -1 ENOENT (No such file or directory)

In the above example, probe fails before we put the module into
buildid-cache. However, user would expect it success in both case
because perf is able to find probe points actually.

The reason is because perf won't utilize module's full path if it failed
to open debuginfo. In:

     convert_to_probe_trace_events ->
        find_probe_trace_events_from_map ->
            get_target_map ->
                kernel_get_module_map ->
                    machine__findnew_module_map ->
                        map_groups__find_by_name

map_groups__find_by_name() is able to find the map of that module, but
this information is found from /proc/module before it knows the real
path of the offline module. Therefore, the map->dso->long_name is set to
something like '[mymodule]', which prevent dso__load() find the real
path of the module file.

In another aspect, if dso__load() can get the offline module through
buildid cache, it can read symble table from that ko. Even if debuginfo
is not available, 'perf probe' can success if the '.symtab' can be
found.

This patch improves machine__findnew_module_map(): when dso->long_name
is leading with '[' (doesn't find path of module when parsing
/proc/modules), fixes it by dso__set_long_name(), so following
dso__load() is possible to find the symbol table.

This patch won't interfere with buildid matching. Here is the test
result:

  # ./perf probe -m ./mymodule.ko my_func
  Added new event:
    probe:my_func        (on my_func in /home/wangnan/kmodule/mymodule.ko)

  You can now use it in all perf tools, such as:

  	perf record -e probe:my_func -aR sleep 1

  # ./perf probe -d '*'
  Removed event: probe:my_func
  # mv ./mymodule.{ko,.bak}
  # mv ./moduleb.ko mymodule.ko
  # ./perf probe -m ./mymodule.ko my_func
  /home/wangnan/kmodule/mymodule.ko with build id 326ab42550ef3d24944f53c817533728367effeb not found, continuing without symbols
  Failed to find symbol my_func in /home/wangnan/kmodule/mymodule.ko
    Error: Failed to add events.

  # ./perf probe -v -m ./mymodule.ko my_func
  probe-definition(0): my_func
  symbol:my_func file:(null) line:0 offset:0 return:0 lazy:(null)
  0 arguments
  Could not open debuginfo. Try to use symbols.
  symsrc__init: build id mismatch for /home/wangnan/kmodule/mymodule.ko.
  /home/wangnan/kmodule/mymodule.ko with build id 326ab42550ef3d24944f53c817533728367effeb not found, continuing without symbols
  Failed to find symbol my_func in /home/wangnan/kmodule/mymodule.ko
    Error: Failed to add events. Reason: No such file or directory (Code: -2)
Signed-off-by: NWang Nan <wangnan0@huawei.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Zefan Li <lizefan@huawei.com>
Cc: pi3orama@163.com
Link: http://lkml.kernel.org/r/1448510397-187965-1-git-send-email-wangnan0@huawei.com
[ Renamed adjust_dso_long_name() do dso__adjust_kmod_long_name() ]
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      c03d5184
    • N
      perf callchain: Honor hide_unresolved · b49a8fe5
      由 Namhyung Kim 提交于 
      If user requested to hide unresolved entries, skip unresolved callchains
as well as hist entries.
Signed-off-by: NNamhyung Kim <namhyung@kernel.org>
Acked-by: NJiri Olsa <jolsa@kernel.org>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Kan Liang <kan.liang@intel.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/r/1448521700-32062-3-git-send-email-namhyung@kernel.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      b49a8fe5
  51. 20 11月, 2015 5 次提交
    • M
      perf machine: Fix machine__findnew_module_map to put dso · 566c69c3
      由 Masami Hiramatsu 提交于 
      Fix machine__findnew_module_map to drop the reference to the dso because
it is already referenced by both machine__findnew_module_dso() and
map__new2().

Refcnt debugger shows:

  ==== [1] ====
  Unreclaimed dso: 0x1ffd980
  Refcount +1 => 1 at
    ./perf(dso__new+0x1ff) [0x4a62df]
    ./perf(__dsos__addnew+0x29) [0x4a6e19]
    ./perf() [0x4b8b91]
    ./perf(modules__parse+0xfc) [0x4a9d5c]
    ./perf() [0x4b8460]
    ./perf(machine__create_kernel_maps+0x150) [0x4bb550]
    ./perf(machine__new_host+0xfa) [0x4bb75a]
    ./perf(init_probe_symbol_maps+0x93) [0x506623]
    ./perf() [0x455ffa]
    ./perf(cmd_probe+0x6c) [0x4566bc]
    ./perf() [0x47abc5]
    ./perf(main+0x610) [0x421f90]
    /lib64/libc.so.6(__libc_start_main+0xf5) [0x7f1345a8eaf5]
    ./perf() [0x4220a9]

This map_groups__insert(0x4b8b91) already gets a reference to the new
dso:

  ----
  eu-addr2line -e ./perf -f 0x4b8b91
  map_groups__insert inlined at util/machine.c:586 in
  machine__create_module
  util/map.h:207
  ----

So this dso refcnt will be released when map_groups gets released.

  [snip]
  Refcount +1 => 2 at
    ./perf(dso__get+0x34) [0x4a65f4]
    ./perf() [0x4b8b35]
    ./perf(modules__parse+0xfc) [0x4a9d5c]
    ./perf() [0x4b8460]
    ./perf(machine__create_kernel_maps+0x150) [0x4bb550]
    ./perf(machine__new_host+0xfa) [0x4bb75a]
    ./perf(init_probe_symbol_maps+0x93) [0x506623]
    ./perf() [0x455ffa]
    ./perf(cmd_probe+0x6c) [0x4566bc]
    ./perf() [0x47abc5]
    ./perf(main+0x610) [0x421f90]
    /lib64/libc.so.6(__libc_start_main+0xf5) [0x7f1345a8eaf5]
    ./perf() [0x4220a9]

Here, machine__findnew_module_dso(0x4b8b35) gets the dso (and stores it
in a local variable):

  ----
  # eu-addr2line -e ./perf -f 0x4b8b35
  machine__findnew_module_dso inlined at util/machine.c:578 in
  machine__create_module
  util/machine.c:514
  ----

  Refcount +1 => 3 at
    ./perf(dso__get+0x34) [0x4a65f4]
    ./perf(map__new2+0x76) [0x4be1c6]
    ./perf() [0x4b8b4f]
    ./perf(modules__parse+0xfc) [0x4a9d5c]
    ./perf() [0x4b8460]
    ./perf(machine__create_kernel_maps+0x150) [0x4bb550]
    ./perf(machine__new_host+0xfa) [0x4bb75a]
    ./perf(init_probe_symbol_maps+0x93) [0x506623]
    ./perf() [0x455ffa]
    ./perf(cmd_probe+0x6c) [0x4566bc]
    ./perf() [0x47abc5]
    ./perf(main+0x610) [0x421f90]
    /lib64/libc.so.6(__libc_start_main+0xf5) [0x7f1345a8eaf5]
    ./perf() [0x4220a9]

But also map__new2() gets the dso which will be put when the map is
released.

So, we have to drop the constructor reference obtained in
machine__findnew_module_dso().
Signed-off-by: NMasami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20151118064035.30709.58824.stgit@localhost.localdomainSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      566c69c3
    • M
      perf tools: Fix machine__create_kernel_maps to put kernel dso refcount · 1154c957
      由 Masami Hiramatsu 提交于 
      Fix machine__create_kernel_maps() to put kernel dso because the dso has
been gotten via __machine__create_kernel_maps().

Refcnt debugger shows:
  ==== [0] ====
  Unreclaimed dso: 0x3036ab0
  Refcount +1 => 1 at
    ./perf(dso__new+0x1ff) [0x4a62df]
    ./perf(__dsos__addnew+0x29) [0x4a6e19]
    ./perf(dsos__findnew+0xd1) [0x4a7181]
    ./perf(machine__findnew_kernel+0x27) [0x4a5e17]
    ./perf() [0x4b8cf2]
    ./perf(machine__create_kernel_maps+0x28) [0x4bb428]
    ./perf(machine__new_host+0xfa) [0x4bb74a]
    ./perf(init_probe_symbol_maps+0x93) [0x506613]
    ./perf() [0x455ffa]
    ./perf(cmd_probe+0x6c) [0x4566bc]
    ./perf() [0x47abc5]
    ./perf(main+0x610) [0x421f90]
    /lib64/libc.so.6(__libc_start_main+0xf5) [0x7ffa6809eaf5]
    ./perf() [0x4220a9]
  [snip]
  Refcount +1 => 2 at
    ./perf(dsos__findnew+0x7e) [0x4a712e]
    ./perf(machine__findnew_kernel+0x27) [0x4a5e17]
    ./perf() [0x4b8cf2]
    ./perf(machine__create_kernel_maps+0x28) [0x4bb428]
    ./perf(machine__new_host+0xfa) [0x4bb74a]
    ./perf(init_probe_symbol_maps+0x93) [0x506613]
    ./perf() [0x455ffa]
    ./perf(cmd_probe+0x6c) [0x4566bc]
    ./perf() [0x47abc5]
    ./perf(main+0x610) [0x421f90]
    /lib64/libc.so.6(__libc_start_main+0xf5) [0x7ffa6809eaf5]
    ./perf() [0x4220a9]
  [snip]
  Refcount -1 => 1 at
    ./perf(dso__put+0x2f) [0x4a664f]
    ./perf(machine__delete+0xfe) [0x4b93ee]
    ./perf(exit_probe_symbol_maps+0x28) [0x5066b8]
    ./perf() [0x45628a]
    ./perf(cmd_probe+0x6c) [0x4566bc]
    ./perf() [0x47abc5]
    ./perf(main+0x610) [0x421f90]
    /lib64/libc.so.6(__libc_start_main+0xf5) [0x7ffa6809eaf5]
    ./perf() [0x4220a9]

Actually, dsos__findnew gets the dso before returning it, so the dso
user (in this case machine__create_kernel_maps) has to put the dso after
used.
Signed-off-by: NMasami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20151118064033.30709.98954.stgit@localhost.localdomainSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      1154c957
    • M
      perf machine: Fix to destroy kernel maps when machine exits · ebe9729c
      由 Masami Hiramatsu 提交于 
      Actually machine__exit forgot to call machine__destroy_kernel_maps.

This fixes some memory leaks on map as below.

Without this fix.
  ----
  ./perf probe vfs_read
  Added new event:
    probe:vfs_read       (on vfs_read)

  You can now use it in all perf tools, such as:

          perf record -e probe:vfs_read -aR sleep 1

  REFCNT: BUG: Unreclaimed objects found.
  REFCNT: Total 4 objects are not reclaimed.
     To see all backtraces, rerun with -v option
  ----
With this fix.
  ----
  ./perf probe vfs_read
  Added new event:
    probe:vfs_read       (on vfs_read)

  You can now use it in all perf tools, such as:

          perf record -e probe:vfs_read -aR sleep 1

  REFCNT: BUG: Unreclaimed objects found.
  REFCNT: Total 2 objects are not reclaimed.
     To see all backtraces, rerun with -v option
  ----
Signed-off-by: NMasami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20151118064024.30709.43577.stgit@localhost.localdomainSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      ebe9729c
    • M
      perf machine: Fix machine__destroy_kernel_maps to drop vmlinux_maps references · e96e4078
      由 Masami Hiramatsu 提交于 
      Fix machine__destroy_kernel_maps() to drop vmlinux_maps references
before filling it with NULL.

Refcnt debugger shows
  ==== [1] ====
  Unreclaimed map: 0x36b1070
  Refcount +1 => 1 at
    ./perf(map__new2+0xb5) [0x4bdec5]
    ./perf(machine__create_kernel_maps+0x72) [0x4bb152]
    ./perf(machine__new_host+0xfa) [0x4bb41a]
    ./perf(init_probe_symbol_maps+0x93) [0x5062d3]
    ./perf() [0x455ffa]
    ./perf(cmd_probe+0x6c) [0x4566bc]
    ./perf() [0x47abc5]
    ./perf(main+0x610) [0x421f90]
    /lib64/libc.so.6(__libc_start_main+0xf5) [0x7f1fc9fc4af5]
    ./perf() [0x4220a9]
  Refcount +1 => 2 at
    ./perf(maps__insert+0x9a) [0x4bfd6a]
    ./perf(machine__create_kernel_maps+0xc3) [0x4bb1a3]
    ./perf(machine__new_host+0xfa) [0x4bb41a]
    ./perf(init_probe_symbol_maps+0x93) [0x5062d3]
    ./perf() [0x455ffa]
    ./perf(cmd_probe+0x6c) [0x4566bc]
    ./perf() [0x47abc5]
    ./perf(main+0x610) [0x421f90]
    /lib64/libc.so.6(__libc_start_main+0xf5) [0x7f1fc9fc4af5]
    ./perf() [0x4220a9]
  Refcount -1 => 1 at
    ./perf(map_groups__exit+0x94) [0x4bea74]
    ./perf(machine__delete+0x3d) [0x4b91fd]
    ./perf(exit_probe_symbol_maps+0x28) [0x506378]
    ./perf() [0x45628a]
    ./perf(cmd_probe+0x6c) [0x4566bc]
    ./perf() [0x47abc5]
    ./perf(main+0x610) [0x421f90]
    /lib64/libc.so.6(__libc_start_main+0xf5) [0x7f1fc9fc4af5]
    ./perf() [0x4220a9]

map__new2() returns map with refcnt = 1, and also map_groups__insert
gets it again in__machine__create_kernel_maps().

machine__destroy_kernel_maps() calls map_groups__remove() to
decrement the refcnt, but before decrement it again (corresponding
to map__new2), it makes vmlinux_maps[type] = NULL. And this may
cause a refcnt leak.
Signed-off-by: NMasami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20151118064022.30709.3897.stgit@localhost.localdomainSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      e96e4078
    • M
      perf machine: Fix machine__findnew_module_map to put registered map · 9afcb420
      由 Masami Hiramatsu 提交于 
      Fix machine object to drop the reference to the map object after it
inserted it into machine->kmaps.

refcnt debugger shows what happened:
  ----
  ==== [2] ====
  Unreclaimed map: 0x346f750
  Refcount +1 => 1 at
    ./perf(map__new2+0xb5) [0x4bdea5]
    ./perf() [0x4b8aaf]
    ./perf(modules__parse+0xfc) [0x4a9cbc]
    ./perf() [0x4b83c0]
    ./perf(machine__create_kernel_maps+0x148) [0x4bb208]
    ./perf(machine__new_host+0xfa) [0x4bb3fa]
    ./perf(init_probe_symbol_maps+0x93) [0x5062b3]
    ./perf() [0x455ffa]
    ./perf(cmd_probe+0x6c) [0x4566bc]
    ./perf() [0x47abc5]
    ./perf(main+0x610) [0x421f90]
    /lib64/libc.so.6(__libc_start_main+0xf5) [0x7f5373899af5]
    ./perf() [0x4220a9]
  Refcount +1 => 2 at
    ./perf(maps__insert+0x9a) [0x4bfd4a]
    ./perf() [0x4b8acb]
    ./perf(modules__parse+0xfc) [0x4a9cbc]
    ./perf() [0x4b83c0]
    ./perf(machine__create_kernel_maps+0x148) [0x4bb208]
    ./perf(machine__new_host+0xfa) [0x4bb3fa]
    ./perf(init_probe_symbol_maps+0x93) [0x5062b3]
    ./perf() [0x455ffa]
    ./perf(cmd_probe+0x6c) [0x4566bc]
    ./perf() [0x47abc5]
    ./perf(main+0x610) [0x421f90]
    /lib64/libc.so.6(__libc_start_main+0xf5) [0x7f5373899af5]
    ./perf() [0x4220a9]
  Refcount -1 => 1 at
    ./perf(map_groups__exit+0x94) [0x4bea54]
    ./perf(machine__delete+0x3d) [0x4b91ed]
    ./perf(exit_probe_symbol_maps+0x28) [0x506358]
    ./perf() [0x45628a]
    ./perf(cmd_probe+0x6c) [0x4566bc]
    ./perf() [0x47abc5]
    ./perf(main+0x610) [0x421f90]
    /lib64/libc.so.6(__libc_start_main+0xf5) [0x7f5373899af5]
    ./perf() [0x4220a9]
  ----

This pattern clearly shows that the refcnt of the map is acquired twice
by map__new2 and maps__insert but released onlu once at
map_groups__exit, when we purge its maps rbtree.

Since maps__insert already reference counted the map, we have to drop
the constructor (map__new2) reference count right after inserting it.

These happened in machine__findnew_module_map, as below.

  ----
  # eu-addr2line -e ./perf -f 0x4b8aaf
  machine__findnew_module_map inlined at util/machine.c:1046
  in machine__create_module
  util/machine.c:582
  # eu-addr2line -e ./perf -f 0x4b8acb
  map_groups__insert inlined at util/machine.c:585
  in machine__create_module
  util/map.h:208
  ----

(note that both are at util/machine.c:58X which is
 machine__findnew_module_map)
Signed-off-by: NMasami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20151118064020.30709.40499.stgit@localhost.localdomainSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      9afcb420
  53. 13 11月, 2015 1 次提交
    • A
      perf symbols: Fix dso lookup by long name and missing buildids · e266a753
      由 Adrian Hunter 提交于 
      Commit 4598a0a6 ("perf symbols: Improve DSO long names lookup speed
with rbtree") Added a tree to lookup dsos by long name.  That tree gets
corrupted whenever a dso long name is changed because the tree is not
updated.

One effect of that is buildid-list does not work with the 'with-hits'
option because dso lookup fails and results in two structs for the same
dso.  The first has the buildid but no hits, the second has hits but no
buildid. e.g.

Before:

  $ tools/perf/perf record ls
  arch     certs    CREDITS  Documentation  firmware  include
  ipc      Kconfig  lib      Makefile       net       REPORTING-BUGS
  scripts  sound    usr      block          COPYING   crypto
  drivers  fs       init     Kbuild         kernel    MAINTAINERS
  mm       README   samples  security       tools     virt
  [ perf record: Woken up 1 times to write data ]
  [ perf record: Captured and wrote 0.012 MB perf.data (11 samples) ]
  $ tools/perf/perf buildid-list
  574da826c66538a8d9060d393a8866289bd06005 [kernel.kallsyms]
  30c94dc66a1fe95180c3d68d2b89e576d5ae213c /lib/x86_64-linux-gnu/libc-2.19.so
  $ tools/perf/perf buildid-list -H
  574da826c66538a8d9060d393a8866289bd06005 [kernel.kallsyms]
  0000000000000000000000000000000000000000 /lib/x86_64-linux-gnu/libc-2.19.so

After:

  $ tools/perf/perf buildid-list -H
  574da826c66538a8d9060d393a8866289bd06005 [kernel.kallsyms]
  30c94dc66a1fe95180c3d68d2b89e576d5ae213c /lib/x86_64-linux-gnu/libc-2.19.so

The fix is to record the root of the tree on the dso so that
dso__set_long_name() can update the tree when the long name changes.
Signed-off-by: NAdrian Hunter <adrian.hunter@intel.com>
Tested-by: NArnaldo Carvalho de Melo <acme@redhat.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Don Zickus <dzickus@redhat.com>
Cc: Douglas Hatch <doug.hatch@hp.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Scott J Norton <scott.norton@hp.com>
Cc: Waiman Long <Waiman.Long@hp.com>
Fixes: 4598a0a6 ("perf symbols: Improve DSO long names lookup speed with rbtree")
Link: http://lkml.kernel.org/r/1447408112-1920-2-git-send-email-adrian.hunter@intel.comSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      e266a753