linux.git/tools/perf/util/dso.c, branch v3.19

perf tools: Add gzip decompression support for kernel module

2014-11-05T13:11:26+00:00

Now my Archlinux box shows module symbols correctly.

Before:
  $ perf report --stdio
  Failed to open /tmp/perf-3477.map, continuing without symbols
  no symbols found in /usr/bin/date, maybe install a debug package?
  No kallsyms or vmlinux with build-id 7b4ea0a49ae2111925857099aaf05c3246ff33e0 was found
  [drm] with build id 7b4ea0a49ae2111925857099aaf05c3246ff33e0 not found, continuing without symbols
  No kallsyms or vmlinux with build-id edd931629094b660ca9dec09a1b635c8d87aa2ee was found
  [jbd2] with build id edd931629094b660ca9dec09a1b635c8d87aa2ee not found, continuing without symbols
  No kallsyms or vmlinux with build-id a7b1eada671c34933e5610bb920b2ca4945a82c3 was found
  [ext4] with build id a7b1eada671c34933e5610bb920b2ca4945a82c3 not found, continuing without symbols
  No kallsyms or vmlinux with build-id d69511fa3e5840e770336ef45b06c83fef8d74e3 was found
  [scsi_mod] with build id d69511fa3e5840e770336ef45b06c83fef8d74e3 not found, continuing without symbols
  No kallsyms or vmlinux with build-id af0430af13461af058770ee9b87afc07922c2e77 was found
  [libata] with build id af0430af13461af058770ee9b87afc07922c2e77 not found, continuing without symbols
  No kallsyms or vmlinux with build-id aaeedff8160ce631a5f0333591c6ff291201d29f was found
  [libahci] with build id aaeedff8160ce631a5f0333591c6ff291201d29f not found, continuing without symbols
  No kallsyms or vmlinux with build-id c57907712becaf662dc4981824bb372c0441d605 was found
  [mac80211] with build id c57907712becaf662dc4981824bb372c0441d605 not found, continuing without symbols
  No kallsyms or vmlinux with build-id e0589077cc0ec8c3e4c40eb9f2d9e69d236bee8f was found
  [iwldvm] with build id e0589077cc0ec8c3e4c40eb9f2d9e69d236bee8f not found, continuing without symbols
  No kallsyms or vmlinux with build-id 2d86086bf136bf374a2f029cf85a48194f9b950b was found
  [cfg80211] with build id 2d86086bf136bf374a2f029cf85a48194f9b950b not found, continuing without symbols
  No kallsyms or vmlinux with build-id 4493c48599bdb3d91d0f8db5150e0be33fdd9221 was found
  [iwlwifi] with build id 4493c48599bdb3d91d0f8db5150e0be33fdd9221 not found, continuing without symbols
  ...
  #
  # Overhead  Command          Shared Object            Symbol
  # ........  ...............  .......................  ........................................................
  #
       0.03%  swapper          [ext4]                   [k] 0x000000000000fe2e
       0.03%  swapper          [kernel.kallsyms]        [k] account_entity_enqueue
       0.03%  swapper          [ext4]                   [k] 0x000000000000fc2b
       0.03%  irq/50-iwlwifi   [iwlwifi]                [k] 0x000000000000200b
       0.03%  swapper          [kernel.kallsyms]        [k] ktime_add_safe
       0.03%  swapper          [kernel.kallsyms]        [k] elv_completed_request
       0.03%  swapper          [libata]                 [k] 0x0000000000003997
       0.03%  swapper          [libahci]                [k] 0x0000000000001f25
       0.03%  swapper          [kernel.kallsyms]        [k] rb_next
       0.03%  swapper          [kernel.kallsyms]        [k] blk_finish_request
       0.03%  swapper          [ext4]                   [k] 0x0000000000010248
       0.00%  perf             [kernel.kallsyms]        [k] native_write_msr_safe

After:
  $ perf report --stdio
  Failed to open /tmp/perf-3477.map, continuing without symbols
  no symbols found in /usr/bin/tr, maybe install a debug package?
  ...
  #
  # Overhead  Command          Shared Object                Symbol
  # ........  ...............  ...........................  ......................................................
  #

       0.04%  kworker/u16:3    [ext4]                       [k] ext4_read_block_bitmap
       0.03%  kworker/u16:0    [mac80211]                   [k] ieee80211_sta_reset_beacon_monitor
       0.02%  irq/50-iwlwifi   [mac80211]                   [k] ieee80211_get_bssid
       0.02%  firefox          [e1000e]                     [k] __ew32_prepare
       0.02%  swapper          [libahci]                    [k] ahci_handle_port_interrupt
       0.02%  emacs            libglib-2.0.so.0.4000.0      [.] g_mutex_unlock
       0.02%  swapper          [e1000e]                     [k] e1000_clean_tx_irq
       0.02%  dwm              [kernel.kallsyms]            [k] __schedule
       0.02%  gnome-terminal-  [vdso]                       [.] __vdso_clock_gettime
       0.02%  swapper          [e1000e]                     [k] e1000_alloc_rx_buffers
       0.02%  irq/50-iwlwifi   [mac80211]                   [k] ieee80211_rx
       0.01%  firefox          [vdso]                       [.] __vdso_gettimeofday
       0.01%  irq/50-iwlwifi   [iwlwifi]                    [k] iwl_pcie_rxq_restock.part.13

Signed-off-by: Namhyung Kim 
Acked-by: Jiri Olsa 
Cc: Jiri Olsa 
Cc: Peter Zijlstra 
Cc: Ingo Molnar 
Cc: Paul Mackerras 
Cc: Namhyung Kim 
Cc: Adrian Hunter 
Cc: David Ahern 
Cc: Stephane Eranian 
Link: http://lkml.kernel.org/r/87h9yexshi.fsf@sejong.aot.lge.com
Signed-off-by: Arnaldo Carvalho de Melo

perf symbols: Preparation for compressed kernel module support

2014-11-04T13:15:53+00:00

This patch adds basic support to handle compressed kernel module as some
distro (such as Archlinux) carries on it now.  The actual work using
compression library will be added later.

Signed-off-by: Namhyung Kim 
Acked-by: Jiri Olsa 
Cc: Adrian Hunter 
Cc: David Ahern 
Cc: Ingo Molnar 
Cc: Jiri Olsa 
Cc: Namhyung Kim 
Cc: Paul Mackerras 
Cc: Peter Zijlstra 
Cc: Stephane Eranian 
Link: http://lkml.kernel.org/r/1415063674-17206-2-git-send-email-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo

perf symbols: Improve DSO long names lookup speed with rbtree

2014-10-01T17:39:57+00:00

With workload that spawns and destroys many threads and processes, it
was found that perf-mem could took a long time to post-process the perf
data after the target workload had completed its operation.

The performance bottleneck was found to be the lookup and insertion of
the new DSO structures (thousands of them in this case).

In a dual-socket Ivy-Bridge E7-4890 v2 machine (30-core, 60-thread), the
perf profile below shows what perf was doing after the profiled AIM7
shared workload completed:

-     83.94%  perf  libc-2.11.3.so     [.] __strcmp_sse42
   - __strcmp_sse42
      - 99.82% map__new
           machine__process_mmap_event
           perf_session_deliver_event
           perf_session__process_event
           __perf_session__process_events
           cmd_record
           cmd_mem
           run_builtin
           main
           __libc_start_main
-     13.17%  perf  perf               [.] __dsos__findnew
     __dsos__findnew
     map__new
     machine__process_mmap_event
     perf_session_deliver_event
     perf_session__process_event
     __perf_session__process_events
     cmd_record
     cmd_mem
     run_builtin
     main
     __libc_start_main

So about 97% of CPU times were spent in the map__new() function trying
to insert new DSO entry into the DSO linked list. The whole
post-processing step took about 9 minutes.

The DSO structures are currently searched linearly. So the total
processing time will be proportional to n^2.

To overcome this performance problem, the DSO code is modified to also
put the DSO structures in a RB tree sorted by its long name in
additional to being in a simple linked list. With this change, the
processing time will become proportional to n*log(n) which will be much
quicker for large n. However, the short name will still be searched
using the old linear searching method.  With that patch in place, the
same perf-mem post-processing step took less than 30 seconds to
complete.

Signed-off-by: Waiman Long 
Cc: Adrian Hunter 
Cc: Don Zickus 
Cc: Douglas Hatch 
Cc: Ingo Molnar 
Cc: Jiri Olsa 
Cc: Namhyung Kim 
Cc: Paul Mackerras 
Cc: Peter Zijlstra 
Cc: Scott J Norton 
Link: http://lkml.kernel.org/r/1412098575-27863-3-git-send-email-Waiman.Long@hp.com
Signed-off-by: Arnaldo Carvalho de Melo

perf symbols: Encapsulate dsos list head into struct dsos

2014-09-30T15:11:49+00:00

This is a precursor patch to enable long name searching of DSOs using
a rbtree.

In this patch, a new dsos structure is created which contains only a
list head structure for the moment.

The new dsos structure is used, in turn, in the machine structure for
the user_dsos and kernel_dsos fields.

Only the following 3 dsos functions are modified to accept the new dsos
structure parameter instead of list_head:

 - dsos__add()
 - dsos__find()
 - __dsos__findnew()

Signed-off-by: Waiman Long 
Cc: Adrian Hunter 
Cc: Don Zickus 
Cc: Douglas Hatch 
Cc: Ingo Molnar 
Cc: Jiri Olsa 
Cc: Namhyung Kim 
Cc: Paul Mackerras 
Cc: Peter Zijlstra 
Cc: Scott J Norton 
Link: http://lkml.kernel.org/r/1412021249-19201-2-git-send-email-Waiman.Long@hp.com
[ Move struct dsos to dso.h to reduce the dso methods depends on machine.h ]
Signed-off-by: Arnaldo Carvalho de Melo

perf util: Replace strerror with strerror_r for thread-safety

2014-08-15T13:58:35+00:00

Replaces all strerror with strerror_r in util for making the perf lib
thread-safe.

Signed-off-by: Masami Hiramatsu 
Cc: Ingo Molnar 
Cc: Namhyung Kim 
Cc: Naohiro Aota 
Cc: Paul Mackerras 
Cc: Peter Zijlstra 
Link: http://lkml.kernel.org/r/20140814022236.3545.3367.stgit@kbuild-fedora.novalocal
Signed-off-by: Arnaldo Carvalho de Melo

perf symbols: Make sure --symfs usage includes the path separator

2014-07-31T12:58:21+00:00

Minchan reported that perf failed to load vmlinux if --symfs argument
doesn't end with '/' character.

Fix it by making sure that the '/' path separator is used when composing
pathnames with a --symfs provided directory name.

Reported-by: Minchan Kim 
Cc: David Ahern 
Cc: Jiri Olsa 
Cc: Minchan Kim 
Cc: Namhyung Kim 
Cc: Paul Mackerras 
Cc: Peter Zijlstra 
Link: http://lkml.kernel.org/n/tip-8n4s6b6zvsez5ktanw006125@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo

perf tools: Add dso__type()

2014-07-23T20:36:36+00:00

dso__type() determines wheather a dso is 32-bit, x32 (32-bit with 64-bit
registers) or 64-bit.

dso__type() will be used to determine the VDSO a program maps.

Reviewed-by: Jiri Olsa 
Signed-off-by: Adrian Hunter 
Cc: David Ahern 
Cc: Frederic Weisbecker 
Cc: Jiri Olsa 
Cc: Namhyung Kim 
Cc: Paul Mackerras 
Cc: Peter Zijlstra 
Cc: Stephane Eranian 
Link: http://lkml.kernel.org/r/1406035081-14301-51-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo

perf tools: Add dso__data_size()

2014-07-23T14:57:27+00:00

Add a function to return the dso data size, for use in estimating the
size an instruction cache.

Signed-off-by: Adrian Hunter 
Cc: David Ahern 
Cc: Frederic Weisbecker 
Cc: Jiri Olsa 
Cc: Namhyung Kim 
Cc: Paul Mackerras 
Cc: Peter Zijlstra 
Cc: Stephane Eranian 
Link: http://lkml.kernel.org/r/1406035081-14301-27-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo

perf tools: Add dso__data_status_seen()

2014-07-23T14:23:44+00:00

Add a function to track whether a caller has seen the data status of a
dso.  This is needed to enable callers to report the error exactly once
only per dso.

Signed-off-by: Adrian Hunter 
Cc: David Ahern 
Cc: Frederic Weisbecker 
Cc: Jiri Olsa 
Cc: Namhyung Kim 
Cc: Paul Mackerras 
Cc: Peter Zijlstra 
Cc: Stephane Eranian 
Link: http://lkml.kernel.org/r/1406035081-14301-11-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo

perf tools: Record whether a dso has data

2014-07-23T14:22:35+00:00

Add 'data.status' to record whether a dso has data (i.e. an object
file).  This is used to avoid repeatedly creating the file name and
attempting to open a file that is not present.

Signed-off-by: Adrian Hunter 
Cc: David Ahern 
Cc: Frederic Weisbecker 
Cc: Jiri Olsa 
Cc: Namhyung Kim 
Cc: Paul Mackerras 
Cc: Peter Zijlstra 
Cc: Stephane Eranian 
Link: http://lkml.kernel.org/r/1406035081-14301-10-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo