1. 06 Dec, 2010 4 commits
    • Masami Hiramatsu's avatar
      kprobes: Support delayed unoptimizing · 6274de49
      Masami Hiramatsu authored
      Unoptimization occurs when a probe is unregistered or disabled,
      and is heavy because it recovers instructions by using
      stop_machine(). This patch delays unoptimization operations and
      unoptimize several probes at once by using
      text_poke_smp_batch(). This can avoid unexpected system slowdown
      coming from stop_machine().
      
      Changes in v5:
      - Split this patch into several cleanup patches and this patch.
      - Fix some text_mutex lock miss.
      - Use bool instead of int for behavior flags.
      - Add additional comment for (un)optimizing path.
      
      Changes in v2:
      - Use dynamic allocated buffers and params.
      Signed-off-by: default avatarMasami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
      Cc: Rusty Russell <rusty@rustcorp.com.au>
      Cc: Frederic Weisbecker <fweisbec@gmail.com>
      Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
      Cc: Jason Baron <jbaron@redhat.com>
      Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
      Cc: 2nddept-manager@sdl.hitachi.co.jp
      LKML-Reference: <20101203095409.2961.82733.stgit@ltc236.sdl.hitachi.co.jp>
      Signed-off-by: default avatarIngo Molnar <mingo@elte.hu>
      6274de49
    • Masami Hiramatsu's avatar
      kprobes: Separate kprobe optimizing code from optimizer · 61f4e13f
      Masami Hiramatsu authored
      Separate kprobe optimizing code from optimizer, this
      will make easy to introducing unoptimizing code in
      optimizer.
      Signed-off-by: default avatarMasami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
      Cc: Rusty Russell <rusty@rustcorp.com.au>
      Cc: Frederic Weisbecker <fweisbec@gmail.com>
      Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
      Cc: Jason Baron <jbaron@redhat.com>
      Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
      Cc: 2nddept-manager@sdl.hitachi.co.jp
      LKML-Reference: <20101203095403.2961.91201.stgit@ltc236.sdl.hitachi.co.jp>
      Signed-off-by: default avatarIngo Molnar <mingo@elte.hu>
      61f4e13f
    • Masami Hiramatsu's avatar
      kprobes: Cleanup disabling and unregistering path · 6f0f1dd7
      Masami Hiramatsu authored
      Merge disabling kprobe to unregistering kprobe function
      and add comments for disabing/unregistring process.
      
      Current unregistering code disables(disarms) kprobes after
      checking target kprobe status. This patch changes it to
      disabling kprobe first after that it changing the kprobe's
      state. This allows to share probe disabling code between
      disable_kprobe() and unregister_kprobe().
      Signed-off-by: default avatarMasami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
      Cc: Rusty Russell <rusty@rustcorp.com.au>
      Cc: Frederic Weisbecker <fweisbec@gmail.com>
      Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
      Cc: Jason Baron <jbaron@redhat.com>
      Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
      Cc: 2nddept-manager@sdl.hitachi.co.jp
      LKML-Reference: <20101203095356.2961.30152.stgit@ltc236.sdl.hitachi.co.jp>
      Signed-off-by: default avatarIngo Molnar <mingo@elte.hu>
      6f0f1dd7
    • Masami Hiramatsu's avatar
      kprobes: Rename old_p to more appropriate name · 6d8e40a8
      Masami Hiramatsu authored
      Rename irrelevant uses of "old_p" to more appropriate names.
      Originally, "old_p" just meant "the old kprobe on given address"
      but current code uses that name as "just another kprobe" or
      something like that. This patch renames those pointer names
      to more appropriate one for maintainability.
      Signed-off-by: default avatarMasami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
      Cc: Rusty Russell <rusty@rustcorp.com.au>
      Cc: Frederic Weisbecker <fweisbec@gmail.com>
      Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
      Cc: Jason Baron <jbaron@redhat.com>
      Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
      Cc: 2nddept-manager@sdl.hitachi.co.jp
      LKML-Reference: <20101203095350.2961.48110.stgit@ltc236.sdl.hitachi.co.jp>
      Signed-off-by: default avatarIngo Molnar <mingo@elte.hu>
      6d8e40a8
  2. 02 Dec, 2010 1 commit
  3. 01 Dec, 2010 19 commits
  4. 30 Nov, 2010 11 commits
  5. 27 Nov, 2010 3 commits
  6. 26 Nov, 2010 2 commits
    • Arnaldo Carvalho de Melo's avatar
      perf events: Default to using event__process_lost · 37982ba0
      Arnaldo Carvalho de Melo authored
      Tool developers have to fill in a 'perf_event_ops' method table to
      specify how to handle each event, so far the ones that were not
      explicitely especified would get a stub that would just discard the
      event.
      
      Change that so that tool developers can get the lost event details and
      the total number of such events at the end of 'perf report -D' output.
      Suggested-by: default avatarThomas Gleixner <tglx@linutronix.de>
      Cc: Frederic Weisbecker <fweisbec@gmail.com>
      Cc: Mike Galbraith <efault@gmx.de>
      Cc: Paul Mackerras <paulus@samba.org>
      Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
      Cc: Stephane Eranian <eranian@google.com>
      CC: Thomas Gleixner <tglx@linutronix.de>
      Cc: Tom Zanussi <tzanussi@gmail.com>
      LKML-Reference: <new-submission>
      Signed-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      37982ba0
    • Arnaldo Carvalho de Melo's avatar
      perf record: Add option to disable collecting build-ids · baa2f6ce
      Arnaldo Carvalho de Melo authored
      Collecting build-ids for long running sessions may take a long time
      because it needs to traverse the whole just collected perf.data stream
      of events, marking the DSOs that had hits and then looking for the
      .note.gnu.build-id ELF section.
      
      For things like the 'trace' tool that records and right away consumes
      the data on systems where its unlikely that the DSOs being monitored
      will change while 'trace' runs, it is desirable to remove build id
      collection, so add a -B/--no-buildid option to perf record to allow such
      use case.
      
      Longer term we'll avoid all this if we, at DSO load time, in the kernel,
      take advantage of this slow code path to collect the build-id and stash
      it somewhere, so that we can insert it in the PERF_RECORD_MMAP event.
      Reported-by: default avatarThomas Gleixner <tglx@linutronix.de>
      Cc: Frederic Weisbecker <fweisbec@gmail.com>
      Cc: Mike Galbraith <efault@gmx.de>
      Cc: Paul Mackerras <paulus@samba.org>
      Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
      Cc: Stephane Eranian <eranian@google.com>
      Cc: Thomas Gleixner <tglx@linutronix.de>
      Cc: Tom Zanussi <tzanussi@gmail.com>
      LKML-Reference: <new-submission>
      Signed-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
      baa2f6ce