Skip to content

K
Kernel

openeuler / Kernel
1 年多 前同步成功

通知 8
Star 0
Fork 0
K
Kernel
切换分支/标签
  1. 29 5月, 2012 1 次提交
  3. 26 5月, 2012 1 次提交
    • A
      perf top: Fix counter name fixup when fallbacking to cpu-clock · 895d9766
      由 Arnaldo Carvalho de Melo 提交于 
      In 40491eaa "perf top: Update event name when falling back to cpu-clock"
we freed counter->name but didn't reset it to NULL, then when setting it
to the result of event_name(), event_name() would use the cached value,
which by now was overwritten and thus we got garbage or a zero lenght
string.

Fix it by just freeing and setting counter->name to NULL, this way
event_name() when called afterwards, will find the right counter name
and cache it again.

Found while trying 'cycles:pp' on a machine were :pp couldn't be
honoured. Probably the best fallback here is to tell the user that that
level of precision is not available on the PMU and then go removing 'p',
levels of precision till we get to play 'cycles' and if even that fails,
_then_ get to 'cpu-clock'.

But that is the matter for another patch, this one just needs to fix the
caching issue, which in the end will show 'cpu-clock' when tools ask for
the event name being used, which clarifies things for the user, that
will see that 'cycles:pp' or whatever not support event is not being
used, some sort of fallback happened.

Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@gmail.com>
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-w1neie2dqli89we1bzwkf4id@git.kernel.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      895d9766
  5. 23 5月, 2012 1 次提交
  7. 22 5月, 2012 1 次提交
  9. 17 5月, 2012 1 次提交
  11. 16 5月, 2012 1 次提交
  13. 09 5月, 2012 1 次提交
  15. 08 5月, 2012 5 次提交
  17. 03 5月, 2012 3 次提交
  19. 06 4月, 2012 2 次提交
  21. 03 3月, 2012 1 次提交
  23. 15 2月, 2012 2 次提交
  25. 14 2月, 2012 2 次提交
  27. 31 1月, 2012 2 次提交
  29. 25 1月, 2012 1 次提交
  31. 08 1月, 2012 1 次提交
  33. 07 1月, 2012 1 次提交
  35. 04 1月, 2012 1 次提交
  37. 29 11月, 2011 1 次提交
  39. 28 11月, 2011 6 次提交
  41. 03 11月, 2011 1 次提交
    • A
      perf top: Fix live annotation in the --stdio interface · f9e3d4b1
      由 Arnaldo Carvalho de Melo 提交于 
      In the old --stdio interface the annotation is done just after one
selects a symbol, while in --tui, now the default when the required libs
are installed, we annotate all symbols with samples so that when
annotation is asked we see what happened recently on that symbol.

To achieve that the --stdio variant checks if the hist_entry being
processed is the one selected by the user via the 's' hotkey. What
happens now that we share the hist_entry abstractions with 'perf report'
is that for minimizing locking contention multiple rb_trees are used,
one for collecting the samples and other to browse/show them after
resorting it by number of samples and decay them, which is done
periodically.

So the simple test in record_precise_ip doesn't work as we move
hist_entries between those rb_trees. To fix it just check that the
underlying struct symbol associated with those hist_entries is the same.
Reported-by: NMike Galbraith <efault@gmx.de>
Tested-by: NMike Galbraith <efault@gmx.de>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
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-bcfnraqkux88fox9ba9767ds@git.kernel.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      f9e3d4b1
  43. 02 11月, 2011 2 次提交
  45. 26 10月, 2011 2 次提交