perf top: Remove lost events checking
authorKan Liang <kan.liang@intel.com>
Thu, 18 Jan 2018 21:26:30 +0000 (13:26 -0800)
committerArnaldo Carvalho de Melo <acme@redhat.com>
Thu, 15 Feb 2018 12:56:43 +0000 (09:56 -0300)
There would be some records lost in overwrite mode because of pausing
the ringbuffer. It has little impact for the accuracy of the snapshot
and could be tolerated by 'perf top'.

Remove the lost events checking.

Signed-off-by: Kan Liang <kan.liang@intel.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jin Yao <yao.jin@linux.intel.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/r/1516310792-208685-16-git-send-email-kan.liang@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
tools/perf/builtin-top.c

index 7def861..5965306 100644 (file)
@@ -283,8 +283,9 @@ static void perf_top__print_sym_table(struct perf_top *top)
 
        printf("%-*.*s\n", win_width, win_width, graph_dotted_line);
 
-       if (hists->stats.nr_lost_warned !=
-           hists->stats.nr_events[PERF_RECORD_LOST]) {
+       if (!top->record_opts.overwrite &&
+           (hists->stats.nr_lost_warned !=
+           hists->stats.nr_events[PERF_RECORD_LOST])) {
                hists->stats.nr_lost_warned =
                              hists->stats.nr_events[PERF_RECORD_LOST];
                color_fprintf(stdout, PERF_COLOR_RED,
@@ -611,7 +612,8 @@ static void *display_thread_tui(void *arg)
 
        perf_evlist__tui_browse_hists(top->evlist, help, &hbt,
                                      top->min_percent,
-                                     &top->session->header.env, true);
+                                     &top->session->header.env,
+                                     !top->record_opts.overwrite);
 
        done = 1;
        return NULL;