Re: [PATCH v3] perf diff: Report noisy for cycles diff

[Jin, Yao]

On 8/12/2019 4:35 PM, Jiri Olsa wrote:
On Sat, Aug 10, 2019 at 07:30:29AM +0800, Jin Yao wrote:

SNIP

+
+               init_stats(&pair->diff.stats);
+               init_spark_values(pair->diff.svals, NUM_SPARKS);
+
+               for (int i = 0; i < pair->block_info->num; i++) {
+                       u64 val;
+
+                       if (i >= he->block_info->num || i >= NUM_SPARKS)
+                               break;
+
+                       val = labs(pair->block_info->cycles_spark[i] -
+                                    he->block_info->cycles_spark[i]);
+
+                       update_spark_value(pair->diff.svals, NUM_SPARKS,
+                                          &pair->diff.stats, val);
+                       update_stats(&pair->diff.stats, val);
+               }
        }
  }
  
@@ -1250,6 +1292,8 @@ static const struct option options[] = {
                    "Show period values."),
        OPT_BOOLEAN('F', "formula", &show_formula,
                    "Show formula."),
+       OPT_BOOLEAN('n', "noisy", &show_noisy,
+                   "Show cycles noisy - WARNING: use only with -c cycles."),

this should be more 'cycles' specific like --cycles-noisy or --cycles-hist?

also I dont think we should use 'n' for this, just the long option

jirka


Yes, it's cycles specific. Maybe --cycles-hist is better and the long 
option should be enough. We can reserve -n for future feature. :)

Thanks
Jin Yao