Re: [PATCH v2 3/5] perf diff: generalize hpp__color_delta for -c

[Ramkumar Ramachandra]

Jiri Olsa wrote:
>>       if (!pair)
>>               goto dummy_print;
>>       if (pair->diff.computed)
>> -             percent = pair->diff.period_ratio_delta;
>> +             switch (comparison_method) {
>> +             case COMPUTE_DELTA:
>> +                     delta = pair->diff.period_ratio_delta;
>> +                     break;
>> +             default:
>> +                     BUG_ON(1);
>> +             }
>>       else
>> -             percent = compute_delta(he, pair);
>> +             switch (comparison_method) {
>> +             case COMPUTE_DELTA:
>> +                     delta = compute_delta(he, pair);
>> +                     break;
>> +             default:
>> +                     BUG_ON(1);
>> +             }
>>
>> -     if (!he->dummy && fabs(percent) >= 0.01) {
>> -             scnprintf(pfmt, 20, "%%%+d.2f%%%%", dfmt->header_width - 1);
>> -             return color_snprintf(hpp->buf, hpp->size,
>> -                             percent > 0 ? PERF_COLOR_GREEN : PERF_COLOR_RED,
>> -                             pfmt, percent);
>> +     if (!he->dummy) {
>
> isn't this check superfluous because of the above (!pair) check?

>From builtin-diff.c:get_pair_data(), we see that `pair' is one of the
pairs in he->pairs. he->dummy is set in
util/hist.c:hists__add_dummy_entry() which is called only when he
doesn't have pairs (util/hist.c:942). Wait, couldn't
util/hist.c:hists__add_dummy_entry() also have returned NULL in the
case when he is NULL? But  __hpp__color_compare wouldn't have been
called with a NULL he in the first place. So yeah, the check is
redundant although it wasn't immediately obvious to me.

Thanks.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/