On 15:51 Mon 23 Oct , Yevgeny Kliteynik wrote: > > Sasha Khapyorsky wrote: > > On 09:02 Mon 23 Oct , Yevgeny Kliteynik wrote: > >> Hi Sasha. > >> > >> The removal of the sm->p_report_buf is a good idea. > >> However, I do have one comment: > >> In several cases this buffer was printed using the osm_log_raw() > >> function, and you replaced this with a plain fprintf(stdout,...). > >> Right now the osm_log_raw function just prints to stdout too, but > >> this doesn't always have to be the case. Besides, osm_log_raw > >> provides verbosity level checking, which is lost when you replace > >> it with printf. > > > > Both functions calls were and still be conditonalized by verbosity > > level, so it is not lost. > > Right, there is a check in the beginning of the function, mea culpa. > Anyway, this wasn't the main point. I grep'ed the osm code, and the > only cases where there is an explicit printing to stdout is when log > is not initialized yet, or in console mode, and I think that this is > a better way to manage logging (even though printing directly to stdout > is more efficient).
I'm not following. Those functions _are_ printing to stdout. If in a future we will want other file we will replace 'stdout' by other stream (and this is why fprintf() and not printf() is used in the patch). Sasha _______________________________________________ openib-general mailing list [email protected] http://openib.org/mailman/listinfo/openib-general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
