On Wed, 2012-09-12 at 07:26 -0700, Andi Kleen wrote: > Peter Zijlstra <[email protected]> writes: > > > On Wed, 2012-09-12 at 13:01 +0200, Robert Richter wrote: > >> + if (notsup) > >> + pr_warn("perf: unsupported attribute flags: %016llx\n", > >> notsup); > > > > This is a dmesg DoS.. > > > > I'm also not sure dmesg is the right way.. could we not somehow change > > the attrs to provide better diagnostic? > > Have a output argument for a string, that is then printed by the user tool? > > I agree it's a problem today, i usually have to resort to ftrace to > figure out what'swrong.
Yeah, its a pain.. something I was thinking of is zero-ing out the entire perf_event_attr and set the field that caused the fail to all 1s (for however many bits that field is wide). It would destroy the user input, but it also would provide better feedback on what exact field you went astray. -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
