On Thu 2019-03-14 23:12:49, Tetsuo Handa wrote: > Petr Mladek wrote: > > It might be even more straightforward when the per-console value > > defines the effective console level. I mean the following semantic: > > > > + "console_loglevel" would define the default loglevel used > > by consoles at runtime. > > > > + the per-console loglevel could override the default > > console_loglevel. > > > > + We would need a custom handler for the sysctl "console_loglevel". > > It would write the given value to the global console_loglevel > > variable and for all already registered consoles (con->loglevel). > > But some functions change console_loglevel without sysctl (e.g. > console_verbose() when reporting hung tasks and panic()). Should > con->loglevel be changed (which might result in too much messages to > slow consoles) when console_loglevel changes? It is about the semantic. We either want to set a hard limit for each console or we want to set per-console loglevel that will get used in normal situations. I prefer the 2nd semantic. IMHO, console_verbose() should be used only in situations when people really want to see all lines, for example, panic, sysrq output when the machine looks deadlocked, ignore_loglevel is set. I believe that they want to see them even on the slow consoles that are there exactly for debugging these critical situations. Best Regards, Petr
