On Thu, Sep 16, 2010 at 1:49 PM, Matthew Dillon <dil...@apollo.backplane.com> wrote: > > :We have quite a few kprintf's in the kernel that do not necessarily > :report error conditions, they are purely informational and highlight > :odd or rare cases that often the end-user of the operating system > :really doesn't need to know about. This has been discussed quite a few > :times on IRC. > : > :I would like to put these under a sysctl like kern.developer and wrap > :them in a function like dkprintf unless anyone has any major > :objections. (I'm open to suggestions on the naming). > : > :Sam > > I'd rather not. If any of them have becoming annoying we can > remove them after a quick code review in irc. > > -Matt > Matthew Dillon > <dil...@backplane.com> >
The issue that has been raised time and again isn't one of annoyance, it is of misunderstanding. Users seem to think these are "errors" when in fact in many cases they are just "interesting to certain developers". Sam
