On 1/27/21 11:11 AM, Petr Mladek wrote: > On Tue 2021-01-26 12:40:32, Steven Rostedt wrote: >> On Tue, 26 Jan 2021 12:39:12 -0500 >> Steven Rostedt <rostedt@xxxxxxxxxxx> wrote: >> >> > On Tue, 26 Jan 2021 11:30:02 -0600 >> > Timur Tabi <timur@xxxxxxxxxx> wrote: >> > >> > > On 1/26/21 11:14 AM, Vlastimil Babka wrote: >> > > > If it was a boot option, I would personally be for leaving hashing enabled by >> > > > default, with opt-in boot option to disable it. >> > > >> > > A boot option would solve all my problems. I wouldn't need to recompile >> > > the kernel, and it would apply to all variations of printk. >> > >> > Should it be called "make-printk-insecure" > > Nit: This makes me feel that printk() might break (block) the system. > Please, make it more clear that it is about unveiling some secret > information, something like: > > "non-secret-printk" > "non-confidental-printk" > "unretricted-printk" > > I do not mind about the words order or using the > "make-printk-non-secret" form. Yeah, let's not be overly dramatic here. >> And even if we make this a boot time option, perhaps we should still >> include that nasty dmesg notice, which will let people know that the kernel >> has unhashed values. > > +1 If it's what it takes to have that option, fine :) > Best Regards, > Petr >
