On Fri, 2016-10-07 at 13:13 -0700, Linus Torvalds wrote: > On Fri, Oct 7, 2016 at 1:04 PM, Joe Perches <j...@perches.com> wrote: > > Any printk without a KERN_<level> prefix, and there > > are still many of those, can cause random interleaving. > How about people actually work on *that* instead of working around it? > Because the above really should not be true. > > Not at all. Until printk KERN_<level> uses are mandated, > > then these newlines are still useful. > The patches literally added those '\n' things to the pr_xyz() routines > that *enforce* KERN_<level>.
No, because any of those can be followed by a bare printk or a pr_cont that continues the original line. > So really. It's a step backwards. We shouldn't need them. We should > *remove* '\n' at the end, and then if that actually causes problems, > we should fix those problems. $ git grep -w printk * | grep -v "^tools" | grep -v KERN | wc -l 13176 Have at it but here are _far_ fewer missing newlines and it'd be much lower churn to fix those than remove all the newlines and fix the missing continuations.
