Re: [RFC PATCH 00/11] printk: safe printing in NMI context

[Frederic Weisbecker <fweisbec@gmail.com>]

On Thu, May 29, 2014 at 10:09:48AM +0200, Jiri Kosina wrote:
> On Thu, 29 May 2014, Frederic Weisbecker wrote:
> 
> > > I am rather surprised that this patchset hasn't received a single review 
> > > comment for 3 weeks.
> > > 
> > > Let me point out that the issues Petr is talking about in the cover letter 
> > > are real -- we've actually seen the lockups triggered by RCU stall 
> > > detector trying to dump stacks on all CPUs, and hard-locking machine up 
> > > while doing so.
> > > 
> > > So this really needs to be solved.
> > 
> > The lack of review may be partly due to a not very appealing changestat 
> > on an old codebase that is already unpopular:
> > 
> >  Documentation/kernel-parameters.txt |   19 +-
> >  kernel/printk/printk.c              | 1218 +++++++++++++++++++++++++----------
> >  2 files changed, 878 insertions(+), 359 deletions(-)
> > 
> > 
> > Your patches look clean and pretty nice actually. They must be seriously 
> > considered if we want to keep the current locked ring buffer design and 
> > extend it to multiple per context buffers. But I wonder if it's worth to 
> > continue that way with the printk ancient design.
> > 
> > If it takes more than 1000 line changes (including 500 added) to make it 
> > finally work correctly with NMIs by working around its fundamental 
> > flaws, shouldn't we rather redesign it to use a lockless ring buffer 
> > like ftrace or perf ones?
> 
> Yeah, printk() has grown over years to a stinking pile of you-know-what, 
> no argument to that.
> 
> I also agree that performing a massive rewrite, which will make it use a 
> lockless buffer, and therefore ultimately solve all its problems 
> (scheduler deadlocks, NMI deadlocks, xtime_lock deadlocks) at once, is 
> necessary in the long run.
> 
> On the other hand, I am completely sure that the diffstat for such rewrite 
> is going to be much more scary :)

Indeed, but probably much more valuable in the long term.

> 
> This is not adding fancy features to printk(), where we really should be 
> saying no; horrible commits like 7ff9554bb5 is exactly something that 
> should be pushed against *heavily*. But bugfixes for hard machine lockups 
> are a completely different story to me (until we have a whole new printk() 
> buffer handling implementation).

Yeah bugfixes are certainly another story. Still it looks like yet another
layer of workaround on a big hack.

But yeah I'm certainly not in a right position to set anyone to do a massive
rewrite on such a boring subsystem :)

There is also a big risk that if we push back this bugfix, nobody will actually do
that desired rewrite.

Lets be crazy and Cc Linus on that.