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

[Jiri Kosina]

On Tue, 10 Jun 2014, Linus Torvalds wrote:

> > Lets be crazy and Cc Linus on that.
> 
> Quite frankly, I hate seeing something like this:
> 
>  kernel/printk/printk.c              | 1218 +++++++++++++++++++++++++----------
> 
> for something that is stupid and broken. Printing from NMI context
> isn't really supposed to work, and we all *know* it's not supposed to
> work.

It's OTOH rather useful in a few scenarios -- particularly it's the only 
way to dump stacktraces from remote CPUs in order to obtain traces that 
actually make sense (in situations like RCU stall); using workqueue-based 
dumping is useless there.

> I'd much rather disallow it, and if there is one or two places that
> really want to print a warning and know that they are in NMI context,
> have a special workaround just for them, with something that does
> *not* try to make printk in general work any better.

Well, that'd mean that at least our stack dumping mechanism would need to 
know both ways of printing; but yes, it'll still probably be less than 880 
lines added.

> Dammit, NMI context is special. I absolutely refuse to buy into the
> broken concept that we should make more stuff work in NMI context.
> Hell no, we should *not* try to make more crap work in NMI. NMI people
> should be careful.

In parallel, I'd for the sake of argument propose to just drop the whole 
_CONT printing (and all the things that followed on top) as that made 
printk() a complete hell to maintain for a disputable gain IMO.

> Make a trivial "printk_nmi()" wrapper that tries to do a trylock on
> logbuf_lock, and *maybe* the existing sequence of
> 
>         if (console_trylock_for_printk())
>                 console_unlock();
> 
> then works for actually triggering the printout. But the wrapper
> should be 15 lines of code for "if possible, try to print things", and
> *not* a thousand lines of changes.

Well, we are carrying much simpler fix for this whole braindamage in our 
enterprise kernel that is from pre-7ff9554bb578 era, and it was rather 
simple fix in principle (the diffstat is much larger than it had to be due 
to code movements):

	http://kernel.suse.com/cgit/kernel/commit/?h=SLE11-SP3&id=8d62ae68ff61d77ae3c4899f05dbd9c9742b14c9

But after the scary 7ff9554bb578 and its successors, things got a lot more 
complicated.

-- 
Jiri Kosina
SUSE Labs
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/