Re: [PATCH v8 4/4] vsprintf: Add command line option debug_boot_weak_hash

[Tobin C. Harding]

On Wed, Jun 20, 2018 at 09:09:49PM -0700, Randy Dunlap wrote:
> On 06/20/2018 08:15 PM, Tobin C. Harding wrote:
> > On Wed, Jun 20, 2018 at 04:38:05PM -0700, Randy Dunlap wrote:
> >> On 06/20/2018 04:22 PM, Tobin C. Harding wrote:
> >>> On Wed, Jun 20, 2018 at 03:36:44PM -0700, Randy Dunlap wrote:
> >>>> On 06/20/2018 03:30 PM, Tobin C. Harding wrote:
> >>>>> On Wed, Jun 20, 2018 at 09:09:49AM -0700, Randy Dunlap wrote:
> >>>>>> On 06/19/2018 09:20 PM, Tobin C. Harding wrote:
> >>>>>>> Currently printing [hashed] pointers requires enough entropy to be
> >>>>>>> available.  Early in the boot sequence this may not be the case
> >>>>>>> resulting in a dummy string '(____ptrval____)' being printed.  This
> >>>>>>> makes debugging the early boot sequence difficult.  We can relax the
> >>>>>>> requirement to use cryptographically secure hashing during debugging.
> >>>>>>> This enables debugging while keeping development/production kernel
> >>>>>>> behaviour the same.
> >>>>>>>
> >>>>>>> If new command line option debug_boot_weak_hash is enabled use
> >>>>>>> cryptographically insecure hashing and hash pointer value immediately.
> >>>>>>>
> >>>>>>> Signed-off-by: Tobin C. Harding <me@xxxxxxxx>
> >>>>>>> Reviewed-by: Steven Rostedt (VMware) <rostedt@xxxxxxxxxxx>
> >>>>>>> ---
> >>>>>>>  Documentation/admin-guide/kernel-parameters.txt |  9 +++++++++
> >>>>>>>  lib/vsprintf.c                                  | 17 +++++++++++++++++
> >>>>>>>  2 files changed, 26 insertions(+)
> >>>>>>>
> >>>>>>> diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt
> >>>>>>> index 638342d0a095..a116fc0366b0 100644
> >>>>>>> --- a/Documentation/admin-guide/kernel-parameters.txt
> >>>>>>> +++ b/Documentation/admin-guide/kernel-parameters.txt
> >>>>>>> @@ -748,6 +748,15 @@
> >>>>>>>  
> >>>>>>>  	debug		[KNL] Enable kernel debugging (events log level).
> >>>>>>>  
> >>>>>>> +	debug_boot_weak_hash
> >>>>>>> +			[KNL] Enable printing pointers early in the boot
> >>>>>>> +			sequence.  If enabled, we use a weak hash instead of
> >>>>>>> +			siphash to hash pointers.  Use this option if you need
> >>>>>>> +			to see pointer values during early boot (i.e you are
> >>>>>>
> >>>>>> maybe:
> >>>>>> 			to see hashed pointer values
> >>>>>> i.e., not raw pointers.
> >>>>>
> >>>>> You cannot see 'raw pointers' anyways?
> >>>>
> >>>> only if using %px ?
> >>>>
> >>>> Maybe it's just terminology.  I don't consider a hashed value as a pointer value.
> >>>> It's just a key or handle or some other number, but it's not a pointer.
> >>>>
> >>>>>>
> >>>>>>> +			seeing instances of '(___ptrval___)').
> >>>>>>> +			Cryptographically insecure, please do not use on
> >>>>>>> +			production kernels.
> >>>>>
> >>>>> thanks for the review, I don't quiet see how to use your suggestion to
> >>>>> make the text clearer.  If you still feel this change is needed perhaps
> >>>>> you could write so I understand i.e 'Use this option if ...'
> >>>>
> >>>>
> >>>> OK, if you are good with it, I am too.  :)
> >>>
> >>> I get you know.  I agree, how about this
> >>>
> >>> 			[KNL] Enable printing pointers early in the boot
> >>> 			sequence.  If enabled, we use a weak hash instead of
> >>> 			siphash to hash pointers.  Use this option if you need
> >>> 			to print pointers with %px during early boot
> >>> 			(i.e you are seeing instances of '(___ptrval___)').
> >>> 			Cryptographically insecure, please do not use on
> >>> 			production kernels.
> >>
> >> Sorry, I'm still confused by this paragraph.  It seems to say two different
> >> things.
> > 
> > My bad, I got totally confused myself.  After all this time you would
> > think I knew which specifier hashed and which didn't.  My apologies,
> > how about this:
> > 
> >  			[KNL] Enable printing [hashed] pointers early in
> > 			the boot sequence.  If enabled, we use a weak hash
> > 			instead of siphash to hash pointers.  Use this option if
> > 			you are seeing instances of '(___ptrval___)') and need
> > 			to see a value (hashed pointer) instead. Cryptographically
> > 			insecure, please do not use on production kernels.
> >  			 			
> > 
> > thanks for your patience,
> > Tobin.
> 
> Yes, that's good.  Thanks.

Awesome, v9 on it's way :)

thanks,
Tobin.