Re: [RFC][PATCH] RANDOM: ATH9K RNG delivers zero bits of entropy

[Stephan Mueller]

Am Samstag, 6. August 2016, 19:45:51 CEST schrieb Jason Cooper:

Hi Jason,

> Hi Stephan,
> 
> On Fri, Aug 05, 2016 at 05:08:14PM +0200, Stephan Mueller wrote:
> > Hi Ted, Herbert,
> > 
> > I sent a question to the ATH9K RNG some time ago to the developers.
> > See
> > https://www.mail-archive.com/linux-crypto@xxxxxxxxxxxxxxx/msg19115.html
> > 
> > I have not yet received a word and I think this issue should be resolved.
> > 
> > Thanks
> > Stephan
> > 
> > ---8<---
> 
> If the above text is placed below the three dashes, "---", below ...
> 
> > The ATH9K driver implements an RNG which is completely bypassing the
> > standard Linux HW generator logic.
> > 
> > The RNG may or may not deliver entropy. Considering the conservative
> > approach in treating entropy with respect to non-auditable sources, this
> > patch changes the delivered entropy value to zero. The RNG still feeds
> > data into the input_pool but it is assumed to have no entropy.
> > 
> > When the ATH9K RNG changes to use the HW RNG framework, it may re-enable
> > the entropy estimation considering that a user can change that value at
> > boot and runtime.
> > 
> > Signed-off-by: Stephan Mueller <smueller@xxxxxxxxxx>
> > ---
> 
> here, then the mail can be applied directly without editing.

Thank you for the hint. I will resend the patch that can be applied.
> 
> >  drivers/net/wireless/ath/ath9k/rng.c | 3 +--
> >  1 file changed, 1 insertion(+), 2 deletions(-)
> > 
> > diff --git a/drivers/net/wireless/ath/ath9k/rng.c
> > b/drivers/net/wireless/ath/ath9k/rng.c index d38e50f..d63dc48 100644
> > --- a/drivers/net/wireless/ath/ath9k/rng.c
> > +++ b/drivers/net/wireless/ath/ath9k/rng.c
> > @@ -92,8 +92,7 @@ static int ath9k_rng_kthread(void *data)
> > 
> >  		fail_stats = 0;
> >  		
> >  		/* sleep until entropy bits under write_wakeup_threshold */
> > 
> > -		add_hwgenerator_randomness((void *)rng_buf, bytes_read,
> > -					   ATH9K_RNG_ENTROPY(bytes_read));
> 
> This is the only use of this macro.  I'd remove the #define on line 25
> as well.

My idea for leaving it was that folks who would bring the RNG into the 
hwrandom framework could reuse the ideas from the original authors.

What about commenting it out with #if 0 ?
> 
> > +		add_hwgenerator_randomness((void *)rng_buf, bytes_read, 0);
> > 
> >  	}
> >  	
> >  	kfree(rng_buf);
> 
> Other than that,
> 
> Reviewed-by: Jason Cooper <jason@xxxxxxxxxxxxxx>

Thank you.
> 
> thx,
> 
> Jason.



-- 
Ciao
Stephan