Re: [PATCH v12 0/6] implement getrandom() in vDSO

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Wed, Dec 21, 2022 at 03:25:49PM +0100, Jason A. Donenfeld wrote:
> On Tue, Dec 20, 2022 at 08:13:14PM +0000, Eric Biggers wrote:
> > On Tue, Dec 20, 2022 at 05:17:52PM +0000, Christophe Leroy wrote:
> > > Hi Jason,
> > > 
> > > Le 12/12/2022 à 19:53, Jason A. Donenfeld a écrit :
> > > > Changes v11->v12:
> > > > ----------------
> > > > - In order to avoid mlock()ing pages, and the related rlimit and fork
> > > >    inheritance issues there, Introduce VM_DROPPABLE to prevent swapping
> > > >    while meeting the cache-like requirements of vDSO getrandom().
> > > > 
> > > >    This has some tenticles in mm/ and arch/x86/ code, so I've marked the
> > > >    two patches for that as still RFC, while the rest of the series is not
> > > >    RFC.
> > > > 
> > > > - Mandate that opaque state blobs don't straddle page boundaries, so
> > > >    that VM_DROPPABLE can work on page-level granularity rather than
> > > >    allocation-level granularity.
> > > > 
> > > > - Add compiler barriers to vDSO getrandom() to prevent theoretical
> > > >    reordering potential.
> > > > 
> > > > - Initialize the trials loop counter in the chacha test.
> > > 
> > > I would have liked to give it a try on powerpc, but the series 
> > > conflicts. I tried both on v6.1 and on linus/master from now:
> > > 
> > 
> > Same here, I can't figure out how to apply this series.
> 
> Rebased v13 posted: https://lore.kernel.org/all/20221221142327.126451-1-Jason@xxxxxxxxx/
> 

Thanks, it is always good to give the *actual* base commit though, preferably
with the --base option to git format-patch.  "Latest random.git master branch"
changes over time.

- Eric



[Index of Archives]     [Kernel]     [Gnu Classpath]     [Gnu Crypto]     [DM Crypt]     [Netfilter]     [Bugtraq]
  Powered by Linux