From mboxrd@z Thu Jan 1 00:00:00 1970 From: Aurelien Jarno Subject: Re: virtio-rng only returns zeros with CONFIG_HW_RANDOM=m Date: Wed, 27 Feb 2013 11:33:36 +0100 Message-ID: <20130227103336.GA16559@ohm.aurel32.net> References: <20130224231154.GA30477@hall.aurel32.net> <8738wifwb2.fsf@rustcorp.com.au> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-15 Cc: kvm@vger.kernel.org, Ian Molton , Matt Mackall , Herbert Xu , Ben Hutchings , Jens Axboe To: Rusty Russell Return-path: Received: from hall.aurel32.net ([88.191.126.93]:40113 "EHLO hall.aurel32.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751937Ab3B0Kdp (ORCPT ); Wed, 27 Feb 2013 05:33:45 -0500 Content-Disposition: inline In-Reply-To: <8738wifwb2.fsf@rustcorp.com.au> Sender: kvm-owner@vger.kernel.org List-ID: On Wed, Feb 27, 2013 at 10:43:37AM +1030, Rusty Russell wrote: > Aurelien Jarno writes: > > Hi, > > > > I have noticed that virtio-rng only returns zero for kernels >= 2.6.33 > > built with CONFIG_HW_RANDOM=m. This is a bit much too predictable for a > > random generator ;-). > > Wow. Fortunately, all of SLES, RHEL, Ubuntu or Fedora set > CONFIG_HW_RANDOM=y. What do they know that we don't? > > Oops, looks like Debian testing: config-3.2.0-4-amd64:CONFIG_HW_RANDOM=m > > > The reason for that is virtio expects guest real addresses, while > > rng_core.ko (ie when built as a module) is passing a vmalloced buffer > > to the virtio-rng read function, declared as such: > > > > static u8 rng_buffer[SMP_CACHE_BYTES < 32 ? 32 : SMP_CACHE_BYTES] > > __cacheline_aligned; > > Yuck... It would be nice if this has oopsed. Jens, what about this patch? > > Cheers, > Rusty. > > Subject: scatterlist: sg_set_buf() argument must be in linear mapping. > > Signed-off-by: Rusty Russell > > diff --git a/include/linux/scatterlist.h b/include/linux/scatterlist.h > index 4bd6c06..9365375 100644 > --- a/include/linux/scatterlist.h > +++ b/include/linux/scatterlist.h > @@ -111,6 +111,9 @@ static inline struct page *sg_page(struct scatterlist *sg) > static inline void sg_set_buf(struct scatterlist *sg, const void *buf, > unsigned int buflen) > { > +#ifdef CONFIG_DEBUG_SG > + BUG_ON(!virt_addr_valid(buf)); > +#endif > sg_set_page(sg, virt_to_page(buf), buflen, offset_in_page(buf)); > } > I confirm this patch catches the issue. Thanks. Tested-by: Aurelien Jarno -- Aurelien Jarno GPG: 1024D/F1BCDB73 aurelien@aurel32.net http://www.aurel32.net