linux-kernel - Re: [PATCH RFC v1 0/2] vhost: ring format independence

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20191012152332-mutt-send-email-mst@kernel.org>
Date:   Sat, 12 Oct 2019 15:26:11 -0400
From:   "Michael S. Tsirkin" <mst@...hat.com>
To:     Jason Wang <jasowang@...hat.com>
Cc:     linux-kernel@...r.kernel.org, kvm@...r.kernel.org,
        virtualization@...ts.linux-foundation.org, netdev@...r.kernel.org
Subject: Re: [PATCH RFC v1 0/2] vhost: ring format independence

On Sat, Oct 12, 2019 at 04:15:42PM +0800, Jason Wang wrote:
> 
> On 2019/10/11 下午9:45, Michael S. Tsirkin wrote:
> > So the idea is as follows: we convert descriptors to an
> > independent format first, and process that converting to
> > iov later.
> > 
> > The point is that we have a tight loop that fetches
> > descriptors, which is good for cache utilization.
> > This will also allow all kind of batching tricks -
> > e.g. it seems possible to keep SMAP disabled while
> > we are fetching multiple descriptors.
> 
> 
> I wonder this may help for performance:

Could you try it out and report please?
Would be very much appreciated.

> - another indirection layer, increased footprint

Seems to be offset off by improved batching.
For sure will be even better if we can move stac/clac out,
or replace some get/put user with bigger copy to/from.

> - won't help or even degrade when there's no batch

I couldn't measure a difference. I'm guessing

> - an extra overhead in the case of in order where we should already had
> tight loop

it's not so tight with translation in there.
this exactly makes the loop tight.

> - need carefully deal with indirect and chain or make it only work for
> packet sit just in a single descriptor
> 
> Thanks

I don't understand this last comment.

> 
> > 
> > And perhaps more importantly, this is a very good fit for the packed
> > ring layout, where we get and put descriptors in order.
> > 
> > This patchset seems to already perform exactly the same as the original
> > code already based on a microbenchmark.  More testing would be very much
> > appreciated.
> > 
> > Biggest TODO before this first step is ready to go in is to
> > batch indirect descriptors as well.
> > 
> > Integrating into vhost-net is basically
> > s/vhost_get_vq_desc/vhost_get_vq_desc_batch/ -
> > or add a module parameter like I did in the test module.
> > 
> > 
> > 
> > Michael S. Tsirkin (2):
> >    vhost: option to fetch descriptors through an independent struct
> >    vhost: batching fetches
> > 
> >   drivers/vhost/test.c  |  19 ++-
> >   drivers/vhost/vhost.c | 333 +++++++++++++++++++++++++++++++++++++++++-
> >   drivers/vhost/vhost.h |  20 ++-
> >   3 files changed, 365 insertions(+), 7 deletions(-)
> >