netdev - Re: [PATCH v2] xps-mp: Transmit Packet Steering for multiqueue

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] [day] [month] [year] [list]

Message-ID: <AANLkTi=KzdzaMJh0d5=v2+d3FL+OSafWyKEWOkw-dUQq@mail.gmail.com>
Date:	Wed, 13 Oct 2010 08:41:16 -0700
From:	Tom Herbert <therbert@...gle.com>
To:	Eric Dumazet <eric.dumazet@...il.com>
Cc:	davem@...emloft.net, netdev@...r.kernel.org
Subject: Re: [PATCH v2] xps-mp: Transmit Packet Steering for multiqueue

On Tue, Oct 12, 2010 at 10:48 PM, Eric Dumazet <eric.dumazet@...il.com> wrote:
>
> Le mardi 12 octobre 2010 à 17:20 -0700, Tom Herbert a écrit :
> > +#ifdef CONFIG_RPS
> > +       struct kobject          kobj;
> > +       struct netdev_queue     *first;
> > +       atomic_t                count;
> > +       struct xps_map          *xps_maps;
> > +#endif
>
> Please put kobj at the end of this block, because its a big object (64
> bytes)
>
Okay.

> +#ifdef CONFIG_RPS
> > +       struct xps_map          *xps_maps;
> > +       struct netdev_queue     *first;
> > +       atomic_t                count;
> > +       struct kobject          kobj;
> > +#endif
> >
>
> This way, we only use one cache line to access hot path fields, instead
> of two cache lines.
>
Okay.

>
>
> Tom, I believe you should split your patch in several parts, its really
> too hard to review.
>
>
> For example the netif_alloc_netdev_queues() stuff should be a patch on
> its own, as this stuff was already discussed on netdev some days ago, so
> that you could copy Ben and me on this one ;)
>
Ahh, that was some discussion about TX queues in the thread about RX
queue allocation.  I seemed to have missed that part.

> (Ie not allocating _tx in alloc_netdev_mq() but in register_netdevice())
>
Sorry about that.  The good changes in rx queue motivated these
changes.  I'll extract the parts that move the TX queue allocation
into a separate patch.

Thanks for reviewing,
Tom

>
>
> Its also a good thing to give diffstat output :
>
>  include/linux/netdevice.h |   28 +++
>  include/linux/skbuff.h    |    1
>  net/core/dev.c            |  191 ++++++++++++++++------
>  net/core/net-sysfs.c      |  305 +++++++++++++++++++++++++++++++++++-
>  net/core/net-sysfs.h      |    3
>  net/ipv4/tcp_output.c     |    4
>  6 files changed, 471 insertions(+), 61 deletions(-)
>
>
>
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html