netdev - Re: 3.7.8/amd64 full interrupt hangs due to iwlwifi under big nfs copies out

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]

Date:	Wed, 20 Feb 2013 08:55:31 -0800
From:	Eric Dumazet <eric.dumazet@...il.com>
To:	Johannes Berg <johannes@...solutions.net>
Cc:	Marc MERLIN <marc@...lins.org>, David Miller <davem@...emloft.net>,
	Larry.Finger@...inger.net, bhutchings@...arflare.com,
	linux-wireless@...r.kernel.org, netdev@...r.kernel.org
Subject: Re: 3.7.8/amd64 full interrupt hangs due to iwlwifi under big nfs
 copies out

On Wed, 2013-02-20 at 17:20 +0100, Johannes Berg wrote:
> On Wed, 2013-02-20 at 07:11 -0800, Eric Dumazet wrote:
> > On Wed, 2013-02-20 at 10:15 +0100, Johannes Berg wrote:
> > 
> > > OTOH, this affects the protocol, and when you really can't allocate any
> > > order-1 pages you pointed out yourself that many other things also won't
> > > work, so I'm not really sure it makes a big difference if we change the
> > > driver?
> > 
> > It will make a huge difference, even on non pressure mode, as TCP
> > receive window will grow twice faster.
> 
> Hmm, why does that depend on the allocation size?

I guess you missed all the patches about skb->truesize on netdev

> 
> > Unless wifi speed reaches 10Gbps, following patch should do the trick
> > 
> > 
> > diff --git a/drivers/net/wireless/iwlwifi/dvm/rx.c b/drivers/net/wireless/iwlwifi/dvm/rx.c
> > index a4eed20..77a3ee3 100644
> > --- a/drivers/net/wireless/iwlwifi/dvm/rx.c
> > +++ b/drivers/net/wireless/iwlwifi/dvm/rx.c
> > @@ -750,7 +750,12 @@ static void iwlagn_pass_packet_to_mac80211(struct iwl_priv *priv,
> >  	/* Dont use dev_alloc_skb(), we'll have enough headroom once
> >  	 * ieee80211_hdr pulled.
> >  	 */
> > -	skb = alloc_skb(128, GFP_ATOMIC);
> > +	fraglen = 128;
> > +	/* if we use order-1 pages, copy to get better TCP performance */
> > +	if (rxb->truesize > PAGE_SIZE)
> > +		fraglen = max_t(unsigned, fraglen, len);
> > +
> > +	skb = alloc_skb(fraglen, GFP_ATOMIC);
> 
> Hmm, I don't quite understand -- that's not doing any copy?
> 
> FWIW if you do the copy you should not "steal" the pages, then they'd be
> recycled in the RX ring right away.

Code should just works, please read the following lines in the same
function....

        /* If frame is small enough to fit in skb->head, pull it completely.
         * If not, only pull ieee80211_hdr so that splice() or TCP coalesce
         * are more efficient.
         */
        hdrlen = (len <= skb_tailroom(skb)) ? len : sizeof(*hdr);

        memcpy(skb_put(skb, hdrlen), hdr, hdrlen);
        fraglen = len - hdrlen;

        if (fraglen) {
                int offset = (void *)hdr + hdrlen -
                             rxb_addr(rxb) + rxb_offset(rxb);

                skb_add_rx_frag(skb, 0, rxb_steal_page(rxb), offset,
                                fraglen, rxb->truesize);
        }



--
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