[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <1361390977.8579.11.camel@jlt4.sipsolutions.net>
Date: Wed, 20 Feb 2013 21:09:37 +0100
From: Johannes Berg <johannes@...solutions.net>
To: Eric Dumazet <eric.dumazet@...il.com>
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 07:11 -0800, Eric Dumazet wrote:
> 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);
Btw, shouldn't this rather be something like
if (rxb->truesize > PAGE_SIZE && len < SKB_WITH_OVERHEAD(PAGE_SIZE))
to avoid allocating large SKBs in the case that this actually is a big
8k A-MSDU? That wouldn't be given to the TCP RX path anyway because
mac80211 will split the A-MSDU into MSDUs and memcpy along the way.
Anyway we're preparing a patch for 4k A-MSDU (which is really just
something like 3782 bytes or something).
johannes
--
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
Powered by blists - more mailing lists