[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <CAE1zotLJ41cOtUVES-CN=kmJzjVTpR_4Mm966aqvEacn21QamQ@mail.gmail.com>
Date: Sun, 8 Jun 2014 19:49:30 +0300
From: Octavian Purdila <octavian.purdila@...el.com>
To: Christoph Paasch <christoph.paasch@...ouvain.be>
Cc: "netdev@...r.kernel.org" <netdev@...r.kernel.org>
Subject: Re: [PATCH net-next] net: add skb allocation flags to __pskb_copy
GOn Sun, Jun 8, 2014 at 6:59 PM, Christoph Paasch
<christoph.paasch@...ouvain.be> wrote:
> Hello Octavian,
>
> On 07/06/14 - 22:19:29, Octavian Purdila wrote:
>> There are several instances where a __pskb_copy is immediately
>> followed by an skb_clone. Add a new parameter to __pskb_copy to allow
>> the skb to be allocated from the fclone cache and thus speed up
>> subsequent skb_clone calls.
>>
>> Signed-off-by: Octavian Purdila <octavian.purdila@...el.com>
>> ---
>> include/linux/skbuff.h | 6 +++---
>> net/bluetooth/hci_sock.c | 5 +++--
>> net/core/skbuff.c | 7 +++++--
>> net/ipv4/tcp_output.c | 2 +-
>> net/nfc/llcp_core.c | 2 +-
>> net/nfc/rawsock.c | 2 +-
>> 6 files changed, 14 insertions(+), 10 deletions(-)
>
> I found two more cases where the stack could benefit from using an FCLONE:
>
> * batadv_nc_skb_store_before_coding() does a pskb_copy and later in
> batadv_nc_skb_store_for_decoding() the skb_clone is done.
>
> * tipc_bcbearer_send() also does a pskb_copy, and the send_msg-callback (in
> tipc_bearer_send()) does a clone too later on in tipc_l2_send_msg().
>
Missed those, I will send a new patch which adds a new parameter to
psbk_copy as well then.
--
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