[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20201201170208.67ba8ac7@kicinski-fedora-pc1c0hjn.DHCP.thefacebook.com>
Date: Tue, 1 Dec 2020 17:02:08 -0800
From: Jakub Kicinski <kuba@...nel.org>
To: Julian Wiedmann <jwi@...ux.ibm.com>
Cc: David Miller <davem@...emloft.net>,
linux-netdev <netdev@...r.kernel.org>,
linux-s390 <linux-s390@...r.kernel.org>,
Heiko Carstens <hca@...ux.ibm.com>,
Karsten Graul <kgraul@...ux.ibm.com>,
Sebastian Andrzej Siewior <bigeasy@...utronix.de>
Subject: Re: [PATCH net-next 1/6] s390/ctcm: Avoid temporary allocation of
struct th_header and th_sweep.
On Mon, 30 Nov 2020 11:09:45 +0100 Julian Wiedmann wrote:
> From: Sebastian Andrzej Siewior <bigeasy@...utronix.de>
>
> The size of struct th_header is 8 byte and the size of struct th_sweep
> is 16 byte. The memory for is allocated, initialized, used and
> deallocated a few lines later.
>
> It is more efficient to avoid the allocation/free dance and assign the
> values directly to skb's data part instead of using memcpy() for it.
>
> Avoid an allocation of struct th_sweep/th_header and use the resulting
> skb pointer instead.
>
> Signed-off-by: Sebastian Andrzej Siewior <bigeasy@...utronix.de>
> [jwi: use skb_put_zero(), instead of skb_put() + memset to 0]
> Signed-off-by: Julian Wiedmann <jwi@...ux.ibm.com>
Stuff like that is usually done when skb data cannot be assumed to be
aligned. I don't see where the skbs are allocated here, so fingers
crossed :)
Powered by blists - more mailing lists