[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20201201170919.08934eed@carbon>
Date: Tue, 1 Dec 2020 17:09:19 +0100
From: Jesper Dangaard Brouer <brouer@...hat.com>
To: Leesoo Ahn <dev@...eel.net>
Cc: brouer@...hat.com, lsahn@...eel.net,
"David S. Miller" <davem@...emloft.net>,
Jakub Kicinski <kuba@...nel.org>,
Alexei Starovoitov <ast@...nel.org>,
Daniel Borkmann <daniel@...earbox.net>,
Jesper Dangaard Brouer <hawk@...nel.org>,
John Fastabend <john.fastabend@...il.com>,
netdev@...r.kernel.org, bpf@...r.kernel.org,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH] net: xdp: Give compiler __always_inline hint for
xdp_rxq_info_init()
On Mon, 30 Nov 2020 20:48:25 +0900
Leesoo Ahn <dev@...eel.net> wrote:
> The function has only a statement of calling memset() to
> clear xdp_rxq object. Let it always be an inline function.
No, this is the wrong approach.
The function is already "static", and the compiler have likely already
inlined this code, but we leave it up to the compiler to choose.
Besides this is slowpath code, why are you even trying to optimize this?
> Signed-off-by: Leesoo Ahn <lsahn@...eel.net>
> ---
> net/core/xdp.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/net/core/xdp.c b/net/core/xdp.c
> index 48aba933a5a8..dab72b9a71a1 100644
> --- a/net/core/xdp.c
> +++ b/net/core/xdp.c
> @@ -151,7 +151,7 @@ void xdp_rxq_info_unreg(struct xdp_rxq_info *xdp_rxq)
> }
> EXPORT_SYMBOL_GPL(xdp_rxq_info_unreg);
>
> -static void xdp_rxq_info_init(struct xdp_rxq_info *xdp_rxq)
> +static __always_inline void xdp_rxq_info_init(struct xdp_rxq_info *xdp_rxq)
> {
> memset(xdp_rxq, 0, sizeof(*xdp_rxq));
> }
--
Best regards,
Jesper Dangaard Brouer
MSc.CS, Principal Kernel Engineer at Red Hat
LinkedIn: http://www.linkedin.com/in/brouer
Powered by blists - more mailing lists