Received-SPF: pass (google.com: domain of linux-kernel+bounces-160304-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.48.161 as permitted sender) client-ip=147.75.48.161;
Precedence: bulk
MIME-Version: 1.0
References: <20240426152011.37069-1-richard120310@gmail.com>
In-Reply-To: <20240426152011.37069-1-richard120310@gmail.com>
From: Eric Dumazet <edumazet@google.com>
Date: Fri, 26 Apr 2024 17:32:57 +0200
Message-ID: <CANn89iKenW5SxMGm753z8eawg+7drUz7oZcTR06habjcFmdqVg@mail.gmail.com>
Subject: Re: [PATCH] tcp_bbr: replace lambda expression with bitwise operation
 for bit flip
To: I Hsin Cheng <richard120310@gmail.com>
Cc: davem@davemloft.net, dsahern@kernel.org, kuba@kernel.org, 
	pabeni@redhat.com, netdev@vger.kernel.org, linux-kernel@vger.kernel.org, 
	bpf@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Fri, Apr 26, 2024 at 5:20=E2=80=AFPM I Hsin Cheng <richard120310@gmail.c=
om> wrote:
>
> In the origin implementation in function bbr_update_ack_aggregation(),
> we utilize a lambda expression to flip the bit value of
> bbr->extra_acked_win_idx. Since the data type of
> bbr->extra_acked_win_idx is simply a single bit, we are actually trying
> to perform a bit flip operation, under the fact we can simply perform a
> bitwise not operation on bbr->extra_acked_win_idx.
>
> This way we can elimate the need of possible branches which generate by
> the lambda function, they could result in branch misses sometimes.
> Perform a bitwise not operation is more straightforward and wouldn't
> generate branches.
>
> Signed-off-by: I Hsin Cheng <richard120310@gmail.com>
> ---
>  net/ipv4/tcp_bbr.c | 3 +--
>  1 file changed, 1 insertion(+), 2 deletions(-)
>
> diff --git a/net/ipv4/tcp_bbr.c b/net/ipv4/tcp_bbr.c
> index 146792cd2..75068ba25 100644
> --- a/net/ipv4/tcp_bbr.c
> +++ b/net/ipv4/tcp_bbr.c
> @@ -829,8 +829,7 @@ static void bbr_update_ack_aggregation(struct sock *s=
k,
>                                                 bbr->extra_acked_win_rtts=
 + 1);
>                 if (bbr->extra_acked_win_rtts >=3D bbr_extra_acked_win_rt=
ts) {
>                         bbr->extra_acked_win_rtts =3D 0;
> -                       bbr->extra_acked_win_idx =3D bbr->extra_acked_win=
_idx ?
> -                                                  0 : 1;
> +                       bbr->extra_acked_win_idx =3D ~(bbr->extra_acked_w=
in_idx);
>                         bbr->extra_acked[bbr->extra_acked_win_idx] =3D 0;
>                 }
>         }

Or

bbr->extra_acked_win_idx ^=3D 1;

Note that C compilers generate the same code, for the 3 variants.

They do not generate branches for something simple like this.