| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <CAK6E8=dDMSKfHPvf5zCASP+X7dYZ7E6UoY8JhGfboqfnYw0abQ@mail.gmail.com>
Date: Wed, 29 Apr 2015 09:10:29 -0700
From: Yuchung Cheng <ycheng@...gle.com>
To: Eric Dumazet <edumazet@...gle.com>
Cc: "David S. Miller" <davem@...emloft.net>,
netdev <netdev@...r.kernel.org>,
Eric Dumazet <eric.dumazet@...il.com>,
Neal Cardwell <ncardwell@...gle.com>
Subject: Re: [PATCH net-next 1/2] tcp: prepare CC get_info() access from getsockopt()
On Tue, Apr 28, 2015 at 4:23 PM, Eric Dumazet <edumazet@...gle.com> wrote:
> We would like that optional info provided by Congestion Control
> modules using netlink can also be read using getsockopt()
>
> This patch changes get_info() to put this information in a buffer,
> instead of skb, like tcp_get_info(), so that following patch
> can reuse this common infrastructure.
>
> Signed-off-by: Eric Dumazet <edumazet@...gle.com>
> Cc: Yuchung Cheng <ycheng@...gle.com>
> Cc: Neal Cardwell <ncardwell@...gle.com>
Acked-by: Yuchung Cheng <ycheng@...gle.com>
> ---
> include/net/tcp.h | 5 ++++-
> include/uapi/linux/inet_diag.h | 4 ++++
> net/ipv4/inet_diag.c | 8 +++++---
> net/ipv4/tcp_dctcp.c | 20 ++++++++++----------
> net/ipv4/tcp_illinois.c | 21 +++++++++++----------
> net/ipv4/tcp_vegas.c | 19 ++++++++++---------
> net/ipv4/tcp_vegas.h | 3 ++-
> 7 files changed, 46 insertions(+), 34 deletions(-)
>
> diff --git a/include/net/tcp.h b/include/net/tcp.h
> index 051dc5c2802d..d5b3089317fb 100644
> --- a/include/net/tcp.h
> +++ b/include/net/tcp.h
> @@ -804,6 +804,8 @@ enum tcp_ca_ack_event_flags {
> /* Requires ECN/ECT set on all packets */
> #define TCP_CONG_NEEDS_ECN 0x2
>
> +union tcp_cc_info;
> +
> struct tcp_congestion_ops {
> struct list_head list;
> u32 key;
> @@ -829,7 +831,8 @@ struct tcp_congestion_ops {
> /* hook for packet ack accounting (optional) */
> void (*pkts_acked)(struct sock *sk, u32 num_acked, s32 rtt_us);
> /* get info for inet_diag (optional) */
> - int (*get_info)(struct sock *sk, u32 ext, struct sk_buff *skb);
> + size_t (*get_info)(struct sock *sk, u32 ext, int *attr,
> + union tcp_cc_info *info);
>
> char name[TCP_CA_NAME_MAX];
> struct module *owner;
> diff --git a/include/uapi/linux/inet_diag.h b/include/uapi/linux/inet_diag.h
> index d65c0a09efd3..c7093c75bdd6 100644
> --- a/include/uapi/linux/inet_diag.h
> +++ b/include/uapi/linux/inet_diag.h
> @@ -143,4 +143,8 @@ struct tcp_dctcp_info {
> __u32 dctcp_ab_tot;
> };
>
> +union tcp_cc_info {
> + struct tcpvegas_info vegas;
> + struct tcp_dctcp_info dctcp;
> +};
> #endif /* _UAPI_INET_DIAG_H_ */
> diff --git a/net/ipv4/inet_diag.c b/net/ipv4/inet_diag.c
> index bb77ebdae3b3..4d32262c7502 100644
> --- a/net/ipv4/inet_diag.c
> +++ b/net/ipv4/inet_diag.c
> @@ -224,14 +224,16 @@ int inet_sk_diag_fill(struct sock *sk, struct inet_connection_sock *icsk,
> handler->idiag_get_info(sk, r, info);
>
> if (sk->sk_state < TCP_TIME_WAIT) {
> - int err = 0;
> + union tcp_cc_info info;
> + size_t sz = 0;
> + int attr;
>
> rcu_read_lock();
> ca_ops = READ_ONCE(icsk->icsk_ca_ops);
> if (ca_ops && ca_ops->get_info)
> - err = ca_ops->get_info(sk, ext, skb);
> + sz = ca_ops->get_info(sk, ext, &attr, &info);
> rcu_read_unlock();
> - if (err < 0)
> + if (sz && nla_put(skb, attr, sz, &info) < 0)
> goto errout;
> }
>
> diff --git a/net/ipv4/tcp_dctcp.c b/net/ipv4/tcp_dctcp.c
> index 4376016f7fa5..4c41c1287197 100644
> --- a/net/ipv4/tcp_dctcp.c
> +++ b/net/ipv4/tcp_dctcp.c
> @@ -277,7 +277,8 @@ static void dctcp_cwnd_event(struct sock *sk, enum tcp_ca_event ev)
> }
> }
>
> -static int dctcp_get_info(struct sock *sk, u32 ext, struct sk_buff *skb)
> +static size_t dctcp_get_info(struct sock *sk, u32 ext, int *attr,
> + union tcp_cc_info *info)
> {
> const struct dctcp *ca = inet_csk_ca(sk);
>
> @@ -286,18 +287,17 @@ static int dctcp_get_info(struct sock *sk, u32 ext, struct sk_buff *skb)
> */
> if (ext & (1 << (INET_DIAG_DCTCPINFO - 1)) ||
> ext & (1 << (INET_DIAG_VEGASINFO - 1))) {
> - struct tcp_dctcp_info info;
> -
> - memset(&info, 0, sizeof(info));
> + memset(info, 0, sizeof(struct tcp_dctcp_info));
> if (inet_csk(sk)->icsk_ca_ops != &dctcp_reno) {
> - info.dctcp_enabled = 1;
> - info.dctcp_ce_state = (u16) ca->ce_state;
> - info.dctcp_alpha = ca->dctcp_alpha;
> - info.dctcp_ab_ecn = ca->acked_bytes_ecn;
> - info.dctcp_ab_tot = ca->acked_bytes_total;
> + info->dctcp.dctcp_enabled = 1;
> + info->dctcp.dctcp_ce_state = (u16) ca->ce_state;
> + info->dctcp.dctcp_alpha = ca->dctcp_alpha;
> + info->dctcp.dctcp_ab_ecn = ca->acked_bytes_ecn;
> + info->dctcp.dctcp_ab_tot = ca->acked_bytes_total;
> }
>
> - return nla_put(skb, INET_DIAG_DCTCPINFO, sizeof(info), &info);
> + *attr = INET_DIAG_DCTCPINFO;
> + return sizeof(*info);
> }
> return 0;
> }
> diff --git a/net/ipv4/tcp_illinois.c b/net/ipv4/tcp_illinois.c
> index 67476f085e48..f71002e4db0b 100644
> --- a/net/ipv4/tcp_illinois.c
> +++ b/net/ipv4/tcp_illinois.c
> @@ -300,24 +300,25 @@ static u32 tcp_illinois_ssthresh(struct sock *sk)
> }
>
> /* Extract info for Tcp socket info provided via netlink. */
> -static int tcp_illinois_info(struct sock *sk, u32 ext, struct sk_buff *skb)
> +static size_t tcp_illinois_info(struct sock *sk, u32 ext, int *attr,
> + union tcp_cc_info *info)
> {
> const struct illinois *ca = inet_csk_ca(sk);
>
> if (ext & (1 << (INET_DIAG_VEGASINFO - 1))) {
> - struct tcpvegas_info info = {
> - .tcpv_enabled = 1,
> - .tcpv_rttcnt = ca->cnt_rtt,
> - .tcpv_minrtt = ca->base_rtt,
> - };
> + info->vegas.tcpv_enabled = 1;
> + info->vegas.tcpv_rttcnt = ca->cnt_rtt;
> + info->vegas.tcpv_minrtt = ca->base_rtt;
> + info->vegas.tcpv_rtt = 0;
>
> - if (info.tcpv_rttcnt > 0) {
> + if (info->vegas.tcpv_rttcnt > 0) {
> u64 t = ca->sum_rtt;
>
> - do_div(t, info.tcpv_rttcnt);
> - info.tcpv_rtt = t;
> + do_div(t, info->vegas.tcpv_rttcnt);
> + info->vegas.tcpv_rtt = t;
> }
> - return nla_put(skb, INET_DIAG_VEGASINFO, sizeof(info), &info);
> + *attr = INET_DIAG_VEGASINFO;
> + return sizeof(struct tcpvegas_info);
> }
> return 0;
> }
> diff --git a/net/ipv4/tcp_vegas.c b/net/ipv4/tcp_vegas.c
> index c71a1b8f7bde..a6cea1d5e20d 100644
> --- a/net/ipv4/tcp_vegas.c
> +++ b/net/ipv4/tcp_vegas.c
> @@ -286,18 +286,19 @@ static void tcp_vegas_cong_avoid(struct sock *sk, u32 ack, u32 acked)
> }
>
> /* Extract info for Tcp socket info provided via netlink. */
> -int tcp_vegas_get_info(struct sock *sk, u32 ext, struct sk_buff *skb)
> +size_t tcp_vegas_get_info(struct sock *sk, u32 ext, int *attr,
> + union tcp_cc_info *info)
> {
> const struct vegas *ca = inet_csk_ca(sk);
> +
> if (ext & (1 << (INET_DIAG_VEGASINFO - 1))) {
> - struct tcpvegas_info info = {
> - .tcpv_enabled = ca->doing_vegas_now,
> - .tcpv_rttcnt = ca->cntRTT,
> - .tcpv_rtt = ca->baseRTT,
> - .tcpv_minrtt = ca->minRTT,
> - };
> -
> - return nla_put(skb, INET_DIAG_VEGASINFO, sizeof(info), &info);
> + info->vegas.tcpv_enabled = ca->doing_vegas_now,
> + info->vegas.tcpv_rttcnt = ca->cntRTT,
> + info->vegas.tcpv_rtt = ca->baseRTT,
> + info->vegas.tcpv_minrtt = ca->minRTT,
> +
> + *attr = INET_DIAG_VEGASINFO;
> + return sizeof(struct tcpvegas_info);
> }
> return 0;
> }
> diff --git a/net/ipv4/tcp_vegas.h b/net/ipv4/tcp_vegas.h
> index e8a6b33cc61d..ef9da5306c68 100644
> --- a/net/ipv4/tcp_vegas.h
> +++ b/net/ipv4/tcp_vegas.h
> @@ -19,6 +19,7 @@ void tcp_vegas_init(struct sock *sk);
> void tcp_vegas_state(struct sock *sk, u8 ca_state);
> void tcp_vegas_pkts_acked(struct sock *sk, u32 cnt, s32 rtt_us);
> void tcp_vegas_cwnd_event(struct sock *sk, enum tcp_ca_event event);
> -int tcp_vegas_get_info(struct sock *sk, u32 ext, struct sk_buff *skb);
> +size_t tcp_vegas_get_info(struct sock *sk, u32 ext, int *attr,
> + union tcp_cc_info *info);
>
> #endif /* __TCP_VEGAS_H */
> --
> 2.2.0.rc0.207.ga3a616c
>
--
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