Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=amazon.com header.i=@amazon.com header.b="ct2lhh5L" Received: from smtp-fw-80007.amazon.com (smtp-fw-80007.amazon.com [99.78.197.218]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 4B0BF19D for ; Mon, 27 Nov 2023 15:00:18 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=amazon.com; i=@amazon.com; q=dns/txt; s=amazon201209; t=1701126019; x=1732662019; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=pjZ9158Gt0Qu6mOdfTkV6CUNjM6yBRdpOuq4OMWc6rg=; b=ct2lhh5LS2HcOKD1pjh9KEtzWl3caKf3wPl4rXZ81kcR6tgF2ePjw9Tv YIbufjWjr+PjVL5IL02+uDt5veYTJ6+BkI5zscZKly6jfZEEyNm70qniz 3trR/IZltH1cAluM1Sh8EeB3zUiBkkfO4i45QRxRF21cq6DTeTkKlcMUZ U=; X-IronPort-AV: E=Sophos;i="6.04,232,1695686400"; d="scan'208";a="255815952" Received: from pdx4-co-svc-p1-lb2-vlan2.amazon.com (HELO email-inbound-relay-pdx-2b-m6i4x-cadc3fbd.us-west-2.amazon.com) ([10.25.36.210]) by smtp-border-fw-80007.pdx80.corp.amazon.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 27 Nov 2023 23:00:17 +0000 Received: from smtpout.prod.us-west-2.prod.farcaster.email.amazon.dev (pdx2-ws-svc-p26-lb5-vlan2.pdx.amazon.com [10.39.38.66]) by email-inbound-relay-pdx-2b-m6i4x-cadc3fbd.us-west-2.amazon.com (Postfix) with ESMTPS id EAB4FA0B96; Mon, 27 Nov 2023 23:00:15 +0000 (UTC) Received: from EX19MTAUWB001.ant.amazon.com [10.0.21.151:27116] by smtpin.naws.us-west-2.prod.farcaster.email.amazon.dev [10.0.12.242:2525] with esmtp (Farcaster) id 50388c63-66f4-4581-b804-d0a6dc63c0f1; Mon, 27 Nov 2023 23:00:15 +0000 (UTC) X-Farcaster-Flow-ID: 50388c63-66f4-4581-b804-d0a6dc63c0f1 Received: from EX19D004ANA001.ant.amazon.com (10.37.240.138) by EX19MTAUWB001.ant.amazon.com (10.250.64.248) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.2.1118.39; Mon, 27 Nov 2023 23:00:15 +0000 Received: from 88665a182662.ant.amazon.com.com (10.106.101.15) by EX19D004ANA001.ant.amazon.com (10.37.240.138) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.2.1118.40; Mon, 27 Nov 2023 23:00:12 +0000 From: Kuniyuki Iwashima To: CC: , , , , , , Subject: Re: [PATCH v2 net-next 4/4] af_unix: Try to run GC async. Date: Mon, 27 Nov 2023 15:00:03 -0800 Message-ID: <20231127230003.53974-1-kuniyu@amazon.com> X-Mailer: git-send-email 2.30.2 In-Reply-To: <0d5145b231d9b7c8d2f32277ce5ab56bb1859bff.camel@redhat.com> References: <0d5145b231d9b7c8d2f32277ce5ab56bb1859bff.camel@redhat.com> Precedence: bulk X-Mailing-List: netdev@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Content-Type: text/plain X-ClientProxiedBy: EX19D033UWC004.ant.amazon.com (10.13.139.225) To EX19D004ANA001.ant.amazon.com (10.37.240.138) Precedence: Bulk From: Paolo Abeni Date: Mon, 27 Nov 2023 10:59:03 +0100 > On Wed, 2023-11-22 at 17:47 -0800, Kuniyuki Iwashima wrote: > > If more than 16000 inflight AF_UNIX sockets exist and the garbage > > collector is not running, unix_(dgram|stream)_sendmsg() call unix_gc(). > > Also, they wait for unix_gc() to complete. > > > > In unix_gc(), all inflight AF_UNIX sockets are traversed at least once, > > and more if they are the GC candidate. Thus, sendmsg() significantly > > slows down with too many inflight AF_UNIX sockets. > > > > However, if a process sends data with no AF_UNIX FD, the sendmsg() call > > does not need to wait for GC. After this change, only the process that > > meets the condition below will be blocked under such a situation. > > > > 1) cmsg contains AF_UNIX socket > > 2) more than 32 AF_UNIX sent by the same user are still inflight > > > > Note that even a sendmsg() call that does not meet the condition but has > > AF_UNIX FD will be blocked later in unix_scm_to_skb() by the spinlock, > > but we allow that as a bonus for sane users. > > > > The results below are the time spent in unix_dgram_sendmsg() sending 1 > > byte of data with no FD 4096 times on a host where 32K inflight AF_UNIX > > sockets exist. > > > > Without series: the sane sendmsg() needs to wait gc unreasonably. > > > > $ sudo /usr/share/bcc/tools/funclatency -p 11165 unix_dgram_sendmsg > > Tracing 1 functions for "unix_dgram_sendmsg"... Hit Ctrl-C to end. > > ^C > > nsecs : count distribution > > [...] > > 524288 -> 1048575 : 0 | | > > 1048576 -> 2097151 : 3881 |****************************************| > > 2097152 -> 4194303 : 214 |** | > > 4194304 -> 8388607 : 1 | | > > > > avg = 1825567 nsecs, total: 7477526027 nsecs, count: 4096 > > > > With series: the sane sendmsg() can finish much faster. > > > > $ sudo /usr/share/bcc/tools/funclatency -p 8702 unix_dgram_sendmsg > > Tracing 1 functions for "unix_dgram_sendmsg"... Hit Ctrl-C to end. > > ^C > > nsecs : count distribution > > [...] > > 128 -> 255 : 0 | | > > 256 -> 511 : 4092 |****************************************| > > 512 -> 1023 : 2 | | > > 1024 -> 2047 : 0 | | > > 2048 -> 4095 : 0 | | > > 4096 -> 8191 : 1 | | > > 8192 -> 16383 : 1 | | > > > > avg = 410 nsecs, total: 1680510 nsecs, count: 4096 > > > > Signed-off-by: Kuniyuki Iwashima > > --- > > include/net/af_unix.h | 2 +- > > include/net/scm.h | 1 + > > net/core/scm.c | 5 +++++ > > net/unix/af_unix.c | 6 ++++-- > > net/unix/garbage.c | 10 ++++++++-- > > 5 files changed, 19 insertions(+), 5 deletions(-) > > > > diff --git a/include/net/af_unix.h b/include/net/af_unix.h > > index c628d30ceb19..f8e654d418e6 100644 > > --- a/include/net/af_unix.h > > +++ b/include/net/af_unix.h > > @@ -13,7 +13,7 @@ void unix_notinflight(struct user_struct *user, struct file *fp); > > void unix_destruct_scm(struct sk_buff *skb); > > void io_uring_destruct_scm(struct sk_buff *skb); > > void unix_gc(void); > > -void wait_for_unix_gc(void); > > +void wait_for_unix_gc(struct scm_fp_list *fpl); > > struct unix_sock *unix_get_socket(struct file *filp); > > struct sock *unix_peer_get(struct sock *sk); > > > > diff --git a/include/net/scm.h b/include/net/scm.h > > index e8c76b4be2fe..1ff6a2855064 100644 > > --- a/include/net/scm.h > > +++ b/include/net/scm.h > > @@ -24,6 +24,7 @@ struct scm_creds { > > > > struct scm_fp_list { > > short count; > > + short count_unix; > > short max; > > struct user_struct *user; > > struct file *fp[SCM_MAX_FD]; > > diff --git a/net/core/scm.c b/net/core/scm.c > > index 880027ecf516..c1aae77d120b 100644 > > --- a/net/core/scm.c > > +++ b/net/core/scm.c > > @@ -35,6 +35,7 @@ > > #include > > #include > > #include > > +#include > > > > > > /* > > @@ -105,6 +106,10 @@ static int scm_fp_copy(struct cmsghdr *cmsg, struct scm_fp_list **fplp) > > return -EBADF; > > *fpp++ = file; > > fpl->count++; > > +#if IS_ENABLED(CONFIG_UNIX) > > + if (unix_get_socket(file)) > > + fpl->count_unix++; > > +#endif > > } > > > > if (!fpl->user) > > diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c > > index 1e6f5aaf1cc9..bbad3959751d 100644 > > --- a/net/unix/af_unix.c > > +++ b/net/unix/af_unix.c > > @@ -1925,11 +1925,12 @@ static int unix_dgram_sendmsg(struct socket *sock, struct msghdr *msg, > > long timeo; > > int err; > > > > - wait_for_unix_gc(); > > err = scm_send(sock, msg, &scm, false); > > if (err < 0) > > return err; > > > > + wait_for_unix_gc(scm.fp); > > + > > err = -EOPNOTSUPP; > > if (msg->msg_flags&MSG_OOB) > > goto out; > > @@ -2201,11 +2202,12 @@ static int unix_stream_sendmsg(struct socket *sock, struct msghdr *msg, > > bool fds_sent = false; > > int data_len; > > > > - wait_for_unix_gc(); > > err = scm_send(sock, msg, &scm, false); > > if (err < 0) > > return err; > > > > + wait_for_unix_gc(scm.fp); > > + > > err = -EOPNOTSUPP; > > if (msg->msg_flags & MSG_OOB) { > > #if IS_ENABLED(CONFIG_AF_UNIX_OOB) > > diff --git a/net/unix/garbage.c b/net/unix/garbage.c > > index 8bc93a7e745f..73091d6b7fc4 100644 > > --- a/net/unix/garbage.c > > +++ b/net/unix/garbage.c > > @@ -184,8 +184,9 @@ static void inc_inflight_move_tail(struct unix_sock *u) > > } > > > > #define UNIX_INFLIGHT_TRIGGER_GC 16000 > > +#define UNIX_INFLIGHT_SANE_USER 32 > > I don't have any relevant usage stats for unix sockets, but out of > sheer ignorance on my side '32' looks a bit low. Why/how did you pick > such value? My take was that the peer should receive the fds in timely manner so that no one will be punished, but I admit 32 is small enough, which can be reached by a single SCM_RIGHTS (SCM_MAX_FD == 253) cmsg. So, probably we can bump it up to 1024 or 2048 (> (4 or 8) * SCM_MAX_FD). > > -void wait_for_unix_gc(void) > > +void wait_for_unix_gc(struct scm_fp_list *fpl) > > { > > /* If number of inflight sockets is insane, kick a > > * garbage collect right now. > > @@ -195,7 +196,12 @@ void wait_for_unix_gc(void) > > if (READ_ONCE(unix_tot_inflight) > UNIX_INFLIGHT_TRIGGER_GC) > > queue_work(system_unbound_wq, &unix_gc_work); > > > > - flush_work(&unix_gc_work); > > + /* Penalise users who want to send AF_UNIX sockets > > + * but whose sockets have not been received yet. > > + */ > > + if (fpl && fpl->count_unix && > > + READ_ONCE(fpl->user->unix_inflight) > UNIX_INFLIGHT_SANE_USER) > > + flush_work(&unix_gc_work); > > flush_work() will be called even when 'unix_tot_inflight' is (much) > less then 'UNIX_INFLIGHT_TRIGGER_GC'. Could that cause some regressions > for workload with moderated numbers of fd in flights, where the GC was > never triggered before this series? Ah exactly, I'll add work_pending() in v3. if (!fpl || !fpl->count_unix) return if (work_pending(&unix_gc_work) && READ_ONCE(fpl->user->unix_inflight) > UNIX_INFLIGHT_SANE_USER) flush_work(&unix_gc_work) Thanks!