From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:59419) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1fPiOy-0005Wg-UW for qemu-devel@nongnu.org; Mon, 04 Jun 2018 01:51:34 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1fPiOv-0005IO-0g for qemu-devel@nongnu.org; Mon, 04 Jun 2018 01:51:33 -0400 Received: from mx3-rdu2.redhat.com ([66.187.233.73]:42286 helo=mx1.redhat.com) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1fPiOu-0005Ht-RJ for qemu-devel@nongnu.org; Mon, 04 Jun 2018 01:51:28 -0400 References: <20180603050546.6827-1-zhangckid@gmail.com> <20180603050546.6827-2-zhangckid@gmail.com> From: Jason Wang Message-ID: Date: Mon, 4 Jun 2018 13:51:19 +0800 MIME-Version: 1.0 In-Reply-To: <20180603050546.6827-2-zhangckid@gmail.com> Content-Type: text/plain; charset=utf-8; format=flowed Content-Language: en-US Content-Transfer-Encoding: quoted-printable Subject: Re: [Qemu-devel] [PATCH V8 01/17] filter-rewriter: fix memory leak for connection in connection_track_table List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Zhang Chen , qemu-devel@nongnu.org, Paolo Bonzini , Juan Quintela , "Dr . David Alan Gilbert" , Eric Blake , Markus Armbruster Cc: zhanghailiang , Li Zhijian On 2018=E5=B9=B406=E6=9C=8803=E6=97=A5 13:05, Zhang Chen wrote: > After a net connection is closed, we didn't clear its releated resource= s > in connection_track_table, which will lead to memory leak. > > Let't track the state of net connection, if it is closed, its related > resources will be cleared up. > > Signed-off-by: zhanghailiang > Signed-off-by: Zhang Chen > --- > net/colo.h | 4 +++ > net/filter-rewriter.c | 69 ++++++++++++++++++++++++++++++++++++++----= - > 2 files changed, 66 insertions(+), 7 deletions(-) > > diff --git a/net/colo.h b/net/colo.h > index da6c36dcf7..cd118510c5 100644 > --- a/net/colo.h > +++ b/net/colo.h > @@ -18,6 +18,7 @@ > #include "slirp/slirp.h" > #include "qemu/jhash.h" > #include "qemu/timer.h" > +#include "slirp/tcp.h" > =20 > #define HASHTABLE_MAX_SIZE 16384 > =20 > @@ -86,6 +87,9 @@ typedef struct Connection { > * run once in independent tcp connection > */ > int syn_flag; > + > + int tcp_state; /* TCP FSM state */ > + tcp_seq fin_ack_seq; /* the seq of 'fin=3D1,ack=3D1' */ So the question is, the state machine is not complete. I suspect there=20 will be corner cases that will be left because of lacking sufficient=20 states. LAST_ACK happens only for passive close. How about active close? So I think we need either maintain a full state machine or not instead=20 of a partial one. We don't want endless bugs. Thanks > } Connection; > =20 > uint32_t connection_key_hash(const void *opaque); > diff --git a/net/filter-rewriter.c b/net/filter-rewriter.c > index 62dad2d773..0909a9a8af 100644 > --- a/net/filter-rewriter.c > +++ b/net/filter-rewriter.c > @@ -59,9 +59,9 @@ static int is_tcp_packet(Packet *pkt) > } > =20 > /* handle tcp packet from primary guest */ > -static int handle_primary_tcp_pkt(NetFilterState *nf, > +static int handle_primary_tcp_pkt(RewriterState *rf, > Connection *conn, > - Packet *pkt) > + Packet *pkt, ConnectionKey *key) > { > struct tcphdr *tcp_pkt; > =20 > @@ -99,15 +99,44 @@ static int handle_primary_tcp_pkt(NetFilterState *n= f, > net_checksum_calculate((uint8_t *)pkt->data + pkt->vnet_h= dr_len, > pkt->size - pkt->vnet_hdr_len); > } > + /* > + * Case 1: > + * The *server* side of this connect is VM, *client* tries to = close > + * the connection. > + * > + * We got 'ack=3D1' packets from client side, it acks 'fin=3D1= , ack=3D1' > + * packet from server side. From this point, we can ensure tha= t there > + * will be no packets in the connection, except that, some err= ors > + * happen between the path of 'filter object' and vNIC, if thi= s rare > + * case really happen, we can still create a new connection, > + * So it is safe to remove the connection from connection_trac= k_table. > + * > + */ > + if ((conn->tcp_state =3D=3D TCPS_LAST_ACK) && > + (ntohl(tcp_pkt->th_ack) =3D=3D (conn->fin_ack_seq + 1))) { > + g_hash_table_remove(rf->connection_track_table, key); > + } > + } > + /* > + * Case 2: > + * The *server* side of this connect is VM, *server* tries to clos= e > + * the connection. > + * > + * We got 'fin=3D1, ack=3D1' packet from client side, we need to > + * record the seq of 'fin=3D1, ack=3D1' packet. > + */ > + if ((tcp_pkt->th_flags & (TH_ACK | TH_FIN)) =3D=3D (TH_ACK | TH_FI= N)) { > + conn->fin_ack_seq =3D htonl(tcp_pkt->th_seq); > + conn->tcp_state =3D TCPS_LAST_ACK; > } > =20 > return 0; > } > =20 > /* handle tcp packet from secondary guest */ > -static int handle_secondary_tcp_pkt(NetFilterState *nf, > +static int handle_secondary_tcp_pkt(RewriterState *rf, > Connection *conn, > - Packet *pkt) > + Packet *pkt, ConnectionKey *key) > { > struct tcphdr *tcp_pkt; > =20 > @@ -139,8 +168,34 @@ static int handle_secondary_tcp_pkt(NetFilterState= *nf, > net_checksum_calculate((uint8_t *)pkt->data + pkt->vnet_h= dr_len, > pkt->size - pkt->vnet_hdr_len); > } > + /* > + * Case 2: > + * The *server* side of this connect is VM, *server* tries to = close > + * the connection. > + * > + * We got 'ack=3D1' packets from server side, it acks 'fin=3D1= , ack=3D1' > + * packet from client side. Like Case 1, there should be no pa= ckets > + * in the connection from now know, But the difference here is > + * if the packet is lost, We will get the resent 'fin=3D1,ack=3D= 1' packet. > + * TODO: Fix above case. > + */ > + if ((conn->tcp_state =3D=3D TCPS_LAST_ACK) && > + (ntohl(tcp_pkt->th_ack) =3D=3D (conn->fin_ack_seq + 1))) { > + g_hash_table_remove(rf->connection_track_table, key); > + } > + } > + /* > + * Case 1: > + * The *server* side of this connect is VM, *client* tries to clos= e > + * the connection. > + * > + * We got 'fin=3D1, ack=3D1' packet from server side, we need to > + * record the seq of 'fin=3D1, ack=3D1' packet. > + */ > + if ((tcp_pkt->th_flags & (TH_ACK | TH_FIN)) =3D=3D (TH_ACK | TH_FI= N)) { > + conn->fin_ack_seq =3D ntohl(tcp_pkt->th_seq); > + conn->tcp_state =3D TCPS_LAST_ACK; > } > - > return 0; > } > =20 > @@ -190,7 +245,7 @@ static ssize_t colo_rewriter_receive_iov(NetFilterS= tate *nf, > =20 > if (sender =3D=3D nf->netdev) { > /* NET_FILTER_DIRECTION_TX */ > - if (!handle_primary_tcp_pkt(nf, conn, pkt)) { > + if (!handle_primary_tcp_pkt(s, conn, pkt, &key)) { > qemu_net_queue_send(s->incoming_queue, sender, 0, > (const uint8_t *)pkt->data, pkt->size, NULL); > packet_destroy(pkt, NULL); > @@ -203,7 +258,7 @@ static ssize_t colo_rewriter_receive_iov(NetFilterS= tate *nf, > } > } else { > /* NET_FILTER_DIRECTION_RX */ > - if (!handle_secondary_tcp_pkt(nf, conn, pkt)) { > + if (!handle_secondary_tcp_pkt(s, conn, pkt, &key)) { > qemu_net_queue_send(s->incoming_queue, sender, 0, > (const uint8_t *)pkt->data, pkt->size, NULL); > packet_destroy(pkt, NULL);