| Message ID | 20210607194102.2883-1-paskripkin@gmail.com (mailing list archive) |
|---|---|
| State | Superseded |
| Headers | show |
| Series | net: rds: fix memory leak in rds_recvmsg | expand |
> On 7 Jun 2021, at 21:41, Pavel Skripkin <paskripkin@gmail.com> wrote: > > Syzbot reported memory leak in rds. The problem > was in unputted refcount in case of error. > > int rds_recvmsg(struct socket *sock, struct msghdr *msg, size_t size, > int msg_flags) > { > ... > > if (!rds_next_incoming(rs, &inc)) { > ... > } > > After this "if" inc refcount incremented and > > if (rds_cmsg_recv(inc, msg, rs)) { > ret = -EFAULT; > goto out; > } > ... > out: > return ret; > } > > in case of rds_cmsg_recv() fail the refcount won't be > decremented. And it's easy to see from ftrace log, that > rds_inc_addref() don't have rds_inc_put() pair in > rds_recvmsg() after rds_cmsg_recv() > > 1) | rds_recvmsg() { > 1) 3.721 us | rds_inc_addref(); > 1) 3.853 us | rds_message_inc_copy_to_user(); > 1) + 10.395 us | rds_cmsg_recv(); > 1) + 34.260 us | } > > Fixes: bdbe6fbc6a2f ("RDS: recv.c") > Reported-and-tested-by: syzbot+5134cdf021c4ed5aaa5f@syzkaller.appspotmail.com > Signed-off-by: Pavel Skripkin <paskripkin@gmail.com> Thank for your commit and analyses. One small nit below. > --- > net/rds/recv.c | 3 ++- > 1 file changed, 2 insertions(+), 1 deletion(-) > > diff --git a/net/rds/recv.c b/net/rds/recv.c > index 4db109fb6ec2..3fa16c339bfe 100644 > --- a/net/rds/recv.c > +++ b/net/rds/recv.c > @@ -714,7 +714,7 @@ int rds_recvmsg(struct socket *sock, struct msghdr *msg, size_t size, > > if (rds_cmsg_recv(inc, msg, rs)) { > ret = -EFAULT; > - goto out; > + goto out_put; Would a simple "break;" do it here and no need for the next hunk? Thxs, Håkon > } > rds_recvmsg_zcookie(rs, msg); > > @@ -740,6 +740,7 @@ int rds_recvmsg(struct socket *sock, struct msghdr *msg, size_t size, > break; > } > > +out_put: > if (inc) > rds_inc_put(inc); > > -- > 2.31.1 >
On Tue, 8 Jun 2021 07:11:27 +0000 Haakon Bugge <haakon.bugge@oracle.com> wrote: > > > > On 7 Jun 2021, at 21:41, Pavel Skripkin <paskripkin@gmail.com> > > wrote: > > > > Syzbot reported memory leak in rds. The problem > > was in unputted refcount in case of error. > > > > int rds_recvmsg(struct socket *sock, struct msghdr *msg, size_t > > size, int msg_flags) > > { > > ... > > > > if (!rds_next_incoming(rs, &inc)) { > > ... > > } > > > > After this "if" inc refcount incremented and > > > > if (rds_cmsg_recv(inc, msg, rs)) { > > ret = -EFAULT; > > goto out; > > } > > ... > > out: > > return ret; > > } > > > > in case of rds_cmsg_recv() fail the refcount won't be > > decremented. And it's easy to see from ftrace log, that > > rds_inc_addref() don't have rds_inc_put() pair in > > rds_recvmsg() after rds_cmsg_recv() > > > > 1) | rds_recvmsg() { > > 1) 3.721 us | rds_inc_addref(); > > 1) 3.853 us | rds_message_inc_copy_to_user(); > > 1) + 10.395 us | rds_cmsg_recv(); > > 1) + 34.260 us | } > > > > Fixes: bdbe6fbc6a2f ("RDS: recv.c") > > Reported-and-tested-by: > > syzbot+5134cdf021c4ed5aaa5f@syzkaller.appspotmail.com > > Signed-off-by: Pavel Skripkin <paskripkin@gmail.com> > > Thank for your commit and analyses. One small nit below. > > > --- > > net/rds/recv.c | 3 ++- > > 1 file changed, 2 insertions(+), 1 deletion(-) > > > > diff --git a/net/rds/recv.c b/net/rds/recv.c > > index 4db109fb6ec2..3fa16c339bfe 100644 > > --- a/net/rds/recv.c > > +++ b/net/rds/recv.c > > @@ -714,7 +714,7 @@ int rds_recvmsg(struct socket *sock, struct > > msghdr *msg, size_t size, > > > > if (rds_cmsg_recv(inc, msg, rs)) { > > ret = -EFAULT; > > - goto out; > > + goto out_put; > > Would a simple "break;" do it here and no need for the next hunk? > > > Thxs, Håkon > Sure! I'll send v2 soon. Thank you for feedback :) With regards, Pavel Skripkin
diff --git a/net/rds/recv.c b/net/rds/recv.c index 4db109fb6ec2..3fa16c339bfe 100644 --- a/net/rds/recv.c +++ b/net/rds/recv.c @@ -714,7 +714,7 @@ int rds_recvmsg(struct socket *sock, struct msghdr *msg, size_t size, if (rds_cmsg_recv(inc, msg, rs)) { ret = -EFAULT; - goto out; + goto out_put; } rds_recvmsg_zcookie(rs, msg); @@ -740,6 +740,7 @@ int rds_recvmsg(struct socket *sock, struct msghdr *msg, size_t size, break; } +out_put: if (inc) rds_inc_put(inc);
Syzbot reported memory leak in rds. The problem was in unputted refcount in case of error. int rds_recvmsg(struct socket *sock, struct msghdr *msg, size_t size, int msg_flags) { ... if (!rds_next_incoming(rs, &inc)) { ... } After this "if" inc refcount incremented and if (rds_cmsg_recv(inc, msg, rs)) { ret = -EFAULT; goto out; } ... out: return ret; } in case of rds_cmsg_recv() fail the refcount won't be decremented. And it's easy to see from ftrace log, that rds_inc_addref() don't have rds_inc_put() pair in rds_recvmsg() after rds_cmsg_recv() 1) | rds_recvmsg() { 1) 3.721 us | rds_inc_addref(); 1) 3.853 us | rds_message_inc_copy_to_user(); 1) + 10.395 us | rds_cmsg_recv(); 1) + 34.260 us | } Fixes: bdbe6fbc6a2f ("RDS: recv.c") Reported-and-tested-by: syzbot+5134cdf021c4ed5aaa5f@syzkaller.appspotmail.com Signed-off-by: Pavel Skripkin <paskripkin@gmail.com> --- net/rds/recv.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)