From 77f300b198f93328c26191b52655ce1b62e202cf Mon Sep 17 00:00:00 2001 From: Daeseok Youn Date: Wed, 16 Apr 2014 14:32:29 +0900 Subject: workqueue: fix bugs in wq_update_unbound_numa() failure path wq_update_unbound_numa() failure path has the following two bugs. - alloc_unbound_pwq() is called without holding wq->mutex; however, if the allocation fails, it jumps to out_unlock which tries to unlock wq->mutex. - The function should switch to dfl_pwq on failure but didn't do so after alloc_unbound_pwq() failure. Fix it by regrabbing wq->mutex and jumping to use_dfl_pwq on alloc_unbound_pwq() failure. Signed-off-by: Daeseok Youn Acked-by: Lai Jiangshan Signed-off-by: Tejun Heo Cc: stable@vger.kernel.org Fixes: 4c16bd327c74 ("workqueue: implement NUMA affinity for unbound workqueues") --- kernel/workqueue.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 0ee63af30bd..3150b217c93 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -4100,7 +4100,8 @@ static void wq_update_unbound_numa(struct workqueue_struct *wq, int cpu, if (!pwq) { pr_warning("workqueue: allocation failed while updating NUMA affinity of \"%s\"\n", wq->name); - goto out_unlock; + mutex_lock(&wq->mutex); + goto use_dfl_pwq; } /* -- cgit v1.2.3-70-g09d2 From e37a06f10994c2ba86f54d8f96734f2483a869b8 Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Thu, 17 Apr 2014 13:53:08 +0800 Subject: cgroup: fix the retry path of cgroup_mount() If we hit the retry path, we'll call parse_cgroupfs_options() again, but the string we pass to it has been modified by the previous call to this function. This bug can be observed by: # mount -t cgroup -o name=foo,cpuset xxx /mnt && umount /mnt && \ mount -t cgroup -o name=foo,cpuset xxx /mnt mount: wrong fs type, bad option, bad superblock on xxx, missing codepage or helper program, or other error ... The second mount passed "name=foo,cpuset" to the parser, and then it hit the retry path and call the parser again, but this time the string passed to the parser is "name=foo". To fix this, we avoid calling parse_cgroupfs_options() again in this case. Signed-off-by: Li Zefan Signed-off-by: Tejun Heo --- kernel/cgroup.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'kernel') diff --git a/kernel/cgroup.c b/kernel/cgroup.c index 9fcdaa705b6..11a03d67635 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -1495,7 +1495,7 @@ static struct dentry *cgroup_mount(struct file_system_type *fs_type, */ if (!use_task_css_set_links) cgroup_enable_task_cg_lists(); -retry: + mutex_lock(&cgroup_tree_mutex); mutex_lock(&cgroup_mutex); @@ -1503,7 +1503,7 @@ retry: ret = parse_cgroupfs_options(data, &opts); if (ret) goto out_unlock; - +retry: /* look for a matching existing root */ if (!opts.subsys_mask && !opts.none && !opts.name) { cgrp_dfl_root_visible = true; @@ -1562,9 +1562,9 @@ retry: if (!atomic_inc_not_zero(&root->cgrp.refcnt)) { mutex_unlock(&cgroup_mutex); mutex_unlock(&cgroup_tree_mutex); - kfree(opts.release_agent); - kfree(opts.name); msleep(10); + mutex_lock(&cgroup_tree_mutex); + mutex_lock(&cgroup_mutex); goto retry; } -- cgit v1.2.3-70-g09d2 From 4d595b866d2c653dc90a492b9973a834eabfa354 Mon Sep 17 00:00:00 2001 From: Lai Jiangshan Date: Fri, 18 Apr 2014 11:04:16 -0400 Subject: workqueue: make rescuer_thread() empty wq->maydays list before exiting After a @pwq is scheduled for emergency execution, other workers may consume the affectd work items before the rescuer gets to them. This means that a workqueue many have pwqs queued on @wq->maydays list while not having any work item pending or in-flight. If destroy_workqueue() executes in such condition, the rescuer may exit without emptying @wq->maydays. This currently doesn't cause any actual harm. destroy_workqueue() can safely destroy all the involved data structures whether @wq->maydays is populated or not as nobody access the list once the rescuer exits. However, this is nasty and makes future development difficult. Let's update rescuer_thread() so that it empties @wq->maydays after seeing should_stop to guarantee that the list is empty on rescuer exit. tj: Updated comment and patch description. Signed-off-by: Lai Jiangshan Signed-off-by: Tejun Heo Cc: stable@vger.kernel.org # v3.10+ --- kernel/workqueue.c | 21 ++++++++++++++++----- 1 file changed, 16 insertions(+), 5 deletions(-) (limited to 'kernel') diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 3150b217c93..6ba0c605422 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -2398,6 +2398,7 @@ static int rescuer_thread(void *__rescuer) struct worker *rescuer = __rescuer; struct workqueue_struct *wq = rescuer->rescue_wq; struct list_head *scheduled = &rescuer->scheduled; + bool should_stop; set_user_nice(current, RESCUER_NICE_LEVEL); @@ -2409,11 +2410,15 @@ static int rescuer_thread(void *__rescuer) repeat: set_current_state(TASK_INTERRUPTIBLE); - if (kthread_should_stop()) { - __set_current_state(TASK_RUNNING); - rescuer->task->flags &= ~PF_WQ_WORKER; - return 0; - } + /* + * By the time the rescuer is requested to stop, the workqueue + * shouldn't have any work pending, but @wq->maydays may still have + * pwq(s) queued. This can happen by non-rescuer workers consuming + * all the work items before the rescuer got to them. Go through + * @wq->maydays processing before acting on should_stop so that the + * list is always empty on exit. + */ + should_stop = kthread_should_stop(); /* see whether any pwq is asking for help */ spin_lock_irq(&wq_mayday_lock); @@ -2459,6 +2464,12 @@ repeat: spin_unlock_irq(&wq_mayday_lock); + if (should_stop) { + __set_current_state(TASK_RUNNING); + rescuer->task->flags &= ~PF_WQ_WORKER; + return 0; + } + /* rescuers should never participate in concurrency management */ WARN_ON_ONCE(!(rescuer->flags & WORKER_NOT_RUNNING)); schedule(); -- cgit v1.2.3-70-g09d2 From 77668c8b559e4fe2acf2a0749c7c83cde49a5025 Mon Sep 17 00:00:00 2001 From: Lai Jiangshan Date: Fri, 18 Apr 2014 11:04:16 -0400 Subject: workqueue: fix a possible race condition between rescuer and pwq-release There is a race condition between rescuer_thread() and pwq_unbound_release_workfn(). Even after a pwq is scheduled for rescue, the associated work items may be consumed by any worker. If all of them are consumed before the rescuer gets to them and the pwq's base ref was put due to attribute change, the pwq may be released while still being linked on @wq->maydays list making the rescuer dereference already freed pwq later. Make send_mayday() pin the target pwq until the rescuer is done with it. tj: Updated comment and patch description. Signed-off-by: Lai Jiangshan Signed-off-by: Tejun Heo Cc: stable@vger.kernel.org # v3.10+ --- kernel/workqueue.c | 12 ++++++++++++ 1 file changed, 12 insertions(+) (limited to 'kernel') diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 6ba0c605422..8edc8718542 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -1916,6 +1916,12 @@ static void send_mayday(struct work_struct *work) /* mayday mayday mayday */ if (list_empty(&pwq->mayday_node)) { + /* + * If @pwq is for an unbound wq, its base ref may be put at + * any time due to an attribute change. Pin @pwq until the + * rescuer is done with it. + */ + get_pwq(pwq); list_add_tail(&pwq->mayday_node, &wq->maydays); wake_up_process(wq->rescuer->task); } @@ -2449,6 +2455,12 @@ repeat: process_scheduled_works(rescuer); + /* + * Put the reference grabbed by send_mayday(). @pool won't + * go away while we're holding its lock. + */ + put_pwq(pwq); + /* * Leave this pool. If keep_working() is %true, notify a * regular worker; otherwise, we end up with 0 concurrency -- cgit v1.2.3-70-g09d2 From 90f62cf30a78721641e08737bda787552428061e Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Wed, 23 Apr 2014 14:29:27 -0700 Subject: net: Use netlink_ns_capable to verify the permisions of netlink messages It is possible by passing a netlink socket to a more privileged executable and then to fool that executable into writing to the socket data that happens to be valid netlink message to do something that privileged executable did not intend to do. To keep this from happening replace bare capable and ns_capable calls with netlink_capable, netlink_net_calls and netlink_ns_capable calls. Which act the same as the previous calls except they verify that the opener of the socket had the desired permissions as well. Reported-by: Andy Lutomirski Signed-off-by: "Eric W. Biederman" Signed-off-by: David S. Miller --- crypto/crypto_user.c | 2 +- drivers/connector/cn_proc.c | 2 +- drivers/scsi/scsi_netlink.c | 2 +- kernel/audit.c | 4 ++-- net/can/gw.c | 4 ++-- net/core/rtnetlink.c | 20 +++++++++++--------- net/dcb/dcbnl.c | 2 +- net/decnet/dn_dev.c | 4 ++-- net/decnet/dn_fib.c | 4 ++-- net/decnet/netfilter/dn_rtmsg.c | 2 +- net/netfilter/nfnetlink.c | 2 +- net/netlink/genetlink.c | 2 +- net/packet/diag.c | 2 +- net/phonet/pn_netlink.c | 8 ++++---- net/sched/act_api.c | 2 +- net/sched/cls_api.c | 2 +- net/sched/sch_api.c | 6 +++--- net/tipc/netlink.c | 2 +- net/xfrm/xfrm_user.c | 2 +- 19 files changed, 38 insertions(+), 36 deletions(-) (limited to 'kernel') diff --git a/crypto/crypto_user.c b/crypto/crypto_user.c index 1512e41cd93..43665d0d090 100644 --- a/crypto/crypto_user.c +++ b/crypto/crypto_user.c @@ -466,7 +466,7 @@ static int crypto_user_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh) type -= CRYPTO_MSG_BASE; link = &crypto_dispatch[type]; - if (!capable(CAP_NET_ADMIN)) + if (!netlink_capable(skb, CAP_NET_ADMIN)) return -EPERM; if ((type == (CRYPTO_MSG_GETALG - CRYPTO_MSG_BASE) && diff --git a/drivers/connector/cn_proc.c b/drivers/connector/cn_proc.c index 148d707a1d4..ccdd4c7e748 100644 --- a/drivers/connector/cn_proc.c +++ b/drivers/connector/cn_proc.c @@ -369,7 +369,7 @@ static void cn_proc_mcast_ctl(struct cn_msg *msg, return; /* Can only change if privileged. */ - if (!capable(CAP_NET_ADMIN)) { + if (!__netlink_ns_capable(nsp, &init_user_ns, CAP_NET_ADMIN)) { err = EPERM; goto out; } diff --git a/drivers/scsi/scsi_netlink.c b/drivers/scsi/scsi_netlink.c index fe30ea94ffe..109802f776e 100644 --- a/drivers/scsi/scsi_netlink.c +++ b/drivers/scsi/scsi_netlink.c @@ -77,7 +77,7 @@ scsi_nl_rcv_msg(struct sk_buff *skb) goto next_msg; } - if (!capable(CAP_SYS_ADMIN)) { + if (!netlink_capable(skb, CAP_SYS_ADMIN)) { err = -EPERM; goto next_msg; } diff --git a/kernel/audit.c b/kernel/audit.c index 7c2893602d0..47845c57eb1 100644 --- a/kernel/audit.c +++ b/kernel/audit.c @@ -643,13 +643,13 @@ static int audit_netlink_ok(struct sk_buff *skb, u16 msg_type) if ((task_active_pid_ns(current) != &init_pid_ns)) return -EPERM; - if (!capable(CAP_AUDIT_CONTROL)) + if (!netlink_capable(skb, CAP_AUDIT_CONTROL)) err = -EPERM; break; case AUDIT_USER: case AUDIT_FIRST_USER_MSG ... AUDIT_LAST_USER_MSG: case AUDIT_FIRST_USER_MSG2 ... AUDIT_LAST_USER_MSG2: - if (!capable(CAP_AUDIT_WRITE)) + if (!netlink_capable(skb, CAP_AUDIT_WRITE)) err = -EPERM; break; default: /* bad msg */ diff --git a/net/can/gw.c b/net/can/gw.c index ac31891967d..050a2110d43 100644 --- a/net/can/gw.c +++ b/net/can/gw.c @@ -804,7 +804,7 @@ static int cgw_create_job(struct sk_buff *skb, struct nlmsghdr *nlh) u8 limhops = 0; int err = 0; - if (!capable(CAP_NET_ADMIN)) + if (!netlink_capable(skb, CAP_NET_ADMIN)) return -EPERM; if (nlmsg_len(nlh) < sizeof(*r)) @@ -893,7 +893,7 @@ static int cgw_remove_job(struct sk_buff *skb, struct nlmsghdr *nlh) u8 limhops = 0; int err = 0; - if (!capable(CAP_NET_ADMIN)) + if (!netlink_capable(skb, CAP_NET_ADMIN)) return -EPERM; if (nlmsg_len(nlh) < sizeof(*r)) diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index d4ff41739b0..64ad17d077e 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -1395,7 +1395,8 @@ static int do_set_master(struct net_device *dev, int ifindex) return 0; } -static int do_setlink(struct net_device *dev, struct ifinfomsg *ifm, +static int do_setlink(const struct sk_buff *skb, + struct net_device *dev, struct ifinfomsg *ifm, struct nlattr **tb, char *ifname, int modified) { const struct net_device_ops *ops = dev->netdev_ops; @@ -1407,7 +1408,7 @@ static int do_setlink(struct net_device *dev, struct ifinfomsg *ifm, err = PTR_ERR(net); goto errout; } - if (!ns_capable(net->user_ns, CAP_NET_ADMIN)) { + if (!netlink_ns_capable(skb, net->user_ns, CAP_NET_ADMIN)) { err = -EPERM; goto errout; } @@ -1661,7 +1662,7 @@ static int rtnl_setlink(struct sk_buff *skb, struct nlmsghdr *nlh) if (err < 0) goto errout; - err = do_setlink(dev, ifm, tb, ifname, 0); + err = do_setlink(skb, dev, ifm, tb, ifname, 0); errout: return err; } @@ -1778,7 +1779,8 @@ err: } EXPORT_SYMBOL(rtnl_create_link); -static int rtnl_group_changelink(struct net *net, int group, +static int rtnl_group_changelink(const struct sk_buff *skb, + struct net *net, int group, struct ifinfomsg *ifm, struct nlattr **tb) { @@ -1787,7 +1789,7 @@ static int rtnl_group_changelink(struct net *net, int group, for_each_netdev(net, dev) { if (dev->group == group) { - err = do_setlink(dev, ifm, tb, NULL, 0); + err = do_setlink(skb, dev, ifm, tb, NULL, 0); if (err < 0) return err; } @@ -1929,12 +1931,12 @@ replay: modified = 1; } - return do_setlink(dev, ifm, tb, ifname, modified); + return do_setlink(skb, dev, ifm, tb, ifname, modified); } if (!(nlh->nlmsg_flags & NLM_F_CREATE)) { if (ifm->ifi_index == 0 && tb[IFLA_GROUP]) - return rtnl_group_changelink(net, + return rtnl_group_changelink(skb, net, nla_get_u32(tb[IFLA_GROUP]), ifm, tb); return -ENODEV; @@ -2321,7 +2323,7 @@ static int rtnl_fdb_del(struct sk_buff *skb, struct nlmsghdr *nlh) int err = -EINVAL; __u8 *addr; - if (!capable(CAP_NET_ADMIN)) + if (!netlink_capable(skb, CAP_NET_ADMIN)) return -EPERM; err = nlmsg_parse(nlh, sizeof(*ndm), tb, NDA_MAX, NULL); @@ -2773,7 +2775,7 @@ static int rtnetlink_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh) sz_idx = type>>2; kind = type&3; - if (kind != 2 && !ns_capable(net->user_ns, CAP_NET_ADMIN)) + if (kind != 2 && !netlink_net_capable(skb, CAP_NET_ADMIN)) return -EPERM; if (kind == 2 && nlh->nlmsg_flags&NLM_F_DUMP) { diff --git a/net/dcb/dcbnl.c b/net/dcb/dcbnl.c index 55364440267..f8b98d89c28 100644 --- a/net/dcb/dcbnl.c +++ b/net/dcb/dcbnl.c @@ -1669,7 +1669,7 @@ static int dcb_doit(struct sk_buff *skb, struct nlmsghdr *nlh) struct nlmsghdr *reply_nlh = NULL; const struct reply_func *fn; - if ((nlh->nlmsg_type == RTM_SETDCB) && !capable(CAP_NET_ADMIN)) + if ((nlh->nlmsg_type == RTM_SETDCB) && !netlink_capable(skb, CAP_NET_ADMIN)) return -EPERM; ret = nlmsg_parse(nlh, sizeof(*dcb), tb, DCB_ATTR_MAX, diff --git a/net/decnet/dn_dev.c b/net/decnet/dn_dev.c index a603823a3e2..3b726f31c64 100644 --- a/net/decnet/dn_dev.c +++ b/net/decnet/dn_dev.c @@ -574,7 +574,7 @@ static int dn_nl_deladdr(struct sk_buff *skb, struct nlmsghdr *nlh) struct dn_ifaddr __rcu **ifap; int err = -EINVAL; - if (!capable(CAP_NET_ADMIN)) + if (!netlink_capable(skb, CAP_NET_ADMIN)) return -EPERM; if (!net_eq(net, &init_net)) @@ -618,7 +618,7 @@ static int dn_nl_newaddr(struct sk_buff *skb, struct nlmsghdr *nlh) struct dn_ifaddr *ifa; int err; - if (!capable(CAP_NET_ADMIN)) + if (!netlink_capable(skb, CAP_NET_ADMIN)) return -EPERM; if (!net_eq(net, &init_net)) diff --git a/net/decnet/dn_fib.c b/net/decnet/dn_fib.c index 57dc159245e..d332aefb084 100644 --- a/net/decnet/dn_fib.c +++ b/net/decnet/dn_fib.c @@ -505,7 +505,7 @@ static int dn_fib_rtm_delroute(struct sk_buff *skb, struct nlmsghdr *nlh) struct nlattr *attrs[RTA_MAX+1]; int err; - if (!capable(CAP_NET_ADMIN)) + if (!netlink_capable(skb, CAP_NET_ADMIN)) return -EPERM; if (!net_eq(net, &init_net)) @@ -530,7 +530,7 @@ static int dn_fib_rtm_newroute(struct sk_buff *skb, struct nlmsghdr *nlh) struct nlattr *attrs[RTA_MAX+1]; int err; - if (!capable(CAP_NET_ADMIN)) + if (!netlink_capable(skb, CAP_NET_ADMIN)) return -EPERM; if (!net_eq(net, &init_net)) diff --git a/net/decnet/netfilter/dn_rtmsg.c b/net/decnet/netfilter/dn_rtmsg.c index e83015cecfa..e4d9560a910 100644 --- a/net/decnet/netfilter/dn_rtmsg.c +++ b/net/decnet/netfilter/dn_rtmsg.c @@ -107,7 +107,7 @@ static inline void dnrmg_receive_user_skb(struct sk_buff *skb) if (nlh->nlmsg_len < sizeof(*nlh) || skb->len < nlh->nlmsg_len) return; - if (!capable(CAP_NET_ADMIN)) + if (!netlink_capable(skb, CAP_NET_ADMIN)) RCV_SKB_FAIL(-EPERM); /* Eventually we might send routing messages too */ diff --git a/net/netfilter/nfnetlink.c b/net/netfilter/nfnetlink.c index e8138da4c14..84392f3237c 100644 --- a/net/netfilter/nfnetlink.c +++ b/net/netfilter/nfnetlink.c @@ -375,7 +375,7 @@ static void nfnetlink_rcv(struct sk_buff *skb) skb->len < nlh->nlmsg_len) return; - if (!ns_capable(net->user_ns, CAP_NET_ADMIN)) { + if (!netlink_net_capable(skb, CAP_NET_ADMIN)) { netlink_ack(skb, nlh, -EPERM); return; } diff --git a/net/netlink/genetlink.c b/net/netlink/genetlink.c index b1dcdb932a8..a3ba3ca0ff9 100644 --- a/net/netlink/genetlink.c +++ b/net/netlink/genetlink.c @@ -561,7 +561,7 @@ static int genl_family_rcv_msg(struct genl_family *family, return -EOPNOTSUPP; if ((ops->flags & GENL_ADMIN_PERM) && - !capable(CAP_NET_ADMIN)) + !netlink_capable(skb, CAP_NET_ADMIN)) return -EPERM; if ((nlh->nlmsg_flags & NLM_F_DUMP) == NLM_F_DUMP) { diff --git a/net/packet/diag.c b/net/packet/diag.c index b34d0de2409..92f2c7107ee 100644 --- a/net/packet/diag.c +++ b/net/packet/diag.c @@ -194,7 +194,7 @@ static int packet_diag_dump(struct sk_buff *skb, struct netlink_callback *cb) net = sock_net(skb->sk); req = nlmsg_data(cb->nlh); - may_report_filterinfo = ns_capable(net->user_ns, CAP_NET_ADMIN); + may_report_filterinfo = netlink_net_capable(cb->skb, CAP_NET_ADMIN); mutex_lock(&net->packet.sklist_lock); sk_for_each(sk, &net->packet.sklist) { diff --git a/net/phonet/pn_netlink.c b/net/phonet/pn_netlink.c index dc15f430080..b64151ade6b 100644 --- a/net/phonet/pn_netlink.c +++ b/net/phonet/pn_netlink.c @@ -70,10 +70,10 @@ static int addr_doit(struct sk_buff *skb, struct nlmsghdr *nlh) int err; u8 pnaddr; - if (!capable(CAP_NET_ADMIN)) + if (!netlink_capable(skb, CAP_NET_ADMIN)) return -EPERM; - if (!capable(CAP_SYS_ADMIN)) + if (!netlink_capable(skb, CAP_SYS_ADMIN)) return -EPERM; ASSERT_RTNL(); @@ -233,10 +233,10 @@ static int route_doit(struct sk_buff *skb, struct nlmsghdr *nlh) int err; u8 dst; - if (!capable(CAP_NET_ADMIN)) + if (!netlink_capable(skb, CAP_NET_ADMIN)) return -EPERM; - if (!capable(CAP_SYS_ADMIN)) + if (!netlink_capable(skb, CAP_SYS_ADMIN)) return -EPERM; ASSERT_RTNL(); diff --git a/net/sched/act_api.c b/net/sched/act_api.c index 8a5ba5add4b..648778aef1a 100644 --- a/net/sched/act_api.c +++ b/net/sched/act_api.c @@ -948,7 +948,7 @@ static int tc_ctl_action(struct sk_buff *skb, struct nlmsghdr *n) u32 portid = skb ? NETLINK_CB(skb).portid : 0; int ret = 0, ovr = 0; - if ((n->nlmsg_type != RTM_GETACTION) && !capable(CAP_NET_ADMIN)) + if ((n->nlmsg_type != RTM_GETACTION) && !netlink_capable(skb, CAP_NET_ADMIN)) return -EPERM; ret = nlmsg_parse(n, sizeof(struct tcamsg), tca, TCA_ACT_MAX, NULL); diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c index 29a30a14c31..bdbdb1a7920 100644 --- a/net/sched/cls_api.c +++ b/net/sched/cls_api.c @@ -134,7 +134,7 @@ static int tc_ctl_tfilter(struct sk_buff *skb, struct nlmsghdr *n) int err; int tp_created = 0; - if ((n->nlmsg_type != RTM_GETTFILTER) && !capable(CAP_NET_ADMIN)) + if ((n->nlmsg_type != RTM_GETTFILTER) && !netlink_capable(skb, CAP_NET_ADMIN)) return -EPERM; replay: diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index a0b84e0e22d..400769014bb 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -1084,7 +1084,7 @@ static int tc_get_qdisc(struct sk_buff *skb, struct nlmsghdr *n) struct Qdisc *p = NULL; int err; - if ((n->nlmsg_type != RTM_GETQDISC) && !capable(CAP_NET_ADMIN)) + if ((n->nlmsg_type != RTM_GETQDISC) && !netlink_capable(skb, CAP_NET_ADMIN)) return -EPERM; err = nlmsg_parse(n, sizeof(*tcm), tca, TCA_MAX, NULL); @@ -1151,7 +1151,7 @@ static int tc_modify_qdisc(struct sk_buff *skb, struct nlmsghdr *n) struct Qdisc *q, *p; int err; - if (!capable(CAP_NET_ADMIN)) + if (!netlink_capable(skb, CAP_NET_ADMIN)) return -EPERM; replay: @@ -1490,7 +1490,7 @@ static int tc_ctl_tclass(struct sk_buff *skb, struct nlmsghdr *n) u32 qid; int err; - if ((n->nlmsg_type != RTM_GETTCLASS) && !capable(CAP_NET_ADMIN)) + if ((n->nlmsg_type != RTM_GETTCLASS) && !netlink_capable(skb, CAP_NET_ADMIN)) return -EPERM; err = nlmsg_parse(n, sizeof(*tcm), tca, TCA_MAX, NULL); diff --git a/net/tipc/netlink.c b/net/tipc/netlink.c index 3aaf73de9e2..ad844d36534 100644 --- a/net/tipc/netlink.c +++ b/net/tipc/netlink.c @@ -47,7 +47,7 @@ static int handle_cmd(struct sk_buff *skb, struct genl_info *info) int hdr_space = nlmsg_total_size(GENL_HDRLEN + TIPC_GENL_HDRLEN); u16 cmd; - if ((req_userhdr->cmd & 0xC000) && (!capable(CAP_NET_ADMIN))) + if ((req_userhdr->cmd & 0xC000) && (!netlink_capable(skb, CAP_NET_ADMIN))) cmd = TIPC_CMD_NOT_NET_ADMIN; else cmd = req_userhdr->cmd; diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index 8f131c10a6f..51398ae6cda 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -2377,7 +2377,7 @@ static int xfrm_user_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh) link = &xfrm_dispatch[type]; /* All operations require privileges, even GET */ - if (!ns_capable(net->user_ns, CAP_NET_ADMIN)) + if (!netlink_net_capable(skb, CAP_NET_ADMIN)) return -EPERM; if ((type == (XFRM_MSG_GETSA - XFRM_MSG_BASE) || -- cgit v1.2.3-70-g09d2 From 79465d2fd48e68940c2bdecddbdecd45bbba06fe Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Mon, 28 Apr 2014 11:05:43 +0930 Subject: module: remove warning about waiting module removal. We remove the waiting module removal in commit 3f2b9c9cdf38 (September 2013), but it turns out that modprobe in kmod (< version 16) was asking for waiting module removal. No one noticed since modprobe would check for 0 usage immediately before trying to remove the module, and the race is unlikely. However, it means that anyone running old (but not ancient) kmod versions is hitting the printk designed to see if anyone was running "rmmod -w". All reports so far have been false positives, so remove the warning. Fixes: 3f2b9c9cdf389e303b2273679af08aab5f153517 Reported-by: Valerio Vanni Cc: Elliott, Robert (Server Storage) Cc: stable@kernel.org Acked-by: Lucas De Marchi Signed-off-by: Rusty Russell --- kernel/module.c | 3 --- 1 file changed, 3 deletions(-) (limited to 'kernel') diff --git a/kernel/module.c b/kernel/module.c index 11869408f79..ae7821898bf 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -815,9 +815,6 @@ SYSCALL_DEFINE2(delete_module, const char __user *, name_user, return -EFAULT; name[MODULE_NAME_LEN-1] = '\0'; - if (!(flags & O_NONBLOCK)) - pr_warn("waiting module removal not supported: please upgrade\n"); - if (mutex_lock_interruptible(&module_mutex) != 0) return -EINTR; -- cgit v1.2.3-70-g09d2 From 62a08ae2a5763aabeee98264605236b001503e0c Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Thu, 24 Apr 2014 09:50:53 +0200 Subject: genirq: x86: Ensure that dynamic irq allocation does not conflict On x86 the allocation of irq descriptors may allocate interrupts which are in the range of the GSI interrupts. That's wrong as those interrupts are hardwired and we don't have the irq domain translation like PPC. So one of these interrupts can be hooked up later to one of the devices which are hard wired to it and the io_apic init code for that particular interrupt line happily reuses that descriptor with a completely different configuration so hell breaks lose. Inside x86 we allocate dynamic interrupts from above nr_gsi_irqs, except for a few usage sites which have not yet blown up in our face for whatever reason. But for drivers which need an irq range, like the GPIO drivers, we have no limit in place and we don't want to expose such a detail to a driver. To cure this introduce a function which an architecture can implement to impose a lower bound on the dynamic interrupt allocations. Implement it for x86 and set the lower bound to nr_gsi_irqs, which is the end of the hardwired interrupt space, so all dynamic allocations happen above. That not only allows the GPIO driver to work sanely, it also protects the bogus callsites of create_irq_nr() in hpet, uv, irq_remapping and htirq code. They need to be cleaned up as well, but that's a separate issue. Reported-by: Jin Yao Signed-off-by: Thomas Gleixner Tested-by: Mika Westerberg Cc: Mathias Nyman Cc: Linus Torvalds Cc: Grant Likely Cc: H. Peter Anvin Cc: Rafael J. Wysocki Cc: Andy Shevchenko Cc: Krogerus Heikki Cc: Linus Walleij Link: http://lkml.kernel.org/r/alpine.DEB.2.02.1404241617360.28206@ionos.tec.linutronix.de Signed-off-by: Thomas Gleixner --- arch/x86/kernel/apic/io_apic.c | 5 +++++ include/linux/irq.h | 2 ++ kernel/irq/irqdesc.c | 7 +++++++ kernel/softirq.c | 5 +++++ 4 files changed, 19 insertions(+) (limited to 'kernel') diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index 6ad4658de70..d23aa82e7a7 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -3425,6 +3425,11 @@ int get_nr_irqs_gsi(void) return nr_irqs_gsi; } +unsigned int arch_dynirq_lower_bound(unsigned int from) +{ + return from < nr_irqs_gsi ? nr_irqs_gsi : from; +} + int __init arch_probe_nr_irqs(void) { int nr; diff --git a/include/linux/irq.h b/include/linux/irq.h index 10a0b1ac4ea..5c57efb863d 100644 --- a/include/linux/irq.h +++ b/include/linux/irq.h @@ -603,6 +603,8 @@ static inline u32 irq_get_trigger_type(unsigned int irq) return d ? irqd_get_trigger_type(d) : 0; } +unsigned int arch_dynirq_lower_bound(unsigned int from); + int __irq_alloc_descs(int irq, unsigned int from, unsigned int cnt, int node, struct module *owner); diff --git a/kernel/irq/irqdesc.c b/kernel/irq/irqdesc.c index a7174617616..bb07f2928f4 100644 --- a/kernel/irq/irqdesc.c +++ b/kernel/irq/irqdesc.c @@ -363,6 +363,13 @@ __irq_alloc_descs(int irq, unsigned int from, unsigned int cnt, int node, if (from > irq) return -EINVAL; from = irq; + } else { + /* + * For interrupts which are freely allocated the + * architecture can force a lower bound to the @from + * argument. x86 uses this to exclude the GSI space. + */ + from = arch_dynirq_lower_bound(from); } mutex_lock(&sparse_irq_lock); diff --git a/kernel/softirq.c b/kernel/softirq.c index b50990a5bea..33e4648ae0e 100644 --- a/kernel/softirq.c +++ b/kernel/softirq.c @@ -779,3 +779,8 @@ int __init __weak arch_early_irq_init(void) { return 0; } + +unsigned int __weak arch_dynirq_lower_bound(unsigned int from) +{ + return from; +} -- cgit v1.2.3-70-g09d2 From a949ae560a511fe4e3adf48fa44fefded93e5c2b Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Red Hat)" Date: Thu, 24 Apr 2014 10:40:12 -0400 Subject: ftrace/module: Hardcode ftrace_module_init() call into load_module() A race exists between module loading and enabling of function tracer. CPU 1 CPU 2 ----- ----- load_module() module->state = MODULE_STATE_COMING register_ftrace_function() mutex_lock(&ftrace_lock); ftrace_startup() update_ftrace_function(); ftrace_arch_code_modify_prepare() set_all_module_text_rw(); ftrace_arch_code_modify_post_process() set_all_module_text_ro(); [ here all module text is set to RO, including the module that is loading!! ] blocking_notifier_call_chain(MODULE_STATE_COMING); ftrace_init_module() [ tries to modify code, but it's RO, and fails! ftrace_bug() is called] When this race happens, ftrace_bug() will produces a nasty warning and all of the function tracing features will be disabled until reboot. The simple solution is to treate module load the same way the core kernel is treated at boot. To hardcode the ftrace function modification of converting calls to mcount into nops. This is done in init/main.c there's no reason it could not be done in load_module(). This gives a better control of the changes and doesn't tie the state of the module to its notifiers as much. Ftrace is special, it needs to be treated as such. The reason this would work, is that the ftrace_module_init() would be called while the module is in MODULE_STATE_UNFORMED, which is ignored by the set_all_module_text_ro() call. Link: http://lkml.kernel.org/r/1395637826-3312-1-git-send-email-indou.takao@jp.fujitsu.com Reported-by: Takao Indoh Acked-by: Rusty Russell Cc: stable@vger.kernel.org # 2.6.38+ Signed-off-by: Steven Rostedt --- include/linux/ftrace.h | 2 ++ kernel/module.c | 3 +++ kernel/trace/ftrace.c | 27 ++++----------------------- 3 files changed, 9 insertions(+), 23 deletions(-) (limited to 'kernel') diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h index 9212b017bc7..ae9504b4b67 100644 --- a/include/linux/ftrace.h +++ b/include/linux/ftrace.h @@ -535,6 +535,7 @@ static inline int ftrace_modify_call(struct dyn_ftrace *rec, unsigned long old_a extern int ftrace_arch_read_dyn_info(char *buf, int size); extern int skip_trace(unsigned long ip); +extern void ftrace_module_init(struct module *mod); extern void ftrace_disable_daemon(void); extern void ftrace_enable_daemon(void); @@ -544,6 +545,7 @@ static inline int ftrace_force_update(void) { return 0; } static inline void ftrace_disable_daemon(void) { } static inline void ftrace_enable_daemon(void) { } static inline void ftrace_release_mod(struct module *mod) {} +static inline void ftrace_module_init(struct module *mod) {} static inline __init int register_ftrace_command(struct ftrace_func_command *cmd) { return -EINVAL; diff --git a/kernel/module.c b/kernel/module.c index 11869408f79..5f14fec9f82 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -3271,6 +3271,9 @@ static int load_module(struct load_info *info, const char __user *uargs, dynamic_debug_setup(info->debug, info->num_debug); + /* Ftrace init must be called in the MODULE_STATE_UNFORMED state */ + ftrace_module_init(mod); + /* Finally it's fully formed, ready to start executing. */ err = complete_formation(mod, info); if (err) diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 1fd4b947921..4a54a25afa2 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -4330,16 +4330,11 @@ static void ftrace_init_module(struct module *mod, ftrace_process_locs(mod, start, end); } -static int ftrace_module_notify_enter(struct notifier_block *self, - unsigned long val, void *data) +void ftrace_module_init(struct module *mod) { - struct module *mod = data; - - if (val == MODULE_STATE_COMING) - ftrace_init_module(mod, mod->ftrace_callsites, - mod->ftrace_callsites + - mod->num_ftrace_callsites); - return 0; + ftrace_init_module(mod, mod->ftrace_callsites, + mod->ftrace_callsites + + mod->num_ftrace_callsites); } static int ftrace_module_notify_exit(struct notifier_block *self, @@ -4353,11 +4348,6 @@ static int ftrace_module_notify_exit(struct notifier_block *self, return 0; } #else -static int ftrace_module_notify_enter(struct notifier_block *self, - unsigned long val, void *data) -{ - return 0; -} static int ftrace_module_notify_exit(struct notifier_block *self, unsigned long val, void *data) { @@ -4365,11 +4355,6 @@ static int ftrace_module_notify_exit(struct notifier_block *self, } #endif /* CONFIG_MODULES */ -struct notifier_block ftrace_module_enter_nb = { - .notifier_call = ftrace_module_notify_enter, - .priority = INT_MAX, /* Run before anything that can use kprobes */ -}; - struct notifier_block ftrace_module_exit_nb = { .notifier_call = ftrace_module_notify_exit, .priority = INT_MIN, /* Run after anything that can remove kprobes */ @@ -4403,10 +4388,6 @@ void __init ftrace_init(void) __start_mcount_loc, __stop_mcount_loc); - ret = register_module_notifier(&ftrace_module_enter_nb); - if (ret) - pr_warning("Failed to register trace ftrace module enter notifier\n"); - ret = register_module_notifier(&ftrace_module_exit_nb); if (ret) pr_warning("Failed to register trace ftrace module exit notifier\n"); -- cgit v1.2.3-70-g09d2 From 6c6c0d5a1c949d2e084706f9e5fb1fccc175b265 Mon Sep 17 00:00:00 2001 From: Stuart Hayes Date: Tue, 29 Apr 2014 17:55:02 -0500 Subject: hrtimer: Prevent all reprogramming if hang detected If the last hrtimer interrupt detected a hang it sets hang_detected=1 and programs the clock event device with a delay to let the system make progress. If hang_detected == 1, we prevent reprogramming of the clock event device in hrtimer_reprogram() but not in hrtimer_force_reprogram(). This can lead to the following situation: hrtimer_interrupt() hang_detected = 1; program ce device to Xms from now (hang delay) We have two timers pending: T1 expires 50ms from now T2 expires 5s from now Now T1 gets canceled, which causes hrtimer_force_reprogram() to be invoked, which in turn programs the clock event device to T2 (5 seconds from now). Any hrtimer_start after that will not reprogram the hardware due to hang_detected still being set. So we effectivly block all timers until the T2 event fires and cleans up the hang situation. Add a check for hang_detected to hrtimer_force_reprogram() which prevents the reprogramming of the hang delay in the hardware timer. The subsequent hrtimer_interrupt will resolve all outstanding issues. [ tglx: Rewrote subject and changelog and fixed up the comment in hrtimer_force_reprogram() ] Signed-off-by: Stuart Hayes Link: http://lkml.kernel.org/r/53602DC6.2060101@gmail.com Cc: stable@vger.kernel.org Signed-off-by: Thomas Gleixner --- kernel/hrtimer.c | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) (limited to 'kernel') diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c index d55092ceee2..e3724fdac2d 100644 --- a/kernel/hrtimer.c +++ b/kernel/hrtimer.c @@ -569,6 +569,23 @@ hrtimer_force_reprogram(struct hrtimer_cpu_base *cpu_base, int skip_equal) cpu_base->expires_next.tv64 = expires_next.tv64; + /* + * If a hang was detected in the last timer interrupt then we + * leave the hang delay active in the hardware. We want the + * system to make progress. That also prevents the following + * scenario: + * T1 expires 50ms from now + * T2 expires 5s from now + * + * T1 is removed, so this code is called and would reprogram + * the hardware to 5s from now. Any hrtimer_start after that + * will not reprogram the hardware due to hang_detected being + * set. So we'd effectivly block all timers until the T2 event + * fires. + */ + if (cpu_base->hang_detected) + return; + if (cpu_base->expires_next.tv64 != KTIME_MAX) tick_program_event(cpu_base->expires_next, 1); } -- cgit v1.2.3-70-g09d2 From 012a45e3f4af68e86d85cce060c6c2fed56498b2 Mon Sep 17 00:00:00 2001 From: Leon Ma Date: Wed, 30 Apr 2014 16:43:10 +0800 Subject: hrtimer: Prevent remote enqueue of leftmost timers If a cpu is idle and starts an hrtimer which is not pinned on that same cpu, the nohz code might target the timer to a different cpu. In the case that we switch the cpu base of the timer we already have a sanity check in place, which determines whether the timer is earlier than the current leftmost timer on the target cpu. In that case we enqueue the timer on the current cpu because we cannot reprogram the clock event device on the target. If the timers base is already the target CPU we do not have this sanity check in place so we enqueue the timer as the leftmost timer in the target cpus rb tree, but we cannot reprogram the clock event device on the target cpu. So the timer expires late and subsequently prevents the reprogramming of the target cpu clock event device until the previously programmed event fires or a timer with an earlier expiry time gets enqueued on the target cpu itself. Add the same target check as we have for the switch base case and start the timer on the current cpu if it would become the leftmost timer on the target. [ tglx: Rewrote subject and changelog ] Signed-off-by: Leon Ma Link: http://lkml.kernel.org/r/1398847391-5994-1-git-send-email-xindong.ma@intel.com Cc: stable@vger.kernel.org Signed-off-by: Thomas Gleixner --- kernel/hrtimer.c | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'kernel') diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c index e3724fdac2d..6b715c0af1b 100644 --- a/kernel/hrtimer.c +++ b/kernel/hrtimer.c @@ -234,6 +234,11 @@ again: goto again; } timer->base = new_base; + } else { + if (cpu != this_cpu && hrtimer_check_target(timer, new_base)) { + cpu = this_cpu; + goto again; + } } return new_base; } -- cgit v1.2.3-70-g09d2 From 98a01e779f3c66b0b11cd7e64d531c0e41c95762 Mon Sep 17 00:00:00 2001 From: Jiri Bohac Date: Fri, 18 Apr 2014 17:23:11 +0200 Subject: timer: Prevent overflow in apply_slack On architectures with sizeof(int) < sizeof (long), the computation of mask inside apply_slack() can be undefined if the computed bit is > 32. E.g. with: expires = 0xffffe6f5 and slack = 25, we get: expires_limit = 0x20000000e bit = 33 mask = (1 << 33) - 1 /* undefined */ On x86, mask becomes 1 and and the slack is not applied properly. On s390, mask is -1, expires is set to 0 and the timer fires immediately. Use 1UL << bit to solve that issue. Suggested-by: Deborah Townsend Signed-off-by: Jiri Bohac Cc: stable@vger.kernel.org Link: http://lkml.kernel.org/r/20140418152310.GA13654@midget.suse.cz Signed-off-by: Thomas Gleixner --- kernel/timer.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/timer.c b/kernel/timer.c index 87bd529879c..3bb01a323b2 100644 --- a/kernel/timer.c +++ b/kernel/timer.c @@ -838,7 +838,7 @@ unsigned long apply_slack(struct timer_list *timer, unsigned long expires) bit = find_last_bit(&mask, BITS_PER_LONG); - mask = (1 << bit) - 1; + mask = (1UL << bit) - 1; expires_limit = expires_limit & ~(mask); -- cgit v1.2.3-70-g09d2 From 561a4fe851ccab9dd0d14989ab566f9392d9f8b5 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Red Hat)" Date: Fri, 2 May 2014 13:30:04 -0400 Subject: tracing: Use rcu_dereference_sched() for trace event triggers As trace event triggers are now part of the mainline kernel, I added my trace event trigger tests to my test suite I run on all my kernels. Now these tests get run under different config options, and one of those options is CONFIG_PROVE_RCU, which checks under lockdep that the rcu locking primitives are being used correctly. This triggered the following splat: =============================== [ INFO: suspicious RCU usage. ] 3.15.0-rc2-test+ #11 Not tainted ------------------------------- kernel/trace/trace_events_trigger.c:80 suspicious rcu_dereference_check() usage! other info that might help us debug this: rcu_scheduler_active = 1, debug_locks = 0 4 locks held by swapper/1/0: #0: ((&(&j_cdbs->work)->timer)){..-...}, at: [] call_timer_fn+0x5/0x1be #1: (&(&pool->lock)->rlock){-.-...}, at: [] __queue_work+0x140/0x283 #2: (&p->pi_lock){-.-.-.}, at: [] try_to_wake_up+0x2e/0x1e8 #3: (&rq->lock){-.-.-.}, at: [] try_to_wake_up+0x1a0/0x1e8 stack backtrace: CPU: 1 PID: 0 Comm: swapper/1 Not tainted 3.15.0-rc2-test+ #11 Hardware name: /DG965MQ, BIOS MQ96510J.86A.0372.2006.0605.1717 06/05/2006 0000000000000001 ffff88007e083b98 ffffffff819f53a5 0000000000000006 ffff88007b0942c0 ffff88007e083bc8 ffffffff81081307 ffff88007ad96d20 0000000000000000 ffff88007af2d840 ffff88007b2e701c ffff88007e083c18 Call Trace: [] dump_stack+0x4f/0x7c [] lockdep_rcu_suspicious+0x107/0x110 [] event_triggers_call+0x99/0x108 [] ftrace_event_buffer_commit+0x42/0xa4 [] ftrace_raw_event_sched_wakeup_template+0x71/0x7c [] ttwu_do_wakeup+0x7f/0xff [] ttwu_do_activate.constprop.126+0x5c/0x61 [] try_to_wake_up+0x1ac/0x1e8 [] wake_up_process+0x36/0x3b [] wake_up_worker+0x24/0x26 [] insert_work+0x5c/0x65 [] __queue_work+0x26c/0x283 [] ? __queue_work+0x283/0x283 [] delayed_work_timer_fn+0x1e/0x20 [] call_timer_fn+0xdf/0x1be^M [] ? call_timer_fn+0x5/0x1be [] ? __queue_work+0x283/0x283 [] run_timer_softirq+0x1a4/0x22f^M [] __do_softirq+0x17b/0x31b^M [] irq_exit+0x42/0x97 [] smp_apic_timer_interrupt+0x37/0x44 [] apic_timer_interrupt+0x6f/0x80 [] ? default_idle+0x21/0x32 [] ? default_idle+0x1f/0x32 [] arch_cpu_idle+0xf/0x11 [] cpu_startup_entry+0x1a3/0x213 [] start_secondary+0x212/0x219 The cause is that the triggers are protected by rcu_read_lock_sched() but the data is dereferenced with rcu_dereference() which expects it to be protected with rcu_read_lock(). The proper reference should be rcu_dereference_sched(). Cc: Tom Zanussi Cc: stable@vger.kernel.org # 3.14+ Signed-off-by: Steven Rostedt --- kernel/trace/trace_events_trigger.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/trace/trace_events_trigger.c b/kernel/trace/trace_events_trigger.c index 925f537f07d..4747b476a03 100644 --- a/kernel/trace/trace_events_trigger.c +++ b/kernel/trace/trace_events_trigger.c @@ -77,7 +77,7 @@ event_triggers_call(struct ftrace_event_file *file, void *rec) data->ops->func(data); continue; } - filter = rcu_dereference(data->filter); + filter = rcu_dereference_sched(data->filter); if (filter && !filter_match_preds(filter, rec)) continue; if (data->cmd_ops->post_trigger) { -- cgit v1.2.3-70-g09d2 From 722a9f9299ca720a3f14660e7c0dce7b76a9cb42 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Fri, 2 May 2014 00:44:38 +0200 Subject: asmlinkage: Add explicit __visible to drivers/*, lib/*, kernel/* As requested by Linus add explicit __visible to the asmlinkage users. This marks functions visible to assembler. Tree sweep for rest of tree. Signed-off-by: Andi Kleen Link: http://lkml.kernel.org/r/1398984278-29319-4-git-send-email-andi@firstfloor.org Signed-off-by: H. Peter Anvin --- drivers/pnp/pnpbios/bioscalls.c | 2 +- init/main.c | 2 +- kernel/context_tracking.c | 2 +- kernel/locking/lockdep.c | 2 +- kernel/power/snapshot.c | 2 +- kernel/printk/printk.c | 4 ++-- kernel/sched/core.c | 10 +++++----- kernel/softirq.c | 4 ++-- lib/dump_stack.c | 4 ++-- 9 files changed, 16 insertions(+), 16 deletions(-) (limited to 'kernel') diff --git a/drivers/pnp/pnpbios/bioscalls.c b/drivers/pnp/pnpbios/bioscalls.c index deb7f4bcdb7..438d4c72c7b 100644 --- a/drivers/pnp/pnpbios/bioscalls.c +++ b/drivers/pnp/pnpbios/bioscalls.c @@ -37,7 +37,7 @@ __visible struct { * kernel begins at offset 3GB... */ -asmlinkage void pnp_bios_callfunc(void); +asmlinkage __visible void pnp_bios_callfunc(void); __asm__(".text \n" __ALIGN_STR "\n" diff --git a/init/main.c b/init/main.c index 9c7fd4c9249..48655ceb66f 100644 --- a/init/main.c +++ b/init/main.c @@ -476,7 +476,7 @@ static void __init mm_init(void) vmalloc_init(); } -asmlinkage void __init start_kernel(void) +asmlinkage __visible void __init start_kernel(void) { char * command_line; extern const struct kernel_param __start___param[], __stop___param[]; diff --git a/kernel/context_tracking.c b/kernel/context_tracking.c index 6cb20d2e7ee..019d4500844 100644 --- a/kernel/context_tracking.c +++ b/kernel/context_tracking.c @@ -120,7 +120,7 @@ void context_tracking_user_enter(void) * instead of preempt_schedule() to exit user context if needed before * calling the scheduler. */ -asmlinkage void __sched notrace preempt_schedule_context(void) +asmlinkage __visible void __sched notrace preempt_schedule_context(void) { enum ctx_state prev_ctx; diff --git a/kernel/locking/lockdep.c b/kernel/locking/lockdep.c index b0e9467922e..d24e4339b46 100644 --- a/kernel/locking/lockdep.c +++ b/kernel/locking/lockdep.c @@ -4188,7 +4188,7 @@ void debug_show_held_locks(struct task_struct *task) } EXPORT_SYMBOL_GPL(debug_show_held_locks); -asmlinkage void lockdep_sys_exit(void) +asmlinkage __visible void lockdep_sys_exit(void) { struct task_struct *curr = current; diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c index 18fb7a2fb14..1ea328aafdc 100644 --- a/kernel/power/snapshot.c +++ b/kernel/power/snapshot.c @@ -1586,7 +1586,7 @@ swsusp_alloc(struct memory_bitmap *orig_bm, struct memory_bitmap *copy_bm, return -ENOMEM; } -asmlinkage int swsusp_save(void) +asmlinkage __visible int swsusp_save(void) { unsigned int nr_pages, nr_highmem; diff --git a/kernel/printk/printk.c b/kernel/printk/printk.c index a45b5096229..7228258b85e 100644 --- a/kernel/printk/printk.c +++ b/kernel/printk/printk.c @@ -1674,7 +1674,7 @@ EXPORT_SYMBOL(printk_emit); * * See the vsnprintf() documentation for format string extensions over C99. */ -asmlinkage int printk(const char *fmt, ...) +asmlinkage __visible int printk(const char *fmt, ...) { va_list args; int r; @@ -1737,7 +1737,7 @@ void early_vprintk(const char *fmt, va_list ap) } } -asmlinkage void early_printk(const char *fmt, ...) +asmlinkage __visible void early_printk(const char *fmt, ...) { va_list ap; diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 268a45ea238..d9d8ece46a1 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -2192,7 +2192,7 @@ static inline void post_schedule(struct rq *rq) * schedule_tail - first thing a freshly forked thread must call. * @prev: the thread we just switched away from. */ -asmlinkage void schedule_tail(struct task_struct *prev) +asmlinkage __visible void schedule_tail(struct task_struct *prev) __releases(rq->lock) { struct rq *rq = this_rq(); @@ -2741,7 +2741,7 @@ static inline void sched_submit_work(struct task_struct *tsk) blk_schedule_flush_plug(tsk); } -asmlinkage void __sched schedule(void) +asmlinkage __visible void __sched schedule(void) { struct task_struct *tsk = current; @@ -2751,7 +2751,7 @@ asmlinkage void __sched schedule(void) EXPORT_SYMBOL(schedule); #ifdef CONFIG_CONTEXT_TRACKING -asmlinkage void __sched schedule_user(void) +asmlinkage __visible void __sched schedule_user(void) { /* * If we come here after a random call to set_need_resched(), @@ -2783,7 +2783,7 @@ void __sched schedule_preempt_disabled(void) * off of preempt_enable. Kernel preemptions off return from interrupt * occur there and call schedule directly. */ -asmlinkage void __sched notrace preempt_schedule(void) +asmlinkage __visible void __sched notrace preempt_schedule(void) { /* * If there is a non-zero preempt_count or interrupts are disabled, @@ -2813,7 +2813,7 @@ EXPORT_SYMBOL(preempt_schedule); * Note, that this is called and return with irqs disabled. This will * protect us against recursive calling from irq. */ -asmlinkage void __sched preempt_schedule_irq(void) +asmlinkage __visible void __sched preempt_schedule_irq(void) { enum ctx_state prev_state; diff --git a/kernel/softirq.c b/kernel/softirq.c index 33e4648ae0e..92f24f5e8d5 100644 --- a/kernel/softirq.c +++ b/kernel/softirq.c @@ -223,7 +223,7 @@ static inline bool lockdep_softirq_start(void) { return false; } static inline void lockdep_softirq_end(bool in_hardirq) { } #endif -asmlinkage void __do_softirq(void) +asmlinkage __visible void __do_softirq(void) { unsigned long end = jiffies + MAX_SOFTIRQ_TIME; unsigned long old_flags = current->flags; @@ -299,7 +299,7 @@ restart: tsk_restore_flags(current, old_flags, PF_MEMALLOC); } -asmlinkage void do_softirq(void) +asmlinkage __visible void do_softirq(void) { __u32 pending; unsigned long flags; diff --git a/lib/dump_stack.c b/lib/dump_stack.c index f23b63f0a1c..6745c6230db 100644 --- a/lib/dump_stack.c +++ b/lib/dump_stack.c @@ -23,7 +23,7 @@ static void __dump_stack(void) #ifdef CONFIG_SMP static atomic_t dump_lock = ATOMIC_INIT(-1); -asmlinkage void dump_stack(void) +asmlinkage __visible void dump_stack(void) { int was_locked; int old; @@ -55,7 +55,7 @@ retry: preempt_enable(); } #else -asmlinkage void dump_stack(void) +asmlinkage __visible void dump_stack(void) { __dump_stack(); } -- cgit v1.2.3-70-g09d2 From 8058bd0faad860e75547cc5cb5d4ade016247a79 Mon Sep 17 00:00:00 2001 From: Mathieu Desnoyers Date: Thu, 8 May 2014 07:47:49 -0400 Subject: tracepoint: Fix use of tracepoint funcs after rcu free Commit de7b2973903c "tracepoint: Use struct pointer instead of name hash for reg/unreg tracepoints" introduces a use after free by calling release_probes on the old struct tracepoint array before the newly allocated array is published with rcu_assign_pointer. There is a race window where tracepoints (RCU readers) can perform a "use-after-grace-period-after-free", which shows up as a GPF in stress-tests. Link: http://lkml.kernel.org/r/53698021.5020108@oracle.com Link: http://lkml.kernel.org/p/1399549669-25465-1-git-send-email-mathieu.desnoyers@efficios.com Reported-by: Sasha Levin CC: Oleg Nesterov CC: Dave Jones Fixes: de7b2973903c "tracepoint: Use struct pointer instead of name hash for reg/unreg tracepoints" Signed-off-by: Mathieu Desnoyers Signed-off-by: Steven Rostedt --- kernel/tracepoint.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/tracepoint.c b/kernel/tracepoint.c index ac5b23cf721..6620e5837ce 100644 --- a/kernel/tracepoint.c +++ b/kernel/tracepoint.c @@ -188,7 +188,6 @@ static int tracepoint_add_func(struct tracepoint *tp, WARN_ON_ONCE(1); return PTR_ERR(old); } - release_probes(old); /* * rcu_assign_pointer has a smp_wmb() which makes sure that the new @@ -200,6 +199,7 @@ static int tracepoint_add_func(struct tracepoint *tp, rcu_assign_pointer(tp->funcs, tp_funcs); if (!static_key_enabled(&tp->key)) static_key_slow_inc(&tp->key); + release_probes(old); return 0; } @@ -221,7 +221,6 @@ static int tracepoint_remove_func(struct tracepoint *tp, WARN_ON_ONCE(1); return PTR_ERR(old); } - release_probes(old); if (!tp_funcs) { /* Removed last function */ @@ -232,6 +231,7 @@ static int tracepoint_remove_func(struct tracepoint *tp, static_key_slow_dec(&tp->key); } rcu_assign_pointer(tp->funcs, tp_funcs); + release_probes(old); return 0; } -- cgit v1.2.3-70-g09d2 From 84ea7fe37908254c3bd90910921f6e1045c1747a Mon Sep 17 00:00:00 2001 From: Viresh Kumar Date: Mon, 12 May 2014 13:42:29 +0530 Subject: hrtimer: Set expiry time before switch_hrtimer_base() switch_hrtimer_base() calls hrtimer_check_target() which ensures that we do not migrate a timer to a remote cpu if the timer expires before the current programmed expiry time on that remote cpu. But __hrtimer_start_range_ns() calls switch_hrtimer_base() before the new expiry time is set. So the sanity check in hrtimer_check_target() is operating on stale or even uninitialized data. Update expiry time before calling switch_hrtimer_base(). [ tglx: Rewrote changelog once again ] Signed-off-by: Viresh Kumar Cc: linaro-kernel@lists.linaro.org Cc: linaro-networking@linaro.org Cc: fweisbec@gmail.com Cc: arvind.chauhan@arm.com Link: http://lkml.kernel.org/r/81999e148745fc51bbcd0615823fbab9b2e87e23.1399882253.git.viresh.kumar@linaro.org Cc: stable@vger.kernel.org Signed-off-by: Thomas Gleixner --- kernel/hrtimer.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'kernel') diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c index 6b715c0af1b..e0501fe7140 100644 --- a/kernel/hrtimer.c +++ b/kernel/hrtimer.c @@ -990,11 +990,8 @@ int __hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim, /* Remove an active timer from the queue: */ ret = remove_hrtimer(timer, base); - /* Switch the timer base, if necessary: */ - new_base = switch_hrtimer_base(timer, base, mode & HRTIMER_MODE_PINNED); - if (mode & HRTIMER_MODE_REL) { - tim = ktime_add_safe(tim, new_base->get_time()); + tim = ktime_add_safe(tim, base->get_time()); /* * CONFIG_TIME_LOW_RES is a temporary way for architectures * to signal that they simply return xtime in @@ -1009,6 +1006,9 @@ int __hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim, hrtimer_set_expires_range_ns(timer, tim, delta_ns); + /* Switch the timer base, if necessary: */ + new_base = switch_hrtimer_base(timer, base, mode & HRTIMER_MODE_PINNED); + timer_stats_hrtimer_set_start_info(timer); leftmost = enqueue_hrtimer(timer, new_base); -- cgit v1.2.3-70-g09d2 From 5024ae29cd285ce9e736776414da645d3a91828c Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Wed, 7 May 2014 21:31:17 -0400 Subject: cgroup: introduce task_css_is_root() Determining the css of a task usually requires RCU read lock as that's the only thing which keeps the returned css accessible till its reference is acquired; however, testing whether a task belongs to the root can be performed without dereferencing the returned css by comparing the returned pointer against the root one in init_css_set[] which never changes. Implement task_css_is_root() which can be invoked in any context. This will be used by the scheduled cgroup_freezer change. v2: cgroup no longer supports modular controllers. No need to export init_css_set. Pointed out by Li. Signed-off-by: Tejun Heo Acked-by: Li Zefan --- include/linux/cgroup.h | 15 +++++++++++++++ kernel/cgroup.c | 2 +- 2 files changed, 16 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index c2515851c1a..d60904b9e50 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -473,6 +473,7 @@ struct cftype { }; extern struct cgroup_root cgrp_dfl_root; +extern struct css_set init_css_set; static inline bool cgroup_on_dfl(const struct cgroup *cgrp) { @@ -700,6 +701,20 @@ static inline struct cgroup_subsys_state *task_css(struct task_struct *task, return task_css_check(task, subsys_id, false); } +/** + * task_css_is_root - test whether a task belongs to the root css + * @task: the target task + * @subsys_id: the target subsystem ID + * + * Test whether @task belongs to the root css on the specified subsystem. + * May be invoked in any context. + */ +static inline bool task_css_is_root(struct task_struct *task, int subsys_id) +{ + return task_css_check(task, subsys_id, true) == + init_css_set.subsys[subsys_id]; +} + static inline struct cgroup *task_cgroup(struct task_struct *task, int subsys_id) { diff --git a/kernel/cgroup.c b/kernel/cgroup.c index 11a03d67635..3f1ca934a23 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -348,7 +348,7 @@ struct cgrp_cset_link { * reference-counted, to improve performance when child cgroups * haven't been created. */ -static struct css_set init_css_set = { +struct css_set init_css_set = { .refcount = ATOMIC_INIT(1), .cgrp_links = LIST_HEAD_INIT(init_css_set.cgrp_links), .tasks = LIST_HEAD_INIT(init_css_set.tasks), -- cgit v1.2.3-70-g09d2 From e5ced8ebb10c20a3b349bd798b69ccabd3b25d21 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Wed, 7 May 2014 21:31:17 -0400 Subject: cgroup_freezer: replace freezer->lock with freezer_mutex After 96d365e0b86e ("cgroup: make css_set_lock a rwsem and rename it to css_set_rwsem"), css task iterators requires sleepable context as it may block on css_set_rwsem. I missed that cgroup_freezer was iterating tasks under IRQ-safe spinlock freezer->lock. This leads to errors like the following on freezer state reads and transitions. BUG: sleeping function called from invalid context at /work /os/work/kernel/locking/rwsem.c:20 in_atomic(): 0, irqs_disabled(): 0, pid: 462, name: bash 5 locks held by bash/462: #0: (sb_writers#7){.+.+.+}, at: [] vfs_write+0x1a3/0x1c0 #1: (&of->mutex){+.+.+.}, at: [] kernfs_fop_write+0xbb/0x170 #2: (s_active#70){.+.+.+}, at: [] kernfs_fop_write+0xc3/0x170 #3: (freezer_mutex){+.+...}, at: [] freezer_write+0x61/0x1e0 #4: (rcu_read_lock){......}, at: [] freezer_write+0x53/0x1e0 Preemption disabled at:[] console_unlock+0x1e4/0x460 CPU: 3 PID: 462 Comm: bash Not tainted 3.15.0-rc1-work+ #10 Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011 ffff88000916a6d0 ffff88000e0a3da0 ffffffff81cf8c96 0000000000000000 ffff88000e0a3dc8 ffffffff810cf4f2 ffffffff82388040 ffff880013aaf740 0000000000000002 ffff88000e0a3de8 ffffffff81d05974 0000000000000246 Call Trace: [] dump_stack+0x4e/0x7a [] __might_sleep+0x162/0x260 [] down_read+0x24/0x60 [] css_task_iter_start+0x27/0x70 [] freezer_apply_state+0x5d/0x130 [] freezer_write+0xf6/0x1e0 [] cgroup_file_write+0xd8/0x230 [] kernfs_fop_write+0xe7/0x170 [] vfs_write+0xb6/0x1c0 [] SyS_write+0x4d/0xc0 [] system_call_fastpath+0x16/0x1b freezer->lock used to be used in hot paths but that time is long gone and there's no reason for the lock to be IRQ-safe spinlock or even per-cgroup. In fact, given the fact that a cgroup may contain large number of tasks, it's not a good idea to iterate over them while holding IRQ-safe spinlock. Let's simplify locking by replacing per-cgroup freezer->lock with global freezer_mutex. This also makes the comments explaining the intricacies of policy inheritance and the locking around it as the states are protected by a common mutex. The conversion is mostly straight-forward. The followings are worth mentioning. * freezer_css_online() no longer needs double locking. * freezer_attach() now performs propagation simply while holding freezer_mutex. update_if_frozen() race no longer exists and the comment is removed. * freezer_fork() now tests whether the task is in root cgroup using the new task_css_is_root() without doing rcu_read_lock/unlock(). If not, it grabs freezer_mutex and performs the operation. * freezer_read() and freezer_change_state() grab freezer_mutex across the whole operation and pin the css while iterating so that each descendant processing happens in sleepable context. Fixes: 96d365e0b86e ("cgroup: make css_set_lock a rwsem and rename it to css_set_rwsem") Signed-off-by: Tejun Heo Acked-by: Li Zefan --- kernel/cgroup_freezer.c | 112 ++++++++++++++++++++---------------------------- 1 file changed, 46 insertions(+), 66 deletions(-) (limited to 'kernel') diff --git a/kernel/cgroup_freezer.c b/kernel/cgroup_freezer.c index 2bc4a225644..12ead0b766e 100644 --- a/kernel/cgroup_freezer.c +++ b/kernel/cgroup_freezer.c @@ -21,6 +21,7 @@ #include #include #include +#include /* * A cgroup is freezing if any FREEZING flags are set. FREEZING_SELF is @@ -42,9 +43,10 @@ enum freezer_state_flags { struct freezer { struct cgroup_subsys_state css; unsigned int state; - spinlock_t lock; }; +static DEFINE_MUTEX(freezer_mutex); + static inline struct freezer *css_freezer(struct cgroup_subsys_state *css) { return css ? container_of(css, struct freezer, css) : NULL; @@ -93,7 +95,6 @@ freezer_css_alloc(struct cgroup_subsys_state *parent_css) if (!freezer) return ERR_PTR(-ENOMEM); - spin_lock_init(&freezer->lock); return &freezer->css; } @@ -110,14 +111,7 @@ static int freezer_css_online(struct cgroup_subsys_state *css) struct freezer *freezer = css_freezer(css); struct freezer *parent = parent_freezer(freezer); - /* - * The following double locking and freezing state inheritance - * guarantee that @cgroup can never escape ancestors' freezing - * states. See css_for_each_descendant_pre() for details. - */ - if (parent) - spin_lock_irq(&parent->lock); - spin_lock_nested(&freezer->lock, SINGLE_DEPTH_NESTING); + mutex_lock(&freezer_mutex); freezer->state |= CGROUP_FREEZER_ONLINE; @@ -126,10 +120,7 @@ static int freezer_css_online(struct cgroup_subsys_state *css) atomic_inc(&system_freezing_cnt); } - spin_unlock(&freezer->lock); - if (parent) - spin_unlock_irq(&parent->lock); - + mutex_unlock(&freezer_mutex); return 0; } @@ -144,14 +135,14 @@ static void freezer_css_offline(struct cgroup_subsys_state *css) { struct freezer *freezer = css_freezer(css); - spin_lock_irq(&freezer->lock); + mutex_lock(&freezer_mutex); if (freezer->state & CGROUP_FREEZING) atomic_dec(&system_freezing_cnt); freezer->state = 0; - spin_unlock_irq(&freezer->lock); + mutex_unlock(&freezer_mutex); } static void freezer_css_free(struct cgroup_subsys_state *css) @@ -175,7 +166,7 @@ static void freezer_attach(struct cgroup_subsys_state *new_css, struct task_struct *task; bool clear_frozen = false; - spin_lock_irq(&freezer->lock); + mutex_lock(&freezer_mutex); /* * Make the new tasks conform to the current state of @new_css. @@ -197,21 +188,13 @@ static void freezer_attach(struct cgroup_subsys_state *new_css, } } - spin_unlock_irq(&freezer->lock); - - /* - * Propagate FROZEN clearing upwards. We may race with - * update_if_frozen(), but as long as both work bottom-up, either - * update_if_frozen() sees child's FROZEN cleared or we clear the - * parent's FROZEN later. No parent w/ !FROZEN children can be - * left FROZEN. - */ + /* propagate FROZEN clearing upwards */ while (clear_frozen && (freezer = parent_freezer(freezer))) { - spin_lock_irq(&freezer->lock); freezer->state &= ~CGROUP_FROZEN; clear_frozen = freezer->state & CGROUP_FREEZING; - spin_unlock_irq(&freezer->lock); } + + mutex_unlock(&freezer_mutex); } /** @@ -228,9 +211,6 @@ static void freezer_fork(struct task_struct *task) { struct freezer *freezer; - rcu_read_lock(); - freezer = task_freezer(task); - /* * The root cgroup is non-freezable, so we can skip locking the * freezer. This is safe regardless of race with task migration. @@ -238,24 +218,18 @@ static void freezer_fork(struct task_struct *task) * to do. If we lost and root is the new cgroup, noop is still the * right thing to do. */ - if (!parent_freezer(freezer)) - goto out; + if (task_css_is_root(task, freezer_cgrp_id)) + return; - /* - * Grab @freezer->lock and freeze @task after verifying @task still - * belongs to @freezer and it's freezing. The former is for the - * case where we have raced against task migration and lost and - * @task is already in a different cgroup which may not be frozen. - * This isn't strictly necessary as freeze_task() is allowed to be - * called spuriously but let's do it anyway for, if nothing else, - * documentation. - */ - spin_lock_irq(&freezer->lock); - if (freezer == task_freezer(task) && (freezer->state & CGROUP_FREEZING)) + mutex_lock(&freezer_mutex); + rcu_read_lock(); + + freezer = task_freezer(task); + if (freezer->state & CGROUP_FREEZING) freeze_task(task); - spin_unlock_irq(&freezer->lock); -out: + rcu_read_unlock(); + mutex_unlock(&freezer_mutex); } /** @@ -281,22 +255,22 @@ static void update_if_frozen(struct cgroup_subsys_state *css) struct css_task_iter it; struct task_struct *task; - WARN_ON_ONCE(!rcu_read_lock_held()); - - spin_lock_irq(&freezer->lock); + lockdep_assert_held(&freezer_mutex); if (!(freezer->state & CGROUP_FREEZING) || (freezer->state & CGROUP_FROZEN)) - goto out_unlock; + return; /* are all (live) children frozen? */ + rcu_read_lock(); css_for_each_child(pos, css) { struct freezer *child = css_freezer(pos); if ((child->state & CGROUP_FREEZER_ONLINE) && !(child->state & CGROUP_FROZEN)) - goto out_unlock; + return; } + rcu_read_unlock(); /* are all tasks frozen? */ css_task_iter_start(css, &it); @@ -317,21 +291,29 @@ static void update_if_frozen(struct cgroup_subsys_state *css) freezer->state |= CGROUP_FROZEN; out_iter_end: css_task_iter_end(&it); -out_unlock: - spin_unlock_irq(&freezer->lock); } static int freezer_read(struct seq_file *m, void *v) { struct cgroup_subsys_state *css = seq_css(m), *pos; + mutex_lock(&freezer_mutex); rcu_read_lock(); /* update states bottom-up */ - css_for_each_descendant_post(pos, css) + css_for_each_descendant_post(pos, css) { + if (!css_tryget(pos)) + continue; + rcu_read_unlock(); + update_if_frozen(pos); + rcu_read_lock(); + css_put(pos); + } + rcu_read_unlock(); + mutex_unlock(&freezer_mutex); seq_puts(m, freezer_state_strs(css_freezer(css)->state)); seq_putc(m, '\n'); @@ -373,7 +355,7 @@ static void freezer_apply_state(struct freezer *freezer, bool freeze, unsigned int state) { /* also synchronizes against task migration, see freezer_attach() */ - lockdep_assert_held(&freezer->lock); + lockdep_assert_held(&freezer_mutex); if (!(freezer->state & CGROUP_FREEZER_ONLINE)) return; @@ -414,31 +396,29 @@ static void freezer_change_state(struct freezer *freezer, bool freeze) * descendant will try to inherit its parent's FREEZING state as * CGROUP_FREEZING_PARENT. */ + mutex_lock(&freezer_mutex); rcu_read_lock(); css_for_each_descendant_pre(pos, &freezer->css) { struct freezer *pos_f = css_freezer(pos); struct freezer *parent = parent_freezer(pos_f); - spin_lock_irq(&pos_f->lock); + if (!css_tryget(pos)) + continue; + rcu_read_unlock(); - if (pos_f == freezer) { + if (pos_f == freezer) freezer_apply_state(pos_f, freeze, CGROUP_FREEZING_SELF); - } else { - /* - * Our update to @parent->state is already visible - * which is all we need. No need to lock @parent. - * For more info on synchronization, see - * freezer_post_create(). - */ + else freezer_apply_state(pos_f, parent->state & CGROUP_FREEZING, CGROUP_FREEZING_PARENT); - } - spin_unlock_irq(&pos_f->lock); + rcu_read_lock(); + css_put(pos); } rcu_read_unlock(); + mutex_unlock(&freezer_mutex); } static int freezer_write(struct cgroup_subsys_state *css, struct cftype *cft, -- cgit v1.2.3-70-g09d2 From 36e9d2ebcc15d029b33f42a36146ab5a5bcfcfe7 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 13 May 2014 11:28:30 -0400 Subject: cgroup: fix rcu_read_lock() leak in update_if_frozen() While updating cgroup_freezer locking, 68fafb77d827 ("cgroup_freezer: replace freezer->lock with freezer_mutex") introduced a bug in update_if_frozen() where it returns with rcu_read_lock() held. Fix it by adding rcu_read_unlock() before returning. Signed-off-by: Tejun Heo Reported-by: kbuild test robot --- kernel/cgroup_freezer.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/cgroup_freezer.c b/kernel/cgroup_freezer.c index 12ead0b766e..345628c78b5 100644 --- a/kernel/cgroup_freezer.c +++ b/kernel/cgroup_freezer.c @@ -267,8 +267,10 @@ static void update_if_frozen(struct cgroup_subsys_state *css) struct freezer *child = css_freezer(pos); if ((child->state & CGROUP_FREEZER_ONLINE) && - !(child->state & CGROUP_FROZEN)) + !(child->state & CGROUP_FROZEN)) { + rcu_read_unlock(); return; + } } rcu_read_unlock(); -- cgit v1.2.3-70-g09d2