From 97e371409da76ebe3282c0ba8bc86a84efc3694c Mon Sep 17 00:00:00 2001 From: Dima Zavin Date: Wed, 2 Aug 2017 13:32:18 -0700 Subject: cpuset: fix a deadlock due to incomplete patching of cpusets_enabled() commit 89affbf5d9ebb15c6460596822e8857ea2f9e735 upstream. In codepaths that use the begin/retry interface for reading mems_allowed_seq with irqs disabled, there exists a race condition that stalls the patch process after only modifying a subset of the static_branch call sites. This problem manifested itself as a deadlock in the slub allocator, inside get_any_partial. The loop reads mems_allowed_seq value (via read_mems_allowed_begin), performs the defrag operation, and then verifies the consistency of mem_allowed via the read_mems_allowed_retry and the cookie returned by xxx_begin. The issue here is that both begin and retry first check if cpusets are enabled via cpusets_enabled() static branch. This branch can be rewritted dynamically (via cpuset_inc) if a new cpuset is created. The x86 jump label code fully synchronizes across all CPUs for every entry it rewrites. If it rewrites only one of the callsites (specifically the one in read_mems_allowed_retry) and then waits for the smp_call_function(do_sync_core) to complete while a CPU is inside the begin/retry section with IRQs off and the mems_allowed value is changed, we can hang. This is because begin() will always return 0 (since it wasn't patched yet) while retry() will test the 0 against the actual value of the seq counter. The fix is to use two different static keys: one for begin (pre_enable_key) and one for retry (enable_key). In cpuset_inc(), we first bump the pre_enable key to ensure that cpuset_mems_allowed_begin() always return a valid seqcount if are enabling cpusets. Similarly, when disabling cpusets via cpuset_dec(), we first ensure that callers of cpuset_mems_allowed_retry() will start ignoring the seqcount value before we let cpuset_mems_allowed_begin() return 0. The relevant stack traces of the two stuck threads: CPU: 1 PID: 1415 Comm: mkdir Tainted: G L 4.9.36-00104-g540c51286237 #4 Hardware name: Default string Default string/Hardware, BIOS 4.29.1-20170526215256 05/26/2017 task: ffff8817f9c28000 task.stack: ffffc9000ffa4000 RIP: smp_call_function_many+0x1f9/0x260 Call Trace: smp_call_function+0x3b/0x70 on_each_cpu+0x2f/0x90 text_poke_bp+0x87/0xd0 arch_jump_label_transform+0x93/0x100 __jump_label_update+0x77/0x90 jump_label_update+0xaa/0xc0 static_key_slow_inc+0x9e/0xb0 cpuset_css_online+0x70/0x2e0 online_css+0x2c/0xa0 cgroup_apply_control_enable+0x27f/0x3d0 cgroup_mkdir+0x2b7/0x420 kernfs_iop_mkdir+0x5a/0x80 vfs_mkdir+0xf6/0x1a0 SyS_mkdir+0xb7/0xe0 entry_SYSCALL_64_fastpath+0x18/0xad ... CPU: 2 PID: 1 Comm: init Tainted: G L 4.9.36-00104-g540c51286237 #4 Hardware name: Default string Default string/Hardware, BIOS 4.29.1-20170526215256 05/26/2017 task: ffff8818087c0000 task.stack: ffffc90000030000 RIP: int3+0x39/0x70 Call Trace: <#DB> ? ___slab_alloc+0x28b/0x5a0 ? copy_process.part.40+0xf7/0x1de0 __slab_alloc.isra.80+0x54/0x90 copy_process.part.40+0xf7/0x1de0 copy_process.part.40+0xf7/0x1de0 kmem_cache_alloc_node+0x8a/0x280 copy_process.part.40+0xf7/0x1de0 _do_fork+0xe7/0x6c0 _raw_spin_unlock_irq+0x2d/0x60 trace_hardirqs_on_caller+0x136/0x1d0 entry_SYSCALL_64_fastpath+0x5/0xad do_syscall_64+0x27/0x350 SyS_clone+0x19/0x20 do_syscall_64+0x60/0x350 entry_SYSCALL64_slow_path+0x25/0x25 Link: http://lkml.kernel.org/r/20170731040113.14197-1-dmitriyz@waymo.com Fixes: 46e700abc44c ("mm, page_alloc: remove unnecessary taking of a seqlock when cpusets are disabled") Signed-off-by: Dima Zavin Reported-by: Cliff Spradlin Acked-by: Vlastimil Babka Cc: Peter Zijlstra Cc: Christopher Lameter Cc: Li Zefan Cc: Pekka Enberg Cc: David Rientjes Cc: Joonsoo Kim Cc: Mel Gorman Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds Signed-off-by: Greg Kroah-Hartman --- include/linux/cpuset.h | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/linux/cpuset.h b/include/linux/cpuset.h index 85a868ccb493..8397dc235e84 100644 --- a/include/linux/cpuset.h +++ b/include/linux/cpuset.h @@ -16,6 +16,7 @@ #ifdef CONFIG_CPUSETS +extern struct static_key cpusets_pre_enable_key; extern struct static_key cpusets_enabled_key; static inline bool cpusets_enabled(void) { @@ -30,12 +31,14 @@ static inline int nr_cpusets(void) static inline void cpuset_inc(void) { + static_key_slow_inc(&cpusets_pre_enable_key); static_key_slow_inc(&cpusets_enabled_key); } static inline void cpuset_dec(void) { static_key_slow_dec(&cpusets_enabled_key); + static_key_slow_dec(&cpusets_pre_enable_key); } extern int cpuset_init(void); @@ -104,7 +107,7 @@ extern void cpuset_print_current_mems_allowed(void); */ static inline unsigned int read_mems_allowed_begin(void) { - if (!cpusets_enabled()) + if (!static_key_false(&cpusets_pre_enable_key)) return 0; return read_seqcount_begin(¤t->mems_allowed_seq); @@ -118,7 +121,7 @@ static inline unsigned int read_mems_allowed_begin(void) */ static inline bool read_mems_allowed_retry(unsigned int seq) { - if (!cpusets_enabled()) + if (!static_key_false(&cpusets_enabled_key)) return false; return read_seqcount_retry(¤t->mems_allowed_seq, seq); -- cgit v1.2.3 From b89e781dab249e1c74e6b49e2664ae53a14c0306 Mon Sep 17 00:00:00 2001 From: Nicholas Bellinger Date: Fri, 4 Aug 2017 23:59:31 -0700 Subject: iscsi-target: Fix iscsi_np reset hung task during parallel delete commit 978d13d60c34818a41fc35962602bdfa5c03f214 upstream. This patch fixes a bug associated with iscsit_reset_np_thread() that can occur during parallel configfs rmdir of a single iscsi_np used across multiple iscsi-target instances, that would result in hung task(s) similar to below where configfs rmdir process context was blocked indefinately waiting for iscsi_np->np_restart_comp to finish: [ 6726.112076] INFO: task dcp_proxy_node_:15550 blocked for more than 120 seconds. [ 6726.119440] Tainted: G W O 4.1.26-3321 #2 [ 6726.125045] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. [ 6726.132927] dcp_proxy_node_ D ffff8803f202bc88 0 15550 1 0x00000000 [ 6726.140058] ffff8803f202bc88 ffff88085c64d960 ffff88083b3b1ad0 ffff88087fffeb08 [ 6726.147593] ffff8803f202c000 7fffffffffffffff ffff88083f459c28 ffff88083b3b1ad0 [ 6726.155132] ffff88035373c100 ffff8803f202bca8 ffffffff8168ced2 ffff8803f202bcb8 [ 6726.162667] Call Trace: [ 6726.165150] [] schedule+0x32/0x80 [ 6726.170156] [] schedule_timeout+0x214/0x290 [ 6726.176030] [] ? __send_signal+0x52/0x4a0 [ 6726.181728] [] wait_for_completion+0x96/0x100 [ 6726.187774] [] ? wake_up_state+0x10/0x10 [ 6726.193395] [] iscsit_reset_np_thread+0x62/0xe0 [iscsi_target_mod] [ 6726.201278] [] iscsit_tpg_disable_portal_group+0x96/0x190 [iscsi_target_mod] [ 6726.210033] [] lio_target_tpg_store_enable+0x4f/0xc0 [iscsi_target_mod] [ 6726.218351] [] configfs_write_file+0xaa/0x110 [ 6726.224392] [] vfs_write+0xa4/0x1b0 [ 6726.229576] [] SyS_write+0x41/0xb0 [ 6726.234659] [] system_call_fastpath+0x12/0x71 It would happen because each iscsit_reset_np_thread() sets state to ISCSI_NP_THREAD_RESET, sends SIGINT, and then blocks waiting for completion on iscsi_np->np_restart_comp. However, if iscsi_np was active processing a login request and more than a single iscsit_reset_np_thread() caller to the same iscsi_np was blocked on iscsi_np->np_restart_comp, iscsi_np kthread process context in __iscsi_target_login_thread() would flush pending signals and only perform a single completion of np->np_restart_comp before going back to sleep within transport specific iscsit_transport->iscsi_accept_np code. To address this bug, add a iscsi_np->np_reset_count and update __iscsi_target_login_thread() to keep completing np->np_restart_comp until ->np_reset_count has reached zero. Reported-by: Gary Guo Tested-by: Gary Guo Cc: Mike Christie Cc: Hannes Reinecke Signed-off-by: Nicholas Bellinger Signed-off-by: Greg Kroah-Hartman --- include/target/iscsi/iscsi_target_core.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include') diff --git a/include/target/iscsi/iscsi_target_core.h b/include/target/iscsi/iscsi_target_core.h index fdda45f26f75..22f442ab85f9 100644 --- a/include/target/iscsi/iscsi_target_core.h +++ b/include/target/iscsi/iscsi_target_core.h @@ -784,6 +784,7 @@ struct iscsi_np { int np_sock_type; enum np_thread_state_table np_thread_state; bool enabled; + atomic_t np_reset_count; enum iscsi_timer_flags_table np_login_timer_flags; u32 np_exports; enum np_flags_table np_flags; -- cgit v1.2.3 From b4cf49024cf412a94121e61f8056636c557ead98 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Mon, 21 Aug 2017 17:35:02 +0200 Subject: pids: make task_tgid_nr_ns() safe commit dd1c1f2f2028a7b851f701fc6a8ebe39dcb95e7c upstream. This was reported many times, and this was even mentioned in commit 52ee2dfdd4f5 ("pids: refactor vnr/nr_ns helpers to make them safe") but somehow nobody bothered to fix the obvious problem: task_tgid_nr_ns() is not safe because task->group_leader points to nowhere after the exiting task passes exit_notify(), rcu_read_lock() can not help. We really need to change __unhash_process() to nullify group_leader, parent, and real_parent, but this needs some cleanups. Until then we can turn task_tgid_nr_ns() into another user of __task_pid_nr_ns() and fix the problem. Reported-by: Troy Kensinger Signed-off-by: Oleg Nesterov Signed-off-by: Linus Torvalds Signed-off-by: Greg Kroah-Hartman --- include/linux/pid.h | 4 +++- include/linux/sched.h | 50 +++++++++++++++++++++++++++----------------------- 2 files changed, 30 insertions(+), 24 deletions(-) (limited to 'include') diff --git a/include/linux/pid.h b/include/linux/pid.h index 23705a53abba..97b745ddece5 100644 --- a/include/linux/pid.h +++ b/include/linux/pid.h @@ -8,7 +8,9 @@ enum pid_type PIDTYPE_PID, PIDTYPE_PGID, PIDTYPE_SID, - PIDTYPE_MAX + PIDTYPE_MAX, + /* only valid to __task_pid_nr_ns() */ + __PIDTYPE_TGID }; /* diff --git a/include/linux/sched.h b/include/linux/sched.h index eff7c1fad26f..e887c8d6f395 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1949,31 +1949,8 @@ static inline pid_t task_tgid_nr(struct task_struct *tsk) return tsk->tgid; } -pid_t task_tgid_nr_ns(struct task_struct *tsk, struct pid_namespace *ns); - -static inline pid_t task_tgid_vnr(struct task_struct *tsk) -{ - return pid_vnr(task_tgid(tsk)); -} - static inline int pid_alive(const struct task_struct *p); -static inline pid_t task_ppid_nr_ns(const struct task_struct *tsk, struct pid_namespace *ns) -{ - pid_t pid = 0; - - rcu_read_lock(); - if (pid_alive(tsk)) - pid = task_tgid_nr_ns(rcu_dereference(tsk->real_parent), ns); - rcu_read_unlock(); - - return pid; -} - -static inline pid_t task_ppid_nr(const struct task_struct *tsk) -{ - return task_ppid_nr_ns(tsk, &init_pid_ns); -} static inline pid_t task_pgrp_nr_ns(struct task_struct *tsk, struct pid_namespace *ns) @@ -1998,6 +1975,33 @@ static inline pid_t task_session_vnr(struct task_struct *tsk) return __task_pid_nr_ns(tsk, PIDTYPE_SID, NULL); } +static inline pid_t task_tgid_nr_ns(struct task_struct *tsk, struct pid_namespace *ns) +{ + return __task_pid_nr_ns(tsk, __PIDTYPE_TGID, ns); +} + +static inline pid_t task_tgid_vnr(struct task_struct *tsk) +{ + return __task_pid_nr_ns(tsk, __PIDTYPE_TGID, NULL); +} + +static inline pid_t task_ppid_nr_ns(const struct task_struct *tsk, struct pid_namespace *ns) +{ + pid_t pid = 0; + + rcu_read_lock(); + if (pid_alive(tsk)) + pid = task_tgid_nr_ns(rcu_dereference(tsk->real_parent), ns); + rcu_read_unlock(); + + return pid; +} + +static inline pid_t task_ppid_nr(const struct task_struct *tsk) +{ + return task_ppid_nr_ns(tsk, &init_pid_ns); +} + /* obsolete, do not use */ static inline pid_t task_pgrp_nr(struct task_struct *tsk) { -- cgit v1.2.3 From 1bd54371388c0c1e24e3ffa8afde9e130c5799b9 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 16 Aug 2017 11:09:12 -0700 Subject: ipv4: better IP_MAX_MTU enforcement [ Upstream commit c780a049f9bf442314335372c9abc4548bfe3e44 ] While working on yet another syzkaller report, I found that our IP_MAX_MTU enforcements were not properly done. gcc seems to reload dev->mtu for min(dev->mtu, IP_MAX_MTU), and final result can be bigger than IP_MAX_MTU :/ This is a problem because device mtu can be changed on other cpus or threads. While this patch does not fix the issue I am working on, it is probably worth addressing it. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- include/net/ip.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/net/ip.h b/include/net/ip.h index b450d8653b30..7476bb10ff37 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -314,7 +314,7 @@ static inline unsigned int ip_dst_mtu_maybe_forward(const struct dst_entry *dst, !forwarding) return dst_mtu(dst); - return min(dst->dev->mtu, IP_MAX_MTU); + return min(READ_ONCE(dst->dev->mtu), IP_MAX_MTU); } static inline unsigned int ip_skb_dst_mtu(const struct sk_buff *skb) @@ -327,7 +327,7 @@ static inline unsigned int ip_skb_dst_mtu(const struct sk_buff *skb) return ip_dst_mtu_maybe_forward(skb_dst(skb), forwarding); } - return min(skb_dst(skb)->dev->mtu, IP_MAX_MTU); + return min(READ_ONCE(skb_dst(skb)->dev->mtu), IP_MAX_MTU); } u32 ip_idents_reserve(u32 hash, int segs); -- cgit v1.2.3 From 58079f56b30227c37709e9db7e9b51204e81f4ea Mon Sep 17 00:00:00 2001 From: Konstantin Khlebnikov Date: Sat, 19 Aug 2017 15:37:07 +0300 Subject: net_sched: fix order of queue length updates in qdisc_replace() [ Upstream commit 68a66d149a8c78ec6720f268597302883e48e9fa ] This important to call qdisc_tree_reduce_backlog() after changing queue length. Parent qdisc should deactivate class in ->qlen_notify() called from qdisc_tree_reduce_backlog() but this happens only if qdisc->q.qlen in zero. Missed class deactivations leads to crashes/warnings at picking packets from empty qdisc and corrupting state at reactivating this class in future. Signed-off-by: Konstantin Khlebnikov Fixes: 86a7996cc8a0 ("net_sched: introduce qdisc_replace() helper") Acked-by: Cong Wang Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- include/net/sch_generic.h | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index e5bba897d206..7a5d6a073165 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -717,8 +717,11 @@ static inline struct Qdisc *qdisc_replace(struct Qdisc *sch, struct Qdisc *new, old = *pold; *pold = new; if (old != NULL) { - qdisc_tree_reduce_backlog(old, old->q.qlen, old->qstats.backlog); + unsigned int qlen = old->q.qlen; + unsigned int backlog = old->qstats.backlog; + qdisc_reset(old); + qdisc_tree_reduce_backlog(old, qlen, backlog); } sch_tree_unlock(sch); -- cgit v1.2.3