From dc434e056fe1dada20df7ba07f32739d3a701adf Mon Sep 17 00:00:00 2001 From: Sebastian Andrzej Siewior Date: Tue, 14 Mar 2017 16:06:45 +0100 Subject: cpu/hotplug: Serialize callback invocations proper The setup/remove_state/instance() functions in the hotplug core code are serialized against concurrent CPU hotplug, but unfortunately not serialized against themself. As a consequence a concurrent invocation of these function results in corruption of the callback machinery because two instances try to invoke callbacks on remote cpus at the same time. This results in missing callback invocations and initiator threads waiting forever on the completion. The obvious solution to replace get_cpu_online() with cpu_hotplug_begin() is not possible because at least one callsite calls into these functions from a get_online_cpu() locked region. Extend the protection scope of the cpuhp_state_mutex from solely protecting the state arrays to cover the callback invocation machinery as well. Fixes: 5b7aa87e0482 ("cpu/hotplug: Implement setup/removal interface") Reported-and-tested-by: Bart Van Assche Signed-off-by: Sebastian Andrzej Siewior Cc: hpa@zytor.com Cc: mingo@kernel.org Cc: akpm@linux-foundation.org Cc: torvalds@linux-foundation.org Link: http://lkml.kernel.org/r/20170314150645.g4tdyoszlcbajmna@linutronix.de Signed-off-by: Thomas Gleixner --- kernel/cpu.c | 28 ++++++++++++++-------------- 1 file changed, 14 insertions(+), 14 deletions(-) (limited to 'kernel') diff --git a/kernel/cpu.c b/kernel/cpu.c index f7c063239fa5..37b223e4fc05 100644 --- a/kernel/cpu.c +++ b/kernel/cpu.c @@ -1335,26 +1335,21 @@ static int cpuhp_store_callbacks(enum cpuhp_state state, const char *name, struct cpuhp_step *sp; int ret = 0; - mutex_lock(&cpuhp_state_mutex); - if (state == CPUHP_AP_ONLINE_DYN || state == CPUHP_BP_PREPARE_DYN) { ret = cpuhp_reserve_state(state); if (ret < 0) - goto out; + return ret; state = ret; } sp = cpuhp_get_step(state); - if (name && sp->name) { - ret = -EBUSY; - goto out; - } + if (name && sp->name) + return -EBUSY; + sp->startup.single = startup; sp->teardown.single = teardown; sp->name = name; sp->multi_instance = multi_instance; INIT_HLIST_HEAD(&sp->list); -out: - mutex_unlock(&cpuhp_state_mutex); return ret; } @@ -1428,6 +1423,7 @@ int __cpuhp_state_add_instance(enum cpuhp_state state, struct hlist_node *node, return -EINVAL; get_online_cpus(); + mutex_lock(&cpuhp_state_mutex); if (!invoke || !sp->startup.multi) goto add_node; @@ -1447,16 +1443,14 @@ int __cpuhp_state_add_instance(enum cpuhp_state state, struct hlist_node *node, if (ret) { if (sp->teardown.multi) cpuhp_rollback_install(cpu, state, node); - goto err; + goto unlock; } } add_node: ret = 0; - mutex_lock(&cpuhp_state_mutex); hlist_add_head(node, &sp->list); +unlock: mutex_unlock(&cpuhp_state_mutex); - -err: put_online_cpus(); return ret; } @@ -1491,6 +1485,7 @@ int __cpuhp_setup_state(enum cpuhp_state state, return -EINVAL; get_online_cpus(); + mutex_lock(&cpuhp_state_mutex); ret = cpuhp_store_callbacks(state, name, startup, teardown, multi_instance); @@ -1524,6 +1519,7 @@ int __cpuhp_setup_state(enum cpuhp_state state, } } out: + mutex_unlock(&cpuhp_state_mutex); put_online_cpus(); /* * If the requested state is CPUHP_AP_ONLINE_DYN, return the @@ -1547,6 +1543,8 @@ int __cpuhp_state_remove_instance(enum cpuhp_state state, return -EINVAL; get_online_cpus(); + mutex_lock(&cpuhp_state_mutex); + if (!invoke || !cpuhp_get_teardown_cb(state)) goto remove; /* @@ -1563,7 +1561,6 @@ int __cpuhp_state_remove_instance(enum cpuhp_state state, } remove: - mutex_lock(&cpuhp_state_mutex); hlist_del(node); mutex_unlock(&cpuhp_state_mutex); put_online_cpus(); @@ -1571,6 +1568,7 @@ remove: return 0; } EXPORT_SYMBOL_GPL(__cpuhp_state_remove_instance); + /** * __cpuhp_remove_state - Remove the callbacks for an hotplug machine state * @state: The state to remove @@ -1589,6 +1587,7 @@ void __cpuhp_remove_state(enum cpuhp_state state, bool invoke) get_online_cpus(); + mutex_lock(&cpuhp_state_mutex); if (sp->multi_instance) { WARN(!hlist_empty(&sp->list), "Error: Removing state %d which has instances left.\n", @@ -1613,6 +1612,7 @@ void __cpuhp_remove_state(enum cpuhp_state state, bool invoke) } remove: cpuhp_store_callbacks(state, NULL, NULL, NULL, false); + mutex_unlock(&cpuhp_state_mutex); put_online_cpus(); } EXPORT_SYMBOL(__cpuhp_remove_state); -- cgit v1.2.3 From c236c8e95a3d395b0494e7108f0d41cf36ec107c Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Sat, 4 Mar 2017 10:27:18 +0100 Subject: futex: Fix potential use-after-free in FUTEX_REQUEUE_PI While working on the futex code, I stumbled over this potential use-after-free scenario. Dmitry triggered it later with syzkaller. pi_mutex is a pointer into pi_state, which we drop the reference on in unqueue_me_pi(). So any access to that pointer after that is bad. Since other sites already do rt_mutex_unlock() with hb->lock held, see for example futex_lock_pi(), simply move the unlock before unqueue_me_pi(). Reported-by: Dmitry Vyukov Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Darren Hart Cc: juri.lelli@arm.com Cc: bigeasy@linutronix.de Cc: xlpang@redhat.com Cc: rostedt@goodmis.org Cc: mathieu.desnoyers@efficios.com Cc: jdesfossez@efficios.com Cc: dvhart@infradead.org Cc: bristot@redhat.com Cc: stable@vger.kernel.org Link: http://lkml.kernel.org/r/20170304093558.801744246@infradead.org Signed-off-by: Thomas Gleixner --- kernel/futex.c | 20 +++++++++++--------- 1 file changed, 11 insertions(+), 9 deletions(-) (limited to 'kernel') diff --git a/kernel/futex.c b/kernel/futex.c index 229a744b1781..3a4775fd7468 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -2815,7 +2815,6 @@ static int futex_wait_requeue_pi(u32 __user *uaddr, unsigned int flags, { struct hrtimer_sleeper timeout, *to = NULL; struct rt_mutex_waiter rt_waiter; - struct rt_mutex *pi_mutex = NULL; struct futex_hash_bucket *hb; union futex_key key2 = FUTEX_KEY_INIT; struct futex_q q = futex_q_init; @@ -2907,6 +2906,8 @@ static int futex_wait_requeue_pi(u32 __user *uaddr, unsigned int flags, spin_unlock(q.lock_ptr); } } else { + struct rt_mutex *pi_mutex; + /* * We have been woken up by futex_unlock_pi(), a timeout, or a * signal. futex_unlock_pi() will not destroy the lock_ptr nor @@ -2930,18 +2931,19 @@ static int futex_wait_requeue_pi(u32 __user *uaddr, unsigned int flags, if (res) ret = (res < 0) ? res : 0; + /* + * If fixup_pi_state_owner() faulted and was unable to handle + * the fault, unlock the rt_mutex and return the fault to + * userspace. + */ + if (ret && rt_mutex_owner(pi_mutex) == current) + rt_mutex_unlock(pi_mutex); + /* Unqueue and drop the lock. */ unqueue_me_pi(&q); } - /* - * If fixup_pi_state_owner() faulted and was unable to handle the - * fault, unlock the rt_mutex and return the fault to userspace. - */ - if (ret == -EFAULT) { - if (pi_mutex && rt_mutex_owner(pi_mutex) == current) - rt_mutex_unlock(pi_mutex); - } else if (ret == -EINTR) { + if (ret == -EINTR) { /* * We've already been requeued, but cannot restart by calling * futex_lock_pi() directly. We could restart this syscall, but -- cgit v1.2.3 From 9bbb25afeb182502ca4f2c4f3f88af0681b34cae Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Sat, 4 Mar 2017 10:27:19 +0100 Subject: futex: Add missing error handling to FUTEX_REQUEUE_PI Thomas spotted that fixup_pi_state_owner() can return errors and we fail to unlock the rt_mutex in that case. Reported-by: Thomas Gleixner Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Darren Hart Cc: juri.lelli@arm.com Cc: bigeasy@linutronix.de Cc: xlpang@redhat.com Cc: rostedt@goodmis.org Cc: mathieu.desnoyers@efficios.com Cc: jdesfossez@efficios.com Cc: dvhart@infradead.org Cc: bristot@redhat.com Cc: stable@vger.kernel.org Link: http://lkml.kernel.org/r/20170304093558.867401760@infradead.org Signed-off-by: Thomas Gleixner --- kernel/futex.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'kernel') diff --git a/kernel/futex.c b/kernel/futex.c index 3a4775fd7468..45858ec73941 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -2898,6 +2898,8 @@ static int futex_wait_requeue_pi(u32 __user *uaddr, unsigned int flags, if (q.pi_state && (q.pi_state->owner != current)) { spin_lock(q.lock_ptr); ret = fixup_pi_state_owner(uaddr2, &q, current); + if (ret && rt_mutex_owner(&q.pi_state->pi_mutex) == current) + rt_mutex_unlock(&q.pi_state->pi_mutex); /* * Drop the reference to the pi state which * the requeue_pi() code acquired for us. -- cgit v1.2.3 From 17fcbd590d0c3e35bd9646e2215f86586378bc42 Mon Sep 17 00:00:00 2001 From: Niklas Cassel Date: Sat, 25 Feb 2017 01:17:53 +0100 Subject: locking/rwsem: Fix down_write_killable() for CONFIG_RWSEM_GENERIC_SPINLOCK=y We hang if SIGKILL has been sent, but the task is stuck in down_read() (after do_exit()), even though no task is doing down_write() on the rwsem in question: INFO: task libupnp:21868 blocked for more than 120 seconds. libupnp D 0 21868 1 0x08100008 ... Call Trace: __schedule() schedule() __down_read() do_exit() do_group_exit() __wake_up_parent() This bug has already been fixed for CONFIG_RWSEM_XCHGADD_ALGORITHM=y in the following commit: 04cafed7fc19 ("locking/rwsem: Fix down_write_killable()") ... however, this bug also exists for CONFIG_RWSEM_GENERIC_SPINLOCK=y. Signed-off-by: Niklas Cassel Signed-off-by: Peter Zijlstra (Intel) Cc: Cc: Cc: Andrew Morton Cc: Linus Torvalds Cc: Niklas Cassel Cc: Paul E. McKenney Cc: Peter Zijlstra Cc: Thomas Gleixner Fixes: d47996082f52 ("locking/rwsem: Introduce basis for down_write_killable()") Link: http://lkml.kernel.org/r/1487981873-12649-1-git-send-email-niklass@axis.com Signed-off-by: Ingo Molnar --- kernel/locking/rwsem-spinlock.c | 16 +++++++++++----- 1 file changed, 11 insertions(+), 5 deletions(-) (limited to 'kernel') diff --git a/kernel/locking/rwsem-spinlock.c b/kernel/locking/rwsem-spinlock.c index 7bc24d477805..c65f7989f850 100644 --- a/kernel/locking/rwsem-spinlock.c +++ b/kernel/locking/rwsem-spinlock.c @@ -213,10 +213,9 @@ int __sched __down_write_common(struct rw_semaphore *sem, int state) */ if (sem->count == 0) break; - if (signal_pending_state(state, current)) { - ret = -EINTR; - goto out; - } + if (signal_pending_state(state, current)) + goto out_nolock; + set_current_state(state); raw_spin_unlock_irqrestore(&sem->wait_lock, flags); schedule(); @@ -224,12 +223,19 @@ int __sched __down_write_common(struct rw_semaphore *sem, int state) } /* got the lock */ sem->count = -1; -out: list_del(&waiter.list); raw_spin_unlock_irqrestore(&sem->wait_lock, flags); return ret; + +out_nolock: + list_del(&waiter.list); + if (!list_empty(&sem->wait_list)) + __rwsem_do_wake(sem, 1); + raw_spin_unlock_irqrestore(&sem->wait_lock, flags); + + return -EINTR; } void __sched __down_write(struct rw_semaphore *sem) -- cgit v1.2.3 From e552a8389aa409e257b7dcba74f67f128f979ccc Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Thu, 16 Mar 2017 13:47:48 +0100 Subject: perf/core: Fix use-after-free in perf_release() Dmitry reported syzcaller tripped a use-after-free in perf_release(). After much puzzlement Oleg spotted the below scenario: Task1 Task2 fork() perf_event_init_task() /* ... */ goto bad_fork_$foo; /* ... */ perf_event_free_task() mutex_lock(ctx->lock) perf_free_event(B) perf_event_release_kernel(A) mutex_lock(A->child_mutex) list_for_each_entry(child, ...) { /* child == B */ ctx = B->ctx; get_ctx(ctx); mutex_unlock(A->child_mutex); mutex_lock(A->child_mutex) list_del_init(B->child_list) mutex_unlock(A->child_mutex) /* ... */ mutex_unlock(ctx->lock); put_ctx() /* >0 */ free_task(); mutex_lock(ctx->lock); mutex_lock(A->child_mutex); /* ... */ mutex_unlock(A->child_mutex); mutex_unlock(ctx->lock) put_ctx() /* 0 */ ctx->task && !TOMBSTONE put_task_struct() /* UAF */ This patch closes the hole by making perf_event_free_task() destroy the task <-> ctx relation such that perf_event_release_kernel() will no longer observe the now dead task. Spotted-by: Oleg Nesterov Reported-by: Dmitry Vyukov Signed-off-by: Peter Zijlstra (Intel) Cc: Alexander Shishkin Cc: Arnaldo Carvalho de Melo Cc: Arnaldo Carvalho de Melo Cc: Jiri Olsa Cc: Linus Torvalds Cc: Mathieu Desnoyers Cc: Peter Zijlstra Cc: Stephane Eranian Cc: Thomas Gleixner Cc: Vince Weaver Cc: fweisbec@gmail.com Cc: oleg@redhat.com Cc: stable@vger.kernel.org Fixes: c6e5b73242d2 ("perf: Synchronously clean up child events") Link: http://lkml.kernel.org/r/20170314155949.GE32474@worktop Link: http://lkml.kernel.org/r/20170316125823.140295131@infradead.org Signed-off-by: Ingo Molnar --- kernel/events/core.c | 11 +++++++++++ 1 file changed, 11 insertions(+) (limited to 'kernel') diff --git a/kernel/events/core.c b/kernel/events/core.c index 1031bdf9f012..4742909c56e6 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -10415,6 +10415,17 @@ void perf_event_free_task(struct task_struct *task) continue; mutex_lock(&ctx->mutex); + raw_spin_lock_irq(&ctx->lock); + /* + * Destroy the task <-> ctx relation and mark the context dead. + * + * This is important because even though the task hasn't been + * exposed yet the context has been (through child_list). + */ + RCU_INIT_POINTER(task->perf_event_ctxp[ctxn], NULL); + WRITE_ONCE(ctx->task, TASK_TOMBSTONE); + put_task_struct(task); /* cannot be last */ + raw_spin_unlock_irq(&ctx->lock); again: list_for_each_entry_safe(event, tmp, &ctx->pinned_groups, group_entry) -- cgit v1.2.3 From e7cc4865f0f31698ef2f7aac01a50e78968985b7 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Thu, 16 Mar 2017 13:47:49 +0100 Subject: perf/core: Fix event inheritance on fork() While hunting for clues to a use-after-free, Oleg spotted that perf_event_init_context() can loose an error value with the result that fork() can succeed even though we did not fully inherit the perf event context. Spotted-by: Oleg Nesterov Signed-off-by: Peter Zijlstra (Intel) Cc: Alexander Shishkin Cc: Arnaldo Carvalho de Melo Cc: Arnaldo Carvalho de Melo Cc: Dmitry Vyukov Cc: Frederic Weisbecker Cc: Jiri Olsa Cc: Linus Torvalds Cc: Mathieu Desnoyers Cc: Peter Zijlstra Cc: Stephane Eranian Cc: Thomas Gleixner Cc: Vince Weaver Cc: oleg@redhat.com Cc: stable@vger.kernel.org Fixes: 889ff0150661 ("perf/core: Split context's event group list into pinned and non-pinned lists") Link: http://lkml.kernel.org/r/20170316125823.190342547@infradead.org Signed-off-by: Ingo Molnar --- kernel/events/core.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/events/core.c b/kernel/events/core.c index 4742909c56e6..fc7c9a85944d 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -10679,7 +10679,7 @@ static int perf_event_init_context(struct task_struct *child, int ctxn) ret = inherit_task_group(event, parent, parent_ctx, child, ctxn, &inherited_all); if (ret) - break; + goto out_unlock; } /* @@ -10695,7 +10695,7 @@ static int perf_event_init_context(struct task_struct *child, int ctxn) ret = inherit_task_group(event, parent, parent_ctx, child, ctxn, &inherited_all); if (ret) - break; + goto out_unlock; } raw_spin_lock_irqsave(&parent_ctx->lock, flags); @@ -10723,6 +10723,7 @@ static int perf_event_init_context(struct task_struct *child, int ctxn) } raw_spin_unlock_irqrestore(&parent_ctx->lock, flags); +out_unlock: mutex_unlock(&parent_ctx->mutex); perf_unpin_context(parent_ctx); -- cgit v1.2.3 From 15121c789e001168decac6483d192bdb7ea29e74 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Thu, 16 Mar 2017 13:47:50 +0100 Subject: perf/core: Simplify perf_event_free_task() We have ctx->event_list that contains all events; no need to repeatedly iterate the group lists to find them all. Signed-off-by: Peter Zijlstra (Intel) Cc: Alexander Shishkin Cc: Arnaldo Carvalho de Melo Cc: Arnaldo Carvalho de Melo Cc: Dmitry Vyukov Cc: Jiri Olsa Cc: Linus Torvalds Cc: Mathieu Desnoyers Cc: Oleg Nesterov Cc: Peter Zijlstra Cc: Stephane Eranian Cc: Thomas Gleixner Cc: Vince Weaver Cc: fweisbec@gmail.com Link: http://lkml.kernel.org/r/20170316125823.239678244@infradead.org Signed-off-by: Ingo Molnar --- kernel/events/core.c | 12 +----------- 1 file changed, 1 insertion(+), 11 deletions(-) (limited to 'kernel') diff --git a/kernel/events/core.c b/kernel/events/core.c index fc7c9a85944d..5f21e5e09ba4 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -10426,21 +10426,11 @@ void perf_event_free_task(struct task_struct *task) WRITE_ONCE(ctx->task, TASK_TOMBSTONE); put_task_struct(task); /* cannot be last */ raw_spin_unlock_irq(&ctx->lock); -again: - list_for_each_entry_safe(event, tmp, &ctx->pinned_groups, - group_entry) - perf_free_event(event, ctx); - list_for_each_entry_safe(event, tmp, &ctx->flexible_groups, - group_entry) + list_for_each_entry_safe(event, tmp, &ctx->event_list, event_entry) perf_free_event(event, ctx); - if (!list_empty(&ctx->pinned_groups) || - !list_empty(&ctx->flexible_groups)) - goto again; - mutex_unlock(&ctx->mutex); - put_ctx(ctx); } } -- cgit v1.2.3 From d8a8cfc76919b6c830305266b23ba671623f37ff Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Thu, 16 Mar 2017 13:47:51 +0100 Subject: perf/core: Better explain the inherit magic While going through the event inheritance code Oleg got confused. Add some comments to better explain the silent dissapearance of orphaned events. So what happens is that at perf_event_release_kernel() time; when an event looses its connection to userspace (and ceases to exist from the user's perspective) we can still have an arbitrary amount of inherited copies of the event. We want to synchronously find and remove all these child events. Since that requires a bit of lock juggling, there is the possibility that concurrent clone()s will create new child events. Therefore we first mark the parent event as DEAD, which marks all the extant child events as orphaned. We then avoid copying orphaned events; in order to avoid getting more of them. Signed-off-by: Peter Zijlstra (Intel) Cc: Alexander Shishkin Cc: Arnaldo Carvalho de Melo Cc: Arnaldo Carvalho de Melo Cc: Dmitry Vyukov Cc: Jiri Olsa Cc: Linus Torvalds Cc: Mathieu Desnoyers Cc: Oleg Nesterov Cc: Peter Zijlstra Cc: Stephane Eranian Cc: Thomas Gleixner Cc: Vince Weaver Cc: fweisbec@gmail.com Link: http://lkml.kernel.org/r/20170316125823.289567442@infradead.org Signed-off-by: Ingo Molnar --- kernel/events/core.c | 36 +++++++++++++++++++++++++++++++++--- 1 file changed, 33 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/kernel/events/core.c b/kernel/events/core.c index 5f21e5e09ba4..7298e149b732 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -4254,7 +4254,7 @@ int perf_event_release_kernel(struct perf_event *event) raw_spin_lock_irq(&ctx->lock); /* - * Mark this even as STATE_DEAD, there is no external reference to it + * Mark this event as STATE_DEAD, there is no external reference to it * anymore. * * Anybody acquiring event->child_mutex after the below loop _must_ @@ -10468,7 +10468,12 @@ const struct perf_event_attr *perf_event_attrs(struct perf_event *event) } /* - * inherit a event from parent task to child task: + * Inherit a event from parent task to child task. + * + * Returns: + * - valid pointer on success + * - NULL for orphaned events + * - IS_ERR() on error */ static struct perf_event * inherit_event(struct perf_event *parent_event, @@ -10562,6 +10567,16 @@ inherit_event(struct perf_event *parent_event, return child_event; } +/* + * Inherits an event group. + * + * This will quietly suppress orphaned events; !inherit_event() is not an error. + * This matches with perf_event_release_kernel() removing all child events. + * + * Returns: + * - 0 on success + * - <0 on error + */ static int inherit_group(struct perf_event *parent_event, struct task_struct *parent, struct perf_event_context *parent_ctx, @@ -10576,6 +10591,11 @@ static int inherit_group(struct perf_event *parent_event, child, NULL, child_ctx); if (IS_ERR(leader)) return PTR_ERR(leader); + /* + * @leader can be NULL here because of is_orphaned_event(). In this + * case inherit_event() will create individual events, similar to what + * perf_group_detach() would do anyway. + */ list_for_each_entry(sub, &parent_event->sibling_list, group_entry) { child_ctr = inherit_event(sub, parent, parent_ctx, child, leader, child_ctx); @@ -10585,6 +10605,17 @@ static int inherit_group(struct perf_event *parent_event, return 0; } +/* + * Creates the child task context and tries to inherit the event-group. + * + * Clears @inherited_all on !attr.inherited or error. Note that we'll leave + * inherited_all set when we 'fail' to inherit an orphaned event; this is + * consistent with perf_event_release_kernel() removing all child events. + * + * Returns: + * - 0 on success + * - <0 on error + */ static int inherit_task_group(struct perf_event *event, struct task_struct *parent, struct perf_event_context *parent_ctx, @@ -10607,7 +10638,6 @@ inherit_task_group(struct perf_event *event, struct task_struct *parent, * First allocate and initialize a context for the * child. */ - child_ctx = alloc_perf_context(parent_ctx->pmu, child); if (!child_ctx) return -ENOMEM; -- cgit v1.2.3 From 55adc1d05dca9e949cdf46c747cb1e91c0e9143d Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Thu, 16 Mar 2017 16:40:30 -0700 Subject: mm: add private lock to serialize memory hotplug operations Commit bfc8c90139eb ("mem-hotplug: implement get/put_online_mems") introduced new functions get/put_online_mems() and mem_hotplug_begin/end() in order to allow similar semantics for memory hotplug like for cpu hotplug. The corresponding functions for cpu hotplug are get/put_online_cpus() and cpu_hotplug_begin/done() for cpu hotplug. The commit however missed to introduce functions that would serialize memory hotplug operations like they are done for cpu hotplug with cpu_maps_update_begin/done(). This basically leaves mem_hotplug.active_writer unprotected and allows concurrent writers to modify it, which may lead to problems as outlined by commit f931ab479dd2 ("mm: fix devm_memremap_pages crash, use mem_hotplug_{begin, done}"). That commit was extended again with commit b5d24fda9c3d ("mm, devm_memremap_pages: hold device_hotplug lock over mem_hotplug_{begin, done}") which serializes memory hotplug operations for some call sites by using the device_hotplug lock. In addition with commit 3fc21924100b ("mm: validate device_hotplug is held for memory hotplug") a sanity check was added to mem_hotplug_begin() to verify that the device_hotplug lock is held. This in turn triggers the following warning on s390: WARNING: CPU: 6 PID: 1 at drivers/base/core.c:643 assert_held_device_hotplug+0x4a/0x58 Call Trace: assert_held_device_hotplug+0x40/0x58) mem_hotplug_begin+0x34/0xc8 add_memory_resource+0x7e/0x1f8 add_memory+0xda/0x130 add_memory_merged+0x15c/0x178 sclp_detect_standby_memory+0x2ae/0x2f8 do_one_initcall+0xa2/0x150 kernel_init_freeable+0x228/0x2d8 kernel_init+0x2a/0x140 kernel_thread_starter+0x6/0xc One possible fix would be to add more lock_device_hotplug() and unlock_device_hotplug() calls around each call site of mem_hotplug_begin/end(). But that would give the device_hotplug lock additional semantics it better should not have (serialize memory hotplug operations). Instead add a new memory_add_remove_lock which has the similar semantics like cpu_add_remove_lock for cpu hotplug. To keep things hopefully a bit easier the lock will be locked and unlocked within the mem_hotplug_begin/end() functions. Link: http://lkml.kernel.org/r/20170314125226.16779-2-heiko.carstens@de.ibm.com Signed-off-by: Heiko Carstens Reported-by: Sebastian Ott Acked-by: Dan Williams Acked-by: Rafael J. Wysocki Cc: Michal Hocko Cc: Vladimir Davydov Cc: Ben Hutchings Cc: Gerald Schaefer Cc: Martin Schwidefsky Cc: Greg Kroah-Hartman Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/memremap.c | 4 ---- mm/memory_hotplug.c | 6 +++++- 2 files changed, 5 insertions(+), 5 deletions(-) (limited to 'kernel') diff --git a/kernel/memremap.c b/kernel/memremap.c index 06123234f118..07e85e5229da 100644 --- a/kernel/memremap.c +++ b/kernel/memremap.c @@ -247,11 +247,9 @@ static void devm_memremap_pages_release(struct device *dev, void *data) align_start = res->start & ~(SECTION_SIZE - 1); align_size = ALIGN(resource_size(res), SECTION_SIZE); - lock_device_hotplug(); mem_hotplug_begin(); arch_remove_memory(align_start, align_size); mem_hotplug_done(); - unlock_device_hotplug(); untrack_pfn(NULL, PHYS_PFN(align_start), align_size); pgmap_radix_release(res); @@ -364,11 +362,9 @@ void *devm_memremap_pages(struct device *dev, struct resource *res, if (error) goto err_pfn_remap; - lock_device_hotplug(); mem_hotplug_begin(); error = arch_add_memory(nid, align_start, align_size, true); mem_hotplug_done(); - unlock_device_hotplug(); if (error) goto err_add_memory; diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c index 295479b792ec..6fa7208bcd56 100644 --- a/mm/memory_hotplug.c +++ b/mm/memory_hotplug.c @@ -125,9 +125,12 @@ void put_online_mems(void) } +/* Serializes write accesses to mem_hotplug.active_writer. */ +static DEFINE_MUTEX(memory_add_remove_lock); + void mem_hotplug_begin(void) { - assert_held_device_hotplug(); + mutex_lock(&memory_add_remove_lock); mem_hotplug.active_writer = current; @@ -147,6 +150,7 @@ void mem_hotplug_done(void) mem_hotplug.active_writer = NULL; mutex_unlock(&mem_hotplug.lock); memhp_lock_release(); + mutex_unlock(&memory_add_remove_lock); } /* add this memory to iomem resource */ -- cgit v1.2.3 From 4296f23ed49a15d36949458adcc66ff993dee2a8 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sun, 19 Mar 2017 14:30:02 +0100 Subject: cpufreq: schedutil: Fix per-CPU structure initialization in sugov_start() sugov_start() only initializes struct sugov_cpu per-CPU structures for shared policies, but it should do that for single-CPU policies too. That in particular makes the IO-wait boost mechanism work in the cases when cpufreq policies correspond to individual CPUs. Fixes: 21ca6d2c52f8 (cpufreq: schedutil: Add iowait boosting) Signed-off-by: Rafael J. Wysocki Acked-by: Viresh Kumar Cc: 4.9+ # 4.9+ --- kernel/sched/cpufreq_schedutil.c | 20 +++++++------------- 1 file changed, 7 insertions(+), 13 deletions(-) (limited to 'kernel') diff --git a/kernel/sched/cpufreq_schedutil.c b/kernel/sched/cpufreq_schedutil.c index cd7cd489f739..54c577578da6 100644 --- a/kernel/sched/cpufreq_schedutil.c +++ b/kernel/sched/cpufreq_schedutil.c @@ -584,20 +584,14 @@ static int sugov_start(struct cpufreq_policy *policy) for_each_cpu(cpu, policy->cpus) { struct sugov_cpu *sg_cpu = &per_cpu(sugov_cpu, cpu); + memset(sg_cpu, 0, sizeof(*sg_cpu)); sg_cpu->sg_policy = sg_policy; - if (policy_is_shared(policy)) { - sg_cpu->util = 0; - sg_cpu->max = 0; - sg_cpu->flags = SCHED_CPUFREQ_RT; - sg_cpu->last_update = 0; - sg_cpu->iowait_boost = 0; - sg_cpu->iowait_boost_max = policy->cpuinfo.max_freq; - cpufreq_add_update_util_hook(cpu, &sg_cpu->update_util, - sugov_update_shared); - } else { - cpufreq_add_update_util_hook(cpu, &sg_cpu->update_util, - sugov_update_single); - } + sg_cpu->flags = SCHED_CPUFREQ_RT; + sg_cpu->iowait_boost_max = policy->cpuinfo.max_freq; + cpufreq_add_update_util_hook(cpu, &sg_cpu->update_util, + policy_is_shared(policy) ? + sugov_update_shared : + sugov_update_single); } return 0; } -- cgit v1.2.3 From 5b52330bbfe63b3305765354d6046c9f7f89c011 Mon Sep 17 00:00:00 2001 From: Paul Moore Date: Tue, 21 Mar 2017 11:26:35 -0400 Subject: audit: fix auditd/kernel connection state tracking What started as a rather straightforward race condition reported by Dmitry using the syzkaller fuzzer ended up revealing some major problems with how the audit subsystem managed its netlink sockets and its connection with the userspace audit daemon. Fixing this properly had quite the cascading effect and what we are left with is this rather large and complicated patch. My initial goal was to try and decompose this patch into multiple smaller patches, but the way these changes are intertwined makes it difficult to split these changes into meaningful pieces that don't break or somehow make things worse for the intermediate states. The patch makes a number of changes, but the most significant are highlighted below: * The auditd tracking variables, e.g. audit_sock, are now gone and replaced by a RCU/spin_lock protected variable auditd_conn which is a structure containing all of the auditd tracking information. * We no longer track the auditd sock directly, instead we track it via the network namespace in which it resides and we use the audit socket associated with that namespace. In spirit, this is what the code was trying to do prior to this patch (at least I think that is what the original authors intended), but it was done rather poorly and added a layer of obfuscation that only masked the underlying problems. * Big backlog queue cleanup, again. In v4.10 we made some pretty big changes to how the audit backlog queues work, here we haven't changed the queue design so much as cleaned up the implementation. Brought about by the locking changes, we've simplified kauditd_thread() quite a bit by consolidating the queue handling into a new helper function, kauditd_send_queue(), which allows us to eliminate a lot of very similar code and makes the looping logic in kauditd_thread() clearer. * All netlink messages sent to auditd are now sent via auditd_send_unicast_skb(). Other than just making sense, this makes the lock handling easier. * Change the audit_log_start() sleep behavior so that we never sleep on auditd events (unchanged) or if the caller is holding the audit_cmd_mutex (changed). Previously we didn't sleep if the caller was auditd or if the message type fell between a certain range; the type check was a poor effort of doing what the cmd_mutex check now does. Richard Guy Briggs originally proposed not sleeping the cmd_mutex owner several years ago but his patch wasn't acceptable at the time. At least the idea lives on here. * A problem with the lost record counter has been resolved. Steve Grubb and I both happened to notice this problem and according to some quick testing by Steve, this problem goes back quite some time. It's largely a harmless problem, although it may have left some careful sysadmins quite puzzled. Cc: # 4.10.x- Reported-by: Dmitry Vyukov Signed-off-by: Paul Moore --- kernel/audit.c | 639 ++++++++++++++++++++++++++++++++++--------------------- kernel/audit.h | 9 +- kernel/auditsc.c | 6 +- 3 files changed, 399 insertions(+), 255 deletions(-) (limited to 'kernel') diff --git a/kernel/audit.c b/kernel/audit.c index e794544f5e63..2f4964cfde0b 100644 --- a/kernel/audit.c +++ b/kernel/audit.c @@ -54,6 +54,10 @@ #include #include #include +#include +#include +#include +#include #include @@ -90,13 +94,34 @@ static u32 audit_default; /* If auditing cannot proceed, audit_failure selects what happens. */ static u32 audit_failure = AUDIT_FAIL_PRINTK; -/* - * If audit records are to be written to the netlink socket, audit_pid - * contains the pid of the auditd process and audit_nlk_portid contains - * the portid to use to send netlink messages to that process. +/* private audit network namespace index */ +static unsigned int audit_net_id; + +/** + * struct audit_net - audit private network namespace data + * @sk: communication socket + */ +struct audit_net { + struct sock *sk; +}; + +/** + * struct auditd_connection - kernel/auditd connection state + * @pid: auditd PID + * @portid: netlink portid + * @net: the associated network namespace + * @lock: spinlock to protect write access + * + * Description: + * This struct is RCU protected; you must either hold the RCU lock for reading + * or the included spinlock for writing. */ -int audit_pid; -static __u32 audit_nlk_portid; +static struct auditd_connection { + int pid; + u32 portid; + struct net *net; + spinlock_t lock; +} auditd_conn; /* If audit_rate_limit is non-zero, limit the rate of sending audit records * to that number per second. This prevents DoS attacks, but results in @@ -123,10 +148,6 @@ u32 audit_sig_sid = 0; */ static atomic_t audit_lost = ATOMIC_INIT(0); -/* The netlink socket. */ -static struct sock *audit_sock; -static unsigned int audit_net_id; - /* Hash for inode-based rules */ struct list_head audit_inode_hash[AUDIT_INODE_BUCKETS]; @@ -139,6 +160,7 @@ static LIST_HEAD(audit_freelist); /* queue msgs to send via kauditd_task */ static struct sk_buff_head audit_queue; +static void kauditd_hold_skb(struct sk_buff *skb); /* queue msgs due to temporary unicast send problems */ static struct sk_buff_head audit_retry_queue; /* queue msgs waiting for new auditd connection */ @@ -192,6 +214,43 @@ struct audit_reply { struct sk_buff *skb; }; +/** + * auditd_test_task - Check to see if a given task is an audit daemon + * @task: the task to check + * + * Description: + * Return 1 if the task is a registered audit daemon, 0 otherwise. + */ +int auditd_test_task(const struct task_struct *task) +{ + int rc; + + rcu_read_lock(); + rc = (auditd_conn.pid && task->tgid == auditd_conn.pid ? 1 : 0); + rcu_read_unlock(); + + return rc; +} + +/** + * audit_get_sk - Return the audit socket for the given network namespace + * @net: the destination network namespace + * + * Description: + * Returns the sock pointer if valid, NULL otherwise. The caller must ensure + * that a reference is held for the network namespace while the sock is in use. + */ +static struct sock *audit_get_sk(const struct net *net) +{ + struct audit_net *aunet; + + if (!net) + return NULL; + + aunet = net_generic(net, audit_net_id); + return aunet->sk; +} + static void audit_set_portid(struct audit_buffer *ab, __u32 portid) { if (ab) { @@ -210,9 +269,7 @@ void audit_panic(const char *message) pr_err("%s\n", message); break; case AUDIT_FAIL_PANIC: - /* test audit_pid since printk is always losey, why bother? */ - if (audit_pid) - panic("audit: %s\n", message); + panic("audit: %s\n", message); break; } } @@ -370,21 +427,87 @@ static int audit_set_failure(u32 state) return audit_do_config_change("audit_failure", &audit_failure, state); } -/* - * For one reason or another this nlh isn't getting delivered to the userspace - * audit daemon, just send it to printk. +/** + * auditd_set - Set/Reset the auditd connection state + * @pid: auditd PID + * @portid: auditd netlink portid + * @net: auditd network namespace pointer + * + * Description: + * This function will obtain and drop network namespace references as + * necessary. + */ +static void auditd_set(int pid, u32 portid, struct net *net) +{ + unsigned long flags; + + spin_lock_irqsave(&auditd_conn.lock, flags); + auditd_conn.pid = pid; + auditd_conn.portid = portid; + if (auditd_conn.net) + put_net(auditd_conn.net); + if (net) + auditd_conn.net = get_net(net); + else + auditd_conn.net = NULL; + spin_unlock_irqrestore(&auditd_conn.lock, flags); +} + +/** + * auditd_reset - Disconnect the auditd connection + * + * Description: + * Break the auditd/kauditd connection and move all the queued records into the + * hold queue in case auditd reconnects. + */ +static void auditd_reset(void) +{ + struct sk_buff *skb; + + /* if it isn't already broken, break the connection */ + rcu_read_lock(); + if (auditd_conn.pid) + auditd_set(0, 0, NULL); + rcu_read_unlock(); + + /* flush all of the main and retry queues to the hold queue */ + while ((skb = skb_dequeue(&audit_retry_queue))) + kauditd_hold_skb(skb); + while ((skb = skb_dequeue(&audit_queue))) + kauditd_hold_skb(skb); +} + +/** + * kauditd_print_skb - Print the audit record to the ring buffer + * @skb: audit record + * + * Whatever the reason, this packet may not make it to the auditd connection + * so write it via printk so the information isn't completely lost. */ static void kauditd_printk_skb(struct sk_buff *skb) { struct nlmsghdr *nlh = nlmsg_hdr(skb); char *data = nlmsg_data(nlh); - if (nlh->nlmsg_type != AUDIT_EOE) { - if (printk_ratelimit()) - pr_notice("type=%d %s\n", nlh->nlmsg_type, data); - else - audit_log_lost("printk limit exceeded"); - } + if (nlh->nlmsg_type != AUDIT_EOE && printk_ratelimit()) + pr_notice("type=%d %s\n", nlh->nlmsg_type, data); +} + +/** + * kauditd_rehold_skb - Handle a audit record send failure in the hold queue + * @skb: audit record + * + * Description: + * This should only be used by the kauditd_thread when it fails to flush the + * hold queue. + */ +static void kauditd_rehold_skb(struct sk_buff *skb) +{ + /* put the record back in the queue at the same place */ + skb_queue_head(&audit_hold_queue, skb); + + /* fail the auditd connection */ + auditd_reset(); } /** @@ -421,6 +544,9 @@ static void kauditd_hold_skb(struct sk_buff *skb) /* we have no other options - drop the message */ audit_log_lost("kauditd hold queue overflow"); kfree_skb(skb); + + /* fail the auditd connection */ + auditd_reset(); } /** @@ -441,51 +567,122 @@ static void kauditd_retry_skb(struct sk_buff *skb) } /** - * auditd_reset - Disconnect the auditd connection + * auditd_send_unicast_skb - Send a record via unicast to auditd + * @skb: audit record * * Description: - * Break the auditd/kauditd connection and move all the records in the retry - * queue into the hold queue in case auditd reconnects. The audit_cmd_mutex - * must be held when calling this function. + * Send a skb to the audit daemon, returns positive/zero values on success and + * negative values on failure; in all cases the skb will be consumed by this + * function. If the send results in -ECONNREFUSED the connection with auditd + * will be reset. This function may sleep so callers should not hold any locks + * where this would cause a problem. */ -static void auditd_reset(void) +static int auditd_send_unicast_skb(struct sk_buff *skb) { - struct sk_buff *skb; - - /* break the connection */ - if (audit_sock) { - sock_put(audit_sock); - audit_sock = NULL; + int rc; + u32 portid; + struct net *net; + struct sock *sk; + + /* NOTE: we can't call netlink_unicast while in the RCU section so + * take a reference to the network namespace and grab local + * copies of the namespace, the sock, and the portid; the + * namespace and sock aren't going to go away while we hold a + * reference and if the portid does become invalid after the RCU + * section netlink_unicast() should safely return an error */ + + rcu_read_lock(); + if (!auditd_conn.pid) { + rcu_read_unlock(); + rc = -ECONNREFUSED; + goto err; } - audit_pid = 0; - audit_nlk_portid = 0; + net = auditd_conn.net; + get_net(net); + sk = audit_get_sk(net); + portid = auditd_conn.portid; + rcu_read_unlock(); - /* flush all of the retry queue to the hold queue */ - while ((skb = skb_dequeue(&audit_retry_queue))) - kauditd_hold_skb(skb); + rc = netlink_unicast(sk, skb, portid, 0); + put_net(net); + if (rc < 0) + goto err; + + return rc; + +err: + if (rc == -ECONNREFUSED) + auditd_reset(); + return rc; } /** - * kauditd_send_unicast_skb - Send a record via unicast to auditd - * @skb: audit record + * kauditd_send_queue - Helper for kauditd_thread to flush skb queues + * @sk: the sending sock + * @portid: the netlink destination + * @queue: the skb queue to process + * @retry_limit: limit on number of netlink unicast failures + * @skb_hook: per-skb hook for additional processing + * @err_hook: hook called if the skb fails the netlink unicast send + * + * Description: + * Run through the given queue and attempt to send the audit records to auditd, + * returns zero on success, negative values on failure. It is up to the caller + * to ensure that the @sk is valid for the duration of this function. + * */ -static int kauditd_send_unicast_skb(struct sk_buff *skb) +static int kauditd_send_queue(struct sock *sk, u32 portid, + struct sk_buff_head *queue, + unsigned int retry_limit, + void (*skb_hook)(struct sk_buff *skb), + void (*err_hook)(struct sk_buff *skb)) { - int rc; + int rc = 0; + struct sk_buff *skb; + static unsigned int failed = 0; - /* if we know nothing is connected, don't even try the netlink call */ - if (!audit_pid) - return -ECONNREFUSED; + /* NOTE: kauditd_thread takes care of all our locking, we just use + * the netlink info passed to us (e.g. sk and portid) */ + + while ((skb = skb_dequeue(queue))) { + /* call the skb_hook for each skb we touch */ + if (skb_hook) + (*skb_hook)(skb); + + /* can we send to anyone via unicast? */ + if (!sk) { + if (err_hook) + (*err_hook)(skb); + continue; + } - /* get an extra skb reference in case we fail to send */ - skb_get(skb); - rc = netlink_unicast(audit_sock, skb, audit_nlk_portid, 0); - if (rc >= 0) { - consume_skb(skb); - rc = 0; + /* grab an extra skb reference in case of error */ + skb_get(skb); + rc = netlink_unicast(sk, skb, portid, 0); + if (rc < 0) { + /* fatal failure for our queue flush attempt? */ + if (++failed >= retry_limit || + rc == -ECONNREFUSED || rc == -EPERM) { + /* yes - error processing for the queue */ + sk = NULL; + if (err_hook) + (*err_hook)(skb); + if (!skb_hook) + goto out; + /* keep processing with the skb_hook */ + continue; + } else + /* no - requeue to preserve ordering */ + skb_queue_head(queue, skb); + } else { + /* it worked - drop the extra reference and continue */ + consume_skb(skb); + failed = 0; + } } - return rc; +out: + return (rc >= 0 ? 0 : rc); } /* @@ -493,16 +690,19 @@ static int kauditd_send_unicast_skb(struct sk_buff *skb) * @skb: audit record * * Description: - * This function doesn't consume an skb as might be expected since it has to - * copy it anyways. + * Write a multicast message to anyone listening in the initial network + * namespace. This function doesn't consume an skb as might be expected since + * it has to copy it anyways. */ static void kauditd_send_multicast_skb(struct sk_buff *skb) { struct sk_buff *copy; - struct audit_net *aunet = net_generic(&init_net, audit_net_id); - struct sock *sock = aunet->nlsk; + struct sock *sock = audit_get_sk(&init_net); struct nlmsghdr *nlh; + /* NOTE: we are not taking an additional reference for init_net since + * we don't have to worry about it going away */ + if (!netlink_has_listeners(sock, AUDIT_NLGRP_READLOG)) return; @@ -526,149 +726,75 @@ static void kauditd_send_multicast_skb(struct sk_buff *skb) } /** - * kauditd_wake_condition - Return true when it is time to wake kauditd_thread - * - * Description: - * This function is for use by the wait_event_freezable() call in - * kauditd_thread(). + * kauditd_thread - Worker thread to send audit records to userspace + * @dummy: unused */ -static int kauditd_wake_condition(void) -{ - static int pid_last = 0; - int rc; - int pid = audit_pid; - - /* wake on new messages or a change in the connected auditd */ - rc = skb_queue_len(&audit_queue) || (pid && pid != pid_last); - if (rc) - pid_last = pid; - - return rc; -} - static int kauditd_thread(void *dummy) { int rc; - int auditd = 0; - int reschedule = 0; - struct sk_buff *skb; - struct nlmsghdr *nlh; + u32 portid = 0; + struct net *net = NULL; + struct sock *sk = NULL; #define UNICAST_RETRIES 5 -#define AUDITD_BAD(x,y) \ - ((x) == -ECONNREFUSED || (x) == -EPERM || ++(y) >= UNICAST_RETRIES) - - /* NOTE: we do invalidate the auditd connection flag on any sending - * errors, but we only "restore" the connection flag at specific places - * in the loop in order to help ensure proper ordering of audit - * records */ set_freezable(); while (!kthread_should_stop()) { - /* NOTE: possible area for future improvement is to look at - * the hold and retry queues, since only this thread - * has access to these queues we might be able to do - * our own queuing and skip some/all of the locking */ - - /* NOTE: it might be a fun experiment to split the hold and - * retry queue handling to another thread, but the - * synchronization issues and other overhead might kill - * any performance gains */ + /* NOTE: see the lock comments in auditd_send_unicast_skb() */ + rcu_read_lock(); + if (!auditd_conn.pid) { + rcu_read_unlock(); + goto main_queue; + } + net = auditd_conn.net; + get_net(net); + sk = audit_get_sk(net); + portid = auditd_conn.portid; + rcu_read_unlock(); /* attempt to flush the hold queue */ - while (auditd && (skb = skb_dequeue(&audit_hold_queue))) { - rc = kauditd_send_unicast_skb(skb); - if (rc) { - /* requeue to the same spot */ - skb_queue_head(&audit_hold_queue, skb); - - auditd = 0; - if (AUDITD_BAD(rc, reschedule)) { - mutex_lock(&audit_cmd_mutex); - auditd_reset(); - mutex_unlock(&audit_cmd_mutex); - reschedule = 0; - } - } else - /* we were able to send successfully */ - reschedule = 0; + rc = kauditd_send_queue(sk, portid, + &audit_hold_queue, UNICAST_RETRIES, + NULL, kauditd_rehold_skb); + if (rc < 0) { + sk = NULL; + goto main_queue; } /* attempt to flush the retry queue */ - while (auditd && (skb = skb_dequeue(&audit_retry_queue))) { - rc = kauditd_send_unicast_skb(skb); - if (rc) { - auditd = 0; - if (AUDITD_BAD(rc, reschedule)) { - kauditd_hold_skb(skb); - mutex_lock(&audit_cmd_mutex); - auditd_reset(); - mutex_unlock(&audit_cmd_mutex); - reschedule = 0; - } else - /* temporary problem (we hope), queue - * to the same spot and retry */ - skb_queue_head(&audit_retry_queue, skb); - } else - /* we were able to send successfully */ - reschedule = 0; + rc = kauditd_send_queue(sk, portid, + &audit_retry_queue, UNICAST_RETRIES, + NULL, kauditd_hold_skb); + if (rc < 0) { + sk = NULL; + goto main_queue; } - /* standard queue processing, try to be as quick as possible */ -quick_loop: - skb = skb_dequeue(&audit_queue); - if (skb) { - /* setup the netlink header, see the comments in - * kauditd_send_multicast_skb() for length quirks */ - nlh = nlmsg_hdr(skb); - nlh->nlmsg_len = skb->len - NLMSG_HDRLEN; - - /* attempt to send to any multicast listeners */ - kauditd_send_multicast_skb(skb); - - /* attempt to send to auditd, queue on failure */ - if (auditd) { - rc = kauditd_send_unicast_skb(skb); - if (rc) { - auditd = 0; - if (AUDITD_BAD(rc, reschedule)) { - mutex_lock(&audit_cmd_mutex); - auditd_reset(); - mutex_unlock(&audit_cmd_mutex); - reschedule = 0; - } - - /* move to the retry queue */ - kauditd_retry_skb(skb); - } else - /* everything is working so go fast! */ - goto quick_loop; - } else if (reschedule) - /* we are currently having problems, move to - * the retry queue */ - kauditd_retry_skb(skb); - else - /* dump the message via printk and hold it */ - kauditd_hold_skb(skb); - } else { - /* we have flushed the backlog so wake everyone */ - wake_up(&audit_backlog_wait); - - /* if everything is okay with auditd (if present), go - * to sleep until there is something new in the queue - * or we have a change in the connected auditd; - * otherwise simply reschedule to give things a chance - * to recover */ - if (reschedule) { - set_current_state(TASK_INTERRUPTIBLE); - schedule(); - } else - wait_event_freezable(kauditd_wait, - kauditd_wake_condition()); - - /* update the auditd connection status */ - auditd = (audit_pid ? 1 : 0); +main_queue: + /* process the main queue - do the multicast send and attempt + * unicast, dump failed record sends to the retry queue; if + * sk == NULL due to previous failures we will just do the + * multicast send and move the record to the retry queue */ + kauditd_send_queue(sk, portid, &audit_queue, 1, + kauditd_send_multicast_skb, + kauditd_retry_skb); + + /* drop our netns reference, no auditd sends past this line */ + if (net) { + put_net(net); + net = NULL; } + sk = NULL; + + /* we have processed all the queues so wake everyone */ + wake_up(&audit_backlog_wait); + + /* NOTE: we want to wake up if there is anything on the queue, + * regardless of if an auditd is connected, as we need to + * do the multicast send and rotate records from the + * main queue to the retry/hold queues */ + wait_event_freezable(kauditd_wait, + (skb_queue_len(&audit_queue) ? 1 : 0)); } return 0; @@ -678,17 +804,16 @@ int audit_send_list(void *_dest) { struct audit_netlink_list *dest = _dest; struct sk_buff *skb; - struct net *net = dest->net; - struct audit_net *aunet = net_generic(net, audit_net_id); + struct sock *sk = audit_get_sk(dest->net); /* wait for parent to finish and send an ACK */ mutex_lock(&audit_cmd_mutex); mutex_unlock(&audit_cmd_mutex); while ((skb = __skb_dequeue(&dest->q)) != NULL) - netlink_unicast(aunet->nlsk, skb, dest->portid, 0); + netlink_unicast(sk, skb, dest->portid, 0); - put_net(net); + put_net(dest->net); kfree(dest); return 0; @@ -722,16 +847,15 @@ out_kfree_skb: static int audit_send_reply_thread(void *arg) { struct audit_reply *reply = (struct audit_reply *)arg; - struct net *net = reply->net; - struct audit_net *aunet = net_generic(net, audit_net_id); + struct sock *sk = audit_get_sk(reply->net); mutex_lock(&audit_cmd_mutex); mutex_unlock(&audit_cmd_mutex); /* Ignore failure. It'll only happen if the sender goes away, because our timeout is set to infinite. */ - netlink_unicast(aunet->nlsk , reply->skb, reply->portid, 0); - put_net(net); + netlink_unicast(sk, reply->skb, reply->portid, 0); + put_net(reply->net); kfree(reply); return 0; } @@ -949,12 +1073,12 @@ static int audit_set_feature(struct sk_buff *skb) static int audit_replace(pid_t pid) { - struct sk_buff *skb = audit_make_reply(0, 0, AUDIT_REPLACE, 0, 0, - &pid, sizeof(pid)); + struct sk_buff *skb; + skb = audit_make_reply(0, 0, AUDIT_REPLACE, 0, 0, &pid, sizeof(pid)); if (!skb) return -ENOMEM; - return netlink_unicast(audit_sock, skb, audit_nlk_portid, 0); + return auditd_send_unicast_skb(skb); } static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) @@ -981,7 +1105,9 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) memset(&s, 0, sizeof(s)); s.enabled = audit_enabled; s.failure = audit_failure; - s.pid = audit_pid; + rcu_read_lock(); + s.pid = auditd_conn.pid; + rcu_read_unlock(); s.rate_limit = audit_rate_limit; s.backlog_limit = audit_backlog_limit; s.lost = atomic_read(&audit_lost); @@ -1014,30 +1140,44 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) * from the initial pid namespace, but something * to keep in mind if this changes */ int new_pid = s.pid; + pid_t auditd_pid; pid_t requesting_pid = task_tgid_vnr(current); - if ((!new_pid) && (requesting_pid != audit_pid)) { - audit_log_config_change("audit_pid", new_pid, audit_pid, 0); + /* test the auditd connection */ + audit_replace(requesting_pid); + + rcu_read_lock(); + auditd_pid = auditd_conn.pid; + /* only the current auditd can unregister itself */ + if ((!new_pid) && (requesting_pid != auditd_pid)) { + rcu_read_unlock(); + audit_log_config_change("audit_pid", new_pid, + auditd_pid, 0); return -EACCES; } - if (audit_pid && new_pid && - audit_replace(requesting_pid) != -ECONNREFUSED) { - audit_log_config_change("audit_pid", new_pid, audit_pid, 0); + /* replacing a healthy auditd is not allowed */ + if (auditd_pid && new_pid) { + rcu_read_unlock(); + audit_log_config_change("audit_pid", new_pid, + auditd_pid, 0); return -EEXIST; } + rcu_read_unlock(); + if (audit_enabled != AUDIT_OFF) - audit_log_config_change("audit_pid", new_pid, audit_pid, 1); + audit_log_config_change("audit_pid", new_pid, + auditd_pid, 1); + if (new_pid) { - if (audit_sock) - sock_put(audit_sock); - audit_pid = new_pid; - audit_nlk_portid = NETLINK_CB(skb).portid; - sock_hold(skb->sk); - audit_sock = skb->sk; - } else { + /* register a new auditd connection */ + auditd_set(new_pid, + NETLINK_CB(skb).portid, + sock_net(NETLINK_CB(skb).sk)); + /* try to process any backlog */ + wake_up_interruptible(&kauditd_wait); + } else + /* unregister the auditd connection */ auditd_reset(); - } - wake_up_interruptible(&kauditd_wait); } if (s.mask & AUDIT_STATUS_RATE_LIMIT) { err = audit_set_rate_limit(s.rate_limit); @@ -1090,7 +1230,6 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) if (err) break; } - mutex_unlock(&audit_cmd_mutex); audit_log_common_recv_msg(&ab, msg_type); if (msg_type != AUDIT_USER_TTY) audit_log_format(ab, " msg='%.*s'", @@ -1108,7 +1247,6 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) } audit_set_portid(ab, NETLINK_CB(skb).portid); audit_log_end(ab); - mutex_lock(&audit_cmd_mutex); } break; case AUDIT_ADD_RULE: @@ -1298,26 +1436,26 @@ static int __net_init audit_net_init(struct net *net) struct audit_net *aunet = net_generic(net, audit_net_id); - aunet->nlsk = netlink_kernel_create(net, NETLINK_AUDIT, &cfg); - if (aunet->nlsk == NULL) { + aunet->sk = netlink_kernel_create(net, NETLINK_AUDIT, &cfg); + if (aunet->sk == NULL) { audit_panic("cannot initialize netlink socket in namespace"); return -ENOMEM; } - aunet->nlsk->sk_sndtimeo = MAX_SCHEDULE_TIMEOUT; + aunet->sk->sk_sndtimeo = MAX_SCHEDULE_TIMEOUT; + return 0; } static void __net_exit audit_net_exit(struct net *net) { struct audit_net *aunet = net_generic(net, audit_net_id); - struct sock *sock = aunet->nlsk; - mutex_lock(&audit_cmd_mutex); - if (sock == audit_sock) + + rcu_read_lock(); + if (net == auditd_conn.net) auditd_reset(); - mutex_unlock(&audit_cmd_mutex); + rcu_read_unlock(); - netlink_kernel_release(sock); - aunet->nlsk = NULL; + netlink_kernel_release(aunet->sk); } static struct pernet_operations audit_net_ops __net_initdata = { @@ -1335,20 +1473,24 @@ static int __init audit_init(void) if (audit_initialized == AUDIT_DISABLED) return 0; - pr_info("initializing netlink subsys (%s)\n", - audit_default ? "enabled" : "disabled"); - register_pernet_subsys(&audit_net_ops); + memset(&auditd_conn, 0, sizeof(auditd_conn)); + spin_lock_init(&auditd_conn.lock); skb_queue_head_init(&audit_queue); skb_queue_head_init(&audit_retry_queue); skb_queue_head_init(&audit_hold_queue); - audit_initialized = AUDIT_INITIALIZED; - audit_enabled = audit_default; - audit_ever_enabled |= !!audit_default; for (i = 0; i < AUDIT_INODE_BUCKETS; i++) INIT_LIST_HEAD(&audit_inode_hash[i]); + pr_info("initializing netlink subsys (%s)\n", + audit_default ? "enabled" : "disabled"); + register_pernet_subsys(&audit_net_ops); + + audit_initialized = AUDIT_INITIALIZED; + audit_enabled = audit_default; + audit_ever_enabled |= !!audit_default; + kauditd_task = kthread_run(kauditd_thread, NULL, "kauditd"); if (IS_ERR(kauditd_task)) { int err = PTR_ERR(kauditd_task); @@ -1519,20 +1661,16 @@ struct audit_buffer *audit_log_start(struct audit_context *ctx, gfp_t gfp_mask, if (unlikely(!audit_filter(type, AUDIT_FILTER_TYPE))) return NULL; - /* don't ever fail/sleep on these two conditions: + /* NOTE: don't ever fail/sleep on these two conditions: * 1. auditd generated record - since we need auditd to drain the * queue; also, when we are checking for auditd, compare PIDs using * task_tgid_vnr() since auditd_pid is set in audit_receive_msg() * using a PID anchored in the caller's namespace - * 2. audit command message - record types 1000 through 1099 inclusive - * are command messages/records used to manage the kernel subsystem - * and the audit userspace, blocking on these messages could cause - * problems under load so don't do it (note: not all of these - * command types are valid as record types, but it is quicker to - * just check two ints than a series of ints in a if/switch stmt) */ - if (!((audit_pid && audit_pid == task_tgid_vnr(current)) || - (type >= 1000 && type <= 1099))) { - long sleep_time = audit_backlog_wait_time; + * 2. generator holding the audit_cmd_mutex - we don't want to block + * while holding the mutex */ + if (!(auditd_test_task(current) || + (current == __mutex_owner(&audit_cmd_mutex)))) { + long stime = audit_backlog_wait_time; while (audit_backlog_limit && (skb_queue_len(&audit_queue) > audit_backlog_limit)) { @@ -1541,14 +1679,13 @@ struct audit_buffer *audit_log_start(struct audit_context *ctx, gfp_t gfp_mask, /* sleep if we are allowed and we haven't exhausted our * backlog wait limit */ - if ((gfp_mask & __GFP_DIRECT_RECLAIM) && - (sleep_time > 0)) { + if (gfpflags_allow_blocking(gfp_mask) && (stime > 0)) { DECLARE_WAITQUEUE(wait, current); add_wait_queue_exclusive(&audit_backlog_wait, &wait); set_current_state(TASK_UNINTERRUPTIBLE); - sleep_time = schedule_timeout(sleep_time); + stime = schedule_timeout(stime); remove_wait_queue(&audit_backlog_wait, &wait); } else { if (audit_rate_check() && printk_ratelimit()) @@ -2127,15 +2264,27 @@ out: */ void audit_log_end(struct audit_buffer *ab) { + struct sk_buff *skb; + struct nlmsghdr *nlh; + if (!ab) return; - if (!audit_rate_check()) { - audit_log_lost("rate limit exceeded"); - } else { - skb_queue_tail(&audit_queue, ab->skb); - wake_up_interruptible(&kauditd_wait); + + if (audit_rate_check()) { + skb = ab->skb; ab->skb = NULL; - } + + /* setup the netlink header, see the comments in + * kauditd_send_multicast_skb() for length quirks */ + nlh = nlmsg_hdr(skb); + nlh->nlmsg_len = skb->len - NLMSG_HDRLEN; + + /* queue the netlink packet and poke the kauditd thread */ + skb_queue_tail(&audit_queue, skb); + wake_up_interruptible(&kauditd_wait); + } else + audit_log_lost("rate limit exceeded"); + audit_buffer_free(ab); } diff --git a/kernel/audit.h b/kernel/audit.h index ca579880303a..0f1cf6d1878a 100644 --- a/kernel/audit.h +++ b/kernel/audit.h @@ -218,7 +218,7 @@ extern void audit_log_name(struct audit_context *context, struct audit_names *n, const struct path *path, int record_num, int *call_panic); -extern int audit_pid; +extern int auditd_test_task(const struct task_struct *task); #define AUDIT_INODE_BUCKETS 32 extern struct list_head audit_inode_hash[AUDIT_INODE_BUCKETS]; @@ -250,10 +250,6 @@ struct audit_netlink_list { int audit_send_list(void *); -struct audit_net { - struct sock *nlsk; -}; - extern int selinux_audit_rule_update(void); extern struct mutex audit_filter_mutex; @@ -340,8 +336,7 @@ extern int audit_filter(int msgtype, unsigned int listtype); extern int __audit_signal_info(int sig, struct task_struct *t); static inline int audit_signal_info(int sig, struct task_struct *t) { - if (unlikely((audit_pid && t->tgid == audit_pid) || - (audit_signals && !audit_dummy_context()))) + if (auditd_test_task(t) || (audit_signals && !audit_dummy_context())) return __audit_signal_info(sig, t); return 0; } diff --git a/kernel/auditsc.c b/kernel/auditsc.c index d6a8de5f8fa3..e59ffc7fc522 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -762,7 +762,7 @@ static enum audit_state audit_filter_syscall(struct task_struct *tsk, struct audit_entry *e; enum audit_state state; - if (audit_pid && tsk->tgid == audit_pid) + if (auditd_test_task(tsk)) return AUDIT_DISABLED; rcu_read_lock(); @@ -816,7 +816,7 @@ void audit_filter_inodes(struct task_struct *tsk, struct audit_context *ctx) { struct audit_names *n; - if (audit_pid && tsk->tgid == audit_pid) + if (auditd_test_task(tsk)) return; rcu_read_lock(); @@ -2256,7 +2256,7 @@ int __audit_signal_info(int sig, struct task_struct *t) struct audit_context *ctx = tsk->audit_context; kuid_t uid = current_uid(), t_uid = task_uid(t); - if (audit_pid && t->tgid == audit_pid) { + if (auditd_test_task(t)) { if (sig == SIGTERM || sig == SIGHUP || sig == SIGUSR1 || sig == SIGUSR2) { audit_sig_pid = task_tgid_nr(tsk); if (uid_valid(tsk->loginuid)) -- cgit v1.2.3 From 8c290e60fa2a51806159522331c9ed41252a8fb3 Mon Sep 17 00:00:00 2001 From: Alexei Starovoitov Date: Tue, 21 Mar 2017 19:05:04 -0700 Subject: bpf: fix hashmap extra_elems logic In both kmalloc and prealloc mode the bpf_map_update_elem() is using per-cpu extra_elems to do atomic update when the map is full. There are two issues with it. The logic can be misused, since it allows max_entries+num_cpus elements to be present in the map. And alloc_extra_elems() at map creation time can fail percpu alloc for large map values with a warn: WARNING: CPU: 3 PID: 2752 at ../mm/percpu.c:892 pcpu_alloc+0x119/0xa60 illegal size (32824) or align (8) for percpu allocation The fixes for both of these issues are different for kmalloc and prealloc modes. For prealloc mode allocate extra num_possible_cpus elements and store their pointers into extra_elems array instead of actual elements. Hence we can use these hidden(spare) elements not only when the map is full but during bpf_map_update_elem() that replaces existing element too. That also improves performance, since pcpu_freelist_pop/push is avoided. Unfortunately this approach cannot be used for kmalloc mode which needs to kfree elements after rcu grace period. Therefore switch it back to normal kmalloc even when full and old element exists like it was prior to commit 6c9059817432 ("bpf: pre-allocate hash map elements"). Add tests to check for over max_entries and large map values. Reported-by: Dave Jones Fixes: 6c9059817432 ("bpf: pre-allocate hash map elements") Signed-off-by: Alexei Starovoitov Acked-by: Daniel Borkmann Acked-by: Martin KaFai Lau Signed-off-by: David S. Miller --- kernel/bpf/hashtab.c | 144 ++++++++++++++++---------------- tools/testing/selftests/bpf/test_maps.c | 29 ++++++- 2 files changed, 97 insertions(+), 76 deletions(-) (limited to 'kernel') diff --git a/kernel/bpf/hashtab.c b/kernel/bpf/hashtab.c index afe5bab376c9..361a69dfe543 100644 --- a/kernel/bpf/hashtab.c +++ b/kernel/bpf/hashtab.c @@ -30,18 +30,12 @@ struct bpf_htab { struct pcpu_freelist freelist; struct bpf_lru lru; }; - void __percpu *extra_elems; + struct htab_elem *__percpu *extra_elems; atomic_t count; /* number of elements in this hashtable */ u32 n_buckets; /* number of hash buckets */ u32 elem_size; /* size of each element in bytes */ }; -enum extra_elem_state { - HTAB_NOT_AN_EXTRA_ELEM = 0, - HTAB_EXTRA_ELEM_FREE, - HTAB_EXTRA_ELEM_USED -}; - /* each htab element is struct htab_elem + key + value */ struct htab_elem { union { @@ -56,7 +50,6 @@ struct htab_elem { }; union { struct rcu_head rcu; - enum extra_elem_state state; struct bpf_lru_node lru_node; }; u32 hash; @@ -77,6 +70,11 @@ static bool htab_is_percpu(const struct bpf_htab *htab) htab->map.map_type == BPF_MAP_TYPE_LRU_PERCPU_HASH; } +static bool htab_is_prealloc(const struct bpf_htab *htab) +{ + return !(htab->map.map_flags & BPF_F_NO_PREALLOC); +} + static inline void htab_elem_set_ptr(struct htab_elem *l, u32 key_size, void __percpu *pptr) { @@ -128,17 +126,20 @@ static struct htab_elem *prealloc_lru_pop(struct bpf_htab *htab, void *key, static int prealloc_init(struct bpf_htab *htab) { + u32 num_entries = htab->map.max_entries; int err = -ENOMEM, i; - htab->elems = bpf_map_area_alloc(htab->elem_size * - htab->map.max_entries); + if (!htab_is_percpu(htab) && !htab_is_lru(htab)) + num_entries += num_possible_cpus(); + + htab->elems = bpf_map_area_alloc(htab->elem_size * num_entries); if (!htab->elems) return -ENOMEM; if (!htab_is_percpu(htab)) goto skip_percpu_elems; - for (i = 0; i < htab->map.max_entries; i++) { + for (i = 0; i < num_entries; i++) { u32 size = round_up(htab->map.value_size, 8); void __percpu *pptr; @@ -166,11 +167,11 @@ skip_percpu_elems: if (htab_is_lru(htab)) bpf_lru_populate(&htab->lru, htab->elems, offsetof(struct htab_elem, lru_node), - htab->elem_size, htab->map.max_entries); + htab->elem_size, num_entries); else pcpu_freelist_populate(&htab->freelist, htab->elems + offsetof(struct htab_elem, fnode), - htab->elem_size, htab->map.max_entries); + htab->elem_size, num_entries); return 0; @@ -191,16 +192,22 @@ static void prealloc_destroy(struct bpf_htab *htab) static int alloc_extra_elems(struct bpf_htab *htab) { - void __percpu *pptr; + struct htab_elem *__percpu *pptr, *l_new; + struct pcpu_freelist_node *l; int cpu; - pptr = __alloc_percpu_gfp(htab->elem_size, 8, GFP_USER | __GFP_NOWARN); + pptr = __alloc_percpu_gfp(sizeof(struct htab_elem *), 8, + GFP_USER | __GFP_NOWARN); if (!pptr) return -ENOMEM; for_each_possible_cpu(cpu) { - ((struct htab_elem *)per_cpu_ptr(pptr, cpu))->state = - HTAB_EXTRA_ELEM_FREE; + l = pcpu_freelist_pop(&htab->freelist); + /* pop will succeed, since prealloc_init() + * preallocated extra num_possible_cpus elements + */ + l_new = container_of(l, struct htab_elem, fnode); + *per_cpu_ptr(pptr, cpu) = l_new; } htab->extra_elems = pptr; return 0; @@ -342,25 +349,25 @@ static struct bpf_map *htab_map_alloc(union bpf_attr *attr) raw_spin_lock_init(&htab->buckets[i].lock); } - if (!percpu && !lru) { - /* lru itself can remove the least used element, so - * there is no need for an extra elem during map_update. - */ - err = alloc_extra_elems(htab); - if (err) - goto free_buckets; - } - if (prealloc) { err = prealloc_init(htab); if (err) - goto free_extra_elems; + goto free_buckets; + + if (!percpu && !lru) { + /* lru itself can remove the least used element, so + * there is no need for an extra elem during map_update. + */ + err = alloc_extra_elems(htab); + if (err) + goto free_prealloc; + } } return &htab->map; -free_extra_elems: - free_percpu(htab->extra_elems); +free_prealloc: + prealloc_destroy(htab); free_buckets: bpf_map_area_free(htab->buckets); free_htab: @@ -575,12 +582,7 @@ static void htab_elem_free_rcu(struct rcu_head *head) static void free_htab_elem(struct bpf_htab *htab, struct htab_elem *l) { - if (l->state == HTAB_EXTRA_ELEM_USED) { - l->state = HTAB_EXTRA_ELEM_FREE; - return; - } - - if (!(htab->map.map_flags & BPF_F_NO_PREALLOC)) { + if (htab_is_prealloc(htab)) { pcpu_freelist_push(&htab->freelist, &l->fnode); } else { atomic_dec(&htab->count); @@ -610,47 +612,43 @@ static void pcpu_copy_value(struct bpf_htab *htab, void __percpu *pptr, static struct htab_elem *alloc_htab_elem(struct bpf_htab *htab, void *key, void *value, u32 key_size, u32 hash, bool percpu, bool onallcpus, - bool old_elem_exists) + struct htab_elem *old_elem) { u32 size = htab->map.value_size; - bool prealloc = !(htab->map.map_flags & BPF_F_NO_PREALLOC); - struct htab_elem *l_new; + bool prealloc = htab_is_prealloc(htab); + struct htab_elem *l_new, **pl_new; void __percpu *pptr; - int err = 0; if (prealloc) { - struct pcpu_freelist_node *l; + if (old_elem) { + /* if we're updating the existing element, + * use per-cpu extra elems to avoid freelist_pop/push + */ + pl_new = this_cpu_ptr(htab->extra_elems); + l_new = *pl_new; + *pl_new = old_elem; + } else { + struct pcpu_freelist_node *l; - l = pcpu_freelist_pop(&htab->freelist); - if (!l) - err = -E2BIG; - else + l = pcpu_freelist_pop(&htab->freelist); + if (!l) + return ERR_PTR(-E2BIG); l_new = container_of(l, struct htab_elem, fnode); - } else { - if (atomic_inc_return(&htab->count) > htab->map.max_entries) { - atomic_dec(&htab->count); - err = -E2BIG; - } else { - l_new = kmalloc(htab->elem_size, - GFP_ATOMIC | __GFP_NOWARN); - if (!l_new) - return ERR_PTR(-ENOMEM); } - } - - if (err) { - if (!old_elem_exists) - return ERR_PTR(err); - - /* if we're updating the existing element and the hash table - * is full, use per-cpu extra elems - */ - l_new = this_cpu_ptr(htab->extra_elems); - if (l_new->state != HTAB_EXTRA_ELEM_FREE) - return ERR_PTR(-E2BIG); - l_new->state = HTAB_EXTRA_ELEM_USED; } else { - l_new->state = HTAB_NOT_AN_EXTRA_ELEM; + if (atomic_inc_return(&htab->count) > htab->map.max_entries) + if (!old_elem) { + /* when map is full and update() is replacing + * old element, it's ok to allocate, since + * old element will be freed immediately. + * Otherwise return an error + */ + atomic_dec(&htab->count); + return ERR_PTR(-E2BIG); + } + l_new = kmalloc(htab->elem_size, GFP_ATOMIC | __GFP_NOWARN); + if (!l_new) + return ERR_PTR(-ENOMEM); } memcpy(l_new->key, key, key_size); @@ -731,7 +729,7 @@ static int htab_map_update_elem(struct bpf_map *map, void *key, void *value, goto err; l_new = alloc_htab_elem(htab, key, value, key_size, hash, false, false, - !!l_old); + l_old); if (IS_ERR(l_new)) { /* all pre-allocated elements are in use or memory exhausted */ ret = PTR_ERR(l_new); @@ -744,7 +742,8 @@ static int htab_map_update_elem(struct bpf_map *map, void *key, void *value, hlist_nulls_add_head_rcu(&l_new->hash_node, head); if (l_old) { hlist_nulls_del_rcu(&l_old->hash_node); - free_htab_elem(htab, l_old); + if (!htab_is_prealloc(htab)) + free_htab_elem(htab, l_old); } ret = 0; err: @@ -856,7 +855,7 @@ static int __htab_percpu_map_update_elem(struct bpf_map *map, void *key, value, onallcpus); } else { l_new = alloc_htab_elem(htab, key, value, key_size, - hash, true, onallcpus, false); + hash, true, onallcpus, NULL); if (IS_ERR(l_new)) { ret = PTR_ERR(l_new); goto err; @@ -1024,8 +1023,7 @@ static void delete_all_elements(struct bpf_htab *htab) hlist_nulls_for_each_entry_safe(l, n, head, hash_node) { hlist_nulls_del_rcu(&l->hash_node); - if (l->state != HTAB_EXTRA_ELEM_USED) - htab_elem_free(htab, l); + htab_elem_free(htab, l); } } } @@ -1045,7 +1043,7 @@ static void htab_map_free(struct bpf_map *map) * not have executed. Wait for them. */ rcu_barrier(); - if (htab->map.map_flags & BPF_F_NO_PREALLOC) + if (!htab_is_prealloc(htab)) delete_all_elements(htab); else prealloc_destroy(htab); diff --git a/tools/testing/selftests/bpf/test_maps.c b/tools/testing/selftests/bpf/test_maps.c index cada17ac00b8..a0aa2009b0e0 100644 --- a/tools/testing/selftests/bpf/test_maps.c +++ b/tools/testing/selftests/bpf/test_maps.c @@ -80,8 +80,9 @@ static void test_hashmap(int task, void *data) assert(bpf_map_update_elem(fd, &key, &value, BPF_EXIST) == 0); key = 2; assert(bpf_map_update_elem(fd, &key, &value, BPF_ANY) == 0); - key = 1; - assert(bpf_map_update_elem(fd, &key, &value, BPF_ANY) == 0); + key = 3; + assert(bpf_map_update_elem(fd, &key, &value, BPF_NOEXIST) == -1 && + errno == E2BIG); /* Check that key = 0 doesn't exist. */ key = 0; @@ -110,6 +111,24 @@ static void test_hashmap(int task, void *data) close(fd); } +static void test_hashmap_sizes(int task, void *data) +{ + int fd, i, j; + + for (i = 1; i <= 512; i <<= 1) + for (j = 1; j <= 1 << 18; j <<= 1) { + fd = bpf_create_map(BPF_MAP_TYPE_HASH, i, j, + 2, map_flags); + if (fd < 0) { + printf("Failed to create hashmap key=%d value=%d '%s'\n", + i, j, strerror(errno)); + exit(1); + } + close(fd); + usleep(10); /* give kernel time to destroy */ + } +} + static void test_hashmap_percpu(int task, void *data) { unsigned int nr_cpus = bpf_num_possible_cpus(); @@ -317,7 +336,10 @@ static void test_arraymap_percpu(int task, void *data) static void test_arraymap_percpu_many_keys(void) { unsigned int nr_cpus = bpf_num_possible_cpus(); - unsigned int nr_keys = 20000; + /* nr_keys is not too large otherwise the test stresses percpu + * allocator more than anything else + */ + unsigned int nr_keys = 2000; long values[nr_cpus]; int key, fd, i; @@ -419,6 +441,7 @@ static void test_map_stress(void) { run_parallel(100, test_hashmap, NULL); run_parallel(100, test_hashmap_percpu, NULL); + run_parallel(100, test_hashmap_sizes, NULL); run_parallel(100, test_arraymap, NULL); run_parallel(100, test_arraymap_percpu, NULL); -- cgit v1.2.3 From 71fdb70eb48784c1f28cdf2e67c4c587dd7f2594 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Mon, 13 Mar 2017 13:46:21 +0100 Subject: sched/clock: Fix clear_sched_clock_stable() preempt wobbly Paul reported a problems with clear_sched_clock_stable(). Since we run all of __clear_sched_clock_stable() from workqueue context, there's a preempt problem. Solve it by only running the static_key_disable() from workqueue. Reported-by: Paul E. McKenney Signed-off-by: Peter Zijlstra (Intel) Cc: Linus Torvalds Cc: Mike Galbraith Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: fweisbec@gmail.com Link: http://lkml.kernel.org/r/20170313124621.GA3328@twins.programming.kicks-ass.net Signed-off-by: Ingo Molnar --- kernel/sched/clock.c | 17 ++++++++++++----- 1 file changed, 12 insertions(+), 5 deletions(-) (limited to 'kernel') diff --git a/kernel/sched/clock.c b/kernel/sched/clock.c index a08795e21628..fec0f58c8dee 100644 --- a/kernel/sched/clock.c +++ b/kernel/sched/clock.c @@ -141,7 +141,14 @@ static void __set_sched_clock_stable(void) tick_dep_clear(TICK_DEP_BIT_CLOCK_UNSTABLE); } -static void __clear_sched_clock_stable(struct work_struct *work) +static void __sched_clock_work(struct work_struct *work) +{ + static_branch_disable(&__sched_clock_stable); +} + +static DECLARE_WORK(sched_clock_work, __sched_clock_work); + +static void __clear_sched_clock_stable(void) { struct sched_clock_data *scd = this_scd(); @@ -160,11 +167,11 @@ static void __clear_sched_clock_stable(struct work_struct *work) scd->tick_gtod, gtod_offset, scd->tick_raw, raw_offset); - static_branch_disable(&__sched_clock_stable); tick_dep_set(TICK_DEP_BIT_CLOCK_UNSTABLE); -} -static DECLARE_WORK(sched_clock_work, __clear_sched_clock_stable); + if (sched_clock_stable()) + schedule_work(&sched_clock_work); +} void clear_sched_clock_stable(void) { @@ -173,7 +180,7 @@ void clear_sched_clock_stable(void) smp_mb(); /* matches sched_clock_init_late() */ if (sched_clock_running == 2) - schedule_work(&sched_clock_work); + __clear_sched_clock_stable(); } void sched_clock_init_late(void) -- cgit v1.2.3 From 698eff6355f735d46d1b7113df8b422874cd7988 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 17 Mar 2017 12:48:18 +0100 Subject: sched/clock, x86/perf: Fix "perf test tsc" People reported that commit: 5680d8094ffa ("sched/clock: Provide better clock continuity") broke "perf test tsc". That commit added another offset to the reported clock value; so take that into account when computing the provided offset values. Reported-by: Adrian Hunter Reported-by: Arnaldo Carvalho de Melo Tested-by: Alexander Shishkin Signed-off-by: Peter Zijlstra (Intel) Cc: Linus Torvalds Cc: Mike Galbraith Cc: Peter Zijlstra Cc: Thomas Gleixner Fixes: 5680d8094ffa ("sched/clock: Provide better clock continuity") Signed-off-by: Ingo Molnar --- arch/x86/events/core.c | 9 ++++++--- arch/x86/include/asm/timer.h | 2 ++ arch/x86/kernel/tsc.c | 4 ++-- include/linux/sched/clock.h | 13 +++++++------ kernel/sched/clock.c | 22 +++++++++++----------- 5 files changed, 28 insertions(+), 22 deletions(-) (limited to 'kernel') diff --git a/arch/x86/events/core.c b/arch/x86/events/core.c index 2aa1ad194db2..580b60f5ac83 100644 --- a/arch/x86/events/core.c +++ b/arch/x86/events/core.c @@ -2256,6 +2256,7 @@ void arch_perf_update_userpage(struct perf_event *event, struct perf_event_mmap_page *userpg, u64 now) { struct cyc2ns_data *data; + u64 offset; userpg->cap_user_time = 0; userpg->cap_user_time_zero = 0; @@ -2263,11 +2264,13 @@ void arch_perf_update_userpage(struct perf_event *event, !!(event->hw.flags & PERF_X86_EVENT_RDPMC_ALLOWED); userpg->pmc_width = x86_pmu.cntval_bits; - if (!sched_clock_stable()) + if (!using_native_sched_clock() || !sched_clock_stable()) return; data = cyc2ns_read_begin(); + offset = data->cyc2ns_offset + __sched_clock_offset; + /* * Internal timekeeping for enabled/running/stopped times * is always in the local_clock domain. @@ -2275,7 +2278,7 @@ void arch_perf_update_userpage(struct perf_event *event, userpg->cap_user_time = 1; userpg->time_mult = data->cyc2ns_mul; userpg->time_shift = data->cyc2ns_shift; - userpg->time_offset = data->cyc2ns_offset - now; + userpg->time_offset = offset - now; /* * cap_user_time_zero doesn't make sense when we're using a different @@ -2283,7 +2286,7 @@ void arch_perf_update_userpage(struct perf_event *event, */ if (!event->attr.use_clockid) { userpg->cap_user_time_zero = 1; - userpg->time_zero = data->cyc2ns_offset; + userpg->time_zero = offset; } cyc2ns_read_end(data); diff --git a/arch/x86/include/asm/timer.h b/arch/x86/include/asm/timer.h index a04eabd43d06..27e9f9d769b8 100644 --- a/arch/x86/include/asm/timer.h +++ b/arch/x86/include/asm/timer.h @@ -12,6 +12,8 @@ extern int recalibrate_cpu_khz(void); extern int no_timer_check; +extern bool using_native_sched_clock(void); + /* * We use the full linear equation: f(x) = a + b*x, in order to allow * a continuous function in the face of dynamic freq changes. diff --git a/arch/x86/kernel/tsc.c b/arch/x86/kernel/tsc.c index c73a7f9e881a..714dfba6a1e7 100644 --- a/arch/x86/kernel/tsc.c +++ b/arch/x86/kernel/tsc.c @@ -328,7 +328,7 @@ unsigned long long sched_clock(void) return paravirt_sched_clock(); } -static inline bool using_native_sched_clock(void) +bool using_native_sched_clock(void) { return pv_time_ops.sched_clock == native_sched_clock; } @@ -336,7 +336,7 @@ static inline bool using_native_sched_clock(void) unsigned long long sched_clock(void) __attribute__((alias("native_sched_clock"))); -static inline bool using_native_sched_clock(void) { return true; } +bool using_native_sched_clock(void) { return true; } #endif int check_tsc_unstable(void) diff --git a/include/linux/sched/clock.h b/include/linux/sched/clock.h index 4a68c6791207..34fe92ce1ebd 100644 --- a/include/linux/sched/clock.h +++ b/include/linux/sched/clock.h @@ -54,15 +54,16 @@ static inline u64 local_clock(void) } #else extern void sched_clock_init_late(void); -/* - * Architectures can set this to 1 if they have specified - * CONFIG_HAVE_UNSTABLE_SCHED_CLOCK in their arch Kconfig, - * but then during bootup it turns out that sched_clock() - * is reliable after all: - */ extern int sched_clock_stable(void); extern void clear_sched_clock_stable(void); +/* + * When sched_clock_stable(), __sched_clock_offset provides the offset + * between local_clock() and sched_clock(). + */ +extern u64 __sched_clock_offset; + + extern void sched_clock_tick(void); extern void sched_clock_idle_sleep_event(void); extern void sched_clock_idle_wakeup_event(u64 delta_ns); diff --git a/kernel/sched/clock.c b/kernel/sched/clock.c index fec0f58c8dee..24a3e01bf8cb 100644 --- a/kernel/sched/clock.c +++ b/kernel/sched/clock.c @@ -96,10 +96,10 @@ static DEFINE_STATIC_KEY_FALSE(__sched_clock_stable); static int __sched_clock_stable_early = 1; /* - * We want: ktime_get_ns() + gtod_offset == sched_clock() + raw_offset + * We want: ktime_get_ns() + __gtod_offset == sched_clock() + __sched_clock_offset */ -static __read_mostly u64 raw_offset; -static __read_mostly u64 gtod_offset; +__read_mostly u64 __sched_clock_offset; +static __read_mostly u64 __gtod_offset; struct sched_clock_data { u64 tick_raw; @@ -131,11 +131,11 @@ static void __set_sched_clock_stable(void) /* * Attempt to make the (initial) unstable->stable transition continuous. */ - raw_offset = (scd->tick_gtod + gtod_offset) - (scd->tick_raw); + __sched_clock_offset = (scd->tick_gtod + __gtod_offset) - (scd->tick_raw); printk(KERN_INFO "sched_clock: Marking stable (%lld, %lld)->(%lld, %lld)\n", - scd->tick_gtod, gtod_offset, - scd->tick_raw, raw_offset); + scd->tick_gtod, __gtod_offset, + scd->tick_raw, __sched_clock_offset); static_branch_enable(&__sched_clock_stable); tick_dep_clear(TICK_DEP_BIT_CLOCK_UNSTABLE); @@ -161,11 +161,11 @@ static void __clear_sched_clock_stable(void) * * Still do what we can. */ - gtod_offset = (scd->tick_raw + raw_offset) - (scd->tick_gtod); + __gtod_offset = (scd->tick_raw + __sched_clock_offset) - (scd->tick_gtod); printk(KERN_INFO "sched_clock: Marking unstable (%lld, %lld)<-(%lld, %lld)\n", - scd->tick_gtod, gtod_offset, - scd->tick_raw, raw_offset); + scd->tick_gtod, __gtod_offset, + scd->tick_raw, __sched_clock_offset); tick_dep_set(TICK_DEP_BIT_CLOCK_UNSTABLE); @@ -238,7 +238,7 @@ again: * scd->tick_gtod + TICK_NSEC); */ - clock = scd->tick_gtod + gtod_offset + delta; + clock = scd->tick_gtod + __gtod_offset + delta; min_clock = wrap_max(scd->tick_gtod, old_clock); max_clock = wrap_max(old_clock, scd->tick_gtod + TICK_NSEC); @@ -324,7 +324,7 @@ u64 sched_clock_cpu(int cpu) u64 clock; if (sched_clock_stable()) - return sched_clock() + raw_offset; + return sched_clock() + __sched_clock_offset; if (unlikely(!sched_clock_running)) return 0ull; -- cgit v1.2.3 From de5540d088fe97ad583cc7d396586437b32149a5 Mon Sep 17 00:00:00 2001 From: "Jason A. Donenfeld" Date: Thu, 23 Mar 2017 12:24:43 +0100 Subject: padata: avoid race in reordering Under extremely heavy uses of padata, crashes occur, and with list debugging turned on, this happens instead: [87487.298728] WARNING: CPU: 1 PID: 882 at lib/list_debug.c:33 __list_add+0xae/0x130 [87487.301868] list_add corruption. prev->next should be next (ffffb17abfc043d0), but was ffff8dba70872c80. (prev=ffff8dba70872b00). [87487.339011] [] dump_stack+0x68/0xa3 [87487.342198] [] ? console_unlock+0x281/0x6d0 [87487.345364] [] __warn+0xff/0x140 [87487.348513] [] warn_slowpath_fmt+0x4a/0x50 [87487.351659] [] __list_add+0xae/0x130 [87487.354772] [] ? _raw_spin_lock+0x64/0x70 [87487.357915] [] padata_reorder+0x1e6/0x420 [87487.361084] [] padata_do_serial+0xa5/0x120 padata_reorder calls list_add_tail with the list to which its adding locked, which seems correct: spin_lock(&squeue->serial.lock); list_add_tail(&padata->list, &squeue->serial.list); spin_unlock(&squeue->serial.lock); This therefore leaves only place where such inconsistency could occur: if padata->list is added at the same time on two different threads. This pdata pointer comes from the function call to padata_get_next(pd), which has in it the following block: next_queue = per_cpu_ptr(pd->pqueue, cpu); padata = NULL; reorder = &next_queue->reorder; if (!list_empty(&reorder->list)) { padata = list_entry(reorder->list.next, struct padata_priv, list); spin_lock(&reorder->lock); list_del_init(&padata->list); atomic_dec(&pd->reorder_objects); spin_unlock(&reorder->lock); pd->processed++; goto out; } out: return padata; I strongly suspect that the problem here is that two threads can race on reorder list. Even though the deletion is locked, call to list_entry is not locked, which means it's feasible that two threads pick up the same padata object and subsequently call list_add_tail on them at the same time. The fix is thus be hoist that lock outside of that block. Signed-off-by: Jason A. Donenfeld Acked-by: Steffen Klassert Signed-off-by: Herbert Xu --- kernel/padata.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/padata.c b/kernel/padata.c index 05316c9f32da..3202aa17492c 100644 --- a/kernel/padata.c +++ b/kernel/padata.c @@ -186,19 +186,20 @@ static struct padata_priv *padata_get_next(struct parallel_data *pd) reorder = &next_queue->reorder; + spin_lock(&reorder->lock); if (!list_empty(&reorder->list)) { padata = list_entry(reorder->list.next, struct padata_priv, list); - spin_lock(&reorder->lock); list_del_init(&padata->list); atomic_dec(&pd->reorder_objects); - spin_unlock(&reorder->lock); pd->processed++; + spin_unlock(&reorder->lock); goto out; } + spin_unlock(&reorder->lock); if (__this_cpu_read(pd->pqueue->cpu_index) == next_queue->cpu_index) { padata = ERR_PTR(-ENODATA); -- cgit v1.2.3 From b1977682a3858b5584ffea7cfb7bd863f68db18d Mon Sep 17 00:00:00 2001 From: Alexei Starovoitov Date: Fri, 24 Mar 2017 15:57:33 -0700 Subject: bpf: improve verifier packet range checks llvm can optimize the 'if (ptr > data_end)' checks to be in the order slightly different than the original C code which will confuse verifier. Like: if (ptr + 16 > data_end) return TC_ACT_SHOT; // may be followed by if (ptr + 14 > data_end) return TC_ACT_SHOT; while llvm can see that 'ptr' is valid for all 16 bytes, the verifier could not. Fix verifier logic to account for such case and add a test. Reported-by: Huapeng Zhou Fixes: 969bf05eb3ce ("bpf: direct packet access") Signed-off-by: Alexei Starovoitov Acked-by: Daniel Borkmann Acked-by: Martin KaFai Lau Signed-off-by: David S. Miller --- kernel/bpf/verifier.c | 5 +++-- tools/testing/selftests/bpf/test_verifier.c | 20 ++++++++++++++++++++ 2 files changed, 23 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index 796b68d00119..5e6202e62265 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -1973,14 +1973,15 @@ static void find_good_pkt_pointers(struct bpf_verifier_state *state, for (i = 0; i < MAX_BPF_REG; i++) if (regs[i].type == PTR_TO_PACKET && regs[i].id == dst_reg->id) - regs[i].range = dst_reg->off; + /* keep the maximum range already checked */ + regs[i].range = max(regs[i].range, dst_reg->off); for (i = 0; i < MAX_BPF_STACK; i += BPF_REG_SIZE) { if (state->stack_slot_type[i] != STACK_SPILL) continue; reg = &state->spilled_regs[i / BPF_REG_SIZE]; if (reg->type == PTR_TO_PACKET && reg->id == dst_reg->id) - reg->range = dst_reg->off; + reg->range = max(reg->range, dst_reg->off); } } diff --git a/tools/testing/selftests/bpf/test_verifier.c b/tools/testing/selftests/bpf/test_verifier.c index d1555e4240c0..7d761d4cc759 100644 --- a/tools/testing/selftests/bpf/test_verifier.c +++ b/tools/testing/selftests/bpf/test_verifier.c @@ -3417,6 +3417,26 @@ static struct bpf_test tests[] = { .result = ACCEPT, .prog_type = BPF_PROG_TYPE_LWT_XMIT, }, + { + "overlapping checks for direct packet access", + .insns = { + BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, + offsetof(struct __sk_buff, data)), + BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, + offsetof(struct __sk_buff, data_end)), + BPF_MOV64_REG(BPF_REG_0, BPF_REG_2), + BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 8), + BPF_JMP_REG(BPF_JGT, BPF_REG_0, BPF_REG_3, 4), + BPF_MOV64_REG(BPF_REG_1, BPF_REG_2), + BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 6), + BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_3, 1), + BPF_LDX_MEM(BPF_H, BPF_REG_0, BPF_REG_2, 6), + BPF_MOV64_IMM(BPF_REG_0, 0), + BPF_EXIT_INSN(), + }, + .result = ACCEPT, + .prog_type = BPF_PROG_TYPE_LWT_XMIT, + }, { "invalid access of tc_classid for LWT_IN", .insns = { -- cgit v1.2.3 From 7b09cc5a9debc86c903c2eff8f8a1fdef773c649 Mon Sep 17 00:00:00 2001 From: Pavel Tatashin Date: Wed, 22 Mar 2017 16:24:17 -0400 Subject: sched/clock: Fix broken stable to unstable transfer When it is determined that the clock is actually unstable, and we switch from stable to unstable, the __clear_sched_clock_stable() function is eventually called. In this function we set gtod_offset so the following holds true: sched_clock() + raw_offset == ktime_get_ns() + gtod_offset But instead of getting the latest timestamps, we use the last values from scd, so instead of sched_clock() we use scd->tick_raw, and instead of ktime_get_ns() we use scd->tick_gtod. However, later, when we use gtod_offset sched_clock_local() we do not add it to scd->tick_gtod to calculate the correct clock value when we determine the boundaries for min/max clocks. This can result in tick granularity sched_clock() values, so fix it. Signed-off-by: Pavel Tatashin Signed-off-by: Peter Zijlstra (Intel) Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: hpa@zytor.com Fixes: 5680d8094ffa ("sched/clock: Provide better clock continuity") Link: http://lkml.kernel.org/r/1490214265-899964-2-git-send-email-pasha.tatashin@oracle.com Signed-off-by: Ingo Molnar --- kernel/sched/clock.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) (limited to 'kernel') diff --git a/kernel/sched/clock.c b/kernel/sched/clock.c index 24a3e01bf8cb..00a45c45beca 100644 --- a/kernel/sched/clock.c +++ b/kernel/sched/clock.c @@ -221,7 +221,7 @@ static inline u64 wrap_max(u64 x, u64 y) */ static u64 sched_clock_local(struct sched_clock_data *scd) { - u64 now, clock, old_clock, min_clock, max_clock; + u64 now, clock, old_clock, min_clock, max_clock, gtod; s64 delta; again: @@ -238,9 +238,10 @@ again: * scd->tick_gtod + TICK_NSEC); */ - clock = scd->tick_gtod + __gtod_offset + delta; - min_clock = wrap_max(scd->tick_gtod, old_clock); - max_clock = wrap_max(old_clock, scd->tick_gtod + TICK_NSEC); + gtod = scd->tick_gtod + __gtod_offset; + clock = gtod + delta; + min_clock = wrap_max(gtod, old_clock); + max_clock = wrap_max(old_clock, gtod + TICK_NSEC); clock = wrap_max(clock, min_clock); clock = wrap_min(clock, max_clock); -- cgit v1.2.3 From ab6434a1377a768a1e6d3e6cf819eb21724a99c2 Mon Sep 17 00:00:00 2001 From: Paul Moore Date: Mon, 27 Mar 2017 14:30:06 -0400 Subject: audit: move audit_signal_info() into kernel/auditsc.c Commit 5b52330bbfe6 ("audit: fix auditd/kernel connection state tracking") made inlining audit_signal_info() a bit pointless as it was always calling into auditd_test_task() so let's remove the inline function in kernel/audit.h and convert __audit_signal_info() in kernel/auditsc.c into audit_signal_info(). Reviewed-by: Richard Guy Briggs Signed-off-by: Paul Moore --- kernel/audit.h | 8 +------- kernel/auditsc.c | 25 +++++++++++++------------ 2 files changed, 14 insertions(+), 19 deletions(-) (limited to 'kernel') diff --git a/kernel/audit.h b/kernel/audit.h index 0f1cf6d1878a..0d87f8ab8778 100644 --- a/kernel/audit.h +++ b/kernel/audit.h @@ -333,13 +333,7 @@ extern u32 audit_sig_sid; extern int audit_filter(int msgtype, unsigned int listtype); #ifdef CONFIG_AUDITSYSCALL -extern int __audit_signal_info(int sig, struct task_struct *t); -static inline int audit_signal_info(int sig, struct task_struct *t) -{ - if (auditd_test_task(t) || (audit_signals && !audit_dummy_context())) - return __audit_signal_info(sig, t); - return 0; -} +extern int audit_signal_info(int sig, struct task_struct *t); extern void audit_filter_inodes(struct task_struct *, struct audit_context *); extern struct list_head *audit_killed_trees(void); #else diff --git a/kernel/auditsc.c b/kernel/auditsc.c index e59ffc7fc522..1c2333155893 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -2249,26 +2249,27 @@ void __audit_ptrace(struct task_struct *t) * If the audit subsystem is being terminated, record the task (pid) * and uid that is doing that. */ -int __audit_signal_info(int sig, struct task_struct *t) +int audit_signal_info(int sig, struct task_struct *t) { struct audit_aux_data_pids *axp; struct task_struct *tsk = current; struct audit_context *ctx = tsk->audit_context; kuid_t uid = current_uid(), t_uid = task_uid(t); - if (auditd_test_task(t)) { - if (sig == SIGTERM || sig == SIGHUP || sig == SIGUSR1 || sig == SIGUSR2) { - audit_sig_pid = task_tgid_nr(tsk); - if (uid_valid(tsk->loginuid)) - audit_sig_uid = tsk->loginuid; - else - audit_sig_uid = uid; - security_task_getsecid(tsk, &audit_sig_sid); - } - if (!audit_signals || audit_dummy_context()) - return 0; + if (auditd_test_task(t) && + (sig == SIGTERM || sig == SIGHUP || + sig == SIGUSR1 || sig == SIGUSR2)) { + audit_sig_pid = task_tgid_nr(tsk); + if (uid_valid(tsk->loginuid)) + audit_sig_uid = tsk->loginuid; + else + audit_sig_uid = uid; + security_task_getsecid(tsk, &audit_sig_sid); } + if (!audit_signals || audit_dummy_context()) + return 0; + /* optimize the common case by putting first signal recipient directly * in audit_context */ if (!ctx->target_pid) { -- cgit v1.2.3 From fce366a9dd0ddc47e7ce05611c266e8574a45116 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Fri, 31 Mar 2017 02:24:02 +0200 Subject: bpf, verifier: fix alu ops against map_value{, _adj} register types While looking into map_value_adj, I noticed that alu operations directly on the map_value() resp. map_value_adj() register (any alu operation on a map_value() register will turn it into a map_value_adj() typed register) are not sufficiently protected against some of the operations. Two non-exhaustive examples are provided that the verifier needs to reject: i) BPF_AND on r0 (map_value_adj): 0: (bf) r2 = r10 1: (07) r2 += -8 2: (7a) *(u64 *)(r2 +0) = 0 3: (18) r1 = 0xbf842a00 5: (85) call bpf_map_lookup_elem#1 6: (15) if r0 == 0x0 goto pc+2 R0=map_value(ks=8,vs=48,id=0),min_value=0,max_value=0 R10=fp 7: (57) r0 &= 8 8: (7a) *(u64 *)(r0 +0) = 22 R0=map_value_adj(ks=8,vs=48,id=0),min_value=0,max_value=8 R10=fp 9: (95) exit from 6 to 9: R0=inv,min_value=0,max_value=0 R10=fp 9: (95) exit processed 10 insns ii) BPF_ADD in 32 bit mode on r0 (map_value_adj): 0: (bf) r2 = r10 1: (07) r2 += -8 2: (7a) *(u64 *)(r2 +0) = 0 3: (18) r1 = 0xc24eee00 5: (85) call bpf_map_lookup_elem#1 6: (15) if r0 == 0x0 goto pc+2 R0=map_value(ks=8,vs=48,id=0),min_value=0,max_value=0 R10=fp 7: (04) (u32) r0 += (u32) 0 8: (7a) *(u64 *)(r0 +0) = 22 R0=map_value_adj(ks=8,vs=48,id=0),min_value=0,max_value=0 R10=fp 9: (95) exit from 6 to 9: R0=inv,min_value=0,max_value=0 R10=fp 9: (95) exit processed 10 insns Issue is, while min_value / max_value boundaries for the access are adjusted appropriately, we change the pointer value in a way that cannot be sufficiently tracked anymore from its origin. Operations like BPF_{AND,OR,DIV,MUL,etc} on a destination register that is PTR_TO_MAP_VALUE{,_ADJ} was probably unintended, in fact, all the test cases coming with 484611357c19 ("bpf: allow access into map value arrays") perform BPF_ADD only on the destination register that is PTR_TO_MAP_VALUE_ADJ. Only for UNKNOWN_VALUE register types such operations make sense, f.e. with unknown memory content fetched initially from a constant offset from the map value memory into a register. That register is then later tested against lower / upper bounds, so that the verifier can then do the tracking of min_value / max_value, and properly check once that UNKNOWN_VALUE register is added to the destination register with type PTR_TO_MAP_VALUE{,_ADJ}. This is also what the original use-case is solving. Note, tracking on what is being added is done through adjust_reg_min_max_vals() and later access to the map value enforced with these boundaries and the given offset from the insn through check_map_access_adj(). Tests will fail for non-root environment due to prohibited pointer arithmetic, in particular in check_alu_op(), we bail out on the is_pointer_value() check on the dst_reg (which is false in root case as we allow for pointer arithmetic via env->allow_ptr_leaks). Similarly to PTR_TO_PACKET, one way to fix it is to restrict the allowed operations on PTR_TO_MAP_VALUE{,_ADJ} registers to 64 bit mode BPF_ADD. The test_verifier suite runs fine after the patch and it also rejects mentioned test cases. Fixes: 484611357c19 ("bpf: allow access into map value arrays") Signed-off-by: Daniel Borkmann Reviewed-by: Josef Bacik Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller --- kernel/bpf/verifier.c | 1 + 1 file changed, 1 insertion(+) (limited to 'kernel') diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index 5e6202e62265..86deddecff25 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -1925,6 +1925,7 @@ static int check_alu_op(struct bpf_verifier_env *env, struct bpf_insn *insn) * register as unknown. */ if (env->allow_ptr_leaks && + BPF_CLASS(insn->code) == BPF_ALU64 && opcode == BPF_ADD && (dst_reg->type == PTR_TO_MAP_VALUE || dst_reg->type == PTR_TO_MAP_VALUE_ADJ)) dst_reg->type = PTR_TO_MAP_VALUE_ADJ; -- cgit v1.2.3 From 79adffcd6489ef43bda2dfded3d637d7fb4fac80 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Fri, 31 Mar 2017 02:24:03 +0200 Subject: bpf, verifier: fix rejection of unaligned access checks for map_value_adj Currently, the verifier doesn't reject unaligned access for map_value_adj register types. Commit 484611357c19 ("bpf: allow access into map value arrays") added logic to check_ptr_alignment() extending it from PTR_TO_PACKET to also PTR_TO_MAP_VALUE_ADJ, but for PTR_TO_MAP_VALUE_ADJ no enforcement is in place, because reg->id for PTR_TO_MAP_VALUE_ADJ reg types is never non-zero, meaning, we can cause BPF_H/_W/_DW-based unaligned access for architectures not supporting efficient unaligned access, and thus worst case could raise exceptions on some archs that are unable to correct the unaligned access or perform a different memory access to the actual requested one and such. i) Unaligned load with !CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS on r0 (map_value_adj): 0: (bf) r2 = r10 1: (07) r2 += -8 2: (7a) *(u64 *)(r2 +0) = 0 3: (18) r1 = 0x42533a00 5: (85) call bpf_map_lookup_elem#1 6: (15) if r0 == 0x0 goto pc+11 R0=map_value(ks=8,vs=48,id=0),min_value=0,max_value=0 R10=fp 7: (61) r1 = *(u32 *)(r0 +0) 8: (35) if r1 >= 0xb goto pc+9 R0=map_value(ks=8,vs=48,id=0),min_value=0,max_value=0 R1=inv,min_value=0,max_value=10 R10=fp 9: (07) r0 += 3 10: (79) r7 = *(u64 *)(r0 +0) R0=map_value_adj(ks=8,vs=48,id=0),min_value=3,max_value=3 R1=inv,min_value=0,max_value=10 R10=fp 11: (79) r7 = *(u64 *)(r0 +2) R0=map_value_adj(ks=8,vs=48,id=0),min_value=3,max_value=3 R1=inv,min_value=0,max_value=10 R7=inv R10=fp [...] ii) Unaligned store with !CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS on r0 (map_value_adj): 0: (bf) r2 = r10 1: (07) r2 += -8 2: (7a) *(u64 *)(r2 +0) = 0 3: (18) r1 = 0x4df16a00 5: (85) call bpf_map_lookup_elem#1 6: (15) if r0 == 0x0 goto pc+19 R0=map_value(ks=8,vs=48,id=0),min_value=0,max_value=0 R10=fp 7: (07) r0 += 3 8: (7a) *(u64 *)(r0 +0) = 42 R0=map_value_adj(ks=8,vs=48,id=0),min_value=3,max_value=3 R10=fp 9: (7a) *(u64 *)(r0 +2) = 43 R0=map_value_adj(ks=8,vs=48,id=0),min_value=3,max_value=3 R10=fp 10: (7a) *(u64 *)(r0 -2) = 44 R0=map_value_adj(ks=8,vs=48,id=0),min_value=3,max_value=3 R10=fp [...] For the PTR_TO_PACKET type, reg->id is initially zero when skb->data was fetched, it later receives a reg->id from env->id_gen generator once another register with UNKNOWN_VALUE type was added to it via check_packet_ptr_add(). The purpose of this reg->id is twofold: i) it is used in find_good_pkt_pointers() for setting the allowed access range for regs with PTR_TO_PACKET of same id once verifier matched on data/data_end tests, and ii) for check_ptr_alignment() to determine that when not having efficient unaligned access and register with UNKNOWN_VALUE was added to PTR_TO_PACKET, that we're only allowed to access the content bytewise due to unknown unalignment. reg->id was never intended for PTR_TO_MAP_VALUE{,_ADJ} types and thus is always zero, the only marking is in PTR_TO_MAP_VALUE_OR_NULL that was added after 484611357c19 via 57a09bf0a416 ("bpf: Detect identical PTR_TO_MAP_VALUE_OR_NULL registers"). Above tests will fail for non-root environment due to prohibited pointer arithmetic. The fix splits register-type specific checks into their own helper instead of keeping them combined, so we don't run into a similar issue in future once we extend check_ptr_alignment() further and forget to add reg->type checks for some of the checks. Fixes: 484611357c19 ("bpf: allow access into map value arrays") Signed-off-by: Daniel Borkmann Reviewed-by: Josef Bacik Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller --- kernel/bpf/verifier.c | 58 +++++++++++++++++++++++++++++++++------------------ 1 file changed, 38 insertions(+), 20 deletions(-) (limited to 'kernel') diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index 86deddecff25..a834068a400e 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -765,38 +765,56 @@ static bool is_pointer_value(struct bpf_verifier_env *env, int regno) } } -static int check_ptr_alignment(struct bpf_verifier_env *env, - struct bpf_reg_state *reg, int off, int size) +static int check_pkt_ptr_alignment(const struct bpf_reg_state *reg, + int off, int size) { - if (reg->type != PTR_TO_PACKET && reg->type != PTR_TO_MAP_VALUE_ADJ) { - if (off % size != 0) { - verbose("misaligned access off %d size %d\n", - off, size); - return -EACCES; - } else { - return 0; - } - } - - if (IS_ENABLED(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS)) - /* misaligned access to packet is ok on x86,arm,arm64 */ - return 0; - if (reg->id && size != 1) { - verbose("Unknown packet alignment. Only byte-sized access allowed\n"); + verbose("Unknown alignment. Only byte-sized access allowed in packet access.\n"); return -EACCES; } /* skb->data is NET_IP_ALIGN-ed */ - if (reg->type == PTR_TO_PACKET && - (NET_IP_ALIGN + reg->off + off) % size != 0) { + if ((NET_IP_ALIGN + reg->off + off) % size != 0) { verbose("misaligned packet access off %d+%d+%d size %d\n", NET_IP_ALIGN, reg->off, off, size); return -EACCES; } + return 0; } +static int check_val_ptr_alignment(const struct bpf_reg_state *reg, + int size) +{ + if (size != 1) { + verbose("Unknown alignment. Only byte-sized access allowed in value access.\n"); + return -EACCES; + } + + return 0; +} + +static int check_ptr_alignment(const struct bpf_reg_state *reg, + int off, int size) +{ + switch (reg->type) { + case PTR_TO_PACKET: + return IS_ENABLED(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS) ? 0 : + check_pkt_ptr_alignment(reg, off, size); + case PTR_TO_MAP_VALUE_ADJ: + return IS_ENABLED(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS) ? 0 : + check_val_ptr_alignment(reg, size); + default: + if (off % size != 0) { + verbose("misaligned access off %d size %d\n", + off, size); + return -EACCES; + } + + return 0; + } +} + /* check whether memory at (regno + off) is accessible for t = (read | write) * if t==write, value_regno is a register which value is stored into memory * if t==read, value_regno is a register which will receive the value from memory @@ -818,7 +836,7 @@ static int check_mem_access(struct bpf_verifier_env *env, u32 regno, int off, if (size < 0) return size; - err = check_ptr_alignment(env, reg, off, size); + err = check_ptr_alignment(reg, off, size); if (err) return err; -- cgit v1.2.3 From 62277de758b155dc04b78f195a1cb5208c37b2df Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Fri, 17 Jun 2016 17:33:59 +0000 Subject: ring-buffer: Fix return value check in test_ringbuffer() In case of error, the function kthread_run() returns ERR_PTR() and never returns NULL. The NULL test in the return value check should be replaced with IS_ERR(). Link: http://lkml.kernel.org/r/1466184839-14927-1-git-send-email-weiyj_lk@163.com Cc: stable@vger.kernel.org Fixes: 6c43e554a ("ring-buffer: Add ring buffer startup selftest") Signed-off-by: Wei Yongjun Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/ring_buffer.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index 96fc3c043ad6..54e7a90db848 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -4826,9 +4826,9 @@ static __init int test_ringbuffer(void) rb_data[cpu].cnt = cpu; rb_threads[cpu] = kthread_create(rb_test, &rb_data[cpu], "rbtester/%d", cpu); - if (WARN_ON(!rb_threads[cpu])) { + if (WARN_ON(IS_ERR(rb_threads[cpu]))) { pr_cont("FAILED\n"); - ret = -1; + ret = PTR_ERR(rb_threads[cpu]); goto out_free; } @@ -4838,9 +4838,9 @@ static __init int test_ringbuffer(void) /* Now create the rb hammer! */ rb_hammer = kthread_run(rb_hammer_test, NULL, "rbhammer"); - if (WARN_ON(!rb_hammer)) { + if (WARN_ON(IS_ERR(rb_hammer))) { pr_cont("FAILED\n"); - ret = -1; + ret = PTR_ERR(rb_hammer); goto out_free; } -- cgit v1.2.3 From 5380e5644afbba9e3d229c36771134976f05c91e Mon Sep 17 00:00:00 2001 From: Liping Zhang Date: Fri, 7 Apr 2017 23:51:06 +0800 Subject: sysctl: don't print negative flag for proc_douintvec I saw some very confusing sysctl output on my system: # cat /proc/sys/net/core/xfrm_aevent_rseqth -2 # cat /proc/sys/net/core/xfrm_aevent_etime -10 # cat /proc/sys/net/ipv4/tcp_notsent_lowat -4294967295 Because we forget to set the *negp flag in proc_douintvec, so it will become a garbage value. Since the value related to proc_douintvec is always an unsigned integer, so we can set *negp to false explictily to fix this issue. Fixes: e7d316a02f68 ("sysctl: handle error writing UINT_MAX to u32 fields") Signed-off-by: Liping Zhang Cc: Subash Abhinov Kasiviswanathan Cc: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/sysctl.c | 1 + 1 file changed, 1 insertion(+) (limited to 'kernel') diff --git a/kernel/sysctl.c b/kernel/sysctl.c index acf0a5a06da7..8cca4c7bb21f 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -2136,6 +2136,7 @@ static int do_proc_douintvec_conv(bool *negp, unsigned long *lvalp, *valp = *lvalp; } else { unsigned int val = *valp; + *negp = false; *lvalp = (unsigned long)val; } return 0; -- cgit v1.2.3 From 5402e97af667e35e54177af8f6575518bf251d51 Mon Sep 17 00:00:00 2001 From: "bsegall@google.com" Date: Fri, 7 Apr 2017 16:04:51 -0700 Subject: ptrace: fix PTRACE_LISTEN race corrupting task->state In PT_SEIZED + LISTEN mode STOP/CONT signals cause a wakeup against __TASK_TRACED. If this races with the ptrace_unfreeze_traced at the end of a PTRACE_LISTEN, this can wake the task /after/ the check against __TASK_TRACED, but before the reset of state to TASK_TRACED. This causes it to instead clobber TASK_WAKING, allowing a subsequent wakeup against TRACED while the task is still on the rq wake_list, corrupting it. Oleg said: "The kernel can crash or this can lead to other hard-to-debug problems. In short, "task->state = TASK_TRACED" in ptrace_unfreeze_traced() assumes that nobody else can wake it up, but PTRACE_LISTEN breaks the contract. Obviusly it is very wrong to manipulate task->state if this task is already running, or WAKING, or it sleeps again" [akpm@linux-foundation.org: coding-style fixes] Fixes: 9899d11f ("ptrace: ensure arch_ptrace/ptrace_request can never race with SIGKILL") Link: http://lkml.kernel.org/r/xm26y3vfhmkp.fsf_-_@bsegall-linux.mtv.corp.google.com Signed-off-by: Ben Segall Acked-by: Oleg Nesterov Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/ptrace.c | 14 ++++++++++---- 1 file changed, 10 insertions(+), 4 deletions(-) (limited to 'kernel') diff --git a/kernel/ptrace.c b/kernel/ptrace.c index 0af928712174..266ddcc1d8bb 100644 --- a/kernel/ptrace.c +++ b/kernel/ptrace.c @@ -184,11 +184,17 @@ static void ptrace_unfreeze_traced(struct task_struct *task) WARN_ON(!task->ptrace || task->parent != current); + /* + * PTRACE_LISTEN can allow ptrace_trap_notify to wake us up remotely. + * Recheck state under the lock to close this race. + */ spin_lock_irq(&task->sighand->siglock); - if (__fatal_signal_pending(task)) - wake_up_state(task, __TASK_TRACED); - else - task->state = TASK_TRACED; + if (task->state == __TASK_TRACED) { + if (__fatal_signal_pending(task)) + wake_up_state(task, __TASK_TRACED); + else + task->state = TASK_TRACED; + } spin_unlock_irq(&task->sighand->siglock); } -- cgit v1.2.3 From 425fffd886bae3d127a08fa6a17f2e31e24ed7ff Mon Sep 17 00:00:00 2001 From: Liping Zhang Date: Fri, 7 Apr 2017 23:51:07 +0800 Subject: sysctl: report EINVAL if value is larger than UINT_MAX for proc_douintvec Currently, inputting the following command will succeed but actually the value will be truncated: # echo 0x12ffffffff > /proc/sys/net/ipv4/tcp_notsent_lowat This is not friendly to the user, so instead, we should report error when the value is larger than UINT_MAX. Fixes: e7d316a02f68 ("sysctl: handle error writing UINT_MAX to u32 fields") Signed-off-by: Liping Zhang Cc: Subash Abhinov Kasiviswanathan Cc: Andrew Morton Cc: Eric W. Biederman Signed-off-by: Linus Torvalds --- kernel/sysctl.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'kernel') diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 8cca4c7bb21f..8c8714fcb53c 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -2133,6 +2133,8 @@ static int do_proc_douintvec_conv(bool *negp, unsigned long *lvalp, if (write) { if (*negp) return -EINVAL; + if (*lvalp > UINT_MAX) + return -EINVAL; *valp = *lvalp; } else { unsigned int val = *valp; -- cgit v1.2.3