From 8cdb878dcb359fd1137e9abdee9322f5e9bcfdf8 Mon Sep 17 00:00:00 2001 From: Christopher Yeoh Date: Thu, 2 Feb 2012 11:34:09 +1030 Subject: Fix race in process_vm_rw_core This fixes the race in process_vm_core found by Oleg (see http://article.gmane.org/gmane.linux.kernel/1235667/ for details). This has been updated since I last sent it as the creation of the new mm_access() function did almost exactly the same thing as parts of the previous version of this patch did. In order to use mm_access() even when /proc isn't enabled, we move it to kernel/fork.c where other related process mm access functions already are. Signed-off-by: Chris Yeoh Signed-off-by: Linus Torvalds --- kernel/fork.c | 20 ++++++++++++++++++++ 1 file changed, 20 insertions(+) (limited to 'kernel/fork.c') diff --git a/kernel/fork.c b/kernel/fork.c index 051f090d40c1..1b2ef3c23ae4 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -647,6 +647,26 @@ struct mm_struct *get_task_mm(struct task_struct *task) } EXPORT_SYMBOL_GPL(get_task_mm); +struct mm_struct *mm_access(struct task_struct *task, unsigned int mode) +{ + struct mm_struct *mm; + int err; + + err = mutex_lock_killable(&task->signal->cred_guard_mutex); + if (err) + return ERR_PTR(err); + + mm = get_task_mm(task); + if (mm && mm != current->mm && + !ptrace_may_access(task, mode)) { + mmput(mm); + mm = ERR_PTR(-EACCES); + } + mutex_unlock(&task->signal->cred_guard_mutex); + + return mm; +} + /* Please note the differences between mmput and mm_release. * mmput is called whenever we stop holding onto a mm_struct, * error success whatever. -- cgit v1.2.3 From 11a3122f6cf2d988a77eb8883d0fc49cd013a6d5 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 7 Feb 2012 07:51:30 +0100 Subject: block: strip out locking optimization in put_io_context() put_io_context() performed a complex trylock dancing to avoid deferring ioc release to workqueue. It was also broken on UP because trylock was always assumed to succeed which resulted in unbalanced preemption count. While there are ways to fix the UP breakage, even the most pathological microbench (forced ioc allocation and tight fork/exit loop) fails to show any appreciable performance benefit of the optimization. Strip it out. If there turns out to be workloads which are affected by this change, simpler optimization from the discussion thread can be applied later. Signed-off-by: Tejun Heo LKML-Reference: <1328514611.21268.66.camel@sli10-conroe> Signed-off-by: Jens Axboe --- block/blk-cgroup.c | 2 +- block/blk-core.c | 2 +- block/blk-ioc.c | 92 ++++++----------------------------------------- block/cfq-iosched.c | 2 +- fs/ioprio.c | 2 +- include/linux/blkdev.h | 3 -- include/linux/iocontext.h | 5 ++- kernel/fork.c | 2 +- 8 files changed, 18 insertions(+), 92 deletions(-) (limited to 'kernel/fork.c') diff --git a/block/blk-cgroup.c b/block/blk-cgroup.c index fa8f26309444..75642a352a8f 100644 --- a/block/blk-cgroup.c +++ b/block/blk-cgroup.c @@ -1659,7 +1659,7 @@ static void blkiocg_attach(struct cgroup_subsys *ss, struct cgroup *cgrp, ioc = get_task_io_context(task, GFP_ATOMIC, NUMA_NO_NODE); if (ioc) { ioc_cgroup_changed(ioc); - put_io_context(ioc, NULL); + put_io_context(ioc); } } } diff --git a/block/blk-core.c b/block/blk-core.c index 636702575118..532b3a21b383 100644 --- a/block/blk-core.c +++ b/block/blk-core.c @@ -642,7 +642,7 @@ static inline void blk_free_request(struct request_queue *q, struct request *rq) if (rq->cmd_flags & REQ_ELVPRIV) { elv_put_request(q, rq); if (rq->elv.icq) - put_io_context(rq->elv.icq->ioc, q); + put_io_context(rq->elv.icq->ioc); } mempool_free(rq, q->rq.rq_pool); diff --git a/block/blk-ioc.c b/block/blk-ioc.c index 7490b6da2453..9884fd7427fe 100644 --- a/block/blk-ioc.c +++ b/block/blk-ioc.c @@ -29,21 +29,6 @@ void get_io_context(struct io_context *ioc) } EXPORT_SYMBOL(get_io_context); -/* - * Releasing ioc may nest into another put_io_context() leading to nested - * fast path release. As the ioc's can't be the same, this is okay but - * makes lockdep whine. Keep track of nesting and use it as subclass. - */ -#ifdef CONFIG_LOCKDEP -#define ioc_release_depth(q) ((q) ? (q)->ioc_release_depth : 0) -#define ioc_release_depth_inc(q) (q)->ioc_release_depth++ -#define ioc_release_depth_dec(q) (q)->ioc_release_depth-- -#else -#define ioc_release_depth(q) 0 -#define ioc_release_depth_inc(q) do { } while (0) -#define ioc_release_depth_dec(q) do { } while (0) -#endif - static void icq_free_icq_rcu(struct rcu_head *head) { struct io_cq *icq = container_of(head, struct io_cq, __rcu_head); @@ -75,11 +60,8 @@ static void ioc_exit_icq(struct io_cq *icq) if (rcu_dereference_raw(ioc->icq_hint) == icq) rcu_assign_pointer(ioc->icq_hint, NULL); - if (et->ops.elevator_exit_icq_fn) { - ioc_release_depth_inc(q); + if (et->ops.elevator_exit_icq_fn) et->ops.elevator_exit_icq_fn(icq); - ioc_release_depth_dec(q); - } /* * @icq->q might have gone away by the time RCU callback runs @@ -149,81 +131,29 @@ static void ioc_release_fn(struct work_struct *work) /** * put_io_context - put a reference of io_context * @ioc: io_context to put - * @locked_q: request_queue the caller is holding queue_lock of (hint) * * Decrement reference count of @ioc and release it if the count reaches - * zero. If the caller is holding queue_lock of a queue, it can indicate - * that with @locked_q. This is an optimization hint and the caller is - * allowed to pass in %NULL even when it's holding a queue_lock. + * zero. */ -void put_io_context(struct io_context *ioc, struct request_queue *locked_q) +void put_io_context(struct io_context *ioc) { - struct request_queue *last_q = locked_q; unsigned long flags; if (ioc == NULL) return; BUG_ON(atomic_long_read(&ioc->refcount) <= 0); - if (locked_q) - lockdep_assert_held(locked_q->queue_lock); - - if (!atomic_long_dec_and_test(&ioc->refcount)) - return; /* - * Destroy @ioc. This is a bit messy because icq's are chained - * from both ioc and queue, and ioc->lock nests inside queue_lock. - * The inner ioc->lock should be held to walk our icq_list and then - * for each icq the outer matching queue_lock should be grabbed. - * ie. We need to do reverse-order double lock dancing. - * - * Another twist is that we are often called with one of the - * matching queue_locks held as indicated by @locked_q, which - * prevents performing double-lock dance for other queues. - * - * So, we do it in two stages. The fast path uses the queue_lock - * the caller is holding and, if other queues need to be accessed, - * uses trylock to avoid introducing locking dependency. This can - * handle most cases, especially if @ioc was performing IO on only - * single device. - * - * If trylock doesn't cut it, we defer to @ioc->release_work which - * can do all the double-locking dancing. + * Releasing ioc requires reverse order double locking and we may + * already be holding a queue_lock. Do it asynchronously from wq. */ - spin_lock_irqsave_nested(&ioc->lock, flags, - ioc_release_depth(locked_q)); - - while (!hlist_empty(&ioc->icq_list)) { - struct io_cq *icq = hlist_entry(ioc->icq_list.first, - struct io_cq, ioc_node); - struct request_queue *this_q = icq->q; - - if (this_q != last_q) { - if (last_q && last_q != locked_q) - spin_unlock(last_q->queue_lock); - last_q = NULL; - - /* spin_trylock() always successes in UP case */ - if (this_q != locked_q && - !spin_trylock(this_q->queue_lock)) - break; - last_q = this_q; - continue; - } - ioc_exit_icq(icq); + if (atomic_long_dec_and_test(&ioc->refcount)) { + spin_lock_irqsave(&ioc->lock, flags); + if (!hlist_empty(&ioc->icq_list)) + schedule_work(&ioc->release_work); + spin_unlock_irqrestore(&ioc->lock, flags); } - - if (last_q && last_q != locked_q) - spin_unlock(last_q->queue_lock); - - spin_unlock_irqrestore(&ioc->lock, flags); - - /* if no icq is left, we're done; otherwise, kick release_work */ - if (hlist_empty(&ioc->icq_list)) - kmem_cache_free(iocontext_cachep, ioc); - else - schedule_work(&ioc->release_work); } EXPORT_SYMBOL(put_io_context); @@ -238,7 +168,7 @@ void exit_io_context(struct task_struct *task) task_unlock(task); atomic_dec(&ioc->nr_tasks); - put_io_context(ioc, NULL); + put_io_context(ioc); } /** diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c index da21c24dbed3..5684df6848bc 100644 --- a/block/cfq-iosched.c +++ b/block/cfq-iosched.c @@ -1794,7 +1794,7 @@ __cfq_slice_expired(struct cfq_data *cfqd, struct cfq_queue *cfqq, cfqd->active_queue = NULL; if (cfqd->active_cic) { - put_io_context(cfqd->active_cic->icq.ioc, cfqd->queue); + put_io_context(cfqd->active_cic->icq.ioc); cfqd->active_cic = NULL; } } diff --git a/fs/ioprio.c b/fs/ioprio.c index f84b380d65e5..0f1b9515213b 100644 --- a/fs/ioprio.c +++ b/fs/ioprio.c @@ -51,7 +51,7 @@ int set_task_ioprio(struct task_struct *task, int ioprio) ioc = get_task_io_context(task, GFP_ATOMIC, NUMA_NO_NODE); if (ioc) { ioc_ioprio_changed(ioc, ioprio); - put_io_context(ioc, NULL); + put_io_context(ioc); } return err; diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 6c6a1f008065..606cf339bb56 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -399,9 +399,6 @@ struct request_queue { /* Throttle data */ struct throtl_data *td; #endif -#ifdef CONFIG_LOCKDEP - int ioc_release_depth; -#endif }; #define QUEUE_FLAG_QUEUED 1 /* uses generic tag queueing */ diff --git a/include/linux/iocontext.h b/include/linux/iocontext.h index 7e1371c4bccf..119773eebe31 100644 --- a/include/linux/iocontext.h +++ b/include/linux/iocontext.h @@ -133,7 +133,7 @@ static inline struct io_context *ioc_task_link(struct io_context *ioc) struct task_struct; #ifdef CONFIG_BLOCK -void put_io_context(struct io_context *ioc, struct request_queue *locked_q); +void put_io_context(struct io_context *ioc); void exit_io_context(struct task_struct *task); struct io_context *get_task_io_context(struct task_struct *task, gfp_t gfp_flags, int node); @@ -141,8 +141,7 @@ void ioc_ioprio_changed(struct io_context *ioc, int ioprio); void ioc_cgroup_changed(struct io_context *ioc); #else struct io_context; -static inline void put_io_context(struct io_context *ioc, - struct request_queue *locked_q) { } +static inline void put_io_context(struct io_context *ioc) { } static inline void exit_io_context(struct task_struct *task) { } #endif diff --git a/kernel/fork.c b/kernel/fork.c index 051f090d40c1..c574aefa8d1b 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -890,7 +890,7 @@ static int copy_io(unsigned long clone_flags, struct task_struct *tsk) return -ENOMEM; new_ioc->ioprio = ioc->ioprio; - put_io_context(new_ioc, NULL); + put_io_context(new_ioc); } #endif return 0; -- cgit v1.2.3 From d80e731ecab420ddcb79ee9d0ac427acbc187b4b Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Fri, 24 Feb 2012 20:07:11 +0100 Subject: epoll: introduce POLLFREE to flush ->signalfd_wqh before kfree() This patch is intentionally incomplete to simplify the review. It ignores ep_unregister_pollwait() which plays with the same wqh. See the next change. epoll assumes that the EPOLL_CTL_ADD'ed file controls everything f_op->poll() needs. In particular it assumes that the wait queue can't go away until eventpoll_release(). This is not true in case of signalfd, the task which does EPOLL_CTL_ADD uses its ->sighand which is not connected to the file. This patch adds the special event, POLLFREE, currently only for epoll. It expects that init_poll_funcptr()'ed hook should do the necessary cleanup. Perhaps it should be defined as EPOLLFREE in eventpoll. __cleanup_sighand() is changed to do wake_up_poll(POLLFREE) if ->signalfd_wqh is not empty, we add the new signalfd_cleanup() helper. ep_poll_callback(POLLFREE) simply does list_del_init(task_list). This make this poll entry inconsistent, but we don't care. If you share epoll fd which contains our sigfd with another process you should blame yourself. signalfd is "really special". I simply do not know how we can define the "right" semantics if it used with epoll. The main problem is, epoll calls signalfd_poll() once to establish the connection with the wait queue, after that signalfd_poll(NULL) returns the different/inconsistent results depending on who does EPOLL_CTL_MOD/signalfd_read/etc. IOW: apart from sigmask, signalfd has nothing to do with the file, it works with the current thread. In short: this patch is the hack which tries to fix the symptoms. It also assumes that nobody can take tasklist_lock under epoll locks, this seems to be true. Note: - we do not have wake_up_all_poll() but wake_up_poll() is fine, poll/epoll doesn't use WQ_FLAG_EXCLUSIVE. - signalfd_cleanup() uses POLLHUP along with POLLFREE, we need a couple of simple changes in eventpoll.c to make sure it can't be "lost". Reported-by: Maxime Bizon Cc: Signed-off-by: Oleg Nesterov Signed-off-by: Linus Torvalds --- fs/eventpoll.c | 4 ++++ fs/signalfd.c | 11 +++++++++++ include/asm-generic/poll.h | 2 ++ include/linux/signalfd.h | 5 ++++- kernel/fork.c | 5 ++++- 5 files changed, 25 insertions(+), 2 deletions(-) (limited to 'kernel/fork.c') diff --git a/fs/eventpoll.c b/fs/eventpoll.c index aabdfc38cf24..34bbfc6dd8dc 100644 --- a/fs/eventpoll.c +++ b/fs/eventpoll.c @@ -842,6 +842,10 @@ static int ep_poll_callback(wait_queue_t *wait, unsigned mode, int sync, void *k struct epitem *epi = ep_item_from_wait(wait); struct eventpoll *ep = epi->ep; + /* the caller holds eppoll_entry->whead->lock */ + if ((unsigned long)key & POLLFREE) + list_del_init(&wait->task_list); + spin_lock_irqsave(&ep->lock, flags); /* diff --git a/fs/signalfd.c b/fs/signalfd.c index 492465b451dd..79c1eea98a3a 100644 --- a/fs/signalfd.c +++ b/fs/signalfd.c @@ -30,6 +30,17 @@ #include #include +void signalfd_cleanup(struct sighand_struct *sighand) +{ + wait_queue_head_t *wqh = &sighand->signalfd_wqh; + + if (likely(!waitqueue_active(wqh))) + return; + + /* wait_queue_t->func(POLLFREE) should do remove_wait_queue() */ + wake_up_poll(wqh, POLLHUP | POLLFREE); +} + struct signalfd_ctx { sigset_t sigmask; }; diff --git a/include/asm-generic/poll.h b/include/asm-generic/poll.h index 44bce836d350..9ce7f44aebd2 100644 --- a/include/asm-generic/poll.h +++ b/include/asm-generic/poll.h @@ -28,6 +28,8 @@ #define POLLRDHUP 0x2000 #endif +#define POLLFREE 0x4000 /* currently only for epoll */ + struct pollfd { int fd; short events; diff --git a/include/linux/signalfd.h b/include/linux/signalfd.h index 3ff4961da9b5..247399b2979a 100644 --- a/include/linux/signalfd.h +++ b/include/linux/signalfd.h @@ -61,13 +61,16 @@ static inline void signalfd_notify(struct task_struct *tsk, int sig) wake_up(&tsk->sighand->signalfd_wqh); } +extern void signalfd_cleanup(struct sighand_struct *sighand); + #else /* CONFIG_SIGNALFD */ static inline void signalfd_notify(struct task_struct *tsk, int sig) { } +static inline void signalfd_cleanup(struct sighand_struct *sighand) { } + #endif /* CONFIG_SIGNALFD */ #endif /* __KERNEL__ */ #endif /* _LINUX_SIGNALFD_H */ - diff --git a/kernel/fork.c b/kernel/fork.c index b77fd559c78e..e2cd3e2a5ae8 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -66,6 +66,7 @@ #include #include #include +#include #include #include @@ -935,8 +936,10 @@ static int copy_sighand(unsigned long clone_flags, struct task_struct *tsk) void __cleanup_sighand(struct sighand_struct *sighand) { - if (atomic_dec_and_test(&sighand->count)) + if (atomic_dec_and_test(&sighand->count)) { + signalfd_cleanup(sighand); kmem_cache_free(sighand_cachep, sighand); + } } -- cgit v1.2.3