From 8c8d2d9670e813d623d8a2cbc881cb57344f4d37 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 17 Mar 2017 16:35:23 -0800 Subject: bcache: move closures to lib/ Prep work for bcachefs - being a fork of bcache it also uses closures Signed-off-by: Kent Overstreet Acked-by: Coly Li Reviewed-by: Randy Dunlap --- drivers/md/bcache/Kconfig | 10 +- drivers/md/bcache/Makefile | 4 +- drivers/md/bcache/bcache.h | 2 +- drivers/md/bcache/closure.c | 207 ------------------------ drivers/md/bcache/closure.h | 378 -------------------------------------------- drivers/md/bcache/super.c | 1 - drivers/md/bcache/util.h | 3 +- include/linux/closure.h | 377 +++++++++++++++++++++++++++++++++++++++++++ lib/Kconfig | 3 + lib/Kconfig.debug | 9 ++ lib/Makefile | 2 + lib/closure.c | 204 ++++++++++++++++++++++++ 12 files changed, 600 insertions(+), 600 deletions(-) delete mode 100644 drivers/md/bcache/closure.c delete mode 100644 drivers/md/bcache/closure.h create mode 100644 include/linux/closure.h create mode 100644 lib/closure.c diff --git a/drivers/md/bcache/Kconfig b/drivers/md/bcache/Kconfig index 529c9d04e9a4..b2d10063d35f 100644 --- a/drivers/md/bcache/Kconfig +++ b/drivers/md/bcache/Kconfig @@ -4,6 +4,7 @@ config BCACHE tristate "Block device as cache" select BLOCK_HOLDER_DEPRECATED if SYSFS select CRC64 + select CLOSURES help Allows a block device to be used as cache for other devices; uses a btree for indexing and the layout is optimized for SSDs. @@ -19,15 +20,6 @@ config BCACHE_DEBUG Enables extra debugging tools, allows expensive runtime checks to be turned on. -config BCACHE_CLOSURES_DEBUG - bool "Debug closures" - depends on BCACHE - select DEBUG_FS - help - Keeps all active closures in a linked list and provides a debugfs - interface to list them, which makes it possible to see asynchronous - operations that get stuck. - config BCACHE_ASYNC_REGISTRATION bool "Asynchronous device registration" depends on BCACHE diff --git a/drivers/md/bcache/Makefile b/drivers/md/bcache/Makefile index 5b87e59676b8..054e8a33a7ab 100644 --- a/drivers/md/bcache/Makefile +++ b/drivers/md/bcache/Makefile @@ -2,6 +2,6 @@ obj-$(CONFIG_BCACHE) += bcache.o -bcache-y := alloc.o bset.o btree.o closure.o debug.o extents.o\ - io.o journal.o movinggc.o request.o stats.o super.o sysfs.o trace.o\ +bcache-y := alloc.o bset.o btree.o debug.o extents.o io.o\ + journal.o movinggc.o request.o stats.o super.o sysfs.o trace.o\ util.o writeback.o features.o diff --git a/drivers/md/bcache/bcache.h b/drivers/md/bcache/bcache.h index 5a79bb3c272f..7c0d00432612 100644 --- a/drivers/md/bcache/bcache.h +++ b/drivers/md/bcache/bcache.h @@ -179,6 +179,7 @@ #define pr_fmt(fmt) "bcache: %s() " fmt, __func__ #include +#include #include #include #include @@ -192,7 +193,6 @@ #include "bcache_ondisk.h" #include "bset.h" #include "util.h" -#include "closure.h" struct bucket { atomic_t pin; diff --git a/drivers/md/bcache/closure.c b/drivers/md/bcache/closure.c deleted file mode 100644 index d8d9394a6beb..000000000000 --- a/drivers/md/bcache/closure.c +++ /dev/null @@ -1,207 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0 -/* - * Asynchronous refcounty things - * - * Copyright 2010, 2011 Kent Overstreet - * Copyright 2012 Google, Inc. - */ - -#include -#include -#include -#include - -#include "closure.h" - -static inline void closure_put_after_sub(struct closure *cl, int flags) -{ - int r = flags & CLOSURE_REMAINING_MASK; - - BUG_ON(flags & CLOSURE_GUARD_MASK); - BUG_ON(!r && (flags & ~CLOSURE_DESTRUCTOR)); - - if (!r) { - if (cl->fn && !(flags & CLOSURE_DESTRUCTOR)) { - atomic_set(&cl->remaining, - CLOSURE_REMAINING_INITIALIZER); - closure_queue(cl); - } else { - struct closure *parent = cl->parent; - closure_fn *destructor = cl->fn; - - closure_debug_destroy(cl); - - if (destructor) - destructor(cl); - - if (parent) - closure_put(parent); - } - } -} - -/* For clearing flags with the same atomic op as a put */ -void closure_sub(struct closure *cl, int v) -{ - closure_put_after_sub(cl, atomic_sub_return(v, &cl->remaining)); -} - -/* - * closure_put - decrement a closure's refcount - */ -void closure_put(struct closure *cl) -{ - closure_put_after_sub(cl, atomic_dec_return(&cl->remaining)); -} - -/* - * closure_wake_up - wake up all closures on a wait list, without memory barrier - */ -void __closure_wake_up(struct closure_waitlist *wait_list) -{ - struct llist_node *list; - struct closure *cl, *t; - struct llist_node *reverse = NULL; - - list = llist_del_all(&wait_list->list); - - /* We first reverse the list to preserve FIFO ordering and fairness */ - reverse = llist_reverse_order(list); - - /* Then do the wakeups */ - llist_for_each_entry_safe(cl, t, reverse, list) { - closure_set_waiting(cl, 0); - closure_sub(cl, CLOSURE_WAITING + 1); - } -} - -/** - * closure_wait - add a closure to a waitlist - * @waitlist: will own a ref on @cl, which will be released when - * closure_wake_up() is called on @waitlist. - * @cl: closure pointer. - * - */ -bool closure_wait(struct closure_waitlist *waitlist, struct closure *cl) -{ - if (atomic_read(&cl->remaining) & CLOSURE_WAITING) - return false; - - closure_set_waiting(cl, _RET_IP_); - atomic_add(CLOSURE_WAITING + 1, &cl->remaining); - llist_add(&cl->list, &waitlist->list); - - return true; -} - -struct closure_syncer { - struct task_struct *task; - int done; -}; - -static void closure_sync_fn(struct closure *cl) -{ - struct closure_syncer *s = cl->s; - struct task_struct *p; - - rcu_read_lock(); - p = READ_ONCE(s->task); - s->done = 1; - wake_up_process(p); - rcu_read_unlock(); -} - -void __sched __closure_sync(struct closure *cl) -{ - struct closure_syncer s = { .task = current }; - - cl->s = &s; - continue_at(cl, closure_sync_fn, NULL); - - while (1) { - set_current_state(TASK_UNINTERRUPTIBLE); - if (s.done) - break; - schedule(); - } - - __set_current_state(TASK_RUNNING); -} - -#ifdef CONFIG_BCACHE_CLOSURES_DEBUG - -static LIST_HEAD(closure_list); -static DEFINE_SPINLOCK(closure_list_lock); - -void closure_debug_create(struct closure *cl) -{ - unsigned long flags; - - BUG_ON(cl->magic == CLOSURE_MAGIC_ALIVE); - cl->magic = CLOSURE_MAGIC_ALIVE; - - spin_lock_irqsave(&closure_list_lock, flags); - list_add(&cl->all, &closure_list); - spin_unlock_irqrestore(&closure_list_lock, flags); -} - -void closure_debug_destroy(struct closure *cl) -{ - unsigned long flags; - - BUG_ON(cl->magic != CLOSURE_MAGIC_ALIVE); - cl->magic = CLOSURE_MAGIC_DEAD; - - spin_lock_irqsave(&closure_list_lock, flags); - list_del(&cl->all); - spin_unlock_irqrestore(&closure_list_lock, flags); -} - -static struct dentry *closure_debug; - -static int debug_show(struct seq_file *f, void *data) -{ - struct closure *cl; - - spin_lock_irq(&closure_list_lock); - - list_for_each_entry(cl, &closure_list, all) { - int r = atomic_read(&cl->remaining); - - seq_printf(f, "%p: %pS -> %pS p %p r %i ", - cl, (void *) cl->ip, cl->fn, cl->parent, - r & CLOSURE_REMAINING_MASK); - - seq_printf(f, "%s%s\n", - test_bit(WORK_STRUCT_PENDING_BIT, - work_data_bits(&cl->work)) ? "Q" : "", - r & CLOSURE_RUNNING ? "R" : ""); - - if (r & CLOSURE_WAITING) - seq_printf(f, " W %pS\n", - (void *) cl->waiting_on); - - seq_printf(f, "\n"); - } - - spin_unlock_irq(&closure_list_lock); - return 0; -} - -DEFINE_SHOW_ATTRIBUTE(debug); - -void __init closure_debug_init(void) -{ - if (!IS_ERR_OR_NULL(bcache_debug)) - /* - * it is unnecessary to check return value of - * debugfs_create_file(), we should not care - * about this. - */ - closure_debug = debugfs_create_file( - "closures", 0400, bcache_debug, NULL, &debug_fops); -} -#endif - -MODULE_AUTHOR("Kent Overstreet "); -MODULE_LICENSE("GPL"); diff --git a/drivers/md/bcache/closure.h b/drivers/md/bcache/closure.h deleted file mode 100644 index c88cdc4ae4ec..000000000000 --- a/drivers/md/bcache/closure.h +++ /dev/null @@ -1,378 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0 */ -#ifndef _LINUX_CLOSURE_H -#define _LINUX_CLOSURE_H - -#include -#include -#include -#include - -/* - * Closure is perhaps the most overused and abused term in computer science, but - * since I've been unable to come up with anything better you're stuck with it - * again. - * - * What are closures? - * - * They embed a refcount. The basic idea is they count "things that are in - * progress" - in flight bios, some other thread that's doing something else - - * anything you might want to wait on. - * - * The refcount may be manipulated with closure_get() and closure_put(). - * closure_put() is where many of the interesting things happen, when it causes - * the refcount to go to 0. - * - * Closures can be used to wait on things both synchronously and asynchronously, - * and synchronous and asynchronous use can be mixed without restriction. To - * wait synchronously, use closure_sync() - you will sleep until your closure's - * refcount hits 1. - * - * To wait asynchronously, use - * continue_at(cl, next_function, workqueue); - * - * passing it, as you might expect, the function to run when nothing is pending - * and the workqueue to run that function out of. - * - * continue_at() also, critically, requires a 'return' immediately following the - * location where this macro is referenced, to return to the calling function. - * There's good reason for this. - * - * To use safely closures asynchronously, they must always have a refcount while - * they are running owned by the thread that is running them. Otherwise, suppose - * you submit some bios and wish to have a function run when they all complete: - * - * foo_endio(struct bio *bio) - * { - * closure_put(cl); - * } - * - * closure_init(cl); - * - * do_stuff(); - * closure_get(cl); - * bio1->bi_endio = foo_endio; - * bio_submit(bio1); - * - * do_more_stuff(); - * closure_get(cl); - * bio2->bi_endio = foo_endio; - * bio_submit(bio2); - * - * continue_at(cl, complete_some_read, system_wq); - * - * If closure's refcount started at 0, complete_some_read() could run before the - * second bio was submitted - which is almost always not what you want! More - * importantly, it wouldn't be possible to say whether the original thread or - * complete_some_read()'s thread owned the closure - and whatever state it was - * associated with! - * - * So, closure_init() initializes a closure's refcount to 1 - and when a - * closure_fn is run, the refcount will be reset to 1 first. - * - * Then, the rule is - if you got the refcount with closure_get(), release it - * with closure_put() (i.e, in a bio->bi_endio function). If you have a refcount - * on a closure because you called closure_init() or you were run out of a - * closure - _always_ use continue_at(). Doing so consistently will help - * eliminate an entire class of particularly pernicious races. - * - * Lastly, you might have a wait list dedicated to a specific event, and have no - * need for specifying the condition - you just want to wait until someone runs - * closure_wake_up() on the appropriate wait list. In that case, just use - * closure_wait(). It will return either true or false, depending on whether the - * closure was already on a wait list or not - a closure can only be on one wait - * list at a time. - * - * Parents: - * - * closure_init() takes two arguments - it takes the closure to initialize, and - * a (possibly null) parent. - * - * If parent is non null, the new closure will have a refcount for its lifetime; - * a closure is considered to be "finished" when its refcount hits 0 and the - * function to run is null. Hence - * - * continue_at(cl, NULL, NULL); - * - * returns up the (spaghetti) stack of closures, precisely like normal return - * returns up the C stack. continue_at() with non null fn is better thought of - * as doing a tail call. - * - * All this implies that a closure should typically be embedded in a particular - * struct (which its refcount will normally control the lifetime of), and that - * struct can very much be thought of as a stack frame. - */ - -struct closure; -struct closure_syncer; -typedef void (closure_fn) (struct closure *); -extern struct dentry *bcache_debug; - -struct closure_waitlist { - struct llist_head list; -}; - -enum closure_state { - /* - * CLOSURE_WAITING: Set iff the closure is on a waitlist. Must be set by - * the thread that owns the closure, and cleared by the thread that's - * waking up the closure. - * - * The rest are for debugging and don't affect behaviour: - * - * CLOSURE_RUNNING: Set when a closure is running (i.e. by - * closure_init() and when closure_put() runs then next function), and - * must be cleared before remaining hits 0. Primarily to help guard - * against incorrect usage and accidentally transferring references. - * continue_at() and closure_return() clear it for you, if you're doing - * something unusual you can use closure_set_dead() which also helps - * annotate where references are being transferred. - */ - - CLOSURE_BITS_START = (1U << 26), - CLOSURE_DESTRUCTOR = (1U << 26), - CLOSURE_WAITING = (1U << 28), - CLOSURE_RUNNING = (1U << 30), -}; - -#define CLOSURE_GUARD_MASK \ - ((CLOSURE_DESTRUCTOR|CLOSURE_WAITING|CLOSURE_RUNNING) << 1) - -#define CLOSURE_REMAINING_MASK (CLOSURE_BITS_START - 1) -#define CLOSURE_REMAINING_INITIALIZER (1|CLOSURE_RUNNING) - -struct closure { - union { - struct { - struct workqueue_struct *wq; - struct closure_syncer *s; - struct llist_node list; - closure_fn *fn; - }; - struct work_struct work; - }; - - struct closure *parent; - - atomic_t remaining; - -#ifdef CONFIG_BCACHE_CLOSURES_DEBUG -#define CLOSURE_MAGIC_DEAD 0xc054dead -#define CLOSURE_MAGIC_ALIVE 0xc054a11e - - unsigned int magic; - struct list_head all; - unsigned long ip; - unsigned long waiting_on; -#endif -}; - -void closure_sub(struct closure *cl, int v); -void closure_put(struct closure *cl); -void __closure_wake_up(struct closure_waitlist *list); -bool closure_wait(struct closure_waitlist *list, struct closure *cl); -void __closure_sync(struct closure *cl); - -/** - * closure_sync - sleep until a closure a closure has nothing left to wait on - * - * Sleeps until the refcount hits 1 - the thread that's running the closure owns - * the last refcount. - */ -static inline void closure_sync(struct closure *cl) -{ - if ((atomic_read(&cl->remaining) & CLOSURE_REMAINING_MASK) != 1) - __closure_sync(cl); -} - -#ifdef CONFIG_BCACHE_CLOSURES_DEBUG - -void closure_debug_init(void); -void closure_debug_create(struct closure *cl); -void closure_debug_destroy(struct closure *cl); - -#else - -static inline void closure_debug_init(void) {} -static inline void closure_debug_create(struct closure *cl) {} -static inline void closure_debug_destroy(struct closure *cl) {} - -#endif - -static inline void closure_set_ip(struct closure *cl) -{ -#ifdef CONFIG_BCACHE_CLOSURES_DEBUG - cl->ip = _THIS_IP_; -#endif -} - -static inline void closure_set_ret_ip(struct closure *cl) -{ -#ifdef CONFIG_BCACHE_CLOSURES_DEBUG - cl->ip = _RET_IP_; -#endif -} - -static inline void closure_set_waiting(struct closure *cl, unsigned long f) -{ -#ifdef CONFIG_BCACHE_CLOSURES_DEBUG - cl->waiting_on = f; -#endif -} - -static inline void closure_set_stopped(struct closure *cl) -{ - atomic_sub(CLOSURE_RUNNING, &cl->remaining); -} - -static inline void set_closure_fn(struct closure *cl, closure_fn *fn, - struct workqueue_struct *wq) -{ - closure_set_ip(cl); - cl->fn = fn; - cl->wq = wq; - /* between atomic_dec() in closure_put() */ - smp_mb__before_atomic(); -} - -static inline void closure_queue(struct closure *cl) -{ - struct workqueue_struct *wq = cl->wq; - /** - * Changes made to closure, work_struct, or a couple of other structs - * may cause work.func not pointing to the right location. - */ - BUILD_BUG_ON(offsetof(struct closure, fn) - != offsetof(struct work_struct, func)); - if (wq) { - INIT_WORK(&cl->work, cl->work.func); - BUG_ON(!queue_work(wq, &cl->work)); - } else - cl->fn(cl); -} - -/** - * closure_get - increment a closure's refcount - */ -static inline void closure_get(struct closure *cl) -{ -#ifdef CONFIG_BCACHE_CLOSURES_DEBUG - BUG_ON((atomic_inc_return(&cl->remaining) & - CLOSURE_REMAINING_MASK) <= 1); -#else - atomic_inc(&cl->remaining); -#endif -} - -/** - * closure_init - Initialize a closure, setting the refcount to 1 - * @cl: closure to initialize - * @parent: parent of the new closure. cl will take a refcount on it for its - * lifetime; may be NULL. - */ -static inline void closure_init(struct closure *cl, struct closure *parent) -{ - memset(cl, 0, sizeof(struct closure)); - cl->parent = parent; - if (parent) - closure_get(parent); - - atomic_set(&cl->remaining, CLOSURE_REMAINING_INITIALIZER); - - closure_debug_create(cl); - closure_set_ip(cl); -} - -static inline void closure_init_stack(struct closure *cl) -{ - memset(cl, 0, sizeof(struct closure)); - atomic_set(&cl->remaining, CLOSURE_REMAINING_INITIALIZER); -} - -/** - * closure_wake_up - wake up all closures on a wait list, - * with memory barrier - */ -static inline void closure_wake_up(struct closure_waitlist *list) -{ - /* Memory barrier for the wait list */ - smp_mb(); - __closure_wake_up(list); -} - -/** - * continue_at - jump to another function with barrier - * - * After @cl is no longer waiting on anything (i.e. all outstanding refs have - * been dropped with closure_put()), it will resume execution at @fn running out - * of @wq (or, if @wq is NULL, @fn will be called by closure_put() directly). - * - * This is because after calling continue_at() you no longer have a ref on @cl, - * and whatever @cl owns may be freed out from under you - a running closure fn - * has a ref on its own closure which continue_at() drops. - * - * Note you are expected to immediately return after using this macro. - */ -#define continue_at(_cl, _fn, _wq) \ -do { \ - set_closure_fn(_cl, _fn, _wq); \ - closure_sub(_cl, CLOSURE_RUNNING + 1); \ -} while (0) - -/** - * closure_return - finish execution of a closure - * - * This is used to indicate that @cl is finished: when all outstanding refs on - * @cl have been dropped @cl's ref on its parent closure (as passed to - * closure_init()) will be dropped, if one was specified - thus this can be - * thought of as returning to the parent closure. - */ -#define closure_return(_cl) continue_at((_cl), NULL, NULL) - -/** - * continue_at_nobarrier - jump to another function without barrier - * - * Causes @fn to be executed out of @cl, in @wq context (or called directly if - * @wq is NULL). - * - * The ref the caller of continue_at_nobarrier() had on @cl is now owned by @fn, - * thus it's not safe to touch anything protected by @cl after a - * continue_at_nobarrier(). - */ -#define continue_at_nobarrier(_cl, _fn, _wq) \ -do { \ - set_closure_fn(_cl, _fn, _wq); \ - closure_queue(_cl); \ -} while (0) - -/** - * closure_return_with_destructor - finish execution of a closure, - * with destructor - * - * Works like closure_return(), except @destructor will be called when all - * outstanding refs on @cl have been dropped; @destructor may be used to safely - * free the memory occupied by @cl, and it is called with the ref on the parent - * closure still held - so @destructor could safely return an item to a - * freelist protected by @cl's parent. - */ -#define closure_return_with_destructor(_cl, _destructor) \ -do { \ - set_closure_fn(_cl, _destructor, NULL); \ - closure_sub(_cl, CLOSURE_RUNNING - CLOSURE_DESTRUCTOR + 1); \ -} while (0) - -/** - * closure_call - execute @fn out of a new, uninitialized closure - * - * Typically used when running out of one closure, and we want to run @fn - * asynchronously out of a new closure - @parent will then wait for @cl to - * finish. - */ -static inline void closure_call(struct closure *cl, closure_fn fn, - struct workqueue_struct *wq, - struct closure *parent) -{ - closure_init(cl, parent); - continue_at_nobarrier(cl, fn, wq); -} - -#endif /* _LINUX_CLOSURE_H */ diff --git a/drivers/md/bcache/super.c b/drivers/md/bcache/super.c index 0ae2b3676293..4affe587586c 100644 --- a/drivers/md/bcache/super.c +++ b/drivers/md/bcache/super.c @@ -2905,7 +2905,6 @@ static int __init bcache_init(void) goto err; bch_debug_init(); - closure_debug_init(); bcache_is_reboot = false; diff --git a/drivers/md/bcache/util.h b/drivers/md/bcache/util.h index 6f3cb7c92130..f61ab1bada6c 100644 --- a/drivers/md/bcache/util.h +++ b/drivers/md/bcache/util.h @@ -4,6 +4,7 @@ #define _BCACHE_UTIL_H #include +#include #include #include #include @@ -13,8 +14,6 @@ #include #include -#include "closure.h" - struct closure; #ifdef CONFIG_BCACHE_DEBUG diff --git a/include/linux/closure.h b/include/linux/closure.h new file mode 100644 index 000000000000..0ec9e7bc8d97 --- /dev/null +++ b/include/linux/closure.h @@ -0,0 +1,377 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef _LINUX_CLOSURE_H +#define _LINUX_CLOSURE_H + +#include +#include +#include +#include + +/* + * Closure is perhaps the most overused and abused term in computer science, but + * since I've been unable to come up with anything better you're stuck with it + * again. + * + * What are closures? + * + * They embed a refcount. The basic idea is they count "things that are in + * progress" - in flight bios, some other thread that's doing something else - + * anything you might want to wait on. + * + * The refcount may be manipulated with closure_get() and closure_put(). + * closure_put() is where many of the interesting things happen, when it causes + * the refcount to go to 0. + * + * Closures can be used to wait on things both synchronously and asynchronously, + * and synchronous and asynchronous use can be mixed without restriction. To + * wait synchronously, use closure_sync() - you will sleep until your closure's + * refcount hits 1. + * + * To wait asynchronously, use + * continue_at(cl, next_function, workqueue); + * + * passing it, as you might expect, the function to run when nothing is pending + * and the workqueue to run that function out of. + * + * continue_at() also, critically, requires a 'return' immediately following the + * location where this macro is referenced, to return to the calling function. + * There's good reason for this. + * + * To use safely closures asynchronously, they must always have a refcount while + * they are running owned by the thread that is running them. Otherwise, suppose + * you submit some bios and wish to have a function run when they all complete: + * + * foo_endio(struct bio *bio) + * { + * closure_put(cl); + * } + * + * closure_init(cl); + * + * do_stuff(); + * closure_get(cl); + * bio1->bi_endio = foo_endio; + * bio_submit(bio1); + * + * do_more_stuff(); + * closure_get(cl); + * bio2->bi_endio = foo_endio; + * bio_submit(bio2); + * + * continue_at(cl, complete_some_read, system_wq); + * + * If closure's refcount started at 0, complete_some_read() could run before the + * second bio was submitted - which is almost always not what you want! More + * importantly, it wouldn't be possible to say whether the original thread or + * complete_some_read()'s thread owned the closure - and whatever state it was + * associated with! + * + * So, closure_init() initializes a closure's refcount to 1 - and when a + * closure_fn is run, the refcount will be reset to 1 first. + * + * Then, the rule is - if you got the refcount with closure_get(), release it + * with closure_put() (i.e, in a bio->bi_endio function). If you have a refcount + * on a closure because you called closure_init() or you were run out of a + * closure - _always_ use continue_at(). Doing so consistently will help + * eliminate an entire class of particularly pernicious races. + * + * Lastly, you might have a wait list dedicated to a specific event, and have no + * need for specifying the condition - you just want to wait until someone runs + * closure_wake_up() on the appropriate wait list. In that case, just use + * closure_wait(). It will return either true or false, depending on whether the + * closure was already on a wait list or not - a closure can only be on one wait + * list at a time. + * + * Parents: + * + * closure_init() takes two arguments - it takes the closure to initialize, and + * a (possibly null) parent. + * + * If parent is non null, the new closure will have a refcount for its lifetime; + * a closure is considered to be "finished" when its refcount hits 0 and the + * function to run is null. Hence + * + * continue_at(cl, NULL, NULL); + * + * returns up the (spaghetti) stack of closures, precisely like normal return + * returns up the C stack. continue_at() with non null fn is better thought of + * as doing a tail call. + * + * All this implies that a closure should typically be embedded in a particular + * struct (which its refcount will normally control the lifetime of), and that + * struct can very much be thought of as a stack frame. + */ + +struct closure; +struct closure_syncer; +typedef void (closure_fn) (struct closure *); +extern struct dentry *bcache_debug; + +struct closure_waitlist { + struct llist_head list; +}; + +enum closure_state { + /* + * CLOSURE_WAITING: Set iff the closure is on a waitlist. Must be set by + * the thread that owns the closure, and cleared by the thread that's + * waking up the closure. + * + * The rest are for debugging and don't affect behaviour: + * + * CLOSURE_RUNNING: Set when a closure is running (i.e. by + * closure_init() and when closure_put() runs then next function), and + * must be cleared before remaining hits 0. Primarily to help guard + * against incorrect usage and accidentally transferring references. + * continue_at() and closure_return() clear it for you, if you're doing + * something unusual you can use closure_set_dead() which also helps + * annotate where references are being transferred. + */ + + CLOSURE_BITS_START = (1U << 26), + CLOSURE_DESTRUCTOR = (1U << 26), + CLOSURE_WAITING = (1U << 28), + CLOSURE_RUNNING = (1U << 30), +}; + +#define CLOSURE_GUARD_MASK \ + ((CLOSURE_DESTRUCTOR|CLOSURE_WAITING|CLOSURE_RUNNING) << 1) + +#define CLOSURE_REMAINING_MASK (CLOSURE_BITS_START - 1) +#define CLOSURE_REMAINING_INITIALIZER (1|CLOSURE_RUNNING) + +struct closure { + union { + struct { + struct workqueue_struct *wq; + struct closure_syncer *s; + struct llist_node list; + closure_fn *fn; + }; + struct work_struct work; + }; + + struct closure *parent; + + atomic_t remaining; + +#ifdef CONFIG_DEBUG_CLOSURES +#define CLOSURE_MAGIC_DEAD 0xc054dead +#define CLOSURE_MAGIC_ALIVE 0xc054a11e + + unsigned int magic; + struct list_head all; + unsigned long ip; + unsigned long waiting_on; +#endif +}; + +void closure_sub(struct closure *cl, int v); +void closure_put(struct closure *cl); +void __closure_wake_up(struct closure_waitlist *list); +bool closure_wait(struct closure_waitlist *list, struct closure *cl); +void __closure_sync(struct closure *cl); + +/** + * closure_sync - sleep until a closure a closure has nothing left to wait on + * + * Sleeps until the refcount hits 1 - the thread that's running the closure owns + * the last refcount. + */ +static inline void closure_sync(struct closure *cl) +{ + if ((atomic_read(&cl->remaining) & CLOSURE_REMAINING_MASK) != 1) + __closure_sync(cl); +} + +#ifdef CONFIG_DEBUG_CLOSURES + +void closure_debug_create(struct closure *cl); +void closure_debug_destroy(struct closure *cl); + +#else + +static inline void closure_debug_create(struct closure *cl) {} +static inline void closure_debug_destroy(struct closure *cl) {} + +#endif + +static inline void closure_set_ip(struct closure *cl) +{ +#ifdef CONFIG_DEBUG_CLOSURES + cl->ip = _THIS_IP_; +#endif +} + +static inline void closure_set_ret_ip(struct closure *cl) +{ +#ifdef CONFIG_DEBUG_CLOSURES + cl->ip = _RET_IP_; +#endif +} + +static inline void closure_set_waiting(struct closure *cl, unsigned long f) +{ +#ifdef CONFIG_DEBUG_CLOSURES + cl->waiting_on = f; +#endif +} + +static inline void closure_set_stopped(struct closure *cl) +{ + atomic_sub(CLOSURE_RUNNING, &cl->remaining); +} + +static inline void set_closure_fn(struct closure *cl, closure_fn *fn, + struct workqueue_struct *wq) +{ + closure_set_ip(cl); + cl->fn = fn; + cl->wq = wq; + /* between atomic_dec() in closure_put() */ + smp_mb__before_atomic(); +} + +static inline void closure_queue(struct closure *cl) +{ + struct workqueue_struct *wq = cl->wq; + /** + * Changes made to closure, work_struct, or a couple of other structs + * may cause work.func not pointing to the right location. + */ + BUILD_BUG_ON(offsetof(struct closure, fn) + != offsetof(struct work_struct, func)); + + if (wq) { + INIT_WORK(&cl->work, cl->work.func); + BUG_ON(!queue_work(wq, &cl->work)); + } else + cl->fn(cl); +} + +/** + * closure_get - increment a closure's refcount + */ +static inline void closure_get(struct closure *cl) +{ +#ifdef CONFIG_DEBUG_CLOSURES + BUG_ON((atomic_inc_return(&cl->remaining) & + CLOSURE_REMAINING_MASK) <= 1); +#else + atomic_inc(&cl->remaining); +#endif +} + +/** + * closure_init - Initialize a closure, setting the refcount to 1 + * @cl: closure to initialize + * @parent: parent of the new closure. cl will take a refcount on it for its + * lifetime; may be NULL. + */ +static inline void closure_init(struct closure *cl, struct closure *parent) +{ + cl->fn = NULL; + cl->parent = parent; + if (parent) + closure_get(parent); + + atomic_set(&cl->remaining, CLOSURE_REMAINING_INITIALIZER); + + closure_debug_create(cl); + closure_set_ip(cl); +} + +static inline void closure_init_stack(struct closure *cl) +{ + memset(cl, 0, sizeof(struct closure)); + atomic_set(&cl->remaining, CLOSURE_REMAINING_INITIALIZER); +} + +/** + * closure_wake_up - wake up all closures on a wait list, + * with memory barrier + */ +static inline void closure_wake_up(struct closure_waitlist *list) +{ + /* Memory barrier for the wait list */ + smp_mb(); + __closure_wake_up(list); +} + +/** + * continue_at - jump to another function with barrier + * + * After @cl is no longer waiting on anything (i.e. all outstanding refs have + * been dropped with closure_put()), it will resume execution at @fn running out + * of @wq (or, if @wq is NULL, @fn will be called by closure_put() directly). + * + * This is because after calling continue_at() you no longer have a ref on @cl, + * and whatever @cl owns may be freed out from under you - a running closure fn + * has a ref on its own closure which continue_at() drops. + * + * Note you are expected to immediately return after using this macro. + */ +#define continue_at(_cl, _fn, _wq) \ +do { \ + set_closure_fn(_cl, _fn, _wq); \ + closure_sub(_cl, CLOSURE_RUNNING + 1); \ +} while (0) + +/** + * closure_return - finish execution of a closure + * + * This is used to indicate that @cl is finished: when all outstanding refs on + * @cl have been dropped @cl's ref on its parent closure (as passed to + * closure_init()) will be dropped, if one was specified - thus this can be + * thought of as returning to the parent closure. + */ +#define closure_return(_cl) continue_at((_cl), NULL, NULL) + +/** + * continue_at_nobarrier - jump to another function without barrier + * + * Causes @fn to be executed out of @cl, in @wq context (or called directly if + * @wq is NULL). + * + * The ref the caller of continue_at_nobarrier() had on @cl is now owned by @fn, + * thus it's not safe to touch anything protected by @cl after a + * continue_at_nobarrier(). + */ +#define continue_at_nobarrier(_cl, _fn, _wq) \ +do { \ + set_closure_fn(_cl, _fn, _wq); \ + closure_queue(_cl); \ +} while (0) + +/** + * closure_return_with_destructor - finish execution of a closure, + * with destructor + * + * Works like closure_return(), except @destructor will be called when all + * outstanding refs on @cl have been dropped; @destructor may be used to safely + * free the memory occupied by @cl, and it is called with the ref on the parent + * closure still held - so @destructor could safely return an item to a + * freelist protected by @cl's parent. + */ +#define closure_return_with_destructor(_cl, _destructor) \ +do { \ + set_closure_fn(_cl, _destructor, NULL); \ + closure_sub(_cl, CLOSURE_RUNNING - CLOSURE_DESTRUCTOR + 1); \ +} while (0) + +/** + * closure_call - execute @fn out of a new, uninitialized closure + * + * Typically used when running out of one closure, and we want to run @fn + * asynchronously out of a new closure - @parent will then wait for @cl to + * finish. + */ +static inline void closure_call(struct closure *cl, closure_fn fn, + struct workqueue_struct *wq, + struct closure *parent) +{ + closure_init(cl, parent); + continue_at_nobarrier(cl, fn, wq); +} + +#endif /* _LINUX_CLOSURE_H */ diff --git a/lib/Kconfig b/lib/Kconfig index c686f4adc124..263aa6ae8d7c 100644 --- a/lib/Kconfig +++ b/lib/Kconfig @@ -506,6 +506,9 @@ config ASSOCIATIVE_ARRAY for more information. +config CLOSURES + bool + config HAS_IOMEM bool depends on !NO_IOMEM diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index fa307f93fa2e..ce3a4abf40f8 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -1720,6 +1720,15 @@ config DEBUG_NOTIFIERS This is a relatively cheap check but if you care about maximum performance, say N. +config DEBUG_CLOSURES + bool "Debug closures (bcache async widgits)" + depends on CLOSURES + select DEBUG_FS + help + Keeps all active closures in a linked list and provides a debugfs + interface to list them, which makes it possible to see asynchronous + operations that get stuck. + config DEBUG_MAPLE_TREE bool "Debug maple trees" depends on DEBUG_KERNEL diff --git a/lib/Makefile b/lib/Makefile index 740109b6e2c8..57d394575919 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -255,6 +255,8 @@ obj-$(CONFIG_ATOMIC64_SELFTEST) += atomic64_test.o obj-$(CONFIG_CPU_RMAP) += cpu_rmap.o +obj-$(CONFIG_CLOSURES) += closure.o + obj-$(CONFIG_DQL) += dynamic_queue_limits.o obj-$(CONFIG_GLOB) += glob.o diff --git a/lib/closure.c b/lib/closure.c new file mode 100644 index 000000000000..b38ded00b9b0 --- /dev/null +++ b/lib/closure.c @@ -0,0 +1,204 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Asynchronous refcounty things + * + * Copyright 2010, 2011 Kent Overstreet + * Copyright 2012 Google, Inc. + */ + +#include +#include +#include +#include +#include + +static inline void closure_put_after_sub(struct closure *cl, int flags) +{ + int r = flags & CLOSURE_REMAINING_MASK; + + BUG_ON(flags & CLOSURE_GUARD_MASK); + BUG_ON(!r && (flags & ~CLOSURE_DESTRUCTOR)); + + if (!r) { + if (cl->fn && !(flags & CLOSURE_DESTRUCTOR)) { + atomic_set(&cl->remaining, + CLOSURE_REMAINING_INITIALIZER); + closure_queue(cl); + } else { + struct closure *parent = cl->parent; + closure_fn *destructor = cl->fn; + + closure_debug_destroy(cl); + + if (destructor) + destructor(cl); + + if (parent) + closure_put(parent); + } + } +} + +/* For clearing flags with the same atomic op as a put */ +void closure_sub(struct closure *cl, int v) +{ + closure_put_after_sub(cl, atomic_sub_return(v, &cl->remaining)); +} +EXPORT_SYMBOL(closure_sub); + +/* + * closure_put - decrement a closure's refcount + */ +void closure_put(struct closure *cl) +{ + closure_put_after_sub(cl, atomic_dec_return(&cl->remaining)); +} +EXPORT_SYMBOL(closure_put); + +/* + * closure_wake_up - wake up all closures on a wait list, without memory barrier + */ +void __closure_wake_up(struct closure_waitlist *wait_list) +{ + struct llist_node *list; + struct closure *cl, *t; + struct llist_node *reverse = NULL; + + list = llist_del_all(&wait_list->list); + + /* We first reverse the list to preserve FIFO ordering and fairness */ + reverse = llist_reverse_order(list); + + /* Then do the wakeups */ + llist_for_each_entry_safe(cl, t, reverse, list) { + closure_set_waiting(cl, 0); + closure_sub(cl, CLOSURE_WAITING + 1); + } +} +EXPORT_SYMBOL(__closure_wake_up); + +/** + * closure_wait - add a closure to a waitlist + * @waitlist: will own a ref on @cl, which will be released when + * closure_wake_up() is called on @waitlist. + * @cl: closure pointer. + * + */ +bool closure_wait(struct closure_waitlist *waitlist, struct closure *cl) +{ + if (atomic_read(&cl->remaining) & CLOSURE_WAITING) + return false; + + closure_set_waiting(cl, _RET_IP_); + atomic_add(CLOSURE_WAITING + 1, &cl->remaining); + llist_add(&cl->list, &waitlist->list); + + return true; +} +EXPORT_SYMBOL(closure_wait); + +struct closure_syncer { + struct task_struct *task; + int done; +}; + +static void closure_sync_fn(struct closure *cl) +{ + struct closure_syncer *s = cl->s; + struct task_struct *p; + + rcu_read_lock(); + p = READ_ONCE(s->task); + s->done = 1; + wake_up_process(p); + rcu_read_unlock(); +} + +void __sched __closure_sync(struct closure *cl) +{ + struct closure_syncer s = { .task = current }; + + cl->s = &s; + continue_at(cl, closure_sync_fn, NULL); + + while (1) { + set_current_state(TASK_UNINTERRUPTIBLE); + if (s.done) + break; + schedule(); + } + + __set_current_state(TASK_RUNNING); +} +EXPORT_SYMBOL(__closure_sync); + +#ifdef CONFIG_DEBUG_CLOSURES + +static LIST_HEAD(closure_list); +static DEFINE_SPINLOCK(closure_list_lock); + +void closure_debug_create(struct closure *cl) +{ + unsigned long flags; + + BUG_ON(cl->magic == CLOSURE_MAGIC_ALIVE); + cl->magic = CLOSURE_MAGIC_ALIVE; + + spin_lock_irqsave(&closure_list_lock, flags); + list_add(&cl->all, &closure_list); + spin_unlock_irqrestore(&closure_list_lock, flags); +} +EXPORT_SYMBOL(closure_debug_create); + +void closure_debug_destroy(struct closure *cl) +{ + unsigned long flags; + + BUG_ON(cl->magic != CLOSURE_MAGIC_ALIVE); + cl->magic = CLOSURE_MAGIC_DEAD; + + spin_lock_irqsave(&closure_list_lock, flags); + list_del(&cl->all); + spin_unlock_irqrestore(&closure_list_lock, flags); +} +EXPORT_SYMBOL(closure_debug_destroy); + +static int debug_show(struct seq_file *f, void *data) +{ + struct closure *cl; + + spin_lock_irq(&closure_list_lock); + + list_for_each_entry(cl, &closure_list, all) { + int r = atomic_read(&cl->remaining); + + seq_printf(f, "%p: %pS -> %pS p %p r %i ", + cl, (void *) cl->ip, cl->fn, cl->parent, + r & CLOSURE_REMAINING_MASK); + + seq_printf(f, "%s%s\n", + test_bit(WORK_STRUCT_PENDING_BIT, + work_data_bits(&cl->work)) ? "Q" : "", + r & CLOSURE_RUNNING ? "R" : ""); + + if (r & CLOSURE_WAITING) + seq_printf(f, " W %pS\n", + (void *) cl->waiting_on); + + seq_puts(f, "\n"); + } + + spin_unlock_irq(&closure_list_lock); + return 0; +} + +DEFINE_SHOW_ATTRIBUTE(debug); + +static int __init closure_debug_init(void) +{ + debugfs_create_file("closures", 0400, NULL, NULL, &debug_fops); + return 0; +} +late_initcall(closure_debug_init) + +#endif -- cgit v1.2.3