From cb65537ee1134d3cc55c1fa83952bc8eb1212833 Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 10 May 2013 19:50:26 +0100 Subject: Add wait_on_atomic_t() and wake_up_atomic_t() Add wait_on_atomic_t() and wake_up_atomic_t() to indicate became-zero events on atomic_t types. This uses the bit-wake waitqueue table. The key is set to a value outside of the number of bits in a long so that wait_on_bit() won't be woken up accidentally. What I'm using this for is: in a following patch I add a counter to struct fscache_cookie to count the number of outstanding operations that need access to netfs data. The way this works is: (1) When a cookie is allocated, the counter is initialised to 1. (2) When an operation wants to access netfs data, it calls atomic_inc_unless() to increment the counter before it does so. If it was 0, then the counter isn't incremented, the operation isn't permitted to access the netfs data (which might by this point no longer exist) and the operation aborts in some appropriate manner. (3) When an operation finishes with the netfs data, it decrements the counter and if it reaches 0, calls wake_up_atomic_t() on it - the assumption being that it was the last blocker. (4) When a cookie is released, the counter is decremented and the releaser uses wait_on_atomic_t() to wait for the counter to become 0 - which should indicate no one is using the netfs data any longer. The netfs data can then be destroyed. There are some alternatives that I have thought of and that have been suggested by Tejun Heo: (A) Using wait_on_bit() to wait on a bit in the counter. This doesn't work because if that bit happens to be 0 then the wait won't happen - even if the counter is non-zero. (B) Using wait_on_bit() to wait on a flag elsewhere which is cleared when the counter reaches 0. Such a flag would be redundant and would add complexity. (C) Adding a waitqueue to fscache_cookie - this would expand that struct by several words for an event that happens just once in each cookie's lifetime. Further, cookies are generally per-file so there are likely to be a lot of them. (D) Similar to (C), but add a pointer to a waitqueue in the cookie instead of a waitqueue. This would add single word per cookie and so would be less of an expansion - but still an expansion. (E) Adding a static waitqueue to the fscache module. Generally this would be fine, but under certain circumstances many cookies will all get added at the same time (eg. NFS umount, cache withdrawal) thereby presenting scaling issues. Note that the wait may be significant as disk I/O may be in progress. So, I think reusing the wait_on_bit() waitqueue set is reasonable. I don't make much use of the waitqueue I need on a per-cookie basis, but sometimes I have a huge flood of the cookies to deal with. I also don't want to add a whole new set of global waitqueue tables specifically for the dec-to-0 event if I can reuse the bit tables. Signed-off-by: David Howells Tested-By: Milosz Tanski Acked-by: Jeff Layton --- include/linux/wait.h | 24 ++++++++++++++++++++++++ 1 file changed, 24 insertions(+) (limited to 'include/linux') diff --git a/include/linux/wait.h b/include/linux/wait.h index ac38be2692d8..5bacfc4b336d 100644 --- a/include/linux/wait.h +++ b/include/linux/wait.h @@ -23,6 +23,7 @@ struct __wait_queue { struct wait_bit_key { void *flags; int bit_nr; +#define WAIT_ATOMIC_T_BIT_NR -1 }; struct wait_bit_queue { @@ -60,6 +61,9 @@ struct task_struct; #define __WAIT_BIT_KEY_INITIALIZER(word, bit) \ { .flags = word, .bit_nr = bit, } +#define __WAIT_ATOMIC_T_KEY_INITIALIZER(p) \ + { .flags = p, .bit_nr = WAIT_ATOMIC_T_BIT_NR, } + extern void __init_waitqueue_head(wait_queue_head_t *q, const char *name, struct lock_class_key *); #define init_waitqueue_head(q) \ @@ -146,8 +150,10 @@ void __wake_up_bit(wait_queue_head_t *, void *, int); int __wait_on_bit(wait_queue_head_t *, struct wait_bit_queue *, int (*)(void *), unsigned); int __wait_on_bit_lock(wait_queue_head_t *, struct wait_bit_queue *, int (*)(void *), unsigned); void wake_up_bit(void *, int); +void wake_up_atomic_t(atomic_t *); int out_of_line_wait_on_bit(void *, int, int (*)(void *), unsigned); int out_of_line_wait_on_bit_lock(void *, int, int (*)(void *), unsigned); +int out_of_line_wait_on_atomic_t(atomic_t *, int (*)(atomic_t *), unsigned); wait_queue_head_t *bit_waitqueue(void *, int); #define wake_up(x) __wake_up(x, TASK_NORMAL, 1, NULL) @@ -896,5 +902,23 @@ static inline int wait_on_bit_lock(void *word, int bit, return 0; return out_of_line_wait_on_bit_lock(word, bit, action, mode); } + +/** + * wait_on_atomic_t - Wait for an atomic_t to become 0 + * @val: The atomic value being waited on, a kernel virtual address + * @action: the function used to sleep, which may take special actions + * @mode: the task state to sleep in + * + * Wait for an atomic_t to become 0. We abuse the bit-wait waitqueue table for + * the purpose of getting a waitqueue, but we set the key to a bit number + * outside of the target 'word'. + */ +static inline +int wait_on_atomic_t(atomic_t *val, int (*action)(atomic_t *), unsigned mode) +{ + if (atomic_read(val) == 0) + return 0; + return out_of_line_wait_on_atomic_t(val, action, mode); +} #endif -- cgit v1.2.3 From 610be24ee434aa89197f06f30fef02be83c006a5 Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 10 May 2013 19:50:25 +0100 Subject: FS-Cache: Uninline fscache_object_init() Uninline fscache_object_init() so as not to expose some of the FS-Cache internals to the cache backend. Signed-off-by: David Howells Tested-By: Milosz Tanski Acked-by: Jeff Layton --- include/linux/fscache-cache.h | 38 ++------------------------------------ 1 file changed, 2 insertions(+), 36 deletions(-) (limited to 'include/linux') diff --git a/include/linux/fscache-cache.h b/include/linux/fscache-cache.h index 5dfa0aa216b6..9b9c1de4a460 100644 --- a/include/linux/fscache-cache.h +++ b/include/linux/fscache-cache.h @@ -426,42 +426,8 @@ extern const char *fscache_object_states[]; (test_bit(FSCACHE_IOERROR, &(obj)->cache->flags) && \ (obj)->state >= FSCACHE_OBJECT_DYING) -extern void fscache_object_work_func(struct work_struct *work); - -/** - * fscache_object_init - Initialise a cache object description - * @object: Object description - * - * Initialise a cache object description to its basic values. - * - * See Documentation/filesystems/caching/backend-api.txt for a complete - * description. - */ -static inline -void fscache_object_init(struct fscache_object *object, - struct fscache_cookie *cookie, - struct fscache_cache *cache) -{ - atomic_inc(&cache->object_count); - - object->state = FSCACHE_OBJECT_INIT; - spin_lock_init(&object->lock); - INIT_LIST_HEAD(&object->cache_link); - INIT_HLIST_NODE(&object->cookie_link); - INIT_WORK(&object->work, fscache_object_work_func); - INIT_LIST_HEAD(&object->dependents); - INIT_LIST_HEAD(&object->dep_link); - INIT_LIST_HEAD(&object->pending_ops); - object->n_children = 0; - object->n_ops = object->n_in_progress = object->n_exclusive = 0; - object->events = object->event_mask = 0; - object->flags = 0; - object->store_limit = 0; - object->store_limit_l = 0; - object->cache = cache; - object->cookie = cookie; - object->parent = NULL; -} +extern void fscache_object_init(struct fscache_object *, struct fscache_cookie *, + struct fscache_cache *); extern void fscache_object_lookup_negative(struct fscache_object *object); extern void fscache_obtained_object(struct fscache_object *object); -- cgit v1.2.3 From 493f7bc11457bc1f6fbf25a4b2bdf215ebaf050f Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 10 May 2013 19:50:26 +0100 Subject: FS-Cache: Wrap checks on object state Wrap checks on object state (mostly outside of fs/fscache/object.c) with inline functions so that the mechanism can be replaced. Some of the state checks within object.c are left as-is as they will be replaced. Signed-off-by: David Howells Tested-By: Milosz Tanski Acked-by: Jeff Layton --- include/linux/fscache-cache.h | 37 ++++++++++++++++++++++++++++--------- 1 file changed, 28 insertions(+), 9 deletions(-) (limited to 'include/linux') diff --git a/include/linux/fscache-cache.h b/include/linux/fscache-cache.h index 9b9c1de4a460..c5f92347cbf8 100644 --- a/include/linux/fscache-cache.h +++ b/include/linux/fscache-cache.h @@ -417,15 +417,6 @@ struct fscache_object { extern const char *fscache_object_states[]; -#define fscache_object_is_active(obj) \ - (!test_bit(FSCACHE_IOERROR, &(obj)->cache->flags) && \ - (obj)->state >= FSCACHE_OBJECT_AVAILABLE && \ - (obj)->state < FSCACHE_OBJECT_DYING) - -#define fscache_object_is_dead(obj) \ - (test_bit(FSCACHE_IOERROR, &(obj)->cache->flags) && \ - (obj)->state >= FSCACHE_OBJECT_DYING) - extern void fscache_object_init(struct fscache_object *, struct fscache_cookie *, struct fscache_cache *); @@ -438,6 +429,34 @@ extern void fscache_object_destroy(struct fscache_object *object); #define fscache_object_destroy(object) do {} while(0) #endif +static inline bool fscache_object_is_live(struct fscache_object *object) +{ + return object->state < FSCACHE_OBJECT_DYING; +} + +static inline bool fscache_object_is_dying(struct fscache_object *object) +{ + return !fscache_object_is_live(object); +} + +static inline bool fscache_object_is_available(struct fscache_object *object) +{ + return object->state >= FSCACHE_OBJECT_AVAILABLE; +} + +static inline bool fscache_object_is_active(struct fscache_object *object) +{ + return fscache_object_is_available(object) && + fscache_object_is_live(object) && + !test_bit(FSCACHE_IOERROR, &object->cache->flags); +} + +static inline bool fscache_object_is_dead(struct fscache_object *object) +{ + return fscache_object_is_dying(object) && + test_bit(FSCACHE_IOERROR, &object->cache->flags); +} + /** * fscache_object_destroyed - Note destruction of an object in a cache * @cache: The cache from which the object came -- cgit v1.2.3 From caaef6900befb45689b1d1831ce3c7e7fb5b504f Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 10 May 2013 19:50:26 +0100 Subject: FS-Cache: Fix object state machine to have separate work and wait states Fix object state machine to have separate work and wait states as that makes it easier to envision. There are now three kinds of state: (1) Work state. This is an execution state. No event processing is performed by a work state. The function attached to a work state returns a pointer indicating the next state to which the OSM should transition. Returning NO_TRANSIT repeats the current state, but goes back to the scheduler first. (2) Wait state. This is an event processing state. No execution is performed by a wait state. Wait states are just tables of "if event X occurs, clear it and transition to state Y". The dispatcher returns to the scheduler if none of the events in which the wait state has an interest are currently pending. (3) Out-of-band state. This is a special work state. Transitions to normal states can be overridden when an unexpected event occurs (eg. I/O error). Instead the dispatcher disables and clears the OOB event and transits to the specified work state. This then acts as an ordinary work state, though object->state points to the overridden destination. Returning NO_TRANSIT resumes the overridden transition. In addition, the states have names in their definitions, so there's no need for tables of state names. Further, the EV_REQUEUE event is no longer necessary as that is automatic for work states. Since the states are now separate structs rather than values in an enum, it's not possible to use comparisons other than (non-)equality between them, so use some object->flags to indicate what phase an object is in. The EV_RELEASE, EV_RETIRE and EV_WITHDRAW events have been squished into one (EV_KILL). An object flag now carries the information about retirement. Similarly, the RELEASING, RECYCLING and WITHDRAWING states have been merged into an KILL_OBJECT state and additional states have been added for handling waiting dependent objects (JUMPSTART_DEPS and KILL_DEPENDENTS). A state has also been added for synchronising with parent object initialisation (WAIT_FOR_PARENT) and another for initiating look up (PARENT_READY). Signed-off-by: David Howells Tested-By: Milosz Tanski Acked-by: Jeff Layton --- include/linux/fscache-cache.h | 66 ++++++++++++++++++++----------------------- 1 file changed, 31 insertions(+), 35 deletions(-) (limited to 'include/linux') diff --git a/include/linux/fscache-cache.h b/include/linux/fscache-cache.h index c5f92347cbf8..9ff516b1b9a0 100644 --- a/include/linux/fscache-cache.h +++ b/include/linux/fscache-cache.h @@ -328,11 +328,9 @@ struct fscache_cookie { #define FSCACHE_COOKIE_LOOKING_UP 0 /* T if non-index cookie being looked up still */ #define FSCACHE_COOKIE_CREATING 1 /* T if non-index object being created still */ #define FSCACHE_COOKIE_NO_DATA_YET 2 /* T if new object with no cached data yet */ -#define FSCACHE_COOKIE_PENDING_FILL 3 /* T if pending initial fill on object */ -#define FSCACHE_COOKIE_FILLING 4 /* T if filling object incrementally */ -#define FSCACHE_COOKIE_UNAVAILABLE 5 /* T if cookie is unavailable (error, etc) */ -#define FSCACHE_COOKIE_WAITING_ON_READS 6 /* T if cookie is waiting on reads */ -#define FSCACHE_COOKIE_INVALIDATING 7 /* T if cookie is being invalidated */ +#define FSCACHE_COOKIE_UNAVAILABLE 3 /* T if cookie is unavailable (error, etc) */ +#define FSCACHE_COOKIE_WAITING_ON_READS 4 /* T if cookie is waiting on reads */ +#define FSCACHE_COOKIE_INVALIDATING 5 /* T if cookie is being invalidated */ }; extern struct fscache_cookie fscache_fsdef_index; @@ -341,45 +339,40 @@ extern struct fscache_cookie fscache_fsdef_index; * Event list for fscache_object::{event_mask,events} */ enum { - FSCACHE_OBJECT_EV_REQUEUE, /* T if object should be requeued */ + FSCACHE_OBJECT_EV_NEW_CHILD, /* T if object has a new child */ + FSCACHE_OBJECT_EV_PARENT_READY, /* T if object's parent is ready */ FSCACHE_OBJECT_EV_UPDATE, /* T if object should be updated */ FSCACHE_OBJECT_EV_INVALIDATE, /* T if cache requested object invalidation */ FSCACHE_OBJECT_EV_CLEARED, /* T if accessors all gone */ FSCACHE_OBJECT_EV_ERROR, /* T if fatal error occurred during processing */ - FSCACHE_OBJECT_EV_RELEASE, /* T if netfs requested object release */ - FSCACHE_OBJECT_EV_RETIRE, /* T if netfs requested object retirement */ - FSCACHE_OBJECT_EV_WITHDRAW, /* T if cache requested object withdrawal */ + FSCACHE_OBJECT_EV_KILL, /* T if netfs relinquished or cache withdrew object */ NR_FSCACHE_OBJECT_EVENTS }; #define FSCACHE_OBJECT_EVENTS_MASK ((1UL << NR_FSCACHE_OBJECT_EVENTS) - 1) +/* + * States for object state machine. + */ +struct fscache_transition { + unsigned long events; + const struct fscache_state *transit_to; +}; + +struct fscache_state { + char name[24]; + char short_name[8]; + const struct fscache_state *(*work)(struct fscache_object *object, + int event); + const struct fscache_transition transitions[]; +}; + /* * on-disk cache file or index handle */ struct fscache_object { - enum fscache_object_state { - FSCACHE_OBJECT_INIT, /* object in initial unbound state */ - FSCACHE_OBJECT_LOOKING_UP, /* looking up object */ - FSCACHE_OBJECT_CREATING, /* creating object */ - - /* active states */ - FSCACHE_OBJECT_AVAILABLE, /* cleaning up object after creation */ - FSCACHE_OBJECT_ACTIVE, /* object is usable */ - FSCACHE_OBJECT_INVALIDATING, /* object is invalidating */ - FSCACHE_OBJECT_UPDATING, /* object is updating */ - - /* terminal states */ - FSCACHE_OBJECT_DYING, /* object waiting for accessors to finish */ - FSCACHE_OBJECT_LC_DYING, /* object cleaning up after lookup/create */ - FSCACHE_OBJECT_ABORT_INIT, /* abort the init state */ - FSCACHE_OBJECT_RELEASING, /* releasing object */ - FSCACHE_OBJECT_RECYCLING, /* retiring object */ - FSCACHE_OBJECT_WITHDRAWING, /* withdrawing object */ - FSCACHE_OBJECT_DEAD, /* object is now dead */ - FSCACHE_OBJECT__NSTATES - } state; - + const struct fscache_state *state; /* Object state machine state */ + const struct fscache_transition *oob_table; /* OOB state transition table */ int debug_id; /* debugging ID */ int n_children; /* number of child objects */ int n_ops; /* number of extant ops on object */ @@ -390,6 +383,7 @@ struct fscache_object { spinlock_t lock; /* state and operations lock */ unsigned long lookup_jif; /* time at which lookup started */ + unsigned long oob_event_mask; /* OOB events this object is interested in */ unsigned long event_mask; /* events this object is interested in */ unsigned long events; /* events to be processed by this object * (order is important - using fls) */ @@ -398,6 +392,10 @@ struct fscache_object { #define FSCACHE_OBJECT_LOCK 0 /* T if object is busy being processed */ #define FSCACHE_OBJECT_PENDING_WRITE 1 /* T if object has pending write */ #define FSCACHE_OBJECT_WAITING 2 /* T if object is waiting on its parent */ +#define FSCACHE_OBJECT_RETIRE 3 /* T if object should be retired */ +#define FSCACHE_OBJECT_IS_LIVE 4 /* T if object is not withdrawn or relinquished */ +#define FSCACHE_OBJECT_IS_LOOKED_UP 5 /* T if object has been looked up */ +#define FSCACHE_OBJECT_IS_AVAILABLE 6 /* T if object has become active */ struct list_head cache_link; /* link in cache->object_list */ struct hlist_node cookie_link; /* link in cookie->backing_objects */ @@ -415,8 +413,6 @@ struct fscache_object { loff_t store_limit_l; /* current storage limit */ }; -extern const char *fscache_object_states[]; - extern void fscache_object_init(struct fscache_object *, struct fscache_cookie *, struct fscache_cache *); @@ -431,7 +427,7 @@ extern void fscache_object_destroy(struct fscache_object *object); static inline bool fscache_object_is_live(struct fscache_object *object) { - return object->state < FSCACHE_OBJECT_DYING; + return test_bit(FSCACHE_OBJECT_IS_LIVE, &object->flags); } static inline bool fscache_object_is_dying(struct fscache_object *object) @@ -441,7 +437,7 @@ static inline bool fscache_object_is_dying(struct fscache_object *object) static inline bool fscache_object_is_available(struct fscache_object *object) { - return object->state >= FSCACHE_OBJECT_AVAILABLE; + return test_bit(FSCACHE_OBJECT_IS_AVAILABLE, &object->flags); } static inline bool fscache_object_is_active(struct fscache_object *object) -- cgit v1.2.3 From 1362729b169b7903c7e739dbe7904994b0d8c47f Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 10 May 2013 19:50:26 +0100 Subject: FS-Cache: Simplify cookie retention for fscache_objects, fixing oops Simplify the way fscache cache objects retain their cookie. The way I implemented the cookie storage handling made synchronisation a pain (ie. the object state machine can't rely on the cookie actually still being there). Instead of the the object being detached from the cookie and the cookie being freed in __fscache_relinquish_cookie(), we defer both operations: (*) The detachment of the object from the list in the cookie now takes place in fscache_drop_object() and is thus governed by the object state machine (fscache_detach_from_cookie() has been removed). (*) The release of the cookie is now in fscache_object_destroy() - which is called by the cache backend just before it frees the object. This means that the fscache_cookie struct is now available to the cache all the way through from ->alloc_object() to ->drop_object() and ->put_object() - meaning that it's no longer necessary to take object->lock to guarantee access. However, __fscache_relinquish_cookie() doesn't wait for the object to go all the way through to destruction before letting the netfs proceed. That would massively slow down the netfs. Since __fscache_relinquish_cookie() leaves the cookie around, in must therefore break all attachments to the netfs - which includes ->def, ->netfs_data and any outstanding page read/writes. To handle this, struct fscache_cookie now has an n_active counter: (1) This starts off initialised to 1. (2) Any time the cache needs to get at the netfs data, it calls fscache_use_cookie() to increment it - if it is not zero. If it was zero, then access is not permitted. (3) When the cache has finished with the data, it calls fscache_unuse_cookie() to decrement it. This does a wake-up on it if it reaches 0. (4) __fscache_relinquish_cookie() decrements n_active and then waits for it to reach 0. The initialisation to 1 in step (1) ensures that we only get wake ups when we're trying to get rid of the cookie. This leaves __fscache_relinquish_cookie() a lot simpler. *** This fixes a problem in the current code whereby if fscache_invalidate() is followed sufficiently quickly by fscache_relinquish_cookie() then it is possible for __fscache_relinquish_cookie() to have detached the cookie from the object and cleared the pointer before a thread is dispatched to process the invalidation state in the object state machine. Since the pending write clearance was deferred to the invalidation state to make it asynchronous, we need to either wait in relinquishment for the stores tree to be cleared in the invalidation state or we need to handle the clearance in relinquishment. Further, if the relinquishment code does clear the tree, then the invalidation state need to make the clearance contingent on still having the cookie to hand (since that's where the tree is rooted) and we have to prevent the cookie from disappearing for the duration. This can lead to an oops like the following: BUG: unable to handle kernel NULL pointer dereference at 000000000000000c ... RIP: 0010:[] _spin_lock+0xe/0x30 ... CR2: 000000000000000c ... ... Process kslowd002 (...) .... Call Trace: [] fscache_invalidate_writes+0x38/0xd0 [fscache] [] ? __switch_to+0xd0/0x320 [] ? find_busiest_queue+0x69/0x150 [] ? slow_work_enqueue+0x104/0x180 [] fscache_object_slow_work_execute+0x5e3/0x9d0 [fscache] [] ? bit_waitqueue+0x17/0xd0 [] slow_work_execute+0x233/0x310 [] slow_work_thread+0x205/0x360 [] ? autoremove_wake_function+0x0/0x40 [] ? slow_work_thread+0x0/0x360 [] kthread+0x96/0xa0 [] child_rip+0xa/0x20 [] ? kthread+0x0/0xa0 [] ? child_rip+0x0/0x20 The parameter to fscache_invalidate_writes() was object->cookie which is NULL. Signed-off-by: David Howells Tested-By: Milosz Tanski Acked-by: Jeff Layton --- include/linux/fscache-cache.h | 55 ++++++++++++++++++++++++++++++------------- 1 file changed, 39 insertions(+), 16 deletions(-) (limited to 'include/linux') diff --git a/include/linux/fscache-cache.h b/include/linux/fscache-cache.h index 9ff516b1b9a0..d32f70611a00 100644 --- a/include/linux/fscache-cache.h +++ b/include/linux/fscache-cache.h @@ -97,7 +97,8 @@ struct fscache_operation { #define FSCACHE_OP_WAITING 4 /* cleared when op is woken */ #define FSCACHE_OP_EXCLUSIVE 5 /* exclusive op, other ops must wait */ #define FSCACHE_OP_DEC_READ_CNT 6 /* decrement object->n_reads on destruction */ -#define FSCACHE_OP_KEEP_FLAGS 0x0070 /* flags to keep when repurposing an op */ +#define FSCACHE_OP_UNUSE_COOKIE 7 /* call fscache_unuse_cookie() on completion */ +#define FSCACHE_OP_KEEP_FLAGS 0x00f0 /* flags to keep when repurposing an op */ enum fscache_operation_state state; atomic_t usage; @@ -314,6 +315,7 @@ struct fscache_cache_ops { struct fscache_cookie { atomic_t usage; /* number of users of this cookie */ atomic_t n_children; /* number of children of this cookie */ + atomic_t n_active; /* number of active users of netfs ptrs */ spinlock_t lock; spinlock_t stores_lock; /* lock on page store tree */ struct hlist_head backing_objects; /* object(s) backing this file/index */ @@ -326,11 +328,11 @@ struct fscache_cookie { unsigned long flags; #define FSCACHE_COOKIE_LOOKING_UP 0 /* T if non-index cookie being looked up still */ -#define FSCACHE_COOKIE_CREATING 1 /* T if non-index object being created still */ -#define FSCACHE_COOKIE_NO_DATA_YET 2 /* T if new object with no cached data yet */ -#define FSCACHE_COOKIE_UNAVAILABLE 3 /* T if cookie is unavailable (error, etc) */ -#define FSCACHE_COOKIE_WAITING_ON_READS 4 /* T if cookie is waiting on reads */ -#define FSCACHE_COOKIE_INVALIDATING 5 /* T if cookie is being invalidated */ +#define FSCACHE_COOKIE_NO_DATA_YET 1 /* T if new object with no cached data yet */ +#define FSCACHE_COOKIE_UNAVAILABLE 2 /* T if cookie is unavailable (error, etc) */ +#define FSCACHE_COOKIE_INVALIDATING 3 /* T if cookie is being invalidated */ +#define FSCACHE_COOKIE_RELINQUISHED 4 /* T if cookie has been relinquished */ +#define FSCACHE_COOKIE_RETIRED 5 /* T if cookie was retired */ }; extern struct fscache_cookie fscache_fsdef_index; @@ -392,10 +394,9 @@ struct fscache_object { #define FSCACHE_OBJECT_LOCK 0 /* T if object is busy being processed */ #define FSCACHE_OBJECT_PENDING_WRITE 1 /* T if object has pending write */ #define FSCACHE_OBJECT_WAITING 2 /* T if object is waiting on its parent */ -#define FSCACHE_OBJECT_RETIRE 3 /* T if object should be retired */ -#define FSCACHE_OBJECT_IS_LIVE 4 /* T if object is not withdrawn or relinquished */ -#define FSCACHE_OBJECT_IS_LOOKED_UP 5 /* T if object has been looked up */ -#define FSCACHE_OBJECT_IS_AVAILABLE 6 /* T if object has become active */ +#define FSCACHE_OBJECT_IS_LIVE 3 /* T if object is not withdrawn or relinquished */ +#define FSCACHE_OBJECT_IS_LOOKED_UP 4 /* T if object has been looked up */ +#define FSCACHE_OBJECT_IS_AVAILABLE 5 /* T if object has become active */ struct list_head cache_link; /* link in cache->object_list */ struct hlist_node cookie_link; /* link in cookie->backing_objects */ @@ -415,16 +416,11 @@ struct fscache_object { extern void fscache_object_init(struct fscache_object *, struct fscache_cookie *, struct fscache_cache *); +extern void fscache_object_destroy(struct fscache_object *); extern void fscache_object_lookup_negative(struct fscache_object *object); extern void fscache_obtained_object(struct fscache_object *object); -#ifdef CONFIG_FSCACHE_OBJECT_LIST -extern void fscache_object_destroy(struct fscache_object *object); -#else -#define fscache_object_destroy(object) do {} while(0) -#endif - static inline bool fscache_object_is_live(struct fscache_object *object) { return test_bit(FSCACHE_OBJECT_IS_LIVE, &object->flags); @@ -512,6 +508,33 @@ static inline void fscache_end_io(struct fscache_retrieval *op, op->end_io_func(page, op->context, error); } +/** + * fscache_use_cookie - Request usage of cookie attached to an object + * @object: Object description + * + * Request usage of the cookie attached to an object. NULL is returned if the + * relinquishment had reduced the cookie usage count to 0. + */ +static inline bool fscache_use_cookie(struct fscache_object *object) +{ + struct fscache_cookie *cookie = object->cookie; + return atomic_inc_not_zero(&cookie->n_active) != 0; +} + +/** + * fscache_unuse_cookie - Cease usage of cookie attached to an object + * @object: Object description + * + * Cease usage of the cookie attached to an object. When the users count + * reaches zero then the cookie relinquishment will be permitted to proceed. + */ +static inline void fscache_unuse_cookie(struct fscache_object *object) +{ + struct fscache_cookie *cookie = object->cookie; + if (atomic_dec_and_test(&cookie->n_active)) + wake_up_atomic_t(&cookie->n_active); +} + /* * out-of-line cache backend functions */ -- cgit v1.2.3 From 1bb4b7f98f361132ea322834515334d95b93c184 Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 21 May 2013 13:44:15 +0100 Subject: FS-Cache: The retrieval remaining-pages counter needs to be atomic_t struct fscache_retrieval contains a count of the number of pages that still need some processing (n_pages). This is decremented as the pages are processed. However, this needs to be atomic as fscache_retrieval_complete() (I think) just occasionally may be called from cachefiles_read_backing_file() and cachefiles_read_copier() simultaneously. This happens when an fscache_read_or_alloc_pages() request containing a lot of pages (say a couple of hundred) is being processed. The read on each backing page is dispatched individually because we need to insert a monitor into the waitqueue to catch when the read completes. However, under low-memory conditions, we might be forced to wait in the allocator - and this gives the I/O on the backing page a chance to complete first. When the I/O completes, fscache_enqueue_retrieval() chucks the retrieval onto the workqueue without waiting for the operation to finish the initial I/O dispatch (we want to release any pages we can as soon as we can), thus both can end up running simultaneously and potentially attempting to partially complete the retrieval simultaneously (ENOMEM may occur, backing pages may already be in the page cache). This was demonstrated by parallelling the non-atomic counter with an atomic counter and printing both of them when the assertion fails. At this point, the atomic counter has reached zero, but the non-atomic counter has not. To fix this, make the counter an atomic_t. This results in the following bug appearing FS-Cache: Assertion failed 3 == 5 is false ------------[ cut here ]------------ kernel BUG at fs/fscache/operation.c:421! or FS-Cache: Assertion failed 3 == 5 is false ------------[ cut here ]------------ kernel BUG at fs/fscache/operation.c:414! With a backtrace like the following: RIP: 0010:[] fscache_put_operation+0x1ad/0x240 [fscache] Call Trace: [] fscache_retrieval_work+0x55/0x270 [fscache] [] ? fscache_retrieval_work+0x0/0x270 [fscache] [] worker_thread+0x170/0x2a0 [] ? autoremove_wake_function+0x0/0x40 [] ? worker_thread+0x0/0x2a0 [] kthread+0x96/0xa0 [] child_rip+0xa/0x20 [] ? kthread+0x0/0xa0 [] ? child_rip+0x0/0x20 Signed-off-by: David Howells Reviewed-and-tested-By: Milosz Tanski Acked-by: Jeff Layton --- include/linux/fscache-cache.h | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/fscache-cache.h b/include/linux/fscache-cache.h index d32f70611a00..a9ff9a36b86d 100644 --- a/include/linux/fscache-cache.h +++ b/include/linux/fscache-cache.h @@ -151,7 +151,7 @@ struct fscache_retrieval { void *context; /* netfs read context (pinned) */ struct list_head to_do; /* list of things to be done by the backend */ unsigned long start_time; /* time at which retrieval started */ - unsigned n_pages; /* number of pages to be retrieved */ + atomic_t n_pages; /* number of pages to be retrieved */ }; typedef int (*fscache_page_retrieval_func_t)(struct fscache_retrieval *op, @@ -195,15 +195,14 @@ static inline void fscache_enqueue_retrieval(struct fscache_retrieval *op) static inline void fscache_retrieval_complete(struct fscache_retrieval *op, int n_pages) { - op->n_pages -= n_pages; - if (op->n_pages <= 0) + atomic_sub(n_pages, &op->n_pages); + if (atomic_read(&op->n_pages) <= 0) fscache_op_complete(&op->op, true); } /** * fscache_put_retrieval - Drop a reference to a retrieval operation * @op: The retrieval operation affected - * @n_pages: The number of pages to account for * * Drop a reference to a retrieval operation. */ -- cgit v1.2.3