From 9f10523f891928330b7529da54c1a3cc65180b1a Mon Sep 17 00:00:00 2001 From: David Howells Date: Thu, 20 Dec 2012 21:52:35 +0000 Subject: FS-Cache: Fix operation state management and accounting Fix the state management of internal fscache operations and the accounting of what operations are in what states. This is done by: (1) Give struct fscache_operation a enum variable that directly represents the state it's currently in, rather than spreading this knowledge over a bunch of flags, who's processing the operation at the moment and whether it is queued or not. This makes it easier to write assertions to check the state at various points and to prevent invalid state transitions. (2) Add an 'operation complete' state and supply a function to indicate the completion of an operation (fscache_op_complete()) and make things call it. The final call to fscache_put_operation() can then check that an op in the appropriate state (complete or cancelled). (3) Adjust the use of object->n_ops, ->n_in_progress, ->n_exclusive to better govern the state of an object: (a) The ->n_ops is now the number of extant operations on the object and is now decremented by fscache_put_operation() only. (b) The ->n_in_progress is simply the number of objects that have been taken off of the object's pending queue for the purposes of being run. This is decremented by fscache_op_complete() only. (c) The ->n_exclusive is the number of exclusive ops that have been submitted and queued or are in progress. It is decremented by fscache_op_complete() and by fscache_cancel_op(). fscache_put_operation() and fscache_operation_gc() now no longer try to clean up ->n_exclusive and ->n_in_progress. That was leading to double decrements against fscache_cancel_op(). fscache_cancel_op() now no longer decrements ->n_ops. That was leading to double decrements against fscache_put_operation(). fscache_submit_exclusive_op() now decides whether it has to queue an op based on ->n_in_progress being > 0 rather than ->n_ops > 0 as the latter will persist in being true even after all preceding operations have been cancelled or completed. Furthermore, if an object is active and there are runnable ops against it, there must be at least one op running. (4) Add a remaining-pages counter (n_pages) to struct fscache_retrieval and provide a function to record completion of the pages as they complete. When n_pages reaches 0, the operation is deemed to be complete and fscache_op_complete() is called. Add calls to fscache_retrieval_complete() anywhere we've finished with a page we've been given to read or allocate for. This includes places where we just return pages to the netfs for reading from the server and where accessing the cache fails and we discard the proposed netfs page. The bugs in the unfixed state management manifest themselves as oopses like the following where the operation completion gets out of sync with return of the cookie by the netfs. This is possible because the cache unlocks and returns all the netfs pages before recording its completion - which means that there's nothing to stop the netfs discarding them and returning the cookie. FS-Cache: Cookie 'NFS.fh' still has outstanding reads ------------[ cut here ]------------ kernel BUG at fs/fscache/cookie.c:519! invalid opcode: 0000 [#1] SMP CPU 1 Modules linked in: cachefiles nfs fscache auth_rpcgss nfs_acl lockd sunrpc Pid: 400, comm: kswapd0 Not tainted 3.1.0-rc7-fsdevel+ #1090 /DG965RY RIP: 0010:[] [] __fscache_relinquish_cookie+0x170/0x343 [fscache] RSP: 0018:ffff8800368cfb00 EFLAGS: 00010282 RAX: 000000000000003c RBX: ffff880023cc8790 RCX: 0000000000000000 RDX: 0000000000002f2e RSI: 0000000000000001 RDI: ffffffff813ab86c RBP: ffff8800368cfb50 R08: 0000000000000002 R09: 0000000000000000 R10: ffff88003a1b7890 R11: ffff88001df6e488 R12: ffff880023d8ed98 R13: ffff880023cc8798 R14: 0000000000000004 R15: ffff88003b8bf370 FS: 0000000000000000(0000) GS:ffff88003bd00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b CR2: 00000000008ba008 CR3: 0000000023d93000 CR4: 00000000000006e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400 Process kswapd0 (pid: 400, threadinfo ffff8800368ce000, task ffff88003b8bf040) Stack: ffff88003b8bf040 ffff88001df6e528 ffff88001df6e528 ffffffffa00b46b0 ffff88003b8bf040 ffff88001df6e488 ffff88001df6e620 ffffffffa00b46b0 ffff88001ebd04c8 0000000000000004 ffff8800368cfb70 ffffffffa00b2c91 Call Trace: [] nfs_fscache_release_inode_cookie+0x3b/0x47 [nfs] [] nfs_clear_inode+0x3c/0x41 [nfs] [] nfs4_evict_inode+0x2f/0x33 [nfs] [] evict+0xa1/0x15c [] dispose_list+0x2c/0x38 [] prune_icache_sb+0x28c/0x29b [] prune_super+0xd5/0x140 [] shrink_slab+0x102/0x1ab [] balance_pgdat+0x2f2/0x595 [] ? process_timeout+0xb/0xb [] kswapd+0x270/0x289 [] ? __init_waitqueue_head+0x46/0x46 [] ? balance_pgdat+0x595/0x595 [] kthread+0x7f/0x87 [] kernel_thread_helper+0x4/0x10 [] ? finish_task_switch+0x45/0xc0 [] ? retint_restore_args+0xe/0xe [] ? __init_kthread_worker+0x53/0x53 [] ? gs_change+0xb/0xb Signed-off-by: David Howells --- fs/fscache/object.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'fs/fscache/object.c') diff --git a/fs/fscache/object.c b/fs/fscache/object.c index b6b897c550ac..773bc798a416 100644 --- a/fs/fscache/object.c +++ b/fs/fscache/object.c @@ -587,8 +587,6 @@ static void fscache_object_available(struct fscache_object *object) if (object->n_in_progress == 0) { if (object->n_ops > 0) { ASSERTCMP(object->n_ops, >=, object->n_obj_ops); - ASSERTIF(object->n_ops > object->n_obj_ops, - !list_empty(&object->pending_ops)); fscache_start_operations(object); } else { ASSERT(list_empty(&object->pending_ops)); -- cgit From ef778e7ae67cd426c30cad43378b908f5eb0bad5 Mon Sep 17 00:00:00 2001 From: David Howells Date: Thu, 20 Dec 2012 21:52:36 +0000 Subject: FS-Cache: Provide proper invalidation Provide a proper invalidation method rather than relying on the netfs retiring the cookie it has and getting a new one. The problem with this is that isn't easy for the netfs to make sure that it has completed/cancelled all its outstanding storage and retrieval operations on the cookie it is retiring. Instead, have the cache provide an invalidation method that will cancel or wait for all currently outstanding operations before invalidating the cache, and will cause new operations to queue up behind that. Whilst invalidation is in progress, some requests will be rejected until the cache can stack a barrier on the operation queue to cause new operations to be deferred behind it. Signed-off-by: David Howells --- fs/fscache/object.c | 72 +++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 72 insertions(+) (limited to 'fs/fscache/object.c') diff --git a/fs/fscache/object.c b/fs/fscache/object.c index 773bc798a416..80b549141ea6 100644 --- a/fs/fscache/object.c +++ b/fs/fscache/object.c @@ -14,6 +14,7 @@ #define FSCACHE_DEBUG_LEVEL COOKIE #include +#include #include "internal.h" const char *fscache_object_states[FSCACHE_OBJECT__NSTATES] = { @@ -22,6 +23,7 @@ const char *fscache_object_states[FSCACHE_OBJECT__NSTATES] = { [FSCACHE_OBJECT_CREATING] = "OBJECT_CREATING", [FSCACHE_OBJECT_AVAILABLE] = "OBJECT_AVAILABLE", [FSCACHE_OBJECT_ACTIVE] = "OBJECT_ACTIVE", + [FSCACHE_OBJECT_INVALIDATING] = "OBJECT_INVALIDATING", [FSCACHE_OBJECT_UPDATING] = "OBJECT_UPDATING", [FSCACHE_OBJECT_DYING] = "OBJECT_DYING", [FSCACHE_OBJECT_LC_DYING] = "OBJECT_LC_DYING", @@ -39,6 +41,7 @@ const char fscache_object_states_short[FSCACHE_OBJECT__NSTATES][5] = { [FSCACHE_OBJECT_CREATING] = "CRTN", [FSCACHE_OBJECT_AVAILABLE] = "AVBL", [FSCACHE_OBJECT_ACTIVE] = "ACTV", + [FSCACHE_OBJECT_INVALIDATING] = "INVL", [FSCACHE_OBJECT_UPDATING] = "UPDT", [FSCACHE_OBJECT_DYING] = "DYNG", [FSCACHE_OBJECT_LC_DYING] = "LCDY", @@ -54,6 +57,7 @@ static void fscache_put_object(struct fscache_object *); static void fscache_initialise_object(struct fscache_object *); static void fscache_lookup_object(struct fscache_object *); static void fscache_object_available(struct fscache_object *); +static void fscache_invalidate_object(struct fscache_object *); static void fscache_release_object(struct fscache_object *); static void fscache_withdraw_object(struct fscache_object *); static void fscache_enqueue_dependents(struct fscache_object *); @@ -78,6 +82,15 @@ static inline void fscache_done_parent_op(struct fscache_object *object) spin_unlock(&parent->lock); } +/* + * Notify netfs of invalidation completion. + */ +static inline void fscache_invalidation_complete(struct fscache_cookie *cookie) +{ + if (test_and_clear_bit(FSCACHE_COOKIE_INVALIDATING, &cookie->flags)) + wake_up_bit(&cookie->flags, FSCACHE_COOKIE_INVALIDATING); +} + /* * process events that have been sent to an object's state machine * - initiates parent lookup @@ -125,6 +138,16 @@ static void fscache_object_state_machine(struct fscache_object *object) case FSCACHE_OBJECT_ACTIVE: goto active_transit; + /* Invalidate an object on disk */ + case FSCACHE_OBJECT_INVALIDATING: + clear_bit(FSCACHE_OBJECT_EV_INVALIDATE, &object->events); + fscache_stat(&fscache_n_invalidates_run); + fscache_stat(&fscache_n_cop_invalidate_object); + fscache_invalidate_object(object); + fscache_stat_d(&fscache_n_cop_invalidate_object); + fscache_raise_event(object, FSCACHE_OBJECT_EV_UPDATE); + goto active_transit; + /* update the object metadata on disk */ case FSCACHE_OBJECT_UPDATING: clear_bit(FSCACHE_OBJECT_EV_UPDATE, &object->events); @@ -275,6 +298,9 @@ active_transit: case FSCACHE_OBJECT_EV_ERROR: new_state = FSCACHE_OBJECT_DYING; goto change_state; + case FSCACHE_OBJECT_EV_INVALIDATE: + new_state = FSCACHE_OBJECT_INVALIDATING; + goto change_state; case FSCACHE_OBJECT_EV_UPDATE: new_state = FSCACHE_OBJECT_UPDATING; goto change_state; @@ -679,6 +705,7 @@ static void fscache_withdraw_object(struct fscache_object *object) if (object->cookie == cookie) { hlist_del_init(&object->cookie_link); object->cookie = NULL; + fscache_invalidation_complete(cookie); detached = true; } spin_unlock(&cookie->lock); @@ -888,3 +915,48 @@ enum fscache_checkaux fscache_check_aux(struct fscache_object *object, return result; } EXPORT_SYMBOL(fscache_check_aux); + +/* + * Asynchronously invalidate an object. + */ +static void fscache_invalidate_object(struct fscache_object *object) +{ + struct fscache_operation *op; + struct fscache_cookie *cookie = object->cookie; + + _enter("{OBJ%x}", object->debug_id); + + /* Reject any new read/write ops and abort any that are pending. */ + fscache_invalidate_writes(cookie); + clear_bit(FSCACHE_OBJECT_PENDING_WRITE, &object->flags); + fscache_cancel_all_ops(object); + + /* Now we have to wait for in-progress reads and writes */ + op = kzalloc(sizeof(*op), GFP_KERNEL); + if (!op) { + fscache_raise_event(object, FSCACHE_OBJECT_EV_ERROR); + _leave(" [ENOMEM]"); + return; + } + + fscache_operation_init(op, object->cache->ops->invalidate_object, NULL); + op->flags = FSCACHE_OP_ASYNC | (1 << FSCACHE_OP_EXCLUSIVE); + + spin_lock(&cookie->lock); + if (fscache_submit_exclusive_op(object, op) < 0) + BUG(); + spin_unlock(&cookie->lock); + fscache_put_operation(op); + + /* Once we've completed the invalidation, we know there will be no data + * stored in the cache and thus we can reinstate the data-check-skip + * optimisation. + */ + set_bit(FSCACHE_COOKIE_NO_DATA_YET, &cookie->flags); + + /* We can allow read and write requests to come in once again. They'll + * queue up behind our exclusive invalidation operation. + */ + fscache_invalidation_complete(cookie); + _leave(""); +} -- cgit From 03acc4be5e479eebc95338cd1d72a9954c128e2b Mon Sep 17 00:00:00 2001 From: David Howells Date: Wed, 5 Dec 2012 13:34:46 +0000 Subject: FS-Cache: Initialise the object event mask with the calculated mask Initialise the object event mask with the calculated mask rather than unmasking undefined events also. Signed-off-by: David Howells --- fs/fscache/object.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'fs/fscache/object.c') diff --git a/fs/fscache/object.c b/fs/fscache/object.c index 80b549141ea6..2ef8a082a272 100644 --- a/fs/fscache/object.c +++ b/fs/fscache/object.c @@ -114,7 +114,8 @@ static void fscache_object_state_machine(struct fscache_object *object) /* wait for the parent object to become ready */ case FSCACHE_OBJECT_INIT: object->event_mask = - ULONG_MAX & ~(1 << FSCACHE_OBJECT_EV_CLEARED); + FSCACHE_OBJECT_EVENTS_MASK & + ~(1 << FSCACHE_OBJECT_EV_CLEARED); fscache_initialise_object(object); goto done; -- cgit From 8d76349d359064859217dc292dc8733e209705af Mon Sep 17 00:00:00 2001 From: David Howells Date: Wed, 5 Dec 2012 13:34:48 +0000 Subject: FS-Cache: Exclusive op submission can BUG if there's been an I/O error The function to submit an exclusive op (fscache_submit_exclusive_op()) can BUG if there's been an I/O error because it may see the parent cache object in an unexpected state. It should only BUG if there hasn't been an I/O error. In this case the problem was produced by remounting the cache partition to be R/O. The EROFS state was detected and the cache was aborted, but not everything handled the aborting correctly. SysRq : Emergency Remount R/O EXT4-fs (sda6): re-mounted. Opts: (null) Emergency Remount complete CacheFiles: I/O Error: Failed to update xattr with error -30 FS-Cache: Cache cachefiles stopped due to I/O error ------------[ cut here ]------------ kernel BUG at fs/fscache/operation.c:128! invalid opcode: 0000 [#1] SMP CPU 0 Modules linked in: cachefiles nfs fscache auth_rpcgss nfs_acl lockd sunrpc Pid: 6612, comm: kworker/u:2 Not tainted 3.1.0-rc8-fsdevel+ #1093 /DG965RY RIP: 0010:[] [] fscache_submit_exclusive_op+0x2ad/0x2c2 [fscache] RSP: 0018:ffff880000853d40 EFLAGS: 00010206 RAX: ffff880038ac72a8 RBX: ffff8800181f2260 RCX: ffffffff81f2b2b0 RDX: 0000000000000001 RSI: ffffffff8179a478 RDI: ffff8800181f2280 RBP: ffff880000853d60 R08: 0000000000000002 R09: 0000000000000000 R10: 0000000000000001 R11: 0000000000000001 R12: ffff880038ac7268 R13: ffff8800181f2280 R14: ffff88003a359190 R15: 000000010122b162 FS: 0000000000000000(0000) GS:ffff88003bc00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b CR2: 00000034cc4a77f0 CR3: 0000000010e96000 CR4: 00000000000006f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400 Process kworker/u:2 (pid: 6612, threadinfo ffff880000852000, task ffff880014c3c040) Stack: ffff8800181f2260 ffff8800181f2310 ffff880038ac7268 ffff8800181f2260 ffff880000853dc0 ffffffffa0072375 ffff880037ecfe00 ffff88003a359198 ffff880000853dc0 0000000000000246 0000000000000000 ffff88000a91d308 Call Trace: [] fscache_object_work_func+0x792/0xe65 [fscache] [] process_one_work+0x1eb/0x37f [] ? process_one_work+0x18d/0x37f [] ? fscache_enqueue_dependents+0xd8/0xd8 [fscache] [] worker_thread+0x15a/0x21a [] ? rescuer_thread+0x188/0x188 [] kthread+0x7f/0x87 [] kernel_thread_helper+0x4/0x10 [] ? finish_task_switch+0x45/0xc0 [] ? retint_restore_args+0xe/0xe [] ? __init_kthread_worker+0x53/0x53 [] ? gs_change+0xb/0xb Signed-off-by: David Howells --- fs/fscache/object.c | 23 +++++++++++++++++------ 1 file changed, 17 insertions(+), 6 deletions(-) (limited to 'fs/fscache/object.c') diff --git a/fs/fscache/object.c b/fs/fscache/object.c index 2ef8a082a272..2c512cbac380 100644 --- a/fs/fscache/object.c +++ b/fs/fscache/object.c @@ -103,6 +103,7 @@ static void fscache_object_state_machine(struct fscache_object *object) { enum fscache_object_state new_state; struct fscache_cookie *cookie; + int event; ASSERT(object != NULL); @@ -275,7 +276,8 @@ static void fscache_object_state_machine(struct fscache_object *object) /* determine the transition from a lookup state */ lookup_transit: - switch (fls(object->events & object->event_mask) - 1) { + event = fls(object->events & object->event_mask) - 1; + switch (event) { case FSCACHE_OBJECT_EV_WITHDRAW: case FSCACHE_OBJECT_EV_RETIRE: case FSCACHE_OBJECT_EV_RELEASE: @@ -292,7 +294,8 @@ lookup_transit: /* determine the transition from an active state */ active_transit: - switch (fls(object->events & object->event_mask) - 1) { + event = fls(object->events & object->event_mask) - 1; + switch (event) { case FSCACHE_OBJECT_EV_WITHDRAW: case FSCACHE_OBJECT_EV_RETIRE: case FSCACHE_OBJECT_EV_RELEASE: @@ -314,7 +317,8 @@ active_transit: /* determine the transition from a terminal state */ terminal_transit: - switch (fls(object->events & object->event_mask) - 1) { + event = fls(object->events & object->event_mask) - 1; + switch (event) { case FSCACHE_OBJECT_EV_WITHDRAW: new_state = FSCACHE_OBJECT_WITHDRAWING; goto change_state; @@ -347,8 +351,8 @@ done: unsupported_event: printk(KERN_ERR "FS-Cache:" - " Unsupported event %lx [mask %lx] in state %s\n", - object->events, object->event_mask, + " Unsupported event %d [%lx/%lx] in state %s\n", + event, object->events, object->event_mask, fscache_object_states[object->state]); BUG(); } @@ -945,7 +949,7 @@ static void fscache_invalidate_object(struct fscache_object *object) spin_lock(&cookie->lock); if (fscache_submit_exclusive_op(object, op) < 0) - BUG(); + goto submit_op_failed; spin_unlock(&cookie->lock); fscache_put_operation(op); @@ -960,4 +964,11 @@ static void fscache_invalidate_object(struct fscache_object *object) */ fscache_invalidation_complete(cookie); _leave(""); + return; + +submit_op_failed: + spin_unlock(&cookie->lock); + kfree(op); + fscache_raise_event(object, FSCACHE_OBJECT_EV_ERROR); + _leave(" [EIO]"); } -- cgit From 969695215f9a865cbf64c4ce3742ac9fc57fffed Mon Sep 17 00:00:00 2001 From: David Howells Date: Wed, 5 Dec 2012 13:34:49 +0000 Subject: FS-Cache: Add transition to handle invalidate immediately after lookup Add a missing transition to the FS-Cache object state machine to handle an invalidation event occuring between the back end completing the object lookup by calling fscache_obtained_object() (which moves to state OBJECT_AVAILABLE) and the backend returning to fscache_lookup_object() and thence to fscache_object_state_machine() which then does a goto lookup_transit to handle the transition - but lookup_transit doesn't handle EV_INVALIDATE. Without this, the following BUG can be logged: FS-Cache: Unsupported event 2 [5/f7] in state OBJECT_AVAILABLE ------------[ cut here ]------------ kernel BUG at fs/fscache/object.c:357! Where event 2 is EV_INVALIDATE. Signed-off-by: David Howells --- fs/fscache/object.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'fs/fscache/object.c') diff --git a/fs/fscache/object.c b/fs/fscache/object.c index 2c512cbac380..50d41c180211 100644 --- a/fs/fscache/object.c +++ b/fs/fscache/object.c @@ -284,6 +284,9 @@ lookup_transit: case FSCACHE_OBJECT_EV_ERROR: new_state = FSCACHE_OBJECT_LC_DYING; goto change_state; + case FSCACHE_OBJECT_EV_INVALIDATE: + new_state = FSCACHE_OBJECT_INVALIDATING; + goto change_state; case FSCACHE_OBJECT_EV_REQUEUE: goto done; case -1: -- cgit