/* * GPL HEADER START * * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER. * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 only, * as published by the Free Software Foundation. * * This program is distributed in the hope that it will be useful, but * WITHOUT ANY WARRANTY; without even the implied warranty of * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU * General Public License version 2 for more details (a copy is included * in the LICENSE file that accompanied this code). * * You should have received a copy of the GNU General Public License * version 2 along with this program; If not, see * http://www.sun.com/software/products/lustre/docs/GPLv2.pdf * * Please contact Sun Microsystems, Inc., 4150 Network Circle, Santa Clara, * CA 95054 USA or visit www.sun.com if you need additional information or * have any questions. * * GPL HEADER END */ /* * Copyright (c) 2008, 2010, Oracle and/or its affiliates. All rights reserved. * Use is subject to license terms. * * Copyright (c) 2011, 2012, Whamcloud, Inc. */ /* * This file is part of Lustre, http://www.lustre.org/ * Lustre is a trademark of Sun Microsystems, Inc. * * Client Lustre Page. * * Author: Nikita Danilov */ #define DEBUG_SUBSYSTEM S_CLASS #include #include #include #include #include #include "cl_internal.h" static void cl_page_delete0(const struct lu_env *env, struct cl_page *pg, int radix); static cfs_mem_cache_t *cl_page_kmem = NULL; static struct lu_kmem_descr cl_page_caches[] = { { .ckd_cache = &cl_page_kmem, .ckd_name = "cl_page_kmem", .ckd_size = sizeof (struct cl_page) }, { .ckd_cache = NULL } }; #ifdef LIBCFS_DEBUG # define PASSERT(env, page, expr) \ do { \ if (unlikely(!(expr))) { \ CL_PAGE_DEBUG(D_ERROR, (env), (page), #expr "\n"); \ LASSERT(0); \ } \ } while (0) #else /* !LIBCFS_DEBUG */ # define PASSERT(env, page, exp) \ ((void)sizeof(env), (void)sizeof(page), (void)sizeof !!(exp)) #endif /* !LIBCFS_DEBUG */ #ifdef INVARIANT_CHECK # define PINVRNT(env, page, expr) \ do { \ if (unlikely(!(expr))) { \ CL_PAGE_DEBUG(D_ERROR, (env), (page), #expr "\n"); \ LINVRNT(0); \ } \ } while (0) #else /* !INVARIANT_CHECK */ # define PINVRNT(env, page, exp) \ ((void)sizeof(env), (void)sizeof(page), (void)sizeof !!(exp)) #endif /* !INVARIANT_CHECK */ /** * Internal version of cl_page_top, it should be called with page referenced, * or coh_page_guard held. */ static struct cl_page *cl_page_top_trusted(struct cl_page *page) { while (page->cp_parent != NULL) page = page->cp_parent; return page; } /** * Internal version of cl_page_get(). * * This function can be used to obtain initial reference to previously * unreferenced cached object. It can be called only if concurrent page * reclamation is somehow prevented, e.g., by locking page radix-tree * (cl_object_header::hdr->coh_page_guard), or by keeping a lock on a VM page, * associated with \a page. * * Use with care! Not exported. */ static void cl_page_get_trust(struct cl_page *page) { /* * Checkless version for trusted users. */ if (cfs_atomic_inc_return(&page->cp_ref) == 1) cfs_atomic_inc(&cl_object_site(page->cp_obj)->cs_pages.cs_busy); } /** * Returns a slice within a page, corresponding to the given layer in the * device stack. * * \see cl_lock_at() */ static const struct cl_page_slice * cl_page_at_trusted(const struct cl_page *page, const struct lu_device_type *dtype) { const struct cl_page_slice *slice; #ifdef INVARIANT_CHECK struct cl_object_header *ch = cl_object_header(page->cp_obj); if (!cfs_atomic_read(&page->cp_ref)) LASSERT_SPIN_LOCKED(&ch->coh_page_guard); #endif ENTRY; page = cl_page_top_trusted((struct cl_page *)page); do { cfs_list_for_each_entry(slice, &page->cp_layers, cpl_linkage) { if (slice->cpl_obj->co_lu.lo_dev->ld_type == dtype) RETURN(slice); } page = page->cp_child; } while (page != NULL); RETURN(NULL); } /** * Returns a page with given index in the given object, or NULL if no page is * found. Acquires a reference on \a page. * * Locking: called under cl_object_header::coh_page_guard spin-lock. */ struct cl_page *cl_page_lookup(struct cl_object_header *hdr, pgoff_t index) { struct cl_page *page; LASSERT_SPIN_LOCKED(&hdr->coh_page_guard); page = radix_tree_lookup(&hdr->coh_tree, index); if (page != NULL) { cl_page_get_trust(page); } return page; } EXPORT_SYMBOL(cl_page_lookup); /** * Returns a list of pages by a given [start, end] of \a obj. * * \param resched If not NULL, then we give up before hogging CPU for too * long and set *resched = 1, in that case caller should implement a retry * logic. * * Gang tree lookup (radix_tree_gang_lookup()) optimization is absolutely * crucial in the face of [offset, EOF] locks. * * Return at least one page in @queue unless there is no covered page. */ int cl_page_gang_lookup(const struct lu_env *env, struct cl_object *obj, struct cl_io *io, pgoff_t start, pgoff_t end, cl_page_gang_cb_t cb, void *cbdata) { struct cl_object_header *hdr; struct cl_page *page; struct cl_page **pvec; const struct cl_page_slice *slice; const struct lu_device_type *dtype; pgoff_t idx; unsigned int nr; unsigned int i; unsigned int j; int res = CLP_GANG_OKAY; int tree_lock = 1; ENTRY; idx = start; hdr = cl_object_header(obj); pvec = cl_env_info(env)->clt_pvec; dtype = cl_object_top(obj)->co_lu.lo_dev->ld_type; cfs_spin_lock(&hdr->coh_page_guard); while ((nr = radix_tree_gang_lookup(&hdr->coh_tree, (void **)pvec, idx, CLT_PVEC_SIZE)) > 0) { int end_of_region = 0; idx = pvec[nr - 1]->cp_index + 1; for (i = 0, j = 0; i < nr; ++i) { page = pvec[i]; pvec[i] = NULL; LASSERT(page->cp_type == CPT_CACHEABLE); if (page->cp_index > end) { end_of_region = 1; break; } if (page->cp_state == CPS_FREEING) continue; slice = cl_page_at_trusted(page, dtype); /* * Pages for lsm-less file has no underneath sub-page * for osc, in case of ... */ PASSERT(env, page, slice != NULL); page = slice->cpl_page; /* * Can safely call cl_page_get_trust() under * radix-tree spin-lock. * * XXX not true, because @page is from object another * than @hdr and protected by different tree lock. */ cl_page_get_trust(page); lu_ref_add_atomic(&page->cp_reference, "gang_lookup", cfs_current()); pvec[j++] = page; } /* * Here a delicate locking dance is performed. Current thread * holds a reference to a page, but has to own it before it * can be placed into queue. Owning implies waiting, so * radix-tree lock is to be released. After a wait one has to * check that pages weren't truncated (cl_page_own() returns * error in the latter case). */ cfs_spin_unlock(&hdr->coh_page_guard); tree_lock = 0; for (i = 0; i < j; ++i) { page = pvec[i]; if (res == CLP_GANG_OKAY) res = (*cb)(env, io, page, cbdata); lu_ref_del(&page->cp_reference, "gang_lookup", cfs_current()); cl_page_put(env, page); } if (nr < CLT_PVEC_SIZE || end_of_region) break; if (res == CLP_GANG_OKAY && cfs_need_resched()) res = CLP_GANG_RESCHED; if (res != CLP_GANG_OKAY) break; cfs_spin_lock(&hdr->coh_page_guard); tree_lock = 1; } if (tree_lock) cfs_spin_unlock(&hdr->coh_page_guard); RETURN(res); } EXPORT_SYMBOL(cl_page_gang_lookup); static void cl_page_free(const struct lu_env *env, struct cl_page *page) { struct cl_object *obj = page->cp_obj; struct cl_site *site = cl_object_site(obj); PASSERT(env, page, cfs_list_empty(&page->cp_batch)); PASSERT(env, page, page->cp_owner == NULL); PASSERT(env, page, page->cp_req == NULL); PASSERT(env, page, page->cp_parent == NULL); PASSERT(env, page, page->cp_state == CPS_FREEING); ENTRY; cfs_might_sleep(); while (!cfs_list_empty(&page->cp_layers)) { struct cl_page_slice *slice; slice = cfs_list_entry(page->cp_layers.next, struct cl_page_slice, cpl_linkage); cfs_list_del_init(page->cp_layers.next); slice->cpl_ops->cpo_fini(env, slice); } cfs_atomic_dec(&site->cs_pages.cs_total); #ifdef LUSTRE_PAGESTATE_TRACKING cfs_atomic_dec(&site->cs_pages_state[page->cp_state]); #endif lu_object_ref_del_at(&obj->co_lu, page->cp_obj_ref, "cl_page", page); cl_object_put(env, obj); lu_ref_fini(&page->cp_reference); OBD_SLAB_FREE_PTR(page, cl_page_kmem); EXIT; } /** * Helper function updating page state. This is the only place in the code * where cl_page::cp_state field is mutated. */ static inline void cl_page_state_set_trust(struct cl_page *page, enum cl_page_state state) { /* bypass const. */ *(enum cl_page_state *)&page->cp_state = state; } static int cl_page_alloc(const struct lu_env *env, struct cl_object *o, pgoff_t ind, struct page *vmpage, enum cl_page_type type, struct cl_page **out) { struct cl_page *page; struct cl_page *err = NULL; struct lu_object_header *head; struct cl_site *site = cl_object_site(o); int result; ENTRY; result = +1; OBD_SLAB_ALLOC_PTR_GFP(page, cl_page_kmem, CFS_ALLOC_IO); if (page != NULL) { cfs_atomic_set(&page->cp_ref, 1); page->cp_obj = o; cl_object_get(o); page->cp_obj_ref = lu_object_ref_add(&o->co_lu, "cl_page", page); page->cp_index = ind; cl_page_state_set_trust(page, CPS_CACHED); page->cp_type = type; CFS_INIT_LIST_HEAD(&page->cp_layers); CFS_INIT_LIST_HEAD(&page->cp_batch); CFS_INIT_LIST_HEAD(&page->cp_flight); cfs_mutex_init(&page->cp_mutex); lu_ref_init(&page->cp_reference); head = o->co_lu.lo_header; cfs_list_for_each_entry(o, &head->loh_layers, co_lu.lo_linkage) { if (o->co_ops->coo_page_init != NULL) { err = o->co_ops->coo_page_init(env, o, page, vmpage); if (err != NULL) { cl_page_delete0(env, page, 0); cl_page_free(env, page); page = err; break; } } } if (err == NULL) { cfs_atomic_inc(&site->cs_pages.cs_busy); cfs_atomic_inc(&site->cs_pages.cs_total); #ifdef LUSTRE_PAGESTATE_TRACKING cfs_atomic_inc(&site->cs_pages_state[CPS_CACHED]); #endif cfs_atomic_inc(&site->cs_pages.cs_created); result = 0; } } else page = ERR_PTR(-ENOMEM); *out = page; RETURN(result); } /** * Returns a cl_page with index \a idx at the object \a o, and associated with * the VM page \a vmpage. * * This is the main entry point into the cl_page caching interface. First, a * cache (implemented as a per-object radix tree) is consulted. If page is * found there, it is returned immediately. Otherwise new page is allocated * and returned. In any case, additional reference to page is acquired. * * \see cl_object_find(), cl_lock_find() */ static struct cl_page *cl_page_find0(const struct lu_env *env, struct cl_object *o, pgoff_t idx, struct page *vmpage, enum cl_page_type type, struct cl_page *parent) { struct cl_page *page = NULL; struct cl_page *ghost = NULL; struct cl_object_header *hdr; struct cl_site *site = cl_object_site(o); int err; LASSERT(type == CPT_CACHEABLE || type == CPT_TRANSIENT); cfs_might_sleep(); ENTRY; hdr = cl_object_header(o); cfs_atomic_inc(&site->cs_pages.cs_lookup); CDEBUG(D_PAGE, "%lu@"DFID" %p %lx %d\n", idx, PFID(&hdr->coh_lu.loh_fid), vmpage, vmpage->private, type); /* fast path. */ if (type == CPT_CACHEABLE) { /* * cl_vmpage_page() can be called here without any locks as * * - "vmpage" is locked (which prevents ->private from * concurrent updates), and * * - "o" cannot be destroyed while current thread holds a * reference on it. */ page = cl_vmpage_page(vmpage, o); PINVRNT(env, page, ergo(page != NULL, cl_page_vmpage(env, page) == vmpage && (void *)radix_tree_lookup(&hdr->coh_tree, idx) == page)); } if (page != NULL) { cfs_atomic_inc(&site->cs_pages.cs_hit); RETURN(page); } /* allocate and initialize cl_page */ err = cl_page_alloc(env, o, idx, vmpage, type, &page); if (err != 0) RETURN(page); if (type == CPT_TRANSIENT) { if (parent) { LASSERT(page->cp_parent == NULL); page->cp_parent = parent; parent->cp_child = page; } RETURN(page); } /* * XXX optimization: use radix_tree_preload() here, and change tree * gfp mask to GFP_KERNEL in cl_object_header_init(). */ cfs_spin_lock(&hdr->coh_page_guard); err = radix_tree_insert(&hdr->coh_tree, idx, page); if (err != 0) { ghost = page; /* * Noted by Jay: a lock on \a vmpage protects cl_page_find() * from this race, but * * 0. it's better to have cl_page interface "locally * consistent" so that its correctness can be reasoned * about without appealing to the (obscure world of) VM * locking. * * 1. handling this race allows ->coh_tree to remain * consistent even when VM locking is somehow busted, * which is very useful during diagnosing and debugging. */ page = ERR_PTR(err); CL_PAGE_DEBUG(D_ERROR, env, ghost, "fail to insert into radix tree: %d\n", err); } else { if (parent) { LASSERT(page->cp_parent == NULL); page->cp_parent = parent; parent->cp_child = page; } hdr->coh_pages++; } cfs_spin_unlock(&hdr->coh_page_guard); if (unlikely(ghost != NULL)) { cfs_atomic_dec(&site->cs_pages.cs_busy); cl_page_delete0(env, ghost, 0); cl_page_free(env, ghost); } RETURN(page); } struct cl_page *cl_page_find(const struct lu_env *env, struct cl_object *o, pgoff_t idx, struct page *vmpage, enum cl_page_type type) { return cl_page_find0(env, o, idx, vmpage, type, NULL); } EXPORT_SYMBOL(cl_page_find); struct cl_page *cl_page_find_sub(const struct lu_env *env, struct cl_object *o, pgoff_t idx, struct page *vmpage, struct cl_page *parent) { return cl_page_find0(env, o, idx, vmpage, parent->cp_type, parent); } EXPORT_SYMBOL(cl_page_find_sub); static inline int cl_page_invariant(const struct cl_page *pg) { struct cl_object_header *header; struct cl_page *parent; struct cl_page *child; struct cl_io *owner; /* * Page invariant is protected by a VM lock. */ LINVRNT(cl_page_is_vmlocked(NULL, pg)); header = cl_object_header(pg->cp_obj); parent = pg->cp_parent; child = pg->cp_child; owner = pg->cp_owner; return cfs_atomic_read(&pg->cp_ref) > 0 && ergo(parent != NULL, parent->cp_child == pg) && ergo(child != NULL, child->cp_parent == pg) && ergo(child != NULL, pg->cp_obj != child->cp_obj) && ergo(parent != NULL, pg->cp_obj != parent->cp_obj) && ergo(owner != NULL && parent != NULL, parent->cp_owner == pg->cp_owner->ci_parent) && ergo(owner != NULL && child != NULL, child->cp_owner->ci_parent == owner) && /* * Either page is early in initialization (has neither child * nor parent yet), or it is in the object radix tree. */ ergo(pg->cp_state < CPS_FREEING && pg->cp_type == CPT_CACHEABLE, (void *)radix_tree_lookup(&header->coh_tree, pg->cp_index) == pg || (child == NULL && parent == NULL)); } static void cl_page_state_set0(const struct lu_env *env, struct cl_page *page, enum cl_page_state state) { enum cl_page_state old; #ifdef LUSTRE_PAGESTATE_TRACKING struct cl_site *site = cl_object_site(page->cp_obj); #endif /* * Matrix of allowed state transitions [old][new], for sanity * checking. */ static const int allowed_transitions[CPS_NR][CPS_NR] = { [CPS_CACHED] = { [CPS_CACHED] = 0, [CPS_OWNED] = 1, /* io finds existing cached page */ [CPS_PAGEIN] = 0, [CPS_PAGEOUT] = 1, /* write-out from the cache */ [CPS_FREEING] = 1, /* eviction on the memory pressure */ }, [CPS_OWNED] = { [CPS_CACHED] = 1, /* release to the cache */ [CPS_OWNED] = 0, [CPS_PAGEIN] = 1, /* start read immediately */ [CPS_PAGEOUT] = 1, /* start write immediately */ [CPS_FREEING] = 1, /* lock invalidation or truncate */ }, [CPS_PAGEIN] = { [CPS_CACHED] = 1, /* io completion */ [CPS_OWNED] = 0, [CPS_PAGEIN] = 0, [CPS_PAGEOUT] = 0, [CPS_FREEING] = 0, }, [CPS_PAGEOUT] = { [CPS_CACHED] = 1, /* io completion */ [CPS_OWNED] = 0, [CPS_PAGEIN] = 0, [CPS_PAGEOUT] = 0, [CPS_FREEING] = 0, }, [CPS_FREEING] = { [CPS_CACHED] = 0, [CPS_OWNED] = 0, [CPS_PAGEIN] = 0, [CPS_PAGEOUT] = 0, [CPS_FREEING] = 0, } }; ENTRY; old = page->cp_state; PASSERT(env, page, allowed_transitions[old][state]); CL_PAGE_HEADER(D_TRACE, env, page, "%d -> %d\n", old, state); for (; page != NULL; page = page->cp_child) { PASSERT(env, page, page->cp_state == old); PASSERT(env, page, equi(state == CPS_OWNED, page->cp_owner != NULL)); #ifdef LUSTRE_PAGESTATE_TRACKING cfs_atomic_dec(&site->cs_pages_state[page->cp_state]); cfs_atomic_inc(&site->cs_pages_state[state]); #endif cl_page_state_set_trust(page, state); } EXIT; } static void cl_page_state_set(const struct lu_env *env, struct cl_page *page, enum cl_page_state state) { cl_page_state_set0(env, page, state); } /** * Acquires an additional reference to a page. * * This can be called only by caller already possessing a reference to \a * page. * * \see cl_object_get(), cl_lock_get(). */ void cl_page_get(struct cl_page *page) { ENTRY; LASSERT(page->cp_state != CPS_FREEING); cl_page_get_trust(page); EXIT; } EXPORT_SYMBOL(cl_page_get); /** * Releases a reference to a page. * * When last reference is released, page is returned to the cache, unless it * is in cl_page_state::CPS_FREEING state, in which case it is immediately * destroyed. * * \see cl_object_put(), cl_lock_put(). */ void cl_page_put(const struct lu_env *env, struct cl_page *page) { struct cl_object_header *hdr; struct cl_site *site = cl_object_site(page->cp_obj); PASSERT(env, page, cfs_atomic_read(&page->cp_ref) > !!page->cp_parent); ENTRY; CL_PAGE_HEADER(D_TRACE, env, page, "%d\n", cfs_atomic_read(&page->cp_ref)); hdr = cl_object_header(cl_object_top(page->cp_obj)); if (cfs_atomic_dec_and_lock(&page->cp_ref, &hdr->coh_page_guard)) { cfs_atomic_dec(&site->cs_pages.cs_busy); /* We're going to access the page w/o a reference, but it's * ok because we have grabbed the lock coh_page_guard, which * means nobody is able to free this page behind us. */ if (page->cp_state == CPS_FREEING) { /* We drop the page reference and check the page state * inside the coh_page_guard. So that if it gets here, * it is the REALLY last reference to this page. */ cfs_spin_unlock(&hdr->coh_page_guard); LASSERT(cfs_atomic_read(&page->cp_ref) == 0); PASSERT(env, page, page->cp_owner == NULL); PASSERT(env, page, cfs_list_empty(&page->cp_batch)); /* * Page is no longer reachable by other threads. Tear * it down. */ cl_page_free(env, page); EXIT; return; } cfs_spin_unlock(&hdr->coh_page_guard); } EXIT; } EXPORT_SYMBOL(cl_page_put); /** * Returns a VM page associated with a given cl_page. */ cfs_page_t *cl_page_vmpage(const struct lu_env *env, struct cl_page *page) { const struct cl_page_slice *slice; /* * Find uppermost layer with ->cpo_vmpage() method, and return its * result. */ page = cl_page_top(page); do { cfs_list_for_each_entry(slice, &page->cp_layers, cpl_linkage) { if (slice->cpl_ops->cpo_vmpage != NULL) RETURN(slice->cpl_ops->cpo_vmpage(env, slice)); } page = page->cp_child; } while (page != NULL); LBUG(); /* ->cpo_vmpage() has to be defined somewhere in the stack */ } EXPORT_SYMBOL(cl_page_vmpage); /** * Returns a cl_page associated with a VM page, and given cl_object. */ struct cl_page *cl_vmpage_page(cfs_page_t *vmpage, struct cl_object *obj) { struct cl_page *page; struct cl_object_header *hdr; ENTRY; KLASSERT(PageLocked(vmpage)); /* * NOTE: absence of races and liveness of data are guaranteed by page * lock on a "vmpage". That works because object destruction has * bottom-to-top pass. */ /* * This loop assumes that ->private points to the top-most page. This * can be rectified easily. */ hdr = cl_object_header(cl_object_top(obj)); cfs_spin_lock(&hdr->coh_page_guard); for (page = (void *)vmpage->private; page != NULL; page = page->cp_child) { if (cl_object_same(page->cp_obj, obj)) { cl_page_get_trust(page); break; } } cfs_spin_unlock(&hdr->coh_page_guard); LASSERT(ergo(page, page->cp_type == CPT_CACHEABLE)); RETURN(page); } EXPORT_SYMBOL(cl_vmpage_page); /** * Returns the top-page for a given page. * * \see cl_object_top(), cl_io_top() */ struct cl_page *cl_page_top(struct cl_page *page) { return cl_page_top_trusted(page); } EXPORT_SYMBOL(cl_page_top); const struct cl_page_slice *cl_page_at(const struct cl_page *page, const struct lu_device_type *dtype) { return cl_page_at_trusted(page, dtype); } EXPORT_SYMBOL(cl_page_at); #define CL_PAGE_OP(opname) offsetof(struct cl_page_operations, opname) #define CL_PAGE_INVOKE(_env, _page, _op, _proto, ...) \ ({ \ const struct lu_env *__env = (_env); \ struct cl_page *__page = (_page); \ const struct cl_page_slice *__scan; \ int __result; \ ptrdiff_t __op = (_op); \ int (*__method)_proto; \ \ __result = 0; \ __page = cl_page_top(__page); \ do { \ cfs_list_for_each_entry(__scan, &__page->cp_layers, \ cpl_linkage) { \ __method = *(void **)((char *)__scan->cpl_ops + \ __op); \ if (__method != NULL) { \ __result = (*__method)(__env, __scan, \ ## __VA_ARGS__); \ if (__result != 0) \ break; \ } \ } \ __page = __page->cp_child; \ } while (__page != NULL && __result == 0); \ if (__result > 0) \ __result = 0; \ __result; \ }) #define CL_PAGE_INVOID(_env, _page, _op, _proto, ...) \ do { \ const struct lu_env *__env = (_env); \ struct cl_page *__page = (_page); \ const struct cl_page_slice *__scan; \ ptrdiff_t __op = (_op); \ void (*__method)_proto; \ \ __page = cl_page_top(__page); \ do { \ cfs_list_for_each_entry(__scan, &__page->cp_layers, \ cpl_linkage) { \ __method = *(void **)((char *)__scan->cpl_ops + \ __op); \ if (__method != NULL) \ (*__method)(__env, __scan, \ ## __VA_ARGS__); \ } \ __page = __page->cp_child; \ } while (__page != NULL); \ } while (0) #define CL_PAGE_INVOID_REVERSE(_env, _page, _op, _proto, ...) \ do { \ const struct lu_env *__env = (_env); \ struct cl_page *__page = (_page); \ const struct cl_page_slice *__scan; \ ptrdiff_t __op = (_op); \ void (*__method)_proto; \ \ /* get to the bottom page. */ \ while (__page->cp_child != NULL) \ __page = __page->cp_child; \ do { \ cfs_list_for_each_entry_reverse(__scan, &__page->cp_layers, \ cpl_linkage) { \ __method = *(void **)((char *)__scan->cpl_ops + \ __op); \ if (__method != NULL) \ (*__method)(__env, __scan, \ ## __VA_ARGS__); \ } \ __page = __page->cp_parent; \ } while (__page != NULL); \ } while (0) static int cl_page_invoke(const struct lu_env *env, struct cl_io *io, struct cl_page *page, ptrdiff_t op) { PINVRNT(env, page, cl_object_same(page->cp_obj, io->ci_obj)); ENTRY; RETURN(CL_PAGE_INVOKE(env, page, op, (const struct lu_env *, const struct cl_page_slice *, struct cl_io *), io)); } static void cl_page_invoid(const struct lu_env *env, struct cl_io *io, struct cl_page *page, ptrdiff_t op) { PINVRNT(env, page, cl_object_same(page->cp_obj, io->ci_obj)); ENTRY; CL_PAGE_INVOID(env, page, op, (const struct lu_env *, const struct cl_page_slice *, struct cl_io *), io); EXIT; } static void cl_page_owner_clear(struct cl_page *page) { ENTRY; for (page = cl_page_top(page); page != NULL; page = page->cp_child) { if (page->cp_owner != NULL) { LASSERT(page->cp_owner->ci_owned_nr > 0); page->cp_owner->ci_owned_nr--; page->cp_owner = NULL; page->cp_task = NULL; } } EXIT; } static void cl_page_owner_set(struct cl_page *page) { ENTRY; for (page = cl_page_top(page); page != NULL; page = page->cp_child) { LASSERT(page->cp_owner != NULL); page->cp_owner->ci_owned_nr++; } EXIT; } void cl_page_disown0(const struct lu_env *env, struct cl_io *io, struct cl_page *pg) { enum cl_page_state state; ENTRY; state = pg->cp_state; PINVRNT(env, pg, state == CPS_OWNED || state == CPS_FREEING); PINVRNT(env, pg, cl_page_invariant(pg)); cl_page_owner_clear(pg); if (state == CPS_OWNED) cl_page_state_set(env, pg, CPS_CACHED); /* * Completion call-backs are executed in the bottom-up order, so that * uppermost layer (llite), responsible for VFS/VM interaction runs * last and can release locks safely. */ CL_PAGE_INVOID_REVERSE(env, pg, CL_PAGE_OP(cpo_disown), (const struct lu_env *, const struct cl_page_slice *, struct cl_io *), io); EXIT; } /** * returns true, iff page is owned by the given io. */ int cl_page_is_owned(const struct cl_page *pg, const struct cl_io *io) { LINVRNT(cl_object_same(pg->cp_obj, io->ci_obj)); ENTRY; RETURN(pg->cp_state == CPS_OWNED && pg->cp_owner == io); } EXPORT_SYMBOL(cl_page_is_owned); /** * Try to own a page by IO. * * Waits until page is in cl_page_state::CPS_CACHED state, and then switch it * into cl_page_state::CPS_OWNED state. * * \pre !cl_page_is_owned(pg, io) * \post result == 0 iff cl_page_is_owned(pg, io) * * \retval 0 success * * \retval -ve failure, e.g., page was destroyed (and landed in * cl_page_state::CPS_FREEING instead of cl_page_state::CPS_CACHED). * or, page was owned by another thread, or in IO. * * \see cl_page_disown() * \see cl_page_operations::cpo_own() * \see cl_page_own_try() * \see cl_page_own */ static int cl_page_own0(const struct lu_env *env, struct cl_io *io, struct cl_page *pg, int nonblock) { int result; PINVRNT(env, pg, !cl_page_is_owned(pg, io)); ENTRY; pg = cl_page_top(pg); io = cl_io_top(io); if (pg->cp_state == CPS_FREEING) { result = -ENOENT; } else { result = CL_PAGE_INVOKE(env, pg, CL_PAGE_OP(cpo_own), (const struct lu_env *, const struct cl_page_slice *, struct cl_io *, int), io, nonblock); if (result == 0) { PASSERT(env, pg, pg->cp_owner == NULL); PASSERT(env, pg, pg->cp_req == NULL); pg->cp_owner = io; pg->cp_task = current; cl_page_owner_set(pg); if (pg->cp_state != CPS_FREEING) { cl_page_state_set(env, pg, CPS_OWNED); } else { cl_page_disown0(env, io, pg); result = -ENOENT; } } } PINVRNT(env, pg, ergo(result == 0, cl_page_invariant(pg))); RETURN(result); } /** * Own a page, might be blocked. * * \see cl_page_own0() */ int cl_page_own(const struct lu_env *env, struct cl_io *io, struct cl_page *pg) { return cl_page_own0(env, io, pg, 0); } EXPORT_SYMBOL(cl_page_own); /** * Nonblock version of cl_page_own(). * * \see cl_page_own0() */ int cl_page_own_try(const struct lu_env *env, struct cl_io *io, struct cl_page *pg) { return cl_page_own0(env, io, pg, 1); } EXPORT_SYMBOL(cl_page_own_try); /** * Assume page ownership. * * Called when page is already locked by the hosting VM. * * \pre !cl_page_is_owned(pg, io) * \post cl_page_is_owned(pg, io) * * \see cl_page_operations::cpo_assume() */ void cl_page_assume(const struct lu_env *env, struct cl_io *io, struct cl_page *pg) { PINVRNT(env, pg, cl_object_same(pg->cp_obj, io->ci_obj)); ENTRY; pg = cl_page_top(pg); io = cl_io_top(io); cl_page_invoid(env, io, pg, CL_PAGE_OP(cpo_assume)); PASSERT(env, pg, pg->cp_owner == NULL); pg->cp_owner = io; pg->cp_task = current; cl_page_owner_set(pg); cl_page_state_set(env, pg, CPS_OWNED); EXIT; } EXPORT_SYMBOL(cl_page_assume); /** * Releases page ownership without unlocking the page. * * Moves page into cl_page_state::CPS_CACHED without releasing a lock on the * underlying VM page (as VM is supposed to do this itself). * * \pre cl_page_is_owned(pg, io) * \post !cl_page_is_owned(pg, io) * * \see cl_page_assume() */ void cl_page_unassume(const struct lu_env *env, struct cl_io *io, struct cl_page *pg) { PINVRNT(env, pg, cl_page_is_owned(pg, io)); PINVRNT(env, pg, cl_page_invariant(pg)); ENTRY; pg = cl_page_top(pg); io = cl_io_top(io); cl_page_owner_clear(pg); cl_page_state_set(env, pg, CPS_CACHED); CL_PAGE_INVOID_REVERSE(env, pg, CL_PAGE_OP(cpo_unassume), (const struct lu_env *, const struct cl_page_slice *, struct cl_io *), io); EXIT; } EXPORT_SYMBOL(cl_page_unassume); /** * Releases page ownership. * * Moves page into cl_page_state::CPS_CACHED. * * \pre cl_page_is_owned(pg, io) * \post !cl_page_is_owned(pg, io) * * \see cl_page_own() * \see cl_page_operations::cpo_disown() */ void cl_page_disown(const struct lu_env *env, struct cl_io *io, struct cl_page *pg) { PINVRNT(env, pg, cl_page_is_owned(pg, io)); ENTRY; pg = cl_page_top(pg); io = cl_io_top(io); cl_page_disown0(env, io, pg); EXIT; } EXPORT_SYMBOL(cl_page_disown); /** * Called when page is to be removed from the object, e.g., as a result of * truncate. * * Calls cl_page_operations::cpo_discard() top-to-bottom. * * \pre cl_page_is_owned(pg, io) * * \see cl_page_operations::cpo_discard() */ void cl_page_discard(const struct lu_env *env, struct cl_io *io, struct cl_page *pg) { PINVRNT(env, pg, cl_page_is_owned(pg, io)); PINVRNT(env, pg, cl_page_invariant(pg)); cl_page_invoid(env, io, pg, CL_PAGE_OP(cpo_discard)); } EXPORT_SYMBOL(cl_page_discard); /** * Version of cl_page_delete() that can be called for not fully constructed * pages, e.g,. in a error handling cl_page_find()->cl_page_delete0() * path. Doesn't check page invariant. */ static void cl_page_delete0(const struct lu_env *env, struct cl_page *pg, int radix) { struct cl_page *tmp = pg; ENTRY; PASSERT(env, pg, pg == cl_page_top(pg)); PASSERT(env, pg, pg->cp_state != CPS_FREEING); /* * Severe all ways to obtain new pointers to @pg. */ cl_page_owner_clear(pg); /* * unexport the page firstly before freeing it so that * the page content is considered to be invalid. * We have to do this because a CPS_FREEING cl_page may * be NOT under the protection of a cl_lock. * Afterwards, if this page is found by other threads, then this * page will be forced to reread. */ cl_page_export(env, pg, 0); cl_page_state_set0(env, pg, CPS_FREEING); if (tmp->cp_type == CPT_CACHEABLE) { if (!radix) /* !radix means that @pg is not yet in the radix tree, * skip removing it. */ tmp = pg->cp_child; for (; tmp != NULL; tmp = tmp->cp_child) { void *value; struct cl_object_header *hdr; hdr = cl_object_header(tmp->cp_obj); cfs_spin_lock(&hdr->coh_page_guard); value = radix_tree_delete(&hdr->coh_tree, tmp->cp_index); PASSERT(env, tmp, value == tmp); PASSERT(env, tmp, hdr->coh_pages > 0); hdr->coh_pages--; cfs_spin_unlock(&hdr->coh_page_guard); } } CL_PAGE_INVOID(env, pg, CL_PAGE_OP(cpo_delete), (const struct lu_env *, const struct cl_page_slice *)); EXIT; } /** * Called when a decision is made to throw page out of memory. * * Notifies all layers about page destruction by calling * cl_page_operations::cpo_delete() method top-to-bottom. * * Moves page into cl_page_state::CPS_FREEING state (this is the only place * where transition to this state happens). * * Eliminates all venues through which new references to the page can be * obtained: * * - removes page from the radix trees, * * - breaks linkage from VM page to cl_page. * * Once page reaches cl_page_state::CPS_FREEING, all remaining references will * drain after some time, at which point page will be recycled. * * \pre pg == cl_page_top(pg) * \pre VM page is locked * \post pg->cp_state == CPS_FREEING * * \see cl_page_operations::cpo_delete() */ void cl_page_delete(const struct lu_env *env, struct cl_page *pg) { PINVRNT(env, pg, cl_page_invariant(pg)); ENTRY; cl_page_delete0(env, pg, 1); EXIT; } EXPORT_SYMBOL(cl_page_delete); /** * Unmaps page from user virtual memory. * * Calls cl_page_operations::cpo_unmap() through all layers top-to-bottom. The * layer responsible for VM interaction has to unmap page from user space * virtual memory. * * \see cl_page_operations::cpo_unmap() */ int cl_page_unmap(const struct lu_env *env, struct cl_io *io, struct cl_page *pg) { PINVRNT(env, pg, cl_page_is_owned(pg, io)); PINVRNT(env, pg, cl_page_invariant(pg)); return cl_page_invoke(env, io, pg, CL_PAGE_OP(cpo_unmap)); } EXPORT_SYMBOL(cl_page_unmap); /** * Marks page up-to-date. * * Call cl_page_operations::cpo_export() through all layers top-to-bottom. The * layer responsible for VM interaction has to mark/clear page as up-to-date * by the \a uptodate argument. * * \see cl_page_operations::cpo_export() */ void cl_page_export(const struct lu_env *env, struct cl_page *pg, int uptodate) { PINVRNT(env, pg, cl_page_invariant(pg)); CL_PAGE_INVOID(env, pg, CL_PAGE_OP(cpo_export), (const struct lu_env *, const struct cl_page_slice *, int), uptodate); } EXPORT_SYMBOL(cl_page_export); /** * Returns true, iff \a pg is VM locked in a suitable sense by the calling * thread. */ int cl_page_is_vmlocked(const struct lu_env *env, const struct cl_page *pg) { int result; const struct cl_page_slice *slice; ENTRY; pg = cl_page_top_trusted((struct cl_page *)pg); slice = container_of(pg->cp_layers.next, const struct cl_page_slice, cpl_linkage); PASSERT(env, pg, slice->cpl_ops->cpo_is_vmlocked != NULL); /* * Call ->cpo_is_vmlocked() directly instead of going through * CL_PAGE_INVOKE(), because cl_page_is_vmlocked() is used by * cl_page_invariant(). */ result = slice->cpl_ops->cpo_is_vmlocked(env, slice); PASSERT(env, pg, result == -EBUSY || result == -ENODATA); RETURN(result == -EBUSY); } EXPORT_SYMBOL(cl_page_is_vmlocked); static enum cl_page_state cl_req_type_state(enum cl_req_type crt) { ENTRY; RETURN(crt == CRT_WRITE ? CPS_PAGEOUT : CPS_PAGEIN); } static void cl_page_io_start(const struct lu_env *env, struct cl_page *pg, enum cl_req_type crt) { /* * Page is queued for IO, change its state. */ ENTRY; cl_page_owner_clear(pg); cl_page_state_set(env, pg, cl_req_type_state(crt)); EXIT; } /** * Prepares page for immediate transfer. cl_page_operations::cpo_prep() is * called top-to-bottom. Every layer either agrees to submit this page (by * returning 0), or requests to omit this page (by returning -EALREADY). Layer * handling interactions with the VM also has to inform VM that page is under * transfer now. */ int cl_page_prep(const struct lu_env *env, struct cl_io *io, struct cl_page *pg, enum cl_req_type crt) { int result; PINVRNT(env, pg, cl_page_is_owned(pg, io)); PINVRNT(env, pg, cl_page_invariant(pg)); PINVRNT(env, pg, crt < CRT_NR); /* * XXX this has to be called bottom-to-top, so that llite can set up * PG_writeback without risking other layers deciding to skip this * page. */ result = cl_page_invoke(env, io, pg, CL_PAGE_OP(io[crt].cpo_prep)); if (result == 0) cl_page_io_start(env, pg, crt); KLASSERT(ergo(crt == CRT_WRITE && pg->cp_type == CPT_CACHEABLE, equi(result == 0, PageWriteback(cl_page_vmpage(env, pg))))); CL_PAGE_HEADER(D_TRACE, env, pg, "%d %d\n", crt, result); return result; } EXPORT_SYMBOL(cl_page_prep); /** * Notify layers about transfer completion. * * Invoked by transfer sub-system (which is a part of osc) to notify layers * that a transfer, of which this page is a part of has completed. * * Completion call-backs are executed in the bottom-up order, so that * uppermost layer (llite), responsible for the VFS/VM interaction runs last * and can release locks safely. * * \pre pg->cp_state == CPS_PAGEIN || pg->cp_state == CPS_PAGEOUT * \post pg->cp_state == CPS_CACHED * * \see cl_page_operations::cpo_completion() */ void cl_page_completion(const struct lu_env *env, struct cl_page *pg, enum cl_req_type crt, int ioret) { struct cl_sync_io *anchor = pg->cp_sync_io; PASSERT(env, pg, crt < CRT_NR); /* cl_page::cp_req already cleared by the caller (osc_completion()) */ PASSERT(env, pg, pg->cp_req == NULL); PASSERT(env, pg, pg->cp_state == cl_req_type_state(crt)); ENTRY; CL_PAGE_HEADER(D_TRACE, env, pg, "%d %d\n", crt, ioret); if (crt == CRT_READ && ioret == 0) { PASSERT(env, pg, !(pg->cp_flags & CPF_READ_COMPLETED)); pg->cp_flags |= CPF_READ_COMPLETED; } cl_page_state_set(env, pg, CPS_CACHED); CL_PAGE_INVOID_REVERSE(env, pg, CL_PAGE_OP(io[crt].cpo_completion), (const struct lu_env *, const struct cl_page_slice *, int), ioret); if (anchor) { LASSERT(cl_page_is_vmlocked(env, pg)); LASSERT(pg->cp_sync_io == anchor); pg->cp_sync_io = NULL; cl_sync_io_note(anchor, ioret); } /* Don't assert the page writeback bit here because the lustre file * may be as a backend of swap space. in this case, the page writeback * is set by VM, and obvious we shouldn't clear it at all. Fortunately * this type of pages are all TRANSIENT pages. */ KLASSERT(ergo(pg->cp_type == CPT_CACHEABLE, !PageWriteback(cl_page_vmpage(env, pg)))); EXIT; } EXPORT_SYMBOL(cl_page_completion); /** * Notify layers that transfer formation engine decided to yank this page from * the cache and to make it a part of a transfer. * * \pre pg->cp_state == CPS_CACHED * \post pg->cp_state == CPS_PAGEIN || pg->cp_state == CPS_PAGEOUT * * \see cl_page_operations::cpo_make_ready() */ int cl_page_make_ready(const struct lu_env *env, struct cl_page *pg, enum cl_req_type crt) { int result; PINVRNT(env, pg, crt < CRT_NR); ENTRY; result = CL_PAGE_INVOKE(env, pg, CL_PAGE_OP(io[crt].cpo_make_ready), (const struct lu_env *, const struct cl_page_slice *)); if (result == 0) { PASSERT(env, pg, pg->cp_state == CPS_CACHED); cl_page_io_start(env, pg, crt); } CL_PAGE_HEADER(D_TRACE, env, pg, "%d %d\n", crt, result); RETURN(result); } EXPORT_SYMBOL(cl_page_make_ready); /** * Notify layers that high level io decided to place this page into a cache * for future transfer. * * The layer implementing transfer engine (osc) has to register this page in * its queues. * * \pre cl_page_is_owned(pg, io) * \post cl_page_is_owned(pg, io) * * \see cl_page_operations::cpo_cache_add() */ int cl_page_cache_add(const struct lu_env *env, struct cl_io *io, struct cl_page *pg, enum cl_req_type crt) { const struct cl_page_slice *scan; int result = 0; PINVRNT(env, pg, crt < CRT_NR); PINVRNT(env, pg, cl_page_is_owned(pg, io)); PINVRNT(env, pg, cl_page_invariant(pg)); ENTRY; cfs_list_for_each_entry(scan, &pg->cp_layers, cpl_linkage) { if (scan->cpl_ops->io[crt].cpo_cache_add == NULL) continue; result = scan->cpl_ops->io[crt].cpo_cache_add(env, scan, io); if (result != 0) break; } CL_PAGE_HEADER(D_TRACE, env, pg, "%d %d\n", crt, result); RETURN(result); } EXPORT_SYMBOL(cl_page_cache_add); /** * Called if a pge is being written back by kernel's intention. * * \pre cl_page_is_owned(pg, io) * \post ergo(result == 0, pg->cp_state == CPS_PAGEOUT) * * \see cl_page_operations::cpo_flush() */ int cl_page_flush(const struct lu_env *env, struct cl_io *io, struct cl_page *pg) { int result; PINVRNT(env, pg, cl_page_is_owned(pg, io)); PINVRNT(env, pg, cl_page_invariant(pg)); ENTRY; result = cl_page_invoke(env, io, pg, CL_PAGE_OP(cpo_flush)); CL_PAGE_HEADER(D_TRACE, env, pg, "%d\n", result); RETURN(result); } EXPORT_SYMBOL(cl_page_flush); /** * Checks whether page is protected by any extent lock is at least required * mode. * * \return the same as in cl_page_operations::cpo_is_under_lock() method. * \see cl_page_operations::cpo_is_under_lock() */ int cl_page_is_under_lock(const struct lu_env *env, struct cl_io *io, struct cl_page *page) { int rc; PINVRNT(env, page, cl_page_invariant(page)); ENTRY; rc = CL_PAGE_INVOKE(env, page, CL_PAGE_OP(cpo_is_under_lock), (const struct lu_env *, const struct cl_page_slice *, struct cl_io *), io); PASSERT(env, page, rc != 0); RETURN(rc); } EXPORT_SYMBOL(cl_page_is_under_lock); static int page_prune_cb(const struct lu_env *env, struct cl_io *io, struct cl_page *page, void *cbdata) { cl_page_own(env, io, page); cl_page_unmap(env, io, page); cl_page_discard(env, io, page); cl_page_disown(env, io, page); return CLP_GANG_OKAY; } /** * Purges all cached pages belonging to the object \a obj. */ int cl_pages_prune(const struct lu_env *env, struct cl_object *clobj) { struct cl_thread_info *info; struct cl_object *obj = cl_object_top(clobj); struct cl_io *io; int result; ENTRY; info = cl_env_info(env); io = &info->clt_io; /* * initialize the io. This is ugly since we never do IO in this * function, we just make cl_page_list functions happy. -jay */ io->ci_obj = obj; result = cl_io_init(env, io, CIT_MISC, obj); if (result != 0) { cl_io_fini(env, io); RETURN(io->ci_result); } do { result = cl_page_gang_lookup(env, obj, io, 0, CL_PAGE_EOF, page_prune_cb, NULL); if (result == CLP_GANG_RESCHED) cfs_cond_resched(); } while (result != CLP_GANG_OKAY); cl_io_fini(env, io); RETURN(result); } EXPORT_SYMBOL(cl_pages_prune); /** * Tells transfer engine that only part of a page is to be transmitted. * * \see cl_page_operations::cpo_clip() */ void cl_page_clip(const struct lu_env *env, struct cl_page *pg, int from, int to) { PINVRNT(env, pg, cl_page_invariant(pg)); CL_PAGE_HEADER(D_TRACE, env, pg, "%d %d\n", from, to); CL_PAGE_INVOID(env, pg, CL_PAGE_OP(cpo_clip), (const struct lu_env *, const struct cl_page_slice *,int, int), from, to); } EXPORT_SYMBOL(cl_page_clip); /** * Prints human readable representation of \a pg to the \a f. */ void cl_page_header_print(const struct lu_env *env, void *cookie, lu_printer_t printer, const struct cl_page *pg) { (*printer)(env, cookie, "page@%p[%d %p:%lu ^%p_%p %d %d %d %p %p %#x]\n", pg, cfs_atomic_read(&pg->cp_ref), pg->cp_obj, pg->cp_index, pg->cp_parent, pg->cp_child, pg->cp_state, pg->cp_error, pg->cp_type, pg->cp_owner, pg->cp_req, pg->cp_flags); } EXPORT_SYMBOL(cl_page_header_print); /** * Prints human readable representation of \a pg to the \a f. */ void cl_page_print(const struct lu_env *env, void *cookie, lu_printer_t printer, const struct cl_page *pg) { struct cl_page *scan; for (scan = cl_page_top((struct cl_page *)pg); scan != NULL; scan = scan->cp_child) cl_page_header_print(env, cookie, printer, scan); CL_PAGE_INVOKE(env, (struct cl_page *)pg, CL_PAGE_OP(cpo_print), (const struct lu_env *env, const struct cl_page_slice *slice, void *cookie, lu_printer_t p), cookie, printer); (*printer)(env, cookie, "end page@%p\n", pg); } EXPORT_SYMBOL(cl_page_print); /** * Cancel a page which is still in a transfer. */ int cl_page_cancel(const struct lu_env *env, struct cl_page *page) { return CL_PAGE_INVOKE(env, page, CL_PAGE_OP(cpo_cancel), (const struct lu_env *, const struct cl_page_slice *)); } EXPORT_SYMBOL(cl_page_cancel); /** * Converts a byte offset within object \a obj into a page index. */ loff_t cl_offset(const struct cl_object *obj, pgoff_t idx) { /* * XXX for now. */ return (loff_t)idx << CFS_PAGE_SHIFT; } EXPORT_SYMBOL(cl_offset); /** * Converts a page index into a byte offset within object \a obj. */ pgoff_t cl_index(const struct cl_object *obj, loff_t offset) { /* * XXX for now. */ return offset >> CFS_PAGE_SHIFT; } EXPORT_SYMBOL(cl_index); int cl_page_size(const struct cl_object *obj) { return 1 << CFS_PAGE_SHIFT; } EXPORT_SYMBOL(cl_page_size); /** * Adds page slice to the compound page. * * This is called by cl_object_operations::coo_page_init() methods to add a * per-layer state to the page. New state is added at the end of * cl_page::cp_layers list, that is, it is at the bottom of the stack. * * \see cl_lock_slice_add(), cl_req_slice_add(), cl_io_slice_add() */ void cl_page_slice_add(struct cl_page *page, struct cl_page_slice *slice, struct cl_object *obj, const struct cl_page_operations *ops) { ENTRY; cfs_list_add_tail(&slice->cpl_linkage, &page->cp_layers); slice->cpl_obj = obj; slice->cpl_ops = ops; slice->cpl_page = page; EXIT; } EXPORT_SYMBOL(cl_page_slice_add); int cl_page_init(void) { return lu_kmem_init(cl_page_caches); } void cl_page_fini(void) { lu_kmem_fini(cl_page_caches); }