X-Git-Url: https://git.whamcloud.com/?a=blobdiff_plain;f=lustre%2Flov%2Flov_object.c;h=afe3b9f532070058b879dcf4cb03486ec53fed25;hb=49b17944e1a61f88bddb5595bb053a555c8c08da;hp=eecfbcbf47b903c36915972f574ff51b72fe4578;hpb=e6136f178b3cba4b1257851a85fd784792297f86;p=fs%2Flustre-release.git diff --git a/lustre/lov/lov_object.c b/lustre/lov/lov_object.c index eecfbcb..afe3b9f 100644 --- a/lustre/lov/lov_object.c +++ b/lustre/lov/lov_object.c @@ -27,7 +27,7 @@ * Copyright (c) 2008, 2010, Oracle and/or its affiliates. All rights reserved. * Use is subject to license terms. * - * Copyright (c) 2011, 2012, Intel Corporation. + * Copyright (c) 2011, 2014, Intel Corporation. */ /* * This file is part of Lustre, http://www.lustre.org/ @@ -42,7 +42,6 @@ #define DEBUG_SUBSYSTEM S_LOV #include "lov_cl_internal.h" -#include /** \addtogroup lov * @{ @@ -68,7 +67,7 @@ struct lov_layout_operations { int (*llo_print)(const struct lu_env *env, void *cookie, lu_printer_t p, const struct lu_object *o); int (*llo_page_init)(const struct lu_env *env, struct cl_object *obj, - struct cl_page *page, cfs_page_t *vmpage); + struct cl_page *page, pgoff_t index); int (*llo_lock_init)(const struct lu_env *env, struct cl_object *obj, struct cl_lock *lock, const struct cl_io *io); @@ -76,10 +75,35 @@ struct lov_layout_operations { struct cl_object *obj, struct cl_io *io); int (*llo_getattr)(const struct lu_env *env, struct cl_object *obj, struct cl_attr *attr); + int (*llo_find_cbdata)(const struct lu_env *env, struct cl_object *obj, + ldlm_iterator_t iter, void *data); }; static int lov_layout_wait(const struct lu_env *env, struct lov_object *lov); +struct lov_stripe_md *lov_lsm_get(struct cl_object *clobj) +{ + struct lu_object *luobj; + struct lov_stripe_md *lsm = NULL; + + if (clobj == NULL) + return NULL; + + luobj = lu_object_locate(&cl_object_header(clobj)->coh_lu, + &lov_device_type); + if (luobj != NULL) + lsm = lov_lsm_addref(lu2lov(luobj)); + return lsm; +} +EXPORT_SYMBOL(lov_lsm_get); + +void lov_lsm_put(struct cl_object *unused, struct lov_stripe_md *lsm) +{ + if (lsm != NULL) + lov_free_memmd(&lsm); +} +EXPORT_SYMBOL(lov_lsm_put); + /***************************************************************************** * * Lov object layout operations. @@ -123,14 +147,14 @@ static struct cl_object *lov_sub_find(const struct lu_env *env, } static int lov_init_sub(const struct lu_env *env, struct lov_object *lov, - struct cl_object *stripe, - struct lov_layout_raid0 *r0, int idx) + struct cl_object *stripe, struct lov_layout_raid0 *r0, + int idx) { - struct cl_object_header *hdr; - struct cl_object_header *subhdr; - struct cl_object_header *parent; - struct lov_oinfo *oinfo; - int result; + struct cl_object_header *hdr; + struct cl_object_header *subhdr; + struct cl_object_header *parent; + struct lov_oinfo *oinfo; + int result; if (OBD_FAIL_CHECK(OBD_FAIL_LOV_INIT)) { /* For sanity:test_206. @@ -143,31 +167,34 @@ static int lov_init_sub(const struct lu_env *env, struct lov_object *lov, return -EIO; } - hdr = cl_object_header(lov2cl(lov)); - subhdr = cl_object_header(stripe); - parent = subhdr->coh_parent; + hdr = cl_object_header(lov2cl(lov)); + subhdr = cl_object_header(stripe); oinfo = lov->lo_lsm->lsm_oinfo[idx]; - CDEBUG(D_INODE, DFID"@%p[%d] -> "DFID"@%p: id: "LPU64" seq: "LPU64 - " idx: %d gen: %d\n", - PFID(&subhdr->coh_lu.loh_fid), subhdr, idx, - PFID(&hdr->coh_lu.loh_fid), hdr, - oinfo->loi_id, oinfo->loi_seq, - oinfo->loi_ost_idx, oinfo->loi_ost_gen); - - if (parent == NULL) { - subhdr->coh_parent = hdr; - subhdr->coh_nesting = hdr->coh_nesting + 1; - lu_object_ref_add(&stripe->co_lu, "lov-parent", lov); - r0->lo_sub[idx] = cl2lovsub(stripe); - r0->lo_sub[idx]->lso_super = lov; - r0->lo_sub[idx]->lso_index = idx; - result = 0; - } else { + CDEBUG(D_INODE, DFID"@%p[%d] -> "DFID"@%p: ostid: "DOSTID + " idx: %d gen: %d\n", + PFID(&subhdr->coh_lu.loh_fid), subhdr, idx, + PFID(&hdr->coh_lu.loh_fid), hdr, POSTID(&oinfo->loi_oi), + oinfo->loi_ost_idx, oinfo->loi_ost_gen); + + /* reuse ->coh_attr_guard to protect coh_parent change */ + spin_lock(&subhdr->coh_attr_guard); + parent = subhdr->coh_parent; + if (parent == NULL) { + subhdr->coh_parent = hdr; + spin_unlock(&subhdr->coh_attr_guard); + subhdr->coh_nesting = hdr->coh_nesting + 1; + lu_object_ref_add(&stripe->co_lu, "lov-parent", lov); + r0->lo_sub[idx] = cl2lovsub(stripe); + r0->lo_sub[idx]->lso_super = lov; + r0->lo_sub[idx]->lso_index = idx; + result = 0; + } else { struct lu_object *old_obj; struct lov_object *old_lov; unsigned int mask = D_INODE; + spin_unlock(&subhdr->coh_attr_guard); old_obj = lu_object_locate(&parent->coh_lu, &lov_device_type); LASSERT(old_obj != NULL); old_lov = cl2lov(lu2cl(old_obj)); @@ -186,8 +213,24 @@ static int lov_init_sub(const struct lu_env *env, struct lov_object *lov, LU_OBJECT_DEBUG(mask, env, old_obj, "owned.\n"); LU_OBJECT_HEADER(mask, env, lov2lu(lov), "try to own.\n"); cl_object_put(env, stripe); - } - return result; + } + return result; +} + +static int lov_page_slice_fixup(struct lov_object *lov, + struct cl_object *stripe) +{ + struct cl_object_header *hdr = cl_object_header(&lov->lo_cl); + struct cl_object *o; + + if (stripe == NULL) + return hdr->coh_page_bufsize - lov->lo_cl.co_slice_off - + cfs_size_round(sizeof(struct lov_page)); + + cl_object_for_each(o, stripe) + o->co_slice_off += hdr->coh_page_bufsize; + + return cl_object_header(stripe)->coh_page_bufsize; } static int lov_init_raid0(const struct lu_env *env, @@ -216,12 +259,16 @@ static int lov_init_raid0(const struct lu_env *env, LASSERT(lov->lo_lsm == NULL); lov->lo_lsm = lsm_addref(lsm); r0->lo_nr = lsm->lsm_stripe_count; - LASSERT(r0->lo_nr <= lov_targets_nr(dev)); + LASSERT(r0->lo_nr <= lov_targets_nr(dev)); - OBD_ALLOC_LARGE(r0->lo_sub, r0->lo_nr * sizeof r0->lo_sub[0]); - if (r0->lo_sub != NULL) { - result = 0; - subconf->coc_inode = conf->coc_inode; + lov->lo_layout_invalid = true; + + OBD_ALLOC_LARGE(r0->lo_sub, r0->lo_nr * sizeof r0->lo_sub[0]); + if (r0->lo_sub != NULL) { + int psz = 0; + + result = 0; + subconf->coc_inode = conf->coc_inode; spin_lock_init(&r0->lo_sub_lock); /* * Create stripe cl_objects. @@ -231,49 +278,108 @@ static int lov_init_raid0(const struct lu_env *env, struct lov_oinfo *oinfo = lsm->lsm_oinfo[i]; int ost_idx = oinfo->loi_ost_idx; - fid_ostid_unpack(ofid, &oinfo->loi_oi, - oinfo->loi_ost_idx); - subdev = lovsub2cl_dev(dev->ld_target[ost_idx]); - subconf->u.coc_oinfo = oinfo; - LASSERTF(subdev != NULL, "not init ost %d\n", ost_idx); + if (lov_oinfo_is_dummy(oinfo)) + continue; + + result = ostid_to_fid(ofid, &oinfo->loi_oi, + oinfo->loi_ost_idx); + if (result != 0) + GOTO(out, result); + + subdev = lovsub2cl_dev(dev->ld_target[ost_idx]); + subconf->u.coc_oinfo = oinfo; + LASSERTF(subdev != NULL, "not init ost %d\n", ost_idx); /* In the function below, .hs_keycmp resolves to * lu_obj_hop_keycmp() */ /* coverity[overrun-buffer-val] */ - stripe = lov_sub_find(env, subdev, ofid, subconf); - if (!IS_ERR(stripe)) { - result = lov_init_sub(env, lov, stripe, r0, i); + stripe = lov_sub_find(env, subdev, ofid, subconf); + if (!IS_ERR(stripe)) { + result = lov_init_sub(env, lov, stripe, r0, i); if (result == -EAGAIN) { /* try again */ --i; result = 0; + continue; } - } else { - result = PTR_ERR(stripe); + } else { + result = PTR_ERR(stripe); + } + + if (result == 0) { + int sz = lov_page_slice_fixup(lov, stripe); + LASSERT(ergo(psz > 0, psz == sz)); + psz = sz; } } - } else - result = -ENOMEM; - RETURN(result); + if (result == 0) + cl_object_header(&lov->lo_cl)->coh_page_bufsize += psz; + } else + result = -ENOMEM; +out: + RETURN(result); +} + +static int lov_init_released(const struct lu_env *env, + struct lov_device *dev, struct lov_object *lov, + const struct cl_object_conf *conf, + union lov_layout_state *state) +{ + struct lov_stripe_md *lsm = conf->u.coc_md->lsm; + + LASSERT(lsm != NULL); + LASSERT(lsm_is_released(lsm)); + LASSERT(lov->lo_lsm == NULL); + + lov->lo_lsm = lsm_addref(lsm); + return 0; +} + +static struct cl_object *lov_find_subobj(const struct lu_env *env, + struct lov_object *lov, + struct lov_stripe_md *lsm, + int stripe_idx) +{ + struct lov_device *dev = lu2lov_dev(lov2lu(lov)->lo_dev); + struct lov_oinfo *oinfo = lsm->lsm_oinfo[stripe_idx]; + struct lov_thread_info *lti = lov_env_info(env); + struct lu_fid *ofid = <i->lti_fid; + struct cl_device *subdev; + int ost_idx; + int rc; + struct cl_object *result; + + if (lov->lo_type != LLT_RAID0) + GOTO(out, result = NULL); + + ost_idx = oinfo->loi_ost_idx; + rc = ostid_to_fid(ofid, &oinfo->loi_oi, ost_idx); + if (rc != 0) + GOTO(out, result = NULL); + + subdev = lovsub2cl_dev(dev->ld_target[ost_idx]); + result = lov_sub_find(env, subdev, ofid, NULL); +out: + if (result == NULL) + result = ERR_PTR(-EINVAL); + return result; } static int lov_delete_empty(const struct lu_env *env, struct lov_object *lov, union lov_layout_state *state) { - LASSERT(lov->lo_type == LLT_EMPTY); + LASSERT(lov->lo_type == LLT_EMPTY || lov->lo_type == LLT_RELEASED); lov_layout_wait(env, lov); - - cl_object_prune(env, &lov->lo_cl); return 0; } static void lov_subobject_kill(const struct lu_env *env, struct lov_object *lov, - struct lovsub_object *los, int idx) + struct lovsub_object *los, int idx) { - struct cl_object *sub; - struct lov_layout_raid0 *r0; - struct lu_site *site; - struct lu_site_bkt_data *bkt; - cfs_waitlink_t *waiter; + struct cl_object *sub; + struct lov_layout_raid0 *r0; + struct lu_site *site; + struct lu_site_bkt_data *bkt; + wait_queue_t *waiter; r0 = &lov->u.raid0; LASSERT(r0->lo_sub[idx] == los); @@ -289,28 +395,28 @@ static void lov_subobject_kill(const struct lu_env *env, struct lov_object *lov, /* ... wait until it is actually destroyed---sub-object clears its * ->lo_sub[] slot in lovsub_object_fini() */ - if (r0->lo_sub[idx] == los) { - waiter = &lov_env_info(env)->lti_waiter; - cfs_waitlink_init(waiter); - cfs_waitq_add(&bkt->lsb_marche_funebre, waiter); - cfs_set_current_state(CFS_TASK_UNINT); - while (1) { - /* this wait-queue is signaled at the end of - * lu_object_free(). */ - cfs_set_current_state(CFS_TASK_UNINT); + if (r0->lo_sub[idx] == los) { + waiter = &lov_env_info(env)->lti_waiter; + init_waitqueue_entry_current(waiter); + add_wait_queue(&bkt->lsb_marche_funebre, waiter); + set_current_state(TASK_UNINTERRUPTIBLE); + while (1) { + /* this wait-queue is signaled at the end of + * lu_object_free(). */ + set_current_state(TASK_UNINTERRUPTIBLE); spin_lock(&r0->lo_sub_lock); if (r0->lo_sub[idx] == los) { spin_unlock(&r0->lo_sub_lock); - cfs_waitq_wait(waiter, CFS_TASK_UNINT); + waitq_wait(waiter, TASK_UNINTERRUPTIBLE); } else { spin_unlock(&r0->lo_sub_lock); - cfs_set_current_state(CFS_TASK_RUNNING); - break; - } - } - cfs_waitq_del(&bkt->lsb_marche_funebre, waiter); - } - LASSERT(r0->lo_sub[idx] == NULL); + set_current_state(TASK_RUNNING); + break; + } + } + remove_wait_queue(&bkt->lsb_marche_funebre, waiter); + } + LASSERT(r0->lo_sub[idx] == NULL); } static int lov_delete_raid0(const struct lu_env *env, struct lov_object *lov, @@ -330,23 +436,22 @@ static int lov_delete_raid0(const struct lu_env *env, struct lov_object *lov, struct lovsub_object *los = r0->lo_sub[i]; if (los != NULL) { - cl_locks_prune(env, &los->lso_cl, 1); + cl_object_prune(env, &los->lso_cl); /* * If top-level object is to be evicted from * the cache, so are its sub-objects. */ lov_subobject_kill(env, lov, los, i); } - } - } - cl_object_prune(env, &lov->lo_cl); + } + } RETURN(0); } static void lov_fini_empty(const struct lu_env *env, struct lov_object *lov, union lov_layout_state *state) { - LASSERT(lov->lo_type == LLT_EMPTY); + LASSERT(lov->lo_type == LLT_EMPTY || lov->lo_type == LLT_RELEASED); } static void lov_fini_raid0(const struct lu_env *env, struct lov_object *lov, @@ -366,6 +471,15 @@ static void lov_fini_raid0(const struct lu_env *env, struct lov_object *lov, EXIT; } +static void lov_fini_released(const struct lu_env *env, struct lov_object *lov, + union lov_layout_state *state) +{ + ENTRY; + dump_lsm(D_INODE, lov->lo_lsm); + lov_free_memmd(&lov->lo_lsm); + EXIT; +} + static int lov_print_empty(const struct lu_env *env, void *cookie, lu_printer_t p, const struct lu_object *o) { @@ -374,27 +488,42 @@ static int lov_print_empty(const struct lu_env *env, void *cookie, } static int lov_print_raid0(const struct lu_env *env, void *cookie, - lu_printer_t p, const struct lu_object *o) + lu_printer_t p, const struct lu_object *o) { - struct lov_object *lov = lu2lov(o); - struct lov_layout_raid0 *r0 = lov_r0(lov); - struct lov_stripe_md *lsm = lov->lo_lsm; - int i; + struct lov_object *lov = lu2lov(o); + struct lov_layout_raid0 *r0 = lov_r0(lov); + struct lov_stripe_md *lsm = lov->lo_lsm; + int i; + + (*p)(env, cookie, "stripes: %d, %s, lsm{%p 0x%08X %d %u %u}:\n", + r0->lo_nr, lov->lo_layout_invalid ? "invalid" : "valid", lsm, + lsm->lsm_magic, atomic_read(&lsm->lsm_refc), + lsm->lsm_stripe_count, lsm->lsm_layout_gen); + for (i = 0; i < r0->lo_nr; ++i) { + struct lu_object *sub; + + if (r0->lo_sub[i] != NULL) { + sub = lovsub2lu(r0->lo_sub[i]); + lu_object_print(env, cookie, p, sub); + } else { + (*p)(env, cookie, "sub %d absent\n", i); + } + } + return 0; +} - (*p)(env, cookie, "stripes: %d, %svalid, lsm{%p 0x%08X %d %u %u}: \n", - r0->lo_nr, lov->lo_layout_invalid ? "in" : "", lsm, - lsm->lsm_magic, cfs_atomic_read(&lsm->lsm_refc), +static int lov_print_released(const struct lu_env *env, void *cookie, + lu_printer_t p, const struct lu_object *o) +{ + struct lov_object *lov = lu2lov(o); + struct lov_stripe_md *lsm = lov->lo_lsm; + + (*p)(env, cookie, + "released: %s, lsm{%p 0x%08X %d %u %u}:\n", + lov->lo_layout_invalid ? "invalid" : "valid", lsm, + lsm->lsm_magic, atomic_read(&lsm->lsm_refc), lsm->lsm_stripe_count, lsm->lsm_layout_gen); - for (i = 0; i < r0->lo_nr; ++i) { - struct lu_object *sub; - - if (r0->lo_sub[i] != NULL) { - sub = lovsub2lu(r0->lo_sub[i]); - lu_object_print(env, cookie, p, sub); - } else - (*p)(env, cookie, "sub %d absent\n", i); - } - return 0; + return 0; } /** @@ -429,7 +558,7 @@ static int lov_attr_get_raid0(const struct lu_env *env, struct cl_object *obj, * hit this assertion. * Anyway, it's still okay to call attr_get w/o type guard as layout * can't go if locks exist. */ - /* LASSERT(cfs_atomic_read(&lsm->lsm_refc) > 1); */ + /* LASSERT(atomic_read(&lsm->lsm_refc) > 1); */ if (!r0->lo_attr_valid) { struct lov_stripe_md *lsm = lov->lo_lsm; @@ -478,6 +607,37 @@ static int lov_attr_get_raid0(const struct lu_env *env, struct cl_object *obj, RETURN(result); } +static int lov_find_cbdata_empty(const struct lu_env *env, + struct cl_object *obj, ldlm_iterator_t iter, + void *data) +{ + return 0; +} + +static int lov_find_cbdata_raid0(const struct lu_env *env, + struct cl_object *obj, ldlm_iterator_t iter, + void *data) +{ + struct lov_object *lov = cl2lov(obj); + struct lov_layout_raid0 *r0 = lov_r0(lov); + struct cl_object *subobj; + int i; + int rc = 0; + + for (i = 0; i < r0->lo_nr; ++i) { + if (r0->lo_sub[i] == NULL) + continue; + + subobj = lovsub2cl(r0->lo_sub[i]); + + rc = cl_object_find_cbdata(env, subobj, iter, data); + if (rc != 0) + break; + } + + return rc; +} + const static struct lov_layout_operations lov_dispatch[] = { [LLT_EMPTY] = { .llo_init = lov_init_empty, @@ -488,7 +648,8 @@ const static struct lov_layout_operations lov_dispatch[] = { .llo_page_init = lov_page_init_empty, .llo_lock_init = lov_lock_init_empty, .llo_io_init = lov_io_init_empty, - .llo_getattr = lov_attr_get_empty + .llo_getattr = lov_attr_get_empty, + .llo_find_cbdata = lov_find_cbdata_empty }, [LLT_RAID0] = { .llo_init = lov_init_raid0, @@ -499,11 +660,23 @@ const static struct lov_layout_operations lov_dispatch[] = { .llo_page_init = lov_page_init_raid0, .llo_lock_init = lov_lock_init_raid0, .llo_io_init = lov_io_init_raid0, - .llo_getattr = lov_attr_get_raid0 + .llo_getattr = lov_attr_get_raid0, + .llo_find_cbdata = lov_find_cbdata_raid0 + }, + [LLT_RELEASED] = { + .llo_init = lov_init_released, + .llo_delete = lov_delete_empty, + .llo_fini = lov_fini_released, + .llo_install = lov_install_empty, + .llo_print = lov_print_released, + .llo_page_init = lov_page_init_empty, + .llo_lock_init = lov_lock_init_empty, + .llo_io_init = lov_io_init_released, + .llo_getattr = lov_attr_get_empty, + .llo_find_cbdata = lov_find_cbdata_empty } }; - /** * Performs a double-dispatch based on the layout type of an object. */ @@ -517,15 +690,27 @@ const static struct lov_layout_operations lov_dispatch[] = { lov_dispatch[__llt].op(__VA_ARGS__); \ }) +/** + * Return lov_layout_type associated with a given lsm + */ +static enum lov_layout_type lov_type(struct lov_stripe_md *lsm) +{ + if (lsm == NULL) + return LLT_EMPTY; + if (lsm_is_released(lsm)) + return LLT_RELEASED; + return LLT_RAID0; +} + static inline void lov_conf_freeze(struct lov_object *lov) { - if (lov->lo_owner != cfs_current()) + if (lov->lo_owner != current) down_read(&lov->lo_type_guard); } static inline void lov_conf_thaw(struct lov_object *lov) { - if (lov->lo_owner != cfs_current()) + if (lov->lo_owner != current) up_read(&lov->lo_type_guard); } @@ -563,10 +748,10 @@ do { \ static void lov_conf_lock(struct lov_object *lov) { - LASSERT(lov->lo_owner != cfs_current()); + LASSERT(lov->lo_owner != current); down_write(&lov->lo_type_guard); LASSERT(lov->lo_owner == NULL); - lov->lo_owner = cfs_current(); + lov->lo_owner = current; } static void lov_conf_unlock(struct lov_object *lov) @@ -580,20 +765,20 @@ static int lov_layout_wait(const struct lu_env *env, struct lov_object *lov) struct l_wait_info lwi = { 0 }; ENTRY; - while (cfs_atomic_read(&lov->lo_active_ios) > 0) { + while (atomic_read(&lov->lo_active_ios) > 0) { CDEBUG(D_INODE, "file:"DFID" wait for active IO, now: %d.\n", PFID(lu_object_fid(lov2lu(lov))), - cfs_atomic_read(&lov->lo_active_ios)); + atomic_read(&lov->lo_active_ios)); l_wait_event(lov->lo_waitq, - cfs_atomic_read(&lov->lo_active_ios) == 0, &lwi); + atomic_read(&lov->lo_active_ios) == 0, &lwi); } RETURN(0); } static int lov_layout_change(const struct lu_env *unused, - struct lov_object *lov, - const struct cl_object_conf *conf) + struct lov_object *lov, + const struct cl_object_conf *conf) { int result; enum lov_layout_type llt = LLT_EMPTY; @@ -601,7 +786,6 @@ static int lov_layout_change(const struct lu_env *unused, const struct lov_layout_operations *old_ops; const struct lov_layout_operations *new_ops; - struct cl_object_header *hdr = cl_object_header(&lov->lo_cl); void *cookie; struct lu_env *env; int refcheck; @@ -609,8 +793,8 @@ static int lov_layout_change(const struct lu_env *unused, LASSERT(0 <= lov->lo_type && lov->lo_type < ARRAY_SIZE(lov_dispatch)); - if (conf->u.coc_md != NULL && conf->u.coc_md->lsm != NULL) - llt = LLT_RAID0; /* only raid0 is supported. */ + if (conf->u.coc_md != NULL) + llt = lov_type(conf->u.coc_md->lsm); LASSERT(0 <= llt && llt < ARRAY_SIZE(lov_dispatch)); cookie = cl_env_reenter(); @@ -620,18 +804,28 @@ static int lov_layout_change(const struct lu_env *unused, RETURN(PTR_ERR(env)); } + CDEBUG(D_INODE, DFID" from %s to %s\n", + PFID(lu_object_fid(lov2lu(lov))), + llt2str(lov->lo_type), llt2str(llt)); + old_ops = &lov_dispatch[lov->lo_type]; new_ops = &lov_dispatch[llt]; + result = cl_object_prune(env, &lov->lo_cl); + if (result != 0) + GOTO(out, result); + result = old_ops->llo_delete(env, lov, &lov->u); if (result == 0) { old_ops->llo_fini(env, lov, &lov->u); - LASSERT(cfs_atomic_read(&lov->lo_active_ios) == 0); - LASSERT(hdr->coh_tree.rnode == NULL); - LASSERT(hdr->coh_pages == 0); + LASSERT(atomic_read(&lov->lo_active_ios) == 0); lov->lo_type = LLT_EMPTY; + /* page bufsize fixup */ + cl_object_header(&lov->lo_cl)->coh_page_bufsize -= + lov_page_slice_fixup(lov, NULL); + result = new_ops->llo_init(env, lu2lov_dev(lov->lo_cl.co_lu.lo_dev), lov, conf, state); @@ -645,6 +839,7 @@ static int lov_layout_change(const struct lu_env *unused, } } +out: cl_env_put(env, &refcheck); cl_env_reexit(cookie); RETURN(result); @@ -655,7 +850,6 @@ static int lov_layout_change(const struct lu_env *unused, * Lov object operations. * */ - int lov_object_init(const struct lu_env *env, struct lu_object *obj, const struct lu_object_conf *conf) { @@ -668,13 +862,13 @@ int lov_object_init(const struct lu_env *env, struct lu_object *obj, ENTRY; init_rwsem(&lov->lo_type_guard); - cfs_atomic_set(&lov->lo_active_ios, 0); - cfs_waitq_init(&lov->lo_waitq); + atomic_set(&lov->lo_active_ios, 0); + init_waitqueue_head(&lov->lo_waitq); cl_object_page_init(lu2cl(obj), sizeof(struct lov_page)); /* no locking is necessary, as object is being created */ - lov->lo_type = cconf->u.coc_md->lsm != NULL ? LLT_RAID0 : LLT_EMPTY; + lov->lo_type = lov_type(cconf->u.coc_md->lsm); ops = &lov_dispatch[lov->lo_type]; result = ops->llo_init(env, dev, lov, cconf, set); if (result == 0) @@ -685,9 +879,9 @@ int lov_object_init(const struct lu_env *env, struct lu_object *obj, static int lov_conf_set(const struct lu_env *env, struct cl_object *obj, const struct cl_object_conf *conf) { - struct lov_stripe_md *lsm = NULL; - struct lov_object *lov = cl2lov(obj); - int result = 0; + struct lov_stripe_md *lsm = NULL; + struct lov_object *lov = cl2lov(obj); + int result = 0; ENTRY; lov_conf_lock(lov); @@ -698,7 +892,7 @@ static int lov_conf_set(const struct lu_env *env, struct cl_object *obj, if (conf->coc_opc == OBJECT_CONF_WAIT) { if (lov->lo_layout_invalid && - cfs_atomic_read(&lov->lo_active_ios) > 0) { + atomic_read(&lov->lo_active_ios) > 0) { lov_conf_unlock(lov); result = lov_layout_wait(env, lov); lov_conf_lock(lov); @@ -711,24 +905,28 @@ static int lov_conf_set(const struct lu_env *env, struct cl_object *obj, if (conf->u.coc_md != NULL) lsm = conf->u.coc_md->lsm; if ((lsm == NULL && lov->lo_lsm == NULL) || - (lsm != NULL && lov->lo_lsm != NULL && - lov->lo_lsm->lsm_layout_gen == lsm->lsm_layout_gen)) { + ((lsm != NULL && lov->lo_lsm != NULL) && + (lov->lo_lsm->lsm_layout_gen == lsm->lsm_layout_gen) && + (lov->lo_lsm->lsm_pattern == lsm->lsm_pattern))) { /* same version of layout */ lov->lo_layout_invalid = false; GOTO(out, result = 0); } /* will change layout - check if there still exists active IO. */ - if (cfs_atomic_read(&lov->lo_active_ios) > 0) { + if (atomic_read(&lov->lo_active_ios) > 0) { lov->lo_layout_invalid = true; GOTO(out, result = -EBUSY); } - lov->lo_layout_invalid = lov_layout_change(env, lov, conf); + result = lov_layout_change(env, lov, conf); + lov->lo_layout_invalid = result != 0; EXIT; out: lov_conf_unlock(lov); + CDEBUG(D_INODE, DFID" lo_layout_invalid=%d\n", + PFID(lu_object_fid(lov2lu(lov))), lov->lo_layout_invalid); RETURN(result); } @@ -759,10 +957,10 @@ static int lov_object_print(const struct lu_env *env, void *cookie, } int lov_page_init(const struct lu_env *env, struct cl_object *obj, - struct cl_page *page, cfs_page_t *vmpage) + struct cl_page *page, pgoff_t index) { - return LOV_2DISPATCH_NOLOCK(cl2lov(obj), - llo_page_init, env, obj, page, vmpage); + return LOV_2DISPATCH_NOLOCK(cl2lov(obj), llo_page_init, env, obj, page, + index); } /** @@ -790,13 +988,13 @@ static int lov_attr_get(const struct lu_env *env, struct cl_object *obj, return LOV_2DISPATCH_NOLOCK(cl2lov(obj), llo_getattr, env, obj, attr); } -static int lov_attr_set(const struct lu_env *env, struct cl_object *obj, - const struct cl_attr *attr, unsigned valid) +static int lov_attr_update(const struct lu_env *env, struct cl_object *obj, + const struct cl_attr *attr, unsigned valid) { - /* - * No dispatch is required here, as no layout implements this. - */ - return 0; + /* + * No dispatch is required here, as no layout implements this. + */ + return 0; } int lov_lock_init(const struct lu_env *env, struct cl_object *obj, @@ -807,13 +1005,611 @@ int lov_lock_init(const struct lu_env *env, struct cl_object *obj, io); } +/** + * We calculate on which OST the mapping will end. If the length of mapping + * is greater than (stripe_size * stripe_count) then the last_stripe will + * will be one just before start_stripe. Else we check if the mapping + * intersects each OST and find last_stripe. + * This function returns the last_stripe and also sets the stripe_count + * over which the mapping is spread + * + * \param lsm [in] striping information for the file + * \param fm_start [in] logical start of mapping + * \param fm_end [in] logical end of mapping + * \param start_stripe [in] starting stripe of the mapping + * \param stripe_count [out] the number of stripes across which to map is + * returned + * + * \retval last_stripe return the last stripe of the mapping + */ +static int fiemap_calc_last_stripe(struct lov_stripe_md *lsm, + loff_t fm_start, loff_t fm_end, + int start_stripe, int *stripe_count) +{ + int last_stripe; + loff_t obd_start; + loff_t obd_end; + int i, j; + + if (fm_end - fm_start > lsm->lsm_stripe_size * lsm->lsm_stripe_count) { + last_stripe = (start_stripe < 1 ? lsm->lsm_stripe_count - 1 : + start_stripe - 1); + *stripe_count = lsm->lsm_stripe_count; + } else { + for (j = 0, i = start_stripe; j < lsm->lsm_stripe_count; + i = (i + 1) % lsm->lsm_stripe_count, j++) { + if ((lov_stripe_intersects(lsm, i, fm_start, fm_end, + &obd_start, &obd_end)) == 0) + break; + } + *stripe_count = j; + last_stripe = (start_stripe + j - 1) % lsm->lsm_stripe_count; + } + + return last_stripe; +} + +/** + * Set fe_device and copy extents from local buffer into main return buffer. + * + * \param fiemap [out] fiemap to hold all extents + * \param lcl_fm_ext [in] array of fiemap extents get from OSC layer + * \param ost_index [in] OST index to be written into the fm_device + * field for each extent + * \param ext_count [in] number of extents to be copied + * \param current_extent [in] where to start copying in the extent array + */ +static void fiemap_prepare_and_copy_exts(struct fiemap *fiemap, + struct ll_fiemap_extent *lcl_fm_ext, + int ost_index, unsigned int ext_count, + int current_extent) +{ + char *to; + unsigned int ext; + + for (ext = 0; ext < ext_count; ext++) { + lcl_fm_ext[ext].fe_device = ost_index; + lcl_fm_ext[ext].fe_flags |= FIEMAP_EXTENT_NET; + } + + /* Copy fm_extent's from fm_local to return buffer */ + to = (char *)fiemap + fiemap_count_to_size(current_extent); + memcpy(to, lcl_fm_ext, ext_count * sizeof(struct ll_fiemap_extent)); +} + +#define FIEMAP_BUFFER_SIZE 4096 + +/** + * Non-zero fe_logical indicates that this is a continuation FIEMAP + * call. The local end offset and the device are sent in the first + * fm_extent. This function calculates the stripe number from the index. + * This function returns a stripe_no on which mapping is to be restarted. + * + * This function returns fm_end_offset which is the in-OST offset at which + * mapping should be restarted. If fm_end_offset=0 is returned then caller + * will re-calculate proper offset in next stripe. + * Note that the first extent is passed to lov_get_info via the value field. + * + * \param fiemap [in] fiemap request header + * \param lsm [in] striping information for the file + * \param fm_start [in] logical start of mapping + * \param fm_end [in] logical end of mapping + * \param start_stripe [out] starting stripe will be returned in this + */ +static loff_t fiemap_calc_fm_end_offset(struct fiemap *fiemap, + struct lov_stripe_md *lsm, + loff_t fm_start, loff_t fm_end, + int *start_stripe) +{ + loff_t local_end = fiemap->fm_extents[0].fe_logical; + loff_t lun_start; + loff_t lun_end; + loff_t fm_end_offset; + int stripe_no = -1; + int i; + + if (fiemap->fm_extent_count == 0 || + fiemap->fm_extents[0].fe_logical == 0) + return 0; + + /* Find out stripe_no from ost_index saved in the fe_device */ + for (i = 0; i < lsm->lsm_stripe_count; i++) { + struct lov_oinfo *oinfo = lsm->lsm_oinfo[i]; + + if (lov_oinfo_is_dummy(oinfo)) + continue; + + if (oinfo->loi_ost_idx == fiemap->fm_extents[0].fe_device) { + stripe_no = i; + break; + } + } + + if (stripe_no == -1) + return -EINVAL; + + /* If we have finished mapping on previous device, shift logical + * offset to start of next device */ + if (lov_stripe_intersects(lsm, stripe_no, fm_start, fm_end, + &lun_start, &lun_end) != 0 && + local_end < lun_end) { + fm_end_offset = local_end; + *start_stripe = stripe_no; + } else { + /* This is a special value to indicate that caller should + * calculate offset in next stripe. */ + fm_end_offset = 0; + *start_stripe = (stripe_no + 1) % lsm->lsm_stripe_count; + } + + return fm_end_offset; +} + +/** + * Break down the FIEMAP request and send appropriate calls to individual OSTs. + * This also handles the restarting of FIEMAP calls in case mapping overflows + * the available number of extents in single call. + * + * \param env [in] lustre environment + * \param obj [in] file object + * \param fmkey [in] fiemap request header and other info + * \param fiemap [out] fiemap buffer holding retrived map extents + * \param buflen [in/out] max buffer length of @fiemap, when iterate + * each OST, it is used to limit max map needed + * \retval 0 success + * \retval < 0 error + */ +static int lov_object_fiemap(const struct lu_env *env, struct cl_object *obj, + struct ll_fiemap_info_key *fmkey, + struct fiemap *fiemap, size_t *buflen) +{ + struct lov_stripe_md *lsm; + struct cl_object *subobj = NULL; + struct lov_obd *lov = lu2lov_dev(obj->co_lu.lo_dev)->ld_lov; + struct fiemap *fm_local = NULL; + struct ll_fiemap_extent *lcl_fm_ext; + loff_t fm_start; + loff_t fm_end; + loff_t fm_length; + loff_t fm_end_offset; + int count_local; + int ost_index = 0; + int start_stripe; + int current_extent = 0; + int rc = 0; + int last_stripe; + int cur_stripe = 0; + int cur_stripe_wrap = 0; + int stripe_count; + unsigned int buffer_size = FIEMAP_BUFFER_SIZE; + /* Whether have we collected enough extents */ + bool enough = false; + /* EOF for object */ + bool ost_eof = false; + /* done with required mapping for this OST? */ + bool ost_done = false; + ENTRY; + + lsm = lov_lsm_addref(cl2lov(obj)); + if (lsm == NULL) + RETURN(-ENODATA); + + /** + * If the stripe_count > 1 and the application does not understand + * DEVICE_ORDER flag, it cannot interpret the extents correctly. + */ + if (lsm->lsm_stripe_count > 1 && !(fiemap->fm_flags & + FIEMAP_FLAG_DEVICE_ORDER)) + GOTO(out, rc = -ENOTSUPP); + + if (lsm_is_released(lsm)) { + if (fiemap->fm_start < fmkey->oa.o_size) { + /** + * released file, return a minimal FIEMAP if + * request fits in file-size. + */ + fiemap->fm_mapped_extents = 1; + fiemap->fm_extents[0].fe_logical = fiemap->fm_start; + if (fiemap->fm_start + fiemap->fm_length < + fmkey->oa.o_size) + fiemap->fm_extents[0].fe_length = + fiemap->fm_length; + else + fiemap->fm_extents[0].fe_length = + fmkey->oa.o_size - fiemap->fm_start; + fiemap->fm_extents[0].fe_flags |= + FIEMAP_EXTENT_UNKNOWN | FIEMAP_EXTENT_LAST; + } + GOTO(out, rc = 0); + } + + if (fiemap_count_to_size(fiemap->fm_extent_count) < buffer_size) + buffer_size = fiemap_count_to_size(fiemap->fm_extent_count); + + OBD_ALLOC_LARGE(fm_local, buffer_size); + if (fm_local == NULL) + GOTO(out, rc = -ENOMEM); + lcl_fm_ext = &fm_local->fm_extents[0]; + count_local = fiemap_size_to_count(buffer_size); + + fm_start = fiemap->fm_start; + fm_length = fiemap->fm_length; + /* Calculate start stripe, last stripe and length of mapping */ + start_stripe = lov_stripe_number(lsm, fm_start); + fm_end = (fm_length == ~0ULL) ? fmkey->oa.o_size : + fm_start + fm_length - 1; + /* If fm_length != ~0ULL but fm_start_fm_length-1 exceeds file size */ + if (fm_end > fmkey->oa.o_size) + fm_end = fmkey->oa.o_size; + + last_stripe = fiemap_calc_last_stripe(lsm, fm_start, fm_end, + start_stripe, &stripe_count); + fm_end_offset = fiemap_calc_fm_end_offset(fiemap, lsm, fm_start, fm_end, + &start_stripe); + if (fm_end_offset == -EINVAL) + GOTO(out, rc = -EINVAL); + + /** + * Requested extent count exceeds the fiemap buffer size, shrink our + * ambition. + */ + if (fiemap_count_to_size(fiemap->fm_extent_count) > *buflen) + fiemap->fm_extent_count = fiemap_size_to_count(*buflen); + if (fiemap->fm_extent_count == 0) + count_local = 0; + + /* Check each stripe */ + for (cur_stripe = start_stripe; stripe_count > 0; + --stripe_count, + cur_stripe = (cur_stripe + 1) % lsm->lsm_stripe_count) { + loff_t req_fm_len; /* Stores length of required mapping */ + loff_t len_mapped_single_call; + loff_t lun_start; + loff_t lun_end; + loff_t obd_object_end; + unsigned int ext_count; + + cur_stripe_wrap = cur_stripe; + + /* Find out range of mapping on this stripe */ + if ((lov_stripe_intersects(lsm, cur_stripe, fm_start, fm_end, + &lun_start, &obd_object_end)) == 0) + continue; + + if (lov_oinfo_is_dummy(lsm->lsm_oinfo[cur_stripe])) + GOTO(out, rc = -EIO); + + /* If this is a continuation FIEMAP call and we are on + * starting stripe then lun_start needs to be set to + * fm_end_offset */ + if (fm_end_offset != 0 && cur_stripe == start_stripe) + lun_start = fm_end_offset; + + if (fm_length != ~0ULL) { + /* Handle fm_start + fm_length overflow */ + if (fm_start + fm_length < fm_start) + fm_length = ~0ULL - fm_start; + lun_end = lov_size_to_stripe(lsm, fm_start + fm_length, + cur_stripe); + } else { + lun_end = ~0ULL; + } + + if (lun_start == lun_end) + continue; + + req_fm_len = obd_object_end - lun_start; + fm_local->fm_length = 0; + len_mapped_single_call = 0; + + /* find lobsub object */ + subobj = lov_find_subobj(env, cl2lov(obj), lsm, + cur_stripe); + if (IS_ERR(subobj)) + GOTO(out, rc = PTR_ERR(subobj)); + /* If the output buffer is very large and the objects have many + * extents we may need to loop on a single OST repeatedly */ + ost_eof = false; + ost_done = false; + do { + if (fiemap->fm_extent_count > 0) { + /* Don't get too many extents. */ + if (current_extent + count_local > + fiemap->fm_extent_count) + count_local = fiemap->fm_extent_count - + current_extent; + } + + lun_start += len_mapped_single_call; + fm_local->fm_length = req_fm_len - + len_mapped_single_call; + req_fm_len = fm_local->fm_length; + fm_local->fm_extent_count = enough ? 1 : count_local; + fm_local->fm_mapped_extents = 0; + fm_local->fm_flags = fiemap->fm_flags; + + ost_index = lsm->lsm_oinfo[cur_stripe]->loi_ost_idx; + + if (ost_index < 0 || + ost_index >= lov->desc.ld_tgt_count) + GOTO(obj_put, rc = -EINVAL); + /* If OST is inactive, return extent with UNKNOWN + * flag. */ + if (!lov->lov_tgts[ost_index]->ltd_active) { + fm_local->fm_flags |= FIEMAP_EXTENT_LAST; + fm_local->fm_mapped_extents = 1; + + lcl_fm_ext[0].fe_logical = lun_start; + lcl_fm_ext[0].fe_length = obd_object_end - + lun_start; + lcl_fm_ext[0].fe_flags |= FIEMAP_EXTENT_UNKNOWN; + + goto inactive_tgt; + } + + fm_local->fm_start = lun_start; + fm_local->fm_flags &= ~FIEMAP_FLAG_DEVICE_ORDER; + memcpy(&fmkey->fiemap, fm_local, sizeof(*fm_local)); + *buflen = fiemap_count_to_size( + fm_local->fm_extent_count); + + rc = cl_object_fiemap(env, subobj, fmkey, fm_local, + buflen); + if (rc != 0) + GOTO(obj_put, rc); +inactive_tgt: + ext_count = fm_local->fm_mapped_extents; + if (ext_count == 0) { + ost_done = true; + /* If last stripe has hold at the end, + * we need to return */ + if (cur_stripe_wrap == last_stripe) { + fiemap->fm_mapped_extents = 0; + goto finish; + } + break; + } else if (enough) { + /* + * We've collected enough extents and there are + * more extents after it. + */ + goto finish; + } + + /* If we just need num of extents, got to next device */ + if (fiemap->fm_extent_count == 0) { + current_extent += ext_count; + break; + } + + /* prepare to copy retrived map extents */ + len_mapped_single_call = + lcl_fm_ext[ext_count - 1].fe_logical - + lun_start + lcl_fm_ext[ext_count - 1].fe_length; + + /* Have we finished mapping on this device? */ + if (req_fm_len <= len_mapped_single_call) + ost_done = true; + + /* Clear the EXTENT_LAST flag which can be present on + * the last extent */ + if (lcl_fm_ext[ext_count - 1].fe_flags & + FIEMAP_EXTENT_LAST) + lcl_fm_ext[ext_count - 1].fe_flags &= + ~FIEMAP_EXTENT_LAST; + if (lov_stripe_size(lsm, + lcl_fm_ext[ext_count - 1].fe_logical + + lcl_fm_ext[ext_count - 1].fe_length, + cur_stripe) >= fmkey->oa.o_size) + ost_eof = true; + + fiemap_prepare_and_copy_exts(fiemap, lcl_fm_ext, + ost_index, ext_count, + current_extent); + current_extent += ext_count; + + /* Ran out of available extents? */ + if (current_extent >= fiemap->fm_extent_count) + enough = true; + } while (!ost_done && !ost_eof); + + cl_object_put(env, subobj); + subobj = NULL; + + if (cur_stripe_wrap == last_stripe) + goto finish; + } /* for each stripe */ +finish: + /* Indicate that we are returning device offsets unless file just has + * single stripe */ + if (lsm->lsm_stripe_count > 1) + fiemap->fm_flags |= FIEMAP_FLAG_DEVICE_ORDER; + + if (fiemap->fm_extent_count == 0) + goto skip_last_device_calc; + + /* Check if we have reached the last stripe and whether mapping for that + * stripe is done. */ + if ((cur_stripe_wrap == last_stripe) && (ost_done || ost_eof)) + fiemap->fm_extents[current_extent - 1].fe_flags |= + FIEMAP_EXTENT_LAST; +skip_last_device_calc: + fiemap->fm_mapped_extents = current_extent; +obj_put: + if (subobj != NULL) + cl_object_put(env, subobj); +out: + if (fm_local != NULL) + OBD_FREE_LARGE(fm_local, buffer_size); + lov_lsm_put(obj, lsm); + RETURN(rc); +} + +static int lov_dispatch_obd_info_get(const struct lu_env *env, + struct cl_object *obj, + struct obd_info *oinfo, + struct ptlrpc_request_set *set) +{ + struct cl_object *subobj = NULL; + struct lov_obd *lov = lu2lov_dev(obj->co_lu.lo_dev)->ld_lov; + struct lov_request_set *lovset; + struct list_head *pos; + struct lov_request *req; + int rc; + int rc2; + ENTRY; + + rc = lov_prep_getattr_set(lov2obd(lov)->obd_self_export, oinfo, + &lovset); + if (rc != 0) + RETURN(rc); + + CDEBUG(D_INFO, "objid "DOSTID": %ux%u byte stripes.\n", + POSTID(&oinfo->oi_md->lsm_oi), + oinfo->oi_md->lsm_stripe_count, + oinfo->oi_md->lsm_stripe_size); + + list_for_each(pos, &lovset->set_list) { + req = list_entry(pos, struct lov_request, rq_link); + + CDEBUG(D_INFO, "objid "DOSTID"[%d] has subobj "DOSTID" at idx" + "%u\n", POSTID(&oinfo->oi_oa->o_oi), req->rq_stripe, + POSTID(&req->rq_oi.oi_oa->o_oi), req->rq_idx); + subobj = lov_find_subobj(env, cl2lov(obj), oinfo->oi_md, + req->rq_stripe); + if (IS_ERR(subobj)) + GOTO(errout, rc = PTR_ERR(subobj)); + + rc = cl_object_obd_info_get(env, subobj, &req->rq_oi, set); + cl_object_put(env, subobj); + if (rc != 0) { + CERROR("%s: getattr objid "DOSTID" subobj" + DOSTID" on OST idx %d: rc = %d\n", + lov2obd(lov)->obd_name, + POSTID(&oinfo->oi_oa->o_oi), + POSTID(&req->rq_oi.oi_oa->o_oi), + req->rq_idx, rc); + GOTO(errout, rc); + } + } + + if (!list_empty(&set->set_requests)) { + LASSERT(rc == 0); + LASSERT(set->set_interpret == NULL); + set->set_interpret = lov_getattr_interpret; + set->set_arg = lovset; + GOTO(out, rc); + } +errout: + if (rc) + atomic_set(&lovset->set_completes, 0); + rc2 = lov_fini_getattr_set(lovset); + rc = rc != 0 ? rc : rc2; +out: + RETURN(rc); +} + +static int lov_object_data_version(const struct lu_env *env, + struct cl_object *obj, __u64 *data_version, + int flags) +{ + struct ptlrpc_request_set *set; + struct obd_info oinfo = { { { 0 } } }; + struct obdo *obdo = NULL; + struct lov_stripe_md *lsm; + int rc; + ENTRY; + + lsm = lov_lsm_addref(cl2lov(obj)); + if (!lsm_has_objects(lsm)) { + /* If no stripe, we consider version is 0. */ + *data_version = 0; + GOTO(out, rc = 0); + } + + OBD_ALLOC_PTR(obdo); + if (obdo == NULL) + GOTO(out, rc = -ENOMEM); + + oinfo.oi_md = lsm; + oinfo.oi_oa = obdo; + obdo->o_oi = lsm->lsm_oi; + obdo->o_mode = S_IFREG; + obdo->o_valid = OBD_MD_FLID | OBD_MD_FLGROUP | OBD_MD_FLTYPE; + if (flags & (LL_DV_RD_FLUSH | LL_DV_WR_FLUSH)) { + obdo->o_valid |= OBD_MD_FLFLAGS; + obdo->o_flags |= OBD_FL_SRVLOCK; + if (flags & LL_DV_WR_FLUSH) + obdo->o_flags |= OBD_FL_FLUSH; + } + + set = ptlrpc_prep_set(); + if (set == NULL) + GOTO(out_obdo, rc = -ENOMEM); + + rc = lov_dispatch_obd_info_get(env, obj, &oinfo, set); + if (rc == 0) + rc = ptlrpc_set_wait(set); + ptlrpc_set_destroy(set); + if (rc == 0) { + oinfo.oi_oa->o_valid &= OBD_MD_FLDATAVERSION | OBD_MD_FLFLAGS; + if (flags & LL_DV_WR_FLUSH && + !(oinfo.oi_oa->o_valid & OBD_MD_FLFLAGS && + oinfo.oi_oa->o_flags & OBD_FL_FLUSH)) + rc = -EOPNOTSUPP; + else if (!(obdo->o_valid & OBD_MD_FLDATAVERSION)) + rc = -EOPNOTSUPP; + else + *data_version = obdo->o_data_version; + } +out_obdo: + OBD_FREE_PTR(obdo); +out: + lov_lsm_put(obj, lsm); + RETURN(rc); +} + +static int lov_object_getstripe(const struct lu_env *env, struct cl_object *obj, + struct lov_user_md __user *lum) +{ + struct lov_object *lov = cl2lov(obj); + struct lov_stripe_md *lsm; + int rc = 0; + ENTRY; + + lsm = lov_lsm_addref(lov); + if (lsm == NULL) + RETURN(-ENODATA); + + rc = lov_getstripe(cl2lov(obj), lsm, lum); + lov_lsm_put(obj, lsm); + RETURN(rc); +} + +static int lov_object_find_cbdata(const struct lu_env *env, + struct cl_object *obj, ldlm_iterator_t iter, + void *data) +{ + int rc; + ENTRY; + + /* call cl_object_find_cbdata for sub obj */ + rc = LOV_2DISPATCH_NOLOCK(cl2lov(obj), llo_find_cbdata, + env, obj, iter, data); + RETURN(rc); +} + static const struct cl_object_operations lov_ops = { - .coo_page_init = lov_page_init, - .coo_lock_init = lov_lock_init, - .coo_io_init = lov_io_init, - .coo_attr_get = lov_attr_get, - .coo_attr_set = lov_attr_set, - .coo_conf_set = lov_conf_set + .coo_page_init = lov_page_init, + .coo_lock_init = lov_lock_init, + .coo_io_init = lov_io_init, + .coo_attr_get = lov_attr_get, + .coo_attr_update = lov_attr_update, + .coo_conf_set = lov_conf_set, + .coo_getstripe = lov_object_getstripe, + .coo_find_cbdata = lov_object_find_cbdata, + .coo_fiemap = lov_object_fiemap, + .coo_data_version = lov_object_data_version, }; static const struct lu_object_operations lov_lu_obj_ops = { @@ -826,28 +1622,28 @@ static const struct lu_object_operations lov_lu_obj_ops = { }; struct lu_object *lov_object_alloc(const struct lu_env *env, - const struct lu_object_header *unused, - struct lu_device *dev) + const struct lu_object_header *unused, + struct lu_device *dev) { - struct lov_object *lov; - struct lu_object *obj; + struct lov_object *lov; + struct lu_object *obj; - ENTRY; - OBD_SLAB_ALLOC_PTR_GFP(lov, lov_object_kmem, CFS_ALLOC_IO); - if (lov != NULL) { - obj = lov2lu(lov); - lu_object_init(obj, NULL, dev); - lov->lo_cl.co_ops = &lov_ops; - lov->lo_type = -1; /* invalid, to catch uninitialized type */ - /* - * object io operation vector (cl_object::co_iop) is installed - * later in lov_object_init(), as different vectors are used - * for object with different layouts. - */ - obj->lo_ops = &lov_lu_obj_ops; - } else - obj = NULL; - RETURN(obj); + ENTRY; + OBD_SLAB_ALLOC_PTR_GFP(lov, lov_object_kmem, GFP_NOFS); + if (lov != NULL) { + obj = lov2lu(lov); + lu_object_init(obj, NULL, dev); + lov->lo_cl.co_ops = &lov_ops; + lov->lo_type = -1; /* invalid, to catch uninitialized type */ + /* + * object io operation vector (cl_object::co_iop) is installed + * later in lov_object_init(), as different vectors are used + * for object with different layouts. + */ + obj->lo_ops = &lov_lu_obj_ops; + } else + obj = NULL; + RETURN(obj); } struct lov_stripe_md *lov_lsm_addref(struct lov_object *lov) @@ -858,47 +1654,13 @@ struct lov_stripe_md *lov_lsm_addref(struct lov_object *lov) if (lov->lo_lsm != NULL) { lsm = lsm_addref(lov->lo_lsm); CDEBUG(D_INODE, "lsm %p addref %d/%d by %p.\n", - lsm, cfs_atomic_read(&lsm->lsm_refc), - lov->lo_layout_invalid, cfs_current()); + lsm, atomic_read(&lsm->lsm_refc), + lov->lo_layout_invalid, current); } lov_conf_thaw(lov); return lsm; } -void lov_lsm_decref(struct lov_object *lov, struct lov_stripe_md *lsm) -{ - if (lsm == NULL) - return; - - CDEBUG(D_INODE, "lsm %p decref %d by %p.\n", - lsm, cfs_atomic_read(&lsm->lsm_refc), cfs_current()); - - lov_free_memmd(&lsm); -} - -struct lov_stripe_md *lov_lsm_get(struct cl_object *clobj) -{ - struct lu_object *luobj; - struct lov_stripe_md *lsm = NULL; - - if (clobj == NULL) - return NULL; - - luobj = lu_object_locate(&cl_object_header(clobj)->coh_lu, - &lov_device_type); - if (luobj != NULL) - lsm = lov_lsm_addref(lu2lov(luobj)); - return lsm; -} -EXPORT_SYMBOL(lov_lsm_get); - -void lov_lsm_put(struct cl_object *unused, struct lov_stripe_md *lsm) -{ - if (lsm != NULL) - lov_free_memmd(&lsm); -} -EXPORT_SYMBOL(lov_lsm_put); - int lov_read_and_clear_async_rc(struct cl_object *clob) { struct lu_object *luobj; @@ -920,11 +1682,16 @@ int lov_read_and_clear_async_rc(struct cl_object *clob) LASSERT(lsm != NULL); for (i = 0; i < lsm->lsm_stripe_count; i++) { struct lov_oinfo *loi = lsm->lsm_oinfo[i]; + + if (lov_oinfo_is_dummy(loi)) + continue; + if (loi->loi_ar.ar_rc && !rc) rc = loi->loi_ar.ar_rc; loi->loi_ar.ar_rc = 0; } } + case LLT_RELEASED: case LLT_EMPTY: break; default: