X-Git-Url: https://git.whamcloud.com/?p=fs%2Flustre-release.git;a=blobdiff_plain;f=lustre%2Flov%2Flov_cl_internal.h;h=c5b3c3ee26318825193f5c39024e81c824e810cf;hp=e16a6aab05719b4606a0a602f71504a37fb3aecf;hb=68fb53ad4bb2dbc2104d19613546ecf2534d97d5;hpb=545f464b68ee21ccc425113db9b57c63ac2840b1 diff --git a/lustre/lov/lov_cl_internal.h b/lustre/lov/lov_cl_internal.h index e16a6aa..c5b3c3e 100644 --- a/lustre/lov/lov_cl_internal.h +++ b/lustre/lov/lov_cl_internal.h @@ -23,7 +23,7 @@ * Copyright (c) 2008, 2010, Oracle and/or its affiliates. All rights reserved. * Use is subject to license terms. * - * Copyright (c) 2012, 2016, Intel Corporation. + * Copyright (c) 2012, 2017, Intel Corporation. */ /* * This file is part of Lustre, http://www.lustre.org/ @@ -81,16 +81,21 @@ struct lovsub_device; struct lovsub_object; -struct lovsub_lock; enum lov_device_flags { - LOV_DEV_INITIALIZED = 1 << 0 + LOV_DEV_INITIALIZED = BIT(0), }; /* * Upper half. */ +/* Data-on-MDT array item in lov_device::ld_md_tgts[] */ +struct lovdom_device { + struct cl_device *ldm_mdc; + int ldm_idx; +}; + struct lov_device { /* * XXX Locking of lov-private data is missing. @@ -101,6 +106,13 @@ struct lov_device { __u32 ld_target_nr; struct lovsub_device **ld_target; __u32 ld_flags; + + /* Data-on-MDT devices */ + __u32 ld_md_tgts_nr; + struct lovdom_device *ld_md_tgts; + struct obd_device *ld_lmv; + /* LU site for subdevices */ + struct lu_site ld_site; }; /** @@ -110,6 +122,7 @@ enum lov_layout_type { LLT_EMPTY, /** empty file without body (mknod + truncate) */ LLT_RELEASED, /** file with no objects (data in HSM) */ LLT_COMP, /** support composite layout */ + LLT_FOREIGN, /** foreign layout */ LLT_NR }; @@ -122,6 +135,8 @@ static inline char *llt2str(enum lov_layout_type llt) return "RELEASED"; case LLT_COMP: return "COMPOSITE"; + case LLT_FOREIGN: + return "FOREIGN"; case LLT_NR: LBUG(); } @@ -129,15 +144,50 @@ static inline char *llt2str(enum lov_layout_type llt) return ""; } +/** + * Return lov_layout_entry_type associated with a given composite layout + * entry. + */ +static inline __u32 lov_entry_type(struct lov_stripe_md_entry *lsme) +{ + if ((lov_pattern(lsme->lsme_pattern) & LOV_PATTERN_RAID0) || + (lov_pattern(lsme->lsme_pattern) == LOV_PATTERN_MDT) || + (lov_pattern(lsme->lsme_pattern) == LOV_PATTERN_FOREIGN)) + return lov_pattern(lsme->lsme_pattern & + ~LOV_PATTERN_OVERSTRIPING); + return 0; +} + +struct lov_layout_entry; +struct lov_object; +struct lov_lock_sub; + +struct lov_comp_layout_entry_ops { + int (*lco_init)(const struct lu_env *env, struct lov_device *dev, + struct lov_object *lov, unsigned int index, + const struct cl_object_conf *conf, + struct lov_layout_entry *lle); + void (*lco_fini)(const struct lu_env *env, + struct lov_layout_entry *lle); + int (*lco_getattr)(const struct lu_env *env, struct lov_object *obj, + unsigned int index, struct lov_layout_entry *lle, + struct cl_attr **attr); +}; + struct lov_layout_raid0 { unsigned lo_nr; /** + * record the stripe no before the truncate size, used for setting OST + * object size for truncate. LU-14128. + */ + int lo_trunc_stripeno; + /** * When this is true, lov_object::lo_attr contains * valid up to date attributes for a top-level * object. This field is reset to 0 when attributes of * any sub-object change. */ - int lo_attr_valid; + bool lo_attr_valid; /** * Array of sub-objects. Allocated when top-object is * created (lov_init_raid0()). @@ -165,6 +215,45 @@ struct lov_layout_raid0 { struct cl_attr lo_attr; }; +struct lov_layout_dom { + /* keep this always at first place so DOM layout entry + * can be addressed also as RAID0 after initialization. + */ + struct lov_layout_raid0 lo_dom_r0; + struct lovsub_object *lo_dom; + struct lov_oinfo *lo_loi; +}; + +struct lov_layout_entry { + __u32 lle_type; + unsigned int lle_valid:1; + struct lu_extent *lle_extent; + struct lov_stripe_md_entry *lle_lsme; + struct lov_comp_layout_entry_ops *lle_comp_ops; + union { + struct lov_layout_raid0 lle_raid0; + struct lov_layout_dom lle_dom; + }; +}; + +struct lov_mirror_entry { + unsigned short lre_mirror_id; + unsigned short lre_preferred:1, + lre_stale:1, /* set if any components is stale */ + lre_valid:1, /* set if at least one of components + * in this mirror is valid */ + lre_foreign:1; /* set if it is a foreign component */ + + unsigned short lre_start; /* index to lo_entries, start index of + * this mirror */ + unsigned short lre_end; /* end index of this mirror */ +}; + +enum lov_object_flags { + /* Layout is invalid, set when layout lock is lost */ + LO_LAYOUT_INVALID = 0x1, +}; + /** * lov-specific file state. * @@ -180,7 +269,7 @@ struct lov_layout_raid0 { * function corresponding to the current layout type. */ struct lov_object { - struct cl_object lo_cl; + struct cl_object lo_cl; /** * Serializes object operations with transitions between layout types. * @@ -195,10 +284,9 @@ struct lov_object { */ enum lov_layout_type lo_type; /** - * True if layout is invalid. This bit is cleared when layout lock - * is lost. + * Object flags. */ - bool lo_layout_invalid; + unsigned long lo_obj_flags; /** * How many IOs are on going on this object. Layout can be changed * only if there is no active IO. @@ -220,13 +308,37 @@ struct lov_object { } released; struct lov_layout_composite { /** - * Current valid entry count of lo_entries. + * flags of lov_comp_md_v1::lcm_flags. Mainly used + * by FLR. + */ + uint32_t lo_flags; + /** + * For FLR: index of preferred mirror to read. + * Preferred mirror is initialized by the preferred + * bit of lsme. It can be changed when the preferred + * is inaccessible. + * In order to make lov_lsm_entry() return the same + * mirror in the same IO context, it's only possible + * to change the preferred mirror when the + * lo_active_ios reaches zero. + */ + int lo_preferred_mirror; + /** + * For FLR: the lock to protect access to + * lo_preferred_mirror. + */ + spinlock_t lo_write_lock; + /** + * For FLR: Number of (valid) mirrors. + */ + unsigned lo_mirror_count; + struct lov_mirror_entry *lo_mirrors; + /** + * Current entry count of lo_entries, include + * invalid entries. */ - unsigned int lo_entry_count; - struct lov_layout_entry { - struct lu_extent lle_extent; - struct lov_layout_raid0 lle_raid0; - } *lo_entries; + unsigned int lo_entry_count; + struct lov_layout_entry *lo_entries; } composite; } u; /** @@ -236,11 +348,82 @@ struct lov_object { struct task_struct *lo_owner; }; -#define lov_foreach_layout_entry(lov, entry) \ - for (entry = &lov->u.composite.lo_entries[0]; \ - entry < &lov->u.composite.lo_entries \ - [lov->u.composite.lo_entry_count]; \ - entry++) +static inline struct lov_layout_raid0 *lov_r0(struct lov_object *lov, int i) +{ + LASSERT(lov->lo_type == LLT_COMP); + LASSERTF(i < lov->u.composite.lo_entry_count, + "entry %d entry_count %d\n", i, + lov->u.composite.lo_entry_count); + + return &lov->u.composite.lo_entries[i].lle_raid0; +} + +static inline struct lov_stripe_md_entry *lov_lse(struct lov_object *lov, int i) +{ + LASSERT(lov->lo_lsm != NULL); + LASSERT(i < lov->lo_lsm->lsm_entry_count); + + return lov->lo_lsm->lsm_entries[i]; +} + +static inline unsigned lov_flr_state(const struct lov_object *lov) +{ + if (lov->lo_type != LLT_COMP) + return LCM_FL_NONE; + + return lov->u.composite.lo_flags & LCM_FL_FLR_MASK; +} + +static inline bool lov_is_flr(const struct lov_object *lov) +{ + return lov_flr_state(lov) != LCM_FL_NONE; +} + +static inline struct lov_layout_entry *lov_entry(struct lov_object *lov, int i) +{ + LASSERT(lov->lo_type == LLT_COMP); + LASSERTF(i < lov->u.composite.lo_entry_count, + "entry %d entry_count %d\n", i, + lov->u.composite.lo_entry_count); + + return &lov->u.composite.lo_entries[i]; +} + +#define lov_for_layout_entry(lov, entry, start, end) \ + for (entry = lov_entry(lov, start); \ + entry <= lov_entry(lov, end); entry++) + +#define lov_foreach_layout_entry(lov, entry) \ + lov_for_layout_entry(lov, entry, 0, \ + (lov)->u.composite.lo_entry_count - 1) + +#define lov_foreach_mirror_layout_entry(lov, entry, lre) \ + lov_for_layout_entry(lov, entry, (lre)->lre_start, (lre)->lre_end) + +static inline struct lov_mirror_entry * +lov_mirror_entry(struct lov_object *lov, int i) +{ + LASSERT(i < lov->u.composite.lo_mirror_count); + return &lov->u.composite.lo_mirrors[i]; +} + +#define lov_foreach_mirror_entry(lov, lre) \ + for (lre = lov_mirror_entry(lov, 0); \ + lre <= lov_mirror_entry(lov, \ + lov->u.composite.lo_mirror_count - 1); \ + lre++) + +static inline unsigned +lov_layout_entry_index(struct lov_object *lov, struct lov_layout_entry *entry) +{ + struct lov_layout_entry *first = &lov->u.composite.lo_entries[0]; + unsigned index = (unsigned)(entry - first); + + LASSERT(entry >= first); + LASSERT(index < lov->u.composite.lo_entry_count); + + return index; +} /** * State lov_lock keeps for each sub-lock. @@ -268,8 +451,8 @@ struct lov_lock { struct lov_page { struct cl_page_slice lps_cl; - /** layout_entry + stripe index, composed using lov_comp_index() */ - unsigned int lps_index; + /* the layout gen when this page was created */ + __u32 lps_layout_gen; }; /* @@ -289,13 +472,6 @@ struct lovsub_object { }; /** - * Lock state at lovsub layer. - */ -struct lovsub_lock { - struct cl_lock_slice lss_cl; -}; - -/** * Describe the environment settings for sublocks. */ struct lov_sublock_env { @@ -303,18 +479,12 @@ struct lov_sublock_env { struct cl_io *lse_io; }; -struct lovsub_page { - struct cl_page_slice lsb_cl; -}; - - struct lov_thread_info { struct cl_object_conf lti_stripe_conf; struct lu_fid lti_fid; struct ost_lvb lti_lvb; struct cl_2queue lti_cl2q; struct cl_page_list lti_plist; - wait_queue_t lti_waiter; }; /** @@ -347,7 +517,6 @@ struct lov_io_sub { * \see cl_env_get() */ __u16 sub_refcheck; - __u16 sub_reenter; }; /** @@ -356,6 +525,26 @@ struct lov_io_sub { struct lov_io { /** super-class */ struct cl_io_slice lis_cl; + + /** + * FLR: index to lo_mirrors. Valid only if lov_is_flr() returns true. + * + * The mirror index of this io. Preserved over cl_io_init() + * if io->ci_ndelay_tried is greater than zero. + */ + int lis_mirror_index; + /** + * FLR: the layout gen when lis_mirror_index was cached. The + * mirror index makes sense only when the layout gen doesn't + * change. + */ + int lis_mirror_layout_gen; + + /** + * fields below this will be initialized in lov_io_init(). + */ + unsigned lis_preserved; + /** * Pointer to the object slice. This is a duplicate of * lov_io::lis_cl::cis_object. @@ -398,6 +587,7 @@ struct lov_io { * All sub-io's created in this lov_io. */ struct list_head lis_subios; + }; struct lov_session { @@ -416,20 +606,8 @@ extern struct kmem_cache *lov_object_kmem; extern struct kmem_cache *lov_thread_kmem; extern struct kmem_cache *lov_session_kmem; -extern struct kmem_cache *lovsub_lock_kmem; extern struct kmem_cache *lovsub_object_kmem; -int lov_object_init (const struct lu_env *env, struct lu_object *obj, - const struct lu_object_conf *conf); -int lovsub_object_init (const struct lu_env *env, struct lu_object *obj, - const struct lu_object_conf *conf); -int lov_lock_init (const struct lu_env *env, struct cl_object *obj, - struct cl_lock *lock, const struct cl_io *io); -int lov_io_init (const struct lu_env *env, struct cl_object *obj, - struct cl_io *io); -int lovsub_lock_init (const struct lu_env *env, struct cl_object *obj, - struct cl_lock *lock, const struct cl_io *io); - int lov_lock_init_composite(const struct lu_env *env, struct cl_object *obj, struct cl_lock *lock, const struct cl_io *io); int lov_lock_init_empty (const struct lu_env *env, struct cl_object *obj, @@ -444,28 +622,42 @@ int lov_io_init_released(const struct lu_env *env, struct cl_object *obj, struct lov_io_sub *lov_sub_get(const struct lu_env *env, struct lov_io *lio, int stripe); -int lov_page_init (const struct lu_env *env, struct cl_object *ob, - struct cl_page *page, pgoff_t index); -int lovsub_page_init (const struct lu_env *env, struct cl_object *ob, - struct cl_page *page, pgoff_t index); int lov_page_init_empty (const struct lu_env *env, struct cl_object *obj, struct cl_page *page, pgoff_t index); int lov_page_init_composite(const struct lu_env *env, struct cl_object *obj, struct cl_page *page, pgoff_t index); +int lov_page_init_foreign(const struct lu_env *env, struct cl_object *obj, + struct cl_page *page, pgoff_t index); struct lu_object *lov_object_alloc (const struct lu_env *env, const struct lu_object_header *hdr, struct lu_device *dev); + struct lu_object *lovsub_object_alloc(const struct lu_env *env, const struct lu_object_header *hdr, struct lu_device *dev); -struct lov_stripe_md *lov_lsm_addref(struct lov_object *lov); int lov_page_stripe(const struct cl_page *page); +bool lov_page_is_empty(const struct cl_page *page); int lov_lsm_entry(const struct lov_stripe_md *lsm, __u64 offset); +int lov_io_layout_at(struct lov_io *lio, __u64 offset); #define lov_foreach_target(lov, var) \ for (var = 0; var < lov_targets_nr(lov); ++var) +static inline struct lu_extent *lov_io_extent(struct lov_io *io, int i) +{ + return &lov_lse(io->lis_object, i)->lsme_extent; +} + +/** + * For layout entries within @ext. + */ +#define lov_foreach_io_layout(ind, lio, ext) \ + for (ind = lov_io_layout_at(lio, (ext)->e_start); \ + ind >= 0 && \ + lu_extent_is_overlapped(lov_io_extent(lio, ind), ext); \ + ind = lov_io_layout_at(lio, lov_io_extent(lio, ind)->e_end)) + /***************************************************************************** * * Type conversions. @@ -505,8 +697,8 @@ static inline struct lu_device *lov2lu_dev(struct lov_device *lov) static inline struct lov_device *lu2lov_dev(const struct lu_device *d) { - LINVRNT(d->ld_type == &lov_device_type); - return container_of0(d, struct lov_device, ld_cl.cd_lu_dev); + LINVRNT(d->ld_type == &lov_device_type); + return container_of(d, struct lov_device, ld_cl.cd_lu_dev); } static inline struct cl_device *lovsub2cl_dev(struct lovsub_device *lovsub) @@ -521,14 +713,14 @@ static inline struct lu_device *lovsub2lu_dev(struct lovsub_device *lovsub) static inline struct lovsub_device *lu2lovsub_dev(const struct lu_device *d) { - LINVRNT(d->ld_type == &lovsub_device_type); - return container_of0(d, struct lovsub_device, acid_cl.cd_lu_dev); + LINVRNT(d->ld_type == &lovsub_device_type); + return container_of(d, struct lovsub_device, acid_cl.cd_lu_dev); } static inline struct lovsub_device *cl2lovsub_dev(const struct cl_device *d) { - LINVRNT(d->cd_lu_dev.ld_type == &lovsub_device_type); - return container_of0(d, struct lovsub_device, acid_cl); + LINVRNT(d->cd_lu_dev.ld_type == &lovsub_device_type); + return container_of(d, struct lovsub_device, acid_cl); } static inline struct lu_object *lov2lu(struct lov_object *lov) @@ -543,71 +735,48 @@ static inline struct cl_object *lov2cl(struct lov_object *lov) static inline struct lov_object *lu2lov(const struct lu_object *obj) { - LINVRNT(lov_is_object(obj)); - return container_of0(obj, struct lov_object, lo_cl.co_lu); + LINVRNT(lov_is_object(obj)); + return container_of(obj, struct lov_object, lo_cl.co_lu); } static inline struct lov_object *cl2lov(const struct cl_object *obj) { - LINVRNT(lov_is_object(&obj->co_lu)); - return container_of0(obj, struct lov_object, lo_cl); + LINVRNT(lov_is_object(&obj->co_lu)); + return container_of(obj, struct lov_object, lo_cl); } static inline struct lu_object *lovsub2lu(struct lovsub_object *los) { - return &los->lso_cl.co_lu; + return &los->lso_cl.co_lu; } static inline struct cl_object *lovsub2cl(struct lovsub_object *los) { - return &los->lso_cl; + return &los->lso_cl; } static inline struct lovsub_object *cl2lovsub(const struct cl_object *obj) { - LINVRNT(lovsub_is_object(&obj->co_lu)); - return container_of0(obj, struct lovsub_object, lso_cl); + LINVRNT(lovsub_is_object(&obj->co_lu)); + return container_of(obj, struct lovsub_object, lso_cl); } static inline struct lovsub_object *lu2lovsub(const struct lu_object *obj) { - LINVRNT(lovsub_is_object(obj)); - return container_of0(obj, struct lovsub_object, lso_cl.co_lu); -} - -static inline struct lovsub_lock * -cl2lovsub_lock(const struct cl_lock_slice *slice) -{ - LINVRNT(lovsub_is_object(&slice->cls_obj->co_lu)); - return container_of(slice, struct lovsub_lock, lss_cl); -} - -static inline struct lovsub_lock *cl2sub_lock(const struct cl_lock *lock) -{ - const struct cl_lock_slice *slice; - - slice = cl_lock_at(lock, &lovsub_device_type); - LASSERT(slice != NULL); - return cl2lovsub_lock(slice); + LINVRNT(lovsub_is_object(obj)); + return container_of(obj, struct lovsub_object, lso_cl.co_lu); } static inline struct lov_lock *cl2lov_lock(const struct cl_lock_slice *slice) { - LINVRNT(lov_is_object(&slice->cls_obj->co_lu)); - return container_of(slice, struct lov_lock, lls_cl); + LINVRNT(lov_is_object(&slice->cls_obj->co_lu)); + return container_of(slice, struct lov_lock, lls_cl); } static inline struct lov_page *cl2lov_page(const struct cl_page_slice *slice) { - LINVRNT(lov_is_object(&slice->cpl_obj->co_lu)); - return container_of0(slice, struct lov_page, lps_cl); -} - -static inline struct lovsub_page * -cl2lovsub_page(const struct cl_page_slice *slice) -{ - LINVRNT(lovsub_is_object(&slice->cpl_obj->co_lu)); - return container_of0(slice, struct lovsub_page, lsb_cl); + LINVRNT(lov_is_object(&slice->cpl_obj->co_lu)); + return container_of(slice, struct lov_page, lps_cl); } static inline struct lov_io *cl2lov_io(const struct lu_env *env, @@ -634,26 +803,10 @@ static inline struct lov_thread_info *lov_env_info(const struct lu_env *env) return info; } -static inline struct lov_layout_raid0 *lov_r0(struct lov_object *lov, int i) -{ - LASSERT(lov->lo_type == LLT_COMP); - LASSERTF(i < lov->u.composite.lo_entry_count, - "entry %d entry_count %d", i, lov->u.composite.lo_entry_count); - - return &lov->u.composite.lo_entries[i].lle_raid0; -} - -static inline struct lov_stripe_md_entry *lov_lse(struct lov_object *lov, int i) -{ - LASSERT(lov->lo_lsm != NULL); - LASSERT(i < lov->lo_lsm->lsm_entry_count); - - return lov->lo_lsm->lsm_entries[i]; -} - /* lov_pack.c */ int lov_getstripe(const struct lu_env *env, struct lov_object *obj, - struct lov_stripe_md *lsm, struct lov_user_md __user *lump); + struct lov_stripe_md *lsm, struct lov_user_md __user *lump, + size_t size); /** @} lov */