diff options
Diffstat (limited to 'fs/erofs/zdata.c')
| -rw-r--r-- | fs/erofs/zdata.c | 2483 |
1 files changed, 1422 insertions, 1061 deletions
diff --git a/fs/erofs/zdata.c b/fs/erofs/zdata.c index 498b7666efe8..65da21504632 100644 --- a/fs/erofs/zdata.c +++ b/fs/erofs/zdata.c @@ -2,13 +2,118 @@ /* * Copyright (C) 2018 HUAWEI, Inc. * https://www.huawei.com/ + * Copyright (C) 2022 Alibaba Cloud */ -#include "zdata.h" #include "compress.h" -#include <linux/prefetch.h> - +#include <linux/psi.h> +#include <linux/cpuhotplug.h> #include <trace/events/erofs.h> +#define Z_EROFS_PCLUSTER_MAX_PAGES (Z_EROFS_PCLUSTER_MAX_SIZE / PAGE_SIZE) +#define Z_EROFS_INLINE_BVECS 2 + +struct z_erofs_bvec { + struct page *page; + int offset; + unsigned int end; +}; + +#define __Z_EROFS_BVSET(name, total) \ +struct name { \ + /* point to the next page which contains the following bvecs */ \ + struct page *nextpage; \ + struct z_erofs_bvec bvec[total]; \ +} +__Z_EROFS_BVSET(z_erofs_bvset,); +__Z_EROFS_BVSET(z_erofs_bvset_inline, Z_EROFS_INLINE_BVECS); + +/* + * Structure fields follow one of the following exclusion rules. + * + * I: Modifiable by initialization/destruction paths and read-only + * for everyone else; + * + * L: Field should be protected by the pcluster lock; + * + * A: Field should be accessed / updated in atomic for parallelized code. + */ +struct z_erofs_pcluster { + struct mutex lock; + struct lockref lockref; + + /* A: point to next chained pcluster or TAILs */ + struct z_erofs_pcluster *next; + + /* I: start physical position of this pcluster */ + erofs_off_t pos; + + /* L: the maximum decompression size of this round */ + unsigned int length; + + /* L: total number of bvecs */ + unsigned int vcnt; + + /* I: pcluster size (compressed size) in bytes */ + unsigned int pclustersize; + + /* I: page offset of start position of decompression */ + unsigned short pageofs_out; + + /* I: page offset of inline compressed data */ + unsigned short pageofs_in; + + union { + /* L: inline a certain number of bvec for bootstrap */ + struct z_erofs_bvset_inline bvset; + + /* I: can be used to free the pcluster by RCU. */ + struct rcu_head rcu; + }; + + /* I: compression algorithm format */ + unsigned char algorithmformat; + + /* I: whether compressed data is in-lined or not */ + bool from_meta; + + /* L: whether partial decompression or not */ + bool partial; + + /* L: whether extra buffer allocations are best-effort */ + bool besteffort; + + /* A: compressed bvecs (can be cached or inplaced pages) */ + struct z_erofs_bvec compressed_bvecs[]; +}; + +/* the end of a chain of pclusters */ +#define Z_EROFS_PCLUSTER_TAIL ((void *) 0x700 + POISON_POINTER_DELTA) + +struct z_erofs_decompressqueue { + struct super_block *sb; + struct z_erofs_pcluster *head; + atomic_t pending_bios; + + union { + struct completion done; + struct work_struct work; + struct kthread_work kthread_work; + } u; + bool eio, sync; +}; + +static inline unsigned int z_erofs_pclusterpages(struct z_erofs_pcluster *pcl) +{ + return PAGE_ALIGN(pcl->pageofs_in + pcl->pclustersize) >> PAGE_SHIFT; +} + +static bool erofs_folio_is_managed(struct erofs_sb_info *sbi, struct folio *fo) +{ + return fo->mapping == MNGD_MAPPING(sbi); +} + +#define Z_EROFS_ONSTACK_PAGES 32 + /* * since pclustersize is variable for big pcluster feature, introduce slab * pools implementation for different pcluster sizes. @@ -23,9 +128,92 @@ struct z_erofs_pcluster_slab { static struct z_erofs_pcluster_slab pcluster_pool[] __read_mostly = { _PCLP(1), _PCLP(4), _PCLP(16), _PCLP(64), _PCLP(128), - _PCLP(Z_EROFS_PCLUSTER_MAX_PAGES) + _PCLP(Z_EROFS_PCLUSTER_MAX_PAGES + 1) }; +struct z_erofs_bvec_iter { + struct page *bvpage; + struct z_erofs_bvset *bvset; + unsigned int nr, cur; +}; + +static struct page *z_erofs_bvec_iter_end(struct z_erofs_bvec_iter *iter) +{ + if (iter->bvpage) + kunmap_local(iter->bvset); + return iter->bvpage; +} + +static struct page *z_erofs_bvset_flip(struct z_erofs_bvec_iter *iter) +{ + unsigned long base = (unsigned long)((struct z_erofs_bvset *)0)->bvec; + /* have to access nextpage in advance, otherwise it will be unmapped */ + struct page *nextpage = iter->bvset->nextpage; + struct page *oldpage; + + DBG_BUGON(!nextpage); + oldpage = z_erofs_bvec_iter_end(iter); + iter->bvpage = nextpage; + iter->bvset = kmap_local_page(nextpage); + iter->nr = (PAGE_SIZE - base) / sizeof(struct z_erofs_bvec); + iter->cur = 0; + return oldpage; +} + +static void z_erofs_bvec_iter_begin(struct z_erofs_bvec_iter *iter, + struct z_erofs_bvset_inline *bvset, + unsigned int bootstrap_nr, + unsigned int cur) +{ + *iter = (struct z_erofs_bvec_iter) { + .nr = bootstrap_nr, + .bvset = (struct z_erofs_bvset *)bvset, + }; + + while (cur > iter->nr) { + cur -= iter->nr; + z_erofs_bvset_flip(iter); + } + iter->cur = cur; +} + +static int z_erofs_bvec_enqueue(struct z_erofs_bvec_iter *iter, + struct z_erofs_bvec *bvec, + struct page **candidate_bvpage, + struct page **pagepool) +{ + if (iter->cur >= iter->nr) { + struct page *nextpage = *candidate_bvpage; + + if (!nextpage) { + nextpage = __erofs_allocpage(pagepool, GFP_KERNEL, + true); + if (!nextpage) + return -ENOMEM; + set_page_private(nextpage, Z_EROFS_SHORTLIVED_PAGE); + } + DBG_BUGON(iter->bvset->nextpage); + iter->bvset->nextpage = nextpage; + z_erofs_bvset_flip(iter); + + iter->bvset->nextpage = NULL; + *candidate_bvpage = NULL; + } + iter->bvset->bvec[iter->cur++] = *bvec; + return 0; +} + +static void z_erofs_bvec_dequeue(struct z_erofs_bvec_iter *iter, + struct z_erofs_bvec *bvec, + struct page **old_bvpage) +{ + if (iter->cur == iter->nr) + *old_bvpage = z_erofs_bvset_flip(iter); + else + *old_bvpage = NULL; + *bvec = iter->bvset->bvec[iter->cur++]; +} + static void z_erofs_destroy_pcluster_pool(void) { int i; @@ -46,7 +234,7 @@ static int z_erofs_create_pcluster_pool(void) for (pcs = pcluster_pool; pcs < pcluster_pool + ARRAY_SIZE(pcluster_pool); ++pcs) { - size = struct_size(a, compressed_pages, pcs->maxpages); + size = struct_size(a, compressed_bvecs, pcs->maxpages); sprintf(pcs->name, "erofs_pcluster-%u", pcs->maxpages); pcs->slab = kmem_cache_create(pcs->name, size, 0, @@ -60,21 +248,20 @@ static int z_erofs_create_pcluster_pool(void) return 0; } -static struct z_erofs_pcluster *z_erofs_alloc_pcluster(unsigned int nrpages) +static struct z_erofs_pcluster *z_erofs_alloc_pcluster(unsigned int size) { - int i; + unsigned int nrpages = PAGE_ALIGN(size) >> PAGE_SHIFT; + struct z_erofs_pcluster_slab *pcs = pcluster_pool; - for (i = 0; i < ARRAY_SIZE(pcluster_pool); ++i) { - struct z_erofs_pcluster_slab *pcs = pcluster_pool + i; + for (; pcs < pcluster_pool + ARRAY_SIZE(pcluster_pool); ++pcs) { struct z_erofs_pcluster *pcl; if (nrpages > pcs->maxpages) continue; - pcl = kmem_cache_zalloc(pcs->slab, GFP_NOFS); + pcl = kmem_cache_zalloc(pcs->slab, GFP_KERNEL); if (!pcl) return ERR_PTR(-ENOMEM); - pcl->pclusterpages = nrpages; return pcl; } return ERR_PTR(-EINVAL); @@ -97,708 +284,821 @@ static void z_erofs_free_pcluster(struct z_erofs_pcluster *pcl) DBG_BUGON(1); } -/* how to allocate cached pages for a pcluster */ -enum z_erofs_cache_alloctype { - DONTALLOC, /* don't allocate any cached pages */ - /* - * try to use cached I/O if page allocation succeeds or fallback - * to in-place I/O instead to avoid any direct reclaim. - */ - TRYALLOC, -}; +static struct workqueue_struct *z_erofs_workqueue __read_mostly; -/* - * tagged pointer with 1-bit tag for all compressed pages - * tag 0 - the page is just found with an extra page reference - */ -typedef tagptr1_t compressed_page_t; +#ifdef CONFIG_EROFS_FS_PCPU_KTHREAD +static struct kthread_worker __rcu **z_erofs_pcpu_workers; +static atomic_t erofs_percpu_workers_initialized = ATOMIC_INIT(0); -#define tag_compressed_page_justfound(page) \ - tagptr_fold(compressed_page_t, page, 1) +static void erofs_destroy_percpu_workers(void) +{ + struct kthread_worker *worker; + unsigned int cpu; + + for_each_possible_cpu(cpu) { + worker = rcu_dereference_protected( + z_erofs_pcpu_workers[cpu], 1); + rcu_assign_pointer(z_erofs_pcpu_workers[cpu], NULL); + if (worker) + kthread_destroy_worker(worker); + } + kfree(z_erofs_pcpu_workers); +} -static struct workqueue_struct *z_erofs_workqueue __read_mostly; +static struct kthread_worker *erofs_init_percpu_worker(int cpu) +{ + struct kthread_worker *worker = + kthread_run_worker_on_cpu(cpu, 0, "erofs_worker/%u"); + + if (IS_ERR(worker)) + return worker; + if (IS_ENABLED(CONFIG_EROFS_FS_PCPU_KTHREAD_HIPRI)) + sched_set_fifo_low(worker->task); + return worker; +} -void z_erofs_exit_zip_subsystem(void) +static int erofs_init_percpu_workers(void) { - destroy_workqueue(z_erofs_workqueue); - z_erofs_destroy_pcluster_pool(); + struct kthread_worker *worker; + unsigned int cpu; + + z_erofs_pcpu_workers = kcalloc(num_possible_cpus(), + sizeof(struct kthread_worker *), GFP_ATOMIC); + if (!z_erofs_pcpu_workers) + return -ENOMEM; + + for_each_online_cpu(cpu) { /* could miss cpu{off,on}line? */ + worker = erofs_init_percpu_worker(cpu); + if (!IS_ERR(worker)) + rcu_assign_pointer(z_erofs_pcpu_workers[cpu], worker); + } + return 0; } -static inline int z_erofs_init_workqueue(void) +#ifdef CONFIG_HOTPLUG_CPU +static DEFINE_SPINLOCK(z_erofs_pcpu_worker_lock); +static enum cpuhp_state erofs_cpuhp_state; + +static int erofs_cpu_online(unsigned int cpu) { - const unsigned int onlinecpus = num_possible_cpus(); + struct kthread_worker *worker, *old; + + worker = erofs_init_percpu_worker(cpu); + if (IS_ERR(worker)) + return PTR_ERR(worker); + + spin_lock(&z_erofs_pcpu_worker_lock); + old = rcu_dereference_protected(z_erofs_pcpu_workers[cpu], + lockdep_is_held(&z_erofs_pcpu_worker_lock)); + if (!old) + rcu_assign_pointer(z_erofs_pcpu_workers[cpu], worker); + spin_unlock(&z_erofs_pcpu_worker_lock); + if (old) + kthread_destroy_worker(worker); + return 0; +} - /* - * no need to spawn too many threads, limiting threads could minimum - * scheduling overhead, perhaps per-CPU threads should be better? - */ - z_erofs_workqueue = alloc_workqueue("erofs_unzipd", - WQ_UNBOUND | WQ_HIGHPRI, - onlinecpus + onlinecpus / 4); - return z_erofs_workqueue ? 0 : -ENOMEM; +static int erofs_cpu_offline(unsigned int cpu) +{ + struct kthread_worker *worker; + + spin_lock(&z_erofs_pcpu_worker_lock); + worker = rcu_dereference_protected(z_erofs_pcpu_workers[cpu], + lockdep_is_held(&z_erofs_pcpu_worker_lock)); + rcu_assign_pointer(z_erofs_pcpu_workers[cpu], NULL); + spin_unlock(&z_erofs_pcpu_worker_lock); + + synchronize_rcu(); + if (worker) + kthread_destroy_worker(worker); + return 0; } -int __init z_erofs_init_zip_subsystem(void) +static int erofs_cpu_hotplug_init(void) { - int err = z_erofs_create_pcluster_pool(); + int state; + + state = cpuhp_setup_state_nocalls(CPUHP_AP_ONLINE_DYN, + "fs/erofs:online", erofs_cpu_online, erofs_cpu_offline); + if (state < 0) + return state; + + erofs_cpuhp_state = state; + return 0; +} + +static void erofs_cpu_hotplug_destroy(void) +{ + if (erofs_cpuhp_state) + cpuhp_remove_state_nocalls(erofs_cpuhp_state); +} +#else /* !CONFIG_HOTPLUG_CPU */ +static inline int erofs_cpu_hotplug_init(void) { return 0; } +static inline void erofs_cpu_hotplug_destroy(void) {} +#endif/* CONFIG_HOTPLUG_CPU */ +static int z_erofs_init_pcpu_workers(struct super_block *sb) +{ + int err; + + if (atomic_xchg(&erofs_percpu_workers_initialized, 1)) + return 0; + + err = erofs_init_percpu_workers(); + if (err) { + erofs_err(sb, "per-cpu workers: failed to allocate."); + goto err_init_percpu_workers; + } + + err = erofs_cpu_hotplug_init(); + if (err < 0) { + erofs_err(sb, "per-cpu workers: failed CPU hotplug init."); + goto err_cpuhp_init; + } + erofs_info(sb, "initialized per-cpu workers successfully."); + return err; + +err_cpuhp_init: + erofs_destroy_percpu_workers(); +err_init_percpu_workers: + atomic_set(&erofs_percpu_workers_initialized, 0); + return err; +} + +static void z_erofs_destroy_pcpu_workers(void) +{ + if (!atomic_xchg(&erofs_percpu_workers_initialized, 0)) + return; + erofs_cpu_hotplug_destroy(); + erofs_destroy_percpu_workers(); +} +#else /* !CONFIG_EROFS_FS_PCPU_KTHREAD */ +static inline int z_erofs_init_pcpu_workers(struct super_block *sb) { return 0; } +static inline void z_erofs_destroy_pcpu_workers(void) {} +#endif/* CONFIG_EROFS_FS_PCPU_KTHREAD */ + +void z_erofs_exit_subsystem(void) +{ + z_erofs_destroy_pcpu_workers(); + destroy_workqueue(z_erofs_workqueue); + z_erofs_destroy_pcluster_pool(); + z_erofs_crypto_disable_all_engines(); + z_erofs_exit_decompressor(); +} + +int __init z_erofs_init_subsystem(void) +{ + int err = z_erofs_init_decompressor(); if (err) - return err; - err = z_erofs_init_workqueue(); + goto err_decompressor; + + err = z_erofs_create_pcluster_pool(); if (err) - z_erofs_destroy_pcluster_pool(); + goto err_pcluster_pool; + + z_erofs_workqueue = alloc_workqueue("erofs_worker", + WQ_UNBOUND | WQ_HIGHPRI, num_possible_cpus()); + if (!z_erofs_workqueue) { + err = -ENOMEM; + goto err_workqueue_init; + } + + return err; + +err_workqueue_init: + z_erofs_destroy_pcluster_pool(); +err_pcluster_pool: + z_erofs_exit_decompressor(); +err_decompressor: return err; } -enum z_erofs_collectmode { - COLLECT_SECONDARY, - COLLECT_PRIMARY, - /* - * The current collection was the tail of an exist chain, in addition - * that the previous processed chained collections are all decided to - * be hooked up to it. - * A new chain will be created for the remaining collections which are - * not processed yet, therefore different from COLLECT_PRIMARY_FOLLOWED, - * the next collection cannot reuse the whole page safely in - * the following scenario: - * ________________________________________________________________ - * | tail (partial) page | head (partial) page | - * | (belongs to the next cl) | (belongs to the current cl) | - * |_______PRIMARY_FOLLOWED_______|________PRIMARY_HOOKED___________| - */ - COLLECT_PRIMARY_HOOKED, +enum z_erofs_pclustermode { + /* It has previously been linked into another processing chain */ + Z_EROFS_PCLUSTER_INFLIGHT, /* - * a weak form of COLLECT_PRIMARY_FOLLOWED, the difference is that it - * could be dispatched into bypass queue later due to uptodated managed - * pages. All related online pages cannot be reused for inplace I/O (or - * pagevec) since it can be directly decoded without I/O submission. + * A weaker form of Z_EROFS_PCLUSTER_FOLLOWED; the difference is that it + * may be dispatched to the bypass queue later due to uptodated managed + * folios. All file-backed folios related to this pcluster cannot be + * reused for in-place I/O (or bvpage) since the pcluster may be decoded + * in a separate queue (and thus out of order). */ - COLLECT_PRIMARY_FOLLOWED_NOINPLACE, + Z_EROFS_PCLUSTER_FOLLOWED_NOINPLACE, /* - * The current collection has been linked with the owned chain, and - * could also be linked with the remaining collections, which means - * if the processing page is the tail page of the collection, thus - * the current collection can safely use the whole page (since - * the previous collection is under control) for in-place I/O, as - * illustrated below: - * ________________________________________________________________ - * | tail (partial) page | head (partial) page | - * | (of the current cl) | (of the previous collection) | - * | PRIMARY_FOLLOWED or | | - * |_____PRIMARY_HOOKED___|____________PRIMARY_FOLLOWED____________| - * - * [ (*) the above page can be used as inplace I/O. ] + * The pcluster has just been linked to our processing chain. + * File-backed folios (except for the head page) related to it can be + * used for in-place I/O (or bvpage). */ - COLLECT_PRIMARY_FOLLOWED, + Z_EROFS_PCLUSTER_FOLLOWED, }; -struct z_erofs_collector { - struct z_erofs_pagevec_ctor vector; +struct z_erofs_frontend { + struct inode *const inode; + struct erofs_map_blocks map; + struct z_erofs_bvec_iter biter; - struct z_erofs_pcluster *pcl, *tailpcl; - struct z_erofs_collection *cl; - /* a pointer used to pick up inplace I/O pages */ - struct page **icpage_ptr; - z_erofs_next_pcluster_t owned_head; + struct page *pagepool; + struct page *candidate_bvpage; + struct z_erofs_pcluster *pcl, *head; + enum z_erofs_pclustermode mode; + + erofs_off_t headoffset; - enum z_erofs_collectmode mode; + /* a pointer used to pick up inplace I/O pages */ + unsigned int icur; }; -struct z_erofs_decompress_frontend { - struct inode *const inode; +#define Z_EROFS_DEFINE_FRONTEND(fe, i, ho) struct z_erofs_frontend fe = { \ + .inode = i, .head = Z_EROFS_PCLUSTER_TAIL, \ + .mode = Z_EROFS_PCLUSTER_FOLLOWED, .headoffset = ho } - struct z_erofs_collector clt; - struct erofs_map_blocks map; +static bool z_erofs_should_alloc_cache(struct z_erofs_frontend *fe) +{ + unsigned int cachestrategy = EROFS_I_SB(fe->inode)->opt.cache_strategy; - bool readahead; - /* used for applying cache strategy on the fly */ - bool backmost; - erofs_off_t headoffset; -}; + if (cachestrategy <= EROFS_ZIP_CACHE_DISABLED) + return false; -#define COLLECTOR_INIT() { \ - .owned_head = Z_EROFS_PCLUSTER_TAIL, \ - .mode = COLLECT_PRIMARY_FOLLOWED } + if (!(fe->map.m_flags & EROFS_MAP_FULL_MAPPED)) + return true; -#define DECOMPRESS_FRONTEND_INIT(__i) { \ - .inode = __i, .clt = COLLECTOR_INIT(), \ - .backmost = true, } + if (cachestrategy >= EROFS_ZIP_CACHE_READAROUND && + fe->map.m_la < fe->headoffset) + return true; -static struct page *z_pagemap_global[Z_EROFS_VMAP_GLOBAL_PAGES]; -static DEFINE_MUTEX(z_pagemap_global_lock); + return false; +} -static void preload_compressed_pages(struct z_erofs_collector *clt, - struct address_space *mc, - enum z_erofs_cache_alloctype type, - struct page **pagepool) +static void z_erofs_bind_cache(struct z_erofs_frontend *fe) { - struct z_erofs_pcluster *pcl = clt->pcl; - bool standalone = true; + struct address_space *mc = MNGD_MAPPING(EROFS_I_SB(fe->inode)); + struct z_erofs_pcluster *pcl = fe->pcl; + unsigned int pclusterpages = z_erofs_pclusterpages(pcl); + bool shouldalloc = z_erofs_should_alloc_cache(fe); + pgoff_t poff = pcl->pos >> PAGE_SHIFT; + bool may_bypass = true; + /* Optimistic allocation, as in-place I/O can be used as a fallback */ gfp_t gfp = (mapping_gfp_mask(mc) & ~__GFP_DIRECT_RECLAIM) | __GFP_NOMEMALLOC | __GFP_NORETRY | __GFP_NOWARN; - struct page **pages; - pgoff_t index; + struct folio *folio, *newfolio; + unsigned int i; - if (clt->mode < COLLECT_PRIMARY_FOLLOWED) + if (i_blocksize(fe->inode) != PAGE_SIZE || + fe->mode < Z_EROFS_PCLUSTER_FOLLOWED) return; - pages = pcl->compressed_pages; - index = pcl->obj.index; - for (; index < pcl->obj.index + pcl->pclusterpages; ++index, ++pages) { - struct page *page; - compressed_page_t t; - struct page *newpage = NULL; - - /* the compressed page was loaded before */ - if (READ_ONCE(*pages)) + for (i = 0; i < pclusterpages; ++i) { + /* Inaccurate check w/o locking to avoid unneeded lookups */ + if (READ_ONCE(pcl->compressed_bvecs[i].page)) continue; - page = find_get_page(mc, index); + folio = filemap_get_folio(mc, poff + i); + if (IS_ERR(folio)) { + may_bypass = false; + if (!shouldalloc) + continue; - if (page) { - t = tag_compressed_page_justfound(page); - } else { - /* I/O is needed, no possible to decompress directly */ - standalone = false; - switch (type) { - case TRYALLOC: - newpage = erofs_allocpage(pagepool, gfp); - if (!newpage) - continue; - set_page_private(newpage, - Z_EROFS_PREALLOCATED_PAGE); - t = tag_compressed_page_justfound(newpage); - break; - default: /* DONTALLOC */ + /* + * Allocate a managed folio for cached I/O, or it may be + * then filled with a file-backed folio for in-place I/O + */ + newfolio = filemap_alloc_folio(gfp, 0, NULL); + if (!newfolio) continue; - } + newfolio->private = Z_EROFS_PREALLOCATED_FOLIO; + folio = NULL; } - - if (!cmpxchg_relaxed(pages, NULL, tagptr_cast_ptr(t))) + spin_lock(&pcl->lockref.lock); + if (!pcl->compressed_bvecs[i].page) { + pcl->compressed_bvecs[i].page = + folio_page(folio ?: newfolio, 0); + spin_unlock(&pcl->lockref.lock); continue; - - if (page) - put_page(page); - else if (newpage) - erofs_pagepool_add(pagepool, newpage); + } + spin_unlock(&pcl->lockref.lock); + folio_put(folio ?: newfolio); } /* - * don't do inplace I/O if all compressed pages are available in - * managed cache since it can be moved to the bypass queue instead. + * Don't perform in-place I/O if all compressed pages are available in + * the managed cache, as the pcluster can be moved to the bypass queue. */ - if (standalone) - clt->mode = COLLECT_PRIMARY_FOLLOWED_NOINPLACE; + if (may_bypass) + fe->mode = Z_EROFS_PCLUSTER_FOLLOWED_NOINPLACE; } -/* called by erofs_shrinker to get rid of all compressed_pages */ -int erofs_try_to_free_all_cached_pages(struct erofs_sb_info *sbi, - struct erofs_workgroup *grp) +/* (erofs_shrinker) disconnect cached encoded data with pclusters */ +static int erofs_try_to_free_all_cached_folios(struct erofs_sb_info *sbi, + struct z_erofs_pcluster *pcl) { - struct z_erofs_pcluster *const pcl = - container_of(grp, struct z_erofs_pcluster, obj); + unsigned int pclusterpages = z_erofs_pclusterpages(pcl); + struct folio *folio; int i; - DBG_BUGON(z_erofs_is_inline_pcluster(pcl)); - /* - * refcount of workgroup is now freezed as 1, - * therefore no need to worry about available decompression users. - */ - for (i = 0; i < pcl->pclusterpages; ++i) { - struct page *page = pcl->compressed_pages[i]; - - if (!page) - continue; - - /* block other users from reclaiming or migrating the page */ - if (!trylock_page(page)) - return -EBUSY; - - if (!erofs_page_is_managed(sbi, page)) - continue; + DBG_BUGON(pcl->from_meta); + /* Each cached folio contains one page unless bs > ps is supported */ + for (i = 0; i < pclusterpages; ++i) { + if (pcl->compressed_bvecs[i].page) { + folio = page_folio(pcl->compressed_bvecs[i].page); + /* Avoid reclaiming or migrating this folio */ + if (!folio_trylock(folio)) + return -EBUSY; - /* barrier is implied in the following 'unlock_page' */ - WRITE_ONCE(pcl->compressed_pages[i], NULL); - detach_page_private(page); - unlock_page(page); + if (!erofs_folio_is_managed(sbi, folio)) + continue; + pcl->compressed_bvecs[i].page = NULL; + folio_detach_private(folio); + folio_unlock(folio); + } } return 0; } -int erofs_try_to_free_cached_page(struct page *page) +static bool z_erofs_cache_release_folio(struct folio *folio, gfp_t gfp) { - struct z_erofs_pcluster *const pcl = (void *)page_private(page); - int ret = 0; /* 0 - busy */ + struct z_erofs_pcluster *pcl = folio_get_private(folio); + struct z_erofs_bvec *bvec = pcl->compressed_bvecs; + struct z_erofs_bvec *end = bvec + z_erofs_pclusterpages(pcl); + bool ret; - if (erofs_workgroup_try_to_freeze(&pcl->obj, 1)) { - unsigned int i; + if (!folio_test_private(folio)) + return true; - DBG_BUGON(z_erofs_is_inline_pcluster(pcl)); - for (i = 0; i < pcl->pclusterpages; ++i) { - if (pcl->compressed_pages[i] == page) { - WRITE_ONCE(pcl->compressed_pages[i], NULL); - ret = 1; + ret = false; + spin_lock(&pcl->lockref.lock); + if (pcl->lockref.count <= 0) { + DBG_BUGON(pcl->from_meta); + for (; bvec < end; ++bvec) { + if (bvec->page && page_folio(bvec->page) == folio) { + bvec->page = NULL; + folio_detach_private(folio); + ret = true; break; } } - erofs_workgroup_unfreeze(&pcl->obj, 1); - - if (ret) - detach_page_private(page); } + spin_unlock(&pcl->lockref.lock); return ret; } -/* page_type must be Z_EROFS_PAGE_TYPE_EXCLUSIVE */ -static bool z_erofs_try_inplace_io(struct z_erofs_collector *clt, - struct page *page) +/* + * It will be called only on inode eviction. In case that there are still some + * decompression requests in progress, wait with rescheduling for a bit here. + * An extra lock could be introduced instead but it seems unnecessary. + */ +static void z_erofs_cache_invalidate_folio(struct folio *folio, + size_t offset, size_t length) { - struct z_erofs_pcluster *const pcl = clt->pcl; + const size_t stop = length + offset; - while (clt->icpage_ptr > pcl->compressed_pages) - if (!cmpxchg(--clt->icpage_ptr, NULL, page)) - return true; - return false; + /* Check for potential overflow in debug mode */ + DBG_BUGON(stop > folio_size(folio) || stop < length); + + if (offset == 0 && stop == folio_size(folio)) + while (!z_erofs_cache_release_folio(folio, 0)) + cond_resched(); } -/* callers must be with collection lock held */ -static int z_erofs_attach_page(struct z_erofs_collector *clt, - struct page *page, enum z_erofs_page_type type, - bool pvec_safereuse) +static const struct address_space_operations z_erofs_cache_aops = { + .release_folio = z_erofs_cache_release_folio, + .invalidate_folio = z_erofs_cache_invalidate_folio, +}; + +int z_erofs_init_super(struct super_block *sb) { - int ret; + struct inode *inode; + int err; - /* give priority for inplaceio */ - if (clt->mode >= COLLECT_PRIMARY && - type == Z_EROFS_PAGE_TYPE_EXCLUSIVE && - z_erofs_try_inplace_io(clt, page)) - return 0; + err = z_erofs_init_pcpu_workers(sb); + if (err) + return err; - ret = z_erofs_pagevec_enqueue(&clt->vector, page, type, - pvec_safereuse); - clt->cl->vcnt += (unsigned int)ret; - return ret ? 0 : -EAGAIN; + inode = new_inode(sb); + if (!inode) + return -ENOMEM; + set_nlink(inode, 1); + inode->i_size = OFFSET_MAX; + inode->i_mapping->a_ops = &z_erofs_cache_aops; + mapping_set_gfp_mask(inode->i_mapping, GFP_KERNEL); + EROFS_SB(sb)->managed_cache = inode; + xa_init(&EROFS_SB(sb)->managed_pslots); + return 0; } -static void z_erofs_try_to_claim_pcluster(struct z_erofs_collector *clt) +/* callers must be with pcluster lock held */ +static int z_erofs_attach_page(struct z_erofs_frontend *fe, + struct z_erofs_bvec *bvec, bool exclusive) { - struct z_erofs_pcluster *pcl = clt->pcl; - z_erofs_next_pcluster_t *owned_head = &clt->owned_head; - - /* type 1, nil pcluster (this pcluster doesn't belong to any chain.) */ - if (cmpxchg(&pcl->next, Z_EROFS_PCLUSTER_NIL, - *owned_head) == Z_EROFS_PCLUSTER_NIL) { - *owned_head = &pcl->next; - /* so we can attach this pcluster to our submission chain. */ - clt->mode = COLLECT_PRIMARY_FOLLOWED; - return; - } + struct z_erofs_pcluster *pcl = fe->pcl; + int ret; - /* - * type 2, link to the end of an existing open chain, be careful - * that its submission is controlled by the original attached chain. - */ - if (cmpxchg(&pcl->next, Z_EROFS_PCLUSTER_TAIL, - *owned_head) == Z_EROFS_PCLUSTER_TAIL) { - *owned_head = Z_EROFS_PCLUSTER_TAIL; - clt->mode = COLLECT_PRIMARY_HOOKED; - clt->tailpcl = NULL; - return; + if (exclusive) { + /* Inplace I/O is limited to one page for uncompressed data */ + if (pcl->algorithmformat < Z_EROFS_COMPRESSION_MAX || + fe->icur <= 1) { + /* Try to prioritize inplace I/O here */ + spin_lock(&pcl->lockref.lock); + while (fe->icur > 0) { + if (pcl->compressed_bvecs[--fe->icur].page) + continue; + pcl->compressed_bvecs[fe->icur] = *bvec; + spin_unlock(&pcl->lockref.lock); + return 0; + } + spin_unlock(&pcl->lockref.lock); + } + + /* otherwise, check if it can be used as a bvpage */ + if (fe->mode >= Z_EROFS_PCLUSTER_FOLLOWED && + !fe->candidate_bvpage) + fe->candidate_bvpage = bvec->page; } - /* type 3, it belongs to a chain, but it isn't the end of the chain */ - clt->mode = COLLECT_PRIMARY; + ret = z_erofs_bvec_enqueue(&fe->biter, bvec, &fe->candidate_bvpage, + &fe->pagepool); + fe->pcl->vcnt += (ret >= 0); + return ret; } -static int z_erofs_lookup_collection(struct z_erofs_collector *clt, - struct inode *inode, - struct erofs_map_blocks *map) +static bool z_erofs_get_pcluster(struct z_erofs_pcluster *pcl) { - struct z_erofs_pcluster *pcl = clt->pcl; - struct z_erofs_collection *cl; - unsigned int length; - - /* to avoid unexpected loop formed by corrupted images */ - if (clt->owned_head == &pcl->next || pcl == clt->tailpcl) { - DBG_BUGON(1); - return -EFSCORRUPTED; - } - - cl = z_erofs_primarycollection(pcl); - if (cl->pageofs != (map->m_la & ~PAGE_MASK)) { - DBG_BUGON(1); - return -EFSCORRUPTED; - } - - length = READ_ONCE(pcl->length); - if (length & Z_EROFS_PCLUSTER_FULL_LENGTH) { - if ((map->m_llen << Z_EROFS_PCLUSTER_LENGTH_BIT) > length) { - DBG_BUGON(1); - return -EFSCORRUPTED; - } - } else { - unsigned int llen = map->m_llen << Z_EROFS_PCLUSTER_LENGTH_BIT; - - if (map->m_flags & EROFS_MAP_FULL_MAPPED) - llen |= Z_EROFS_PCLUSTER_FULL_LENGTH; + if (lockref_get_not_zero(&pcl->lockref)) + return true; - while (llen > length && - length != cmpxchg_relaxed(&pcl->length, length, llen)) { - cpu_relax(); - length = READ_ONCE(pcl->length); - } + spin_lock(&pcl->lockref.lock); + if (__lockref_is_dead(&pcl->lockref)) { + spin_unlock(&pcl->lockref.lock); + return false; } - mutex_lock(&cl->lock); - /* used to check tail merging loop due to corrupted images */ - if (clt->owned_head == Z_EROFS_PCLUSTER_TAIL) - clt->tailpcl = pcl; - z_erofs_try_to_claim_pcluster(clt); - clt->cl = cl; - return 0; + if (!pcl->lockref.count++) + atomic_long_dec(&erofs_global_shrink_cnt); + spin_unlock(&pcl->lockref.lock); + return true; } -static int z_erofs_register_collection(struct z_erofs_collector *clt, - struct inode *inode, - struct erofs_map_blocks *map) +static int z_erofs_register_pcluster(struct z_erofs_frontend *fe) { - bool ztailpacking = map->m_flags & EROFS_MAP_META; - struct z_erofs_pcluster *pcl; - struct z_erofs_collection *cl; - struct erofs_workgroup *grp; + struct erofs_map_blocks *map = &fe->map; + struct super_block *sb = fe->inode->i_sb; + struct erofs_sb_info *sbi = EROFS_SB(sb); + struct z_erofs_pcluster *pcl, *pre; + unsigned int pageofs_in; int err; - if (!(map->m_flags & EROFS_MAP_ENCODED)) { - DBG_BUGON(1); - return -EFSCORRUPTED; - } - - /* no available pcluster, let's allocate one */ - pcl = z_erofs_alloc_pcluster(ztailpacking ? 1 : - map->m_plen >> PAGE_SHIFT); + pageofs_in = erofs_blkoff(sb, map->m_pa); + pcl = z_erofs_alloc_pcluster(pageofs_in + map->m_plen); if (IS_ERR(pcl)) return PTR_ERR(pcl); - atomic_set(&pcl->obj.refcount, 1); + lockref_init(&pcl->lockref); /* one ref for this request */ pcl->algorithmformat = map->m_algorithmformat; - pcl->length = (map->m_llen << Z_EROFS_PCLUSTER_LENGTH_BIT) | - (map->m_flags & EROFS_MAP_FULL_MAPPED ? - Z_EROFS_PCLUSTER_FULL_LENGTH : 0); - - /* new pclusters should be claimed as type 1, primary and followed */ - pcl->next = clt->owned_head; - clt->mode = COLLECT_PRIMARY_FOLLOWED; - - cl = z_erofs_primarycollection(pcl); - cl->pageofs = map->m_la & ~PAGE_MASK; + pcl->pclustersize = map->m_plen; + pcl->length = 0; + pcl->partial = true; + pcl->next = fe->head; + pcl->pos = map->m_pa; + pcl->pageofs_in = pageofs_in; + pcl->pageofs_out = map->m_la & ~PAGE_MASK; + pcl->from_meta = map->m_flags & EROFS_MAP_META; + fe->mode = Z_EROFS_PCLUSTER_FOLLOWED; /* * lock all primary followed works before visible to others * and mutex_trylock *never* fails for a new pcluster. */ - mutex_init(&cl->lock); - DBG_BUGON(!mutex_trylock(&cl->lock)); - - if (ztailpacking) { - pcl->obj.index = 0; /* which indicates ztailpacking */ - pcl->pageofs_in = erofs_blkoff(map->m_pa); - pcl->tailpacking_size = map->m_plen; - } else { - pcl->obj.index = map->m_pa >> PAGE_SHIFT; - - grp = erofs_insert_workgroup(inode->i_sb, &pcl->obj); - if (IS_ERR(grp)) { - err = PTR_ERR(grp); - goto err_out; + mutex_init(&pcl->lock); + DBG_BUGON(!mutex_trylock(&pcl->lock)); + + if (!pcl->from_meta) { + while (1) { + xa_lock(&sbi->managed_pslots); + pre = __xa_cmpxchg(&sbi->managed_pslots, pcl->pos, + NULL, pcl, GFP_KERNEL); + if (!pre || xa_is_err(pre) || z_erofs_get_pcluster(pre)) { + xa_unlock(&sbi->managed_pslots); + break; + } + /* try to legitimize the current in-tree one */ + xa_unlock(&sbi->managed_pslots); + cond_resched(); } - - if (grp != &pcl->obj) { - clt->pcl = container_of(grp, - struct z_erofs_pcluster, obj); + if (xa_is_err(pre)) { + err = xa_err(pre); + goto err_out; + } else if (pre) { + fe->pcl = pre; err = -EEXIST; goto err_out; } } - /* used to check tail merging loop due to corrupted images */ - if (clt->owned_head == Z_EROFS_PCLUSTER_TAIL) - clt->tailpcl = pcl; - clt->owned_head = &pcl->next; - clt->pcl = pcl; - clt->cl = cl; + fe->head = fe->pcl = pcl; return 0; err_out: - mutex_unlock(&cl->lock); + mutex_unlock(&pcl->lock); z_erofs_free_pcluster(pcl); return err; } -static int z_erofs_collector_begin(struct z_erofs_collector *clt, - struct inode *inode, - struct erofs_map_blocks *map) +static int z_erofs_pcluster_begin(struct z_erofs_frontend *fe) { - struct erofs_workgroup *grp; + struct erofs_map_blocks *map = &fe->map; + struct super_block *sb = fe->inode->i_sb; + struct z_erofs_pcluster *pcl = NULL; + void *ptr; int ret; - DBG_BUGON(clt->cl); - - /* must be Z_EROFS_PCLUSTER_TAIL or pointed to previous collection */ - DBG_BUGON(clt->owned_head == Z_EROFS_PCLUSTER_NIL); - DBG_BUGON(clt->owned_head == Z_EROFS_PCLUSTER_TAIL_CLOSED); - - if (map->m_flags & EROFS_MAP_META) { - if ((map->m_pa & ~PAGE_MASK) + map->m_plen > PAGE_SIZE) { - DBG_BUGON(1); - return -EFSCORRUPTED; + DBG_BUGON(fe->pcl); + /* must be Z_EROFS_PCLUSTER_TAIL or pointed to previous pcluster */ + DBG_BUGON(!fe->head); + + if (!(map->m_flags & EROFS_MAP_META)) { + while (1) { + rcu_read_lock(); + pcl = xa_load(&EROFS_SB(sb)->managed_pslots, map->m_pa); + if (!pcl || z_erofs_get_pcluster(pcl)) { + DBG_BUGON(pcl && map->m_pa != pcl->pos); + rcu_read_unlock(); + break; + } + rcu_read_unlock(); } - goto tailpacking; } - grp = erofs_find_workgroup(inode->i_sb, map->m_pa >> PAGE_SHIFT); - if (grp) { - clt->pcl = container_of(grp, struct z_erofs_pcluster, obj); + if (pcl) { + fe->pcl = pcl; + ret = -EEXIST; } else { -tailpacking: - ret = z_erofs_register_collection(clt, inode, map); - if (!ret) - goto out; - if (ret != -EEXIST) - return ret; + ret = z_erofs_register_pcluster(fe); } - ret = z_erofs_lookup_collection(clt, inode, map); - if (ret) { - erofs_workgroup_put(&clt->pcl->obj); + if (ret == -EEXIST) { + mutex_lock(&fe->pcl->lock); + /* check if this pcluster hasn't been linked into any chain. */ + if (!cmpxchg(&fe->pcl->next, NULL, fe->head)) { + /* .. so it can be attached to our submission chain */ + fe->head = fe->pcl; + fe->mode = Z_EROFS_PCLUSTER_FOLLOWED; + } else { /* otherwise, it belongs to an inflight chain */ + fe->mode = Z_EROFS_PCLUSTER_INFLIGHT; + } + } else if (ret) { return ret; } -out: - z_erofs_pagevec_ctor_init(&clt->vector, Z_EROFS_NR_INLINE_PAGEVECS, - clt->cl->pagevec, clt->cl->vcnt); - /* since file-backed online pages are traversed in reverse order */ - clt->icpage_ptr = clt->pcl->compressed_pages + - z_erofs_pclusterpages(clt->pcl); + z_erofs_bvec_iter_begin(&fe->biter, &fe->pcl->bvset, + Z_EROFS_INLINE_BVECS, fe->pcl->vcnt); + if (!fe->pcl->from_meta) { + /* bind cache first when cached decompression is preferred */ + z_erofs_bind_cache(fe); + } else { + ret = erofs_init_metabuf(&map->buf, sb, + erofs_inode_in_metabox(fe->inode)); + if (ret) + return ret; + ptr = erofs_bread(&map->buf, map->m_pa, false); + if (IS_ERR(ptr)) { + ret = PTR_ERR(ptr); + erofs_err(sb, "failed to get inline folio %d", ret); + return ret; + } + folio_get(page_folio(map->buf.page)); + WRITE_ONCE(fe->pcl->compressed_bvecs[0].page, map->buf.page); + fe->pcl->pageofs_in = map->m_pa & ~PAGE_MASK; + fe->mode = Z_EROFS_PCLUSTER_FOLLOWED_NOINPLACE; + } + /* file-backed inplace I/O pages are traversed in reverse order */ + fe->icur = z_erofs_pclusterpages(fe->pcl); return 0; } -/* - * keep in mind that no referenced pclusters will be freed - * only after a RCU grace period. - */ static void z_erofs_rcu_callback(struct rcu_head *head) { - struct z_erofs_collection *const cl = - container_of(head, struct z_erofs_collection, rcu); - - z_erofs_free_pcluster(container_of(cl, struct z_erofs_pcluster, - primary_collection)); -} - -void erofs_workgroup_free_rcu(struct erofs_workgroup *grp) -{ - struct z_erofs_pcluster *const pcl = - container_of(grp, struct z_erofs_pcluster, obj); - struct z_erofs_collection *const cl = z_erofs_primarycollection(pcl); - - call_rcu(&cl->rcu, z_erofs_rcu_callback); -} - -static void z_erofs_collection_put(struct z_erofs_collection *cl) -{ - struct z_erofs_pcluster *const pcl = - container_of(cl, struct z_erofs_pcluster, primary_collection); - - erofs_workgroup_put(&pcl->obj); + z_erofs_free_pcluster(container_of(head, struct z_erofs_pcluster, rcu)); } -static bool z_erofs_collector_end(struct z_erofs_collector *clt) +static bool __erofs_try_to_release_pcluster(struct erofs_sb_info *sbi, + struct z_erofs_pcluster *pcl) { - struct z_erofs_collection *cl = clt->cl; - - if (!cl) + if (pcl->lockref.count) return false; - z_erofs_pagevec_ctor_exit(&clt->vector, false); - mutex_unlock(&cl->lock); + /* + * Note that all cached folios should be detached before deleted from + * the XArray. Otherwise some folios could be still attached to the + * orphan old pcluster when the new one is available in the tree. + */ + if (erofs_try_to_free_all_cached_folios(sbi, pcl)) + return false; /* - * if all pending pages are added, don't hold its reference - * any longer if the pcluster isn't hosted by ourselves. + * It's impossible to fail after the pcluster is freezed, but in order + * to avoid some race conditions, add a DBG_BUGON to observe this. */ - if (clt->mode < COLLECT_PRIMARY_FOLLOWED_NOINPLACE) - z_erofs_collection_put(cl); + DBG_BUGON(__xa_erase(&sbi->managed_pslots, pcl->pos) != pcl); - clt->cl = NULL; + lockref_mark_dead(&pcl->lockref); return true; } -static bool should_alloc_managed_pages(struct z_erofs_decompress_frontend *fe, - unsigned int cachestrategy, - erofs_off_t la) +static bool erofs_try_to_release_pcluster(struct erofs_sb_info *sbi, + struct z_erofs_pcluster *pcl) { - if (cachestrategy <= EROFS_ZIP_CACHE_DISABLED) - return false; - - if (fe->backmost) - return true; - - return cachestrategy >= EROFS_ZIP_CACHE_READAROUND && - la < fe->headoffset; + bool free; + + spin_lock(&pcl->lockref.lock); + free = __erofs_try_to_release_pcluster(sbi, pcl); + spin_unlock(&pcl->lockref.lock); + if (free) { + atomic_long_dec(&erofs_global_shrink_cnt); + call_rcu(&pcl->rcu, z_erofs_rcu_callback); + } + return free; } -static int z_erofs_do_read_page(struct z_erofs_decompress_frontend *fe, - struct page *page, struct page **pagepool) +unsigned long z_erofs_shrink_scan(struct erofs_sb_info *sbi, unsigned long nr) { - struct inode *const inode = fe->inode; - struct erofs_sb_info *const sbi = EROFS_I_SB(inode); - struct erofs_map_blocks *const map = &fe->map; - struct z_erofs_collector *const clt = &fe->clt; - const loff_t offset = page_offset(page); - bool tight = true; - - enum z_erofs_cache_alloctype cache_strategy; - enum z_erofs_page_type page_type; - unsigned int cur, end, spiltted, index; - int err = 0; + struct z_erofs_pcluster *pcl; + unsigned long index, freed = 0; - /* register locked file pages as online pages in pack */ - z_erofs_onlinepage_init(page); + xa_lock(&sbi->managed_pslots); + xa_for_each(&sbi->managed_pslots, index, pcl) { + /* try to shrink each valid pcluster */ + if (!erofs_try_to_release_pcluster(sbi, pcl)) + continue; + xa_unlock(&sbi->managed_pslots); - spiltted = 0; - end = PAGE_SIZE; -repeat: - cur = end - 1; - - /* lucky, within the range of the current map_blocks */ - if (offset + cur >= map->m_la && - offset + cur < map->m_la + map->m_llen) { - /* didn't get a valid collection previously (very rare) */ - if (!clt->cl) - goto restart_now; - goto hitted; + ++freed; + if (!--nr) + return freed; + xa_lock(&sbi->managed_pslots); } + xa_unlock(&sbi->managed_pslots); + return freed; +} - /* go ahead the next map_blocks */ - erofs_dbg("%s: [out-of-range] pos %llu", __func__, offset + cur); +static void z_erofs_put_pcluster(struct erofs_sb_info *sbi, + struct z_erofs_pcluster *pcl, bool try_free) +{ + bool free = false; - if (z_erofs_collector_end(clt)) - fe->backmost = false; + if (lockref_put_or_lock(&pcl->lockref)) + return; - map->m_la = offset + cur; - map->m_llen = 0; - err = z_erofs_map_blocks_iter(inode, map, 0); - if (err) - goto err_out; + DBG_BUGON(__lockref_is_dead(&pcl->lockref)); + if (!--pcl->lockref.count) { + if (try_free && xa_trylock(&sbi->managed_pslots)) { + free = __erofs_try_to_release_pcluster(sbi, pcl); + xa_unlock(&sbi->managed_pslots); + } + atomic_long_add(!free, &erofs_global_shrink_cnt); + } + spin_unlock(&pcl->lockref.lock); + if (free) + call_rcu(&pcl->rcu, z_erofs_rcu_callback); +} -restart_now: - if (!(map->m_flags & EROFS_MAP_MAPPED)) - goto hitted; +static void z_erofs_pcluster_end(struct z_erofs_frontend *fe) +{ + struct z_erofs_pcluster *pcl = fe->pcl; - err = z_erofs_collector_begin(clt, inode, map); - if (err) - goto err_out; + if (!pcl) + return; - if (z_erofs_is_inline_pcluster(clt->pcl)) { - void *mp; + z_erofs_bvec_iter_end(&fe->biter); + mutex_unlock(&pcl->lock); - mp = erofs_read_metabuf(&fe->map.buf, inode->i_sb, - erofs_blknr(map->m_pa), EROFS_NO_KMAP); - if (IS_ERR(mp)) { - err = PTR_ERR(mp); - erofs_err(inode->i_sb, - "failed to get inline page, err %d", err); - goto err_out; - } - get_page(fe->map.buf.page); - WRITE_ONCE(clt->pcl->compressed_pages[0], fe->map.buf.page); - clt->mode = COLLECT_PRIMARY_FOLLOWED_NOINPLACE; - } else { - /* preload all compressed pages (can change mode if needed) */ - if (should_alloc_managed_pages(fe, sbi->opt.cache_strategy, - map->m_la)) - cache_strategy = TRYALLOC; - else - cache_strategy = DONTALLOC; + if (fe->candidate_bvpage) + fe->candidate_bvpage = NULL; - preload_compressed_pages(clt, MNGD_MAPPING(sbi), - cache_strategy, pagepool); - } + /* Drop refcount if it doesn't belong to our processing chain */ + if (fe->mode < Z_EROFS_PCLUSTER_FOLLOWED_NOINPLACE) + z_erofs_put_pcluster(EROFS_I_SB(fe->inode), pcl, false); + fe->pcl = NULL; +} -hitted: - /* - * Ensure the current partial page belongs to this submit chain rather - * than other concurrent submit chains or the noio(bypass) chain since - * those chains are handled asynchronously thus the page cannot be used - * for inplace I/O or pagevec (should be processed in strict order.) - */ - tight &= (clt->mode >= COLLECT_PRIMARY_HOOKED && - clt->mode != COLLECT_PRIMARY_FOLLOWED_NOINPLACE); +static int z_erofs_read_fragment(struct super_block *sb, struct folio *folio, + unsigned int cur, unsigned int end, erofs_off_t pos) +{ + struct inode *packed_inode = EROFS_SB(sb)->packed_inode; + struct erofs_buf buf = __EROFS_BUF_INITIALIZER; + unsigned int cnt; + u8 *src; - cur = end - min_t(unsigned int, offset + end - map->m_la, end); - if (!(map->m_flags & EROFS_MAP_MAPPED)) { - zero_user_segment(page, cur, end); - goto next_part; - } + if (!packed_inode) + return -EFSCORRUPTED; - /* let's derive page type */ - page_type = cur ? Z_EROFS_VLE_PAGE_TYPE_HEAD : - (!spiltted ? Z_EROFS_PAGE_TYPE_EXCLUSIVE : - (tight ? Z_EROFS_PAGE_TYPE_EXCLUSIVE : - Z_EROFS_VLE_PAGE_TYPE_TAIL_SHARED)); - - if (cur) - tight &= (clt->mode >= COLLECT_PRIMARY_FOLLOWED); - -retry: - err = z_erofs_attach_page(clt, page, page_type, - clt->mode >= COLLECT_PRIMARY_FOLLOWED); - /* should allocate an additional short-lived page for pagevec */ - if (err == -EAGAIN) { - struct page *const newpage = - alloc_page(GFP_NOFS | __GFP_NOFAIL); - - set_page_private(newpage, Z_EROFS_SHORTLIVED_PAGE); - err = z_erofs_attach_page(clt, newpage, - Z_EROFS_PAGE_TYPE_EXCLUSIVE, true); - if (!err) - goto retry; + buf.mapping = packed_inode->i_mapping; + for (; cur < end; cur += cnt, pos += cnt) { + cnt = min(end - cur, sb->s_blocksize - erofs_blkoff(sb, pos)); + src = erofs_bread(&buf, pos, true); + if (IS_ERR(src)) { + erofs_put_metabuf(&buf); + return PTR_ERR(src); + } + memcpy_to_folio(folio, cur, src, cnt); } + erofs_put_metabuf(&buf); + return 0; +} - if (err) - goto err_out; - - index = page->index - (map->m_la >> PAGE_SHIFT); +static int z_erofs_scan_folio(struct z_erofs_frontend *f, + struct folio *folio, bool ra) +{ + struct inode *const inode = f->inode; + struct erofs_map_blocks *const map = &f->map; + const loff_t offset = folio_pos(folio); + const unsigned int bs = i_blocksize(inode); + unsigned int end = folio_size(folio), split = 0, cur, pgs; + bool tight, excl; + int err = 0; - z_erofs_onlinepage_fixup(page, index, true); + tight = (bs == PAGE_SIZE); + erofs_onlinefolio_init(folio); + do { + if (offset + end - 1 < map->m_la || + offset + end - 1 >= map->m_la + map->m_llen) { + z_erofs_pcluster_end(f); + map->m_la = offset + end - 1; + map->m_llen = 0; + err = z_erofs_map_blocks_iter(inode, map, 0); + if (err) + break; + } - /* bump up the number of spiltted parts of a page */ - ++spiltted; - /* also update nr_pages */ - clt->cl->nr_pages = max_t(pgoff_t, clt->cl->nr_pages, index + 1); -next_part: - /* can be used for verification */ - map->m_llen = offset + cur - map->m_la; + cur = offset > map->m_la ? 0 : map->m_la - offset; + pgs = round_down(cur, PAGE_SIZE); + /* bump split parts first to avoid several separate cases */ + ++split; + + if (!(map->m_flags & EROFS_MAP_MAPPED)) { + folio_zero_segment(folio, cur, end); + tight = false; + } else if (map->m_flags & __EROFS_MAP_FRAGMENT) { + erofs_off_t fpos = offset + cur - map->m_la; + + err = z_erofs_read_fragment(inode->i_sb, folio, cur, + cur + min(map->m_llen - fpos, end - cur), + EROFS_I(inode)->z_fragmentoff + fpos); + if (err) + break; + tight = false; + } else { + if (!f->pcl) { + err = z_erofs_pcluster_begin(f); + if (err) + break; + f->pcl->besteffort |= !ra; + } - end = cur; - if (end > 0) - goto repeat; + pgs = round_down(end - 1, PAGE_SIZE); + /* + * Ensure this partial page belongs to this submit chain + * rather than other concurrent submit chains or + * noio(bypass) chains since those chains are handled + * asynchronously thus it cannot be used for inplace I/O + * or bvpage (should be processed in the strict order.) + */ + tight &= (f->mode >= Z_EROFS_PCLUSTER_FOLLOWED); + excl = false; + if (cur <= pgs) { + excl = (split <= 1) || tight; + cur = pgs; + } -out: - z_erofs_onlinepage_endio(page); + err = z_erofs_attach_page(f, &((struct z_erofs_bvec) { + .page = folio_page(folio, pgs >> PAGE_SHIFT), + .offset = offset + pgs - map->m_la, + .end = end - pgs, }), excl); + if (err) + break; - erofs_dbg("%s, finish page: %pK spiltted: %u map->m_llen %llu", - __func__, page, spiltted, map->m_llen); + erofs_onlinefolio_split(folio); + if (f->pcl->length < offset + end - map->m_la) { + f->pcl->length = offset + end - map->m_la; + f->pcl->pageofs_out = map->m_la & ~PAGE_MASK; + } + if ((map->m_flags & EROFS_MAP_FULL_MAPPED) && + !(map->m_flags & EROFS_MAP_PARTIAL_REF) && + f->pcl->length == map->m_llen) + f->pcl->partial = false; + } + /* shorten the remaining extent to update progress */ + map->m_llen = offset + cur - map->m_la; + map->m_flags &= ~EROFS_MAP_FULL_MAPPED; + if (cur <= pgs) { + split = cur < pgs; + tight = (bs == PAGE_SIZE); + } + } while ((end = cur) > 0); + erofs_onlinefolio_end(folio, err, false); return err; - - /* if some error occurred while processing this page */ -err_out: - SetPageError(page); - goto out; } -static bool z_erofs_get_sync_decompress_policy(struct erofs_sb_info *sbi, +static bool z_erofs_is_sync_decompress(struct erofs_sb_info *sbi, unsigned int readahead_pages) { - /* auto: enable for readpage, disable for readahead */ + /* auto: enable for read_folio, disable for readahead */ if ((sbi->opt.sync_decompress == EROFS_SYNC_DECOMPRESS_AUTO) && !readahead_pages) return true; @@ -810,304 +1110,317 @@ static bool z_erofs_get_sync_decompress_policy(struct erofs_sb_info *sbi, return false; } -static void z_erofs_decompressqueue_work(struct work_struct *work); -static void z_erofs_decompress_kickoff(struct z_erofs_decompressqueue *io, - bool sync, int bios) +static bool z_erofs_page_is_invalidated(struct page *page) { - struct erofs_sb_info *const sbi = EROFS_SB(io->sb); + return !page_folio(page)->mapping && !z_erofs_is_shortlived_page(page); +} - /* wake up the caller thread for sync decompression */ - if (sync) { - unsigned long flags; +struct z_erofs_backend { + struct page *onstack_pages[Z_EROFS_ONSTACK_PAGES]; + struct super_block *sb; + struct z_erofs_pcluster *pcl; + /* pages with the longest decompressed length for deduplication */ + struct page **decompressed_pages; + /* pages to keep the compressed data */ + struct page **compressed_pages; + + struct list_head decompressed_secondary_bvecs; + struct page **pagepool; + unsigned int onstack_used, nr_pages; + /* indicate if temporary copies should be preserved for later use */ + bool keepxcpy; +}; - spin_lock_irqsave(&io->u.wait.lock, flags); - if (!atomic_add_return(bios, &io->pending_bios)) - wake_up_locked(&io->u.wait); - spin_unlock_irqrestore(&io->u.wait.lock, flags); - return; - } +struct z_erofs_bvec_item { + struct z_erofs_bvec bvec; + struct list_head list; +}; - if (atomic_add_return(bios, &io->pending_bios)) - return; - /* Use workqueue and sync decompression for atomic contexts only */ - if (in_atomic() || irqs_disabled()) { - queue_work(z_erofs_workqueue, &io->u.work); - /* enable sync decompression for readahead */ - if (sbi->opt.sync_decompress == EROFS_SYNC_DECOMPRESS_AUTO) - sbi->opt.sync_decompress = EROFS_SYNC_DECOMPRESS_FORCE_ON; - return; +static void z_erofs_do_decompressed_bvec(struct z_erofs_backend *be, + struct z_erofs_bvec *bvec) +{ + int poff = bvec->offset + be->pcl->pageofs_out; + struct z_erofs_bvec_item *item; + struct page **page; + + if (!(poff & ~PAGE_MASK) && (bvec->end == PAGE_SIZE || + bvec->offset + bvec->end == be->pcl->length)) { + DBG_BUGON((poff >> PAGE_SHIFT) >= be->nr_pages); + page = be->decompressed_pages + (poff >> PAGE_SHIFT); + if (!*page) { + *page = bvec->page; + return; + } + } else { + be->keepxcpy = true; } - z_erofs_decompressqueue_work(&io->u.work); -} -static bool z_erofs_page_is_invalidated(struct page *page) -{ - return !page->mapping && !z_erofs_is_shortlived_page(page); + /* (cold path) one pcluster is requested multiple times */ + item = kmalloc(sizeof(*item), GFP_KERNEL | __GFP_NOFAIL); + item->bvec = *bvec; + list_add(&item->list, &be->decompressed_secondary_bvecs); } -static void z_erofs_decompressqueue_endio(struct bio *bio) +static void z_erofs_fill_other_copies(struct z_erofs_backend *be, int err) { - tagptr1_t t = tagptr_init(tagptr1_t, bio->bi_private); - struct z_erofs_decompressqueue *q = tagptr_unfold_ptr(t); - blk_status_t err = bio->bi_status; - struct bio_vec *bvec; - struct bvec_iter_all iter_all; - - bio_for_each_segment_all(bvec, bio, iter_all) { - struct page *page = bvec->bv_page; - - DBG_BUGON(PageUptodate(page)); - DBG_BUGON(z_erofs_page_is_invalidated(page)); - - if (err) - SetPageError(page); - - if (erofs_page_is_managed(EROFS_SB(q->sb), page)) { - if (!err) - SetPageUptodate(page); - unlock_page(page); + unsigned int off0 = be->pcl->pageofs_out; + struct list_head *p, *n; + + list_for_each_safe(p, n, &be->decompressed_secondary_bvecs) { + struct z_erofs_bvec_item *bvi; + unsigned int end, cur; + void *dst, *src; + + bvi = container_of(p, struct z_erofs_bvec_item, list); + cur = bvi->bvec.offset < 0 ? -bvi->bvec.offset : 0; + end = min_t(unsigned int, be->pcl->length - bvi->bvec.offset, + bvi->bvec.end); + dst = kmap_local_page(bvi->bvec.page); + while (cur < end) { + unsigned int pgnr, scur, len; + + pgnr = (bvi->bvec.offset + cur + off0) >> PAGE_SHIFT; + DBG_BUGON(pgnr >= be->nr_pages); + + scur = bvi->bvec.offset + cur - + ((pgnr << PAGE_SHIFT) - off0); + len = min_t(unsigned int, end - cur, PAGE_SIZE - scur); + if (!be->decompressed_pages[pgnr]) { + err = -EFSCORRUPTED; + cur += len; + continue; + } + src = kmap_local_page(be->decompressed_pages[pgnr]); + memcpy(dst + cur, src + scur, len); + kunmap_local(src); + cur += len; } + kunmap_local(dst); + erofs_onlinefolio_end(page_folio(bvi->bvec.page), err, true); + list_del(p); + kfree(bvi); } - z_erofs_decompress_kickoff(q, tagptr_unfold_tags(t), -1); - bio_put(bio); } -static int z_erofs_decompress_pcluster(struct super_block *sb, - struct z_erofs_pcluster *pcl, - struct page **pagepool) +static void z_erofs_parse_out_bvecs(struct z_erofs_backend *be) { - struct erofs_sb_info *const sbi = EROFS_SB(sb); - unsigned int pclusterpages = z_erofs_pclusterpages(pcl); - struct z_erofs_pagevec_ctor ctor; - unsigned int i, inputsize, outputsize, llen, nr_pages; - struct page *pages_onstack[Z_EROFS_VMAP_ONSTACK_PAGES]; - struct page **pages, **compressed_pages, *page; - - enum z_erofs_page_type page_type; - bool overlapped, partial; - struct z_erofs_collection *cl; - int err; - - might_sleep(); - cl = z_erofs_primarycollection(pcl); - DBG_BUGON(!READ_ONCE(cl->nr_pages)); - - mutex_lock(&cl->lock); - nr_pages = cl->nr_pages; + struct z_erofs_pcluster *pcl = be->pcl; + struct z_erofs_bvec_iter biter; + struct page *old_bvpage; + int i; - if (nr_pages <= Z_EROFS_VMAP_ONSTACK_PAGES) { - pages = pages_onstack; - } else if (nr_pages <= Z_EROFS_VMAP_GLOBAL_PAGES && - mutex_trylock(&z_pagemap_global_lock)) { - pages = z_pagemap_global; - } else { - gfp_t gfp_flags = GFP_KERNEL; + z_erofs_bvec_iter_begin(&biter, &pcl->bvset, Z_EROFS_INLINE_BVECS, 0); + for (i = 0; i < pcl->vcnt; ++i) { + struct z_erofs_bvec bvec; - if (nr_pages > Z_EROFS_VMAP_GLOBAL_PAGES) - gfp_flags |= __GFP_NOFAIL; + z_erofs_bvec_dequeue(&biter, &bvec, &old_bvpage); - pages = kvmalloc_array(nr_pages, sizeof(struct page *), - gfp_flags); + if (old_bvpage) + z_erofs_put_shortlivedpage(be->pagepool, old_bvpage); - /* fallback to global pagemap for the lowmem scenario */ - if (!pages) { - mutex_lock(&z_pagemap_global_lock); - pages = z_pagemap_global; - } + DBG_BUGON(z_erofs_page_is_invalidated(bvec.page)); + z_erofs_do_decompressed_bvec(be, &bvec); } - for (i = 0; i < nr_pages; ++i) - pages[i] = NULL; - - err = 0; - z_erofs_pagevec_ctor_init(&ctor, Z_EROFS_NR_INLINE_PAGEVECS, - cl->pagevec, 0); - - for (i = 0; i < cl->vcnt; ++i) { - unsigned int pagenr; + old_bvpage = z_erofs_bvec_iter_end(&biter); + if (old_bvpage) + z_erofs_put_shortlivedpage(be->pagepool, old_bvpage); +} - page = z_erofs_pagevec_dequeue(&ctor, &page_type); +static int z_erofs_parse_in_bvecs(struct z_erofs_backend *be, bool *overlapped) +{ + struct z_erofs_pcluster *pcl = be->pcl; + unsigned int pclusterpages = z_erofs_pclusterpages(pcl); + int i, err = 0; - /* all pages in pagevec ought to be valid */ - DBG_BUGON(!page); - DBG_BUGON(z_erofs_page_is_invalidated(page)); + *overlapped = false; + for (i = 0; i < pclusterpages; ++i) { + struct z_erofs_bvec *bvec = &pcl->compressed_bvecs[i]; + struct page *page = bvec->page; - if (z_erofs_put_shortlivedpage(pagepool, page)) + /* compressed data ought to be valid when decompressing */ + if (IS_ERR(page) || !page) { + bvec->page = NULL; /* clear the failure reason */ + err = page ? PTR_ERR(page) : -EIO; continue; - - if (page_type == Z_EROFS_VLE_PAGE_TYPE_HEAD) - pagenr = 0; - else - pagenr = z_erofs_onlinepage_index(page); - - DBG_BUGON(pagenr >= nr_pages); - - /* - * currently EROFS doesn't support multiref(dedup), - * so here erroring out one multiref page. - */ - if (pages[pagenr]) { - DBG_BUGON(1); - SetPageError(pages[pagenr]); - z_erofs_onlinepage_endio(pages[pagenr]); - err = -EFSCORRUPTED; } - pages[pagenr] = page; - } - z_erofs_pagevec_ctor_exit(&ctor, true); + be->compressed_pages[i] = page; - overlapped = false; - compressed_pages = pcl->compressed_pages; - - for (i = 0; i < pclusterpages; ++i) { - unsigned int pagenr; - - page = compressed_pages[i]; - /* all compressed pages ought to be valid */ - DBG_BUGON(!page); - - if (z_erofs_is_inline_pcluster(pcl)) { + if (pcl->from_meta || + erofs_folio_is_managed(EROFS_SB(be->sb), page_folio(page))) { if (!PageUptodate(page)) err = -EIO; continue; } DBG_BUGON(z_erofs_page_is_invalidated(page)); - if (!z_erofs_is_shortlived_page(page)) { - if (erofs_page_is_managed(sbi, page)) { - if (!PageUptodate(page)) - err = -EIO; - continue; - } - - /* - * only if non-head page can be selected - * for inplace decompression - */ - pagenr = z_erofs_onlinepage_index(page); - - DBG_BUGON(pagenr >= nr_pages); - if (pages[pagenr]) { - DBG_BUGON(1); - SetPageError(pages[pagenr]); - z_erofs_onlinepage_endio(pages[pagenr]); - err = -EFSCORRUPTED; - } - pages[pagenr] = page; - - overlapped = true; - } - - /* PG_error needs checking for all non-managed pages */ - if (PageError(page)) { - DBG_BUGON(PageUptodate(page)); - err = -EIO; - } + if (z_erofs_is_shortlived_page(page)) + continue; + z_erofs_do_decompressed_bvec(be, bvec); + *overlapped = true; } + return err; +} - if (err) - goto out; - - llen = pcl->length >> Z_EROFS_PCLUSTER_LENGTH_BIT; - if (nr_pages << PAGE_SHIFT >= cl->pageofs + llen) { - outputsize = llen; - partial = !(pcl->length & Z_EROFS_PCLUSTER_FULL_LENGTH); - } else { - outputsize = (nr_pages << PAGE_SHIFT) - cl->pageofs; - partial = true; +static int z_erofs_decompress_pcluster(struct z_erofs_backend *be, int err) +{ + struct erofs_sb_info *const sbi = EROFS_SB(be->sb); + struct z_erofs_pcluster *pcl = be->pcl; + unsigned int pclusterpages = z_erofs_pclusterpages(pcl); + const struct z_erofs_decompressor *alg = + z_erofs_decomp[pcl->algorithmformat]; + bool try_free = true; + int i, j, jtop, err2; + struct page *page; + bool overlapped; + const char *reason; + + mutex_lock(&pcl->lock); + be->nr_pages = PAGE_ALIGN(pcl->length + pcl->pageofs_out) >> PAGE_SHIFT; + + /* allocate (de)compressed page arrays if cannot be kept on stack */ + be->decompressed_pages = NULL; + be->compressed_pages = NULL; + be->onstack_used = 0; + if (be->nr_pages <= Z_EROFS_ONSTACK_PAGES) { + be->decompressed_pages = be->onstack_pages; + be->onstack_used = be->nr_pages; + memset(be->decompressed_pages, 0, + sizeof(struct page *) * be->nr_pages); } - if (z_erofs_is_inline_pcluster(pcl)) - inputsize = pcl->tailpacking_size; - else - inputsize = pclusterpages * PAGE_SIZE; - - err = z_erofs_decompress(&(struct z_erofs_decompress_req) { - .sb = sb, - .in = compressed_pages, - .out = pages, + if (pclusterpages + be->onstack_used <= Z_EROFS_ONSTACK_PAGES) + be->compressed_pages = be->onstack_pages + be->onstack_used; + + if (!be->decompressed_pages) + be->decompressed_pages = + kvcalloc(be->nr_pages, sizeof(struct page *), + GFP_KERNEL | __GFP_NOFAIL); + if (!be->compressed_pages) + be->compressed_pages = + kvcalloc(pclusterpages, sizeof(struct page *), + GFP_KERNEL | __GFP_NOFAIL); + + z_erofs_parse_out_bvecs(be); + err2 = z_erofs_parse_in_bvecs(be, &overlapped); + if (err2) + err = err2; + if (!err) { + reason = alg->decompress(&(struct z_erofs_decompress_req) { + .sb = be->sb, + .in = be->compressed_pages, + .out = be->decompressed_pages, + .inpages = pclusterpages, + .outpages = be->nr_pages, .pageofs_in = pcl->pageofs_in, - .pageofs_out = cl->pageofs, - .inputsize = inputsize, - .outputsize = outputsize, + .pageofs_out = pcl->pageofs_out, + .inputsize = pcl->pclustersize, + .outputsize = pcl->length, .alg = pcl->algorithmformat, .inplace_io = overlapped, - .partial_decoding = partial - }, pagepool); + .partial_decoding = pcl->partial, + .fillgaps = be->keepxcpy, + .gfp = pcl->besteffort ? GFP_KERNEL : + GFP_NOWAIT | __GFP_NORETRY + }, be->pagepool); + if (IS_ERR(reason)) { + erofs_err(be->sb, "failed to decompress (%s) %ld @ pa %llu size %u => %u", + alg->name, PTR_ERR(reason), pcl->pos, + pcl->pclustersize, pcl->length); + err = PTR_ERR(reason); + } else if (unlikely(reason)) { + erofs_err(be->sb, "failed to decompress (%s) %s @ pa %llu size %u => %u", + alg->name, reason, pcl->pos, + pcl->pclustersize, pcl->length); + err = -EFSCORRUPTED; + } + } -out: /* must handle all compressed pages before actual file pages */ - if (z_erofs_is_inline_pcluster(pcl)) { - page = compressed_pages[0]; - WRITE_ONCE(compressed_pages[0], NULL); - put_page(page); + if (pcl->from_meta) { + folio_put(page_folio(pcl->compressed_bvecs[0].page)); + WRITE_ONCE(pcl->compressed_bvecs[0].page, NULL); } else { + /* managed folios are still left in compressed_bvecs[] */ for (i = 0; i < pclusterpages; ++i) { - page = compressed_pages[i]; - - if (erofs_page_is_managed(sbi, page)) + page = be->compressed_pages[i]; + if (!page) continue; - - /* recycle all individual short-lived pages */ - (void)z_erofs_put_shortlivedpage(pagepool, page); - WRITE_ONCE(compressed_pages[i], NULL); + if (erofs_folio_is_managed(sbi, page_folio(page))) { + try_free = false; + continue; + } + (void)z_erofs_put_shortlivedpage(be->pagepool, page); + WRITE_ONCE(pcl->compressed_bvecs[i].page, NULL); } } - - for (i = 0; i < nr_pages; ++i) { - page = pages[i]; + if (be->compressed_pages < be->onstack_pages || + be->compressed_pages >= be->onstack_pages + Z_EROFS_ONSTACK_PAGES) + kvfree(be->compressed_pages); + + jtop = 0; + z_erofs_fill_other_copies(be, err); + for (i = 0; i < be->nr_pages; ++i) { + page = be->decompressed_pages[i]; if (!page) continue; DBG_BUGON(z_erofs_page_is_invalidated(page)); - - /* recycle all individual short-lived pages */ - if (z_erofs_put_shortlivedpage(pagepool, page)) + if (!z_erofs_is_shortlived_page(page)) { + erofs_onlinefolio_end(page_folio(page), err, true); continue; - - if (err < 0) - SetPageError(page); - - z_erofs_onlinepage_endio(page); + } + if (pcl->algorithmformat != Z_EROFS_COMPRESSION_LZ4) { + erofs_pagepool_add(be->pagepool, page); + continue; + } + for (j = 0; j < jtop && be->decompressed_pages[j] != page; ++j) + ; + if (j >= jtop) /* this bounce page is newly detected */ + be->decompressed_pages[jtop++] = page; } - - if (pages == z_pagemap_global) - mutex_unlock(&z_pagemap_global_lock); - else if (pages != pages_onstack) - kvfree(pages); - - cl->nr_pages = 0; - cl->vcnt = 0; - - /* all cl locks MUST be taken before the following line */ - WRITE_ONCE(pcl->next, Z_EROFS_PCLUSTER_NIL); - - /* all cl locks SHOULD be released right now */ - mutex_unlock(&cl->lock); - - z_erofs_collection_put(cl); + while (jtop) + erofs_pagepool_add(be->pagepool, + be->decompressed_pages[--jtop]); + if (be->decompressed_pages != be->onstack_pages) + kvfree(be->decompressed_pages); + + pcl->length = 0; + pcl->partial = true; + pcl->besteffort = false; + pcl->bvset.nextpage = NULL; + pcl->vcnt = 0; + + /* pcluster lock MUST be taken before the following line */ + WRITE_ONCE(pcl->next, NULL); + mutex_unlock(&pcl->lock); + + if (pcl->from_meta) + z_erofs_free_pcluster(pcl); + else + z_erofs_put_pcluster(sbi, pcl, try_free); return err; } -static void z_erofs_decompress_queue(const struct z_erofs_decompressqueue *io, - struct page **pagepool) +static int z_erofs_decompress_queue(const struct z_erofs_decompressqueue *io, + struct page **pagepool) { - z_erofs_next_pcluster_t owned = io->head; - - while (owned != Z_EROFS_PCLUSTER_TAIL_CLOSED) { - struct z_erofs_pcluster *pcl; - - /* no possible that 'owned' equals Z_EROFS_WORK_TPTR_TAIL */ - DBG_BUGON(owned == Z_EROFS_PCLUSTER_TAIL); - - /* no possible that 'owned' equals NULL */ - DBG_BUGON(owned == Z_EROFS_PCLUSTER_NIL); - - pcl = container_of(owned, struct z_erofs_pcluster, next); - owned = READ_ONCE(pcl->next); - - z_erofs_decompress_pcluster(io->sb, pcl, pagepool); + struct z_erofs_backend be = { + .sb = io->sb, + .pagepool = pagepool, + .decompressed_secondary_bvecs = + LIST_HEAD_INIT(be.decompressed_secondary_bvecs), + .pcl = io->head, + }; + struct z_erofs_pcluster *next; + int err = io->eio ? -EIO : 0; + + for (; be.pcl != Z_EROFS_PCLUSTER_TAIL; be.pcl = next) { + DBG_BUGON(!be.pcl); + next = READ_ONCE(be.pcl->next); + err = z_erofs_decompress_pcluster(&be, err) ?: err; } + return err; } static void z_erofs_decompressqueue_work(struct work_struct *work) @@ -1116,129 +1429,174 @@ static void z_erofs_decompressqueue_work(struct work_struct *work) container_of(work, struct z_erofs_decompressqueue, u.work); struct page *pagepool = NULL; - DBG_BUGON(bgq->head == Z_EROFS_PCLUSTER_TAIL_CLOSED); + DBG_BUGON(bgq->head == Z_EROFS_PCLUSTER_TAIL); z_erofs_decompress_queue(bgq, &pagepool); - erofs_release_pages(&pagepool); kvfree(bgq); } -static struct page *pickup_page_for_submission(struct z_erofs_pcluster *pcl, - unsigned int nr, - struct page **pagepool, - struct address_space *mc, - gfp_t gfp) +#ifdef CONFIG_EROFS_FS_PCPU_KTHREAD +static void z_erofs_decompressqueue_kthread_work(struct kthread_work *work) { - const pgoff_t index = pcl->obj.index; - bool tocache = false; + z_erofs_decompressqueue_work((struct work_struct *)work); +} +#endif - struct address_space *mapping; - struct page *oldpage, *page; +/* Use (kthread_)work in atomic contexts to minimize scheduling overhead */ +static inline bool z_erofs_in_atomic(void) +{ + if (IS_ENABLED(CONFIG_PREEMPTION) && rcu_preempt_depth()) + return true; + if (!IS_ENABLED(CONFIG_PREEMPT_COUNT)) + return true; + return !preemptible(); +} - compressed_page_t t; - int justfound; +static void z_erofs_decompress_kickoff(struct z_erofs_decompressqueue *io, + int bios) +{ + struct erofs_sb_info *const sbi = EROFS_SB(io->sb); -repeat: - page = READ_ONCE(pcl->compressed_pages[nr]); - oldpage = page; + /* wake up the caller thread for sync decompression */ + if (io->sync) { + if (!atomic_add_return(bios, &io->pending_bios)) + complete(&io->u.done); + return; + } - if (!page) - goto out_allocpage; + if (atomic_add_return(bios, &io->pending_bios)) + return; + if (z_erofs_in_atomic()) { +#ifdef CONFIG_EROFS_FS_PCPU_KTHREAD + struct kthread_worker *worker; + + rcu_read_lock(); + worker = rcu_dereference( + z_erofs_pcpu_workers[raw_smp_processor_id()]); + if (!worker) { + INIT_WORK(&io->u.work, z_erofs_decompressqueue_work); + queue_work(z_erofs_workqueue, &io->u.work); + } else { + kthread_queue_work(worker, &io->u.kthread_work); + } + rcu_read_unlock(); +#else + queue_work(z_erofs_workqueue, &io->u.work); +#endif + /* enable sync decompression for readahead */ + if (sbi->opt.sync_decompress == EROFS_SYNC_DECOMPRESS_AUTO) + sbi->opt.sync_decompress = EROFS_SYNC_DECOMPRESS_FORCE_ON; + return; + } + z_erofs_decompressqueue_work(&io->u.work); +} - /* process the target tagged pointer */ - t = tagptr_init(compressed_page_t, page); - justfound = tagptr_unfold_tags(t); - page = tagptr_unfold_ptr(t); +static void z_erofs_fill_bio_vec(struct bio_vec *bvec, + struct z_erofs_frontend *f, + struct z_erofs_pcluster *pcl, + unsigned int nr, + struct address_space *mc) +{ + gfp_t gfp = mapping_gfp_mask(mc); + bool tocache = false; + struct z_erofs_bvec zbv; + struct address_space *mapping; + struct folio *folio; + struct page *page; + int bs = i_blocksize(f->inode); - /* - * preallocated cached pages, which is used to avoid direct reclaim - * otherwise, it will go inplace I/O path instead. - */ - if (page->private == Z_EROFS_PREALLOCATED_PAGE) { - WRITE_ONCE(pcl->compressed_pages[nr], page); - set_page_private(page, 0); + /* Except for inplace folios, the entire folio can be used for I/Os */ + bvec->bv_offset = 0; + bvec->bv_len = PAGE_SIZE; +repeat: + spin_lock(&pcl->lockref.lock); + zbv = pcl->compressed_bvecs[nr]; + spin_unlock(&pcl->lockref.lock); + if (!zbv.page) + goto out_allocfolio; + + bvec->bv_page = zbv.page; + DBG_BUGON(z_erofs_is_shortlived_page(bvec->bv_page)); + + folio = page_folio(zbv.page); + /* For preallocated managed folios, add them to page cache here */ + if (folio->private == Z_EROFS_PREALLOCATED_FOLIO) { tocache = true; goto out_tocache; } - mapping = READ_ONCE(page->mapping); + mapping = READ_ONCE(folio->mapping); /* - * file-backed online pages in plcuster are all locked steady, - * therefore it is impossible for `mapping' to be NULL. + * File-backed folios for inplace I/Os are all locked steady, + * therefore it is impossible for `mapping` to be NULL. */ - if (mapping && mapping != mc) - /* ought to be unmanaged pages */ - goto out; - - /* directly return for shortlived page as well */ - if (z_erofs_is_shortlived_page(page)) - goto out; - - lock_page(page); - - /* only true if page reclaim goes wrong, should never happen */ - DBG_BUGON(justfound && PagePrivate(page)); - - /* the page is still in manage cache */ - if (page->mapping == mc) { - WRITE_ONCE(pcl->compressed_pages[nr], page); - - ClearPageError(page); - if (!PagePrivate(page)) { - /* - * impossible to be !PagePrivate(page) for - * the current restriction as well if - * the page is already in compressed_pages[]. - */ - DBG_BUGON(!justfound); + if (mapping && mapping != mc) { + if (zbv.offset < 0) + bvec->bv_offset = round_up(-zbv.offset, bs); + bvec->bv_len = round_up(zbv.end, bs) - bvec->bv_offset; + return; + } - justfound = 0; - set_page_private(page, (unsigned long)pcl); - SetPagePrivate(page); + folio_lock(folio); + if (likely(folio->mapping == mc)) { + /* + * The cached folio is still in managed cache but without + * a valid `->private` pcluster hint. Let's reconnect them. + */ + if (!folio_test_private(folio)) { + folio_attach_private(folio, pcl); + /* compressed_bvecs[] already takes a ref before */ + folio_put(folio); } - - /* no need to submit io if it is already up-to-date */ - if (PageUptodate(page)) { - unlock_page(page); - page = NULL; + if (likely(folio->private == pcl)) { + /* don't submit cache I/Os again if already uptodate */ + if (folio_test_uptodate(folio)) { + folio_unlock(folio); + bvec->bv_page = NULL; + } + return; } - goto out; + /* + * Already linked with another pcluster, which only appears in + * crafted images by fuzzers for now. But handle this anyway. + */ + tocache = false; /* use temporary short-lived pages */ + } else { + DBG_BUGON(1); /* referenced managed folios can't be truncated */ + tocache = true; } - - /* - * the managed page has been truncated, it's unsafe to - * reuse this one, let's allocate a new cache-managed page. - */ - DBG_BUGON(page->mapping); - DBG_BUGON(!justfound); - - tocache = true; - unlock_page(page); - put_page(page); -out_allocpage: - page = erofs_allocpage(pagepool, gfp | __GFP_NOFAIL); - if (oldpage != cmpxchg(&pcl->compressed_pages[nr], oldpage, page)) { - erofs_pagepool_add(pagepool, page); + folio_unlock(folio); + folio_put(folio); +out_allocfolio: + page = __erofs_allocpage(&f->pagepool, gfp, true); + spin_lock(&pcl->lockref.lock); + if (unlikely(pcl->compressed_bvecs[nr].page != zbv.page)) { + if (page) + erofs_pagepool_add(&f->pagepool, page); + spin_unlock(&pcl->lockref.lock); cond_resched(); goto repeat; } + pcl->compressed_bvecs[nr].page = page ? page : ERR_PTR(-ENOMEM); + spin_unlock(&pcl->lockref.lock); + bvec->bv_page = page; + if (!page) + return; + folio = page_folio(page); out_tocache: - if (!tocache || add_to_page_cache_lru(page, mc, index + nr, gfp)) { - /* turn into temporary page if fails (1 ref) */ - set_page_private(page, Z_EROFS_SHORTLIVED_PAGE); - goto out; + if (!tocache || bs != PAGE_SIZE || + filemap_add_folio(mc, folio, (pcl->pos >> PAGE_SHIFT) + nr, gfp)) { + /* turn into a temporary shortlived folio (1 ref) */ + folio->private = (void *)Z_EROFS_SHORTLIVED_PAGE; + return; } - attach_page_private(page, pcl); - /* drop a refcount added by allocpage (then we have 2 refs here) */ - put_page(page); - -out: /* the only exit (for tracing and debugging) */ - return page; + folio_attach_private(folio, pcl); + /* drop a refcount added by allocpage (then 2 refs in total here) */ + folio_put(folio); } -static struct z_erofs_decompressqueue * -jobqueue_init(struct super_block *sb, - struct z_erofs_decompressqueue *fgq, bool *fg) +static struct z_erofs_decompressqueue *jobqueue_init(struct super_block *sb, + struct z_erofs_decompressqueue *fgq, bool *fg) { struct z_erofs_decompressqueue *q; @@ -1248,15 +1606,22 @@ jobqueue_init(struct super_block *sb, *fg = true; goto fg_out; } +#ifdef CONFIG_EROFS_FS_PCPU_KTHREAD + kthread_init_work(&q->u.kthread_work, + z_erofs_decompressqueue_kthread_work); +#else INIT_WORK(&q->u.work, z_erofs_decompressqueue_work); +#endif } else { fg_out: q = fgq; - init_waitqueue_head(&fgq->u.wait); + init_completion(&fgq->u.done); atomic_set(&fgq->pending_bios, 0); + q->eio = false; + q->sync = true; } q->sb = sb; - q->head = Z_EROFS_PCLUSTER_TAIL_CLOSED; + q->head = Z_EROFS_PCLUSTER_TAIL; return q; } @@ -1267,316 +1632,312 @@ enum { NR_JOBQUEUES, }; -static void *jobqueueset_init(struct super_block *sb, - struct z_erofs_decompressqueue *q[], - struct z_erofs_decompressqueue *fgq, bool *fg) +static void z_erofs_move_to_bypass_queue(struct z_erofs_pcluster *pcl, + struct z_erofs_pcluster *next, + struct z_erofs_pcluster **qtail[]) { - /* - * if managed cache is enabled, bypass jobqueue is needed, - * no need to read from device for all pclusters in this queue. - */ - q[JQ_BYPASS] = jobqueue_init(sb, fgq + JQ_BYPASS, NULL); - q[JQ_SUBMIT] = jobqueue_init(sb, fgq + JQ_SUBMIT, fg); - - return tagptr_cast_ptr(tagptr_fold(tagptr1_t, q[JQ_SUBMIT], *fg)); + WRITE_ONCE(pcl->next, Z_EROFS_PCLUSTER_TAIL); + WRITE_ONCE(*qtail[JQ_SUBMIT], next); + WRITE_ONCE(*qtail[JQ_BYPASS], pcl); + qtail[JQ_BYPASS] = &pcl->next; } -static void move_to_bypass_jobqueue(struct z_erofs_pcluster *pcl, - z_erofs_next_pcluster_t qtail[], - z_erofs_next_pcluster_t owned_head) +static void z_erofs_endio(struct bio *bio) { - z_erofs_next_pcluster_t *const submit_qtail = qtail[JQ_SUBMIT]; - z_erofs_next_pcluster_t *const bypass_qtail = qtail[JQ_BYPASS]; - - DBG_BUGON(owned_head == Z_EROFS_PCLUSTER_TAIL_CLOSED); - if (owned_head == Z_EROFS_PCLUSTER_TAIL) - owned_head = Z_EROFS_PCLUSTER_TAIL_CLOSED; + struct z_erofs_decompressqueue *q = bio->bi_private; + blk_status_t err = bio->bi_status; + struct folio_iter fi; - WRITE_ONCE(pcl->next, Z_EROFS_PCLUSTER_TAIL_CLOSED); + bio_for_each_folio_all(fi, bio) { + struct folio *folio = fi.folio; - WRITE_ONCE(*submit_qtail, owned_head); - WRITE_ONCE(*bypass_qtail, &pcl->next); + DBG_BUGON(folio_test_uptodate(folio)); + DBG_BUGON(z_erofs_page_is_invalidated(&folio->page)); + if (!erofs_folio_is_managed(EROFS_SB(q->sb), folio)) + continue; - qtail[JQ_BYPASS] = &pcl->next; + if (!err) + folio_mark_uptodate(folio); + folio_unlock(folio); + } + if (err) + q->eio = true; + z_erofs_decompress_kickoff(q, -1); + if (bio->bi_bdev) + bio_put(bio); } -static void z_erofs_submit_queue(struct super_block *sb, - struct z_erofs_decompress_frontend *f, - struct page **pagepool, +static void z_erofs_submit_queue(struct z_erofs_frontend *f, struct z_erofs_decompressqueue *fgq, - bool *force_fg) + bool *force_fg, bool readahead) { - struct erofs_sb_info *const sbi = EROFS_SB(sb); - z_erofs_next_pcluster_t qtail[NR_JOBQUEUES]; + struct super_block *sb = f->inode->i_sb; + struct address_space *mc = MNGD_MAPPING(EROFS_SB(sb)); + struct z_erofs_pcluster **qtail[NR_JOBQUEUES]; struct z_erofs_decompressqueue *q[NR_JOBQUEUES]; - void *bi_private; - z_erofs_next_pcluster_t owned_head = f->clt.owned_head; + struct z_erofs_pcluster *pcl, *next; /* bio is NULL initially, so no need to initialize last_{index,bdev} */ - pgoff_t last_index; - struct block_device *last_bdev; + erofs_off_t last_pa; unsigned int nr_bios = 0; struct bio *bio = NULL; + unsigned long pflags; + int memstall = 0; + + /* No need to read from device for pclusters in the bypass queue. */ + q[JQ_BYPASS] = jobqueue_init(sb, fgq + JQ_BYPASS, NULL); + q[JQ_SUBMIT] = jobqueue_init(sb, fgq + JQ_SUBMIT, force_fg); - bi_private = jobqueueset_init(sb, q, fgq, force_fg); qtail[JQ_BYPASS] = &q[JQ_BYPASS]->head; qtail[JQ_SUBMIT] = &q[JQ_SUBMIT]->head; /* by default, all need io submission */ - q[JQ_SUBMIT]->head = owned_head; + q[JQ_SUBMIT]->head = next = f->head; do { struct erofs_map_dev mdev; - struct z_erofs_pcluster *pcl; - pgoff_t cur, end; + erofs_off_t cur, end; + struct bio_vec bvec; unsigned int i = 0; bool bypass = true; - /* no possible 'owned_head' equals the following */ - DBG_BUGON(owned_head == Z_EROFS_PCLUSTER_TAIL_CLOSED); - DBG_BUGON(owned_head == Z_EROFS_PCLUSTER_NIL); - - pcl = container_of(owned_head, struct z_erofs_pcluster, next); - - /* close the main owned chain at first */ - owned_head = cmpxchg(&pcl->next, Z_EROFS_PCLUSTER_TAIL, - Z_EROFS_PCLUSTER_TAIL_CLOSED); - if (z_erofs_is_inline_pcluster(pcl)) { - move_to_bypass_jobqueue(pcl, qtail, owned_head); + pcl = next; + next = READ_ONCE(pcl->next); + if (pcl->from_meta) { + z_erofs_move_to_bypass_queue(pcl, next, qtail); continue; } /* no device id here, thus it will always succeed */ mdev = (struct erofs_map_dev) { - .m_pa = blknr_to_addr(pcl->obj.index), + .m_pa = round_down(pcl->pos, sb->s_blocksize), }; (void)erofs_map_dev(sb, &mdev); - cur = erofs_blknr(mdev.m_pa); - end = cur + pcl->pclusterpages; - + cur = mdev.m_pa; + end = round_up(cur + pcl->pageofs_in + pcl->pclustersize, + sb->s_blocksize); do { - struct page *page; + bvec.bv_page = NULL; + if (bio && (cur != last_pa || + bio->bi_bdev != mdev.m_bdev)) { +drain_io: + if (erofs_is_fileio_mode(EROFS_SB(sb))) + erofs_fileio_submit_bio(bio); + else if (erofs_is_fscache_mode(sb)) + erofs_fscache_submit_bio(bio); + else + submit_bio(bio); + + if (memstall) { + psi_memstall_leave(&pflags); + memstall = 0; + } + bio = NULL; + } - page = pickup_page_for_submission(pcl, i++, pagepool, - MNGD_MAPPING(sbi), - GFP_NOFS); - if (!page) - continue; + if (!bvec.bv_page) { + z_erofs_fill_bio_vec(&bvec, f, pcl, i++, mc); + if (!bvec.bv_page) + continue; + if (cur + bvec.bv_len > end) + bvec.bv_len = end - cur; + DBG_BUGON(bvec.bv_len < sb->s_blocksize); + } - if (bio && (cur != last_index + 1 || - last_bdev != mdev.m_bdev)) { -submit_bio_retry: - submit_bio(bio); - bio = NULL; + if (unlikely(PageWorkingset(bvec.bv_page)) && + !memstall) { + psi_memstall_enter(&pflags); + memstall = 1; } if (!bio) { - bio = bio_alloc(GFP_NOIO, BIO_MAX_VECS); - bio->bi_end_io = z_erofs_decompressqueue_endio; - - bio_set_dev(bio, mdev.m_bdev); - last_bdev = mdev.m_bdev; - bio->bi_iter.bi_sector = (sector_t)cur << - LOG_SECTORS_PER_BLOCK; - bio->bi_private = bi_private; - bio->bi_opf = REQ_OP_READ; - if (f->readahead) + if (erofs_is_fileio_mode(EROFS_SB(sb))) + bio = erofs_fileio_bio_alloc(&mdev); + else if (erofs_is_fscache_mode(sb)) + bio = erofs_fscache_bio_alloc(&mdev); + else + bio = bio_alloc(mdev.m_bdev, BIO_MAX_VECS, + REQ_OP_READ, GFP_NOIO); + bio->bi_end_io = z_erofs_endio; + bio->bi_iter.bi_sector = + (mdev.m_dif->fsoff + cur) >> 9; + bio->bi_private = q[JQ_SUBMIT]; + if (readahead) bio->bi_opf |= REQ_RAHEAD; ++nr_bios; } - if (bio_add_page(bio, page, PAGE_SIZE, 0) < PAGE_SIZE) - goto submit_bio_retry; - - last_index = cur; + if (!bio_add_page(bio, bvec.bv_page, bvec.bv_len, + bvec.bv_offset)) + goto drain_io; + last_pa = cur + bvec.bv_len; bypass = false; - } while (++cur < end); + } while ((cur += bvec.bv_len) < end); if (!bypass) qtail[JQ_SUBMIT] = &pcl->next; else - move_to_bypass_jobqueue(pcl, qtail, owned_head); - } while (owned_head != Z_EROFS_PCLUSTER_TAIL); - - if (bio) - submit_bio(bio); + z_erofs_move_to_bypass_queue(pcl, next, qtail); + } while (next != Z_EROFS_PCLUSTER_TAIL); + + if (bio) { + if (erofs_is_fileio_mode(EROFS_SB(sb))) + erofs_fileio_submit_bio(bio); + else if (erofs_is_fscache_mode(sb)) + erofs_fscache_submit_bio(bio); + else + submit_bio(bio); + } + if (memstall) + psi_memstall_leave(&pflags); /* * although background is preferred, no one is pending for submission. - * don't issue workqueue for decompression but drop it directly instead. + * don't issue decompression but drop it directly instead. */ if (!*force_fg && !nr_bios) { kvfree(q[JQ_SUBMIT]); return; } - z_erofs_decompress_kickoff(q[JQ_SUBMIT], *force_fg, nr_bios); + z_erofs_decompress_kickoff(q[JQ_SUBMIT], nr_bios); } -static void z_erofs_runqueue(struct super_block *sb, - struct z_erofs_decompress_frontend *f, - struct page **pagepool, bool force_fg) +static int z_erofs_runqueue(struct z_erofs_frontend *f, unsigned int rapages) { struct z_erofs_decompressqueue io[NR_JOBQUEUES]; + struct erofs_sb_info *sbi = EROFS_I_SB(f->inode); + bool force_fg = z_erofs_is_sync_decompress(sbi, rapages); + int err; - if (f->clt.owned_head == Z_EROFS_PCLUSTER_TAIL) - return; - z_erofs_submit_queue(sb, f, pagepool, io, &force_fg); + if (f->head == Z_EROFS_PCLUSTER_TAIL) + return 0; + z_erofs_submit_queue(f, io, &force_fg, !!rapages); /* handle bypass queue (no i/o pclusters) immediately */ - z_erofs_decompress_queue(&io[JQ_BYPASS], pagepool); - + err = z_erofs_decompress_queue(&io[JQ_BYPASS], &f->pagepool); if (!force_fg) - return; + return err; /* wait until all bios are completed */ - io_wait_event(io[JQ_SUBMIT].u.wait, - !atomic_read(&io[JQ_SUBMIT].pending_bios)); + wait_for_completion_io(&io[JQ_SUBMIT].u.done); /* handle synchronous decompress queue in the caller context */ - z_erofs_decompress_queue(&io[JQ_SUBMIT], pagepool); + return z_erofs_decompress_queue(&io[JQ_SUBMIT], &f->pagepool) ?: err; } /* * Since partial uptodate is still unimplemented for now, we have to use * approximate readmore strategies as a start. */ -static void z_erofs_pcluster_readmore(struct z_erofs_decompress_frontend *f, - struct readahead_control *rac, - erofs_off_t end, - struct page **pagepool, - bool backmost) +static void z_erofs_pcluster_readmore(struct z_erofs_frontend *f, + struct readahead_control *rac, bool backmost) { struct inode *inode = f->inode; struct erofs_map_blocks *map = &f->map; - erofs_off_t cur; + erofs_off_t cur, end, headoffset = f->headoffset; int err; if (backmost) { + if (rac) + end = headoffset + readahead_length(rac) - 1; + else + end = headoffset + PAGE_SIZE - 1; map->m_la = end; err = z_erofs_map_blocks_iter(inode, map, EROFS_GET_BLOCKS_READMORE); - if (err) + if (err || !(map->m_flags & EROFS_MAP_ENCODED)) return; - /* expend ra for the trailing edge if readahead */ + /* expand ra for the trailing edge if readahead */ if (rac) { - loff_t newstart = readahead_pos(rac); - cur = round_up(map->m_la + map->m_llen, PAGE_SIZE); - readahead_expand(rac, newstart, cur - newstart); + readahead_expand(rac, headoffset, cur - headoffset); return; } end = round_up(end, PAGE_SIZE); } else { end = round_up(map->m_la, PAGE_SIZE); - - if (!map->m_llen) + if (!(map->m_flags & EROFS_MAP_ENCODED) || !map->m_llen) return; } cur = map->m_la + map->m_llen - 1; - while (cur >= end) { + while ((cur >= end) && (cur < i_size_read(inode))) { pgoff_t index = cur >> PAGE_SHIFT; - struct page *page; - - page = erofs_grab_cache_page_nowait(inode->i_mapping, index); - if (!page) - goto skip; - - if (PageUptodate(page)) { - unlock_page(page); - put_page(page); - goto skip; + struct folio *folio; + + folio = erofs_grab_folio_nowait(inode->i_mapping, index); + if (!IS_ERR_OR_NULL(folio)) { + if (folio_test_uptodate(folio)) + folio_unlock(folio); + else + z_erofs_scan_folio(f, folio, !!rac); + folio_put(folio); } - err = z_erofs_do_read_page(f, page, pagepool); - if (err) - erofs_err(inode->i_sb, - "readmore error at page %lu @ nid %llu", - index, EROFS_I(inode)->nid); - put_page(page); -skip: if (cur < PAGE_SIZE) break; cur = (index << PAGE_SHIFT) - 1; } } -static int z_erofs_readpage(struct file *file, struct page *page) +static int z_erofs_read_folio(struct file *file, struct folio *folio) { - struct inode *const inode = page->mapping->host; - struct erofs_sb_info *const sbi = EROFS_I_SB(inode); - struct z_erofs_decompress_frontend f = DECOMPRESS_FRONTEND_INIT(inode); - struct page *pagepool = NULL; + struct inode *const inode = folio->mapping->host; + Z_EROFS_DEFINE_FRONTEND(f, inode, folio_pos(folio)); int err; - trace_erofs_readpage(page, false); - f.headoffset = (erofs_off_t)page->index << PAGE_SHIFT; - - z_erofs_pcluster_readmore(&f, NULL, f.headoffset + PAGE_SIZE - 1, - &pagepool, true); - err = z_erofs_do_read_page(&f, page, &pagepool); - z_erofs_pcluster_readmore(&f, NULL, 0, &pagepool, false); + trace_erofs_read_folio(folio, false); + z_erofs_pcluster_readmore(&f, NULL, true); + err = z_erofs_scan_folio(&f, folio, false); + z_erofs_pcluster_readmore(&f, NULL, false); + z_erofs_pcluster_end(&f); - (void)z_erofs_collector_end(&f.clt); - - /* if some compressed cluster ready, need submit them anyway */ - z_erofs_runqueue(inode->i_sb, &f, &pagepool, - z_erofs_get_sync_decompress_policy(sbi, 0)); - - if (err) - erofs_err(inode->i_sb, "failed to read, err [%d]", err); + /* if some pclusters are ready, need submit them anyway */ + err = z_erofs_runqueue(&f, 0) ?: err; + if (err && err != -EINTR) + erofs_err(inode->i_sb, "read error %d @ %lu of nid %llu", + err, folio->index, EROFS_I(inode)->nid); erofs_put_metabuf(&f.map.buf); - erofs_release_pages(&pagepool); + erofs_release_pages(&f.pagepool); return err; } static void z_erofs_readahead(struct readahead_control *rac) { struct inode *const inode = rac->mapping->host; - struct erofs_sb_info *const sbi = EROFS_I_SB(inode); - struct z_erofs_decompress_frontend f = DECOMPRESS_FRONTEND_INIT(inode); - struct page *pagepool = NULL, *head = NULL, *page; - unsigned int nr_pages; - - f.readahead = true; - f.headoffset = readahead_pos(rac); - - z_erofs_pcluster_readmore(&f, rac, f.headoffset + - readahead_length(rac) - 1, &pagepool, true); - nr_pages = readahead_count(rac); - trace_erofs_readpages(inode, readahead_index(rac), nr_pages, false); - - while ((page = readahead_page(rac))) { - set_page_private(page, (unsigned long)head); - head = page; + Z_EROFS_DEFINE_FRONTEND(f, inode, readahead_pos(rac)); + unsigned int nrpages = readahead_count(rac); + struct folio *head = NULL, *folio; + int err; + + trace_erofs_readahead(inode, readahead_index(rac), nrpages, false); + z_erofs_pcluster_readmore(&f, rac, true); + while ((folio = readahead_folio(rac))) { + folio->private = head; + head = folio; } + /* traverse in reverse order for best metadata I/O performance */ while (head) { - struct page *page = head; - int err; - - /* traversal in reverse order */ - head = (void *)page_private(page); - - err = z_erofs_do_read_page(&f, page, &pagepool); - if (err) - erofs_err(inode->i_sb, - "readahead error at page %lu @ nid %llu", - page->index, EROFS_I(inode)->nid); - put_page(page); + folio = head; + head = folio_get_private(folio); + + err = z_erofs_scan_folio(&f, folio, true); + if (err && err != -EINTR) + erofs_err(inode->i_sb, "readahead error at folio %lu @ nid %llu", + folio->index, EROFS_I(inode)->nid); } - z_erofs_pcluster_readmore(&f, rac, 0, &pagepool, false); - (void)z_erofs_collector_end(&f.clt); + z_erofs_pcluster_readmore(&f, rac, false); + z_erofs_pcluster_end(&f); - z_erofs_runqueue(inode->i_sb, &f, &pagepool, - z_erofs_get_sync_decompress_policy(sbi, nr_pages)); + (void)z_erofs_runqueue(&f, nrpages); erofs_put_metabuf(&f.map.buf); - erofs_release_pages(&pagepool); + erofs_release_pages(&f.pagepool); } const struct address_space_operations z_erofs_aops = { - .readpage = z_erofs_readpage, + .read_folio = z_erofs_read_folio, .readahead = z_erofs_readahead, }; |