diff options
Diffstat (limited to 'fs')
113 files changed, 2167 insertions, 851 deletions
diff --git a/fs/bcachefs/Kconfig b/fs/bcachefs/Kconfig index 5455412b2b75..df383b97aacc 100644 --- a/fs/bcachefs/Kconfig +++ b/fs/bcachefs/Kconfig @@ -21,7 +21,6 @@ config BCACHEFS_FS select XOR_BLOCKS select XXHASH select SYMBOLIC_ERRNAME - select MIN_HEAP help The bcachefs filesystem - a modern, copy on write filesystem, with support for multiple devices, compression, checksumming, etc. @@ -34,7 +33,6 @@ config BCACHEFS_QUOTA config BCACHEFS_ERASURE_CODING bool "bcachefs erasure coding (RAID5/6) support (EXPERIMENTAL)" depends on BCACHEFS_FS - select QUOTACTL help This enables the "erasure_code" filesysystem and inode option, which organizes data into reed-solomon stripes instead of ordinary @@ -43,11 +41,6 @@ config BCACHEFS_ERASURE_CODING WARNING: this feature is still undergoing on disk format changes, and should only be enabled for testing purposes. -config BCACHEFS_POSIX_ACL - bool "bcachefs POSIX ACL support" - depends on BCACHEFS_FS - select FS_POSIX_ACL - config BCACHEFS_DEBUG bool "bcachefs debugging" depends on BCACHEFS_FS diff --git a/fs/bcachefs/Makefile b/fs/bcachefs/Makefile index 1e87eee962ec..98acb3dc66a3 100644 --- a/fs/bcachefs/Makefile +++ b/fs/bcachefs/Makefile @@ -1,4 +1,9 @@ +ifdef BCACHEFS_DKMS + CONFIG_BCACHEFS_FS := m + # Enable other features here? +endif + obj-$(CONFIG_BCACHEFS_FS) += bcachefs.o bcachefs-y := \ @@ -99,9 +104,16 @@ bcachefs-y := \ util.o \ varint.o \ xattr.o \ - vendor/closure.o + vendor/closure.o \ + vendor/min_heap.o -obj-$(CONFIG_MEAN_AND_VARIANCE_UNIT_TEST) += mean_and_variance_test.o +ifndef BCACHEFS_DKMS + obj-$(CONFIG_MEAN_AND_VARIANCE_UNIT_TEST) += mean_and_variance_test.o +endif # Silence "note: xyz changed in GCC X.X" messages subdir-ccflags-y += $(call cc-disable-warning, psabi) + +ifdef BCACHEFS_DKMS + subdir-ccflags-y += -I$(src) +endif diff --git a/fs/bcachefs/acl.c b/fs/bcachefs/acl.c index 3befa1f36e72..a1df0ec2d511 100644 --- a/fs/bcachefs/acl.c +++ b/fs/bcachefs/acl.c @@ -57,7 +57,7 @@ void bch2_acl_to_text(struct printbuf *out, const void *value, size_t size) } } -#ifdef CONFIG_BCACHEFS_POSIX_ACL +#ifndef NO_BCACHEFS_FS #include "fs.h" @@ -437,4 +437,4 @@ err: return ret; } -#endif /* CONFIG_BCACHEFS_POSIX_ACL */ +#endif /* NO_BCACHEFS_FS */ diff --git a/fs/bcachefs/acl.h b/fs/bcachefs/acl.h index fe730a6bf0c1..01cf21c5a0b6 100644 --- a/fs/bcachefs/acl.h +++ b/fs/bcachefs/acl.h @@ -26,7 +26,7 @@ typedef struct { void bch2_acl_to_text(struct printbuf *, const void *, size_t); -#ifdef CONFIG_BCACHEFS_POSIX_ACL +#ifndef NO_BCACHEFS_FS struct posix_acl *bch2_get_acl(struct inode *, int, bool); @@ -55,6 +55,6 @@ static inline int bch2_acl_chmod(struct btree_trans *trans, subvol_inum inum, return 0; } -#endif /* CONFIG_BCACHEFS_POSIX_ACL */ +#endif /* NO_BCACHEFS_FS */ #endif /* _BCACHEFS_ACL_H */ diff --git a/fs/bcachefs/bcachefs.h b/fs/bcachefs/bcachefs.h index 3ccca855f05e..3673290586e2 100644 --- a/fs/bcachefs/bcachefs.h +++ b/fs/bcachefs/bcachefs.h @@ -686,6 +686,7 @@ struct btree_debug { unsigned id; }; +#define BCH_LINK_MAX U32_MAX #define BCH_TRANSACTIONS_NR 128 struct btree_transaction_stats { diff --git a/fs/bcachefs/btree_gc.c b/fs/bcachefs/btree_gc.c index 63dc0836bf08..638c2a9268a9 100644 --- a/fs/bcachefs/btree_gc.c +++ b/fs/bcachefs/btree_gc.c @@ -204,7 +204,7 @@ static int btree_check_node_boundaries(struct btree_trans *trans, struct btree * if (bpos_eq(expected_start, cur->data->min_key)) return 0; - prt_printf(&buf, " at "); + prt_printf(&buf, " at "); bch2_btree_id_level_to_text(&buf, b->c.btree_id, b->c.level); prt_printf(&buf, ":\nparent: "); bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(&b->key)); @@ -229,8 +229,8 @@ static int btree_check_node_boundaries(struct btree_trans *trans, struct btree * *pulled_from_scan = cur->data->min_key; ret = bch_err_throw(c, topology_repair_did_fill_from_scan); } else { - if (mustfix_fsck_err(trans, btree_node_topology_bad_min_key, - "btree node with incorrect min_key%s", buf.buf)) + if (mustfix_fsck_err(trans, btree_node_topology_gap_between_nodes, + "gap between btree nodes%s", buf.buf)) ret = set_node_min(c, cur, expected_start); } } else { /* overlap */ diff --git a/fs/bcachefs/btree_io.c b/fs/bcachefs/btree_io.c index 52d21259ed6f..3808c41dda84 100644 --- a/fs/bcachefs/btree_io.c +++ b/fs/bcachefs/btree_io.c @@ -1318,6 +1318,7 @@ int bch2_btree_node_read_done(struct bch_fs *c, struct bch_dev *ca, set_btree_bset_end(b, b->set); set_btree_node_need_rewrite(b); set_btree_node_need_rewrite_error(b); + ret = 0; continue; } if (ret) diff --git a/fs/bcachefs/btree_journal_iter.c b/fs/bcachefs/btree_journal_iter.c index a6f344faf751..73fa8513ee61 100644 --- a/fs/bcachefs/btree_journal_iter.c +++ b/fs/bcachefs/btree_journal_iter.c @@ -838,8 +838,10 @@ void bch2_shoot_down_journal_keys(struct bch_fs *c, enum btree_id btree, bpos_ge(k->k.p, start) && bpos_le(k->k.p, end))) keys->data[dst++] = *i; - else if (i->allocated) + else if (i->allocated) { kfree(i->allocated_k); + i->allocated_k = NULL; + } } keys->nr = keys->gap = dst; } diff --git a/fs/bcachefs/btree_node_scan.c b/fs/bcachefs/btree_node_scan.c index c0dff992ad60..433d498540fd 100644 --- a/fs/bcachefs/btree_node_scan.c +++ b/fs/bcachefs/btree_node_scan.c @@ -12,7 +12,6 @@ #include "recovery_passes.h" #include <linux/kthread.h> -#include <linux/min_heap.h> #include <linux/sched/sysctl.h> #include <linux/sort.h> diff --git a/fs/bcachefs/clock_types.h b/fs/bcachefs/clock_types.h index 37554e4514fe..d12c11dc6601 100644 --- a/fs/bcachefs/clock_types.h +++ b/fs/bcachefs/clock_types.h @@ -2,7 +2,7 @@ #ifndef _BCACHEFS_CLOCK_TYPES_H #define _BCACHEFS_CLOCK_TYPES_H -#include "util.h" +#include "vendor/min_heap.h" #define NR_IO_TIMERS (BCH_SB_MEMBERS_MAX * 3) diff --git a/fs/bcachefs/data_update.c b/fs/bcachefs/data_update.c index 62d5d17d681e..6615c868d066 100644 --- a/fs/bcachefs/data_update.c +++ b/fs/bcachefs/data_update.c @@ -55,63 +55,6 @@ static bool bkey_get_dev_refs(struct bch_fs *c, struct bkey_s_c k) return true; } -static void bkey_nocow_unlock(struct bch_fs *c, struct bkey_s_c k) -{ - struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(k); - - bkey_for_each_ptr(ptrs, ptr) { - struct bch_dev *ca = bch2_dev_have_ref(c, ptr->dev); - struct bpos bucket = PTR_BUCKET_POS(ca, ptr); - - bch2_bucket_nocow_unlock(&c->nocow_locks, bucket, 0); - } -} - -static noinline_for_stack -bool __bkey_nocow_lock(struct bch_fs *c, struct moving_context *ctxt, struct bkey_ptrs_c ptrs, - const struct bch_extent_ptr *start) -{ - if (!ctxt) { - bkey_for_each_ptr(ptrs, ptr) { - if (ptr == start) - break; - - struct bch_dev *ca = bch2_dev_have_ref(c, ptr->dev); - struct bpos bucket = PTR_BUCKET_POS(ca, ptr); - bch2_bucket_nocow_unlock(&c->nocow_locks, bucket, 0); - } - return false; - } - - __bkey_for_each_ptr(start, ptrs.end, ptr) { - struct bch_dev *ca = bch2_dev_have_ref(c, ptr->dev); - struct bpos bucket = PTR_BUCKET_POS(ca, ptr); - - bool locked; - move_ctxt_wait_event(ctxt, - (locked = bch2_bucket_nocow_trylock(&c->nocow_locks, bucket, 0)) || - list_empty(&ctxt->ios)); - if (!locked) { - bch2_trans_unlock(ctxt->trans); - bch2_bucket_nocow_lock(&c->nocow_locks, bucket, 0); - } - } - return true; -} - -static bool bkey_nocow_lock(struct bch_fs *c, struct moving_context *ctxt, struct bkey_ptrs_c ptrs) -{ - bkey_for_each_ptr(ptrs, ptr) { - struct bch_dev *ca = bch2_dev_have_ref(c, ptr->dev); - struct bpos bucket = PTR_BUCKET_POS(ca, ptr); - - if (!bch2_bucket_nocow_trylock(&c->nocow_locks, bucket, 0)) - return __bkey_nocow_lock(c, ctxt, ptrs, ptr); - } - - return true; -} - noinline_for_stack static void trace_io_move_finish2(struct data_update *u, struct bkey_i *new, @@ -538,7 +481,7 @@ void bch2_data_update_exit(struct data_update *update) update->bvecs = NULL; if (c->opts.nocow_enabled) - bkey_nocow_unlock(c, k); + bch2_bkey_nocow_unlock(c, k, 0); bkey_put_dev_refs(c, k); bch2_disk_reservation_put(c, &update->op.res); bch2_bkey_buf_exit(&update->k, c); @@ -1018,10 +961,19 @@ int bch2_data_update_init(struct btree_trans *trans, goto out; } - if (c->opts.nocow_enabled && - !bkey_nocow_lock(c, ctxt, ptrs)) { - ret = bch_err_throw(c, nocow_lock_blocked); - goto out_put_dev_refs; + if (c->opts.nocow_enabled) { + if (!bch2_bkey_nocow_trylock(c, ptrs, 0)) { + bool locked = false; + if (ctxt) + move_ctxt_wait_event(ctxt, + (locked = bch2_bkey_nocow_trylock(c, ptrs, 0)) || + list_empty(&ctxt->ios)); + if (!locked) { + if (ctxt) + bch2_trans_unlock(ctxt->trans); + bch2_bkey_nocow_lock(c, ptrs, 0); + } + } } if (unwritten) { @@ -1039,8 +991,7 @@ int bch2_data_update_init(struct btree_trans *trans, return 0; out_nocow_unlock: if (c->opts.nocow_enabled) - bkey_nocow_unlock(c, k); -out_put_dev_refs: + bch2_bkey_nocow_unlock(c, k, 0); bkey_put_dev_refs(c, k); out: bch2_disk_reservation_put(c, &m->op.res); diff --git a/fs/bcachefs/disk_accounting.c b/fs/bcachefs/disk_accounting.c index a99f821c6a1c..9fd32ca15d19 100644 --- a/fs/bcachefs/disk_accounting.c +++ b/fs/bcachefs/disk_accounting.c @@ -814,6 +814,8 @@ int bch2_accounting_read(struct bch_fs *c) struct journal_keys *keys = &c->journal_keys; struct journal_key *jk = keys->data; + move_gap(keys, keys->nr); + while (jk < &darray_top(*keys) && __journal_key_cmp(c, BTREE_ID_accounting, 0, POS_MIN, jk) > 0) jk++; diff --git a/fs/bcachefs/errcode.h b/fs/bcachefs/errcode.h index cbf1eedddad7..2712a9754100 100644 --- a/fs/bcachefs/errcode.h +++ b/fs/bcachefs/errcode.h @@ -228,6 +228,7 @@ x(BCH_ERR_topology_repair, topology_repair_drop_this_node) \ x(BCH_ERR_topology_repair, topology_repair_drop_prev_node) \ x(BCH_ERR_topology_repair, topology_repair_did_fill_from_scan) \ + x(EMLINK, too_many_links) \ x(EOPNOTSUPP, may_not_use_incompat_feature) \ x(EOPNOTSUPP, no_casefolding_without_utf8) \ x(EOPNOTSUPP, casefolding_disabled) \ @@ -367,7 +368,10 @@ x(BCH_ERR_nopromote, nopromote_enomem) \ x(0, invalid_snapshot_node) \ x(0, option_needs_open_fs) \ - x(0, remove_disk_accounting_entry) + x(0, remove_disk_accounting_entry) \ + x(0, nocow_trylock_fail) \ + x(BCH_ERR_nocow_trylock_fail, nocow_trylock_contended) \ + x(BCH_ERR_nocow_trylock_fail, nocow_trylock_bucket_full) \ enum bch_errcode { BCH_ERR_START = 2048, diff --git a/fs/bcachefs/fs.c b/fs/bcachefs/fs.c index f1849eb8327d..414adf86efdc 100644 --- a/fs/bcachefs/fs.c +++ b/fs/bcachefs/fs.c @@ -541,11 +541,9 @@ __bch2_create(struct mnt_idmap *idmap, * preallocate acls + vfs inode before btree transaction, so that * nothing can fail after the transaction succeeds: */ -#ifdef CONFIG_BCACHEFS_POSIX_ACL ret = posix_acl_create(&dir->v, &mode, &default_acl, &acl); if (ret) return ERR_PTR(ret); -#endif inode = __bch2_new_inode(c, GFP_NOFS); if (unlikely(!inode)) { @@ -1751,10 +1749,8 @@ static const struct inode_operations bch_file_inode_operations = { .setattr = bch2_setattr, .fiemap = bch2_fiemap, .listxattr = bch2_xattr_list, -#ifdef CONFIG_BCACHEFS_POSIX_ACL .get_inode_acl = bch2_get_acl, .set_acl = bch2_set_acl, -#endif .fileattr_get = bch2_fileattr_get, .fileattr_set = bch2_fileattr_set, }; @@ -1773,10 +1769,8 @@ static const struct inode_operations bch_dir_inode_operations = { .setattr = bch2_setattr, .tmpfile = bch2_tmpfile, .listxattr = bch2_xattr_list, -#ifdef CONFIG_BCACHEFS_POSIX_ACL .get_inode_acl = bch2_get_acl, .set_acl = bch2_set_acl, -#endif .fileattr_get = bch2_fileattr_get, .fileattr_set = bch2_fileattr_set, }; @@ -1797,10 +1791,8 @@ static const struct inode_operations bch_symlink_inode_operations = { .getattr = bch2_getattr, .setattr = bch2_setattr, .listxattr = bch2_xattr_list, -#ifdef CONFIG_BCACHEFS_POSIX_ACL .get_inode_acl = bch2_get_acl, .set_acl = bch2_set_acl, -#endif .fileattr_get = bch2_fileattr_get, .fileattr_set = bch2_fileattr_set, }; @@ -1809,10 +1801,8 @@ static const struct inode_operations bch_special_inode_operations = { .getattr = bch2_getattr, .setattr = bch2_setattr, .listxattr = bch2_xattr_list, -#ifdef CONFIG_BCACHEFS_POSIX_ACL .get_inode_acl = bch2_get_acl, .set_acl = bch2_set_acl, -#endif .fileattr_get = bch2_fileattr_get, .fileattr_set = bch2_fileattr_set, }; @@ -2546,10 +2536,8 @@ got_sb: c->dev = sb->s_dev; -#ifdef CONFIG_BCACHEFS_POSIX_ACL if (c->opts.acl) sb->s_flags |= SB_POSIXACL; -#endif sb->s_shrink->seeks = 0; diff --git a/fs/bcachefs/inode.c b/fs/bcachefs/inode.c index 543627fb58be..fda4ca783848 100644 --- a/fs/bcachefs/inode.c +++ b/fs/bcachefs/inode.c @@ -1184,8 +1184,8 @@ int bch2_inode_nlink_inc(struct bch_inode_unpacked *bi) if (bi->bi_flags & BCH_INODE_unlinked) bi->bi_flags &= ~BCH_INODE_unlinked; else { - if (bi->bi_nlink == U32_MAX) - return -EINVAL; + if (bi->bi_nlink == BCH_LINK_MAX - nlink_bias(bi->bi_mode)) + return -BCH_ERR_too_many_links; bi->bi_nlink++; } diff --git a/fs/bcachefs/io_write.c b/fs/bcachefs/io_write.c index 6a5da02ce266..ed4a9440143a 100644 --- a/fs/bcachefs/io_write.c +++ b/fs/bcachefs/io_write.c @@ -33,7 +33,6 @@ #include <linux/blkdev.h> #include <linux/moduleparam.h> -#include <linux/prefetch.h> #include <linux/random.h> #include <linux/sched/mm.h> @@ -1323,11 +1322,25 @@ static CLOSURE_CALLBACK(bch2_nocow_write_done) bch2_write_done(cl); } -struct bucket_to_lock { - struct bpos b; - unsigned gen; - struct nocow_lock_bucket *l; -}; +static bool bkey_get_dev_iorefs(struct bch_fs *c, struct bkey_ptrs_c ptrs) +{ + bkey_for_each_ptr(ptrs, ptr) { + struct bch_dev *ca = bch2_dev_get_ioref(c, ptr->dev, WRITE, + BCH_DEV_WRITE_REF_io_write); + if (unlikely(!ca)) { + bkey_for_each_ptr(ptrs, ptr2) { + if (ptr2 == ptr) + break; + enumerated_ref_put(&bch2_dev_have_ref(c, ptr2->dev)->io_ref[WRITE], + BCH_DEV_WRITE_REF_io_write); + } + + return false; + } + } + + return true; +} static void bch2_nocow_write(struct bch_write_op *op) { @@ -1335,15 +1348,14 @@ static void bch2_nocow_write(struct bch_write_op *op) struct btree_trans *trans; struct btree_iter iter; struct bkey_s_c k; - DARRAY_PREALLOCATED(struct bucket_to_lock, 3) buckets; + struct bkey_ptrs_c ptrs; u32 snapshot; - struct bucket_to_lock *stale_at; + const struct bch_extent_ptr *stale_at; int stale, ret; if (op->flags & BCH_WRITE_move) return; - darray_init(&buckets); trans = bch2_trans_get(c); retry: bch2_trans_begin(trans); @@ -1358,8 +1370,6 @@ retry: while (1) { struct bio *bio = &op->wbio.bio; - buckets.nr = 0; - ret = bch2_trans_relock(trans); if (ret) break; @@ -1381,50 +1391,42 @@ retry: break; /* Get iorefs before dropping btree locks: */ - struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(k); - bkey_for_each_ptr(ptrs, ptr) { - struct bch_dev *ca = bch2_dev_get_ioref(c, ptr->dev, WRITE, - BCH_DEV_WRITE_REF_io_write); - if (unlikely(!ca)) - goto err_get_ioref; - - struct bpos b = PTR_BUCKET_POS(ca, ptr); - struct nocow_lock_bucket *l = - bucket_nocow_lock(&c->nocow_locks, bucket_to_u64(b)); - prefetch(l); - - /* XXX allocating memory with btree locks held - rare */ - darray_push_gfp(&buckets, ((struct bucket_to_lock) { - .b = b, .gen = ptr->gen, .l = l, - }), GFP_KERNEL|__GFP_NOFAIL); - - if (ptr->unwritten) - op->flags |= BCH_WRITE_convert_unwritten; - } + ptrs = bch2_bkey_ptrs_c(k); + if (!bkey_get_dev_iorefs(c, ptrs)) + goto out; /* Unlock before taking nocow locks, doing IO: */ bkey_reassemble(op->insert_keys.top, k); - bch2_trans_unlock(trans); + k = bkey_i_to_s_c(op->insert_keys.top); + ptrs = bch2_bkey_ptrs_c(k); - bch2_cut_front(op->pos, op->insert_keys.top); - if (op->flags & BCH_WRITE_convert_unwritten) - bch2_cut_back(POS(op->pos.inode, op->pos.offset + bio_sectors(bio)), op->insert_keys.top); + bch2_trans_unlock(trans); - darray_for_each(buckets, i) { - struct bch_dev *ca = bch2_dev_have_ref(c, i->b.inode); + bch2_bkey_nocow_lock(c, ptrs, BUCKET_NOCOW_LOCK_UPDATE); - __bch2_bucket_nocow_lock(&c->nocow_locks, i->l, - bucket_to_u64(i->b), - BUCKET_NOCOW_LOCK_UPDATE); + /* + * This could be handled better: If we're able to trylock the + * nocow locks with btree locks held we know dirty pointers + * can't be stale + */ + bkey_for_each_ptr(ptrs, ptr) { + struct bch_dev *ca = bch2_dev_have_ref(c, ptr->dev); - int gen = bucket_gen_get(ca, i->b.offset); - stale = gen < 0 ? gen : gen_after(gen, i->gen); + int gen = bucket_gen_get(ca, PTR_BUCKET_NR(ca, ptr)); + stale = gen < 0 ? gen : gen_after(gen, ptr->gen); if (unlikely(stale)) { - stale_at = i; + stale_at = ptr; goto err_bucket_stale; } + + if (ptr->unwritten) + op->flags |= BCH_WRITE_convert_unwritten; } + bch2_cut_front(op->pos, op->insert_keys.top); + if (op->flags & BCH_WRITE_convert_unwritten) + bch2_cut_back(POS(op->pos.inode, op->pos.offset + bio_sectors(bio)), op->insert_keys.top); + bio = &op->wbio.bio; if (k.k->p.offset < op->pos.offset + bio_sectors(bio)) { bio = bio_split(bio, k.k->p.offset - op->pos.offset, @@ -1458,8 +1460,6 @@ err: goto retry; bch2_trans_put(trans); - darray_exit(&buckets); - if (ret) { bch2_write_op_error(op, op->pos.offset, "%s(): btree lookup error: %s", __func__, bch2_err_str(ret)); @@ -1484,24 +1484,11 @@ err: continue_at(&op->cl, bch2_nocow_write_done, index_update_wq(op)); } return; -err_get_ioref: - darray_for_each(buckets, i) - enumerated_ref_put(&bch2_dev_have_ref(c, i->b.inode)->io_ref[WRITE], - BCH_DEV_WRITE_REF_io_write); - - /* Fall back to COW path: */ - goto out; err_bucket_stale: - darray_for_each(buckets, i) { - bch2_bucket_nocow_unlock(&c->nocow_locks, i->b, BUCKET_NOCOW_LOCK_UPDATE); - if (i == stale_at) - break; - } - CLASS(printbuf, buf)(); if (bch2_fs_inconsistent_on(stale < 0, c, - "pointer to invalid bucket in nocow path on device %llu\n %s", - stale_at->b.inode, + "pointer to invalid bucket in nocow path on device %u\n %s", + stale_at->dev, (bch2_bkey_val_to_text(&buf, c, k), buf.buf))) { ret = bch_err_throw(c, data_write_invalid_ptr); } else { @@ -1509,7 +1496,13 @@ err_bucket_stale: ret = bch_err_throw(c, transaction_restart); } - goto err_get_ioref; + bch2_bkey_nocow_unlock(c, k, BUCKET_NOCOW_LOCK_UPDATE); + bkey_for_each_ptr(ptrs, ptr) + enumerated_ref_put(&bch2_dev_have_ref(c, ptr->dev)->io_ref[WRITE], + BCH_DEV_WRITE_REF_io_write); + + /* Fall back to COW path: */ + goto out; } static void __bch2_write(struct bch_write_op *op) diff --git a/fs/bcachefs/journal_reclaim.c b/fs/bcachefs/journal_reclaim.c index ae747c87fcf9..f7b0fdd99c75 100644 --- a/fs/bcachefs/journal_reclaim.c +++ b/fs/bcachefs/journal_reclaim.c @@ -766,6 +766,9 @@ static int bch2_journal_reclaim_thread(void *arg) set_freezable(); + kthread_wait_freezable(test_bit(BCH_FS_rw, &c->flags) || + kthread_should_stop()); + j->last_flushed = jiffies; while (!ret && !kthread_should_stop()) { @@ -826,8 +829,10 @@ int bch2_journal_reclaim_start(struct journal *j) struct task_struct *p; int ret; - if (j->reclaim_thread) + if (j->reclaim_thread) { + wake_up_process(j->reclaim_thread); return 0; + } p = kthread_create(bch2_journal_reclaim_thread, j, "bch-reclaim/%s", c->name); diff --git a/fs/bcachefs/nocow_locking.c b/fs/bcachefs/nocow_locking.c index 71b17f18e90c..c8907070796f 100644 --- a/fs/bcachefs/nocow_locking.c +++ b/fs/bcachefs/nocow_locking.c @@ -6,6 +6,16 @@ #include "nocow_locking.h" #include "util.h" +#include <linux/prefetch.h> + +static bool nocow_bucket_empty(struct nocow_lock_bucket *l) +{ + for (unsigned i = 0; i < ARRAY_SIZE(l->b); i++) + if (atomic_read(&l->l[i])) + return false; + return true; +} + bool bch2_bucket_nocow_is_locked(struct bucket_nocow_lock_table *t, struct bpos bucket) { u64 dev_bucket = bucket_to_u64(bucket); @@ -20,14 +30,12 @@ bool bch2_bucket_nocow_is_locked(struct bucket_nocow_lock_table *t, struct bpos #define sign(v) (v < 0 ? -1 : v > 0 ? 1 : 0) -void bch2_bucket_nocow_unlock(struct bucket_nocow_lock_table *t, struct bpos bucket, int flags) +void __bch2_bucket_nocow_unlock(struct bucket_nocow_lock_table *t, u64 dev_bucket, int flags) { - u64 dev_bucket = bucket_to_u64(bucket); struct nocow_lock_bucket *l = bucket_nocow_lock(t, dev_bucket); int lock_val = flags ? 1 : -1; - unsigned i; - for (i = 0; i < ARRAY_SIZE(l->b); i++) + for (unsigned i = 0; i < ARRAY_SIZE(l->b); i++) if (l->b[i] == dev_bucket) { int v = atomic_sub_return(lock_val, &l->l[i]); @@ -40,8 +48,8 @@ void bch2_bucket_nocow_unlock(struct bucket_nocow_lock_table *t, struct bpos buc BUG(); } -bool __bch2_bucket_nocow_trylock(struct nocow_lock_bucket *l, - u64 dev_bucket, int flags) +static int __bch2_bucket_nocow_trylock(struct bch_fs *c, struct nocow_lock_bucket *l, + u64 dev_bucket, int flags) { int v, lock_val = flags ? 1 : -1; unsigned i; @@ -58,32 +66,128 @@ bool __bch2_bucket_nocow_trylock(struct nocow_lock_bucket *l, goto take_lock; } - return false; + return bch_err_throw(c, nocow_trylock_bucket_full); got_entry: v = atomic_read(&l->l[i]); if (lock_val > 0 ? v < 0 : v > 0) - return false; + return bch_err_throw(c, nocow_trylock_contended); take_lock: v = atomic_read(&l->l[i]); /* Overflow? */ if (v && sign(v + lock_val) != sign(v)) - return false; + return bch_err_throw(c, nocow_trylock_contended); atomic_add(lock_val, &l->l[i]); + return 0; +} + +static inline bool bch2_bucket_nocow_trylock(struct bch_fs *c, struct bpos bucket, int flags) +{ + struct bucket_nocow_lock_table *t = &c->nocow_locks; + u64 dev_bucket = bucket_to_u64(bucket); + struct nocow_lock_bucket *l = bucket_nocow_lock(t, dev_bucket); + + return !__bch2_bucket_nocow_trylock(c, l, dev_bucket, flags); +} + +void bch2_bkey_nocow_unlock(struct bch_fs *c, struct bkey_s_c k, int flags) +{ + struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(k); + + bkey_for_each_ptr(ptrs, ptr) { + struct bch_dev *ca = bch2_dev_have_ref(c, ptr->dev); + struct bpos bucket = PTR_BUCKET_POS(ca, ptr); + + bch2_bucket_nocow_unlock(&c->nocow_locks, bucket, flags); + } +} + +bool bch2_bkey_nocow_trylock(struct bch_fs *c, struct bkey_ptrs_c ptrs, int flags) +{ + bkey_for_each_ptr(ptrs, ptr) { + struct bch_dev *ca = bch2_dev_have_ref(c, ptr->dev); + struct bpos bucket = PTR_BUCKET_POS(ca, ptr); + + if (unlikely(!bch2_bucket_nocow_trylock(c, bucket, flags))) { + bkey_for_each_ptr(ptrs, ptr2) { + if (ptr2 == ptr) + break; + + struct bch_dev *ca = bch2_dev_have_ref(c, ptr2->dev); + struct bpos bucket = PTR_BUCKET_POS(ca, ptr2); + bch2_bucket_nocow_unlock(&c->nocow_locks, bucket, flags); + } + return false; + } + } + return true; } -void __bch2_bucket_nocow_lock(struct bucket_nocow_lock_table *t, - struct nocow_lock_bucket *l, - u64 dev_bucket, int flags) +struct bucket_to_lock { + u64 b; + struct nocow_lock_bucket *l; +}; + +static inline int bucket_to_lock_cmp(struct bucket_to_lock l, + struct bucket_to_lock r) { - if (!__bch2_bucket_nocow_trylock(l, dev_bucket, flags)) { - struct bch_fs *c = container_of(t, struct bch_fs, nocow_locks); + return cmp_int(l.l, r.l); +} + +void bch2_bkey_nocow_lock(struct bch_fs *c, struct bkey_ptrs_c ptrs, int flags) +{ + if (bch2_bkey_nocow_trylock(c, ptrs, flags)) + return; + + DARRAY_PREALLOCATED(struct bucket_to_lock, 3) buckets; + darray_init(&buckets); + + bkey_for_each_ptr(ptrs, ptr) { + struct bch_dev *ca = bch2_dev_have_ref(c, ptr->dev); + u64 b = bucket_to_u64(PTR_BUCKET_POS(ca, ptr)); + struct nocow_lock_bucket *l = + bucket_nocow_lock(&c->nocow_locks, b); + prefetch(l); + + /* XXX allocating memory with btree locks held - rare */ + darray_push_gfp(&buckets, ((struct bucket_to_lock) { .b = b, .l = l, }), + GFP_KERNEL|__GFP_NOFAIL); + } + + WARN_ON_ONCE(buckets.nr > NOCOW_LOCK_BUCKET_SIZE); + + bubble_sort(buckets.data, buckets.nr, bucket_to_lock_cmp); +retake_all: + darray_for_each(buckets, i) { + int ret = __bch2_bucket_nocow_trylock(c, i->l, i->b, flags); + if (!ret) + continue; + u64 start_time = local_clock(); - __closure_wait_event(&l->wait, __bch2_bucket_nocow_trylock(l, dev_bucket, flags)); + if (ret == -BCH_ERR_nocow_trylock_contended) + __closure_wait_event(&i->l->wait, + (ret = __bch2_bucket_nocow_trylock(c, i->l, i->b, flags)) != -BCH_ERR_nocow_trylock_contended); + if (!ret) { + bch2_time_stats_update(&c->times[BCH_TIME_nocow_lock_contended], start_time); + continue; + } + + BUG_ON(ret != -BCH_ERR_nocow_trylock_bucket_full); + + darray_for_each(buckets, i2) { + if (i2 == i) + break; + __bch2_bucket_nocow_unlock(&c->nocow_locks, i2->b, flags); + } + + __closure_wait_event(&i->l->wait, nocow_bucket_empty(i->l)); bch2_time_stats_update(&c->times[BCH_TIME_nocow_lock_contended], start_time); + goto retake_all; } + + darray_exit(&buckets); } void bch2_nocow_locks_to_text(struct printbuf *out, struct bucket_nocow_lock_table *t) diff --git a/fs/bcachefs/nocow_locking.h b/fs/bcachefs/nocow_locking.h index 48b8a003c0d2..972c914780f9 100644 --- a/fs/bcachefs/nocow_locking.h +++ b/fs/bcachefs/nocow_locking.h @@ -19,29 +19,19 @@ static inline struct nocow_lock_bucket *bucket_nocow_lock(struct bucket_nocow_lo #define BUCKET_NOCOW_LOCK_UPDATE (1 << 0) bool bch2_bucket_nocow_is_locked(struct bucket_nocow_lock_table *, struct bpos); -void bch2_bucket_nocow_unlock(struct bucket_nocow_lock_table *, struct bpos, int); -bool __bch2_bucket_nocow_trylock(struct nocow_lock_bucket *, u64, int); -void __bch2_bucket_nocow_lock(struct bucket_nocow_lock_table *, - struct nocow_lock_bucket *, u64, int); -static inline void bch2_bucket_nocow_lock(struct bucket_nocow_lock_table *t, - struct bpos bucket, int flags) -{ - u64 dev_bucket = bucket_to_u64(bucket); - struct nocow_lock_bucket *l = bucket_nocow_lock(t, dev_bucket); - - __bch2_bucket_nocow_lock(t, l, dev_bucket, flags); -} +void __bch2_bucket_nocow_unlock(struct bucket_nocow_lock_table *, u64, int); -static inline bool bch2_bucket_nocow_trylock(struct bucket_nocow_lock_table *t, - struct bpos bucket, int flags) +static inline void bch2_bucket_nocow_unlock(struct bucket_nocow_lock_table *t, struct bpos bucket, + int flags) { - u64 dev_bucket = bucket_to_u64(bucket); - struct nocow_lock_bucket *l = bucket_nocow_lock(t, dev_bucket); - - return __bch2_bucket_nocow_trylock(l, dev_bucket, flags); + __bch2_bucket_nocow_unlock(t, bucket_to_u64(bucket), flags); } +void bch2_bkey_nocow_unlock(struct bch_fs *, struct bkey_s_c, int); +bool bch2_bkey_nocow_trylock(struct bch_fs *, struct bkey_ptrs_c, int); +void bch2_bkey_nocow_lock(struct bch_fs *, struct bkey_ptrs_c, int); + void bch2_nocow_locks_to_text(struct printbuf *, struct bucket_nocow_lock_table *); void bch2_fs_nocow_locking_exit(struct bch_fs *); diff --git a/fs/bcachefs/nocow_locking_types.h b/fs/bcachefs/nocow_locking_types.h index bd12bf677924..3fed8e957e20 100644 --- a/fs/bcachefs/nocow_locking_types.h +++ b/fs/bcachefs/nocow_locking_types.h @@ -5,11 +5,13 @@ #define BUCKET_NOCOW_LOCKS_BITS 10 #define BUCKET_NOCOW_LOCKS (1U << BUCKET_NOCOW_LOCKS_BITS) +#define NOCOW_LOCK_BUCKET_SIZE 6 + struct nocow_lock_bucket { struct closure_waitlist wait; spinlock_t lock; - u64 b[4]; - atomic_t l[4]; + u64 b[NOCOW_LOCK_BUCKET_SIZE]; + atomic_t l[NOCOW_LOCK_BUCKET_SIZE]; } __aligned(SMP_CACHE_BYTES); struct bucket_nocow_lock_table { diff --git a/fs/bcachefs/opts.c b/fs/bcachefs/opts.c index bd5faafc9aa7..b9bac9fbbafb 100644 --- a/fs/bcachefs/opts.c +++ b/fs/bcachefs/opts.c @@ -649,10 +649,6 @@ int bch2_parse_one_mount_opt(struct bch_fs *c, struct bch_opts *opts, if (!(bch2_opt_table[id].flags & OPT_MOUNT)) return -BCH_ERR_option_name; - if (id == Opt_acl && - !IS_ENABLED(CONFIG_BCACHEFS_POSIX_ACL)) - return -BCH_ERR_option_name; - if ((id == Opt_usrquota || id == Opt_grpquota) && !IS_ENABLED(CONFIG_BCACHEFS_QUOTA)) diff --git a/fs/bcachefs/replicas_format.h b/fs/bcachefs/replicas_format.h index b7eff904acdb..898caf943b34 100644 --- a/fs/bcachefs/replicas_format.h +++ b/fs/bcachefs/replicas_format.h @@ -17,7 +17,8 @@ struct bch_replicas_entry_v1 { __u8 data_type; __u8 nr_devs; __u8 nr_required; - __u8 devs[] __counted_by(nr_devs); + /* No counted_by: bch_replicas_cpu entries are all the size of the biggest entry */ + __u8 devs[]; } __packed; struct bch_sb_field_replicas { diff --git a/fs/bcachefs/sb-errors_format.h b/fs/bcachefs/sb-errors_format.h index 77e3fc92e39b..fbdb7b7df23e 100644 --- a/fs/bcachefs/sb-errors_format.h +++ b/fs/bcachefs/sb-errors_format.h @@ -74,6 +74,7 @@ enum bch_fsck_flags { x(btree_root_bad_min_key, 60, 0) \ x(btree_root_bad_max_key, 61, 0) \ x(btree_node_read_error, 62, FSCK_AUTOFIX) \ + x(btree_node_topology_gap_between_nodes, 328, FSCK_AUTOFIX) \ x(btree_node_topology_bad_min_key, 63, FSCK_AUTOFIX) \ x(btree_node_topology_bad_max_key, 64, FSCK_AUTOFIX) \ x(btree_node_topology_bad_root_min_key, 323, FSCK_AUTOFIX) \ @@ -339,7 +340,7 @@ enum bch_fsck_flags { x(dirent_stray_data_after_cf_name, 305, 0) \ x(rebalance_work_incorrectly_set, 309, FSCK_AUTOFIX) \ x(rebalance_work_incorrectly_unset, 310, FSCK_AUTOFIX) \ - x(MAX, 328, 0) + x(MAX, 329, 0) enum bch_sb_error_id { #define x(t, n, ...) BCH_FSCK_ERR_##t = n, diff --git a/fs/bcachefs/super.c b/fs/bcachefs/super.c index 5cd308a68035..3038b5e6c1f0 100644 --- a/fs/bcachefs/super.c +++ b/fs/bcachefs/super.c @@ -564,15 +564,17 @@ static int __bch2_fs_read_write(struct bch_fs *c, bool early) * successfully marked the filesystem dirty */ - ret = bch2_journal_reclaim_start(&c->journal); - if (ret) - goto err; - set_bit(BCH_FS_rw, &c->flags); set_bit(BCH_FS_was_rw, &c->flags); enumerated_ref_start(&c->writes); + ret = bch2_journal_reclaim_start(&c->journal); + if (ret) { + bch_err_msg(c, ret, "error starting journal reclaim thread"); + goto err; + } + ret = bch2_copygc_start(c); if (ret) { bch_err_msg(c, ret, "error starting copygc thread"); @@ -852,7 +854,8 @@ int bch2_fs_init_rw(struct bch_fs *c) bch2_fs_btree_write_buffer_init(c) ?: bch2_fs_fs_io_buffered_init(c) ?: bch2_fs_io_write_init(c) ?: - bch2_fs_journal_init(&c->journal); + bch2_fs_journal_init(&c->journal) ?: + bch2_journal_reclaim_start(&c->journal); if (ret) return ret; diff --git a/fs/bcachefs/util.h b/fs/bcachefs/util.h index 555e0d8f3cf0..eb4222f0f34b 100644 --- a/fs/bcachefs/util.h +++ b/fs/bcachefs/util.h @@ -7,7 +7,6 @@ #include <linux/errno.h> #include <linux/freezer.h> #include <linux/kernel.h> -#include <linux/min_heap.h> #include <linux/sched/clock.h> #include <linux/llist.h> #include <linux/log2.h> diff --git a/fs/bcachefs/vendor/min_heap.c b/fs/bcachefs/vendor/min_heap.c new file mode 100644 index 000000000000..21f744549d66 --- /dev/null +++ b/fs/bcachefs/vendor/min_heap.c @@ -0,0 +1,59 @@ +// SPDX-License-Identifier: GPL-2.0 +#include "min_heap.h" + +void __bch2_min_heap_init(min_heap_char *heap, void *data, size_t size) +{ + __min_heap_init_inline(heap, data, size); +} + +void *__bch2_min_heap_peek(struct min_heap_char *heap) +{ + return __min_heap_peek_inline(heap); +} + +bool __bch2_min_heap_full(min_heap_char *heap) +{ + return __min_heap_full_inline(heap); +} + +void __bch2_min_heap_sift_down(min_heap_char *heap, size_t pos, size_t elem_size, + const struct min_heap_callbacks *func, void *args) +{ + __min_heap_sift_down_inline(heap, pos, elem_size, func, args); +} + +void __bch2_min_heap_sift_up(min_heap_char *heap, size_t elem_size, size_t idx, + const struct min_heap_callbacks *func, void *args) +{ + __min_heap_sift_up_inline(heap, elem_size, idx, func, args); +} + +void __bch2_min_heapify_all(min_heap_char *heap, size_t elem_size, + const struct min_heap_callbacks *func, void *args) +{ + __min_heapify_all_inline(heap, elem_size, func, args); +} + +bool __bch2_min_heap_pop(min_heap_char *heap, size_t elem_size, + const struct min_heap_callbacks *func, void *args) +{ + return __min_heap_pop_inline(heap, elem_size, func, args); +} + +void __bch2_min_heap_pop_push(min_heap_char *heap, const void *element, size_t elem_size, + const struct min_heap_callbacks *func, void *args) +{ + __min_heap_pop_push_inline(heap, element, elem_size, func, args); +} + +bool __bch2_min_heap_push(min_heap_char *heap, const void *element, size_t elem_size, + const struct min_heap_callbacks *func, void *args) +{ + return __min_heap_push_inline(heap, element, elem_size, func, args); +} + +bool __bch2_min_heap_del(min_heap_char *heap, size_t elem_size, size_t idx, + const struct min_heap_callbacks *func, void *args) +{ + return __min_heap_del_inline(heap, elem_size, idx, func, args); +} diff --git a/fs/bcachefs/vendor/min_heap.h b/fs/bcachefs/vendor/min_heap.h new file mode 100644 index 000000000000..865c8b33b11f --- /dev/null +++ b/fs/bcachefs/vendor/min_heap.h @@ -0,0 +1,477 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef _LINUX_MIN_HEAP_H +#define _LINUX_MIN_HEAP_H + +#include <linux/bug.h> +#include <linux/string.h> +#include <linux/types.h> + +/* + * The Min Heap API provides utilities for managing min-heaps, a binary tree + * structure where each node's value is less than or equal to its children's + * values, ensuring the smallest element is at the root. + * + * Users should avoid directly calling functions prefixed with __min_heap_*(). + * Instead, use the provided macro wrappers. + * + * For further details and examples, refer to Documentation/core-api/min_heap.rst. + */ + +/** + * Data structure to hold a min-heap. + * @nr: Number of elements currently in the heap. + * @size: Maximum number of elements that can be held in current storage. + * @data: Pointer to the start of array holding the heap elements. + * @preallocated: Start of the static preallocated array holding the heap elements. + */ +#define MIN_HEAP_PREALLOCATED(_type, _name, _nr) \ +struct _name { \ + size_t nr; \ + size_t size; \ + _type *data; \ + _type preallocated[_nr]; \ +} + +#define DEFINE_MIN_HEAP(_type, _name) MIN_HEAP_PREALLOCATED(_type, _name, 0) + +typedef DEFINE_MIN_HEAP(char, min_heap_char) min_heap_char; + +#define __minheap_cast(_heap) (typeof((_heap)->data[0]) *) +#define __minheap_obj_size(_heap) sizeof((_heap)->data[0]) + +/** + * struct min_heap_callbacks - Data/functions to customise the min_heap. + * @less: Partial order function for this heap. + * @swp: Swap elements function. + */ +struct min_heap_callbacks { + bool (*less)(const void *lhs, const void *rhs, void *args); + void (*swp)(void *lhs, void *rhs, void *args); +}; + +/** + * is_aligned - is this pointer & size okay for word-wide copying? + * @base: pointer to data + * @size: size of each element + * @align: required alignment (typically 4 or 8) + * + * Returns true if elements can be copied using word loads and stores. + * The size must be a multiple of the alignment, and the base address must + * be if we do not have CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS. + * + * For some reason, gcc doesn't know to optimize "if (a & mask || b & mask)" + * to "if ((a | b) & mask)", so we do that by hand. + */ +__attribute_const__ __always_inline +static bool is_aligned(const void *base, size_t size, unsigned char align) +{ + unsigned char lsbits = (unsigned char)size; + + (void)base; +#ifndef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS + lsbits |= (unsigned char)(uintptr_t)base; +#endif + return (lsbits & (align - 1)) == 0; +} + +/** + * swap_words_32 - swap two elements in 32-bit chunks + * @a: pointer to the first element to swap + * @b: pointer to the second element to swap + * @n: element size (must be a multiple of 4) + * + * Exchange the two objects in memory. This exploits base+index addressing, + * which basically all CPUs have, to minimize loop overhead computations. + * + * For some reason, on x86 gcc 7.3.0 adds a redundant test of n at the + * bottom of the loop, even though the zero flag is still valid from the + * subtract (since the intervening mov instructions don't alter the flags). + * Gcc 8.1.0 doesn't have that problem. + */ +static __always_inline +void swap_words_32(void *a, void *b, size_t n) +{ + do { + u32 t = *(u32 *)(a + (n -= 4)); + *(u32 *)(a + n) = *(u32 *)(b + n); + *(u32 *)(b + n) = t; + } while (n); +} + +/** + * swap_words_64 - swap two elements in 64-bit chunks + * @a: pointer to the first element to swap + * @b: pointer to the second element to swap + * @n: element size (must be a multiple of 8) + * + * Exchange the two objects in memory. This exploits base+index + * addressing, which basically all CPUs have, to minimize loop overhead + * computations. + * + * We'd like to use 64-bit loads if possible. If they're not, emulating + * one requires base+index+4 addressing which x86 has but most other + * processors do not. If CONFIG_64BIT, we definitely have 64-bit loads, + * but it's possible to have 64-bit loads without 64-bit pointers (e.g. + * x32 ABI). Are there any cases the kernel needs to worry about? + */ +static __always_inline +void swap_words_64(void *a, void *b, size_t n) +{ + do { +#ifdef CONFIG_64BIT + u64 t = *(u64 *)(a + (n -= 8)); + *(u64 *)(a + n) = *(u64 *)(b + n); + *(u64 *)(b + n) = t; +#else + /* Use two 32-bit transfers to avoid base+index+4 addressing */ + u32 t = *(u32 *)(a + (n -= 4)); + *(u32 *)(a + n) = *(u32 *)(b + n); + *(u32 *)(b + n) = t; + + t = *(u32 *)(a + (n -= 4)); + *(u32 *)(a + n) = *(u32 *)(b + n); + *(u32 *)(b + n) = t; +#endif + } while (n); +} + +/** + * swap_bytes - swap two elements a byte at a time + * @a: pointer to the first element to swap + * @b: pointer to the second element to swap + * @n: element size + * + * This is the fallback if alignment doesn't allow using larger chunks. + */ +static __always_inline +void swap_bytes(void *a, void *b, size_t n) +{ + do { + char t = ((char *)a)[--n]; + ((char *)a)[n] = ((char *)b)[n]; + ((char *)b)[n] = t; + } while (n); +} + +/* + * The values are arbitrary as long as they can't be confused with + * a pointer, but small integers make for the smallest compare + * instructions. + */ +#define SWAP_WORDS_64 ((void (*)(void *, void *, void *))0) +#define SWAP_WORDS_32 ((void (*)(void *, void *, void *))1) +#define SWAP_BYTES ((void (*)(void *, void *, void *))2) + +/* + * Selects the appropriate swap function based on the element size. + */ +static __always_inline +void *select_swap_func(const void *base, size_t size) +{ + if (is_aligned(base, size, 8)) + return SWAP_WORDS_64; + else if (is_aligned(base, size, 4)) + return SWAP_WORDS_32; + else + return SWAP_BYTES; +} + +static __always_inline +void do_swap(void *a, void *b, size_t size, void (*swap_func)(void *lhs, void *rhs, void *args), + void *priv) +{ + if (swap_func == SWAP_WORDS_64) + swap_words_64(a, b, size); + else if (swap_func == SWAP_WORDS_32) + swap_words_32(a, b, size); + else if (swap_func == SWAP_BYTES) + swap_bytes(a, b, size); + else + swap_func(a, b, priv); +} + +/** + * parent - given the offset of the child, find the offset of the parent. + * @i: the offset of the heap element whose parent is sought. Non-zero. + * @lsbit: a precomputed 1-bit mask, equal to "size & -size" + * @size: size of each element + * + * In terms of array indexes, the parent of element j = @i/@size is simply + * (j-1)/2. But when working in byte offsets, we can't use implicit + * truncation of integer divides. + * + * Fortunately, we only need one bit of the quotient, not the full divide. + * @size has a least significant bit. That bit will be clear if @i is + * an even multiple of @size, and set if it's an odd multiple. + * + * Logically, we're doing "if (i & lsbit) i -= size;", but since the + * branch is unpredictable, it's done with a bit of clever branch-free + * code instead. + */ +__attribute_const__ __always_inline +static size_t parent(size_t i, unsigned int lsbit, size_t size) +{ + i -= size; + i -= size & -(i & lsbit); + return i / 2; +} + +/* Initialize a min-heap. */ +static __always_inline +void __min_heap_init_inline(min_heap_char *heap, void *data, size_t size) +{ + heap->nr = 0; + heap->size = size; + if (data) + heap->data = data; + else + heap->data = heap->preallocated; +} + +#define min_heap_init_inline(_heap, _data, _size) \ + __min_heap_init_inline(container_of(&(_heap)->nr, min_heap_char, nr), _data, _size) + +/* Get the minimum element from the heap. */ +static __always_inline +void *__min_heap_peek_inline(struct min_heap_char *heap) +{ + return heap->nr ? heap->data : NULL; +} + +#define min_heap_peek_inline(_heap) \ + (__minheap_cast(_heap) \ + __min_heap_peek_inline(container_of(&(_heap)->nr, min_heap_char, nr))) + +/* Check if the heap is full. */ +static __always_inline +bool __min_heap_full_inline(min_heap_char *heap) +{ + return heap->nr == heap->size; +} + +#define min_heap_full_inline(_heap) \ + __min_heap_full_inline(container_of(&(_heap)->nr, min_heap_char, nr)) + +/* Sift the element at pos down the heap. */ +static __always_inline +void __min_heap_sift_down_inline(min_heap_char *heap, size_t pos, size_t elem_size, + const struct min_heap_callbacks *func, void *args) +{ + const unsigned long lsbit = elem_size & -elem_size; + void *data = heap->data; + void (*swp)(void *lhs, void *rhs, void *args) = func->swp; + /* pre-scale counters for performance */ + size_t a = pos * elem_size; + size_t b, c, d; + size_t n = heap->nr * elem_size; + + if (!swp) + swp = select_swap_func(data, elem_size); + + /* Find the sift-down path all the way to the leaves. */ + for (b = a; c = 2 * b + elem_size, (d = c + elem_size) < n;) + b = func->less(data + c, data + d, args) ? c : d; + + /* Special case for the last leaf with no sibling. */ + if (d == n) + b = c; + + /* Backtrack to the correct location. */ + while (b != a && func->less(data + a, data + b, args)) + b = parent(b, lsbit, elem_size); + + /* Shift the element into its correct place. */ + c = b; + while (b != a) { + b = parent(b, lsbit, elem_size); + do_swap(data + b, data + c, elem_size, swp, args); + } +} + +#define min_heap_sift_down_inline(_heap, _pos, _func, _args) \ + __min_heap_sift_down_inline(container_of(&(_heap)->nr, min_heap_char, nr), _pos, \ + __minheap_obj_size(_heap), _func, _args) + +/* Sift up ith element from the heap, O(log2(nr)). */ +static __always_inline +void __min_heap_sift_up_inline(min_heap_char *heap, size_t elem_size, size_t idx, + const struct min_heap_callbacks *func, void *args) +{ + const unsigned long lsbit = elem_size & -elem_size; + void *data = heap->data; + void (*swp)(void *lhs, void *rhs, void *args) = func->swp; + /* pre-scale counters for performance */ + size_t a = idx * elem_size, b; + + if (!swp) + swp = select_swap_func(data, elem_size); + + while (a) { + b = parent(a, lsbit, elem_size); + if (func->less(data + b, data + a, args)) + break; + do_swap(data + a, data + b, elem_size, swp, args); + a = b; + } +} + +#define min_heap_sift_up_inline(_heap, _idx, _func, _args) \ + __min_heap_sift_up_inline(container_of(&(_heap)->nr, min_heap_char, nr), \ + __minheap_obj_size(_heap), _idx, _func, _args) + +/* Floyd's approach to heapification that is O(nr). */ +static __always_inline +void __min_heapify_all_inline(min_heap_char *heap, size_t elem_size, + const struct min_heap_callbacks *func, void *args) +{ + ssize_t i; + + for (i = heap->nr / 2 - 1; i >= 0; i--) + __min_heap_sift_down_inline(heap, i, elem_size, func, args); +} + +#define min_heapify_all_inline(_heap, _func, _args) \ + __min_heapify_all_inline(container_of(&(_heap)->nr, min_heap_char, nr), \ + __minheap_obj_size(_heap), _func, _args) + +/* Remove minimum element from the heap, O(log2(nr)). */ +static __always_inline +bool __min_heap_pop_inline(min_heap_char *heap, size_t elem_size, + const struct min_heap_callbacks *func, void *args) +{ + void *data = heap->data; + + if (WARN_ONCE(heap->nr <= 0, "Popping an empty heap")) + return false; + + /* Place last element at the root (position 0) and then sift down. */ + heap->nr--; + memcpy(data, data + (heap->nr * elem_size), elem_size); + __min_heap_sift_down_inline(heap, 0, elem_size, func, args); + + return true; +} + +#define min_heap_pop_inline(_heap, _func, _args) \ + __min_heap_pop_inline(container_of(&(_heap)->nr, min_heap_char, nr), \ + __minheap_obj_size(_heap), _func, _args) + +/* + * Remove the minimum element and then push the given element. The + * implementation performs 1 sift (O(log2(nr))) and is therefore more + * efficient than a pop followed by a push that does 2. + */ +static __always_inline +void __min_heap_pop_push_inline(min_heap_char *heap, const void *element, size_t elem_size, + const struct min_heap_callbacks *func, void *args) +{ + memcpy(heap->data, element, elem_size); + __min_heap_sift_down_inline(heap, 0, elem_size, func, args); +} + +#define min_heap_pop_push_inline(_heap, _element, _func, _args) \ + __min_heap_pop_push_inline(container_of(&(_heap)->nr, min_heap_char, nr), _element, \ + __minheap_obj_size(_heap), _func, _args) + +/* Push an element on to the heap, O(log2(nr)). */ +static __always_inline +bool __min_heap_push_inline(min_heap_char *heap, const void *element, size_t elem_size, + const struct min_heap_callbacks *func, void *args) +{ + void *data = heap->data; + size_t pos; + + if (WARN_ONCE(heap->nr >= heap->size, "Pushing on a full heap")) + return false; + + /* Place at the end of data. */ + pos = heap->nr; + memcpy(data + (pos * elem_size), element, elem_size); + heap->nr++; + + /* Sift child at pos up. */ + __min_heap_sift_up_inline(heap, elem_size, pos, func, args); + + return true; +} + +#define min_heap_push_inline(_heap, _element, _func, _args) \ + __min_heap_push_inline(container_of(&(_heap)->nr, min_heap_char, nr), _element, \ + __minheap_obj_size(_heap), _func, _args) + +/* Remove ith element from the heap, O(log2(nr)). */ +static __always_inline +bool __min_heap_del_inline(min_heap_char *heap, size_t elem_size, size_t idx, + const struct min_heap_callbacks *func, void *args) +{ + void *data = heap->data; + void (*swp)(void *lhs, void *rhs, void *args) = func->swp; + + if (WARN_ONCE(heap->nr <= 0, "Popping an empty heap")) + return false; + + if (!swp) + swp = select_swap_func(data, elem_size); + + /* Place last element at the root (position 0) and then sift down. */ + heap->nr--; + if (idx == heap->nr) + return true; + do_swap(data + (idx * elem_size), data + (heap->nr * elem_size), elem_size, swp, args); + __min_heap_sift_up_inline(heap, elem_size, idx, func, args); + __min_heap_sift_down_inline(heap, idx, elem_size, func, args); + + return true; +} + +#define min_heap_del_inline(_heap, _idx, _func, _args) \ + __min_heap_del_inline(container_of(&(_heap)->nr, min_heap_char, nr), \ + __minheap_obj_size(_heap), _idx, _func, _args) + +void __bch2_min_heap_init(min_heap_char *heap, void *data, size_t size); +void *__bch2_min_heap_peek(struct min_heap_char *heap); +bool __bch2_min_heap_full(min_heap_char *heap); +void __bch2_min_heap_sift_down(min_heap_char *heap, size_t pos, size_t elem_size, + const struct min_heap_callbacks *func, void *args); +void __bch2_min_heap_sift_up(min_heap_char *heap, size_t elem_size, size_t idx, + const struct min_heap_callbacks *func, void *args); +void __bch2_min_heapify_all(min_heap_char *heap, size_t elem_size, + const struct min_heap_callbacks *func, void *args); +bool __bch2_min_heap_pop(min_heap_char *heap, size_t elem_size, + const struct min_heap_callbacks *func, void *args); +void __bch2_min_heap_pop_push(min_heap_char *heap, const void *element, size_t elem_size, + const struct min_heap_callbacks *func, void *args); +bool __bch2_min_heap_push(min_heap_char *heap, const void *element, size_t elem_size, + const struct min_heap_callbacks *func, void *args); +bool __bch2_min_heap_del(min_heap_char *heap, size_t elem_size, size_t idx, + const struct min_heap_callbacks *func, void *args); + +#define min_heap_init(_heap, _data, _size) \ + __bch2_min_heap_init(container_of(&(_heap)->nr, min_heap_char, nr), _data, _size) +#define min_heap_peek(_heap) \ + (__minheap_cast(_heap) __bch2_min_heap_peek(container_of(&(_heap)->nr, min_heap_char, nr))) +#define min_heap_full(_heap) \ + __bch2_min_heap_full(container_of(&(_heap)->nr, min_heap_char, nr)) +#define min_heap_sift_down(_heap, _pos, _func, _args) \ + __bch2_min_heap_sift_down(container_of(&(_heap)->nr, min_heap_char, nr), _pos, \ + __minheap_obj_size(_heap), _func, _args) +#define min_heap_sift_up(_heap, _idx, _func, _args) \ + __bch2_min_heap_sift_up(container_of(&(_heap)->nr, min_heap_char, nr), \ + __minheap_obj_size(_heap), _idx, _func, _args) +#define min_heapify_all(_heap, _func, _args) \ + __bch2_min_heapify_all(container_of(&(_heap)->nr, min_heap_char, nr), \ + __minheap_obj_size(_heap), _func, _args) +#define min_heap_pop(_heap, _func, _args) \ + __bch2_min_heap_pop(container_of(&(_heap)->nr, min_heap_char, nr), \ + __minheap_obj_size(_heap), _func, _args) +#define min_heap_pop_push(_heap, _element, _func, _args) \ + __bch2_min_heap_pop_push(container_of(&(_heap)->nr, min_heap_char, nr), _element, \ + __minheap_obj_size(_heap), _func, _args) +#define min_heap_push(_heap, _element, _func, _args) \ + __bch2_min_heap_push(container_of(&(_heap)->nr, min_heap_char, nr), _element, \ + __minheap_obj_size(_heap), _func, _args) +#define min_heap_del(_heap, _idx, _func, _args) \ + __bch2_min_heap_del(container_of(&(_heap)->nr, min_heap_char, nr), \ + __minheap_obj_size(_heap), _idx, _func, _args) + +#endif /* _LINUX_MIN_HEAP_H */ diff --git a/fs/btrfs/block-group.c b/fs/btrfs/block-group.c index 9bf282d2453c..499a9edf0ca3 100644 --- a/fs/btrfs/block-group.c +++ b/fs/btrfs/block-group.c @@ -1795,7 +1795,14 @@ static int reclaim_bgs_cmp(void *unused, const struct list_head *a, bg1 = list_entry(a, struct btrfs_block_group, bg_list); bg2 = list_entry(b, struct btrfs_block_group, bg_list); - return bg1->used > bg2->used; + /* + * Some other task may be updating the ->used field concurrently, but it + * is not serious if we get a stale value or load/store tearing issues, + * as sorting the list of block groups to reclaim is not critical and an + * occasional imperfect order is ok. So silence KCSAN and avoid the + * overhead of locking or any other synchronization. + */ + return data_race(bg1->used > bg2->used); } static inline bool btrfs_should_reclaim(const struct btrfs_fs_info *fs_info) diff --git a/fs/btrfs/btrfs_inode.h b/fs/btrfs/btrfs_inode.h index b99fb0273292..0387b9f43a52 100644 --- a/fs/btrfs/btrfs_inode.h +++ b/fs/btrfs/btrfs_inode.h @@ -248,7 +248,7 @@ struct btrfs_inode { u64 new_delalloc_bytes; /* * The offset of the last dir index key that was logged. - * This is used only for directories. + * This is used only for directories. Protected by 'log_mutex'. */ u64 last_dir_index_offset; }; diff --git a/fs/btrfs/compression.c b/fs/btrfs/compression.c index d09d622016ef..35e3071cec06 100644 --- a/fs/btrfs/compression.c +++ b/fs/btrfs/compression.c @@ -1616,25 +1616,29 @@ out: } /* - * Convert the compression suffix (eg. after "zlib" starting with ":") to - * level, unrecognized string will set the default level. Negative level - * numbers are allowed. + * Convert the compression suffix (eg. after "zlib" starting with ":") to level. + * + * If the resulting level exceeds the algo's supported levels, it will be clamped. + * + * Return <0 if no valid string can be found. + * Return 0 if everything is fine. */ -int btrfs_compress_str2level(unsigned int type, const char *str) +int btrfs_compress_str2level(unsigned int type, const char *str, int *level_ret) { int level = 0; int ret; - if (!type) + if (!type) { + *level_ret = btrfs_compress_set_level(type, level); return 0; + } if (str[0] == ':') { ret = kstrtoint(str + 1, 10, &level); if (ret) - level = 0; + return ret; } - level = btrfs_compress_set_level(type, level); - - return level; + *level_ret = btrfs_compress_set_level(type, level); + return 0; } diff --git a/fs/btrfs/compression.h b/fs/btrfs/compression.h index 1b38e707bbd9..7b41b2b5ff44 100644 --- a/fs/btrfs/compression.h +++ b/fs/btrfs/compression.h @@ -102,7 +102,7 @@ void btrfs_submit_compressed_write(struct btrfs_ordered_extent *ordered, bool writeback); void btrfs_submit_compressed_read(struct btrfs_bio *bbio); -int btrfs_compress_str2level(unsigned int type, const char *str); +int btrfs_compress_str2level(unsigned int type, const char *str, int *level_ret); struct folio *btrfs_alloc_compr_folio(void); void btrfs_free_compr_folio(struct folio *folio); diff --git a/fs/btrfs/delayed-inode.c b/fs/btrfs/delayed-inode.c index 0f8d8e275143..c0c1ddd46b67 100644 --- a/fs/btrfs/delayed-inode.c +++ b/fs/btrfs/delayed-inode.c @@ -1843,7 +1843,6 @@ static void fill_stack_inode_item(struct btrfs_trans_handle *trans, int btrfs_fill_inode(struct btrfs_inode *inode, u32 *rdev) { - struct btrfs_fs_info *fs_info = inode->root->fs_info; struct btrfs_delayed_node *delayed_node; struct btrfs_inode_item *inode_item; struct inode *vfs_inode = &inode->vfs_inode; @@ -1864,8 +1863,6 @@ int btrfs_fill_inode(struct btrfs_inode *inode, u32 *rdev) i_uid_write(vfs_inode, btrfs_stack_inode_uid(inode_item)); i_gid_write(vfs_inode, btrfs_stack_inode_gid(inode_item)); btrfs_i_size_write(inode, btrfs_stack_inode_size(inode_item)); - btrfs_inode_set_file_extent_range(inode, 0, - round_up(i_size_read(vfs_inode), fs_info->sectorsize)); vfs_inode->i_mode = btrfs_stack_inode_mode(inode_item); set_nlink(vfs_inode, btrfs_stack_inode_nlink(inode_item)); inode_set_bytes(vfs_inode, btrfs_stack_inode_nbytes(inode_item)); diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index c953297aa89a..b21cb72835cc 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -111,6 +111,24 @@ struct btrfs_bio_ctrl { */ unsigned long submit_bitmap; struct readahead_control *ractl; + + /* + * The start offset of the last used extent map by a read operation. + * + * This is for proper compressed read merge. + * U64_MAX means we are starting the read and have made no progress yet. + * + * The current btrfs_bio_is_contig() only uses disk_bytenr as + * the condition to check if the read can be merged with previous + * bio, which is not correct. E.g. two file extents pointing to the + * same extent but with different offset. + * + * So here we need to do extra checks to only merge reads that are + * covered by the same extent map. + * Just extent_map::start will be enough, as they are unique + * inside the same inode. + */ + u64 last_em_start; }; static void submit_one_bio(struct btrfs_bio_ctrl *bio_ctrl) @@ -909,7 +927,7 @@ static void btrfs_readahead_expand(struct readahead_control *ractl, * return 0 on success, otherwise return error */ static int btrfs_do_readpage(struct folio *folio, struct extent_map **em_cached, - struct btrfs_bio_ctrl *bio_ctrl, u64 *prev_em_start) + struct btrfs_bio_ctrl *bio_ctrl) { struct inode *inode = folio->mapping->host; struct btrfs_fs_info *fs_info = inode_to_fs_info(inode); @@ -1019,12 +1037,11 @@ static int btrfs_do_readpage(struct folio *folio, struct extent_map **em_cached, * non-optimal behavior (submitting 2 bios for the same extent). */ if (compress_type != BTRFS_COMPRESS_NONE && - prev_em_start && *prev_em_start != (u64)-1 && - *prev_em_start != em->start) + bio_ctrl->last_em_start != U64_MAX && + bio_ctrl->last_em_start != em->start) force_bio_submit = true; - if (prev_em_start) - *prev_em_start = em->start; + bio_ctrl->last_em_start = em->start; btrfs_free_extent_map(em); em = NULL; @@ -1238,12 +1255,15 @@ int btrfs_read_folio(struct file *file, struct folio *folio) const u64 start = folio_pos(folio); const u64 end = start + folio_size(folio) - 1; struct extent_state *cached_state = NULL; - struct btrfs_bio_ctrl bio_ctrl = { .opf = REQ_OP_READ }; + struct btrfs_bio_ctrl bio_ctrl = { + .opf = REQ_OP_READ, + .last_em_start = U64_MAX, + }; struct extent_map *em_cached = NULL; int ret; lock_extents_for_read(inode, start, end, &cached_state); - ret = btrfs_do_readpage(folio, &em_cached, &bio_ctrl, NULL); + ret = btrfs_do_readpage(folio, &em_cached, &bio_ctrl); btrfs_unlock_extent(&inode->io_tree, start, end, &cached_state); btrfs_free_extent_map(em_cached); @@ -2583,7 +2603,8 @@ void btrfs_readahead(struct readahead_control *rac) { struct btrfs_bio_ctrl bio_ctrl = { .opf = REQ_OP_READ | REQ_RAHEAD, - .ractl = rac + .ractl = rac, + .last_em_start = U64_MAX, }; struct folio *folio; struct btrfs_inode *inode = BTRFS_I(rac->mapping->host); @@ -2591,12 +2612,11 @@ void btrfs_readahead(struct readahead_control *rac) const u64 end = start + readahead_length(rac) - 1; struct extent_state *cached_state = NULL; struct extent_map *em_cached = NULL; - u64 prev_em_start = (u64)-1; lock_extents_for_read(inode, start, end, &cached_state); while ((folio = readahead_folio(rac)) != NULL) - btrfs_do_readpage(folio, &em_cached, &bio_ctrl, &prev_em_start); + btrfs_do_readpage(folio, &em_cached, &bio_ctrl); btrfs_unlock_extent(&inode->io_tree, start, end, &cached_state); diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 9e4aec7330cb..18db1053cdf0 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -3885,10 +3885,6 @@ static int btrfs_read_locked_inode(struct btrfs_inode *inode, struct btrfs_path bool filled = false; int first_xattr_slot; - ret = btrfs_init_file_extent_tree(inode); - if (ret) - goto out; - ret = btrfs_fill_inode(inode, &rdev); if (!ret) filled = true; @@ -3920,8 +3916,6 @@ static int btrfs_read_locked_inode(struct btrfs_inode *inode, struct btrfs_path i_uid_write(vfs_inode, btrfs_inode_uid(leaf, inode_item)); i_gid_write(vfs_inode, btrfs_inode_gid(leaf, inode_item)); btrfs_i_size_write(inode, btrfs_inode_size(leaf, inode_item)); - btrfs_inode_set_file_extent_range(inode, 0, - round_up(i_size_read(vfs_inode), fs_info->sectorsize)); inode_set_atime(vfs_inode, btrfs_timespec_sec(leaf, &inode_item->atime), btrfs_timespec_nsec(leaf, &inode_item->atime)); @@ -3953,6 +3947,11 @@ static int btrfs_read_locked_inode(struct btrfs_inode *inode, struct btrfs_path btrfs_set_inode_mapping_order(inode); cache_index: + ret = btrfs_init_file_extent_tree(inode); + if (ret) + goto out; + btrfs_inode_set_file_extent_range(inode, 0, + round_up(i_size_read(vfs_inode), fs_info->sectorsize)); /* * If we were modified in the current generation and evicted from memory * and then re-read we need to do a full sync since we don't have any @@ -5696,7 +5695,17 @@ static void btrfs_del_inode_from_root(struct btrfs_inode *inode) bool empty = false; xa_lock(&root->inodes); - entry = __xa_erase(&root->inodes, btrfs_ino(inode)); + /* + * This btrfs_inode is being freed and has already been unhashed at this + * point. It's possible that another btrfs_inode has already been + * allocated for the same inode and inserted itself into the root, so + * don't delete it in that case. + * + * Note that this shouldn't need to allocate memory, so the gfp flags + * don't really matter. + */ + entry = __xa_cmpxchg(&root->inodes, btrfs_ino(inode), inode, NULL, + GFP_ATOMIC); if (entry == inode) empty = xa_empty(&root->inodes); xa_unlock(&root->inodes); @@ -6805,7 +6814,6 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir, struct fscrypt_name fname; u64 index; int ret; - int drop_inode = 0; /* do not allow sys_link's with other subvols of the same device */ if (btrfs_root_id(root) != btrfs_root_id(BTRFS_I(inode)->root)) @@ -6837,44 +6845,44 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir, /* There are several dir indexes for this inode, clear the cache. */ BTRFS_I(inode)->dir_index = 0ULL; - inc_nlink(inode); inode_inc_iversion(inode); inode_set_ctime_current(inode); - ihold(inode); set_bit(BTRFS_INODE_COPY_EVERYTHING, &BTRFS_I(inode)->runtime_flags); ret = btrfs_add_link(trans, BTRFS_I(dir), BTRFS_I(inode), &fname.disk_name, 1, index); + if (ret) + goto fail; + /* Link added now we update the inode item with the new link count. */ + inc_nlink(inode); + ret = btrfs_update_inode(trans, BTRFS_I(inode)); if (ret) { - drop_inode = 1; - } else { - struct dentry *parent = dentry->d_parent; + btrfs_abort_transaction(trans, ret); + goto fail; + } - ret = btrfs_update_inode(trans, BTRFS_I(inode)); - if (ret) + if (inode->i_nlink == 1) { + /* + * If the new hard link count is 1, it's a file created with the + * open(2) O_TMPFILE flag. + */ + ret = btrfs_orphan_del(trans, BTRFS_I(inode)); + if (ret) { + btrfs_abort_transaction(trans, ret); goto fail; - if (inode->i_nlink == 1) { - /* - * If new hard link count is 1, it's a file created - * with open(2) O_TMPFILE flag. - */ - ret = btrfs_orphan_del(trans, BTRFS_I(inode)); - if (ret) - goto fail; } - d_instantiate(dentry, inode); - btrfs_log_new_name(trans, old_dentry, NULL, 0, parent); } + /* Grab reference for the new dentry passed to d_instantiate(). */ + ihold(inode); + d_instantiate(dentry, inode); + btrfs_log_new_name(trans, old_dentry, NULL, 0, dentry->d_parent); + fail: fscrypt_free_filename(&fname); if (trans) btrfs_end_transaction(trans); - if (drop_inode) { - inode_dec_link_count(inode); - iput(inode); - } btrfs_btree_balance_dirty(fs_info); return ret; } @@ -7830,6 +7838,7 @@ struct inode *btrfs_alloc_inode(struct super_block *sb) ei->last_sub_trans = 0; ei->logged_trans = 0; ei->delalloc_bytes = 0; + /* new_delalloc_bytes and last_dir_index_offset are in a union. */ ei->new_delalloc_bytes = 0; ei->defrag_bytes = 0; ei->disk_i_size = 0; diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c index ccaa9a3cf1ce..da102da169fd 100644 --- a/fs/btrfs/qgroup.c +++ b/fs/btrfs/qgroup.c @@ -1455,6 +1455,7 @@ static int __qgroup_excl_accounting(struct btrfs_fs_info *fs_info, u64 ref_root, struct btrfs_qgroup *qgroup; LIST_HEAD(qgroup_list); u64 num_bytes = src->excl; + u64 num_bytes_cmpr = src->excl_cmpr; int ret = 0; qgroup = find_qgroup_rb(fs_info, ref_root); @@ -1466,11 +1467,12 @@ static int __qgroup_excl_accounting(struct btrfs_fs_info *fs_info, u64 ref_root, struct btrfs_qgroup_list *glist; qgroup->rfer += sign * num_bytes; - qgroup->rfer_cmpr += sign * num_bytes; + qgroup->rfer_cmpr += sign * num_bytes_cmpr; WARN_ON(sign < 0 && qgroup->excl < num_bytes); + WARN_ON(sign < 0 && qgroup->excl_cmpr < num_bytes_cmpr); qgroup->excl += sign * num_bytes; - qgroup->excl_cmpr += sign * num_bytes; + qgroup->excl_cmpr += sign * num_bytes_cmpr; if (sign > 0) qgroup_rsv_add_by_qgroup(fs_info, qgroup, src); diff --git a/fs/btrfs/ref-verify.c b/fs/btrfs/ref-verify.c index 3871c3a6c743..9f1858b42c0e 100644 --- a/fs/btrfs/ref-verify.c +++ b/fs/btrfs/ref-verify.c @@ -980,11 +980,18 @@ int btrfs_build_ref_tree(struct btrfs_fs_info *fs_info) if (!btrfs_test_opt(fs_info, REF_VERIFY)) return 0; + extent_root = btrfs_extent_root(fs_info, 0); + /* If the extent tree is damaged we cannot ignore it (IGNOREBADROOTS). */ + if (IS_ERR(extent_root)) { + btrfs_warn(fs_info, "ref-verify: extent tree not available, disabling"); + btrfs_clear_opt(fs_info->mount_opt, REF_VERIFY); + return 0; + } + path = btrfs_alloc_path(); if (!path) return -ENOMEM; - extent_root = btrfs_extent_root(fs_info, 0); eb = btrfs_read_lock_root_node(extent_root); level = btrfs_header_level(eb); path->nodes[level] = eb; diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c index a262b494a89f..b06b8f325537 100644 --- a/fs/btrfs/super.c +++ b/fs/btrfs/super.c @@ -276,6 +276,7 @@ static int btrfs_parse_compress(struct btrfs_fs_context *ctx, const struct fs_parameter *param, int opt) { const char *string = param->string; + int ret; /* * Provide the same semantics as older kernels that don't use fs @@ -294,21 +295,30 @@ static int btrfs_parse_compress(struct btrfs_fs_context *ctx, btrfs_clear_opt(ctx->mount_opt, NODATASUM); } else if (btrfs_match_compress_type(string, "zlib", true)) { ctx->compress_type = BTRFS_COMPRESS_ZLIB; - ctx->compress_level = btrfs_compress_str2level(BTRFS_COMPRESS_ZLIB, - string + 4); + ret = btrfs_compress_str2level(BTRFS_COMPRESS_ZLIB, string + 4, + &ctx->compress_level); + if (ret < 0) + goto error; btrfs_set_opt(ctx->mount_opt, COMPRESS); btrfs_clear_opt(ctx->mount_opt, NODATACOW); btrfs_clear_opt(ctx->mount_opt, NODATASUM); - } else if (btrfs_match_compress_type(string, "lzo", false)) { + } else if (btrfs_match_compress_type(string, "lzo", true)) { ctx->compress_type = BTRFS_COMPRESS_LZO; - ctx->compress_level = 0; + ret = btrfs_compress_str2level(BTRFS_COMPRESS_LZO, string + 3, + &ctx->compress_level); + if (ret < 0) + goto error; + if (string[3] == ':' && string[4]) + btrfs_warn(NULL, "Compression level ignored for LZO"); btrfs_set_opt(ctx->mount_opt, COMPRESS); btrfs_clear_opt(ctx->mount_opt, NODATACOW); btrfs_clear_opt(ctx->mount_opt, NODATASUM); } else if (btrfs_match_compress_type(string, "zstd", true)) { ctx->compress_type = BTRFS_COMPRESS_ZSTD; - ctx->compress_level = btrfs_compress_str2level(BTRFS_COMPRESS_ZSTD, - string + 4); + ret = btrfs_compress_str2level(BTRFS_COMPRESS_ZSTD, string + 4, + &ctx->compress_level); + if (ret < 0) + goto error; btrfs_set_opt(ctx->mount_opt, COMPRESS); btrfs_clear_opt(ctx->mount_opt, NODATACOW); btrfs_clear_opt(ctx->mount_opt, NODATASUM); @@ -319,10 +329,14 @@ static int btrfs_parse_compress(struct btrfs_fs_context *ctx, btrfs_clear_opt(ctx->mount_opt, COMPRESS); btrfs_clear_opt(ctx->mount_opt, FORCE_COMPRESS); } else { - btrfs_err(NULL, "unrecognized compression value %s", string); - return -EINVAL; + ret = -EINVAL; + goto error; } return 0; +error: + btrfs_err(NULL, "failed to parse compression option '%s'", string); + return ret; + } static int btrfs_parse_param(struct fs_context *fc, struct fs_parameter *param) @@ -1079,7 +1093,7 @@ static int btrfs_show_options(struct seq_file *seq, struct dentry *dentry) seq_printf(seq, ",compress-force=%s", compress_type); else seq_printf(seq, ",compress=%s", compress_type); - if (info->compress_level) + if (info->compress_level && info->compress_type != BTRFS_COMPRESS_LZO) seq_printf(seq, ":%d", info->compress_level); } if (btrfs_test_opt(info, NOSSD)) diff --git a/fs/btrfs/tree-checker.c b/fs/btrfs/tree-checker.c index 0f556f4de3f9..a997c7cc35a2 100644 --- a/fs/btrfs/tree-checker.c +++ b/fs/btrfs/tree-checker.c @@ -1756,10 +1756,10 @@ static int check_inode_ref(struct extent_buffer *leaf, while (ptr < end) { u16 namelen; - if (unlikely(ptr + sizeof(iref) > end)) { + if (unlikely(ptr + sizeof(*iref) > end)) { inode_ref_err(leaf, slot, "inode ref overflow, ptr %lu end %lu inode_ref_size %zu", - ptr, end, sizeof(iref)); + ptr, end, sizeof(*iref)); return -EUCLEAN; } diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c index 69e11557fd13..7a63afedd01e 100644 --- a/fs/btrfs/tree-log.c +++ b/fs/btrfs/tree-log.c @@ -1964,7 +1964,7 @@ static noinline int replay_one_name(struct btrfs_trans_handle *trans, search_key.objectid = log_key.objectid; search_key.type = BTRFS_INODE_EXTREF_KEY; - search_key.offset = key->objectid; + search_key.offset = btrfs_extref_hash(key->objectid, name.name, name.len); ret = backref_in_log(root->log_root, &search_key, key->objectid, &name); if (ret < 0) { goto out; @@ -3340,6 +3340,31 @@ int btrfs_free_log_root_tree(struct btrfs_trans_handle *trans, return 0; } +static bool mark_inode_as_not_logged(const struct btrfs_trans_handle *trans, + struct btrfs_inode *inode) +{ + bool ret = false; + + /* + * Do this only if ->logged_trans is still 0 to prevent races with + * concurrent logging as we may see the inode not logged when + * inode_logged() is called but it gets logged after inode_logged() did + * not find it in the log tree and we end up setting ->logged_trans to a + * value less than trans->transid after the concurrent logging task has + * set it to trans->transid. As a consequence, subsequent rename, unlink + * and link operations may end up not logging new names and removing old + * names from the log. + */ + spin_lock(&inode->lock); + if (inode->logged_trans == 0) + inode->logged_trans = trans->transid - 1; + else if (inode->logged_trans == trans->transid) + ret = true; + spin_unlock(&inode->lock); + + return ret; +} + /* * Check if an inode was logged in the current transaction. This correctly deals * with the case where the inode was logged but has a logged_trans of 0, which @@ -3357,15 +3382,32 @@ static int inode_logged(const struct btrfs_trans_handle *trans, struct btrfs_key key; int ret; - if (inode->logged_trans == trans->transid) + /* + * Quick lockless call, since once ->logged_trans is set to the current + * transaction, we never set it to a lower value anywhere else. + */ + if (data_race(inode->logged_trans) == trans->transid) return 1; /* - * If logged_trans is not 0, then we know the inode logged was not logged - * in this transaction, so we can return false right away. + * If logged_trans is not 0 and not trans->transid, then we know the + * inode was not logged in this transaction, so we can return false + * right away. We take the lock to avoid a race caused by load/store + * tearing with a concurrent btrfs_log_inode() call or a concurrent task + * in this function further below - an update to trans->transid can be + * teared into two 32 bits updates for example, in which case we could + * see a positive value that is not trans->transid and assume the inode + * was not logged when it was. */ - if (inode->logged_trans > 0) + spin_lock(&inode->lock); + if (inode->logged_trans == trans->transid) { + spin_unlock(&inode->lock); + return 1; + } else if (inode->logged_trans > 0) { + spin_unlock(&inode->lock); return 0; + } + spin_unlock(&inode->lock); /* * If no log tree was created for this root in this transaction, then @@ -3374,10 +3416,8 @@ static int inode_logged(const struct btrfs_trans_handle *trans, * transaction's ID, to avoid the search below in a future call in case * a log tree gets created after this. */ - if (!test_bit(BTRFS_ROOT_HAS_LOG_TREE, &inode->root->state)) { - inode->logged_trans = trans->transid - 1; - return 0; - } + if (!test_bit(BTRFS_ROOT_HAS_LOG_TREE, &inode->root->state)) + return mark_inode_as_not_logged(trans, inode); /* * We have a log tree and the inode's logged_trans is 0. We can't tell @@ -3431,8 +3471,7 @@ static int inode_logged(const struct btrfs_trans_handle *trans, * Set logged_trans to a value greater than 0 and less then the * current transaction to avoid doing the search in future calls. */ - inode->logged_trans = trans->transid - 1; - return 0; + return mark_inode_as_not_logged(trans, inode); } /* @@ -3440,20 +3479,9 @@ static int inode_logged(const struct btrfs_trans_handle *trans, * the current transacion's ID, to avoid future tree searches as long as * the inode is not evicted again. */ + spin_lock(&inode->lock); inode->logged_trans = trans->transid; - - /* - * If it's a directory, then we must set last_dir_index_offset to the - * maximum possible value, so that the next attempt to log the inode does - * not skip checking if dir index keys found in modified subvolume tree - * leaves have been logged before, otherwise it would result in attempts - * to insert duplicate dir index keys in the log tree. This must be done - * because last_dir_index_offset is an in-memory only field, not persisted - * in the inode item or any other on-disk structure, so its value is lost - * once the inode is evicted. - */ - if (S_ISDIR(inode->vfs_inode.i_mode)) - inode->last_dir_index_offset = (u64)-1; + spin_unlock(&inode->lock); return 1; } @@ -4045,7 +4073,7 @@ done: /* * If the inode was logged before and it was evicted, then its - * last_dir_index_offset is (u64)-1, so we don't the value of the last index + * last_dir_index_offset is 0, so we don't know the value of the last index * key offset. If that's the case, search for it and update the inode. This * is to avoid lookups in the log tree every time we try to insert a dir index * key from a leaf changed in the current transaction, and to allow us to always @@ -4061,7 +4089,7 @@ static int update_last_dir_index_offset(struct btrfs_inode *inode, lockdep_assert_held(&inode->log_mutex); - if (inode->last_dir_index_offset != (u64)-1) + if (inode->last_dir_index_offset != 0) return 0; if (!ctx->logged_before) { diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index fa7a929a0461..c6e3efd6f602 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -2722,6 +2722,11 @@ int btrfs_init_new_device(struct btrfs_fs_info *fs_info, const char *device_path goto error; } + if (bdev_nr_bytes(file_bdev(bdev_file)) <= BTRFS_DEVICE_RANGE_RESERVED) { + ret = -EINVAL; + goto error; + } + if (fs_devices->seeding) { seeding_dev = true; down_write(&sb->s_umount); diff --git a/fs/btrfs/zoned.c b/fs/btrfs/zoned.c index ea662036f441..efc2a81f50e5 100644 --- a/fs/btrfs/zoned.c +++ b/fs/btrfs/zoned.c @@ -2582,9 +2582,9 @@ again: spin_lock(&space_info->lock); space_info->total_bytes -= bg->length; space_info->disk_total -= bg->length * factor; + space_info->disk_total -= bg->zone_unusable; /* There is no allocation ever happened. */ ASSERT(bg->used == 0); - ASSERT(bg->zone_unusable == 0); /* No super block in a block group on the zoned setup. */ ASSERT(bg->bytes_super == 0); spin_unlock(&space_info->lock); diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c index 8b202d789e93..322ed268f14a 100644 --- a/fs/ceph/addr.c +++ b/fs/ceph/addr.c @@ -1264,7 +1264,9 @@ static inline int move_dirty_folio_in_page_array(struct address_space *mapping, 0, gfp_flags); if (IS_ERR(pages[index])) { - if (PTR_ERR(pages[index]) == -EINVAL) { + int err = PTR_ERR(pages[index]); + + if (err == -EINVAL) { pr_err_client(cl, "inode->i_blkbits=%hhu\n", inode->i_blkbits); } @@ -1273,7 +1275,7 @@ static inline int move_dirty_folio_in_page_array(struct address_space *mapping, BUG_ON(ceph_wbc->locked_pages == 0); pages[index] = NULL; - return PTR_ERR(pages[index]); + return err; } } else { pages[index] = &folio->page; @@ -1687,6 +1689,7 @@ get_more_pages: process_folio_batch: rc = ceph_process_folio_batch(mapping, wbc, &ceph_wbc); + ceph_shift_unused_folios_left(&ceph_wbc.fbatch); if (rc) goto release_folios; @@ -1695,8 +1698,6 @@ process_folio_batch: goto release_folios; if (ceph_wbc.processed_in_fbatch) { - ceph_shift_unused_folios_left(&ceph_wbc.fbatch); - if (folio_batch_count(&ceph_wbc.fbatch) == 0 && ceph_wbc.locked_pages < ceph_wbc.max_pages) { doutc(cl, "reached end fbatch, trying for more\n"); diff --git a/fs/ceph/debugfs.c b/fs/ceph/debugfs.c index fdd404fc8112..f3fe786b4143 100644 --- a/fs/ceph/debugfs.c +++ b/fs/ceph/debugfs.c @@ -55,8 +55,6 @@ static int mdsc_show(struct seq_file *s, void *p) struct ceph_mds_client *mdsc = fsc->mdsc; struct ceph_mds_request *req; struct rb_node *rp; - int pathlen = 0; - u64 pathbase; char *path; mutex_lock(&mdsc->mutex); @@ -81,8 +79,8 @@ static int mdsc_show(struct seq_file *s, void *p) if (req->r_inode) { seq_printf(s, " #%llx", ceph_ino(req->r_inode)); } else if (req->r_dentry) { - path = ceph_mdsc_build_path(mdsc, req->r_dentry, &pathlen, - &pathbase, 0); + struct ceph_path_info path_info; + path = ceph_mdsc_build_path(mdsc, req->r_dentry, &path_info, 0); if (IS_ERR(path)) path = NULL; spin_lock(&req->r_dentry->d_lock); @@ -91,7 +89,7 @@ static int mdsc_show(struct seq_file *s, void *p) req->r_dentry, path ? path : ""); spin_unlock(&req->r_dentry->d_lock); - ceph_mdsc_free_path(path, pathlen); + ceph_mdsc_free_path_info(&path_info); } else if (req->r_path1) { seq_printf(s, " #%llx/%s", req->r_ino1.ino, req->r_path1); @@ -100,8 +98,8 @@ static int mdsc_show(struct seq_file *s, void *p) } if (req->r_old_dentry) { - path = ceph_mdsc_build_path(mdsc, req->r_old_dentry, &pathlen, - &pathbase, 0); + struct ceph_path_info path_info; + path = ceph_mdsc_build_path(mdsc, req->r_old_dentry, &path_info, 0); if (IS_ERR(path)) path = NULL; spin_lock(&req->r_old_dentry->d_lock); @@ -111,7 +109,7 @@ static int mdsc_show(struct seq_file *s, void *p) req->r_old_dentry, path ? path : ""); spin_unlock(&req->r_old_dentry->d_lock); - ceph_mdsc_free_path(path, pathlen); + ceph_mdsc_free_path_info(&path_info); } else if (req->r_path2 && req->r_op != CEPH_MDS_OP_SYMLINK) { if (req->r_ino2.ino) seq_printf(s, " #%llx/%s", req->r_ino2.ino, diff --git a/fs/ceph/dir.c b/fs/ceph/dir.c index 8478e7e75df6..32973c62c1a2 100644 --- a/fs/ceph/dir.c +++ b/fs/ceph/dir.c @@ -1271,10 +1271,8 @@ static void ceph_async_unlink_cb(struct ceph_mds_client *mdsc, /* If op failed, mark everyone involved for errors */ if (result) { - int pathlen = 0; - u64 base = 0; - char *path = ceph_mdsc_build_path(mdsc, dentry, &pathlen, - &base, 0); + struct ceph_path_info path_info = {0}; + char *path = ceph_mdsc_build_path(mdsc, dentry, &path_info, 0); /* mark error on parent + clear complete */ mapping_set_error(req->r_parent->i_mapping, result); @@ -1288,8 +1286,8 @@ static void ceph_async_unlink_cb(struct ceph_mds_client *mdsc, mapping_set_error(req->r_old_inode->i_mapping, result); pr_warn_client(cl, "failure path=(%llx)%s result=%d!\n", - base, IS_ERR(path) ? "<<bad>>" : path, result); - ceph_mdsc_free_path(path, pathlen); + path_info.vino.ino, IS_ERR(path) ? "<<bad>>" : path, result); + ceph_mdsc_free_path_info(&path_info); } out: iput(req->r_old_inode); @@ -1347,8 +1345,6 @@ static int ceph_unlink(struct inode *dir, struct dentry *dentry) int err = -EROFS; int op; char *path; - int pathlen; - u64 pathbase; if (ceph_snap(dir) == CEPH_SNAPDIR) { /* rmdir .snap/foo is RMSNAP */ @@ -1367,14 +1363,15 @@ static int ceph_unlink(struct inode *dir, struct dentry *dentry) if (!dn) { try_async = false; } else { - path = ceph_mdsc_build_path(mdsc, dn, &pathlen, &pathbase, 0); + struct ceph_path_info path_info; + path = ceph_mdsc_build_path(mdsc, dn, &path_info, 0); if (IS_ERR(path)) { try_async = false; err = 0; } else { err = ceph_mds_check_access(mdsc, path, MAY_WRITE); } - ceph_mdsc_free_path(path, pathlen); + ceph_mdsc_free_path_info(&path_info); dput(dn); /* For none EACCES cases will let the MDS do the mds auth check */ diff --git a/fs/ceph/file.c b/fs/ceph/file.c index c02f100f8552..978acd3d4b32 100644 --- a/fs/ceph/file.c +++ b/fs/ceph/file.c @@ -368,8 +368,6 @@ int ceph_open(struct inode *inode, struct file *file) int flags, fmode, wanted; struct dentry *dentry; char *path; - int pathlen; - u64 pathbase; bool do_sync = false; int mask = MAY_READ; @@ -399,14 +397,15 @@ int ceph_open(struct inode *inode, struct file *file) if (!dentry) { do_sync = true; } else { - path = ceph_mdsc_build_path(mdsc, dentry, &pathlen, &pathbase, 0); + struct ceph_path_info path_info; + path = ceph_mdsc_build_path(mdsc, dentry, &path_info, 0); if (IS_ERR(path)) { do_sync = true; err = 0; } else { err = ceph_mds_check_access(mdsc, path, mask); } - ceph_mdsc_free_path(path, pathlen); + ceph_mdsc_free_path_info(&path_info); dput(dentry); /* For none EACCES cases will let the MDS do the mds auth check */ @@ -614,15 +613,13 @@ static void ceph_async_create_cb(struct ceph_mds_client *mdsc, mapping_set_error(req->r_parent->i_mapping, result); if (result) { - int pathlen = 0; - u64 base = 0; - char *path = ceph_mdsc_build_path(mdsc, req->r_dentry, &pathlen, - &base, 0); + struct ceph_path_info path_info = {0}; + char *path = ceph_mdsc_build_path(mdsc, req->r_dentry, &path_info, 0); pr_warn_client(cl, "async create failure path=(%llx)%s result=%d!\n", - base, IS_ERR(path) ? "<<bad>>" : path, result); - ceph_mdsc_free_path(path, pathlen); + path_info.vino.ino, IS_ERR(path) ? "<<bad>>" : path, result); + ceph_mdsc_free_path_info(&path_info); ceph_dir_clear_complete(req->r_parent); if (!d_unhashed(dentry)) @@ -791,8 +788,6 @@ int ceph_atomic_open(struct inode *dir, struct dentry *dentry, int mask; int err; char *path; - int pathlen; - u64 pathbase; doutc(cl, "%p %llx.%llx dentry %p '%pd' %s flags %d mode 0%o\n", dir, ceph_vinop(dir), dentry, dentry, @@ -814,7 +809,8 @@ int ceph_atomic_open(struct inode *dir, struct dentry *dentry, if (!dn) { try_async = false; } else { - path = ceph_mdsc_build_path(mdsc, dn, &pathlen, &pathbase, 0); + struct ceph_path_info path_info; + path = ceph_mdsc_build_path(mdsc, dn, &path_info, 0); if (IS_ERR(path)) { try_async = false; err = 0; @@ -826,7 +822,7 @@ int ceph_atomic_open(struct inode *dir, struct dentry *dentry, mask |= MAY_WRITE; err = ceph_mds_check_access(mdsc, path, mask); } - ceph_mdsc_free_path(path, pathlen); + ceph_mdsc_free_path_info(&path_info); dput(dn); /* For none EACCES cases will let the MDS do the mds auth check */ diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c index fc543075b827..f67025465de0 100644 --- a/fs/ceph/inode.c +++ b/fs/ceph/inode.c @@ -55,6 +55,52 @@ static int ceph_set_ino_cb(struct inode *inode, void *data) return 0; } +/* + * Check if the parent inode matches the vino from directory reply info + */ +static inline bool ceph_vino_matches_parent(struct inode *parent, + struct ceph_vino vino) +{ + return ceph_ino(parent) == vino.ino && ceph_snap(parent) == vino.snap; +} + +/* + * Validate that the directory inode referenced by @req->r_parent matches the + * inode number and snapshot id contained in the reply's directory record. If + * they do not match – which can theoretically happen if the parent dentry was + * moved between the time the request was issued and the reply arrived – fall + * back to looking up the correct inode in the inode cache. + * + * A reference is *always* returned. Callers that receive a different inode + * than the original @parent are responsible for dropping the extra reference + * once the reply has been processed. + */ +static struct inode *ceph_get_reply_dir(struct super_block *sb, + struct inode *parent, + struct ceph_mds_reply_info_parsed *rinfo) +{ + struct ceph_vino vino; + + if (unlikely(!rinfo->diri.in)) + return parent; /* nothing to compare against */ + + /* If we didn't have a cached parent inode to begin with, just bail out. */ + if (!parent) + return NULL; + + vino.ino = le64_to_cpu(rinfo->diri.in->ino); + vino.snap = le64_to_cpu(rinfo->diri.in->snapid); + + if (likely(ceph_vino_matches_parent(parent, vino))) + return parent; /* matches – use the original reference */ + + /* Mismatch – this should be rare. Emit a WARN and obtain the correct inode. */ + WARN_ONCE(1, "ceph: reply dir mismatch (parent valid %llx.%llx reply %llx.%llx)\n", + ceph_ino(parent), ceph_snap(parent), vino.ino, vino.snap); + + return ceph_get_inode(sb, vino, NULL); +} + /** * ceph_new_inode - allocate a new inode in advance of an expected create * @dir: parent directory for new inode @@ -1523,6 +1569,7 @@ int ceph_fill_trace(struct super_block *sb, struct ceph_mds_request *req) struct ceph_vino tvino, dvino; struct ceph_fs_client *fsc = ceph_sb_to_fs_client(sb); struct ceph_client *cl = fsc->client; + struct inode *parent_dir = NULL; int err = 0; doutc(cl, "%p is_dentry %d is_target %d\n", req, @@ -1536,10 +1583,17 @@ int ceph_fill_trace(struct super_block *sb, struct ceph_mds_request *req) } if (rinfo->head->is_dentry) { - struct inode *dir = req->r_parent; - - if (dir) { - err = ceph_fill_inode(dir, NULL, &rinfo->diri, + /* + * r_parent may be stale, in cases when R_PARENT_LOCKED is not set, + * so we need to get the correct inode + */ + parent_dir = ceph_get_reply_dir(sb, req->r_parent, rinfo); + if (unlikely(IS_ERR(parent_dir))) { + err = PTR_ERR(parent_dir); + goto done; + } + if (parent_dir) { + err = ceph_fill_inode(parent_dir, NULL, &rinfo->diri, rinfo->dirfrag, session, -1, &req->r_caps_reservation); if (err < 0) @@ -1548,14 +1602,14 @@ int ceph_fill_trace(struct super_block *sb, struct ceph_mds_request *req) WARN_ON_ONCE(1); } - if (dir && req->r_op == CEPH_MDS_OP_LOOKUPNAME && + if (parent_dir && req->r_op == CEPH_MDS_OP_LOOKUPNAME && test_bit(CEPH_MDS_R_PARENT_LOCKED, &req->r_req_flags) && !test_bit(CEPH_MDS_R_ABORTED, &req->r_req_flags)) { bool is_nokey = false; struct qstr dname; struct dentry *dn, *parent; struct fscrypt_str oname = FSTR_INIT(NULL, 0); - struct ceph_fname fname = { .dir = dir, + struct ceph_fname fname = { .dir = parent_dir, .name = rinfo->dname, .ctext = rinfo->altname, .name_len = rinfo->dname_len, @@ -1564,10 +1618,10 @@ int ceph_fill_trace(struct super_block *sb, struct ceph_mds_request *req) BUG_ON(!rinfo->head->is_target); BUG_ON(req->r_dentry); - parent = d_find_any_alias(dir); + parent = d_find_any_alias(parent_dir); BUG_ON(!parent); - err = ceph_fname_alloc_buffer(dir, &oname); + err = ceph_fname_alloc_buffer(parent_dir, &oname); if (err < 0) { dput(parent); goto done; @@ -1576,7 +1630,7 @@ int ceph_fill_trace(struct super_block *sb, struct ceph_mds_request *req) err = ceph_fname_to_usr(&fname, NULL, &oname, &is_nokey); if (err < 0) { dput(parent); - ceph_fname_free_buffer(dir, &oname); + ceph_fname_free_buffer(parent_dir, &oname); goto done; } dname.name = oname.name; @@ -1595,7 +1649,7 @@ retry_lookup: dname.len, dname.name, dn); if (!dn) { dput(parent); - ceph_fname_free_buffer(dir, &oname); + ceph_fname_free_buffer(parent_dir, &oname); err = -ENOMEM; goto done; } @@ -1610,12 +1664,12 @@ retry_lookup: ceph_snap(d_inode(dn)) != tvino.snap)) { doutc(cl, " dn %p points to wrong inode %p\n", dn, d_inode(dn)); - ceph_dir_clear_ordered(dir); + ceph_dir_clear_ordered(parent_dir); d_delete(dn); dput(dn); goto retry_lookup; } - ceph_fname_free_buffer(dir, &oname); + ceph_fname_free_buffer(parent_dir, &oname); req->r_dentry = dn; dput(parent); @@ -1794,6 +1848,9 @@ retry_lookup: &dvino, ptvino); } done: + /* Drop extra ref from ceph_get_reply_dir() if it returned a new inode */ + if (unlikely(!IS_ERR_OR_NULL(parent_dir) && parent_dir != req->r_parent)) + iput(parent_dir); doutc(cl, "done err=%d\n", err); return err; } @@ -2487,22 +2544,21 @@ int __ceph_setattr(struct mnt_idmap *idmap, struct inode *inode, int truncate_retry = 20; /* The RMW will take around 50ms */ struct dentry *dentry; char *path; - int pathlen; - u64 pathbase; bool do_sync = false; dentry = d_find_alias(inode); if (!dentry) { do_sync = true; } else { - path = ceph_mdsc_build_path(mdsc, dentry, &pathlen, &pathbase, 0); + struct ceph_path_info path_info; + path = ceph_mdsc_build_path(mdsc, dentry, &path_info, 0); if (IS_ERR(path)) { do_sync = true; err = 0; } else { err = ceph_mds_check_access(mdsc, path, MAY_WRITE); } - ceph_mdsc_free_path(path, pathlen); + ceph_mdsc_free_path_info(&path_info); dput(dentry); /* For none EACCES cases will let the MDS do the mds auth check */ diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c index 0f497c39ff82..3bc72b47fe4d 100644 --- a/fs/ceph/mds_client.c +++ b/fs/ceph/mds_client.c @@ -2681,8 +2681,7 @@ static u8 *get_fscrypt_altname(const struct ceph_mds_request *req, u32 *plen) * ceph_mdsc_build_path - build a path string to a given dentry * @mdsc: mds client * @dentry: dentry to which path should be built - * @plen: returned length of string - * @pbase: returned base inode number + * @path_info: output path, length, base ino+snap, and freepath ownership flag * @for_wire: is this path going to be sent to the MDS? * * Build a string that represents the path to the dentry. This is mostly called @@ -2700,7 +2699,7 @@ static u8 *get_fscrypt_altname(const struct ceph_mds_request *req, u32 *plen) * foo/.snap/bar -> foo//bar */ char *ceph_mdsc_build_path(struct ceph_mds_client *mdsc, struct dentry *dentry, - int *plen, u64 *pbase, int for_wire) + struct ceph_path_info *path_info, int for_wire) { struct ceph_client *cl = mdsc->fsc->client; struct dentry *cur; @@ -2810,16 +2809,28 @@ retry: return ERR_PTR(-ENAMETOOLONG); } - *pbase = base; - *plen = PATH_MAX - 1 - pos; + /* Initialize the output structure */ + memset(path_info, 0, sizeof(*path_info)); + + path_info->vino.ino = base; + path_info->pathlen = PATH_MAX - 1 - pos; + path_info->path = path + pos; + path_info->freepath = true; + + /* Set snap from dentry if available */ + if (d_inode(dentry)) + path_info->vino.snap = ceph_snap(d_inode(dentry)); + else + path_info->vino.snap = CEPH_NOSNAP; + doutc(cl, "on %p %d built %llx '%.*s'\n", dentry, d_count(dentry), - base, *plen, path + pos); + base, PATH_MAX - 1 - pos, path + pos); return path + pos; } static int build_dentry_path(struct ceph_mds_client *mdsc, struct dentry *dentry, - struct inode *dir, const char **ppath, int *ppathlen, - u64 *pino, bool *pfreepath, bool parent_locked) + struct inode *dir, struct ceph_path_info *path_info, + bool parent_locked) { char *path; @@ -2828,41 +2839,47 @@ static int build_dentry_path(struct ceph_mds_client *mdsc, struct dentry *dentry dir = d_inode_rcu(dentry->d_parent); if (dir && parent_locked && ceph_snap(dir) == CEPH_NOSNAP && !IS_ENCRYPTED(dir)) { - *pino = ceph_ino(dir); + path_info->vino.ino = ceph_ino(dir); + path_info->vino.snap = ceph_snap(dir); rcu_read_unlock(); - *ppath = dentry->d_name.name; - *ppathlen = dentry->d_name.len; + path_info->path = dentry->d_name.name; + path_info->pathlen = dentry->d_name.len; + path_info->freepath = false; return 0; } rcu_read_unlock(); - path = ceph_mdsc_build_path(mdsc, dentry, ppathlen, pino, 1); + path = ceph_mdsc_build_path(mdsc, dentry, path_info, 1); if (IS_ERR(path)) return PTR_ERR(path); - *ppath = path; - *pfreepath = true; + /* + * ceph_mdsc_build_path already fills path_info, including snap handling. + */ return 0; } -static int build_inode_path(struct inode *inode, - const char **ppath, int *ppathlen, u64 *pino, - bool *pfreepath) +static int build_inode_path(struct inode *inode, struct ceph_path_info *path_info) { struct ceph_mds_client *mdsc = ceph_sb_to_mdsc(inode->i_sb); struct dentry *dentry; char *path; if (ceph_snap(inode) == CEPH_NOSNAP) { - *pino = ceph_ino(inode); - *ppathlen = 0; + path_info->vino.ino = ceph_ino(inode); + path_info->vino.snap = ceph_snap(inode); + path_info->pathlen = 0; + path_info->freepath = false; return 0; } dentry = d_find_alias(inode); - path = ceph_mdsc_build_path(mdsc, dentry, ppathlen, pino, 1); + path = ceph_mdsc_build_path(mdsc, dentry, path_info, 1); dput(dentry); if (IS_ERR(path)) return PTR_ERR(path); - *ppath = path; - *pfreepath = true; + /* + * ceph_mdsc_build_path already fills path_info, including snap from dentry. + * Override with inode's snap since that's what this function is for. + */ + path_info->vino.snap = ceph_snap(inode); return 0; } @@ -2872,26 +2889,32 @@ static int build_inode_path(struct inode *inode, */ static int set_request_path_attr(struct ceph_mds_client *mdsc, struct inode *rinode, struct dentry *rdentry, struct inode *rdiri, - const char *rpath, u64 rino, const char **ppath, - int *pathlen, u64 *ino, bool *freepath, + const char *rpath, u64 rino, + struct ceph_path_info *path_info, bool parent_locked) { struct ceph_client *cl = mdsc->fsc->client; int r = 0; + /* Initialize the output structure */ + memset(path_info, 0, sizeof(*path_info)); + if (rinode) { - r = build_inode_path(rinode, ppath, pathlen, ino, freepath); + r = build_inode_path(rinode, path_info); doutc(cl, " inode %p %llx.%llx\n", rinode, ceph_ino(rinode), ceph_snap(rinode)); } else if (rdentry) { - r = build_dentry_path(mdsc, rdentry, rdiri, ppath, pathlen, ino, - freepath, parent_locked); - doutc(cl, " dentry %p %llx/%.*s\n", rdentry, *ino, *pathlen, *ppath); + r = build_dentry_path(mdsc, rdentry, rdiri, path_info, parent_locked); + doutc(cl, " dentry %p %llx/%.*s\n", rdentry, path_info->vino.ino, + path_info->pathlen, path_info->path); } else if (rpath || rino) { - *ino = rino; - *ppath = rpath; - *pathlen = rpath ? strlen(rpath) : 0; - doutc(cl, " path %.*s\n", *pathlen, rpath); + path_info->vino.ino = rino; + path_info->vino.snap = CEPH_NOSNAP; + path_info->path = rpath; + path_info->pathlen = rpath ? strlen(rpath) : 0; + path_info->freepath = false; + + doutc(cl, " path %.*s\n", path_info->pathlen, rpath); } return r; @@ -2968,11 +2991,8 @@ static struct ceph_msg *create_request_message(struct ceph_mds_session *session, struct ceph_client *cl = mdsc->fsc->client; struct ceph_msg *msg; struct ceph_mds_request_head_legacy *lhead; - const char *path1 = NULL; - const char *path2 = NULL; - u64 ino1 = 0, ino2 = 0; - int pathlen1 = 0, pathlen2 = 0; - bool freepath1 = false, freepath2 = false; + struct ceph_path_info path_info1 = {0}; + struct ceph_path_info path_info2 = {0}; struct dentry *old_dentry = NULL; int len; u16 releases; @@ -2982,25 +3002,49 @@ static struct ceph_msg *create_request_message(struct ceph_mds_session *session, u16 request_head_version = mds_supported_head_version(session); kuid_t caller_fsuid = req->r_cred->fsuid; kgid_t caller_fsgid = req->r_cred->fsgid; + bool parent_locked = test_bit(CEPH_MDS_R_PARENT_LOCKED, &req->r_req_flags); ret = set_request_path_attr(mdsc, req->r_inode, req->r_dentry, - req->r_parent, req->r_path1, req->r_ino1.ino, - &path1, &pathlen1, &ino1, &freepath1, - test_bit(CEPH_MDS_R_PARENT_LOCKED, - &req->r_req_flags)); + req->r_parent, req->r_path1, req->r_ino1.ino, + &path_info1, parent_locked); if (ret < 0) { msg = ERR_PTR(ret); goto out; } + /* + * When the parent directory's i_rwsem is *not* locked, req->r_parent may + * have become stale (e.g. after a concurrent rename) between the time the + * dentry was looked up and now. If we detect that the stored r_parent + * does not match the inode number we just encoded for the request, switch + * to the correct inode so that the MDS receives a valid parent reference. + */ + if (!parent_locked && req->r_parent && path_info1.vino.ino && + ceph_ino(req->r_parent) != path_info1.vino.ino) { + struct inode *old_parent = req->r_parent; + struct inode *correct_dir = ceph_get_inode(mdsc->fsc->sb, path_info1.vino, NULL); + if (!IS_ERR(correct_dir)) { + WARN_ONCE(1, "ceph: r_parent mismatch (had %llx wanted %llx) - updating\n", + ceph_ino(old_parent), path_info1.vino.ino); + /* + * Transfer CEPH_CAP_PIN from the old parent to the new one. + * The pin was taken earlier in ceph_mdsc_submit_request(). + */ + ceph_put_cap_refs(ceph_inode(old_parent), CEPH_CAP_PIN); + iput(old_parent); + req->r_parent = correct_dir; + ceph_get_cap_refs(ceph_inode(req->r_parent), CEPH_CAP_PIN); + } + } + /* If r_old_dentry is set, then assume that its parent is locked */ if (req->r_old_dentry && !(req->r_old_dentry->d_flags & DCACHE_DISCONNECTED)) old_dentry = req->r_old_dentry; ret = set_request_path_attr(mdsc, NULL, old_dentry, - req->r_old_dentry_dir, - req->r_path2, req->r_ino2.ino, - &path2, &pathlen2, &ino2, &freepath2, true); + req->r_old_dentry_dir, + req->r_path2, req->r_ino2.ino, + &path_info2, true); if (ret < 0) { msg = ERR_PTR(ret); goto out_free1; @@ -3031,7 +3075,7 @@ static struct ceph_msg *create_request_message(struct ceph_mds_session *session, /* filepaths */ len += 2 * (1 + sizeof(u32) + sizeof(u64)); - len += pathlen1 + pathlen2; + len += path_info1.pathlen + path_info2.pathlen; /* cap releases */ len += sizeof(struct ceph_mds_request_release) * @@ -3039,9 +3083,9 @@ static struct ceph_msg *create_request_message(struct ceph_mds_session *session, !!req->r_old_inode_drop + !!req->r_old_dentry_drop); if (req->r_dentry_drop) - len += pathlen1; + len += path_info1.pathlen; if (req->r_old_dentry_drop) - len += pathlen2; + len += path_info2.pathlen; /* MClientRequest tail */ @@ -3154,8 +3198,8 @@ static struct ceph_msg *create_request_message(struct ceph_mds_session *session, lhead->ino = cpu_to_le64(req->r_deleg_ino); lhead->args = req->r_args; - ceph_encode_filepath(&p, end, ino1, path1); - ceph_encode_filepath(&p, end, ino2, path2); + ceph_encode_filepath(&p, end, path_info1.vino.ino, path_info1.path); + ceph_encode_filepath(&p, end, path_info2.vino.ino, path_info2.path); /* make note of release offset, in case we need to replay */ req->r_request_release_offset = p - msg->front.iov_base; @@ -3218,11 +3262,9 @@ static struct ceph_msg *create_request_message(struct ceph_mds_session *session, msg->hdr.data_off = cpu_to_le16(0); out_free2: - if (freepath2) - ceph_mdsc_free_path((char *)path2, pathlen2); + ceph_mdsc_free_path_info(&path_info2); out_free1: - if (freepath1) - ceph_mdsc_free_path((char *)path1, pathlen1); + ceph_mdsc_free_path_info(&path_info1); out: return msg; out_err: @@ -4579,24 +4621,20 @@ static int reconnect_caps_cb(struct inode *inode, int mds, void *arg) struct ceph_pagelist *pagelist = recon_state->pagelist; struct dentry *dentry; struct ceph_cap *cap; - char *path; - int pathlen = 0, err; - u64 pathbase; + struct ceph_path_info path_info = {0}; + int err; u64 snap_follows; dentry = d_find_primary(inode); if (dentry) { /* set pathbase to parent dir when msg_version >= 2 */ - path = ceph_mdsc_build_path(mdsc, dentry, &pathlen, &pathbase, + char *path = ceph_mdsc_build_path(mdsc, dentry, &path_info, recon_state->msg_version >= 2); dput(dentry); if (IS_ERR(path)) { err = PTR_ERR(path); goto out_err; } - } else { - path = NULL; - pathbase = 0; } spin_lock(&ci->i_ceph_lock); @@ -4629,7 +4667,7 @@ static int reconnect_caps_cb(struct inode *inode, int mds, void *arg) rec.v2.wanted = cpu_to_le32(__ceph_caps_wanted(ci)); rec.v2.issued = cpu_to_le32(cap->issued); rec.v2.snaprealm = cpu_to_le64(ci->i_snap_realm->ino); - rec.v2.pathbase = cpu_to_le64(pathbase); + rec.v2.pathbase = cpu_to_le64(path_info.vino.ino); rec.v2.flock_len = (__force __le32) ((ci->i_ceph_flags & CEPH_I_ERROR_FILELOCK) ? 0 : 1); } else { @@ -4644,7 +4682,7 @@ static int reconnect_caps_cb(struct inode *inode, int mds, void *arg) ts = inode_get_atime(inode); ceph_encode_timespec64(&rec.v1.atime, &ts); rec.v1.snaprealm = cpu_to_le64(ci->i_snap_realm->ino); - rec.v1.pathbase = cpu_to_le64(pathbase); + rec.v1.pathbase = cpu_to_le64(path_info.vino.ino); } if (list_empty(&ci->i_cap_snaps)) { @@ -4706,7 +4744,7 @@ encode_again: sizeof(struct ceph_filelock); rec.v2.flock_len = cpu_to_le32(struct_len); - struct_len += sizeof(u32) + pathlen + sizeof(rec.v2); + struct_len += sizeof(u32) + path_info.pathlen + sizeof(rec.v2); if (struct_v >= 2) struct_len += sizeof(u64); /* snap_follows */ @@ -4730,7 +4768,7 @@ encode_again: ceph_pagelist_encode_8(pagelist, 1); ceph_pagelist_encode_32(pagelist, struct_len); } - ceph_pagelist_encode_string(pagelist, path, pathlen); + ceph_pagelist_encode_string(pagelist, (char *)path_info.path, path_info.pathlen); ceph_pagelist_append(pagelist, &rec, sizeof(rec.v2)); ceph_locks_to_pagelist(flocks, pagelist, num_fcntl_locks, num_flock_locks); @@ -4741,17 +4779,17 @@ out_freeflocks: } else { err = ceph_pagelist_reserve(pagelist, sizeof(u64) + sizeof(u32) + - pathlen + sizeof(rec.v1)); + path_info.pathlen + sizeof(rec.v1)); if (err) goto out_err; ceph_pagelist_encode_64(pagelist, ceph_ino(inode)); - ceph_pagelist_encode_string(pagelist, path, pathlen); + ceph_pagelist_encode_string(pagelist, (char *)path_info.path, path_info.pathlen); ceph_pagelist_append(pagelist, &rec, sizeof(rec.v1)); } out_err: - ceph_mdsc_free_path(path, pathlen); + ceph_mdsc_free_path_info(&path_info); if (!err) recon_state->nr_caps++; return err; diff --git a/fs/ceph/mds_client.h b/fs/ceph/mds_client.h index 3e2a6fa7c19a..0428a5eaf28c 100644 --- a/fs/ceph/mds_client.h +++ b/fs/ceph/mds_client.h @@ -617,14 +617,24 @@ extern int ceph_mds_check_access(struct ceph_mds_client *mdsc, char *tpath, extern void ceph_mdsc_pre_umount(struct ceph_mds_client *mdsc); -static inline void ceph_mdsc_free_path(char *path, int len) +/* + * Structure to group path-related output parameters for build_*_path functions + */ +struct ceph_path_info { + const char *path; + int pathlen; + struct ceph_vino vino; + bool freepath; +}; + +static inline void ceph_mdsc_free_path_info(const struct ceph_path_info *path_info) { - if (!IS_ERR_OR_NULL(path)) - __putname(path - (PATH_MAX - 1 - len)); + if (path_info && path_info->freepath && !IS_ERR_OR_NULL(path_info->path)) + __putname((char *)path_info->path - (PATH_MAX - 1 - path_info->pathlen)); } extern char *ceph_mdsc_build_path(struct ceph_mds_client *mdsc, - struct dentry *dentry, int *plen, u64 *base, + struct dentry *dentry, struct ceph_path_info *path_info, int for_wire); extern void __ceph_mdsc_drop_dentry_lease(struct dentry *dentry); diff --git a/fs/coredump.c b/fs/coredump.c index 5dce257c67fc..60bc9685e149 100644 --- a/fs/coredump.c +++ b/fs/coredump.c @@ -1466,11 +1466,15 @@ static int proc_dostring_coredump(const struct ctl_table *table, int write, ssize_t retval; char old_core_pattern[CORENAME_MAX_SIZE]; + if (write) + return proc_dostring(table, write, buffer, lenp, ppos); + retval = strscpy(old_core_pattern, core_pattern, CORENAME_MAX_SIZE); error = proc_dostring(table, write, buffer, lenp, ppos); if (error) return error; + if (!check_coredump_socket()) { strscpy(core_pattern, old_core_pattern, retval + 1); return -EINVAL; diff --git a/fs/efivarfs/super.c b/fs/efivarfs/super.c index c4a139911356..4bb4002e3cdf 100644 --- a/fs/efivarfs/super.c +++ b/fs/efivarfs/super.c @@ -152,6 +152,10 @@ static int efivarfs_d_compare(const struct dentry *dentry, { int guid = len - EFI_VARIABLE_GUID_LEN; + /* Parallel lookups may produce a temporary invalid filename */ + if (guid <= 0) + return 1; + if (name->len != len) return 1; diff --git a/fs/erofs/erofs_fs.h b/fs/erofs/erofs_fs.h index 377ee12b8b96..3d5738f80072 100644 --- a/fs/erofs/erofs_fs.h +++ b/fs/erofs/erofs_fs.h @@ -12,10 +12,12 @@ /* to allow for x86 boot sectors and other oddities. */ #define EROFS_SUPER_OFFSET 1024 -#define EROFS_FEATURE_COMPAT_SB_CHKSUM 0x00000001 -#define EROFS_FEATURE_COMPAT_MTIME 0x00000002 -#define EROFS_FEATURE_COMPAT_XATTR_FILTER 0x00000004 +#define EROFS_FEATURE_COMPAT_SB_CHKSUM 0x00000001 +#define EROFS_FEATURE_COMPAT_MTIME 0x00000002 +#define EROFS_FEATURE_COMPAT_XATTR_FILTER 0x00000004 #define EROFS_FEATURE_COMPAT_SHARED_EA_IN_METABOX 0x00000008 +#define EROFS_FEATURE_COMPAT_PLAIN_XATTR_PFX 0x00000010 + /* * Any bits that aren't in EROFS_ALL_FEATURE_INCOMPAT should diff --git a/fs/erofs/internal.h b/fs/erofs/internal.h index 4ccc5f0ee8df..9319c66e86c3 100644 --- a/fs/erofs/internal.h +++ b/fs/erofs/internal.h @@ -234,6 +234,7 @@ EROFS_FEATURE_FUNCS(metabox, incompat, INCOMPAT_METABOX) EROFS_FEATURE_FUNCS(sb_chksum, compat, COMPAT_SB_CHKSUM) EROFS_FEATURE_FUNCS(xattr_filter, compat, COMPAT_XATTR_FILTER) EROFS_FEATURE_FUNCS(shared_ea_in_metabox, compat, COMPAT_SHARED_EA_IN_METABOX) +EROFS_FEATURE_FUNCS(plain_xattr_pfx, compat, COMPAT_PLAIN_XATTR_PFX) static inline u64 erofs_nid_to_ino64(struct erofs_sb_info *sbi, erofs_nid_t nid) { diff --git a/fs/erofs/super.c b/fs/erofs/super.c index 1b529ace4db0..db13b40a78e0 100644 --- a/fs/erofs/super.c +++ b/fs/erofs/super.c @@ -1018,10 +1018,22 @@ static int erofs_show_options(struct seq_file *seq, struct dentry *root) return 0; } +static void erofs_evict_inode(struct inode *inode) +{ +#ifdef CONFIG_FS_DAX + if (IS_DAX(inode)) + dax_break_layout_final(inode); +#endif + + truncate_inode_pages_final(&inode->i_data); + clear_inode(inode); +} + const struct super_operations erofs_sops = { .put_super = erofs_put_super, .alloc_inode = erofs_alloc_inode, .free_inode = erofs_free_inode, + .evict_inode = erofs_evict_inode, .statfs = erofs_statfs, .show_options = erofs_show_options, }; diff --git a/fs/erofs/xattr.c b/fs/erofs/xattr.c index eaa9efd766ee..396536d9a862 100644 --- a/fs/erofs/xattr.c +++ b/fs/erofs/xattr.c @@ -482,6 +482,7 @@ int erofs_xattr_prefixes_init(struct super_block *sb) erofs_off_t pos = (erofs_off_t)sbi->xattr_prefix_start << 2; struct erofs_xattr_prefix_item *pfs; int ret = 0, i, len; + bool plain = erofs_sb_has_plain_xattr_pfx(sbi); if (!sbi->xattr_prefix_count) return 0; @@ -490,9 +491,15 @@ int erofs_xattr_prefixes_init(struct super_block *sb) if (!pfs) return -ENOMEM; - if (sbi->packed_inode) - buf.mapping = sbi->packed_inode->i_mapping; - else + if (!plain) { + if (erofs_sb_has_metabox(sbi)) + (void)erofs_init_metabuf(&buf, sb, true); + else if (sbi->packed_inode) + buf.mapping = sbi->packed_inode->i_mapping; + else + plain = true; + } + if (plain) (void)erofs_init_metabuf(&buf, sb, false); for (i = 0; i < sbi->xattr_prefix_count; i++) { diff --git a/fs/erofs/zmap.c b/fs/erofs/zmap.c index a93efd95c555..798223e6da9c 100644 --- a/fs/erofs/zmap.c +++ b/fs/erofs/zmap.c @@ -394,10 +394,10 @@ static int z_erofs_map_blocks_fo(struct inode *inode, .map = map, .in_mbox = erofs_inode_in_metabox(inode), }; - int err = 0; - unsigned int endoff, afmt; + unsigned int endoff; unsigned long initial_lcn; unsigned long long ofs, end; + int err; ofs = flags & EROFS_GET_BLOCKS_FINDTAIL ? inode->i_size - 1 : map->m_la; if (fragment && !(flags & EROFS_GET_BLOCKS_FINDTAIL) && @@ -482,20 +482,15 @@ static int z_erofs_map_blocks_fo(struct inode *inode, err = -EFSCORRUPTED; goto unmap_out; } - afmt = vi->z_advise & Z_EROFS_ADVISE_INTERLACED_PCLUSTER ? - Z_EROFS_COMPRESSION_INTERLACED : - Z_EROFS_COMPRESSION_SHIFTED; + if (vi->z_advise & Z_EROFS_ADVISE_INTERLACED_PCLUSTER) + map->m_algorithmformat = Z_EROFS_COMPRESSION_INTERLACED; + else + map->m_algorithmformat = Z_EROFS_COMPRESSION_SHIFTED; + } else if (m.headtype == Z_EROFS_LCLUSTER_TYPE_HEAD2) { + map->m_algorithmformat = vi->z_algorithmtype[1]; } else { - afmt = m.headtype == Z_EROFS_LCLUSTER_TYPE_HEAD2 ? - vi->z_algorithmtype[1] : vi->z_algorithmtype[0]; - if (!(EROFS_I_SB(inode)->available_compr_algs & (1 << afmt))) { - erofs_err(sb, "inconsistent algorithmtype %u for nid %llu", - afmt, vi->nid); - err = -EFSCORRUPTED; - goto unmap_out; - } + map->m_algorithmformat = vi->z_algorithmtype[0]; } - map->m_algorithmformat = afmt; if ((flags & EROFS_GET_BLOCKS_FIEMAP) || ((flags & EROFS_GET_BLOCKS_READMORE) && @@ -626,9 +621,9 @@ static int z_erofs_fill_inode(struct inode *inode, struct erofs_map_blocks *map) { struct erofs_inode *const vi = EROFS_I(inode); struct super_block *const sb = inode->i_sb; - int err, headnr; - erofs_off_t pos; struct z_erofs_map_header *h; + erofs_off_t pos; + int err = 0; if (test_bit(EROFS_I_Z_INITED_BIT, &vi->flags)) { /* @@ -642,7 +637,6 @@ static int z_erofs_fill_inode(struct inode *inode, struct erofs_map_blocks *map) if (wait_on_bit_lock(&vi->flags, EROFS_I_BL_Z_BIT, TASK_KILLABLE)) return -ERESTARTSYS; - err = 0; if (test_bit(EROFS_I_Z_INITED_BIT, &vi->flags)) goto out_unlock; @@ -679,15 +673,6 @@ static int z_erofs_fill_inode(struct inode *inode, struct erofs_map_blocks *map) else if (vi->z_advise & Z_EROFS_ADVISE_INLINE_PCLUSTER) vi->z_idata_size = le16_to_cpu(h->h_idata_size); - headnr = 0; - if (vi->z_algorithmtype[0] >= Z_EROFS_COMPRESSION_MAX || - vi->z_algorithmtype[++headnr] >= Z_EROFS_COMPRESSION_MAX) { - erofs_err(sb, "unknown HEAD%u format %u for nid %llu, please upgrade kernel", - headnr + 1, vi->z_algorithmtype[headnr], vi->nid); - err = -EOPNOTSUPP; - goto out_unlock; - } - if (!erofs_sb_has_big_pcluster(EROFS_SB(sb)) && vi->z_advise & (Z_EROFS_ADVISE_BIG_PCLUSTER_1 | Z_EROFS_ADVISE_BIG_PCLUSTER_2)) { @@ -726,6 +711,30 @@ out_unlock: return err; } +static int z_erofs_map_sanity_check(struct inode *inode, + struct erofs_map_blocks *map) +{ + struct erofs_sb_info *sbi = EROFS_I_SB(inode); + + if (!(map->m_flags & EROFS_MAP_ENCODED)) + return 0; + if (unlikely(map->m_algorithmformat >= Z_EROFS_COMPRESSION_RUNTIME_MAX)) { + erofs_err(inode->i_sb, "unknown algorithm %d @ pos %llu for nid %llu, please upgrade kernel", + map->m_algorithmformat, map->m_la, EROFS_I(inode)->nid); + return -EOPNOTSUPP; + } + if (unlikely(map->m_algorithmformat < Z_EROFS_COMPRESSION_MAX && + !(sbi->available_compr_algs & (1 << map->m_algorithmformat)))) { + erofs_err(inode->i_sb, "inconsistent algorithmtype %u for nid %llu", + map->m_algorithmformat, EROFS_I(inode)->nid); + return -EFSCORRUPTED; + } + if (unlikely(map->m_plen > Z_EROFS_PCLUSTER_MAX_SIZE || + map->m_llen > Z_EROFS_PCLUSTER_MAX_DSIZE)) + return -EOPNOTSUPP; + return 0; +} + int z_erofs_map_blocks_iter(struct inode *inode, struct erofs_map_blocks *map, int flags) { @@ -746,10 +755,8 @@ int z_erofs_map_blocks_iter(struct inode *inode, struct erofs_map_blocks *map, else err = z_erofs_map_blocks_fo(inode, map, flags); } - if (!err && (map->m_flags & EROFS_MAP_ENCODED) && - unlikely(map->m_plen > Z_EROFS_PCLUSTER_MAX_SIZE || - map->m_llen > Z_EROFS_PCLUSTER_MAX_DSIZE)) - err = -EOPNOTSUPP; + if (!err) + err = z_erofs_map_sanity_check(inode, map); if (err) map->m_llen = 0; } diff --git a/fs/exec.c b/fs/exec.c index 2a1e5e4042a1..e861a4b7ffda 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -2048,7 +2048,7 @@ static int proc_dointvec_minmax_coredump(const struct ctl_table *table, int writ { int error = proc_dointvec_minmax(table, write, buffer, lenp, ppos); - if (!error) + if (!error && !write) validate_coredump_safety(); return error; } diff --git a/fs/fhandle.c b/fs/fhandle.c index 68a7d2861c58..a907ddfac4d5 100644 --- a/fs/fhandle.c +++ b/fs/fhandle.c @@ -208,6 +208,14 @@ static int vfs_dentry_acceptable(void *context, struct dentry *dentry) return 1; /* + * Verify that the decoded dentry itself has a valid id mapping. + * In case the decoded dentry is the mountfd root itself, this + * verifies that the mountfd inode itself has a valid id mapping. + */ + if (!privileged_wrt_inode_uidgid(user_ns, idmap, d_inode(dentry))) + return 0; + + /* * It's racy as we're not taking rename_lock but we're able to ignore * permissions and we just need an approximation whether we were able * to follow a path to the file. diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c index e80cd8f2c049..5150aa25e64b 100644 --- a/fs/fuse/dev.c +++ b/fs/fuse/dev.c @@ -1893,7 +1893,7 @@ static int fuse_retrieve(struct fuse_mount *fm, struct inode *inode, index = outarg->offset >> PAGE_SHIFT; - while (num) { + while (num && ap->num_folios < num_pages) { struct folio *folio; unsigned int folio_offset; unsigned int nr_bytes; diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c index 2d817d7cab26..5c569c3cb53f 100644 --- a/fs/fuse/dir.c +++ b/fs/fuse/dir.c @@ -1199,7 +1199,7 @@ static void fuse_fillattr(struct mnt_idmap *idmap, struct inode *inode, if (attr->blksize != 0) blkbits = ilog2(attr->blksize); else - blkbits = inode->i_sb->s_blocksize_bits; + blkbits = fc->blkbits; stat->blksize = 1 << blkbits; } @@ -1377,6 +1377,7 @@ retry: generic_fillattr(idmap, request_mask, inode, stat); stat->mode = fi->orig_i_mode; stat->ino = fi->orig_ino; + stat->blksize = 1 << fi->cached_i_blkbits; if (test_bit(FUSE_I_BTIME, &fi->state)) { stat->btime = fi->i_btime; stat->result_mask |= STATX_BTIME; diff --git a/fs/fuse/file.c b/fs/fuse/file.c index 5525a4520b0f..4adcf09d4b01 100644 --- a/fs/fuse/file.c +++ b/fs/fuse/file.c @@ -2960,7 +2960,7 @@ static ssize_t __fuse_copy_file_range(struct file *file_in, loff_t pos_in, .nodeid_out = ff_out->nodeid, .fh_out = ff_out->fh, .off_out = pos_out, - .len = len, + .len = min_t(size_t, len, UINT_MAX & PAGE_MASK), .flags = flags }; struct fuse_write_out outarg; @@ -3026,6 +3026,9 @@ static ssize_t __fuse_copy_file_range(struct file *file_in, loff_t pos_in, fc->no_copy_file_range = 1; err = -EOPNOTSUPP; } + if (!err && outarg.size > len) + err = -EIO; + if (err) goto out; diff --git a/fs/fuse/fuse_i.h b/fs/fuse/fuse_i.h index ec248d13c8bf..cc428d04be3e 100644 --- a/fs/fuse/fuse_i.h +++ b/fs/fuse/fuse_i.h @@ -210,6 +210,12 @@ struct fuse_inode { /** Reference to backing file in passthrough mode */ struct fuse_backing *fb; #endif + + /* + * The underlying inode->i_blkbits value will not be modified, + * so preserve the blocksize specified by the server. + */ + u8 cached_i_blkbits; }; /** FUSE inode state bits */ @@ -969,6 +975,14 @@ struct fuse_conn { /* Request timeout (in jiffies). 0 = no timeout */ unsigned int req_timeout; } timeout; + + /* + * This is a workaround until fuse uses iomap for reads. + * For fuseblk servers, this represents the blocksize passed in at + * mount time and for regular fuse servers, this is equivalent to + * inode->i_blkbits. + */ + u8 blkbits; }; /* diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c index 67c2318bfc42..7ddfd2b3cc9c 100644 --- a/fs/fuse/inode.c +++ b/fs/fuse/inode.c @@ -289,6 +289,11 @@ void fuse_change_attributes_common(struct inode *inode, struct fuse_attr *attr, } } + if (attr->blksize) + fi->cached_i_blkbits = ilog2(attr->blksize); + else + fi->cached_i_blkbits = fc->blkbits; + /* * Don't set the sticky bit in i_mode, unless we want the VFS * to check permissions. This prevents failures due to the @@ -1805,10 +1810,21 @@ int fuse_fill_super_common(struct super_block *sb, struct fuse_fs_context *ctx) err = -EINVAL; if (!sb_set_blocksize(sb, ctx->blksize)) goto err; + /* + * This is a workaround until fuse hooks into iomap for reads. + * Use PAGE_SIZE for the blocksize else if the writeback cache + * is enabled, buffered writes go through iomap and a read may + * overwrite partially written data if blocksize < PAGE_SIZE + */ + fc->blkbits = sb->s_blocksize_bits; + if (ctx->blksize != PAGE_SIZE && + !sb_set_blocksize(sb, PAGE_SIZE)) + goto err; #endif } else { sb->s_blocksize = PAGE_SIZE; sb->s_blocksize_bits = PAGE_SHIFT; + fc->blkbits = sb->s_blocksize_bits; } sb->s_subtype = ctx->subtype; diff --git a/fs/fuse/passthrough.c b/fs/fuse/passthrough.c index 607ef735ad4a..eb97ac009e75 100644 --- a/fs/fuse/passthrough.c +++ b/fs/fuse/passthrough.c @@ -237,6 +237,11 @@ int fuse_backing_open(struct fuse_conn *fc, struct fuse_backing_map *map) if (!file) goto out; + /* read/write/splice/mmap passthrough only relevant for regular files */ + res = d_is_dir(file->f_path.dentry) ? -EISDIR : -EINVAL; + if (!d_is_reg(file->f_path.dentry)) + goto out_fput; + backing_sb = file_inode(file)->i_sb; res = -ELOOP; if (backing_sb->s_stack_depth >= fc->max_stack_depth) diff --git a/fs/fuse/virtio_fs.c b/fs/fuse/virtio_fs.c index c826e7ca49f5..76c8fd0bfc75 100644 --- a/fs/fuse/virtio_fs.c +++ b/fs/fuse/virtio_fs.c @@ -1016,7 +1016,7 @@ static long virtio_fs_direct_access(struct dax_device *dax_dev, pgoff_t pgoff, if (kaddr) *kaddr = fs->window_kaddr + offset; if (pfn) - *pfn = fs->window_phys_addr + offset; + *pfn = PHYS_PFN(fs->window_phys_addr + offset); return nr_pages > max_nr_pages ? max_nr_pages : nr_pages; } diff --git a/fs/kernfs/file.c b/fs/kernfs/file.c index a6c692cac616..9adf36e6364b 100644 --- a/fs/kernfs/file.c +++ b/fs/kernfs/file.c @@ -70,6 +70,24 @@ static struct kernfs_open_node *of_on(struct kernfs_open_file *of) !list_empty(&of->list)); } +/* Get active reference to kernfs node for an open file */ +static struct kernfs_open_file *kernfs_get_active_of(struct kernfs_open_file *of) +{ + /* Skip if file was already released */ + if (unlikely(of->released)) + return NULL; + + if (!kernfs_get_active(of->kn)) + return NULL; + + return of; +} + +static void kernfs_put_active_of(struct kernfs_open_file *of) +{ + return kernfs_put_active(of->kn); +} + /** * kernfs_deref_open_node_locked - Get kernfs_open_node corresponding to @kn * @@ -139,7 +157,7 @@ static void kernfs_seq_stop_active(struct seq_file *sf, void *v) if (ops->seq_stop) ops->seq_stop(sf, v); - kernfs_put_active(of->kn); + kernfs_put_active_of(of); } static void *kernfs_seq_start(struct seq_file *sf, loff_t *ppos) @@ -152,7 +170,7 @@ static void *kernfs_seq_start(struct seq_file *sf, loff_t *ppos) * the ops aren't called concurrently for the same open file. */ mutex_lock(&of->mutex); - if (!kernfs_get_active(of->kn)) + if (!kernfs_get_active_of(of)) return ERR_PTR(-ENODEV); ops = kernfs_ops(of->kn); @@ -238,7 +256,7 @@ static ssize_t kernfs_file_read_iter(struct kiocb *iocb, struct iov_iter *iter) * the ops aren't called concurrently for the same open file. */ mutex_lock(&of->mutex); - if (!kernfs_get_active(of->kn)) { + if (!kernfs_get_active_of(of)) { len = -ENODEV; mutex_unlock(&of->mutex); goto out_free; @@ -252,7 +270,7 @@ static ssize_t kernfs_file_read_iter(struct kiocb *iocb, struct iov_iter *iter) else len = -EINVAL; - kernfs_put_active(of->kn); + kernfs_put_active_of(of); mutex_unlock(&of->mutex); if (len < 0) @@ -323,7 +341,7 @@ static ssize_t kernfs_fop_write_iter(struct kiocb *iocb, struct iov_iter *iter) * the ops aren't called concurrently for the same open file. */ mutex_lock(&of->mutex); - if (!kernfs_get_active(of->kn)) { + if (!kernfs_get_active_of(of)) { mutex_unlock(&of->mutex); len = -ENODEV; goto out_free; @@ -335,7 +353,7 @@ static ssize_t kernfs_fop_write_iter(struct kiocb *iocb, struct iov_iter *iter) else len = -EINVAL; - kernfs_put_active(of->kn); + kernfs_put_active_of(of); mutex_unlock(&of->mutex); if (len > 0) @@ -357,13 +375,13 @@ static void kernfs_vma_open(struct vm_area_struct *vma) if (!of->vm_ops) return; - if (!kernfs_get_active(of->kn)) + if (!kernfs_get_active_of(of)) return; if (of->vm_ops->open) of->vm_ops->open(vma); - kernfs_put_active(of->kn); + kernfs_put_active_of(of); } static vm_fault_t kernfs_vma_fault(struct vm_fault *vmf) @@ -375,14 +393,14 @@ static vm_fault_t kernfs_vma_fault(struct vm_fault *vmf) if (!of->vm_ops) return VM_FAULT_SIGBUS; - if (!kernfs_get_active(of->kn)) + if (!kernfs_get_active_of(of)) return VM_FAULT_SIGBUS; ret = VM_FAULT_SIGBUS; if (of->vm_ops->fault) ret = of->vm_ops->fault(vmf); - kernfs_put_active(of->kn); + kernfs_put_active_of(of); return ret; } @@ -395,7 +413,7 @@ static vm_fault_t kernfs_vma_page_mkwrite(struct vm_fault *vmf) if (!of->vm_ops) return VM_FAULT_SIGBUS; - if (!kernfs_get_active(of->kn)) + if (!kernfs_get_active_of(of)) return VM_FAULT_SIGBUS; ret = 0; @@ -404,7 +422,7 @@ static vm_fault_t kernfs_vma_page_mkwrite(struct vm_fault *vmf) else file_update_time(file); - kernfs_put_active(of->kn); + kernfs_put_active_of(of); return ret; } @@ -418,14 +436,14 @@ static int kernfs_vma_access(struct vm_area_struct *vma, unsigned long addr, if (!of->vm_ops) return -EINVAL; - if (!kernfs_get_active(of->kn)) + if (!kernfs_get_active_of(of)) return -EINVAL; ret = -EINVAL; if (of->vm_ops->access) ret = of->vm_ops->access(vma, addr, buf, len, write); - kernfs_put_active(of->kn); + kernfs_put_active_of(of); return ret; } @@ -455,7 +473,7 @@ static int kernfs_fop_mmap(struct file *file, struct vm_area_struct *vma) mutex_lock(&of->mutex); rc = -ENODEV; - if (!kernfs_get_active(of->kn)) + if (!kernfs_get_active_of(of)) goto out_unlock; ops = kernfs_ops(of->kn); @@ -490,7 +508,7 @@ static int kernfs_fop_mmap(struct file *file, struct vm_area_struct *vma) } vma->vm_ops = &kernfs_vm_ops; out_put: - kernfs_put_active(of->kn); + kernfs_put_active_of(of); out_unlock: mutex_unlock(&of->mutex); @@ -852,7 +870,7 @@ static __poll_t kernfs_fop_poll(struct file *filp, poll_table *wait) struct kernfs_node *kn = kernfs_dentry_node(filp->f_path.dentry); __poll_t ret; - if (!kernfs_get_active(kn)) + if (!kernfs_get_active_of(of)) return DEFAULT_POLLMASK|EPOLLERR|EPOLLPRI; if (kn->attr.ops->poll) @@ -860,7 +878,7 @@ static __poll_t kernfs_fop_poll(struct file *filp, poll_table *wait) else ret = kernfs_generic_poll(of, wait); - kernfs_put_active(kn); + kernfs_put_active_of(of); return ret; } @@ -875,7 +893,7 @@ static loff_t kernfs_fop_llseek(struct file *file, loff_t offset, int whence) * the ops aren't called concurrently for the same open file. */ mutex_lock(&of->mutex); - if (!kernfs_get_active(of->kn)) { + if (!kernfs_get_active_of(of)) { mutex_unlock(&of->mutex); return -ENODEV; } @@ -886,7 +904,7 @@ static loff_t kernfs_fop_llseek(struct file *file, loff_t offset, int whence) else ret = generic_file_llseek(file, offset, whence); - kernfs_put_active(of->kn); + kernfs_put_active_of(of); mutex_unlock(&of->mutex); return ret; } diff --git a/fs/namespace.c b/fs/namespace.c index ae6d1312b184..51f77c65c0c6 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -2455,7 +2455,7 @@ struct vfsmount *clone_private_mount(const struct path *path) return ERR_PTR(-EINVAL); } - if (!ns_capable(old_mnt->mnt_ns->user_ns, CAP_SYS_ADMIN)) + if (!ns_capable(old_mnt->mnt_ns->user_ns, CAP_SYS_ADMIN)) return ERR_PTR(-EPERM); if (__has_locked_children(old_mnt, path->dentry)) diff --git a/fs/nfs/client.c b/fs/nfs/client.c index 8fb4a950dd55..4e3dcc157a83 100644 --- a/fs/nfs/client.c +++ b/fs/nfs/client.c @@ -888,6 +888,8 @@ static void nfs_server_set_fsinfo(struct nfs_server *server, if (fsinfo->xattr_support) server->caps |= NFS_CAP_XATTR; + else + server->caps &= ~NFS_CAP_XATTR; #endif } diff --git a/fs/nfs/file.c b/fs/nfs/file.c index 86e36c630f09..8059ece82468 100644 --- a/fs/nfs/file.c +++ b/fs/nfs/file.c @@ -28,6 +28,7 @@ #include <linux/mm.h> #include <linux/pagemap.h> #include <linux/gfp.h> +#include <linux/rmap.h> #include <linux/swap.h> #include <linux/compaction.h> @@ -280,6 +281,37 @@ nfs_file_fsync(struct file *file, loff_t start, loff_t end, int datasync) } EXPORT_SYMBOL_GPL(nfs_file_fsync); +void nfs_truncate_last_folio(struct address_space *mapping, loff_t from, + loff_t to) +{ + struct folio *folio; + + if (from >= to) + return; + + folio = filemap_lock_folio(mapping, from >> PAGE_SHIFT); + if (IS_ERR(folio)) + return; + + if (folio_mkclean(folio)) + folio_mark_dirty(folio); + + if (folio_test_uptodate(folio)) { + loff_t fpos = folio_pos(folio); + size_t offset = from - fpos; + size_t end = folio_size(folio); + + if (to - fpos < end) + end = to - fpos; + folio_zero_segment(folio, offset, end); + trace_nfs_size_truncate_folio(mapping->host, to); + } + + folio_unlock(folio); + folio_put(folio); +} +EXPORT_SYMBOL_GPL(nfs_truncate_last_folio); + /* * Decide whether a read/modify/write cycle may be more efficient * then a modify/write/read cycle when writing to a page in the @@ -356,6 +388,7 @@ static int nfs_write_begin(const struct kiocb *iocb, dfprintk(PAGECACHE, "NFS: write_begin(%pD2(%lu), %u@%lld)\n", file, mapping->host->i_ino, len, (long long) pos); + nfs_truncate_last_folio(mapping, i_size_read(mapping->host), pos); fgp |= fgf_set_order(len); start: @@ -442,10 +475,11 @@ static void nfs_invalidate_folio(struct folio *folio, size_t offset, dfprintk(PAGECACHE, "NFS: invalidate_folio(%lu, %zu, %zu)\n", folio->index, offset, length); - if (offset != 0 || length < folio_size(folio)) - return; /* Cancel any unstarted writes on this page */ - nfs_wb_folio_cancel(inode, folio); + if (offset != 0 || length < folio_size(folio)) + nfs_wb_folio(inode, folio); + else + nfs_wb_folio_cancel(inode, folio); folio_wait_private_2(folio); /* [DEPRECATED] */ trace_nfs_invalidate_folio(inode, folio_pos(folio) + offset, length); } diff --git a/fs/nfs/flexfilelayout/flexfilelayout.c b/fs/nfs/flexfilelayout/flexfilelayout.c index 8dc921d83538..9edb5f9b0c4e 100644 --- a/fs/nfs/flexfilelayout/flexfilelayout.c +++ b/fs/nfs/flexfilelayout/flexfilelayout.c @@ -293,7 +293,7 @@ ff_lseg_match_mirrors(struct pnfs_layout_segment *l1, struct pnfs_layout_segment *l2) { const struct nfs4_ff_layout_segment *fl1 = FF_LAYOUT_LSEG(l1); - const struct nfs4_ff_layout_segment *fl2 = FF_LAYOUT_LSEG(l1); + const struct nfs4_ff_layout_segment *fl2 = FF_LAYOUT_LSEG(l2); u32 i; if (fl1->mirror_array_cnt != fl2->mirror_array_cnt) @@ -773,8 +773,11 @@ ff_layout_choose_ds_for_read(struct pnfs_layout_segment *lseg, continue; if (check_device && - nfs4_test_deviceid_unavailable(&mirror->mirror_ds->id_node)) + nfs4_test_deviceid_unavailable(&mirror->mirror_ds->id_node)) { + // reinitialize the error state in case if this is the last iteration + ds = ERR_PTR(-EINVAL); continue; + } *best_idx = idx; break; @@ -804,7 +807,7 @@ ff_layout_choose_best_ds_for_read(struct pnfs_layout_segment *lseg, struct nfs4_pnfs_ds *ds; ds = ff_layout_choose_valid_ds_for_read(lseg, start_idx, best_idx); - if (ds) + if (!IS_ERR(ds)) return ds; return ff_layout_choose_any_ds_for_read(lseg, start_idx, best_idx); } @@ -818,7 +821,7 @@ ff_layout_get_ds_for_read(struct nfs_pageio_descriptor *pgio, ds = ff_layout_choose_best_ds_for_read(lseg, pgio->pg_mirror_idx, best_idx); - if (ds || !pgio->pg_mirror_idx) + if (!IS_ERR(ds) || !pgio->pg_mirror_idx) return ds; return ff_layout_choose_best_ds_for_read(lseg, 0, best_idx); } @@ -868,7 +871,7 @@ retry: req->wb_nio = 0; ds = ff_layout_get_ds_for_read(pgio, &ds_idx); - if (!ds) { + if (IS_ERR(ds)) { if (!ff_layout_no_fallback_to_mds(pgio->pg_lseg)) goto out_mds; pnfs_generic_pg_cleanup(pgio); @@ -1072,11 +1075,13 @@ static void ff_layout_resend_pnfs_read(struct nfs_pgio_header *hdr) { u32 idx = hdr->pgio_mirror_idx + 1; u32 new_idx = 0; + struct nfs4_pnfs_ds *ds; - if (ff_layout_choose_any_ds_for_read(hdr->lseg, idx, &new_idx)) - ff_layout_send_layouterror(hdr->lseg); - else + ds = ff_layout_choose_any_ds_for_read(hdr->lseg, idx, &new_idx); + if (IS_ERR(ds)) pnfs_error_mark_layout_for_return(hdr->inode, hdr->lseg); + else + ff_layout_send_layouterror(hdr->lseg); pnfs_read_resend_pnfs(hdr, new_idx); } diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index 338ef77ae423..49df9debb1a6 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -716,6 +716,7 @@ nfs_setattr(struct mnt_idmap *idmap, struct dentry *dentry, { struct inode *inode = d_inode(dentry); struct nfs_fattr *fattr; + loff_t oldsize = i_size_read(inode); int error = 0; nfs_inc_stats(inode, NFSIOS_VFSSETATTR); @@ -731,7 +732,7 @@ nfs_setattr(struct mnt_idmap *idmap, struct dentry *dentry, if (error) return error; - if (attr->ia_size == i_size_read(inode)) + if (attr->ia_size == oldsize) attr->ia_valid &= ~ATTR_SIZE; } @@ -767,8 +768,10 @@ nfs_setattr(struct mnt_idmap *idmap, struct dentry *dentry, trace_nfs_setattr_enter(inode); /* Write all dirty data */ - if (S_ISREG(inode->i_mode)) + if (S_ISREG(inode->i_mode)) { + nfs_file_block_o_direct(NFS_I(inode)); nfs_sync_inode(inode); + } fattr = nfs_alloc_fattr_with_label(NFS_SERVER(inode)); if (fattr == NULL) { @@ -777,8 +780,12 @@ nfs_setattr(struct mnt_idmap *idmap, struct dentry *dentry, } error = NFS_PROTO(inode)->setattr(dentry, fattr, attr); - if (error == 0) + if (error == 0) { + if (attr->ia_valid & ATTR_SIZE) + nfs_truncate_last_folio(inode->i_mapping, oldsize, + attr->ia_size); error = nfs_refresh_inode(inode, fattr); + } nfs_free_fattr(fattr); out: trace_nfs_setattr_exit(inode, error); diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h index 74d712b58423..c0a44f389f8f 100644 --- a/fs/nfs/internal.h +++ b/fs/nfs/internal.h @@ -437,6 +437,8 @@ int nfs_file_release(struct inode *, struct file *); int nfs_lock(struct file *, int, struct file_lock *); int nfs_flock(struct file *, int, struct file_lock *); int nfs_check_flags(int); +void nfs_truncate_last_folio(struct address_space *mapping, loff_t from, + loff_t to); /* inode.c */ extern struct workqueue_struct *nfsiod_workqueue; @@ -530,6 +532,16 @@ static inline bool nfs_file_io_is_buffered(struct nfs_inode *nfsi) return test_bit(NFS_INO_ODIRECT, &nfsi->flags) == 0; } +/* Must be called with exclusively locked inode->i_rwsem */ +static inline void nfs_file_block_o_direct(struct nfs_inode *nfsi) +{ + if (test_bit(NFS_INO_ODIRECT, &nfsi->flags)) { + clear_bit(NFS_INO_ODIRECT, &nfsi->flags); + inode_dio_wait(&nfsi->vfs_inode); + } +} + + /* namespace.c */ #define NFS_PATH_CANONICAL 1 extern char *nfs_path(char **p, struct dentry *dentry, diff --git a/fs/nfs/io.c b/fs/nfs/io.c index 3388faf2acb9..d275b0a250bf 100644 --- a/fs/nfs/io.c +++ b/fs/nfs/io.c @@ -14,15 +14,6 @@ #include "internal.h" -/* Call with exclusively locked inode->i_rwsem */ -static void nfs_block_o_direct(struct nfs_inode *nfsi, struct inode *inode) -{ - if (test_bit(NFS_INO_ODIRECT, &nfsi->flags)) { - clear_bit(NFS_INO_ODIRECT, &nfsi->flags); - inode_dio_wait(inode); - } -} - /** * nfs_start_io_read - declare the file is being used for buffered reads * @inode: file inode @@ -57,7 +48,7 @@ nfs_start_io_read(struct inode *inode) err = down_write_killable(&inode->i_rwsem); if (err) return err; - nfs_block_o_direct(nfsi, inode); + nfs_file_block_o_direct(nfsi); downgrade_write(&inode->i_rwsem); return 0; @@ -90,7 +81,7 @@ nfs_start_io_write(struct inode *inode) err = down_write_killable(&inode->i_rwsem); if (!err) - nfs_block_o_direct(NFS_I(inode), inode); + nfs_file_block_o_direct(NFS_I(inode)); return err; } diff --git a/fs/nfs/localio.c b/fs/nfs/localio.c index bd5fca285899..97abf62f109d 100644 --- a/fs/nfs/localio.c +++ b/fs/nfs/localio.c @@ -180,10 +180,8 @@ static void nfs_local_probe(struct nfs_client *clp) return; } - if (nfs_client_is_local(clp)) { - /* If already enabled, disable and re-enable */ - nfs_localio_disable_client(clp); - } + if (nfs_client_is_local(clp)) + return; if (!nfs_uuid_begin(&clp->cl_uuid)) return; @@ -244,7 +242,8 @@ __nfs_local_open_fh(struct nfs_client *clp, const struct cred *cred, case -ENOMEM: case -ENXIO: case -ENOENT: - /* Revalidate localio, will disable if unsupported */ + /* Revalidate localio */ + nfs_localio_disable_client(clp); nfs_local_probe(clp); } } @@ -453,12 +452,13 @@ static void nfs_local_call_read(struct work_struct *work) nfs_local_iter_init(&iter, iocb, READ); status = filp->f_op->read_iter(&iocb->kiocb, &iter); + + revert_creds(save_cred); + if (status != -EIOCBQUEUED) { nfs_local_read_done(iocb, status); nfs_local_pgio_release(iocb); } - - revert_creds(save_cred); } static int @@ -648,14 +648,15 @@ static void nfs_local_call_write(struct work_struct *work) file_start_write(filp); status = filp->f_op->write_iter(&iocb->kiocb, &iter); file_end_write(filp); + + revert_creds(save_cred); + current->flags = old_flags; + if (status != -EIOCBQUEUED) { nfs_local_write_done(iocb, status); nfs_local_vfs_getattr(iocb); nfs_local_pgio_release(iocb); } - - revert_creds(save_cred); - current->flags = old_flags; } static int diff --git a/fs/nfs/nfs42proc.c b/fs/nfs/nfs42proc.c index 01c01f45358b..6a0b5871ba3b 100644 --- a/fs/nfs/nfs42proc.c +++ b/fs/nfs/nfs42proc.c @@ -114,6 +114,7 @@ static int nfs42_proc_fallocate(struct rpc_message *msg, struct file *filep, exception.inode = inode; exception.state = lock->open_context->state; + nfs_file_block_o_direct(NFS_I(inode)); err = nfs_sync_inode(inode); if (err) goto out; @@ -137,6 +138,7 @@ int nfs42_proc_allocate(struct file *filep, loff_t offset, loff_t len) .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_ALLOCATE], }; struct inode *inode = file_inode(filep); + loff_t oldsize = i_size_read(inode); int err; if (!nfs_server_capable(inode, NFS_CAP_ALLOCATE)) @@ -145,7 +147,11 @@ int nfs42_proc_allocate(struct file *filep, loff_t offset, loff_t len) inode_lock(inode); err = nfs42_proc_fallocate(&msg, filep, offset, len); - if (err == -EOPNOTSUPP) + + if (err == 0) + nfs_truncate_last_folio(inode->i_mapping, oldsize, + offset + len); + else if (err == -EOPNOTSUPP) NFS_SERVER(inode)->caps &= ~(NFS_CAP_ALLOCATE | NFS_CAP_ZERO_RANGE); @@ -183,6 +189,7 @@ int nfs42_proc_zero_range(struct file *filep, loff_t offset, loff_t len) .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_ZERO_RANGE], }; struct inode *inode = file_inode(filep); + loff_t oldsize = i_size_read(inode); int err; if (!nfs_server_capable(inode, NFS_CAP_ZERO_RANGE)) @@ -191,9 +198,11 @@ int nfs42_proc_zero_range(struct file *filep, loff_t offset, loff_t len) inode_lock(inode); err = nfs42_proc_fallocate(&msg, filep, offset, len); - if (err == 0) + if (err == 0) { + nfs_truncate_last_folio(inode->i_mapping, oldsize, + offset + len); truncate_pagecache_range(inode, offset, (offset + len) -1); - if (err == -EOPNOTSUPP) + } else if (err == -EOPNOTSUPP) NFS_SERVER(inode)->caps &= ~NFS_CAP_ZERO_RANGE; inode_unlock(inode); @@ -354,22 +363,27 @@ out: /** * nfs42_copy_dest_done - perform inode cache updates after clone/copy offload - * @inode: pointer to destination inode + * @file: pointer to destination file * @pos: destination offset * @len: copy length + * @oldsize: length of the file prior to clone/copy * * Punch a hole in the inode page cache, so that the NFS client will * know to retrieve new data. * Update the file size if necessary, and then mark the inode as having * invalid cached values for change attribute, ctime, mtime and space used. */ -static void nfs42_copy_dest_done(struct inode *inode, loff_t pos, loff_t len) +static void nfs42_copy_dest_done(struct file *file, loff_t pos, loff_t len, + loff_t oldsize) { + struct inode *inode = file_inode(file); + struct address_space *mapping = file->f_mapping; loff_t newsize = pos + len; loff_t end = newsize - 1; - WARN_ON_ONCE(invalidate_inode_pages2_range(inode->i_mapping, - pos >> PAGE_SHIFT, end >> PAGE_SHIFT)); + nfs_truncate_last_folio(mapping, oldsize, pos); + WARN_ON_ONCE(invalidate_inode_pages2_range(mapping, pos >> PAGE_SHIFT, + end >> PAGE_SHIFT)); spin_lock(&inode->i_lock); if (newsize > i_size_read(inode)) @@ -402,6 +416,7 @@ static ssize_t _nfs42_proc_copy(struct file *src, struct nfs_server *src_server = NFS_SERVER(src_inode); loff_t pos_src = args->src_pos; loff_t pos_dst = args->dst_pos; + loff_t oldsize_dst = i_size_read(dst_inode); size_t count = args->count; ssize_t status; @@ -430,6 +445,7 @@ static ssize_t _nfs42_proc_copy(struct file *src, return status; } + nfs_file_block_o_direct(NFS_I(dst_inode)); status = nfs_sync_inode(dst_inode); if (status) return status; @@ -475,7 +491,7 @@ static ssize_t _nfs42_proc_copy(struct file *src, goto out; } - nfs42_copy_dest_done(dst_inode, pos_dst, res->write_res.count); + nfs42_copy_dest_done(dst, pos_dst, res->write_res.count, oldsize_dst); nfs_invalidate_atime(src_inode); status = res->write_res.count; out: @@ -1242,6 +1258,7 @@ static int _nfs42_proc_clone(struct rpc_message *msg, struct file *src_f, struct nfs42_clone_res res = { .server = server, }; + loff_t oldsize_dst = i_size_read(dst_inode); int status; msg->rpc_argp = &args; @@ -1276,7 +1293,7 @@ static int _nfs42_proc_clone(struct rpc_message *msg, struct file *src_f, /* a zero-length count means clone to EOF in src */ if (count == 0 && res.dst_fattr->valid & NFS_ATTR_FATTR_SIZE) count = nfs_size_to_loff_t(res.dst_fattr->size) - dst_offset; - nfs42_copy_dest_done(dst_inode, dst_offset, count); + nfs42_copy_dest_done(dst_f, dst_offset, count, oldsize_dst); status = nfs_post_op_update_inode(dst_inode, res.dst_fattr); } diff --git a/fs/nfs/nfs4file.c b/fs/nfs/nfs4file.c index 1d6b5f4230c9..c9a0d1e420c6 100644 --- a/fs/nfs/nfs4file.c +++ b/fs/nfs/nfs4file.c @@ -278,9 +278,11 @@ static loff_t nfs42_remap_file_range(struct file *src_file, loff_t src_off, lock_two_nondirectories(src_inode, dst_inode); /* flush all pending writes on both src and dst so that server * has the latest data */ + nfs_file_block_o_direct(NFS_I(src_inode)); ret = nfs_sync_inode(src_inode); if (ret) goto out_unlock; + nfs_file_block_o_direct(NFS_I(dst_inode)); ret = nfs_sync_inode(dst_inode); if (ret) goto out_unlock; diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 7d2b67e06cc3..ce61253efd45 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -4013,8 +4013,10 @@ static int _nfs4_server_capabilities(struct nfs_server *server, struct nfs_fh *f res.attr_bitmask[2]; } memcpy(server->attr_bitmask, res.attr_bitmask, sizeof(server->attr_bitmask)); - server->caps &= ~(NFS_CAP_ACLS | NFS_CAP_HARDLINKS | - NFS_CAP_SYMLINKS| NFS_CAP_SECURITY_LABEL); + server->caps &= + ~(NFS_CAP_ACLS | NFS_CAP_HARDLINKS | NFS_CAP_SYMLINKS | + NFS_CAP_SECURITY_LABEL | NFS_CAP_FS_LOCATIONS | + NFS_CAP_OPEN_XOR | NFS_CAP_DELEGTIME); server->fattr_valid = NFS_ATTR_FATTR_V4; if (res.attr_bitmask[0] & FATTR4_WORD0_ACL && res.acl_bitmask & ACL4_SUPPORT_ALLOW_ACL) @@ -4092,7 +4094,6 @@ int nfs4_server_capabilities(struct nfs_server *server, struct nfs_fh *fhandle) }; int err; - nfs_server_set_init_caps(server); do { err = nfs4_handle_exception(server, _nfs4_server_capabilities(server, fhandle), diff --git a/fs/nfs/nfstrace.h b/fs/nfs/nfstrace.h index 96b1323318c2..627115179795 100644 --- a/fs/nfs/nfstrace.h +++ b/fs/nfs/nfstrace.h @@ -272,6 +272,7 @@ DECLARE_EVENT_CLASS(nfs_update_size_class, TP_ARGS(inode, new_size)) DEFINE_NFS_UPDATE_SIZE_EVENT(truncate); +DEFINE_NFS_UPDATE_SIZE_EVENT(truncate_folio); DEFINE_NFS_UPDATE_SIZE_EVENT(wcc); DEFINE_NFS_UPDATE_SIZE_EVENT(update); DEFINE_NFS_UPDATE_SIZE_EVENT(grow); diff --git a/fs/nfs/write.c b/fs/nfs/write.c index 8b7c04737967..647c53d1418a 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -237,59 +237,17 @@ static void nfs_mapping_set_error(struct folio *folio, int error) } /* - * nfs_page_group_search_locked - * @head - head request of page group - * @page_offset - offset into page + * nfs_page_covers_folio + * @req: struct nfs_page * - * Search page group with head @head to find a request that contains the - * page offset @page_offset. - * - * Returns a pointer to the first matching nfs request, or NULL if no - * match is found. - * - * Must be called with the page group lock held - */ -static struct nfs_page * -nfs_page_group_search_locked(struct nfs_page *head, unsigned int page_offset) -{ - struct nfs_page *req; - - req = head; - do { - if (page_offset >= req->wb_pgbase && - page_offset < (req->wb_pgbase + req->wb_bytes)) - return req; - - req = req->wb_this_page; - } while (req != head); - - return NULL; -} - -/* - * nfs_page_group_covers_page - * @head - head request of page group - * - * Return true if the page group with head @head covers the whole page, - * returns false otherwise + * Return true if the request covers the whole folio. + * Note that the caller should ensure all subrequests have been joined */ static bool nfs_page_group_covers_page(struct nfs_page *req) { unsigned int len = nfs_folio_length(nfs_page_to_folio(req)); - struct nfs_page *tmp; - unsigned int pos = 0; - - nfs_page_group_lock(req); - for (;;) { - tmp = nfs_page_group_search_locked(req->wb_head, pos); - if (!tmp) - break; - pos = tmp->wb_pgbase + tmp->wb_bytes; - } - - nfs_page_group_unlock(req); - return pos >= len; + return req->wb_pgbase == 0 && req->wb_bytes == len; } /* We can set the PG_uptodate flag if we see that a write request @@ -2045,6 +2003,7 @@ int nfs_wb_folio_cancel(struct inode *inode, struct folio *folio) * release it */ nfs_inode_remove_request(req); nfs_unlock_and_release_request(req); + folio_cancel_dirty(folio); } return ret; diff --git a/fs/nilfs2/sysfs.c b/fs/nilfs2/sysfs.c index 14868a3dd592..bc52afbfc5c7 100644 --- a/fs/nilfs2/sysfs.c +++ b/fs/nilfs2/sysfs.c @@ -1075,7 +1075,7 @@ void nilfs_sysfs_delete_device_group(struct the_nilfs *nilfs) ************************************************************************/ static ssize_t nilfs_feature_revision_show(struct kobject *kobj, - struct attribute *attr, char *buf) + struct kobj_attribute *attr, char *buf) { return sysfs_emit(buf, "%d.%d\n", NILFS_CURRENT_REV, NILFS_MINOR_REV); @@ -1087,7 +1087,7 @@ static const char features_readme_str[] = "(1) revision\n\tshow current revision of NILFS file system driver.\n"; static ssize_t nilfs_feature_README_show(struct kobject *kobj, - struct attribute *attr, + struct kobj_attribute *attr, char *buf) { return sysfs_emit(buf, features_readme_str); diff --git a/fs/nilfs2/sysfs.h b/fs/nilfs2/sysfs.h index 78a87a016928..d370cd5cce3f 100644 --- a/fs/nilfs2/sysfs.h +++ b/fs/nilfs2/sysfs.h @@ -50,16 +50,16 @@ struct nilfs_sysfs_dev_subgroups { struct completion sg_segments_kobj_unregister; }; -#define NILFS_COMMON_ATTR_STRUCT(name) \ +#define NILFS_KOBJ_ATTR_STRUCT(name) \ struct nilfs_##name##_attr { \ struct attribute attr; \ - ssize_t (*show)(struct kobject *, struct attribute *, \ + ssize_t (*show)(struct kobject *, struct kobj_attribute *, \ char *); \ - ssize_t (*store)(struct kobject *, struct attribute *, \ + ssize_t (*store)(struct kobject *, struct kobj_attribute *, \ const char *, size_t); \ } -NILFS_COMMON_ATTR_STRUCT(feature); +NILFS_KOBJ_ATTR_STRUCT(feature); #define NILFS_DEV_ATTR_STRUCT(name) \ struct nilfs_##name##_attr { \ diff --git a/fs/ocfs2/extent_map.c b/fs/ocfs2/extent_map.c index 930150ed5db1..ef147e8b3271 100644 --- a/fs/ocfs2/extent_map.c +++ b/fs/ocfs2/extent_map.c @@ -706,6 +706,8 @@ out: * it not only handles the fiemap for inlined files, but also deals * with the fast symlink, cause they have no difference for extent * mapping per se. + * + * Must be called with ip_alloc_sem semaphore held. */ static int ocfs2_fiemap_inline(struct inode *inode, struct buffer_head *di_bh, struct fiemap_extent_info *fieinfo, @@ -717,6 +719,7 @@ static int ocfs2_fiemap_inline(struct inode *inode, struct buffer_head *di_bh, u64 phys; u32 flags = FIEMAP_EXTENT_DATA_INLINE|FIEMAP_EXTENT_LAST; struct ocfs2_inode_info *oi = OCFS2_I(inode); + lockdep_assert_held_read(&oi->ip_alloc_sem); di = (struct ocfs2_dinode *)di_bh->b_data; if (ocfs2_inode_is_fast_symlink(inode)) @@ -732,8 +735,11 @@ static int ocfs2_fiemap_inline(struct inode *inode, struct buffer_head *di_bh, phys += offsetof(struct ocfs2_dinode, id2.i_data.id_data); + /* Release the ip_alloc_sem to prevent deadlock on page fault */ + up_read(&OCFS2_I(inode)->ip_alloc_sem); ret = fiemap_fill_next_extent(fieinfo, 0, phys, id_count, flags); + down_read(&OCFS2_I(inode)->ip_alloc_sem); if (ret < 0) return ret; } @@ -802,9 +808,11 @@ int ocfs2_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, len_bytes = (u64)le16_to_cpu(rec.e_leaf_clusters) << osb->s_clustersize_bits; phys_bytes = le64_to_cpu(rec.e_blkno) << osb->sb->s_blocksize_bits; virt_bytes = (u64)le32_to_cpu(rec.e_cpos) << osb->s_clustersize_bits; - + /* Release the ip_alloc_sem to prevent deadlock on page fault */ + up_read(&OCFS2_I(inode)->ip_alloc_sem); ret = fiemap_fill_next_extent(fieinfo, virt_bytes, phys_bytes, len_bytes, fe_flags); + down_read(&OCFS2_I(inode)->ip_alloc_sem); if (ret) break; diff --git a/fs/ocfs2/inode.c b/fs/ocfs2/inode.c index 14bf440ea4df..6c4f78f473fb 100644 --- a/fs/ocfs2/inode.c +++ b/fs/ocfs2/inode.c @@ -1281,6 +1281,9 @@ static void ocfs2_clear_inode(struct inode *inode) * the journal is flushed before journal shutdown. Thus it is safe to * have inodes get cleaned up after journal shutdown. */ + if (!osb->journal) + return; + jbd2_journal_release_jbd_inode(osb->journal->j_journal, &oi->ip_jinode); } diff --git a/fs/proc/generic.c b/fs/proc/generic.c index 76e800e38c8f..176281112273 100644 --- a/fs/proc/generic.c +++ b/fs/proc/generic.c @@ -367,6 +367,25 @@ static const struct inode_operations proc_dir_inode_operations = { .setattr = proc_notify_change, }; +static void pde_set_flags(struct proc_dir_entry *pde) +{ + const struct proc_ops *proc_ops = pde->proc_ops; + + if (!proc_ops) + return; + + if (proc_ops->proc_flags & PROC_ENTRY_PERMANENT) + pde->flags |= PROC_ENTRY_PERMANENT; + if (proc_ops->proc_read_iter) + pde->flags |= PROC_ENTRY_proc_read_iter; +#ifdef CONFIG_COMPAT + if (proc_ops->proc_compat_ioctl) + pde->flags |= PROC_ENTRY_proc_compat_ioctl; +#endif + if (proc_ops->proc_lseek) + pde->flags |= PROC_ENTRY_proc_lseek; +} + /* returns the registered entry, or frees dp and returns NULL on failure */ struct proc_dir_entry *proc_register(struct proc_dir_entry *dir, struct proc_dir_entry *dp) @@ -374,6 +393,9 @@ struct proc_dir_entry *proc_register(struct proc_dir_entry *dir, if (proc_alloc_inum(&dp->low_ino)) goto out_free_entry; + if (!S_ISDIR(dp->mode)) + pde_set_flags(dp); + write_lock(&proc_subdir_lock); dp->parent = dir; if (pde_subdir_insert(dir, dp) == false) { @@ -561,20 +583,6 @@ struct proc_dir_entry *proc_create_reg(const char *name, umode_t mode, return p; } -static void pde_set_flags(struct proc_dir_entry *pde) -{ - if (pde->proc_ops->proc_flags & PROC_ENTRY_PERMANENT) - pde->flags |= PROC_ENTRY_PERMANENT; - if (pde->proc_ops->proc_read_iter) - pde->flags |= PROC_ENTRY_proc_read_iter; -#ifdef CONFIG_COMPAT - if (pde->proc_ops->proc_compat_ioctl) - pde->flags |= PROC_ENTRY_proc_compat_ioctl; -#endif - if (pde->proc_ops->proc_lseek) - pde->flags |= PROC_ENTRY_proc_lseek; -} - struct proc_dir_entry *proc_create_data(const char *name, umode_t mode, struct proc_dir_entry *parent, const struct proc_ops *proc_ops, void *data) @@ -585,7 +593,6 @@ struct proc_dir_entry *proc_create_data(const char *name, umode_t mode, if (!p) return NULL; p->proc_ops = proc_ops; - pde_set_flags(p); return proc_register(parent, p); } EXPORT_SYMBOL(proc_create_data); @@ -636,7 +643,6 @@ struct proc_dir_entry *proc_create_seq_private(const char *name, umode_t mode, p->proc_ops = &proc_seq_ops; p->seq_ops = ops; p->state_size = state_size; - pde_set_flags(p); return proc_register(parent, p); } EXPORT_SYMBOL(proc_create_seq_private); @@ -667,7 +673,6 @@ struct proc_dir_entry *proc_create_single_data(const char *name, umode_t mode, return NULL; p->proc_ops = &proc_single_ops; p->single_show = show; - pde_set_flags(p); return proc_register(parent, p); } EXPORT_SYMBOL(proc_create_single_data); diff --git a/fs/resctrl/ctrlmondata.c b/fs/resctrl/ctrlmondata.c index d98e0d2de09f..3c39cfacb251 100644 --- a/fs/resctrl/ctrlmondata.c +++ b/fs/resctrl/ctrlmondata.c @@ -625,11 +625,11 @@ int rdtgroup_mondata_show(struct seq_file *m, void *arg) */ list_for_each_entry(d, &r->mon_domains, hdr.list) { if (d->ci_id == domid) { - rr.ci_id = d->ci_id; cpu = cpumask_any(&d->hdr.cpu_mask); ci = get_cpu_cacheinfo_level(cpu, RESCTRL_L3_CACHE); if (!ci) continue; + rr.ci = ci; mon_event_read(&rr, r, NULL, rdtgrp, &ci->shared_cpu_map, evtid, false); goto checkresult; diff --git a/fs/resctrl/internal.h b/fs/resctrl/internal.h index 0a1eedba2b03..9a8cf6f11151 100644 --- a/fs/resctrl/internal.h +++ b/fs/resctrl/internal.h @@ -98,7 +98,7 @@ struct mon_data { * domains in @r sharing L3 @ci.id * @evtid: Which monitor event to read. * @first: Initialize MBM counter when true. - * @ci_id: Cacheinfo id for L3. Only set when @d is NULL. Used when summing domains. + * @ci: Cacheinfo for L3. Only set when @d is NULL. Used when summing domains. * @err: Error encountered when reading counter. * @val: Returned value of event counter. If @rgrp is a parent resource group, * @val includes the sum of event counts from its child resource groups. @@ -112,7 +112,7 @@ struct rmid_read { struct rdt_mon_domain *d; enum resctrl_event_id evtid; bool first; - unsigned int ci_id; + struct cacheinfo *ci; int err; u64 val; void *arch_mon_ctx; diff --git a/fs/resctrl/monitor.c b/fs/resctrl/monitor.c index f5637855c3ac..7326c28a7908 100644 --- a/fs/resctrl/monitor.c +++ b/fs/resctrl/monitor.c @@ -361,7 +361,6 @@ static int __mon_event_count(u32 closid, u32 rmid, struct rmid_read *rr) { int cpu = smp_processor_id(); struct rdt_mon_domain *d; - struct cacheinfo *ci; struct mbm_state *m; int err, ret; u64 tval = 0; @@ -389,8 +388,7 @@ static int __mon_event_count(u32 closid, u32 rmid, struct rmid_read *rr) } /* Summing domains that share a cache, must be on a CPU for that cache. */ - ci = get_cpu_cacheinfo_level(cpu, RESCTRL_L3_CACHE); - if (!ci || ci->id != rr->ci_id) + if (!cpumask_test_cpu(cpu, &rr->ci->shared_cpu_map)) return -EINVAL; /* @@ -402,7 +400,7 @@ static int __mon_event_count(u32 closid, u32 rmid, struct rmid_read *rr) */ ret = -EINVAL; list_for_each_entry(d, &rr->r->mon_domains, hdr.list) { - if (d->ci_id != rr->ci_id) + if (d->ci_id != rr->ci->id) continue; err = resctrl_arch_rmid_read(rr->r, d, closid, rmid, rr->evtid, &tval, rr->arch_mon_ctx); diff --git a/fs/smb/client/cifs_debug.c b/fs/smb/client/cifs_debug.c index beb4f18f05ef..2337cf795db3 100644 --- a/fs/smb/client/cifs_debug.c +++ b/fs/smb/client/cifs_debug.c @@ -304,6 +304,8 @@ static int cifs_debug_dirs_proc_show(struct seq_file *m, void *v) list_for_each(tmp1, &ses->tcon_list) { tcon = list_entry(tmp1, struct cifs_tcon, tcon_list); cfids = tcon->cfids; + if (!cfids) + continue; spin_lock(&cfids->cfid_list_lock); /* check lock ordering */ seq_printf(m, "Num entries: %d\n", cfids->num_entries); list_for_each_entry(cfid, &cfids->entries, entry) { @@ -319,8 +321,6 @@ static int cifs_debug_dirs_proc_show(struct seq_file *m, void *v) seq_printf(m, "\n"); } spin_unlock(&cfids->cfid_list_lock); - - } } } @@ -347,6 +347,22 @@ static __always_inline const char *compression_alg_str(__le16 alg) } } +static __always_inline const char *cipher_alg_str(__le16 cipher) +{ + switch (cipher) { + case SMB2_ENCRYPTION_AES128_CCM: + return "AES128-CCM"; + case SMB2_ENCRYPTION_AES128_GCM: + return "AES128-GCM"; + case SMB2_ENCRYPTION_AES256_CCM: + return "AES256-CCM"; + case SMB2_ENCRYPTION_AES256_GCM: + return "AES256-GCM"; + default: + return "UNKNOWN"; + } +} + static int cifs_debug_data_proc_show(struct seq_file *m, void *v) { struct mid_q_entry *mid_entry; @@ -539,6 +555,11 @@ skip_rdma: else seq_puts(m, "disabled (not supported by this server)"); + /* Show negotiated encryption cipher, even if not required */ + seq_puts(m, "\nEncryption: "); + if (server->cipher_type) + seq_printf(m, "Negotiated cipher (%s)", cipher_alg_str(server->cipher_type)); + seq_printf(m, "\n\n\tSessions: "); i = 0; list_for_each_entry(ses, &server->smb_ses_list, smb_ses_list) { @@ -576,12 +597,8 @@ skip_rdma: /* dump session id helpful for use with network trace */ seq_printf(m, " SessionId: 0x%llx", ses->Suid); - if (ses->session_flags & SMB2_SESSION_FLAG_ENCRYPT_DATA) { + if (ses->session_flags & SMB2_SESSION_FLAG_ENCRYPT_DATA) seq_puts(m, " encrypted"); - /* can help in debugging to show encryption type */ - if (server->cipher_type == SMB2_ENCRYPTION_AES256_GCM) - seq_puts(m, "(gcm256)"); - } if (ses->sign) seq_puts(m, " signed"); diff --git a/fs/smb/client/cifs_unicode.c b/fs/smb/client/cifs_unicode.c index 4cc6e0896fad..f8659d36793f 100644 --- a/fs/smb/client/cifs_unicode.c +++ b/fs/smb/client/cifs_unicode.c @@ -629,6 +629,9 @@ cifs_strndup_to_utf16(const char *src, const int maxlen, int *utf16_len, int len; __le16 *dst; + if (!src) + return NULL; + len = cifs_local_to_utf16_bytes(src, maxlen, cp); len += 2; /* NULL */ dst = kmalloc(len, GFP_KERNEL); diff --git a/fs/smb/client/cifsfs.c b/fs/smb/client/cifsfs.c index 3bd85ab2deb1..e1848276bab4 100644 --- a/fs/smb/client/cifsfs.c +++ b/fs/smb/client/cifsfs.c @@ -1358,6 +1358,20 @@ static loff_t cifs_remap_file_range(struct file *src_file, loff_t off, truncate_setsize(target_inode, new_size); fscache_resize_cookie(cifs_inode_cookie(target_inode), new_size); + } else if (rc == -EOPNOTSUPP) { + /* + * copy_file_range syscall man page indicates EINVAL + * is returned e.g when "fd_in and fd_out refer to the + * same file and the source and target ranges overlap." + * Test generic/157 was what showed these cases where + * we need to remap EOPNOTSUPP to EINVAL + */ + if (off >= src_inode->i_size) { + rc = -EINVAL; + } else if (src_inode == target_inode) { + if (off + len > destoff) + rc = -EINVAL; + } } if (rc == 0 && new_size > target_cifsi->netfs.zero_point) target_cifsi->netfs.zero_point = new_size; diff --git a/fs/smb/client/cifsglob.h b/fs/smb/client/cifsglob.h index 1e64a4fb6af0..0fae95cf81c4 100644 --- a/fs/smb/client/cifsglob.h +++ b/fs/smb/client/cifsglob.h @@ -87,7 +87,7 @@ #define SMB_INTERFACE_POLL_INTERVAL 600 /* maximum number of PDUs in one compound */ -#define MAX_COMPOUND 7 +#define MAX_COMPOUND 10 /* * Default number of credits to keep available for SMB3. @@ -1882,9 +1882,12 @@ static inline bool is_replayable_error(int error) /* cifs_get_writable_file() flags */ -#define FIND_WR_ANY 0 -#define FIND_WR_FSUID_ONLY 1 -#define FIND_WR_WITH_DELETE 2 +enum cifs_writable_file_flags { + FIND_WR_ANY = 0U, + FIND_WR_FSUID_ONLY = (1U << 0), + FIND_WR_WITH_DELETE = (1U << 1), + FIND_WR_NO_PENDING_DELETE = (1U << 2), +}; #define MID_FREE 0 #define MID_REQUEST_ALLOCATED 1 @@ -2343,6 +2346,8 @@ struct smb2_compound_vars { struct kvec qi_iov; struct kvec io_iov[SMB2_IOCTL_IOV_SIZE]; struct kvec si_iov[SMB2_SET_INFO_IOV_SIZE]; + struct kvec unlink_iov[SMB2_SET_INFO_IOV_SIZE]; + struct kvec rename_iov[SMB2_SET_INFO_IOV_SIZE]; struct kvec close_iov; struct smb2_file_rename_info_hdr rename_info; struct smb2_file_link_info_hdr link_info; diff --git a/fs/smb/client/cifsproto.h b/fs/smb/client/cifsproto.h index c34c533b2efa..e8fba98690ce 100644 --- a/fs/smb/client/cifsproto.h +++ b/fs/smb/client/cifsproto.h @@ -312,8 +312,8 @@ extern void cifs_close_deferred_file(struct cifsInodeInfo *cifs_inode); extern void cifs_close_all_deferred_files(struct cifs_tcon *cifs_tcon); -extern void cifs_close_deferred_file_under_dentry(struct cifs_tcon *cifs_tcon, - const char *path); +void cifs_close_deferred_file_under_dentry(struct cifs_tcon *cifs_tcon, + struct dentry *dentry); extern void cifs_mark_open_handles_for_deleted_file(struct inode *inode, const char *path); diff --git a/fs/smb/client/file.c b/fs/smb/client/file.c index 186e061068be..cb907e18cc35 100644 --- a/fs/smb/client/file.c +++ b/fs/smb/client/file.c @@ -998,7 +998,10 @@ int cifs_open(struct inode *inode, struct file *file) /* Get the cached handle as SMB2 close is deferred */ if (OPEN_FMODE(file->f_flags) & FMODE_WRITE) { - rc = cifs_get_writable_path(tcon, full_path, FIND_WR_FSUID_ONLY, &cfile); + rc = cifs_get_writable_path(tcon, full_path, + FIND_WR_FSUID_ONLY | + FIND_WR_NO_PENDING_DELETE, + &cfile); } else { rc = cifs_get_readable_path(tcon, full_path, &cfile); } @@ -2530,6 +2533,9 @@ refind_writable: continue; if (with_delete && !(open_file->fid.access & DELETE)) continue; + if ((flags & FIND_WR_NO_PENDING_DELETE) && + open_file->status_file_deleted) + continue; if (OPEN_FMODE(open_file->f_flags) & FMODE_WRITE) { if (!open_file->invalidHandle) { /* found a good writable file */ @@ -2647,6 +2653,16 @@ cifs_get_readable_path(struct cifs_tcon *tcon, const char *name, spin_unlock(&tcon->open_file_lock); free_dentry_path(page); *ret_file = find_readable_file(cinode, 0); + if (*ret_file) { + spin_lock(&cinode->open_file_lock); + if ((*ret_file)->status_file_deleted) { + spin_unlock(&cinode->open_file_lock); + cifsFileInfo_put(*ret_file); + *ret_file = NULL; + } else { + spin_unlock(&cinode->open_file_lock); + } + } return *ret_file ? 0 : -ENOENT; } diff --git a/fs/smb/client/inode.c b/fs/smb/client/inode.c index fe453a4b3dc8..0f0d2dae6283 100644 --- a/fs/smb/client/inode.c +++ b/fs/smb/client/inode.c @@ -1931,7 +1931,7 @@ cifs_drop_nlink(struct inode *inode) * but will return the EACCES to the caller. Note that the VFS does not call * unlink on negative dentries currently. */ -int cifs_unlink(struct inode *dir, struct dentry *dentry) +static int __cifs_unlink(struct inode *dir, struct dentry *dentry, bool sillyrename) { int rc = 0; unsigned int xid; @@ -1984,7 +1984,7 @@ int cifs_unlink(struct inode *dir, struct dentry *dentry) } netfs_wait_for_outstanding_io(inode); - cifs_close_deferred_file_under_dentry(tcon, full_path); + cifs_close_deferred_file_under_dentry(tcon, dentry); #ifdef CONFIG_CIFS_ALLOW_INSECURE_LEGACY if (cap_unix(tcon->ses) && (CIFS_UNIX_POSIX_PATH_OPS_CAP & le64_to_cpu(tcon->fsUnixInfo.Capability))) { @@ -2003,7 +2003,24 @@ retry_std_delete: goto psx_del_no_retry; } - rc = server->ops->unlink(xid, tcon, full_path, cifs_sb, dentry); + /* For SMB2+, if the file is open, we always perform a silly rename. + * + * We check for d_count() right after calling + * cifs_close_deferred_file_under_dentry() to make sure that the + * dentry's refcount gets dropped in case the file had any deferred + * close. + */ + if (!sillyrename && server->vals->protocol_id > SMB10_PROT_ID) { + spin_lock(&dentry->d_lock); + if (d_count(dentry) > 1) + sillyrename = true; + spin_unlock(&dentry->d_lock); + } + + if (sillyrename) + rc = -EBUSY; + else + rc = server->ops->unlink(xid, tcon, full_path, cifs_sb, dentry); psx_del_no_retry: if (!rc) { @@ -2071,6 +2088,11 @@ unlink_out: return rc; } +int cifs_unlink(struct inode *dir, struct dentry *dentry) +{ + return __cifs_unlink(dir, dentry, false); +} + static int cifs_mkdir_qinfo(struct inode *parent, struct dentry *dentry, umode_t mode, const char *full_path, struct cifs_sb_info *cifs_sb, @@ -2358,14 +2380,16 @@ int cifs_rmdir(struct inode *inode, struct dentry *direntry) rc = server->ops->rmdir(xid, tcon, full_path, cifs_sb); cifs_put_tlink(tlink); + cifsInode = CIFS_I(d_inode(direntry)); + if (!rc) { + set_bit(CIFS_INO_DELETE_PENDING, &cifsInode->flags); spin_lock(&d_inode(direntry)->i_lock); i_size_write(d_inode(direntry), 0); clear_nlink(d_inode(direntry)); spin_unlock(&d_inode(direntry)->i_lock); } - cifsInode = CIFS_I(d_inode(direntry)); /* force revalidate to go get info when needed */ cifsInode->time = 0; @@ -2458,8 +2482,11 @@ cifs_do_rename(const unsigned int xid, struct dentry *from_dentry, } #endif /* CONFIG_CIFS_ALLOW_INSECURE_LEGACY */ do_rename_exit: - if (rc == 0) + if (rc == 0) { d_move(from_dentry, to_dentry); + /* Force a new lookup */ + d_drop(from_dentry); + } cifs_put_tlink(tlink); return rc; } @@ -2470,6 +2497,7 @@ cifs_rename2(struct mnt_idmap *idmap, struct inode *source_dir, struct dentry *target_dentry, unsigned int flags) { const char *from_name, *to_name; + struct TCP_Server_Info *server; void *page1, *page2; struct cifs_sb_info *cifs_sb; struct tcon_link *tlink; @@ -2505,6 +2533,7 @@ cifs_rename2(struct mnt_idmap *idmap, struct inode *source_dir, if (IS_ERR(tlink)) return PTR_ERR(tlink); tcon = tlink_tcon(tlink); + server = tcon->ses->server; page1 = alloc_dentry_path(); page2 = alloc_dentry_path(); @@ -2522,10 +2551,10 @@ cifs_rename2(struct mnt_idmap *idmap, struct inode *source_dir, goto cifs_rename_exit; } - cifs_close_deferred_file_under_dentry(tcon, from_name); + cifs_close_deferred_file_under_dentry(tcon, source_dentry); if (d_inode(target_dentry) != NULL) { netfs_wait_for_outstanding_io(d_inode(target_dentry)); - cifs_close_deferred_file_under_dentry(tcon, to_name); + cifs_close_deferred_file_under_dentry(tcon, target_dentry); } rc = cifs_do_rename(xid, source_dentry, from_name, target_dentry, @@ -2591,19 +2620,53 @@ cifs_rename2(struct mnt_idmap *idmap, struct inode *source_dir, unlink_target: #endif /* CONFIG_CIFS_ALLOW_INSECURE_LEGACY */ - - /* Try unlinking the target dentry if it's not negative */ - if (d_really_is_positive(target_dentry) && (rc == -EACCES || rc == -EEXIST)) { - if (d_is_dir(target_dentry)) - tmprc = cifs_rmdir(target_dir, target_dentry); - else - tmprc = cifs_unlink(target_dir, target_dentry); - if (tmprc) - goto cifs_rename_exit; - rc = cifs_do_rename(xid, source_dentry, from_name, - target_dentry, to_name); - if (!rc) - rehash = false; + if (d_really_is_positive(target_dentry)) { + if (!rc) { + struct inode *inode = d_inode(target_dentry); + /* + * Samba and ksmbd servers allow renaming a target + * directory that is open, so make sure to update + * ->i_nlink and then mark it as delete pending. + */ + if (S_ISDIR(inode->i_mode)) { + drop_cached_dir_by_name(xid, tcon, to_name, cifs_sb); + spin_lock(&inode->i_lock); + i_size_write(inode, 0); + clear_nlink(inode); + spin_unlock(&inode->i_lock); + set_bit(CIFS_INO_DELETE_PENDING, &CIFS_I(inode)->flags); + CIFS_I(inode)->time = 0; /* force reval */ + inode_set_ctime_current(inode); + inode_set_mtime_to_ts(inode, inode_set_ctime_current(inode)); + } + } else if (rc == -EACCES || rc == -EEXIST) { + /* + * Rename failed, possibly due to a busy target. + * Retry it by unliking the target first. + */ + if (d_is_dir(target_dentry)) { + tmprc = cifs_rmdir(target_dir, target_dentry); + } else { + tmprc = __cifs_unlink(target_dir, target_dentry, + server->vals->protocol_id > SMB10_PROT_ID); + } + if (tmprc) { + /* + * Some servers will return STATUS_ACCESS_DENIED + * or STATUS_DIRECTORY_NOT_EMPTY when failing to + * rename a non-empty directory. Make sure to + * propagate the appropriate error back to + * userspace. + */ + if (tmprc == -EEXIST || tmprc == -ENOTEMPTY) + rc = tmprc; + goto cifs_rename_exit; + } + rc = cifs_do_rename(xid, source_dentry, from_name, + target_dentry, to_name); + if (!rc) + rehash = false; + } } /* force revalidate to go get info when needed */ @@ -2629,6 +2692,8 @@ cifs_dentry_needs_reval(struct dentry *dentry) struct cifs_tcon *tcon = cifs_sb_master_tcon(cifs_sb); struct cached_fid *cfid = NULL; + if (test_bit(CIFS_INO_DELETE_PENDING, &cifs_i->flags)) + return false; if (cifs_i->time == 0) return true; diff --git a/fs/smb/client/misc.c b/fs/smb/client/misc.c index da23cc12a52c..dda6dece802a 100644 --- a/fs/smb/client/misc.c +++ b/fs/smb/client/misc.c @@ -832,33 +832,28 @@ cifs_close_all_deferred_files(struct cifs_tcon *tcon) kfree(tmp_list); } } -void -cifs_close_deferred_file_under_dentry(struct cifs_tcon *tcon, const char *path) + +void cifs_close_deferred_file_under_dentry(struct cifs_tcon *tcon, + struct dentry *dentry) { - struct cifsFileInfo *cfile; struct file_list *tmp_list, *tmp_next_list; - void *page; - const char *full_path; + struct cifsFileInfo *cfile; LIST_HEAD(file_head); - page = alloc_dentry_path(); spin_lock(&tcon->open_file_lock); list_for_each_entry(cfile, &tcon->openFileList, tlist) { - full_path = build_path_from_dentry(cfile->dentry, page); - if (strstr(full_path, path)) { - if (delayed_work_pending(&cfile->deferred)) { - if (cancel_delayed_work(&cfile->deferred)) { - spin_lock(&CIFS_I(d_inode(cfile->dentry))->deferred_lock); - cifs_del_deferred_close(cfile); - spin_unlock(&CIFS_I(d_inode(cfile->dentry))->deferred_lock); + if ((cfile->dentry == dentry) && + delayed_work_pending(&cfile->deferred) && + cancel_delayed_work(&cfile->deferred)) { + spin_lock(&CIFS_I(d_inode(cfile->dentry))->deferred_lock); + cifs_del_deferred_close(cfile); + spin_unlock(&CIFS_I(d_inode(cfile->dentry))->deferred_lock); - tmp_list = kmalloc(sizeof(struct file_list), GFP_ATOMIC); - if (tmp_list == NULL) - break; - tmp_list->cfile = cfile; - list_add_tail(&tmp_list->list, &file_head); - } - } + tmp_list = kmalloc(sizeof(struct file_list), GFP_ATOMIC); + if (tmp_list == NULL) + break; + tmp_list->cfile = cfile; + list_add_tail(&tmp_list->list, &file_head); } } spin_unlock(&tcon->open_file_lock); @@ -868,7 +863,6 @@ cifs_close_deferred_file_under_dentry(struct cifs_tcon *tcon, const char *path) list_del(&tmp_list->list); kfree(tmp_list); } - free_dentry_path(page); } /* diff --git a/fs/smb/client/reparse.c b/fs/smb/client/reparse.c index 7869cec58f52..10c84c095fe7 100644 --- a/fs/smb/client/reparse.c +++ b/fs/smb/client/reparse.c @@ -278,7 +278,7 @@ static int detect_directory_symlink_target(struct cifs_sb_info *cifs_sb, } /* - * For absolute symlinks it is not possible to determinate + * For absolute symlinks it is not possible to determine * if it should point to directory or file. */ if (symname[0] == '/') { diff --git a/fs/smb/client/smb1ops.c b/fs/smb/client/smb1ops.c index 893a1ea8c000..a02d41d1ce4a 100644 --- a/fs/smb/client/smb1ops.c +++ b/fs/smb/client/smb1ops.c @@ -1005,7 +1005,7 @@ smb_set_file_info(struct inode *inode, const char *full_path, rc = -EOPNOTSUPP; } - /* Fallback to SMB_COM_SETATTR command when absolutelty needed. */ + /* Fallback to SMB_COM_SETATTR command when absolutely needed. */ if (rc == -EOPNOTSUPP) { cifs_dbg(FYI, "calling SetInformation since SetPathInfo for attrs/times not supported by this server\n"); rc = SMBSetInformation(xid, tcon, full_path, @@ -1039,7 +1039,7 @@ set_via_filehandle: cifsFileInfo_put(open_file); /* - * Setting the read-only bit is not honered on non-NT servers when done + * Setting the read-only bit is not honored on non-NT servers when done * via open-semantics. So for setting it, use SMB_COM_SETATTR command. * This command works only after the file is closed, so use it only when * operation was called without the filehandle. diff --git a/fs/smb/client/smb2glob.h b/fs/smb/client/smb2glob.h index 224495322a05..e56e4d402f13 100644 --- a/fs/smb/client/smb2glob.h +++ b/fs/smb/client/smb2glob.h @@ -30,10 +30,9 @@ enum smb2_compound_ops { SMB2_OP_QUERY_DIR, SMB2_OP_MKDIR, SMB2_OP_RENAME, - SMB2_OP_DELETE, SMB2_OP_HARDLINK, SMB2_OP_SET_EOF, - SMB2_OP_RMDIR, + SMB2_OP_UNLINK, SMB2_OP_POSIX_QUERY_INFO, SMB2_OP_SET_REPARSE, SMB2_OP_GET_REPARSE, diff --git a/fs/smb/client/smb2inode.c b/fs/smb/client/smb2inode.c index 2a0316c514e4..7cadc8ca4f55 100644 --- a/fs/smb/client/smb2inode.c +++ b/fs/smb/client/smb2inode.c @@ -207,8 +207,10 @@ replay_again: server = cifs_pick_channel(ses); vars = kzalloc(sizeof(*vars), GFP_ATOMIC); - if (vars == NULL) - return -ENOMEM; + if (vars == NULL) { + rc = -ENOMEM; + goto out; + } rqst = &vars->rqst[0]; rsp_iov = &vars->rsp_iov[0]; @@ -344,9 +346,6 @@ replay_again: trace_smb3_posix_query_info_compound_enter(xid, tcon->tid, ses->Suid, full_path); break; - case SMB2_OP_DELETE: - trace_smb3_delete_enter(xid, tcon->tid, ses->Suid, full_path); - break; case SMB2_OP_MKDIR: /* * Directories are created through parameters in the @@ -354,23 +353,40 @@ replay_again: */ trace_smb3_mkdir_enter(xid, tcon->tid, ses->Suid, full_path); break; - case SMB2_OP_RMDIR: - rqst[num_rqst].rq_iov = &vars->si_iov[0]; + case SMB2_OP_UNLINK: + rqst[num_rqst].rq_iov = vars->unlink_iov; rqst[num_rqst].rq_nvec = 1; size[0] = 1; /* sizeof __u8 See MS-FSCC section 2.4.11 */ data[0] = &delete_pending[0]; - rc = SMB2_set_info_init(tcon, server, - &rqst[num_rqst], COMPOUND_FID, - COMPOUND_FID, current->tgid, - FILE_DISPOSITION_INFORMATION, - SMB2_O_INFO_FILE, 0, data, size); - if (rc) + if (cfile) { + rc = SMB2_set_info_init(tcon, server, + &rqst[num_rqst], + cfile->fid.persistent_fid, + cfile->fid.volatile_fid, + current->tgid, + FILE_DISPOSITION_INFORMATION, + SMB2_O_INFO_FILE, 0, + data, size); + } else { + rc = SMB2_set_info_init(tcon, server, + &rqst[num_rqst], + COMPOUND_FID, + COMPOUND_FID, + current->tgid, + FILE_DISPOSITION_INFORMATION, + SMB2_O_INFO_FILE, 0, + data, size); + } + if (!rc && (!cfile || num_rqst > 1)) { + smb2_set_next_command(tcon, &rqst[num_rqst]); + smb2_set_related(&rqst[num_rqst]); + } else if (rc) { goto finished; - smb2_set_next_command(tcon, &rqst[num_rqst]); - smb2_set_related(&rqst[num_rqst++]); - trace_smb3_rmdir_enter(xid, tcon->tid, ses->Suid, full_path); + } + num_rqst++; + trace_smb3_unlink_enter(xid, tcon->tid, ses->Suid, full_path); break; case SMB2_OP_SET_EOF: rqst[num_rqst].rq_iov = &vars->si_iov[0]; @@ -440,7 +456,7 @@ replay_again: ses->Suid, full_path); break; case SMB2_OP_RENAME: - rqst[num_rqst].rq_iov = &vars->si_iov[0]; + rqst[num_rqst].rq_iov = vars->rename_iov; rqst[num_rqst].rq_nvec = 2; len = in_iov[i].iov_len; @@ -730,19 +746,6 @@ finished: trace_smb3_posix_query_info_compound_done(xid, tcon->tid, ses->Suid); break; - case SMB2_OP_DELETE: - if (rc) - trace_smb3_delete_err(xid, tcon->tid, ses->Suid, rc); - else { - /* - * If dentry (hence, inode) is NULL, lease break is going to - * take care of degrading leases on handles for deleted files. - */ - if (inode) - cifs_mark_open_handles_for_deleted_file(inode, full_path); - trace_smb3_delete_done(xid, tcon->tid, ses->Suid); - } - break; case SMB2_OP_MKDIR: if (rc) trace_smb3_mkdir_err(xid, tcon->tid, ses->Suid, rc); @@ -763,11 +766,11 @@ finished: trace_smb3_rename_done(xid, tcon->tid, ses->Suid); SMB2_set_info_free(&rqst[num_rqst++]); break; - case SMB2_OP_RMDIR: - if (rc) - trace_smb3_rmdir_err(xid, tcon->tid, ses->Suid, rc); + case SMB2_OP_UNLINK: + if (!rc) + trace_smb3_unlink_done(xid, tcon->tid, ses->Suid); else - trace_smb3_rmdir_done(xid, tcon->tid, ses->Suid); + trace_smb3_unlink_err(xid, tcon->tid, ses->Suid, rc); SMB2_set_info_free(&rqst[num_rqst++]); break; case SMB2_OP_SET_EOF: @@ -864,6 +867,7 @@ finished: smb2_should_replay(tcon, &retries, &cur_sleep)) goto replay_again; +out: if (cfile) cifsFileInfo_put(cfile); @@ -1163,7 +1167,7 @@ smb2_rmdir(const unsigned int xid, struct cifs_tcon *tcon, const char *name, FILE_OPEN, CREATE_NOT_FILE, ACL_NO_MODE); return smb2_compound_op(xid, tcon, cifs_sb, name, &oparms, NULL, - &(int){SMB2_OP_RMDIR}, 1, + &(int){SMB2_OP_UNLINK}, 1, NULL, NULL, NULL, NULL); } @@ -1172,20 +1176,29 @@ smb2_unlink(const unsigned int xid, struct cifs_tcon *tcon, const char *name, struct cifs_sb_info *cifs_sb, struct dentry *dentry) { struct cifs_open_parms oparms; + struct inode *inode = NULL; + int rc; - oparms = CIFS_OPARMS(cifs_sb, tcon, name, - DELETE, FILE_OPEN, - CREATE_DELETE_ON_CLOSE | OPEN_REPARSE_POINT, - ACL_NO_MODE); - int rc = smb2_compound_op(xid, tcon, cifs_sb, name, &oparms, - NULL, &(int){SMB2_OP_DELETE}, 1, - NULL, NULL, NULL, dentry); + if (dentry) + inode = d_inode(dentry); + + oparms = CIFS_OPARMS(cifs_sb, tcon, name, DELETE, + FILE_OPEN, OPEN_REPARSE_POINT, ACL_NO_MODE); + rc = smb2_compound_op(xid, tcon, cifs_sb, name, &oparms, + NULL, &(int){SMB2_OP_UNLINK}, + 1, NULL, NULL, NULL, dentry); if (rc == -EINVAL) { cifs_dbg(FYI, "invalid lease key, resending request without lease"); rc = smb2_compound_op(xid, tcon, cifs_sb, name, &oparms, - NULL, &(int){SMB2_OP_DELETE}, 1, - NULL, NULL, NULL, NULL); + NULL, &(int){SMB2_OP_UNLINK}, + 1, NULL, NULL, NULL, NULL); } + /* + * If dentry (hence, inode) is NULL, lease break is going to + * take care of degrading leases on handles for deleted files. + */ + if (!rc && inode) + cifs_mark_open_handles_for_deleted_file(inode, name); return rc; } @@ -1438,3 +1451,113 @@ out: cifs_free_open_info(&data); return rc; } + +static inline __le16 *utf16_smb2_path(struct cifs_sb_info *cifs_sb, + const char *name, size_t namelen) +{ + int len; + + if (*name == '\\' || + (cifs_sb_master_tlink(cifs_sb) && + cifs_sb_master_tcon(cifs_sb)->posix_extensions && *name == '/')) + name++; + return cifs_strndup_to_utf16(name, namelen, &len, + cifs_sb->local_nls, + cifs_remap(cifs_sb)); +} + +int smb2_rename_pending_delete(const char *full_path, + struct dentry *dentry, + const unsigned int xid) +{ + struct cifs_sb_info *cifs_sb = CIFS_SB(d_inode(dentry)->i_sb); + struct cifsInodeInfo *cinode = CIFS_I(d_inode(dentry)); + __le16 *utf16_path __free(kfree) = NULL; + __u32 co = file_create_options(dentry); + int cmds[] = { + SMB2_OP_SET_INFO, + SMB2_OP_RENAME, + SMB2_OP_UNLINK, + }; + const int num_cmds = ARRAY_SIZE(cmds); + char *to_name __free(kfree) = NULL; + __u32 attrs = cinode->cifsAttrs; + struct cifs_open_parms oparms; + static atomic_t sillycounter; + struct cifsFileInfo *cfile; + struct tcon_link *tlink; + struct cifs_tcon *tcon; + struct kvec iov[2]; + const char *ppath; + void *page; + size_t len; + int rc; + + tlink = cifs_sb_tlink(cifs_sb); + if (IS_ERR(tlink)) + return PTR_ERR(tlink); + tcon = tlink_tcon(tlink); + + page = alloc_dentry_path(); + + ppath = build_path_from_dentry(dentry->d_parent, page); + if (IS_ERR(ppath)) { + rc = PTR_ERR(ppath); + goto out; + } + + len = strlen(ppath) + strlen("/.__smb1234") + 1; + to_name = kmalloc(len, GFP_KERNEL); + if (!to_name) { + rc = -ENOMEM; + goto out; + } + + scnprintf(to_name, len, "%s%c.__smb%04X", ppath, CIFS_DIR_SEP(cifs_sb), + atomic_inc_return(&sillycounter) & 0xffff); + + utf16_path = utf16_smb2_path(cifs_sb, to_name, len); + if (!utf16_path) { + rc = -ENOMEM; + goto out; + } + + drop_cached_dir_by_name(xid, tcon, full_path, cifs_sb); + oparms = CIFS_OPARMS(cifs_sb, tcon, full_path, + DELETE | FILE_WRITE_ATTRIBUTES, + FILE_OPEN, co, ACL_NO_MODE); + + attrs &= ~ATTR_READONLY; + if (!attrs) + attrs = ATTR_NORMAL; + if (d_inode(dentry)->i_nlink <= 1) + attrs |= ATTR_HIDDEN; + iov[0].iov_base = &(FILE_BASIC_INFO) { + .Attributes = cpu_to_le32(attrs), + }; + iov[0].iov_len = sizeof(FILE_BASIC_INFO); + iov[1].iov_base = utf16_path; + iov[1].iov_len = sizeof(*utf16_path) * UniStrlen((wchar_t *)utf16_path); + + cifs_get_writable_path(tcon, full_path, FIND_WR_WITH_DELETE, &cfile); + rc = smb2_compound_op(xid, tcon, cifs_sb, full_path, &oparms, iov, + cmds, num_cmds, cfile, NULL, NULL, dentry); + if (rc == -EINVAL) { + cifs_dbg(FYI, "invalid lease key, resending request without lease\n"); + cifs_get_writable_path(tcon, full_path, + FIND_WR_WITH_DELETE, &cfile); + rc = smb2_compound_op(xid, tcon, cifs_sb, full_path, &oparms, iov, + cmds, num_cmds, cfile, NULL, NULL, NULL); + } + if (!rc) { + set_bit(CIFS_INO_DELETE_PENDING, &cinode->flags); + } else { + cifs_tcon_dbg(FYI, "%s: failed to rename '%s' to '%s': %d\n", + __func__, full_path, to_name, rc); + rc = -EIO; + } +out: + cifs_put_tlink(tlink); + free_dentry_path(page); + return rc; +} diff --git a/fs/smb/client/smb2misc.c b/fs/smb/client/smb2misc.c index cddf273c14ae..89d933b4a8bc 100644 --- a/fs/smb/client/smb2misc.c +++ b/fs/smb/client/smb2misc.c @@ -614,6 +614,15 @@ smb2_is_valid_lease_break(char *buffer, struct TCP_Server_Info *server) struct cifs_tcon *tcon; struct cifs_pending_open *open; + /* Trace receipt of lease break request from server */ + trace_smb3_lease_break_enter(le32_to_cpu(rsp->CurrentLeaseState), + le32_to_cpu(rsp->Flags), + le16_to_cpu(rsp->Epoch), + le32_to_cpu(rsp->hdr.Id.SyncId.TreeId), + le64_to_cpu(rsp->hdr.SessionId), + *((u64 *)rsp->LeaseKey), + *((u64 *)&rsp->LeaseKey[8])); + cifs_dbg(FYI, "Checking for lease break\n"); /* If server is a channel, select the primary channel */ @@ -660,10 +669,12 @@ smb2_is_valid_lease_break(char *buffer, struct TCP_Server_Info *server) spin_unlock(&cifs_tcp_ses_lock); cifs_dbg(FYI, "Can not process lease break - no lease matched\n"); trace_smb3_lease_not_found(le32_to_cpu(rsp->CurrentLeaseState), - le32_to_cpu(rsp->hdr.Id.SyncId.TreeId), - le64_to_cpu(rsp->hdr.SessionId), - *((u64 *)rsp->LeaseKey), - *((u64 *)&rsp->LeaseKey[8])); + le32_to_cpu(rsp->Flags), + le16_to_cpu(rsp->Epoch), + le32_to_cpu(rsp->hdr.Id.SyncId.TreeId), + le64_to_cpu(rsp->hdr.SessionId), + *((u64 *)rsp->LeaseKey), + *((u64 *)&rsp->LeaseKey[8])); return false; } diff --git a/fs/smb/client/smb2ops.c b/fs/smb/client/smb2ops.c index 94b1d7a395d5..e586f3f4b5c9 100644 --- a/fs/smb/client/smb2ops.c +++ b/fs/smb/client/smb2ops.c @@ -2640,13 +2640,35 @@ smb2_set_next_command(struct cifs_tcon *tcon, struct smb_rqst *rqst) } /* SMB headers in a compound are 8 byte aligned. */ - if (!IS_ALIGNED(len, 8)) { - num_padding = 8 - (len & 7); + if (IS_ALIGNED(len, 8)) + goto out; + + num_padding = 8 - (len & 7); + if (smb3_encryption_required(tcon)) { + int i; + + /* + * Flatten request into a single buffer with required padding as + * the encryption layer can't handle the padding iovs. + */ + for (i = 1; i < rqst->rq_nvec; i++) { + memcpy(rqst->rq_iov[0].iov_base + + rqst->rq_iov[0].iov_len, + rqst->rq_iov[i].iov_base, + rqst->rq_iov[i].iov_len); + rqst->rq_iov[0].iov_len += rqst->rq_iov[i].iov_len; + } + memset(rqst->rq_iov[0].iov_base + rqst->rq_iov[0].iov_len, + 0, num_padding); + rqst->rq_iov[0].iov_len += num_padding; + rqst->rq_nvec = 1; + } else { rqst->rq_iov[rqst->rq_nvec].iov_base = smb2_padding; rqst->rq_iov[rqst->rq_nvec].iov_len = num_padding; rqst->rq_nvec++; - len += num_padding; } + len += num_padding; +out: shdr->NextCommand = cpu_to_le32(len); } @@ -5376,6 +5398,7 @@ struct smb_version_operations smb20_operations = { .llseek = smb3_llseek, .is_status_io_timeout = smb2_is_status_io_timeout, .is_network_name_deleted = smb2_is_network_name_deleted, + .rename_pending_delete = smb2_rename_pending_delete, }; #endif /* CIFS_ALLOW_INSECURE_LEGACY */ @@ -5481,6 +5504,7 @@ struct smb_version_operations smb21_operations = { .llseek = smb3_llseek, .is_status_io_timeout = smb2_is_status_io_timeout, .is_network_name_deleted = smb2_is_network_name_deleted, + .rename_pending_delete = smb2_rename_pending_delete, }; struct smb_version_operations smb30_operations = { @@ -5597,6 +5621,7 @@ struct smb_version_operations smb30_operations = { .llseek = smb3_llseek, .is_status_io_timeout = smb2_is_status_io_timeout, .is_network_name_deleted = smb2_is_network_name_deleted, + .rename_pending_delete = smb2_rename_pending_delete, }; struct smb_version_operations smb311_operations = { @@ -5713,6 +5738,7 @@ struct smb_version_operations smb311_operations = { .llseek = smb3_llseek, .is_status_io_timeout = smb2_is_status_io_timeout, .is_network_name_deleted = smb2_is_network_name_deleted, + .rename_pending_delete = smb2_rename_pending_delete, }; #ifdef CONFIG_CIFS_ALLOW_INSECURE_LEGACY diff --git a/fs/smb/client/smb2pdu.c b/fs/smb/client/smb2pdu.c index 2df93a75e3b8..c3b9d3f6210f 100644 --- a/fs/smb/client/smb2pdu.c +++ b/fs/smb/client/smb2pdu.c @@ -6192,11 +6192,11 @@ SMB2_lease_break(const unsigned int xid, struct cifs_tcon *tcon, please_key_high = (__u64 *)(lease_key+8); if (rc) { cifs_stats_fail_inc(tcon, SMB2_OPLOCK_BREAK_HE); - trace_smb3_lease_err(le32_to_cpu(lease_state), tcon->tid, + trace_smb3_lease_ack_err(le32_to_cpu(lease_state), tcon->tid, ses->Suid, *please_key_low, *please_key_high, rc); cifs_dbg(FYI, "Send error in Lease Break = %d\n", rc); } else - trace_smb3_lease_done(le32_to_cpu(lease_state), tcon->tid, + trace_smb3_lease_ack_done(le32_to_cpu(lease_state), tcon->tid, ses->Suid, *please_key_low, *please_key_high); return rc; diff --git a/fs/smb/client/smb2proto.h b/fs/smb/client/smb2proto.h index 6e805ece6a7b..b3f1398c9f79 100644 --- a/fs/smb/client/smb2proto.h +++ b/fs/smb/client/smb2proto.h @@ -317,5 +317,8 @@ int posix_info_sid_size(const void *beg, const void *end); int smb2_make_nfs_node(unsigned int xid, struct inode *inode, struct dentry *dentry, struct cifs_tcon *tcon, const char *full_path, umode_t mode, dev_t dev); +int smb2_rename_pending_delete(const char *full_path, + struct dentry *dentry, + const unsigned int xid); #endif /* _SMB2PROTO_H */ diff --git a/fs/smb/client/smbdirect.c b/fs/smb/client/smbdirect.c index 02d6db431fd4..e0fce5033004 100644 --- a/fs/smb/client/smbdirect.c +++ b/fs/smb/client/smbdirect.c @@ -453,9 +453,12 @@ static void recv_done(struct ib_cq *cq, struct ib_wc *wc) struct smbdirect_recv_io *response = container_of(wc->wr_cqe, struct smbdirect_recv_io, cqe); struct smbdirect_socket *sc = response->socket; + struct smbdirect_socket_parameters *sp = &sc->parameters; struct smbd_connection *info = container_of(sc, struct smbd_connection, socket); - int data_length = 0; + u32 data_offset = 0; + u32 data_length = 0; + u32 remaining_data_length = 0; log_rdma_recv(INFO, "response=0x%p type=%d wc status=%d wc opcode %d byte_len=%d pkey_index=%u\n", response, sc->recv_io.expected, wc->status, wc->opcode, @@ -487,7 +490,22 @@ static void recv_done(struct ib_cq *cq, struct ib_wc *wc) /* SMBD data transfer packet */ case SMBDIRECT_EXPECT_DATA_TRANSFER: data_transfer = smbdirect_recv_io_payload(response); + + if (wc->byte_len < + offsetof(struct smbdirect_data_transfer, padding)) + goto error; + + remaining_data_length = le32_to_cpu(data_transfer->remaining_data_length); + data_offset = le32_to_cpu(data_transfer->data_offset); data_length = le32_to_cpu(data_transfer->data_length); + if (wc->byte_len < data_offset || + (u64)wc->byte_len < (u64)data_offset + data_length) + goto error; + + if (remaining_data_length > sp->max_fragmented_recv_size || + data_length > sp->max_fragmented_recv_size || + (u64)remaining_data_length + (u64)data_length > (u64)sp->max_fragmented_recv_size) + goto error; if (data_length) { if (sc->recv_io.reassembly.full_packet_received) @@ -1090,8 +1108,10 @@ static int smbd_negotiate(struct smbd_connection *info) log_rdma_event(INFO, "smbd_post_recv rc=%d iov.addr=0x%llx iov.length=%u iov.lkey=0x%x\n", rc, response->sge.addr, response->sge.length, response->sge.lkey); - if (rc) + if (rc) { + put_receive_buffer(info, response); return rc; + } init_completion(&info->negotiate_completion); info->negotiate_done = false; @@ -1329,13 +1349,16 @@ void smbd_destroy(struct TCP_Server_Info *server) sc->status == SMBDIRECT_SOCKET_DISCONNECTED); } + log_rdma_event(INFO, "cancelling post_send_credits_work\n"); + disable_work_sync(&info->post_send_credits_work); + log_rdma_event(INFO, "destroying qp\n"); ib_drain_qp(sc->ib.qp); rdma_destroy_qp(sc->rdma.cm_id); sc->ib.qp = NULL; log_rdma_event(INFO, "cancelling idle timer\n"); - cancel_delayed_work_sync(&info->idle_timer_work); + disable_delayed_work_sync(&info->idle_timer_work); /* It's not possible for upper layer to get to reassembly */ log_rdma_event(INFO, "drain the reassembly queue\n"); @@ -1708,7 +1731,7 @@ allocate_mr_failed: return NULL; negotiation_failed: - cancel_delayed_work_sync(&info->idle_timer_work); + disable_delayed_work_sync(&info->idle_timer_work); destroy_caches_and_workqueue(info); sc->status = SMBDIRECT_SOCKET_NEGOTIATE_FAILED; rdma_disconnect(sc->rdma.cm_id); @@ -2067,7 +2090,7 @@ static void destroy_mr_list(struct smbd_connection *info) struct smbdirect_socket *sc = &info->socket; struct smbd_mr *mr, *tmp; - cancel_work_sync(&info->mr_recovery_work); + disable_work_sync(&info->mr_recovery_work); list_for_each_entry_safe(mr, tmp, &info->mr_list, list) { if (mr->state == MR_INVALIDATED) ib_dma_unmap_sg(sc->ib.dev, mr->sgt.sgl, diff --git a/fs/smb/client/trace.h b/fs/smb/client/trace.h index 93e5b2bb9f28..fd650e2afc76 100644 --- a/fs/smb/client/trace.h +++ b/fs/smb/client/trace.h @@ -669,13 +669,12 @@ DEFINE_SMB3_INF_COMPOUND_ENTER_EVENT(query_info_compound_enter); DEFINE_SMB3_INF_COMPOUND_ENTER_EVENT(posix_query_info_compound_enter); DEFINE_SMB3_INF_COMPOUND_ENTER_EVENT(hardlink_enter); DEFINE_SMB3_INF_COMPOUND_ENTER_EVENT(rename_enter); -DEFINE_SMB3_INF_COMPOUND_ENTER_EVENT(rmdir_enter); +DEFINE_SMB3_INF_COMPOUND_ENTER_EVENT(unlink_enter); DEFINE_SMB3_INF_COMPOUND_ENTER_EVENT(set_eof_enter); DEFINE_SMB3_INF_COMPOUND_ENTER_EVENT(set_info_compound_enter); DEFINE_SMB3_INF_COMPOUND_ENTER_EVENT(set_reparse_compound_enter); DEFINE_SMB3_INF_COMPOUND_ENTER_EVENT(get_reparse_compound_enter); DEFINE_SMB3_INF_COMPOUND_ENTER_EVENT(query_wsl_ea_compound_enter); -DEFINE_SMB3_INF_COMPOUND_ENTER_EVENT(delete_enter); DEFINE_SMB3_INF_COMPOUND_ENTER_EVENT(mkdir_enter); DEFINE_SMB3_INF_COMPOUND_ENTER_EVENT(tdis_enter); DEFINE_SMB3_INF_COMPOUND_ENTER_EVENT(mknod_enter); @@ -710,13 +709,12 @@ DEFINE_SMB3_INF_COMPOUND_DONE_EVENT(query_info_compound_done); DEFINE_SMB3_INF_COMPOUND_DONE_EVENT(posix_query_info_compound_done); DEFINE_SMB3_INF_COMPOUND_DONE_EVENT(hardlink_done); DEFINE_SMB3_INF_COMPOUND_DONE_EVENT(rename_done); -DEFINE_SMB3_INF_COMPOUND_DONE_EVENT(rmdir_done); +DEFINE_SMB3_INF_COMPOUND_DONE_EVENT(unlink_done); DEFINE_SMB3_INF_COMPOUND_DONE_EVENT(set_eof_done); DEFINE_SMB3_INF_COMPOUND_DONE_EVENT(set_info_compound_done); DEFINE_SMB3_INF_COMPOUND_DONE_EVENT(set_reparse_compound_done); DEFINE_SMB3_INF_COMPOUND_DONE_EVENT(get_reparse_compound_done); DEFINE_SMB3_INF_COMPOUND_DONE_EVENT(query_wsl_ea_compound_done); -DEFINE_SMB3_INF_COMPOUND_DONE_EVENT(delete_done); DEFINE_SMB3_INF_COMPOUND_DONE_EVENT(mkdir_done); DEFINE_SMB3_INF_COMPOUND_DONE_EVENT(tdis_done); DEFINE_SMB3_INF_COMPOUND_DONE_EVENT(mknod_done); @@ -756,14 +754,13 @@ DEFINE_SMB3_INF_COMPOUND_ERR_EVENT(query_info_compound_err); DEFINE_SMB3_INF_COMPOUND_ERR_EVENT(posix_query_info_compound_err); DEFINE_SMB3_INF_COMPOUND_ERR_EVENT(hardlink_err); DEFINE_SMB3_INF_COMPOUND_ERR_EVENT(rename_err); -DEFINE_SMB3_INF_COMPOUND_ERR_EVENT(rmdir_err); +DEFINE_SMB3_INF_COMPOUND_ERR_EVENT(unlink_err); DEFINE_SMB3_INF_COMPOUND_ERR_EVENT(set_eof_err); DEFINE_SMB3_INF_COMPOUND_ERR_EVENT(set_info_compound_err); DEFINE_SMB3_INF_COMPOUND_ERR_EVENT(set_reparse_compound_err); DEFINE_SMB3_INF_COMPOUND_ERR_EVENT(get_reparse_compound_err); DEFINE_SMB3_INF_COMPOUND_ERR_EVENT(query_wsl_ea_compound_err); DEFINE_SMB3_INF_COMPOUND_ERR_EVENT(mkdir_err); -DEFINE_SMB3_INF_COMPOUND_ERR_EVENT(delete_err); DEFINE_SMB3_INF_COMPOUND_ERR_EVENT(tdis_err); DEFINE_SMB3_INF_COMPOUND_ERR_EVENT(mknod_err); @@ -1171,8 +1168,54 @@ DEFINE_EVENT(smb3_lease_done_class, smb3_##name, \ __u64 lease_key_high), \ TP_ARGS(lease_state, tid, sesid, lease_key_low, lease_key_high)) -DEFINE_SMB3_LEASE_DONE_EVENT(lease_done); -DEFINE_SMB3_LEASE_DONE_EVENT(lease_not_found); +DEFINE_SMB3_LEASE_DONE_EVENT(lease_ack_done); +/* Tracepoint when a lease break request is received/entered (includes epoch and flags) */ +DECLARE_EVENT_CLASS(smb3_lease_enter_class, + TP_PROTO(__u32 lease_state, + __u32 flags, + __u16 epoch, + __u32 tid, + __u64 sesid, + __u64 lease_key_low, + __u64 lease_key_high), + TP_ARGS(lease_state, flags, epoch, tid, sesid, lease_key_low, lease_key_high), + TP_STRUCT__entry( + __field(__u32, lease_state) + __field(__u32, flags) + __field(__u16, epoch) + __field(__u32, tid) + __field(__u64, sesid) + __field(__u64, lease_key_low) + __field(__u64, lease_key_high) + ), + TP_fast_assign( + __entry->lease_state = lease_state; + __entry->flags = flags; + __entry->epoch = epoch; + __entry->tid = tid; + __entry->sesid = sesid; + __entry->lease_key_low = lease_key_low; + __entry->lease_key_high = lease_key_high; + ), + TP_printk("sid=0x%llx tid=0x%x lease_key=0x%llx%llx lease_state=0x%x flags=0x%x epoch=%u", + __entry->sesid, __entry->tid, __entry->lease_key_high, + __entry->lease_key_low, __entry->lease_state, __entry->flags, __entry->epoch) +) + +#define DEFINE_SMB3_LEASE_ENTER_EVENT(name) \ +DEFINE_EVENT(smb3_lease_enter_class, smb3_##name, \ + TP_PROTO(__u32 lease_state, \ + __u32 flags, \ + __u16 epoch, \ + __u32 tid, \ + __u64 sesid, \ + __u64 lease_key_low, \ + __u64 lease_key_high), \ + TP_ARGS(lease_state, flags, epoch, tid, sesid, lease_key_low, lease_key_high)) + +DEFINE_SMB3_LEASE_ENTER_EVENT(lease_break_enter); +/* Lease not found: reuse lease_enter payload (includes epoch and flags) */ +DEFINE_SMB3_LEASE_ENTER_EVENT(lease_not_found); DECLARE_EVENT_CLASS(smb3_lease_err_class, TP_PROTO(__u32 lease_state, @@ -1213,7 +1256,7 @@ DEFINE_EVENT(smb3_lease_err_class, smb3_##name, \ int rc), \ TP_ARGS(lease_state, tid, sesid, lease_key_low, lease_key_high, rc)) -DEFINE_SMB3_LEASE_ERR_EVENT(lease_err); +DEFINE_SMB3_LEASE_ERR_EVENT(lease_ack_err); DECLARE_EVENT_CLASS(smb3_connect_class, TP_PROTO(char *hostname, diff --git a/fs/smb/server/smb2pdu.c b/fs/smb/server/smb2pdu.c index 0d92ce49aed7..a565fc36cee6 100644 --- a/fs/smb/server/smb2pdu.c +++ b/fs/smb/server/smb2pdu.c @@ -2951,18 +2951,19 @@ int smb2_open(struct ksmbd_work *work) } ksmbd_debug(SMB, "converted name = %s\n", name); - if (strchr(name, ':')) { - if (!test_share_config_flag(work->tcon->share_conf, - KSMBD_SHARE_FLAG_STREAMS)) { - rc = -EBADF; - goto err_out2; - } - rc = parse_stream_name(name, &stream_name, &s_type); - if (rc < 0) - goto err_out2; - } if (posix_ctxt == false) { + if (strchr(name, ':')) { + if (!test_share_config_flag(work->tcon->share_conf, + KSMBD_SHARE_FLAG_STREAMS)) { + rc = -EBADF; + goto err_out2; + } + rc = parse_stream_name(name, &stream_name, &s_type); + if (rc < 0) + goto err_out2; + } + rc = ksmbd_validate_filename(name); if (rc < 0) goto err_out2; @@ -3443,6 +3444,8 @@ int smb2_open(struct ksmbd_work *work) fp->attrib_only = !(req->DesiredAccess & ~(FILE_READ_ATTRIBUTES_LE | FILE_WRITE_ATTRIBUTES_LE | FILE_SYNCHRONIZE_LE)); + fp->is_posix_ctxt = posix_ctxt; + /* fp should be searchable through ksmbd_inode.m_fp_list * after daccess, saccess, attrib_only, and stream are * initialized. @@ -5988,7 +5991,7 @@ static int smb2_rename(struct ksmbd_work *work, if (IS_ERR(new_name)) return PTR_ERR(new_name); - if (strchr(new_name, ':')) { + if (fp->is_posix_ctxt == false && strchr(new_name, ':')) { int s_type; char *xattr_stream_name, *stream_name = NULL; size_t xattr_stream_size; diff --git a/fs/smb/server/transport_rdma.c b/fs/smb/server/transport_rdma.c index 5466aa8c39b1..6550bd9f002c 100644 --- a/fs/smb/server/transport_rdma.c +++ b/fs/smb/server/transport_rdma.c @@ -554,7 +554,7 @@ static void recv_done(struct ib_cq *cq, struct ib_wc *wc) case SMB_DIRECT_MSG_DATA_TRANSFER: { struct smb_direct_data_transfer *data_transfer = (struct smb_direct_data_transfer *)recvmsg->packet; - unsigned int data_length; + u32 remaining_data_length, data_offset, data_length; int avail_recvmsg_count, receive_credits; if (wc->byte_len < @@ -564,15 +564,25 @@ static void recv_done(struct ib_cq *cq, struct ib_wc *wc) return; } + remaining_data_length = le32_to_cpu(data_transfer->remaining_data_length); data_length = le32_to_cpu(data_transfer->data_length); - if (data_length) { - if (wc->byte_len < sizeof(struct smb_direct_data_transfer) + - (u64)data_length) { - put_recvmsg(t, recvmsg); - smb_direct_disconnect_rdma_connection(t); - return; - } + data_offset = le32_to_cpu(data_transfer->data_offset); + if (wc->byte_len < data_offset || + wc->byte_len < (u64)data_offset + data_length) { + put_recvmsg(t, recvmsg); + smb_direct_disconnect_rdma_connection(t); + return; + } + if (remaining_data_length > t->max_fragmented_recv_size || + data_length > t->max_fragmented_recv_size || + (u64)remaining_data_length + (u64)data_length > + (u64)t->max_fragmented_recv_size) { + put_recvmsg(t, recvmsg); + smb_direct_disconnect_rdma_connection(t); + return; + } + if (data_length) { if (t->full_packet_received) recvmsg->first_segment = true; @@ -1209,78 +1219,130 @@ static int smb_direct_writev(struct ksmbd_transport *t, bool need_invalidate, unsigned int remote_key) { struct smb_direct_transport *st = smb_trans_direct_transfort(t); - int remaining_data_length; - int start, i, j; - int max_iov_size = st->max_send_size - + size_t remaining_data_length; + size_t iov_idx; + size_t iov_ofs; + size_t max_iov_size = st->max_send_size - sizeof(struct smb_direct_data_transfer); int ret; - struct kvec vec; struct smb_direct_send_ctx send_ctx; + int error = 0; if (st->status != SMB_DIRECT_CS_CONNECTED) return -ENOTCONN; //FIXME: skip RFC1002 header.. + if (WARN_ON_ONCE(niovs <= 1 || iov[0].iov_len != 4)) + return -EINVAL; buflen -= 4; + iov_idx = 1; + iov_ofs = 0; remaining_data_length = buflen; ksmbd_debug(RDMA, "Sending smb (RDMA): smb_len=%u\n", buflen); smb_direct_send_ctx_init(st, &send_ctx, need_invalidate, remote_key); - start = i = 1; - buflen = 0; - while (true) { - buflen += iov[i].iov_len; - if (buflen > max_iov_size) { - if (i > start) { - remaining_data_length -= - (buflen - iov[i].iov_len); - ret = smb_direct_post_send_data(st, &send_ctx, - &iov[start], i - start, - remaining_data_length); - if (ret) - goto done; - } else { - /* iov[start] is too big, break it */ - int nvec = (buflen + max_iov_size - 1) / - max_iov_size; + while (remaining_data_length) { + struct kvec vecs[SMB_DIRECT_MAX_SEND_SGES - 1]; /* minus smbdirect hdr */ + size_t possible_bytes = max_iov_size; + size_t possible_vecs; + size_t bytes = 0; + size_t nvecs = 0; + + /* + * For the last message remaining_data_length should be + * have been 0 already! + */ + if (WARN_ON_ONCE(iov_idx >= niovs)) { + error = -EINVAL; + goto done; + } + + /* + * We have 2 factors which limit the arguments we pass + * to smb_direct_post_send_data(): + * + * 1. The number of supported sges for the send, + * while one is reserved for the smbdirect header. + * And we currently need one SGE per page. + * 2. The number of negotiated payload bytes per send. + */ + possible_vecs = min_t(size_t, ARRAY_SIZE(vecs), niovs - iov_idx); + + while (iov_idx < niovs && possible_vecs && possible_bytes) { + struct kvec *v = &vecs[nvecs]; + int page_count; - for (j = 0; j < nvec; j++) { - vec.iov_base = - (char *)iov[start].iov_base + - j * max_iov_size; - vec.iov_len = - min_t(int, max_iov_size, - buflen - max_iov_size * j); - remaining_data_length -= vec.iov_len; - ret = smb_direct_post_send_data(st, &send_ctx, &vec, 1, - remaining_data_length); - if (ret) - goto done; + v->iov_base = ((u8 *)iov[iov_idx].iov_base) + iov_ofs; + v->iov_len = min_t(size_t, + iov[iov_idx].iov_len - iov_ofs, + possible_bytes); + page_count = get_buf_page_count(v->iov_base, v->iov_len); + if (page_count > possible_vecs) { + /* + * If the number of pages in the buffer + * is to much (because we currently require + * one SGE per page), we need to limit the + * length. + * + * We know possible_vecs is at least 1, + * so we always keep the first page. + * + * We need to calculate the number extra + * pages (epages) we can also keep. + * + * We calculate the number of bytes in the + * first page (fplen), this should never be + * larger than v->iov_len because page_count is + * at least 2, but adding a limitation feels + * better. + * + * Then we calculate the number of bytes (elen) + * we can keep for the extra pages. + */ + size_t epages = possible_vecs - 1; + size_t fpofs = offset_in_page(v->iov_base); + size_t fplen = min_t(size_t, PAGE_SIZE - fpofs, v->iov_len); + size_t elen = min_t(size_t, v->iov_len - fplen, epages*PAGE_SIZE); + + v->iov_len = fplen + elen; + page_count = get_buf_page_count(v->iov_base, v->iov_len); + if (WARN_ON_ONCE(page_count > possible_vecs)) { + /* + * Something went wrong in the above + * logic... + */ + error = -EINVAL; + goto done; } - i++; - if (i == niovs) - break; } - start = i; - buflen = 0; - } else { - i++; - if (i == niovs) { - /* send out all remaining vecs */ - remaining_data_length -= buflen; - ret = smb_direct_post_send_data(st, &send_ctx, - &iov[start], i - start, - remaining_data_length); - if (ret) - goto done; - break; + possible_vecs -= page_count; + nvecs += 1; + possible_bytes -= v->iov_len; + bytes += v->iov_len; + + iov_ofs += v->iov_len; + if (iov_ofs >= iov[iov_idx].iov_len) { + iov_idx += 1; + iov_ofs = 0; } } + + remaining_data_length -= bytes; + + ret = smb_direct_post_send_data(st, &send_ctx, + vecs, nvecs, + remaining_data_length); + if (unlikely(ret)) { + error = ret; + goto done; + } } done: ret = smb_direct_flush_send_list(st, &send_ctx, true); + if (unlikely(!ret && error)) + ret = error; /* * As an optimization, we don't wait for individual I/O to finish @@ -1744,6 +1806,11 @@ static int smb_direct_init_params(struct smb_direct_transport *t, return -EINVAL; } + if (device->attrs.max_send_sge < SMB_DIRECT_MAX_SEND_SGES) { + pr_err("warning: device max_send_sge = %d too small\n", + device->attrs.max_send_sge); + return -EINVAL; + } if (device->attrs.max_recv_sge < SMB_DIRECT_MAX_RECV_SGES) { pr_err("warning: device max_recv_sge = %d too small\n", device->attrs.max_recv_sge); @@ -1767,7 +1834,7 @@ static int smb_direct_init_params(struct smb_direct_transport *t, cap->max_send_wr = max_send_wrs; cap->max_recv_wr = t->recv_credit_max; - cap->max_send_sge = max_sge_per_wr; + cap->max_send_sge = SMB_DIRECT_MAX_SEND_SGES; cap->max_recv_sge = SMB_DIRECT_MAX_RECV_SGES; cap->max_inline_data = 0; cap->max_rdma_ctxs = t->max_rw_credits; diff --git a/fs/smb/server/vfs_cache.h b/fs/smb/server/vfs_cache.h index 0708155b5caf..78b506c5ef03 100644 --- a/fs/smb/server/vfs_cache.h +++ b/fs/smb/server/vfs_cache.h @@ -112,6 +112,8 @@ struct ksmbd_file { bool is_durable; bool is_persistent; bool is_resilient; + + bool is_posix_ctxt; }; static inline void set_ctx_actor(struct dir_context *ctx, diff --git a/fs/xfs/Kconfig b/fs/xfs/Kconfig index ae0ca6858496..065953475cf5 100644 --- a/fs/xfs/Kconfig +++ b/fs/xfs/Kconfig @@ -105,6 +105,7 @@ config XFS_POSIX_ACL config XFS_RT bool "XFS Realtime subvolume support" depends on XFS_FS + default BLK_DEV_ZONED help If you say Y here you will be able to mount and use XFS filesystems which contain a realtime subvolume. The realtime subvolume is a diff --git a/fs/xfs/libxfs/xfs_attr_remote.c b/fs/xfs/libxfs/xfs_attr_remote.c index 4c44ce1c8a64..bff3dc226f81 100644 --- a/fs/xfs/libxfs/xfs_attr_remote.c +++ b/fs/xfs/libxfs/xfs_attr_remote.c @@ -435,6 +435,13 @@ xfs_attr_rmtval_get( 0, &bp, &xfs_attr3_rmt_buf_ops); if (xfs_metadata_is_sick(error)) xfs_dirattr_mark_sick(args->dp, XFS_ATTR_FORK); + /* + * ENODATA from disk implies a disk medium failure; + * ENODATA for xattrs means attribute not found, so + * disambiguate that here. + */ + if (error == -ENODATA) + error = -EIO; if (error) return error; diff --git a/fs/xfs/libxfs/xfs_da_btree.c b/fs/xfs/libxfs/xfs_da_btree.c index 17d9e6154f19..723a0643b838 100644 --- a/fs/xfs/libxfs/xfs_da_btree.c +++ b/fs/xfs/libxfs/xfs_da_btree.c @@ -2833,6 +2833,12 @@ xfs_da_read_buf( &bp, ops); if (xfs_metadata_is_sick(error)) xfs_dirattr_mark_sick(dp, whichfork); + /* + * ENODATA from disk implies a disk medium failure; ENODATA for + * xattrs means attribute not found, so disambiguate that here. + */ + if (error == -ENODATA && whichfork == XFS_ATTR_FORK) + error = -EIO; if (error) goto out_free; diff --git a/fs/xfs/xfs_aops.c b/fs/xfs/xfs_aops.c index 1ee4f835ac3c..a26f79815533 100644 --- a/fs/xfs/xfs_aops.c +++ b/fs/xfs/xfs_aops.c @@ -760,6 +760,9 @@ xfs_vm_swap_activate( { struct xfs_inode *ip = XFS_I(file_inode(swap_file)); + if (xfs_is_zoned_inode(ip)) + return -EINVAL; + /* * Swap file activation can race against concurrent shared extent * removal in files that have been cloned. If this happens, diff --git a/fs/xfs/xfs_zone_alloc.c b/fs/xfs/xfs_zone_alloc.c index f8bd6d741755..f28214c28ab5 100644 --- a/fs/xfs/xfs_zone_alloc.c +++ b/fs/xfs/xfs_zone_alloc.c @@ -374,44 +374,6 @@ xfs_zone_free_blocks( return 0; } -/* - * Check if the zone containing the data just before the offset we are - * writing to is still open and has space. - */ -static struct xfs_open_zone * -xfs_last_used_zone( - struct iomap_ioend *ioend) -{ - struct xfs_inode *ip = XFS_I(ioend->io_inode); - struct xfs_mount *mp = ip->i_mount; - xfs_fileoff_t offset_fsb = XFS_B_TO_FSB(mp, ioend->io_offset); - struct xfs_rtgroup *rtg = NULL; - struct xfs_open_zone *oz = NULL; - struct xfs_iext_cursor icur; - struct xfs_bmbt_irec got; - - xfs_ilock(ip, XFS_ILOCK_SHARED); - if (!xfs_iext_lookup_extent_before(ip, &ip->i_df, &offset_fsb, - &icur, &got)) { - xfs_iunlock(ip, XFS_ILOCK_SHARED); - return NULL; - } - xfs_iunlock(ip, XFS_ILOCK_SHARED); - - rtg = xfs_rtgroup_grab(mp, xfs_rtb_to_rgno(mp, got.br_startblock)); - if (!rtg) - return NULL; - - xfs_ilock(rtg_rmap(rtg), XFS_ILOCK_SHARED); - oz = READ_ONCE(rtg->rtg_open_zone); - if (oz && (oz->oz_is_gc || !atomic_inc_not_zero(&oz->oz_ref))) - oz = NULL; - xfs_iunlock(rtg_rmap(rtg), XFS_ILOCK_SHARED); - - xfs_rtgroup_rele(rtg); - return oz; -} - static struct xfs_group * xfs_find_free_zone( struct xfs_mount *mp, @@ -918,12 +880,9 @@ xfs_zone_alloc_and_submit( goto out_error; /* - * If we don't have a cached zone in this write context, see if the - * last extent before the one we are writing to points to an active - * zone. If so, just continue writing to it. + * If we don't have a locally cached zone in this write context, see if + * the inode is still associated with a zone and use that if so. */ - if (!*oz && ioend->io_offset) - *oz = xfs_last_used_zone(ioend); if (!*oz) *oz = xfs_cached_zone(mp, ip); diff --git a/fs/xfs/xfs_zone_space_resv.c b/fs/xfs/xfs_zone_space_resv.c index 1313c55b8cbe..9cd38716fd25 100644 --- a/fs/xfs/xfs_zone_space_resv.c +++ b/fs/xfs/xfs_zone_space_resv.c @@ -10,6 +10,7 @@ #include "xfs_mount.h" #include "xfs_inode.h" #include "xfs_rtbitmap.h" +#include "xfs_icache.h" #include "xfs_zone_alloc.h" #include "xfs_zone_priv.h" #include "xfs_zones.h" @@ -230,6 +231,11 @@ xfs_zoned_space_reserve( error = xfs_dec_freecounter(mp, XC_FREE_RTEXTENTS, count_fsb, flags & XFS_ZR_RESERVED); + if (error == -ENOSPC && !(flags & XFS_ZR_NOWAIT)) { + xfs_inodegc_flush(mp); + error = xfs_dec_freecounter(mp, XC_FREE_RTEXTENTS, count_fsb, + flags & XFS_ZR_RESERVED); + } if (error == -ENOSPC && (flags & XFS_ZR_GREEDY) && count_fsb > 1) error = xfs_zoned_reserve_extents_greedy(mp, &count_fsb, flags); if (error) |