diff options
Diffstat (limited to 'fs')
87 files changed, 1440 insertions, 859 deletions
diff --git a/fs/Kconfig b/fs/Kconfig index d621f02a3f9..aa195265362 100644 --- a/fs/Kconfig +++ b/fs/Kconfig @@ -4,6 +4,10 @@ menu "File systems" +# Use unaligned word dcache accesses +config DCACHE_WORD_ACCESS + bool + if BLOCK source "fs/ext2/Kconfig" diff --git a/fs/afs/file.c b/fs/afs/file.c index 14d89fa58fe..8f6e9234d56 100644 --- a/fs/afs/file.c +++ b/fs/afs/file.c @@ -251,7 +251,7 @@ static int afs_readpages(struct file *file, struct address_space *mapping, ASSERT(key != NULL); vnode = AFS_FS_I(mapping->host); - if (vnode->flags & AFS_VNODE_DELETED) { + if (test_bit(AFS_VNODE_DELETED, &vnode->flags)) { _leave(" = -ESTALE"); return -ESTALE; } diff --git a/fs/afs/internal.h b/fs/afs/internal.h index d2b0888126d..a306bb6d88d 100644 --- a/fs/afs/internal.h +++ b/fs/afs/internal.h @@ -109,7 +109,7 @@ struct afs_call { unsigned reply_size; /* current size of reply */ unsigned first_offset; /* offset into mapping[first] */ unsigned last_to; /* amount of mapping[last] */ - unsigned short offset; /* offset into received data store */ + unsigned offset; /* offset into received data store */ unsigned char unmarshall; /* unmarshalling phase */ bool incoming; /* T if incoming call */ bool send_pages; /* T if data from mapping should be sent */ diff --git a/fs/afs/rxrpc.c b/fs/afs/rxrpc.c index e45a323aebb..8ad8c2a0703 100644 --- a/fs/afs/rxrpc.c +++ b/fs/afs/rxrpc.c @@ -314,6 +314,7 @@ int afs_make_call(struct in_addr *addr, struct afs_call *call, gfp_t gfp, struct msghdr msg; struct kvec iov[1]; int ret; + struct sk_buff *skb; _enter("%x,{%d},", addr->s_addr, ntohs(call->port)); @@ -380,6 +381,8 @@ int afs_make_call(struct in_addr *addr, struct afs_call *call, gfp_t gfp, error_do_abort: rxrpc_kernel_abort_call(rxcall, RX_USER_ABORT); + while ((skb = skb_dequeue(&call->rx_queue))) + afs_free_skb(skb); rxrpc_kernel_end_call(rxcall); call->rxcall = NULL; error_kill_call: @@ -228,12 +228,6 @@ static void __put_ioctx(struct kioctx *ctx) call_rcu(&ctx->rcu_head, ctx_rcu_free); } -static inline void get_ioctx(struct kioctx *kioctx) -{ - BUG_ON(atomic_read(&kioctx->users) <= 0); - atomic_inc(&kioctx->users); -} - static inline int try_get_ioctx(struct kioctx *kioctx) { return atomic_inc_not_zero(&kioctx->users); @@ -273,7 +267,7 @@ static struct kioctx *ioctx_alloc(unsigned nr_events) mm = ctx->mm = current->mm; atomic_inc(&mm->mm_count); - atomic_set(&ctx->users, 1); + atomic_set(&ctx->users, 2); spin_lock_init(&ctx->ctx_lock); spin_lock_init(&ctx->ring_info.ring_lock); init_waitqueue_head(&ctx->wait); @@ -490,6 +484,8 @@ static void kiocb_batch_free(struct kioctx *ctx, struct kiocb_batch *batch) kmem_cache_free(kiocb_cachep, req); ctx->reqs_active--; } + if (unlikely(!ctx->reqs_active && ctx->dead)) + wake_up_all(&ctx->wait); spin_unlock_irq(&ctx->ctx_lock); } @@ -607,11 +603,16 @@ static void aio_fput_routine(struct work_struct *data) fput(req->ki_filp); /* Link the iocb into the context's free list */ + rcu_read_lock(); spin_lock_irq(&ctx->ctx_lock); really_put_req(ctx, req); + /* + * at that point ctx might've been killed, but actual + * freeing is RCU'd + */ spin_unlock_irq(&ctx->ctx_lock); + rcu_read_unlock(); - put_ioctx(ctx); spin_lock_irq(&fput_lock); } spin_unlock_irq(&fput_lock); @@ -642,7 +643,6 @@ static int __aio_put_req(struct kioctx *ctx, struct kiocb *req) * this function will be executed w/out any aio kthread wakeup. */ if (unlikely(!fput_atomic(req->ki_filp))) { - get_ioctx(ctx); spin_lock(&fput_lock); list_add(&req->ki_list, &fput_head); spin_unlock(&fput_lock); @@ -1336,10 +1336,10 @@ SYSCALL_DEFINE2(io_setup, unsigned, nr_events, aio_context_t __user *, ctxp) ret = PTR_ERR(ioctx); if (!IS_ERR(ioctx)) { ret = put_user(ioctx->user_id, ctxp); - if (!ret) + if (!ret) { + put_ioctx(ioctx); return 0; - - get_ioctx(ioctx); /* io_destroy() expects us to hold a ref */ + } io_destroy(ioctx); } diff --git a/fs/autofs4/autofs_i.h b/fs/autofs4/autofs_i.h index d8d8e7ba6a1..eb1cc92cd67 100644 --- a/fs/autofs4/autofs_i.h +++ b/fs/autofs4/autofs_i.h @@ -110,6 +110,7 @@ struct autofs_sb_info { int sub_version; int min_proto; int max_proto; + int compat_daemon; unsigned long exp_timeout; unsigned int type; int reghost_enabled; diff --git a/fs/autofs4/dev-ioctl.c b/fs/autofs4/dev-ioctl.c index 76741d8d778..85f1fcdb30e 100644 --- a/fs/autofs4/dev-ioctl.c +++ b/fs/autofs4/dev-ioctl.c @@ -385,6 +385,7 @@ static int autofs_dev_ioctl_setpipefd(struct file *fp, sbi->pipefd = pipefd; sbi->pipe = pipe; sbi->catatonic = 0; + sbi->compat_daemon = is_compat_task(); } out: mutex_unlock(&sbi->wq_mutex); diff --git a/fs/autofs4/expire.c b/fs/autofs4/expire.c index 450f529a4ea..1feb68ecef9 100644 --- a/fs/autofs4/expire.c +++ b/fs/autofs4/expire.c @@ -124,6 +124,7 @@ start: /* Negative dentry - try next */ if (!simple_positive(q)) { spin_unlock(&p->d_lock); + lock_set_subclass(&q->d_lock.dep_map, 0, _RET_IP_); p = q; goto again; } @@ -186,6 +187,7 @@ again: /* Negative dentry - try next */ if (!simple_positive(ret)) { spin_unlock(&p->d_lock); + lock_set_subclass(&ret->d_lock.dep_map, 0, _RET_IP_); p = ret; goto again; } diff --git a/fs/autofs4/inode.c b/fs/autofs4/inode.c index e16980b00b8..06858d95512 100644 --- a/fs/autofs4/inode.c +++ b/fs/autofs4/inode.c @@ -19,6 +19,7 @@ #include <linux/parser.h> #include <linux/bitops.h> #include <linux/magic.h> +#include <linux/compat.h> #include "autofs_i.h" #include <linux/module.h> @@ -224,6 +225,7 @@ int autofs4_fill_super(struct super_block *s, void *data, int silent) set_autofs_type_indirect(&sbi->type); sbi->min_proto = 0; sbi->max_proto = 0; + sbi->compat_daemon = is_compat_task(); mutex_init(&sbi->wq_mutex); mutex_init(&sbi->pipe_mutex); spin_lock_init(&sbi->fs_lock); diff --git a/fs/autofs4/waitq.c b/fs/autofs4/waitq.c index da8876d38a7..9c098db4334 100644 --- a/fs/autofs4/waitq.c +++ b/fs/autofs4/waitq.c @@ -91,7 +91,24 @@ static int autofs4_write(struct autofs_sb_info *sbi, return (bytes > 0); } - + +/* + * The autofs_v5 packet was misdesigned. + * + * The packets are identical on x86-32 and x86-64, but have different + * alignment. Which means that 'sizeof()' will give different results. + * Fix it up for the case of running 32-bit user mode on a 64-bit kernel. + */ +static noinline size_t autofs_v5_packet_size(struct autofs_sb_info *sbi) +{ + size_t pktsz = sizeof(struct autofs_v5_packet); +#if defined(CONFIG_X86_64) && defined(CONFIG_COMPAT) + if (sbi->compat_daemon > 0) + pktsz -= 4; +#endif + return pktsz; +} + static void autofs4_notify_daemon(struct autofs_sb_info *sbi, struct autofs_wait_queue *wq, int type) @@ -155,8 +172,7 @@ static void autofs4_notify_daemon(struct autofs_sb_info *sbi, { struct autofs_v5_packet *packet = &pkt.v5_pkt.v5_packet; - pktsz = sizeof(*packet); - + pktsz = autofs_v5_packet_size(sbi); packet->wait_queue_token = wq->wait_queue_token; packet->len = wq->name.len; memcpy(packet->name, wq->name.name, wq->name.len); diff --git a/fs/binfmt_aout.c b/fs/binfmt_aout.c index a6395bdb26a..1ff94054d35 100644 --- a/fs/binfmt_aout.c +++ b/fs/binfmt_aout.c @@ -259,6 +259,13 @@ static int load_aout_binary(struct linux_binprm * bprm, struct pt_regs * regs) current->mm->free_area_cache = current->mm->mmap_base; current->mm->cached_hole_size = 0; + retval = setup_arg_pages(bprm, STACK_TOP, EXSTACK_DEFAULT); + if (retval < 0) { + /* Someone check-me: is this error path enough? */ + send_sig(SIGKILL, current, 0); + return retval; + } + install_exec_creds(bprm); current->flags &= ~PF_FORKNOEXEC; @@ -352,13 +359,6 @@ beyond_if: return retval; } - retval = setup_arg_pages(bprm, STACK_TOP, EXSTACK_DEFAULT); - if (retval < 0) { - /* Someone check-me: is this error path enough? */ - send_sig(SIGKILL, current, 0); - return retval; - } - current->mm->start_stack = (unsigned long) create_aout_tables((char __user *) bprm->p, bprm); #ifdef __alpha__ diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c index bcb884e2d61..07d096c4992 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -1421,7 +1421,7 @@ static int fill_thread_core_info(struct elf_thread_core_info *t, for (i = 1; i < view->n; ++i) { const struct user_regset *regset = &view->regsets[i]; do_thread_regset_writeback(t->task, regset); - if (regset->core_note_type && + if (regset->core_note_type && regset->get && (!regset->active || regset->active(t->task, regset))) { int ret; size_t size = regset->n * regset->size; @@ -505,13 +505,9 @@ EXPORT_SYMBOL(bio_clone); int bio_get_nr_vecs(struct block_device *bdev) { struct request_queue *q = bdev_get_queue(bdev); - int nr_pages; - - nr_pages = ((queue_max_sectors(q) << 9) + PAGE_SIZE - 1) >> PAGE_SHIFT; - if (nr_pages > queue_max_segments(q)) - nr_pages = queue_max_segments(q); - - return nr_pages; + return min_t(unsigned, + queue_max_segments(q), + queue_max_sectors(q) / (PAGE_SIZE >> 9) + 1); } EXPORT_SYMBOL(bio_get_nr_vecs); diff --git a/fs/block_dev.c b/fs/block_dev.c index 0e575d1304b..5e9f198f771 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -1183,8 +1183,12 @@ static int __blkdev_get(struct block_device *bdev, fmode_t mode, int for_part) * The latter is necessary to prevent ghost * partitions on a removed medium. */ - if (bdev->bd_invalidated && (!ret || ret == -ENOMEDIUM)) - rescan_partitions(disk, bdev); + if (bdev->bd_invalidated) { + if (!ret) + rescan_partitions(disk, bdev); + else if (ret == -ENOMEDIUM) + invalidate_partitions(disk, bdev); + } if (ret) goto out_clear; } else { @@ -1214,8 +1218,12 @@ static int __blkdev_get(struct block_device *bdev, fmode_t mode, int for_part) if (bdev->bd_disk->fops->open) ret = bdev->bd_disk->fops->open(bdev, mode); /* the same as first opener case, read comment there */ - if (bdev->bd_invalidated && (!ret || ret == -ENOMEDIUM)) - rescan_partitions(bdev->bd_disk, bdev); + if (bdev->bd_invalidated) { + if (!ret) + rescan_partitions(bdev->bd_disk, bdev); + else if (ret == -ENOMEDIUM) + invalidate_partitions(bdev->bd_disk, bdev); + } if (ret) goto out_unlock_bdev; } diff --git a/fs/btrfs/backref.c b/fs/btrfs/backref.c index 633c701a287..0436c12da8c 100644 --- a/fs/btrfs/backref.c +++ b/fs/btrfs/backref.c @@ -583,7 +583,7 @@ static int find_parent_nodes(struct btrfs_trans_handle *trans, struct btrfs_path *path; struct btrfs_key info_key = { 0 }; struct btrfs_delayed_ref_root *delayed_refs = NULL; - struct btrfs_delayed_ref_head *head = NULL; + struct btrfs_delayed_ref_head *head; int info_level = 0; int ret; struct list_head prefs_delayed; @@ -607,6 +607,8 @@ static int find_parent_nodes(struct btrfs_trans_handle *trans, * at a specified point in time */ again: + head = NULL; + ret = btrfs_search_slot(trans, fs_info->extent_root, &key, path, 0, 0); if (ret < 0) goto out; @@ -635,8 +637,10 @@ again: goto again; } ret = __add_delayed_refs(head, seq, &info_key, &prefs_delayed); - if (ret) + if (ret) { + spin_unlock(&delayed_refs->lock); goto out; + } } spin_unlock(&delayed_refs->lock); @@ -892,6 +896,8 @@ static char *iref_to_path(struct btrfs_root *fs_root, struct btrfs_path *path, if (eb != eb_in) free_extent_buffer(eb); ret = inode_ref_info(parent, 0, fs_root, path, &found_key); + if (ret > 0) + ret = -ENOENT; if (ret) break; next_inum = found_key.offset; diff --git a/fs/btrfs/check-integrity.c b/fs/btrfs/check-integrity.c index b669a7d8e49..d986824bb2b 100644 --- a/fs/btrfs/check-integrity.c +++ b/fs/btrfs/check-integrity.c @@ -644,7 +644,7 @@ static struct btrfsic_dev_state *btrfsic_dev_state_hashtable_lookup( static int btrfsic_process_superblock(struct btrfsic_state *state, struct btrfs_fs_devices *fs_devices) { - int ret; + int ret = 0; struct btrfs_super_block *selected_super; struct list_head *dev_head = &fs_devices->devices; struct btrfs_device *device; diff --git a/fs/btrfs/compression.c b/fs/btrfs/compression.c index 14f1c5a0b2d..d02c27cd14c 100644 --- a/fs/btrfs/compression.c +++ b/fs/btrfs/compression.c @@ -588,6 +588,8 @@ int btrfs_submit_compressed_read(struct inode *inode, struct bio *bio, page_offset(bio->bi_io_vec->bv_page), PAGE_CACHE_SIZE); read_unlock(&em_tree->lock); + if (!em) + return -EIO; compressed_len = em->block_len; cb = kmalloc(compressed_bio_size(root, compressed_len), GFP_NOFS); diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index 27ebe61d3cc..80b6486fd5e 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -886,7 +886,7 @@ struct btrfs_block_rsv { u64 reserved; struct btrfs_space_info *space_info; spinlock_t lock; - unsigned int full:1; + unsigned int full; }; /* diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 811d9f918b1..534266fe505 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -2260,6 +2260,12 @@ int open_ctree(struct super_block *sb, goto fail_sb_buffer; } + if (sectorsize < PAGE_SIZE) { + printk(KERN_WARNING "btrfs: Incompatible sector size " + "found on %s\n", sb->s_id); + goto fail_sb_buffer; + } + mutex_lock(&fs_info->chunk_mutex); ret = btrfs_read_sys_array(tree_root); mutex_unlock(&fs_info->chunk_mutex); @@ -2301,6 +2307,12 @@ int open_ctree(struct super_block *sb, btrfs_close_extra_devices(fs_devices); + if (!fs_devices->latest_bdev) { + printk(KERN_CRIT "btrfs: failed to read devices on %s\n", + sb->s_id); + goto fail_tree_roots; + } + retry_root_backup: blocksize = btrfs_level_size(tree_root, btrfs_super_root_level(disk_super)); diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 283af7a676a..37e0a800d34 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -3312,7 +3312,8 @@ commit_trans: } data_sinfo->bytes_may_use += bytes; trace_btrfs_space_reservation(root->fs_info, "space_info", - (u64)data_sinfo, bytes, 1); + (u64)(unsigned long)data_sinfo, + bytes, 1); spin_unlock(&data_sinfo->lock); return 0; @@ -3333,7 +3334,8 @@ void btrfs_free_reserved_data_space(struct inode *inode, u64 bytes) spin_lock(&data_sinfo->lock); data_sinfo->bytes_may_use -= bytes; trace_btrfs_space_reservation(root->fs_info, "space_info", - (u64)data_sinfo, bytes, 0); + (u64)(unsigned long)data_sinfo, + bytes, 0); spin_unlock(&data_sinfo->lock); } @@ -3611,12 +3613,15 @@ static int may_commit_transaction(struct btrfs_root *root, if (space_info != delayed_rsv->space_info) return -ENOSPC; + spin_lock(&space_info->lock); spin_lock(&delayed_rsv->lock); - if (delayed_rsv->size < bytes) { + if (space_info->bytes_pinned + delayed_rsv->size < bytes) { spin_unlock(&delayed_rsv->lock); + spin_unlock(&space_info->lock); return -ENOSPC; } spin_unlock(&delayed_rsv->lock); + spin_unlock(&space_info->lock); commit: trans = btrfs_join_transaction(root); @@ -3695,9 +3700,9 @@ again: if (used + orig_bytes <= space_info->total_bytes) { space_info->bytes_may_use += orig_bytes; trace_btrfs_space_reservation(root->fs_info, - "space_info", - (u64)space_info, - orig_bytes, 1); + "space_info", + (u64)(unsigned long)space_info, + orig_bytes, 1); ret = 0; } else { /* @@ -3766,9 +3771,9 @@ again: if (used + num_bytes < space_info->total_bytes + avail) { space_info->bytes_may_use += orig_bytes; trace_btrfs_space_reservation(root->fs_info, - "space_info", - (u64)space_info, - orig_bytes, 1); + "space_info", + (u64)(unsigned long)space_info, + orig_bytes, 1); ret = 0; } else { wait_ordered = true; @@ -3913,8 +3918,8 @@ static void block_rsv_release_bytes(struct btrfs_fs_info *fs_info, spin_lock(&space_info->lock); space_info->bytes_may_use -= num_bytes; trace_btrfs_space_reservation(fs_info, "space_info", - (u64)space_info, - num_bytes, 0); + (u64)(unsigned long)space_info, + num_bytes, 0); space_info->reservation_progress++; spin_unlock(&space_info->lock); } @@ -4105,7 +4110,7 @@ static u64 calc_global_metadata_size(struct btrfs_fs_info *fs_info) num_bytes += div64_u64(data_used + meta_used, 50); if (num_bytes * 3 > meta_used) - num_bytes = div64_u64(meta_used, 3); + num_bytes = div64_u64(meta_used, 3) * 2; return ALIGN(num_bytes, fs_info->extent_root->leafsize << 10); } @@ -4132,14 +4137,14 @@ static void update_global_block_rsv(struct btrfs_fs_info *fs_info) block_rsv->reserved += num_bytes; sinfo->bytes_may_use += num_bytes; trace_btrfs_space_reservation(fs_info, "space_info", - (u64)sinfo, num_bytes, 1); + (u64)(unsigned long)sinfo, num_bytes, 1); } if (block_rsv->reserved >= block_rsv->size) { num_bytes = block_rsv->reserved - block_rsv->size; sinfo->bytes_may_use -= num_bytes; trace_btrfs_space_reservation(fs_info, "space_info", - (u64)sinfo, num_bytes, 0); + (u64)(unsigned long)sinfo, num_bytes, 0); sinfo->reservation_progress++; block_rsv->reserved = block_rsv->size; block_rsv->full = 1; @@ -4192,7 +4197,8 @@ void btrfs_trans_release_metadata(struct btrfs_trans_handle *trans, if (!trans->bytes_reserved) return; - trace_btrfs_space_reservation(root->fs_info, "transaction", (u64)trans, + trace_btrfs_space_reservation(root->fs_info, "transaction", + (u64)(unsigned long)trans, trans->bytes_reserved, 0); btrfs_block_rsv_release(root, trans->block_rsv, trans->bytes_reserved); trans->bytes_reserved = 0; @@ -4710,9 +4716,9 @@ static int btrfs_update_reserved_bytes(struct btrfs_block_group_cache *cache, space_info->bytes_reserved += num_bytes; if (reserve == RESERVE_ALLOC) { trace_btrfs_space_reservation(cache->fs_info, - "space_info", - (u64)space_info, - num_bytes, 0); + "space_info", + (u64)(unsigned long)space_info, + num_bytes, 0); space_info->bytes_may_use -= num_bytes; } } @@ -7886,9 +7892,16 @@ int btrfs_trim_fs(struct btrfs_root *root, struct fstrim_range *range) u64 start; u64 end; u64 trimmed = 0; + u64 total_bytes = btrfs_super_total_bytes(fs_info->super_copy); int ret = 0; - cache = btrfs_lookup_block_group(fs_info, range->start); + /* + * try to trim all FS space, our block group may start from non-zero. + */ + if (range->len == total_bytes) + cache = btrfs_lookup_first_block_group(fs_info, range->start); + else + cache = btrfs_lookup_block_group(fs_info, range->start); while (cache) { if (cache->key.objectid >= (range->start + range->len)) { diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index fcf77e1ded4..a55fbe6252d 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -513,6 +513,15 @@ hit_next: WARN_ON(state->end < start); last_end = state->end; + if (state->end < end && !need_resched()) + next_node = rb_next(&state->rb_node); + else + next_node = NULL; + + /* the state doesn't have the wanted bits, go ahead */ + if (!(state->state & bits)) + goto next; + /* * | ---- desired range ---- | * | state | or @@ -565,20 +574,15 @@ hit_next: goto out; } - if (state->end < end && prealloc && !need_resched()) - next_node = rb_next(&state->rb_node); - else - next_node = NULL; - set |= clear_state_bit(tree, state, &bits, wake); +next: if (last_end == (u64)-1) goto out; start = last_end + 1; if (start <= end && next_node) { state = rb_entry(next_node, struct extent_state, rb_node); - if (state->start == start) - goto hit_next; + goto hit_next; } goto search_again; @@ -961,8 +965,6 @@ hit_next: set_state_bits(tree, state, &bits); clear_state_bit(tree, state, &clear_bits, 0); - - merge_state(tree, state); if (last_end == (u64)-1) goto out; @@ -1007,7 +1009,6 @@ hit_next: if (state->end <= end) { set_state_bits(tree, state, &bits); clear_state_bit(tree, state, &clear_bits, 0); - merge_state(tree, state); if (last_end == (u64)-1) goto out; start = last_end + 1; @@ -1068,8 +1069,6 @@ hit_next: set_state_bits(tree, prealloc, &bits); clear_state_bit(tree, prealloc, &clear_bits, 0); - - merge_state(tree, prealloc); prealloc = NULL; goto out; } @@ -2154,13 +2153,46 @@ static int bio_readpage_error(struct bio *failed_bio, struct page *page, "this_mirror=%d, num_copies=%d, in_validation=%d\n", read_mode, failrec->this_mirror, num_copies, failrec->in_validation); - tree->ops->submit_bio_hook(inode, read_mode, bio, failrec->this_mirror, - failrec->bio_flags, 0); - return 0; + ret = tree->ops->submit_bio_hook(inode, read_mode, bio, + failrec->this_mirror, + failrec->bio_flags, 0); + return ret; } /* lots and lots of room for performance fixes in the end_bio funcs */ +int end_extent_writepage(struct page *page, int err, u64 start, u64 end) +{ + int uptodate = (err == 0); + struct extent_io_tree *tree; + int ret; + + tree = &BTRFS_I(page->mapping->host)->io_tree; + + if (tree->ops && tree->ops->writepage_end_io_hook) { + ret = tree->ops->writepage_end_io_hook(page, start, + end, NULL, uptodate); + if (ret) + uptodate = 0; + } + + if (!uptodate && tree->ops && + tree->ops->writepage_io_failed_hook) { + ret = tree->ops->writepage_io_failed_hook(NULL, page, + start, end, NULL); + /* Writeback already completed */ + if (ret == 0) + return 1; + } + + if (!uptodate) { + clear_extent_uptodate(tree, start, end, NULL, GFP_NOFS); + ClearPageUptodate(page); + SetPageError(page); + } + return 0; +} + /* * after a writepage IO is done, we need to: * clear the uptodate bits on error @@ -2172,13 +2204,11 @@ static int bio_readpage_error(struct bio *failed_bio, struct page *page, */ static void end_bio_extent_writepage(struct bio *bio, int err) { - int uptodate = err == 0; struct bio_vec *bvec = bio->bi_io_vec + bio->bi_vcnt - 1; struct extent_io_tree *tree; u64 start; u64 end; int whole_page; - int ret; do { struct page *page = bvec->bv_page; @@ -2195,28 +2225,9 @@ static void end_bio_extent_writepage(struct bio *bio, int err) if (--bvec >= bio->bi_io_vec) prefetchw(&bvec->bv_page->flags); - if (tree->ops && tree->ops->writepage_end_io_hook) { - ret = tree->ops->writepage_end_io_hook(page, start, - end, NULL, uptodate); - if (ret) - uptodate = 0; - } - - if (!uptodate && tree->ops && - tree->ops->writepage_io_failed_hook) { - ret = tree->ops->writepage_io_failed_hook(bio, page, - start, end, NULL); - if (ret == 0) { - uptodate = (err == 0); - continue; - } - } - if (!uptodate) { - clear_extent_uptodate(tree, start, end, NULL, GFP_NOFS); - ClearPageUptodate(page); - SetPageError(page); - } + if (end_extent_writepage(page, err, start, end)) + continue; if (whole_page) end_page_writeback(page); @@ -2779,9 +2790,12 @@ static int __extent_writepage(struct page *page, struct writeback_control *wbc, delalloc_start = delalloc_end + 1; continue; } - tree->ops->fill_delalloc(inode, page, delalloc_start, - delalloc_end, &page_started, - &nr_written); + ret = tree->ops->fill_delalloc(inode, page, + delalloc_start, + delalloc_end, + &page_started, + &nr_written); + BUG_ON(ret); /* * delalloc_end is already one less than the total * length, so we don't subtract one from @@ -2818,8 +2832,12 @@ static int __extent_writepage(struct page *page, struct writeback_control *wbc, if (tree->ops && tree->ops->writepage_start_hook) { ret = tree->ops->writepage_start_hook(page, start, page_end); - if (ret == -EAGAIN) { - redirty_page_for_writepage(wbc, page); + if (ret) { + /* Fixup worker will requeue */ + if (ret == -EBUSY) + wbc->pages_skipped++; + else + redirty_page_for_writepage(wbc, page); update_nr_written(page, wbc, nr_written); unlock_page(page); ret = 0; @@ -3289,7 +3307,7 @@ int try_release_extent_mapping(struct extent_map_tree *map, len = end - start + 1; write_lock(&map->lock); em = lookup_extent_mapping(map, start, len); - if (IS_ERR_OR_NULL(em)) { + if (!em) { write_unlock(&map->lock); break; } @@ -3853,10 +3871,9 @@ int clear_extent_buffer_uptodate(struct extent_io_tree *tree, num_pages = num_extent_pages(eb->start, eb->len); clear_bit(EXTENT_BUFFER_UPTODATE, &eb->bflags); - if (eb_straddles_pages(eb)) { - clear_extent_uptodate(tree, eb->start, eb->start + eb->len - 1, - cached_state, GFP_NOFS); - } + clear_extent_uptodate(tree, eb->start, eb->start + eb->len - 1, + cached_state, GFP_NOFS); + for (i = 0; i < num_pages; i++) { page = extent_buffer_page(eb, i); if (page) diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h index bc6a042cb6f..cecc3518c12 100644 --- a/fs/btrfs/extent_io.h +++ b/fs/btrfs/extent_io.h @@ -319,4 +319,5 @@ struct btrfs_mapping_tree; int repair_io_failure(struct btrfs_mapping_tree *map_tree, u64 start, u64 length, u64 logical, struct page *page, int mirror_num); +int end_extent_writepage(struct page *page, int err, u64 start, u64 end); #endif diff --git a/fs/btrfs/extent_map.h b/fs/btrfs/extent_map.h index 33a7890b1f4..1195f09761f 100644 --- a/fs/btrfs/extent_map.h +++ b/fs/btrfs/extent_map.h @@ -26,8 +26,8 @@ struct extent_map { unsigned long flags; struct block_device *bdev; atomic_t refs; - unsigned int in_tree:1; - unsigned int compress_type:4; + unsigned int in_tree; + unsigned int compress_type; }; struct extent_map_tree { diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index 859ba2dd889..e8d06b6b919 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -1605,6 +1605,14 @@ static long btrfs_fallocate(struct file *file, int mode, return -EOPNOTSUPP; /* + * Make sure we have enough space before we do the + * allocation. + */ + ret = btrfs_check_data_free_space(inode, len); + if (ret) + return ret; + + /* * wait for ordered IO before we have any locks. We'll loop again * below with the locks held. */ @@ -1667,27 +1675,12 @@ static long btrfs_fallocate(struct file *file, int mode, if (em->block_start == EXTENT_MAP_HOLE || (cur_offset >= inode->i_size && !test_bit(EXTENT_FLAG_PREALLOC, &em->flags))) { - - /* - * Make sure we have enough space before we do the - * allocation. - */ - ret = btrfs_check_data_free_space(inode, last_byte - - cur_offset); - if (ret) { - free_extent_map(em); - break; - } - ret = btrfs_prealloc_file_range(inode, mode, cur_offset, last_byte - cur_offset, 1 << inode->i_blkbits, offset + len, &alloc_hint); - /* Let go of our reservation. */ - btrfs_free_reserved_data_space(inode, last_byte - - cur_offset); if (ret < 0) { free_extent_map(em); break; @@ -1715,6 +1708,8 @@ static long btrfs_fallocate(struct file *file, int mode, &cached_state, GFP_NOFS); out: mutex_unlock(&inode->i_mutex); + /* Let go of our reservation. */ + btrfs_free_reserved_data_space(inode, len); return ret; } @@ -1761,7 +1756,7 @@ static int find_desired_extent(struct inode *inode, loff_t *offset, int origin) start - root->sectorsize, root->sectorsize, 0); if (IS_ERR(em)) { - ret = -ENXIO; + ret = PTR_ERR(em); goto out; } last_end = em->start + em->len; @@ -1773,7 +1768,7 @@ static int find_desired_extent(struct inode *inode, loff_t *offset, int origin) while (1) { em = btrfs_get_extent_fiemap(inode, NULL, 0, start, len, 0); if (IS_ERR(em)) { - ret = -ENXIO; + ret = PTR_ERR(em); break; } diff --git a/fs/btrfs/free-space-cache.c b/fs/btrfs/free-space-cache.c index c2f20594c9f..710ea380c7e 100644 --- a/fs/btrfs/free-space-cache.c +++ b/fs/btrfs/free-space-cache.c @@ -777,6 +777,7 @@ int load_free_space_cache(struct btrfs_fs_info *fs_info, spin_lock(&block_group->lock); if (block_group->disk_cache_state != BTRFS_DC_WRITTEN) { spin_unlock(&block_group->lock); + btrfs_free_path(path); goto out; } spin_unlock(&block_group->lock); diff --git a/fs/btrfs/inode-map.c b/fs/btrfs/inode-map.c index 213ffa86ce1..ee15d88b33d 100644 --- a/fs/btrfs/inode-map.c +++ b/fs/btrfs/inode-map.c @@ -438,7 +438,8 @@ int btrfs_save_ino_cache(struct btrfs_root *root, trans->bytes_reserved); if (ret) goto out; - trace_btrfs_space_reservation(root->fs_info, "ino_cache", (u64)trans, + trace_btrfs_space_reservation(root->fs_info, "ino_cache", + (u64)(unsigned long)trans, trans->bytes_reserved, 1); again: inode = lookup_free_ino_inode(root, path); @@ -500,7 +501,8 @@ again: out_put: iput(inode); out_release: - trace_btrfs_space_reservation(root->fs_info, "ino_cache", (u64)trans, + trace_btrfs_space_reservation(root->fs_info, "ino_cache", + (u64)(unsigned long)trans, trans->bytes_reserved, 0); btrfs_block_rsv_release(root, trans->block_rsv, trans->bytes_reserved); out: diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 32214fe0f7e..892b34785cc 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -1555,6 +1555,7 @@ static void btrfs_writepage_fixup_worker(struct btrfs_work *work) struct inode *inode; u64 page_start; u64 page_end; + int ret; fixup = container_of(work, struct btrfs_writepage_fixup, work); page = fixup->page; @@ -1582,12 +1583,21 @@ again: page_end, &cached_state, GFP_NOFS); unlock_page(page); btrfs_start_ordered_extent(inode, ordered, 1); + btrfs_put_ordered_extent(ordered); goto again; } - BUG(); + ret = btrfs_delalloc_reserve_space(inode, PAGE_CACHE_SIZE); + if (ret) { + mapping_set_error(page->mapping, ret); + end_extent_writepage(page, ret, page_start, page_end); + ClearPageChecked(page); + goto out; + } + btrfs_set_extent_delalloc(inode, page_start, page_end, &cached_state); ClearPageChecked(page); + set_page_dirty(page); out: unlock_extent_cached(&BTRFS_I(inode)->io_tree, page_start, page_end, &cached_state, GFP_NOFS); @@ -1630,7 +1640,7 @@ static int btrfs_writepage_start_hook(struct page *page, u64 start, u64 end) fixup->work.func = btrfs_writepage_fixup_worker; fixup->page = page; btrfs_queue_worker(&root->fs_info->fixup_workers, &fixup->work); - return -EAGAIN; + return -EBUSY; } static int insert_reserved_file_extent(struct btrfs_trans_handle *trans, @@ -4575,7 +4585,8 @@ int btrfs_add_link(struct btrfs_trans_handle *trans, ret = btrfs_insert_dir_item(trans, root, name, name_len, parent_inode, &key, btrfs_inode_type(inode), index); - BUG_ON(ret); + if (ret) + goto fail_dir_item; btrfs_i_size_write(parent_inode, parent_inode->i_size + name_len * 2); @@ -4583,6 +4594,23 @@ int btrfs_add_link(struct btrfs_trans_handle *trans, ret = btrfs_update_inode(trans, root, parent_inode); } return ret; + +fail_dir_item: + if (unlikely(ino == BTRFS_FIRST_FREE_OBJECTID)) { + u64 local_index; + int err; + err = btrfs_del_root_ref(trans, root->fs_info->tree_root, + key.objectid, root->root_key.objectid, + parent_ino, &local_index, name, name_len); + + } else if (add_backref) { + u64 local_index; + int err; + + err = btrfs_del_inode_ref(trans, root, name, name_len, + ino, parent_ino, &local_index); + } + return ret; } static int btrfs_add_nondir(struct btrfs_trans_handle *trans, @@ -6696,8 +6724,10 @@ int btrfs_create_subvol_root(struct btrfs_trans_handle *trans, int err; u64 index = 0; - inode = btrfs_new_inode(trans, new_root, NULL, "..", 2, new_dirid, - new_dirid, S_IFDIR | 0700, &index); + inode = btrfs_new_inode(trans, new_root, NULL, "..", 2, + new_dirid, new_dirid, + S_IFDIR | (~current_umask() & S_IRWXUGO), + &index); if (IS_ERR(inode)) return PTR_ERR(inode); inode->i_op = &btrfs_dir_inode_operations; diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index 03bb62a9ee2..d8b54715c2d 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -861,6 +861,7 @@ static int cluster_pages_for_defrag(struct inode *inode, int i_done; struct btrfs_ordered_extent *ordered; struct extent_state *cached_state = NULL; + struct extent_io_tree *tree; gfp_t mask = btrfs_alloc_write_mask(inode->i_mapping); if (isize == 0) @@ -871,18 +872,34 @@ static int cluster_pages_for_defrag(struct inode *inode, num_pages << PAGE_CACHE_SHIFT); if (ret) return ret; -again: - ret = 0; i_done = 0; + tree = &BTRFS_I(inode)->io_tree; /* step one, lock all the pages */ for (i = 0; i < num_pages; i++) { struct page *page; +again: page = find_or_create_page(inode->i_mapping, - start_index + i, mask); + start_index + i, mask); if (!page) break; + page_start = page_offset(page); + page_end = page_start + PAGE_CACHE_SIZE - 1; + while (1) { + lock_extent(tree, page_start, page_end, GFP_NOFS); + ordered = btrfs_lookup_ordered_extent(inode, + page_start); + unlock_extent(tree, page_start, page_end, GFP_NOFS); + if (!ordered) + break; + + unlock_page(page); + btrfs_start_ordered_extent(inode, ordered, 1); + btrfs_put_ordered_extent(ordered); + lock_page(page); + } + if (!PageUptodate(page)) { btrfs_readpage(NULL, page); lock_page(page); @@ -893,15 +910,22 @@ again: break; } } + isize = i_size_read(inode); file_end = (isize - 1) >> PAGE_CACHE_SHIFT; - if (!isize || page->index > file_end || - page->mapping != inode->i_mapping) { + if (!isize || page->index > file_end) { /* whoops, we blew past eof, skip this page */ unlock_page(page); page_cache_release(page); break; } + + if (page->mapping != inode->i_mapping) { + unlock_page(page); + page_cache_release(page); + goto again; + } + pages[i] = page; i_done++; } @@ -924,25 +948,6 @@ again: lock_extent_bits(&BTRFS_I(inode)->io_tree, page_start, page_end - 1, 0, &cached_state, GFP_NOFS); - ordered = btrfs_lookup_first_ordered_extent(inode, page_end - 1); - if (ordered && - ordered->file_offset + ordered->len > page_start && - ordered->file_offset < page_end) { - btrfs_put_ordered_extent(ordered); - unlock_extent_cached(&BTRFS_I(inode)->io_tree, - page_start, page_end - 1, - &cached_state, GFP_NOFS); - for (i = 0; i < i_done; i++) { - unlock_page(pages[i]); - page_cache_release(pages[i]); - } - btrfs_wait_ordered_range(inode, page_start, - page_end - page_start); - goto again; - } - if (ordered) - btrfs_put_ordered_extent(ordered); - clear_extent_bit(&BTRFS_I(inode)->io_tree, page_start, page_end - 1, EXTENT_DIRTY | EXTENT_DELALLOC | EXTENT_DO_ACCOUNTING, 0, 0, &cached_state, @@ -1327,6 +1332,12 @@ static noinline int btrfs_ioctl_snap_create_transid(struct file *file, goto out; } + if (name[0] == '.' && + (namelen == 1 || (name[1] == '.' && namelen == 2))) { + ret = -EEXIST; + goto out; + } + if (subvol) { ret = btrfs_mksubvol(&file->f_path, name, namelen, NULL, transid, readonly); diff --git a/fs/btrfs/reada.c b/fs/btrfs/reada.c index 2373b39a132..22db04550f6 100644 --- a/fs/btrfs/reada.c +++ b/fs/btrfs/reada.c @@ -305,7 +305,7 @@ again: spin_lock(&fs_info->reada_lock); ret = radix_tree_insert(&dev->reada_zones, - (unsigned long)zone->end >> PAGE_CACHE_SHIFT, + (unsigned long)(zone->end >> PAGE_CACHE_SHIFT), zone); spin_unlock(&fs_info->reada_lock); diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c index 9770cc5bfb7..abc0fbffa51 100644 --- a/fs/btrfs/scrub.c +++ b/fs/btrfs/scrub.c @@ -1367,7 +1367,8 @@ out: } static noinline_for_stack int scrub_chunk(struct scrub_dev *sdev, - u64 chunk_tree, u64 chunk_objectid, u64 chunk_offset, u64 length) + u64 chunk_tree, u64 chunk_objectid, u64 chunk_offset, u64 length, + u64 dev_offset) { struct btrfs_mapping_tree *map_tree = &sdev->dev->dev_root->fs_info->mapping_tree; @@ -1391,7 +1392,8 @@ static noinline_for_stack int scrub_chunk(struct scrub_dev *sdev, goto out; for (i = 0; i < map->num_stripes; ++i) { - if (map->stripes[i].dev == sdev->dev) { + if (map->stripes[i].dev == sdev->dev && + map->stripes[i].physical == dev_offset) { ret = scrub_stripe(sdev, map, i, chunk_offset, length); if (ret) goto out; @@ -1487,7 +1489,7 @@ int scrub_enumerate_chunks(struct scrub_dev *sdev, u64 start, u64 end) break; } ret = scrub_chunk(sdev, chunk_tree, chunk_objectid, - chunk_offset, length); + chunk_offset, length, found_key.offset); btrfs_put_block_group(cache); if (ret) break; diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c index 287a6728b1a..04b77e3ceb7 100644 --- a/fs/btrfs/transaction.c +++ b/fs/btrfs/transaction.c @@ -327,7 +327,8 @@ again: if (num_bytes) { trace_btrfs_space_reservation(root->fs_info, "transaction", - (u64)h, num_bytes, 1); + (u64)(unsigned long)h, + num_bytes, 1); h->block_rsv = &root->fs_info->trans_block_rsv; h->bytes_reserved = num_bytes; } @@ -915,7 +916,11 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans, dentry->d_name.name, dentry->d_name.len, parent_inode, &key, BTRFS_FT_DIR, index); - BUG_ON(ret); + if (ret) { + pending->error = -EEXIST; + dput(parent); + goto fail; + } btrfs_i_size_write(parent_inode, parent_inode->i_size + dentry->d_name.len * 2); @@ -993,12 +998,9 @@ static noinline int create_pending_snapshots(struct btrfs_trans_handle *trans, { struct btrfs_pending_snapshot *pending; struct list_head *head = &trans->transaction->pending_snapshots; - int ret; - list_for_each_entry(pending, head, list) { - ret = create_pending_snapshot(trans, fs_info, pending); - BUG_ON(ret); - } + list_for_each_entry(pending, head, list) + create_pending_snapshot(trans, fs_info, pending); return 0; } diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index 0b4e2af7954..ef41f285a47 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -459,12 +459,23 @@ int btrfs_close_extra_devices(struct btrfs_fs_devices *fs_devices) { struct btrfs_device *device, *next; + struct block_device *latest_bdev = NULL; + u64 latest_devid = 0; + u64 latest_transid = 0; + mutex_lock(&uuid_mutex); again: /* This is the initialized path, it is safe to release the devices. */ list_for_each_entry_safe(device, next, &fs_devices->devices, dev_list) { - if (device->in_fs_metadata) + if (device->in_fs_metadata) { + if (!latest_transid || + device->generation > latest_transid) { + latest_devid = device->devid; + latest_transid = device->generation; + latest_bdev = device->bdev; + } continue; + } if (device->bdev) { blkdev_put(device->bdev, device->mode); @@ -487,6 +498,10 @@ again: goto again; } + fs_devices->latest_bdev = latest_bdev; + fs_devices->latest_devid = latest_devid; + fs_devices->latest_trans = latest_transid; + mutex_unlock(&uuid_mutex); return 0; } @@ -1953,7 +1968,7 @@ static int btrfs_relocate_chunk(struct btrfs_root *root, em = lookup_extent_mapping(em_tree, chunk_offset, 1); read_unlock(&em_tree->lock); - BUG_ON(em->start > chunk_offset || + BUG_ON(!em || em->start > chunk_offset || em->start + em->len < chunk_offset); map = (struct map_lookup *)em->bdev; @@ -4356,6 +4371,20 @@ int btrfs_read_sys_array(struct btrfs_root *root) return -ENOMEM; btrfs_set_buffer_uptodate(sb); btrfs_set_buffer_lockdep_class(root->root_key.objectid, sb, 0); + /* + * The sb extent buffer is artifical and just used to read the system array. + * btrfs_set_buffer_uptodate() call does not properly mark all it's + * pages up-to-date when the page is larger: extent does not cover the + * whole page and consequently check_page_uptodate does not find all + * the page's extents up-to-date (the hole beyond sb), + * write_extent_buffer then triggers a WARN_ON. + * + * Regular short extents go through mark_extent_buffer_dirty/writeback cycle, + * but sb spans only this function. Add an explicit SetPageUptodate call + * to silence the warning eg. on PowerPC 64. + */ + if (PAGE_CACHE_SIZE > BTRFS_SUPER_INFO_SIZE) + SetPageUptodate(sb->first_page); write_extent_buffer(sb, super_copy, 0, BTRFS_SUPER_INFO_SIZE); array_size = btrfs_super_sys_array_size(super_copy); diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index 9c288653e6d..602f77c304c 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -773,10 +773,11 @@ standard_receive3(struct TCP_Server_Info *server, struct mid_q_entry *mid) cifs_dump_mem("Bad SMB: ", buf, min_t(unsigned int, server->total_read, 48)); - if (mid) - handle_mid(mid, server, smb_buffer, length); + if (!mid) + return length; - return length; + handle_mid(mid, server, smb_buffer, length); + return 0; } static int @@ -2125,7 +2126,7 @@ cifs_set_cifscreds(struct smb_vol *vol, struct cifs_ses *ses) down_read(&key->sem); upayload = key->payload.data; if (IS_ERR_OR_NULL(upayload)) { - rc = PTR_ERR(key); + rc = upayload ? PTR_ERR(upayload) : -EINVAL; goto out_key_put; } diff --git a/fs/cifs/dir.c b/fs/cifs/dir.c index df8fecb5b99..bc7e24420ac 100644 --- a/fs/cifs/dir.c +++ b/fs/cifs/dir.c @@ -492,7 +492,7 @@ cifs_lookup(struct inode *parent_dir_inode, struct dentry *direntry, { int xid; int rc = 0; /* to get around spurious gcc warning, set to zero here */ - __u32 oplock = 0; + __u32 oplock = enable_oplocks ? REQ_OPLOCK : 0; __u16 fileHandle = 0; bool posix_open = false; struct cifs_sb_info *cifs_sb; @@ -584,10 +584,26 @@ cifs_lookup(struct inode *parent_dir_inode, struct dentry *direntry, * If either that or op not supported returned, follow * the normal lookup. */ - if ((rc == 0) || (rc == -ENOENT)) + switch (rc) { + case 0: + /* + * The server may allow us to open things like + * FIFOs, but the client isn't set up to deal + * with that. If it's not a regular file, just + * close it and proceed as if it were a normal + * lookup. + */ + if (newInode && !S_ISREG(newInode->i_mode)) { + CIFSSMBClose(xid, pTcon, fileHandle); + break; + } + case -ENOENT: posix_open = true; - else if ((rc == -EINVAL) || (rc != -EOPNOTSUPP)) + case -EOPNOTSUPP: + break; + default: pTcon->broken_posix_open = true; + } } if (!posix_open) rc = cifs_get_inode_info_unix(&newInode, full_path, diff --git a/fs/cifs/file.c b/fs/cifs/file.c index 4dd9283885e..5e64748a291 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -920,16 +920,26 @@ cifs_push_mandatory_locks(struct cifsFileInfo *cfile) for (lockp = &inode->i_flock; *lockp != NULL; \ lockp = &(*lockp)->fl_next) +struct lock_to_push { + struct list_head llist; + __u64 offset; + __u64 length; + __u32 pid; + __u16 netfid; + __u8 type; +}; + static int cifs_push_posix_locks(struct cifsFileInfo *cfile) { struct cifsInodeInfo *cinode = CIFS_I(cfile->dentry->d_inode); struct cifs_tcon *tcon = tlink_tcon(cfile->tlink); struct file_lock *flock, **before; - struct cifsLockInfo *lck, *tmp; + unsigned int count = 0, i = 0; int rc = 0, xid, type; + struct list_head locks_to_send, *el; + struct lock_to_push *lck, *tmp; __u64 length; - struct list_head locks_to_send; xid = GetXid(); @@ -940,29 +950,55 @@ cifs_push_posix_locks(struct cifsFileInfo *cfile) return rc; } + lock_flocks(); + cifs_for_each_lock(cfile->dentry->d_inode, before) { + if ((*before)->fl_flags & FL_POSIX) + count++; + } + unlock_flocks(); + INIT_LIST_HEAD(&locks_to_send); + /* + * Allocating count locks is enough because no locks can be added to + * the list while we are holding cinode->lock_mutex that protects + * locking operations of this inode. + */ + for (; i < count; i++) { + lck = kmalloc(sizeof(struct lock_to_push), GFP_KERNEL); + if (!lck) { + rc = -ENOMEM; + goto err_out; + } + list_add_tail(&lck->llist, &locks_to_send); + } + + i = 0; + el = locks_to_send.next; lock_flocks(); cifs_for_each_lock(cfile->dentry->d_inode, before) { + if (el == &locks_to_send) { + /* something is really wrong */ + cERROR(1, "Can't push all brlocks!"); + break; + } flock = *before; + if ((flock->fl_flags & FL_POSIX) == 0) + continue; length = 1 + flock->fl_end - flock->fl_start; if (flock->fl_type == F_RDLCK || flock->fl_type == F_SHLCK) type = CIFS_RDLCK; else type = CIFS_WRLCK; - - lck = cifs_lock_init(flock->fl_start, length, type, - cfile->netfid); - if (!lck) { - rc = -ENOMEM; - goto send_locks; - } + lck = list_entry(el, struct lock_to_push, llist); lck->pid = flock->fl_pid; - - list_add_tail(&lck->llist, &locks_to_send); + lck->netfid = cfile->netfid; + lck->length = length; + lck->type = type; + lck->offset = flock->fl_start; + i++; + el = el->next; } - -send_locks: unlock_flocks(); list_for_each_entry_safe(lck, tmp, &locks_to_send, llist) { @@ -979,11 +1015,18 @@ send_locks: kfree(lck); } +out: cinode->can_cache_brlcks = false; mutex_unlock(&cinode->lock_mutex); FreeXid(xid); return rc; +err_out: + list_for_each_entry_safe(lck, tmp, &locks_to_send, llist) { + list_del(&lck->llist); + kfree(lck); + } + goto out; } static int diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c index a5f54b7d982..745da3d0653 100644 --- a/fs/cifs/inode.c +++ b/fs/cifs/inode.c @@ -534,6 +534,11 @@ cifs_all_info_to_fattr(struct cifs_fattr *fattr, FILE_ALL_INFO *info, if (fattr->cf_cifsattrs & ATTR_DIRECTORY) { fattr->cf_mode = S_IFDIR | cifs_sb->mnt_dir_mode; fattr->cf_dtype = DT_DIR; + /* + * Server can return wrong NumberOfLinks value for directories + * when Unix extensions are disabled - fake it. + */ + fattr->cf_nlink = 2; } else { fattr->cf_mode = S_IFREG | cifs_sb->mnt_file_mode; fattr->cf_dtype = DT_REG; @@ -541,9 +546,9 @@ cifs_all_info_to_fattr(struct cifs_fattr *fattr, FILE_ALL_INFO *info, /* clear write bits if ATTR_READONLY is set */ if (fattr->cf_cifsattrs & ATTR_READONLY) fattr->cf_mode &= ~(S_IWUGO); - } - fattr->cf_nlink = le32_to_cpu(info->NumberOfLinks); + fattr->cf_nlink = le32_to_cpu(info->NumberOfLinks); + } fattr->cf_uid = cifs_sb->mnt_uid; fattr->cf_gid = cifs_sb->mnt_gid; @@ -1322,7 +1327,6 @@ int cifs_mkdir(struct inode *inode, struct dentry *direntry, umode_t mode) } /*BB check (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SET_UID ) to see if need to set uid/gid */ - inc_nlink(inode); cifs_unix_basic_to_fattr(&fattr, pInfo, cifs_sb); cifs_fill_uniqueid(inode->i_sb, &fattr); @@ -1355,7 +1359,6 @@ mkdir_retry_old: d_drop(direntry); } else { mkdir_get_info: - inc_nlink(inode); if (pTcon->unix_ext) rc = cifs_get_inode_info_unix(&newinode, full_path, inode->i_sb, xid); @@ -1436,6 +1439,11 @@ mkdir_get_info: } } mkdir_out: + /* + * Force revalidate to get parent dir info when needed since cached + * attributes are invalid now. + */ + CIFS_I(inode)->time = 0; kfree(full_path); FreeXid(xid); cifs_put_tlink(tlink); @@ -1475,7 +1483,6 @@ int cifs_rmdir(struct inode *inode, struct dentry *direntry) cifs_put_tlink(tlink); if (!rc) { - drop_nlink(inode); spin_lock(&direntry->d_inode->i_lock); i_size_write(direntry->d_inode, 0); clear_nlink(direntry->d_inode); @@ -1483,12 +1490,15 @@ int cifs_rmdir(struct inode *inode, struct dentry *direntry) } cifsInode = CIFS_I(direntry->d_inode); - cifsInode->time = 0; /* force revalidate to go get info when - needed */ + /* force revalidate to go get info when needed */ + cifsInode->time = 0; cifsInode = CIFS_I(inode); - cifsInode->time = 0; /* force revalidate to get parent dir info - since cached search results now invalid */ + /* + * Force revalidate to get parent dir info when needed since cached + * attributes are invalid now. + */ + cifsInode->time = 0; direntry->d_inode->i_ctime = inode->i_ctime = inode->i_mtime = current_fs_time(inode->i_sb); diff --git a/fs/cifs/xattr.c b/fs/cifs/xattr.c index 45f07c46f3e..10d92cf57ab 100644 --- a/fs/cifs/xattr.c +++ b/fs/cifs/xattr.c @@ -105,7 +105,6 @@ int cifs_setxattr(struct dentry *direntry, const char *ea_name, struct cifs_tcon *pTcon; struct super_block *sb; char *full_path; - struct cifs_ntsd *pacl; if (direntry == NULL) return -EIO; @@ -164,23 +163,24 @@ int cifs_setxattr(struct dentry *direntry, const char *ea_name, cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); } else if (strncmp(ea_name, CIFS_XATTR_CIFS_ACL, strlen(CIFS_XATTR_CIFS_ACL)) == 0) { +#ifdef CONFIG_CIFS_ACL + struct cifs_ntsd *pacl; pacl = kmalloc(value_size, GFP_KERNEL); if (!pacl) { cFYI(1, "%s: Can't allocate memory for ACL", __func__); rc = -ENOMEM; } else { -#ifdef CONFIG_CIFS_ACL memcpy(pacl, ea_value, value_size); rc = set_cifs_acl(pacl, value_size, direntry->d_inode, full_path, CIFS_ACL_DACL); if (rc == 0) /* force revalidate of the inode */ CIFS_I(direntry->d_inode)->time = 0; kfree(pacl); + } #else cFYI(1, "Set CIFS ACL not supported yet"); #endif /* CONFIG_CIFS_ACL */ - } } else { int temp; temp = strncmp(ea_name, POSIX_ACL_XATTR_ACCESS, diff --git a/fs/compat.c b/fs/compat.c index fa9d721ecfe..07880bae28a 100644 --- a/fs/compat.c +++ b/fs/compat.c @@ -131,41 +131,35 @@ asmlinkage long compat_sys_utimes(const char __user *filename, struct compat_tim static int cp_compat_stat(struct kstat *stat, struct compat_stat __user *ubuf) { - compat_ino_t ino = stat->ino; - typeof(ubuf->st_uid) uid = 0; - typeof(ubuf->st_gid) gid = 0; - int err; + struct compat_stat tmp; - SET_UID(uid, stat->uid); - SET_GID(gid, stat->gid); + if (!old_valid_dev(stat->dev) || !old_valid_dev(stat->rdev)) + return -EOVERFLOW; - if ((u64) stat->size > MAX_NON_LFS || - !old_valid_dev(stat->dev) || - !old_valid_dev(stat->rdev)) + memset(&tmp, 0, sizeof(tmp)); + tmp.st_dev = old_encode_dev(stat->dev); + tmp.st_ino = stat->ino; + if (sizeof(tmp.st_ino) < sizeof(stat->ino) && tmp.st_ino != stat->ino) return -EOVERFLOW; - if (sizeof(ino) < sizeof(stat->ino) && ino != stat->ino) + tmp.st_mode = stat->mode; + tmp.st_nlink = stat->nlink; + if (tmp.st_nlink != stat->nlink) return -EOVERFLOW; - - if (clear_user(ubuf, sizeof(*ubuf))) - return -EFAULT; - - err = __put_user(old_encode_dev(stat->dev), &ubuf->st_dev); - err |= __put_user(ino, &ubuf->st_ino); - err |= __put_user(stat->mode, &ubuf->st_mode); - err |= __put_user(stat->nlink, &ubuf->st_nlink); - err |= __put_user(uid, &ubuf->st_uid); - err |= __put_user(gid, &ubuf->st_gid); - err |= __put_user(old_encode_dev(stat->rdev), &ubuf->st_rdev); - err |= __put_user(stat->size, &ubuf->st_size); - err |= __put_user(stat->atime.tv_sec, &ubuf->st_atime); - err |= __put_user(stat->atime.tv_nsec, &ubuf->st_atime_nsec); - err |= __put_user(stat->mtime.tv_sec, &ubuf->st_mtime); - err |= __put_user(stat->mtime.tv_nsec, &ubuf->st_mtime_nsec); - err |= __put_user(stat->ctime.tv_sec, &ubuf->st_ctime); - err |= __put_user(stat->ctime.tv_nsec, &ubuf->st_ctime_nsec); - err |= __put_user(stat->blksize, &ubuf->st_blksize); - err |= __put_user(stat->blocks, &ubuf->st_blocks); - return err; + SET_UID(tmp.st_uid, stat->uid); + SET_GID(tmp.st_gid, stat->gid); + tmp.st_rdev = old_encode_dev(stat->rdev); + if ((u64) stat->size > MAX_NON_LFS) + return -EOVERFLOW; + tmp.st_size = stat->size; + tmp.st_atime = stat->atime.tv_sec; + tmp.st_atime_nsec = stat->atime.tv_nsec; + tmp.st_mtime = stat->mtime.tv_sec; + tmp.st_mtime_nsec = stat->mtime.tv_nsec; + tmp.st_ctime = stat->ctime.tv_sec; + tmp.st_ctime_nsec = stat->ctime.tv_nsec; + tmp.st_blocks = stat->blocks; + tmp.st_blksize = stat->blksize; + return copy_to_user(ubuf, &tmp, sizeof(tmp)) ? -EFAULT : 0; } asmlinkage long compat_sys_newstat(const char __user * filename, diff --git a/fs/dcache.c b/fs/dcache.c index 16a53cc2cc0..11828de68dc 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -104,11 +104,11 @@ static unsigned int d_hash_shift __read_mostly; static struct hlist_bl_head *dentry_hashtable __read_mostly; -static inline struct hlist_bl_head *d_hash(struct dentry *parent, - unsigned long hash) +static inline struct hlist_bl_head *d_hash(const struct dentry *parent, + unsigned int hash) { - hash += ((unsigned long) parent ^ GOLDEN_RATIO_PRIME) / L1_CACHE_BYTES; - hash = hash ^ ((hash ^ GOLDEN_RATIO_PRIME) >> D_HASHBITS); + hash += (unsigned long) parent / L1_CACHE_BYTES; + hash = hash + (hash >> D_HASHBITS); return dentry_hashtable + (hash & D_HASHMASK); } @@ -137,6 +137,49 @@ int proc_nr_dentry(ctl_table *table, int write, void __user *buffer, } #endif +/* + * Compare 2 name strings, return 0 if they match, otherwise non-zero. + * The strings are both count bytes long, and count is non-zero. + */ +static inline int dentry_cmp(const unsigned char *cs, size_t scount, + const unsigned char *ct, size_t tcount) +{ +#ifdef CONFIG_DCACHE_WORD_ACCESS + unsigned long a,b,mask; + + if (unlikely(scount != tcount)) + return 1; + + for (;;) { + a = *(unsigned long *)cs; + b = *(unsigned long *)ct; + if (tcount < sizeof(unsigned long)) + break; + if (unlikely(a != b)) + return 1; + cs += sizeof(unsigned long); + ct += sizeof(unsigned long); + tcount -= sizeof(unsigned long); + if (!tcount) + return 0; + } + mask = ~(~0ul << tcount*8); + return unlikely(!!((a ^ b) & mask)); +#else + if (scount != tcount) + return 1; + + do { + if (*cs != *ct) + return 1; + cs++; + ct++; + tcount--; + } while (tcount); + return 0; +#endif +} + static void __d_free(struct rcu_head *head) { struct dentry *dentry = container_of(head, struct dentry, d_u.d_rcu); @@ -1717,8 +1760,9 @@ EXPORT_SYMBOL(d_add_ci); * child is looked up. Thus, an interlocking stepping of sequence lock checks * is formed, giving integrity down the path walk. */ -struct dentry *__d_lookup_rcu(struct dentry *parent, struct qstr *name, - unsigned *seq, struct inode **inode) +struct dentry *__d_lookup_rcu(const struct dentry *parent, + const struct qstr *name, + unsigned *seqp, struct inode **inode) { unsigned int len = name->len; unsigned int hash = name->hash; @@ -1748,6 +1792,7 @@ struct dentry *__d_lookup_rcu(struct dentry *parent, struct qstr *name, * See Documentation/filesystems/path-lookup.txt for more details. */ hlist_bl_for_each_entry_rcu(dentry, node, b, d_hash) { + unsigned seq; struct inode *i; const char *tname; int tlen; @@ -1756,7 +1801,7 @@ struct dentry *__d_lookup_rcu(struct dentry *parent, struct qstr *name, continue; seqretry: - *seq = read_seqcount_begin(&dentry->d_seq); + seq = read_seqcount_begin(&dentry->d_seq); if (dentry->d_parent != parent) continue; if (d_unhashed(dentry)) @@ -1771,7 +1816,7 @@ seqretry: * edge of memory when walking. If we could load this * atomically some other way, we could drop this check. */ - if (read_seqcount_retry(&dentry->d_seq, *seq)) + if (read_seqcount_retry(&dentry->d_seq, seq)) goto seqretry; if (unlikely(parent->d_flags & DCACHE_OP_COMPARE)) { if (parent->d_op->d_compare(parent, *inode, @@ -1788,6 +1833,7 @@ seqretry: * order to do anything useful with the returned dentry * anyway. */ + *seqp = seq; *inode = i; return dentry; } @@ -2968,7 +3014,7 @@ __setup("dhash_entries=", set_dhash_entries); static void __init dcache_init_early(void) { - int loop; + unsigned int loop; /* If hashes are distributed across NUMA nodes, defer * hash allocation until vmalloc space is available. @@ -2986,13 +3032,13 @@ static void __init dcache_init_early(void) &d_hash_mask, 0); - for (loop = 0; loop < (1 << d_hash_shift); loop++) + for (loop = 0; loop < (1U << d_hash_shift); loop++) INIT_HLIST_BL_HEAD(dentry_hashtable + loop); } static void __init dcache_init(void) { - int loop; + unsigned int loop; /* * A constructor could be added for stable state like the lists, @@ -3016,7 +3062,7 @@ static void __init dcache_init(void) &d_hash_mask, 0); - for (loop = 0; loop < (1 << d_hash_shift); loop++) + for (loop = 0; loop < (1U << d_hash_shift); loop++) INIT_HLIST_BL_HEAD(dentry_hashtable + loop); } diff --git a/fs/debugfs/inode.c b/fs/debugfs/inode.c index 956d5ddddf6..b80bc846a15 100644 --- a/fs/debugfs/inode.c +++ b/fs/debugfs/inode.c @@ -23,9 +23,13 @@ #include <linux/debugfs.h> #include <linux/fsnotify.h> #include <linux/string.h> +#include <linux/seq_file.h> +#include <linux/parser.h> #include <linux/magic.h> #include <linux/slab.h> +#define DEBUGFS_DEFAULT_MODE 0755 + static struct vfsmount *debugfs_mount; static int debugfs_mount_count; static bool debugfs_registered; @@ -125,11 +129,154 @@ static inline int debugfs_positive(struct dentry *dentry) return dentry->d_inode && !d_unhashed(dentry); } +struct debugfs_mount_opts { + uid_t uid; + gid_t gid; + umode_t mode; +}; + +enum { + Opt_uid, + Opt_gid, + Opt_mode, + Opt_err +}; + +static const match_table_t tokens = { + {Opt_uid, "uid=%u"}, + {Opt_gid, "gid=%u"}, + {Opt_mode, "mode=%o"}, + {Opt_err, NULL} +}; + +struct debugfs_fs_info { + struct debugfs_mount_opts mount_opts; +}; + +static int debugfs_parse_options(char *data, struct debugfs_mount_opts *opts) +{ + substring_t args[MAX_OPT_ARGS]; + int option; + int token; + char *p; + + opts->mode = DEBUGFS_DEFAULT_MODE; + + while ((p = strsep(&data, ",")) != NULL) { + if (!*p) + continue; + + token = match_token(p, tokens, args); + switch (token) { + case Opt_uid: + if (match_int(&args[0], &option)) + return -EINVAL; + opts->uid = option; + break; + case Opt_gid: + if (match_octal(&args[0], &option)) + return -EINVAL; + opts->gid = option; + break; + case Opt_mode: + if (match_octal(&args[0], &option)) + return -EINVAL; + opts->mode = option & S_IALLUGO; + break; + /* + * We might like to report bad mount options here; + * but traditionally debugfs has ignored all mount options + */ + } + } + + return 0; +} + +static int debugfs_apply_options(struct super_block *sb) +{ + struct debugfs_fs_info *fsi = sb->s_fs_info; + struct inode *inode = sb->s_root->d_inode; + struct debugfs_mount_opts *opts = &fsi->mount_opts; + + inode->i_mode &= ~S_IALLUGO; + inode->i_mode |= opts->mode; + + inode->i_uid = opts->uid; + inode->i_gid = opts->gid; + + return 0; +} + +static int debugfs_remount(struct super_block *sb, int *flags, char *data) +{ + int err; + struct debugfs_fs_info *fsi = sb->s_fs_info; + + err = debugfs_parse_options(data, &fsi->mount_opts); + if (err) + goto fail; + + debugfs_apply_options(sb); + +fail: + return err; +} + +static int debugfs_show_options(struct seq_file *m, struct dentry *root) +{ + struct debugfs_fs_info *fsi = root->d_sb->s_fs_info; + struct debugfs_mount_opts *opts = &fsi->mount_opts; + + if (opts->uid != 0) + seq_printf(m, ",uid=%u", opts->uid); + if (opts->gid != 0) + seq_printf(m, ",gid=%u", opts->gid); + if (opts->mode != DEBUGFS_DEFAULT_MODE) + seq_printf(m, ",mode=%o", opts->mode); + + return 0; +} + +static const struct super_operations debugfs_super_operations = { + .statfs = simple_statfs, + .remount_fs = debugfs_remount, + .show_options = debugfs_show_options, +}; + static int debug_fill_super(struct super_block *sb, void *data, int silent) { static struct tree_descr debug_files[] = {{""}}; + struct debugfs_fs_info *fsi; + int err; + + save_mount_options(sb, data); + + fsi = kzalloc(sizeof(struct debugfs_fs_info), GFP_KERNEL); + sb->s_fs_info = fsi; + if (!fsi) { + err = -ENOMEM; + goto fail; + } + + err = debugfs_parse_options(data, &fsi->mount_opts); + if (err) + goto fail; + + err = simple_fill_super(sb, DEBUGFS_MAGIC, debug_files); + if (err) + goto fail; + + sb->s_op = &debugfs_super_operations; + + debugfs_apply_options(sb); + + return 0; - return simple_fill_super(sb, DEBUGFS_MAGIC, debug_files); +fail: + kfree(fsi); + sb->s_fs_info = NULL; + return err; } static struct dentry *debug_mount(struct file_system_type *fs_type, diff --git a/fs/direct-io.c b/fs/direct-io.c index 4a588dbd11b..f4aadd15b61 100644 --- a/fs/direct-io.c +++ b/fs/direct-io.c @@ -173,7 +173,7 @@ void inode_dio_wait(struct inode *inode) if (atomic_read(&inode->i_dio_count)) __inode_dio_wait(inode); } -EXPORT_SYMBOL_GPL(inode_dio_wait); +EXPORT_SYMBOL(inode_dio_wait); /* * inode_dio_done - signal finish of a direct I/O requests @@ -187,7 +187,7 @@ void inode_dio_done(struct inode *inode) if (atomic_dec_and_test(&inode->i_dio_count)) wake_up_bit(&inode->i_state, __I_DIO_WAKEUP); } -EXPORT_SYMBOL_GPL(inode_dio_done); +EXPORT_SYMBOL(inode_dio_done); /* * How many pages are in the queue? diff --git a/fs/ecryptfs/crypto.c b/fs/ecryptfs/crypto.c index 63ab2451064..ea993128155 100644 --- a/fs/ecryptfs/crypto.c +++ b/fs/ecryptfs/crypto.c @@ -1990,6 +1990,17 @@ out: return; } +static size_t ecryptfs_max_decoded_size(size_t encoded_size) +{ + /* Not exact; conservatively long. Every block of 4 + * encoded characters decodes into a block of 3 + * decoded characters. This segment of code provides + * the caller with the maximum amount of allocated + * space that @dst will need to point to in a + * subsequent call. */ + return ((encoded_size + 1) * 3) / 4; +} + /** * ecryptfs_decode_from_filename * @dst: If NULL, this function only sets @dst_size and returns. If @@ -2008,13 +2019,7 @@ ecryptfs_decode_from_filename(unsigned char *dst, size_t *dst_size, size_t dst_byte_offset = 0; if (dst == NULL) { - /* Not exact; conservatively long. Every block of 4 - * encoded characters decodes into a block of 3 - * decoded characters. This segment of code provides - * the caller with the maximum amount of allocated - * space that @dst will need to point to in a - * subsequent call. */ - (*dst_size) = (((src_size + 1) * 3) / 4); + (*dst_size) = ecryptfs_max_decoded_size(src_size); goto out; } while (src_byte_offset < src_size) { @@ -2239,3 +2244,52 @@ out_free: out: return rc; } + +#define ENC_NAME_MAX_BLOCKLEN_8_OR_16 143 + +int ecryptfs_set_f_namelen(long *namelen, long lower_namelen, + struct ecryptfs_mount_crypt_stat *mount_crypt_stat) +{ + struct blkcipher_desc desc; + struct mutex *tfm_mutex; + size_t cipher_blocksize; + int rc; + + if (!(mount_crypt_stat->flags & ECRYPTFS_GLOBAL_ENCRYPT_FILENAMES)) { + (*namelen) = lower_namelen; + return 0; + } + + rc = ecryptfs_get_tfm_and_mutex_for_cipher_name(&desc.tfm, &tfm_mutex, + mount_crypt_stat->global_default_fn_cipher_name); + if (unlikely(rc)) { + (*namelen) = 0; + return rc; + } + + mutex_lock(tfm_mutex); + cipher_blocksize = crypto_blkcipher_blocksize(desc.tfm); + mutex_unlock(tfm_mutex); + + /* Return an exact amount for the common cases */ + if (lower_namelen == NAME_MAX + && (cipher_blocksize == 8 || cipher_blocksize == 16)) { + (*namelen) = ENC_NAME_MAX_BLOCKLEN_8_OR_16; + return 0; + } + + /* Return a safe estimate for the uncommon cases */ + (*namelen) = lower_namelen; + (*namelen) -= ECRYPTFS_FNEK_ENCRYPTED_FILENAME_PREFIX_SIZE; + /* Since this is the max decoded size, subtract 1 "decoded block" len */ + (*namelen) = ecryptfs_max_decoded_size(*namelen) - 3; + (*namelen) -= ECRYPTFS_TAG_70_MAX_METADATA_SIZE; + (*namelen) -= ECRYPTFS_FILENAME_MIN_RANDOM_PREPEND_BYTES; + /* Worst case is that the filename is padded nearly a full block size */ + (*namelen) -= cipher_blocksize - 1; + + if ((*namelen) < 0) + (*namelen) = 0; + + return 0; +} diff --git a/fs/ecryptfs/ecryptfs_kernel.h b/fs/ecryptfs/ecryptfs_kernel.h index a2362df58ae..867b64c5d84 100644 --- a/fs/ecryptfs/ecryptfs_kernel.h +++ b/fs/ecryptfs/ecryptfs_kernel.h @@ -162,6 +162,10 @@ ecryptfs_get_key_payload_data(struct key *key) #define ECRYPTFS_NON_NULL 0x42 /* A reasonable substitute for NULL */ #define MD5_DIGEST_SIZE 16 #define ECRYPTFS_TAG_70_DIGEST_SIZE MD5_DIGEST_SIZE +#define ECRYPTFS_TAG_70_MIN_METADATA_SIZE (1 + ECRYPTFS_MIN_PKT_LEN_SIZE \ + + ECRYPTFS_SIG_SIZE + 1 + 1) +#define ECRYPTFS_TAG_70_MAX_METADATA_SIZE (1 + ECRYPTFS_MAX_PKT_LEN_SIZE \ + + ECRYPTFS_SIG_SIZE + 1 + 1) #define ECRYPTFS_FEK_ENCRYPTED_FILENAME_PREFIX "ECRYPTFS_FEK_ENCRYPTED." #define ECRYPTFS_FEK_ENCRYPTED_FILENAME_PREFIX_SIZE 23 #define ECRYPTFS_FNEK_ENCRYPTED_FILENAME_PREFIX "ECRYPTFS_FNEK_ENCRYPTED." @@ -701,6 +705,8 @@ ecryptfs_parse_tag_70_packet(char **filename, size_t *filename_size, size_t *packet_size, struct ecryptfs_mount_crypt_stat *mount_crypt_stat, char *data, size_t max_packet_size); +int ecryptfs_set_f_namelen(long *namelen, long lower_namelen, + struct ecryptfs_mount_crypt_stat *mount_crypt_stat); int ecryptfs_derive_iv(char *iv, struct ecryptfs_crypt_stat *crypt_stat, loff_t offset); diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c index 19892d7d2ed..ab35b113003 100644 --- a/fs/ecryptfs/inode.c +++ b/fs/ecryptfs/inode.c @@ -1085,6 +1085,8 @@ ecryptfs_setxattr(struct dentry *dentry, const char *name, const void *value, } rc = vfs_setxattr(lower_dentry, name, value, size, flags); + if (!rc) + fsstack_copy_attr_all(dentry->d_inode, lower_dentry->d_inode); out: return rc; } diff --git a/fs/ecryptfs/keystore.c b/fs/ecryptfs/keystore.c index 8e3b943e330..2333203a120 100644 --- a/fs/ecryptfs/keystore.c +++ b/fs/ecryptfs/keystore.c @@ -679,10 +679,7 @@ ecryptfs_write_tag_70_packet(char *dest, size_t *remaining_bytes, * Octets N3-N4: Block-aligned encrypted filename * - Consists of a minimum number of random characters, a \0 * separator, and then the filename */ - s->max_packet_size = (1 /* Tag 70 identifier */ - + 3 /* Max Tag 70 packet size */ - + ECRYPTFS_SIG_SIZE /* FNEK sig */ - + 1 /* Cipher identifier */ + s->max_packet_size = (ECRYPTFS_TAG_70_MAX_METADATA_SIZE + s->block_aligned_filename_size); if (dest == NULL) { (*packet_size) = s->max_packet_size; @@ -934,10 +931,10 @@ ecryptfs_parse_tag_70_packet(char **filename, size_t *filename_size, goto out; } s->desc.flags = CRYPTO_TFM_REQ_MAY_SLEEP; - if (max_packet_size < (1 + 1 + ECRYPTFS_SIG_SIZE + 1 + 1)) { + if (max_packet_size < ECRYPTFS_TAG_70_MIN_METADATA_SIZE) { printk(KERN_WARNING "%s: max_packet_size is [%zd]; it must be " "at least [%d]\n", __func__, max_packet_size, - (1 + 1 + ECRYPTFS_SIG_SIZE + 1 + 1)); + ECRYPTFS_TAG_70_MIN_METADATA_SIZE); rc = -EINVAL; goto out; } diff --git a/fs/ecryptfs/miscdev.c b/fs/ecryptfs/miscdev.c index 349209dc6a9..3a06f4043df 100644 --- a/fs/ecryptfs/miscdev.c +++ b/fs/ecryptfs/miscdev.c @@ -429,7 +429,7 @@ ecryptfs_miscdev_write(struct file *file, const char __user *buf, goto memdup; } else if (count < MIN_MSG_PKT_SIZE || count > MAX_MSG_PKT_SIZE) { printk(KERN_WARNING "%s: Acceptable packet size range is " - "[%d-%lu], but amount of data written is [%zu].", + "[%d-%zu], but amount of data written is [%zu].", __func__, MIN_MSG_PKT_SIZE, MAX_MSG_PKT_SIZE, count); return -EINVAL; } diff --git a/fs/ecryptfs/mmap.c b/fs/ecryptfs/mmap.c index 10ec695ccd6..a46b3a8fee1 100644 --- a/fs/ecryptfs/mmap.c +++ b/fs/ecryptfs/mmap.c @@ -150,7 +150,7 @@ ecryptfs_copy_up_encrypted_with_header(struct page *page, /* This is a header extent */ char *page_virt; - page_virt = kmap_atomic(page, KM_USER0); + page_virt = kmap_atomic(page); memset(page_virt, 0, PAGE_CACHE_SIZE); /* TODO: Support more than one header extent */ if (view_extent_num == 0) { @@ -163,7 +163,7 @@ ecryptfs_copy_up_encrypted_with_header(struct page *page, crypt_stat, &written); } - kunmap_atomic(page_virt, KM_USER0); + kunmap_atomic(page_virt); flush_dcache_page(page); if (rc) { printk(KERN_ERR "%s: Error reading xattr " diff --git a/fs/ecryptfs/read_write.c b/fs/ecryptfs/read_write.c index 5c0106f7577..b2a34a192f4 100644 --- a/fs/ecryptfs/read_write.c +++ b/fs/ecryptfs/read_write.c @@ -156,7 +156,7 @@ int ecryptfs_write(struct inode *ecryptfs_inode, char *data, loff_t offset, ecryptfs_page_idx, rc); goto out; } - ecryptfs_page_virt = kmap_atomic(ecryptfs_page, KM_USER0); + ecryptfs_page_virt = kmap_atomic(ecryptfs_page); /* * pos: where we're now writing, offset: where the request was @@ -179,7 +179,7 @@ int ecryptfs_write(struct inode *ecryptfs_inode, char *data, loff_t offset, (data + data_offset), num_bytes); data_offset += num_bytes; } - kunmap_atomic(ecryptfs_page_virt, KM_USER0); + kunmap_atomic(ecryptfs_page_virt); flush_dcache_page(ecryptfs_page); SetPageUptodate(ecryptfs_page); unlock_page(ecryptfs_page); diff --git a/fs/ecryptfs/super.c b/fs/ecryptfs/super.c index 9df7fd6e0c3..cf152823bbf 100644 --- a/fs/ecryptfs/super.c +++ b/fs/ecryptfs/super.c @@ -30,6 +30,8 @@ #include <linux/seq_file.h> #include <linux/file.h> #include <linux/crypto.h> +#include <linux/statfs.h> +#include <linux/magic.h> #include "ecryptfs_kernel.h" struct kmem_cache *ecryptfs_inode_info_cache; @@ -102,10 +104,20 @@ static void ecryptfs_destroy_inode(struct inode *inode) static int ecryptfs_statfs(struct dentry *dentry, struct kstatfs *buf) { struct dentry *lower_dentry = ecryptfs_dentry_to_lower(dentry); + int rc; if (!lower_dentry->d_sb->s_op->statfs) return -ENOSYS; - return lower_dentry->d_sb->s_op->statfs(lower_dentry, buf); + + rc = lower_dentry->d_sb->s_op->statfs(lower_dentry, buf); + if (rc) + return rc; + + buf->f_type = ECRYPTFS_SUPER_MAGIC; + rc = ecryptfs_set_f_namelen(&buf->f_namelen, buf->f_namelen, + &ecryptfs_superblock_to_private(dentry->d_sb)->mount_crypt_stat); + + return rc; } /** diff --git a/fs/eventpoll.c b/fs/eventpoll.c index aabdfc38cf2..4d9d3a45e35 100644 --- a/fs/eventpoll.c +++ b/fs/eventpoll.c @@ -320,6 +320,11 @@ static inline int ep_is_linked(struct list_head *p) return !list_empty(p); } +static inline struct eppoll_entry *ep_pwq_from_wait(wait_queue_t *p) +{ + return container_of(p, struct eppoll_entry, wait); +} + /* Get the "struct epitem" from a wait queue pointer */ static inline struct epitem *ep_item_from_wait(wait_queue_t *p) { @@ -467,6 +472,18 @@ static void ep_poll_safewake(wait_queue_head_t *wq) put_cpu(); } +static void ep_remove_wait_queue(struct eppoll_entry *pwq) +{ + wait_queue_head_t *whead; + + rcu_read_lock(); + /* If it is cleared by POLLFREE, it should be rcu-safe */ + whead = rcu_dereference(pwq->whead); + if (whead) + remove_wait_queue(whead, &pwq->wait); + rcu_read_unlock(); +} + /* * This function unregisters poll callbacks from the associated file * descriptor. Must be called with "mtx" held (or "epmutex" if called from @@ -481,7 +498,7 @@ static void ep_unregister_pollwait(struct eventpoll *ep, struct epitem *epi) pwq = list_first_entry(lsthead, struct eppoll_entry, llink); list_del(&pwq->llink); - remove_wait_queue(pwq->whead, &pwq->wait); + ep_remove_wait_queue(pwq); kmem_cache_free(pwq_cache, pwq); } } @@ -842,6 +859,17 @@ static int ep_poll_callback(wait_queue_t *wait, unsigned mode, int sync, void *k struct epitem *epi = ep_item_from_wait(wait); struct eventpoll *ep = epi->ep; + if ((unsigned long)key & POLLFREE) { + ep_pwq_from_wait(wait)->whead = NULL; + /* + * whead = NULL above can race with ep_remove_wait_queue() + * which can do another remove_wait_queue() after us, so we + * can't use __remove_wait_queue(). whead->lock is held by + * the caller. + */ + list_del_init(&wait->task_list); + } + spin_lock_irqsave(&ep->lock, flags); /* @@ -960,6 +988,10 @@ static int path_count[PATH_ARR_SIZE]; static int path_count_inc(int nests) { + /* Allow an arbitrary number of depth 1 paths */ + if (nests == 0) + return 0; + if (++path_count[nests] > path_limits[nests]) return -1; return 0; diff --git a/fs/exec.c b/fs/exec.c index 92ce83a11e9..b0695a9900e 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -63,6 +63,8 @@ #include <trace/events/task.h> #include "internal.h" +#include <trace/events/sched.h> + int core_uses_pid; char core_pattern[CORENAME_MAX_SIZE] = "core"; unsigned int core_pipe_limit; @@ -1402,9 +1404,10 @@ int search_binary_handler(struct linux_binprm *bprm,struct pt_regs *regs) */ bprm->recursion_depth = depth; if (retval >= 0) { - if (depth == 0) - ptrace_event(PTRACE_EVENT_EXEC, - old_pid); + if (depth == 0) { + trace_sched_process_exec(current, old_pid, bprm); + ptrace_event(PTRACE_EVENT_EXEC, old_pid); + } put_binfmt(fmt); allow_write_access(bprm->file); if (bprm->file) @@ -1915,7 +1918,6 @@ static int coredump_wait(int exit_code, struct core_state *core_state) { struct task_struct *tsk = current; struct mm_struct *mm = tsk->mm; - struct completion *vfork_done; int core_waiters = -EBUSY; init_completion(&core_state->startup); @@ -1927,22 +1929,9 @@ static int coredump_wait(int exit_code, struct core_state *core_state) core_waiters = zap_threads(tsk, mm, core_state, exit_code); up_write(&mm->mmap_sem); - if (unlikely(core_waiters < 0)) - goto fail; - - /* - * Make sure nobody is waiting for us to release the VM, - * otherwise we can deadlock when we wait on each other - */ - vfork_done = tsk->vfork_done; - if (vfork_done) { - tsk->vfork_done = NULL; - complete(vfork_done); - } - - if (core_waiters) + if (core_waiters > 0) wait_for_completion(&core_state->startup); -fail: + return core_waiters; } diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index f855916657b..5b4a9362d5a 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -53,14 +53,6 @@ struct wb_writeback_work { }; /* - * Include the creation of the trace points after defining the - * wb_writeback_work structure so that the definition remains local to this - * file. - */ -#define CREATE_TRACE_POINTS -#include <trace/events/writeback.h> - -/* * We don't actually have pdflush, but this one is exported though /proc... */ int nr_pdflush_threads; @@ -92,6 +84,14 @@ static inline struct inode *wb_inode(struct list_head *head) return list_entry(head, struct inode, i_wb_list); } +/* + * Include the creation of the trace points after defining the + * wb_writeback_work structure and inline functions so that the definition + * remains local to this file. + */ +#define CREATE_TRACE_POINTS +#include <trace/events/writeback.h> + /* Wakeup flusher thread or forker thread to fork it. Requires bdi->wb_lock. */ static void bdi_wakeup_flusher(struct backing_dev_info *bdi) { diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index 376816fcd04..351a3e79778 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -167,14 +167,19 @@ void gfs2_glock_add_to_lru(struct gfs2_glock *gl) spin_unlock(&lru_lock); } -static void gfs2_glock_remove_from_lru(struct gfs2_glock *gl) +static void __gfs2_glock_remove_from_lru(struct gfs2_glock *gl) { - spin_lock(&lru_lock); if (!list_empty(&gl->gl_lru)) { list_del_init(&gl->gl_lru); atomic_dec(&lru_count); clear_bit(GLF_LRU, &gl->gl_flags); } +} + +static void gfs2_glock_remove_from_lru(struct gfs2_glock *gl) +{ + spin_lock(&lru_lock); + __gfs2_glock_remove_from_lru(gl); spin_unlock(&lru_lock); } @@ -217,11 +222,12 @@ void gfs2_glock_put(struct gfs2_glock *gl) struct gfs2_sbd *sdp = gl->gl_sbd; struct address_space *mapping = gfs2_glock2aspace(gl); - if (atomic_dec_and_test(&gl->gl_ref)) { + if (atomic_dec_and_lock(&gl->gl_ref, &lru_lock)) { + __gfs2_glock_remove_from_lru(gl); + spin_unlock(&lru_lock); spin_lock_bucket(gl->gl_hash); hlist_bl_del_rcu(&gl->gl_list); spin_unlock_bucket(gl->gl_hash); - gfs2_glock_remove_from_lru(gl); GLOCK_BUG_ON(gl, !list_empty(&gl->gl_holders)); GLOCK_BUG_ON(gl, mapping && mapping->nrpages); trace_gfs2_glock_put(gl); diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c index a7d611b93f0..56987460cda 100644 --- a/fs/gfs2/inode.c +++ b/fs/gfs2/inode.c @@ -391,10 +391,6 @@ static int alloc_dinode(struct gfs2_inode *dip, u64 *no_addr, u64 *generation) int error; int dblocks = 1; - error = gfs2_rindex_update(sdp); - if (error) - fs_warn(sdp, "rindex update returns %d\n", error); - error = gfs2_inplace_reserve(dip, RES_DINODE); if (error) goto out; @@ -1043,6 +1039,7 @@ static int gfs2_unlink(struct inode *dir, struct dentry *dentry) rgd = gfs2_blk2rgrpd(sdp, ip->i_no_addr); if (!rgd) goto out_inodes; + gfs2_holder_init(rgd->rd_gl, LM_ST_EXCLUSIVE, 0, ghs + 2); diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c index 6aacf3f230a..24f609c9ef9 100644 --- a/fs/gfs2/ops_fstype.c +++ b/fs/gfs2/ops_fstype.c @@ -800,6 +800,11 @@ static int init_inodes(struct gfs2_sbd *sdp, int undo) fs_err(sdp, "can't get quota file inode: %d\n", error); goto fail_rindex; } + + error = gfs2_rindex_update(sdp); + if (error) + goto fail_qinode; + return 0; fail_qinode: diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c index 981bfa32121..49ada95209d 100644 --- a/fs/gfs2/rgrp.c +++ b/fs/gfs2/rgrp.c @@ -683,16 +683,21 @@ int gfs2_rindex_update(struct gfs2_sbd *sdp) struct gfs2_glock *gl = ip->i_gl; struct gfs2_holder ri_gh; int error = 0; + int unlock_required = 0; /* Read new copy from disk if we don't have the latest */ if (!sdp->sd_rindex_uptodate) { mutex_lock(&sdp->sd_rindex_mutex); - error = gfs2_glock_nq_init(gl, LM_ST_SHARED, 0, &ri_gh); - if (error) - return error; + if (!gfs2_glock_is_locked_by_me(gl)) { + error = gfs2_glock_nq_init(gl, LM_ST_SHARED, 0, &ri_gh); + if (error) + return error; + unlock_required = 1; + } if (!sdp->sd_rindex_uptodate) error = gfs2_ri_update(ip); - gfs2_glock_dq_uninit(&ri_gh); + if (unlock_required) + gfs2_glock_dq_uninit(&ri_gh); mutex_unlock(&sdp->sd_rindex_mutex); } diff --git a/fs/inode.c b/fs/inode.c index fb10d86ffad..83ab215baab 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -938,8 +938,7 @@ void lockdep_annotate_inode_mutex_key(struct inode *inode) struct file_system_type *type = inode->i_sb->s_type; /* Set new key only if filesystem hasn't already changed it */ - if (!lockdep_match_class(&inode->i_mutex, - &type->i_mutex_key)) { + if (lockdep_match_class(&inode->i_mutex, &type->i_mutex_key)) { /* * ensure nobody is actually holding i_mutex */ @@ -966,6 +965,7 @@ void unlock_new_inode(struct inode *inode) spin_lock(&inode->i_lock); WARN_ON(!(inode->i_state & I_NEW)); inode->i_state &= ~I_NEW; + smp_mb(); wake_up_bit(&inode->i_state, __I_NEW); spin_unlock(&inode->i_lock); } @@ -1651,7 +1651,7 @@ __setup("ihash_entries=", set_ihash_entries); */ void __init inode_init_early(void) { - int loop; + unsigned int loop; /* If hashes are distributed across NUMA nodes, defer * hash allocation until vmalloc space is available. @@ -1669,13 +1669,13 @@ void __init inode_init_early(void) &i_hash_mask, 0); - for (loop = 0; loop < (1 << i_hash_shift); loop++) + for (loop = 0; loop < (1U << i_hash_shift); loop++) INIT_HLIST_HEAD(&inode_hashtable[loop]); } void __init inode_init(void) { - int loop; + unsigned int loop; /* inode slab cache */ inode_cachep = kmem_cache_create("inode_cache", @@ -1699,7 +1699,7 @@ void __init inode_init(void) &i_hash_mask, 0); - for (loop = 0; loop < (1 << i_hash_shift); loop++) + for (loop = 0; loop < (1U << i_hash_shift); loop++) INIT_HLIST_HEAD(&inode_hashtable[loop]); } diff --git a/fs/ioprio.c b/fs/ioprio.c index f84b380d65e..0f1b9515213 100644 --- a/fs/ioprio.c +++ b/fs/ioprio.c @@ -51,7 +51,7 @@ int set_task_ioprio(struct task_struct *task, int ioprio) ioc = get_task_io_context(task, GFP_ATOMIC, NUMA_NO_NODE); if (ioc) { ioc_ioprio_changed(ioc, ioprio); - put_io_context(ioc, NULL); + put_io_context(ioc); } return err; diff --git a/fs/namei.c b/fs/namei.c index 208c6aa4a98..fa96a26d329 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -1095,8 +1095,10 @@ static struct dentry *d_inode_lookup(struct dentry *parent, struct dentry *dentr struct dentry *old; /* Don't create child dentry for a dead directory. */ - if (unlikely(IS_DEADDIR(inode))) + if (unlikely(IS_DEADDIR(inode))) { + dput(dentry); return ERR_PTR(-ENOENT); + } old = inode->i_op->lookup(inode, dentry, nd); if (unlikely(old)) { @@ -1373,6 +1375,156 @@ static inline int can_lookup(struct inode *inode) } /* + * We can do the critical dentry name comparison and hashing + * operations one word at a time, but we are limited to: + * + * - Architectures with fast unaligned word accesses. We could + * do a "get_unaligned()" if this helps and is sufficiently + * fast. + * + * - Little-endian machines (so that we can generate the mask + * of low bytes efficiently). Again, we *could* do a byte + * swapping load on big-endian architectures if that is not + * expensive enough to make the optimization worthless. + * + * - non-CONFIG_DEBUG_PAGEALLOC configurations (so that we + * do not trap on the (extremely unlikely) case of a page + * crossing operation. + * + * - Furthermore, we need an efficient 64-bit compile for the + * 64-bit case in order to generate the "number of bytes in + * the final mask". Again, that could be replaced with a + * efficient population count instruction or similar. + */ +#ifdef CONFIG_DCACHE_WORD_ACCESS + +#ifdef CONFIG_64BIT + +/* + * Jan Achrenius on G+: microoptimized version of + * the simpler "(mask & ONEBYTES) * ONEBYTES >> 56" + * that works for the bytemasks without having to + * mask them first. + */ +static inline long count_masked_bytes(unsigned long mask) +{ + return mask*0x0001020304050608 >> 56; +} + +static inline unsigned int fold_hash(unsigned long hash) +{ + hash += hash >> (8*sizeof(int)); + return hash; +} + +#else /* 32-bit case */ + +/* Carl Chatfield / Jan Achrenius G+ version for 32-bit */ +static inline long count_masked_bytes(long mask) +{ + /* (000000 0000ff 00ffff ffffff) -> ( 1 1 2 3 ) */ + long a = (0x0ff0001+mask) >> 23; + /* Fix the 1 for 00 case */ + return a & mask; +} + +#define fold_hash(x) (x) + +#endif + +unsigned int full_name_hash(const unsigned char *name, unsigned int len) +{ + unsigned long a, mask; + unsigned long hash = 0; + + for (;;) { + a = *(unsigned long *)name; + hash *= 9; + if (len < sizeof(unsigned long)) + break; + hash += a; + name += sizeof(unsigned long); + len -= sizeof(unsigned long); + if (!len) + goto done; + } + mask = ~(~0ul << len*8); + hash += mask & a; +done: + return fold_hash(hash); +} +EXPORT_SYMBOL(full_name_hash); + +#define ONEBYTES 0x0101010101010101ul +#define SLASHBYTES 0x2f2f2f2f2f2f2f2ful +#define HIGHBITS 0x8080808080808080ul + +/* Return the high bit set in the first byte that is a zero */ +static inline unsigned long has_zero(unsigned long a) +{ + return ((a - ONEBYTES) & ~a) & HIGHBITS; +} + +/* + * Calculate the length and hash of the path component, and + * return the length of the component; + */ +static inline unsigned long hash_name(const char *name, unsigned int *hashp) +{ + unsigned long a, mask, hash, len; + + hash = a = 0; + len = -sizeof(unsigned long); + do { + hash = (hash + a) * 9; + len += sizeof(unsigned long); + a = *(unsigned long *)(name+len); + /* Do we have any NUL or '/' bytes in this word? */ + mask = has_zero(a) | has_zero(a ^ SLASHBYTES); + } while (!mask); + + /* The mask *below* the first high bit set */ + mask = (mask - 1) & ~mask; + mask >>= 7; + hash += a & mask; + *hashp = fold_hash(hash); + + return len + count_masked_bytes(mask); +} + +#else + +unsigned int full_name_hash(const unsigned char *name, unsigned int len) +{ + unsigned long hash = init_name_hash(); + while (len--) + hash = partial_name_hash(*name++, hash); + return end_name_hash(hash); +} +EXPORT_SYMBOL(full_name_hash); + +/* + * We know there's a real path component here of at least + * one character. + */ +static inline unsigned long hash_name(const char *name, unsigned int *hashp) +{ + unsigned long hash = init_name_hash(); + unsigned long len = 0, c; + + c = (unsigned char)*name; + do { + len++; + hash = partial_name_hash(c, hash); + c = (unsigned char)name[len]; + } while (c && c != '/'); + *hashp = end_name_hash(hash); + return len; +} + +#endif + +/* * Name resolution. * This is the basic name resolution function, turning a pathname into * the final dentry. We expect 'base' to be positive and a directory. @@ -1392,31 +1544,22 @@ static int link_path_walk(const char *name, struct nameidata *nd) /* At this point we know we have a real path component. */ for(;;) { - unsigned long hash; struct qstr this; - unsigned int c; + long len; int type; err = may_lookup(nd); if (err) break; + len = hash_name(name, &this.hash); this.name = name; - c = *(const unsigned char *)name; - - hash = init_name_hash(); - do { - name++; - hash = partial_name_hash(c, hash); - c = *(const unsigned char *)name; - } while (c && (c != '/')); - this.len = name - (const char *) this.name; - this.hash = end_name_hash(hash); + this.len = len; type = LAST_NORM; - if (this.name[0] == '.') switch (this.len) { + if (name[0] == '.') switch (len) { case 2: - if (this.name[1] == '.') { + if (name[1] == '.') { type = LAST_DOTDOT; nd->flags |= LOOKUP_JUMPED; } @@ -1435,12 +1578,18 @@ static int link_path_walk(const char *name, struct nameidata *nd) } } - /* remove trailing slashes? */ - if (!c) + if (!name[len]) goto last_component; - while (*++name == '/'); - if (!*name) + /* + * If it wasn't NUL, we know it was '/'. Skip that + * slash, and continue until no more slashes. + */ + do { + len++; + } while (unlikely(name[len] == '/')); + if (!name[len]) goto last_component; + name += len; err = walk_component(nd, &next, &this, type, LOOKUP_FOLLOW); if (err < 0) @@ -1773,24 +1922,21 @@ static struct dentry *lookup_hash(struct nameidata *nd) struct dentry *lookup_one_len(const char *name, struct dentry *base, int len) { struct qstr this; - unsigned long hash; unsigned int c; WARN_ON_ONCE(!mutex_is_locked(&base->d_inode->i_mutex)); this.name = name; this.len = len; + this.hash = full_name_hash(name, len); if (!len) return ERR_PTR(-EACCES); - hash = init_name_hash(); while (len--) { c = *(const unsigned char *)name++; if (c == '/' || c == '\0') return ERR_PTR(-EACCES); - hash = partial_name_hash(c, hash); } - this.hash = end_name_hash(hash); /* * See if the low-level filesystem might want * to use its own hash.. @@ -2138,7 +2284,7 @@ static struct file *do_last(struct nameidata *nd, struct path *path, /* sayonara */ error = complete_walk(nd); if (error) - return ERR_PTR(-ECHILD); + return ERR_PTR(error); error = -ENOTDIR; if (nd->flags & LOOKUP_DIRECTORY) { @@ -2237,7 +2383,7 @@ static struct file *do_last(struct nameidata *nd, struct path *path, /* Why this, you ask? _Now_ we might have grown LOOKUP_JUMPED... */ error = complete_walk(nd); if (error) - goto exit; + return ERR_PTR(error); error = -EISDIR; if (S_ISDIR(nd->inode->i_mode)) goto exit; diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index f0c849c98fe..ec9f6ef6c5d 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -3575,8 +3575,8 @@ static ssize_t __nfs4_get_acl_uncached(struct inode *inode, void *buf, size_t bu } if (npages > 1) { /* for decoding across pages */ - args.acl_scratch = alloc_page(GFP_KERNEL); - if (!args.acl_scratch) + res.acl_scratch = alloc_page(GFP_KERNEL); + if (!res.acl_scratch) goto out_free; } args.acl_len = npages * PAGE_SIZE; @@ -3612,8 +3612,8 @@ out_free: for (i = 0; i < npages; i++) if (pages[i]) __free_page(pages[i]); - if (args.acl_scratch) - __free_page(args.acl_scratch); + if (res.acl_scratch) + __free_page(res.acl_scratch); return ret; } @@ -4883,8 +4883,10 @@ int nfs4_proc_exchange_id(struct nfs_client *clp, struct rpc_cred *cred) clp->cl_rpcclient->cl_auth->au_flavor); res.server_scope = kzalloc(sizeof(struct server_scope), GFP_KERNEL); - if (unlikely(!res.server_scope)) - return -ENOMEM; + if (unlikely(!res.server_scope)) { + status = -ENOMEM; + goto out; + } status = rpc_call_sync(clp->cl_rpcclient, &msg, RPC_TASK_TIMEOUT); if (!status) @@ -4901,12 +4903,13 @@ int nfs4_proc_exchange_id(struct nfs_client *clp, struct rpc_cred *cred) clp->server_scope = NULL; } - if (!clp->server_scope) + if (!clp->server_scope) { clp->server_scope = res.server_scope; - else - kfree(res.server_scope); + goto out; + } } - + kfree(res.server_scope); +out: dprintk("<-- %s status= %d\n", __func__, status); return status; } @@ -5008,37 +5011,53 @@ int nfs4_proc_get_lease_time(struct nfs_client *clp, struct nfs_fsinfo *fsinfo) return status; } +static struct nfs4_slot *nfs4_alloc_slots(u32 max_slots, gfp_t gfp_flags) +{ + return kcalloc(max_slots, sizeof(struct nfs4_slot), gfp_flags); +} + +static void nfs4_add_and_init_slots(struct nfs4_slot_table *tbl, + struct nfs4_slot *new, + u32 max_slots, + u32 ivalue) +{ + struct nfs4_slot *old = NULL; + u32 i; + + spin_lock(&tbl->slot_tbl_lock); + if (new) { + old = tbl->slots; + tbl->slots = new; + tbl->max_slots = max_slots; + } + tbl->highest_used_slotid = -1; /* no slot is currently used */ + for (i = 0; i < tbl->max_slots; i++) + tbl->slots[i].seq_nr = ivalue; + spin_unlock(&tbl->slot_tbl_lock); + kfree(old); +} + /* - * Reset a slot table + * (re)Initialise a slot table */ -static int nfs4_reset_slot_table(struct nfs4_slot_table *tbl, u32 max_reqs, - int ivalue) +static int nfs4_realloc_slot_table(struct nfs4_slot_table *tbl, u32 max_reqs, + u32 ivalue) { struct nfs4_slot *new = NULL; - int i; - int ret = 0; + int ret = -ENOMEM; dprintk("--> %s: max_reqs=%u, tbl->max_slots %d\n", __func__, max_reqs, tbl->max_slots); /* Does the newly negotiated max_reqs match the existing slot table? */ if (max_reqs != tbl->max_slots) { - ret = -ENOMEM; - new = kmalloc(max_reqs * sizeof(struct nfs4_slot), - GFP_NOFS); + new = nfs4_alloc_slots(max_reqs, GFP_NOFS); if (!new) goto out; - ret = 0; - kfree(tbl->slots); } - spin_lock(&tbl->slot_tbl_lock); - if (new) { - tbl->slots = new; - tbl->max_slots = max_reqs; - } - for (i = 0; i < tbl->max_slots; ++i) - tbl->slots[i].seq_nr = ivalue; - spin_unlock(&tbl->slot_tbl_lock); + ret = 0; + + nfs4_add_and_init_slots(tbl, new, max_reqs, ivalue); dprintk("%s: tbl=%p slots=%p max_slots=%d\n", __func__, tbl, tbl->slots, tbl->max_slots); out: @@ -5061,36 +5080,6 @@ static void nfs4_destroy_slot_tables(struct nfs4_session *session) } /* - * Initialize slot table - */ -static int nfs4_init_slot_table(struct nfs4_slot_table *tbl, - int max_slots, int ivalue) -{ - struct nfs4_slot *slot; - int ret = -ENOMEM; - - BUG_ON(max_slots > NFS4_MAX_SLOT_TABLE); - - dprintk("--> %s: max_reqs=%u\n", __func__, max_slots); - - slot = kcalloc(max_slots, sizeof(struct nfs4_slot), GFP_NOFS); - if (!slot) - goto out; - ret = 0; - - spin_lock(&tbl->slot_tbl_lock); - tbl->max_slots = max_slots; - tbl->slots = slot; - tbl->highest_used_slotid = -1; /* no slot is currently used */ - spin_unlock(&tbl->slot_tbl_lock); - dprintk("%s: tbl=%p slots=%p max_slots=%d\n", __func__, - tbl, tbl->slots, tbl->max_slots); -out: - dprintk("<-- %s: return %d\n", __func__, ret); - return ret; -} - -/* * Initialize or reset the forechannel and backchannel tables */ static int nfs4_setup_session_slot_tables(struct nfs4_session *ses) @@ -5101,25 +5090,16 @@ static int nfs4_setup_session_slot_tables(struct nfs4_session *ses) dprintk("--> %s\n", __func__); /* Fore channel */ tbl = &ses->fc_slot_table; - if (tbl->slots == NULL) { - status = nfs4_init_slot_table(tbl, ses->fc_attrs.max_reqs, 1); - if (status) /* -ENOMEM */ - return status; - } else { - status = nfs4_reset_slot_table(tbl, ses->fc_attrs.max_reqs, 1); - if (status) - return status; - } + status = nfs4_realloc_slot_table(tbl, ses->fc_attrs.max_reqs, 1); + if (status) /* -ENOMEM */ + return status; /* Back channel */ tbl = &ses->bc_slot_table; - if (tbl->slots == NULL) { - status = nfs4_init_slot_table(tbl, ses->bc_attrs.max_reqs, 0); - if (status) - /* Fore and back channel share a connection so get - * both slot tables or neither */ - nfs4_destroy_slot_tables(ses); - } else - status = nfs4_reset_slot_table(tbl, ses->bc_attrs.max_reqs, 0); + status = nfs4_realloc_slot_table(tbl, ses->bc_attrs.max_reqs, 0); + if (status && tbl->slots == NULL) + /* Fore and back channel share a connection so get + * both slot tables or neither */ + nfs4_destroy_slot_tables(ses); return status; } diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c index a53f33b4ac3..45392032e7b 100644 --- a/fs/nfs/nfs4state.c +++ b/fs/nfs/nfs4state.c @@ -1132,6 +1132,8 @@ void nfs4_schedule_stateid_recovery(const struct nfs_server *server, struct nfs4 { struct nfs_client *clp = server->nfs_client; + if (test_and_clear_bit(NFS_DELEGATED_STATE, &state->flags)) + nfs_async_inode_return_delegation(state->inode, &state->stateid); nfs4_state_mark_reclaim_nograce(clp, state); nfs4_schedule_state_manager(clp); } diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index 95e92e43840..33bd8d0f745 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -2522,7 +2522,6 @@ static void nfs4_xdr_enc_getacl(struct rpc_rqst *req, struct xdr_stream *xdr, xdr_inline_pages(&req->rq_rcv_buf, replen << 2, args->acl_pages, args->acl_pgbase, args->acl_len); - xdr_set_scratch_buffer(xdr, page_address(args->acl_scratch), PAGE_SIZE); encode_nops(&hdr); } @@ -6032,6 +6031,10 @@ nfs4_xdr_dec_getacl(struct rpc_rqst *rqstp, struct xdr_stream *xdr, struct compound_hdr hdr; int status; + if (res->acl_scratch != NULL) { + void *p = page_address(res->acl_scratch); + xdr_set_scratch_buffer(xdr, p, PAGE_SIZE); + } status = decode_compound_hdr(xdr, &hdr); if (status) goto out; diff --git a/fs/nilfs2/the_nilfs.c b/fs/nilfs2/the_nilfs.c index d3271409437..501b7f8b739 100644 --- a/fs/nilfs2/the_nilfs.c +++ b/fs/nilfs2/the_nilfs.c @@ -409,6 +409,12 @@ static int nilfs_store_disk_layout(struct the_nilfs *nilfs, nilfs->ns_first_data_block = le64_to_cpu(sbp->s_first_data_block); nilfs->ns_r_segments_percentage = le32_to_cpu(sbp->s_r_segments_percentage); + if (nilfs->ns_r_segments_percentage < 1 || + nilfs->ns_r_segments_percentage > 99) { + printk(KERN_ERR "NILFS: invalid reserved segments percentage.\n"); + return -EINVAL; + } + nilfs_set_nsegments(nilfs, le64_to_cpu(sbp->s_nsegments)); nilfs->ns_crc_seed = le32_to_cpu(sbp->s_crc_seed); return 0; @@ -515,6 +521,7 @@ static int nilfs_load_super_block(struct the_nilfs *nilfs, brelse(sbh[1]); sbh[1] = NULL; sbp[1] = NULL; + valid[1] = 0; swp = 0; } if (!valid[swp]) { diff --git a/fs/ntfs/attrib.c b/fs/ntfs/attrib.c index f14fde2b03d..e0281992ddc 100644 --- a/fs/ntfs/attrib.c +++ b/fs/ntfs/attrib.c @@ -1,7 +1,7 @@ /** * attrib.c - NTFS attribute operations. Part of the Linux-NTFS project. * - * Copyright (c) 2001-2007 Anton Altaparmakov + * Copyright (c) 2001-2012 Anton Altaparmakov and Tuxera Inc. * Copyright (c) 2002 Richard Russon * * This program/include file is free software; you can redistribute it and/or @@ -345,10 +345,10 @@ LCN ntfs_attr_vcn_to_lcn_nolock(ntfs_inode *ni, const VCN vcn, unsigned long flags; bool is_retry = false; + BUG_ON(!ni); ntfs_debug("Entering for i_ino 0x%lx, vcn 0x%llx, %s_locked.", ni->mft_no, (unsigned long long)vcn, write_locked ? "write" : "read"); - BUG_ON(!ni); BUG_ON(!NInoNonResident(ni)); BUG_ON(vcn < 0); if (!ni->runlist.rl) { @@ -469,9 +469,9 @@ runlist_element *ntfs_attr_find_vcn_nolock(ntfs_inode *ni, const VCN vcn, int err = 0; bool is_retry = false; + BUG_ON(!ni); ntfs_debug("Entering for i_ino 0x%lx, vcn 0x%llx, with%s ctx.", ni->mft_no, (unsigned long long)vcn, ctx ? "" : "out"); - BUG_ON(!ni); BUG_ON(!NInoNonResident(ni)); BUG_ON(vcn < 0); if (!ni->runlist.rl) { diff --git a/fs/ntfs/mft.c b/fs/ntfs/mft.c index 382857f9c7d..3014a36a255 100644 --- a/fs/ntfs/mft.c +++ b/fs/ntfs/mft.c @@ -1,7 +1,7 @@ /** * mft.c - NTFS kernel mft record operations. Part of the Linux-NTFS project. * - * Copyright (c) 2001-2011 Anton Altaparmakov and Tuxera Inc. + * Copyright (c) 2001-2012 Anton Altaparmakov and Tuxera Inc. * Copyright (c) 2002 Richard Russon * * This program/include file is free software; you can redistribute it and/or @@ -1367,7 +1367,7 @@ static int ntfs_mft_bitmap_extend_allocation_nolock(ntfs_volume *vol) ntfs_error(vol->sb, "Failed to merge runlists for mft " "bitmap."); if (ntfs_cluster_free_from_rl(vol, rl2)) { - ntfs_error(vol->sb, "Failed to dealocate " + ntfs_error(vol->sb, "Failed to deallocate " "allocated cluster.%s", es); NVolSetErrors(vol); } @@ -1805,7 +1805,7 @@ static int ntfs_mft_data_extend_allocation_nolock(ntfs_volume *vol) ntfs_error(vol->sb, "Failed to merge runlists for mft data " "attribute."); if (ntfs_cluster_free_from_rl(vol, rl2)) { - ntfs_error(vol->sb, "Failed to dealocate clusters " + ntfs_error(vol->sb, "Failed to deallocate clusters " "from the mft data attribute.%s", es); NVolSetErrors(vol); } diff --git a/fs/ntfs/super.c b/fs/ntfs/super.c index 5a4a8af5c40..f907611cca7 100644 --- a/fs/ntfs/super.c +++ b/fs/ntfs/super.c @@ -1,7 +1,7 @@ /* * super.c - NTFS kernel super block handling. Part of the Linux-NTFS project. * - * Copyright (c) 2001-2011 Anton Altaparmakov and Tuxera Inc. + * Copyright (c) 2001-2012 Anton Altaparmakov and Tuxera Inc. * Copyright (c) 2001,2002 Richard Russon * * This program/include file is free software; you can redistribute it and/or @@ -1239,7 +1239,6 @@ static int check_windows_hibernation_status(ntfs_volume *vol) { MFT_REF mref; struct inode *vi; - ntfs_inode *ni; struct page *page; u32 *kaddr, *kend; ntfs_name *name = NULL; @@ -1290,7 +1289,6 @@ static int check_windows_hibernation_status(ntfs_volume *vol) "is not the system volume.", i_size_read(vi)); goto iput_out; } - ni = NTFS_I(vi); page = ntfs_map_page(vi->i_mapping, 0); if (IS_ERR(page)) { ntfs_error(vol->sb, "Failed to read from hiberfil.sys."); diff --git a/fs/ocfs2/namei.c b/fs/ocfs2/namei.c index be244692550..a9856e3eaaf 100644 --- a/fs/ocfs2/namei.c +++ b/fs/ocfs2/namei.c @@ -1053,7 +1053,7 @@ static int ocfs2_rename(struct inode *old_dir, handle_t *handle = NULL; struct buffer_head *old_dir_bh = NULL; struct buffer_head *new_dir_bh = NULL; - nlink_t old_dir_nlink = old_dir->i_nlink; + u32 old_dir_nlink = old_dir->i_nlink; struct ocfs2_dinode *old_di; struct ocfs2_dir_lookup_result old_inode_dot_dot_res = { NULL, }; struct ocfs2_dir_lookup_result target_lookup_res = { NULL, }; diff --git a/fs/proc/base.c b/fs/proc/base.c index d4548dd49b0..965d4bde3a3 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -1310,8 +1310,7 @@ sched_autogroup_write(struct file *file, const char __user *buf, if (!p) return -ESRCH; - err = nice; - err = proc_sched_autogroup_set_nice(p, &err); + err = proc_sched_autogroup_set_nice(p, nice); if (err) count = err; diff --git a/fs/quota/quota.c b/fs/quota/quota.c index 7898cd688a0..fc2c4388d12 100644 --- a/fs/quota/quota.c +++ b/fs/quota/quota.c @@ -292,11 +292,26 @@ static int do_quotactl(struct super_block *sb, int type, int cmd, qid_t id, } } +/* Return 1 if 'cmd' will block on frozen filesystem */ +static int quotactl_cmd_write(int cmd) +{ + switch (cmd) { + case Q_GETFMT: + case Q_GETINFO: + case Q_SYNC: + case Q_XGETQSTAT: + case Q_XGETQUOTA: + case Q_XQUOTASYNC: + return 0; + } + return 1; +} + /* * look up a superblock on which quota ops will be performed * - use the name of a block device to find the superblock thereon */ -static struct super_block *quotactl_block(const char __user *special) +static struct super_block *quotactl_block(const char __user *special, int cmd) { #ifdef CONFIG_BLOCK struct block_device *bdev; @@ -309,7 +324,10 @@ static struct super_block *quotactl_block(const char __user *special) putname(tmp); if (IS_ERR(bdev)) return ERR_CAST(bdev); - sb = get_super(bdev); + if (quotactl_cmd_write(cmd)) + sb = get_super_thawed(bdev); + else + sb = get_super(bdev); bdput(bdev); if (!sb) return ERR_PTR(-ENODEV); @@ -361,7 +379,7 @@ SYSCALL_DEFINE4(quotactl, unsigned int, cmd, const char __user *, special, pathp = &path; } - sb = quotactl_block(special); + sb = quotactl_block(special, cmds); if (IS_ERR(sb)) { ret = PTR_ERR(sb); goto out; diff --git a/fs/select.c b/fs/select.c index d33418fdc85..e782258d0de 100644 --- a/fs/select.c +++ b/fs/select.c @@ -912,7 +912,7 @@ static long do_restart_poll(struct restart_block *restart_block) } SYSCALL_DEFINE3(poll, struct pollfd __user *, ufds, unsigned int, nfds, - long, timeout_msecs) + int, timeout_msecs) { struct timespec end_time, *to = NULL; int ret; diff --git a/fs/signalfd.c b/fs/signalfd.c index 492465b451d..7ae2a574cb2 100644 --- a/fs/signalfd.c +++ b/fs/signalfd.c @@ -30,6 +30,21 @@ #include <linux/signalfd.h> #include <linux/syscalls.h> +void signalfd_cleanup(struct sighand_struct *sighand) +{ + wait_queue_head_t *wqh = &sighand->signalfd_wqh; + /* + * The lockless check can race with remove_wait_queue() in progress, + * but in this case its caller should run under rcu_read_lock() and + * sighand_cachep is SLAB_DESTROY_BY_RCU, we can safely return. + */ + if (likely(!waitqueue_active(wqh))) + return; + + /* wait_queue_t->func(POLLFREE) should do remove_wait_queue() */ + wake_up_poll(wqh, POLLHUP | POLLFREE); +} + struct signalfd_ctx { sigset_t sigmask; }; diff --git a/fs/super.c b/fs/super.c index 6015c02296b..6277ec6cb60 100644 --- a/fs/super.c +++ b/fs/super.c @@ -634,6 +634,28 @@ rescan: EXPORT_SYMBOL(get_super); /** + * get_super_thawed - get thawed superblock of a device + * @bdev: device to get the superblock for + * + * Scans the superblock list and finds the superblock of the file system + * mounted on the device. The superblock is returned once it is thawed + * (or immediately if it was not frozen). %NULL is returned if no match + * is found. + */ +struct super_block *get_super_thawed(struct block_device *bdev) +{ + while (1) { + struct super_block *s = get_super(bdev); + if (!s || s->s_frozen == SB_UNFROZEN) + return s; + up_read(&s->s_umount); + vfs_check_frozen(s, SB_FREEZE_WRITE); + put_super(s); + } +} +EXPORT_SYMBOL(get_super_thawed); + +/** * get_active_super - get an active reference to the superblock of a device * @bdev: device to get the superblock for * diff --git a/fs/sysfs/dir.c b/fs/sysfs/dir.c index 7fdf6a7b743..2a7a3f5d1ca 100644 --- a/fs/sysfs/dir.c +++ b/fs/sysfs/dir.c @@ -22,76 +22,103 @@ #include <linux/mutex.h> #include <linux/slab.h> #include <linux/security.h> +#include <linux/hash.h> #include "sysfs.h" DEFINE_MUTEX(sysfs_mutex); DEFINE_SPINLOCK(sysfs_assoc_lock); +#define to_sysfs_dirent(X) rb_entry((X), struct sysfs_dirent, s_rb); + static DEFINE_SPINLOCK(sysfs_ino_lock); static DEFINE_IDA(sysfs_ino_ida); /** - * sysfs_link_sibling - link sysfs_dirent into sibling list + * sysfs_name_hash + * @ns: Namespace tag to hash + * @name: Null terminated string to hash + * + * Returns 31 bit hash of ns + name (so it fits in an off_t ) + */ +static unsigned int sysfs_name_hash(const void *ns, const char *name) +{ + unsigned long hash = init_name_hash(); + unsigned int len = strlen(name); + while (len--) + hash = partial_name_hash(*name++, hash); + hash = ( end_name_hash(hash) ^ hash_ptr( (void *)ns, 31 ) ); + hash &= 0x7fffffffU; + /* Reserve hash numbers 0, 1 and INT_MAX for magic directory entries */ + if (hash < 1) + hash += 2; + if (hash >= INT_MAX) + hash = INT_MAX - 1; + return hash; +} + +static int sysfs_name_compare(unsigned int hash, const void *ns, + const char *name, const struct sysfs_dirent *sd) +{ + if (hash != sd->s_hash) + return hash - sd->s_hash; + if (ns != sd->s_ns) + return ns - sd->s_ns; + return strcmp(name, sd->s_name); +} + +static int sysfs_sd_compare(const struct sysfs_dirent *left, + const struct sysfs_dirent *right) +{ + return sysfs_name_compare(left->s_hash, left->s_ns, left->s_name, + right); +} + +/** + * sysfs_link_subling - link sysfs_dirent into sibling rbtree * @sd: sysfs_dirent of interest * - * Link @sd into its sibling list which starts from + * Link @sd into its sibling rbtree which starts from * sd->s_parent->s_dir.children. * * Locking: * mutex_lock(sysfs_mutex) + * + * RETURNS: + * 0 on susccess -EEXIST on failure. */ -static void sysfs_link_sibling(struct sysfs_dirent *sd) +static int sysfs_link_sibling(struct sysfs_dirent *sd) { - struct sysfs_dirent *parent_sd = sd->s_parent; - - struct rb_node **p; - struct rb_node *parent; + struct rb_node **node = &sd->s_parent->s_dir.children.rb_node; + struct rb_node *parent = NULL; if (sysfs_type(sd) == SYSFS_DIR) - parent_sd->s_dir.subdirs++; - - p = &parent_sd->s_dir.inode_tree.rb_node; - parent = NULL; - while (*p) { - parent = *p; -#define node rb_entry(parent, struct sysfs_dirent, inode_node) - if (sd->s_ino < node->s_ino) { - p = &node->inode_node.rb_left; - } else if (sd->s_ino > node->s_ino) { - p = &node->inode_node.rb_right; - } else { - printk(KERN_CRIT "sysfs: inserting duplicate inode '%lx'\n", - (unsigned long) sd->s_ino); - BUG(); - } -#undef node - } - rb_link_node(&sd->inode_node, parent, p); - rb_insert_color(&sd->inode_node, &parent_sd->s_dir.inode_tree); - - p = &parent_sd->s_dir.name_tree.rb_node; - parent = NULL; - while (*p) { - int c; - parent = *p; -#define node rb_entry(parent, struct sysfs_dirent, name_node) - c = strcmp(sd->s_name, node->s_name); - if (c < 0) { - p = &node->name_node.rb_left; - } else { - p = &node->name_node.rb_right; - } -#undef node + sd->s_parent->s_dir.subdirs++; + + while (*node) { + struct sysfs_dirent *pos; + int result; + + pos = to_sysfs_dirent(*node); + parent = *node; + result = sysfs_sd_compare(sd, pos); + if (result < 0) + node = &pos->s_rb.rb_left; + else if (result > 0) + node = &pos->s_rb.rb_right; + else + return -EEXIST; } - rb_link_node(&sd->name_node, parent, p); - rb_insert_color(&sd->name_node, &parent_sd->s_dir.name_tree); + /* add new node and rebalance the tree */ + rb_link_node(&sd->s_rb, parent, node); + rb_insert_color(&sd->s_rb, &sd->s_parent->s_dir.children); + return 0; } /** - * sysfs_unlink_sibling - unlink sysfs_dirent from sibling list + * sysfs_unlink_sibling - unlink sysfs_dirent from sibling rbtree * @sd: sysfs_dirent of interest * - * Unlink @sd from its sibling list which starts from + * Unlink @sd from its sibling rbtree which starts from * sd->s_parent->s_dir.children. * * Locking: @@ -102,8 +129,7 @@ static void sysfs_unlink_sibling(struct sysfs_dirent *sd) if (sysfs_type(sd) == SYSFS_DIR) sd->s_parent->s_dir.subdirs--; - rb_erase(&sd->inode_node, &sd->s_parent->s_dir.inode_tree); - rb_erase(&sd->name_node, &sd->s_parent->s_dir.name_tree); + rb_erase(&sd->s_rb, &sd->s_parent->s_dir.children); } /** @@ -198,7 +224,7 @@ static void sysfs_deactivate(struct sysfs_dirent *sd) rwsem_release(&sd->dep_map, 1, _RET_IP_); } -static int sysfs_alloc_ino(ino_t *pino) +static int sysfs_alloc_ino(unsigned int *pino) { int ino, rc; @@ -217,7 +243,7 @@ static int sysfs_alloc_ino(ino_t *pino) return rc; } -static void sysfs_free_ino(ino_t ino) +static void sysfs_free_ino(unsigned int ino) { spin_lock(&sysfs_ino_lock); ida_remove(&sysfs_ino_ida, ino); @@ -402,6 +428,7 @@ void sysfs_addrm_start(struct sysfs_addrm_cxt *acxt, int __sysfs_add_one(struct sysfs_addrm_cxt *acxt, struct sysfs_dirent *sd) { struct sysfs_inode_attrs *ps_iattr; + int ret; if (!!sysfs_ns_type(acxt->parent_sd) != !!sd->s_ns) { WARN(1, KERN_WARNING "sysfs: ns %s in '%s' for '%s'\n", @@ -410,12 +437,12 @@ int __sysfs_add_one(struct sysfs_addrm_cxt *acxt, struct sysfs_dirent *sd) return -EINVAL; } - if (sysfs_find_dirent(acxt->parent_sd, sd->s_ns, sd->s_name)) - return -EEXIST; - + sd->s_hash = sysfs_name_hash(sd->s_ns, sd->s_name); sd->s_parent = sysfs_get(acxt->parent_sd); - sysfs_link_sibling(sd); + ret = sysfs_link_sibling(sd); + if (ret) + return ret; /* Update timestamps on the parent */ ps_iattr = acxt->parent_sd->s_iattr; @@ -565,8 +592,8 @@ struct sysfs_dirent *sysfs_find_dirent(struct sysfs_dirent *parent_sd, const void *ns, const unsigned char *name) { - struct rb_node *p = parent_sd->s_dir.name_tree.rb_node; - struct sysfs_dirent *found = NULL; + struct rb_node *node = parent_sd->s_dir.children.rb_node; + unsigned int hash; if (!!sysfs_ns_type(parent_sd) != !!ns) { WARN(1, KERN_WARNING "sysfs: ns %s in '%s' for '%s'\n", @@ -575,33 +602,21 @@ struct sysfs_dirent *sysfs_find_dirent(struct sysfs_dirent *parent_sd, return NULL; } - while (p) { - int c; -#define node rb_entry(p, struct sysfs_dirent, name_node) - c = strcmp(name, node->s_name); - if (c < 0) { - p = node->name_node.rb_left; - } else if (c > 0) { - p = node->name_node.rb_right; - } else { - found = node; - p = node->name_node.rb_left; - } -#undef node - } - - if (found) { - while (found->s_ns != ns) { - p = rb_next(&found->name_node); - if (!p) - return NULL; - found = rb_entry(p, struct sysfs_dirent, name_node); - if (strcmp(name, found->s_name)) - return NULL; - } + hash = sysfs_name_hash(ns, name); + while (node) { + struct sysfs_dirent *sd; + int result; + + sd = to_sysfs_dirent(node); + result = sysfs_name_compare(hash, ns, name, sd); + if (result < 0) + node = node->rb_left; + else if (result > 0) + node = node->rb_right; + else + return sd; } - - return found; + return NULL; } /** @@ -804,9 +819,9 @@ static void __sysfs_remove_dir(struct sysfs_dirent *dir_sd) pr_debug("sysfs %s: removing dir\n", dir_sd->s_name); sysfs_addrm_start(&acxt, dir_sd); - pos = rb_first(&dir_sd->s_dir.inode_tree); + pos = rb_first(&dir_sd->s_dir.children); while (pos) { - struct sysfs_dirent *sd = rb_entry(pos, struct sysfs_dirent, inode_node); + struct sysfs_dirent *sd = to_sysfs_dirent(pos); pos = rb_next(pos); if (sysfs_type(sd) != SYSFS_DIR) sysfs_remove_one(&acxt, sd); @@ -863,6 +878,7 @@ int sysfs_rename(struct sysfs_dirent *sd, dup_name = sd->s_name; sd->s_name = new_name; + sd->s_hash = sysfs_name_hash(sd->s_ns, sd->s_name); } /* Move to the appropriate place in the appropriate directories rbtree. */ @@ -919,38 +935,36 @@ static int sysfs_dir_release(struct inode *inode, struct file *filp) } static struct sysfs_dirent *sysfs_dir_pos(const void *ns, - struct sysfs_dirent *parent_sd, ino_t ino, struct sysfs_dirent *pos) + struct sysfs_dirent *parent_sd, loff_t hash, struct sysfs_dirent *pos) { if (pos) { int valid = !(pos->s_flags & SYSFS_FLAG_REMOVED) && pos->s_parent == parent_sd && - ino == pos->s_ino; + hash == pos->s_hash; sysfs_put(pos); if (!valid) pos = NULL; } - if (!pos && (ino > 1) && (ino < INT_MAX)) { - struct rb_node *p = parent_sd->s_dir.inode_tree.rb_node; - while (p) { -#define node rb_entry(p, struct sysfs_dirent, inode_node) - if (ino < node->s_ino) { - pos = node; - p = node->inode_node.rb_left; - } else if (ino > node->s_ino) { - p = node->inode_node.rb_right; - } else { - pos = node; + if (!pos && (hash > 1) && (hash < INT_MAX)) { + struct rb_node *node = parent_sd->s_dir.children.rb_node; + while (node) { + pos = to_sysfs_dirent(node); + + if (hash < pos->s_hash) + node = node->rb_left; + else if (hash > pos->s_hash) + node = node->rb_right; + else break; - } -#undef node } } + /* Skip over entries in the wrong namespace */ while (pos && pos->s_ns != ns) { - struct rb_node *p = rb_next(&pos->inode_node); - if (!p) + struct rb_node *node = rb_next(&pos->s_rb); + if (!node) pos = NULL; else - pos = rb_entry(p, struct sysfs_dirent, inode_node); + pos = to_sysfs_dirent(node); } return pos; } @@ -960,11 +974,11 @@ static struct sysfs_dirent *sysfs_dir_next_pos(const void *ns, { pos = sysfs_dir_pos(ns, parent_sd, ino, pos); if (pos) do { - struct rb_node *p = rb_next(&pos->inode_node); - if (!p) + struct rb_node *node = rb_next(&pos->s_rb); + if (!node) pos = NULL; else - pos = rb_entry(p, struct sysfs_dirent, inode_node); + pos = to_sysfs_dirent(node); } while (pos && pos->s_ns != ns); return pos; } @@ -1006,7 +1020,7 @@ static int sysfs_readdir(struct file * filp, void * dirent, filldir_t filldir) len = strlen(name); ino = pos->s_ino; type = dt_type(pos); - filp->f_pos = ino; + filp->f_pos = pos->s_hash; filp->private_data = sysfs_get(pos); mutex_unlock(&sysfs_mutex); diff --git a/fs/sysfs/inode.c b/fs/sysfs/inode.c index 85eb81683a2..feb2d69396c 100644 --- a/fs/sysfs/inode.c +++ b/fs/sysfs/inode.c @@ -136,12 +136,13 @@ static int sysfs_sd_setsecdata(struct sysfs_dirent *sd, void **secdata, u32 *sec void *old_secdata; size_t old_secdata_len; - iattrs = sd->s_iattr; - if (!iattrs) - iattrs = sysfs_init_inode_attrs(sd); - if (!iattrs) - return -ENOMEM; + if (!sd->s_iattr) { + sd->s_iattr = sysfs_init_inode_attrs(sd); + if (!sd->s_iattr) + return -ENOMEM; + } + iattrs = sd->s_iattr; old_secdata = iattrs->ia_secdata; old_secdata_len = iattrs->ia_secdata_len; diff --git a/fs/sysfs/mount.c b/fs/sysfs/mount.c index e34f0d99ea4..140f26a3428 100644 --- a/fs/sysfs/mount.c +++ b/fs/sysfs/mount.c @@ -36,7 +36,7 @@ struct sysfs_dirent sysfs_root = { .s_name = "", .s_count = ATOMIC_INIT(1), .s_flags = SYSFS_DIR | (KOBJ_NS_TYPE_NONE << SYSFS_NS_TYPE_SHIFT), - .s_mode = S_IFDIR | S_IRWXU | S_IRUGO | S_IXUGO, + .s_mode = S_IFDIR | S_IRUGO | S_IXUGO, .s_ino = 1, }; diff --git a/fs/sysfs/sysfs.h b/fs/sysfs/sysfs.h index 7484a36ee67..661a9639570 100644 --- a/fs/sysfs/sysfs.h +++ b/fs/sysfs/sysfs.h @@ -20,9 +20,8 @@ struct sysfs_elem_dir { struct kobject *kobj; unsigned long subdirs; - - struct rb_root inode_tree; - struct rb_root name_tree; + /* children rbtree starts here and goes through sd->s_rb */ + struct rb_root children; }; struct sysfs_elem_symlink { @@ -62,8 +61,7 @@ struct sysfs_dirent { struct sysfs_dirent *s_parent; const char *s_name; - struct rb_node inode_node; - struct rb_node name_node; + struct rb_node s_rb; union { struct completion *completion; @@ -71,6 +69,7 @@ struct sysfs_dirent { } u; const void *s_ns; /* namespace tag */ + unsigned int s_hash; /* ns + name hash */ union { struct sysfs_elem_dir s_dir; struct sysfs_elem_symlink s_symlink; @@ -78,9 +77,9 @@ struct sysfs_dirent { struct sysfs_elem_bin_attr s_bin_attr; }; - unsigned int s_flags; + unsigned short s_flags; umode_t s_mode; - ino_t s_ino; + unsigned int s_ino; struct sysfs_inode_attrs *s_iattr; }; @@ -95,11 +94,11 @@ struct sysfs_dirent { #define SYSFS_ACTIVE_REF (SYSFS_KOBJ_ATTR | SYSFS_KOBJ_BIN_ATTR) /* identify any namespace tag on sysfs_dirents */ -#define SYSFS_NS_TYPE_MASK 0xff00 +#define SYSFS_NS_TYPE_MASK 0xf00 #define SYSFS_NS_TYPE_SHIFT 8 #define SYSFS_FLAG_MASK ~(SYSFS_NS_TYPE_MASK|SYSFS_TYPE_MASK) -#define SYSFS_FLAG_REMOVED 0x020000 +#define SYSFS_FLAG_REMOVED 0x02000 static inline unsigned int sysfs_type(struct sysfs_dirent *sd) { diff --git a/fs/udf/file.c b/fs/udf/file.c index dca0c3881e8..d567b8448df 100644 --- a/fs/udf/file.c +++ b/fs/udf/file.c @@ -201,12 +201,10 @@ out: static int udf_release_file(struct inode *inode, struct file *filp) { if (filp->f_mode & FMODE_WRITE) { - mutex_lock(&inode->i_mutex); down_write(&UDF_I(inode)->i_data_sem); udf_discard_prealloc(inode); udf_truncate_tail_extent(inode); up_write(&UDF_I(inode)->i_data_sem); - mutex_unlock(&inode->i_mutex); } return 0; } diff --git a/fs/xfs/kmem.h b/fs/xfs/kmem.h index 292eff19803..ab7c53fe346 100644 --- a/fs/xfs/kmem.h +++ b/fs/xfs/kmem.h @@ -110,10 +110,4 @@ kmem_zone_destroy(kmem_zone_t *zone) extern void *kmem_zone_alloc(kmem_zone_t *, unsigned int __nocast); extern void *kmem_zone_zalloc(kmem_zone_t *, unsigned int __nocast); -static inline int -kmem_shake_allow(gfp_t gfp_mask) -{ - return ((gfp_mask & __GFP_WAIT) && (gfp_mask & __GFP_FS)); -} - #endif /* __XFS_SUPPORT_KMEM_H__ */ diff --git a/fs/xfs/xfs_dquot.c b/fs/xfs/xfs_dquot.c index b4ff40b5f91..53db20ee3e7 100644 --- a/fs/xfs/xfs_dquot.c +++ b/fs/xfs/xfs_dquot.c @@ -63,82 +63,6 @@ int xfs_dqerror_mod = 33; static struct lock_class_key xfs_dquot_other_class; /* - * Allocate and initialize a dquot. We don't always allocate fresh memory; - * we try to reclaim a free dquot if the number of incore dquots are above - * a threshold. - * The only field inside the core that gets initialized at this point - * is the d_id field. The idea is to fill in the entire q_core - * when we read in the on disk dquot. - */ -STATIC xfs_dquot_t * -xfs_qm_dqinit( - xfs_mount_t *mp, - xfs_dqid_t id, - uint type) -{ - xfs_dquot_t *dqp; - boolean_t brandnewdquot; - - brandnewdquot = xfs_qm_dqalloc_incore(&dqp); - dqp->dq_flags = type; - dqp->q_core.d_id = cpu_to_be32(id); - dqp->q_mount = mp; - - /* - * No need to re-initialize these if this is a reclaimed dquot. - */ - if (brandnewdquot) { - INIT_LIST_HEAD(&dqp->q_freelist); - mutex_init(&dqp->q_qlock); - init_waitqueue_head(&dqp->q_pinwait); - - /* - * Because we want to use a counting completion, complete - * the flush completion once to allow a single access to - * the flush completion without blocking. - */ - init_completion(&dqp->q_flush); - complete(&dqp->q_flush); - - trace_xfs_dqinit(dqp); - } else { - /* - * Only the q_core portion was zeroed in dqreclaim_one(). - * So, we need to reset others. - */ - dqp->q_nrefs = 0; - dqp->q_blkno = 0; - INIT_LIST_HEAD(&dqp->q_mplist); - INIT_LIST_HEAD(&dqp->q_hashlist); - dqp->q_bufoffset = 0; - dqp->q_fileoffset = 0; - dqp->q_transp = NULL; - dqp->q_gdquot = NULL; - dqp->q_res_bcount = 0; - dqp->q_res_icount = 0; - dqp->q_res_rtbcount = 0; - atomic_set(&dqp->q_pincount, 0); - dqp->q_hash = NULL; - ASSERT(list_empty(&dqp->q_freelist)); - - trace_xfs_dqreuse(dqp); - } - - /* - * In either case we need to make sure group quotas have a different - * lock class than user quotas, to make sure lockdep knows we can - * locks of one of each at the same time. - */ - if (!(type & XFS_DQ_USER)) - lockdep_set_class(&dqp->q_qlock, &xfs_dquot_other_class); - - /* - * log item gets initialized later - */ - return (dqp); -} - -/* * This is called to free all the memory associated with a dquot */ void @@ -215,10 +139,10 @@ xfs_qm_adjust_dqtimers( if (!d->d_btimer) { if ((d->d_blk_softlimit && - (be64_to_cpu(d->d_bcount) >= + (be64_to_cpu(d->d_bcount) > be64_to_cpu(d->d_blk_softlimit))) || (d->d_blk_hardlimit && - (be64_to_cpu(d->d_bcount) >= + (be64_to_cpu(d->d_bcount) > be64_to_cpu(d->d_blk_hardlimit)))) { d->d_btimer = cpu_to_be32(get_seconds() + mp->m_quotainfo->qi_btimelimit); @@ -227,10 +151,10 @@ xfs_qm_adjust_dqtimers( } } else { if ((!d->d_blk_softlimit || - (be64_to_cpu(d->d_bcount) < + (be64_to_cpu(d->d_bcount) <= be64_to_cpu(d->d_blk_softlimit))) && (!d->d_blk_hardlimit || - (be64_to_cpu(d->d_bcount) < + (be64_to_cpu(d->d_bcount) <= be64_to_cpu(d->d_blk_hardlimit)))) { d->d_btimer = 0; } @@ -238,10 +162,10 @@ xfs_qm_adjust_dqtimers( if (!d->d_itimer) { if ((d->d_ino_softlimit && - (be64_to_cpu(d->d_icount) >= + (be64_to_cpu(d->d_icount) > be64_to_cpu(d->d_ino_softlimit))) || (d->d_ino_hardlimit && - (be64_to_cpu(d->d_icount) >= + (be64_to_cpu(d->d_icount) > be64_to_cpu(d->d_ino_hardlimit)))) { d->d_itimer = cpu_to_be32(get_seconds() + mp->m_quotainfo->qi_itimelimit); @@ -250,10 +174,10 @@ xfs_qm_adjust_dqtimers( } } else { if ((!d->d_ino_softlimit || - (be64_to_cpu(d->d_icount) < + (be64_to_cpu(d->d_icount) <= be64_to_cpu(d->d_ino_softlimit))) && (!d->d_ino_hardlimit || - (be64_to_cpu(d->d_icount) < + (be64_to_cpu(d->d_icount) <= be64_to_cpu(d->d_ino_hardlimit)))) { d->d_itimer = 0; } @@ -261,10 +185,10 @@ xfs_qm_adjust_dqtimers( if (!d->d_rtbtimer) { if ((d->d_rtb_softlimit && - (be64_to_cpu(d->d_rtbcount) >= + (be64_to_cpu(d->d_rtbcount) > be64_to_cpu(d->d_rtb_softlimit))) || (d->d_rtb_hardlimit && - (be64_to_cpu(d->d_rtbcount) >= + (be64_to_cpu(d->d_rtbcount) > be64_to_cpu(d->d_rtb_hardlimit)))) { d->d_rtbtimer = cpu_to_be32(get_seconds() + mp->m_quotainfo->qi_rtbtimelimit); @@ -273,10 +197,10 @@ xfs_qm_adjust_dqtimers( } } else { if ((!d->d_rtb_softlimit || - (be64_to_cpu(d->d_rtbcount) < + (be64_to_cpu(d->d_rtbcount) <= be64_to_cpu(d->d_rtb_softlimit))) && (!d->d_rtb_hardlimit || - (be64_to_cpu(d->d_rtbcount) < + (be64_to_cpu(d->d_rtbcount) <= be64_to_cpu(d->d_rtb_hardlimit)))) { d->d_rtbtimer = 0; } @@ -567,7 +491,32 @@ xfs_qm_dqread( int error; int cancelflags = 0; - dqp = xfs_qm_dqinit(mp, id, type); + + dqp = kmem_zone_zalloc(xfs_Gqm->qm_dqzone, KM_SLEEP); + + dqp->dq_flags = type; + dqp->q_core.d_id = cpu_to_be32(id); + dqp->q_mount = mp; + INIT_LIST_HEAD(&dqp->q_freelist); + mutex_init(&dqp->q_qlock); + init_waitqueue_head(&dqp->q_pinwait); + + /* + * Because we want to use a counting completion, complete + * the flush completion once to allow a single access to + * the flush completion without blocking. + */ + init_completion(&dqp->q_flush); + complete(&dqp->q_flush); + + /* + * Make sure group quotas have a different lock class than user + * quotas. + */ + if (!(type & XFS_DQ_USER)) + lockdep_set_class(&dqp->q_qlock, &xfs_dquot_other_class); + + atomic_inc(&xfs_Gqm->qm_totaldquots); trace_xfs_dqread(dqp); diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c index 541a508adea..0ed9ee77937 100644 --- a/fs/xfs/xfs_log_recover.c +++ b/fs/xfs/xfs_log_recover.c @@ -1489,7 +1489,7 @@ xlog_recover_add_to_cont_trans( old_ptr = item->ri_buf[item->ri_cnt-1].i_addr; old_len = item->ri_buf[item->ri_cnt-1].i_len; - ptr = kmem_realloc(old_ptr, len+old_len, old_len, 0u); + ptr = kmem_realloc(old_ptr, len+old_len, old_len, KM_SLEEP); memcpy(&ptr[old_len], dp, len); /* d, s, l */ item->ri_buf[item->ri_cnt-1].i_len += len; item->ri_buf[item->ri_cnt-1].i_addr = ptr; @@ -1981,7 +1981,7 @@ xfs_qm_dqcheck( if (!errs && ddq->d_id) { if (ddq->d_blk_softlimit && - be64_to_cpu(ddq->d_bcount) >= + be64_to_cpu(ddq->d_bcount) > be64_to_cpu(ddq->d_blk_softlimit)) { if (!ddq->d_btimer) { if (flags & XFS_QMOPT_DOWARN) @@ -1992,7 +1992,7 @@ xfs_qm_dqcheck( } } if (ddq->d_ino_softlimit && - be64_to_cpu(ddq->d_icount) >= + be64_to_cpu(ddq->d_icount) > be64_to_cpu(ddq->d_ino_softlimit)) { if (!ddq->d_itimer) { if (flags & XFS_QMOPT_DOWARN) @@ -2003,7 +2003,7 @@ xfs_qm_dqcheck( } } if (ddq->d_rtb_softlimit && - be64_to_cpu(ddq->d_rtbcount) >= + be64_to_cpu(ddq->d_rtbcount) > be64_to_cpu(ddq->d_rtb_softlimit)) { if (!ddq->d_rtbtimer) { if (flags & XFS_QMOPT_DOWARN) diff --git a/fs/xfs/xfs_qm.c b/fs/xfs/xfs_qm.c index 671f37eae1c..c436def733b 100644 --- a/fs/xfs/xfs_qm.c +++ b/fs/xfs/xfs_qm.c @@ -50,7 +50,6 @@ */ struct mutex xfs_Gqm_lock; struct xfs_qm *xfs_Gqm; -uint ndquot; kmem_zone_t *qm_dqzone; kmem_zone_t *qm_dqtrxzone; @@ -93,7 +92,6 @@ xfs_Gqm_init(void) goto out_free_udqhash; hsize /= sizeof(xfs_dqhash_t); - ndquot = hsize << 8; xqm = kmem_zalloc(sizeof(xfs_qm_t), KM_SLEEP); xqm->qm_dqhashmask = hsize - 1; @@ -137,7 +135,6 @@ xfs_Gqm_init(void) xqm->qm_dqtrxzone = qm_dqtrxzone; atomic_set(&xqm->qm_totaldquots, 0); - xqm->qm_dqfree_ratio = XFS_QM_DQFREE_RATIO; xqm->qm_nrefs = 0; return xqm; @@ -1600,216 +1597,150 @@ xfs_qm_init_quotainos( return 0; } +STATIC void +xfs_qm_dqfree_one( + struct xfs_dquot *dqp) +{ + struct xfs_mount *mp = dqp->q_mount; + struct xfs_quotainfo *qi = mp->m_quotainfo; + mutex_lock(&dqp->q_hash->qh_lock); + list_del_init(&dqp->q_hashlist); + dqp->q_hash->qh_version++; + mutex_unlock(&dqp->q_hash->qh_lock); -/* - * Pop the least recently used dquot off the freelist and recycle it. - */ -STATIC struct xfs_dquot * -xfs_qm_dqreclaim_one(void) + mutex_lock(&qi->qi_dqlist_lock); + list_del_init(&dqp->q_mplist); + qi->qi_dquots--; + qi->qi_dqreclaims++; + mutex_unlock(&qi->qi_dqlist_lock); + + xfs_qm_dqdestroy(dqp); +} + +STATIC void +xfs_qm_dqreclaim_one( + struct xfs_dquot *dqp, + struct list_head *dispose_list) { - struct xfs_dquot *dqp; - int restarts = 0; + struct xfs_mount *mp = dqp->q_mount; + int error; - mutex_lock(&xfs_Gqm->qm_dqfrlist_lock); -restart: - list_for_each_entry(dqp, &xfs_Gqm->qm_dqfrlist, q_freelist) { - struct xfs_mount *mp = dqp->q_mount; + if (!xfs_dqlock_nowait(dqp)) + goto out_busy; - if (!xfs_dqlock_nowait(dqp)) - continue; + /* + * This dquot has acquired a reference in the meantime remove it from + * the freelist and try again. + */ + if (dqp->q_nrefs) { + xfs_dqunlock(dqp); - /* - * This dquot has already been grabbed by dqlookup. - * Remove it from the freelist and try again. - */ - if (dqp->q_nrefs) { - trace_xfs_dqreclaim_want(dqp); - XQM_STATS_INC(xqmstats.xs_qm_dqwants); - - list_del_init(&dqp->q_freelist); - xfs_Gqm->qm_dqfrlist_cnt--; - restarts++; - goto dqunlock; - } + trace_xfs_dqreclaim_want(dqp); + XQM_STATS_INC(xqmstats.xs_qm_dqwants); - ASSERT(dqp->q_hash); - ASSERT(!list_empty(&dqp->q_mplist)); + list_del_init(&dqp->q_freelist); + xfs_Gqm->qm_dqfrlist_cnt--; + return; + } - /* - * Try to grab the flush lock. If this dquot is in the process - * of getting flushed to disk, we don't want to reclaim it. - */ - if (!xfs_dqflock_nowait(dqp)) - goto dqunlock; + ASSERT(dqp->q_hash); + ASSERT(!list_empty(&dqp->q_mplist)); - /* - * We have the flush lock so we know that this is not in the - * process of being flushed. So, if this is dirty, flush it - * DELWRI so that we don't get a freelist infested with - * dirty dquots. - */ - if (XFS_DQ_IS_DIRTY(dqp)) { - int error; + /* + * Try to grab the flush lock. If this dquot is in the process of + * getting flushed to disk, we don't want to reclaim it. + */ + if (!xfs_dqflock_nowait(dqp)) + goto out_busy; - trace_xfs_dqreclaim_dirty(dqp); + /* + * We have the flush lock so we know that this is not in the + * process of being flushed. So, if this is dirty, flush it + * DELWRI so that we don't get a freelist infested with + * dirty dquots. + */ + if (XFS_DQ_IS_DIRTY(dqp)) { + trace_xfs_dqreclaim_dirty(dqp); - /* - * We flush it delayed write, so don't bother - * releasing the freelist lock. - */ - error = xfs_qm_dqflush(dqp, SYNC_TRYLOCK); - if (error) { - xfs_warn(mp, "%s: dquot %p flush failed", - __func__, dqp); - } - goto dqunlock; + /* + * We flush it delayed write, so don't bother releasing the + * freelist lock. + */ + error = xfs_qm_dqflush(dqp, 0); + if (error) { + xfs_warn(mp, "%s: dquot %p flush failed", + __func__, dqp); } - xfs_dqfunlock(dqp); /* - * Prevent lookup now that we are going to reclaim the dquot. - * Once XFS_DQ_FREEING is set lookup won't touch the dquot, - * thus we can drop the lock now. + * Give the dquot another try on the freelist, as the + * flushing will take some time. */ - dqp->dq_flags |= XFS_DQ_FREEING; - xfs_dqunlock(dqp); - - mutex_lock(&dqp->q_hash->qh_lock); - list_del_init(&dqp->q_hashlist); - dqp->q_hash->qh_version++; - mutex_unlock(&dqp->q_hash->qh_lock); - - mutex_lock(&mp->m_quotainfo->qi_dqlist_lock); - list_del_init(&dqp->q_mplist); - mp->m_quotainfo->qi_dquots--; - mp->m_quotainfo->qi_dqreclaims++; - mutex_unlock(&mp->m_quotainfo->qi_dqlist_lock); + goto out_busy; + } + xfs_dqfunlock(dqp); - ASSERT(dqp->q_nrefs == 0); - list_del_init(&dqp->q_freelist); - xfs_Gqm->qm_dqfrlist_cnt--; + /* + * Prevent lookups now that we are past the point of no return. + */ + dqp->dq_flags |= XFS_DQ_FREEING; + xfs_dqunlock(dqp); - mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock); - return dqp; -dqunlock: - xfs_dqunlock(dqp); - if (restarts >= XFS_QM_RECLAIM_MAX_RESTARTS) - break; - goto restart; - } + ASSERT(dqp->q_nrefs == 0); + list_move_tail(&dqp->q_freelist, dispose_list); + xfs_Gqm->qm_dqfrlist_cnt--; - mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock); - return NULL; -} + trace_xfs_dqreclaim_done(dqp); + XQM_STATS_INC(xqmstats.xs_qm_dqreclaims); + return; -/* - * Traverse the freelist of dquots and attempt to reclaim a maximum of - * 'howmany' dquots. This operation races with dqlookup(), and attempts to - * favor the lookup function ... - */ -STATIC int -xfs_qm_shake_freelist( - int howmany) -{ - int nreclaimed = 0; - xfs_dquot_t *dqp; +out_busy: + xfs_dqunlock(dqp); - if (howmany <= 0) - return 0; + /* + * Move the dquot to the tail of the list so that we don't spin on it. + */ + list_move_tail(&dqp->q_freelist, &xfs_Gqm->qm_dqfrlist); - while (nreclaimed < howmany) { - dqp = xfs_qm_dqreclaim_one(); - if (!dqp) - return nreclaimed; - xfs_qm_dqdestroy(dqp); - nreclaimed++; - } - return nreclaimed; + trace_xfs_dqreclaim_busy(dqp); + XQM_STATS_INC(xqmstats.xs_qm_dqreclaim_misses); } -/* - * The kmem_shake interface is invoked when memory is running low. - */ -/* ARGSUSED */ STATIC int xfs_qm_shake( - struct shrinker *shrink, - struct shrink_control *sc) + struct shrinker *shrink, + struct shrink_control *sc) { - int ndqused, nfree, n; - gfp_t gfp_mask = sc->gfp_mask; - - if (!kmem_shake_allow(gfp_mask)) - return 0; - if (!xfs_Gqm) - return 0; - - nfree = xfs_Gqm->qm_dqfrlist_cnt; /* free dquots */ - /* incore dquots in all f/s's */ - ndqused = atomic_read(&xfs_Gqm->qm_totaldquots) - nfree; - - ASSERT(ndqused >= 0); + int nr_to_scan = sc->nr_to_scan; + LIST_HEAD (dispose_list); + struct xfs_dquot *dqp; - if (nfree <= ndqused && nfree < ndquot) + if ((sc->gfp_mask & (__GFP_FS|__GFP_WAIT)) != (__GFP_FS|__GFP_WAIT)) return 0; + if (!nr_to_scan) + goto out; - ndqused *= xfs_Gqm->qm_dqfree_ratio; /* target # of free dquots */ - n = nfree - ndqused - ndquot; /* # over target */ - - return xfs_qm_shake_freelist(MAX(nfree, n)); -} - - -/*------------------------------------------------------------------*/ - -/* - * Return a new incore dquot. Depending on the number of - * dquots in the system, we either allocate a new one on the kernel heap, - * or reclaim a free one. - * Return value is B_TRUE if we allocated a new dquot, B_FALSE if we managed - * to reclaim an existing one from the freelist. - */ -boolean_t -xfs_qm_dqalloc_incore( - xfs_dquot_t **O_dqpp) -{ - xfs_dquot_t *dqp; - - /* - * Check against high water mark to see if we want to pop - * a nincompoop dquot off the freelist. - */ - if (atomic_read(&xfs_Gqm->qm_totaldquots) >= ndquot) { - /* - * Try to recycle a dquot from the freelist. - */ - if ((dqp = xfs_qm_dqreclaim_one())) { - XQM_STATS_INC(xqmstats.xs_qm_dqreclaims); - /* - * Just zero the core here. The rest will get - * reinitialized by caller. XXX we shouldn't even - * do this zero ... - */ - memset(&dqp->q_core, 0, sizeof(dqp->q_core)); - *O_dqpp = dqp; - return B_FALSE; - } - XQM_STATS_INC(xqmstats.xs_qm_dqreclaim_misses); + mutex_lock(&xfs_Gqm->qm_dqfrlist_lock); + while (!list_empty(&xfs_Gqm->qm_dqfrlist)) { + if (nr_to_scan-- <= 0) + break; + dqp = list_first_entry(&xfs_Gqm->qm_dqfrlist, struct xfs_dquot, + q_freelist); + xfs_qm_dqreclaim_one(dqp, &dispose_list); } + mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock); - /* - * Allocate a brand new dquot on the kernel heap and return it - * to the caller to initialize. - */ - ASSERT(xfs_Gqm->qm_dqzone != NULL); - *O_dqpp = kmem_zone_zalloc(xfs_Gqm->qm_dqzone, KM_SLEEP); - atomic_inc(&xfs_Gqm->qm_totaldquots); - - return B_TRUE; + while (!list_empty(&dispose_list)) { + dqp = list_first_entry(&dispose_list, struct xfs_dquot, + q_freelist); + list_del_init(&dqp->q_freelist); + xfs_qm_dqfree_one(dqp); + } +out: + return (xfs_Gqm->qm_dqfrlist_cnt / 100) * sysctl_vfs_cache_pressure; } - /* * Start a transaction and write the incore superblock changes to * disk. flags parameter indicates which fields have changed. diff --git a/fs/xfs/xfs_qm.h b/fs/xfs/xfs_qm.h index 9b4f3adefbc..9a9b997e1a0 100644 --- a/fs/xfs/xfs_qm.h +++ b/fs/xfs/xfs_qm.h @@ -26,24 +26,12 @@ struct xfs_qm; struct xfs_inode; -extern uint ndquot; extern struct mutex xfs_Gqm_lock; extern struct xfs_qm *xfs_Gqm; extern kmem_zone_t *qm_dqzone; extern kmem_zone_t *qm_dqtrxzone; /* - * Ditto, for xfs_qm_dqreclaim_one. - */ -#define XFS_QM_RECLAIM_MAX_RESTARTS 4 - -/* - * Ideal ratio of free to in use dquots. Quota manager makes an attempt - * to keep this balance. - */ -#define XFS_QM_DQFREE_RATIO 2 - -/* * Dquot hashtable constants/threshold values. */ #define XFS_QM_HASHSIZE_LOW (PAGE_SIZE / sizeof(xfs_dqhash_t)) @@ -74,7 +62,6 @@ typedef struct xfs_qm { int qm_dqfrlist_cnt; atomic_t qm_totaldquots; /* total incore dquots */ uint qm_nrefs; /* file systems with quota on */ - int qm_dqfree_ratio;/* ratio of free to inuse dquots */ kmem_zone_t *qm_dqzone; /* dquot mem-alloc zone */ kmem_zone_t *qm_dqtrxzone; /* t_dqinfo of transactions */ } xfs_qm_t; @@ -143,7 +130,6 @@ extern int xfs_qm_quotacheck(xfs_mount_t *); extern int xfs_qm_write_sb_changes(xfs_mount_t *, __int64_t); /* dquot stuff */ -extern boolean_t xfs_qm_dqalloc_incore(xfs_dquot_t **); extern int xfs_qm_dqpurge_all(xfs_mount_t *, uint); extern void xfs_qm_dqrele_all_inodes(xfs_mount_t *, uint); diff --git a/fs/xfs/xfs_qm_stats.c b/fs/xfs/xfs_qm_stats.c index 8671a0b3264..5729ba57087 100644 --- a/fs/xfs/xfs_qm_stats.c +++ b/fs/xfs/xfs_qm_stats.c @@ -42,9 +42,9 @@ static int xqm_proc_show(struct seq_file *m, void *v) { /* maximum; incore; ratio free to inuse; freelist */ seq_printf(m, "%d\t%d\t%d\t%u\n", - ndquot, + 0, xfs_Gqm? atomic_read(&xfs_Gqm->qm_totaldquots) : 0, - xfs_Gqm? xfs_Gqm->qm_dqfree_ratio : 0, + 0, xfs_Gqm? xfs_Gqm->qm_dqfrlist_cnt : 0); return 0; } diff --git a/fs/xfs/xfs_qm_syscalls.c b/fs/xfs/xfs_qm_syscalls.c index eafbcff81f3..711a86e39ff 100644 --- a/fs/xfs/xfs_qm_syscalls.c +++ b/fs/xfs/xfs_qm_syscalls.c @@ -813,11 +813,11 @@ xfs_qm_export_dquot( (XFS_IS_OQUOTA_ENFORCED(mp) && (dst->d_flags & (FS_PROJ_QUOTA | FS_GROUP_QUOTA)))) && dst->d_id != 0) { - if (((int) dst->d_bcount >= (int) dst->d_blk_softlimit) && + if (((int) dst->d_bcount > (int) dst->d_blk_softlimit) && (dst->d_blk_softlimit > 0)) { ASSERT(dst->d_btimer != 0); } - if (((int) dst->d_icount >= (int) dst->d_ino_softlimit) && + if (((int) dst->d_icount > (int) dst->d_ino_softlimit) && (dst->d_ino_softlimit > 0)) { ASSERT(dst->d_itimer != 0); } diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h index 6b6df5802e9..bb134a81993 100644 --- a/fs/xfs/xfs_trace.h +++ b/fs/xfs/xfs_trace.h @@ -733,11 +733,10 @@ DEFINE_EVENT(xfs_dquot_class, name, \ DEFINE_DQUOT_EVENT(xfs_dqadjust); DEFINE_DQUOT_EVENT(xfs_dqreclaim_want); DEFINE_DQUOT_EVENT(xfs_dqreclaim_dirty); -DEFINE_DQUOT_EVENT(xfs_dqreclaim_unlink); +DEFINE_DQUOT_EVENT(xfs_dqreclaim_busy); +DEFINE_DQUOT_EVENT(xfs_dqreclaim_done); DEFINE_DQUOT_EVENT(xfs_dqattach_found); DEFINE_DQUOT_EVENT(xfs_dqattach_get); -DEFINE_DQUOT_EVENT(xfs_dqinit); -DEFINE_DQUOT_EVENT(xfs_dqreuse); DEFINE_DQUOT_EVENT(xfs_dqalloc); DEFINE_DQUOT_EVENT(xfs_dqtobp_read); DEFINE_DQUOT_EVENT(xfs_dqread); diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c index 329b06aba1c..7adcdf15ae0 100644 --- a/fs/xfs/xfs_trans.c +++ b/fs/xfs/xfs_trans.c @@ -1151,8 +1151,8 @@ xfs_trans_add_item( { struct xfs_log_item_desc *lidp; - ASSERT(lip->li_mountp = tp->t_mountp); - ASSERT(lip->li_ailp = tp->t_mountp->m_ail); + ASSERT(lip->li_mountp == tp->t_mountp); + ASSERT(lip->li_ailp == tp->t_mountp->m_ail); lidp = kmem_zone_zalloc(xfs_log_item_desc_zone, KM_SLEEP | KM_NOFS); diff --git a/fs/xfs/xfs_trans_dquot.c b/fs/xfs/xfs_trans_dquot.c index 4d00ee67792..c4ba366d24e 100644 --- a/fs/xfs/xfs_trans_dquot.c +++ b/fs/xfs/xfs_trans_dquot.c @@ -649,12 +649,12 @@ xfs_trans_dqresv( * nblks. */ if (hardlimit > 0ULL && - hardlimit <= nblks + *resbcountp) { + hardlimit < nblks + *resbcountp) { xfs_quota_warn(mp, dqp, QUOTA_NL_BHARDWARN); goto error_return; } if (softlimit > 0ULL && - softlimit <= nblks + *resbcountp) { + softlimit < nblks + *resbcountp) { if ((timer != 0 && get_seconds() > timer) || (warns != 0 && warns >= warnlimit)) { xfs_quota_warn(mp, dqp, @@ -677,11 +677,13 @@ xfs_trans_dqresv( if (!softlimit) softlimit = q->qi_isoftlimit; - if (hardlimit > 0ULL && count >= hardlimit) { + if (hardlimit > 0ULL && + hardlimit < ninos + count) { xfs_quota_warn(mp, dqp, QUOTA_NL_IHARDWARN); goto error_return; } - if (softlimit > 0ULL && count >= softlimit) { + if (softlimit > 0ULL && + softlimit < ninos + count) { if ((timer != 0 && get_seconds() > timer) || (warns != 0 && warns >= warnlimit)) { xfs_quota_warn(mp, dqp, |