diff options
author | Paul Mackerras <paulus@samba.org> | 2007-10-03 15:33:17 +1000 |
---|---|---|
committer | Paul Mackerras <paulus@samba.org> | 2007-10-03 15:33:17 +1000 |
commit | 70f227d8846a8a9b1f36f71c42e11cc7c6e9408d (patch) | |
tree | fb4dd5c8240bdaada819fb569c01a392b52847b9 /fs | |
parent | a0c7ce9c877ceef8428798ac91fb794f83609aed (diff) | |
parent | f778089cb2445dfc6dfd30a7a567925fd8589f1e (diff) | |
download | op-kernel-dev-70f227d8846a8a9b1f36f71c42e11cc7c6e9408d.zip op-kernel-dev-70f227d8846a8a9b1f36f71c42e11cc7c6e9408d.tar.gz |
Merge branch 'linux-2.6' into for-2.6.24
Diffstat (limited to 'fs')
-rw-r--r-- | fs/compat_ioctl.c | 2 | ||||
-rw-r--r-- | fs/exec.c | 3 | ||||
-rw-r--r-- | fs/lockd/svclock.c | 29 | ||||
-rw-r--r-- | fs/nfs/client.c | 29 | ||||
-rw-r--r-- | fs/nfs/dir.c | 2 | ||||
-rw-r--r-- | fs/nfs/getroot.c | 3 | ||||
-rw-r--r-- | fs/ocfs2/aops.c | 33 | ||||
-rw-r--r-- | fs/ocfs2/file.c | 4 | ||||
-rw-r--r-- | fs/ocfs2/localalloc.c | 4 | ||||
-rw-r--r-- | fs/ocfs2/localalloc.h | 2 | ||||
-rw-r--r-- | fs/ocfs2/suballoc.c | 29 | ||||
-rw-r--r-- | fs/ocfs2/suballoc.h | 11 | ||||
-rw-r--r-- | fs/ocfs2/vote.c | 4 | ||||
-rw-r--r-- | fs/signalfd.c | 190 | ||||
-rw-r--r-- | fs/splice.c | 46 | ||||
-rw-r--r-- | fs/ufs/super.c | 4 | ||||
-rw-r--r-- | fs/xfs/xfs_buf_item.h | 5 | ||||
-rw-r--r-- | fs/xfs/xfs_filestream.c | 7 | ||||
-rw-r--r-- | fs/xfs/xfs_log_recover.c | 51 | ||||
-rw-r--r-- | fs/xfs/xfs_trans_buf.c | 1 |
20 files changed, 172 insertions, 287 deletions
diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c index 5a5b711..37310b0 100644 --- a/fs/compat_ioctl.c +++ b/fs/compat_ioctl.c @@ -3190,6 +3190,8 @@ COMPATIBLE_IOCTL(SIOCSIWRETRY) COMPATIBLE_IOCTL(SIOCGIWRETRY) COMPATIBLE_IOCTL(SIOCSIWPOWER) COMPATIBLE_IOCTL(SIOCGIWPOWER) +COMPATIBLE_IOCTL(SIOCSIWAUTH) +COMPATIBLE_IOCTL(SIOCGIWAUTH) /* hiddev */ COMPATIBLE_IOCTL(HIDIOCGVERSION) COMPATIBLE_IOCTL(HIDIOCAPPLICATION) @@ -50,7 +50,6 @@ #include <linux/tsacct_kern.h> #include <linux/cn_proc.h> #include <linux/audit.h> -#include <linux/signalfd.h> #include <asm/uaccess.h> #include <asm/mmu_context.h> @@ -784,7 +783,6 @@ static int de_thread(struct task_struct *tsk) * and we can just re-use it all. */ if (atomic_read(&oldsighand->count) <= 1) { - signalfd_detach(tsk); exit_itimers(sig); return 0; } @@ -923,7 +921,6 @@ static int de_thread(struct task_struct *tsk) sig->flags = 0; no_thread_group: - signalfd_detach(tsk); exit_itimers(sig); if (leader) release_task(leader); diff --git a/fs/lockd/svclock.c b/fs/lockd/svclock.c index a21e4bc..d098c7a 100644 --- a/fs/lockd/svclock.c +++ b/fs/lockd/svclock.c @@ -171,19 +171,14 @@ found: * GRANTED_RES message by cookie, without having to rely on the client's IP * address. --okir */ -static inline struct nlm_block * -nlmsvc_create_block(struct svc_rqst *rqstp, struct nlm_file *file, - struct nlm_lock *lock, struct nlm_cookie *cookie) +static struct nlm_block * +nlmsvc_create_block(struct svc_rqst *rqstp, struct nlm_host *host, + struct nlm_file *file, struct nlm_lock *lock, + struct nlm_cookie *cookie) { struct nlm_block *block; - struct nlm_host *host; struct nlm_rqst *call = NULL; - /* Create host handle for callback */ - host = nlmsvc_lookup_host(rqstp, lock->caller, lock->len); - if (host == NULL) - return NULL; - call = nlm_alloc_call(host); if (call == NULL) return NULL; @@ -366,6 +361,7 @@ nlmsvc_lock(struct svc_rqst *rqstp, struct nlm_file *file, struct nlm_lock *lock, int wait, struct nlm_cookie *cookie) { struct nlm_block *block = NULL; + struct nlm_host *host; int error; __be32 ret; @@ -377,6 +373,10 @@ nlmsvc_lock(struct svc_rqst *rqstp, struct nlm_file *file, (long long)lock->fl.fl_end, wait); + /* Create host handle for callback */ + host = nlmsvc_lookup_host(rqstp, lock->caller, lock->len); + if (host == NULL) + return nlm_lck_denied_nolocks; /* Lock file against concurrent access */ mutex_lock(&file->f_mutex); @@ -385,7 +385,8 @@ nlmsvc_lock(struct svc_rqst *rqstp, struct nlm_file *file, */ block = nlmsvc_lookup_block(file, lock); if (block == NULL) { - block = nlmsvc_create_block(rqstp, file, lock, cookie); + block = nlmsvc_create_block(rqstp, nlm_get_host(host), file, + lock, cookie); ret = nlm_lck_denied_nolocks; if (block == NULL) goto out; @@ -449,6 +450,7 @@ nlmsvc_lock(struct svc_rqst *rqstp, struct nlm_file *file, out: mutex_unlock(&file->f_mutex); nlmsvc_release_block(block); + nlm_release_host(host); dprintk("lockd: nlmsvc_lock returned %u\n", ret); return ret; } @@ -477,10 +479,15 @@ nlmsvc_testlock(struct svc_rqst *rqstp, struct nlm_file *file, if (block == NULL) { struct file_lock *conf = kzalloc(sizeof(*conf), GFP_KERNEL); + struct nlm_host *host; if (conf == NULL) return nlm_granted; - block = nlmsvc_create_block(rqstp, file, lock, cookie); + /* Create host handle for callback */ + host = nlmsvc_lookup_host(rqstp, lock->caller, lock->len); + if (host == NULL) + return nlm_lck_denied_nolocks; + block = nlmsvc_create_block(rqstp, host, file, lock, cookie); if (block == NULL) { kfree(conf); return nlm_granted; diff --git a/fs/nfs/client.c b/fs/nfs/client.c index a49f9fe..a204484 100644 --- a/fs/nfs/client.c +++ b/fs/nfs/client.c @@ -588,16 +588,6 @@ static int nfs_init_server(struct nfs_server *server, const struct nfs_mount_dat server->namelen = data->namlen; /* Create a client RPC handle for the NFSv3 ACL management interface */ nfs_init_server_aclclient(server); - if (clp->cl_nfsversion == 3) { - if (server->namelen == 0 || server->namelen > NFS3_MAXNAMLEN) - server->namelen = NFS3_MAXNAMLEN; - if (!(data->flags & NFS_MOUNT_NORDIRPLUS)) - server->caps |= NFS_CAP_READDIRPLUS; - } else { - if (server->namelen == 0 || server->namelen > NFS2_MAXNAMLEN) - server->namelen = NFS2_MAXNAMLEN; - } - dprintk("<-- nfs_init_server() = 0 [new %p]\n", clp); return 0; @@ -794,6 +784,16 @@ struct nfs_server *nfs_create_server(const struct nfs_mount_data *data, error = nfs_probe_fsinfo(server, mntfh, &fattr); if (error < 0) goto error; + if (server->nfs_client->rpc_ops->version == 3) { + if (server->namelen == 0 || server->namelen > NFS3_MAXNAMLEN) + server->namelen = NFS3_MAXNAMLEN; + if (!(data->flags & NFS_MOUNT_NORDIRPLUS)) + server->caps |= NFS_CAP_READDIRPLUS; + } else { + if (server->namelen == 0 || server->namelen > NFS2_MAXNAMLEN) + server->namelen = NFS2_MAXNAMLEN; + } + if (!(fattr.valid & NFS_ATTR_FATTR)) { error = server->nfs_client->rpc_ops->getattr(server, mntfh, &fattr); if (error < 0) { @@ -984,6 +984,9 @@ struct nfs_server *nfs4_create_server(const struct nfs4_mount_data *data, if (error < 0) goto error; + if (server->namelen == 0 || server->namelen > NFS4_MAXNAMLEN) + server->namelen = NFS4_MAXNAMLEN; + BUG_ON(!server->nfs_client); BUG_ON(!server->nfs_client->rpc_ops); BUG_ON(!server->nfs_client->rpc_ops->file_inode_ops); @@ -1056,6 +1059,9 @@ struct nfs_server *nfs4_create_referral_server(struct nfs_clone_mount *data, if (error < 0) goto error; + if (server->namelen == 0 || server->namelen > NFS4_MAXNAMLEN) + server->namelen = NFS4_MAXNAMLEN; + dprintk("Referral FSID: %llx:%llx\n", (unsigned long long) server->fsid.major, (unsigned long long) server->fsid.minor); @@ -1115,6 +1121,9 @@ struct nfs_server *nfs_clone_server(struct nfs_server *source, if (error < 0) goto out_free_server; + if (server->namelen == 0 || server->namelen > NFS4_MAXNAMLEN) + server->namelen = NFS4_MAXNAMLEN; + dprintk("Cloned FSID: %llx:%llx\n", (unsigned long long) server->fsid.major, (unsigned long long) server->fsid.minor); diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index ea97408..e4a04d1 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -1162,6 +1162,8 @@ static struct dentry *nfs_readdir_lookup(nfs_readdir_descriptor_t *desc) } if (!desc->plus || !(entry->fattr->valid & NFS_ATTR_FATTR)) return NULL; + if (name.len > NFS_SERVER(dir)->namelen) + return NULL; /* Note: caller is already holding the dir->i_mutex! */ dentry = d_alloc(parent, &name); if (dentry == NULL) diff --git a/fs/nfs/getroot.c b/fs/nfs/getroot.c index d1cbf0a..522e5ad 100644 --- a/fs/nfs/getroot.c +++ b/fs/nfs/getroot.c @@ -175,6 +175,9 @@ next_component: path++; name.len = path - (const char *) name.name; + if (name.len > NFS4_MAXNAMLEN) + return -ENAMETOOLONG; + eat_dot_dir: while (*path == '/') path++; diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c index 50cd8a2..f37f25c 100644 --- a/fs/ocfs2/aops.c +++ b/fs/ocfs2/aops.c @@ -930,18 +930,11 @@ static void ocfs2_write_failure(struct inode *inode, loff_t user_pos, unsigned user_len) { int i; - unsigned from, to; + unsigned from = user_pos & (PAGE_CACHE_SIZE - 1), + to = user_pos + user_len; struct page *tmppage; - ocfs2_zero_new_buffers(wc->w_target_page, user_pos, user_len); - - if (wc->w_large_pages) { - from = wc->w_target_from; - to = wc->w_target_to; - } else { - from = 0; - to = PAGE_CACHE_SIZE; - } + ocfs2_zero_new_buffers(wc->w_target_page, from, to); for(i = 0; i < wc->w_num_pages; i++) { tmppage = wc->w_pages[i]; @@ -991,9 +984,6 @@ static int ocfs2_prepare_page_for_write(struct inode *inode, u64 *p_blkno, map_from = cluster_start; map_to = cluster_end; } - - wc->w_target_from = map_from; - wc->w_target_to = map_to; } else { /* * If we haven't allocated the new page yet, we @@ -1211,18 +1201,33 @@ static int ocfs2_write_cluster_by_desc(struct address_space *mapping, loff_t pos, unsigned len) { int ret, i; + loff_t cluster_off; + unsigned int local_len = len; struct ocfs2_write_cluster_desc *desc; + struct ocfs2_super *osb = OCFS2_SB(mapping->host->i_sb); for (i = 0; i < wc->w_clen; i++) { desc = &wc->w_desc[i]; + /* + * We have to make sure that the total write passed in + * doesn't extend past a single cluster. + */ + local_len = len; + cluster_off = pos & (osb->s_clustersize - 1); + if ((cluster_off + local_len) > osb->s_clustersize) + local_len = osb->s_clustersize - cluster_off; + ret = ocfs2_write_cluster(mapping, desc->c_phys, desc->c_unwritten, data_ac, meta_ac, - wc, desc->c_cpos, pos, len); + wc, desc->c_cpos, pos, local_len); if (ret) { mlog_errno(ret); goto out; } + + len -= local_len; + pos += local_len; } ret = 0; diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index 7e34e66..f3bc365 100644 --- a/fs/ocfs2/file.c +++ b/fs/ocfs2/file.c @@ -491,8 +491,8 @@ int ocfs2_do_extend_allocation(struct ocfs2_super *osb, goto leave; } - status = ocfs2_claim_clusters(osb, handle, data_ac, 1, - &bit_off, &num_bits); + status = __ocfs2_claim_clusters(osb, handle, data_ac, 1, + clusters_to_add, &bit_off, &num_bits); if (status < 0) { if (status != -ENOSPC) mlog_errno(status); diff --git a/fs/ocfs2/localalloc.c b/fs/ocfs2/localalloc.c index 545f789..de984d2 100644 --- a/fs/ocfs2/localalloc.c +++ b/fs/ocfs2/localalloc.c @@ -524,13 +524,12 @@ bail: int ocfs2_claim_local_alloc_bits(struct ocfs2_super *osb, handle_t *handle, struct ocfs2_alloc_context *ac, - u32 min_bits, + u32 bits_wanted, u32 *bit_off, u32 *num_bits) { int status, start; struct inode *local_alloc_inode; - u32 bits_wanted; void *bitmap; struct ocfs2_dinode *alloc; struct ocfs2_local_alloc *la; @@ -538,7 +537,6 @@ int ocfs2_claim_local_alloc_bits(struct ocfs2_super *osb, mlog_entry_void(); BUG_ON(ac->ac_which != OCFS2_AC_USE_LOCAL); - bits_wanted = ac->ac_bits_wanted - ac->ac_bits_given; local_alloc_inode = ac->ac_inode; alloc = (struct ocfs2_dinode *) osb->local_alloc_bh->b_data; la = OCFS2_LOCAL_ALLOC(alloc); diff --git a/fs/ocfs2/localalloc.h b/fs/ocfs2/localalloc.h index 385a101..3f76631 100644 --- a/fs/ocfs2/localalloc.h +++ b/fs/ocfs2/localalloc.h @@ -48,7 +48,7 @@ int ocfs2_reserve_local_alloc_bits(struct ocfs2_super *osb, int ocfs2_claim_local_alloc_bits(struct ocfs2_super *osb, handle_t *handle, struct ocfs2_alloc_context *ac, - u32 min_bits, + u32 bits_wanted, u32 *bit_off, u32 *num_bits); diff --git a/fs/ocfs2/suballoc.c b/fs/ocfs2/suballoc.c index d9c5c9f..8f09f52 100644 --- a/fs/ocfs2/suballoc.c +++ b/fs/ocfs2/suballoc.c @@ -1486,21 +1486,21 @@ static inline void ocfs2_block_to_cluster_group(struct inode *inode, * contig. allocation, set to '1' to indicate we can deal with extents * of any size. */ -int ocfs2_claim_clusters(struct ocfs2_super *osb, - handle_t *handle, - struct ocfs2_alloc_context *ac, - u32 min_clusters, - u32 *cluster_start, - u32 *num_clusters) +int __ocfs2_claim_clusters(struct ocfs2_super *osb, + handle_t *handle, + struct ocfs2_alloc_context *ac, + u32 min_clusters, + u32 max_clusters, + u32 *cluster_start, + u32 *num_clusters) { int status; - unsigned int bits_wanted = ac->ac_bits_wanted - ac->ac_bits_given; + unsigned int bits_wanted = max_clusters; u64 bg_blkno = 0; u16 bg_bit_off; mlog_entry_void(); - BUG_ON(!ac); BUG_ON(ac->ac_bits_given >= ac->ac_bits_wanted); BUG_ON(ac->ac_which != OCFS2_AC_USE_LOCAL @@ -1557,6 +1557,19 @@ bail: return status; } +int ocfs2_claim_clusters(struct ocfs2_super *osb, + handle_t *handle, + struct ocfs2_alloc_context *ac, + u32 min_clusters, + u32 *cluster_start, + u32 *num_clusters) +{ + unsigned int bits_wanted = ac->ac_bits_wanted - ac->ac_bits_given; + + return __ocfs2_claim_clusters(osb, handle, ac, min_clusters, + bits_wanted, cluster_start, num_clusters); +} + static inline int ocfs2_block_group_clear_bits(handle_t *handle, struct inode *alloc_inode, struct ocfs2_group_desc *bg, diff --git a/fs/ocfs2/suballoc.h b/fs/ocfs2/suballoc.h index f212dc0..cafe937 100644 --- a/fs/ocfs2/suballoc.h +++ b/fs/ocfs2/suballoc.h @@ -85,6 +85,17 @@ int ocfs2_claim_clusters(struct ocfs2_super *osb, u32 min_clusters, u32 *cluster_start, u32 *num_clusters); +/* + * Use this variant of ocfs2_claim_clusters to specify a maxiumum + * number of clusters smaller than the allocation reserved. + */ +int __ocfs2_claim_clusters(struct ocfs2_super *osb, + handle_t *handle, + struct ocfs2_alloc_context *ac, + u32 min_clusters, + u32 max_clusters, + u32 *cluster_start, + u32 *num_clusters); int ocfs2_free_suballoc_bits(handle_t *handle, struct inode *alloc_inode, diff --git a/fs/ocfs2/vote.c b/fs/ocfs2/vote.c index 66a13ee..c053585 100644 --- a/fs/ocfs2/vote.c +++ b/fs/ocfs2/vote.c @@ -66,7 +66,7 @@ struct ocfs2_vote_msg { struct ocfs2_msg_hdr v_hdr; __be32 v_reserved1; -}; +} __attribute__ ((packed)); /* Responses are given these values to maintain backwards * compatibility with older ocfs2 versions */ @@ -78,7 +78,7 @@ struct ocfs2_response_msg { struct ocfs2_msg_hdr r_hdr; __be32 r_response; -}; +} __attribute__ ((packed)); struct ocfs2_vote_work { struct list_head w_list; diff --git a/fs/signalfd.c b/fs/signalfd.c index a8e293d..aefb0be 100644 --- a/fs/signalfd.c +++ b/fs/signalfd.c @@ -11,8 +11,10 @@ * Now using anonymous inode source. * Thanks to Oleg Nesterov for useful code review and suggestions. * More comments and suggestions from Arnd Bergmann. - * Sat May 19, 2007: Davi E. M. Arnaut <davi@haxent.com.br> + * Sat May 19, 2007: Davi E. M. Arnaut <davi@haxent.com.br> * Retrieve multiple signals with one read() call + * Sun Jul 15, 2007: Davide Libenzi <davidel@xmailserver.org> + * Attach to the sighand only during read() and poll(). */ #include <linux/file.h> @@ -27,102 +29,12 @@ #include <linux/signalfd.h> struct signalfd_ctx { - struct list_head lnk; - wait_queue_head_t wqh; sigset_t sigmask; - struct task_struct *tsk; }; -struct signalfd_lockctx { - struct task_struct *tsk; - unsigned long flags; -}; - -/* - * Tries to acquire the sighand lock. We do not increment the sighand - * use count, and we do not even pin the task struct, so we need to - * do it inside an RCU read lock, and we must be prepared for the - * ctx->tsk going to NULL (in signalfd_deliver()), and for the sighand - * being detached. We return 0 if the sighand has been detached, or - * 1 if we were able to pin the sighand lock. - */ -static int signalfd_lock(struct signalfd_ctx *ctx, struct signalfd_lockctx *lk) -{ - struct sighand_struct *sighand = NULL; - - rcu_read_lock(); - lk->tsk = rcu_dereference(ctx->tsk); - if (likely(lk->tsk != NULL)) - sighand = lock_task_sighand(lk->tsk, &lk->flags); - rcu_read_unlock(); - - if (!sighand) - return 0; - - if (!ctx->tsk) { - unlock_task_sighand(lk->tsk, &lk->flags); - return 0; - } - - if (lk->tsk->tgid == current->tgid) - lk->tsk = current; - - return 1; -} - -static void signalfd_unlock(struct signalfd_lockctx *lk) -{ - unlock_task_sighand(lk->tsk, &lk->flags); -} - -/* - * This must be called with the sighand lock held. - */ -void signalfd_deliver(struct task_struct *tsk, int sig) -{ - struct sighand_struct *sighand = tsk->sighand; - struct signalfd_ctx *ctx, *tmp; - - BUG_ON(!sig); - list_for_each_entry_safe(ctx, tmp, &sighand->signalfd_list, lnk) { - /* - * We use a negative signal value as a way to broadcast that the - * sighand has been orphaned, so that we can notify all the - * listeners about this. Remember the ctx->sigmask is inverted, - * so if the user is interested in a signal, that corresponding - * bit will be zero. - */ - if (sig < 0) { - if (ctx->tsk == tsk) { - ctx->tsk = NULL; - list_del_init(&ctx->lnk); - wake_up(&ctx->wqh); - } - } else { - if (!sigismember(&ctx->sigmask, sig)) - wake_up(&ctx->wqh); - } - } -} - -static void signalfd_cleanup(struct signalfd_ctx *ctx) -{ - struct signalfd_lockctx lk; - - /* - * This is tricky. If the sighand is gone, we do not need to remove - * context from the list, the list itself won't be there anymore. - */ - if (signalfd_lock(ctx, &lk)) { - list_del(&ctx->lnk); - signalfd_unlock(&lk); - } - kfree(ctx); -} - static int signalfd_release(struct inode *inode, struct file *file) { - signalfd_cleanup(file->private_data); + kfree(file->private_data); return 0; } @@ -130,23 +42,15 @@ static unsigned int signalfd_poll(struct file *file, poll_table *wait) { struct signalfd_ctx *ctx = file->private_data; unsigned int events = 0; - struct signalfd_lockctx lk; - poll_wait(file, &ctx->wqh, wait); + poll_wait(file, ¤t->sighand->signalfd_wqh, wait); - /* - * Let the caller get a POLLIN in this case, ala socket recv() when - * the peer disconnects. - */ - if (signalfd_lock(ctx, &lk)) { - if ((lk.tsk == current && - next_signal(&lk.tsk->pending, &ctx->sigmask) > 0) || - next_signal(&lk.tsk->signal->shared_pending, - &ctx->sigmask) > 0) - events |= POLLIN; - signalfd_unlock(&lk); - } else + spin_lock_irq(¤t->sighand->siglock); + if (next_signal(¤t->pending, &ctx->sigmask) || + next_signal(¤t->signal->shared_pending, + &ctx->sigmask)) events |= POLLIN; + spin_unlock_irq(¤t->sighand->siglock); return events; } @@ -219,59 +123,46 @@ static ssize_t signalfd_dequeue(struct signalfd_ctx *ctx, siginfo_t *info, int nonblock) { ssize_t ret; - struct signalfd_lockctx lk; DECLARE_WAITQUEUE(wait, current); - if (!signalfd_lock(ctx, &lk)) - return 0; - - ret = dequeue_signal(lk.tsk, &ctx->sigmask, info); + spin_lock_irq(¤t->sighand->siglock); + ret = dequeue_signal(current, &ctx->sigmask, info); switch (ret) { case 0: if (!nonblock) break; ret = -EAGAIN; default: - signalfd_unlock(&lk); + spin_unlock_irq(¤t->sighand->siglock); return ret; } - add_wait_queue(&ctx->wqh, &wait); + add_wait_queue(¤t->sighand->signalfd_wqh, &wait); for (;;) { set_current_state(TASK_INTERRUPTIBLE); - ret = dequeue_signal(lk.tsk, &ctx->sigmask, info); - signalfd_unlock(&lk); + ret = dequeue_signal(current, &ctx->sigmask, info); if (ret != 0) break; if (signal_pending(current)) { ret = -ERESTARTSYS; break; } + spin_unlock_irq(¤t->sighand->siglock); schedule(); - ret = signalfd_lock(ctx, &lk); - if (unlikely(!ret)) { - /* - * Let the caller read zero byte, ala socket - * recv() when the peer disconnect. This test - * must be done before doing a dequeue_signal(), - * because if the sighand has been orphaned, - * the dequeue_signal() call is going to crash - * because ->sighand will be long gone. - */ - break; - } + spin_lock_irq(¤t->sighand->siglock); } + spin_unlock_irq(¤t->sighand->siglock); - remove_wait_queue(&ctx->wqh, &wait); + remove_wait_queue(¤t->sighand->signalfd_wqh, &wait); __set_current_state(TASK_RUNNING); return ret; } /* - * Returns either the size of a "struct signalfd_siginfo", or zero if the - * sighand we are attached to, has been orphaned. The "count" parameter - * must be at least the size of a "struct signalfd_siginfo". + * Returns a multiple of the size of a "struct signalfd_siginfo", or a negative + * error code. The "count" parameter must be at least the size of a + * "struct signalfd_siginfo". */ static ssize_t signalfd_read(struct file *file, char __user *buf, size_t count, loff_t *ppos) @@ -287,7 +178,6 @@ static ssize_t signalfd_read(struct file *file, char __user *buf, size_t count, return -EINVAL; siginfo = (struct signalfd_siginfo __user *) buf; - do { ret = signalfd_dequeue(ctx, &info, nonblock); if (unlikely(ret <= 0)) @@ -300,7 +190,7 @@ static ssize_t signalfd_read(struct file *file, char __user *buf, size_t count, nonblock = 1; } while (--count); - return total ? total : ret; + return total ? total: ret; } static const struct file_operations signalfd_fops = { @@ -309,20 +199,13 @@ static const struct file_operations signalfd_fops = { .read = signalfd_read, }; -/* - * Create a file descriptor that is associated with our signal - * state. We can pass it around to others if we want to, but - * it will always be _our_ signal state. - */ asmlinkage long sys_signalfd(int ufd, sigset_t __user *user_mask, size_t sizemask) { int error; sigset_t sigmask; struct signalfd_ctx *ctx; - struct sighand_struct *sighand; struct file *file; struct inode *inode; - struct signalfd_lockctx lk; if (sizemask != sizeof(sigset_t) || copy_from_user(&sigmask, user_mask, sizeof(sigmask))) @@ -335,17 +218,7 @@ asmlinkage long sys_signalfd(int ufd, sigset_t __user *user_mask, size_t sizemas if (!ctx) return -ENOMEM; - init_waitqueue_head(&ctx->wqh); ctx->sigmask = sigmask; - ctx->tsk = current->group_leader; - - sighand = current->sighand; - /* - * Add this fd to the list of signal listeners. - */ - spin_lock_irq(&sighand->siglock); - list_add_tail(&ctx->lnk, &sighand->signalfd_list); - spin_unlock_irq(&sighand->siglock); /* * When we call this, the initialization must be complete, since @@ -364,23 +237,18 @@ asmlinkage long sys_signalfd(int ufd, sigset_t __user *user_mask, size_t sizemas fput(file); return -EINVAL; } - /* - * We need to be prepared of the fact that the sighand this fd - * is attached to, has been detched. In that case signalfd_lock() - * will return 0, and we'll just skip setting the new mask. - */ - if (signalfd_lock(ctx, &lk)) { - ctx->sigmask = sigmask; - signalfd_unlock(&lk); - } - wake_up(&ctx->wqh); + spin_lock_irq(¤t->sighand->siglock); + ctx->sigmask = sigmask; + spin_unlock_irq(¤t->sighand->siglock); + + wake_up(¤t->sighand->signalfd_wqh); fput(file); } return ufd; err_fdalloc: - signalfd_cleanup(ctx); + kfree(ctx); return error; } diff --git a/fs/splice.c b/fs/splice.c index c010a72..e95a362 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -1224,6 +1224,33 @@ static long do_splice(struct file *in, loff_t __user *off_in, } /* + * Do a copy-from-user while holding the mmap_semaphore for reading, in a + * manner safe from deadlocking with simultaneous mmap() (grabbing mmap_sem + * for writing) and page faulting on the user memory pointed to by src. + * This assumes that we will very rarely hit the partial != 0 path, or this + * will not be a win. + */ +static int copy_from_user_mmap_sem(void *dst, const void __user *src, size_t n) +{ + int partial; + + pagefault_disable(); + partial = __copy_from_user_inatomic(dst, src, n); + pagefault_enable(); + + /* + * Didn't copy everything, drop the mmap_sem and do a faulting copy + */ + if (unlikely(partial)) { + up_read(¤t->mm->mmap_sem); + partial = copy_from_user(dst, src, n); + down_read(¤t->mm->mmap_sem); + } + + return partial; +} + +/* * Map an iov into an array of pages and offset/length tupples. With the * partial_page structure, we can map several non-contiguous ranges into * our ones pages[] map instead of splitting that operation into pieces. @@ -1236,31 +1263,26 @@ static int get_iovec_page_array(const struct iovec __user *iov, { int buffers = 0, error = 0; - /* - * It's ok to take the mmap_sem for reading, even - * across a "get_user()". - */ down_read(¤t->mm->mmap_sem); while (nr_vecs) { unsigned long off, npages; + struct iovec entry; void __user *base; size_t len; int i; - /* - * Get user address base and length for this iovec. - */ - error = get_user(base, &iov->iov_base); - if (unlikely(error)) - break; - error = get_user(len, &iov->iov_len); - if (unlikely(error)) + error = -EFAULT; + if (copy_from_user_mmap_sem(&entry, iov, sizeof(entry))) break; + base = entry.iov_base; + len = entry.iov_len; + /* * Sanity check this iovec. 0 read succeeds. */ + error = 0; if (unlikely(!len)) break; error = -EFAULT; diff --git a/fs/ufs/super.c b/fs/ufs/super.c index 73402c5..38eb0b7 100644 --- a/fs/ufs/super.c +++ b/fs/ufs/super.c @@ -894,7 +894,7 @@ magic_found: goto again; } - + sbi->s_flags = flags;/*after that line some functions use s_flags*/ ufs_print_super_stuff(sb, usb1, usb2, usb3); /* @@ -1025,8 +1025,6 @@ magic_found: UFS_MOUNT_UFSTYPE_44BSD) uspi->s_maxsymlinklen = fs32_to_cpu(sb, usb3->fs_un2.fs_44.fs_maxsymlinklen); - - sbi->s_flags = flags; inode = iget(sb, UFS_ROOTINO); if (!inode || is_bad_inode(inode)) diff --git a/fs/xfs/xfs_buf_item.h b/fs/xfs/xfs_buf_item.h index fa25b7d..d7e1361 100644 --- a/fs/xfs/xfs_buf_item.h +++ b/fs/xfs/xfs_buf_item.h @@ -52,11 +52,6 @@ typedef struct xfs_buf_log_format_t { #define XFS_BLI_UDQUOT_BUF 0x4 #define XFS_BLI_PDQUOT_BUF 0x8 #define XFS_BLI_GDQUOT_BUF 0x10 -/* - * This flag indicates that the buffer contains newly allocated - * inodes. - */ -#define XFS_BLI_INODE_NEW_BUF 0x20 #define XFS_BLI_CHUNK 128 #define XFS_BLI_SHIFT 7 diff --git a/fs/xfs/xfs_filestream.c b/fs/xfs/xfs_filestream.c index 16f8e17..36d8f6a 100644 --- a/fs/xfs/xfs_filestream.c +++ b/fs/xfs/xfs_filestream.c @@ -350,9 +350,10 @@ _xfs_filestream_update_ag( /* xfs_fstrm_free_func(): callback for freeing cached stream items. */ void xfs_fstrm_free_func( - xfs_ino_t ino, - fstrm_item_t *item) + unsigned long ino, + void *data) { + fstrm_item_t *item = (fstrm_item_t *)data; xfs_inode_t *ip = item->ip; int ref; @@ -438,7 +439,7 @@ xfs_filestream_mount( grp_count = 10; err = xfs_mru_cache_create(&mp->m_filestream, lifetime, grp_count, - (xfs_mru_cache_free_func_t)xfs_fstrm_free_func); + xfs_fstrm_free_func); return err; } diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c index dacb197..8ae6e8e 100644 --- a/fs/xfs/xfs_log_recover.c +++ b/fs/xfs/xfs_log_recover.c @@ -1874,7 +1874,6 @@ xlog_recover_do_inode_buffer( /*ARGSUSED*/ STATIC void xlog_recover_do_reg_buffer( - xfs_mount_t *mp, xlog_recover_item_t *item, xfs_buf_t *bp, xfs_buf_log_format_t *buf_f) @@ -1885,50 +1884,6 @@ xlog_recover_do_reg_buffer( unsigned int *data_map = NULL; unsigned int map_size = 0; int error; - int stale_buf = 1; - - /* - * Scan through the on-disk inode buffer and attempt to - * determine if it has been written to since it was logged. - * - * - If any of the magic numbers are incorrect then the buffer is stale - * - If any of the modes are non-zero then the buffer is not stale - * - If all of the modes are zero and at least one of the generation - * counts is non-zero then the buffer is stale - * - * If the end result is a stale buffer then the log buffer is replayed - * otherwise it is skipped. - * - * This heuristic is not perfect. It can be improved by scanning the - * entire inode chunk for evidence that any of the inode clusters have - * been updated. To fix this problem completely we will need a major - * architectural change to the logging system. - */ - if (buf_f->blf_flags & XFS_BLI_INODE_NEW_BUF) { - xfs_dinode_t *dip; - int inodes_per_buf; - int mode_count = 0; - int gen_count = 0; - - stale_buf = 0; - inodes_per_buf = XFS_BUF_COUNT(bp) >> mp->m_sb.sb_inodelog; - for (i = 0; i < inodes_per_buf; i++) { - dip = (xfs_dinode_t *)xfs_buf_offset(bp, - i * mp->m_sb.sb_inodesize); - if (be16_to_cpu(dip->di_core.di_magic) != - XFS_DINODE_MAGIC) { - stale_buf = 1; - break; - } - if (be16_to_cpu(dip->di_core.di_mode)) - mode_count++; - if (be16_to_cpu(dip->di_core.di_gen)) - gen_count++; - } - - if (!mode_count && gen_count) - stale_buf = 1; - } switch (buf_f->blf_type) { case XFS_LI_BUF: @@ -1962,7 +1917,7 @@ xlog_recover_do_reg_buffer( -1, 0, XFS_QMOPT_DOWARN, "dquot_buf_recover"); } - if (!error && stale_buf) + if (!error) memcpy(xfs_buf_offset(bp, (uint)bit << XFS_BLI_SHIFT), /* dest */ item->ri_buf[i].i_addr, /* source */ @@ -2134,7 +2089,7 @@ xlog_recover_do_dquot_buffer( if (log->l_quotaoffs_flag & type) return; - xlog_recover_do_reg_buffer(mp, item, bp, buf_f); + xlog_recover_do_reg_buffer(item, bp, buf_f); } /* @@ -2235,7 +2190,7 @@ xlog_recover_do_buffer_trans( (XFS_BLI_UDQUOT_BUF|XFS_BLI_PDQUOT_BUF|XFS_BLI_GDQUOT_BUF)) { xlog_recover_do_dquot_buffer(mp, log, item, bp, buf_f); } else { - xlog_recover_do_reg_buffer(mp, item, bp, buf_f); + xlog_recover_do_reg_buffer(item, bp, buf_f); } if (error) return XFS_ERROR(error); diff --git a/fs/xfs/xfs_trans_buf.c b/fs/xfs/xfs_trans_buf.c index 95fff68..60b6b89 100644 --- a/fs/xfs/xfs_trans_buf.c +++ b/fs/xfs/xfs_trans_buf.c @@ -966,7 +966,6 @@ xfs_trans_inode_alloc_buf( ASSERT(atomic_read(&bip->bli_refcount) > 0); bip->bli_flags |= XFS_BLI_INODE_ALLOC_BUF; - bip->bli_format.blf_flags |= XFS_BLI_INODE_NEW_BUF; } |