1 files changed, 102 insertions, 22 deletions

diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index 099b3518feea..314f57164602 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -48,6 +48,7 @@
 #include "internal.h"
 #include "fscache.h"
 #include "dns_resolve.h"
+#include "pnfs.h"
 
 #define NFSDBG_FACILITY         NFSDBG_VFS
 
@@ -98,7 +99,7 @@ u64 nfs_compat_user_ino64(u64 fileid)
         return ino;
 }
 
-void nfs_clear_inode(struct inode *inode)
+static void nfs_clear_inode(struct inode *inode)
 {
         /*
          * The following should never happen...
@@ -110,6 +111,13 @@ void nfs_clear_inode(struct inode *inode)
         nfs_fscache_release_inode_cookie(inode);
 }
 
+void nfs_evict_inode(struct inode *inode)
+{
+        truncate_inode_pages(&inode->i_data, 0);
+        end_writeback(inode);
+        nfs_clear_inode(inode);
+}
+
 /**
  * nfs_sync_mapping - helper to flush all mmapped dirty data to disk
  */
@@ -227,9 +235,6 @@ nfs_init_locked(struct inode *inode, void *opaque)
         return 0;
 }
 
-/* Don't use READDIRPLUS on directories that we believe are too large */
-#define NFS_LIMIT_READDIRPLUS (8*PAGE_SIZE)
-
 /*
  * This is our front-end to iget that looks up inodes by file handle
  * instead of inode number.
@@ -284,8 +289,7 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr)
                 } else if (S_ISDIR(inode->i_mode)) {
                         inode->i_op = NFS_SB(sb)->nfs_client->rpc_ops->dir_inode_ops;
                         inode->i_fop = &nfs_dir_operations;
-                        if (nfs_server_capable(inode, NFS_CAP_READDIRPLUS)
-                            && fattr->size <= NFS_LIMIT_READDIRPLUS)
+                        if (nfs_server_capable(inode, NFS_CAP_READDIRPLUS))
                                 set_bit(NFS_INO_ADVISE_RDPLUS, &NFS_I(inode)->flags);
                         /* Deal with crossing mountpoints */
                         if ((fattr->valid & NFS_ATTR_FATTR_FSID)
@@ -413,10 +417,8 @@ nfs_setattr(struct dentry *dentry, struct iattr *attr)
                 return 0;
 
         /* Write all dirty data */
-        if (S_ISREG(inode->i_mode)) {
-                filemap_write_and_wait(inode->i_mapping);
+        if (S_ISREG(inode->i_mode))
                 nfs_wb_all(inode);
-        }
 
         fattr = nfs_alloc_fattr();
         if (fattr == NULL)
@@ -530,6 +532,68 @@ out:
         return err;
 }
 
+static void nfs_init_lock_context(struct nfs_lock_context *l_ctx)
+{
+        atomic_set(&l_ctx->count, 1);
+        l_ctx->lockowner = current->files;
+        l_ctx->pid = current->tgid;
+        INIT_LIST_HEAD(&l_ctx->list);
+}
+
+static struct nfs_lock_context *__nfs_find_lock_context(struct nfs_open_context *ctx)
+{
+        struct nfs_lock_context *pos;
+
+        list_for_each_entry(pos, &ctx->lock_context.list, list) {
+                if (pos->lockowner != current->files)
+                        continue;
+                if (pos->pid != current->tgid)
+                        continue;
+                atomic_inc(&pos->count);
+                return pos;
+        }
+        return NULL;
+}
+
+struct nfs_lock_context *nfs_get_lock_context(struct nfs_open_context *ctx)
+{
+        struct nfs_lock_context *res, *new = NULL;
+        struct inode *inode = ctx->path.dentry->d_inode;
+
+        spin_lock(&inode->i_lock);
+        res = __nfs_find_lock_context(ctx);
+        if (res == NULL) {
+                spin_unlock(&inode->i_lock);
+                new = kmalloc(sizeof(*new), GFP_KERNEL);
+                if (new == NULL)
+                        return NULL;
+                nfs_init_lock_context(new);
+                spin_lock(&inode->i_lock);
+                res = __nfs_find_lock_context(ctx);
+                if (res == NULL) {
+                        list_add_tail(&new->list, &ctx->lock_context.list);
+                        new->open_context = ctx;
+                        res = new;
+                        new = NULL;
+                }
+        }
+        spin_unlock(&inode->i_lock);
+        kfree(new);
+        return res;
+}
+
+void nfs_put_lock_context(struct nfs_lock_context *l_ctx)
+{
+        struct nfs_open_context *ctx = l_ctx->open_context;
+        struct inode *inode = ctx->path.dentry->d_inode;
+
+        if (!atomic_dec_and_lock(&l_ctx->count, &inode->i_lock))
+                return;
+        list_del(&l_ctx->list);
+        spin_unlock(&inode->i_lock);
+        kfree(l_ctx);
+}
+
 /**
  * nfs_close_context - Common close_context() routine NFSv2/v3
  * @ctx: pointer to context
@@ -556,7 +620,7 @@ void nfs_close_context(struct nfs_open_context *ctx, int is_sync)
         nfs_revalidate_inode(server, inode);
 }
 
-static struct nfs_open_context *alloc_nfs_open_context(struct path *path, struct rpc_cred *cred)
+struct nfs_open_context *alloc_nfs_open_context(struct path *path, struct rpc_cred *cred, fmode_t f_mode)
 {
         struct nfs_open_context *ctx;
 
@@ -566,11 +630,13 @@ static struct nfs_open_context *alloc_nfs_open_context(struct path *path, struct
                 path_get(&ctx->path);
                 ctx->cred = get_rpccred(cred);
                 ctx->state = NULL;
-                ctx->lockowner = current->files;
+                ctx->mode = f_mode;
                 ctx->flags = 0;
                 ctx->error = 0;
                 ctx->dir_cookie = 0;
-                atomic_set(&ctx->count, 1);
+                nfs_init_lock_context(&ctx->lock_context);
+                ctx->lock_context.open_context = ctx;
+                INIT_LIST_HEAD(&ctx->list);
         }
         return ctx;
 }
@@ -578,7 +644,7 @@ static struct nfs_open_context *alloc_nfs_open_context(struct path *path, struct
 struct nfs_open_context *get_nfs_open_context(struct nfs_open_context *ctx)
 {
         if (ctx != NULL)
-                atomic_inc(&ctx->count);
+                atomic_inc(&ctx->lock_context.count);
         return ctx;
 }
 
@@ -586,11 +652,15 @@ static void __put_nfs_open_context(struct nfs_open_context *ctx, int is_sync)
 {
         struct inode *inode = ctx->path.dentry->d_inode;
 
-        if (!atomic_dec_and_lock(&ctx->count, &inode->i_lock))
+        if (!list_empty(&ctx->list)) {
+                if (!atomic_dec_and_lock(&ctx->lock_context.count, &inode->i_lock))
+                        return;
+                list_del(&ctx->list);
+                spin_unlock(&inode->i_lock);
+        } else if (!atomic_dec_and_test(&ctx->lock_context.count))
                 return;
-        list_del(&ctx->list);
-        spin_unlock(&inode->i_lock);
-        NFS_PROTO(inode)->close_context(ctx, is_sync);
+        if (inode != NULL)
+                NFS_PROTO(inode)->close_context(ctx, is_sync);
         if (ctx->cred != NULL)
                 put_rpccred(ctx->cred);
         path_put(&ctx->path);
@@ -606,7 +676,7 @@ void put_nfs_open_context(struct nfs_open_context *ctx)
  * Ensure that mmap has a recent RPC credential for use when writing out
  * shared pages
  */
-static void nfs_file_set_open_context(struct file *filp, struct nfs_open_context *ctx)
+void nfs_file_set_open_context(struct file *filp, struct nfs_open_context *ctx)
 {
         struct inode *inode = filp->f_path.dentry->d_inode;
         struct nfs_inode *nfsi = NFS_I(inode);
@@ -663,11 +733,10 @@ int nfs_open(struct inode *inode, struct file *filp)
         cred = rpc_lookup_cred();
         if (IS_ERR(cred))
                 return PTR_ERR(cred);
-        ctx = alloc_nfs_open_context(&filp->f_path, cred);
+        ctx = alloc_nfs_open_context(&filp->f_path, cred, filp->f_mode);
         put_rpccred(cred);
         if (ctx == NULL)
                 return -ENOMEM;
-        ctx->mode = filp->f_mode;
         nfs_file_set_open_context(filp, ctx);
         put_nfs_open_context(ctx);
         nfs_fscache_set_inode_cookie(inode, filp);
@@ -1338,8 +1407,11 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr)
  * to open() calls that passed nfs_atomic_lookup, but failed to call
  * nfs_open().
  */
-void nfs4_clear_inode(struct inode *inode)
+void nfs4_evict_inode(struct inode *inode)
 {
+        truncate_inode_pages(&inode->i_data, 0);
+        end_writeback(inode);
+        pnfs_destroy_layout(NFS_I(inode));
         /* If we are holding a delegation, return it! */
         nfs_inode_return_delegation_noreclaim(inode);
         /* First call standard NFS clear_inode() code */
@@ -1377,6 +1449,7 @@ static inline void nfs4_init_once(struct nfs_inode *nfsi)
         nfsi->delegation = NULL;
         nfsi->delegation_state = 0;
         init_rwsem(&nfsi->rwsem);
+        nfsi->layout = NULL;
 #endif
 }
 
@@ -1424,7 +1497,7 @@ static int nfsiod_start(void)
 {
         struct workqueue_struct *wq;
         dprintk("RPC:       creating workqueue nfsiod\n");
-        wq = create_singlethread_workqueue("nfsiod");
+        wq = alloc_workqueue("nfsiod", WQ_RESCUER, 0);
         if (wq == NULL)
                 return -ENOMEM;
         nfsiod_workqueue = wq;
@@ -1452,6 +1525,10 @@ static int __init init_nfs_fs(void)
 {
         int err;
 
+        err = nfs_idmap_init();
+        if (err < 0)
+                goto out9;
+
         err = nfs_dns_resolver_init();
         if (err < 0)
                 goto out8;
@@ -1516,6 +1593,8 @@ out6:
 out7:
         nfs_dns_resolver_destroy();
 out8:
+        nfs_idmap_quit();
+out9:
         return err;
 }
 
@@ -1528,6 +1607,7 @@ static void __exit exit_nfs_fs(void)
         nfs_destroy_nfspagecache();
         nfs_fscache_unregister();
         nfs_dns_resolver_destroy();
+        nfs_idmap_quit();
 #ifdef CONFIG_PROC_FS
         rpc_proc_unregister("nfs");
 #endif