From 1c3c07e9f6cc50dab2aeb8051325e317d4f6c70e Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Tue, 25 Jul 2006 11:28:18 -0400 Subject: NFS: Add a new ACCESS rpc call cache to the linux nfs client The current access cache only allows one entry at a time to be cached for each inode. Add a per-inode red-black tree in order to allow more than one to be cached at a time. Should significantly cut down the time spent in path traversal for shared directories such as ${PATH}, /usr/share, etc. Signed-off-by: Trond Myklebust --- include/linux/nfs_fs.h | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 6c2066caeaa..cc013ed2e52 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -42,6 +42,7 @@ #include #include #include +#include #include #include @@ -69,6 +70,7 @@ * NFSv3/v4 Access mode cache entry */ struct nfs_access_entry { + struct rb_node rb_node; unsigned long jiffies; struct rpc_cred * cred; int mask; @@ -145,7 +147,7 @@ struct nfs_inode { */ atomic_t data_updates; - struct nfs_access_entry cache_access; + struct rb_root access_cache; #ifdef CONFIG_NFS_V3_ACL struct posix_acl *acl_access; struct posix_acl *acl_default; @@ -297,6 +299,7 @@ extern int nfs_getattr(struct vfsmount *, struct dentry *, struct kstat *); extern int nfs_permission(struct inode *, int, struct nameidata *); extern int nfs_access_get_cached(struct inode *, struct rpc_cred *, struct nfs_access_entry *); extern void nfs_access_add_cache(struct inode *, struct nfs_access_entry *); +extern void nfs_access_zap_cache(struct inode *inode); extern int nfs_open(struct inode *, struct file *); extern int nfs_release(struct inode *, struct file *); extern int nfs_attribute_timeout(struct inode *inode); -- cgit v1.2.3 From cfcea3e8c66c2dcde98d5c2693d4bff50b5cac97 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Tue, 25 Jul 2006 11:28:18 -0400 Subject: NFS: Add a global LRU list for the ACCESS cache ...in order to allow the addition of a memory shrinker. Signed-off-by: Trond Myklebust --- include/linux/nfs_fs.h | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'include/linux') diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index cc013ed2e52..a36e01cd632 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -71,6 +71,7 @@ */ struct nfs_access_entry { struct rb_node rb_node; + struct list_head lru; unsigned long jiffies; struct rpc_cred * cred; int mask; @@ -148,6 +149,8 @@ struct nfs_inode { atomic_t data_updates; struct rb_root access_cache; + struct list_head access_cache_entry_lru; + struct list_head access_cache_inode_lru; #ifdef CONFIG_NFS_V3_ACL struct posix_acl *acl_access; struct posix_acl *acl_default; @@ -201,6 +204,7 @@ struct nfs_inode { #define NFS_INO_REVALIDATING (0) /* revalidating attrs */ #define NFS_INO_ADVISE_RDPLUS (1) /* advise readdirplus */ #define NFS_INO_STALE (2) /* possible stale inode */ +#define NFS_INO_ACL_LRU_SET (3) /* Inode is on the LRU list */ static inline struct nfs_inode *NFS_I(struct inode *inode) { -- cgit v1.2.3 From 770bfad846ab6628444428467b11fa6773ae9ea1 Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 22 Aug 2006 20:06:07 -0400 Subject: NFS: Add dentry materialisation op The attached patch adds a new directory cache management function that prepares a disconnected anonymous function to be connected into the dentry tree. The anonymous dentry is transferred the name and parentage from another dentry. The following changes were made in [try #2]: (*) d_materialise_dentry() now switches the parentage of the two nodes around correctly when one or other of them is self-referential. The following changes were made in [try #7]: (*) d_instantiate_unique() has had the interior part split out as function __d_instantiate_unique(). Callers of this latter function must be holding the appropriate locks. (*) _d_rehash() has been added as a wrapper around __d_rehash() to call it with the most obvious hash list (the one from the name). d_rehash() now calls _d_rehash(). (*) d_materialise_dentry() is now __d_materialise_dentry() and is static. (*) d_materialise_unique() added to perform the combination of d_find_alias(), d_materialise_dentry() and d_add_unique() that the NFS client was doing twice, all within a single dcache_lock critical section. This reduces the number of times two different spinlocks were being accessed. The following further changes were made: (*) Add the dentries onto their parents d_subdirs lists. Signed-Off-By: David Howells Signed-off-by: Trond Myklebust --- include/linux/dcache.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/dcache.h b/include/linux/dcache.h index 471781ffeab..44605be5940 100644 --- a/include/linux/dcache.h +++ b/include/linux/dcache.h @@ -221,6 +221,7 @@ static inline int dname_external(struct dentry *dentry) */ extern void d_instantiate(struct dentry *, struct inode *); extern struct dentry * d_instantiate_unique(struct dentry *, struct inode *); +extern struct dentry * d_materialise_unique(struct dentry *, struct inode *); extern void d_delete(struct dentry *); /* allocate/de-allocate */ -- cgit v1.2.3 From adfa6f980bd46974e6b32b22dd0c45e3f52063f4 Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 22 Aug 2006 20:06:08 -0400 Subject: NFS: Rename struct nfs4_client to struct nfs_client Rename struct nfs4_client to struct nfs_client so that it can become the basis for a general client record for NFS2 and NFS3 in addition to NFS4. Signed-Off-By: David Howells Signed-off-by: Trond Myklebust --- include/linux/nfs_fs_sb.h | 2 +- include/linux/nfs_idmap.h | 14 +++++++------- 2 files changed, 8 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h index 6b4a13c7947..4db90df2aed 100644 --- a/include/linux/nfs_fs_sb.h +++ b/include/linux/nfs_fs_sb.h @@ -43,7 +43,7 @@ struct nfs_server { */ char ip_addr[16]; char * mnt_path; - struct nfs4_client * nfs4_state; /* all NFSv4 state starts here */ + struct nfs_client * nfs4_state; /* all NFSv4 state starts here */ struct list_head nfs4_siblings; /* List of other nfs_server structs * that share the same clientid */ diff --git a/include/linux/nfs_idmap.h b/include/linux/nfs_idmap.h index 102e5609429..678fe68982e 100644 --- a/include/linux/nfs_idmap.h +++ b/include/linux/nfs_idmap.h @@ -62,15 +62,15 @@ struct idmap_msg { #ifdef __KERNEL__ /* Forward declaration to make this header independent of others */ -struct nfs4_client; +struct nfs_client; -void nfs_idmap_new(struct nfs4_client *); -void nfs_idmap_delete(struct nfs4_client *); +void nfs_idmap_new(struct nfs_client *); +void nfs_idmap_delete(struct nfs_client *); -int nfs_map_name_to_uid(struct nfs4_client *, const char *, size_t, __u32 *); -int nfs_map_group_to_gid(struct nfs4_client *, const char *, size_t, __u32 *); -int nfs_map_uid_to_name(struct nfs4_client *, __u32, char *); -int nfs_map_gid_to_group(struct nfs4_client *, __u32, char *); +int nfs_map_name_to_uid(struct nfs_client *, const char *, size_t, __u32 *); +int nfs_map_group_to_gid(struct nfs_client *, const char *, size_t, __u32 *); +int nfs_map_uid_to_name(struct nfs_client *, __u32, char *); +int nfs_map_gid_to_group(struct nfs_client *, __u32, char *); extern unsigned int nfs_idmap_cache_timeout; #endif /* __KERNEL__ */ -- cgit v1.2.3 From 7539bbab8062aadc1db95a22b377146843cfa88f Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 22 Aug 2006 20:06:09 -0400 Subject: NFS: Rename nfs_server::nfs4_state Rename nfs_server::nfs4_state to nfs_client as it will be used to represent the client state for NFS2 and NFS3 also. Signed-Off-By: David Howells Signed-off-by: Trond Myklebust --- include/linux/nfs_fs_sb.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h index 4db90df2aed..fc20d6b934f 100644 --- a/include/linux/nfs_fs_sb.h +++ b/include/linux/nfs_fs_sb.h @@ -43,7 +43,7 @@ struct nfs_server { */ char ip_addr[16]; char * mnt_path; - struct nfs_client * nfs4_state; /* all NFSv4 state starts here */ + struct nfs_client * nfs_client; /* all NFSv4 state starts here */ struct list_head nfs4_siblings; /* List of other nfs_server structs * that share the same clientid */ -- cgit v1.2.3 From b7162792b5c0e0f6e91b8997f8e6bbc76ec5420a Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 22 Aug 2006 20:06:09 -0400 Subject: NFS: Return an error when starting the idmapping pipe Return an error when starting the idmapping pipe so that we can detect it failing. Signed-Off-By: David Howells Signed-off-by: Trond Myklebust --- include/linux/nfs_idmap.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/nfs_idmap.h b/include/linux/nfs_idmap.h index 678fe68982e..15a9f3b7289 100644 --- a/include/linux/nfs_idmap.h +++ b/include/linux/nfs_idmap.h @@ -64,7 +64,7 @@ struct idmap_msg { /* Forward declaration to make this header independent of others */ struct nfs_client; -void nfs_idmap_new(struct nfs_client *); +int nfs_idmap_new(struct nfs_client *); void nfs_idmap_delete(struct nfs_client *); int nfs_map_name_to_uid(struct nfs_client *, const char *, size_t, __u32 *); -- cgit v1.2.3 From 2b3de4411b3ccaeb00018c99d1bbe7203554cf7f Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 22 Aug 2006 20:06:09 -0400 Subject: NFS: Add a lookupfh NFS RPC op Add a lookup filehandle NFS RPC op so that a file handle can be looked up without requiring dentries and inodes and other VFS stuff when doing an NFS4 pathwalk during mounting. Signed-Off-By: David Howells Signed-off-by: Trond Myklebust --- include/linux/nfs_xdr.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include/linux') diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index 41e5a19199e..26879771831 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -770,6 +770,9 @@ struct nfs_rpc_ops { int (*getroot) (struct nfs_server *, struct nfs_fh *, struct nfs_fsinfo *); + int (*lookupfh)(struct nfs_server *, struct nfs_fh *, + struct qstr *, struct nfs_fh *, + struct nfs_fattr *); int (*getattr) (struct nfs_server *, struct nfs_fh *, struct nfs_fattr *); int (*setattr) (struct dentry *, struct nfs_fattr *, -- cgit v1.2.3 From e9326dcab413848e70ab746c7c5363da13e5f801 Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 22 Aug 2006 20:06:10 -0400 Subject: NFS: Add a server capabilities NFS RPC op Add a set_capabilities NFS RPC op so that the server capabilities can be set. Signed-Off-By: David Howells Signed-off-by: Trond Myklebust --- include/linux/nfs_xdr.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index 26879771831..dd9ae6761f7 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -809,6 +809,7 @@ struct nfs_rpc_ops { struct nfs_fsinfo *); int (*pathconf) (struct nfs_server *, struct nfs_fh *, struct nfs_pathconf *); + int (*set_capabilities)(struct nfs_server *, struct nfs_fh *); u32 * (*decode_dirent)(u32 *, struct nfs_entry *, int plus); void (*read_setup) (struct nfs_read_data *); int (*read_done) (struct rpc_task *, struct nfs_read_data *); -- cgit v1.2.3 From 24c8dbbb5f777187d660393599641ab3307b4b97 Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 22 Aug 2006 20:06:10 -0400 Subject: NFS: Generalise the nfs_client structure Generalise the nfs_client structure by: (1) Moving nfs_client to a more general place (nfs_fs_sb.h). (2) Renaming its maintenance routines to be non-NFS4 specific. (3) Move those maintenance routines to a new non-NFS4 specific file (client.c) and move the declarations to internal.h. (4) Make nfs_find/get_client() take a full sockaddr_in to include the port number (will be required for NFS2/3). (5) Make nfs_find/get_client() take the NFS protocol version (again will be required to differentiate NFS2, 3 & 4 client records). Also: (6) Make nfs_client construction proceed akin to inodes, marking them as under construction and providing a function to indicate completion. (7) Make nfs_get_client() wait interruptibly if it finds a client that it can share, but that client is currently being constructed. (8) Make nfs4_create_client() use (6) and (7) instead of locking cl_sem. Signed-Off-By: David Howells Signed-off-by: Trond Myklebust --- include/linux/nfs_fs.h | 1 + include/linux/nfs_fs_sb.h | 60 +++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 61 insertions(+) (limited to 'include/linux') diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index a36e01cd632..70e1dc9162e 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -586,6 +586,7 @@ extern void * nfs_root_data(void); #define NFSDBG_FILE 0x0040 #define NFSDBG_ROOT 0x0080 #define NFSDBG_CALLBACK 0x0100 +#define NFSDBG_CLIENT 0x0200 #define NFSDBG_ALL 0xFFFF #ifdef __KERNEL__ diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h index fc20d6b934f..a727657e0ad 100644 --- a/include/linux/nfs_fs_sb.h +++ b/include/linux/nfs_fs_sb.h @@ -6,6 +6,66 @@ struct nfs_iostats; +/* + * The nfs_client identifies our client state to the server. + */ +struct nfs_client { + atomic_t cl_count; + int cl_cons_state; /* current construction state (-ve: init error) */ +#define NFS_CS_READY 0 /* ready to be used */ +#define NFS_CS_INITING 1 /* busy initialising */ + int cl_nfsversion; /* NFS protocol version */ + unsigned long cl_res_state; /* NFS resources state */ +#define NFS_CS_RPCIOD 0 /* - rpciod started */ +#define NFS_CS_CALLBACK 1 /* - callback started */ +#define NFS_CS_IDMAP 2 /* - idmap started */ + struct sockaddr_in cl_addr; /* server identifier */ + char * cl_hostname; /* hostname of server */ + struct list_head cl_share_link; /* link in global client list */ + struct list_head cl_superblocks; /* List of nfs_server structs */ + + struct rpc_clnt * cl_rpcclient; + +#ifdef CONFIG_NFS_V4 + u64 cl_clientid; /* constant */ + nfs4_verifier cl_confirm; + unsigned long cl_state; + + u32 cl_lockowner_id; + + /* + * The following rwsem ensures exclusive access to the server + * while we recover the state following a lease expiration. + */ + struct rw_semaphore cl_sem; + + struct list_head cl_delegations; + struct list_head cl_state_owners; + struct list_head cl_unused; + int cl_nunused; + spinlock_t cl_lock; + + unsigned long cl_lease_time; + unsigned long cl_last_renewal; + struct work_struct cl_renewd; + struct work_struct cl_recoverd; + + struct rpc_wait_queue cl_rpcwaitq; + + /* used for the setclientid verifier */ + struct timespec cl_boot_time; + + /* idmapper */ + struct idmap * cl_idmap; + + /* Our own IP address, as a null-terminated string. + * This is used to generate the clientid, and the callback address. + */ + char cl_ipaddr[16]; + unsigned char cl_id_uniquifier; +#endif +}; + /* * NFS client parameters stored in the superblock. */ -- cgit v1.2.3 From 509de8111656a7d89b4a1a5f430f4460ce510f0f Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 22 Aug 2006 20:06:11 -0400 Subject: NFS: Add extra const qualifiers Add some extra const qualifiers into NFS. Signed-Off-By: David Howells Signed-off-by: Trond Myklebust --- include/linux/nfs_fs_sb.h | 2 +- include/linux/nfs_xdr.h | 6 +++--- 2 files changed, 4 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h index a727657e0ad..95f32d5f6e9 100644 --- a/include/linux/nfs_fs_sb.h +++ b/include/linux/nfs_fs_sb.h @@ -73,7 +73,7 @@ struct nfs_server { struct rpc_clnt * client; /* RPC client handle */ struct rpc_clnt * client_sys; /* 2nd handle for FSINFO */ struct rpc_clnt * client_acl; /* ACL RPC client handle */ - struct nfs_rpc_ops * rpc_ops; /* NFS protocol vector */ + const struct nfs_rpc_ops *rpc_ops; /* NFS protocol vector */ struct nfs_iostats * io_stats; /* I/O statistics */ struct backing_dev_info backing_dev_info; int flags; /* various flags */ diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index dd9ae6761f7..2426b11b6cc 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -833,9 +833,9 @@ struct nfs_rpc_ops { /* * Function vectors etc. for the NFS client */ -extern struct nfs_rpc_ops nfs_v2_clientops; -extern struct nfs_rpc_ops nfs_v3_clientops; -extern struct nfs_rpc_ops nfs_v4_clientops; +extern const struct nfs_rpc_ops nfs_v2_clientops; +extern const struct nfs_rpc_ops nfs_v3_clientops; +extern const struct nfs_rpc_ops nfs_v4_clientops; extern struct rpc_version nfs_version2; extern struct rpc_version nfs_version3; extern struct rpc_version nfs_version4; -- cgit v1.2.3 From 27951bd26031f6c27d38df9e94623bbe208a2464 Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 22 Aug 2006 20:06:11 -0400 Subject: NFS: Maintain a common server record for NFS2/3 as well as for NFS4 Maintain a common server record for NFS2/3 as well as for NFS4 so that common stuff can be moved there from struct nfs_server. Signed-Off-By: David Howells Signed-off-by: Trond Myklebust --- include/linux/nfs_fs_sb.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h index 95f32d5f6e9..e7d7662f51f 100644 --- a/include/linux/nfs_fs_sb.h +++ b/include/linux/nfs_fs_sb.h @@ -70,6 +70,7 @@ struct nfs_client { * NFS client parameters stored in the superblock. */ struct nfs_server { + struct nfs_client * nfs_client; /* shared client and NFS4 state */ struct rpc_clnt * client; /* RPC client handle */ struct rpc_clnt * client_sys; /* 2nd handle for FSINFO */ struct rpc_clnt * client_acl; /* ACL RPC client handle */ @@ -103,7 +104,6 @@ struct nfs_server { */ char ip_addr[16]; char * mnt_path; - struct nfs_client * nfs_client; /* all NFSv4 state starts here */ struct list_head nfs4_siblings; /* List of other nfs_server structs * that share the same clientid */ -- cgit v1.2.3 From 8fa5c000d7f986ef9cdc6d95f9f7fcee20e0a7d6 Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 22 Aug 2006 20:06:12 -0400 Subject: NFS: Move rpc_ops from nfs_server to nfs_client Move the rpc_ops from the nfs_server struct to the nfs_client struct as they're common to all server records of a particular NFS protocol version. Signed-Off-By: David Howells Signed-off-by: Trond Myklebust --- include/linux/nfs_fs.h | 2 +- include/linux/nfs_fs_sb.h | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 70e1dc9162e..51e9bd90ded 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -215,7 +215,7 @@ static inline struct nfs_inode *NFS_I(struct inode *inode) #define NFS_FH(inode) (&NFS_I(inode)->fh) #define NFS_SERVER(inode) (NFS_SB(inode->i_sb)) #define NFS_CLIENT(inode) (NFS_SERVER(inode)->client) -#define NFS_PROTO(inode) (NFS_SERVER(inode)->rpc_ops) +#define NFS_PROTO(inode) (NFS_SERVER(inode)->nfs_client->rpc_ops) #define NFS_ADDR(inode) (RPC_PEERADDR(NFS_CLIENT(inode))) #define NFS_COOKIEVERF(inode) (NFS_I(inode)->cookieverf) #define NFS_READTIME(inode) (NFS_I(inode)->read_cache_jiffies) diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h index e7d7662f51f..aae7c117597 100644 --- a/include/linux/nfs_fs_sb.h +++ b/include/linux/nfs_fs_sb.h @@ -25,6 +25,7 @@ struct nfs_client { struct list_head cl_superblocks; /* List of nfs_server structs */ struct rpc_clnt * cl_rpcclient; + const struct nfs_rpc_ops *rpc_ops; /* NFS protocol vector */ #ifdef CONFIG_NFS_V4 u64 cl_clientid; /* constant */ @@ -74,7 +75,6 @@ struct nfs_server { struct rpc_clnt * client; /* RPC client handle */ struct rpc_clnt * client_sys; /* 2nd handle for FSINFO */ struct rpc_clnt * client_acl; /* ACL RPC client handle */ - const struct nfs_rpc_ops *rpc_ops; /* NFS protocol vector */ struct nfs_iostats * io_stats; /* I/O statistics */ struct backing_dev_info backing_dev_info; int flags; /* various flags */ -- cgit v1.2.3 From 5006a76cca8f86c6975c16fcf67e83b8b0eee2b6 Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 22 Aug 2006 20:06:12 -0400 Subject: NFS: Eliminate client_sys in favour of cl_rpcclient Eliminate nfs_server::client_sys in favour of nfs_client::cl_rpcclient as we only really need one per server that we're talking to since it doesn't have any security on it. The retransmission management variables are also moved to the common struct as they're required to set up the cl_rpcclient connection. The NFS2/3 client and client_acl connections are thenceforth derived by cloning the cl_rpcclient connection and post-applying the authorisation flavour. The code for setting up the initial common connection has been moved to client.c as nfs_create_rpc_client(). All the NFS program definition tables are also moved there as that's where they're now required rather than super.c. Signed-Off-By: David Howells Signed-off-by: Trond Myklebust --- include/linux/nfs_fs_sb.h | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h index aae7c117597..d404ceca916 100644 --- a/include/linux/nfs_fs_sb.h +++ b/include/linux/nfs_fs_sb.h @@ -26,6 +26,8 @@ struct nfs_client { struct rpc_clnt * cl_rpcclient; const struct nfs_rpc_ops *rpc_ops; /* NFS protocol vector */ + unsigned long retrans_timeo; /* retransmit timeout */ + unsigned int retrans_count; /* number of retransmit tries */ #ifdef CONFIG_NFS_V4 u64 cl_clientid; /* constant */ @@ -73,7 +75,6 @@ struct nfs_client { struct nfs_server { struct nfs_client * nfs_client; /* shared client and NFS4 state */ struct rpc_clnt * client; /* RPC client handle */ - struct rpc_clnt * client_sys; /* 2nd handle for FSINFO */ struct rpc_clnt * client_acl; /* ACL RPC client handle */ struct nfs_iostats * io_stats; /* I/O statistics */ struct backing_dev_info backing_dev_info; @@ -90,8 +91,6 @@ struct nfs_server { unsigned int acregmax; unsigned int acdirmin; unsigned int acdirmax; - unsigned long retrans_timeo; /* retransmit timeout */ - unsigned int retrans_count; /* number of retransmit tries */ unsigned int namelen; char * hostname; /* remote hostname */ struct nfs_fh fh; -- cgit v1.2.3 From 54ceac4515986030c2502960be620198dd8fe25b Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 22 Aug 2006 20:06:13 -0400 Subject: NFS: Share NFS superblocks per-protocol per-server per-FSID The attached patch makes NFS share superblocks between mounts from the same server and FSID over the same protocol. It does this by creating each superblock with a false root and returning the real root dentry in the vfsmount presented by get_sb(). The root dentry set starts off as an anonymous dentry if we don't already have the dentry for its inode, otherwise it simply returns the dentry we already have. We may thus end up with several trees of dentries in the superblock, and if at some later point one of anonymous tree roots is discovered by normal filesystem activity to be located in another tree within the superblock, the anonymous root is named and materialises attached to the second tree at the appropriate point. Why do it this way? Why not pass an extra argument to the mount() syscall to indicate the subpath and then pathwalk from the server root to the desired directory? You can't guarantee this will work for two reasons: (1) The root and intervening nodes may not be accessible to the client. With NFS2 and NFS3, for instance, mountd is called on the server to get the filehandle for the tip of a path. mountd won't give us handles for anything we don't have permission to access, and so we can't set up NFS inodes for such nodes, and so can't easily set up dentries (we'd have to have ghost inodes or something). With this patch we don't actually create dentries until we get handles from the server that we can use to set up their inodes, and we don't actually bind them into the tree until we know for sure where they go. (2) Inaccessible symbolic links. If we're asked to mount two exports from the server, eg: mount warthog:/warthog/aaa/xxx /mmm mount warthog:/warthog/bbb/yyy /nnn We may not be able to access anything nearer the root than xxx and yyy, but we may find out later that /mmm/www/yyy, say, is actually the same directory as the one mounted on /nnn. What we might then find out, for example, is that /warthog/bbb was actually a symbolic link to /warthog/aaa/xxx/www, but we can't actually determine that by talking to the server until /warthog is made available by NFS. This would lead to having constructed an errneous dentry tree which we can't easily fix. We can end up with a dentry marked as a directory when it should actually be a symlink, or we could end up with an apparently hardlinked directory. With this patch we need not make assumptions about the type of a dentry for which we can't retrieve information, nor need we assume we know its place in the grand scheme of things until we actually see that place. This patch reduces the possibility of aliasing in the inode and page caches for inodes that may be accessed by more than one NFS export. It also reduces the number of superblocks required for NFS where there are many NFS exports being used from a server (home directory server + autofs for example). This in turn makes it simpler to do local caching of network filesystems, as it can then be guaranteed that there won't be links from multiple inodes in separate superblocks to the same cache file. Obviously, cache aliasing between different levels of NFS protocol could still be a problem, but at least that gives us another key to use when indexing the cache. This patch makes the following changes: (1) The server record construction/destruction has been abstracted out into its own set of functions to make things easier to get right. These have been moved into fs/nfs/client.c. All the code in fs/nfs/client.c has to do with the management of connections to servers, and doesn't touch superblocks in any way; the remaining code in fs/nfs/super.c has to do with VFS superblock management. (2) The sequence of events undertaken by NFS mount is now reordered: (a) A volume representation (struct nfs_server) is allocated. (b) A server representation (struct nfs_client) is acquired. This may be allocated or shared, and is keyed on server address, port and NFS version. (c) If allocated, the client representation is initialised. The state member variable of nfs_client is used to prevent a race during initialisation from two mounts. (d) For NFS4 a simple pathwalk is performed, walking from FH to FH to find the root filehandle for the mount (fs/nfs/getroot.c). For NFS2/3 we are given the root FH in advance. (e) The volume FSID is probed for on the root FH. (f) The volume representation is initialised from the FSINFO record retrieved on the root FH. (g) sget() is called to acquire a superblock. This may be allocated or shared, keyed on client pointer and FSID. (h) If allocated, the superblock is initialised. (i) If the superblock is shared, then the new nfs_server record is discarded. (j) The root dentry for this mount is looked up from the root FH. (k) The root dentry for this mount is assigned to the vfsmount. (3) nfs_readdir_lookup() creates dentries for each of the entries readdir() returns; this function now attaches disconnected trees from alternate roots that happen to be discovered attached to a directory being read (in the same way nfs_lookup() is made to do for lookup ops). The new d_materialise_unique() function is now used to do this, thus permitting the whole thing to be done under one set of locks, and thus avoiding any race between mount and lookup operations on the same directory. (4) The client management code uses a new debug facility: NFSDBG_CLIENT which is set by echoing 1024 to /proc/net/sunrpc/nfs_debug. (5) Clone mounts are now called xdev mounts. (6) Use the dentry passed to the statfs() op as the handle for retrieving fs statistics rather than the root dentry of the superblock (which is now a dummy). Signed-Off-By: David Howells Signed-off-by: Trond Myklebust --- include/linux/nfs_fs_sb.h | 21 +++++++++------------ 1 file changed, 9 insertions(+), 12 deletions(-) (limited to 'include/linux') diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h index d404ceca916..6d0be0efd1b 100644 --- a/include/linux/nfs_fs_sb.h +++ b/include/linux/nfs_fs_sb.h @@ -51,7 +51,6 @@ struct nfs_client { unsigned long cl_lease_time; unsigned long cl_last_renewal; struct work_struct cl_renewd; - struct work_struct cl_recoverd; struct rpc_wait_queue cl_rpcwaitq; @@ -74,6 +73,10 @@ struct nfs_client { */ struct nfs_server { struct nfs_client * nfs_client; /* shared client and NFS4 state */ + struct list_head client_link; /* List of other nfs_server structs + * that share the same client + */ + struct list_head master_link; /* link in master servers list */ struct rpc_clnt * client; /* RPC client handle */ struct rpc_clnt * client_acl; /* ACL RPC client handle */ struct nfs_iostats * io_stats; /* I/O statistics */ @@ -92,20 +95,13 @@ struct nfs_server { unsigned int acdirmin; unsigned int acdirmax; unsigned int namelen; - char * hostname; /* remote hostname */ - struct nfs_fh fh; - struct sockaddr_in addr; + struct nfs_fsid fsid; + __u64 maxfilesize; /* maximum file size */ unsigned long mount_time; /* when this fs was mounted */ + dev_t s_dev; /* superblock dev numbers */ + #ifdef CONFIG_NFS_V4 - /* Our own IP address, as a null-terminated string. - * This is used to generate the clientid, and the callback address. - */ - char ip_addr[16]; - char * mnt_path; - struct list_head nfs4_siblings; /* List of other nfs_server structs - * that share the same clientid - */ u32 attr_bitmask[2];/* V4 bitmask representing the set of attributes supported on this filesystem */ @@ -113,6 +109,7 @@ struct nfs_server { that are supported on this filesystem */ #endif + void (*destroy)(struct nfs_server *); }; /* Server capabilities */ -- cgit v1.2.3 From ec739ef03dc926d05051c8c5838971445504470a Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Tue, 22 Aug 2006 20:06:15 -0400 Subject: SUNRPC: Create a helper to tell whether a transport is bound Hide the contents and format of xprt->addr by eliminating direct uses of the xprt->addr.sin_port field. This change is required to support alternate RPC host address formats (eg IPv6). Test-plan: Destructive testing (unplugging the network temporarily). Repeated runs of Connectathon locking suite with UDP and TCP. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xprt.h | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) (limited to 'include/linux') diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index 3a0cca255b7..a71106723d7 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -269,6 +269,7 @@ int xs_setup_tcp(struct rpc_xprt *xprt, struct rpc_timeout *to); #define XPRT_CONNECTED (1) #define XPRT_CONNECTING (2) #define XPRT_CLOSE_WAIT (3) +#define XPRT_BOUND (4) static inline void xprt_set_connected(struct rpc_xprt *xprt) { @@ -312,6 +313,21 @@ static inline int xprt_test_and_set_connecting(struct rpc_xprt *xprt) return test_and_set_bit(XPRT_CONNECTING, &xprt->state); } +static inline void xprt_set_bound(struct rpc_xprt *xprt) +{ + test_and_set_bit(XPRT_BOUND, &xprt->state); +} + +static inline int xprt_bound(struct rpc_xprt *xprt) +{ + return test_bit(XPRT_BOUND, &xprt->state); +} + +static inline void xprt_clear_bound(struct rpc_xprt *xprt) +{ + clear_bit(XPRT_BOUND, &xprt->state); +} + #endif /* __KERNEL__*/ #endif /* _LINUX_SUNRPC_XPRT_H */ -- cgit v1.2.3 From 4a68179d38874c37be2802442a71b847f5d1a2a9 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Tue, 22 Aug 2006 20:06:15 -0400 Subject: SUNRPC: Make RPC portmapper use per-transport storage Move connection and bind state that was maintained in the rpc_clnt structure to the rpc_xprt structure. This will allow the creation of a clean API for plugging in different types of bind mechanisms. This brings improvements such as the elimination of a single spin lock to control serialization for all in-kernel RPC binding. A set of per-xprt bitops is used to serialize tasks during RPC binding, just like it now works for making RPC transport connections. Test-plan: Destructive testing (unplugging the network temporarily). Connectathon with UDP and TCP. NFSv2/3 and NFSv4 mounting should be carefully checked. Probably need to rig a server where certain services aren't running, or that returns an error for some typical operation. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/clnt.h | 23 +++-------------------- include/linux/sunrpc/xprt.h | 14 ++++++++++++++ 2 files changed, 17 insertions(+), 20 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/clnt.h b/include/linux/sunrpc/clnt.h index 8fe9f35eba3..00e9dbaec9c 100644 --- a/include/linux/sunrpc/clnt.h +++ b/include/linux/sunrpc/clnt.h @@ -18,18 +18,6 @@ #include #include -/* - * This defines an RPC port mapping - */ -struct rpc_portmap { - __u32 pm_prog; - __u32 pm_vers; - __u32 pm_prot; - __u16 pm_port; - unsigned char pm_binding : 1; /* doing a getport() */ - struct rpc_wait_queue pm_bindwait; /* waiting on getport() */ -}; - struct rpc_inode; /* @@ -40,7 +28,9 @@ struct rpc_clnt { atomic_t cl_users; /* number of references */ struct rpc_xprt * cl_xprt; /* transport */ struct rpc_procinfo * cl_procinfo; /* procedure info */ - u32 cl_maxproc; /* max procedure number */ + u32 cl_prog, /* RPC program number */ + cl_vers, /* RPC version number */ + cl_maxproc; /* max procedure number */ char * cl_server; /* server machine name */ char * cl_protname; /* protocol name */ @@ -55,7 +45,6 @@ struct rpc_clnt { cl_dead : 1;/* abandoned */ struct rpc_rtt * cl_rtt; /* RTO estimator data */ - struct rpc_portmap * cl_pmap; /* port mapping */ int cl_nodelen; /* nodename length */ char cl_nodename[UNX_MAXNODENAME]; @@ -64,14 +53,8 @@ struct rpc_clnt { struct dentry * cl_dentry; /* inode */ struct rpc_clnt * cl_parent; /* Points to parent of clones */ struct rpc_rtt cl_rtt_default; - struct rpc_portmap cl_pmap_default; char cl_inline_name[32]; }; -#define cl_timeout cl_xprt->timeout -#define cl_prog cl_pmap->pm_prog -#define cl_vers cl_pmap->pm_vers -#define cl_port cl_pmap->pm_port -#define cl_prot cl_pmap->pm_prot /* * General RPC program info diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index a71106723d7..4ce82616873 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -138,6 +138,7 @@ struct rpc_xprt { unsigned int tsh_size; /* size of transport specific header */ + struct rpc_wait_queue binding; /* requests waiting on rpcbind */ struct rpc_wait_queue sending; /* requests waiting to send */ struct rpc_wait_queue resend; /* requests waiting to resend */ struct rpc_wait_queue pending; /* requests in flight */ @@ -270,6 +271,7 @@ int xs_setup_tcp(struct rpc_xprt *xprt, struct rpc_timeout *to); #define XPRT_CONNECTING (2) #define XPRT_CLOSE_WAIT (3) #define XPRT_BOUND (4) +#define XPRT_BINDING (5) static inline void xprt_set_connected(struct rpc_xprt *xprt) { @@ -328,6 +330,18 @@ static inline void xprt_clear_bound(struct rpc_xprt *xprt) clear_bit(XPRT_BOUND, &xprt->state); } +static inline void xprt_clear_binding(struct rpc_xprt *xprt) +{ + smp_mb__before_clear_bit(); + clear_bit(XPRT_BINDING, &xprt->state); + smp_mb__after_clear_bit(); +} + +static inline int xprt_test_and_set_binding(struct rpc_xprt *xprt) +{ + return test_and_set_bit(XPRT_BINDING, &xprt->state); +} + #endif /* __KERNEL__*/ #endif /* _LINUX_SUNRPC_XPRT_H */ -- cgit v1.2.3 From 5b1eacbcd78930d976eb50a93f1779d311b553d1 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Tue, 22 Aug 2006 20:06:16 -0400 Subject: SUNRPC: Support for RPC child tasks no longer needed The previous patches removed the last user of RPC child tasks, so we can remove support for child tasks from net/sunrpc/sched.c now. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/sched.h | 5 ----- 1 file changed, 5 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/sched.h b/include/linux/sunrpc/sched.h index 82a91bb2236..f399c138f79 100644 --- a/include/linux/sunrpc/sched.h +++ b/include/linux/sunrpc/sched.h @@ -127,7 +127,6 @@ struct rpc_call_ops { */ #define RPC_TASK_ASYNC 0x0001 /* is an async task */ #define RPC_TASK_SWAPPER 0x0002 /* is swapping in/out */ -#define RPC_TASK_CHILD 0x0008 /* is child of other task */ #define RPC_CALL_MAJORSEEN 0x0020 /* major timeout seen */ #define RPC_TASK_ROOTCREDS 0x0040 /* force root creds */ #define RPC_TASK_DYNAMIC 0x0080 /* task was kmalloc'ed */ @@ -136,7 +135,6 @@ struct rpc_call_ops { #define RPC_TASK_NOINTR 0x0400 /* uninterruptible task */ #define RPC_IS_ASYNC(t) ((t)->tk_flags & RPC_TASK_ASYNC) -#define RPC_IS_CHILD(t) ((t)->tk_flags & RPC_TASK_CHILD) #define RPC_IS_SWAPPER(t) ((t)->tk_flags & RPC_TASK_SWAPPER) #define RPC_DO_ROOTOVERRIDE(t) ((t)->tk_flags & RPC_TASK_ROOTCREDS) #define RPC_ASSASSINATED(t) ((t)->tk_flags & RPC_TASK_KILLED) @@ -253,7 +251,6 @@ struct rpc_task *rpc_new_task(struct rpc_clnt *, int flags, const struct rpc_call_ops *ops, void *data); struct rpc_task *rpc_run_task(struct rpc_clnt *clnt, int flags, const struct rpc_call_ops *ops, void *data); -struct rpc_task *rpc_new_child(struct rpc_clnt *, struct rpc_task *parent); void rpc_init_task(struct rpc_task *task, struct rpc_clnt *clnt, int flags, const struct rpc_call_ops *ops, void *data); @@ -261,8 +258,6 @@ void rpc_release_task(struct rpc_task *); void rpc_exit_task(struct rpc_task *); void rpc_killall_tasks(struct rpc_clnt *); int rpc_execute(struct rpc_task *); -void rpc_run_child(struct rpc_task *parent, struct rpc_task *child, - rpc_action action); void rpc_init_priority_wait_queue(struct rpc_wait_queue *, const char *); void rpc_init_wait_queue(struct rpc_wait_queue *, const char *); void rpc_sleep_on(struct rpc_wait_queue *, struct rpc_task *, -- cgit v1.2.3 From bbf7c1dd2ae2b4040b41b1065ee9b1b6905b1605 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Tue, 22 Aug 2006 20:06:16 -0400 Subject: SUNRPC: Introduce transport switch callout for pluggable rpcbind Introduce a clean transport switch API for plugging in different types of rpcbind mechanisms. For instance, rpcbind can cleanly replace the existing portmapper client, or a transport can choose to implement RPC binding any way it likes. Test plan: Destructive testing (unplugging the network temporarily). Connectathon with UDP and TCP. NFSv2/3 and NFSv4 mounting should be carefully checked. Probably need to rig a server where certain services aren't running, or that returns an error for some typical operation. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/clnt.h | 2 +- include/linux/sunrpc/xprt.h | 1 + 2 files changed, 2 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/clnt.h b/include/linux/sunrpc/clnt.h index 00e9dbaec9c..2e68ac0aa02 100644 --- a/include/linux/sunrpc/clnt.h +++ b/include/linux/sunrpc/clnt.h @@ -106,7 +106,7 @@ struct rpc_clnt *rpc_clone_client(struct rpc_clnt *); int rpc_shutdown_client(struct rpc_clnt *); int rpc_destroy_client(struct rpc_clnt *); void rpc_release_client(struct rpc_clnt *); -void rpc_getport(struct rpc_task *, struct rpc_clnt *); +void rpc_getport(struct rpc_task *); int rpc_register(u32, u32, int, unsigned short, int *); void rpc_call_setup(struct rpc_task *, struct rpc_message *, int); diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index 4ce82616873..84122559fa1 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -105,6 +105,7 @@ struct rpc_xprt_ops { void (*set_buffer_size)(struct rpc_xprt *xprt, size_t sndsize, size_t rcvsize); int (*reserve_xprt)(struct rpc_task *task); void (*release_xprt)(struct rpc_xprt *xprt, struct rpc_task *task); + void (*rpcbind)(struct rpc_task *task); void (*set_port)(struct rpc_xprt *xprt, unsigned short port); void (*connect)(struct rpc_task *task); void * (*buf_alloc)(struct rpc_task *task, size_t size); -- cgit v1.2.3 From ed39440a2573abc926f230267000f21fa5a87822 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Tue, 22 Aug 2006 20:06:17 -0400 Subject: SUNRPC: create API for getting remote peer address Provide an API for retrieving the remote peer address without allowing direct access to the rpc_xprt struct. Test-plan: Compile kernel with CONFIG_NFS enabled. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/clnt.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/sunrpc/clnt.h b/include/linux/sunrpc/clnt.h index 2e68ac0aa02..65196b03f0a 100644 --- a/include/linux/sunrpc/clnt.h +++ b/include/linux/sunrpc/clnt.h @@ -123,6 +123,7 @@ void rpc_setbufsize(struct rpc_clnt *, unsigned int, unsigned int); size_t rpc_max_payload(struct rpc_clnt *); void rpc_force_rebind(struct rpc_clnt *); int rpc_ping(struct rpc_clnt *clnt, int flags); +size_t rpc_peeraddr(struct rpc_clnt *, struct sockaddr *, size_t); /* * Helper function for NFSroot support -- cgit v1.2.3 From 39d7bbcb5ba5e9d8d658b70903dd7939400e57db Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Tue, 22 Aug 2006 20:06:18 -0400 Subject: SUNRPC: remove extraneous header inclusions include/linux/sunrpc/clnt.h already includes include/linux/sunrpc/xprt.h. We can remove xprt.h from source files that already include clnt.h. Likewise include/linux/sunrpc/timer.h. Test plan: Compile kernel with CONFIG_NFS enabled. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/nfs_xdr.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index 2426b11b6cc..0f33e621892 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -1,7 +1,6 @@ #ifndef _LINUX_NFS_XDR_H #define _LINUX_NFS_XDR_H -#include #include /* -- cgit v1.2.3 From edb267a688fcee5335d596752f117a30c7152e44 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Tue, 22 Aug 2006 20:06:18 -0400 Subject: SUNRPC: add xprt switch API for printing formatted remote peer addresses Add a new method to the transport switch API to provide a way to convert the opaque contents of xprt->addr to a human-readable string. Test plan: Compile kernel with CONFIG_NFS enabled. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xprt.h | 11 +++++++++++ 1 file changed, 11 insertions(+) (limited to 'include/linux') diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index 84122559fa1..8372ab8fc9b 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -51,6 +51,14 @@ struct rpc_timeout { unsigned char to_exponential; }; +enum rpc_display_format_t { + RPC_DISPLAY_ADDR = 0, + RPC_DISPLAY_PORT, + RPC_DISPLAY_PROTO, + RPC_DISPLAY_ALL, + RPC_DISPLAY_MAX, +}; + struct rpc_task; struct rpc_xprt; struct seq_file; @@ -103,6 +111,7 @@ struct rpc_rqst { struct rpc_xprt_ops { void (*set_buffer_size)(struct rpc_xprt *xprt, size_t sndsize, size_t rcvsize); + char * (*print_addr)(struct rpc_xprt *xprt, enum rpc_display_format_t format); int (*reserve_xprt)(struct rpc_task *task); void (*release_xprt)(struct rpc_xprt *xprt, struct rpc_task *task); void (*rpcbind)(struct rpc_task *task); @@ -207,6 +216,8 @@ struct rpc_xprt { void (*old_data_ready)(struct sock *, int); void (*old_state_change)(struct sock *); void (*old_write_space)(struct sock *); + + char * address_strings[RPC_DISPLAY_MAX]; }; #define XPRT_LAST_FRAG (1 << 0) -- cgit v1.2.3 From f425eba437f0051bde979ea2eef8bc875a77cd00 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Tue, 22 Aug 2006 20:06:18 -0400 Subject: SUNRPC: Create API for displaying remote peer address Provide an API for formatting the remote peer address for printing without exposing its internal structure. The address could be dynamic, so we support a function call to get the address rather than reading it straight out of a structure. Test-plan: Destructive testing (unplugging the network temporarily). Probably need to rig a server where certain services aren't running, or that returns an error for some typical operation. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/clnt.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/sunrpc/clnt.h b/include/linux/sunrpc/clnt.h index 65196b03f0a..b7d47f01835 100644 --- a/include/linux/sunrpc/clnt.h +++ b/include/linux/sunrpc/clnt.h @@ -124,6 +124,7 @@ size_t rpc_max_payload(struct rpc_clnt *); void rpc_force_rebind(struct rpc_clnt *); int rpc_ping(struct rpc_clnt *clnt, int flags); size_t rpc_peeraddr(struct rpc_clnt *, struct sockaddr *, size_t); +char * rpc_peeraddr2str(struct rpc_clnt *, enum rpc_display_format_t); /* * Helper function for NFSroot support -- cgit v1.2.3 From c4efcb1d3e0bc76aeb9ca6301d19a5079893c6c9 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Tue, 22 Aug 2006 20:06:19 -0400 Subject: SUNRPC: Use "sockaddr_storage" for storing RPC client's remote peer address IPv6 addresses are big (128 bytes). Now that no RPC client consumers treat the addr field in rpc_xprt structs as an opaque, and access it only via the API calls, we can safely widen the field in the rpc_xprt struct to accomodate larger addresses. Test plan: Compile kernel with CONFIG_NFS enabled. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xprt.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index 8372ab8fc9b..fc05cfbd580 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -134,7 +134,8 @@ struct rpc_xprt { struct sock * inet; /* INET layer */ struct rpc_timeout timeout; /* timeout parms */ - struct sockaddr_in addr; /* server address */ + struct sockaddr_storage addr; /* server address */ + size_t addrlen; /* size of server address */ int prot; /* IP protocol */ unsigned long cong; /* current congestion */ -- cgit v1.2.3 From 6ca948238724c945bd353f51d54ae7d285f3889f Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Tue, 22 Aug 2006 20:06:19 -0400 Subject: SUNRPC: Clean-up after previous patches. Remove some unused macros related to accessing an RPC peer address Test plan: Compile kernel with CONFIG_NFS option enabled. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/nfs_fs.h | 1 - include/linux/sunrpc/clnt.h | 3 --- 2 files changed, 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 51e9bd90ded..3b5b04193fe 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -216,7 +216,6 @@ static inline struct nfs_inode *NFS_I(struct inode *inode) #define NFS_SERVER(inode) (NFS_SB(inode->i_sb)) #define NFS_CLIENT(inode) (NFS_SERVER(inode)->client) #define NFS_PROTO(inode) (NFS_SERVER(inode)->nfs_client->rpc_ops) -#define NFS_ADDR(inode) (RPC_PEERADDR(NFS_CLIENT(inode))) #define NFS_COOKIEVERF(inode) (NFS_I(inode)->cookieverf) #define NFS_READTIME(inode) (NFS_I(inode)->read_cache_jiffies) #define NFS_CHANGE_ATTR(inode) (NFS_I(inode)->change_attr) diff --git a/include/linux/sunrpc/clnt.h b/include/linux/sunrpc/clnt.h index b7d47f01835..a26d69583c7 100644 --- a/include/linux/sunrpc/clnt.h +++ b/include/linux/sunrpc/clnt.h @@ -89,9 +89,6 @@ struct rpc_procinfo { char * p_name; /* name of procedure */ }; -#define RPC_CONGESTED(clnt) (RPCXPRT_CONGESTED((clnt)->cl_xprt)) -#define RPC_PEERADDR(clnt) (&(clnt)->cl_xprt->addr) - #ifdef __KERNEL__ struct rpc_clnt *rpc_create_client(struct rpc_xprt *xprt, char *servname, -- cgit v1.2.3 From c2866763b4029411d166040306691773c12d4caf Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Tue, 22 Aug 2006 20:06:20 -0400 Subject: SUNRPC: use sockaddr + size when creating remote transport endpoints Prepare for more generic transport endpoint handling needed by transports that might use different forms of addressing, such as IPv6. Introduce a single function call to replace the two-call xprt_create_proto/rpc_create_client API. Define a new rpc_create_args structure that allows callers to pass in remote endpoint addresses of varying length. Test-plan: Compile kernel with CONFIG_NFS enabled. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/clnt.h | 22 ++++++++++++++++++++++ include/linux/sunrpc/xprt.h | 1 + 2 files changed, 23 insertions(+) (limited to 'include/linux') diff --git a/include/linux/sunrpc/clnt.h b/include/linux/sunrpc/clnt.h index a26d69583c7..7817ba82f1b 100644 --- a/include/linux/sunrpc/clnt.h +++ b/include/linux/sunrpc/clnt.h @@ -97,6 +97,28 @@ struct rpc_clnt *rpc_create_client(struct rpc_xprt *xprt, char *servname, struct rpc_clnt *rpc_new_client(struct rpc_xprt *xprt, char *servname, struct rpc_program *info, u32 version, rpc_authflavor_t authflavor); + +struct rpc_create_args { + int protocol; + struct sockaddr *address; + size_t addrsize; + struct rpc_timeout *timeout; + char *servername; + struct rpc_program *program; + u32 version; + rpc_authflavor_t authflavor; + unsigned long flags; +}; + +/* Values for "flags" field */ +#define RPC_CLNT_CREATE_HARDRTRY (1UL << 0) +#define RPC_CLNT_CREATE_INTR (1UL << 1) +#define RPC_CLNT_CREATE_AUTOBIND (1UL << 2) +#define RPC_CLNT_CREATE_ONESHOT (1UL << 3) +#define RPC_CLNT_CREATE_NONPRIVPORT (1UL << 4) +#define RPC_CLNT_CREATE_NOPING (1UL << 5) + +struct rpc_clnt *rpc_create(struct rpc_create_args *args); struct rpc_clnt *rpc_bind_new_program(struct rpc_clnt *, struct rpc_program *, int); struct rpc_clnt *rpc_clone_client(struct rpc_clnt *); diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index fc05cfbd580..bc80fcfdd89 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -237,6 +237,7 @@ void xprt_set_timeout(struct rpc_timeout *to, unsigned int retr, unsigned long /* * Generic internal transport functions */ +struct rpc_xprt * xprt_create_transport(int proto, struct sockaddr *addr, size_t size, struct rpc_timeout *toparms); void xprt_connect(struct rpc_task *task); void xprt_reserve(struct rpc_task *task); int xprt_reserve_xprt(struct rpc_task *task); -- cgit v1.2.3 From ff9aa5e56df60cc8565a93cc868fe25ae3f20e49 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Tue, 22 Aug 2006 20:06:21 -0400 Subject: SUNRPC: Eliminate xprt_create_proto and rpc_create_client The two function call API for creating a new RPC client is now obsolete. Remove it. Also, remove an unnecessary check to see whether the caller is capable of using privileged network services. The kernel RPC client always uses a privileged ephemeral port by default; callers are responsible for checking the authority of users to make use of any RPC service, or for specifying that a nonprivileged port is acceptable. Test plan: Repeated runs of Connectathon locking suite. Check network trace to ensure correctness of NLM requests and replies. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/clnt.h | 7 ------- include/linux/sunrpc/xprt.h | 1 - 2 files changed, 8 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/clnt.h b/include/linux/sunrpc/clnt.h index 7817ba82f1b..f6d1d646ce0 100644 --- a/include/linux/sunrpc/clnt.h +++ b/include/linux/sunrpc/clnt.h @@ -91,13 +91,6 @@ struct rpc_procinfo { #ifdef __KERNEL__ -struct rpc_clnt *rpc_create_client(struct rpc_xprt *xprt, char *servname, - struct rpc_program *info, - u32 version, rpc_authflavor_t authflavor); -struct rpc_clnt *rpc_new_client(struct rpc_xprt *xprt, char *servname, - struct rpc_program *info, - u32 version, rpc_authflavor_t authflavor); - struct rpc_create_args { int protocol; struct sockaddr *address; diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index bc80fcfdd89..de4efea7c85 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -231,7 +231,6 @@ struct rpc_xprt { /* * Transport operations used by ULPs */ -struct rpc_xprt * xprt_create_proto(int proto, struct sockaddr_in *addr, struct rpc_timeout *to); void xprt_set_timeout(struct rpc_timeout *to, unsigned int retr, unsigned long incr); /* -- cgit v1.2.3 From 4f390c152bc87165da4b1f5b7d870b46fb106d4e Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Tue, 22 Aug 2006 20:06:22 -0400 Subject: NFS: Fix double d_drop in nfs_instantiate() error path If the LOOKUP or GETATTR in nfs_instantiate fail, nfs_instantiate will do a d_drop before returning. But some callers already do a d_drop in the case of an error return. Make certain we do only one d_drop in all error paths. This issue was introduced because over time, the symlink proc API diverged slightly from the create/mkdir/mknod proc API. To prevent other coding mistakes of this type, change the symlink proc API to be more like create/mkdir/mknod and move the nfs_instantiate call into the symlink proc routines so it is used in exactly the same way for create, mkdir, mknod, and symlink. Test plan: Connectathon, all versions of NFS. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/nfs_xdr.h | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index 0f33e621892..ddf5d75e97a 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -793,9 +793,8 @@ struct nfs_rpc_ops { int (*rename) (struct inode *, struct qstr *, struct inode *, struct qstr *); int (*link) (struct inode *, struct inode *, struct qstr *); - int (*symlink) (struct inode *, struct qstr *, struct qstr *, - struct iattr *, struct nfs_fh *, - struct nfs_fattr *); + int (*symlink) (struct inode *, struct dentry *, struct qstr *, + struct iattr *); int (*mkdir) (struct inode *, struct dentry *, struct iattr *); int (*rmdir) (struct inode *, struct qstr *); int (*readdir) (struct dentry *, struct rpc_cred *, -- cgit v1.2.3 From 94a6d75320b3681e6e728b70e18bd186cb55e682 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Tue, 22 Aug 2006 20:06:23 -0400 Subject: NFS: Use cached page as buffer for NFS symlink requests Now that we have a copy of the symlink path in the page cache, we can pass a struct page down to the XDR routines instead of a string buffer. Test plan: Connectathon, all NFS versions. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/nfs_xdr.h | 17 ++++++++++------- 1 file changed, 10 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index ddf5d75e97a..dc5397d9d23 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -358,8 +358,8 @@ struct nfs_symlinkargs { struct nfs_fh * fromfh; const char * fromname; unsigned int fromlen; - const char * topath; - unsigned int tolen; + struct page ** pages; + unsigned int pathlen; struct iattr * sattr; }; @@ -434,8 +434,8 @@ struct nfs3_symlinkargs { struct nfs_fh * fromfh; const char * fromname; unsigned int fromlen; - const char * topath; - unsigned int tolen; + struct page ** pages; + unsigned int pathlen; struct iattr * sattr; }; @@ -533,7 +533,10 @@ struct nfs4_accessres { struct nfs4_create_arg { u32 ftype; union { - struct qstr * symlink; /* NF4LNK */ + struct { + struct page ** pages; + unsigned int len; + } symlink; /* NF4LNK */ struct { u32 specdata1; u32 specdata2; @@ -793,8 +796,8 @@ struct nfs_rpc_ops { int (*rename) (struct inode *, struct qstr *, struct inode *, struct qstr *); int (*link) (struct inode *, struct inode *, struct qstr *); - int (*symlink) (struct inode *, struct dentry *, struct qstr *, - struct iattr *); + int (*symlink) (struct inode *, struct dentry *, struct page *, + unsigned int, struct iattr *); int (*mkdir) (struct inode *, struct dentry *, struct iattr *); int (*rmdir) (struct inode *, struct qstr *); int (*readdir) (struct dentry *, struct rpc_cred *, -- cgit v1.2.3 From 275a082fe9308e710324e26ccb5363c53d8fd45f Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Tue, 22 Aug 2006 20:06:24 -0400 Subject: Add a real API for dealing with blk_congestion_wait() Signed-off-by: Trond Myklebust --- include/linux/blkdev.h | 1 + include/linux/writeback.h | 1 + 2 files changed, 2 insertions(+) (limited to 'include/linux') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index aafe82788b4..96c9040c00a 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -746,6 +746,7 @@ extern void blk_queue_free_tags(request_queue_t *); extern int blk_queue_resize_tags(request_queue_t *, int); extern void blk_queue_invalidate_tags(request_queue_t *); extern long blk_congestion_wait(int rw, long timeout); +extern void blk_congestion_end(int rw); extern void blk_rq_bio_prep(request_queue_t *, struct request *, struct bio *); extern int blkdev_issue_flush(struct block_device *, sector_t *); diff --git a/include/linux/writeback.h b/include/linux/writeback.h index 9e38b566d0e..0422036af4e 100644 --- a/include/linux/writeback.h +++ b/include/linux/writeback.h @@ -85,6 +85,7 @@ int wakeup_pdflush(long nr_pages); void laptop_io_completion(void); void laptop_sync_completion(void); void throttle_vm_writeout(void); +void writeback_congestion_end(void); /* These are exported to sysctl. */ extern int dirty_background_ratio; -- cgit v1.2.3 From 5dd3177ae5012c1e2ad7a9ffdbd0e0d0de2f60e4 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Thu, 24 Aug 2006 01:03:05 -0400 Subject: NFSv4: Fix a use-after-free issue with the nfs server. Signed-off-by: Trond Myklebust --- include/linux/nfs_fs_sb.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h index 6d0be0efd1b..7ccfc7ef0a8 100644 --- a/include/linux/nfs_fs_sb.h +++ b/include/linux/nfs_fs_sb.h @@ -19,6 +19,7 @@ struct nfs_client { #define NFS_CS_RPCIOD 0 /* - rpciod started */ #define NFS_CS_CALLBACK 1 /* - callback started */ #define NFS_CS_IDMAP 2 /* - idmap started */ +#define NFS_CS_RENEWD 3 /* - renewd started */ struct sockaddr_in cl_addr; /* server identifier */ char * cl_hostname; /* hostname of server */ struct list_head cl_share_link; /* link in global client list */ -- cgit v1.2.3 From 158998b6fe36f6acef087f574c96d44713499cc9 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Thu, 24 Aug 2006 01:03:17 -0400 Subject: SUNRPC: Make rpc_mkpipe() take the parent dentry as an argument Signed-off-by: Trond Myklebust --- include/linux/sunrpc/rpc_pipe_fs.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/rpc_pipe_fs.h b/include/linux/sunrpc/rpc_pipe_fs.h index a481472c948..a2eb9b4a9de 100644 --- a/include/linux/sunrpc/rpc_pipe_fs.h +++ b/include/linux/sunrpc/rpc_pipe_fs.h @@ -43,7 +43,7 @@ extern int rpc_queue_upcall(struct inode *, struct rpc_pipe_msg *); extern struct dentry *rpc_mkdir(char *, struct rpc_clnt *); extern int rpc_rmdir(struct dentry *); -extern struct dentry *rpc_mkpipe(char *, void *, struct rpc_pipe_ops *, int flags); +extern struct dentry *rpc_mkpipe(struct dentry *, const char *, void *, struct rpc_pipe_ops *, int flags); extern int rpc_unlink(struct dentry *); extern struct vfsmount *rpc_get_mount(void); extern void rpc_put_mount(void); -- cgit v1.2.3 From 6b6ca86b77b62b798cf9ca2599036420abce7796 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Tue, 5 Sep 2006 12:55:57 -0400 Subject: SUNRPC: Add refcounting to the struct rpc_xprt In a subsequent patch, this will allow the portmapper to take a reference to the rpc_xprt for which it is updating the port number, fixing an Oops. Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xprt.h | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index de4efea7c85..bdeba8538c7 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -12,6 +12,7 @@ #include #include #include +#include #include #include @@ -129,6 +130,7 @@ struct rpc_xprt_ops { }; struct rpc_xprt { + struct kref kref; /* Reference count */ struct rpc_xprt_ops * ops; /* transport methods */ struct socket * sock; /* BSD socket layer */ struct sock * inet; /* INET layer */ @@ -248,7 +250,8 @@ int xprt_adjust_timeout(struct rpc_rqst *req); void xprt_release_xprt(struct rpc_xprt *xprt, struct rpc_task *task); void xprt_release_xprt_cong(struct rpc_xprt *xprt, struct rpc_task *task); void xprt_release(struct rpc_task *task); -int xprt_destroy(struct rpc_xprt *xprt); +struct rpc_xprt * xprt_get(struct rpc_xprt *xprt); +void xprt_put(struct rpc_xprt *xprt); static inline u32 *xprt_skip_transport_header(struct rpc_xprt *xprt, u32 *p) { -- cgit v1.2.3