From fe315e76fc3a3f9f7e1581dc22fec7e7719f0896 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Wed, 11 Mar 2009 14:10:21 -0400 Subject: [PATCH 01/58] SUNRPC: Avoid spurious wake-up during UDP connect processing To clear out old state, the UDP connect workers unconditionally invoke xs_close() before proceeding with a new connect. Nowadays this causes a spurious wake-up of the task waiting for the connect to complete. This is a little racey, but usually harmless. The waiting task immediately retries the connect via a call_bind/call_connect sequence, which usually finds the transport already in the connected state because the connect worker has finished in the background. To avoid a spurious wake-up, factor the xs_close() logic that resets the underlying socket into a helper, and have the UDP connect workers call that helper instead of xs_close(). Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- net/sunrpc/xprtsock.c | 44 +++++++++++++++++++++++++------------------ 1 file changed, 26 insertions(+), 18 deletions(-) diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index 29c71e645b27..1127eb934136 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -767,23 +767,13 @@ static void xs_restore_old_callbacks(struct sock_xprt *transport, struct sock *s sk->sk_error_report = transport->old_error_report; } -/** - * xs_close - close a socket - * @xprt: transport - * - * This is used when all requests are complete; ie, no DRC state remains - * on the server we want to save. - */ -static void xs_close(struct rpc_xprt *xprt) +static void xs_reset_transport(struct sock_xprt *transport) { - struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt); struct socket *sock = transport->sock; struct sock *sk = transport->inet; - if (!sk) - goto clear_close_wait; - - dprintk("RPC: xs_close xprt %p\n", xprt); + if (sk == NULL) + return; write_lock_bh(&sk->sk_callback_lock); transport->inet = NULL; @@ -797,7 +787,23 @@ static void xs_close(struct rpc_xprt *xprt) sk->sk_no_check = 0; sock_release(sock); -clear_close_wait: +} + +/** + * xs_close - close a socket + * @xprt: transport + * + * This is used when all requests are complete; ie, no DRC state remains + * on the server we want to save. + */ +static void xs_close(struct rpc_xprt *xprt) +{ + struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt); + + dprintk("RPC: xs_close xprt %p\n", xprt); + + xs_reset_transport(transport); + smp_mb__before_clear_bit(); clear_bit(XPRT_CLOSE_WAIT, &xprt->state); clear_bit(XPRT_CLOSING, &xprt->state); @@ -1537,9 +1543,10 @@ static void xs_udp_connect_worker4(struct work_struct *work) goto out; /* Start by resetting any existing state */ - xs_close(xprt); + xs_reset_transport(transport); - if ((err = sock_create_kern(PF_INET, SOCK_DGRAM, IPPROTO_UDP, &sock)) < 0) { + err = sock_create_kern(PF_INET, SOCK_DGRAM, IPPROTO_UDP, &sock); + if (err < 0) { dprintk("RPC: can't create UDP transport socket (%d).\n", -err); goto out; } @@ -1578,9 +1585,10 @@ static void xs_udp_connect_worker6(struct work_struct *work) goto out; /* Start by resetting any existing state */ - xs_close(xprt); + xs_reset_transport(transport); - if ((err = sock_create_kern(PF_INET6, SOCK_DGRAM, IPPROTO_UDP, &sock)) < 0) { + err = sock_create_kern(PF_INET6, SOCK_DGRAM, IPPROTO_UDP, &sock); + if (err < 0) { dprintk("RPC: can't create UDP transport socket (%d).\n", -err); goto out; } From 2b57dc6cf9bf31edc0df430ea18dd1dbd3028975 Mon Sep 17 00:00:00 2001 From: Suresh Jayaraman Date: Wed, 11 Mar 2009 14:10:22 -0400 Subject: [PATCH 02/58] NFS: Minor __nfs_revalidate_inode cleanup Remove redundant NFS_STALE() check, a leftover due to the commit 691beb13cdc88358334ef0ba867c080a247a760f Signed-off-by: Suresh Jayaraman Signed-off-by: Trond Myklebust --- fs/nfs/inode.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index 0c381686171e..acaaa7c7efa4 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -670,9 +670,6 @@ __nfs_revalidate_inode(struct nfs_server *server, struct inode *inode) if (NFS_STALE(inode)) goto out; - if (NFS_STALE(inode)) - goto out; - nfs_inc_stats(inode, NFSIOS_INODEREVALIDATE); status = NFS_PROTO(inode)->getattr(server, NFS_FH(inode), &fattr); if (status != 0) { From 37d9d76d8b3a2ac5817e1fa3263cfe0fdb439e51 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Wed, 11 Mar 2009 14:10:23 -0400 Subject: [PATCH 03/58] NFS: flush cached directory information slightly more readily. If cached directory contents becomes incorrect, there is no way to flush the contents. This contrasts with files where file locking is the recommended way to ensure cache consistency between multiple applications (a read-lock always flushes the cache). Also while changes to files often change the size of the file (thus triggering a cache flush), changes to directories often do not change the apparent size (as the size is often rounded to a block size). So it is particularly important with directories to avoid the possibility of an incorrect cache wherever possible. When the link count on a directory changes it implies a change in the number of child directories, and so a change in the contents of this directory. So use that as a trigger to flush cached contents. When the ctime changes but the mtime does not, there are two possible reasons. 1/ The owner/mode information has been changed. 2/ utimes has been used to set the mtime backwards. In the first case, a data-cache flush is not required. In the second case it is. So on the basis that correctness trumps performance, flush the directory contents cache in this case also. Signed-off-by: NeilBrown Signed-off-by: Trond Myklebust --- fs/nfs/inode.c | 15 +++++++++++++-- 1 file changed, 13 insertions(+), 2 deletions(-) diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index acaaa7c7efa4..268ce3a46220 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -1113,8 +1113,16 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) nfs_force_lookup_revalidate(inode); } /* If ctime has changed we should definitely clear access+acl caches */ - if (!timespec_equal(&inode->i_ctime, &fattr->ctime)) + if (!timespec_equal(&inode->i_ctime, &fattr->ctime)) { invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL; + /* and probably clear data for a directory too as utimes can cause + * havoc with our cache. + */ + if (S_ISDIR(inode->i_mode)) { + invalid |= NFS_INO_INVALID_DATA; + nfs_force_lookup_revalidate(inode); + } + } } else if (nfsi->change_attr != fattr->change_attr) { dprintk("NFS: change_attr change on server for file %s/%ld\n", inode->i_sb->s_id, inode->i_ino); @@ -1148,8 +1156,11 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) inode->i_gid != fattr->gid) invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL; - if (inode->i_nlink != fattr->nlink) + if (inode->i_nlink != fattr->nlink) { invalid |= NFS_INO_INVALID_ATTR; + if (S_ISDIR(inode->i_mode)) + invalid |= NFS_INO_INVALID_DATA; + } inode->i_mode = fattr->mode; inode->i_nlink = fattr->nlink; From 78f945f88ef83dcc7c962614a080e0a9a2db5889 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 11 Mar 2009 14:10:23 -0400 Subject: [PATCH 04/58] NFSv4: Ignore errors on the post-op attributes in SETATTR calls There is no need to fail or retry a SETATTR call just because the post-op GETATTR failed. Signed-off-by: Trond Myklebust --- fs/nfs/nfs4xdr.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index d1e4c8f8a0a9..5f0ee3e2bd84 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -4078,9 +4078,7 @@ static int nfs4_xdr_dec_setattr(struct rpc_rqst *rqstp, __be32 *p, struct nfs_se status = decode_setattr(&xdr, res); if (status) goto out; - status = decode_getfattr(&xdr, res->fattr, res->server); - if (status == NFS4ERR_DELAY) - status = 0; + decode_getfattr(&xdr, res->fattr, res->server); out: return status; } From 9e6e70f8d8b6698e0017c56b86525aabe9c7cd4c Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 11 Mar 2009 14:10:24 -0400 Subject: [PATCH 05/58] NFSv4: Support NFSv4 optional attributes in the struct nfs_fattr Currently, filling struct nfs_fattr is more or less an all or nothing operation, since NFSv2 and NFSv3 have only mandatory attributes. In NFSv4, some attributes are optional, and so we may simply not be able to fill in those fields. Furthermore, NFSv4 allows you to specify which attributes you are interested in retrieving, thus permitting you to optimise away retrieval of attributes that you know will no change... Signed-off-by: Trond Myklebust --- fs/nfs/inode.c | 243 +++++++++++++++++++++++++--------------- fs/nfs/nfs2xdr.c | 2 +- fs/nfs/nfs3xdr.c | 6 +- fs/nfs/nfs4xdr.c | 6 +- include/linux/nfs_xdr.h | 48 ++++++-- 5 files changed, 202 insertions(+), 103 deletions(-) diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index 268ce3a46220..b7656bd3706f 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -249,13 +249,10 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr) struct inode *inode = ERR_PTR(-ENOENT); unsigned long hash; - if ((fattr->valid & NFS_ATTR_FATTR) == 0) + if ((fattr->valid & NFS_ATTR_FATTR_FILEID) == 0) goto out_no_inode; - - if (!fattr->nlink) { - printk("NFS: Buggy server - nlink == 0!\n"); + if ((fattr->valid & NFS_ATTR_FATTR_TYPE) == 0) goto out_no_inode; - } hash = nfs_fattr_to_ino_t(fattr); @@ -291,7 +288,8 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr) && fattr->size <= NFS_LIMIT_READDIRPLUS) set_bit(NFS_INO_ADVISE_RDPLUS, &NFS_I(inode)->flags); /* Deal with crossing mountpoints */ - if (!nfs_fsid_equal(&NFS_SB(sb)->fsid, &fattr->fsid)) { + if ((fattr->valid & NFS_ATTR_FATTR_FSID) + && !nfs_fsid_equal(&NFS_SB(sb)->fsid, &fattr->fsid)) { if (fattr->valid & NFS_ATTR_FATTR_V4_REFERRAL) inode->i_op = &nfs_referral_inode_operations; else @@ -304,28 +302,45 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr) else init_special_inode(inode, inode->i_mode, fattr->rdev); + memset(&inode->i_atime, 0, sizeof(inode->i_atime)); + memset(&inode->i_mtime, 0, sizeof(inode->i_mtime)); + memset(&inode->i_ctime, 0, sizeof(inode->i_ctime)); + nfsi->change_attr = 0; + inode->i_size = 0; + inode->i_nlink = 0; + inode->i_uid = -2; + inode->i_gid = -2; + inode->i_blocks = 0; + memset(nfsi->cookieverf, 0, sizeof(nfsi->cookieverf)); + nfsi->read_cache_jiffies = fattr->time_start; nfsi->attr_gencount = fattr->gencount; - inode->i_atime = fattr->atime; - inode->i_mtime = fattr->mtime; - inode->i_ctime = fattr->ctime; - if (fattr->valid & NFS_ATTR_FATTR_V4) + if (fattr->valid & NFS_ATTR_FATTR_ATIME) + inode->i_atime = fattr->atime; + if (fattr->valid & NFS_ATTR_FATTR_MTIME) + inode->i_mtime = fattr->mtime; + if (fattr->valid & NFS_ATTR_FATTR_CTIME) + inode->i_ctime = fattr->ctime; + if (fattr->valid & NFS_ATTR_FATTR_CHANGE) nfsi->change_attr = fattr->change_attr; - inode->i_size = nfs_size_to_loff_t(fattr->size); - inode->i_nlink = fattr->nlink; - inode->i_uid = fattr->uid; - inode->i_gid = fattr->gid; - if (fattr->valid & (NFS_ATTR_FATTR_V3 | NFS_ATTR_FATTR_V4)) { + if (fattr->valid & NFS_ATTR_FATTR_SIZE) + inode->i_size = nfs_size_to_loff_t(fattr->size); + if (fattr->valid & NFS_ATTR_FATTR_NLINK) + inode->i_nlink = fattr->nlink; + if (fattr->valid & NFS_ATTR_FATTR_OWNER) + inode->i_uid = fattr->uid; + if (fattr->valid & NFS_ATTR_FATTR_GROUP) + inode->i_gid = fattr->gid; + if (fattr->valid & NFS_ATTR_FATTR_BLOCKS_USED) + inode->i_blocks = fattr->du.nfs2.blocks; + if (fattr->valid & NFS_ATTR_FATTR_SPACE_USED) { /* * report the blocks in 512byte units */ inode->i_blocks = nfs_calc_block_size(fattr->du.nfs3.used); - } else { - inode->i_blocks = fattr->du.nfs2.blocks; } nfsi->attrtimeo = NFS_MINATTRTIMEO(inode); nfsi->attrtimeo_timestamp = now; - memset(nfsi->cookieverf, 0, sizeof(nfsi->cookieverf)); nfsi->access_cache = RB_ROOT; unlock_new_inode(inode); @@ -812,25 +827,31 @@ static void nfs_wcc_update_inode(struct inode *inode, struct nfs_fattr *fattr) { struct nfs_inode *nfsi = NFS_I(inode); - if ((fattr->valid & NFS_ATTR_WCC_V4) != 0 && - nfsi->change_attr == fattr->pre_change_attr) { + if ((fattr->valid & NFS_ATTR_FATTR_PRECHANGE) + && (fattr->valid & NFS_ATTR_FATTR_CHANGE) + && nfsi->change_attr == fattr->pre_change_attr) { nfsi->change_attr = fattr->change_attr; if (S_ISDIR(inode->i_mode)) nfsi->cache_validity |= NFS_INO_INVALID_DATA; } /* If we have atomic WCC data, we may update some attributes */ - if ((fattr->valid & NFS_ATTR_WCC) != 0) { - if (timespec_equal(&inode->i_ctime, &fattr->pre_ctime)) + if ((fattr->valid & NFS_ATTR_FATTR_PRECTIME) + && (fattr->valid & NFS_ATTR_FATTR_CTIME) + && timespec_equal(&inode->i_ctime, &fattr->pre_ctime)) memcpy(&inode->i_ctime, &fattr->ctime, sizeof(inode->i_ctime)); - if (timespec_equal(&inode->i_mtime, &fattr->pre_mtime)) { + + if ((fattr->valid & NFS_ATTR_FATTR_PREMTIME) + && (fattr->valid & NFS_ATTR_FATTR_MTIME) + && timespec_equal(&inode->i_mtime, &fattr->pre_mtime)) { memcpy(&inode->i_mtime, &fattr->mtime, sizeof(inode->i_mtime)); if (S_ISDIR(inode->i_mode)) nfsi->cache_validity |= NFS_INO_INVALID_DATA; - } - if (i_size_read(inode) == nfs_size_to_loff_t(fattr->pre_size) && - nfsi->npages == 0) - i_size_write(inode, nfs_size_to_loff_t(fattr->size)); } + if ((fattr->valid & NFS_ATTR_FATTR_PRESIZE) + && (fattr->valid & NFS_ATTR_FATTR_SIZE) + && i_size_read(inode) == nfs_size_to_loff_t(fattr->pre_size) + && nfsi->npages == 0) + i_size_write(inode, nfs_size_to_loff_t(fattr->size)); } /** @@ -850,35 +871,39 @@ static int nfs_check_inode_attributes(struct inode *inode, struct nfs_fattr *fat /* Has the inode gone and changed behind our back? */ - if (nfsi->fileid != fattr->fileid - || (inode->i_mode & S_IFMT) != (fattr->mode & S_IFMT)) { + if ((fattr->valid & NFS_ATTR_FATTR_FILEID) && nfsi->fileid != fattr->fileid) + return -EIO; + if ((fattr->valid & NFS_ATTR_FATTR_TYPE) && (inode->i_mode & S_IFMT) != (fattr->mode & S_IFMT)) return -EIO; - } - if ((fattr->valid & NFS_ATTR_FATTR_V4) != 0 && + if ((fattr->valid & NFS_ATTR_FATTR_CHANGE) != 0 && nfsi->change_attr != fattr->change_attr) invalid |= NFS_INO_INVALID_ATTR|NFS_INO_REVAL_PAGECACHE; /* Verify a few of the more important attributes */ - if (!timespec_equal(&inode->i_mtime, &fattr->mtime)) + if ((fattr->valid & NFS_ATTR_FATTR_MTIME) && !timespec_equal(&inode->i_mtime, &fattr->mtime)) invalid |= NFS_INO_INVALID_ATTR|NFS_INO_REVAL_PAGECACHE; - cur_size = i_size_read(inode); - new_isize = nfs_size_to_loff_t(fattr->size); - if (cur_size != new_isize && nfsi->npages == 0) - invalid |= NFS_INO_INVALID_ATTR|NFS_INO_REVAL_PAGECACHE; + if (fattr->valid & NFS_ATTR_FATTR_SIZE) { + cur_size = i_size_read(inode); + new_isize = nfs_size_to_loff_t(fattr->size); + if (cur_size != new_isize && nfsi->npages == 0) + invalid |= NFS_INO_INVALID_ATTR|NFS_INO_REVAL_PAGECACHE; + } /* Have any file permissions changed? */ - if ((inode->i_mode & S_IALLUGO) != (fattr->mode & S_IALLUGO) - || inode->i_uid != fattr->uid - || inode->i_gid != fattr->gid) + if ((fattr->valid & NFS_ATTR_FATTR_MODE) && (inode->i_mode & S_IALLUGO) != (fattr->mode & S_IALLUGO)) + invalid |= NFS_INO_INVALID_ATTR | NFS_INO_INVALID_ACCESS | NFS_INO_INVALID_ACL; + if ((fattr->valid & NFS_ATTR_FATTR_OWNER) && inode->i_uid != fattr->uid) + invalid |= NFS_INO_INVALID_ATTR | NFS_INO_INVALID_ACCESS | NFS_INO_INVALID_ACL; + if ((fattr->valid & NFS_ATTR_FATTR_GROUP) && inode->i_gid != fattr->gid) invalid |= NFS_INO_INVALID_ATTR | NFS_INO_INVALID_ACCESS | NFS_INO_INVALID_ACL; /* Has the link count changed? */ - if (inode->i_nlink != fattr->nlink) + if ((fattr->valid & NFS_ATTR_FATTR_NLINK) && inode->i_nlink != fattr->nlink) invalid |= NFS_INO_INVALID_ATTR; - if (!timespec_equal(&inode->i_atime, &fattr->atime)) + if ((fattr->valid & NFS_ATTR_FATTR_ATIME) && !timespec_equal(&inode->i_atime, &fattr->atime)) invalid |= NFS_INO_INVALID_ATIME; if (invalid != 0) @@ -890,11 +915,15 @@ static int nfs_check_inode_attributes(struct inode *inode, struct nfs_fattr *fat static int nfs_ctime_need_update(const struct inode *inode, const struct nfs_fattr *fattr) { + if (!(fattr->valid & NFS_ATTR_FATTR_CTIME)) + return 0; return timespec_compare(&fattr->ctime, &inode->i_ctime) > 0; } static int nfs_size_need_update(const struct inode *inode, const struct nfs_fattr *fattr) { + if (!(fattr->valid & NFS_ATTR_FATTR_SIZE)) + return 0; return nfs_size_to_loff_t(fattr->size) > i_size_read(inode); } @@ -1030,20 +1059,31 @@ int nfs_post_op_update_inode_force_wcc(struct inode *inode, struct nfs_fattr *fa /* Don't do a WCC update if these attributes are already stale */ if ((fattr->valid & NFS_ATTR_FATTR) == 0 || !nfs_inode_attrs_need_update(inode, fattr)) { - fattr->valid &= ~(NFS_ATTR_WCC_V4|NFS_ATTR_WCC); + fattr->valid &= ~(NFS_ATTR_FATTR_PRECHANGE + | NFS_ATTR_FATTR_PRESIZE + | NFS_ATTR_FATTR_PREMTIME + | NFS_ATTR_FATTR_PRECTIME); goto out_noforce; } - if ((fattr->valid & NFS_ATTR_FATTR_V4) != 0 && - (fattr->valid & NFS_ATTR_WCC_V4) == 0) { + if ((fattr->valid & NFS_ATTR_FATTR_CHANGE) != 0 && + (fattr->valid & NFS_ATTR_FATTR_PRECHANGE) == 0) { fattr->pre_change_attr = NFS_I(inode)->change_attr; - fattr->valid |= NFS_ATTR_WCC_V4; + fattr->valid |= NFS_ATTR_FATTR_PRECHANGE; } - if ((fattr->valid & NFS_ATTR_FATTR) != 0 && - (fattr->valid & NFS_ATTR_WCC) == 0) { + if ((fattr->valid & NFS_ATTR_FATTR_CTIME) != 0 && + (fattr->valid & NFS_ATTR_FATTR_PRECTIME) == 0) { memcpy(&fattr->pre_ctime, &inode->i_ctime, sizeof(fattr->pre_ctime)); + fattr->valid |= NFS_ATTR_FATTR_PRECTIME; + } + if ((fattr->valid & NFS_ATTR_FATTR_MTIME) != 0 && + (fattr->valid & NFS_ATTR_FATTR_PREMTIME) == 0) { memcpy(&fattr->pre_mtime, &inode->i_mtime, sizeof(fattr->pre_mtime)); + fattr->valid |= NFS_ATTR_FATTR_PREMTIME; + } + if ((fattr->valid & NFS_ATTR_FATTR_SIZE) != 0 && + (fattr->valid & NFS_ATTR_FATTR_PRESIZE) == 0) { fattr->pre_size = i_size_read(inode); - fattr->valid |= NFS_ATTR_WCC; + fattr->valid |= NFS_ATTR_FATTR_PRESIZE; } out_noforce: status = nfs_post_op_update_inode_locked(inode, fattr); @@ -1075,18 +1115,18 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) __func__, inode->i_sb->s_id, inode->i_ino, atomic_read(&inode->i_count), fattr->valid); - if (nfsi->fileid != fattr->fileid) + if ((fattr->valid & NFS_ATTR_FATTR_FILEID) && nfsi->fileid != fattr->fileid) goto out_fileid; /* * Make sure the inode's type hasn't changed. */ - if ((inode->i_mode & S_IFMT) != (fattr->mode & S_IFMT)) + if ((fattr->valid & NFS_ATTR_FATTR_TYPE) && (inode->i_mode & S_IFMT) != (fattr->mode & S_IFMT)) goto out_changed; server = NFS_SERVER(inode); /* Update the fsid? */ - if (S_ISDIR(inode->i_mode) && + if (S_ISDIR(inode->i_mode) && (fattr->valid & NFS_ATTR_FATTR_FSID) && !nfs_fsid_equal(&server->fsid, &fattr->fsid) && !test_bit(NFS_INO_MOUNTPOINT, &nfsi->flags)) server->fsid = fattr->fsid; @@ -1096,14 +1136,27 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) */ nfsi->read_cache_jiffies = fattr->time_start; - nfsi->cache_validity &= ~(NFS_INO_INVALID_ATTR | NFS_INO_INVALID_ATIME - | NFS_INO_REVAL_PAGECACHE); + if ((fattr->valid & NFS_ATTR_FATTR_CHANGE) || (fattr->valid & (NFS_ATTR_FATTR_MTIME|NFS_ATTR_FATTR_CTIME))) + nfsi->cache_validity &= ~(NFS_INO_INVALID_ATTR + | NFS_INO_INVALID_ATIME + | NFS_INO_REVAL_PAGECACHE); /* Do atomic weak cache consistency updates */ nfs_wcc_update_inode(inode, fattr); /* More cache consistency checks */ - if (!(fattr->valid & NFS_ATTR_FATTR_V4)) { + if (fattr->valid & NFS_ATTR_FATTR_CHANGE) { + if (nfsi->change_attr != fattr->change_attr) { + dprintk("NFS: change_attr change on server for file %s/%ld\n", + inode->i_sb->s_id, inode->i_ino); + invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL; + if (S_ISDIR(inode->i_mode)) + nfs_force_lookup_revalidate(inode); + nfsi->change_attr = fattr->change_attr; + } + } + + if (fattr->valid & NFS_ATTR_FATTR_MTIME) { /* NFSv2/v3: Check if the mtime agrees */ if (!timespec_equal(&inode->i_mtime, &fattr->mtime)) { dprintk("NFS: mtime change on server for file %s/%ld\n", @@ -1111,7 +1164,10 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA; if (S_ISDIR(inode->i_mode)) nfs_force_lookup_revalidate(inode); + memcpy(&inode->i_mtime, &fattr->mtime, sizeof(inode->i_mtime)); } + } + if (fattr->valid & NFS_ATTR_FATTR_CTIME) { /* If ctime has changed we should definitely clear access+acl caches */ if (!timespec_equal(&inode->i_ctime, &fattr->ctime)) { invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL; @@ -1122,59 +1178,66 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) invalid |= NFS_INO_INVALID_DATA; nfs_force_lookup_revalidate(inode); } + memcpy(&inode->i_ctime, &fattr->ctime, sizeof(inode->i_ctime)); } - } else if (nfsi->change_attr != fattr->change_attr) { - dprintk("NFS: change_attr change on server for file %s/%ld\n", - inode->i_sb->s_id, inode->i_ino); - invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL; - if (S_ISDIR(inode->i_mode)) - nfs_force_lookup_revalidate(inode); } /* Check if our cached file size is stale */ - new_isize = nfs_size_to_loff_t(fattr->size); - cur_isize = i_size_read(inode); - if (new_isize != cur_isize) { - /* Do we perhaps have any outstanding writes, or has - * the file grown beyond our last write? */ - if (nfsi->npages == 0 || new_isize > cur_isize) { - i_size_write(inode, new_isize); - invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA; + if (fattr->valid & NFS_ATTR_FATTR_SIZE) { + new_isize = nfs_size_to_loff_t(fattr->size); + cur_isize = i_size_read(inode); + if (new_isize != cur_isize) { + /* Do we perhaps have any outstanding writes, or has + * the file grown beyond our last write? */ + if (nfsi->npages == 0 || new_isize > cur_isize) { + i_size_write(inode, new_isize); + invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA; + } + dprintk("NFS: isize change on server for file %s/%ld\n", + inode->i_sb->s_id, inode->i_ino); } - dprintk("NFS: isize change on server for file %s/%ld\n", - inode->i_sb->s_id, inode->i_ino); } - memcpy(&inode->i_mtime, &fattr->mtime, sizeof(inode->i_mtime)); - memcpy(&inode->i_ctime, &fattr->ctime, sizeof(inode->i_ctime)); - memcpy(&inode->i_atime, &fattr->atime, sizeof(inode->i_atime)); - nfsi->change_attr = fattr->change_attr; + if (fattr->valid & NFS_ATTR_FATTR_ATIME) + memcpy(&inode->i_atime, &fattr->atime, sizeof(inode->i_atime)); - if ((inode->i_mode & S_IALLUGO) != (fattr->mode & S_IALLUGO) || - inode->i_uid != fattr->uid || - inode->i_gid != fattr->gid) - invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL; - - if (inode->i_nlink != fattr->nlink) { - invalid |= NFS_INO_INVALID_ATTR; - if (S_ISDIR(inode->i_mode)) - invalid |= NFS_INO_INVALID_DATA; + if (fattr->valid & NFS_ATTR_FATTR_MODE) { + if ((inode->i_mode & S_IALLUGO) != (fattr->mode & S_IALLUGO)) { + invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL; + inode->i_mode = fattr->mode; + } + } + if (fattr->valid & NFS_ATTR_FATTR_OWNER) { + if (inode->i_uid != fattr->uid) { + invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL; + inode->i_uid = fattr->uid; + } + } + if (fattr->valid & NFS_ATTR_FATTR_GROUP) { + if (inode->i_gid != fattr->gid) { + invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL; + inode->i_gid = fattr->gid; + } } - inode->i_mode = fattr->mode; - inode->i_nlink = fattr->nlink; - inode->i_uid = fattr->uid; - inode->i_gid = fattr->gid; + if (fattr->valid & NFS_ATTR_FATTR_NLINK) { + if (inode->i_nlink != fattr->nlink) { + invalid |= NFS_INO_INVALID_ATTR; + if (S_ISDIR(inode->i_mode)) + invalid |= NFS_INO_INVALID_DATA; + inode->i_nlink = fattr->nlink; + } + } - if (fattr->valid & (NFS_ATTR_FATTR_V3 | NFS_ATTR_FATTR_V4)) { + if (fattr->valid & NFS_ATTR_FATTR_SPACE_USED) { /* * report the blocks in 512byte units */ inode->i_blocks = nfs_calc_block_size(fattr->du.nfs3.used); - } else { - inode->i_blocks = fattr->du.nfs2.blocks; } + if (fattr->valid & NFS_ATTR_FATTR_BLOCKS_USED) + inode->i_blocks = fattr->du.nfs2.blocks; /* Update attrtimeo value if we're out of the unstable period */ if (invalid & NFS_INO_INVALID_ATTR) { diff --git a/fs/nfs/nfs2xdr.c b/fs/nfs/nfs2xdr.c index 28bab67d1519..bea99992c302 100644 --- a/fs/nfs/nfs2xdr.c +++ b/fs/nfs/nfs2xdr.c @@ -136,7 +136,7 @@ xdr_decode_fattr(__be32 *p, struct nfs_fattr *fattr) p = xdr_decode_time(p, &fattr->atime); p = xdr_decode_time(p, &fattr->mtime); p = xdr_decode_time(p, &fattr->ctime); - fattr->valid |= NFS_ATTR_FATTR; + fattr->valid |= NFS_ATTR_FATTR_V2; fattr->rdev = new_decode_dev(rdev); if (fattr->type == NFCHR && rdev == NFS2_FIFO_DEV) { fattr->type = NFFIFO; diff --git a/fs/nfs/nfs3xdr.c b/fs/nfs/nfs3xdr.c index 6cdeacffde46..c0f7d02aced9 100644 --- a/fs/nfs/nfs3xdr.c +++ b/fs/nfs/nfs3xdr.c @@ -177,7 +177,7 @@ xdr_decode_fattr(__be32 *p, struct nfs_fattr *fattr) p = xdr_decode_time3(p, &fattr->ctime); /* Update the mode bits */ - fattr->valid |= (NFS_ATTR_FATTR | NFS_ATTR_FATTR_V3); + fattr->valid |= NFS_ATTR_FATTR_V3; return p; } @@ -233,7 +233,9 @@ xdr_decode_wcc_attr(__be32 *p, struct nfs_fattr *fattr) p = xdr_decode_hyper(p, &fattr->pre_size); p = xdr_decode_time3(p, &fattr->pre_mtime); p = xdr_decode_time3(p, &fattr->pre_ctime); - fattr->valid |= NFS_ATTR_WCC; + fattr->valid |= NFS_ATTR_FATTR_PRESIZE + | NFS_ATTR_FATTR_PREMTIME + | NFS_ATTR_FATTR_PRECTIME; return p; } diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index 5f0ee3e2bd84..7d220da3db36 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -3012,7 +3012,7 @@ static int decode_getfattr(struct xdr_stream *xdr, struct nfs_fattr *fattr, cons if ((status = decode_attr_type(xdr, bitmap, &type)) != 0) goto xdr_error; fattr->type = nfs_type2fmt[type].nfs2type; - fmode = nfs_type2fmt[type].mode; + fattr->mode = nfs_type2fmt[type].mode; if ((status = decode_attr_change(xdr, bitmap, &fattr->change_attr)) != 0) goto xdr_error; @@ -3026,7 +3026,7 @@ static int decode_getfattr(struct xdr_stream *xdr, struct nfs_fattr *fattr, cons struct nfs4_fs_locations, fattr))) != 0) goto xdr_error; - if ((status = decode_attr_mode(xdr, bitmap, &fattr->mode)) != 0) + if ((status = decode_attr_mode(xdr, bitmap, &fmode)) != 0) goto xdr_error; fattr->mode |= fmode; if ((status = decode_attr_nlink(xdr, bitmap, &fattr->nlink)) != 0) @@ -3050,7 +3050,7 @@ static int decode_getfattr(struct xdr_stream *xdr, struct nfs_fattr *fattr, cons if (fattr->fileid == 0 && fileid != 0) fattr->fileid = fileid; if ((status = verify_attr_len(xdr, savep, attrlen)) == 0) - fattr->valid = NFS_ATTR_FATTR | NFS_ATTR_FATTR_V3 | NFS_ATTR_FATTR_V4; + fattr->valid = NFS_ATTR_FATTR_V4; xdr_error: dprintk("%s: xdr returned %d\n", __func__, -status); return status; diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index 2e5f00066afd..b99295e07cdf 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -27,7 +27,7 @@ static inline int nfs_fsid_equal(const struct nfs_fsid *a, const struct nfs_fsid } struct nfs_fattr { - unsigned short valid; /* which fields are valid */ + unsigned int valid; /* which fields are valid */ __u64 pre_size; /* pre_op_attr.size */ struct timespec pre_mtime; /* pre_op_attr.mtime */ struct timespec pre_ctime; /* pre_op_attr.ctime */ @@ -59,12 +59,46 @@ struct nfs_fattr { unsigned long gencount; }; -#define NFS_ATTR_WCC 0x0001 /* pre-op WCC data */ -#define NFS_ATTR_FATTR 0x0002 /* post-op attributes */ -#define NFS_ATTR_FATTR_V3 0x0004 /* NFSv3 attributes */ -#define NFS_ATTR_FATTR_V4 0x0008 /* NFSv4 change attribute */ -#define NFS_ATTR_WCC_V4 0x0010 /* pre-op change attribute */ -#define NFS_ATTR_FATTR_V4_REFERRAL 0x0020 /* NFSv4 referral */ +#define NFS_ATTR_FATTR_TYPE (1U << 0) +#define NFS_ATTR_FATTR_MODE (1U << 1) +#define NFS_ATTR_FATTR_NLINK (1U << 2) +#define NFS_ATTR_FATTR_OWNER (1U << 3) +#define NFS_ATTR_FATTR_GROUP (1U << 4) +#define NFS_ATTR_FATTR_RDEV (1U << 5) +#define NFS_ATTR_FATTR_SIZE (1U << 6) +#define NFS_ATTR_FATTR_PRESIZE (1U << 7) +#define NFS_ATTR_FATTR_BLOCKS_USED (1U << 8) +#define NFS_ATTR_FATTR_SPACE_USED (1U << 9) +#define NFS_ATTR_FATTR_FSID (1U << 10) +#define NFS_ATTR_FATTR_FILEID (1U << 11) +#define NFS_ATTR_FATTR_ATIME (1U << 12) +#define NFS_ATTR_FATTR_MTIME (1U << 13) +#define NFS_ATTR_FATTR_CTIME (1U << 14) +#define NFS_ATTR_FATTR_PREMTIME (1U << 15) +#define NFS_ATTR_FATTR_PRECTIME (1U << 16) +#define NFS_ATTR_FATTR_CHANGE (1U << 17) +#define NFS_ATTR_FATTR_PRECHANGE (1U << 18) +#define NFS_ATTR_FATTR_V4_REFERRAL (1U << 19) /* NFSv4 referral */ + +#define NFS_ATTR_FATTR (NFS_ATTR_FATTR_TYPE \ + | NFS_ATTR_FATTR_MODE \ + | NFS_ATTR_FATTR_NLINK \ + | NFS_ATTR_FATTR_OWNER \ + | NFS_ATTR_FATTR_GROUP \ + | NFS_ATTR_FATTR_RDEV \ + | NFS_ATTR_FATTR_SIZE \ + | NFS_ATTR_FATTR_FSID \ + | NFS_ATTR_FATTR_FILEID \ + | NFS_ATTR_FATTR_ATIME \ + | NFS_ATTR_FATTR_MTIME \ + | NFS_ATTR_FATTR_CTIME) +#define NFS_ATTR_FATTR_V2 (NFS_ATTR_FATTR \ + | NFS_ATTR_FATTR_BLOCKS_USED) +#define NFS_ATTR_FATTR_V3 (NFS_ATTR_FATTR \ + | NFS_ATTR_FATTR_SPACE_USED) +#define NFS_ATTR_FATTR_V4 (NFS_ATTR_FATTR \ + | NFS_ATTR_FATTR_SPACE_USED \ + | NFS_ATTR_FATTR_CHANGE) /* * Info on the file system From 1ca277d88dafdbc3c5a69d32590e7184b9af6371 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 11 Mar 2009 14:10:25 -0400 Subject: [PATCH 06/58] NFS: Shrink the struct nfs_fattr We don't need the bitmap[] field anymore, since the 'valid' field tells us all we need to know about which attributes were filled in... Also move the pre-op attributes in order to improve the structure packing. Signed-off-by: Trond Myklebust --- fs/nfs/nfs4xdr.c | 3 --- include/linux/nfs_xdr.h | 7 +++---- 2 files changed, 3 insertions(+), 7 deletions(-) diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index 7d220da3db36..9f1df8361974 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -3002,9 +3002,6 @@ static int decode_getfattr(struct xdr_stream *xdr, struct nfs_fattr *fattr, cons if ((status = decode_attr_bitmap(xdr, bitmap)) != 0) goto xdr_error; - fattr->bitmap[0] = bitmap[0]; - fattr->bitmap[1] = bitmap[1]; - if ((status = decode_attr_length(xdr, &attrlen, &savep)) != 0) goto xdr_error; diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index b99295e07cdf..6013acb0131f 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -28,9 +28,6 @@ static inline int nfs_fsid_equal(const struct nfs_fsid *a, const struct nfs_fsid struct nfs_fattr { unsigned int valid; /* which fields are valid */ - __u64 pre_size; /* pre_op_attr.size */ - struct timespec pre_mtime; /* pre_op_attr.mtime */ - struct timespec pre_ctime; /* pre_op_attr.ctime */ enum nfs_ftype type; /* always use NFSv2 types */ __u32 mode; __u32 nlink; @@ -52,9 +49,11 @@ struct nfs_fattr { struct timespec atime; struct timespec mtime; struct timespec ctime; - __u32 bitmap[2]; /* NFSv4 returned attribute bitmap */ __u64 change_attr; /* NFSv4 change attribute */ __u64 pre_change_attr;/* pre-op NFSv4 change attribute */ + __u64 pre_size; /* pre_op_attr.size */ + struct timespec pre_mtime; /* pre_op_attr.mtime */ + struct timespec pre_ctime; /* pre_op_attr.ctime */ unsigned long time_start; unsigned long gencount; }; From bca794785c2c12ecddeb09e70165b8ff80baa6ae Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 11 Mar 2009 14:10:26 -0400 Subject: [PATCH 07/58] NFS: Fix the type of struct nfs_fattr->mode There is no point in using anything other than umode_t, since we copy the content pretty much directly into inode->i_mode. Signed-off-by: Trond Myklebust --- fs/nfs/getroot.c | 4 ++-- fs/nfs/nfs2xdr.c | 7 +++---- fs/nfs/nfs3xdr.c | 31 +++++++++++++------------------ fs/nfs/nfs4xdr.c | 40 +++++++++++++++++++--------------------- include/linux/nfs_xdr.h | 3 +-- 5 files changed, 38 insertions(+), 47 deletions(-) diff --git a/fs/nfs/getroot.c b/fs/nfs/getroot.c index b7c9b2df1f29..46177cb87064 100644 --- a/fs/nfs/getroot.c +++ b/fs/nfs/getroot.c @@ -156,7 +156,7 @@ int nfs4_path_walk(struct nfs_server *server, return ret; } - if (fattr.type != NFDIR) { + if (!S_ISDIR(fattr.mode)) { printk(KERN_ERR "nfs4_get_root:" " getroot encountered non-directory\n"); return -ENOTDIR; @@ -213,7 +213,7 @@ eat_dot_dir: return ret; } - if (fattr.type != NFDIR) { + if (!S_ISDIR(fattr.mode)) { printk(KERN_ERR "nfs4_get_root:" " lookupfh encountered non-directory\n"); return -ENOTDIR; diff --git a/fs/nfs/nfs2xdr.c b/fs/nfs/nfs2xdr.c index bea99992c302..c862c9340f9a 100644 --- a/fs/nfs/nfs2xdr.c +++ b/fs/nfs/nfs2xdr.c @@ -120,8 +120,8 @@ xdr_decode_time(__be32 *p, struct timespec *timep) static __be32 * xdr_decode_fattr(__be32 *p, struct nfs_fattr *fattr) { - u32 rdev; - fattr->type = (enum nfs_ftype) ntohl(*p++); + u32 rdev, type; + type = ntohl(*p++); fattr->mode = ntohl(*p++); fattr->nlink = ntohl(*p++); fattr->uid = ntohl(*p++); @@ -138,8 +138,7 @@ xdr_decode_fattr(__be32 *p, struct nfs_fattr *fattr) p = xdr_decode_time(p, &fattr->ctime); fattr->valid |= NFS_ATTR_FATTR_V2; fattr->rdev = new_decode_dev(rdev); - if (fattr->type == NFCHR && rdev == NFS2_FIFO_DEV) { - fattr->type = NFFIFO; + if (type == NFCHR && rdev == NFS2_FIFO_DEV) { fattr->mode = (fattr->mode & ~S_IFMT) | S_IFIFO; fattr->rdev = 0; } diff --git a/fs/nfs/nfs3xdr.c b/fs/nfs/nfs3xdr.c index c0f7d02aced9..e6a1932c7110 100644 --- a/fs/nfs/nfs3xdr.c +++ b/fs/nfs/nfs3xdr.c @@ -91,19 +91,15 @@ /* * Map file type to S_IFMT bits */ -static struct { - unsigned int mode; - unsigned int nfs2type; -} nfs_type2fmt[] = { - { 0, NFNON }, - { S_IFREG, NFREG }, - { S_IFDIR, NFDIR }, - { S_IFBLK, NFBLK }, - { S_IFCHR, NFCHR }, - { S_IFLNK, NFLNK }, - { S_IFSOCK, NFSOCK }, - { S_IFIFO, NFFIFO }, - { 0, NFBAD } +static const umode_t nfs_type2fmt[] = { + [NF3BAD] = 0, + [NF3REG] = S_IFREG, + [NF3DIR] = S_IFDIR, + [NF3BLK] = S_IFBLK, + [NF3CHR] = S_IFCHR, + [NF3LNK] = S_IFLNK, + [NF3SOCK] = S_IFSOCK, + [NF3FIFO] = S_IFIFO, }; /* @@ -148,13 +144,12 @@ static __be32 * xdr_decode_fattr(__be32 *p, struct nfs_fattr *fattr) { unsigned int type, major, minor; - int fmode; + umode_t fmode; type = ntohl(*p++); - if (type >= NF3BAD) - type = NF3BAD; - fmode = nfs_type2fmt[type].mode; - fattr->type = nfs_type2fmt[type].nfs2type; + if (type > NF3FIFO) + type = NF3NON; + fmode = nfs_type2fmt[type]; fattr->mode = (ntohl(*p++) & ~S_IFMT) | fmode; fattr->nlink = ntohl(*p++); fattr->uid = ntohl(*p++); diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index 9f1df8361974..c1906d2a226b 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -522,20 +522,17 @@ static int nfs4_stat_to_errno(int); decode_lookup_maxsz + \ decode_fs_locations_maxsz) -static struct { - unsigned int mode; - unsigned int nfs2type; -} nfs_type2fmt[] = { - { 0, NFNON }, - { S_IFREG, NFREG }, - { S_IFDIR, NFDIR }, - { S_IFBLK, NFBLK }, - { S_IFCHR, NFCHR }, - { S_IFLNK, NFLNK }, - { S_IFSOCK, NFSOCK }, - { S_IFIFO, NFFIFO }, - { 0, NFNON }, - { 0, NFNON }, +static const umode_t nfs_type2fmt[] = { + [NF4BAD] = 0, + [NF4REG] = S_IFREG, + [NF4DIR] = S_IFDIR, + [NF4BLK] = S_IFBLK, + [NF4CHR] = S_IFCHR, + [NF4LNK] = S_IFLNK, + [NF4SOCK] = S_IFSOCK, + [NF4FIFO] = S_IFIFO, + [NF4ATTRDIR] = 0, + [NF4NAMEDATTR] = 0, }; struct compound_hdr { @@ -2173,7 +2170,7 @@ static int decode_attr_type(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t * } bitmap[0] &= ~FATTR4_WORD0_TYPE; } - dprintk("%s: type=0%o\n", __func__, nfs_type2fmt[*type].nfs2type); + dprintk("%s: type=0%o\n", __func__, nfs_type2fmt[*type]); return 0; } @@ -2580,8 +2577,9 @@ static int decode_attr_maxwrite(struct xdr_stream *xdr, uint32_t *bitmap, uint32 return status; } -static int decode_attr_mode(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t *mode) +static int decode_attr_mode(struct xdr_stream *xdr, uint32_t *bitmap, umode_t *mode) { + uint32_t tmp; __be32 *p; *mode = 0; @@ -2589,8 +2587,8 @@ static int decode_attr_mode(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t * return -EIO; if (likely(bitmap[1] & FATTR4_WORD1_MODE)) { READ_BUF(4); - READ32(*mode); - *mode &= ~S_IFMT; + READ32(tmp); + *mode = tmp & ~S_IFMT; bitmap[1] &= ~FATTR4_WORD1_MODE; } dprintk("%s: file mode=0%o\n", __func__, (unsigned int)*mode); @@ -2994,7 +2992,8 @@ static int decode_getfattr(struct xdr_stream *xdr, struct nfs_fattr *fattr, cons uint32_t attrlen, bitmap[2] = {0}, type; - int status, fmode = 0; + int status; + umode_t fmode = 0; uint64_t fileid; if ((status = decode_op_hdr(xdr, OP_GETATTR)) != 0) @@ -3008,8 +3007,7 @@ static int decode_getfattr(struct xdr_stream *xdr, struct nfs_fattr *fattr, cons if ((status = decode_attr_type(xdr, bitmap, &type)) != 0) goto xdr_error; - fattr->type = nfs_type2fmt[type].nfs2type; - fattr->mode = nfs_type2fmt[type].mode; + fattr->mode = nfs_type2fmt[type]; if ((status = decode_attr_change(xdr, bitmap, &fattr->change_attr)) != 0) goto xdr_error; diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index 6013acb0131f..0691b9c188d9 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -28,8 +28,7 @@ static inline int nfs_fsid_equal(const struct nfs_fsid *a, const struct nfs_fsid struct nfs_fattr { unsigned int valid; /* which fields are valid */ - enum nfs_ftype type; /* always use NFSv2 types */ - __u32 mode; + umode_t mode; __u32 nlink; __u32 uid; __u32 gid; From f26c7a78876ccd6c9b477ab4ca127aa1a4ef68c7 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 11 Mar 2009 14:10:26 -0400 Subject: [PATCH 08/58] NFSv4: Clean up decode_getfattr() Signed-off-by: Trond Myklebust --- fs/nfs/nfs4xdr.c | 80 +++++++++++++++++++++++++++++++++++------------- 1 file changed, 58 insertions(+), 22 deletions(-) diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index c1906d2a226b..43c6e50ff173 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -2996,55 +2996,91 @@ static int decode_getfattr(struct xdr_stream *xdr, struct nfs_fattr *fattr, cons umode_t fmode = 0; uint64_t fileid; - if ((status = decode_op_hdr(xdr, OP_GETATTR)) != 0) - goto xdr_error; - if ((status = decode_attr_bitmap(xdr, bitmap)) != 0) + status = decode_op_hdr(xdr, OP_GETATTR); + if (status < 0) goto xdr_error; - if ((status = decode_attr_length(xdr, &attrlen, &savep)) != 0) + status = decode_attr_bitmap(xdr, bitmap); + if (status < 0) + goto xdr_error; + + status = decode_attr_length(xdr, &attrlen, &savep); + if (status < 0) goto xdr_error; - if ((status = decode_attr_type(xdr, bitmap, &type)) != 0) + status = decode_attr_type(xdr, bitmap, &type); + if (status < 0) goto xdr_error; fattr->mode = nfs_type2fmt[type]; - if ((status = decode_attr_change(xdr, bitmap, &fattr->change_attr)) != 0) + status = decode_attr_change(xdr, bitmap, &fattr->change_attr); + if (status < 0) goto xdr_error; - if ((status = decode_attr_size(xdr, bitmap, &fattr->size)) != 0) + + status = decode_attr_size(xdr, bitmap, &fattr->size); + if (status < 0) goto xdr_error; - if ((status = decode_attr_fsid(xdr, bitmap, &fattr->fsid)) != 0) + + status = decode_attr_fsid(xdr, bitmap, &fattr->fsid); + if (status < 0) goto xdr_error; - if ((status = decode_attr_fileid(xdr, bitmap, &fattr->fileid)) != 0) + + status = decode_attr_fileid(xdr, bitmap, &fattr->fileid); + if (status < 0) goto xdr_error; - if ((status = decode_attr_fs_locations(xdr, bitmap, container_of(fattr, + + status = decode_attr_fs_locations(xdr, bitmap, container_of(fattr, struct nfs4_fs_locations, - fattr))) != 0) + fattr)); + if (status < 0) goto xdr_error; - if ((status = decode_attr_mode(xdr, bitmap, &fmode)) != 0) + + status = decode_attr_mode(xdr, bitmap, &fmode); + if (status < 0) goto xdr_error; fattr->mode |= fmode; - if ((status = decode_attr_nlink(xdr, bitmap, &fattr->nlink)) != 0) + + status = decode_attr_nlink(xdr, bitmap, &fattr->nlink); + if (status < 0) goto xdr_error; - if ((status = decode_attr_owner(xdr, bitmap, server->nfs_client, &fattr->uid)) != 0) + + status = decode_attr_owner(xdr, bitmap, server->nfs_client, &fattr->uid); + if (status < 0) goto xdr_error; - if ((status = decode_attr_group(xdr, bitmap, server->nfs_client, &fattr->gid)) != 0) + + status = decode_attr_group(xdr, bitmap, server->nfs_client, &fattr->gid); + if (status < 0) goto xdr_error; - if ((status = decode_attr_rdev(xdr, bitmap, &fattr->rdev)) != 0) + + status = decode_attr_rdev(xdr, bitmap, &fattr->rdev); + if (status < 0) goto xdr_error; - if ((status = decode_attr_space_used(xdr, bitmap, &fattr->du.nfs3.used)) != 0) + + status = decode_attr_space_used(xdr, bitmap, &fattr->du.nfs3.used); + if (status < 0) goto xdr_error; - if ((status = decode_attr_time_access(xdr, bitmap, &fattr->atime)) != 0) + + status = decode_attr_time_access(xdr, bitmap, &fattr->atime); + if (status < 0) goto xdr_error; - if ((status = decode_attr_time_metadata(xdr, bitmap, &fattr->ctime)) != 0) + + status = decode_attr_time_metadata(xdr, bitmap, &fattr->ctime); + if (status < 0) goto xdr_error; - if ((status = decode_attr_time_modify(xdr, bitmap, &fattr->mtime)) != 0) + + status = decode_attr_time_modify(xdr, bitmap, &fattr->mtime); + if (status < 0) goto xdr_error; - if ((status = decode_attr_mounted_on_fileid(xdr, bitmap, &fileid)) != 0) + + status = decode_attr_mounted_on_fileid(xdr, bitmap, &fileid); + if (status < 0) goto xdr_error; if (fattr->fileid == 0 && fileid != 0) fattr->fileid = fileid; - if ((status = verify_attr_len(xdr, savep, attrlen)) == 0) + + status = verify_attr_len(xdr, savep, attrlen); + if (status == 0) fattr->valid = NFS_ATTR_FATTR_V4; xdr_error: dprintk("%s: xdr returned %d\n", __func__, -status); From 409924e4c943072a63c43bb6b77576bf12f1896b Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 11 Mar 2009 14:10:27 -0400 Subject: [PATCH 09/58] NFSv4: Make decode_getfattr() set fattr->valid to reflect what was decoded Signed-off-by: Trond Myklebust --- fs/nfs/nfs4xdr.c | 92 ++++++++++++++++++++++++++++++++++++++---------- 1 file changed, 73 insertions(+), 19 deletions(-) diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index 43c6e50ff173..1690f0e44b91 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -2157,6 +2157,7 @@ static int decode_attr_supported(struct xdr_stream *xdr, uint32_t *bitmap, uint3 static int decode_attr_type(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t *type) { __be32 *p; + int ret = 0; *type = 0; if (unlikely(bitmap[0] & (FATTR4_WORD0_TYPE - 1U))) @@ -2169,14 +2170,16 @@ static int decode_attr_type(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t * return -EIO; } bitmap[0] &= ~FATTR4_WORD0_TYPE; + ret = NFS_ATTR_FATTR_TYPE; } dprintk("%s: type=0%o\n", __func__, nfs_type2fmt[*type]); - return 0; + return ret; } static int decode_attr_change(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *change) { __be32 *p; + int ret = 0; *change = 0; if (unlikely(bitmap[0] & (FATTR4_WORD0_CHANGE - 1U))) @@ -2185,15 +2188,17 @@ static int decode_attr_change(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t READ_BUF(8); READ64(*change); bitmap[0] &= ~FATTR4_WORD0_CHANGE; + ret = NFS_ATTR_FATTR_CHANGE; } dprintk("%s: change attribute=%Lu\n", __func__, (unsigned long long)*change); - return 0; + return ret; } static int decode_attr_size(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *size) { __be32 *p; + int ret = 0; *size = 0; if (unlikely(bitmap[0] & (FATTR4_WORD0_SIZE - 1U))) @@ -2202,9 +2207,10 @@ static int decode_attr_size(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t * READ_BUF(8); READ64(*size); bitmap[0] &= ~FATTR4_WORD0_SIZE; + ret = NFS_ATTR_FATTR_SIZE; } dprintk("%s: file size=%Lu\n", __func__, (unsigned long long)*size); - return 0; + return ret; } static int decode_attr_link_support(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t *res) @@ -2242,6 +2248,7 @@ static int decode_attr_symlink_support(struct xdr_stream *xdr, uint32_t *bitmap, static int decode_attr_fsid(struct xdr_stream *xdr, uint32_t *bitmap, struct nfs_fsid *fsid) { __be32 *p; + int ret = 0; fsid->major = 0; fsid->minor = 0; @@ -2252,11 +2259,12 @@ static int decode_attr_fsid(struct xdr_stream *xdr, uint32_t *bitmap, struct nfs READ64(fsid->major); READ64(fsid->minor); bitmap[0] &= ~FATTR4_WORD0_FSID; + ret = NFS_ATTR_FATTR_FSID; } dprintk("%s: fsid=(0x%Lx/0x%Lx)\n", __func__, (unsigned long long)fsid->major, (unsigned long long)fsid->minor); - return 0; + return ret; } static int decode_attr_lease_time(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t *res) @@ -2294,6 +2302,7 @@ static int decode_attr_aclsupport(struct xdr_stream *xdr, uint32_t *bitmap, uint static int decode_attr_fileid(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *fileid) { __be32 *p; + int ret = 0; *fileid = 0; if (unlikely(bitmap[0] & (FATTR4_WORD0_FILEID - 1U))) @@ -2302,14 +2311,16 @@ static int decode_attr_fileid(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t READ_BUF(8); READ64(*fileid); bitmap[0] &= ~FATTR4_WORD0_FILEID; + ret = NFS_ATTR_FATTR_FILEID; } dprintk("%s: fileid=%Lu\n", __func__, (unsigned long long)*fileid); - return 0; + return ret; } static int decode_attr_mounted_on_fileid(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *fileid) { __be32 *p; + int ret = 0; *fileid = 0; if (unlikely(bitmap[1] & (FATTR4_WORD1_MOUNTED_ON_FILEID - 1U))) @@ -2318,9 +2329,10 @@ static int decode_attr_mounted_on_fileid(struct xdr_stream *xdr, uint32_t *bitma READ_BUF(8); READ64(*fileid); bitmap[1] &= ~FATTR4_WORD1_MOUNTED_ON_FILEID; + ret = NFS_ATTR_FATTR_FILEID; } dprintk("%s: fileid=%Lu\n", __func__, (unsigned long long)*fileid); - return 0; + return ret; } static int decode_attr_files_avail(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *res) @@ -2476,6 +2488,8 @@ static int decode_attr_fs_locations(struct xdr_stream *xdr, uint32_t *bitmap, st if (res->nlocations < NFS4_FS_LOCATIONS_MAXENTRIES) res->nlocations++; } + if (res->nlocations != 0) + status = NFS_ATTR_FATTR_V4_REFERRAL; out: dprintk("%s: fs_locations done, error = %d\n", __func__, status); return status; @@ -2581,6 +2595,7 @@ static int decode_attr_mode(struct xdr_stream *xdr, uint32_t *bitmap, umode_t *m { uint32_t tmp; __be32 *p; + int ret = 0; *mode = 0; if (unlikely(bitmap[1] & (FATTR4_WORD1_MODE - 1U))) @@ -2590,14 +2605,16 @@ static int decode_attr_mode(struct xdr_stream *xdr, uint32_t *bitmap, umode_t *m READ32(tmp); *mode = tmp & ~S_IFMT; bitmap[1] &= ~FATTR4_WORD1_MODE; + ret = NFS_ATTR_FATTR_MODE; } dprintk("%s: file mode=0%o\n", __func__, (unsigned int)*mode); - return 0; + return ret; } static int decode_attr_nlink(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t *nlink) { __be32 *p; + int ret = 0; *nlink = 1; if (unlikely(bitmap[1] & (FATTR4_WORD1_NUMLINKS - 1U))) @@ -2606,15 +2623,17 @@ static int decode_attr_nlink(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t READ_BUF(4); READ32(*nlink); bitmap[1] &= ~FATTR4_WORD1_NUMLINKS; + ret = NFS_ATTR_FATTR_NLINK; } dprintk("%s: nlink=%u\n", __func__, (unsigned int)*nlink); - return 0; + return ret; } static int decode_attr_owner(struct xdr_stream *xdr, uint32_t *bitmap, struct nfs_client *clp, uint32_t *uid) { uint32_t len; __be32 *p; + int ret = 0; *uid = -2; if (unlikely(bitmap[1] & (FATTR4_WORD1_OWNER - 1U))) @@ -2624,7 +2643,9 @@ static int decode_attr_owner(struct xdr_stream *xdr, uint32_t *bitmap, struct nf READ32(len); READ_BUF(len); if (len < XDR_MAX_NETOBJ) { - if (nfs_map_name_to_uid(clp, (char *)p, len, uid) != 0) + if (nfs_map_name_to_uid(clp, (char *)p, len, uid) == 0) + ret = NFS_ATTR_FATTR_OWNER; + else dprintk("%s: nfs_map_name_to_uid failed!\n", __func__); } else @@ -2633,13 +2654,14 @@ static int decode_attr_owner(struct xdr_stream *xdr, uint32_t *bitmap, struct nf bitmap[1] &= ~FATTR4_WORD1_OWNER; } dprintk("%s: uid=%d\n", __func__, (int)*uid); - return 0; + return ret; } static int decode_attr_group(struct xdr_stream *xdr, uint32_t *bitmap, struct nfs_client *clp, uint32_t *gid) { uint32_t len; __be32 *p; + int ret = 0; *gid = -2; if (unlikely(bitmap[1] & (FATTR4_WORD1_OWNER_GROUP - 1U))) @@ -2649,7 +2671,9 @@ static int decode_attr_group(struct xdr_stream *xdr, uint32_t *bitmap, struct nf READ32(len); READ_BUF(len); if (len < XDR_MAX_NETOBJ) { - if (nfs_map_group_to_gid(clp, (char *)p, len, gid) != 0) + if (nfs_map_group_to_gid(clp, (char *)p, len, gid) == 0) + ret = NFS_ATTR_FATTR_GROUP; + else dprintk("%s: nfs_map_group_to_gid failed!\n", __func__); } else @@ -2658,13 +2682,14 @@ static int decode_attr_group(struct xdr_stream *xdr, uint32_t *bitmap, struct nf bitmap[1] &= ~FATTR4_WORD1_OWNER_GROUP; } dprintk("%s: gid=%d\n", __func__, (int)*gid); - return 0; + return ret; } static int decode_attr_rdev(struct xdr_stream *xdr, uint32_t *bitmap, dev_t *rdev) { uint32_t major = 0, minor = 0; __be32 *p; + int ret = 0; *rdev = MKDEV(0,0); if (unlikely(bitmap[1] & (FATTR4_WORD1_RAWDEV - 1U))) @@ -2679,9 +2704,10 @@ static int decode_attr_rdev(struct xdr_stream *xdr, uint32_t *bitmap, dev_t *rde if (MAJOR(tmp) == major && MINOR(tmp) == minor) *rdev = tmp; bitmap[1] &= ~ FATTR4_WORD1_RAWDEV; + ret = NFS_ATTR_FATTR_RDEV; } dprintk("%s: rdev=(0x%x:0x%x)\n", __func__, major, minor); - return 0; + return ret; } static int decode_attr_space_avail(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *res) @@ -2738,6 +2764,7 @@ static int decode_attr_space_total(struct xdr_stream *xdr, uint32_t *bitmap, uin static int decode_attr_space_used(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *used) { __be32 *p; + int ret = 0; *used = 0; if (unlikely(bitmap[1] & (FATTR4_WORD1_SPACE_USED - 1U))) @@ -2746,10 +2773,11 @@ static int decode_attr_space_used(struct xdr_stream *xdr, uint32_t *bitmap, uint READ_BUF(8); READ64(*used); bitmap[1] &= ~FATTR4_WORD1_SPACE_USED; + ret = NFS_ATTR_FATTR_SPACE_USED; } dprintk("%s: space used=%Lu\n", __func__, (unsigned long long)*used); - return 0; + return ret; } static int decode_attr_time(struct xdr_stream *xdr, struct timespec *time) @@ -2776,6 +2804,8 @@ static int decode_attr_time_access(struct xdr_stream *xdr, uint32_t *bitmap, str return -EIO; if (likely(bitmap[1] & FATTR4_WORD1_TIME_ACCESS)) { status = decode_attr_time(xdr, time); + if (status == 0) + status = NFS_ATTR_FATTR_ATIME; bitmap[1] &= ~FATTR4_WORD1_TIME_ACCESS; } dprintk("%s: atime=%ld\n", __func__, (long)time->tv_sec); @@ -2792,6 +2822,8 @@ static int decode_attr_time_metadata(struct xdr_stream *xdr, uint32_t *bitmap, s return -EIO; if (likely(bitmap[1] & FATTR4_WORD1_TIME_METADATA)) { status = decode_attr_time(xdr, time); + if (status == 0) + status = NFS_ATTR_FATTR_CTIME; bitmap[1] &= ~FATTR4_WORD1_TIME_METADATA; } dprintk("%s: ctime=%ld\n", __func__, (long)time->tv_sec); @@ -2808,6 +2840,8 @@ static int decode_attr_time_modify(struct xdr_stream *xdr, uint32_t *bitmap, str return -EIO; if (likely(bitmap[1] & FATTR4_WORD1_TIME_MODIFY)) { status = decode_attr_time(xdr, time); + if (status == 0) + status = NFS_ATTR_FATTR_MTIME; bitmap[1] &= ~FATTR4_WORD1_TIME_MODIFY; } dprintk("%s: mtime=%ld\n", __func__, (long)time->tv_sec); @@ -3012,76 +3046,96 @@ static int decode_getfattr(struct xdr_stream *xdr, struct nfs_fattr *fattr, cons status = decode_attr_type(xdr, bitmap, &type); if (status < 0) goto xdr_error; - fattr->mode = nfs_type2fmt[type]; + fattr->mode = 0; + if (status != 0) { + fattr->mode |= nfs_type2fmt[type]; + fattr->valid |= status; + } status = decode_attr_change(xdr, bitmap, &fattr->change_attr); if (status < 0) goto xdr_error; + fattr->valid |= status; status = decode_attr_size(xdr, bitmap, &fattr->size); if (status < 0) goto xdr_error; + fattr->valid |= status; status = decode_attr_fsid(xdr, bitmap, &fattr->fsid); if (status < 0) goto xdr_error; + fattr->valid |= status; status = decode_attr_fileid(xdr, bitmap, &fattr->fileid); if (status < 0) goto xdr_error; + fattr->valid |= status; status = decode_attr_fs_locations(xdr, bitmap, container_of(fattr, struct nfs4_fs_locations, fattr)); if (status < 0) goto xdr_error; + fattr->valid |= status; status = decode_attr_mode(xdr, bitmap, &fmode); if (status < 0) goto xdr_error; - fattr->mode |= fmode; + if (status != 0) { + fattr->mode |= fmode; + fattr->valid |= status; + } status = decode_attr_nlink(xdr, bitmap, &fattr->nlink); if (status < 0) goto xdr_error; + fattr->valid |= status; status = decode_attr_owner(xdr, bitmap, server->nfs_client, &fattr->uid); if (status < 0) goto xdr_error; + fattr->valid |= status; status = decode_attr_group(xdr, bitmap, server->nfs_client, &fattr->gid); if (status < 0) goto xdr_error; + fattr->valid |= status; status = decode_attr_rdev(xdr, bitmap, &fattr->rdev); if (status < 0) goto xdr_error; + fattr->valid |= status; status = decode_attr_space_used(xdr, bitmap, &fattr->du.nfs3.used); if (status < 0) goto xdr_error; + fattr->valid |= status; status = decode_attr_time_access(xdr, bitmap, &fattr->atime); if (status < 0) goto xdr_error; + fattr->valid |= status; status = decode_attr_time_metadata(xdr, bitmap, &fattr->ctime); if (status < 0) goto xdr_error; + fattr->valid |= status; status = decode_attr_time_modify(xdr, bitmap, &fattr->mtime); if (status < 0) goto xdr_error; + fattr->valid |= status; status = decode_attr_mounted_on_fileid(xdr, bitmap, &fileid); if (status < 0) goto xdr_error; - if (fattr->fileid == 0 && fileid != 0) + if (status != 0 && !(fattr->valid & status)) { fattr->fileid = fileid; + fattr->valid |= status; + } status = verify_attr_len(xdr, savep, attrlen); - if (status == 0) - fattr->valid = NFS_ATTR_FATTR_V4; xdr_error: dprintk("%s: xdr returned %d\n", __func__, -status); return status; From 69aaaae18f7027d9594bce100378f102926cc0be Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 11 Mar 2009 14:10:28 -0400 Subject: [PATCH 10/58] NFSv4: A referral is assumed to always point to a directory. Fix a bug whereby we would fail to create a mount point for a referral. Signed-off-by: Trond Myklebust --- fs/nfs/nfs4proc.c | 14 ++++++++++++++ 1 file changed, 14 insertions(+) diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 8dde84b988d9..aa433d077945 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -3678,6 +3678,19 @@ ssize_t nfs4_listxattr(struct dentry *dentry, char *buf, size_t buflen) return len; } +static void nfs_fixup_referral_attributes(struct nfs_fattr *fattr) +{ + if (!((fattr->valid & NFS_ATTR_FATTR_FILEID) && + (fattr->valid & NFS_ATTR_FATTR_FSID) && + (fattr->valid & NFS_ATTR_FATTR_V4_REFERRAL))) + return; + + fattr->valid |= NFS_ATTR_FATTR_TYPE | NFS_ATTR_FATTR_MODE | + NFS_ATTR_FATTR_NLINK; + fattr->mode = S_IFDIR | S_IRUGO | S_IXUGO; + fattr->nlink = 2; +} + int nfs4_proc_fs_locations(struct inode *dir, const struct qstr *name, struct nfs4_fs_locations *fs_locations, struct page *page) { @@ -3704,6 +3717,7 @@ int nfs4_proc_fs_locations(struct inode *dir, const struct qstr *name, fs_locations->server = server; fs_locations->nlocations = 0; status = rpc_call_sync(server->client, &msg, 0); + nfs_fixup_referral_attributes(&fs_locations->fattr); dprintk("%s: returned status = %d\n", __func__, status); return status; } From a65318bf3afc93ce49227e849d213799b072c5fd Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 11 Mar 2009 14:10:28 -0400 Subject: [PATCH 11/58] NFSv4: Simplify some cache consistency post-op GETATTRs Certain asynchronous operations such as write() do not expect (or care) that other metadata such as the file owner, mode, acls, ... change. All they want to do is update and/or check the change attribute, ctime, and mtime. By skipping the file owner and group update, we also avoid having to do a potential idmapper upcall for these asynchronous RPC calls. Signed-off-by: Trond Myklebust --- fs/nfs/nfs4proc.c | 13 ++++++++----- include/linux/nfs_fs_sb.h | 5 +++++ 2 files changed, 13 insertions(+), 5 deletions(-) diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index aa433d077945..101f5f4c304f 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -1439,7 +1439,7 @@ int nfs4_do_close(struct path *path, struct nfs4_state *state, int wait) if (calldata->arg.seqid == NULL) goto out_free_calldata; calldata->arg.fmode = 0; - calldata->arg.bitmask = server->attr_bitmask; + calldata->arg.bitmask = server->cache_consistency_bitmask; calldata->res.fattr = &calldata->fattr; calldata->res.seqid = calldata->arg.seqid; calldata->res.server = server; @@ -1600,6 +1600,9 @@ static int _nfs4_server_capabilities(struct nfs_server *server, struct nfs_fh *f server->caps |= NFS_CAP_HARDLINKS; if (res.has_symlinks != 0) server->caps |= NFS_CAP_SYMLINKS; + memcpy(server->cache_consistency_bitmask, res.attr_bitmask, sizeof(server->cache_consistency_bitmask)); + server->cache_consistency_bitmask[0] &= FATTR4_WORD0_CHANGE|FATTR4_WORD0_SIZE; + server->cache_consistency_bitmask[1] &= FATTR4_WORD1_TIME_METADATA|FATTR4_WORD1_TIME_MODIFY; server->acl_bitmask = res.acl_bitmask; } return status; @@ -2079,7 +2082,7 @@ static void nfs4_proc_unlink_setup(struct rpc_message *msg, struct inode *dir) struct nfs_removeargs *args = msg->rpc_argp; struct nfs_removeres *res = msg->rpc_resp; - args->bitmask = server->attr_bitmask; + args->bitmask = server->cache_consistency_bitmask; res->server = server; msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_REMOVE]; } @@ -2323,7 +2326,7 @@ static int _nfs4_proc_readdir(struct dentry *dentry, struct rpc_cred *cred, .pages = &page, .pgbase = 0, .count = count, - .bitmask = NFS_SERVER(dentry->d_inode)->attr_bitmask, + .bitmask = NFS_SERVER(dentry->d_inode)->cache_consistency_bitmask, }; struct nfs4_readdir_res res; struct rpc_message msg = { @@ -2552,7 +2555,7 @@ static void nfs4_proc_write_setup(struct nfs_write_data *data, struct rpc_messag { struct nfs_server *server = NFS_SERVER(data->inode); - data->args.bitmask = server->attr_bitmask; + data->args.bitmask = server->cache_consistency_bitmask; data->res.server = server; data->timestamp = jiffies; @@ -2575,7 +2578,7 @@ static void nfs4_proc_commit_setup(struct nfs_write_data *data, struct rpc_messa { struct nfs_server *server = NFS_SERVER(data->inode); - data->args.bitmask = server->attr_bitmask; + data->args.bitmask = server->cache_consistency_bitmask; data->res.server = server; msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_COMMIT]; } diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h index 9bb81aec91cf..29b1e40dce99 100644 --- a/include/linux/nfs_fs_sb.h +++ b/include/linux/nfs_fs_sb.h @@ -106,6 +106,11 @@ struct nfs_server { u32 attr_bitmask[2];/* V4 bitmask representing the set of attributes supported on this filesystem */ + u32 cache_consistency_bitmask[2]; + /* V4 bitmask representing the subset + of change attribute, size, ctime + and mtime attributes supported by + the server */ u32 acl_bitmask; /* V4 bitmask representing the ACEs that are supported on this filesystem */ From fb8a1f11b64e213d94dfa1cebb2a42a7b8c115c4 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 11 Mar 2009 14:10:29 -0400 Subject: [PATCH 12/58] NFS: cleanup - remove struct nfs_inode->ncommit Signed-off-by: Trond Myklebust --- fs/nfs/inode.c | 1 - fs/nfs/write.c | 25 ++++++++++++++++--------- include/linux/nfs_fs.h | 3 +-- 3 files changed, 17 insertions(+), 12 deletions(-) diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index b7656bd3706f..00f116cdadc6 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -1345,7 +1345,6 @@ static void init_once(void *foo) INIT_LIST_HEAD(&nfsi->access_cache_entry_lru); INIT_LIST_HEAD(&nfsi->access_cache_inode_lru); INIT_RADIX_TREE(&nfsi->nfs_page_tree, GFP_ATOMIC); - nfsi->ncommit = 0; nfsi->npages = 0; atomic_set(&nfsi->silly_count, 1); INIT_HLIST_HEAD(&nfsi->silly_list); diff --git a/fs/nfs/write.c b/fs/nfs/write.c index 9f9845859fc1..1a999939fedf 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -404,7 +404,6 @@ nfs_mark_request_commit(struct nfs_page *req) struct nfs_inode *nfsi = NFS_I(inode); spin_lock(&inode->i_lock); - nfsi->ncommit++; set_bit(PG_CLEAN, &(req)->wb_flags); radix_tree_tag_set(&nfsi->nfs_page_tree, req->wb_index, @@ -523,6 +522,12 @@ static void nfs_cancel_commit_list(struct list_head *head) } } +static int +nfs_need_commit(struct nfs_inode *nfsi) +{ + return radix_tree_tagged(&nfsi->nfs_page_tree, NFS_PAGE_TAG_COMMIT); +} + #if defined(CONFIG_NFS_V3) || defined(CONFIG_NFS_V4) /* * nfs_scan_commit - Scan an inode for commit requests @@ -538,16 +543,18 @@ static int nfs_scan_commit(struct inode *inode, struct list_head *dst, pgoff_t idx_start, unsigned int npages) { struct nfs_inode *nfsi = NFS_I(inode); - int res = 0; - if (nfsi->ncommit != 0) { - res = nfs_scan_list(nfsi, dst, idx_start, npages, - NFS_PAGE_TAG_COMMIT); - nfsi->ncommit -= res; - } - return res; + if (!nfs_need_commit(nfsi)) + return 0; + + return nfs_scan_list(nfsi, dst, idx_start, npages, NFS_PAGE_TAG_COMMIT); } #else +static inline int nfs_need_commit(struct nfs_inode *nfsi) +{ + return 0; +} + static inline int nfs_scan_commit(struct inode *inode, struct list_head *dst, pgoff_t idx_start, unsigned int npages) { return 0; @@ -820,7 +827,7 @@ static int nfs_write_rpcsetup(struct nfs_page *req, data->args.stable = NFS_UNSTABLE; if (how & FLUSH_STABLE) { data->args.stable = NFS_DATA_SYNC; - if (!NFS_I(inode)->ncommit) + if (!nfs_need_commit(NFS_I(inode))) data->args.stable = NFS_FILE_SYNC; } diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index db867b04ac3c..c9fecd3e8f0f 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -166,8 +166,7 @@ struct nfs_inode { */ struct radix_tree_root nfs_page_tree; - unsigned long ncommit, - npages; + unsigned long npages; /* Open contexts for shared mmap writes */ struct list_head open_files; From 72cb77f4a5ace37b12dcb47a0e8637a2c28ad881 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 11 Mar 2009 14:10:30 -0400 Subject: [PATCH 13/58] NFS: Throttle page dirtying while we're flushing to disk The following patch is a combination of a patch by myself and Peter Staubach. Trond: If we allow other processes to dirty pages while a process is doing a consistency sync to disk, we can end up never making progress. Peter: Attached is a patch which addresses a continuing problem with the NFS client generating out of order WRITE requests. While this is compliant with all of the current protocol specifications, there are servers in the market which can not handle out of order WRITE requests very well. Also, this may lead to sub-optimal block allocations in the underlying file system on the server. This may cause the read throughputs to be reduced when reading the file from the server. Peter: There has been a lot of work recently done to address out of order issues on a systemic level. However, the NFS client is still susceptible to the problem. Out of order WRITE requests can occur when pdflush is in the middle of writing out pages while the process dirtying the pages calls generic_file_buffered_write which calls generic_perform_write which calls balance_dirty_pages_rate_limited which ends up calling writeback_inodes which ends up calling back into the NFS client to writes out dirty pages for the same file that pdflush happens to be working with. Signed-off-by: Peter Staubach [modification by Trond to merge the two similar patches] Signed-off-by: Trond Myklebust --- fs/nfs/file.c | 9 +++++++++ fs/nfs/inode.c | 12 ++++++++++++ fs/nfs/internal.h | 1 + fs/nfs/nfs4proc.c | 10 +--------- fs/nfs/pagelist.c | 11 ----------- fs/nfs/write.c | 28 +++++++++++++++++++--------- include/linux/nfs_fs.h | 1 + 7 files changed, 43 insertions(+), 29 deletions(-) diff --git a/fs/nfs/file.c b/fs/nfs/file.c index 90f292b520d2..404c19c866a7 100644 --- a/fs/nfs/file.c +++ b/fs/nfs/file.c @@ -354,6 +354,15 @@ static int nfs_write_begin(struct file *file, struct address_space *mapping, file->f_path.dentry->d_name.name, mapping->host->i_ino, len, (long long) pos); + /* + * Prevent starvation issues if someone is doing a consistency + * sync-to-disk + */ + ret = wait_on_bit(&NFS_I(mapping->host)->flags, NFS_INO_FLUSHING, + nfs_wait_bit_killable, TASK_KILLABLE); + if (ret) + return ret; + page = grab_cache_page_write_begin(mapping, index, flags); if (!page) return -ENOMEM; diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index 00f116cdadc6..c40adc5dd609 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -65,6 +65,18 @@ nfs_fattr_to_ino_t(struct nfs_fattr *fattr) return nfs_fileid_to_ino_t(fattr->fileid); } +/** + * nfs_wait_bit_killable - helper for functions that are sleeping on bit locks + * @word: long word containing the bit lock + */ +int nfs_wait_bit_killable(void *word) +{ + if (fatal_signal_pending(current)) + return -ERESTARTSYS; + schedule(); + return 0; +} + /** * nfs_compat_user_ino64 - returns the user-visible inode number * @fileid: 64-bit fileid diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h index 340ede8f608f..a55e69aa52e5 100644 --- a/fs/nfs/internal.h +++ b/fs/nfs/internal.h @@ -165,6 +165,7 @@ extern void nfs_clear_inode(struct inode *); extern void nfs4_clear_inode(struct inode *); #endif void nfs_zap_acl_cache(struct inode *inode); +extern int nfs_wait_bit_killable(void *word); /* super.c */ void nfs_parse_ip_address(char *, size_t, struct sockaddr *, size_t *); diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 101f5f4c304f..95f171e7e05a 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -193,14 +193,6 @@ static void nfs4_setup_readdir(u64 cookie, __be32 *verifier, struct dentry *dent kunmap_atomic(start, KM_USER0); } -static int nfs4_wait_bit_killable(void *word) -{ - if (fatal_signal_pending(current)) - return -ERESTARTSYS; - schedule(); - return 0; -} - static int nfs4_wait_clnt_recover(struct nfs_client *clp) { int res; @@ -208,7 +200,7 @@ static int nfs4_wait_clnt_recover(struct nfs_client *clp) might_sleep(); res = wait_on_bit(&clp->cl_state, NFS4CLNT_MANAGER_RUNNING, - nfs4_wait_bit_killable, TASK_KILLABLE); + nfs_wait_bit_killable, TASK_KILLABLE); return res; } diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c index 7f079209d70a..e2975939126a 100644 --- a/fs/nfs/pagelist.c +++ b/fs/nfs/pagelist.c @@ -176,17 +176,6 @@ void nfs_release_request(struct nfs_page *req) kref_put(&req->wb_kref, nfs_free_request); } -static int nfs_wait_bit_killable(void *word) -{ - int ret = 0; - - if (fatal_signal_pending(current)) - ret = -ERESTARTSYS; - else - schedule(); - return ret; -} - /** * nfs_wait_on_request - Wait for a request to complete. * @req: request to wait upon. diff --git a/fs/nfs/write.c b/fs/nfs/write.c index 1a999939fedf..36fd35e0de83 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -313,19 +313,34 @@ static int nfs_writepages_callback(struct page *page, struct writeback_control * int nfs_writepages(struct address_space *mapping, struct writeback_control *wbc) { struct inode *inode = mapping->host; + unsigned long *bitlock = &NFS_I(inode)->flags; struct nfs_pageio_descriptor pgio; int err; + /* Stop dirtying of new pages while we sync */ + err = wait_on_bit_lock(bitlock, NFS_INO_FLUSHING, + nfs_wait_bit_killable, TASK_KILLABLE); + if (err) + goto out_err; + nfs_inc_stats(inode, NFSIOS_VFSWRITEPAGES); nfs_pageio_init_write(&pgio, inode, wb_priority(wbc)); err = write_cache_pages(mapping, wbc, nfs_writepages_callback, &pgio); nfs_pageio_complete(&pgio); + + clear_bit_unlock(NFS_INO_FLUSHING, bitlock); + smp_mb__after_clear_bit(); + wake_up_bit(bitlock, NFS_INO_FLUSHING); + if (err < 0) - return err; - if (pgio.pg_error < 0) - return pgio.pg_error; + goto out_err; + err = pgio.pg_error; + if (err < 0) + goto out_err; return 0; +out_err: + return err; } /* @@ -1432,18 +1447,13 @@ static int nfs_write_mapping(struct address_space *mapping, int how) { struct writeback_control wbc = { .bdi = mapping->backing_dev_info, - .sync_mode = WB_SYNC_NONE, + .sync_mode = WB_SYNC_ALL, .nr_to_write = LONG_MAX, .range_start = 0, .range_end = LLONG_MAX, .for_writepages = 1, }; - int ret; - ret = __nfs_write_mapping(mapping, &wbc, how); - if (ret < 0) - return ret; - wbc.sync_mode = WB_SYNC_ALL; return __nfs_write_mapping(mapping, &wbc, how); } diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index c9fecd3e8f0f..933bc261c0df 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -206,6 +206,7 @@ struct nfs_inode { #define NFS_INO_STALE (1) /* possible stale inode */ #define NFS_INO_ACL_LRU_SET (2) /* Inode is on the LRU list */ #define NFS_INO_MOUNTPOINT (3) /* inode is remote mountpoint */ +#define NFS_INO_FLUSHING (4) /* inode is flushing out data */ static inline struct nfs_inode *NFS_I(const struct inode *inode) { From e1ebfd33be068ec933f8954060a499bd22ad6f69 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 11 Mar 2009 14:37:54 -0400 Subject: [PATCH 14/58] NFS: Kill the "defined but not used" compile error on nommu machines Bryan Wu reports that when compiling NFS on nommu machines he gets a "defined but not used" error on nfs_file_mmap(). The easiest fix is simply to get rid of the special casing in NFS, and just always call generic_file_mmap() to set up the file. Signed-off-by: Trond Myklebust --- fs/nfs/file.c | 12 +++++------- 1 file changed, 5 insertions(+), 7 deletions(-) diff --git a/fs/nfs/file.c b/fs/nfs/file.c index 404c19c866a7..1eab9c9ad242 100644 --- a/fs/nfs/file.c +++ b/fs/nfs/file.c @@ -64,11 +64,7 @@ const struct file_operations nfs_file_operations = { .write = do_sync_write, .aio_read = nfs_file_read, .aio_write = nfs_file_write, -#ifdef CONFIG_MMU .mmap = nfs_file_mmap, -#else - .mmap = generic_file_mmap, -#endif .open = nfs_file_open, .flush = nfs_file_flush, .release = nfs_file_release, @@ -304,11 +300,13 @@ nfs_file_mmap(struct file * file, struct vm_area_struct * vma) dprintk("NFS: mmap(%s/%s)\n", dentry->d_parent->d_name.name, dentry->d_name.name); - status = nfs_revalidate_mapping(inode, file->f_mapping); + /* Note: generic_file_mmap() returns ENOSYS on nommu systems + * so we call that before revalidating the mapping + */ + status = generic_file_mmap(file, vma); if (!status) { vma->vm_ops = &nfs_file_vm_ops; - vma->vm_flags |= VM_CAN_NONLINEAR; - file_accessed(file); + status = nfs_revalidate_mapping(inode, file->f_mapping); } return status; } From b1e1e158779f1d99c2cc18e466f6bf9099fc0853 Mon Sep 17 00:00:00 2001 From: Tom Talpey Date: Wed, 11 Mar 2009 14:37:55 -0400 Subject: [PATCH 15/58] SVCRDMA: remove faulty assertions in rpc/rdma chunk validation. Certain client-provided RPCRDMA chunk alignments result in an additional scatter/gather entry, which triggered nfs/rdma server assertions incorrectly. OpenSolaris nfs/rdma client connectathon testing was blocked by these in the special/locking section. Signed-off-by: Tom Talpey Cc: Tom Tucker Signed-off-by: Trond Myklebust --- net/sunrpc/xprtrdma/svc_rdma_sendto.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/net/sunrpc/xprtrdma/svc_rdma_sendto.c b/net/sunrpc/xprtrdma/svc_rdma_sendto.c index a3334e3b73cc..d0bea987d80e 100644 --- a/net/sunrpc/xprtrdma/svc_rdma_sendto.c +++ b/net/sunrpc/xprtrdma/svc_rdma_sendto.c @@ -191,7 +191,6 @@ static int map_xdr(struct svcxprt_rdma *xprt, struct xdr_buf *xdr, struct svc_rdma_req_map *vec) { - int sge_max = (xdr->len+PAGE_SIZE-1) / PAGE_SIZE + 3; int sge_no; u32 sge_bytes; u32 page_bytes; @@ -235,7 +234,11 @@ static int map_xdr(struct svcxprt_rdma *xprt, sge_no++; } - BUG_ON(sge_no > sge_max); + dprintk("svcrdma: map_xdr: sge_no %d page_no %d " + "page_base %zd page_len %zd head_len %d tail_len %d\n", + sge_no, page_no, xdr->page_base, xdr->page_len, + xdr->head[0].iov_len, xdr->tail[0].iov_len); + vec->count = sge_no; return 0; } @@ -579,7 +582,6 @@ static int send_reply(struct svcxprt_rdma *rdma, ctxt->sge[page_no+1].length = 0; } BUG_ON(sge_no > rdma->sc_max_sge); - BUG_ON(sge_no > ctxt->count); memset(&send_wr, 0, sizeof send_wr); ctxt->wr_op = IB_WR_SEND; send_wr.wr_id = (unsigned long)ctxt; From b38ab40ad58c1fc43ea590d6342f6a6763ac8fb6 Mon Sep 17 00:00:00 2001 From: Tom Talpey Date: Wed, 11 Mar 2009 14:37:55 -0400 Subject: [PATCH 16/58] XPRTRDMA: correct an rpc/rdma inline send marshaling error Certain client rpc's which contain both lengthy page-contained metadata and a non-empty xdr_tail buffer require careful handling to avoid overlapped memory copying. Rearranging of existing rpcrdma marshaling code avoids it; this fixes an NFSv4 symlink creation error detected with connectathon basic/test8 to multiple servers. Signed-off-by: Tom Talpey Signed-off-by: Trond Myklebust --- net/sunrpc/xprtrdma/rpc_rdma.c | 26 ++++++++++++++------------ 1 file changed, 14 insertions(+), 12 deletions(-) diff --git a/net/sunrpc/xprtrdma/rpc_rdma.c b/net/sunrpc/xprtrdma/rpc_rdma.c index 14106d26bb95..e5e28d1946a4 100644 --- a/net/sunrpc/xprtrdma/rpc_rdma.c +++ b/net/sunrpc/xprtrdma/rpc_rdma.c @@ -310,6 +310,19 @@ rpcrdma_inline_pullup(struct rpc_rqst *rqst, int pad) __func__, pad, destp, rqst->rq_slen, curlen); copy_len = rqst->rq_snd_buf.page_len; + + if (rqst->rq_snd_buf.tail[0].iov_len) { + curlen = rqst->rq_snd_buf.tail[0].iov_len; + if (destp + copy_len != rqst->rq_snd_buf.tail[0].iov_base) { + memmove(destp + copy_len, + rqst->rq_snd_buf.tail[0].iov_base, curlen); + r_xprt->rx_stats.pullup_copy_count += curlen; + } + dprintk("RPC: %s: tail destp 0x%p len %d\n", + __func__, destp + copy_len, curlen); + rqst->rq_svec[0].iov_len += curlen; + } + r_xprt->rx_stats.pullup_copy_count += copy_len; npages = PAGE_ALIGN(rqst->rq_snd_buf.page_base+copy_len) >> PAGE_SHIFT; for (i = 0; copy_len && i < npages; i++) { @@ -332,17 +345,6 @@ rpcrdma_inline_pullup(struct rpc_rqst *rqst, int pad) destp += curlen; copy_len -= curlen; } - if (rqst->rq_snd_buf.tail[0].iov_len) { - curlen = rqst->rq_snd_buf.tail[0].iov_len; - if (destp != rqst->rq_snd_buf.tail[0].iov_base) { - memcpy(destp, - rqst->rq_snd_buf.tail[0].iov_base, curlen); - r_xprt->rx_stats.pullup_copy_count += curlen; - } - dprintk("RPC: %s: tail destp 0x%p len %d curlen %d\n", - __func__, destp, copy_len, curlen); - rqst->rq_svec[0].iov_len += curlen; - } /* header now contains entire send message */ return pad; } @@ -656,7 +658,7 @@ rpcrdma_inline_fixup(struct rpc_rqst *rqst, char *srcp, int copy_len, int pad) if (curlen > rqst->rq_rcv_buf.tail[0].iov_len) curlen = rqst->rq_rcv_buf.tail[0].iov_len; if (rqst->rq_rcv_buf.tail[0].iov_base != srcp) - memcpy(rqst->rq_rcv_buf.tail[0].iov_base, srcp, curlen); + memmove(rqst->rq_rcv_buf.tail[0].iov_base, srcp, curlen); dprintk("RPC: %s: tail srcp 0x%p len %d curlen %d\n", __func__, srcp, copy_len, curlen); rqst->rq_rcv_buf.tail[0].iov_len = curlen; From 441e3e242903f9b190d5764bed73edb58f977413 Mon Sep 17 00:00:00 2001 From: Tom Talpey Date: Wed, 11 Mar 2009 14:37:56 -0400 Subject: [PATCH 17/58] SUNRPC: dynamically load RPC transport modules on-demand Provide an api to attempt to load any necessary kernel RPC client transport module automatically. By convention, the desired module name is "xprt"+"transport name". For example, when NFS mounting with "-o proto=rdma", attempt to load the "xprtrdma" module. Signed-off-by: Tom Talpey Cc: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xprt.h | 1 + net/sunrpc/xprt.c | 31 +++++++++++++++++++++++++++++++ 2 files changed, 32 insertions(+) diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index 11fc71d50c1e..2b0d960603b9 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -235,6 +235,7 @@ static inline __be32 *xprt_skip_transport_header(struct rpc_xprt *xprt, __be32 * */ int xprt_register_transport(struct xprt_class *type); int xprt_unregister_transport(struct xprt_class *type); +int xprt_load_transport(const char *); void xprt_set_retrans_timeout_def(struct rpc_task *task); void xprt_set_retrans_timeout_rtt(struct rpc_task *task); void xprt_wake_pending_tasks(struct rpc_xprt *xprt, int status); diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c index 62098d101a1f..d1afec640394 100644 --- a/net/sunrpc/xprt.c +++ b/net/sunrpc/xprt.c @@ -151,6 +151,37 @@ out: } EXPORT_SYMBOL_GPL(xprt_unregister_transport); +/** + * xprt_load_transport - load a transport implementation + * @transport_name: transport to load + * + * Returns: + * 0: transport successfully loaded + * -ENOENT: transport module not available + */ +int xprt_load_transport(const char *transport_name) +{ + struct xprt_class *t; + char module_name[sizeof t->name + 5]; + int result; + + result = 0; + spin_lock(&xprt_list_lock); + list_for_each_entry(t, &xprt_list, list) { + if (strcmp(t->name, transport_name) == 0) { + spin_unlock(&xprt_list_lock); + goto out; + } + } + spin_unlock(&xprt_list_lock); + strcpy(module_name, "xprt"); + strncat(module_name, transport_name, sizeof t->name); + result = request_module(module_name); +out: + return result; +} +EXPORT_SYMBOL_GPL(xprt_load_transport); + /** * xprt_reserve_xprt - serialize write access to transports * @task: task that is requesting access to the transport From a67d18f89f5782806135aad4ee012ff78d45aae7 Mon Sep 17 00:00:00 2001 From: Tom Talpey Date: Wed, 11 Mar 2009 14:37:56 -0400 Subject: [PATCH 18/58] NFS: load the rpc/rdma transport module automatically When mounting an NFS/RDMA server with the "-o proto=rdma" or "-o rdma" options, attempt to dynamically load the necessary "xprtrdma" client transport module. Doing so improves usability, while avoiding a static module dependency and any unnecesary resources. Signed-off-by: Tom Talpey Cc: Chuck Lever Signed-off-by: Trond Myklebust --- fs/nfs/super.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/fs/nfs/super.c b/fs/nfs/super.c index d6686f4786dc..0942fcbbad3c 100644 --- a/fs/nfs/super.c +++ b/fs/nfs/super.c @@ -1018,6 +1018,7 @@ static int nfs_parse_mount_options(char *raw, case Opt_rdma: mnt->flags |= NFS_MOUNT_TCP; /* for side protocols */ mnt->nfs_server.protocol = XPRT_TRANSPORT_RDMA; + xprt_load_transport(p); break; case Opt_acl: mnt->flags &= ~NFS_MOUNT_NOACL; @@ -1205,12 +1206,14 @@ static int nfs_parse_mount_options(char *raw, /* vector side protocols to TCP */ mnt->flags |= NFS_MOUNT_TCP; mnt->nfs_server.protocol = XPRT_TRANSPORT_RDMA; + xprt_load_transport(string); break; default: errors++; dfprintk(MOUNT, "NFS: unrecognized " "transport protocol\n"); } + kfree(string); break; case Opt_mountproto: string = match_strdup(args); @@ -1218,7 +1221,6 @@ static int nfs_parse_mount_options(char *raw, goto out_nomem; token = match_token(string, nfs_xprt_protocol_tokens, args); - kfree(string); switch (token) { case Opt_xprt_udp: From 15f081ca8ddfe150fb639c591b18944a539da0fc Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 11 Mar 2009 14:37:57 -0400 Subject: [PATCH 19/58] SUNRPC: Avoid an unnecessary task reschedule on ENOTCONN If the socket is unconnected, and xprt_transmit() returns ENOTCONN, we currently give up the lock on the transport channel. Doing so means that the lock automatically gets assigned to the next task in the xprt->sending queue, and so that task needs to be woken up to do the actual connect. The following patch aims to avoid that unnecessary task switch. Signed-off-by: Trond Myklebust --- net/sunrpc/clnt.c | 26 ++++++++++++++++++-------- 1 file changed, 18 insertions(+), 8 deletions(-) diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c index 836f15c0c4a3..07e9b05321e6 100644 --- a/net/sunrpc/clnt.c +++ b/net/sunrpc/clnt.c @@ -1105,14 +1105,24 @@ static void call_transmit_status(struct rpc_task *task) { task->tk_action = call_status; - /* - * Special case: if we've been waiting on the socket's write_space() - * callback, then don't call xprt_end_transmit(). - */ - if (task->tk_status == -EAGAIN) - return; - xprt_end_transmit(task); - rpc_task_force_reencode(task); + switch (task->tk_status) { + case -EAGAIN: + break; + default: + xprt_end_transmit(task); + /* + * Special cases: if we've been waiting on the + * socket's write_space() callback, or if the + * socket just returned a connection error, + * then hold onto the transport lock. + */ + case -ECONNREFUSED: + case -ENOTCONN: + case -EHOSTDOWN: + case -EHOSTUNREACH: + case -ENETUNREACH: + rpc_task_force_reencode(task); + } } /* From 670f94573104b4a25525d3fcdcd6496c678df172 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 11 Mar 2009 14:37:58 -0400 Subject: [PATCH 20/58] SUNRPC: Ensure we set XPRT_CLOSING only after we've sent a tcp FIN... ...so that we can distinguish between when we need to shutdown and when we don't. Also remove the call to xs_tcp_shutdown() from xs_tcp_connect(), since xprt_connect() makes the same test. Signed-off-by: Trond Myklebust --- net/sunrpc/xprtsock.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index 1127eb934136..cb4bd93b9211 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -1180,7 +1180,6 @@ static void xs_tcp_state_change(struct sock *sk) break; case TCP_CLOSE_WAIT: /* The server initiated a shutdown of the socket */ - set_bit(XPRT_CLOSING, &xprt->state); xprt_force_disconnect(xprt); case TCP_SYN_SENT: xprt->connect_cookie++; @@ -1193,6 +1192,7 @@ static void xs_tcp_state_change(struct sock *sk) xprt->reestablish_timeout = XS_TCP_INIT_REEST_TO; break; case TCP_LAST_ACK: + set_bit(XPRT_CLOSING, &xprt->state); smp_mb__before_clear_bit(); clear_bit(XPRT_CONNECTED, &xprt->state); smp_mb__after_clear_bit(); @@ -1836,9 +1836,6 @@ static void xs_tcp_connect(struct rpc_task *task) { struct rpc_xprt *xprt = task->tk_xprt; - /* Initiate graceful shutdown of the socket if not already done */ - if (test_bit(XPRT_CONNECTED, &xprt->state)) - xs_tcp_shutdown(xprt); /* Exit if we need to wait for socket shutdown to complete */ if (test_bit(XPRT_CLOSING, &xprt->state)) return; From 40d2549db5f515e415894def98b49db7d4c56714 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 11 Mar 2009 14:37:58 -0400 Subject: [PATCH 21/58] SUNRPC: Don't disconnect if a connection is still in progress. Signed-off-by: Trond Myklebust --- net/sunrpc/xprtsock.c | 18 ++++++++++++++---- 1 file changed, 14 insertions(+), 4 deletions(-) diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index cb4bd93b9211..9d1898f6ee87 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -1613,10 +1613,9 @@ out: * We need to preserve the port number so the reply cache on the server can * find our cached RPC replies when we get around to reconnecting. */ -static void xs_tcp_reuse_connection(struct rpc_xprt *xprt) +static void xs_abort_connection(struct rpc_xprt *xprt, struct sock_xprt *transport) { int result; - struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt); struct sockaddr any; dprintk("RPC: disconnecting xprt %p to reuse port\n", xprt); @@ -1633,6 +1632,17 @@ static void xs_tcp_reuse_connection(struct rpc_xprt *xprt) result); } +static void xs_tcp_reuse_connection(struct rpc_xprt *xprt, struct sock_xprt *transport) +{ + unsigned int state = transport->inet->sk_state; + + if (state == TCP_CLOSE && transport->sock->state == SS_UNCONNECTED) + return; + if ((1 << state) & (TCPF_ESTABLISHED|TCPF_SYN_SENT)) + return; + xs_abort_connection(xprt, transport); +} + static int xs_tcp_finish_connecting(struct rpc_xprt *xprt, struct socket *sock) { struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt); @@ -1706,7 +1716,7 @@ static void xs_tcp_connect_worker4(struct work_struct *work) } } else /* "close" the socket, preserving the local port */ - xs_tcp_reuse_connection(xprt); + xs_tcp_reuse_connection(xprt, transport); dprintk("RPC: worker connecting xprt %p to address: %s\n", xprt, xprt->address_strings[RPC_DISPLAY_ALL]); @@ -1766,7 +1776,7 @@ static void xs_tcp_connect_worker6(struct work_struct *work) } } else /* "close" the socket, preserving the local port */ - xs_tcp_reuse_connection(xprt); + xs_tcp_reuse_connection(xprt, transport); dprintk("RPC: worker connecting xprt %p to address: %s\n", xprt, xprt->address_strings[RPC_DISPLAY_ALL]); From c8485e4d634f6df155040293928707f127f0d06d Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 11 Mar 2009 14:37:59 -0400 Subject: [PATCH 22/58] SUNRPC: Handle ECONNREFUSED correctly in xprt_transmit() If we get an ECONNREFUSED error, we currently go to sleep on the 'xprt->sending' wait queue. The problem is that no timeout is set there, and there is nothing else that will wake the task up later. We should deal with ECONNREFUSED in call_status, given that is where we also deal with -EHOSTDOWN, and friends. Signed-off-by: Trond Myklebust --- net/sunrpc/clnt.c | 7 ++++++- net/sunrpc/xprt.c | 40 +++++++++++++++++----------------------- net/sunrpc/xprtsock.c | 26 ++++++++++++-------------- 3 files changed, 35 insertions(+), 38 deletions(-) diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c index 07e9b05321e6..145715b53115 100644 --- a/net/sunrpc/clnt.c +++ b/net/sunrpc/clnt.c @@ -1117,10 +1117,12 @@ call_transmit_status(struct rpc_task *task) * then hold onto the transport lock. */ case -ECONNREFUSED: + case -ECONNRESET: case -ENOTCONN: case -EHOSTDOWN: case -EHOSTUNREACH: case -ENETUNREACH: + case -EPIPE: rpc_task_force_reencode(task); } } @@ -1162,9 +1164,12 @@ call_status(struct rpc_task *task) xprt_conditional_disconnect(task->tk_xprt, req->rq_connect_cookie); break; + case -ECONNRESET: case -ECONNREFUSED: - case -ENOTCONN: rpc_force_rebind(clnt); + rpc_delay(task, 3*HZ); + case -EPIPE: + case -ENOTCONN: task->tk_action = call_bind; break; case -EAGAIN: diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c index d1afec640394..d588e755e107 100644 --- a/net/sunrpc/xprt.c +++ b/net/sunrpc/xprt.c @@ -901,32 +901,26 @@ void xprt_transmit(struct rpc_task *task) req->rq_connect_cookie = xprt->connect_cookie; req->rq_xtime = jiffies; status = xprt->ops->send_request(task); - if (status == 0) { - dprintk("RPC: %5u xmit complete\n", task->tk_pid); - spin_lock_bh(&xprt->transport_lock); - - xprt->ops->set_retrans_timeout(task); - - xprt->stat.sends++; - xprt->stat.req_u += xprt->stat.sends - xprt->stat.recvs; - xprt->stat.bklog_u += xprt->backlog.qlen; - - /* Don't race with disconnect */ - if (!xprt_connected(xprt)) - task->tk_status = -ENOTCONN; - else if (!req->rq_received) - rpc_sleep_on(&xprt->pending, task, xprt_timer); - spin_unlock_bh(&xprt->transport_lock); + if (status != 0) { + task->tk_status = status; return; } - /* Note: at this point, task->tk_sleeping has not yet been set, - * hence there is no danger of the waking up task being put on - * schedq, and being picked up by a parallel run of rpciod(). - */ - task->tk_status = status; - if (status == -ECONNREFUSED) - rpc_sleep_on(&xprt->sending, task, NULL); + dprintk("RPC: %5u xmit complete\n", task->tk_pid); + spin_lock_bh(&xprt->transport_lock); + + xprt->ops->set_retrans_timeout(task); + + xprt->stat.sends++; + xprt->stat.req_u += xprt->stat.sends - xprt->stat.recvs; + xprt->stat.bklog_u += xprt->backlog.qlen; + + /* Don't race with disconnect */ + if (!xprt_connected(xprt)) + task->tk_status = -ENOTCONN; + else if (!req->rq_received) + rpc_sleep_on(&xprt->pending, task, xprt_timer); + spin_unlock_bh(&xprt->transport_lock); } static inline void do_xprt_reserve(struct rpc_task *task) diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index 9d1898f6ee87..5e8198bede81 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -594,6 +594,8 @@ static int xs_udp_send_request(struct rpc_task *task) /* Still some bytes left; set up for a retry later. */ status = -EAGAIN; } + if (!transport->sock) + goto out; switch (status) { case -ENOTSOCK: @@ -603,19 +605,17 @@ static int xs_udp_send_request(struct rpc_task *task) case -EAGAIN: xs_nospace(task); break; + default: + dprintk("RPC: sendmsg returned unrecognized error %d\n", + -status); case -ENETUNREACH: case -EPIPE: case -ECONNREFUSED: /* When the server has died, an ICMP port unreachable message * prompts ECONNREFUSED. */ clear_bit(SOCK_ASYNC_NOSPACE, &transport->sock->flags); - break; - default: - clear_bit(SOCK_ASYNC_NOSPACE, &transport->sock->flags); - dprintk("RPC: sendmsg returned unrecognized error %d\n", - -status); } - +out: return status; } @@ -697,6 +697,8 @@ static int xs_tcp_send_request(struct rpc_task *task) status = -EAGAIN; break; } + if (!transport->sock) + goto out; switch (status) { case -ENOTSOCK: @@ -706,21 +708,17 @@ static int xs_tcp_send_request(struct rpc_task *task) case -EAGAIN: xs_nospace(task); break; + default: + dprintk("RPC: sendmsg returned unrecognized error %d\n", + -status); case -ECONNRESET: xs_tcp_shutdown(xprt); case -ECONNREFUSED: case -ENOTCONN: case -EPIPE: - status = -ENOTCONN; clear_bit(SOCK_ASYNC_NOSPACE, &transport->sock->flags); - break; - default: - dprintk("RPC: sendmsg returned unrecognized error %d\n", - -status); - clear_bit(SOCK_ASYNC_NOSPACE, &transport->sock->flags); - xs_tcp_shutdown(xprt); } - +out: return status; } From 482f32e65d31cbf88d08306fa5d397cc945c3c26 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 11 Mar 2009 14:38:00 -0400 Subject: [PATCH 23/58] SUNRPC: Handle socket errors correctly Ensure that we pick up and handle socket errors as they occur. Signed-off-by: Trond Myklebust --- net/sunrpc/xprtsock.c | 12 +++++------- 1 file changed, 5 insertions(+), 7 deletions(-) diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index 5e8198bede81..879af6f27b4c 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -1208,23 +1208,20 @@ static void xs_tcp_state_change(struct sock *sk) } /** - * xs_tcp_error_report - callback mainly for catching RST events + * xs_error_report - callback mainly for catching socket errors * @sk: socket */ -static void xs_tcp_error_report(struct sock *sk) +static void xs_error_report(struct sock *sk) { struct rpc_xprt *xprt; read_lock(&sk->sk_callback_lock); - if (sk->sk_err != ECONNRESET || sk->sk_state != TCP_ESTABLISHED) - goto out; if (!(xprt = xprt_from_sock(sk))) goto out; dprintk("RPC: %s client %p...\n" "RPC: error %d\n", __func__, xprt, sk->sk_err); - - xprt_force_disconnect(xprt); + xprt_wake_pending_tasks(xprt, -EAGAIN); out: read_unlock(&sk->sk_callback_lock); } @@ -1509,6 +1506,7 @@ static void xs_udp_finish_connecting(struct rpc_xprt *xprt, struct socket *sock) sk->sk_user_data = xprt; sk->sk_data_ready = xs_udp_data_ready; sk->sk_write_space = xs_udp_write_space; + sk->sk_error_report = xs_error_report; sk->sk_no_check = UDP_CSUM_NORCV; sk->sk_allocation = GFP_ATOMIC; @@ -1656,7 +1654,7 @@ static int xs_tcp_finish_connecting(struct rpc_xprt *xprt, struct socket *sock) sk->sk_data_ready = xs_tcp_data_ready; sk->sk_state_change = xs_tcp_state_change; sk->sk_write_space = xs_tcp_write_space; - sk->sk_error_report = xs_tcp_error_report; + sk->sk_error_report = xs_error_report; sk->sk_allocation = GFP_ATOMIC; /* socket options */ From 2a4919919a97911b0aa4b9f5ac1eab90ba87652b Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 11 Mar 2009 14:38:00 -0400 Subject: [PATCH 24/58] SUNRPC: Return EAGAIN instead of ENOTCONN when waking up xprt->pending While we should definitely return socket errors to the task that is currently trying to send data, there is no need to propagate the same error to all the other tasks on xprt->pending. Doing so actually slows down recovery, since it causes more than one tasks to attempt socket recovery. Signed-off-by: Trond Myklebust --- net/sunrpc/clnt.c | 15 +++-------- net/sunrpc/xprt.c | 20 +++++---------- net/sunrpc/xprtsock.c | 58 +++++++++++++++++++++++-------------------- 3 files changed, 41 insertions(+), 52 deletions(-) diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c index 145715b53115..5abab094441f 100644 --- a/net/sunrpc/clnt.c +++ b/net/sunrpc/clnt.c @@ -1032,27 +1032,20 @@ call_connect_status(struct rpc_task *task) dprint_status(task); task->tk_status = 0; - if (status >= 0) { + if (status >= 0 || status == -EAGAIN) { clnt->cl_stats->netreconn++; task->tk_action = call_transmit; return; } - /* Something failed: remote service port may have changed */ - rpc_force_rebind(clnt); - switch (status) { - case -ENOTCONN: - case -EAGAIN: - task->tk_action = call_bind; - if (!RPC_IS_SOFT(task)) - return; /* if soft mounted, test if we've timed out */ case -ETIMEDOUT: task->tk_action = call_timeout; - return; + break; + default: + rpc_exit(task, -EIO); } - rpc_exit(task, -EIO); } /* diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c index d588e755e107..a0bfe53f1621 100644 --- a/net/sunrpc/xprt.c +++ b/net/sunrpc/xprt.c @@ -611,7 +611,7 @@ void xprt_disconnect_done(struct rpc_xprt *xprt) dprintk("RPC: disconnected transport %p\n", xprt); spin_lock_bh(&xprt->transport_lock); xprt_clear_connected(xprt); - xprt_wake_pending_tasks(xprt, -ENOTCONN); + xprt_wake_pending_tasks(xprt, -EAGAIN); spin_unlock_bh(&xprt->transport_lock); } EXPORT_SYMBOL_GPL(xprt_disconnect_done); @@ -629,7 +629,7 @@ void xprt_force_disconnect(struct rpc_xprt *xprt) /* Try to schedule an autoclose RPC call */ if (test_and_set_bit(XPRT_LOCKED, &xprt->state) == 0) queue_work(rpciod_workqueue, &xprt->task_cleanup); - xprt_wake_pending_tasks(xprt, -ENOTCONN); + xprt_wake_pending_tasks(xprt, -EAGAIN); spin_unlock_bh(&xprt->transport_lock); } @@ -656,7 +656,7 @@ void xprt_conditional_disconnect(struct rpc_xprt *xprt, unsigned int cookie) /* Try to schedule an autoclose RPC call */ if (test_and_set_bit(XPRT_LOCKED, &xprt->state) == 0) queue_work(rpciod_workqueue, &xprt->task_cleanup); - xprt_wake_pending_tasks(xprt, -ENOTCONN); + xprt_wake_pending_tasks(xprt, -EAGAIN); out: spin_unlock_bh(&xprt->transport_lock); } @@ -726,9 +726,8 @@ static void xprt_connect_status(struct rpc_task *task) } switch (task->tk_status) { - case -ENOTCONN: - dprintk("RPC: %5u xprt_connect_status: connection broken\n", - task->tk_pid); + case -EAGAIN: + dprintk("RPC: %5u xprt_connect_status: retrying\n", task->tk_pid); break; case -ETIMEDOUT: dprintk("RPC: %5u xprt_connect_status: connect attempt timed " @@ -849,15 +848,8 @@ int xprt_prepare_transmit(struct rpc_task *task) err = req->rq_received; goto out_unlock; } - if (!xprt->ops->reserve_xprt(task)) { + if (!xprt->ops->reserve_xprt(task)) err = -EAGAIN; - goto out_unlock; - } - - if (!xprt_connected(xprt)) { - err = -ENOTCONN; - goto out_unlock; - } out_unlock: spin_unlock_bh(&xprt->transport_lock); return err; diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index 879af6f27b4c..8e58b0b5460b 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -1162,7 +1162,7 @@ static void xs_tcp_state_change(struct sock *sk) transport->tcp_flags = TCP_RCV_COPY_FRAGHDR | TCP_RCV_COPY_XID; - xprt_wake_pending_tasks(xprt, 0); + xprt_wake_pending_tasks(xprt, -EAGAIN); } spin_unlock_bh(&xprt->transport_lock); break; @@ -1721,20 +1721,22 @@ static void xs_tcp_connect_worker4(struct work_struct *work) dprintk("RPC: %p connect status %d connected %d sock state %d\n", xprt, -status, xprt_connected(xprt), sock->sk->sk_state); - if (status < 0) { - switch (status) { - case -EINPROGRESS: - case -EALREADY: - goto out_clear; - case -ECONNREFUSED: - case -ECONNRESET: - /* retry with existing socket, after a delay */ - break; - default: - /* get rid of existing socket, and retry */ - xs_tcp_shutdown(xprt); - } + switch (status) { + case 0: + case -EINPROGRESS: + case -EALREADY: + goto out_clear; + case -ECONNREFUSED: + case -ECONNRESET: + /* retry with existing socket, after a delay */ + break; + default: + /* get rid of existing socket, and retry */ + xs_tcp_shutdown(xprt); + printk("%s: connect returned unhandled error %d\n", + __func__, status); } + status = -EAGAIN; out: xprt_wake_pending_tasks(xprt, status); out_clear: @@ -1780,20 +1782,22 @@ static void xs_tcp_connect_worker6(struct work_struct *work) status = xs_tcp_finish_connecting(xprt, sock); dprintk("RPC: %p connect status %d connected %d sock state %d\n", xprt, -status, xprt_connected(xprt), sock->sk->sk_state); - if (status < 0) { - switch (status) { - case -EINPROGRESS: - case -EALREADY: - goto out_clear; - case -ECONNREFUSED: - case -ECONNRESET: - /* retry with existing socket, after a delay */ - break; - default: - /* get rid of existing socket, and retry */ - xs_tcp_shutdown(xprt); - } + switch (status) { + case 0: + case -EINPROGRESS: + case -EALREADY: + goto out_clear; + case -ECONNREFUSED: + case -ECONNRESET: + /* retry with existing socket, after a delay */ + break; + default: + /* get rid of existing socket, and retry */ + xs_tcp_shutdown(xprt); + printk("%s: connect returned unhandled error %d\n", + __func__, status); } + status = -EAGAIN; out: xprt_wake_pending_tasks(xprt, status); out_clear: From 8a2cec295f4499cc9d4452e9b02d4ed071bb42d3 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 11 Mar 2009 14:38:01 -0400 Subject: [PATCH 25/58] SUNRPC: Delay, then retry on connection errors. Enforce the comment in xs_tcp_connect_worker4/xs_tcp_connect_worker6 that we should delay, then retry on certain connection errors. Signed-off-by: Trond Myklebust --- net/sunrpc/xprtsock.c | 34 ++++++++++++++++------------------ 1 file changed, 16 insertions(+), 18 deletions(-) diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index 8e58b0b5460b..9f3e615d3e09 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -1722,20 +1722,19 @@ static void xs_tcp_connect_worker4(struct work_struct *work) xprt, -status, xprt_connected(xprt), sock->sk->sk_state); switch (status) { + case -ECONNREFUSED: + case -ECONNRESET: + case -ENETUNREACH: + /* retry with existing socket, after a delay */ case 0: case -EINPROGRESS: case -EALREADY: goto out_clear; - case -ECONNREFUSED: - case -ECONNRESET: - /* retry with existing socket, after a delay */ - break; - default: - /* get rid of existing socket, and retry */ - xs_tcp_shutdown(xprt); - printk("%s: connect returned unhandled error %d\n", - __func__, status); } + /* get rid of existing socket, and retry */ + xs_tcp_shutdown(xprt); + printk("%s: connect returned unhandled error %d\n", + __func__, status); status = -EAGAIN; out: xprt_wake_pending_tasks(xprt, status); @@ -1783,20 +1782,19 @@ static void xs_tcp_connect_worker6(struct work_struct *work) dprintk("RPC: %p connect status %d connected %d sock state %d\n", xprt, -status, xprt_connected(xprt), sock->sk->sk_state); switch (status) { + case -ECONNREFUSED: + case -ECONNRESET: + case -ENETUNREACH: + /* retry with existing socket, after a delay */ case 0: case -EINPROGRESS: case -EALREADY: goto out_clear; - case -ECONNREFUSED: - case -ECONNRESET: - /* retry with existing socket, after a delay */ - break; - default: - /* get rid of existing socket, and retry */ - xs_tcp_shutdown(xprt); - printk("%s: connect returned unhandled error %d\n", - __func__, status); } + /* get rid of existing socket, and retry */ + xs_tcp_shutdown(xprt); + printk("%s: connect returned unhandled error %d\n", + __func__, status); status = -EAGAIN; out: xprt_wake_pending_tasks(xprt, status); From 5e3771ce2d6a69e10fcc870cdf226d121d868491 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 11 Mar 2009 14:38:01 -0400 Subject: [PATCH 26/58] SUNRPC: Ensure that xs_nospace return values are propagated If xs_nospace() finds that the socket has disconnected, it attempts to return ENOTCONN, however that value is then squashed by the callers. Signed-off-by: Trond Myklebust --- net/sunrpc/xprtsock.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index 9f3e615d3e09..2e070679ab4a 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -521,11 +521,12 @@ static void xs_nospace_callback(struct rpc_task *task) * @task: task to put to sleep * */ -static void xs_nospace(struct rpc_task *task) +static int xs_nospace(struct rpc_task *task) { struct rpc_rqst *req = task->tk_rqstp; struct rpc_xprt *xprt = req->rq_xprt; struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt); + int ret = 0; dprintk("RPC: %5u xmit incomplete (%u left of %u)\n", task->tk_pid, req->rq_slen - req->rq_bytes_sent, @@ -537,6 +538,7 @@ static void xs_nospace(struct rpc_task *task) /* Don't race with disconnect */ if (xprt_connected(xprt)) { if (test_bit(SOCK_ASYNC_NOSPACE, &transport->sock->flags)) { + ret = -EAGAIN; /* * Notify TCP that we're limited by the application * window size @@ -548,10 +550,11 @@ static void xs_nospace(struct rpc_task *task) } } else { clear_bit(SOCK_ASYNC_NOSPACE, &transport->sock->flags); - task->tk_status = -ENOTCONN; + ret = -ENOTCONN; } spin_unlock_bh(&xprt->transport_lock); + return ret; } /** @@ -603,7 +606,7 @@ static int xs_udp_send_request(struct rpc_task *task) /* Should we call xs_close() here? */ break; case -EAGAIN: - xs_nospace(task); + status = xs_nospace(task); break; default: dprintk("RPC: sendmsg returned unrecognized error %d\n", @@ -706,7 +709,7 @@ static int xs_tcp_send_request(struct rpc_task *task) /* Should we call xs_close() here? */ break; case -EAGAIN: - xs_nospace(task); + status = xs_nospace(task); break; default: dprintk("RPC: sendmsg returned unrecognized error %d\n", From 7d1e8255cf959fba7ee2317550dfde39f0b936ae Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 11 Mar 2009 14:38:03 -0400 Subject: [PATCH 27/58] SUNRPC: Add the equivalent of the linger and linger2 timeouts to RPC sockets This fixes a regression against FreeBSD servers as reported by Tomas Kasparek. Apparently when using RPC over a TCP socket, the FreeBSD servers don't ever react to the client closing the socket, and so commit e06799f958bf7f9f8fae15f0c6f519953fb0257c (SUNRPC: Use shutdown() instead of close() when disconnecting a TCP socket) causes the setup to hang forever whenever the client attempts to close and then reconnect. We break the deadlock by adding a 'linger2' style timeout to the socket, after which, the client will abort the connection using a TCP 'RST'. The default timeout is set to 15 seconds. A subsequent patch will put it under user control by means of a systctl. Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xprt.h | 1 + net/sunrpc/xprtsock.c | 98 ++++++++++++++++++++++++++++++------- 2 files changed, 82 insertions(+), 17 deletions(-) diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index 2b0d960603b9..1758d9f5b5c3 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -260,6 +260,7 @@ void xprt_conditional_disconnect(struct rpc_xprt *xprt, unsigned int cookie); #define XPRT_BOUND (4) #define XPRT_BINDING (5) #define XPRT_CLOSING (6) +#define XPRT_CONNECTION_ABORT (7) static inline void xprt_set_connected(struct rpc_xprt *xprt) { diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index 2e070679ab4a..b51f58b95c39 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -49,6 +49,8 @@ unsigned int xprt_tcp_slot_table_entries = RPC_DEF_SLOT_TABLE; unsigned int xprt_min_resvport = RPC_DEF_MIN_RESVPORT; unsigned int xprt_max_resvport = RPC_DEF_MAX_RESVPORT; +#define XS_TCP_LINGER_TO (15U * HZ) + /* * We can register our own files under /proc/sys/sunrpc by * calling register_sysctl_table() again. The files in that @@ -806,6 +808,7 @@ static void xs_close(struct rpc_xprt *xprt) xs_reset_transport(transport); smp_mb__before_clear_bit(); + clear_bit(XPRT_CONNECTION_ABORT, &xprt->state); clear_bit(XPRT_CLOSE_WAIT, &xprt->state); clear_bit(XPRT_CLOSING, &xprt->state); smp_mb__after_clear_bit(); @@ -1133,6 +1136,47 @@ out: read_unlock(&sk->sk_callback_lock); } +/* + * Do the equivalent of linger/linger2 handling for dealing with + * broken servers that don't close the socket in a timely + * fashion + */ +static void xs_tcp_schedule_linger_timeout(struct rpc_xprt *xprt, + unsigned long timeout) +{ + struct sock_xprt *transport; + + if (xprt_test_and_set_connecting(xprt)) + return; + set_bit(XPRT_CONNECTION_ABORT, &xprt->state); + transport = container_of(xprt, struct sock_xprt, xprt); + queue_delayed_work(rpciod_workqueue, &transport->connect_worker, + timeout); +} + +static void xs_tcp_cancel_linger_timeout(struct rpc_xprt *xprt) +{ + struct sock_xprt *transport; + + transport = container_of(xprt, struct sock_xprt, xprt); + + if (!test_bit(XPRT_CONNECTION_ABORT, &xprt->state) || + !cancel_delayed_work(&transport->connect_worker)) + return; + clear_bit(XPRT_CONNECTION_ABORT, &xprt->state); + xprt_clear_connecting(xprt); +} + +static void xs_sock_mark_closed(struct rpc_xprt *xprt) +{ + smp_mb__before_clear_bit(); + clear_bit(XPRT_CLOSE_WAIT, &xprt->state); + clear_bit(XPRT_CLOSING, &xprt->state); + smp_mb__after_clear_bit(); + /* Mark transport as closed and wake up all pending tasks */ + xprt_disconnect_done(xprt); +} + /** * xs_tcp_state_change - callback to handle TCP socket state changes * @sk: socket whose state has changed @@ -1178,6 +1222,7 @@ static void xs_tcp_state_change(struct sock *sk) clear_bit(XPRT_CONNECTED, &xprt->state); clear_bit(XPRT_CLOSE_WAIT, &xprt->state); smp_mb__after_clear_bit(); + xs_tcp_schedule_linger_timeout(xprt, XS_TCP_LINGER_TO); break; case TCP_CLOSE_WAIT: /* The server initiated a shutdown of the socket */ @@ -1194,17 +1239,14 @@ static void xs_tcp_state_change(struct sock *sk) break; case TCP_LAST_ACK: set_bit(XPRT_CLOSING, &xprt->state); + xs_tcp_schedule_linger_timeout(xprt, XS_TCP_LINGER_TO); smp_mb__before_clear_bit(); clear_bit(XPRT_CONNECTED, &xprt->state); smp_mb__after_clear_bit(); break; case TCP_CLOSE: - smp_mb__before_clear_bit(); - clear_bit(XPRT_CLOSE_WAIT, &xprt->state); - clear_bit(XPRT_CLOSING, &xprt->state); - smp_mb__after_clear_bit(); - /* Mark transport as closed and wake up all pending tasks */ - xprt_disconnect_done(xprt); + xs_tcp_cancel_linger_timeout(xprt); + xs_sock_mark_closed(xprt); } out: read_unlock(&sk->sk_callback_lock); @@ -1562,8 +1604,8 @@ static void xs_udp_connect_worker4(struct work_struct *work) xs_udp_finish_connecting(xprt, sock); status = 0; out: - xprt_wake_pending_tasks(xprt, status); xprt_clear_connecting(xprt); + xprt_wake_pending_tasks(xprt, status); } /** @@ -1604,8 +1646,8 @@ static void xs_udp_connect_worker6(struct work_struct *work) xs_udp_finish_connecting(xprt, sock); status = 0; out: - xprt_wake_pending_tasks(xprt, status); xprt_clear_connecting(xprt); + xprt_wake_pending_tasks(xprt, status); } /* @@ -1626,7 +1668,9 @@ static void xs_abort_connection(struct rpc_xprt *xprt, struct sock_xprt *transpo memset(&any, 0, sizeof(any)); any.sa_family = AF_UNSPEC; result = kernel_connect(transport->sock, &any, sizeof(any), 0); - if (result) + if (!result) + xs_sock_mark_closed(xprt); + else dprintk("RPC: AF_UNSPEC connect return code %d\n", result); } @@ -1702,6 +1746,7 @@ static void xs_tcp_connect_worker4(struct work_struct *work) goto out; if (!sock) { + clear_bit(XPRT_CONNECTION_ABORT, &xprt->state); /* start from scratch */ if ((err = sock_create_kern(PF_INET, SOCK_STREAM, IPPROTO_TCP, &sock)) < 0) { dprintk("RPC: can't create TCP transport socket (%d).\n", -err); @@ -1713,10 +1758,18 @@ static void xs_tcp_connect_worker4(struct work_struct *work) sock_release(sock); goto out; } - } else + } else { + int abort_and_exit; + + abort_and_exit = test_and_clear_bit(XPRT_CONNECTION_ABORT, + &xprt->state); /* "close" the socket, preserving the local port */ xs_tcp_reuse_connection(xprt, transport); + if (abort_and_exit) + goto out_eagain; + } + dprintk("RPC: worker connecting xprt %p to address: %s\n", xprt, xprt->address_strings[RPC_DISPLAY_ALL]); @@ -1732,17 +1785,18 @@ static void xs_tcp_connect_worker4(struct work_struct *work) case 0: case -EINPROGRESS: case -EALREADY: - goto out_clear; + xprt_clear_connecting(xprt); + return; } /* get rid of existing socket, and retry */ xs_tcp_shutdown(xprt); printk("%s: connect returned unhandled error %d\n", __func__, status); +out_eagain: status = -EAGAIN; out: - xprt_wake_pending_tasks(xprt, status); -out_clear: xprt_clear_connecting(xprt); + xprt_wake_pending_tasks(xprt, status); } /** @@ -1763,6 +1817,7 @@ static void xs_tcp_connect_worker6(struct work_struct *work) goto out; if (!sock) { + clear_bit(XPRT_CONNECTION_ABORT, &xprt->state); /* start from scratch */ if ((err = sock_create_kern(PF_INET6, SOCK_STREAM, IPPROTO_TCP, &sock)) < 0) { dprintk("RPC: can't create TCP transport socket (%d).\n", -err); @@ -1774,10 +1829,18 @@ static void xs_tcp_connect_worker6(struct work_struct *work) sock_release(sock); goto out; } - } else + } else { + int abort_and_exit; + + abort_and_exit = test_and_clear_bit(XPRT_CONNECTION_ABORT, + &xprt->state); /* "close" the socket, preserving the local port */ xs_tcp_reuse_connection(xprt, transport); + if (abort_and_exit) + goto out_eagain; + } + dprintk("RPC: worker connecting xprt %p to address: %s\n", xprt, xprt->address_strings[RPC_DISPLAY_ALL]); @@ -1792,17 +1855,18 @@ static void xs_tcp_connect_worker6(struct work_struct *work) case 0: case -EINPROGRESS: case -EALREADY: - goto out_clear; + xprt_clear_connecting(xprt); + return; } /* get rid of existing socket, and retry */ xs_tcp_shutdown(xprt); printk("%s: connect returned unhandled error %d\n", __func__, status); +out_eagain: status = -EAGAIN; out: - xprt_wake_pending_tasks(xprt, status); -out_clear: xprt_clear_connecting(xprt); + xprt_wake_pending_tasks(xprt, status); } /** From 25fe6142a57c720452c5e9ddbc1f32309c1e5c19 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 11 Mar 2009 14:38:03 -0400 Subject: [PATCH 28/58] SUNRPC: Add a sysctl to control the duration of the socket linger timeout Signed-off-by: Trond Myklebust --- net/sunrpc/xprtsock.c | 13 +++++++++++-- 1 file changed, 11 insertions(+), 2 deletions(-) diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index b51f58b95c39..42222b4dd76d 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -50,6 +50,7 @@ unsigned int xprt_min_resvport = RPC_DEF_MIN_RESVPORT; unsigned int xprt_max_resvport = RPC_DEF_MAX_RESVPORT; #define XS_TCP_LINGER_TO (15U * HZ) +static unsigned int xs_tcp_fin_timeout __read_mostly = XS_TCP_LINGER_TO; /* * We can register our own files under /proc/sys/sunrpc by @@ -118,6 +119,14 @@ static ctl_table xs_tunables_table[] = { .extra1 = &xprt_min_resvport_limit, .extra2 = &xprt_max_resvport_limit }, + { + .procname = "tcp_fin_timeout", + .data = &xs_tcp_fin_timeout, + .maxlen = sizeof(xs_tcp_fin_timeout), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + .strategy = sysctl_jiffies + }, { .ctl_name = 0, }, @@ -1222,7 +1231,7 @@ static void xs_tcp_state_change(struct sock *sk) clear_bit(XPRT_CONNECTED, &xprt->state); clear_bit(XPRT_CLOSE_WAIT, &xprt->state); smp_mb__after_clear_bit(); - xs_tcp_schedule_linger_timeout(xprt, XS_TCP_LINGER_TO); + xs_tcp_schedule_linger_timeout(xprt, xs_tcp_fin_timeout); break; case TCP_CLOSE_WAIT: /* The server initiated a shutdown of the socket */ @@ -1239,7 +1248,7 @@ static void xs_tcp_state_change(struct sock *sk) break; case TCP_LAST_ACK: set_bit(XPRT_CLOSING, &xprt->state); - xs_tcp_schedule_linger_timeout(xprt, XS_TCP_LINGER_TO); + xs_tcp_schedule_linger_timeout(xprt, xs_tcp_fin_timeout); smp_mb__before_clear_bit(); clear_bit(XPRT_CONNECTED, &xprt->state); smp_mb__after_clear_bit(); From b61d59fffd3e5b6037c92b4c840605831de8a251 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 11 Mar 2009 14:38:04 -0400 Subject: [PATCH 29/58] =?UTF-8?q?SUNRPC:=C2=A0xs=5Ftcp=5Fconnect=5Fworker{?= =?UTF-8?q?4,6}:=20merge=20common=20code?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Signed-off-by: Trond Myklebust --- net/sunrpc/xprtsock.c | 148 ++++++++++++++++++++++-------------------- 1 file changed, 76 insertions(+), 72 deletions(-) diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index 42222b4dd76d..f05a56e597ef 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -1738,33 +1738,29 @@ static int xs_tcp_finish_connecting(struct rpc_xprt *xprt, struct socket *sock) } /** - * xs_tcp_connect_worker4 - connect a TCP socket to a remote endpoint - * @work: RPC transport to connect + * xs_tcp_setup_socket - create a TCP socket and connect to a remote endpoint + * @xprt: RPC transport to connect + * @transport: socket transport to connect + * @create_sock: function to create a socket of the correct type * * Invoked by a work queue tasklet. */ -static void xs_tcp_connect_worker4(struct work_struct *work) +static void xs_tcp_setup_socket(struct rpc_xprt *xprt, + struct sock_xprt *transport, + struct socket *(*create_sock)(struct rpc_xprt *, + struct sock_xprt *)) { - struct sock_xprt *transport = - container_of(work, struct sock_xprt, connect_worker.work); - struct rpc_xprt *xprt = &transport->xprt; struct socket *sock = transport->sock; - int err, status = -EIO; + int status = -EIO; if (xprt->shutdown) goto out; if (!sock) { clear_bit(XPRT_CONNECTION_ABORT, &xprt->state); - /* start from scratch */ - if ((err = sock_create_kern(PF_INET, SOCK_STREAM, IPPROTO_TCP, &sock)) < 0) { - dprintk("RPC: can't create TCP transport socket (%d).\n", -err); - goto out; - } - xs_reclassify_socket4(sock); - - if (xs_bind4(transport, sock) < 0) { - sock_release(sock); + sock = create_sock(xprt, transport); + if (IS_ERR(sock)) { + status = PTR_ERR(sock); goto out; } } else { @@ -1808,6 +1804,69 @@ out: xprt_wake_pending_tasks(xprt, status); } +static struct socket *xs_create_tcp_sock4(struct rpc_xprt *xprt, + struct sock_xprt *transport) +{ + struct socket *sock; + int err; + + /* start from scratch */ + err = sock_create_kern(PF_INET, SOCK_STREAM, IPPROTO_TCP, &sock); + if (err < 0) { + dprintk("RPC: can't create TCP transport socket (%d).\n", + -err); + goto out_err; + } + xs_reclassify_socket4(sock); + + if (xs_bind4(transport, sock) < 0) { + sock_release(sock); + goto out_err; + } + return sock; +out_err: + return ERR_PTR(-EIO); +} + +/** + * xs_tcp_connect_worker4 - connect a TCP socket to a remote endpoint + * @work: RPC transport to connect + * + * Invoked by a work queue tasklet. + */ +static void xs_tcp_connect_worker4(struct work_struct *work) +{ + struct sock_xprt *transport = + container_of(work, struct sock_xprt, connect_worker.work); + struct rpc_xprt *xprt = &transport->xprt; + + xs_tcp_setup_socket(xprt, transport, xs_create_tcp_sock4); +} + +static struct socket *xs_create_tcp_sock6(struct rpc_xprt *xprt, + struct sock_xprt *transport) +{ + struct socket *sock; + int err; + + /* start from scratch */ + err = sock_create_kern(PF_INET6, SOCK_STREAM, IPPROTO_TCP, &sock); + if (err < 0) { + dprintk("RPC: can't create TCP transport socket (%d).\n", + -err); + goto out_err; + } + xs_reclassify_socket6(sock); + + if (xs_bind6(transport, sock) < 0) { + sock_release(sock); + goto out_err; + } + return sock; +out_err: + return ERR_PTR(-EIO); +} + /** * xs_tcp_connect_worker6 - connect a TCP socket to a remote endpoint * @work: RPC transport to connect @@ -1819,63 +1878,8 @@ static void xs_tcp_connect_worker6(struct work_struct *work) struct sock_xprt *transport = container_of(work, struct sock_xprt, connect_worker.work); struct rpc_xprt *xprt = &transport->xprt; - struct socket *sock = transport->sock; - int err, status = -EIO; - if (xprt->shutdown) - goto out; - - if (!sock) { - clear_bit(XPRT_CONNECTION_ABORT, &xprt->state); - /* start from scratch */ - if ((err = sock_create_kern(PF_INET6, SOCK_STREAM, IPPROTO_TCP, &sock)) < 0) { - dprintk("RPC: can't create TCP transport socket (%d).\n", -err); - goto out; - } - xs_reclassify_socket6(sock); - - if (xs_bind6(transport, sock) < 0) { - sock_release(sock); - goto out; - } - } else { - int abort_and_exit; - - abort_and_exit = test_and_clear_bit(XPRT_CONNECTION_ABORT, - &xprt->state); - /* "close" the socket, preserving the local port */ - xs_tcp_reuse_connection(xprt, transport); - - if (abort_and_exit) - goto out_eagain; - } - - dprintk("RPC: worker connecting xprt %p to address: %s\n", - xprt, xprt->address_strings[RPC_DISPLAY_ALL]); - - status = xs_tcp_finish_connecting(xprt, sock); - dprintk("RPC: %p connect status %d connected %d sock state %d\n", - xprt, -status, xprt_connected(xprt), sock->sk->sk_state); - switch (status) { - case -ECONNREFUSED: - case -ECONNRESET: - case -ENETUNREACH: - /* retry with existing socket, after a delay */ - case 0: - case -EINPROGRESS: - case -EALREADY: - xprt_clear_connecting(xprt); - return; - } - /* get rid of existing socket, and retry */ - xs_tcp_shutdown(xprt); - printk("%s: connect returned unhandled error %d\n", - __func__, status); -out_eagain: - status = -EAGAIN; -out: - xprt_clear_connecting(xprt); - xprt_wake_pending_tasks(xprt, status); + xs_tcp_setup_socket(xprt, transport, xs_create_tcp_sock6); } /** From 55420c24a0d4d1fce70ca713f84aa00b6b74a70e Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 11 Mar 2009 15:29:24 -0400 Subject: [PATCH 30/58] SUNRPC: Ensure we close the socket on EPIPE errors too... As long as one task is holding the socket lock, then calls to xprt_force_disconnect(xprt) will not succeed in shutting down the socket. In particular, this would mean that a server initiated shutdown will not succeed until the lock is relinquished. In order to avoid the deadlock, we should ensure that xs_tcp_send_request() closes the socket on EPIPE errors too. Signed-off-by: Trond Myklebust --- net/sunrpc/xprtsock.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index f05a56e597ef..fbc8725c20cb 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -726,10 +726,10 @@ static int xs_tcp_send_request(struct rpc_task *task) dprintk("RPC: sendmsg returned unrecognized error %d\n", -status); case -ECONNRESET: + case -EPIPE: xs_tcp_shutdown(xprt); case -ECONNREFUSED: case -ENOTCONN: - case -EPIPE: clear_bit(SOCK_ASYNC_NOSPACE, &transport->sock->flags); } out: From 2e3c230bc7149a6af65d26a0c312e230e0c33cc3 Mon Sep 17 00:00:00 2001 From: Tom Talpey Date: Thu, 12 Mar 2009 22:21:21 -0400 Subject: [PATCH 31/58] SVCRDMA: fix recent printk format warnings. printk formats in prior commit were reversed/incorrect. Compiled without warning on x86 and x86_64, but detected on ppc. Signed-off-by: Tom Talpey Signed-off-by: Trond Myklebust --- net/sunrpc/xprtrdma/svc_rdma_sendto.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/sunrpc/xprtrdma/svc_rdma_sendto.c b/net/sunrpc/xprtrdma/svc_rdma_sendto.c index d0bea987d80e..6c26a675435a 100644 --- a/net/sunrpc/xprtrdma/svc_rdma_sendto.c +++ b/net/sunrpc/xprtrdma/svc_rdma_sendto.c @@ -235,7 +235,7 @@ static int map_xdr(struct svcxprt_rdma *xprt, } dprintk("svcrdma: map_xdr: sge_no %d page_no %d " - "page_base %zd page_len %zd head_len %d tail_len %d\n", + "page_base %u page_len %u head_len %zu tail_len %zu\n", sge_no, page_no, xdr->page_base, xdr->page_len, xdr->head[0].iov_len, xdr->tail[0].iov_len); From 47c62564200609b6de60f535f61f0c73dd10c7c9 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Mon, 16 Mar 2009 08:13:41 -0400 Subject: [PATCH 32/58] NFS: Fix up a mismerged patch Move the definition of nfs_need_commit() into the #ifdef CONFIG_NFS_V3 section as originally intended in the patch "NFS: cleanup - remove struct nfs_inode->ncommit" Signed-off-by: Trond Myklebust --- fs/nfs/write.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/nfs/write.c b/fs/nfs/write.c index 36fd35e0de83..e560a78995a3 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -537,13 +537,13 @@ static void nfs_cancel_commit_list(struct list_head *head) } } +#if defined(CONFIG_NFS_V3) || defined(CONFIG_NFS_V4) static int nfs_need_commit(struct nfs_inode *nfsi) { return radix_tree_tagged(&nfsi->nfs_page_tree, NFS_PAGE_TAG_COMMIT); } -#if defined(CONFIG_NFS_V3) || defined(CONFIG_NFS_V4) /* * nfs_scan_commit - Scan an inode for commit requests * @inode: NFS inode to scan From b1e4adf4ea41bb8b5a7bfc1a7001f137e65495df Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Thu, 19 Mar 2009 15:35:49 -0400 Subject: [PATCH 33/58] NFS: Fix the notifications when renaming onto an existing file NFS appears to be returning an unnecessary "delete" notification when we're doing an atomic rename. See http://bugzilla.gnome.org/show_bug.cgi?id=575684 The fix is to get rid of the redundant call to d_delete(). Signed-off-by: Trond Myklebust --- fs/nfs/dir.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index 672368f865ca..3b2f6973e7c5 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -1624,8 +1624,7 @@ static int nfs_rename(struct inode *old_dir, struct dentry *old_dentry, } else if (atomic_read(&new_dentry->d_count) > 1) /* dentry still busy? */ goto out; - } else - nfs_drop_nlink(new_inode); + } go_ahead: /* @@ -1638,10 +1637,8 @@ go_ahead: } nfs_inode_return_delegation(old_inode); - if (new_inode != NULL) { + if (new_inode != NULL) nfs_inode_return_delegation(new_inode); - d_delete(new_dentry); - } error = NFS_PROTO(old_dir)->rename(old_dir, &old_dentry->d_name, new_dir, &new_dentry->d_name); @@ -1650,6 +1647,8 @@ out: if (rehash) d_rehash(rehash); if (!error) { + if (new_inode != NULL) + nfs_drop_nlink(new_inode); d_move(old_dentry, new_dentry); nfs_set_verifier(new_dentry, nfs_save_change_attribute(new_dir)); From 7fe5c398fc2186ed586db11106a6692d871d0d58 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Thu, 19 Mar 2009 15:35:50 -0400 Subject: [PATCH 34/58] NFS: Optimise NFS close() Close-to-open cache consistency rules really only require us to flush out writes on calls to close(), and require us to revalidate attributes on the very last close of the file. Currently we appear to be doing a lot of extra attribute revalidation and cache flushes. Signed-off-by: Trond Myklebust --- fs/nfs/file.c | 11 ++--------- fs/nfs/inode.c | 41 +++++++++++++++++++++++++++++------------ fs/nfs/internal.h | 3 +++ fs/nfs/nfs3proc.c | 1 + fs/nfs/nfs4proc.c | 10 ++++++++++ fs/nfs/proc.c | 1 + include/linux/nfs_xdr.h | 1 + 7 files changed, 47 insertions(+), 21 deletions(-) diff --git a/fs/nfs/file.c b/fs/nfs/file.c index 1eab9c9ad242..d451073c4947 100644 --- a/fs/nfs/file.c +++ b/fs/nfs/file.c @@ -137,9 +137,6 @@ nfs_file_release(struct inode *inode, struct file *filp) dentry->d_parent->d_name.name, dentry->d_name.name); - /* Ensure that dirty pages are flushed out with the right creds */ - if (filp->f_mode & FMODE_WRITE) - nfs_wb_all(dentry->d_inode); nfs_inc_stats(inode, NFSIOS_VFSRELEASE); return nfs_release(inode, filp); } @@ -231,7 +228,6 @@ nfs_file_flush(struct file *file, fl_owner_t id) struct nfs_open_context *ctx = nfs_file_open_context(file); struct dentry *dentry = file->f_path.dentry; struct inode *inode = dentry->d_inode; - int status; dprintk("NFS: flush(%s/%s)\n", dentry->d_parent->d_name.name, @@ -241,11 +237,8 @@ nfs_file_flush(struct file *file, fl_owner_t id) return 0; nfs_inc_stats(inode, NFSIOS_VFSFLUSH); - /* Ensure that data+attribute caches are up to date after close() */ - status = nfs_do_fsync(ctx, inode); - if (!status) - nfs_revalidate_inode(NFS_SERVER(inode), inode); - return status; + /* Flush writes to the server and return any errors */ + return nfs_do_fsync(ctx, inode); } static ssize_t diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index c40adc5dd609..a834d1d850b7 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -541,6 +541,32 @@ int nfs_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat) return err; } +/** + * nfs_close_context - Common close_context() routine NFSv2/v3 + * @ctx: pointer to context + * @is_sync: is this a synchronous close + * + * always ensure that the attributes are up to date if we're mounted + * with close-to-open semantics + */ +void nfs_close_context(struct nfs_open_context *ctx, int is_sync) +{ + struct inode *inode; + struct nfs_server *server; + + if (!(ctx->mode & FMODE_WRITE)) + return; + if (!is_sync) + return; + inode = ctx->path.dentry->d_inode; + if (!list_empty(&NFS_I(inode)->open_files)) + return; + server = NFS_SERVER(inode); + if (server->flags & NFS_MOUNT_NOCTO) + return; + nfs_revalidate_inode(server, inode); +} + static struct nfs_open_context *alloc_nfs_open_context(struct vfsmount *mnt, struct dentry *dentry, struct rpc_cred *cred) { struct nfs_open_context *ctx; @@ -567,24 +593,15 @@ struct nfs_open_context *get_nfs_open_context(struct nfs_open_context *ctx) return ctx; } -static void __put_nfs_open_context(struct nfs_open_context *ctx, int wait) +static void __put_nfs_open_context(struct nfs_open_context *ctx, int is_sync) { - struct inode *inode; + struct inode *inode = ctx->path.dentry->d_inode; - if (ctx == NULL) - return; - - inode = ctx->path.dentry->d_inode; if (!atomic_dec_and_lock(&ctx->count, &inode->i_lock)) return; list_del(&ctx->list); spin_unlock(&inode->i_lock); - if (ctx->state != NULL) { - if (wait) - nfs4_close_sync(&ctx->path, ctx->state, ctx->mode); - else - nfs4_close_state(&ctx->path, ctx->state, ctx->mode); - } + NFS_PROTO(inode)->close_context(ctx, is_sync); if (ctx->cred != NULL) put_rpccred(ctx->cred); path_put(&ctx->path); diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h index a55e69aa52e5..2041f68ff1cc 100644 --- a/fs/nfs/internal.h +++ b/fs/nfs/internal.h @@ -152,6 +152,9 @@ extern __be32 *nfs4_decode_dirent(__be32 *p, struct nfs_entry *entry, int plus); extern struct rpc_procinfo nfs4_procedures[]; #endif +/* proc.c */ +void nfs_close_context(struct nfs_open_context *ctx, int is_sync); + /* dir.c */ extern int nfs_access_cache_shrinker(int nr_to_scan, gfp_t gfp_mask); diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c index c55be7a7679e..b82fe6847f14 100644 --- a/fs/nfs/nfs3proc.c +++ b/fs/nfs/nfs3proc.c @@ -834,4 +834,5 @@ const struct nfs_rpc_ops nfs_v3_clientops = { .commit_done = nfs3_commit_done, .lock = nfs3_proc_lock, .clear_acl_cache = nfs3_forget_cached_acls, + .close_context = nfs_close_context, }; diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 95f171e7e05a..97bacccff579 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -1572,6 +1572,15 @@ out_drop: return 0; } +void nfs4_close_context(struct nfs_open_context *ctx, int is_sync) +{ + if (ctx->state == NULL) + return; + if (is_sync) + nfs4_close_sync(&ctx->path, ctx->state, ctx->mode); + else + nfs4_close_state(&ctx->path, ctx->state, ctx->mode); +} static int _nfs4_server_capabilities(struct nfs_server *server, struct nfs_fh *fhandle) { @@ -3776,6 +3785,7 @@ const struct nfs_rpc_ops nfs_v4_clientops = { .commit_done = nfs4_commit_done, .lock = nfs4_proc_lock, .clear_acl_cache = nfs4_zap_acl_attr, + .close_context = nfs4_close_context, }; /* diff --git a/fs/nfs/proc.c b/fs/nfs/proc.c index 193465210d7c..7be72d90d49d 100644 --- a/fs/nfs/proc.c +++ b/fs/nfs/proc.c @@ -663,4 +663,5 @@ const struct nfs_rpc_ops nfs_v2_clientops = { .commit_setup = nfs_proc_commit_setup, .lock = nfs_proc_lock, .lock_check_bounds = nfs_lock_check_bounds, + .close_context = nfs_close_context, }; diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index 0691b9c188d9..9708e78a4d49 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -868,6 +868,7 @@ struct nfs_rpc_ops { int (*lock)(struct file *, int, struct file_lock *); int (*lock_check_bounds)(const struct file_lock *); void (*clear_acl_cache)(struct inode *); + void (*close_context)(struct nfs_open_context *ctx, int); }; /* From 776bd5c7a207de546918f805090bfc823d2660c8 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Wed, 18 Mar 2009 20:45:28 -0400 Subject: [PATCH 35/58] SUNRPC: Don't flag empty RPCB_GETADDR reply as bogus In 2007, commit e65fe3976f594603ed7b1b4a99d3e9b867f573ea added additional sanity checking to rpcb_decode_getaddr() to make sure we were getting a reply that was long enough to be an actual universal address. If the uaddr string isn't long enough, the XDR decoder returns EIO. However, an empty string is a valid RPCB_GETADDR response if the requested service isn't registered. Moreover, "::.n.m" is also a valid RPCB_GETADDR response for IPv6 addresses that is shorter than rpcb_decode_getaddr()'s lower limit of 11. So this sanity check introduced a regression for rpcbind requests against IPv6 remotes. So revert the lower bound check added by commit e65fe3976f594603ed7b1b4a99d3e9b867f573ea, and add an explicit check for an empty uaddr string, similar to libtirpc's rpcb_getaddr(3). Pointed-out-by: Jeff Layton Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- net/sunrpc/rpcb_clnt.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) diff --git a/net/sunrpc/rpcb_clnt.c b/net/sunrpc/rpcb_clnt.c index 03ae007641e4..2caa7edeeaba 100644 --- a/net/sunrpc/rpcb_clnt.c +++ b/net/sunrpc/rpcb_clnt.c @@ -703,11 +703,16 @@ static int rpcb_decode_getaddr(struct rpc_rqst *req, __be32 *p, *portp = 0; addr_len = ntohl(*p++); + if (addr_len == 0) { + dprintk("RPC: rpcb_decode_getaddr: " + "service is not registered\n"); + return 0; + } + /* - * Simple sanity check. The smallest possible universal - * address is an IPv4 address string containing 11 bytes. + * Simple sanity check. */ - if (addr_len < 11 || addr_len > RPCBIND_MAXUADDRLEN) + if (addr_len > RPCBIND_MAXUADDRLEN) goto out_err; /* From efb3288b423d7e3533a68dccecaa05a56a281a4e Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Wed, 18 Mar 2009 20:45:43 -0400 Subject: [PATCH 36/58] SUNRPC: Clean up static inline functions in svc_xprt.h Clean up: Enable the use of const arguments in higher level svc_ APIs by adding const to the arguments of the helper functions in svc_xprt.h Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/svc_xprt.h | 46 +++++++++++++++++++-------------- 1 file changed, 26 insertions(+), 20 deletions(-) diff --git a/include/linux/sunrpc/svc_xprt.h b/include/linux/sunrpc/svc_xprt.h index 0127daca4354..959b931b6053 100644 --- a/include/linux/sunrpc/svc_xprt.h +++ b/include/linux/sunrpc/svc_xprt.h @@ -88,29 +88,32 @@ static inline void svc_xprt_get(struct svc_xprt *xprt) kref_get(&xprt->xpt_ref); } static inline void svc_xprt_set_local(struct svc_xprt *xprt, - struct sockaddr *sa, int salen) + const struct sockaddr *sa, + const size_t salen) { memcpy(&xprt->xpt_local, sa, salen); xprt->xpt_locallen = salen; } static inline void svc_xprt_set_remote(struct svc_xprt *xprt, - struct sockaddr *sa, int salen) + const struct sockaddr *sa, + const size_t salen) { memcpy(&xprt->xpt_remote, sa, salen); xprt->xpt_remotelen = salen; } -static inline unsigned short svc_addr_port(struct sockaddr *sa) +static inline unsigned short svc_addr_port(const struct sockaddr *sa) { - unsigned short ret = 0; + const struct sockaddr_in *sin = (const struct sockaddr_in *)sa; + const struct sockaddr_in6 *sin6 = (const struct sockaddr_in6 *)sa; + switch (sa->sa_family) { case AF_INET: - ret = ntohs(((struct sockaddr_in *)sa)->sin_port); - break; + return ntohs(sin->sin_port); case AF_INET6: - ret = ntohs(((struct sockaddr_in6 *)sa)->sin6_port); - break; + return ntohs(sin6->sin6_port); } - return ret; + + return 0; } static inline size_t svc_addr_len(struct sockaddr *sa) @@ -124,36 +127,39 @@ static inline size_t svc_addr_len(struct sockaddr *sa) return -EAFNOSUPPORT; } -static inline unsigned short svc_xprt_local_port(struct svc_xprt *xprt) +static inline unsigned short svc_xprt_local_port(const struct svc_xprt *xprt) { - return svc_addr_port((struct sockaddr *)&xprt->xpt_local); + return svc_addr_port((const struct sockaddr *)&xprt->xpt_local); } -static inline unsigned short svc_xprt_remote_port(struct svc_xprt *xprt) +static inline unsigned short svc_xprt_remote_port(const struct svc_xprt *xprt) { - return svc_addr_port((struct sockaddr *)&xprt->xpt_remote); + return svc_addr_port((const struct sockaddr *)&xprt->xpt_remote); } -static inline char *__svc_print_addr(struct sockaddr *addr, - char *buf, size_t len) +static inline char *__svc_print_addr(const struct sockaddr *addr, + char *buf, const size_t len) { + const struct sockaddr_in *sin = (const struct sockaddr_in *)addr; + const struct sockaddr_in6 *sin6 = (const struct sockaddr_in6 *)addr; + switch (addr->sa_family) { case AF_INET: - snprintf(buf, len, "%pI4, port=%u", - &((struct sockaddr_in *)addr)->sin_addr, - ntohs(((struct sockaddr_in *) addr)->sin_port)); + snprintf(buf, len, "%pI4, port=%u", &sin->sin_addr, + ntohs(sin->sin_port)); break; case AF_INET6: snprintf(buf, len, "%pI6, port=%u", - &((struct sockaddr_in6 *)addr)->sin6_addr, - ntohs(((struct sockaddr_in6 *) addr)->sin6_port)); + &sin6->sin6_addr, + ntohs(sin6->sin6_port)); break; default: snprintf(buf, len, "unknown address type: %d", addr->sa_family); break; } + return buf; } #endif /* SUNRPC_SVC_XPRT_H */ From adbbe929569e6eec8ff9feca23f1f2b40b42853d Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Wed, 18 Mar 2009 20:45:51 -0400 Subject: [PATCH 37/58] NFSD: If port value written to /proc/fs/nfsd/portlist is invalid, return EINVAL Make sure port value read from user space by write_ports is valid before passing it to svc_find_xprt(). If it wasn't, the writer would get ENOENT instead of EINVAL. Noticed-by: J. Bruce Fields Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- fs/nfsd/nfsctl.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c index 3d93b2064ce5..5a936c14f6ff 100644 --- a/fs/nfsd/nfsctl.c +++ b/fs/nfsd/nfsctl.c @@ -938,6 +938,8 @@ static ssize_t __write_ports(struct file *file, char *buf, size_t size) char transport[16]; int port; if (sscanf(buf, "%15s %4d", transport, &port) == 2) { + if (port < 1 || port > 65535) + return -EINVAL; err = nfsd_create_serv(); if (!err) { err = svc_create_xprt(nfsd_serv, @@ -960,7 +962,7 @@ static ssize_t __write_ports(struct file *file, char *buf, size_t size) char transport[16]; int port; if (sscanf(&buf[1], "%15s %4d", transport, &port) == 2) { - if (port == 0) + if (port < 1 || port > 65535) return -EINVAL; if (nfsd_serv) { xprt = svc_find_xprt(nfsd_serv, transport, From 156e62094a74cf43f02f56ef96b6cda567501357 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Wed, 18 Mar 2009 20:45:58 -0400 Subject: [PATCH 38/58] SUNRPC: Clean up svc_find_xprt() calling sequence Clean up: add documentating comment and use appropriate data types for svc_find_xprt()'s arguments. This also eliminates a mixed sign comparison: @port was an int, while the return value of svc_xprt_local_port() is an unsigned short. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/svc_xprt.h | 3 ++- net/sunrpc/svc_xprt.c | 16 +++++++++++----- 2 files changed, 13 insertions(+), 6 deletions(-) diff --git a/include/linux/sunrpc/svc_xprt.h b/include/linux/sunrpc/svc_xprt.h index 959b931b6053..55b68582c5d9 100644 --- a/include/linux/sunrpc/svc_xprt.h +++ b/include/linux/sunrpc/svc_xprt.h @@ -80,7 +80,8 @@ void svc_close_xprt(struct svc_xprt *xprt); void svc_delete_xprt(struct svc_xprt *xprt); int svc_port_is_privileged(struct sockaddr *sin); int svc_print_xprts(char *buf, int maxlen); -struct svc_xprt *svc_find_xprt(struct svc_serv *, char *, int, int); +struct svc_xprt *svc_find_xprt(struct svc_serv *serv, const char *xcl_name, + const sa_family_t af, const unsigned short port); int svc_xprt_names(struct svc_serv *serv, char *buf, int buflen); static inline void svc_xprt_get(struct svc_xprt *xprt) diff --git a/net/sunrpc/svc_xprt.c b/net/sunrpc/svc_xprt.c index e588df5d6b34..c947c93dbc24 100644 --- a/net/sunrpc/svc_xprt.c +++ b/net/sunrpc/svc_xprt.c @@ -1033,7 +1033,13 @@ static struct svc_deferred_req *svc_deferred_dequeue(struct svc_xprt *xprt) return dr; } -/* +/** + * svc_find_xprt - find an RPC transport instance + * @serv: pointer to svc_serv to search + * @xcl_name: C string containing transport's class name + * @af: Address family of transport's local address + * @port: transport's IP port number + * * Return the transport instance pointer for the endpoint accepting * connections/peer traffic from the specified transport class, * address family and port. @@ -1042,14 +1048,14 @@ static struct svc_deferred_req *svc_deferred_dequeue(struct svc_xprt *xprt) * wild-card, and will result in matching the first transport in the * service's list that has a matching class name. */ -struct svc_xprt *svc_find_xprt(struct svc_serv *serv, char *xcl_name, - int af, int port) +struct svc_xprt *svc_find_xprt(struct svc_serv *serv, const char *xcl_name, + const sa_family_t af, const unsigned short port) { struct svc_xprt *xprt; struct svc_xprt *found = NULL; /* Sanity check the args */ - if (!serv || !xcl_name) + if (serv == NULL || xcl_name == NULL) return found; spin_lock_bh(&serv->sv_lock); @@ -1058,7 +1064,7 @@ struct svc_xprt *svc_find_xprt(struct svc_serv *serv, char *xcl_name, continue; if (af != AF_UNSPEC && af != xprt->xpt_local.ss_family) continue; - if (port && port != svc_xprt_local_port(xprt)) + if (port != 0 && port != svc_xprt_local_port(xprt)) continue; found = xprt; svc_xprt_get(xprt); From 4b62e58cccff9c5e7ffc7023f7ec24c75fbd549b Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Wed, 18 Mar 2009 20:46:06 -0400 Subject: [PATCH 39/58] SUNRPC: Pass a family argument to svc_register() The sv_family field is going away. Instead of using sv_family, have the svc_register() function take a protocol family argument. Since this argument represents a protocol family, and not an address family, this argument takes an int, as this is what is passed to sock_create_kern(). Also make sure svc_register's helpers are checking for PF_FOO instead of AF_FOO. The value of [AP]F_FOO are equivalent; this is simply a symbolic change to reflect the semantics of the value stored in that variable. sock_create_kern() should return EPFNOSUPPORT if the passed-in protocol family isn't supported, but it uses EAFNOSUPPORT for this case. We will stick with that tradition here, as svc_register() is called by the RPC server in the same path as sock_create_kern(). Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/svc.h | 4 ++-- net/sunrpc/svc.c | 21 +++++++++++---------- net/sunrpc/svcsock.c | 2 +- 3 files changed, 14 insertions(+), 13 deletions(-) diff --git a/include/linux/sunrpc/svc.h b/include/linux/sunrpc/svc.h index 3435d24bfe55..1f18fc728cba 100644 --- a/include/linux/sunrpc/svc.h +++ b/include/linux/sunrpc/svc.h @@ -396,8 +396,8 @@ struct svc_serv * svc_create_pooled(struct svc_program *, unsigned int, int svc_set_num_threads(struct svc_serv *, struct svc_pool *, int); void svc_destroy(struct svc_serv *); int svc_process(struct svc_rqst *); -int svc_register(const struct svc_serv *, const unsigned short, - const unsigned short); +int svc_register(const struct svc_serv *, const int, + const unsigned short, const unsigned short); void svc_wake_up(struct svc_serv *); void svc_reserve(struct svc_rqst *rqstp, int space); diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c index c51fed4d1af1..41bc36ea2224 100644 --- a/net/sunrpc/svc.c +++ b/net/sunrpc/svc.c @@ -800,17 +800,17 @@ static int __svc_rpcb_register6(const u32 program, const u32 version, * if any error occurs. */ static int __svc_register(const u32 program, const u32 version, - const sa_family_t family, + const int family, const unsigned short protocol, const unsigned short port) { int error; switch (family) { - case AF_INET: + case PF_INET: return __svc_rpcb_register4(program, version, protocol, port); - case AF_INET6: + case PF_INET6: error = __svc_rpcb_register6(program, version, protocol, port); if (error < 0) @@ -840,11 +840,11 @@ static int __svc_register(const u32 program, const u32 version, * if any error occurs. */ static int __svc_register(const u32 program, const u32 version, - sa_family_t family, + const int family, const unsigned short protocol, const unsigned short port) { - if (family != AF_INET) + if (family != PF_INET) return -EAFNOSUPPORT; return rpcb_register(program, version, protocol, port); @@ -855,13 +855,14 @@ static int __svc_register(const u32 program, const u32 version, /** * svc_register - register an RPC service with the local portmapper * @serv: svc_serv struct for the service to register + * @family: protocol family of service's listener socket * @proto: transport protocol number to advertise * @port: port to advertise * - * Service is registered for any address in serv's address family + * Service is registered for any address in the passed-in protocol family */ -int svc_register(const struct svc_serv *serv, const unsigned short proto, - const unsigned short port) +int svc_register(const struct svc_serv *serv, const int family, + const unsigned short proto, const unsigned short port) { struct svc_program *progp; unsigned int i; @@ -879,7 +880,7 @@ int svc_register(const struct svc_serv *serv, const unsigned short proto, i, proto == IPPROTO_UDP? "udp" : "tcp", port, - serv->sv_family, + family, progp->pg_vers[i]->vs_hidden? " (but not telling portmap)" : ""); @@ -887,7 +888,7 @@ int svc_register(const struct svc_serv *serv, const unsigned short proto, continue; error = __svc_register(progp->pg_prog, i, - serv->sv_family, proto, port); + family, proto, port); if (error < 0) break; } diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c index 5763e6460fea..d00583c1cd04 100644 --- a/net/sunrpc/svcsock.c +++ b/net/sunrpc/svcsock.c @@ -1122,7 +1122,7 @@ static struct svc_sock *svc_setup_socket(struct svc_serv *serv, /* Register socket with portmapper */ if (*errp >= 0 && pmap_register) - *errp = svc_register(serv, inet->sk_protocol, + *errp = svc_register(serv, serv->sv_family, inet->sk_protocol, ntohs(inet_sk(inet)->sport)); if (*errp < 0) { From baf01caf09e87579c2d157e5ee29975db8551522 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Wed, 18 Mar 2009 20:46:13 -0400 Subject: [PATCH 40/58] SUNRPC: svc_setup_socket() gets protocol family from socket Since the sv_family field is going away, modify svc_setup_socket() to extract the protocol family from the passed-in socket instead of from the passed-in svc_serv struct. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- net/sunrpc/svcsock.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c index d00583c1cd04..d00bc3307745 100644 --- a/net/sunrpc/svcsock.c +++ b/net/sunrpc/svcsock.c @@ -1122,7 +1122,7 @@ static struct svc_sock *svc_setup_socket(struct svc_serv *serv, /* Register socket with portmapper */ if (*errp >= 0 && pmap_register) - *errp = svc_register(serv, serv->sv_family, inet->sk_protocol, + *errp = svc_register(serv, inet->sk_family, inet->sk_protocol, ntohs(inet_sk(inet)->sport)); if (*errp < 0) { @@ -1145,13 +1145,13 @@ static struct svc_sock *svc_setup_socket(struct svc_serv *serv, /* * We start one listener per sv_serv. We want AF_INET - * requests to be automatically shunted to our AF_INET6 + * requests to be automatically shunted to our PF_INET6 * listener using a mapped IPv4 address. Make sure * no-one starts an equivalent IPv4 listener, which * would steal our incoming connections. */ val = 0; - if (serv->sv_family == AF_INET6) + if (inet->sk_family == PF_INET6) kernel_setsockopt(sock, SOL_IPV6, IPV6_V6ONLY, (char *)&val, sizeof(val)); From 9652ada3fb5914a67d8422114e8a76388330fa79 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Wed, 18 Mar 2009 20:46:21 -0400 Subject: [PATCH 41/58] SUNRPC: Change svc_create_xprt() to take a @family argument The sv_family field is going away. Pass a protocol family argument to svc_create_xprt() instead of extracting the family from the passed-in svc_serv struct. Again, as this is a listener socket and not an address, we make this new argument an "int" protocol family, instead of an "sa_family_t." Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- fs/lockd/svc.c | 3 ++- fs/nfs/callback.c | 4 ++-- fs/nfsd/nfsctl.c | 2 +- fs/nfsd/nfssvc.c | 4 ++-- include/linux/sunrpc/svc_xprt.h | 3 ++- net/sunrpc/svc_xprt.c | 15 +++++++++------ 6 files changed, 18 insertions(+), 13 deletions(-) diff --git a/fs/lockd/svc.c b/fs/lockd/svc.c index 64f1c31b5853..390c5593655c 100644 --- a/fs/lockd/svc.c +++ b/fs/lockd/svc.c @@ -211,7 +211,8 @@ static int create_lockd_listener(struct svc_serv *serv, char *name, xprt = svc_find_xprt(serv, name, 0, 0); if (xprt == NULL) - return svc_create_xprt(serv, name, port, SVC_SOCK_DEFAULTS); + return svc_create_xprt(serv, name, nlmsvc_family, + port, SVC_SOCK_DEFAULTS); svc_xprt_put(xprt); return 0; diff --git a/fs/nfs/callback.c b/fs/nfs/callback.c index 3e634f2a1083..fb35cab63c8a 100644 --- a/fs/nfs/callback.c +++ b/fs/nfs/callback.c @@ -122,8 +122,8 @@ int nfs_callback_up(void) if (!serv) goto out_err; - ret = svc_create_xprt(serv, "tcp", nfs_callback_set_tcpport, - SVC_SOCK_ANONYMOUS); + ret = svc_create_xprt(serv, "tcp", nfs_callback_family, + nfs_callback_set_tcpport, SVC_SOCK_ANONYMOUS); if (ret <= 0) goto out_err; nfs_callback_tcpport = ret; diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c index 5a936c14f6ff..a4ed8644d69c 100644 --- a/fs/nfsd/nfsctl.c +++ b/fs/nfsd/nfsctl.c @@ -943,7 +943,7 @@ static ssize_t __write_ports(struct file *file, char *buf, size_t size) err = nfsd_create_serv(); if (!err) { err = svc_create_xprt(nfsd_serv, - transport, port, + transport, PF_INET, port, SVC_SOCK_ANONYMOUS); if (err == -ENOENT) /* Give a reasonable perror msg for diff --git a/fs/nfsd/nfssvc.c b/fs/nfsd/nfssvc.c index 07e4f5d7baa8..ab7f249055b5 100644 --- a/fs/nfsd/nfssvc.c +++ b/fs/nfsd/nfssvc.c @@ -244,7 +244,7 @@ static int nfsd_init_socks(int port) if (!list_empty(&nfsd_serv->sv_permsocks)) return 0; - error = svc_create_xprt(nfsd_serv, "udp", port, + error = svc_create_xprt(nfsd_serv, "udp", PF_INET, port, SVC_SOCK_DEFAULTS); if (error < 0) return error; @@ -253,7 +253,7 @@ static int nfsd_init_socks(int port) if (error < 0) return error; - error = svc_create_xprt(nfsd_serv, "tcp", port, + error = svc_create_xprt(nfsd_serv, "tcp", PF_INET, port, SVC_SOCK_DEFAULTS); if (error < 0) return error; diff --git a/include/linux/sunrpc/svc_xprt.h b/include/linux/sunrpc/svc_xprt.h index 55b68582c5d9..0d9cb6ef28b0 100644 --- a/include/linux/sunrpc/svc_xprt.h +++ b/include/linux/sunrpc/svc_xprt.h @@ -71,7 +71,8 @@ int svc_reg_xprt_class(struct svc_xprt_class *); void svc_unreg_xprt_class(struct svc_xprt_class *); void svc_xprt_init(struct svc_xprt_class *, struct svc_xprt *, struct svc_serv *); -int svc_create_xprt(struct svc_serv *, char *, unsigned short, int); +int svc_create_xprt(struct svc_serv *, const char *, const int, + const unsigned short, int); void svc_xprt_enqueue(struct svc_xprt *xprt); void svc_xprt_received(struct svc_xprt *); void svc_xprt_put(struct svc_xprt *xprt); diff --git a/net/sunrpc/svc_xprt.c b/net/sunrpc/svc_xprt.c index c947c93dbc24..2819ee093f36 100644 --- a/net/sunrpc/svc_xprt.c +++ b/net/sunrpc/svc_xprt.c @@ -161,7 +161,9 @@ EXPORT_SYMBOL_GPL(svc_xprt_init); static struct svc_xprt *__svc_xpo_create(struct svc_xprt_class *xcl, struct svc_serv *serv, - unsigned short port, int flags) + const int family, + const unsigned short port, + int flags) { struct sockaddr_in sin = { .sin_family = AF_INET, @@ -176,12 +178,12 @@ static struct svc_xprt *__svc_xpo_create(struct svc_xprt_class *xcl, struct sockaddr *sap; size_t len; - switch (serv->sv_family) { - case AF_INET: + switch (family) { + case PF_INET: sap = (struct sockaddr *)&sin; len = sizeof(sin); break; - case AF_INET6: + case PF_INET6: sap = (struct sockaddr *)&sin6; len = sizeof(sin6); break; @@ -192,7 +194,8 @@ static struct svc_xprt *__svc_xpo_create(struct svc_xprt_class *xcl, return xcl->xcl_ops->xpo_create(serv, sap, len, flags); } -int svc_create_xprt(struct svc_serv *serv, char *xprt_name, unsigned short port, +int svc_create_xprt(struct svc_serv *serv, const char *xprt_name, + const int family, const unsigned short port, int flags) { struct svc_xprt_class *xcl; @@ -209,7 +212,7 @@ int svc_create_xprt(struct svc_serv *serv, char *xprt_name, unsigned short port, goto err; spin_unlock(&svc_xprt_class_lock); - newxprt = __svc_xpo_create(xcl, serv, port, flags); + newxprt = __svc_xpo_create(xcl, serv, family, port, flags); if (IS_ERR(newxprt)) { module_put(xcl->xcl_owner); return PTR_ERR(newxprt); From 49a9072f29a1039f142ec98b44a72d7173651c02 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Wed, 18 Mar 2009 20:46:29 -0400 Subject: [PATCH 42/58] SUNRPC: Remove @family argument from svc_create() and svc_create_pooled() Since an RPC service listener's protocol family is specified now via svc_create_xprt(), it no longer needs to be passed to svc_create() or svc_create_pooled(). Remove that argument from the synopsis of those functions, and remove the sv_family field from the svc_serv struct. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- fs/lockd/svc.c | 2 +- fs/nfs/callback.c | 3 +-- fs/nfsd/nfssvc.c | 1 - include/linux/sunrpc/svc.h | 5 ++--- net/sunrpc/svc.c | 11 +++++------ 5 files changed, 9 insertions(+), 13 deletions(-) diff --git a/fs/lockd/svc.c b/fs/lockd/svc.c index 390c5593655c..d30920038cb6 100644 --- a/fs/lockd/svc.c +++ b/fs/lockd/svc.c @@ -275,7 +275,7 @@ int lockd_up(void) "lockd_up: no pid, %d users??\n", nlmsvc_users); error = -ENOMEM; - serv = svc_create(&nlmsvc_program, LOCKD_BUFSIZE, nlmsvc_family, NULL); + serv = svc_create(&nlmsvc_program, LOCKD_BUFSIZE, NULL); if (!serv) { printk(KERN_WARNING "lockd_up: create service failed\n"); goto out; diff --git a/fs/nfs/callback.c b/fs/nfs/callback.c index fb35cab63c8a..ddf4b4ae6967 100644 --- a/fs/nfs/callback.c +++ b/fs/nfs/callback.c @@ -116,8 +116,7 @@ int nfs_callback_up(void) mutex_lock(&nfs_callback_mutex); if (nfs_callback_info.users++ || nfs_callback_info.task != NULL) goto out; - serv = svc_create(&nfs4_callback_program, NFS4_CALLBACK_BUFSIZE, - nfs_callback_family, NULL); + serv = svc_create(&nfs4_callback_program, NFS4_CALLBACK_BUFSIZE, NULL); ret = -ENOMEM; if (!serv) goto out_err; diff --git a/fs/nfsd/nfssvc.c b/fs/nfsd/nfssvc.c index ab7f249055b5..bc3567bab8c4 100644 --- a/fs/nfsd/nfssvc.c +++ b/fs/nfsd/nfssvc.c @@ -229,7 +229,6 @@ int nfsd_create_serv(void) atomic_set(&nfsd_busy, 0); nfsd_serv = svc_create_pooled(&nfsd_program, nfsd_max_blksize, - AF_INET, nfsd_last_thread, nfsd, THIS_MODULE); if (nfsd_serv == NULL) err = -ENOMEM; diff --git a/include/linux/sunrpc/svc.h b/include/linux/sunrpc/svc.h index 1f18fc728cba..d3a4c0231933 100644 --- a/include/linux/sunrpc/svc.h +++ b/include/linux/sunrpc/svc.h @@ -69,7 +69,6 @@ struct svc_serv { struct list_head sv_tempsocks; /* all temporary sockets */ int sv_tmpcnt; /* count of temporary sockets */ struct timer_list sv_temptimer; /* timer for aging temporary sockets */ - sa_family_t sv_family; /* listener's address family */ char * sv_name; /* service name */ @@ -385,13 +384,13 @@ struct svc_procedure { /* * Function prototypes. */ -struct svc_serv *svc_create(struct svc_program *, unsigned int, sa_family_t, +struct svc_serv *svc_create(struct svc_program *, unsigned int, void (*shutdown)(struct svc_serv *)); struct svc_rqst *svc_prepare_thread(struct svc_serv *serv, struct svc_pool *pool); void svc_exit_thread(struct svc_rqst *); struct svc_serv * svc_create_pooled(struct svc_program *, unsigned int, - sa_family_t, void (*shutdown)(struct svc_serv *), + void (*shutdown)(struct svc_serv *), svc_thread_fn, struct module *); int svc_set_num_threads(struct svc_serv *, struct svc_pool *, int); void svc_destroy(struct svc_serv *); diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c index 41bc36ea2224..d72ff44826d8 100644 --- a/net/sunrpc/svc.c +++ b/net/sunrpc/svc.c @@ -359,7 +359,7 @@ svc_pool_for_cpu(struct svc_serv *serv, int cpu) */ static struct svc_serv * __svc_create(struct svc_program *prog, unsigned int bufsize, int npools, - sa_family_t family, void (*shutdown)(struct svc_serv *serv)) + void (*shutdown)(struct svc_serv *serv)) { struct svc_serv *serv; unsigned int vers; @@ -368,7 +368,6 @@ __svc_create(struct svc_program *prog, unsigned int bufsize, int npools, if (!(serv = kzalloc(sizeof(*serv), GFP_KERNEL))) return NULL; - serv->sv_family = family; serv->sv_name = prog->pg_name; serv->sv_program = prog; serv->sv_nrthreads = 1; @@ -427,21 +426,21 @@ __svc_create(struct svc_program *prog, unsigned int bufsize, int npools, struct svc_serv * svc_create(struct svc_program *prog, unsigned int bufsize, - sa_family_t family, void (*shutdown)(struct svc_serv *serv)) + void (*shutdown)(struct svc_serv *serv)) { - return __svc_create(prog, bufsize, /*npools*/1, family, shutdown); + return __svc_create(prog, bufsize, /*npools*/1, shutdown); } EXPORT_SYMBOL_GPL(svc_create); struct svc_serv * svc_create_pooled(struct svc_program *prog, unsigned int bufsize, - sa_family_t family, void (*shutdown)(struct svc_serv *serv), + void (*shutdown)(struct svc_serv *serv), svc_thread_fn func, struct module *mod) { struct svc_serv *serv; unsigned int npools = svc_pool_map_get(); - serv = __svc_create(prog, bufsize, npools, family, shutdown); + serv = __svc_create(prog, bufsize, npools, shutdown); if (serv != NULL) { serv->sv_function = func; From 26298caacac3e4754194b13aef377706d5de6cf6 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Wed, 18 Mar 2009 20:46:36 -0400 Subject: [PATCH 43/58] NFS: Revert creation of IPv6 listeners for lockd and NFSv4 callbacks We're about to convert over to using separate PF_INET and PF_INET6 listeners, instead of a single PF_INET6 listener that also receives AF_INET requests and maps them to AF_INET6. Clear the way by removing the logic in lockd and the NFSv4 callback server that creates an AF_INET6 service listener. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- fs/lockd/svc.c | 13 +------------ fs/nfs/callback.c | 14 ++------------ 2 files changed, 3 insertions(+), 24 deletions(-) diff --git a/fs/lockd/svc.c b/fs/lockd/svc.c index d30920038cb6..566932b98fd3 100644 --- a/fs/lockd/svc.c +++ b/fs/lockd/svc.c @@ -52,17 +52,6 @@ static struct task_struct *nlmsvc_task; static struct svc_rqst *nlmsvc_rqst; unsigned long nlmsvc_timeout; -/* - * If the kernel has IPv6 support available, always listen for - * both AF_INET and AF_INET6 requests. - */ -#if (defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)) && \ - defined(CONFIG_SUNRPC_REGISTER_V4) -static const sa_family_t nlmsvc_family = AF_INET6; -#else /* (CONFIG_IPV6 || CONFIG_IPV6_MODULE) && CONFIG_SUNRPC_REGISTER_V4 */ -static const sa_family_t nlmsvc_family = AF_INET; -#endif /* (CONFIG_IPV6 || CONFIG_IPV6_MODULE) && CONFIG_SUNRPC_REGISTER_V4 */ - /* * These can be set at insmod time (useful for NFS as root filesystem), * and also changed through the sysctl interface. -- Jamie Lokier, Aug 2003 @@ -211,7 +200,7 @@ static int create_lockd_listener(struct svc_serv *serv, char *name, xprt = svc_find_xprt(serv, name, 0, 0); if (xprt == NULL) - return svc_create_xprt(serv, name, nlmsvc_family, + return svc_create_xprt(serv, name, PF_INET, port, SVC_SOCK_DEFAULTS); svc_xprt_put(xprt); diff --git a/fs/nfs/callback.c b/fs/nfs/callback.c index ddf4b4ae6967..0ef47dff89be 100644 --- a/fs/nfs/callback.c +++ b/fs/nfs/callback.c @@ -41,16 +41,6 @@ unsigned short nfs_callback_tcpport; static const int nfs_set_port_min = 0; static const int nfs_set_port_max = 65535; -/* - * If the kernel has IPv6 support available, always listen for - * both AF_INET and AF_INET6 requests. - */ -#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) -static const sa_family_t nfs_callback_family = AF_INET6; -#else -static const sa_family_t nfs_callback_family = AF_INET; -#endif - static int param_set_port(const char *val, struct kernel_param *kp) { char *endp; @@ -121,13 +111,13 @@ int nfs_callback_up(void) if (!serv) goto out_err; - ret = svc_create_xprt(serv, "tcp", nfs_callback_family, + ret = svc_create_xprt(serv, "tcp", PF_INET, nfs_callback_set_tcpport, SVC_SOCK_ANONYMOUS); if (ret <= 0) goto out_err; nfs_callback_tcpport = ret; dprintk("NFS: Callback listener port = %u (af %u)\n", - nfs_callback_tcpport, nfs_callback_family); + nfs_callback_tcpport, PF_INET); nfs_callback_info.rqst = svc_prepare_thread(serv, &serv->sv_pools[0]); if (IS_ERR(nfs_callback_info.rqst)) { From 7d21c0f9845f0ce4e81baac3519fbb2c6c2cc908 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Wed, 18 Mar 2009 20:46:44 -0400 Subject: [PATCH 44/58] SUNRPC: Set IPV6ONLY flag on PF_INET6 RPC listener sockets We are about to convert to using separate RPC listener sockets for PF_INET and PF_INET6. This echoes the way IPv6 is handled in user space by TI-RPC, and eliminates the need for ULPs to worry about mapped IPv4 AF_INET6 addresses when doing address comparisons. Start by setting the IPV6ONLY flag on PF_INET6 RPC listener sockets. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- net/sunrpc/svcsock.c | 10 ++++------ 1 file changed, 4 insertions(+), 6 deletions(-) diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c index d00bc3307745..ac6cd65220c7 100644 --- a/net/sunrpc/svcsock.c +++ b/net/sunrpc/svcsock.c @@ -1144,13 +1144,11 @@ static struct svc_sock *svc_setup_socket(struct svc_serv *serv, svc_tcp_init(svsk, serv); /* - * We start one listener per sv_serv. We want AF_INET - * requests to be automatically shunted to our PF_INET6 - * listener using a mapped IPv4 address. Make sure - * no-one starts an equivalent IPv4 listener, which - * would steal our incoming connections. + * If this is a PF_INET6 listener, we want to avoid + * getting requests from IPv4 remotes. Those should + * be shunted to a PF_INET listener via rpcbind. */ - val = 0; + val = 1; if (inet->sk_family == PF_INET6) kernel_setsockopt(sock, SOL_IPV6, IPV6_V6ONLY, (char *)&val, sizeof(val)); From fc28decdc93633a65d54e42498e9e819d466329c Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Wed, 18 Mar 2009 20:46:51 -0400 Subject: [PATCH 45/58] SUNRPC: Use IPv4 loopback for registering AF_INET6 kernel RPC services The kernel uses an IPv6 loopback address when registering its AF_INET6 RPC services so that it can tell whether the local portmapper is actually IPv6-enabled. Since the legacy portmapper doesn't listen on IPv6, however, this causes a long timeout on older systems if the kernel happens to try creating and registering an AF_INET6 RPC service. Originally I wanted to use a connected transport (either TCP or connected UDP) so that the upcall would fail immediately if the portmapper wasn't listening on IPv6, but we never agreed on what transport to use. In the end, it's of little consequence to the kernel whether the local portmapper is listening on IPv6. It's only important whether the portmapper supports rpcbind v4. And the kernel can't tell that at all if it is sending requests via IPv6 -- the portmapper will just ignore them. So, send both rpcbind v2 and v4 SET/UNSET requests via IPv4 loopback to maintain better backwards compatibility between new kernels and legacy user space, and prevent multi-second hangs in some cases when the kernel attempts to register RPC services. This patch is part of a series that addresses http://bugzilla.kernel.org/show_bug.cgi?id=12256 Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- net/sunrpc/rpcb_clnt.c | 23 ++++++----------------- 1 file changed, 6 insertions(+), 17 deletions(-) diff --git a/net/sunrpc/rpcb_clnt.c b/net/sunrpc/rpcb_clnt.c index 2caa7edeeaba..ebce7a5976c9 100644 --- a/net/sunrpc/rpcb_clnt.c +++ b/net/sunrpc/rpcb_clnt.c @@ -124,12 +124,6 @@ static const struct sockaddr_in rpcb_inaddr_loopback = { .sin_port = htons(RPCBIND_PORT), }; -static const struct sockaddr_in6 rpcb_in6addr_loopback = { - .sin6_family = AF_INET6, - .sin6_addr = IN6ADDR_LOOPBACK_INIT, - .sin6_port = htons(RPCBIND_PORT), -}; - static struct rpc_clnt *rpcb_create_local(struct sockaddr *addr, size_t addrlen, u32 version) { @@ -176,9 +170,10 @@ static struct rpc_clnt *rpcb_create(char *hostname, struct sockaddr *srvaddr, return rpc_create(&args); } -static int rpcb_register_call(struct sockaddr *addr, size_t addrlen, - u32 version, struct rpc_message *msg) +static int rpcb_register_call(u32 version, struct rpc_message *msg) { + struct sockaddr *addr = (struct sockaddr *)&rpcb_inaddr_loopback; + size_t addrlen = sizeof(rpcb_inaddr_loopback); struct rpc_clnt *rpcb_clnt; int result, error = 0; @@ -254,9 +249,7 @@ int rpcb_register(u32 prog, u32 vers, int prot, unsigned short port) if (port) msg.rpc_proc = &rpcb_procedures2[RPCBPROC_SET]; - return rpcb_register_call((struct sockaddr *)&rpcb_inaddr_loopback, - sizeof(rpcb_inaddr_loopback), - RPCBVERS_2, &msg); + return rpcb_register_call(RPCBVERS_2, &msg); } /* @@ -284,9 +277,7 @@ static int rpcb_register_netid4(struct sockaddr_in *address_to_register, if (port) msg->rpc_proc = &rpcb_procedures4[RPCBPROC_SET]; - return rpcb_register_call((struct sockaddr *)&rpcb_inaddr_loopback, - sizeof(rpcb_inaddr_loopback), - RPCBVERS_4, msg); + return rpcb_register_call(RPCBVERS_4, msg); } /* @@ -318,9 +309,7 @@ static int rpcb_register_netid6(struct sockaddr_in6 *address_to_register, if (port) msg->rpc_proc = &rpcb_procedures4[RPCBPROC_SET]; - return rpcb_register_call((struct sockaddr *)&rpcb_in6addr_loopback, - sizeof(rpcb_in6addr_loopback), - RPCBVERS_4, msg); + return rpcb_register_call(RPCBVERS_4, msg); } /** From ba5c35e0c7e30b095636cd58b0854fdbd3c32947 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Wed, 18 Mar 2009 20:46:59 -0400 Subject: [PATCH 46/58] SUNRPC: Don't return EPROTONOSUPPORT in svc_register()'s helpers The RPC client returns -EPROTONOSUPPORT if there is a protocol version mismatch (ie the remote RPC server doesn't support the RPC protocol version sent by the client). Helpers for the svc_register() function return -EPROTONOSUPPORT if they don't recognize the passed-in IPPROTO_ value. These are two entirely different failure modes. Have the helpers return -ENOPROTOOPT instead of -EPROTONOSUPPORT. This will allow callers to determine more precisely what the underlying problem is, and decide to report or recover appropriately. This patch is part of a series that addresses http://bugzilla.kernel.org/show_bug.cgi?id=12256 Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- net/sunrpc/svc.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c index d72ff44826d8..17e0d7265dfd 100644 --- a/net/sunrpc/svc.c +++ b/net/sunrpc/svc.c @@ -749,7 +749,7 @@ static int __svc_rpcb_register4(const u32 program, const u32 version, netid = RPCBIND_NETID_TCP; break; default: - return -EPROTONOSUPPORT; + return -ENOPROTOOPT; } return rpcb_v4_register(program, version, @@ -785,7 +785,7 @@ static int __svc_rpcb_register6(const u32 program, const u32 version, netid = RPCBIND_NETID_TCP6; break; default: - return -EPROTONOSUPPORT; + return -ENOPROTOOPT; } return rpcb_v4_register(program, version, From 3aba45536fe8f92aa07bcdfd2fb1cf17eec7d786 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Wed, 18 Mar 2009 20:47:06 -0400 Subject: [PATCH 47/58] SUNRPC: Clean up address type casts in rpcb_v4_register() Clean up: Simplify rpcb_v4_register() and its helpers by moving the details of sockaddr type casting to rpcb_v4_register()'s helper functions. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- net/sunrpc/rpcb_clnt.c | 26 ++++++++++++-------------- 1 file changed, 12 insertions(+), 14 deletions(-) diff --git a/net/sunrpc/rpcb_clnt.c b/net/sunrpc/rpcb_clnt.c index ebce7a5976c9..44d0732ba874 100644 --- a/net/sunrpc/rpcb_clnt.c +++ b/net/sunrpc/rpcb_clnt.c @@ -170,7 +170,7 @@ static struct rpc_clnt *rpcb_create(char *hostname, struct sockaddr *srvaddr, return rpc_create(&args); } -static int rpcb_register_call(u32 version, struct rpc_message *msg) +static int rpcb_register_call(const u32 version, struct rpc_message *msg) { struct sockaddr *addr = (struct sockaddr *)&rpcb_inaddr_loopback; size_t addrlen = sizeof(rpcb_inaddr_loopback); @@ -255,17 +255,17 @@ int rpcb_register(u32 prog, u32 vers, int prot, unsigned short port) /* * Fill in AF_INET family-specific arguments to register */ -static int rpcb_register_netid4(struct sockaddr_in *address_to_register, +static int rpcb_register_netid4(const struct sockaddr *sap, struct rpc_message *msg) { + const struct sockaddr_in *sin = (const struct sockaddr_in *)sap; struct rpcbind_args *map = msg->rpc_argp; - unsigned short port = ntohs(address_to_register->sin_port); + unsigned short port = ntohs(sin->sin_port); char buf[32]; /* Construct AF_INET universal address */ snprintf(buf, sizeof(buf), "%pI4.%u.%u", - &address_to_register->sin_addr.s_addr, - port >> 8, port & 0xff); + &sin->sin_addr.s_addr, port >> 8, port & 0xff); map->r_addr = buf; dprintk("RPC: %sregistering [%u, %u, %s, '%s'] with " @@ -283,21 +283,21 @@ static int rpcb_register_netid4(struct sockaddr_in *address_to_register, /* * Fill in AF_INET6 family-specific arguments to register */ -static int rpcb_register_netid6(struct sockaddr_in6 *address_to_register, +static int rpcb_register_netid6(const struct sockaddr *sap, struct rpc_message *msg) { + const struct sockaddr_in6 *sin6 = (const struct sockaddr_in6 *)sap; struct rpcbind_args *map = msg->rpc_argp; - unsigned short port = ntohs(address_to_register->sin6_port); + unsigned short port = ntohs(sin6->sin6_port); char buf[64]; /* Construct AF_INET6 universal address */ - if (ipv6_addr_any(&address_to_register->sin6_addr)) + if (ipv6_addr_any(&sin6->sin6_addr)) snprintf(buf, sizeof(buf), "::.%u.%u", port >> 8, port & 0xff); else snprintf(buf, sizeof(buf), "%pI6.%u.%u", - &address_to_register->sin6_addr, - port >> 8, port & 0xff); + &sin6->sin6_addr, port >> 8, port & 0xff); map->r_addr = buf; dprintk("RPC: %sregistering [%u, %u, %s, '%s'] with " @@ -369,11 +369,9 @@ int rpcb_v4_register(const u32 program, const u32 version, switch (address->sa_family) { case AF_INET: - return rpcb_register_netid4((struct sockaddr_in *)address, - &msg); + return rpcb_register_netid4(address, &msg); case AF_INET6: - return rpcb_register_netid6((struct sockaddr_in6 *)address, - &msg); + return rpcb_register_netid6(address, &msg); } return -EAFNOSUPPORT; From 126e4bc3b3b446482696377f67a634c76eaf2e9c Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Wed, 18 Mar 2009 20:47:14 -0400 Subject: [PATCH 48/58] SUNRPC: rpcbind actually interprets r_owner string RFC 1833 has little to say about the contents of r_owner; it only specifies that it is a string, and states that it is used to control who can UNSET an entry. Our port of rpcbind (from Sun) assumes this string contains a numeric UID value, not alphabetical or symbolic characters, but checks this value only for AF_LOCAL RPCB_SET or RPCB_UNSET requests. In all other cases, rpcbind ignores the contents of the r_owner string. The reference user space implementation of rpcb_set(3) uses a numeric UID for all SET/UNSET requests (even via the network) and an empty string for all other requests. We emulate that behavior here to maintain bug-for-bug compatibility. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- net/sunrpc/rpcb_clnt.c | 13 ++++++++++--- 1 file changed, 10 insertions(+), 3 deletions(-) diff --git a/net/sunrpc/rpcb_clnt.c b/net/sunrpc/rpcb_clnt.c index 44d0732ba874..d550d0b967db 100644 --- a/net/sunrpc/rpcb_clnt.c +++ b/net/sunrpc/rpcb_clnt.c @@ -63,9 +63,16 @@ enum { * r_owner * * The "owner" is allowed to unset a service in the rpcbind database. - * We always use the following (arbitrary) fixed string. + * + * For AF_LOCAL SET/UNSET requests, rpcbind treats this string as a + * UID which it maps to a local user name via a password lookup. + * In all other cases it is ignored. + * + * For SET/UNSET requests, user space provides a value, even for + * network requests, and GETADDR uses an empty string. We follow + * those precedents here. */ -#define RPCB_OWNER_STRING "rpcb" +#define RPCB_OWNER_STRING "0" #define RPCB_MAXOWNERLEN sizeof(RPCB_OWNER_STRING) static void rpcb_getport_done(struct rpc_task *, void *); @@ -566,7 +573,7 @@ void rpcb_getport_async(struct rpc_task *task) map->r_xprt = xprt_get(xprt); map->r_netid = rpc_peeraddr2str(clnt, RPC_DISPLAY_NETID); map->r_addr = rpc_peeraddr2str(rpcb_clnt, RPC_DISPLAY_UNIVERSAL_ADDR); - map->r_owner = RPCB_OWNER_STRING; /* ignored for GETADDR */ + map->r_owner = ""; map->r_status = -EIO; child = rpcb_call_async(rpcb_clnt, map, proc); From 1673d0de40ab46cac3b456ad50e1c8d6a31bfd66 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Wed, 18 Mar 2009 20:47:21 -0400 Subject: [PATCH 49/58] SUNRPC: Allow callers to pass rpcb_v4_register a NULL address The user space TI-RPC library uses an empty string for the universal address when unregistering all target addresses for [program, version]. The kernel's rpcb client should behave the same way. Here, we are switching between several registration methods based on the protocol family of the incoming address. Rename the other rpcbind v4 registration functions to make it clear that they, as well, are switched on protocol family. In /etc/netconfig, this is either "inet" or "inet6". NB: The loopback protocol families are not supported in the kernel. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- net/sunrpc/rpcb_clnt.c | 38 ++++++++++++++++++++++++++++---------- 1 file changed, 28 insertions(+), 10 deletions(-) diff --git a/net/sunrpc/rpcb_clnt.c b/net/sunrpc/rpcb_clnt.c index d550d0b967db..8ea8907d4b8d 100644 --- a/net/sunrpc/rpcb_clnt.c +++ b/net/sunrpc/rpcb_clnt.c @@ -262,8 +262,8 @@ int rpcb_register(u32 prog, u32 vers, int prot, unsigned short port) /* * Fill in AF_INET family-specific arguments to register */ -static int rpcb_register_netid4(const struct sockaddr *sap, - struct rpc_message *msg) +static int rpcb_register_inet4(const struct sockaddr *sap, + struct rpc_message *msg) { const struct sockaddr_in *sin = (const struct sockaddr_in *)sap; struct rpcbind_args *map = msg->rpc_argp; @@ -290,8 +290,8 @@ static int rpcb_register_netid4(const struct sockaddr *sap, /* * Fill in AF_INET6 family-specific arguments to register */ -static int rpcb_register_netid6(const struct sockaddr *sap, - struct rpc_message *msg) +static int rpcb_register_inet6(const struct sockaddr *sap, + struct rpc_message *msg) { const struct sockaddr_in6 *sin6 = (const struct sockaddr_in6 *)sap; struct rpcbind_args *map = msg->rpc_argp; @@ -319,6 +319,20 @@ static int rpcb_register_netid6(const struct sockaddr *sap, return rpcb_register_call(RPCBVERS_4, msg); } +static int rpcb_unregister_all_protofamilies(struct rpc_message *msg) +{ + struct rpcbind_args *map = msg->rpc_argp; + + dprintk("RPC: unregistering [%u, %u, '%s'] with " + "local rpcbind\n", + map->r_prog, map->r_vers, map->r_netid); + + map->r_addr = ""; + msg->rpc_proc = &rpcb_procedures4[RPCBPROC_UNSET]; + + return rpcb_register_call(RPCBVERS_4, msg); +} + /** * rpcb_v4_register - set or unset a port registration with the local rpcbind * @program: RPC program number of service to (un)register @@ -336,10 +350,11 @@ static int rpcb_register_netid6(const struct sockaddr *sap, * invoke this function once for each [program, version, address, * netid] tuple they wish to advertise. * - * Callers may also unregister RPC services that are no longer - * available by setting the port number in the passed-in address - * to zero. Callers pass a netid of "" to unregister all - * transport netids associated with [program, version, address]. + * Callers may also unregister RPC services that are registered at a + * specific address by setting the port number in @address to zero. + * They may unregister all registered protocol families at once for + * a service by passing a NULL @address argument. If @netid is "" + * then all netids for [program, version, address] are unregistered. * * This function uses rpcbind protocol version 4 to contact the * local rpcbind daemon. The local rpcbind daemon must support @@ -374,11 +389,14 @@ int rpcb_v4_register(const u32 program, const u32 version, .rpc_argp = &map, }; + if (address == NULL) + return rpcb_unregister_all_protofamilies(&msg); + switch (address->sa_family) { case AF_INET: - return rpcb_register_netid4(address, &msg); + return rpcb_register_inet4(address, &msg); case AF_INET6: - return rpcb_register_netid6(address, &msg); + return rpcb_register_inet6(address, &msg); } return -EAFNOSUPPORT; From d5a8620f7c8a5bcade730e2fa1224191f289fb00 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Wed, 18 Mar 2009 20:47:29 -0400 Subject: [PATCH 50/58] SUNRPC: Simplify svc_unregister() Our initial implementation of svc_unregister() assumed that PMAP_UNSET cleared all rpcbind registrations for a [program, version] tuple. However, we now have evidence that PMAP_UNSET clears only "inet" entries, and not "inet6" entries, in the rpcbind database. For backwards compatibility with the legacy portmapper, the svc_unregister() function also must work if user space doesn't support rpcbind version 4 at all. Thus we'll send an rpcbind v4 UNSET, and if that fails, we'll send a PMAP_UNSET. This simplifies the code in svc_unregister() and provides better backwards compatibility with legacy user space that does not support rpcbind version 4. We can get rid of the conditional compilation in here as well. This patch is part of a series that addresses http://bugzilla.kernel.org/show_bug.cgi?id=12256 Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- net/sunrpc/svc.c | 39 ++++++++++++++++----------------------- 1 file changed, 16 insertions(+), 23 deletions(-) diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c index 17e0d7265dfd..bd0ee312dac9 100644 --- a/net/sunrpc/svc.c +++ b/net/sunrpc/svc.c @@ -896,38 +896,31 @@ int svc_register(const struct svc_serv *serv, const int family, return error; } -#ifdef CONFIG_SUNRPC_REGISTER_V4 - -static void __svc_unregister(const u32 program, const u32 version, - const char *progname) -{ - struct sockaddr_in6 sin6 = { - .sin6_family = AF_INET6, - .sin6_addr = IN6ADDR_ANY_INIT, - .sin6_port = 0, - }; - int error; - - error = rpcb_v4_register(program, version, - (struct sockaddr *)&sin6, ""); - dprintk("svc: %s(%sv%u), error %d\n", - __func__, progname, version, error); -} - -#else /* CONFIG_SUNRPC_REGISTER_V4 */ - +/* + * If user space is running rpcbind, it should take the v4 UNSET + * and clear everything for this [program, version]. If user space + * is running portmap, it will reject the v4 UNSET, but won't have + * any "inet6" entries anyway. So a PMAP_UNSET should be sufficient + * in this case to clear all existing entries for [program, version]. + */ static void __svc_unregister(const u32 program, const u32 version, const char *progname) { int error; - error = rpcb_register(program, version, 0, 0); + error = rpcb_v4_register(program, version, NULL, ""); + + /* + * User space didn't support rpcbind v4, so retry this + * request with the legacy rpcbind v2 protocol. + */ + if (error == -EPROTONOSUPPORT) + error = rpcb_register(program, version, 0, 0); + dprintk("svc: %s(%sv%u), error %d\n", __func__, progname, version, error); } -#endif /* CONFIG_SUNRPC_REGISTER_V4 */ - /* * All netids, bind addresses and ports registered for [program, version] * are removed from the local rpcbind database (if the service is not From cadc0fa534e51e20fdffe1623913c163a18d71b1 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Wed, 18 Mar 2009 20:47:36 -0400 Subject: [PATCH 51/58] SUNRPC: Simplify kernel RPC service registration The kernel registers RPC services with the local portmapper with an rpcbind SET upcall to the local portmapper. Traditionally, this used rpcbind v2 (PMAP), but registering RPC services that support IPv6 requires rpcbind v3 or v4. Since we now want separate PF_INET and PF_INET6 listeners for each kernel RPC service, svc_register() will do only one of those registrations at a time. For PF_INET, it tries an rpcb v4 SET upcall first; if that fails, it does a legacy portmap SET. This makes it entirely backwards compatible with legacy user space, but allows a proper v4 SET to be used if rpcbind is available. For PF_INET6, it does an rpcb v4 SET upcall. If that fails, it fails the registration, and thus the transport creation. This let's the kernel detect if user space is able to support IPv6 RPC services, and thus whether it should maintain a PF_INET6 listener for each service at all. This provides complete backwards compatibilty with legacy user space that only supports rpcbind v2. The only down-side is that registering a new kernel RPC service may take an extra exchange with the local portmapper on legacy systems, but this is an infrequent operation and is done over UDP (no lingering sockets in TIMEWAIT), so it shouldn't be consequential. This patch is part of a series that addresses http://bugzilla.kernel.org/show_bug.cgi?id=12256 Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- net/sunrpc/svc.c | 79 +++++++++++++++++++++--------------------------- 1 file changed, 34 insertions(+), 45 deletions(-) diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c index bd0ee312dac9..142f64745fba 100644 --- a/net/sunrpc/svc.c +++ b/net/sunrpc/svc.c @@ -718,8 +718,6 @@ svc_exit_thread(struct svc_rqst *rqstp) } EXPORT_SYMBOL_GPL(svc_exit_thread); -#ifdef CONFIG_SUNRPC_REGISTER_V4 - /* * Register an "inet" protocol family netid with the local * rpcbind daemon via an rpcbind v4 SET request. @@ -734,12 +732,13 @@ static int __svc_rpcb_register4(const u32 program, const u32 version, const unsigned short protocol, const unsigned short port) { - struct sockaddr_in sin = { + const struct sockaddr_in sin = { .sin_family = AF_INET, .sin_addr.s_addr = htonl(INADDR_ANY), .sin_port = htons(port), }; - char *netid; + const char *netid; + int error; switch (protocol) { case IPPROTO_UDP: @@ -752,10 +751,20 @@ static int __svc_rpcb_register4(const u32 program, const u32 version, return -ENOPROTOOPT; } - return rpcb_v4_register(program, version, - (struct sockaddr *)&sin, netid); + error = rpcb_v4_register(program, version, + (const struct sockaddr *)&sin, netid); + + /* + * User space didn't support rpcbind v4, so retry this + * registration request with the legacy rpcbind v2 protocol. + */ + if (error == -EPROTONOSUPPORT) + error = rpcb_register(program, version, protocol, port); + + return error; } +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) /* * Register an "inet6" protocol family netid with the local * rpcbind daemon via an rpcbind v4 SET request. @@ -770,12 +779,13 @@ static int __svc_rpcb_register6(const u32 program, const u32 version, const unsigned short protocol, const unsigned short port) { - struct sockaddr_in6 sin6 = { + const struct sockaddr_in6 sin6 = { .sin6_family = AF_INET6, .sin6_addr = IN6ADDR_ANY_INIT, .sin6_port = htons(port), }; - char *netid; + const char *netid; + int error; switch (protocol) { case IPPROTO_UDP: @@ -788,9 +798,19 @@ static int __svc_rpcb_register6(const u32 program, const u32 version, return -ENOPROTOOPT; } - return rpcb_v4_register(program, version, - (struct sockaddr *)&sin6, netid); + error = rpcb_v4_register(program, version, + (const struct sockaddr *)&sin6, netid); + + /* + * User space didn't support rpcbind version 4, so we won't + * use a PF_INET6 listener. + */ + if (error == -EPROTONOSUPPORT) + error = -EAFNOSUPPORT; + + return error; } +#endif /* defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) */ /* * Register a kernel RPC service via rpcbind version 4. @@ -809,48 +829,17 @@ static int __svc_register(const u32 program, const u32 version, case PF_INET: return __svc_rpcb_register4(program, version, protocol, port); + break; +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) case PF_INET6: - error = __svc_rpcb_register6(program, version, + return__svc_rpcb_register6(program, version, protocol, port); - if (error < 0) - return error; - - /* - * Work around bug in some versions of Linux rpcbind - * which don't allow registration of both inet and - * inet6 netids. - * - * Error return ignored for now. - */ - __svc_rpcb_register4(program, version, - protocol, port); - return 0; +#endif /* defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) */ } return -EAFNOSUPPORT; } -#else /* CONFIG_SUNRPC_REGISTER_V4 */ - -/* - * Register a kernel RPC service via rpcbind version 2. - * - * Returns zero on success; a negative errno value is returned - * if any error occurs. - */ -static int __svc_register(const u32 program, const u32 version, - const int family, - const unsigned short protocol, - const unsigned short port) -{ - if (family != PF_INET) - return -EAFNOSUPPORT; - - return rpcb_register(program, version, protocol, port); -} - -#endif /* CONFIG_SUNRPC_REGISTER_V4 */ - /** * svc_register - register an RPC service with the local portmapper * @serv: svc_serv struct for the service to register From 363f724cdd3d2ae554e261be995abdeb15f7bdd9 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Wed, 18 Mar 2009 20:47:44 -0400 Subject: [PATCH 52/58] SUNRPC: rpcb_register() should handle errors silently Move error reporting for RPC registration to rpcb_register's caller. This way the caller can choose to recover silently from certain errors, but report errors it does not recognize. Error reporting for kernel RPC service registration is now handled in one place. This patch is part of a series that addresses http://bugzilla.kernel.org/show_bug.cgi?id=12256 Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- net/sunrpc/rpcb_clnt.c | 2 +- net/sunrpc/svc.c | 18 +++++++++++------- 2 files changed, 12 insertions(+), 8 deletions(-) diff --git a/net/sunrpc/rpcb_clnt.c b/net/sunrpc/rpcb_clnt.c index 8ea8907d4b8d..beee6da33035 100644 --- a/net/sunrpc/rpcb_clnt.c +++ b/net/sunrpc/rpcb_clnt.c @@ -194,7 +194,7 @@ static int rpcb_register_call(const u32 version, struct rpc_message *msg) error = PTR_ERR(rpcb_clnt); if (error < 0) { - printk(KERN_WARNING "RPC: failed to contact local rpcbind " + dprintk("RPC: failed to contact local rpcbind " "server (errno %d).\n", -error); return error; } diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c index 142f64745fba..8ba654bdd608 100644 --- a/net/sunrpc/svc.c +++ b/net/sunrpc/svc.c @@ -818,26 +818,30 @@ static int __svc_rpcb_register6(const u32 program, const u32 version, * Returns zero on success; a negative errno value is returned * if any error occurs. */ -static int __svc_register(const u32 program, const u32 version, +static int __svc_register(const char *progname, + const u32 program, const u32 version, const int family, const unsigned short protocol, const unsigned short port) { - int error; + int error = -EAFNOSUPPORT; switch (family) { case PF_INET: - return __svc_rpcb_register4(program, version, + error = __svc_rpcb_register4(program, version, protocol, port); break; #if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) case PF_INET6: - return__svc_rpcb_register6(program, version, + error = __svc_rpcb_register6(program, version, protocol, port); #endif /* defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) */ } - return -EAFNOSUPPORT; + if (error < 0) + printk(KERN_WARNING "svc: failed to register %sv%u RPC " + "service (errno %d).\n", progname, version, -error); + return error; } /** @@ -875,8 +879,8 @@ int svc_register(const struct svc_serv *serv, const int family, if (progp->pg_vers[i]->vs_hidden) continue; - error = __svc_register(progp->pg_prog, i, - family, proto, port); + error = __svc_register(progp->pg_name, progp->pg_prog, + i, family, proto, port); if (error < 0) break; } From 9355982830ad67dca35e0f3d43319f3d438f82b4 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Wed, 18 Mar 2009 20:47:51 -0400 Subject: [PATCH 53/58] SUNRPC: Remove CONFIG_SUNRPC_REGISTER_V4 We just augmented the kernel's RPC service registration code so that it automatically adjusts to what is supported in user space. Thus we no longer need the kernel configuration option to enable registering RPC services with v4 -- it's all done automatically. This patch is part of a series that addresses http://bugzilla.kernel.org/show_bug.cgi?id=12256 Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- net/sunrpc/Kconfig | 22 ---------------------- 1 file changed, 22 deletions(-) diff --git a/net/sunrpc/Kconfig b/net/sunrpc/Kconfig index 5592883e1e4a..afd91c78ce8e 100644 --- a/net/sunrpc/Kconfig +++ b/net/sunrpc/Kconfig @@ -17,28 +17,6 @@ config SUNRPC_XPRT_RDMA If unsure, say N. -config SUNRPC_REGISTER_V4 - bool "Register local RPC services via rpcbind v4 (EXPERIMENTAL)" - depends on SUNRPC && EXPERIMENTAL - default n - help - Sun added support for registering RPC services at an IPv6 - address by creating two new versions of the rpcbind protocol - (RFC 1833). - - This option enables support in the kernel RPC server for - registering kernel RPC services via version 4 of the rpcbind - protocol. If you enable this option, you must run a portmapper - daemon that supports rpcbind protocol version 4. - - Serving NFS over IPv6 from knfsd (the kernel's NFS server) - requires that you enable this option and use a portmapper that - supports rpcbind version 4. - - If unsure, say N to get traditional behavior (register kernel - RPC services using only rpcbind version 2). Distributions - using the legacy Linux portmapper daemon must say N here. - config RPCSEC_GSS_KRB5 tristate "Secure RPC: Kerberos V mechanism (EXPERIMENTAL)" depends on SUNRPC && EXPERIMENTAL From eb16e907781a9da7f272a3e8284c26bc4e4aeb9d Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Wed, 18 Mar 2009 20:47:59 -0400 Subject: [PATCH 54/58] lockd: Start PF_INET6 listener only if IPv6 support is available Apparently a lot of people need to disable IPv6 completely on their distributor-built systems, which have CONFIG_IPV6_MODULE enabled at build time. They do this by blacklisting the ipv6.ko module. This causes the creation of the lockd service listener to fail if CONFIG_IPV6_MODULE is set, but the module cannot be loaded. Now that the kernel's PF_INET6 RPC listeners are completely separate from PF_INET listeners, we can always start PF_INET. Then lockd can try to start PF_INET6, but it isn't required to be available. Note this has the added benefit that NLM callbacks from AF_INET6 servers will never come from AF_INET remotes. We no longer have to worry about matching mapped IPv4 addresses to AF_INET when comparing addresses. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- fs/lockd/clntlock.c | 51 +-------------------------------------------- fs/lockd/svc.c | 30 ++++++++++++++++++-------- 2 files changed, 22 insertions(+), 59 deletions(-) diff --git a/fs/lockd/clntlock.c b/fs/lockd/clntlock.c index aedc47a264c1..1f3b0fc0d351 100644 --- a/fs/lockd/clntlock.c +++ b/fs/lockd/clntlock.c @@ -139,55 +139,6 @@ int nlmclnt_block(struct nlm_wait *block, struct nlm_rqst *req, long timeout) return 0; } -#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) -static const struct in6_addr *nlmclnt_map_v4addr(const struct sockaddr *sap, - struct in6_addr *addr_mapped) -{ - const struct sockaddr_in *sin = (const struct sockaddr_in *)sap; - - switch (sap->sa_family) { - case AF_INET6: - return &((const struct sockaddr_in6 *)sap)->sin6_addr; - case AF_INET: - ipv6_addr_set_v4mapped(sin->sin_addr.s_addr, addr_mapped); - return addr_mapped; - } - - return NULL; -} - -/* - * If lockd is using a PF_INET6 listener, all incoming requests appear - * to come from AF_INET6 remotes. The address of AF_INET remotes are - * mapped to AF_INET6 automatically by the network layer. In case the - * user passed an AF_INET server address at mount time, ensure both - * addresses are AF_INET6 before comparing them. - */ -static int nlmclnt_cmp_addr(const struct nlm_host *host, - const struct sockaddr *sap) -{ - const struct in6_addr *addr1; - const struct in6_addr *addr2; - struct in6_addr addr1_mapped; - struct in6_addr addr2_mapped; - - addr1 = nlmclnt_map_v4addr(nlm_addr(host), &addr1_mapped); - if (likely(addr1 != NULL)) { - addr2 = nlmclnt_map_v4addr(sap, &addr2_mapped); - if (likely(addr2 != NULL)) - return ipv6_addr_equal(addr1, addr2); - } - - return 0; -} -#else /* !(CONFIG_IPV6 || CONFIG_IPV6_MODULE) */ -static int nlmclnt_cmp_addr(const struct nlm_host *host, - const struct sockaddr *sap) -{ - return nlm_cmp_addr(nlm_addr(host), sap); -} -#endif /* !(CONFIG_IPV6 || CONFIG_IPV6_MODULE) */ - /* * The server lockd has called us back to tell us the lock was granted */ @@ -215,7 +166,7 @@ __be32 nlmclnt_grant(const struct sockaddr *addr, const struct nlm_lock *lock) */ if (fl_blocked->fl_u.nfs_fl.owner->pid != lock->svid) continue; - if (!nlmclnt_cmp_addr(block->b_host, addr)) + if (!nlm_cmp_addr(nlm_addr(block->b_host), addr)) continue; if (nfs_compare_fh(NFS_FH(fl_blocked->fl_file->f_path.dentry->d_inode) ,fh) != 0) continue; diff --git a/fs/lockd/svc.c b/fs/lockd/svc.c index 566932b98fd3..abf83881f68a 100644 --- a/fs/lockd/svc.c +++ b/fs/lockd/svc.c @@ -193,20 +193,30 @@ lockd(void *vrqstp) return 0; } -static int create_lockd_listener(struct svc_serv *serv, char *name, - unsigned short port) +static int create_lockd_listener(struct svc_serv *serv, const char *name, + const int family, const unsigned short port) { struct svc_xprt *xprt; - xprt = svc_find_xprt(serv, name, 0, 0); + xprt = svc_find_xprt(serv, name, family, 0); if (xprt == NULL) - return svc_create_xprt(serv, name, PF_INET, - port, SVC_SOCK_DEFAULTS); - + return svc_create_xprt(serv, name, family, port, + SVC_SOCK_DEFAULTS); svc_xprt_put(xprt); return 0; } +static int create_lockd_family(struct svc_serv *serv, const int family) +{ + int err; + + err = create_lockd_listener(serv, "udp", family, nlm_udpport); + if (err < 0) + return err; + + return create_lockd_listener(serv, "tcp", family, nlm_tcpport); +} + /* * Ensure there are active UDP and TCP listeners for lockd. * @@ -222,13 +232,15 @@ static int make_socks(struct svc_serv *serv) static int warned; int err; - err = create_lockd_listener(serv, "udp", nlm_udpport); + err = create_lockd_family(serv, PF_INET); if (err < 0) goto out_err; - err = create_lockd_listener(serv, "tcp", nlm_tcpport); - if (err < 0) +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + err = create_lockd_family(serv, PF_INET6); + if (err < 0 && err != -EAFNOSUPPORT) goto out_err; +#endif /* CONFIG_IPV6 || CONFIG_IPV6_MODULE */ warned = 0; return 0; From f738f5170367b367e38b2d75a413e7b3c52d46a5 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Wed, 18 Mar 2009 20:48:06 -0400 Subject: [PATCH 55/58] NFS: Start PF_INET6 callback listener only if IPv6 support is available Apparently a lot of people need to disable IPv6 completely on their distributor-built systems, which have CONFIG_IPV6_MODULE enabled at build time. They do this by blacklisting the ipv6.ko module. This causes the creation of the NFSv4 callback service listener to fail if CONFIG_IPV6_MODULE is set, but the module cannot be loaded. Now that the kernel's PF_INET6 RPC listeners are completely separate from PF_INET listeners, we can always start PF_INET. Then the NFS client can try to start a PF_INET6 listener, but it isn't required to be available. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- fs/nfs/callback.c | 12 ++++++++++++ fs/nfs/callback.h | 1 + fs/nfs/nfs4state.c | 10 ++++++++-- 3 files changed, 21 insertions(+), 2 deletions(-) diff --git a/fs/nfs/callback.c b/fs/nfs/callback.c index 0ef47dff89be..a886e692ddd0 100644 --- a/fs/nfs/callback.c +++ b/fs/nfs/callback.c @@ -38,6 +38,7 @@ static struct svc_program nfs4_callback_program; unsigned int nfs_callback_set_tcpport; unsigned short nfs_callback_tcpport; +unsigned short nfs_callback_tcpport6; static const int nfs_set_port_min = 0; static const int nfs_set_port_max = 65535; @@ -119,6 +120,17 @@ int nfs_callback_up(void) dprintk("NFS: Callback listener port = %u (af %u)\n", nfs_callback_tcpport, PF_INET); +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + ret = svc_create_xprt(serv, "tcp", PF_INET6, + nfs_callback_set_tcpport, SVC_SOCK_ANONYMOUS); + if (ret > 0) { + nfs_callback_tcpport6 = ret; + dprintk("NFS: Callback listener port = %u (af %u)\n", + nfs_callback_tcpport6, PF_INET6); + } else if (ret != -EAFNOSUPPORT) + goto out_err; +#endif /* defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) */ + nfs_callback_info.rqst = svc_prepare_thread(serv, &serv->sv_pools[0]); if (IS_ERR(nfs_callback_info.rqst)) { ret = PTR_ERR(nfs_callback_info.rqst); diff --git a/fs/nfs/callback.h b/fs/nfs/callback.h index bb25d2135ff1..e110e286a262 100644 --- a/fs/nfs/callback.h +++ b/fs/nfs/callback.h @@ -72,5 +72,6 @@ extern void nfs_callback_down(void); extern unsigned int nfs_callback_set_tcpport; extern unsigned short nfs_callback_tcpport; +extern unsigned short nfs_callback_tcpport6; #endif /* __LINUX_FS_NFS_CALLBACK_H */ diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c index 2022fe47966f..0298e909559f 100644 --- a/fs/nfs/nfs4state.c +++ b/fs/nfs/nfs4state.c @@ -62,8 +62,14 @@ static LIST_HEAD(nfs4_clientid_list); static int nfs4_init_client(struct nfs_client *clp, struct rpc_cred *cred) { - int status = nfs4_proc_setclientid(clp, NFS4_CALLBACK, - nfs_callback_tcpport, cred); + unsigned short port; + int status; + + port = nfs_callback_tcpport; + if (clp->cl_addr.ss_family == AF_INET6) + port = nfs_callback_tcpport6; + + status = nfs4_proc_setclientid(clp, NFS4_CALLBACK, port, cred); if (status == 0) status = nfs4_proc_setclientid_confirm(clp, cred); if (status == 0) From 3c8c45dfab78a1919f6f8a3ea46998c487eb7e12 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Wed, 18 Mar 2009 20:48:14 -0400 Subject: [PATCH 56/58] NFS: Simplify logic to compare socket addresses in client.c Callback requests from IPv4 servers are now always guaranteed to be AF_INET, and never mapped IPv4 AF_INET6 addresses. Both nfs_match_client() and nfs_find_client() can now share the same address comparison logic, so fold them together. We can also dispense with of most of the conditional compilation in here. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- fs/nfs/client.c | 118 ++++++++++++++++++++++-------------------------- 1 file changed, 53 insertions(+), 65 deletions(-) diff --git a/fs/nfs/client.c b/fs/nfs/client.c index 574158ae2398..855daac0f246 100644 --- a/fs/nfs/client.c +++ b/fs/nfs/client.c @@ -224,38 +224,6 @@ void nfs_put_client(struct nfs_client *clp) } #if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) -static const struct in6_addr *nfs_map_ipv4_addr(const struct sockaddr *sa, struct in6_addr *addr_mapped) -{ - switch (sa->sa_family) { - default: - return NULL; - case AF_INET6: - return &((const struct sockaddr_in6 *)sa)->sin6_addr; - break; - case AF_INET: - ipv6_addr_set_v4mapped(((const struct sockaddr_in *)sa)->sin_addr.s_addr, - addr_mapped); - return addr_mapped; - } -} - -static int nfs_sockaddr_match_ipaddr(const struct sockaddr *sa1, - const struct sockaddr *sa2) -{ - const struct in6_addr *addr1; - const struct in6_addr *addr2; - struct in6_addr addr1_mapped; - struct in6_addr addr2_mapped; - - addr1 = nfs_map_ipv4_addr(sa1, &addr1_mapped); - if (likely(addr1 != NULL)) { - addr2 = nfs_map_ipv4_addr(sa2, &addr2_mapped); - if (likely(addr2 != NULL)) - return ipv6_addr_equal(addr1, addr2); - } - return 0; -} - /* * Test if two ip6 socket addresses refer to the same socket by * comparing relevant fields. The padding bytes specifically, are not @@ -267,38 +235,21 @@ static int nfs_sockaddr_match_ipaddr(const struct sockaddr *sa1, * * The caller should ensure both socket addresses are AF_INET6. */ -static int nfs_sockaddr_cmp_ip6(const struct sockaddr *sa1, - const struct sockaddr *sa2) +static int nfs_sockaddr_match_ipaddr6(const struct sockaddr *sa1, + const struct sockaddr *sa2) { - const struct sockaddr_in6 *saddr1 = (const struct sockaddr_in6 *)sa1; - const struct sockaddr_in6 *saddr2 = (const struct sockaddr_in6 *)sa2; + const struct sockaddr_in6 *sin1 = (const struct sockaddr_in6 *)sa1; + const struct sockaddr_in6 *sin2 = (const struct sockaddr_in6 *)sa2; - if (!ipv6_addr_equal(&saddr1->sin6_addr, - &saddr1->sin6_addr)) + if (ipv6_addr_scope(&sin1->sin6_addr) == IPV6_ADDR_SCOPE_LINKLOCAL && + sin1->sin6_scope_id != sin2->sin6_scope_id) return 0; - if (ipv6_addr_scope(&saddr1->sin6_addr) == IPV6_ADDR_SCOPE_LINKLOCAL && - saddr1->sin6_scope_id != saddr2->sin6_scope_id) - return 0; - return saddr1->sin6_port == saddr2->sin6_port; -} -#else -static int nfs_sockaddr_match_ipaddr4(const struct sockaddr_in *sa1, - const struct sockaddr_in *sa2) -{ - return sa1->sin_addr.s_addr == sa2->sin_addr.s_addr; -} -static int nfs_sockaddr_match_ipaddr(const struct sockaddr *sa1, - const struct sockaddr *sa2) -{ - if (unlikely(sa1->sa_family != AF_INET || sa2->sa_family != AF_INET)) - return 0; - return nfs_sockaddr_match_ipaddr4((const struct sockaddr_in *)sa1, - (const struct sockaddr_in *)sa2); + return ipv6_addr_equal(&sin1->sin6_addr, &sin1->sin6_addr); } - -static int nfs_sockaddr_cmp_ip6(const struct sockaddr * sa1, - const struct sockaddr * sa2) +#else /* !defined(CONFIG_IPV6) && !defined(CONFIG_IPV6_MODULE) */ +static int nfs_sockaddr_match_ipaddr6(const struct sockaddr *sa1, + const struct sockaddr *sa2) { return 0; } @@ -311,20 +262,57 @@ static int nfs_sockaddr_cmp_ip6(const struct sockaddr * sa1, * * The caller should ensure both socket addresses are AF_INET. */ +static int nfs_sockaddr_match_ipaddr4(const struct sockaddr *sa1, + const struct sockaddr *sa2) +{ + const struct sockaddr_in *sin1 = (const struct sockaddr_in *)sa1; + const struct sockaddr_in *sin2 = (const struct sockaddr_in *)sa2; + + return sin1->sin_addr.s_addr == sin2->sin_addr.s_addr; +} + +static int nfs_sockaddr_cmp_ip6(const struct sockaddr *sa1, + const struct sockaddr *sa2) +{ + const struct sockaddr_in6 *sin1 = (const struct sockaddr_in6 *)sa1; + const struct sockaddr_in6 *sin2 = (const struct sockaddr_in6 *)sa2; + + return nfs_sockaddr_match_ipaddr6(sa1, sa2) && + (sin1->sin6_port == sin2->sin6_port); +} + static int nfs_sockaddr_cmp_ip4(const struct sockaddr *sa1, const struct sockaddr *sa2) { - const struct sockaddr_in *saddr1 = (const struct sockaddr_in *)sa1; - const struct sockaddr_in *saddr2 = (const struct sockaddr_in *)sa2; + const struct sockaddr_in *sin1 = (const struct sockaddr_in *)sa1; + const struct sockaddr_in *sin2 = (const struct sockaddr_in *)sa2; - if (saddr1->sin_addr.s_addr != saddr2->sin_addr.s_addr) - return 0; - return saddr1->sin_port == saddr2->sin_port; + return nfs_sockaddr_match_ipaddr4(sa1, sa2) && + (sin1->sin_port == sin2->sin_port); } /* * Test if two socket addresses represent the same actual socket, - * by comparing (only) relevant fields. + * by comparing (only) relevant fields, excluding the port number. + */ +static int nfs_sockaddr_match_ipaddr(const struct sockaddr *sa1, + const struct sockaddr *sa2) +{ + if (sa1->sa_family != sa2->sa_family) + return 0; + + switch (sa1->sa_family) { + case AF_INET: + return nfs_sockaddr_match_ipaddr4(sa1, sa2); + case AF_INET6: + return nfs_sockaddr_match_ipaddr6(sa1, sa2); + } + return 0; +} + +/* + * Test if two socket addresses represent the same actual socket, + * by comparing (only) relevant fields, including the port number. */ static int nfs_sockaddr_cmp(const struct sockaddr *sa1, const struct sockaddr *sa2) From ad5b365c1266b0c9e8e254a3c1cc4ef66bf33cba Mon Sep 17 00:00:00 2001 From: Mans Rullgard Date: Sat, 28 Mar 2009 19:55:20 +0000 Subject: [PATCH 57/58] NSM: Fix unaligned accesses in nsm_init_private() This fixes unaligned accesses in nsm_init_private() when creating nlm_reboot keys. Signed-off-by: Mans Rullgard Reviewed-by: Chuck Lever Signed-off-by: Trond Myklebust --- fs/lockd/mon.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/fs/lockd/mon.c b/fs/lockd/mon.c index 5e2c4d5ac827..6d5d4a4169e5 100644 --- a/fs/lockd/mon.c +++ b/fs/lockd/mon.c @@ -16,6 +16,8 @@ #include #include +#include + #define NLMDBG_FACILITY NLMDBG_MONITOR #define NSM_PROGRAM 100024 #define NSM_VERSION 1 @@ -274,10 +276,12 @@ static void nsm_init_private(struct nsm_handle *nsm) { u64 *p = (u64 *)&nsm->sm_priv.data; struct timespec ts; + s64 ns; ktime_get_ts(&ts); - *p++ = timespec_to_ns(&ts); - *p = (unsigned long)nsm; + ns = timespec_to_ns(&ts); + put_unaligned(ns, p); + put_unaligned((unsigned long)nsm, p + 1); } static struct nsm_handle *nsm_create_handle(const struct sockaddr *sap, From c69da774b28e01e062e0a3aba7509f2dcfd2a11a Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Mon, 30 Mar 2009 18:59:17 -0400 Subject: [PATCH 58/58] SUNRPC: Ensure IPV6_V6ONLY is set on the socket before binding to a port Also ensure that we use the protocol family instead of the address family when calling sock_create_kern(). Signed-off-by: Trond Myklebust --- net/sunrpc/svcsock.c | 38 +++++++++++++++++++++++++------------- 1 file changed, 25 insertions(+), 13 deletions(-) diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c index ac6cd65220c7..9d504234af4a 100644 --- a/net/sunrpc/svcsock.c +++ b/net/sunrpc/svcsock.c @@ -1110,7 +1110,6 @@ static struct svc_sock *svc_setup_socket(struct svc_serv *serv, struct svc_sock *svsk; struct sock *inet; int pmap_register = !(flags & SVC_SOCK_ANONYMOUS); - int val; dprintk("svc: svc_setup_socket %p\n", sock); if (!(svsk = kzalloc(sizeof(*svsk), GFP_KERNEL))) { @@ -1143,16 +1142,6 @@ static struct svc_sock *svc_setup_socket(struct svc_serv *serv, else svc_tcp_init(svsk, serv); - /* - * If this is a PF_INET6 listener, we want to avoid - * getting requests from IPv4 remotes. Those should - * be shunted to a PF_INET listener via rpcbind. - */ - val = 1; - if (inet->sk_family == PF_INET6) - kernel_setsockopt(sock, SOL_IPV6, IPV6_V6ONLY, - (char *)&val, sizeof(val)); - dprintk("svc: svc_setup_socket created %p (inet %p)\n", svsk, svsk->sk_sk); @@ -1220,6 +1209,8 @@ static struct svc_xprt *svc_create_socket(struct svc_serv *serv, struct sockaddr_storage addr; struct sockaddr *newsin = (struct sockaddr *)&addr; int newlen; + int family; + int val; RPC_IFDEBUG(char buf[RPC_MAX_ADDRBUFLEN]); dprintk("svc: svc_create_socket(%s, %d, %s)\n", @@ -1231,14 +1222,35 @@ static struct svc_xprt *svc_create_socket(struct svc_serv *serv, "sockets supported\n"); return ERR_PTR(-EINVAL); } - type = (protocol == IPPROTO_UDP)? SOCK_DGRAM : SOCK_STREAM; - error = sock_create_kern(sin->sa_family, type, protocol, &sock); + type = (protocol == IPPROTO_UDP)? SOCK_DGRAM : SOCK_STREAM; + switch (sin->sa_family) { + case AF_INET6: + family = PF_INET6; + break; + case AF_INET: + family = PF_INET; + break; + default: + return ERR_PTR(-EINVAL); + } + + error = sock_create_kern(family, type, protocol, &sock); if (error < 0) return ERR_PTR(error); svc_reclassify_socket(sock); + /* + * If this is an PF_INET6 listener, we want to avoid + * getting requests from IPv4 remotes. Those should + * be shunted to a PF_INET listener via rpcbind. + */ + val = 1; + if (family == PF_INET6) + kernel_setsockopt(sock, SOL_IPV6, IPV6_V6ONLY, + (char *)&val, sizeof(val)); + if (type == SOCK_STREAM) sock->sk->sk_reuse = 1; /* allow address reuse */ error = kernel_bind(sock, sin, len);