Commit dc59250c authored by Chuck Lever's avatar Chuck Lever Committed by Linus Torvalds

[PATCH] NFS: Introduce the use of inode->i_lock to protect fields in nfsi

Down the road we want to eliminate the use of the global kernel lock entirely
from the NFS client.  To do this, we need to protect the fields in the
nfs_inode structure adequately.  Start by serializing updates to the
"cache_validity" field.

Note this change addresses an SMP hang found by njw@osdl.org, where processes
deadlock because nfs_end_data_update and nfs_revalidate_mapping update the
"cache_validity" field without proper serialization.

Test plan:
 Millions of fsx ops on SMP clients.  Run Nick Wilson's breaknfs program on
 large SMP clients.
Signed-off-by: default avatarChuck Lever <cel@netapp.com>
Cc: Trond Myklebust <trond.myklebust@fys.uio.no>
Signed-off-by: default avatarAndrew Morton <akpm@osdl.org>
Signed-off-by: default avatarLinus Torvalds <torvalds@osdl.org>
parent 412d582e
...@@ -189,7 +189,9 @@ int nfs_readdir_filler(nfs_readdir_descriptor_t *desc, struct page *page) ...@@ -189,7 +189,9 @@ int nfs_readdir_filler(nfs_readdir_descriptor_t *desc, struct page *page)
goto error; goto error;
} }
SetPageUptodate(page); SetPageUptodate(page);
spin_lock(&inode->i_lock);
NFS_I(inode)->cache_validity |= NFS_INO_INVALID_ATIME; NFS_I(inode)->cache_validity |= NFS_INO_INVALID_ATIME;
spin_unlock(&inode->i_lock);
/* Ensure consistent page alignment of the data. /* Ensure consistent page alignment of the data.
* Note: assumes we have exclusive access to this mapping either * Note: assumes we have exclusive access to this mapping either
* through inode->i_sem or some other mechanism. * through inode->i_sem or some other mechanism.
...@@ -462,7 +464,9 @@ int uncached_readdir(nfs_readdir_descriptor_t *desc, void *dirent, ...@@ -462,7 +464,9 @@ int uncached_readdir(nfs_readdir_descriptor_t *desc, void *dirent,
page, page,
NFS_SERVER(inode)->dtsize, NFS_SERVER(inode)->dtsize,
desc->plus); desc->plus);
spin_lock(&inode->i_lock);
NFS_I(inode)->cache_validity |= NFS_INO_INVALID_ATIME; NFS_I(inode)->cache_validity |= NFS_INO_INVALID_ATIME;
spin_unlock(&inode->i_lock);
desc->page = page; desc->page = page;
desc->ptr = kmap(page); /* matching kunmap in nfs_do_filldir */ desc->ptr = kmap(page); /* matching kunmap in nfs_do_filldir */
if (desc->error >= 0) { if (desc->error >= 0) {
...@@ -1596,7 +1600,10 @@ void nfs_access_add_cache(struct inode *inode, struct nfs_access_entry *set) ...@@ -1596,7 +1600,10 @@ void nfs_access_add_cache(struct inode *inode, struct nfs_access_entry *set)
put_rpccred(cache->cred); put_rpccred(cache->cred);
cache->cred = get_rpccred(set->cred); cache->cred = get_rpccred(set->cred);
} }
/* FIXME: replace current access_cache BKL reliance with inode->i_lock */
spin_lock(&inode->i_lock);
nfsi->cache_validity &= ~NFS_INO_INVALID_ACCESS; nfsi->cache_validity &= ~NFS_INO_INVALID_ACCESS;
spin_unlock(&inode->i_lock);
cache->jiffies = set->jiffies; cache->jiffies = set->jiffies;
cache->mask = set->mask; cache->mask = set->mask;
} }
......
...@@ -615,6 +615,8 @@ nfs_zap_caches(struct inode *inode) ...@@ -615,6 +615,8 @@ nfs_zap_caches(struct inode *inode)
struct nfs_inode *nfsi = NFS_I(inode); struct nfs_inode *nfsi = NFS_I(inode);
int mode = inode->i_mode; int mode = inode->i_mode;
spin_lock(&inode->i_lock);
NFS_ATTRTIMEO(inode) = NFS_MINATTRTIMEO(inode); NFS_ATTRTIMEO(inode) = NFS_MINATTRTIMEO(inode);
NFS_ATTRTIMEO_UPDATE(inode) = jiffies; NFS_ATTRTIMEO_UPDATE(inode) = jiffies;
...@@ -623,6 +625,8 @@ nfs_zap_caches(struct inode *inode) ...@@ -623,6 +625,8 @@ nfs_zap_caches(struct inode *inode)
nfsi->cache_validity |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL|NFS_INO_REVAL_PAGECACHE; nfsi->cache_validity |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL|NFS_INO_REVAL_PAGECACHE;
else else
nfsi->cache_validity |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL|NFS_INO_REVAL_PAGECACHE; nfsi->cache_validity |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL|NFS_INO_REVAL_PAGECACHE;
spin_unlock(&inode->i_lock);
} }
static void nfs_zap_acl_cache(struct inode *inode) static void nfs_zap_acl_cache(struct inode *inode)
...@@ -632,7 +636,9 @@ static void nfs_zap_acl_cache(struct inode *inode) ...@@ -632,7 +636,9 @@ static void nfs_zap_acl_cache(struct inode *inode)
clear_acl_cache = NFS_PROTO(inode)->clear_acl_cache; clear_acl_cache = NFS_PROTO(inode)->clear_acl_cache;
if (clear_acl_cache != NULL) if (clear_acl_cache != NULL)
clear_acl_cache(inode); clear_acl_cache(inode);
spin_lock(&inode->i_lock);
NFS_I(inode)->cache_validity &= ~NFS_INO_INVALID_ACL; NFS_I(inode)->cache_validity &= ~NFS_INO_INVALID_ACL;
spin_unlock(&inode->i_lock);
} }
/* /*
...@@ -841,7 +847,9 @@ void nfs_setattr_update_inode(struct inode *inode, struct iattr *attr) ...@@ -841,7 +847,9 @@ void nfs_setattr_update_inode(struct inode *inode, struct iattr *attr)
inode->i_uid = attr->ia_uid; inode->i_uid = attr->ia_uid;
if ((attr->ia_valid & ATTR_GID) != 0) if ((attr->ia_valid & ATTR_GID) != 0)
inode->i_gid = attr->ia_gid; inode->i_gid = attr->ia_gid;
spin_lock(&inode->i_lock);
NFS_I(inode)->cache_validity |= NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL; NFS_I(inode)->cache_validity |= NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL;
spin_unlock(&inode->i_lock);
} }
if ((attr->ia_valid & ATTR_SIZE) != 0) { if ((attr->ia_valid & ATTR_SIZE) != 0) {
inode->i_size = attr->ia_size; inode->i_size = attr->ia_size;
...@@ -1082,6 +1090,7 @@ __nfs_revalidate_inode(struct nfs_server *server, struct inode *inode) ...@@ -1082,6 +1090,7 @@ __nfs_revalidate_inode(struct nfs_server *server, struct inode *inode)
(long long)NFS_FILEID(inode), status); (long long)NFS_FILEID(inode), status);
goto out; goto out;
} }
spin_lock(&inode->i_lock);
cache_validity = nfsi->cache_validity; cache_validity = nfsi->cache_validity;
nfsi->cache_validity &= ~NFS_INO_REVAL_PAGECACHE; nfsi->cache_validity &= ~NFS_INO_REVAL_PAGECACHE;
...@@ -1091,6 +1100,7 @@ __nfs_revalidate_inode(struct nfs_server *server, struct inode *inode) ...@@ -1091,6 +1100,7 @@ __nfs_revalidate_inode(struct nfs_server *server, struct inode *inode)
*/ */
if (verifier == nfsi->cache_change_attribute) if (verifier == nfsi->cache_change_attribute)
nfsi->cache_validity &= ~(NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ATIME); nfsi->cache_validity &= ~(NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ATIME);
spin_unlock(&inode->i_lock);
nfs_revalidate_mapping(inode, inode->i_mapping); nfs_revalidate_mapping(inode, inode->i_mapping);
...@@ -1149,12 +1159,16 @@ void nfs_revalidate_mapping(struct inode *inode, struct address_space *mapping) ...@@ -1149,12 +1159,16 @@ void nfs_revalidate_mapping(struct inode *inode, struct address_space *mapping)
nfs_wb_all(inode); nfs_wb_all(inode);
} }
invalidate_inode_pages2(mapping); invalidate_inode_pages2(mapping);
spin_lock(&inode->i_lock);
nfsi->cache_validity &= ~NFS_INO_INVALID_DATA; nfsi->cache_validity &= ~NFS_INO_INVALID_DATA;
if (S_ISDIR(inode->i_mode)) { if (S_ISDIR(inode->i_mode)) {
memset(nfsi->cookieverf, 0, sizeof(nfsi->cookieverf)); memset(nfsi->cookieverf, 0, sizeof(nfsi->cookieverf));
/* This ensures we revalidate child dentries */ /* This ensures we revalidate child dentries */
nfsi->cache_change_attribute++; nfsi->cache_change_attribute++;
} }
spin_unlock(&inode->i_lock);
dfprintk(PAGECACHE, "NFS: (%s/%Ld) data cache invalidated\n", dfprintk(PAGECACHE, "NFS: (%s/%Ld) data cache invalidated\n",
inode->i_sb->s_id, inode->i_sb->s_id,
(long long)NFS_FILEID(inode)); (long long)NFS_FILEID(inode));
...@@ -1184,10 +1198,12 @@ void nfs_end_data_update(struct inode *inode) ...@@ -1184,10 +1198,12 @@ void nfs_end_data_update(struct inode *inode)
if (!nfs_have_delegation(inode, FMODE_READ)) { if (!nfs_have_delegation(inode, FMODE_READ)) {
/* Mark the attribute cache for revalidation */ /* Mark the attribute cache for revalidation */
spin_lock(&inode->i_lock);
nfsi->cache_validity |= NFS_INO_INVALID_ATTR; nfsi->cache_validity |= NFS_INO_INVALID_ATTR;
/* Directories and symlinks: invalidate page cache too */ /* Directories and symlinks: invalidate page cache too */
if (S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode)) if (S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode))
nfsi->cache_validity |= NFS_INO_INVALID_DATA; nfsi->cache_validity |= NFS_INO_INVALID_DATA;
spin_unlock(&inode->i_lock);
} }
nfsi->cache_change_attribute ++; nfsi->cache_change_attribute ++;
atomic_dec(&nfsi->data_updates); atomic_dec(&nfsi->data_updates);
...@@ -1212,6 +1228,8 @@ int nfs_refresh_inode(struct inode *inode, struct nfs_fattr *fattr) ...@@ -1212,6 +1228,8 @@ int nfs_refresh_inode(struct inode *inode, struct nfs_fattr *fattr)
if (nfs_have_delegation(inode, FMODE_READ)) if (nfs_have_delegation(inode, FMODE_READ))
return 0; return 0;
spin_lock(&inode->i_lock);
/* Are we in the process of updating data on the server? */ /* Are we in the process of updating data on the server? */
data_unstable = nfs_caches_unstable(inode); data_unstable = nfs_caches_unstable(inode);
...@@ -1226,13 +1244,17 @@ int nfs_refresh_inode(struct inode *inode, struct nfs_fattr *fattr) ...@@ -1226,13 +1244,17 @@ int nfs_refresh_inode(struct inode *inode, struct nfs_fattr *fattr)
} }
} }
if ((fattr->valid & NFS_ATTR_FATTR) == 0) if ((fattr->valid & NFS_ATTR_FATTR) == 0) {
spin_unlock(&inode->i_lock);
return 0; return 0;
}
/* Has the inode gone and changed behind our back? */ /* Has the inode gone and changed behind our back? */
if (nfsi->fileid != fattr->fileid if (nfsi->fileid != fattr->fileid
|| (inode->i_mode & S_IFMT) != (fattr->mode & S_IFMT)) || (inode->i_mode & S_IFMT) != (fattr->mode & S_IFMT)) {
spin_unlock(&inode->i_lock);
return -EIO; return -EIO;
}
cur_size = i_size_read(inode); cur_size = i_size_read(inode);
new_isize = nfs_size_to_loff_t(fattr->size); new_isize = nfs_size_to_loff_t(fattr->size);
...@@ -1271,6 +1293,7 @@ int nfs_refresh_inode(struct inode *inode, struct nfs_fattr *fattr) ...@@ -1271,6 +1293,7 @@ int nfs_refresh_inode(struct inode *inode, struct nfs_fattr *fattr)
nfsi->cache_validity |= NFS_INO_INVALID_ATIME; nfsi->cache_validity |= NFS_INO_INVALID_ATIME;
nfsi->read_cache_jiffies = fattr->timestamp; nfsi->read_cache_jiffies = fattr->timestamp;
spin_unlock(&inode->i_lock);
return 0; return 0;
} }
...@@ -1309,11 +1332,15 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr, unsign ...@@ -1309,11 +1332,15 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr, unsign
goto out_err; goto out_err;
} }
spin_lock(&inode->i_lock);
/* /*
* Make sure the inode's type hasn't changed. * Make sure the inode's type hasn't changed.
*/ */
if ((inode->i_mode & S_IFMT) != (fattr->mode & S_IFMT)) if ((inode->i_mode & S_IFMT) != (fattr->mode & S_IFMT)) {
spin_unlock(&inode->i_lock);
goto out_changed; goto out_changed;
}
/* /*
* Update the read time so we don't revalidate too often. * Update the read time so we don't revalidate too often.
...@@ -1406,6 +1433,7 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr, unsign ...@@ -1406,6 +1433,7 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr, unsign
if (!nfs_have_delegation(inode, FMODE_READ)) if (!nfs_have_delegation(inode, FMODE_READ))
nfsi->cache_validity |= invalid; nfsi->cache_validity |= invalid;
spin_unlock(&inode->i_lock);
return 0; return 0;
out_changed: out_changed:
/* /*
......
...@@ -308,7 +308,9 @@ static int nfs3_proc_setacls(struct inode *inode, struct posix_acl *acl, ...@@ -308,7 +308,9 @@ static int nfs3_proc_setacls(struct inode *inode, struct posix_acl *acl,
nfs_begin_data_update(inode); nfs_begin_data_update(inode);
status = rpc_call(server->client_acl, ACLPROC3_SETACL, status = rpc_call(server->client_acl, ACLPROC3_SETACL,
&args, &fattr, 0); &args, &fattr, 0);
spin_lock(&inode->i_lock);
NFS_I(inode)->cache_validity |= NFS_INO_INVALID_ACCESS; NFS_I(inode)->cache_validity |= NFS_INO_INVALID_ACCESS;
spin_unlock(&inode->i_lock);
nfs_end_data_update(inode); nfs_end_data_update(inode);
dprintk("NFS reply setacl: %d\n", status); dprintk("NFS reply setacl: %d\n", status);
......
...@@ -140,7 +140,9 @@ static int nfs_readpage_sync(struct nfs_open_context *ctx, struct inode *inode, ...@@ -140,7 +140,9 @@ static int nfs_readpage_sync(struct nfs_open_context *ctx, struct inode *inode,
if (rdata->res.eof != 0 || result == 0) if (rdata->res.eof != 0 || result == 0)
break; break;
} while (count); } while (count);
spin_lock(&inode->i_lock);
NFS_I(inode)->cache_validity |= NFS_INO_INVALID_ATIME; NFS_I(inode)->cache_validity |= NFS_INO_INVALID_ATIME;
spin_unlock(&inode->i_lock);
if (count) if (count)
memclear_highpage_flush(page, rdata->args.pgbase, count); memclear_highpage_flush(page, rdata->args.pgbase, count);
...@@ -473,7 +475,9 @@ void nfs_readpage_result(struct rpc_task *task) ...@@ -473,7 +475,9 @@ void nfs_readpage_result(struct rpc_task *task)
} }
task->tk_status = -EIO; task->tk_status = -EIO;
} }
spin_lock(&data->inode->i_lock);
NFS_I(data->inode)->cache_validity |= NFS_INO_INVALID_ATIME; NFS_I(data->inode)->cache_validity |= NFS_INO_INVALID_ATIME;
spin_unlock(&data->inode->i_lock);
data->complete(data, status); data->complete(data, status);
} }
......
...@@ -238,8 +238,11 @@ static inline int nfs_caches_unstable(struct inode *inode) ...@@ -238,8 +238,11 @@ static inline int nfs_caches_unstable(struct inode *inode)
static inline void NFS_CACHEINV(struct inode *inode) static inline void NFS_CACHEINV(struct inode *inode)
{ {
if (!nfs_caches_unstable(inode)) if (!nfs_caches_unstable(inode)) {
spin_lock(&inode->i_lock);
NFS_I(inode)->cache_validity |= NFS_INO_INVALID_ATTR | NFS_INO_INVALID_ACCESS; NFS_I(inode)->cache_validity |= NFS_INO_INVALID_ATTR | NFS_INO_INVALID_ACCESS;
spin_unlock(&inode->i_lock);
}
} }
static inline int nfs_server_capable(struct inode *inode, int cap) static inline int nfs_server_capable(struct inode *inode, int cap)
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment