diff options
author | Linus Torvalds <torvalds@linux-foundation.org> | 2014-08-13 18:13:19 -0600 |
---|---|---|
committer | Linus Torvalds <torvalds@linux-foundation.org> | 2014-08-13 18:13:19 -0600 |
commit | 06b8ab55289345ab191bf4bf0e4acc6d4bdf293d (patch) | |
tree | 9af9215097e26c026f30a58c6ca3092ec15d1e1e /fs | |
parent | dc1cc85133120e49c223f36aa77d398b8abac727 (diff) | |
parent | 71a6ec8ac587418ceb6b420def1ca44b334c1ff7 (diff) |
Merge tag 'nfs-for-3.17-1' of git://git.linux-nfs.org/projects/trondmy/linux-nfs
Pull NFS client updates from Trond Myklebust:
"Highlights include:
- stable fix for a bug in nfs3_list_one_acl()
- speed up NFS path walks by supporting LOOKUP_RCU
- more read/write code cleanups
- pNFS fixes for layout return on close
- fixes for the RCU handling in the rpcsec_gss code
- more NFS/RDMA fixes"
* tag 'nfs-for-3.17-1' of git://git.linux-nfs.org/projects/trondmy/linux-nfs: (79 commits)
nfs: reject changes to resvport and sharecache during remount
NFS: Avoid infinite loop when RELEASE_LOCKOWNER getting expired error
SUNRPC: remove all refcounting of groupinfo from rpcauth_lookupcred
NFS: fix two problems in lookup_revalidate in RCU-walk
NFS: allow lockless access to access_cache
NFS: teach nfs_lookup_verify_inode to handle LOOKUP_RCU
NFS: teach nfs_neg_need_reval to understand LOOKUP_RCU
NFS: support RCU_WALK in nfs_permission()
sunrpc/auth: allow lockless (rcu) lookup of credential cache.
NFS: prepare for RCU-walk support but pushing tests later in code.
NFS: nfs4_lookup_revalidate: only evaluate parent if it will be used.
NFS: add checks for returned value of try_module_get()
nfs: clear_request_commit while holding i_lock
pnfs: add pnfs_put_lseg_async
pnfs: find swapped pages on pnfs commit lists too
nfs: fix comment and add warn_on for PG_INODE_REF
nfs: check wait_on_bit_lock err in page_group_lock
sunrpc: remove "ec" argument from encrypt_v2 operation
sunrpc: clean up sparse endianness warnings in gss_krb5_wrap.c
sunrpc: clean up sparse endianness warnings in gss_krb5_seal.c
...
Diffstat (limited to 'fs')
-rw-r--r-- | fs/nfs/blocklayout/blocklayout.c | 101 | ||||
-rw-r--r-- | fs/nfs/callback.c | 12 | ||||
-rw-r--r-- | fs/nfs/client.c | 18 | ||||
-rw-r--r-- | fs/nfs/delegation.c | 34 | ||||
-rw-r--r-- | fs/nfs/delegation.h | 1 | ||||
-rw-r--r-- | fs/nfs/dir.c | 208 | ||||
-rw-r--r-- | fs/nfs/direct.c | 33 | ||||
-rw-r--r-- | fs/nfs/filelayout/filelayout.c | 298 | ||||
-rw-r--r-- | fs/nfs/filelayout/filelayoutdev.c | 2 | ||||
-rw-r--r-- | fs/nfs/inode.c | 9 | ||||
-rw-r--r-- | fs/nfs/internal.h | 11 | ||||
-rw-r--r-- | fs/nfs/nfs3acl.c | 2 | ||||
-rw-r--r-- | fs/nfs/nfs3proc.c | 21 | ||||
-rw-r--r-- | fs/nfs/nfs4_fs.h | 32 | ||||
-rw-r--r-- | fs/nfs/nfs4client.c | 5 | ||||
-rw-r--r-- | fs/nfs/nfs4proc.c | 248 | ||||
-rw-r--r-- | fs/nfs/nfs4state.c | 69 | ||||
-rw-r--r-- | fs/nfs/nfs4trace.h | 28 | ||||
-rw-r--r-- | fs/nfs/nfs4xdr.c | 2 | ||||
-rw-r--r-- | fs/nfs/objlayout/objio_osd.c | 24 | ||||
-rw-r--r-- | fs/nfs/objlayout/objlayout.c | 81 | ||||
-rw-r--r-- | fs/nfs/objlayout/objlayout.h | 8 | ||||
-rw-r--r-- | fs/nfs/pagelist.c | 276 | ||||
-rw-r--r-- | fs/nfs/pnfs.c | 178 | ||||
-rw-r--r-- | fs/nfs/pnfs.h | 45 | ||||
-rw-r--r-- | fs/nfs/proc.c | 27 | ||||
-rw-r--r-- | fs/nfs/read.c | 54 | ||||
-rw-r--r-- | fs/nfs/super.c | 12 | ||||
-rw-r--r-- | fs/nfs/write.c | 150 | ||||
-rw-r--r-- | fs/nfs_common/nfsacl.c | 5 |
30 files changed, 1124 insertions, 870 deletions
diff --git a/fs/nfs/blocklayout/blocklayout.c b/fs/nfs/blocklayout/blocklayout.c index 9b431f44fad9..cbb1797149d5 100644 --- a/fs/nfs/blocklayout/blocklayout.c +++ b/fs/nfs/blocklayout/blocklayout.c @@ -210,8 +210,7 @@ static void bl_end_io_read(struct bio *bio, int err) SetPageUptodate(bvec->bv_page); if (err) { - struct nfs_pgio_data *rdata = par->data; - struct nfs_pgio_header *header = rdata->header; + struct nfs_pgio_header *header = par->data; if (!header->pnfs_error) header->pnfs_error = -EIO; @@ -224,43 +223,44 @@ static void bl_end_io_read(struct bio *bio, int err) static void bl_read_cleanup(struct work_struct *work) { struct rpc_task *task; - struct nfs_pgio_data *rdata; + struct nfs_pgio_header *hdr; dprintk("%s enter\n", __func__); task = container_of(work, struct rpc_task, u.tk_work); - rdata = container_of(task, struct nfs_pgio_data, task); - pnfs_ld_read_done(rdata); + hdr = container_of(task, struct nfs_pgio_header, task); + pnfs_ld_read_done(hdr); } static void bl_end_par_io_read(void *data, int unused) { - struct nfs_pgio_data *rdata = data; + struct nfs_pgio_header *hdr = data; - rdata->task.tk_status = rdata->header->pnfs_error; - INIT_WORK(&rdata->task.u.tk_work, bl_read_cleanup); - schedule_work(&rdata->task.u.tk_work); + hdr->task.tk_status = hdr->pnfs_error; + INIT_WORK(&hdr->task.u.tk_work, bl_read_cleanup); + schedule_work(&hdr->task.u.tk_work); } static enum pnfs_try_status -bl_read_pagelist(struct nfs_pgio_data *rdata) +bl_read_pagelist(struct nfs_pgio_header *hdr) { - struct nfs_pgio_header *header = rdata->header; + struct nfs_pgio_header *header = hdr; int i, hole; struct bio *bio = NULL; struct pnfs_block_extent *be = NULL, *cow_read = NULL; sector_t isect, extent_length = 0; struct parallel_io *par; - loff_t f_offset = rdata->args.offset; - size_t bytes_left = rdata->args.count; + loff_t f_offset = hdr->args.offset; + size_t bytes_left = hdr->args.count; unsigned int pg_offset, pg_len; - struct page **pages = rdata->args.pages; - int pg_index = rdata->args.pgbase >> PAGE_CACHE_SHIFT; + struct page **pages = hdr->args.pages; + int pg_index = hdr->args.pgbase >> PAGE_CACHE_SHIFT; const bool is_dio = (header->dreq != NULL); dprintk("%s enter nr_pages %u offset %lld count %u\n", __func__, - rdata->pages.npages, f_offset, (unsigned int)rdata->args.count); + hdr->page_array.npages, f_offset, + (unsigned int)hdr->args.count); - par = alloc_parallel(rdata); + par = alloc_parallel(hdr); if (!par) goto use_mds; par->pnfs_callback = bl_end_par_io_read; @@ -268,7 +268,7 @@ bl_read_pagelist(struct nfs_pgio_data *rdata) isect = (sector_t) (f_offset >> SECTOR_SHIFT); /* Code assumes extents are page-aligned */ - for (i = pg_index; i < rdata->pages.npages; i++) { + for (i = pg_index; i < hdr->page_array.npages; i++) { if (!extent_length) { /* We've used up the previous extent */ bl_put_extent(be); @@ -317,7 +317,8 @@ bl_read_pagelist(struct nfs_pgio_data *rdata) struct pnfs_block_extent *be_read; be_read = (hole && cow_read) ? cow_read : be; - bio = do_add_page_to_bio(bio, rdata->pages.npages - i, + bio = do_add_page_to_bio(bio, + hdr->page_array.npages - i, READ, isect, pages[i], be_read, bl_end_io_read, par, @@ -332,10 +333,10 @@ bl_read_pagelist(struct nfs_pgio_data *rdata) extent_length -= PAGE_CACHE_SECTORS; } if ((isect << SECTOR_SHIFT) >= header->inode->i_size) { - rdata->res.eof = 1; - rdata->res.count = header->inode->i_size - rdata->args.offset; + hdr->res.eof = 1; + hdr->res.count = header->inode->i_size - hdr->args.offset; } else { - rdata->res.count = (isect << SECTOR_SHIFT) - rdata->args.offset; + hdr->res.count = (isect << SECTOR_SHIFT) - hdr->args.offset; } out: bl_put_extent(be); @@ -390,8 +391,7 @@ static void bl_end_io_write_zero(struct bio *bio, int err) } if (unlikely(err)) { - struct nfs_pgio_data *data = par->data; - struct nfs_pgio_header *header = data->header; + struct nfs_pgio_header *header = par->data; if (!header->pnfs_error) header->pnfs_error = -EIO; @@ -405,8 +405,7 @@ static void bl_end_io_write(struct bio *bio, int err) { struct parallel_io *par = bio->bi_private; const int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags); - struct nfs_pgio_data *data = par->data; - struct nfs_pgio_header *header = data->header; + struct nfs_pgio_header *header = par->data; if (!uptodate) { if (!header->pnfs_error) @@ -423,32 +422,32 @@ static void bl_end_io_write(struct bio *bio, int err) static void bl_write_cleanup(struct work_struct *work) { struct rpc_task *task; - struct nfs_pgio_data *wdata; + struct nfs_pgio_header *hdr; dprintk("%s enter\n", __func__); task = container_of(work, struct rpc_task, u.tk_work); - wdata = container_of(task, struct nfs_pgio_data, task); - if (likely(!wdata->header->pnfs_error)) { + hdr = container_of(task, struct nfs_pgio_header, task); + if (likely(!hdr->pnfs_error)) { /* Marks for LAYOUTCOMMIT */ - mark_extents_written(BLK_LSEG2EXT(wdata->header->lseg), - wdata->args.offset, wdata->args.count); + mark_extents_written(BLK_LSEG2EXT(hdr->lseg), + hdr->args.offset, hdr->args.count); } - pnfs_ld_write_done(wdata); + pnfs_ld_write_done(hdr); } /* Called when last of bios associated with a bl_write_pagelist call finishes */ static void bl_end_par_io_write(void *data, int num_se) { - struct nfs_pgio_data *wdata = data; + struct nfs_pgio_header *hdr = data; - if (unlikely(wdata->header->pnfs_error)) { - bl_free_short_extents(&BLK_LSEG2EXT(wdata->header->lseg)->bl_inval, + if (unlikely(hdr->pnfs_error)) { + bl_free_short_extents(&BLK_LSEG2EXT(hdr->lseg)->bl_inval, num_se); } - wdata->task.tk_status = wdata->header->pnfs_error; - wdata->verf.committed = NFS_FILE_SYNC; - INIT_WORK(&wdata->task.u.tk_work, bl_write_cleanup); - schedule_work(&wdata->task.u.tk_work); + hdr->task.tk_status = hdr->pnfs_error; + hdr->verf.committed = NFS_FILE_SYNC; + INIT_WORK(&hdr->task.u.tk_work, bl_write_cleanup); + schedule_work(&hdr->task.u.tk_work); } /* FIXME STUB - mark intersection of layout and page as bad, so is not @@ -673,18 +672,17 @@ check_page: } static enum pnfs_try_status -bl_write_pagelist(struct nfs_pgio_data *wdata, int sync) +bl_write_pagelist(struct nfs_pgio_header *header, int sync) { - struct nfs_pgio_header *header = wdata->header; int i, ret, npg_zero, pg_index, last = 0; struct bio *bio = NULL; struct pnfs_block_extent *be = NULL, *cow_read = NULL; sector_t isect, last_isect = 0, extent_length = 0; struct parallel_io *par = NULL; - loff_t offset = wdata->args.offset; - size_t count = wdata->args.count; + loff_t offset = header->args.offset; + size_t count = header->args.count; unsigned int pg_offset, pg_len, saved_len; - struct page **pages = wdata->args.pages; + struct page **pages = header->args.pages; struct page *page; pgoff_t index; u64 temp; @@ -699,11 +697,11 @@ bl_write_pagelist(struct nfs_pgio_data *wdata, int sync) dprintk("pnfsblock nonblock aligned DIO writes. Resend MDS\n"); goto out_mds; } - /* At this point, wdata->pages is a (sequential) list of nfs_pages. + /* At this point, header->page_aray is a (sequential) list of nfs_pages. * We want to write each, and if there is an error set pnfs_error * to have it redone using nfs. */ - par = alloc_parallel(wdata); + par = alloc_parallel(header); if (!par) goto out_mds; par->pnfs_callback = bl_end_par_io_write; @@ -790,8 +788,8 @@ next_page: bio = bl_submit_bio(WRITE, bio); /* Middle pages */ - pg_index = wdata->args.pgbase >> PAGE_CACHE_SHIFT; - for (i = pg_index; i < wdata->pages.npages; i++) { + pg_index = header->args.pgbase >> PAGE_CACHE_SHIFT; + for (i = pg_index; i < header->page_array.npages; i++) { if (!extent_length) { /* We've used up the previous extent */ bl_put_extent(be); @@ -862,7 +860,8 @@ next_page: } - bio = do_add_page_to_bio(bio, wdata->pages.npages - i, WRITE, + bio = do_add_page_to_bio(bio, header->page_array.npages - i, + WRITE, isect, pages[i], be, bl_end_io_write, par, pg_offset, pg_len); @@ -890,7 +889,7 @@ next_page: } write_done: - wdata->res.count = wdata->args.count; + header->res.count = header->args.count; out: bl_put_extent(be); bl_put_extent(cow_read); @@ -1063,7 +1062,7 @@ nfs4_blk_get_deviceinfo(struct nfs_server *server, const struct nfs_fh *fh, return ERR_PTR(-ENOMEM); } - pages = kzalloc(max_pages * sizeof(struct page *), GFP_NOFS); + pages = kcalloc(max_pages, sizeof(struct page *), GFP_NOFS); if (pages == NULL) { kfree(dev); return ERR_PTR(-ENOMEM); diff --git a/fs/nfs/callback.c b/fs/nfs/callback.c index 073b4cf67ed9..54de482143cc 100644 --- a/fs/nfs/callback.c +++ b/fs/nfs/callback.c @@ -428,6 +428,18 @@ check_gss_callback_principal(struct nfs_client *clp, struct svc_rqst *rqstp) if (p == NULL) return 0; + /* + * Did we get the acceptor from userland during the SETCLIENID + * negotiation? + */ + if (clp->cl_acceptor) + return !strcmp(p, clp->cl_acceptor); + + /* + * Otherwise try to verify it using the cl_hostname. Note that this + * doesn't work if a non-canonical hostname was used in the devname. + */ + /* Expect a GSS_C_NT_HOSTBASED_NAME like "nfs@serverhostname" */ if (memcmp(p, "nfs@", 4) != 0) diff --git a/fs/nfs/client.c b/fs/nfs/client.c index 180d1ec9c32e..1c5ff6d58385 100644 --- a/fs/nfs/client.c +++ b/fs/nfs/client.c @@ -110,8 +110,8 @@ struct nfs_subversion *get_nfs_version(unsigned int version) mutex_unlock(&nfs_version_mutex); } - if (!IS_ERR(nfs)) - try_module_get(nfs->owner); + if (!IS_ERR(nfs) && !try_module_get(nfs->owner)) + return ERR_PTR(-EAGAIN); return nfs; } @@ -158,7 +158,8 @@ struct nfs_client *nfs_alloc_client(const struct nfs_client_initdata *cl_init) goto error_0; clp->cl_nfs_mod = cl_init->nfs_mod; - try_module_get(clp->cl_nfs_mod->owner); + if (!try_module_get(clp->cl_nfs_mod->owner)) + goto error_dealloc; clp->rpc_ops = clp->cl_nfs_mod->rpc_ops; @@ -190,6 +191,7 @@ struct nfs_client *nfs_alloc_client(const struct nfs_client_initdata *cl_init) error_cleanup: put_nfs_version(clp->cl_nfs_mod); +error_dealloc: kfree(clp); error_0: return ERR_PTR(err); @@ -252,6 +254,7 @@ void nfs_free_client(struct nfs_client *clp) put_net(clp->cl_net); put_nfs_version(clp->cl_nfs_mod); kfree(clp->cl_hostname); + kfree(clp->cl_acceptor); kfree(clp); dprintk("<-- nfs_free_client()\n"); @@ -482,8 +485,13 @@ nfs_get_client(const struct nfs_client_initdata *cl_init, struct nfs_net *nn = net_generic(cl_init->net, nfs_net_id); const struct nfs_rpc_ops *rpc_ops = cl_init->nfs_mod->rpc_ops; + if (cl_init->hostname == NULL) { + WARN_ON(1); + return NULL; + } + dprintk("--> nfs_get_client(%s,v%u)\n", - cl_init->hostname ?: "", rpc_ops->version); + cl_init->hostname, rpc_ops->version); /* see if the client already exists */ do { @@ -510,7 +518,7 @@ nfs_get_client(const struct nfs_client_initdata *cl_init, } while (!IS_ERR(new)); dprintk("<-- nfs_get_client() Failed to find %s (%ld)\n", - cl_init->hostname ?: "", PTR_ERR(new)); + cl_init->hostname, PTR_ERR(new)); return new; } EXPORT_SYMBOL_GPL(nfs_get_client); diff --git a/fs/nfs/delegation.c b/fs/nfs/delegation.c index 5d8ccecf5f5c..5853f53db732 100644 --- a/fs/nfs/delegation.c +++ b/fs/nfs/delegation.c @@ -41,14 +41,8 @@ void nfs_mark_delegation_referenced(struct nfs_delegation *delegation) set_bit(NFS_DELEGATION_REFERENCED, &delegation->flags); } -/** - * nfs_have_delegation - check if inode has a delegation - * @inode: inode to check - * @flags: delegation types to check for - * - * Returns one if inode has the indicated delegation, otherwise zero. - */ -int nfs4_have_delegation(struct inode *inode, fmode_t flags) +static int +nfs4_do_check_delegation(struct inode *inode, fmode_t flags, bool mark) { struct nfs_delegation *delegation; int ret = 0; @@ -58,12 +52,34 @@ int nfs4_have_delegation(struct inode *inode, fmode_t flags) delegation = rcu_dereference(NFS_I(inode)->delegation); if (delegation != NULL && (delegation->type & flags) == flags && !test_bit(NFS_DELEGATION_RETURNING, &delegation->flags)) { - nfs_mark_delegation_referenced(delegation); + if (mark) + nfs_mark_delegation_referenced(delegation); ret = 1; } rcu_read_unlock(); return ret; } +/** + * nfs_have_delegation - check if inode has a delegation, mark it + * NFS_DELEGATION_REFERENCED if there is one. + * @inode: inode to check + * @flags: delegation types to check for + * + * Returns one if inode has the indicated delegation, otherwise zero. + */ +int nfs4_have_delegation(struct inode *inode, fmode_t flags) +{ + return nfs4_do_check_delegation(inode, flags, true); +} + +/* + * nfs4_check_delegation - check if inode has a delegation, do not mark + * NFS_DELEGATION_REFERENCED if it has one. + */ +int nfs4_check_delegation(struct inode *inode, fmode_t flags) +{ + return nfs4_do_check_delegation(inode, flags, false); +} static int nfs_delegation_claim_locks(struct nfs_open_context *ctx, struct nfs4_state *state, const nfs4_stateid *stateid) { diff --git a/fs/nfs/delegation.h b/fs/nfs/delegation.h index 9a79c7a99d6d..5c1cce39297f 100644 --- a/fs/nfs/delegation.h +++ b/fs/nfs/delegation.h @@ -59,6 +59,7 @@ bool nfs4_copy_delegation_stateid(nfs4_stateid *dst, struct inode *inode, fmode_ void nfs_mark_delegation_referenced(struct nfs_delegation *delegation); int nfs4_have_delegation(struct inode *inode, fmode_t flags); +int nfs4_check_delegation(struct inode *inode, fmode_t flags); #endif diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index 4a3d4ef76127..36d921f0c602 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -988,9 +988,13 @@ EXPORT_SYMBOL_GPL(nfs_force_lookup_revalidate); * A check for whether or not the parent directory has changed. * In the case it has, we assume that the dentries are untrustworthy * and may need to be looked up again. + * If rcu_walk prevents us from performing a full check, return 0. */ -static int nfs_check_verifier(struct inode *dir, struct dentry *dentry) +static int nfs_check_verifier(struct inode *dir, struct dentry *dentry, + int rcu_walk) { + int ret; + if (IS_ROOT(dentry)) return 1; if (NFS_SERVER(dir)->flags & NFS_MOUNT_LOOKUP_CACHE_NONE) @@ -998,7 +1002,11 @@ static int nfs_check_verifier(struct inode *dir, struct dentry *dentry) if (!nfs_verify_change_attribute(dir, dentry->d_time)) return 0; /* Revalidate nfsi->cache_change_attribute before we declare a match */ - if (nfs_revalidate_inode(NFS_SERVER(dir), dir) < 0) + if (rcu_walk) + ret = nfs_revalidate_inode_rcu(NFS_SERVER(dir), dir); + else + ret = nfs_revalidate_inode(NFS_SERVER(dir), dir); + if (ret < 0) return 0; if (!nfs_verify_change_attribute(dir, dentry->d_time)) return 0; @@ -1042,6 +1050,8 @@ int nfs_lookup_verify_inode(struct inode *inode, unsigned int flags) out: return (inode->i_nlink == 0) ? -ENOENT : 0; out_force: + if (flags & LOOKUP_RCU) + return -ECHILD; ret = __nfs_revalidate_inode(server, inode); if (ret != 0) return ret; @@ -1054,6 +1064,9 @@ out_force: * * If parent mtime has changed, we revalidate, else we wait for a * period corresponding to the parent's attribute cache timeout value. + * + * If LOOKUP_RCU prevents us from performing a full check, return 1 + * suggesting a reval is needed. */ static inline int nfs_neg_need_reval(struct inode *dir, struct dentry *dentry, @@ -1064,7 +1077,7 @@ int nfs_neg_need_reval(struct inode *dir, struct dentry *dentry, return 0; if (NFS_SERVER(dir)->flags & NFS_MOUNT_LOOKUP_CACHE_NONEG) return 1; - return !nfs_check_verifier(dir, dentry); + return !nfs_check_verifier(dir, dentry, flags & LOOKUP_RCU); } /* @@ -1088,21 +1101,30 @@ static int nfs_lookup_revalidate(struct dentry *dentry, unsigned int flags) struct nfs4_label *label = NULL; int error; - if (flags & LOOKUP_RCU) - return -ECHILD; - - parent = dget_parent(dentry); - dir = parent->d_inode; + if (flags & LOOKUP_RCU) { + parent = ACCESS_ONCE(dentry->d_parent); + dir = ACCESS_ONCE(parent->d_inode); + if (!dir) + return -ECHILD; + } else { + parent = dget_parent(dentry); + dir = parent->d_inode; + } nfs_inc_stats(dir, NFSIOS_DENTRYREVALIDATE); inode = dentry->d_inode; if (!inode) { - if (nfs_neg_need_reval(dir, dentry, flags)) + if (nfs_neg_need_reval(dir, dentry, flags)) { + if (flags & LOOKUP_RCU) + return -ECHILD; goto out_bad; + } goto out_valid_noent; } if (is_bad_inode(inode)) { + if (flags & LOOKUP_RCU) + return -ECHILD; dfprintk(LOOKUPCACHE, "%s: %pd2 has dud inode\n", __func__, dentry); goto out_bad; @@ -1112,12 +1134,20 @@ static int nfs_lookup_revalidate(struct dentry *dentry, unsigned int flags) goto out_set_verifier; /* Force a full look up iff the parent directory has changed */ - if (!nfs_is_exclusive_create(dir, flags) && nfs_check_verifier(dir, dentry)) { - if (nfs_lookup_verify_inode(inode, flags)) + if (!nfs_is_exclusive_create(dir, flags) && + nfs_check_verifier(dir, dentry, flags & LOOKUP_RCU)) { + + if (nfs_lookup_verify_inode(inode, flags)) { + if (flags & LOOKUP_RCU) + return -ECHILD; goto out_zap_parent; + } goto out_valid; } + if (flags & LOOKUP_RCU) + return -ECHILD; + if (NFS_STALE(inode)) goto out_bad; @@ -1153,13 +1183,18 @@ out_set_verifier: /* Success: notify readdir to use READDIRPLUS */ nfs_advise_use_readdirplus(dir); out_valid_noent: - dput(parent); + if (flags & LOOKUP_RCU) { + if (parent != ACCESS_ONCE(dentry->d_parent)) + return -ECHILD; + } else + dput(parent); dfprintk(LOOKUPCACHE, "NFS: %s(%pd2) is valid\n", __func__, dentry); return 1; out_zap_parent: nfs_zap_caches(dir); out_bad: + WARN_ON(flags & LOOKUP_RCU); nfs_free_fattr(fattr); nfs_free_fhandle(fhandle); nfs4_label_free(label); @@ -1185,6 +1220,7 @@ out_zap_parent: __func__, dentry); return 0; out_error: + WARN_ON(flags & LOOKUP_RCU); nfs_free_fattr(fattr); nfs_free_fhandle(fhandle); nfs4_label_free(label); @@ -1529,14 +1565,9 @@ EXPORT_SYMBOL_GPL(nfs_atomic_open); static int nfs4_lookup_revalidate(struct dentry *dentry, unsigned int flags) { - struct dentry *parent = NULL; struct inode *inode; - struct inode *dir; int ret = 0; - if (flags & LOOKUP_RCU) - return -ECHILD; - if (!(flags & LOOKUP_OPEN) || (flags & LOOKUP_DIRECTORY)) goto no_open; if (d_mountpoint(dentry)) @@ -1545,34 +1576,47 @@ static int nfs4_lookup_revalidate(struct dentry *dentry, unsigned int flags) goto no_open; inode = dentry->d_inode; - parent = dget_parent(dentry); - dir = parent->d_inode; /* We can't create new files in nfs_open_revalidate(), so we * optimize away revalidation of negative dentries. */ if (inode == NULL) { + struct dentry *parent; + struct inode *dir; + + if (flags & LOOKUP_RCU) { + parent = ACCESS_ONCE(dentry->d_parent); + dir = ACCESS_ONCE(parent->d_inode); + if (!dir) + return -ECHILD; + } else { + parent = dget_parent(dentry); + dir = parent->d_inode; + } if (!nfs_neg_need_reval(dir, dentry, flags)) ret = 1; + else if (flags & LOOKUP_RCU) + ret = -ECHILD; + if (!(flags & LOOKUP_RCU)) + dput(parent); + else if (parent != ACCESS_ONCE(dentry->d_parent)) + return -ECHILD; goto out; } /* NFS only supports OPEN on regular files */ if (!S_ISREG(inode->i_mode)) - goto no_open_dput; + goto no_open; /* We cannot do exclusive creation on a positive dentry */ if (flags & LOOKUP_EXCL) - goto no_open_dput; + goto no_open; /* Let f_op->open() actually open (and revalidate) the file */ ret = 1; out: - dput(parent); return ret; -no_open_dput: - dput(parent); no_open: return nfs_lookup_revalidate(dentry, flags); } @@ -2028,10 +2072,14 @@ static DEFINE_SPINLOCK(nfs_access_lru_lock); static LIST_HEAD(nfs_access_lru_list); static atomic_long_t nfs_access_nr_entries; +static unsigned long nfs_access_max_cachesize = ULONG_MAX; +module_param(nfs_access_max_cachesize, ulong, 0644); +MODULE_PARM_DESC(nfs_access_max_cachesize, "NFS access maximum total cache length"); + static void nfs_access_free_entry(struct nfs_access_entry *entry) { put_rpccred(entry->cred); - kfree(entry); + kfree_rcu(entry, rcu_head); smp_mb__before_atomic(); atomic_long_dec(&nfs_access_nr_entries); smp_mb__after_atomic(); @@ -2048,19 +2096,14 @@ static void nfs_access_free_list(struct list_head *head) } } -unsigned long -nfs_access_cache_scan(struct shrinker *shrink, struct shrink_control *sc) +static unsigned long +nfs_do_access_cache_scan(unsigned int nr_to_scan) { LIST_HEAD(head); struct nfs_inode *nfsi, *next; struct nfs_access_entry *cache; - int nr_to_scan = sc->nr_to_scan; - gfp_t gfp_mask = sc->gfp_mask; long freed = 0; - if ((gfp_mask & GFP_KERNEL) != GFP_KERNEL) - return SHRINK_STOP; - spin_lock(&nfs_access_lru_lock); list_for_each_entry_safe(nfsi, next, &nfs_access_lru_list, access_cache_inode_lru) { struct inode *inode; @@ -2094,11 +2137,39 @@ remove_lru_entry: } unsigned long +nfs_access_cache_scan(struct shrinker *shrink, struct shrink_control *sc) +{ + int nr_to_scan = sc->nr_to_scan; + gfp_t gfp_mask = sc->gfp_mask; + + if ((gfp_mask & GFP_KERNEL) != GFP_KERNEL) + return SHRINK_STOP; + return nfs_do_access_cache_scan(nr_to_scan); +} + + +unsigned long nfs_access_cache_count(struct shrinker *shrink, struct shrink_control *sc) { return vfs_pressure_ratio(atomic_long_read(&nfs_access_nr_entries)); } +static void +nfs_access_cache_enforce_limit(void) +{ + long nr_entries = atomic_long_read(&nfs_access_nr_entries); + unsigned long diff; + unsigned int nr_to_scan; + + if (nr_entries < 0 || nr_entries <= nfs_access_max_cachesize) + return; + nr_to_scan = 100; + diff = nr_entries - nfs_access_max_cachesize; + if (diff < nr_to_scan) + nr_to_scan = diff; + nfs_do_access_cache_scan(nr_to_scan); +} + static void __nfs_access_zap_cache(struct nfs_inode *nfsi, struct list_head *head) { struct rb_root *root_node = &nfsi->access_cache; @@ -2186,6 +2257,38 @@ out_zap: return -ENOENT; } +static int nfs_access_get_cached_rcu(struct inode *inode, struct rpc_cred *cred, struct nfs_access_entry *res) +{ + /* Only check the most recently returned cache entry, + * but do it without locking. + */ + struct nfs_inode *nfsi = NFS_I(inode); + struct nfs_access_entry *cache; + int err = -ECHILD; + struct list_head *lh; + + rcu_read_lock(); + if (nfsi->cache_validity & NFS_INO_INVALID_ACCESS) + goto out; + lh = rcu_dereference(nfsi->access_cache_entry_lru.prev); + cache = list_entry(lh, struct nfs_access_entry, lru); + if (lh == &nfsi->access_cache_entry_lru || + cred != cache->cred) + cache = NULL; + if (cache == NULL) + goto out; + if (!nfs_have_delegated_attributes(inode) && + !time_in_range_open(jiffies, cache->jiffies, cache->jiffies + nfsi->attrtimeo)) + goto out; + res->jiffies = cache->jiffies; + res->cred = cache->cred; + res->mask = cache->mask; + err = 0; +out: + rcu_read_unlock(); + return err; +} + static void nfs_access_add_rbtree(struct inode *inode, struct nfs_access_entry *set) { struct nfs_inode *nfsi = NFS_I(inode); @@ -2229,6 +2332,11 @@ void nfs_access_add_cache(struct inode *inode, struct nfs_access_entry *set) cache->cred = get_rpccred(set->cred); cache->mask = set->mask; + /* The above field assignments must be visible + * before this item appears on the lru. We cannot easily + * use rcu_assign_pointer, so just force the memory barrier. + */ + smp_wmb(); nfs_access_add_rbtree(inode, cache); /* Update accounting */ @@ -2244,6 +2352,7 @@ void nfs_access_add_cache(struct inode *inode, struct nfs_access_entry *set) &nfs_access_lru_list); spin_unlock(&nfs_access_lru_lock); } + nfs_access_cache_enforce_limit(); } EXPORT_SYMBOL_GPL(nfs_access_add_cache); @@ -2267,10 +2376,16 @@ static int nfs_do_access(struct inode *inode, struct rpc_cred *cred, int mask) trace_nfs_access_enter(inode); - status = nfs_access_get_cached(inode, cred, &cache); + status = nfs_access_get_cached_rcu(inode, cred, &cache); + if (status != 0) + status = nfs_access_get_cached(inode, cred, &cache); if (status == 0) goto out_cached; + status = -ECHILD; + if (mask & MAY_NOT_BLOCK) + goto out; + /* Be clever: ask server to check for all possible rights */ cache.mask = MAY_EXEC | MAY_WRITE | MAY_READ; cache.cred = cred; @@ -2321,9 +2436,6 @@ int nfs_permission(struct inode *inode, int mask) struct rpc_cred *cred; int res = 0; - if (mask & MAY_NOT_BLOCK) - return -ECHILD; - nfs_inc_stats(inode, NFSIOS_VFSACCESS); if ((mask & (MAY_READ | MAY_WRITE | MAY_EXEC)) == 0) @@ -2350,12 +2462,23 @@ force_lookup: if (!NFS_PROTO(inode)->access) goto out_notsup; - cred = rpc_lookup_cred(); - if (!IS_ERR(cred)) { - res = nfs_do_access(inode, cred, mask); - put_rpccred(cred); - } else + /* Always try fast lookups first */ + rcu_read_lock(); + cred = rpc_lookup_cred_nonblock(); + if (!IS_ERR(cred)) + res = nfs_do_access(inode, cred, mask|MAY_NOT_BLOCK); + |