diff options
author | Linus Torvalds <torvalds@linux-foundation.org> | 2019-05-10 00:33:15 +0300 |
---|---|---|
committer | Linus Torvalds <torvalds@linux-foundation.org> | 2019-05-10 00:33:15 +0300 |
commit | 06cbd26d312edfe4a83ff541c23f8f866265eb24 (patch) | |
tree | 45046d0daca202df6c590390df05fbd07ed84a5d /fs/nfs | |
parent | abde77eb5c66b2f98539c4644b54f34b7e179e6b (diff) | |
parent | 5940d1cf9f42f67e9cc3f7df9eda39f5888d6e9e (diff) | |
download | linux-06cbd26d312edfe4a83ff541c23f8f866265eb24.tar.xz |
Merge tag 'nfs-for-5.2-1' of git://git.linux-nfs.org/projects/anna/linux-nfs
Pull NFS client updates from Anna Schumaker:
"Highlights include:
Stable bugfixes:
- Fall back to MDS if no deviceid is found rather than aborting # v4.11+
- NFS4: Fix v4.0 client state corruption when mount
Features:
- Much improved handling of soft mounts with NFS v4.0:
- Reduce risk of false positive timeouts
- Faster failover of reads and writes after a timeout
- Added a "softerr" mount option to return ETIMEDOUT instead of
EIO to the application after a timeout
- Increase number of xprtrdma backchannel requests
- Add additional xprtrdma tracepoints
- Improved send completion batching for xprtrdma
Other bugfixes and cleanups:
- Return -EINVAL when NFS v4.2 is passed an invalid dedup mode
- Reduce usage of GFP_ATOMIC pages in SUNRPC
- Various minor NFS over RDMA cleanups and bugfixes
- Use the correct container namespace for upcalls
- Don't share superblocks between user namespaces
- Various other container fixes
- Make nfs_match_client() killable to prevent soft lockups
- Don't mark all open state for recovery when handling recallable
state revoked flag"
* tag 'nfs-for-5.2-1' of git://git.linux-nfs.org/projects/anna/linux-nfs: (69 commits)
SUNRPC: Rebalance a kref in auth_gss.c
NFS: Fix a double unlock from nfs_match,get_client
nfs: pass the correct prototype to read_cache_page
NFSv4: don't mark all open state for recovery when handling recallable state revoked flag
SUNRPC: Fix an error code in gss_alloc_msg()
SUNRPC: task should be exit if encode return EKEYEXPIRED more times
NFS4: Fix v4.0 client state corruption when mount
PNFS fallback to MDS if no deviceid found
NFS: make nfs_match_client killable
lockd: Store the lockd client credential in struct nlm_host
NFS: When mounting, don't share filesystems between different user namespaces
NFS: Convert NFSv2 to use the container user namespace
NFSv4: Convert the NFS client idmapper to use the container user namespace
NFS: Convert NFSv3 to use the container user namespace
SUNRPC: Use namespace of listening daemon in the client AUTH_GSS upcall
SUNRPC: Use the client user namespace when encoding creds
NFS: Store the credential of the mount process in the nfs_server
SUNRPC: Cache cred of process creating the rpc_client
xprtrdma: Remove stale comment
xprtrdma: Update comments that reference ib_drain_qp
...
Diffstat (limited to 'fs/nfs')
-rw-r--r-- | fs/nfs/client.c | 16 | ||||
-rw-r--r-- | fs/nfs/delegation.c | 12 | ||||
-rw-r--r-- | fs/nfs/delegation.h | 1 | ||||
-rw-r--r-- | fs/nfs/dir.c | 7 | ||||
-rw-r--r-- | fs/nfs/direct.c | 11 | ||||
-rw-r--r-- | fs/nfs/file.c | 31 | ||||
-rw-r--r-- | fs/nfs/filelayout/filelayout.c | 6 | ||||
-rw-r--r-- | fs/nfs/flexfilelayout/flexfilelayout.c | 14 | ||||
-rw-r--r-- | fs/nfs/inode.c | 13 | ||||
-rw-r--r-- | fs/nfs/internal.h | 8 | ||||
-rw-r--r-- | fs/nfs/mount_clnt.c | 2 | ||||
-rw-r--r-- | fs/nfs/nfs2xdr.c | 58 | ||||
-rw-r--r-- | fs/nfs/nfs3client.c | 1 | ||||
-rw-r--r-- | fs/nfs/nfs3xdr.c | 142 | ||||
-rw-r--r-- | fs/nfs/nfs4_fs.h | 1 | ||||
-rw-r--r-- | fs/nfs/nfs4client.c | 6 | ||||
-rw-r--r-- | fs/nfs/nfs4file.c | 4 | ||||
-rw-r--r-- | fs/nfs/nfs4idmap.c | 27 | ||||
-rw-r--r-- | fs/nfs/nfs4proc.c | 159 | ||||
-rw-r--r-- | fs/nfs/nfs4state.c | 7 | ||||
-rw-r--r-- | fs/nfs/pagelist.c | 123 | ||||
-rw-r--r-- | fs/nfs/pnfs.c | 4 | ||||
-rw-r--r-- | fs/nfs/pnfs.h | 4 | ||||
-rw-r--r-- | fs/nfs/read.c | 6 | ||||
-rw-r--r-- | fs/nfs/super.c | 32 | ||||
-rw-r--r-- | fs/nfs/symlink.c | 7 | ||||
-rw-r--r-- | fs/nfs/write.c | 70 |
27 files changed, 507 insertions, 265 deletions
diff --git a/fs/nfs/client.c b/fs/nfs/client.c index 90d71fda65ce..da74c4c4a244 100644 --- a/fs/nfs/client.c +++ b/fs/nfs/client.c @@ -284,6 +284,7 @@ static struct nfs_client *nfs_match_client(const struct nfs_client_initdata *dat struct nfs_client *clp; const struct sockaddr *sap = data->addr; struct nfs_net *nn = net_generic(data->net, nfs_net_id); + int error; again: list_for_each_entry(clp, &nn->nfs_client_list, cl_share_link) { @@ -296,9 +297,11 @@ again: if (clp->cl_cons_state > NFS_CS_READY) { refcount_inc(&clp->cl_count); spin_unlock(&nn->nfs_client_lock); - nfs_wait_client_init_complete(clp); + error = nfs_wait_client_init_complete(clp); nfs_put_client(clp); spin_lock(&nn->nfs_client_lock); + if (error < 0) + return ERR_PTR(error); goto again; } @@ -407,6 +410,8 @@ struct nfs_client *nfs_get_client(const struct nfs_client_initdata *cl_init) clp = nfs_match_client(cl_init); if (clp) { spin_unlock(&nn->nfs_client_lock); + if (IS_ERR(clp)) + return clp; if (new) new->rpc_ops->free_client(new); return nfs_found_client(cl_init, clp); @@ -500,6 +505,7 @@ int nfs_create_rpc_client(struct nfs_client *clp, .program = &nfs_program, .version = clp->rpc_ops->version, .authflavor = flavor, + .cred = cl_init->cred, }; if (test_bit(NFS_CS_DISCRTRY, &clp->cl_flags)) @@ -598,6 +604,8 @@ int nfs_init_server_rpcclient(struct nfs_server *server, sizeof(server->client->cl_timeout_default)); server->client->cl_timeout = &server->client->cl_timeout_default; server->client->cl_softrtry = 0; + if (server->flags & NFS_MOUNT_SOFTERR) + server->client->cl_softerr = 1; if (server->flags & NFS_MOUNT_SOFT) server->client->cl_softrtry = 1; @@ -652,6 +660,7 @@ static int nfs_init_server(struct nfs_server *server, .proto = data->nfs_server.protocol, .net = data->net, .timeparms = &timeparms, + .cred = server->cred, }; struct nfs_client *clp; int error; @@ -920,6 +929,7 @@ void nfs_free_server(struct nfs_server *server) ida_destroy(&server->lockowner_id); ida_destroy(&server->openowner_id); nfs_free_iostats(server->io_stats); + put_cred(server->cred); kfree(server); nfs_release_automount_timer(); } @@ -940,6 +950,8 @@ struct nfs_server *nfs_create_server(struct nfs_mount_info *mount_info, if (!server) return ERR_PTR(-ENOMEM); + server->cred = get_cred(current_cred()); + error = -ENOMEM; fattr = nfs_alloc_fattr(); if (fattr == NULL) @@ -1006,6 +1018,8 @@ struct nfs_server *nfs_clone_server(struct nfs_server *source, if (!server) return ERR_PTR(-ENOMEM); + server->cred = get_cred(source->cred); + error = -ENOMEM; fattr_fsinfo = nfs_alloc_fattr(); if (fattr_fsinfo == NULL) diff --git a/fs/nfs/delegation.c b/fs/nfs/delegation.c index 2f6b447cdd82..8b78274e3e56 100644 --- a/fs/nfs/delegation.c +++ b/fs/nfs/delegation.c @@ -1034,6 +1034,18 @@ void nfs_mark_test_expired_all_delegations(struct nfs_client *clp) } /** + * nfs_test_expired_all_delegations - test all delegations for a client + * @clp: nfs_client to process + * + * Helper for handling "recallable state revoked" status from server. + */ +void nfs_test_expired_all_delegations(struct nfs_client *clp) +{ + nfs_mark_test_expired_all_delegations(clp); + nfs4_schedule_state_manager(clp); +} + +/** * nfs_reap_expired_delegations - reap expired delegations * @clp: nfs_client to process * diff --git a/fs/nfs/delegation.h b/fs/nfs/delegation.h index 35b4b02c1ae0..5799777df5ec 100644 --- a/fs/nfs/delegation.h +++ b/fs/nfs/delegation.h @@ -58,6 +58,7 @@ void nfs_delegation_mark_reclaim(struct nfs_client *clp); void nfs_delegation_reap_unclaimed(struct nfs_client *clp); void nfs_mark_test_expired_all_delegations(struct nfs_client *clp); +void nfs_test_expired_all_delegations(struct nfs_client *clp); void nfs_reap_expired_delegations(struct nfs_client *clp); /* NFSv4 delegation-related procedures */ diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index a71d0b42d160..47d445bec8c9 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -714,8 +714,9 @@ out: * We only need to convert from xdr once so future lookups are much simpler */ static -int nfs_readdir_filler(nfs_readdir_descriptor_t *desc, struct page* page) +int nfs_readdir_filler(void *data, struct page* page) { + nfs_readdir_descriptor_t *desc = data; struct inode *inode = file_inode(desc->file); int ret; @@ -762,8 +763,8 @@ void cache_page_release(nfs_readdir_descriptor_t *desc) static struct page *get_cache_page(nfs_readdir_descriptor_t *desc) { - return read_cache_page(desc->file->f_mapping, - desc->page_index, (filler_t *)nfs_readdir_filler, desc); + return read_cache_page(desc->file->f_mapping, desc->page_index, + nfs_readdir_filler, desc); } /* diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c index 0fd811ac08b5..2436bd92bc00 100644 --- a/fs/nfs/direct.c +++ b/fs/nfs/direct.c @@ -492,7 +492,7 @@ static ssize_t nfs_direct_read_schedule_iovec(struct nfs_direct_req *dreq, struct nfs_page *req; unsigned int req_len = min_t(size_t, bytes, PAGE_SIZE - pgbase); /* XXX do we need to do the eof zeroing found in async_filler? */ - req = nfs_create_request(dreq->ctx, pagevec[i], NULL, + req = nfs_create_request(dreq->ctx, pagevec[i], pgbase, req_len); if (IS_ERR(req)) { result = PTR_ERR(req); @@ -663,6 +663,8 @@ static void nfs_direct_write_reschedule(struct nfs_direct_req *dreq) } list_for_each_entry_safe(req, tmp, &reqs, wb_list) { + /* Bump the transmission count */ + req->wb_nio++; if (!nfs_pageio_add_request(&desc, req)) { nfs_list_move_request(req, &failed); spin_lock(&cinfo.inode->i_lock); @@ -703,6 +705,11 @@ static void nfs_direct_commit_complete(struct nfs_commit_data *data) req = nfs_list_entry(data->pages.next); nfs_list_remove_request(req); if (dreq->flags == NFS_ODIRECT_RESCHED_WRITES) { + /* + * Despite the reboot, the write was successful, + * so reset wb_nio. + */ + req->wb_nio = 0; /* Note the rewrite will go through mds */ nfs_mark_request_commit(req, NULL, &cinfo, 0); } else @@ -899,7 +906,7 @@ static ssize_t nfs_direct_write_schedule_iovec(struct nfs_direct_req *dreq, struct nfs_page *req; unsigned int req_len = min_t(size_t, bytes, PAGE_SIZE - pgbase); - req = nfs_create_request(dreq->ctx, pagevec[i], NULL, + req = nfs_create_request(dreq->ctx, pagevec[i], pgbase, req_len); if (IS_ERR(req)) { result = PTR_ERR(req); diff --git a/fs/nfs/file.c b/fs/nfs/file.c index 4899b85f9b3c..144e183250c3 100644 --- a/fs/nfs/file.c +++ b/fs/nfs/file.c @@ -147,7 +147,7 @@ nfs_file_flush(struct file *file, fl_owner_t id) return 0; /* Flush writes to the server and return any errors */ - return vfs_fsync(file, 0); + return nfs_wb_all(inode); } ssize_t @@ -199,13 +199,6 @@ EXPORT_SYMBOL_GPL(nfs_file_mmap); * Flush any dirty pages for this process, and check for write errors. * The return status from this call provides a reliable indication of * whether any write errors occurred for this process. - * - * Notice that it clears the NFS_CONTEXT_ERROR_WRITE before synching to - * disk, but it retrieves and clears ctx->error after synching, despite - * the two being set at the same time in nfs_context_set_write_error(). - * This is because the former is used to notify the _next_ call to - * nfs_file_write() that a write error occurred, and hence cause it to - * fall back to doing a synchronous write. */ static int nfs_file_fsync_commit(struct file *file, int datasync) @@ -220,11 +213,8 @@ nfs_file_fsync_commit(struct file *file, int datasync) nfs_inc_stats(inode, NFSIOS_VFSFSYNC); do_resend = test_and_clear_bit(NFS_CONTEXT_RESEND_WRITES, &ctx->flags); status = nfs_commit_inode(inode, FLUSH_SYNC); - if (test_bit(NFS_CONTEXT_ERROR_WRITE, &ctx->flags)) { - ret = xchg(&ctx->error, 0); - if (ret) - goto out; - } + if (status == 0) + status = file_check_and_advance_wb_err(file); if (status < 0) { ret = status; goto out; @@ -245,13 +235,7 @@ nfs_file_fsync(struct file *file, loff_t start, loff_t end, int datasync) trace_nfs_fsync_enter(inode); do { - struct nfs_open_context *ctx = nfs_file_open_context(file); - ret = filemap_write_and_wait_range(inode->i_mapping, start, end); - if (test_and_clear_bit(NFS_CONTEXT_ERROR_WRITE, &ctx->flags)) { - int ret2 = xchg(&ctx->error, 0); - if (ret2) - ret = ret2; - } + ret = file_write_and_wait_range(file, start, end); if (ret != 0) break; ret = nfs_file_fsync_commit(file, datasync); @@ -600,8 +584,7 @@ static int nfs_need_check_write(struct file *filp, struct inode *inode) struct nfs_open_context *ctx; ctx = nfs_file_open_context(filp); - if (test_bit(NFS_CONTEXT_ERROR_WRITE, &ctx->flags) || - nfs_ctx_key_to_expire(ctx, inode)) + if (nfs_ctx_key_to_expire(ctx, inode)) return 1; return 0; } @@ -655,7 +638,7 @@ ssize_t nfs_file_write(struct kiocb *iocb, struct iov_iter *from) /* Return error values */ if (nfs_need_check_write(file, inode)) { - int err = vfs_fsync(file, 0); + int err = nfs_wb_all(inode); if (err < 0) result = err; } @@ -709,7 +692,7 @@ do_unlk(struct file *filp, int cmd, struct file_lock *fl, int is_local) * Flush all pending writes before doing anything * with locks.. */ - vfs_fsync(filp, 0); + nfs_wb_all(inode); l_ctx = nfs_get_lock_context(nfs_file_open_context(filp)); if (!IS_ERR(l_ctx)) { diff --git a/fs/nfs/filelayout/filelayout.c b/fs/nfs/filelayout/filelayout.c index 61f46facb39c..3cb073c50fa6 100644 --- a/fs/nfs/filelayout/filelayout.c +++ b/fs/nfs/filelayout/filelayout.c @@ -904,7 +904,7 @@ fl_pnfs_update_layout(struct inode *ino, status = filelayout_check_deviceid(lo, fl, gfp_flags); if (status) { pnfs_put_lseg(lseg); - lseg = ERR_PTR(status); + lseg = NULL; } out: return lseg; @@ -917,7 +917,7 @@ filelayout_pg_init_read(struct nfs_pageio_descriptor *pgio, pnfs_generic_pg_check_layout(pgio); if (!pgio->pg_lseg) { pgio->pg_lseg = fl_pnfs_update_layout(pgio->pg_inode, - req->wb_context, + nfs_req_openctx(req), 0, NFS4_MAX_UINT64, IOMODE_READ, @@ -944,7 +944,7 @@ filelayout_pg_init_write(struct nfs_pageio_descriptor *pgio, pnfs_generic_pg_check_layout(pgio); if (!pgio->pg_lseg) { pgio->pg_lseg = fl_pnfs_update_layout(pgio->pg_inode, - req->wb_context, + nfs_req_openctx(req), 0, NFS4_MAX_UINT64, IOMODE_RW, diff --git a/fs/nfs/flexfilelayout/flexfilelayout.c b/fs/nfs/flexfilelayout/flexfilelayout.c index 6673d4ff5a2a..9920c52bd0cd 100644 --- a/fs/nfs/flexfilelayout/flexfilelayout.c +++ b/fs/nfs/flexfilelayout/flexfilelayout.c @@ -28,6 +28,8 @@ #define FF_LAYOUT_POLL_RETRY_MAX (15*HZ) #define FF_LAYOUTRETURN_MAXERR 20 +static unsigned short io_maxretrans; + static void ff_layout_read_record_layoutstats_done(struct rpc_task *task, struct nfs_pgio_header *hdr); static int ff_layout_mirror_prepare_stats(struct pnfs_layout_hdr *lo, @@ -871,7 +873,7 @@ ff_layout_pg_get_read(struct nfs_pageio_descriptor *pgio, { pnfs_put_lseg(pgio->pg_lseg); pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode, - req->wb_context, + nfs_req_openctx(req), 0, NFS4_MAX_UINT64, IOMODE_READ, @@ -925,6 +927,7 @@ retry: pgm = &pgio->pg_mirrors[0]; pgm->pg_bsize = mirror->mirror_ds->ds_versions[0].rsize; + pgio->pg_maxretrans = io_maxretrans; return; out_nolseg: if (pgio->pg_error < 0) @@ -950,7 +953,7 @@ retry: pnfs_generic_pg_check_layout(pgio); if (!pgio->pg_lseg) { pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode, - req->wb_context, + nfs_req_openctx(req), 0, NFS4_MAX_UINT64, IOMODE_RW, @@ -992,6 +995,7 @@ retry: pgm->pg_bsize = mirror->mirror_ds->ds_versions[0].wsize; } + pgio->pg_maxretrans = io_maxretrans; return; out_mds: @@ -1006,7 +1010,7 @@ ff_layout_pg_get_mirror_count_write(struct nfs_pageio_descriptor *pgio, { if (!pgio->pg_lseg) { pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode, - req->wb_context, + nfs_req_openctx(req), 0, NFS4_MAX_UINT64, IOMODE_RW, @@ -2515,3 +2519,7 @@ MODULE_DESCRIPTION("The NFSv4 flexfile layout driver"); module_init(nfs4flexfilelayout_init); module_exit(nfs4flexfilelayout_exit); + +module_param(io_maxretrans, ushort, 0644); +MODULE_PARM_DESC(io_maxretrans, "The number of times the NFSv4.1 client " + "retries an I/O request before returning an error. "); diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index f61af8307dc8..3bc2550cfe4e 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -885,10 +885,14 @@ struct nfs_lock_context *nfs_get_lock_context(struct nfs_open_context *ctx) spin_lock(&inode->i_lock); res = __nfs_find_lock_context(ctx); if (res == NULL) { - list_add_tail_rcu(&new->list, &ctx->lock_context.list); - new->open_context = ctx; - res = new; - new = NULL; + new->open_context = get_nfs_open_context(ctx); + if (new->open_context) { + list_add_tail_rcu(&new->list, + &ctx->lock_context.list); + res = new; + new = NULL; + } else + res = ERR_PTR(-EBADF); } spin_unlock(&inode->i_lock); kfree(new); @@ -906,6 +910,7 @@ void nfs_put_lock_context(struct nfs_lock_context *l_ctx) return; list_del_rcu(&l_ctx->list); spin_unlock(&inode->i_lock); + put_nfs_open_context(ctx); kfree_rcu(l_ctx, rcu_head); } EXPORT_SYMBOL_GPL(nfs_put_lock_context); diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h index 331a0504eaf8..498fab72f70b 100644 --- a/fs/nfs/internal.h +++ b/fs/nfs/internal.h @@ -84,6 +84,7 @@ struct nfs_client_initdata { u32 minorversion; struct net *net; const struct rpc_timeout *timeparms; + const struct cred *cred; }; /* @@ -766,15 +767,10 @@ static inline bool nfs_error_is_fatal(int err) case -ESTALE: case -E2BIG: case -ENOMEM: + case -ETIMEDOUT: return true; default: return false; } } -static inline void nfs_context_set_write_error(struct nfs_open_context *ctx, int error) -{ - ctx->error = error; - smp_wmb(); - set_bit(NFS_CONTEXT_ERROR_WRITE, &ctx->flags); -} diff --git a/fs/nfs/mount_clnt.c b/fs/nfs/mount_clnt.c index d979ff4fee7e..cb7c10e9721e 100644 --- a/fs/nfs/mount_clnt.c +++ b/fs/nfs/mount_clnt.c @@ -163,6 +163,7 @@ int nfs_mount(struct nfs_mount_request *info) .program = &mnt_program, .version = info->version, .authflavor = RPC_AUTH_UNIX, + .cred = current_cred(), }; struct rpc_clnt *mnt_clnt; int status; @@ -249,6 +250,7 @@ void nfs_umount(const struct nfs_mount_request *info) .version = info->version, .authflavor = RPC_AUTH_UNIX, .flags = RPC_CLNT_CREATE_NOPING, + .cred = current_cred(), }; struct rpc_message msg = { .rpc_argp = info->dirpath, diff --git a/fs/nfs/nfs2xdr.c b/fs/nfs/nfs2xdr.c index a7ed29de0a40..572794dab4b1 100644 --- a/fs/nfs/nfs2xdr.c +++ b/fs/nfs/nfs2xdr.c @@ -76,6 +76,20 @@ static int nfs_stat_to_errno(enum nfs_stat); * or decoded inline. */ +static struct user_namespace *rpc_userns(const struct rpc_clnt *clnt) +{ + if (clnt && clnt->cl_cred) + return clnt->cl_cred->user_ns; + return &init_user_ns; +} + +static struct user_namespace *rpc_rqst_userns(const struct rpc_rqst *rqstp) +{ + if (rqstp->rq_task) + return rpc_userns(rqstp->rq_task->tk_client); + return &init_user_ns; +} + /* * typedef opaque nfsdata<>; */ @@ -248,7 +262,8 @@ static __be32 *xdr_decode_time(__be32 *p, struct timespec *timep) * }; * */ -static int decode_fattr(struct xdr_stream *xdr, struct nfs_fattr *fattr) +static int decode_fattr(struct xdr_stream *xdr, struct nfs_fattr *fattr, + struct user_namespace *userns) { u32 rdev, type; __be32 *p; @@ -263,10 +278,10 @@ static int decode_fattr(struct xdr_stream *xdr, struct nfs_fattr *fattr) fattr->mode = be32_to_cpup(p++); fattr->nlink = be32_to_cpup(p++); - fattr->uid = make_kuid(&init_user_ns, be32_to_cpup(p++)); + fattr->uid = make_kuid(userns, be32_to_cpup(p++)); if (!uid_valid(fattr->uid)) goto out_uid; - fattr->gid = make_kgid(&init_user_ns, be32_to_cpup(p++)); + fattr->gid = make_kgid(userns, be32_to_cpup(p++)); if (!gid_valid(fattr->gid)) goto out_gid; @@ -321,7 +336,8 @@ static __be32 *xdr_time_not_set(__be32 *p) return p; } -static void encode_sattr(struct xdr_stream *xdr, const struct iattr *attr) +static void encode_sattr(struct xdr_stream *xdr, const struct iattr *attr, + struct user_namespace *userns) { struct timespec ts; __be32 *p; @@ -333,11 +349,11 @@ static void encode_sattr(struct xdr_stream *xdr, const struct iattr *attr) else *p++ = cpu_to_be32(NFS2_SATTR_NOT_SET); if (attr->ia_valid & ATTR_UID) - *p++ = cpu_to_be32(from_kuid(&init_user_ns, attr->ia_uid)); + *p++ = cpu_to_be32(from_kuid_munged(userns, attr->ia_uid)); else *p++ = cpu_to_be32(NFS2_SATTR_NOT_SET); if (attr->ia_valid & ATTR_GID) - *p++ = cpu_to_be32(from_kgid(&init_user_ns, attr->ia_gid)); + *p++ = cpu_to_be32(from_kgid_munged(userns, attr->ia_gid)); else *p++ = cpu_to_be32(NFS2_SATTR_NOT_SET); if (attr->ia_valid & ATTR_SIZE) @@ -451,7 +467,8 @@ out_cheating: * }; */ static int decode_attrstat(struct xdr_stream *xdr, struct nfs_fattr *result, - __u32 *op_status) + __u32 *op_status, + struct user_namespace *userns) { enum nfs_stat status; int error; @@ -463,7 +480,7 @@ static int decode_attrstat(struct xdr_stream *xdr, struct nfs_fattr *result, *op_status = status; if (status != NFS_OK) goto out_default; - error = decode_fattr(xdr, result); + error = decode_fattr(xdr, result, userns); out: return error; out_default: @@ -498,19 +515,21 @@ static void encode_diropargs(struct xdr_stream *xdr, const struct nfs_fh *fh, * void; * }; */ -static int decode_diropok(struct xdr_stream *xdr, struct nfs_diropok *result) +static int decode_diropok(struct xdr_stream *xdr, struct nfs_diropok *result, + struct user_namespace *userns) { int error; error = decode_fhandle(xdr, result->fh); if (unlikely(error)) goto out; - error = decode_fattr(xdr, result->fattr); + error = decode_fattr(xdr, result->fattr, userns); out: return error; } -static int decode_diropres(struct xdr_stream *xdr, struct nfs_diropok *result) +static int decode_diropres(struct xdr_stream *xdr, struct nfs_diropok *result, + struct user_namespace *userns) { enum nfs_stat status; int error; @@ -520,7 +539,7 @@ static int decode_diropres(struct xdr_stream *xdr, struct nfs_diropok *result) goto out; if (status != NFS_OK) goto out_default; - error = decode_diropok(xdr, result); + error = decode_diropok(xdr, result, userns); out: return error; out_default: @@ -559,7 +578,7 @@ static void nfs2_xdr_enc_sattrargs(struct rpc_rqst *req, const struct nfs_sattrargs *args = data; encode_fhandle(xdr, args->fh); - encode_sattr(xdr, args->sattr); + encode_sattr(xdr, args->sattr, rpc_rqst_userns(req)); } static void nfs2_xdr_enc_diropargs(struct rpc_rqst *req, @@ -674,7 +693,7 @@ static void nfs2_xdr_enc_createargs(struct rpc_rqst *req, const struct nfs_createargs *args = data; encode_diropargs(xdr, args->fh, args->name, args->len); - encode_sattr(xdr, args->sattr); + encode_sattr(xdr, args->sattr, rpc_rqst_userns(req)); } static void nfs2_xdr_enc_removeargs(struct rpc_rqst *req, @@ -741,7 +760,7 @@ static void nfs2_xdr_enc_symlinkargs(struct rpc_rqst *req, encode_diropargs(xdr, args->fromfh, args->fromname, args->fromlen); encode_path(xdr, args->pages, args->pathlen); - encode_sattr(xdr, args->sattr); + encode_sattr(xdr, args->sattr, rpc_rqst_userns(req)); } /* @@ -803,13 +822,13 @@ out_default: static int nfs2_xdr_dec_attrstat(struct rpc_rqst *req, struct xdr_stream *xdr, void *result) { - return decode_attrstat(xdr, result, NULL); + return decode_attrstat(xdr, result, NULL, rpc_rqst_userns(req)); } static int nfs2_xdr_dec_diropres(struct rpc_rqst *req, struct xdr_stream *xdr, void *result) { - return decode_diropres(xdr, result); + return decode_diropres(xdr, result, rpc_rqst_userns(req)); } /* @@ -864,7 +883,7 @@ static int nfs2_xdr_dec_readres(struct rpc_rqst *req, struct xdr_stream *xdr, result->op_status = status; if (status != NFS_OK) goto out_default; - error = decode_fattr(xdr, result->fattr); + error = decode_fattr(xdr, result->fattr, rpc_rqst_userns(req)); if (unlikely(error)) goto out; error = decode_nfsdata(xdr, result); @@ -881,7 +900,8 @@ static int nfs2_xdr_dec_writeres(struct rpc_rqst *req, struct xdr_stream *xdr, /* All NFSv2 writes are "file sync" writes */ result->verf->committed = NFS_FILE_SYNC; - return decode_attrstat(xdr, result->fattr, &result->op_status); + return decode_attrstat(xdr, result->fattr, &result->op_status, + rpc_rqst_userns(req)); } /** diff --git a/fs/nfs/nfs3client.c b/fs/nfs/nfs3client.c index 7879f2a0fcfd..1afdb0f7473f 100644 --- a/fs/nfs/nfs3client.c +++ b/fs/nfs/nfs3client.c @@ -91,6 +91,7 @@ struct nfs_client *nfs3_set_ds_client(struct nfs_server *mds_srv, .proto = ds_proto, .net = mds_clp->cl_net, .timeparms = &ds_timeout, + .cred = mds_srv->cred, }; struct nfs_client *clp; char buf[INET6_ADDRSTRLEN + 1]; diff --git a/fs/nfs/nfs3xdr.c b/fs/nfs/nfs3xdr.c index 110358f4986d..abbbdde97e31 100644 --- a/fs/nfs/nfs3xdr.c +++ b/fs/nfs/nfs3xdr.c @@ -104,6 +104,20 @@ static const umode_t nfs_type2fmt[] = { [NF3FIFO] = S_IFIFO, }; +static struct user_namespace *rpc_userns(const struct rpc_clnt *clnt) +{ + if (clnt && clnt->cl_cred) + return clnt->cl_cred->user_ns; + return &init_user_ns; +} + +static struct user_namespace *rpc_rqst_userns(const struct rpc_rqst *rqstp) +{ + if (rqstp->rq_task) + return rpc_userns(rqstp->rq_task->tk_client); + return &init_user_ns; +} + /* * Encode/decode NFSv3 basic data types * @@ -516,7 +530,8 @@ static __be32 *xdr_decode_nfstime3(__be32 *p, struct timespec *timep) * set_mtime mtime; * }; */ -static void encode_sattr3(struct xdr_stream *xdr, const struct iattr *attr) +static void encode_sattr3(struct xdr_stream *xdr, const struct iattr *attr, + struct user_namespace *userns) { struct timespec ts; u32 nbytes; @@ -551,13 +566,13 @@ static void encode_sattr3(struct xdr_stream *xdr, const struct iattr *attr) if (attr->ia_valid & ATTR_UID) { *p++ = xdr_one; - *p++ = cpu_to_be32(from_kuid(&init_user_ns, attr->ia_uid)); + *p++ = cpu_to_be32(from_kuid_munged(userns, attr->ia_uid)); } else *p++ = xdr_zero; if (attr->ia_valid & ATTR_GID) { *p++ = xdr_one; - *p++ = cpu_to_be32(from_kgid(&init_user_ns, attr->ia_gid)); + *p++ = cpu_to_be32(from_kgid_munged(userns, attr->ia_gid)); } else *p++ = xdr_zero; @@ -606,7 +621,8 @@ static void encode_sattr3(struct xdr_stream *xdr, const struct iattr *attr) * nfstime3 ctime; * }; */ -static int decode_fattr3(struct xdr_stream *xdr, struct nfs_fattr *fattr) +static int decode_fattr3(struct xdr_stream *xdr, struct nfs_fattr *fattr, + struct user_namespace *userns) { umode_t fmode; __be32 *p; @@ -619,10 +635,10 @@ static int decode_fattr3(struct xdr_stream *xdr, struct nfs_fattr *fattr) fattr->mode = (be32_to_cpup(p++) & ~S_IFMT) | fmode; fattr->nlink = be32_to_cpup(p++); - fattr->uid = make_kuid(&init_user_ns, be32_to_cpup(p++)); + fattr->uid = make_kuid(userns, be32_to_cpup(p++)); if (!uid_valid(fattr->uid)) goto out_uid; - fattr->gid = make_kgid(&init_user_ns, be32_to_cpup(p++)); + fattr->gid = make_kgid(userns, be32_to_cpup(p++)); if (!gid_valid(fattr->gid)) goto out_gid; @@ -659,7 +675,8 @@ out_gid: * void; * }; */ -static int decode_post_op_attr(struct xdr_stream *xdr, struct nfs_fattr *fattr) +static int decode_post_op_attr(struct xdr_stream *xdr, struct nfs_fattr *fattr, + struct user_namespace *userns) { __be32 *p; @@ -667,7 +684,7 @@ static int decode_post_op_attr(struct xdr_stream *xdr, struct nfs_fattr *fattr) if (unlikely(!p)) return -EIO; if (*p != xdr_zero) - return decode_fattr3(xdr, fattr); + return decode_fattr3(xdr, fattr, userns); return 0; } @@ -728,14 +745,15 @@ static int decode_pre_op_attr(struct xdr_stream *xdr, struct nfs_fattr *fattr) return 0; } -static int decode_wcc_data(struct xdr_stream *xdr, struct nfs_fattr *fattr) +static int decode_wcc_data(struct xdr_stream *xdr, struct nfs_fattr *fattr, + struct user_namespace *userns) { int error; error = decode_pre_op_attr(xdr, fattr); if (unlikely(error)) goto out; - error = decode_post_op_attr(xdr, fattr); + error = decode_post_op_attr(xdr, fattr, userns); out: return error; } @@ -837,7 +855,7 @@ static void nfs3_xdr_enc_setattr3args(struct rpc_rqst *req, { const struct nfs3_sattrargs *args = data; encode_nfs_fh3(xdr, args->fh); - encode_sattr3(xdr, args->sattr); + encode_sattr3(xdr, args->sattr, rpc_rqst_userns(req)); encode_sattrguard3(xdr, args); } @@ -998,13 +1016,14 @@ static void nfs3_xdr_enc_write3args(struct rpc_rqst *req, * }; */ static void encode_createhow3(struct xdr_stream *xdr, - const struct nfs3_createargs *args) + const struct nfs3_createargs *args, + struct user_namespace *userns) { encode_uint32(xdr, args->createmode); switch (args->createmode) { case NFS3_CREATE_UNCHECKED: case NFS3_CREATE_GUARDED: - encode_sattr3(xdr, args->sattr); + encode_sattr3(xdr, args->sattr, userns); break; case NFS3_CREATE_EXCLUSIVE: encode_createverf3(xdr, args->verifier); @@ -1021,7 +1040,7 @@ static void nfs3_xdr_enc_create3args(struct rpc_rqst *req, const struct nfs3_createargs *args = data; encode_diropargs3(xdr, args->fh, args->name, args->len); - encode_createhow3(xdr, args); + encode_createhow3(xdr, args, rpc_rqst_userns(req)); } /* @@ -1039,7 +1058,7 @@ static void nfs3_xdr_enc_mkdir3args(struct rpc_rqst *req, const struct nfs3_mkdirargs *args = data; encode_diropargs3(xdr, args->fh, args->name, args->len); - encode_sattr3(xdr, args->sattr); + encode_sattr3(xdr, args->sattr, rpc_rqst_userns(req)); } /* @@ -1056,11 +1075,12 @@ static void nfs3_xdr_enc_mkdir3args(struct rpc_rqst *req, * }; */ static void encode_symlinkdata3(struct xdr_stream *xdr, - const void *data) + const void *data, + struct user_namespace *userns) { const struct nfs3_symlinkargs *args = data; - encode_sattr3(xdr, args->sattr); + encode_sattr3(xdr, args->sattr, userns); encode_nfspath3(xdr, args->pages, args->pathlen); } @@ -1071,7 +1091,7 @@ static void nfs3_xdr_enc_symlink3args(struct rpc_rqst *req, const struct nfs3_symlinkargs *args = data; encode_diropargs3(xdr, args->fromfh, args->fromname, args->fromlen); - encode_symlinkdata3(xdr, args); + encode_symlinkdata3(xdr, args, rpc_rqst_userns(req)); xdr->buf->flags |= XDRBUF_WRITE; } @@ -1100,24 +1120,26 @@ static void nfs3_xdr_enc_symlink3args(struct rpc_rqst *req, * }; */ static void encode_devicedata3(struct xdr_stream *xdr, - const struct nfs3_mknodargs *args) + const struct nfs3_mknodargs *args, + struct user_namespace *userns) { - encode_sattr3(xdr, args->sattr); + encode_sattr3(xdr, args->sattr, userns); encode_specdata3(xdr, args->rdev); } static void encode_mknoddata3(struct xdr_stream *xdr, - const struct nfs3_mknodargs *args) + const struct nfs3_mknodargs *args, + struct user_namespace *userns) { encode_ftype3(xdr, args->type); switch (args->type) { case NF3CHR: case NF3BLK: - encode_devicedata3(xdr, args); + encode_devicedata3(xdr, args, userns); break; case NF3SOCK: case NF3FIFO: - encode_sattr3(xdr, args->sattr); + encode_sattr3(xdr, args->sattr, userns); break; case NF3REG: case NF3DIR: @@ -1134,7 +1156,7 @@ static void nfs3_xdr_enc_mknod3args(struct rpc_rqst *req, const struct nfs3_mknodargs *args = data; encode_diropargs3(xdr, args->fh, args->name, args->len); - encode_mknoddata3(xdr, args); + encode_mknoddata3(xdr, args, rpc_rqst_userns(req)); } /* @@ -1379,7 +1401,7 @@ static int nfs3_xdr_dec_getattr3res(struct rpc_rqst *req, goto out; if (status != NFS3_OK) goto out_default; - error = decode_fattr3(xdr, result); + error = decode_fattr3(xdr, result, rpc_rqst_userns(req)); out: return error; out_default: @@ -1414,7 +1436,7 @@ static int nfs3_xdr_dec_setattr3res(struct rpc_rqst *req, error = decode_nfsstat3(xdr, &status); if (unlikely(error)) goto out; - error = decode_wcc_data(xdr, result); + error = decode_wcc_data(xdr, result, rpc_rqst_userns(req)); if (unlikely(error)) goto out; if (status != NFS3_OK) @@ -1449,6 +1471,7 @@ static int nfs3_xdr_dec_lookup3res(struct rpc_rqst *req, struct xdr_stream *xdr, void *data) { + struct user_namespace *userns = rpc_rqst_userns(req); struct nfs3_diropres *result = data; enum nfs_stat status; int error; @@ -1461,14 +1484,14 @@ static int nfs3_xdr_dec_lookup3res(struct rpc_rqst *req, error = decode_nfs_fh3(xdr, result->fh); if (unlikely(error)) goto out; - error = decode_post_op_attr(xdr, result->fattr); + error = decode_post_op_attr(xdr, result->fattr, userns); if (unlikely(error)) goto out; - error = decode_post_op_attr(xdr, result->dir_attr); + error = decode_post_op_attr(xdr, result->dir_attr, userns); out: return error; out_default: - error = decode_post_op_attr(xdr, result->dir_attr); + error = decode_post_op_attr(xdr, result->dir_attr, userns); if (unlikely(error)) goto out; return nfs3_stat_to_errno(status); @@ -1504,7 +1527,7 @@ static int nfs3_xdr_dec_access3res(struct rpc_rqst *req, error = decode_nfsstat3(xdr, &status); if (unlikely(error)) goto out; - error = decode_post_op_attr(xdr, result->fattr); + error = decode_post_op_attr(xdr, result->fattr, rpc_rqst_userns(req)); if (unlikely(error)) goto out; if (status != NFS3_OK) @@ -1545,7 +1568,7 @@ static int nfs3_xdr_dec_readlink3res(struct rpc_rqst *req, error = decode_nfsstat3(xdr, &status); if (unlikely(error)) goto out; - error = decode_post_op_attr(xdr, result); + error = decode_post_op_attr(xdr, result, rpc_rqst_userns(req)); if (unlikely(error)) goto out; if (status != NFS3_OK) @@ -1623,7 +1646,7 @@ static int nfs3_xdr_dec_read3res(struct rpc_rqst *req, struct xdr_stream *xdr, error = decode_nfsstat3(xdr, &status); if (unlikely(error)) goto out; - error = decode_post_op_attr(xdr, result->fattr); + error = decode_post_op_attr(xdr, result->fattr, rpc_rqst_userns(req)); if (unlikely(error)) goto out; result->op_status = status; @@ -1694,7 +1717,7 @@ static int nfs3_xdr_dec_write3res(struct rpc_rqst *req, struct xdr_stream *xdr, error = decode_nfsstat3(xdr, &status); if (unlikely(error)) goto out; - error = decode_wcc_data(xdr, result->fattr); + error = decode_wcc_data(xdr, result->fattr, rpc_rqst_userns(req)); if (unlikely(error)) goto out; result->op_status = status; @@ -1728,14 +1751,15 @@ out_status: * }; */ static int decode_create3resok(struct xdr_stream *xdr, - struct nfs3_diropres *result) + struct nfs3_diropres *result, + struct user_namespace *userns) { int error; error = decode_post_op_fh3(xdr, result->fh); if (unlikely(error)) goto out; - error = decode_post_op_attr(xdr, result->fattr); + error = decode_post_op_attr(xdr, result->fattr, userns); if (unlikely(error)) goto out; /* The server isn't required to return a file handle. @@ -1744,7 +1768,7 @@ static int decode_create3resok(struct xdr_stream *xdr, * values for the new object. */ if (result->fh->size == 0) result->fattr->valid = 0; - error = decode_wcc_data(xdr, result->dir_attr); + error = decode_wcc_data(xdr, result->dir_attr, userns); out: return error; } @@ -1753,6 +1777,7 @@ static int nfs3_xdr_dec_create3res(struct rpc_rqst *req, struct xdr_stream *xdr, void *data) { + struct user_namespace *userns = rpc_rqst_userns(req); struct nfs3_diropres *result = data; enum nfs_stat status; int error; @@ -1762,11 +1787,11 @@ static int nfs3_xdr_dec_create3res(struct rpc_rqst *req, goto out; if (status != NFS3_OK) goto out_default; - error = decode_create3resok(xdr, result); + error = decode_create3resok(xdr, result, userns); out: return error; out_default: - error = decode_wcc_data(xdr, result->dir_attr); + error = decode_wcc_data(xdr, result->dir_attr, userns); if (unlikely(error)) goto out; return nfs3_stat_to_errno(status); @@ -1801,7 +1826,7 @@ static int nfs3_xdr_dec_remove3res(struct rpc_rqst *req, error = decode_nfsstat3(xdr, &status); if (unlikely(error)) goto out; - error = decode_wcc_data(xdr, result->dir_attr); + error = decode_wcc_data(xdr, result->dir_attr, rpc_rqst_userns(req)); if (unlikely(error)) goto out; if (status != NFS3_OK) @@ -1836,6 +1861,7 @@ static int nfs3_xdr_dec_rename3res(struct rpc_rqst *req, struct xdr_stream *xdr, void *data) { + struct user_namespace *userns = rpc_rqst_userns(req); struct nfs_renameres *result = data; enum nfs_stat status; int error; @@ -1843,10 +1869,10 @@ static int nfs3_xdr_dec_rename3res(struct rpc_rqst *req, error = decode_nfsstat3(xdr, &status); if (unlikely(error)) goto out; - error = decode_wcc_data(xdr, result->old_fattr); + error = decode_wcc_data(xdr, result->old_fattr, userns); if (unlikely(error)) goto out; - error = decode_wcc_data(xdr, result->new_fattr); + error = decode_wcc_data(xdr, result->new_fattr, userns); if (unlikely(error)) goto out; if (status != NFS3_OK) @@ -1880,6 +1906,7 @@ out_status: static int nfs3_xdr_dec_link3res(struct rpc_rqst *req, struct xdr_stream *xdr, void *data) { + struct user_namespace *userns = rpc_rqst_userns(req); struct nfs3_linkres *result = data; enum nfs_stat status; int error; @@ -1887,10 +1914,10 @@ static int nfs3_xdr_dec_link3res(struct rpc_rqst *req, struct xdr_stream *xdr, error = decode_nfsstat3(xdr, &status); if (unlikely(error)) goto out; - error = decode_post_op_attr(xdr, result->fattr); + error = decode_post_op_attr(xdr, result->fattr, userns); if (unlikely(error)) goto out; - error = decode_wcc_data(xdr, result->dir_attr); + error = decode_wcc_data(xdr, result->dir_attr, userns); if (unlikely(error)) goto out; if (status != NFS3_OK) @@ -1939,6 +1966,7 @@ out_status: int nfs3_decode_dirent(struct xdr_stream *xdr, struct nfs_entry *entry, bool plus) { + struct user_namespace *userns = rpc_userns(entry->server->client); struct nfs_entry old = *entry; __be32 *p; int error; @@ -1973,7 +2001,7 @@ int nfs3_decode_dirent(struct xdr_stream *xdr, struct nfs_entry *entry, if (plus) { entry->fattr->valid = 0; - error = decode_post_op_attr(xdr, entry->fattr); + error = decode_post_op_attr(xdr, entry->fattr, userns); if (unlikely(error)) return error; if (entry->fattr->valid & NFS_ATTR_FATTR_V3) @@ -2045,11 +2073,12 @@ static int decode_dirlist3(struct xdr_stream *xdr) } static int decode_readdir3resok(struct xdr_stream *xdr, - struct nfs3_readdirres *result) + struct nfs3_readdirres *result, + struct user_namespace *userns) { int error; - error = decode_post_op_attr(xdr, result->dir_attr); + error = decode_post_op_attr(xdr, result->dir_attr, userns); if (unlikely(error)) goto out; /* XXX: do we need to check if result->verf != NULL ? */ @@ -2074,11 +2103,11 @@ static int nfs3_xdr_dec_readdir3res(struct rpc_rqst *req, goto out; if (status != NFS3_OK) goto out_default; - error = decode_readdir3resok(xdr, result); + error = decode_readdir3resok(xdr, result, rpc_rqst_userns(req)); out: return error; out_default: - error = decode_post_op_attr(xdr, result->dir_attr); + error = decode_post_op_attr(xdr, result->dir_attr, rpc_rqst_userns(req)); if (unlikely(error)) goto out; return nfs3_stat_to_errno(status); @@ -2138,7 +2167,7 @@ static int nfs3_xdr_dec_fsstat3res(struct rpc_rqst *req, error = decode_nfsstat3(xdr, &status); if (unlikely(error)) goto out; - error = decode_post_op_attr(xdr, result->fattr); + error = decode_post_op_attr(xdr, result->fattr, rpc_rqst_userns(req)); if (unlikely(error)) goto out; if (status != NFS3_OK) @@ -2212,7 +2241,7 @@ static int nfs3_xdr_dec_fsinfo3res(struct rpc_rqst *req, error = decode_nfsstat3(xdr, &status); if (unlikely(error)) goto out; - error = decode_post_op_attr(xdr, result->fattr); + error = decode_post_op_attr(xdr, result->fattr, rpc_rqst_userns(req)); if (unlikely(error)) goto out; if (status != NFS3_OK) @@ -2273,7 +2302,7 @@ static int nfs3_xdr_dec_pathconf3res(struct rpc_rqst *req, error = decode_nfsstat3(xdr, &status); if (unlikely(error)) goto out; - error = decode_post_op_attr(xdr, result->fattr); + error = decode_post_op_attr(xdr, result->fattr, rpc_rqst_userns(req)); if (unlikely(error)) goto out; if (status != NFS3_OK) @@ -2315,7 +2344,7 @@ static int nfs3_xdr_dec_commit3res(struct rpc_rqst *req, error = decode_nfsstat3(xdr, &status); if (unlikely(error)) goto out; - error = decode_wcc_data(xdr, result->fattr); + error = decode_wcc_data(xdr, result->fattr, rpc_rqst_userns(req)); if (unlikely(error)) goto out; result->op_status = status; @@ -2331,14 +2360,15 @@ out_status: #ifdef CONFIG_NFS_V3_ACL static inline int decode_getacl3resok(struct xdr_stream *xdr, - struct nfs3_getaclres *result) + struct nfs3_getaclres *result, + struct user_namespace *userns) { struct posix_acl **acl; unsigned int *aclcnt; size_t hdrlen; int error; - error = decode_post_op_attr(xdr, result->fattr); + error = decode_post_op_attr(xdr, result->fattr, userns); if (unlikely(error)) goto out; error = decode_uint32(xdr, &result->mask); @@ -2386,7 +2416,7 @@ static int nfs3_xdr_dec_getacl3res(struct rpc_rqst *req, goto out; if (status != NFS3_OK) goto out_default; - error = decode_getacl3resok(xdr, result); + error = decode_getacl3resok(xdr, result, rpc_rqst_userns(req)); out: return error; out_default: @@ -2405,7 +2435,7 @@ static int nfs3_xdr_dec_setacl3res(struct rpc_rqst *req, goto out; if (status != NFS3_OK) goto out_default; - error = decode_post_op_attr(xdr, result); + error = decode_post_op_attr(xdr, result, rpc_rqst_userns(req)); out: return error; out_default: diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h index 06ac3d9ac7c6..8a38a254f516 100644 --- a/fs/nfs/nfs4_fs.h +++ b/fs/nfs/nfs4_fs.h @@ -206,6 +206,7 @@ struct nfs4_exception { unsigned char delay : 1, recovering : 1, retry : 1; + bool interruptible; }; struct nfs4_state_recovery_ops { diff --git a/fs/nfs/nfs4client.c b/fs/nfs/nfs4client.c index 1339ede979af..3ce246346f02 100644 --- a/fs/nfs/nfs4client.c +++ b/fs/nfs/nfs4client.c @@ -870,6 +870,7 @@ static int nfs4_set_client(struct nfs_server *server, .minorversion = minorversion, .net = net, .timeparms = timeparms, + .cred = server->cred, }; struct nfs_client *clp; @@ -931,6 +932,7 @@ struct nfs_client *nfs4_set_ds_client(struct nfs_server *mds_srv, .minorversion = minor_version, .net = mds_clp->cl_net, .timeparms = &ds_timeout, + .cred = mds_srv->cred, }; char buf[INET6_ADDRSTRLEN + 1]; @@ -1107,6 +1109,8 @@ struct nfs_server *nfs4_create_server(struct nfs_mount_info *mount_info, if (!server) return ERR_PTR(-ENOMEM); + server->cred = get_cred(current_cred()); + auth_probe = mount_info->parsed->auth_info.flavor_len < 1; /* set up the general RPC client */ @@ -1143,6 +1147,8 @@ struct nfs_server *nfs4_create_referral_server(struct nfs_clone_mount *data, parent_server = NFS_SB(data->sb); parent_client = parent_server->nfs_client; + server->cred = get_cred(parent_server->cred); + /* Initialise the client representation from the parent server */ nfs_server_copy_userdata(server, parent_server); diff --git a/fs/nfs/nfs4file.c b/fs/nfs/nfs4file.c index 00d17198ee12..cf42a8b939e3 100644 --- a/fs/nfs/nfs4file.c +++ b/fs/nfs/nfs4file.c @@ -125,7 +125,7 @@ nfs4_file_flush(struct file *file, fl_owner_t id) return filemap_fdatawrite(file->f_mapping); /* Flush writes to the server and return any errors */ - return vfs_fsync(file, 0); + return nfs_wb_all(inode); } #ifdef CONFIG_NFS_V4_2 @@ -187,7 +187,7 @@ static loff_t nfs42_remap_file_range(struct file *src_file, loff_t src_off, bool same_inode = false; int ret; - if (remap_flags & ~REMAP_FILE_ADVISORY) + if (remap_flags & ~(REMAP_FILE_DEDUP | REMAP_FILE_ADVISORY)) return -EINVAL; /* check alignment w.r.t. clone_blksize */ diff --git a/fs/nfs/nfs4idmap.c b/fs/nfs/nfs4idmap.c index bf34ddaa2ad7..4884fdae28fb 100644 --- a/fs/nfs/nfs4idmap.c +++ b/fs/nfs/nfs4idmap.c @@ -69,8 +69,16 @@ struct idmap { struct rpc_pipe *idmap_pipe; struct idmap_legacy_upcalldata *idmap_upcall_data; struct mutex idmap_mutex; + const struct cred *cred; }; +static struct user_namespace *idmap_userns(const struct idmap *idmap) +{ + if (idmap && idmap->cred) + return idmap->cred->user_ns; + return &init_user_ns; +} + /** * nfs_fattr_init_names - initialise the nfs_fattr owner_name/group_name fields * @fattr: fully initialised struct nfs_fattr @@ -271,14 +279,15 @@ static struct key *nfs_idmap_request_key(const char *name, size_t namelen, const char *type, struct idmap *idmap) { char *desc; - struct key *rkey; + struct key *rkey = ERR_PTR(-EAGAIN); ssize_t ret; ret = nfs_idmap_get_desc(name, namelen, type, strlen(type), &desc); if (ret < 0) return ERR_PTR(ret); - rkey = request_key(&key_type_id_resolver, desc, ""); + if (!idmap->cred || idmap->cred->user_ns == &init_user_ns) + rkey = request_key(&key_type_id_resolver, desc, ""); if (IS_ERR(rkey)) { mutex_lock(&idmap->idmap_mutex); rkey = request_key_with_auxdata(&key_type_id_resolver_legacy, @@ -452,6 +461,9 @@ nfs_idmap_new(struct nfs_client *clp) if (idmap == NULL) return -ENOMEM; + mutex_init(&idmap->idmap_mutex); + idmap->cred = get_cred(clp->cl_rpcclient->cl_cred); + rpc_init_pipe_dir_object(&idmap->idmap_pdo, &nfs_idmap_pipe_dir_object_ops, idmap); @@ -462,7 +474,6 @@ nfs_idmap_new(struct nfs_client *clp) goto err; } idmap->idmap_pipe = pipe; - mutex_init(&idmap->idmap_mutex); error = rpc_add_pipe_dir_object(clp->cl_net, &clp->cl_rpcclient->cl_pipedir_objects, @@ -475,6 +486,7 @@ nfs_idmap_new(struct nfs_client *clp) err_destroy_pipe: rpc_destroy_pipe_data(idmap->idmap_pipe); err: + put_cred(idmap->cred); kfree(idmap); return error; } @@ -491,6 +503,7 @@ nfs_idmap_delete(struct nfs_client *clp) &clp->cl_rpcclient->cl_pipedir_objects, &idmap->idmap_pdo); rpc_destroy_pipe_data(idmap->idmap_pipe); + put_cred(idmap->cred); kfree(idmap); } @@ -735,7 +748,7 @@ int nfs_map_name_to_uid(const struct nfs_server *server, const char *name, size_ if (!nfs_map_string_to_numeric(name, namelen, &id)) ret = nfs_idmap_lookup_id(name, namelen, "uid", &id, idmap); if (ret == 0) { - *uid = make_kuid(&init_user_ns, id); + *uid = make_kuid(idmap_userns(idmap), id); if (!uid_valid(*uid)) ret = -ERANGE; } @@ -752,7 +765,7 @@ int nfs_map_group_to_gid(const struct nfs_server *server, const char *name, size if (!nfs_map_string_to_numeric(name, namelen, &id)) ret = nfs_idmap_lookup_id(name, namelen, "gid", &id, idmap); if (ret == 0) { - *gid = make_kgid(&init_user_ns, id); + *gid = make_kgid(idmap_userns(idmap), id); if (!gid_valid(*gid)) ret = -ERANGE; } @@ -766,7 +779,7 @@ int nfs_map_uid_to_name(const struct nfs_server *server, kuid_t uid, char *buf, int ret = -EINVAL; __u32 id; - id = from_kuid(&init_user_ns, uid); + id = from_kuid_munged(idmap_userns(idmap), uid); if (!(server->caps & NFS_CAP_UIDGID_NOMAP)) ret = nfs_idmap_lookup_name(id, "user", buf, buflen, idmap); if (ret < 0) @@ -780,7 +793,7 @@ int nfs_map_gid_to_group(const struct nfs_server *server, kgid_t gid, char *buf, int ret = -EINVAL; __u32 id; - id = from_kgid(&init_user_ns, gid); + id = from_kgid_munged(idmap_userns(idmap), gid); if (!(server->caps & NFS_CAP_UIDGID_NOMAP)) ret = nfs_idmap_lookup_name(id, "group", buf, buflen, idmap); if (ret < 0) diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 741ff8c9c6ed..c29cbef6b53f 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -400,17 +400,32 @@ static long nfs4_update_delay(long *timeout) return ret; } -static int nfs4_delay(struct rpc_clnt *clnt, long *timeout) +static int nfs4_delay_killable(long *timeout) { - int res = 0; - might_sleep(); freezable_schedule_timeout_killable_unsafe( nfs4_update_delay(timeout)); - if (fatal_signal_pending(current)) - res = -ERESTARTSYS; - return res; + if (!__fatal_signal_pending(current)) + return 0; + return -EINTR; +} + +static int nfs4_delay_interruptible(long *timeout) +{ + might_sleep(); + + freezable_schedule_timeout_interruptible(nfs4_update_delay(timeout)); + if (!signal_pending(current)) + return 0; + return __fatal_signal_pending(current) ? -EINTR :-ERESTARTSYS; +} + +static int nfs4_delay(long *timeout, bool interruptible) +{ + if (interruptible) + return nfs4_delay_interruptible(timeout); + return nfs4_delay_killable(timeout); } /* This is the error handling routine for processes that are allowed @@ -546,7 +561,8 @@ int nfs4_handle_exception(struct nfs_server *server, int errorcode, struct nfs4_ ret = nfs4_do_handle_exception(server, errorcode, exception); if (exception->delay) { - ret = nfs4_delay(server->client, &exception->timeout); + ret = nfs4_delay(&exception->timeout, + exception->interruptible); goto out_retry; } if (exception->recovering) { @@ -978,10 +994,8 @@ int nfs4_setup_sequence(struct nfs_client *client, if (res->sr_slot != NULL) goto out_start; - if (session) { + if (session) tbl = &session->fc_slot_table; - task->tk_timeout = 0; - } spin_lock(&tbl->slot_tbl_lock); /* The state manager will wait until the slot table is empty */ @@ -990,9 +1004,8 @@ int nfs4_setup_sequence(struct nfs_client *client, slot = nfs4_alloc_slot(tbl); if (IS_ERR(slot)) { - /* Try again in 1/4 second */ if (slot == ERR_PTR(-ENOMEM)) - task->tk_timeout = HZ >> 2; + goto out_sleep_timeout; goto out_sleep; } spin_unlock(&tbl->slot_tbl_lock); @@ -1004,11 +1017,20 @@ out_start: nfs41_sequence_res_init(res); rpc_call_start(task); return 0; - +out_sleep_timeout: + /* Try again in 1/4 second */ + if (args->sa_privileged) + rpc_sleep_on_priority_timeout(&tbl->slot_tbl_waitq, task, + jiffies + (HZ >> 2), RPC_PRIORITY_PRIVILEGED); + else + rpc_sleep_on_timeout(&tbl->slot_tbl_waitq, task, + NULL, jiffies + (HZ >> 2)); + spin_unlock(&tbl->slot_tbl_lock); + return -EAGAIN; out_sleep: if (args->sa_privileged) rpc_sleep_on_priority(&tbl->slot_tbl_waitq, task, - NULL, RPC_PRIORITY_PRIVILEGED); + RPC_PRIORITY_PRIVILEGED); else rpc_sleep_on(&tbl->slot_tbl_waitq, task, NULL); spin_unlock(&tbl->slot_tbl_lock); @@ -3060,7 +3082,9 @@ static struct nfs4_state *nfs4_do_open(struct inode *dir, int *opened) { struct nfs_server *server = NFS_SERVER(dir); - struct nfs4_exception exception = { }; + struct nfs4_exception exception = { + .interruptible = true, + }; struct nfs4_state *res; struct nfs4_open_createattrs c = { .label = label, @@ -3673,7 +3697,9 @@ static int _nfs4_server_capabilities(struct nfs_server *server, struct nfs_fh *f int nfs4_server_capabilities(struct nfs_server *server, struct nfs_fh *fhandle) { - struct nfs4_exception exception = { }; + struct nfs4_exception exception = { + .interruptible = true, + }; int err; do { err = nfs4_handle_exception(server, @@ -3715,7 +3741,9 @@ static int _nfs4_lookup_root(struct nfs_server *server, struct nfs_fh *fhandle, static int nfs4_lookup_root(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_fsinfo *info) { - struct nfs4_exception exception = { }; + struct nfs4_exception exception = { + .interruptible = true, + }; int err; do { err = _nfs4_lookup_root(server, fhandle, info); @@ -3942,7 +3970,9 @@ static int nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_fattr *fattr, struct nfs4_label *label, struct inode *inode) { - struct nfs4_exception exception = { }; + struct nfs4_exception exception = { + .interruptible = true, + }; int err; do { err = _nfs4_proc_getattr(server, fhandle, fattr, label, inode); @@ -4065,7 +4095,9 @@ static int nfs4_proc_lookup_common(struct rpc_clnt **clnt, struct inode *dir, const struct qstr *name, struct nfs_fh *fhandle, struct nfs_fattr *fattr, struct nfs4_label *label) { - struct nfs4_exception exception = { }; + struct nfs4_exception exception = { + .interruptible = true, + }; struct rpc_clnt *client = *clnt; int err; do { @@ -4169,7 +4201,9 @@ static int _nfs4_proc_lookupp(struct inode *inode, static int nfs4_proc_lookupp(struct inode *inode, struct nfs_fh *fhandle, struct nfs_fattr *fattr, struct nfs4_label *label) { - struct nfs4_exception exception = { }; + struct nfs4_exception exception = { + .interruptible = true, + }; int err; do { err = _nfs4_proc_lookupp(inode, fhandle, fattr, label); @@ -4216,7 +4250,9 @@ static int _nfs4_proc_access(struct inode *inode, struct nfs_access_entry *entry static int nfs4_proc_access(struct inode *inode, struct nfs_access_entry *entry) { - struct nfs4_exception exception = { }; + struct nfs4_exception exception = { + .interruptible = true, + }; int err; do { err = _nfs4_proc_access(inode, entry); @@ -4271,7 +4307,9 @@ static int _nfs4_proc_readlink(struct inode *inode, struct page *page, static int nfs4_proc_readlink(struct inode *inode, struct page *page, unsigned int pgbase, unsigned int pglen) { - struct nfs4_exception exception = { }; + struct nfs4_exception exception = { + .interruptible = true, + }; int err; do { err = _nfs4_proc_readlink(inode, page, pgbase, pglen); @@ -4347,7 +4385,9 @@ _nfs4_proc_remove(struct inode *dir, const struct qstr *name, u32 ftype) static int nfs4_proc_remove(struct inode *dir, struct dentry *dentry) { - struct nfs4_exception exception = { }; + struct nfs4_exception exception = { + .interruptible = true, + }; struct inode *inode = d_inode(dentry); int err; @@ -4368,7 +4408,9 @@ static int nfs4_proc_remove(struct inode *dir, struct dentry *dentry) static int nfs4_proc_rmdir(struct inode *dir, const struct qstr *name) { - struct nfs4_exception exception = { }; + struct nfs4_exception exception = { + .interruptible = true, + }; int err; do { @@ -4527,7 +4569,9 @@ out: static int nfs4_proc_link(struct inode *inode, struct inode *dir, const struct qstr *name) { - struct nfs4_exception exception = { }; + struct nfs4_exception exception = { + .interruptible = true, + }; int err; do { err = nfs4_handle_exception(NFS_SERVER(inode), @@ -4634,7 +4678,9 @@ out: static int nfs4_proc_symlink(struct inode *dir, struct dentry *dentry, struct page *page, unsigned int len, struct iattr *sattr) { - struct nfs4_exception exception = { }; + struct nfs4_exception exception = { + .interruptible = true, + }; struct nfs4_label l, *label = NULL; int err; @@ -4673,7 +4719,9 @@ static int nfs4_proc_mkdir(struct inode *dir, struct dentry *dentry, struct iattr *sattr) { struct nfs_server *server = NFS_SERVER(dir); - struct nfs4_exception exception = { }; + struct nfs4_exception exception = { + .interruptible = true, + }; struct nfs4_label l, *label = NULL; int err; @@ -4733,7 +4781,9 @@ static int _nfs4_proc_readdir(struct dentry *dentry, const struct cred *cred, static int nfs4_proc_readdir(struct dentry *dentry, const struct cred *cred, u64 cookie, struct page **pages, unsigned int count, bool plus) { - struct nfs4_exception exception = { }; + struct nfs4_exception exception = { + .interruptible = true, + }; int err; do { err = _nfs4_proc_readdir(dentry, cred, cookie, @@ -4784,7 +4834,9 @@ static int nfs4_proc_mknod(struct inode *dir, struct dentry *dentry, struct iattr *sattr, dev_t rdev) { struct nfs_server *server = NFS_SERVER(dir); - struct nfs4_exception exception = { }; + struct nfs4_exception exception = { + .interruptible = true, + }; struct nfs4_label l, *label = NULL; int err; @@ -4826,7 +4878,9 @@ static int _nfs4_proc_statfs(struct nfs_server *server, struct nfs_fh *fhandle, static int nfs4_proc_statfs(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_fsstat *fsstat) { - struct nfs4_exception exception = { }; + struct nfs4_exception exception = { + .interruptible = true, + }; int err; do { err = nfs4_handle_exception(server, @@ -4857,7 +4911,9 @@ static int _nfs4_do_fsinfo(struct nfs_server *server, struct nfs_fh *fhandle, static int nfs4_do_fsinfo(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_fsinfo *fsinfo) { - struct nfs4_exception exception = { }; + struct nfs4_exception exception = { + .interruptible = true, + }; unsigned long now = jiffies; int err; @@ -4919,7 +4975,9 @@ static int _nfs4_proc_pathconf(struct nfs_server *server, struct nfs_fh *fhandle static int nfs4_proc_pathconf(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_pathconf *pathconf) { - struct nfs4_exception exception = { }; + struct nfs4_exception exception = { + .interruptible = true, + }; int err; do { @@ -5488,7 +5546,9 @@ out_free: static ssize_t nfs4_get_acl_uncached(struct inode *inode, void *buf, size_t buflen) { - struct nfs4_exception exception = { }; + struct nfs4_exception exception = { + .interruptible = true, + }; ssize_t ret; do { ret = __nfs4_get_acl_uncached(inode, buf, buflen); @@ -5622,7 +5682,9 @@ static int _nfs4_get_security_label(struct inode *inode, void *buf, static int nfs4_get_security_label(struct inode *inode, void *buf, size_t buflen) { - struct nfs4_exception exception = { }; + struct nfs4_exception exception = { + .interruptible = true, + }; int err; if (!nfs_server_capable(inode, NFS_CAP_SECURITY_LABEL)) @@ -6263,7 +6325,9 @@ out: static int nfs4_proc_getlk(struct nfs4_state *state, int cmd, struct file_lock *request) { - struct nfs4_exception exception = { }; + struct nfs4_exception exception = { + .interruptible = true, + }; int err; do { @@ -6827,6 +6891,7 @@ static int nfs4_proc_setlk(struct nfs4_state *state, int cmd, struct file_lock * struct nfs4_exception exception = { .state = state, .inode = state->inode, + .interruptible = true, }; int err; @@ -7240,7 +7305,9 @@ int nfs4_proc_fs_locations(struct rpc_clnt *client, struct inode *dir, struct nfs4_fs_locations *fs_locations, struct page *page) { - struct nfs4_exception exception = { }; + struct nfs4_exception exception = { + .interruptible = true, + }; int err; do { err = _nfs4_proc_fs_locations(client, dir, name, @@ -7383,7 +7450,9 @@ int nfs4_proc_get_locations(struct inode *inode, struct nfs_client *clp = server->nfs_client; const struct nfs4_mig_recovery_ops *ops = clp->cl_mvops->mig_recovery_ops; - struct nfs4_exception exception = { }; + struct nfs4_exception exception = { + .interruptible = true, + }; int status; dprintk("%s: FSID %llx:%llx on \"%s\"\n", __func__, @@ -7507,7 +7576,9 @@ int nfs4_proc_fsid_present(struct inode *inode, const struct cred *cred) struct nfs_client *clp = server->nfs_client; const struct nfs4_mig_recovery_ops *ops = clp->cl_mvops->mig_recovery_ops; - struct nfs4_exception exception = { }; + struct nfs4_exception exception = { + .interruptible = true, + }; int status; dprintk("%s: FSID %llx:%llx on \"%s\"\n", __func__, @@ -7573,7 +7644,9 @@ static int _nfs4_proc_secinfo(struct inode *dir, const struct qstr *name, struct int nfs4_proc_secinfo(struct inode *dir, const struct qstr *name, struct nfs4_secinfo_flavors *flavors) { - struct nfs4_exception exception = { }; + struct nfs4_exception exception = { + .interruptible = true, + }; int err; do { err = -NFS4ERR_WRONGSEC; @@ -9263,7 +9336,9 @@ static int nfs41_proc_secinfo_no_name(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_fsinfo *info, struct nfs4_secinfo_flavors *flavors) { - struct nfs4_exception exception = { }; + struct nfs4_exception exception = { + .interruptible = true, + }; int err; do { /* first try using integrity protection */ @@ -9430,7 +9505,9 @@ static int nfs41_test_stateid(struct nfs_server *server, nfs4_stateid *stateid, const struct cred *cred) { - struct nfs4_exception exception = { }; + struct nfs4_exception exception = { + .interruptible = true, + }; int err; do { err = _nfs41_test_stateid(server, stateid, cred); diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c index 3de36479ed7a..e2e3c4f04d3e 100644 --- a/fs/nfs/nfs4state.c +++ b/fs/nfs/nfs4state.c @@ -159,6 +159,10 @@ int nfs40_discover_server_trunking(struct nfs_client *clp, /* Sustain the lease, even if it's empty. If the clientid4 * goes stale it's of no use for trunking discovery. */ nfs4_schedule_state_renewal(*result); + + /* If the client state need to recover, do it. */ + if (clp->cl_state) + nfs4_schedule_state_manager(clp); } out: return status; @@ -2346,8 +2350,7 @@ static void nfs41_handle_recallable_state_revoked(struct nfs_client *clp) { /* FIXME: For now, we destroy all layouts. */ pnfs_destroy_all_layouts(clp); - /* FIXME: For now, we test all delegations+open state+locks. */ - nfs41_handle_some_state_revoked(clp); + nfs_test_expired_all_delegations(clp); dprintk("%s: Recallable state revoked on server %s!\n", __func__, clp->cl_hostname); } diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c index e9f39fa5964b..6ec30014a439 100644 --- a/fs/nfs/pagelist.c +++ b/fs/nfs/pagelist.c @@ -16,8 +16,8 @@ #include <linux/nfs.h> #include <linux/nfs3.h> #include <linux/nfs4.h> -#include <linux/nfs_page.h> #include <linux/nfs_fs.h> +#include <linux/nfs_page.h> #include <linux/nfs_mount.h> #include <linux/export.h> @@ -47,7 +47,7 @@ void nfs_pgheader_init(struct nfs_pageio_descriptor *desc, hdr->req = nfs_list_entry(mirror->pg_list.next); hdr->inode = desc->pg_inode; - hdr->cred = hdr->req->wb_context->cred; + hdr->cred = nfs_req_openctx(hdr->req)->cred; hdr->io_start = req_offset(hdr->req); hdr->good_bytes = mirror->pg_count; hdr->io_completion = desc->pg_io_completion; @@ -295,25 +295,13 @@ out: nfs_release_request(head); } -/** - * nfs_create_request - Create an NFS read/write request. - * @ctx: open context to use - * @page: page to write - * @last: last nfs request created for this page group or NULL if head - * @offset: starting offset within the page for the write - * @count: number of bytes to read/write - * - * The page must be locked by the caller. This makes sure we never - * create two different requests for the same page. - * User should ensure it is safe to sleep in this function. - */ -struct nfs_page * -nfs_create_request(struct nfs_open_context *ctx, struct page *page, - struct nfs_page *last, unsigned int offset, +static struct nfs_page * +__nfs_create_request(struct nfs_lock_context *l_ctx, struct page *page, + unsigned int pgbase, unsigned int offset, unsigned int count) { struct nfs_page *req; - struct nfs_lock_context *l_ctx; + struct nfs_open_context *ctx = l_ctx->open_context; if (test_bit(NFS_CONTEXT_BAD, &ctx->flags)) return ERR_PTR(-EBADF); @@ -322,13 +310,8 @@ nfs_create_request(struct nfs_open_context *ctx, struct page *page, if (req == NULL) return ERR_PTR(-ENOMEM); - /* get lock context early so we can deal with alloc failures */ - l_ctx = nfs_get_lock_context(ctx); - if (IS_ERR(l_ctx)) { - nfs_page_free(req); - return ERR_CAST(l_ctx); - } req->wb_lock_context = l_ctx; + refcount_inc(&l_ctx->count); atomic_inc(&l_ctx->io_count); /* Initialize the request struct. Initially, we assume a @@ -340,15 +323,59 @@ nfs_create_request(struct nfs_open_context *ctx, struct page *page, get_page(page); } req->wb_offset = offset; - req->wb_pgbase = offset; + req->wb_pgbase = pgbase; req->wb_bytes = count; - req->wb_context = get_nfs_open_context(ctx); kref_init(&req->wb_kref); - nfs_page_group_init(req, last); + req->wb_nio = 0; return req; } /** + * nfs_create_request - Create an NFS read/write request. + * @ctx: open context to use + * @page: page to write + * @offset: starting offset within the page for the write + * @count: number of bytes to read/write + * + * The page must be locked by the caller. This makes sure we never + * create two different requests for the same page. + * User should ensure it is safe to sleep in this function. + */ +struct nfs_page * +nfs_create_request(struct nfs_open_context *ctx, struct page *page, + unsigned int offset, unsigned int count) +{ + struct nfs_lock_context *l_ctx = nfs_get_lock_context(ctx); + struct nfs_page *ret; + + if (IS_ERR(l_ctx)) + return ERR_CAST(l_ctx); + ret = __nfs_create_request(l_ctx, page, offset, offset, count); + if (!IS_ERR(ret)) + nfs_page_group_init(ret, NULL); + nfs_put_lock_context(l_ctx); + return ret; +} + +static struct nfs_page * +nfs_create_subreq(struct nfs_page *req, struct nfs_page *last, + unsigned int pgbase, unsigned int offset, + unsigned int count) +{ + struct nfs_page *ret; + + ret = __nfs_create_request(req->wb_lock_context, req->wb_page, + pgbase, offset, count); + if (!IS_ERR(ret)) { + nfs_lock_request(ret); + ret->wb_index = req->wb_index; + nfs_page_group_init(ret, last); + ret->wb_nio = req->wb_nio; + } + return ret; +} + +/** * nfs_unlock_request - Unlock request and wake up sleepers. * @req: pointer to request */ @@ -386,8 +413,8 @@ void nfs_unlock_and_release_request(struct nfs_page *req) static void nfs_clear_request(struct nfs_page *req) { struct page *page = req->wb_page; - struct nfs_open_context *ctx = req->wb_context; struct nfs_lock_context *l_ctx = req->wb_lock_context; + struct nfs_open_context *ctx; if (page != NULL) { put_page(page); @@ -396,16 +423,13 @@ static void nfs_clear_request(struct nfs_page *req) if (l_ctx != NULL) { if (atomic_dec_and_test(&l_ctx->io_count)) { wake_up_var(&l_ctx->io_count); + ctx = l_ctx->open_context; if (test_bit(NFS_CONTEXT_UNLOCK, &ctx->flags)) rpc_wake_up(&NFS_SERVER(d_inode(ctx->dentry))->uoc_rpcwaitq); } nfs_put_lock_context(l_ctx); req->wb_lock_context = NULL; } - if (ctx != NULL) { - put_nfs_open_context(ctx); - req->wb_context = NULL; - } } /** @@ -550,7 +574,7 @@ static void nfs_pgio_rpcsetup(struct nfs_pgio_header *hdr, hdr->args.pgbase = req->wb_pgbase; hdr->args.pages = hdr->page_array.pagevec; hdr->args.count = count; - hdr->args.context = get_nfs_open_context(req->wb_context); + hdr->args.context = get_nfs_open_context(nfs_req_openctx(req)); hdr->args.lock_context = req->wb_lock_context; hdr->args.stable = NFS_UNSTABLE; switch (how & (FLUSH_STABLE | FLUSH_COND_STABLE)) { @@ -698,6 +722,7 @@ void nfs_pageio_init(struct nfs_pageio_descriptor *desc, desc->pg_mirrors_dynamic = NULL; desc->pg_mirrors = desc->pg_mirrors_static; nfs_pageio_mirror_init(&desc->pg_mirrors[0], bsize); + desc->pg_maxretrans = 0; } /** @@ -906,9 +931,9 @@ static bool nfs_can_coalesce_requests(struct nfs_page *prev, struct file_lock_context *flctx; if (prev) { - if (!nfs_match_open_context(req->wb_context, prev->wb_context)) + if (!nfs_match_open_context(nfs_req_openctx(req), nfs_req_openctx(prev))) return false; - flctx = d_inode(req->wb_context->dentry)->i_flctx; + flctx = d_inode(nfs_req_openctx(req)->dentry)->i_flctx; if (flctx != NULL && !(list_empty_careful(&flctx->flc_posix) && list_empty_careful(&flctx->flc_flock)) && @@ -957,6 +982,15 @@ static int nfs_pageio_do_add_request(struct nfs_pageio_descriptor *desc, return 0; mirror->pg_base = req->wb_pgbase; } + + if (desc->pg_maxretrans && req->wb_nio > desc->pg_maxretrans) { + if (NFS_SERVER(desc->pg_inode)->flags & NFS_MOUNT_SOFTERR) + desc->pg_error = -ETIMEDOUT; + else + desc->pg_error = -EIO; + return 0; + } + if (!nfs_can_coalesce_requests(prev, req, desc)) return 0; nfs_list_move_request(req, &mirror->pg_list); @@ -1049,14 +1083,10 @@ static int __nfs_pageio_add_request(struct nfs_pageio_descriptor *desc, pgbase += subreq->wb_bytes; if (bytes_left) { - subreq = nfs_create_request(req->wb_context, - req->wb_page, - subreq, pgbase, bytes_left); + subreq = nfs_create_subreq(req, subreq, pgbase, + offset, bytes_left); if (IS_ERR(subreq)) goto err_ptr; - nfs_lock_request(subreq); - subreq->wb_offset = offset; - subreq->wb_index = req->wb_index; } } while (bytes_left > 0); @@ -1158,19 +1188,14 @@ int nfs_pageio_add_request(struct nfs_pageio_descriptor *desc, lastreq = lastreq->wb_this_page) ; - dupreq = nfs_create_request(req->wb_context, - req->wb_page, lastreq, pgbase, bytes); + dupreq = nfs_create_subreq(req, lastreq, + pgbase, offset, bytes); + nfs_page_group_unlock(req); if (IS_ERR(dupreq)) { - nfs_page_group_unlock(req); desc->pg_error = PTR_ERR(dupreq); goto out_failed; } - - nfs_lock_request(dupreq); - nfs_page_group_unlock(req); - dupreq->wb_offset = offset; - dupreq->wb_index = req->wb_index; } else dupreq = req; diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c index 7066cd7c7aff..83722e936b4a 100644 --- a/fs/nfs/pnfs.c +++ b/fs/nfs/pnfs.c @@ -2436,7 +2436,7 @@ pnfs_generic_pg_init_read(struct nfs_pageio_descriptor *pgio, struct nfs_page *r rd_size = nfs_dreq_bytes_left(pgio->pg_dreq); pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode, - req->wb_context, + nfs_req_openctx(req), req_offset(req), rd_size, IOMODE_READ, @@ -2463,7 +2463,7 @@ pnfs_generic_pg_init_write(struct nfs_pageio_descriptor *pgio, pnfs_generic_pg_check_range(pgio, req); if (pgio->pg_lseg == NULL) { pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode, - req->wb_context, + nfs_req_openctx(req), req_offset(req), wb_size, IOMODE_RW, diff --git a/fs/nfs/pnfs.h b/fs/nfs/pnfs.h index c0420b979d88..f15609c003d8 100644 --- a/fs/nfs/pnfs.h +++ b/fs/nfs/pnfs.h @@ -459,7 +459,7 @@ static inline bool pnfs_mark_request_commit(struct nfs_page *req, struct pnfs_layout_segment *lseg, struct nfs_commit_info *cinfo, u32 ds_commit_idx) { - struct inode *inode = d_inode(req->wb_context->dentry); + struct inode *inode = d_inode(nfs_req_openctx(req)->dentry); struct pnfs_layoutdriver_type *ld = NFS_SERVER(inode)->pnfs_curr_ld; if (lseg == NULL || ld->mark_request_commit == NULL) @@ -471,7 +471,7 @@ pnfs_mark_request_commit(struct nfs_page *req, struct pnfs_layout_segment *lseg, static inline bool pnfs_clear_request_commit(struct nfs_page *req, struct nfs_commit_info *cinfo) { - struct inode *inode = d_inode(req->wb_context->dentry); + struct inode *inode = d_inode(nfs_req_openctx(req)->dentry); struct pnfs_layoutdriver_type *ld = NFS_SERVER(inode)->pnfs_curr_ld; if (ld == NULL || ld->clear_request_commit == NULL) diff --git a/fs/nfs/read.c b/fs/nfs/read.c index 1d95a60b2586..c799e540ed1e 100644 --- a/fs/nfs/read.c +++ b/fs/nfs/read.c @@ -92,7 +92,7 @@ EXPORT_SYMBOL_GPL(nfs_pageio_reset_read_mds); static void nfs_readpage_release(struct nfs_page *req) { - struct inode *inode = d_inode(req->wb_context->dentry); + struct inode *inode = d_inode(nfs_req_openctx(req)->dentry); dprintk("NFS: read done (%s/%llu %d@%lld)\n", inode->i_sb->s_id, (unsigned long long)NFS_FILEID(inode), req->wb_bytes, @@ -118,7 +118,7 @@ int nfs_readpage_async(struct nfs_open_context *ctx, struct inode *inode, len = nfs_page_length(page); if (len == 0) return nfs_return_empty_page(page); - new = nfs_create_request(ctx, page, NULL, 0, len); + new = nfs_create_request(ctx, page, 0, len); if (IS_ERR(new)) { unlock_page(page); return PTR_ERR(new); @@ -363,7 +363,7 @@ readpage_async_filler(void *data, struct page *page) if (len == 0) return nfs_return_empty_page(page); - new = nfs_create_request(desc->ctx, page, NULL, 0, len); + new = nfs_create_request(desc->ctx, page, 0, len); if (IS_ERR(new)) goto out_error; diff --git a/fs/nfs/super.c b/fs/nfs/super.c index 450ae77d19bf..d6c687419a81 100644 --- a/fs/nfs/super.c +++ b/fs/nfs/super.c @@ -78,7 +78,7 @@ enum { /* Mount options that take no arguments */ - Opt_soft, Opt_hard, + Opt_soft, Opt_softerr, Opt_hard, Opt_posix, Opt_noposix, Opt_cto, Opt_nocto, Opt_ac, Opt_noac, @@ -125,6 +125,7 @@ static const match_table_t nfs_mount_option_tokens = { { Opt_sloppy, "sloppy" }, { Opt_soft, "soft" }, + { Opt_softerr, "softerr" }, { Opt_hard, "hard" }, { Opt_deprecated, "intr" }, { Opt_deprecated, "nointr" }, @@ -628,7 +629,8 @@ static void nfs_show_mount_options(struct seq_file *m, struct nfs_server *nfss, const char *str; const char *nostr; } nfs_info[] = { - { NFS_MOUNT_SOFT, ",soft", ",hard" }, + { NFS_MOUNT_SOFT, ",soft", "" }, + { NFS_MOUNT_SOFTERR, ",softerr", "" }, { NFS_MOUNT_POSIX, ",posix", "" }, { NFS_MOUNT_NOCTO, ",nocto", "" }, { NFS_MOUNT_NOAC, ",noac", "" }, @@ -658,6 +660,8 @@ static void nfs_show_mount_options(struct seq_file *m, struct nfs_server *nfss, seq_printf(m, ",acdirmin=%u", nfss->acdirmin/HZ); if (nfss->acdirmax != NFS_DEF_ACDIRMAX*HZ || showdefaults) seq_printf(m, ",acdirmax=%u", nfss->acdirmax/HZ); + if (!(nfss->flags & (NFS_MOUNT_SOFT|NFS_MOUNT_SOFTERR))) + seq_puts(m, ",hard"); for (nfs_infop = nfs_info; nfs_infop->flag; nfs_infop++) { if (nfss->flags & nfs_infop->flag) seq_puts(m, nfs_infop->str); @@ -1239,10 +1243,15 @@ static int nfs_parse_mount_options(char *raw, */ case Opt_soft: mnt->flags |= NFS_MOUNT_SOFT; + mnt->flags &= ~NFS_MOUNT_SOFTERR; break; - case Opt_hard: + case Opt_softerr: + mnt->flags |= NFS_MOUNT_SOFTERR; mnt->flags &= ~NFS_MOUNT_SOFT; break; + case Opt_hard: + mnt->flags &= ~(NFS_MOUNT_SOFT|NFS_MOUNT_SOFTERR); + break; case Opt_posix: mnt->flags |= NFS_MOUNT_POSIX; break; @@ -2476,6 +2485,21 @@ static int nfs_compare_super_address(struct nfs_server *server1, return 1; } +static int nfs_compare_userns(const struct nfs_server *old, + const struct nfs_server *new) +{ + const struct user_namespace *oldns = &init_user_ns; + const struct user_namespace *newns = &init_user_ns; + + if (old->client && old->client->cl_cred) + oldns = old->client->cl_cred->user_ns; + if (new->client && new->client->cl_cred) + newns = new->client->cl_cred->user_ns; + if (oldns != newns) + return 0; + return 1; +} + static int nfs_compare_super(struct super_block *sb, void *data) { struct nfs_sb_mountdata *sb_mntdata = data; @@ -2489,6 +2513,8 @@ static int nfs_compare_super(struct super_block *sb, void *data) return 0; if (memcmp(&old->fsid, &server->fsid, sizeof(old->fsid)) != 0) return 0; + if (!nfs_compare_userns(old, server)) + return 0; return nfs_compare_mount_options(sb, server, mntflags); } diff --git a/fs/nfs/symlink.c b/fs/nfs/symlink.c index 06eb44b47885..25ba299fdac2 100644 --- a/fs/nfs/symlink.c +++ b/fs/nfs/symlink.c @@ -26,8 +26,9 @@ * and straight-forward than readdir caching. */ -static int nfs_symlink_filler(struct inode *inode, struct page *page) +static int nfs_symlink_filler(void *data, struct page *page) { + struct inode *inode = data; int error; error = NFS_PROTO(inode)->readlink(inode, page, 0, PAGE_SIZE); @@ -65,8 +66,8 @@ static const char *nfs_get_link(struct dentry *dentry, err = ERR_PTR(nfs_revalidate_mapping(inode, inode->i_mapping)); if (err) return err; - page = read_cache_page(&inode->i_data, 0, - (filler_t *)nfs_symlink_filler, inode); + page = read_cache_page(&inode->i_data, 0, nfs_symlink_filler, + inode); if (IS_ERR(page)) return ERR_CAST(page); } diff --git a/fs/nfs/write.c b/fs/nfs/write.c index f3ebabaa291d..bc5bb9323412 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -244,6 +244,12 @@ static void nfs_set_pageerror(struct address_space *mapping) nfs_zap_mapping(mapping->host, mapping); } +static void nfs_mapping_set_error(struct page *page, int error) +{ + SetPageError(page); + mapping_set_error(page_file_mapping(page), error); +} + /* * nfs_page_group_search_locked * @head - head request of page group @@ -582,11 +588,10 @@ release_request: return ERR_PTR(ret); } -static void nfs_write_error_remove_page(struct nfs_page *req) +static void nfs_write_error(struct nfs_page *req, int error) { + nfs_mapping_set_error(req->wb_page, error); nfs_end_page_writeback(req); - generic_error_remove_page(page_file_mapping(req->wb_page), - req->wb_page); nfs_release_request(req); } @@ -609,6 +614,7 @@ nfs_error_is_fatal_on_server(int err) static int nfs_page_async_flush(struct nfs_pageio_descriptor *pgio, struct page *page) { + struct address_space *mapping; struct nfs_page *req; int ret = 0; @@ -622,19 +628,19 @@ static int nfs_page_async_flush(struct nfs_pageio_descriptor *pgio, nfs_set_page_writeback(page); WARN_ON_ONCE(test_bit(PG_CLEAN, &req->wb_flags)); - ret = req->wb_context->error; /* If there is a fatal error that covers this write, just exit */ - if (nfs_error_is_fatal_on_server(ret)) + ret = 0; + mapping = page_file_mapping(page); + if (test_bit(AS_ENOSPC, &mapping->flags) || + test_bit(AS_EIO, &mapping->flags)) goto out_launder; - ret = 0; if (!nfs_pageio_add_request(pgio, req)) { ret = pgio->pg_error; /* * Remove the problematic req upon fatal errors on the server */ if (nfs_error_is_fatal(ret)) { - nfs_context_set_write_error(req->wb_context, ret); if (nfs_error_is_fatal_on_server(ret)) goto out_launder; } else @@ -646,8 +652,8 @@ static int nfs_page_async_flush(struct nfs_pageio_descriptor *pgio, out: return ret; out_launder: - nfs_write_error_remove_page(req); - return ret; + nfs_write_error(req, ret); + return 0; } static int nfs_do_writepage(struct page *page, struct writeback_control *wbc, @@ -958,7 +964,8 @@ static void nfs_clear_request_commit(struct nfs_page *req) { if (test_bit(PG_CLEAN, &req->wb_flags)) { - struct inode *inode = d_inode(req->wb_context->dentry); + struct nfs_open_context *ctx = nfs_req_openctx(req); + struct inode *inode = d_inode(ctx->dentry); struct nfs_commit_info cinfo; nfs_init_cinfo_from_inode(&cinfo, inode); @@ -999,10 +1006,12 @@ static void nfs_write_completion(struct nfs_pgio_header *hdr) if (test_bit(NFS_IOHDR_ERROR, &hdr->flags) && (hdr->good_bytes < bytes)) { nfs_set_pageerror(page_file_mapping(req->wb_page)); - nfs_context_set_write_error(req->wb_context, hdr->error); + nfs_mapping_set_error(req->wb_page, hdr->error); goto remove_req; } if (nfs_write_need_commit(hdr)) { + /* Reset wb_nio, since the write was successful. */ + req->wb_nio = 0; memcpy(&req->wb_verf, &hdr->verf.verifier, sizeof(req->wb_verf)); nfs_mark_request_commit(req, hdr->lseg, &cinfo, hdr->pgio_mirror_idx); @@ -1136,6 +1145,7 @@ static struct nfs_page *nfs_try_to_update_request(struct inode *inode, req->wb_bytes = end - req->wb_offset; else req->wb_bytes = rqend - req->wb_offset; + req->wb_nio = 0; return req; out_flushme: /* @@ -1165,7 +1175,7 @@ static struct nfs_page * nfs_setup_write_request(struct nfs_open_context* ctx, req = nfs_try_to_update_request(inode, page, offset, bytes); if (req != NULL) goto out; - req = nfs_create_request(ctx, page, NULL, offset, bytes); + req = nfs_create_request(ctx, page, offset, bytes); if (IS_ERR(req)) goto out; nfs_inode_add_request(inode, req); @@ -1210,7 +1220,7 @@ int nfs_flush_incompatible(struct file *file, struct page *page) return 0; l_ctx = req->wb_lock_context; do_flush = req->wb_page != page || - !nfs_match_open_context(req->wb_context, ctx); + !nfs_match_open_context(nfs_req_openctx(req), ctx); if (l_ctx && flctx && !(list_empty_careful(&flctx->flc_posix) && list_empty_careful(&flctx->flc_flock))) { @@ -1410,8 +1420,10 @@ static void nfs_initiate_write(struct nfs_pgio_header *hdr, */ static void nfs_redirty_request(struct nfs_page *req) { + /* Bump the transmission count */ + req->wb_nio++; nfs_mark_request_dirty(req); - set_bit(NFS_CONTEXT_RESEND_WRITES, &req->wb_context->flags); + set_bit(NFS_CONTEXT_RESEND_WRITES, &nfs_req_openctx(req)->flags); nfs_end_page_writeback(req); nfs_release_request(req); } @@ -1423,14 +1435,10 @@ static void nfs_async_write_error(struct list_head *head, int error) while (!list_empty(head)) { req = nfs_list_entry(head->next); nfs_list_remove_request(req); - if (nfs_error_is_fatal(error)) { - nfs_context_set_write_error(req->wb_context, error); - if (nfs_error_is_fatal_on_server(error)) { - nfs_write_error_remove_page(req); - continue; - } - } - nfs_redirty_request(req); + if (nfs_error_is_fatal(error)) + nfs_write_error(req, error); + else + nfs_redirty_request(req); } } @@ -1735,7 +1743,8 @@ void nfs_init_commit(struct nfs_commit_data *data, struct nfs_commit_info *cinfo) { struct nfs_page *first = nfs_list_entry(head->next); - struct inode *inode = d_inode(first->wb_context->dentry); + struct nfs_open_context *ctx = nfs_req_openctx(first); + struct inode *inode = d_inode(ctx->dentry); /* Set up the RPC argument and reply structs * NB: take care not to mess about with data->commit et al. */ @@ -1743,7 +1752,7 @@ void nfs_init_commit(struct nfs_commit_data *data, list_splice_init(head, &data->pages); data->inode = inode; - data->cred = first->wb_context->cred; + data->cred = ctx->cred; data->lseg = lseg; /* reference transferred */ /* only set lwb for pnfs commit */ if (lseg) @@ -1756,7 +1765,7 @@ void nfs_init_commit(struct nfs_commit_data *data, /* Note: we always request a commit of the entire inode */ data->args.offset = 0; data->args.count = 0; - data->context = get_nfs_open_context(first->wb_context); + data->context = get_nfs_open_context(ctx); data->res.fattr = &data->fattr; data->res.verf = &data->verf; nfs_fattr_init(&data->fattr); @@ -1839,14 +1848,15 @@ static void nfs_commit_release_pages(struct nfs_commit_data *data) nfs_clear_page_commit(req->wb_page); dprintk("NFS: commit (%s/%llu %d@%lld)", - req->wb_context->dentry->d_sb->s_id, - (unsigned long long)NFS_FILEID(d_inode(req->wb_context->dentry)), + nfs_req_openctx(req)->dentry->d_sb->s_id, + (unsigned long long)NFS_FILEID(d_inode(nfs_req_openctx(req)->dentry)), req->wb_bytes, (long long)req_offset(req)); if (status < 0) { - nfs_context_set_write_error(req->wb_context, status); - if (req->wb_page) + if (req->wb_page) { + nfs_mapping_set_error(req->wb_page, status); nfs_inode_remove_request(req); + } dprintk_cont(", error = %d\n", status); goto next; } @@ -1863,7 +1873,7 @@ static void nfs_commit_release_pages(struct nfs_commit_data *data) /* We have a mismatch. Write the page again */ dprintk_cont(" mismatch\n"); nfs_mark_request_dirty(req); - set_bit(NFS_CONTEXT_RESEND_WRITES, &req->wb_context->flags); + set_bit(NFS_CONTEXT_RESEND_WRITES, &nfs_req_openctx(req)->flags); next: nfs_unlock_and_release_request(req); /* Latency breaker */ |